START-INFO-DIR-ENTRY
* Gambit-C: (gambit-c).		A portable implementation of Scheme.
* gsi: (gambit-c) interpreter.	Gambit interpreter.
* gsc: (gambit-c) compiler.	Gambit compiler.
END-INFO-DIR-ENTRY

   This file documents Gambit-C, a portable implementation of Scheme.

   Copyright (C) 1994-2006 Marc Feeley.

   Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.

   Permission is granted to copy and distribute modified versions of
this manual under the conditions for verbatim copying, provided that
the entire resulting derived work is distributed under the terms of a
permission notice identical to this one.

   Permission is granted to copy and distribute translations of this
manual into another language, under the above conditions for modified
versions, except that this permission notice may be stated in a
translation approved by the copyright holder.

Gambit-C
********

This manual documents Gambit-C.  It covers release 4.0 beta 20.

1 The Gambit-C system
*********************

The Gambit programming system is a full implementation of the Scheme
language which conforms to the R4RS and IEEE Scheme standards.  It
consists of two main programs: `gsi', the Gambit Scheme interpreter,
and `gsc', the Gambit Scheme compiler.

   Gambit-C is a version of the Gambit programming system in which the
compiler generates portable C code, making the whole Gambit-C system and
the programs compiled with it easily portable to many computer
architectures for which a C compiler is available.  With appropriate
declarations in the source code the executable programs generated by
the compiler run roughly as fast as equivalent C programs.

   For the most up to date information on Gambit and add-on packages
please check the Gambit web page at
`http://www.iro.umontreal.ca/~gambit' or the mailing list at
`http://mailman.iro.umontreal.ca/mailman/listinfo/gambit-list'.  Bug
reports and inquiries should be sent to `gambit@iro.umontreal.ca'.

1.1 Accessing the system files
==============================

The "Gambit installation directory" is where all system files are
installed.  This directory is `prefix/version', where `version' is the
system version number (e.g. 4.1.55 for Gambit 4.1.55) and `prefix' is
`/usr/local/Gambit-C' under UNIX and Mac OS X and `C:/Gambit-C' under
Microsoft Windows.  The prefix can be overridden when the system is
built with the command `configure --prefix=/my/own/Gambit-C'.
Moreover, `prefix/current' is a symbolic link which points to the
installation directory.

   Executable programs such as the interpreter `gsi' and compiler `gsc'
can be found in the `bin' subdirectory of the installation directory.
Adding this directory to the `PATH' environment variable allows these
programs to be started by simply entering their name.

   The runtime library is located in the `lib' subdirectory.  When the
system's runtime library is built as a shared-library (with the command
`configure --enable-shared') all programs built with Gambit-C,
including the interpreter and compiler, need to find this library when
they are executed and consequently this directory must be in the path
searched by the system for shared-libraries.  This path is normally
specified through an environment variable which is `LD_LIBRARY_PATH' on
most versions of UNIX, `LIBPATH' on AIX, `SHLIB_PATH' on HPUX,
`DYLD_LIBRARY_PATH' on Mac OS X, and `PATH' on Microsoft Windows.  If
the shell is `sh', the setting of the path can be made for a single
execution by prefixing the program name with the environment variable
assignment, as in:

     $ LD_LIBRARY_PATH=/usr/local/Gambit-C/current/lib gsi

   A similar problem exists with the Gambit header file `gambit.h',
located in the `include' subdirectory.  This header file is needed for
compiling Scheme programs with the Gambit-C compiler.  When the C
compiler is being called explicitly it may be necessary to use a
`-I<DIR>' command line option to indicate where to find header files
and a `-L<DIR>' command line option to indicate where to find
libraries.  Access to both of these files can be simplified by creating
a link to them in the appropriate system directories (special
privileges may however be required):

     $ ln -s /usr/local/Gambit-C/current/lib/libgambc.a /usr/lib # name may vary
     $ ln -s /usr/local/Gambit-C/current/include/gambit.h /usr/include

   This is not done by the installation process.  Alternatively these
files can also be copied or linked in the directory where the C
compiler is invoked (this requires no special privileges).

2 The Gambit Scheme interpreter
*******************************

Synopsis:

     gsi [-:RUNTIMEOPTION,...] [-i] [-f] [-v] [[-] [-e EXPRESSIONS] [FILE]]...

   The interpreter is executed in "interactive mode" when no file or
`-' or `-e' option is given on the command line.  When at least one
file or `-' or `-e' option is present the interpreter is executed in
"batch mode".  The `-i' option is ignored by the interpreter.  The
initialization file will be examined unless the `-f' option is present
(*note GSI customization::).  The `-v' option prints the system version
string on standard output and exits.  Runtime options are explained in
*Note Runtime options::.

2.1 Interactive mode
====================

In interactive mode a read-eval-print loop (REPL) is started for the
user to interact with the interpreter.  At each iteration of this loop
the interpreter displays a prompt, reads a command and executes it.
The commands can be expressions to evaluate (the typical case) or
special commands related to debugging, for example `,q' to terminate
the current thread (for a complete list of commands see *Note
Debugging::).  Most commands produce some output, such as the value or
error message resulting from an evaluation.

   The input and output of the interaction is done on the "interaction
channel".  The interaction channel can be specified through the runtime
options but if none is specified the system uses a reasonable default
that depends on the system's configuration.  When the system's runtime
library was built with support for GUIDE, the Gambit Universal IDE
(with the command `configure --enable-guide') the interaction channel
corresponds to the "console window" of the primordial thread (for
details see *Note GUIDE::), otherwise the interaction channel is the
user's "console", also known as the "controlling terminal" in the UNIX
world.  When the REPL starts, the ports associated with
`(current-input-port)', `(current-output-port)' and
`(current-error-port)' all refer to the interaction channel.

   Expressions are evaluated in the global "interaction environment".
The interpreter adds to this environment any definition entered using
the `define' and `define-macro' special forms.  Once the evaluation of
an expression is completed, the value or values resulting from the
evaluation are output to the interaction channel by the pretty printer.
The special "void" object is not output.  This object is returned by
most procedures and special forms which the Scheme standard defines as
returning an unspecified value (e.g. `write', `set!', `define').

   Here is a sample interaction with `gsi':

     $ gsi
     Gambit Version 4.0 beta 20

     > (define (fact n) (if (< n 2) 1 (* n (fact (- n 1)))))
     > (map fact '(1 2 3 4 5 6))
     (1 2 6 24 120 720)
     > (values (fact 10) (fact 40))
     3628800
     815915283247897734345611269596115894272000000000
     > ,q

   What happens when errors occur is explained in *Note Debugging::.

2.2 Batch mode
==============

In batch mode the command line arguments denote files to be loaded,
REPL interactions to start (`-' option), and expressions to be
evaluated (`-e' option).  Note that the `-' and `-e' options can be
interspersed with the files on the command line and can occur multiple
times.  The interpreter processes the command line arguments from left
to right, loading files with the `load' procedure and evaluating
expressions with the `eval' procedure in the global interaction
environment.  After this processing the interpreter exits.

   When the file name has no extension the `load' procedure first
attempts to load the file with no extension as a Scheme source file.
If that file doesn't exist it completes the file name with a `.oN'
extension with the highest consecutive version number starting with 1,
and loads that file as an object file.  If that file doesn't exist the
file extensions `.scm' and `.six' will be tried in that order.  When
the file name has an extension, the `load' procedure will only attempt
to load the file with that specific name.

   When the extension of the file loaded is `.scm' the content of the
file will be parsed using the normal Scheme prefix syntax.  When the
extension of the file loaded is `.six' the content of the file will be
parsed using the Scheme infix syntax extension (see *Note Scheme infix
syntax extension::).  Otherwise, `gsi' will parse the file using the
normal Scheme prefix syntax.

   The ports associated with `(current-input-port)',
`(current-output-port)' and `(current-error-port)' initially refer
respectively to the standard input (`stdin'), standard output
(`stdout') and the standard error (`stderr') of the interpreter.  This
is true even in REPLs started with the `-' option.  The usual
interaction channel (console or IDE's console window) is still used to
read expressions and commands and to display results.  This makes it
possible to use REPLs to debug programs which read the standard input
and write to the standard output, even when these have been redirected.

   Here is a sample use of the interpreter in batch mode, under UNIX:

     $ cat h.scm
     (display "hello") (newline)
     $ cat w.six
     display("world"); newline();
     $ gsi h.scm - w.six -e "(pretty-print 1)(pretty-print 2)"
     hello
     > (define (display x) (write (reverse (string->list x))))
     > ,(c 0)
     (#\d #\l #\r #\o #\w)
     1
     2

2.3 Customization
=================

There are two ways to customize the interpreter.  When the interpreter
starts off it tries to execute a `(load "~~/gambcext")' (for an
explanation of how file names are interpreted see *Note Host
environment::).  An error is not signaled when the file does not exist.
Interpreter extensions and patches that are meant to apply to all
users and all modes should go in that file.

   Extensions which are meant to apply to a single user or to a specific
working directory are best placed in the "initialization file", which
is a file containing Scheme code.  In all modes, the interpreter first
tries to locate the initialization file by searching the following
locations: `gambcini' and `~/gambcini' (with no extension, a `.scm'
extension, and a `.six' extension in that order).  The first file that
is found is examined as though the expression `(include
INITIALIZATION-FILE)' had been entered at the read-eval-print loop
where INITIALIZATION-FILE is the file that was found.  Note that by
using an `include' the macros defined in the initialization file will
be visible from the read-eval-print loop (this would not have been the
case if `load' had been used).  The initialization file is not searched
for or examined when the `-f' option is specified.

2.4 Process exit status
=======================

The status is zero when the interpreter exits normally and is nonzero
when the interpreter exits due to an error.  Here is the meaning of the
exit statuses:

`0'
     The execution of the primordial thread (i.e. the main thread) did
     not encounter any error.  It is however possible that other threads
     terminated abnormally (by default threads other than the primordial
     thread terminate silently when they raise an exception that is not
     handled).

`64'
     The runtime options or the environment variable `GAMBCOPT'
     contained a syntax error or were invalid.

`70'
     This normally indicates that an exception was raised in the
     primordial thread and the exception was not handled.

`71'
     There was a problem initializing the runtime system, for example
     insufficient memory to allocate critical tables.


   For example, if the shell is `sh':

     $ gsi -:d0 -e "(pretty-print (expt 2 100))"
     1267650600228229401496703205376
     $ echo $?
     0
     $ gsi -:d0,unknown # try to use an unknown runtime option
     $ echo $?
     64
     $ gsi -:d0 nonexistent.scm # try to load a file that does not exist
     $ echo $?
     70
     $ gsi nonexistent.scm
     *** ERROR IN ##main -- No such file or directory
     (load "nonexistent.scm")
     $ echo $?
     70

     $ gsi -:m4000000 # ask for a 4 gigabyte heap
     *** malloc: vm_allocate(size=528384) failed (error code=3)
     *** malloc[15068]: error: Can't allocate region
     $ echo $?
     71

   Note the use of the runtime option `-:d0' that prevents error
messages from being output, and the runtime option `-:m4000000' which
sets the minimum heap size to 4 gigabytes.

2.5 Scheme scripts
==================

The `load' procedure treats specially files that begin with the two
characters `#!' and `@;'.  Such files are called "script files".  In
addition to indicating that the file is a script, the first line
provides information about the source code language to be used by the
`load' procedure.  After the two characters `#!' and `@;' the system
will search for the first substring matching one of the following
language specifying tokens:

`scheme-r4rs'
     R4RS language with prefix syntax, case-insensitivity, keyword
     syntax not supported

`scheme-r5rs'
     R5RS language with prefix syntax, case-insensitivity, keyword
     syntax not supported

`scheme-ieee-1178-1990'
     IEEE 1178-1990 language with prefix syntax, case-insensitivity,
     keyword syntax not supported

`scheme-srfi-0'
     R5RS language with prefix syntax and SRFI 0 support (i.e.
     `cond-expand' special form), case-insensitivity, keyword syntax
     not supported

`gsi-script'
     Full Gambit Scheme language with prefix syntax, case-sensitivity,
     keyword syntax supported

`gsc-script'
     Full Gambit Scheme language with prefix syntax, case-sensitivity,
     keyword syntax supported

`six-script'
     Full Gambit Scheme language with infix syntax, case-sensitivity,
     keyword syntax supported


   If a language specifying token is not found, `load' will use the
same language as a nonscript file (i.e. it uses the file extension and
runtime system options to determine the language).

   After processing the first line, `load' will parse the rest of the
file (using the syntax of the language indicated) and then execute it.
When the file is being loaded because it is an argument on the
interpreter's command line, the interpreter will:

   * Setup the `command-line' procedure so that it returns a list
     containing the expanded file name of the script file and the
     arguments following the script file on the command line.  This is
     done before the script is executed.  The expanded file name of the
     script file can be used to determine the directory that contains
     the script (i.e. `(path-directory (car (command-line)))').

   * After the script is loaded the procedure `main' is called with the
     command-line arguments.  The way this is done depends on the
     language specifying token.  For `scheme-r4rs', `scheme-r5rs',
     `scheme-ieee-1178-1990', and `scheme-srfi-0', the `main' procedure
     is called with the equivalent of `(main (cdr (command-line)))' and
     `main' is expected to return a process exit status code in the
     range 0 to 255.  This conforms to the "Running Scheme Scripts on
     Unix SRFI" (SRFI 22).  For `gsi-script' and `six-script' the `main'
     procedure is called with the equivalent of `(apply main (cdr
     (command-line)))' and the process exit status code is 0 (`main''s
     result is ignored).  The Gambit-C system has a predefined `main'
     procedure which accepts any number of arguments and returns 0, so
     it is perfectly valid for a script to not define `main' and to do
     all its processing with top-level expressions (examples are given
     in the next section).

   * When `main' returns, the interpreter exits.  The command-line
     arguments after a script file are consequently not processed
     (however they do appear in the list returned by the `command-line'
     procedure, after the script file's expanded file name, so it is up
     to the script to process them).


2.5.1 Scripts under UNIX and Mac OS X
-------------------------------------

Under UNIX and Mac OS X, the Gambit-C installation process creates the
executable `gsi' and also the executables `six', `gsi-script',
`six-script', `scheme-r5rs', `scheme-srfi-0', etc as links to `gsi'.  A
Scheme script need only start with the name of the desired Scheme
language variant prefixed with `#!' and the directory where the Gambit-C
executables are stored.  This script should be made executable by
setting the execute permission bits (with a `chmod +x SCRIPT').  Here
is an example of a script which lists on standard output the files in
the current directory:

     #!/usr/local/Gambit-C/current/bin/gsi-script
     (for-each pretty-print (directory-files))

   Here is another UNIX script, using the Scheme infix syntax extension,
which takes a single integer argument and prints on standard output the
numbers from 1 to that integer:

     #!/usr/local/Gambit-C/current/bin/six-script

     void main (obj n_str)
     {
       int n = \string->number(n_str);
       for (int i=1; i<=n; i++)
         \pretty-print(i);
     }

   For maximal portability it is a good idea to start scripts indirectly
through the `/usr/bin/env' program, so that the executable of the
interpreter will be searched in the user's `PATH'.  This is what SRFI
22 recommends.  For example here is a script that mimics the UNIX `cat'
utility for text files:

     #!/usr/bin/env gsi-script

     (define (display-file filename)
       (display (call-with-input-file filename
                  (lambda (port)
                    (read-line port #f)))))

     (for-each display-file (cdr (command-line)))

2.5.2 Scripts under Microsoft Windows
-------------------------------------

Under Microsoft Windows, the Gambit-C installation process creates the
executable `gsi.exe' and `six.exe' and also the batch files
`gsi-script.bat', `six-script.bat', `scheme-r5rs.bat',
`scheme-srfi-0.bat', etc which simply invoke `gsi.exe' with the same
command line arguments.  A Scheme script need only start with the name
of the desired Scheme language variant prefixed with `@;'.  A UNIX
script can be converted to a Microsoft Windows script simply by
changing the first line and storing the script in a file whose name has
a `.bat' or `.cmd' extension:

     @;gsi-script %~f0 %*
     (display "files:\n")
     (pretty-print (directory-files))

   Note that Microsoft Windows always searches executables in the user's
`PATH', so there is no need for an indirection such as the UNIX
`/usr/bin/env'.  However the first line must end with `%~f0 %*' to pass
the expanded filename of the script and command line arguments to the
interpreter.

2.5.3 Compiling scripts
-----------------------

A script file can be compiled using the Gambit Scheme compiler (*note
GSC::) into a dynamically loadable object file or into a standalone
executable.  The first line of the script will provide information to
the compiler on which language to use.  The first line also provides
information on which runtime options to use when executing the script.
The compiled script will be executed similarly to an interpreted script
(i.e. the list of command line arguments returned by the `command-line'
procedure and the invocation of the `main' procedure).

   For example:

     $ cat square.scm
     #!/usr/local/Gambit-C/current/bin/gsi-script
     (define (main arg)
       (pretty-print (expt (string->number arg) 2)))
     $ gsi square 30   # will load square.scm
     900
     $ gsc square
     $ gsi square 30   # will load square.o1
     900

3 The Gambit Scheme compiler
****************************

Synopsis:

     gsc [-:RUNTIMEOPTION,...] [-i] [-f] [-v]
         [-prelude EXPRESSIONS] [-postlude EXPRESSIONS]
         [-dynamic] [-cc-options OPTIONS] [-ld-options OPTIONS]
         [-warnings] [-verbose] [-report] [-expansion]
         [-gvm] [-debug] [-track-scheme]
         [-o OUTPUT] [-c] [-link] [-flat] [-l BASE]
         [[-] [-e EXPRESSIONS] [FILE]]...

3.1 Interactive mode
====================

When no command line argument is present other than options the
compiler behaves like the interpreter in interactive mode.  The only
difference with the interpreter is that the compilation related
procedures listed in this chapter are also available (i.e.
`compile-file', `compile-file-to-c', etc).

3.2 Customization
=================

Like the interpreter, the compiler will examine the initialization file
unless the `-f' option is specified.

3.3 Batch mode
==============

In batch mode `gsc' takes a set of file names (with either no
extension, or a `.c' extension, or some other extension) on the command
line and compiles each Scheme file into a C file.  The extension can be
omitted from FILE when the Scheme file has a `.scm' or `.six'
extension.  When the extension of the Scheme file is `.six' the content
of the file will be parsed using the Scheme infix syntax extension (see
*Note Scheme infix syntax extension::). Otherwise, `gsc' will parse the
Scheme file using the normal Scheme prefix syntax.  Files with a `.c'
extension must have been previously produced by `gsc' and are used by
Gambit's linker.

   For each Scheme file a C file `FILE.c' will be produced.  The C
file's name is the same as the Scheme file, but the extension is
changed to `.c' and it is stripped of its directory (i.e. the C file is
created in the current working directory).

   The C files produced by the compiler serve two purposes.  They will
be processed by a C compiler to generate object files, and they also
contain information to be read by Gambit's linker to generate a "link
file".  The link file is a C file that collects various linking
information for a group of modules, such as the set of all symbols and
global variables used by the modules.  The linker is only invoked when
the `-link' option appears on the command line.

   Compiler options must be specified before the first file name and
after the `-:' runtime option (*note Runtime options::).  If present,
the `-i', `-f', and `-v' compiler options must come first.  The
available options are:

`-i'
     Force interpreter mode.

`-f'
     Do not examine the initialization file.

`-v'
     Print the system version number on standard output and exit.

`-prelude EXPRESSIONS'
     Add expressions to the top of the source code being compiled.

`-postlude EXPRESSIONS'
     Add expressions to the bottom of the source code being compiled.

`-cc-options OPTIONS'
     Add options to the command that invokes the C compiler.

`-ld-options OPTIONS'
     Add options to the command that invokes the C linker.

`-warnings'
     Display warnings.

`-verbose'
     Display a trace of the compiler's activity.

`-report'
     Display a global variable usage report.

`-expansion'
     Display the source code after expansion.

`-gvm'
     Generate a listing of the GVM code.

`-debug'
     Include debugging information in the code generated.

`-track-scheme'
     Generate `#line' directives referring back to the Scheme code.

`-o OUTPUT'
     Set name of output file.

`-dynamic'
     Compile Scheme source files to dynamically loadable object files
     (this is the default).

`-c'
     Compile Scheme source files to C without generating link file.

`-link'
     Compile Scheme source files to C and generate a link file.

`-flat'
     Generate a flat link file instead of the default incremental link
     file.

`-l BASE'
     Specify the link file of the base library to use for the link.

`-'
     Start REPL interaction.

`-e EXPRESSIONS'
     Evaluate expressions in the interaction environment.

   The `-i' option forces the compiler to process the remaining command
line arguments like the interpreter.

   The `-prelude' option adds the specified expressions to the top of
the source code being compiled.  The main use of this option is to
supply declarations on the command line.  For example the following
invocation of the compiler will compile the file `bench.scm' in unsafe
mode:

     $ gsc -prelude "(declare (not safe))" bench.scm

   The `-postlude' option adds the specified expressions to the bottom
of the source code being compiled.  The main use of this option is to
supply the expression that will start the execution of the program.  For
example:

     $ gsc -postlude "(start-bench)" bench.scm

   The `-cc-options' option is only meaningful when a dynamically
loadable object file is being generated (neither the `-c' or `-link'
options are used).  The `-cc-options' option adds the specified options
to the command that invokes the C compiler.  The main use of this
option is to specify the include path, some symbols to define or
undefine, the optimization level, or any C compiler option that is
different from the default.  For example:

     $ gsc -cc-options "-U___SINGLE_HOST -O2 -I../include" bench.scm

   The `-ld-options' option is only meaningful when a dynamically
loadable object file is being generated (neither the `-c' or `-link'
options are used).  The `-ld-options' option adds the specified options
to the command that invokes the C linker.  The main use of this option
is to specify additional object files or libraries that need to be
linked, or any C linker option that is different from the default (such
as the library search path and flags to select between static and
dynamic linking).  For example:

     $ gsc -ld-options "-L/usr/X11R6/lib -lX11 -dynamic" bench.scm

   The `-warnings' option displays on standard output all warnings that
the compiler may have.

   The `-verbose' option displays on standard output a trace of the
compiler's activity.

   The `-report' option displays on standard output a global variable
usage report.  Each global variable used in the program is listed with
4 flags that indicate whether the global variable is defined,
referenced, mutated and called.

   The `-expansion' option displays on standard output the source code
after expansion and inlining by the front end.

   The `-gvm' option generates a listing of the intermediate code for
the "Gambit Virtual Machine" (GVM) of each Scheme file on `FILE.gvm'.

   The `-debug' option causes debugging information to be saved in the
code generated.  With this option run time error messages indicate the
source code and its location, the backtraces are more precise, and the
`pp' procedure will display the source code of compiled procedures.
The debugging information is large (the size of the object file is
typically 2 to 4 times bigger).

   The `-track-scheme' options causes the generation of `#line'
directives that refer back to the Scheme source code.  This allows the
use of a C debugger to debug Scheme code.

   The `-o' option sets the name of the output file generated by the
compiler.  When a link file is being generated the name specified is
that of the link file.  Otherwise the name specified is that of the C
file (this option is ignored when the compiler is generating more than
one output file or is generating a dynamically loadable object file).

   If the `-link' option appears on the command line, the Gambit linker
is invoked to generate the link file from the set of C files specified
on the command line or produced by the Gambit compiler.  Unless the
name is specified explicitly with the `-o' option, the link file is
named `LAST_.c', where `LAST.c' is the last file in the set of C files.
When the `-c' option is specified, the Scheme source files are
compiled to C files.  If neither the `-link' or `-c' options appear on
the command line, the Scheme source files are compiled to dynamically
loadable object files (`.oN' extension).

   The `-flat' option is only meaningful when a link file is being
generated (i.e. the `-link' option also appears on the command line).
The `-flat' option directs the Gambit linker to generate a flat link
file.  By default, the linker generates an incremental link file (see
the next section for a description of the two types of link files).

   The `-l' option is only meaningful when an incremental link file is
being generated (i.e. the `-link' option appears on the command line
and the `-flat' option is absent).  The `-l' option specifies the link
file (without the `.c' extension) of the base library to use for the
incremental link.  By default the link file of the Gambit runtime
library is used (i.e. `~~/lib/_gambc.c').

   The `-' option starts a REPL interaction.

   The `-e' option evaluates the specified expressions in the
interaction environment.

3.4 Link files
==============

Gambit can be used to create programs and libraries of Scheme modules.
This section explains the steps required to do so and the role played
by the link files.

   In general, a program is composed of a set of Scheme modules and C
modules.  Some of the modules are part of the Gambit runtime library and
the other modules are supplied by the user.  When the program is
started it must setup various global tables (including the symbol table
and the global variable table) and then sequentially execute the Scheme
modules (more or less as though they were being loaded one after
another).  The information required for this is contained in one or
more "link files" generated by the Gambit linker from the C files
produced by the Gambit compiler.

   The order of execution of the Scheme modules corresponds to the
order of the modules on the command line which produced the link file.
The order is usually important because most modules define variables and
procedures which are used by other modules (for this reason the
program's main computation is normally started by the last module).

   When a single link file is used to contain the linking information of
all the Scheme modules it is called a "flat link file".  Thus a program
built with a flat link file contains in its link file both information
on the user modules and on the runtime library.  This is fine if the
program is to be statically linked but is wasteful in a shared-library
context because the linking information of the runtime library can't be
shared and will be duplicated in all programs (this linking information
typically takes hundreds of kilobytes).

   Flat link files are mainly useful to bundle multiple Scheme modules
to make a runtime library (such as the Gambit runtime library) or to
make a single file that can be loaded with the `load' procedure.

   An "incremental link file" contains only the linking information
that is not already contained in a second link file (the "base" link
file).  Assuming that a flat link file was produced when the runtime
library was linked, a program can be built by linking the user modules
with the runtime library's link file, producing an incremental link
file.  This allows the creation of a shared-library which contains the
modules of the runtime library and its flat link file.  The program is
dynamically linked with this shared-library and only contains the user
modules and the incremental link file.  For small programs this
approach greatly reduces the size of the program because the
incremental link file is small.  A "hello world" program built this way
can be as small as 5 Kbytes.  Note that it is perfectly fine to use an
incremental link file for statically linked programs (there is very
little loss compared to a single flat link file).

   Incremental link files may be built from other incremental link
files.  This allows the creation of shared-libraries which extend the
functionality of the Gambit runtime library.

3.4.1 Building an executable program
------------------------------------

The simplest way to create an executable program is to call up `gsc' to
compile each Scheme module into a C file and create an incremental link
file.  The C files and the link file must then be compiled with a C
compiler and linked (at the object file level) with the Gambit runtime
library and possibly other libraries (such as the math library and the
dynamic loading library).

   Here is for example how a program with three modules (one in C and
two in Scheme) can be built.  The content of the three source files
(`m1.c', `m2.scm' and `m3.scm') is:

     /* File: "m1.c" */
     int power_of_2 (int x) { return 1<<x; }

     ; File: "m2.scm"
     (c-declare "extern int power_of_2 ();")
     (define pow2 (c-lambda (int) int "power_of_2"))
     (define (twice x) (cons x x))

     ; File: "m3.scm"
     (write (map twice (map pow2 '(1 2 3 4)))) (newline)

   The compilation of the two Scheme source files can be done with
three invocations of `gsc':

     $ gsc -c m2.scm        # create m2.c (note: .scm is optional)
     $ gsc -c m3.scm        # create m3.c (note: .scm is optional)
     $ gsc -link m2.c m3.c  # create the incremental link file m3_.c

   Alternatively, the three invocations of `gsc' can be replaced by a
single invocation:

     $ gsc -link m2 m3
     m2:
     m3:

   At this point there will be 4 C files: `m1.c', `m2.c', `m3.c', and
`m3_.c'.  To produce an executable program these files must be compiled
with a C compiler and linked with the Gambit-C runtime library.  The C
compiler options needed will depend on the C compiler and the operating
system (in particular it may be necessary to add the options
`-I/usr/local/Gambit-C/current/include
-L/usr/local/Gambit-C/current/lib' to access the `gambit.h' header file
and the Gambit-C runtime library).

   Here is an example under Mac OS X:

     $ uname -srmp
     Darwin 8.1.0 Power Macintosh powerpc
     $ gcc m1.c m2.c m3.c m3_.c -lgambc
     $ ./a.out
     ((2 . 2) (4 . 4) (8 . 8) (16 . 16))

   Here is an example under Linux:

     $ uname -srmp
     Linux 2.6.8-1.521 i686 athlon
     $ gcc m1.c m2.c m3.c m3_.c -lgambc -lm -ldl -lutil
     $ ./a.out
     ((2 . 2) (4 . 4) (8 . 8) (16 . 16))

3.4.2 Building a loadable library
---------------------------------

To bundle multiple modules into a single object file that can be
dynamically loaded with the `load' procedure, a flat link file is
needed.  The compiler's `-o' option must be used to name the C file
generated.  If the dynamically loadable object file is to be named
`MYFILE.oN' then the `-o' option must set the name of the link file
generated to `MYFILE.oN.c' (note that the `.c' extension could also be
`.cc', `.cpp' or whatever extension is appropriate for C/C++ source
files).  The three modules of the previous example can be bundled by
generating a link file in this way:

     $ gsc -link -flat -o foo.o1.c m2 m3
     m2:
     m3:
     *** WARNING -- "cons" is not defined,
     ***            referenced in: ("m2.c")
     *** WARNING -- "map" is not defined,
     ***            referenced in: ("m3.c")
     *** WARNING -- "newline" is not defined,
     ***            referenced in: ("m3.c")
     *** WARNING -- "write" is not defined,
     ***            referenced in: ("m3.c")

   The warnings indicate that there are no definitions (`define's or
`set!'s) of the variables `cons', `map', `newline' and `write' in the
set of modules being linked.  Before `foo.o1' is loaded, these
variables will have to be bound; either implicitly (by the runtime
library) or explicitly.

   When compiling the C files and link file generated, the flag
`-D___DYNAMIC' must be passed to the C compiler and the C compiler and
linker must be told to generate a dynamically loadable shared library.

   Here is an example under Mac OS X:

     $ uname -srmp
     Darwin 8.1.0 Power Macintosh powerpc
     $ gsc -link -flat -o foo.o1.c m2 m3 > /dev/null
     m2:
     m3:
     $ gcc -bundle -D___DYNAMIC m1.c m2.c m3.c foo.o1.c -o foo.o1
     $ gsi foo.o1
     ((2 . 2) (4 . 4) (8 . 8) (16 . 16))

   Here is an example under Linux:

     $ uname -srmp
     Linux 2.6.8-1.521 i686 athlon
     $ gsc -link -flat -o foo.o1.c m2 m3 > /dev/null
     m2:
     m3:
     $ gcc -shared -D___DYNAMIC m1.c m2.c m3.c foo.o1.c -o foo.o1
     $ gsi foo.o1
     ((2 . 2) (4 . 4) (8 . 8) (16 . 16))

   Here is a more complex example, under Solaris, which shows how to
build a loadable library `mymod.o1' composed of the files `m4.scm',
`m5.scm' and `x.c' that links to system shared libraries (for
X-windows):

     $ uname -srmp
     SunOS ungava 5.6 Generic_105181-05 sun4m sparc SUNW,SPARCstation-20
     $ gsc -link -flat -o mymod.o1.c m4 m5
     m4:
     m5:
     *** WARNING -- "*" is not defined,
     ***            referenced in: ("m4.c")
     *** WARNING -- "+" is not defined,
     ***            referenced in: ("m5.c")
     *** WARNING -- "display" is not defined,
     ***            referenced in: ("m5.c" "m4.c")
     *** WARNING -- "newline" is not defined,
     ***            referenced in: ("m5.c" "m4.c")
     *** WARNING -- "write" is not defined,
     ***            referenced in: ("m5.c")
     $ gcc -fPIC -c -D___DYNAMIC mymod.o1.c m4.c m5.c x.c
     $ /usr/ccs/bin/ld -G -o mymod.o1 mymod.o1.o m4.o m5.o x.o -lX11 -lsocket
     $ gsi mymod.o1
     hello from m4
     hello from m5
     (f1 10) = 22
     $ cat m4.scm
     (define (f1 x) (* 2 (f2 x)))
     (display "hello from m4")
     (newline)

     (c-declare #<<c-declare-end
     #include "x.h"
     c-declare-end
     )
     (define x-initialize (c-lambda (char-string) bool "x_initialize"))
     (define x-display-name (c-lambda () char-string "x_display_name"))
     (define x-bell (c-lambda (int) void "x_bell"))
     $ cat m5.scm
     (define (f2 x) (+ x 1))
     (display "hello from m5")
     (newline)

     (display "(f1 10) = ")
     (write (f1 10))
     (newline)

     (x-initialize (x-display-name))
     (x-bell 50) ; sound the bell at 50%
     $ cat x.c
     #include <X11/Xlib.h>

     static Display *display;

     int x_initialize (char *display_name)
     {
       display = XOpenDisplay (display_name);
       return display != NULL;
     }

     char *x_display_name (void)
     {
       return XDisplayName (NULL);
     }

     void x_bell (int volume)
     {
       XBell (display, volume);
       XFlush (display);
     }
     $ cat x.h
     int x_initialize (char *display_name);
     char *x_display_name (void);
     void x_bell (int);

3.4.3 Building a shared-library
-------------------------------

A shared-library can be built using an incremental link file or a flat
link file.  An incremental link file is normally used when the Gambit
runtime library (or some other library) is to be extended with new
procedures.  A flat link file is mainly useful when building a "primal"
runtime library, which is a library (such as the Gambit runtime
library) that does not extend another library.  When compiling the C
files and link file generated, the flags `-D___LIBRARY' and
`-D___SHARED' must be passed to the C compiler.  The flag `-D___PRIMAL'
must also be passed to the C compiler when a primal library is being
built.

   A shared-library `mylib.so' containing the two first modules of the
previous example can be built this way:

     $ uname -srmp
     Linux bailey 1.2.13 #2 Wed Aug 28 16:29:41 GMT 1996 i586
     $ gsc -link -o mylib.c m2
     $ gcc -shared -fPIC -D___LIBRARY -D___SHARED m1.c m2.c mylib.c -o mylib.so

   Note that this shared-library is built using an incremental link file
(it extends the Gambit runtime library with the procedures `pow2' and
`twice').  This shared-library can in turn be used to build an
executable program from the third module of the previous example:

     $ gsc -link -l mylib m3
     $ gcc m3.c m3_.c mylib.so -lgambc
     $ LD_LIBRARY_PATH=.:/usr/local/lib ./a.out
     ((2 . 2) (4 . 4) (8 . 8) (16 . 16))

3.4.4 Other compilation options
-------------------------------

The performance of the code can be increased by passing the
`-D___SINGLE_HOST' flag to the C compiler.  This will merge all the
procedures of a module into a single C procedure, which reduces the
cost of intra-module procedure calls.  In addition the `-O' option can
be passed to the C compiler.  For large modules, it will not be
practical to specify both `-O' and `-D___SINGLE_HOST' for typical C
compilers because the compile time will be high and the C compiler
might even fail to compile the program for lack of memory.  It has been
observed that lower levels of optimization (e.g. `-O1') often give
faster compilation and also generate faster code.  It is a good idea to
experiment.

   Normally C compilers will not automatically search
`/usr/local/Gambit-C/current/include' for header files so the flag
`-I/usr/local/Gambit-C/current/include' should be passed to the C
compiler.  Similarly, C compilers/linkers will not automatically search
`/usr/local/Gambit-C/current/lib' for libraries so the flag
`-L/usr/local/Gambit-C/current/lib' should be passed to the C
compiler/linker.  Alternatives are given in *Note Accessing the system
files::.

   A variety of flags are needed by some C compilers when compiling a
shared-library or a dynamically loadable library.  Some of these flags
are: `-shared', `-call_shared', `-rdynamic', `-fpic', `-fPIC', `-Kpic',
`-KPIC', `-pic', `+z', `-G'.  Check your compiler's documentation to see
which flag you need.

3.5 Procedures specific to compiler
===================================

The Gambit Scheme compiler features the following procedures that are
not available in the Gambit Scheme interpreter.

 -- procedure: compile-file-to-c FILE [OPTIONS [OUTPUT]]
     The FILE argument must be a string naming an existing file
     containing Scheme source code.  The extension can be omitted from
     FILE when the Scheme file has a `.scm' or `.six' extension.  This
     procedure compiles the source file into a file containing C code.
     By default, this file is named after FILE with the extension
     replaced with `.c'.  However, when OUTPUT is supplied the file is
     named `OUTPUT'.

     Compilation options are given as a list of symbols after the file
     name.  Any combination of the following options can be used:
     `verbose', `report', `expansion', `gvm', and `debug'.


 -- procedure: compile-file FILE [OPTIONS]
     The arguments of `compile-file' are the same as the first two
     arguments of `compile-file-to-c'.  The `compile-file' procedure
     compiles the source file into an object file by first generating a
     C file and then compiling it with the C compiler.  The object file
     is named after FILE with the extension replaced with `.oN', where
     N is a positive integer that acts as a version number.  The next
     available version number is generated automatically by
     `compile-file'.  Object files can be loaded dynamically by using
     the `load' procedure.  The `.oN' extension can be specified (to
     select a particular version) or omitted (to load the highest
     numbered version).  When older versions are no longer needed, all
     versions must be deleted and the compilation must be repeated
     (this is necessary because the file name, including the extension,
     is used to name some of the exported symbols of the object file).

     Note that this procedure is only available on host operating
     systems that support dynamic loading.


 -- procedure: link-incremental MODULE-LIST [OUTPUT [BASE]]
     The first argument must be a non empty list of strings naming
     Scheme modules to link (extensions must be omitted).  The
     remaining optional arguments must be strings.  An incremental link
     file is generated for the modules specified in MODULE-LIST.  By
     default the link file generated is named `LAST_.c', where LAST is
     the name of the last module.  However, when OUTPUT is supplied the
     link file is named `OUTPUT'.  The base link file is specified by
     the BASE parameter.  By default the base link file is the Gambit
     runtime library link file `~~/lib/_gambc.c'.  However, when BASE
     is supplied the base link file is named `BASE.c'.

     The following example shows how to build the executable program
     `hello' which contains the two Scheme modules `h.scm' and `w.six'.

          $ uname -srmp
          Darwin 8.1.0 Power Macintosh powerpc
          $ cat h.scm
          (display "hello") (newline)
          $ cat w.six
          display("world"); newline();
          $ gsc
          Gambit Version 4.0 beta 20

          > (compile-file-to-c "h")
          #t
          > (compile-file-to-c "w")
          #t
          > (link-incremental '("h" "w") "hello.c")
          > ,q
          $ gcc h.c w.c hello.c -lgambc -o hello
          $ ./hello
          hello
          world


 -- procedure: link-flat MODULE-LIST [OUTPUT]
     The first argument must be a non empty list of strings.  The first
     string must be the name of a Scheme module or the name of a link
     file and the remaining strings must name Scheme modules (in all
     cases extensions must be omitted).  If it is supplied, the second
     argument must be a string.  A flat link file is generated for the
     modules specified in MODULE-LIST.  By default the link file
     generated is named `LAST_.c', where LAST is the name of the last
     module.  However, when OUTPUT is supplied the link file is named
     `OUTPUT'.

     The following example shows how to build the dynamically loadable
     Scheme library `lib.o1' which contains the two Scheme modules
     `m6.scm' and `m7.scm'.

          $ uname -srmp
          Darwin 8.1.0 Power Macintosh powerpc
          $ cat m6.scm
          (define (f x) (g (* x x)))
          $ cat m7.scm
          (define (g y) (+ n y))
          $ gsc
          Gambit Version 4.0 beta 20

          > (compile-file-to-c "m6")
          #t
          > (compile-file-to-c "m7")
          #t
          > (link-flat '("m6" "m7") "lib.o1.c")
          *** WARNING -- "*" is not defined,
          ***            referenced in: ("m6.c")
          *** WARNING -- "+" is not defined,
          ***            referenced in: ("m7.c")
          *** WARNING -- "n" is not defined,
          ***            referenced in: ("m7.c")
          > ,q
          $ gcc -bundle -D___DYNAMIC m6.c m7.c lib.o1.c -o lib.o1
          $ gsc
          Gambit Version 4.0 beta 20

          > (load "lib")
          *** WARNING -- Variable "n" used in module "m7" is undefined
          "/Users/feeley/gambit/doc/lib.o1"
          > (define n 10)
          > (f 5)
          35
          > ,q

     The warnings indicate that there are no definitions (`define's or
     `set!'s) of the variables `*', `+' and `n' in the modules
     contained in the library.  Before the library is used, these
     variables will have to be bound; either implicitly (by the runtime
     library) or explicitly.


4 Runtime options for all programs
**********************************

Both `gsi' and `gsc' as well as executable programs compiled and linked
using `gsc' take a `-:' option which supplies parameters to the runtime
system.  This option must appear first on the command line.  The colon
is followed by a comma separated list of options with no intervening
spaces.  The available options are:

`mHEAPSIZE'
     Set minimum heap size in kilobytes.

`hHEAPSIZE'
     Set maximum heap size in kilobytes.

`lLIVEPERCENT'
     Set heap occupation after garbage collection.

`s'
     Select standard Scheme mode.

`S'
     Select Gambit Scheme mode.

`d[OPT...]'
     Set debugging options.

`=DIRECTORY'
     Override the Gambit installation directory.

`+ARGUMENT'
     Add ARGUMENT to the command line before other arguments.

`f[OPT...]'
     Set file options.

`t[OPT...]'
     Set terminal options.

`-[OPT...]'
     Set standard input and output options.

   The `m' option specifies the minimum size of the heap.  The `m' is
immediately followed by an integer indicating the number of kilobytes
of memory.  The heap will not shrink lower than this size.  By default,
the minimum size is 0.

   The `h' option specifies the maximum size of the heap.  The `h' is
immediately followed by an integer indicating the number of kilobytes
of memory.  The heap will not grow larger than this size.  By default,
there is no limit (i.e.  the heap will grow until the virtual memory is
exhausted).

   The `l' option specifies the percentage of the heap that will be
occupied with live objects after the heap is resized at the end of a
garbage collection.  The `l' is immediately followed by an integer
between 1 and 100 inclusively indicating the desired percentage.  The
garbage collector resizes the heap to reach this percentage occupation.
By default, the percentage is 50.

   The `s' option selects standard Scheme mode.  In this mode the
reader is case-insensitive and keywords are not recognized.  The `S'
option selects Gambit Scheme mode (the reader is case-sensitive and
recognizes keywords which end with a colon).  By default Gambit Scheme
mode is used.

   The `d' option sets various debugging options.  The letter `d' is
followed by a sequence of letters indicating suboptions.

`p'
     Uncaught exceptions will be treated as "errors" in the primordial
     thread only.

`a'
     Uncaught exceptions will be treated as "errors" in all threads.

`r'
     When an "error" occurs a new REPL will be started.

`s'
     When an "error" occurs a new REPL will be started.  Moreover the
     program starts in single-stepping mode.

`q'
     When an "error" occurs the program will terminate with a nonzero
     exit status.

`i'
     The REPL interaction channel will be the IDE REPL window (if the
     IDE is available).

`c'
     The REPL interaction channel will be the console.

`-'
     The REPL interaction channel will be standard input and standard
     output.

`LEVEL'
     The verbosity level is set to LEVEL (a digit from 0 to 9).  At
     level 0 the runtime system will not display error messages and
     warnings.


   The default debugging options are equivalent to `-:dpqi1' (i.e. an
uncaught exception in the primordial thread terminates the program after
displaying an error message).  When the letter `d' is not followed by
suboptions, it is equivalent to `-:dpri1' (i.e. a new REPL is started
only when an uncaught exception occurs in the primordial thread).

   The `=' option overrides the setting of the Gambit installation
directory.

   The `+' option adds the text that follows to the command line before
other arguments.

   The `f', `t' and `-' options specify the default settings of the
ports created for files, terminals and standard input and output
respectively.  The default character encoding, end-of-line encoding and
buffering can be set.  Moreover, for terminals the line-editing feature
can be enabled or disabled.  The `f', `t' and `-' must be followed by a
sequence of these options:

`A'
     ASCII character encoding.

`1'
     ISO-8859-1 character encoding.

`2'
     UCS-2 character encoding.

`4'
     UCS-4 character encoding.

`6'
     UTF-16 character encoding.

`8'
     UTF-8 character encoding.

`c'
     End-of-line is encoded as CR (carriage-return).

`l'
     End-of-line is encoded as LF (linefeed)

`cl'
     End-of-line is encoded as CR-LF.

`u'
     Unbuffered I/O.

`n'
     Line buffered I/O (`n' for "newline").

`f'
     Fully buffered I/O.

`e'
     Enable line-editing (applies to terminals only).

`E'
     Disable line-editing (applies to terminals only).

   When the environment variable `GAMBCOPT' is defined, the runtime
system will take its options from that environment variable.  A `-:'
option can be used to override some or all of the runtime system
options.  For example:

     $ GAMBCOPT=d0,=~/my-gambit2
     $ export GAMBCOPT
     $ gsi -e '(pretty-print (path-expand "~~")) (/ 1 0)'
     "/Users/feeley/my-gambit2/4.0b20/"
     $ echo $?
     70
     $ gsi -:d1 -e '(pretty-print (path-expand "~~")) (/ 1 0)'
     "/Users/feeley/my-gambit2/4.0b20/"
     *** ERROR IN (string)@1.3 -- Divide by zero
     (/ 1 0)

5 Debugging
***********

5.1 Debugging model
===================

The evaluation of an expression may stop before it is completed for the
following reasons:

  a. An evaluation error has occured, such as attempting to divide by
     zero.

  b. The user has interrupted the evaluation (usually by typing <^C>).

  c. A breakpoint has been reached or `(step)' was evaluated.

  d. Single-stepping mode is enabled.


   When an evaluation stops, a message is displayed indicating the
reason and location where the evaluation was stopped.  The location
information includes, if known, the name of the procedure where the
evaluation was stopped and the source code location in the format
`STREAM@LINE.COLUMN', where STREAM is either a string naming a file or
a symbol within parentheses, such as `(console)'.

   A "nested REPL" is then initiated in the context of the point of
execution where the evaluation was stopped.  The nested REPL's
continuation and evaluation environment are the same as the point where
the evaluation was stopped.  For example when evaluating the expression
`(let ((y (- 1 1))) (* (/ x y) 2))', a "divide by zero" error is
reported and the nested REPL's continuation is the one that takes the
result and multiplies it by two.  The REPL's lexical environment
includes the lexical variable `y'.  This allows the inspection of the
evaluation context (i.e. the lexical and dynamic environments and
continuation), which is particularly useful to determine the exact
location and cause of an error.

   The prompt of nested REPLs includes the nesting level; `1>' is the
prompt at the first nesting level, `2>' at the second nesting level,
and so on.  An end of file (usually <^D>) will cause the current REPL
to be terminated and the enclosing REPL (one nesting level less) to be
resumed.

   At any time the user can examine the frames in the REPL's
continuation, which is useful to determine which chain of procedure
calls lead to an error.  A backtrace that lists the chain of active
continuation frames in the REPL's continuation can be obtained with the
`,b' command.  The frames are numbered from 0, that is frame 0 is the
most recent frame of the continuation where execution stopped, frame 1
is the parent frame of frame 0, and so on.  It is also possible to move
the REPL to a specific parent continuation (i.e. a specific frame of
the continuation where execution stopped) with the `,+', `,-' and `,N'
commands (where N is the frame number).  When the frame number of the
frame being examined is not zero, it is shown in the prompt after the
nesting level, for example `1\5>' is the prompt when the REPL nesting
level is 1 and the frame number is 5.

   Expressions entered at a nested REPL are evaluated in the environment
(both lexical and dynamic) of the continuation frame currently being
examined if that frame was created by interpreted Scheme code.  If the
frame was created by compiled Scheme code then expressions get evaluated
in the global interaction environment.  This feature may be used in
interpreted code to fetch the value of a variable in the current frame
or to change its value with `set!'.  Note that some special forms
(`define' in particular) can only be evaluated in the global
interaction environment.

5.2 Debugging commands
======================

In addition to expressions, the REPL accepts the following special
"comma" commands:

`,?'
     Give a summary of the REPL commands.

`,q'
     Terminate the current thread (note that terminating the primordial
     thread terminates the program).  To terminate the program from any
     thread, call the `exit' procedure.

`,t'
     Return to the outermost REPL, also known as the "top-level REPL".

`,d'
     Leave the current REPL and resume the enclosing REPL.  This
     command does nothing in the top-level REPL.

`,(c EXPR)'
     Leave the current REPL and continue the computation that initiated
     the REPL with a specific value.  This command can only be used to
     continue a computation that signaled an error.  The expression
     EXPR is evaluated in the current context and the resulting value
     is returned as the value of the expression which signaled the
     error.  For example, if the evaluation of the expression `(* (/ x
     y) 2)' signaled an error because `y' is zero, then in the nested
     REPL a `,(c (+ 4 y))' will resume the computation of `(* (/ x y)
     2)' as though the value of `(/ x y)' was 4.  This command must be
     used carefully because the context where the error occured may
     rely on the result being of a particular type.  For instance a
     `,(c #f)' in the previous example will cause `*' to signal a type
     error (this problem is the most troublesome when debugging Scheme
     code that was compiled with type checking turned off so be
     careful).

`,c'
     Leave the current REPL and continue the computation that initiated
     the REPL.  This command can only be used to continue a computation
     that was stopped due to a user interrupt, breakpoint or a
     single-step.

`,s'
     Leave the current REPL and continue the computation that initiated
     the REPL in single-stepping mode.  The computation will perform an
     evaluation step (as defined by `step-level-set!') and then stop,
     causing a nested REPL to be entered.  Just before the evaluation
     step is performed, a line is displayed (in the same format as
     `trace') which indicates the expression that is being evaluated.
     If the evaluation step produces a result, the result is also
     displayed on another line.  A nested REPL is then entered after
     displaying a message which describes the next step of the
     computation.  This command can only be used to continue a
     computation that was stopped due to a user interrupt, breakpoint
     or a single-step.

`,l'
     This command is similar to `,s' except that it "leaps" over
     procedure calls, that is procedure calls are treated like a single
     step.  Single-stepping mode will resume when the procedure call
     returns, or if and when the execution of the called procedure
     encounters a breakpoint.

`,N'
     Move to frame number N of the continuation.  After changing the
     current frame, a one-line summary of the frame is displayed as if
     the `,y' command was entered.

`,+'
     Move to the next frame in the chain of continuation frames (i.e.
     towards older continuation frames).  After changing the current
     frame, a one-line summary of the frame is displayed as if the `,y'
     command was entered.

`,-'
     Move to the previous frame in the chain of continuation frames
     (i.e.  towards more recently created continuation frames).  After
     changing the current frame, a one-line summary of the frame is
     displayed as if the `,y' command was entered.

`,y'
     Display a one-line summary of the current frame.  The information
     is displayed in four fields.  The first field is the frame number.
     The second field is the procedure that created the frame or
     `(interaction)' if the frame was created by an expression entered
     at the REPL.  The remaining fields describe the subproblem
     associated with the frame, that is the expression whose value is
     being computed.  The third field is the location of the
     subproblem's source code and the fourth field is a reproduction of
     the source code, possibly truncated to fit on the line.  The last
     two fields may be missing if that information is not available.
     In particular, the third field is missing when the frame was
     created by a user call to the `eval' procedure, and the last two
     fields are missing when the frame was created by a compiled
     procedure not compiled with the `-debug' option.

`,b'
     Display a backtrace summarizing each frame in the chain of
     continuation frames starting with the current frame.  For each
     frame, the same information as for the `,y' command is displayed
     (except that location information is displayed in the format
     `STREAM@LINE:COLUMN').  If there are more that 15 frames in the
     chain of continuation frames, some of the middle frames will be
     omitted.

`,i'
     Pretty print the procedure that created the current frame or
     `(interaction)' if the frame was created by an expression entered
     at the REPL.  Compiled procedures will only be pretty printed when
     they are compiled with the `-debug' option.

`,e'
     Display the environment which is accessible from the current frame.
     Both the lexical and dynamic environments are displayed.  However,
     only non-global lexical variables are displayed and only if the
     frame was created by interpreted code or code compiled with the
     `-debug' option.  Due to space safety considerations and compiler
     optimizations, some of the lexical variable bindings may be
     missing.  Lexical variable bindings are displayed using the format
     `VARIABLE = EXPRESSION' and dynamically-bound parameter bindings
     are displayed using the format `(PARAMETER) = EXPRESSION'.  Note
     that EXPRESSION can be a self-evaluating expression (number,
     string, boolean, character, ...), a quoted expression, a lambda
     expression or a global variable (the last two cases, which are
     only used when the value of the variable or parameter is a
     procedure, simplifies the debugging of higher-order procedures).
     A PARAMETER can be a quoted expression or a global variable.
     Lexical bindings are displayed in inverse binding order (most
     deeply nested first) and shadowed variables are included in the
     list.


   Here is a sample interaction with `gsi':

     $ gsi
     Gambit Version 4.0 beta 20

     > (define (invsqr x) (/ 1 (expt x 2)))
     > (define (mymap fn lst)
         (define (mm in)
           (if (null? in)
               '()
               (cons (fn (car in)) (mm (cdr in)))))
         (mm lst))
     > (mymap invsqr '(5 2 hello 9 1))
     *** ERROR IN invsqr, (console)@1.25 -- (Argument 1) NUMBER expected
     (expt 'hello 2)
     1> ,i
     #<procedure #2 invsqr> =
     (lambda (x) (/ 1 (expt x 2)))
     1> ,e
     x = 'hello
     (current-exception-handler) = primordial-exception-handler
     (current-input-port) = '#<input-output-port #3 (console)>
     (current-output-port) = '#<input-output-port #3 (console)>
     (current-directory) = "/Users/feeley/gambit/doc/"
     1> ,b
     0  invsqr                    (console)@1:25          (expt x 2)
     1  #<procedure #4>           (console)@6:17          (fn (car in))
     2  #<procedure #4>           (console)@6:31          (mm (cdr in))
     3  #<procedure #4>           (console)@6:31          (mm (cdr in))
     4  (interaction)             (console)@8:1           (mymap invsqr '(5 2 hel...
     1> ,+
     1  #<procedure #4>           (console)@6.17          (fn (car in))
     1\1> (pp #4)
     (lambda (in) (if (null? in) '() (cons (fn (car in)) (mm (cdr in)))))
     1\1> ,e
     in = '(hello 9 1)
     mm = (lambda (in) (if (null? in) '() (cons (fn (car in)) (mm (cdr in)))))
     fn = invsqr
     lst = '(5 2 hello 9 1)
     (current-exception-handler) = primordial-exception-handler
     (current-input-port) = '#<input-output-port #3 (console)>
     (current-output-port) = '#<input-output-port #3 (console)>
     (current-directory) = "/Users/feeley/gambit/doc/"
     1\1> fn
     #<procedure #2 invsqr>
     1\1> (pp fn)
     (lambda (x) (/ 1 (expt x 2)))
     1\1> ,+
     2  #<procedure #4>           (console)@6.31          (mm (cdr in))
     1\2> ,e
     in = '(2 hello 9 1)
     mm = (lambda (in) (if (null? in) '() (cons (fn (car in)) (mm (cdr in)))))
     fn = invsqr
     lst = '(5 2 hello 9 1)
     (current-exception-handler) = primordial-exception-handler
     (current-input-port) = '#<input-output-port #3 (console)>
     (current-output-port) = '#<input-output-port #3 (console)>
     (current-directory) = "/Users/feeley/gambit/doc/"
     1\2> ,(c (list 3 4 5))
     (1/25 1/4 3 4 5)
     > ,q

5.3 Procedures related to debugging
===================================

 -- procedure: trace PROC...
 -- procedure: untrace PROC...
     The `trace' procedure starts tracing calls to the specified
     procedures.  When a traced procedure is called, a line containing
     the procedure and its arguments is displayed (using the procedure
     call expression syntax).  The line is indented with a sequence of
     vertical bars which indicate the nesting depth of the procedure's
     continuation.  After the vertical bars is a greater-than sign
     which indicates that the evaluation of the call is starting.

     When a traced procedure returns a result, it is displayed with the
     same indentation as the call but without the greater-than sign.
     This makes it easy to match calls and results (the result of a
     given call is the value at the same indentation as the
     greater-than sign).  If a traced procedure P1 performs a tail call
     to a traced procedure P2, then P2 will use the same indentation as
     P1.  This makes it easy to spot tail calls.  The special handling
     for tail calls is needed to preserve the space complexity of the
     program (i.e. tail calls are implemented as required by Scheme
     even when they involve traced procedures).

     The `untrace' procedure stops tracing calls to the specified
     procedures.  When no arguments is passed to the `trace' procedure,
     the list of procedures currently being traced is returned.  The
     void object is returned by the `trace' procedure when it is passed
     one or more arguments.  When no argument is passed to the
     `untrace' procedure stops all tracing and returns the void object.
     A compiled procedure may be traced but only if it is bound to a
     global variable.

     For example:

          > (define (fact n) (if (< n 2) 1 (* n (fact (- n 1)))))
          > (trace fact)
          > (fact 5)
          | > (fact 5)
          | | > (fact 4)
          | | | > (fact 3)
          | | | | > (fact 2)
          | | | | | > (fact 1)
          | | | | | 1
          | | | | 2
          | | | 6
          | | 24
          | 120
          120
          > (trace -)
          *** WARNING -- Rebinding global variable "-" to an interpreted procedure
          > (define (fact-iter n r) (if (< n 2) r (fact-iter (- n 1) (* n r))))
          > (trace fact-iter)
          > (fact-iter 5 1)
          | > (fact-iter 5 1)
          | | > (- 5 1)
          | | 4
          | > (fact-iter 4 5)
          | | > (- 4 1)
          | | 3
          | > (fact-iter 3 20)
          | | > (- 3 1)
          | | 2
          | > (fact-iter 2 60)
          | | > (- 2 1)
          | | 1
          | > (fact-iter 1 120)
          | 120
          120
          > (trace)
          (#<procedure #2 fact-iter> #<procedure #3 -> #<procedure #4 fact>)
          > (untrace)
          > (fact 5)
          120


 -- procedure: step
 -- procedure: step-level-set! LEVEL
     The `step' procedure enables single-stepping mode.  After the call
     to `step' the computation will stop just before the interpreter
     executes the next evaluation step (as defined by
     `step-level-set!').  A nested REPL is then started.  Note that
     because single-stepping is stopped by the REPL whenever the prompt
     is displayed it is pointless to enter `(step)' by itself.  On the
     other hand entering `(begin (step) EXPR)' will evaluate EXPR in
     single-stepping mode.

     The procedure `step-level-set!' sets the stepping level which
     determines the granularity of the evaluation steps when
     single-stepping is enabled.  The stepping level LEVEL must be an
     exact integer in the range 0 to 7.  At a level of 0, the
     interpreter ignores single-stepping mode.  At higher levels the
     interpreter stops the computation just before it performs the
     following operations, depending on the stepping level:

       1. procedure call

       2. `delay' special form and operations at lower levels

       3. `lambda' special form and operations at lower levels

       4. `define' special form and operations at lower levels

       5. `set!' special form and operations at lower levels

       6. variable reference and operations at lower levels

       7. constant reference and operations at lower levels


     The default stepping level is 7.

     For example:

          > (define (fact n) (if (< n 2) 1 (* n (fact (- n 1)))))
          > (step-level-set! 1)
          > (begin (step) (fact 5))
          *** STOPPED IN (console)@3.15
          1> ,s
          | > (fact 5)
          *** STOPPED IN fact, (console)@1.22
          1> ,s
          | | > (< n 2)
          | | #f
          *** STOPPED IN fact, (console)@1.43
          1> ,s
          | | > (- n 1)
          | | 4
          *** STOPPED IN fact, (console)@1.37
          1> ,s
          | | > (fact (- n 1))
          *** STOPPED IN fact, (console)@1.22
          1> ,s
          | | | > (< n 2)
          | | | #f
          *** STOPPED IN fact, (console)@1.43
          1> ,s
          | | | > (- n 1)
          | | | 3
          *** STOPPED IN fact, (console)@1.37
          1> ,l
          | | | > (fact (- n 1))
          *** STOPPED IN fact, (console)@1.22
          1> ,l
          | | > (* n (fact (- n 1)))
          | | 24
          *** STOPPED IN fact, (console)@1.32
          1> ,l
          | > (* n (fact (- n 1)))
          | 120
          120


 -- procedure: break PROC...
 -- procedure: unbreak PROC...
     The `break' procedure places a breakpoint on each of the specified
     procedures.  When a procedure is called that has a breakpoint, the
     interpreter will enable single-stepping mode (as if `step' had
     been called).  This typically causes the computation to stop soon
     inside the procedure if the stepping level is high enough.

     The `unbreak' procedure removes the breakpoints on the specified
     procedures.  With no argument, `break' returns the list of
     procedures currently containing breakpoints.  The void object is
     returned by `break' if it is passed one or more arguments.  With
     no argument `unbreak' removes all the breakpoints and returns the
     void object.  A breakpoint can be placed on a compiled procedure
     but only if it is bound to a global variable.

     For example:

          > (define (double x) (+ x x))
          > (define (triple y) (- (double (double y)) y))
          > (define (f z) (* (triple z) 10))
          > (break double)
          > (break -)
          *** WARNING -- Rebinding global variable "-" to an interpreted procedure
          > (f 5)
          *** STOPPED IN double, (console)@1.21
          1> ,b
          0  double                    (console)@1:21          +
          1  triple                    (console)@2:31          (double y)
          2  f                         (console)@3:18          (triple z)
          3  (interaction)             (console)@6:1           (f 5)
          1> ,e
          x = 5
          (current-exception-handler) = primordial-exception-handler
          (current-input-port) = '#<input-output-port #2 (console)>
          (current-output-port) = '#<input-output-port #2 (console)>
          (current-directory) = "/Users/feeley/gambit/doc/"
          1> ,c
          *** STOPPED IN double, (console)@1.21
          1> ,c
          *** STOPPED IN f, (console)@3.29
          1> ,c
          150
          > (break)
          (#<procedure #3 -> #<procedure #4 double>)
          > (unbreak)
          > (f 5)
          150


 -- procedure: generate-proper-tail-calls [NEW-VALUE]
     The parameter object `generate-proper-tail-calls' is bound to a
     boolean value controlling how the interpreter handles tail calls.
     When it is bound to `#f' the interpreter will treat tail calls
     like nontail calls, that is a new continuation will be created for
     the call.  This setting is useful for debugging, because when a
     primitive signals an error the location information will point to
     the call site of the primitive even if this primitive was called
     with a tail call.  The initial value of this parameter object is
     `#t', which means that a tail call will reuse the continuation of
     the calling function.

     This parameter object only affects code that is subsequently
     processed by `load' or `eval', or entered at the REPL.

     For example:

          > (generate-proper-tail-calls)
          #t
          > (let loop ((i 1)) (if (< i 10) (loop (* i 2)) oops))
          *** ERROR IN #<procedure #2>, (console)@2.47 -- Unbound variable: oops
          1> ,b
          0  #<procedure #2>           (console)@2:47          oops
          1  (interaction)             (console)@2:1           ((letrec ((loop (lambda...
          1> ,t
          > (generate-proper-tail-calls #f)
          > (let loop ((i 1)) (if (< i 10) (loop (* i 2)) oops))
          *** ERROR IN #<procedure #3>, (console)@6.47 -- Unbound variable: oops
          1> ,b
          0  #<procedure #3>           (console)@6:47          oops
          1  #<procedure #3>           (console)@6:32          (loop (* i 2))
          2  #<procedure #3>           (console)@6:32          (loop (* i 2))
          3  #<procedure #3>           (console)@6:32          (loop (* i 2))
          4  #<procedure #3>           (console)@6:32          (loop (* i 2))
          5  (interaction)             (console)@6:1           ((letrec ((loop (lambda...


 -- procedure: display-environment-set! DISPLAY?
     This procedure sets a flag that controls the automatic display of
     the environment by the REPL.  If DISPLAY? is true, the environment
     is displayed by the REPL before the prompt.  The default setting is
     not to display the environment.


 -- procedure: pretty-print OBJ [PORT]
     This procedure pretty-prints OBJ on the port PORT.  If it is not
     specified, PORT defaults to the current output-port.

     For example:

          > (pretty-print
              (let* ((x '(1 2 3 4)) (y (list x x x))) (list y y y)))
          (((1 2 3 4) (1 2 3 4) (1 2 3 4))
           ((1 2 3 4) (1 2 3 4) (1 2 3 4))
           ((1 2 3 4) (1 2 3 4) (1 2 3 4)))


 -- procedure: pp OBJ [PORT]
     This procedure pretty-prints OBJ on the port PORT.  When OBJ is a
     procedure created by the interpreter or a procedure created by
     code compiled with the `-debug' option, the procedure's source
     code is displayed.  If it is not specified, PORT defaults to the
     interaction channel (i.e. the output will appear at the REPL).

     For example:

          > (define (f g) (+ (time (g 100)) (time (g 1000))))
          > (pp f)
          (lambda (g)
            (+ (##time (lambda () (g 100)) '(g 100))
               (##time (lambda () (g 1000)) '(g 1000))))


 -- procedure: gc-report-set! REPORT?
     This procedure controls the generation of reports during garbage
     collections.  If the argument is true, a brief report of memory
     usage is generated after every garbage collection.  It contains:
     the time taken for this garbage collection, the amount of memory
     allocated in megabytes since the program was started, the size of
     the heap in megabytes, the heap memory in megabytes occupied by
     live data, the proportion of the heap occupied by live data, and
     the number of bytes occupied by movable and nonmovable objects.


5.4 Console line-editing
========================

The console implements a simple Scheme-friendly line-editing
user-interface that is enabled by default.  It offers parentheses
balancing, a history of previous commands, and several emacs-compatible
keyboard commands.  The user's input is displayed in a bold font and
the output produced by the system is in a plain font.  Here are the
keyboard commands available (where the ``M-'' prefix means the escape
key is typed and the ``C-'' prefix means the control key is pressed):

`C-d'
     Generate an end-of-file when the line is empty, otherwise delete
     character at cursor.

`C-a'
     Move cursor to beginning of line.

`C-e'
     Move cursor to end of line.

`C-b or left-arrow'
     Move cursor left one character.

`M-C-b or `M-'left-arrow'
     Move cursor left one S-expression.

`C-f or right-arrow'
     Move cursor right one character.

`M-C-f or `M-'right-arrow'
     Move cursor right one S-expression.

`C-p or up-arrow'
     Move to previous line in history.

`C-n or down-arrow'
     Move to next line in history.

`C-t'
     Transpose character at cursor with previous character.

`M-C-t'
     Transpose S-expression at cursor with previous S-expression.

`C-l'
     Clear console and redraw line being edited.

`C-nul'
     Set the mark to the cursor.

`C-w'
     Delete the text between the cursor and the mark and keep a copy of
     this text on the internal clipboard.

`C-k'
     Delete the text from the cursor to the end of the line and keep a
     copy of this text on the internal clipboard.

`C-y'
     Paste the text that is on the internal clipboard.

`F8'
     Same as typing `#||#,c;' (REPL command to continue the
     computation).

`F9'
     Same as typing `#||#,-;' (REPL command to move to newer frame).

`F10'
     Same as typing `#||#,+;' (REPL command to move to older frame).

`F11'
     Same as typing `#||#,s;' (REPL command to step the computation).

`F12'
     Same as typing `#||#,l;' (REPL command to leap the computation).


   Note that on Mac OS X, depending on your configuration, you may have
to press the `fn' key to access the function key `F12' and the `option'
key to access the other function keys.

5.5 Emacs interface
===================

Gambit comes with the Emacs package `gambit.el' which provides a nice
environment for running Gambit from within the Emacs editor.  This
package filters the standard output of the Gambit process and when it
intercepts a location information (in the format `STREAM@LINE.COLUMN'
where STREAM is either `(stdin)' when the expression was obtained from
standard input, `(console)' when the expression was obtained from the
console, or a string naming a file) it opens a window to highlight the
corresponding expression.

   To use this package, make sure the file `gambit.el' is accessible
from your load-path and that the following lines are in your `.emacs'
file:

     (autoload 'gambit-inferior-mode "gambit" "Hook Gambit mode into cmuscheme.")
     (autoload 'gambit-mode "gambit" "Hook Gambit mode into scheme.")
     (add-hook 'inferior-scheme-mode-hook (function gambit-inferior-mode))
     (add-hook 'scheme-mode-hook (function gambit-mode))
     (setq scheme-program-name "gsi -:d-")

   Alternatively, if you don't mind always loading this package, you
can simply add this line to your `.emacs' file:

     (require 'gambit)

   You can then start an inferior Gambit process by typing `M-x
run-scheme'.  The commands provided in `cmuscheme' mode will be
available in the Gambit interaction buffer (i.e. `*scheme*') and in
buffers attached to Scheme source files.  Here is a list of the most
useful commands (for a complete list type `C-h m' in the Gambit
interaction buffer):
`C-x C-e'
     Evaluate the expression which is before the cursor (the expression
     will be copied to the Gambit interaction buffer).

`C-c C-z'
     Switch to Gambit interaction buffer.

`C-c C-l'
     Load a file (file attached to current buffer is default) using
     `(load FILE)'.

`C-c C-k'
     Compile a file (file attached to current buffer is default) using
     `(compile-file FILE)'.

   The file `gambit.el' provides these additional commands:

`F8 or C-c c'
     Continue the computation (same as typing `#||#,c;' to the REPL).

`F9 or C-c ]'
     Move to newer frame (same as typing `#||#,-;' to the REPL).

`F10 or C-c ['
     Move to older frame (same as typing `#||#,+;' to the REPL).

`F11 or C-c s'
     Step the computation (same as typing `#||#,s;' to the REPL).

`F12 or C-c l'
     Leap the computation (same as typing `#||#,l;' to the REPL).

`C-c _'
     Removes the last window that was opened to highlight an expression.

   The two keystroke version of these commands can be shortened to
`M-c', `M-[', `M-]', `M-s', `M-l', and `M-_' respectively by adding
this line to your `.emacs' file:

     (setq gambit-repl-command-prefix "\e")

   This is more convenient to type than the two keystroke `C-c' based
sequences but the purist may not like this because it does not follow
normal Emacs conventions.

   Here is what a typical `.emacs' file will look like:

     (setq load-path ; add directory containing gambit.el
       (cons "/usr/local/Gambit-C/current/share/emacs/site-lisp"
             load-path))
     (setq scheme-program-name "/tmp/gsi -:d-") ; if gsi not in executable path
     (setq gambit-highlight-color "gray") ; if you don't like the default
     (setq gambit-repl-command-prefix "\e") ; if you want M-c, M-s, etc
     (require 'gambit)

5.6 GUIDE
=========

The implementation and documentation for GUIDE, the Gambit Universal
IDE, are not yet complete.

6 Scheme extensions
*******************

6.1 Extensions to standard procedures
=====================================

 -- procedure: transcript-on FILE
 -- procedure: transcript-off
     These procedures do nothing.


 -- procedure: call-with-current-continuation PROC
 -- procedure: call/cc PROC
     The procedure `call-with-current-continuation' is bound to the
     global variables `call-with-current-continuation' and `call/cc'.


6.2 Extensions to standard special forms
========================================

 -- special form: lambda lambda-formals body
 -- special form: define (variable define-formals) body
          lambda-formals = `(' formal-argument-list `)' |
          r4rs-lambda-formals

          define-formals = formal-argument-list | r4rs-define-formals

          formal-argument-list = dsssl-formal-argument-list |
          rest-at-end-formal-argument-list

          dsssl-formal-argument-list = reqs opts rest keys

          rest-at-end-formal-argument-list = reqs opts keys rest | reqs
          opts keys `.' rest-formal-argument

          reqs = required-formal-argument*

          required-formal-argument = variable

          opts = `#!optional' optional-formal-argument* | empty

          optional-formal-argument = variable | `(' variable
          initializer `)'

          rest = `#!rest' rest-formal-argument | empty

          rest-formal-argument = variable

          keys = `#!key' keyword-formal-argument* | empty

          keyword-formal-argument = variable | `(' variable initializer
          `)'

          initializer = expression

          r4rs-lambda-formals = `(' variable* `)' | `(' variable+ `.'
          variable `)' | variable

          r4rs-define-formals = variable* | variable* `.' variable

     These forms are extended versions of the `lambda' and `define'
     special forms of standard Scheme.  They allow the use of optional
     formal arguments, either positional or named, and support the
     syntax and semantics of the DSSSL standard.

     When the procedure introduced by a `lambda' (or `define') is
     applied to a list of actual arguments, the formal and actual
     arguments are processed as specified in the R4RS if the
     lambda-formals (or define-formals) is a r4rs-lambda-formals (or
     r4rs-define-formals).

     If the formal-argument-list matches dsssl-formal-argument-list or
     extended-formal-argument-list they are processed as follows:

       a. Variables in required-formal-arguments are bound to
          successive actual arguments starting with the first actual
          argument.  It shall be an error if there are fewer actual
          arguments than required-formal-arguments.

       b. Next variables in optional-formal-arguments are bound to
          remaining actual arguments.  If there are fewer remaining
          actual arguments than optional-formal-arguments, then the
          variables are bound to the result of evaluating initializer,
          if one was specified, and otherwise to `#f'.  The initializer
          is evaluated in an environment in which all previous formal
          arguments have been bound.

       c. If `#!key' does not appear in the formal-argument-list and
          there is no rest-formal-argument then it shall be an error if
          there are any remaining actual arguments.

       d. If `#!key' does not appear in the formal-argument-list and
          there is a rest-formal-argument then the rest-formal-argument
          is bound to a list of all remaining actual arguments.

       e. If `#!key' appears in the formal-argument-list and there is
          no rest-formal-argument then there shall be an even number of
          remaining actual arguments.  These are interpreted as a
          series of pairs, where the first member of each pair is a
          keyword specifying the argument name, and the second is the
          corresponding value.  It shall be an error if the first
          member of a pair is not a keyword.  It shall be an error if
          the argument name is not the same as a variable in a
          keyword-formal-argument.  If the same argument name occurs
          more than once in the list of actual arguments, then the
          first value is used.  If there is no actual argument for a
          particular keyword-formal-argument, then the variable is
          bound to the result of evaluating initializer if one was
          specified, and otherwise to `#f'.  The initializer is
          evaluated in an environment in which all previous formal
          arguments have been bound.

       f. If `#!key' appears in the formal-argument-list and there is a
          rest-formal-argument before the `#!key' then there may be an
          even or odd number of remaining actual arguments and the
          rest-formal-argument is bound to a list of all remaining
          actual arguments.  Then, these remaining actual arguments are
          scanned from left to right in pairs, stopping at the first
          pair whose first element is not a keyword.  Each pair whose
          first element is a keyword matching the name of a
          keyword-formal-argument gives the value (i.e. the second
          element of the pair) of the corresponding formal argument.  If
          the same argument name occurs more than once in the list of
          actual arguments, then the first value is used.  If there is
          no actual argument for a particular keyword-formal-argument,
          then the variable is bound to the result of evaluating
          initializer if one was specified, and otherwise to `#f'.  The
          initializer is evaluated in an environment in which all
          previous formal arguments have been bound.

       g. If `#!key' appears in the formal-argument-list and there is a
          rest-formal-argument after the `#!key' then there may be an
          even or odd number of remaining actual arguments.  The
          remaining actual arguments are scanned from left to right in
          pairs, stopping at the first pair whose first element is not
          a keyword.  Each pair shall have as its first element a
          keyword matching the name of a keyword-formal-argument; the
          second element gives the value of the corresponding formal
          argument.  If the same argument name occurs more than once in
          the list of actual arguments, then the first value is used.
          If there is no actual argument for a particular
          keyword-formal-argument, then the variable is bound to the
          result of evaluating initializer if one was specified, and
          otherwise to `#f'.  The initializer is evaluated in an
          environment in which all previous formal arguments have been
          bound.  Finally, the rest-formal-argument is bound to the
          list of the actual arguments that were not scanned (i.e.
          after the last keyword/value pair).

     In all cases it is an error for a variable to appear more than
     once in a formal-argument-list.

     Note that this specification is compatible with the DSSSL language
     standard (i.e. a correct DSSSL program will have the same semantics
     when run with Gambit).

     It is unspecified whether variables receive their value by binding
     or by assignment.  Currently the compiler and interpreter use
     different methods, which can lead to different semantics if
     `call-with-current-continuation' is used in an initializer.  Note
     that this is irrelevant for DSSSL programs because
     `call-with-current-continuation' does not exist in DSSSL.

     For example:

          > ((lambda (#!rest x) x) 1 2 3)
          (1 2 3)
          > (define (f a #!optional b) (list a b))
          > (define (g a #!optional (b a) #!key (k (* a b))) (list a b k))
          > (define (h1 a #!rest r #!key k) (list a k r))
          > (define (h2 a #!key k #!rest r) (list a k r))
          > (f 1)
          (1 #f)
          > (f 1 2)
          (1 2)
          > (g 3)
          (3 3 9)
          > (g 3 4)
          (3 4 12)
          > (g 3 4 k: 5)
          (3 4 5)
          > (g 3 4 k: 5 k: 6)
          (3 4 5)
          > (h1 7)
          (7 #f ())
          > (h1 7 k: 8 9)
          (7 8 (k: 8 9))
          > (h1 7 k: 8 z: 9)
          (7 8 (k: 8 z: 9))
          > (h2 7)
          (7 #f ())
          > (h2 7 k: 8 9)
          (7 8 (9))
          > (h2 7 k: 8 z: 9)
          *** ERROR IN (console)@17.1 -- Unknown keyword argument passed to procedure
          (h2 7 k: 8 z: 9)


6.3 Miscellaneous extensions
============================

 -- procedure: vector-copy VECTOR
     This procedure returns a newly allocated vector with the same
     content as the vector VECTOR.  Note that the elements are not
     recursively copied.


 -- procedure: vector-append VECTOR...
     This procedure is the vector analog of the `string-append'
     procedure.  It returns a newly allocated vector whose elements
     form the concatenation of the given vectors.


 -- procedure: subvector VECTOR START END
     This procedure is the vector analog of the `substring' procedure.
     It returns a newly allocated vector formed from the elements of
     the vector VECTOR beginning with index START (inclusive) and
     ending with index END (exclusive).


 -- procedure: box OBJ
 -- procedure: box? OBJ
 -- procedure: unbox BOX
 -- procedure: set-box! BOX OBJ
     These procedures implement the "box" data type.  A box is a cell
     containing a single mutable field.  The lexical syntax of a box
     containing the object OBJ is `#&OBJ' (*note Box syntax::).

     The procedure `box' returns a new box object whose content is
     initialized to OBJ.  The procedure `box?' returns `#t' if OBJ is a
     box, and otherwise returns `#f'.  The procedure `unbox' returns
     the content of the box BOX.  The procedure `set-box!' changes the
     content of the box BOX to OBJ.  The procedure `set-box!'  returns
     an unspecified value.

     For example:

          > (define b (box 0))
          > b
          #&0
          > (define (inc!) (set-box! b (+ (unbox b) 1)))
          > (inc!)
          > b
          #&1
          > (unbox b)
          1


 -- procedure: keyword? OBJ
 -- procedure: keyword->string KEYWORD
 -- procedure: string->keyword STRING
     These procedures implement the "keyword" data type.  Keywords are
     similar to symbols but are self evaluating and distinct from the
     symbol data type.  The lexical syntax of keywords is specified in
     *Note Keyword syntax::.

     The procedure `keyword?' returns `#t' if OBJ is a keyword, and
     otherwise returns `#f'.  The procedure `keyword->string' returns
     the name of KEYWORD as a string.  The procedure `string->keyword'
     returns the keyword whose name is STRING.

     For example:

          > (keyword? 'color)
          #f
          > (keyword? color:)
          #t
          > (keyword->string color:)
          "color"
          > (string->keyword "color")
          color:


 -- procedure: gensym [PREFIX]
     This procedure returns a new "uninterned symbol".  Uninterned
     symbols are guaranteed to be distinct from the symbols generated
     by the procedures `read' and `string->symbol'.  The symbol PREFIX
     is the prefix used to generate the new symbol's name.  If it is
     not specified, the prefix defaults to `g'.

     For example:

          > (gensym)
          #:g0
          > (gensym)
          #:g1
          > (gensym 'star-trek-)
          #:star-trek-2


 -- procedure: make-uninterned-symbol NAME [HASH]
 -- procedure: uninterned-symbol? OBJ
     The procedure `make-uninterned-symbol' returns a new uninterned
     symbol whose name is NAME and hash is HASH.  The name must be a
     string and the hash must be a nonnegative fixnum.

     The procedure `uninterned-symbol?' returns `#t' when OBJ is a
     symbol that is uninterned and `#f' otherwise.

     For example:

          > (uninterned-symbol? (gensym))
          #t
          > (make-uninterned-symbol "foo")
          #:foo:
          > (uninterned-symbol? (make-uninterned-symbol "foo"))
          #t
          > (uninterned-symbol? 'hello)
          #f
          > (uninterned-symbol? 123)
          #f


 -- procedure: make-uninterned-keyword NAME [HASH]
 -- procedure: uninterned-keyword? OBJ
     The procedure `make-uninterned-keyword' returns a new uninterned
     keyword whose name is NAME and hash is HASH.  The name must be a
     string and the hash must be a nonnegative fixnum.

     The procedure `uninterned-keyword?' returns `#t' when OBJ is a
     keyword that is uninterned and `#f' otherwise.

     For example:

          > (make-uninterned-keyword "foo")
          #:foo:
          > (uninterned-keyword? (make-uninterned-keyword "foo"))
          #t
          > (uninterned-keyword? hello:)
          #f
          > (uninterned-keyword? 123)
          #f


 -- procedure: void
     This procedure returns the void object.  The read-eval-print loop
     prints nothing when the result is the void object.


 -- procedure: eval EXPR [ENV]
     The first argument is a datum representing an expression.  The
     `eval' procedure evaluates this expression in the global
     interaction environment and returns the result.  If present, the
     second argument is ignored (it is provided for compatibility with
     R5RS).

     For example:

          > (eval '(+ 1 2))
          3
          > ((eval 'car) '(1 2))
          1
          > (eval '(define x 5))
          > x
          5


 -- special form: include file
     The file argument must be a string naming an existing file
     containing Scheme source code.  The `include' special form splices
     the content of the specified source file.  This form can only
     appear where a `define' form is acceptable.

     For example:

          (include "macros.scm")

          (define (f lst)
            (include "sort.scm")
            (map sqrt (sort lst)))


 -- special form: define-macro (name define-formals) body
     Define name as a macro special form which expands into body.  This
     form can only appear where a `define' form is acceptable.  Macros
     are lexically scoped.  The scope of a local macro definition
     extends from the definition to the end of the body of the
     surrounding binding construct.  Macros defined at the top level of
     a Scheme module are only visible in that module.  To have access
     to the macro definitions contained in a file, that file must be
     included using the `include' special form.  Macros which are
     visible from the REPL are also visible during the compilation of
     Scheme source files.

     For example:

          (define-macro (unless test . body)
            `(if ,test #f (begin ,@body)))

          (define-macro (push var #!optional val)
            `(set! ,var (cons ,val ,var)))

     To examine the code into which a macro expands you can use the
     compiler's `-expansion' option or the `pp' procedure.  For example:

          > (define-macro (push var #!optional val)
              `(set! ,var (cons ,val ,var)))
          > (pp (lambda () (push stack 1) (push stack) (push stack 3)))
          (lambda ()
            (set! stack (cons 1 stack))
            (set! stack (cons #f stack))
            (set! stack (cons 3 stack)))


 -- special form: define-syntax name expander
     Define name as a macro special form whose expansion is specified
     by expander.  This form is available only after evaluating `(load
     "~~/syntax-case")', which can be done at the REPL or in the
     initialization file.  This file contains Hieb and Dybvig's
     portable `syntax-case' implementation that has been ported to the
     Gambit interpreter and compiler.  Note that this implementation of
     `syntax-case' does not correctly track source code location
     information, so the error messages will be much less precise.

     For example:

          > (load "~~/syntax-case")
          "/usr/local/Gambit-C/4.0b20/syntax-case.scm"
          > (define-syntax unless
              (syntax-rules ()
                ((unless test body ...)
                 (if test #f (begin body ...)))))
          > (let ((test 111)) (unless (= 1 2) (list test test)))
          (111 111)
          > (pp (lambda () (let ((test 111)) (unless (= 1 2) (list test test)))))
          (lambda () ((lambda (%%test14) (if (= 1 2) #f (list %%test14 %%test14))) 111))
          > (unless #f (pp xxx))
          *** ERROR IN (console)@8.16 -- Unbound variable: xxx


 -- special form: declare declaration...
     This form introduces declarations to be used by the compiler
     (currently the interpreter ignores the declarations).  This form
     can only appear where a `define' form is acceptable.  Declarations
     are lexically scoped in the same way as macros.  The following
     declarations are accepted by the compiler:

    `(DIALECT)'
          Use the given dialect's semantics.  DIALECT can be:
          `ieee-scheme' or `r4rs-scheme'.

    `(STRATEGY)'
          Select block compilation or separate compilation.  In block
          compilation, the compiler assumes that global variables
          defined in the current file that are not mutated in the file
          will never be mutated.  STRATEGY can be: `block' or
          `separate'.

    `([not] inline)'
          Allow (or disallow) inlining of user procedures.

    `([not] inline-primitives PRIMITIVE...)'
          The given primitives should (or should not) be inlined if
          possible (all primitives if none specified).

    `(inlining-limit N)'
          Select the degree to which the compiler inlines user
          procedures.  N is the upper-bound, in percent, on code
          expansion that will result from inlining.  Thus, a value of
          300 indicates that the size of the program will not grow by
          more than 300 percent (i.e. it will be at most 4 times the
          size of the original).  A value of 0 disables inlining.  The
          size of a program is the total number of subexpressions it
          contains (i.e. the size of an expression is one plus the size
          of its immediate subexpressions).  The following conditions
          must hold for a procedure to be inlined: inlining the
          procedure must not cause the size of the call site to grow
          more than specified by the inlining limit, the site of
          definition (the `define' or `lambda') and the call site must
          be declared as `(inline)', and the compiler must be able to
          find the definition of the procedure referred to at the call
          site (if the procedure is bound to a global variable, the
          definition site must have a `(block)' declaration).  Note
          that inlining usually causes much less code expansion than
          specified by the inlining limit (an expansion around 10% is
          common for N=350).

    `([not] lambda-lift)'
          Lambda-lift (or don't lambda-lift) locally defined procedures.

    `([not] constant-fold)'
          Allow (or disallow) constant-folding of primitive procedures.

    `([not] standard-bindings VAR...)'
          The given global variables are known (or not known) to be
          equal to the value defined for them in the dialect (all
          variables defined in the standard if none specified).

    `([not] extended-bindings VAR...)'
          The given global variables are known (or not known) to be
          equal to the value defined for them in the runtime system
          (all variables defined in the runtime if none specified).

    `([not] run-time-bindings VAR...)'
          The given global variables will be tested at run time to see
          if they are equal to the value defined for them in the
          runtime system (all variables defined in the runtime if none
          specified).

    `([not] safe)'
          Generate (or don't generate) code that will prevent fatal
          errors at run time.  Note that in `safe' mode certain
          semantic errors will not be checked as long as they can't
          crash the system.  For example the primitive `char=?' may
          disregard the type of its arguments in `safe' as well as `not
          safe' mode.

    `([not] interrupts-enabled)'
          Generate (or don't generate) interrupt checks.  Interrupt
          checks are used to detect user interrupts and also to check
          for stack overflows.  Interrupt checking should not be turned
          off casually.

    `(NUMBER-TYPE PRIMITIVE...)'
          Numeric arguments and result of the specified primitives are
          known to be of the given type (all primitives if none
          specified).  NUMBER-TYPE can be: `generic', `fixnum', or
          `flonum'.

    `(MOSTLY-NUMBER-TYPE PRIMITIVE...)'
          Numeric arguments and result of the specified primitives are
          expected to be most often of the given type (all primitives
          if none specified).  MOSTLY-NUMBER-TYPE can be:
          `mostly-generic', `mostly-fixnum', `mostly-fixnum-flonum',
          `mostly-flonum', or `mostly-flonum-fixnum'.


     The default declarations used by the compiler are equivalent to:

          (declare
            (ieee-scheme)
            (separate)
            (inline)
            (inline-primitives)
            (inlining-limit 350)
            (constant-fold)
            (lambda-lift)
            (not standard-bindings)
            (not extended-bindings)
            (run-time-bindings)
            (safe)
            (interrupts-enabled)
            (generic)
            (mostly-fixnum-flonum)
          )

     These declarations are compatible with the semantics of R5RS
     Scheme.  Typically used declarations that enhance performance, at
     the cost of violating the R5RS Scheme semantics, are:
     `(standard-bindings)', `(block)', `(not safe)' and `(fixnum)'.


6.4 Undocumented extensions
===========================

The procedures in this section are not yet documented.

 -- procedure: print [`port:' PORT] OBJ...
 -- procedure: println [`port:' PORT] OBJ...

 -- procedure: make-thread-group [NAME [THREAD-GROUP]]
 -- procedure: thread-group? OBJ
 -- procedure: thread-group-name THREAD-GROUP
 -- procedure: thread-group-parent THREAD-GROUP
 -- procedure: thread-group-resume! THREAD-GROUP
 -- procedure: thread-group-suspend! THREAD-GROUP
 -- procedure: thread-group-terminate! THREAD-GROUP

 -- procedure: thread-suspend! THREAD
 -- procedure: thread-resume! THREAD

 -- procedure: thread-thread-group THREAD

 -- special form: define-type-of-thread name field...

 -- procedure: thread-init! THREAD THUNK [NAME [THREAD-GROUP]]

 -- procedure: initialized-thread-exception? OBJ
 -- procedure: initialized-thread-exception-procedure EXC
 -- procedure: initialized-thread-exception-arguments EXC

 -- procedure: uninitialized-thread-exception? OBJ
 -- procedure: uninitialized-thread-exception-procedure EXC
 -- procedure: uninitialized-thread-exception-arguments EXC

 -- procedure: process-pid PROCESS-PORT

 -- procedure: process-status PROCESS-PORT [TIMEOUT [TIMEOUT-VAL]]

 -- procedure: unterminated-process-exception? OBJ
 -- procedure: unterminated-process-exception-procedure EXC
 -- procedure: unterminated-process-exception-arguments EXC

 -- procedure: timeout->time TIMEOUT

 -- procedure: open-dummy

 -- procedure: port-settings-set! PORT SETTINGS

 -- procedure: input-port-bytes-buffered PORT

 -- procedure: input-port-characters-buffered PORT

 -- procedure: nonempty-input-port-character-buffer-exception? OBJ
 -- procedure: nonempty-input-port-character-buffer-exception-arguments
          EXC
 -- procedure: nonempty-input-port-character-buffer-exception-procedure
          EXC

 -- procedure: repl-input-port
 -- procedure: repl-output-port
 -- procedure: console-port

 -- procedure: current-user-interrupt-handler [HANDLER]

 -- procedure: primordial-exception-handler EXC

 -- procedure: err-code->string CODE

 -- procedure: foreign-address FOREIGN
 -- procedure: foreign-release! FOREIGN
 -- procedure: foreign-released? FOREIGN

 -- procedure: invalid-hash-number-exception? OBJ
 -- procedure: invalid-hash-number-exception-procedure EXC
 -- procedure: invalid-hash-number-exception-arguments EXC

 -- procedure: network-info NETWORK
 -- procedure: network-info? OBJ
 -- procedure: network-info-name NETWORK-INFO
 -- procedure: network-info-net NETWORK-INFO
 -- procedure: network-info-aliases NETWORK-INFO

 -- procedure: protocol-info PROTOCOL
 -- procedure: protocol-info? OBJ
 -- procedure: protocol-info-name PROTOCOL-INFO
 -- procedure: protocol-info-number PROTOCOL-INFO
 -- procedure: protocol-info-aliases PROTOCOL-INFO

 -- procedure: service-info SERVICE [PROTOCOL]
 -- procedure: service-info? OBJ
 -- procedure: service-info-name SERVICE-INFO
 -- procedure: service-info-port SERVICE-INFO
 -- procedure: service-info-protocol SERVICE-INFO
 -- procedure: service-info-aliases SERVICE-INFO

 -- procedure: tcp-client-peer-socket-info TCP-CLIENT-PORT
 -- procedure: tcp-client-self-socket-info TCP-CLIENT-PORT

 -- procedure: socket-info? OBJ
 -- procedure: socket-info-address SOCKET-INFO
 -- procedure: socket-info-family SOCKET-INFO
 -- procedure: socket-info-port-number SOCKET-INFO

 -- procedure: six.make-array ...

 -- procedure: system-version
 -- procedure: system-version-string

 -- procedure: touch OBJ

 -- procedure: tty? OBJ
 -- procedure: tty-history TTY
 -- procedure: tty-history-set! TTY HISTORY
 -- procedure: tty-max-history-length-set! TTY N
 -- procedure: tty-paren-balance-duration-set! TTY DURATION
 -- procedure: tty-text-attributes-set! TTY ATTRIBUTES
 -- procedure: tty-mode-set! TTY MODE
 -- procedure: tty-type-set! TTY TYPE

 -- procedure: with-input-from-port PORT THUNK
 -- procedure: with-output-to-port PORT THUNK

 -- procedure: input-port-char-position PORT
 -- procedure: output-port-char-position PORT

 -- procedure: open-event-queue N

 -- procedure: main ...

7 Modules
*********

TO DO!

8 Characters and strings
************************

Gambit supports the Unicode character encoding standard
(ISO/IEC-10646-1).  Scheme characters can be any of the characters in
the 16 bit subset of Unicode known as UCS-2.  Scheme strings can
contain any character in UCS-2.  Source code can also contain any
character in UCS-2.  However, to read such source code properly `gsi'
and `gsc' must be told which character encoding to use for reading the
source code (i.e. UTF-8, UCS-2, or UCS-4).  This can be done by
specifying the runtime option `-:f' when `gsi' and `gsc' are started.

8.1 Extensions to character procedures
======================================

 -- procedure: char->integer CHAR
 -- procedure: integer->char N
     The procedure `char->integer' returns the Unicode encoding of the
     character CHAR.

     The procedure `integer->char' returns the character whose Unicode
     encoding is the exact integer N.

     For example:

          > (char->integer #\!)
          33
          > (integer->char 65)
          #\A
          > (integer->char (char->integer #\u1234))
          #\u1234


 -- procedure: char=? CHAR1...
 -- procedure: char<? CHAR1...
 -- procedure: char>? CHAR1...
 -- procedure: char<=? CHAR1...
 -- procedure: char>=? CHAR1...
 -- procedure: char-ci=? CHAR1...
 -- procedure: char-ci<? CHAR1...
 -- procedure: char-ci>? CHAR1...
 -- procedure: char-ci<=? CHAR1...
 -- procedure: char-ci>=? CHAR1...
     These procedures take any number of arguments including no
     argument.  This is useful to test if the elements of a list are
     sorted in a particular order.  For example, testing that the list
     of characters `lst' is sorted in nondecreasing order can be done
     with the call `(apply char<? lst)'.


8.2 Extensions to string procedures
===================================

 -- procedure: string=? STRING1...
 -- procedure: string<? STRING1...
 -- procedure: string>? STRING1...
 -- procedure: string<=? STRING1...
 -- procedure: string>=? STRING1...
 -- procedure: string-ci=? STRING1...
 -- procedure: string-ci<? STRING1...
 -- procedure: string-ci>? STRING1...
 -- procedure: string-ci<=? STRING1...
 -- procedure: string-ci>=? STRING1...
     These procedures take any number of arguments including no
     argument.  This is useful to test if the elements of a list are
     sorted in a particular order.  For example, testing that the list
     of strings `lst' is sorted in nondecreasing order can be done with
     the call `(apply string<? lst)'.


9 Numbers
*********

9.1 Extensions to numeric procedures
====================================

 -- procedure: = Z1...
 -- procedure: < X1...
 -- procedure: > X1...
 -- procedure: <= X1...
 -- procedure: >= X1...
     These procedures take any number of arguments including no
     argument.  This is useful to test if the elements of a list are
     sorted in a particular order.  For example, testing that the list
     of numbers `lst' is sorted in nondecreasing order can be done with
     the call `(apply < lst)'.


9.2 IEEE floating point arithmetic
==================================

To better conform to IEEE floating point arithmetic the standard
numeric tower is extended with these special inexact reals:

`+inf.0'
     positive infinity

`-inf.0'
     negative infinity

`+nan.0'
     "not a number"

`-0.'
     negative zero (`0.' is the positive zero)

   The infinities and "not a number" are reals (i.e. `(real?  +inf.0)'
is `#t') but are not rational (i.e. `(rational?  +inf.0)' is `#f').

   Both zeros are numerically equal (i.e. `(= -0. 0.)' is `#t') but are
not equivalent (i.e. `(eqv? -0. 0.)' and `(equal?  -0. 0.)' are `#f').
All numerical comparisons with "not a number", including `(= +nan.0
+nan.0)', are `#f'.

9.3 Integer square root and nth root
====================================

 -- procedure: integer-sqrt N
     This procedure returns the integer part of the square root of the
     nonnegative exact integer N.

     For example:

          > (integer-sqrt 123)
          11


 -- procedure: integer-nth-root N1 N2
     This procedure returns the integer part of N1 raised to the power
     1/N2, where N1 is a nonnegative exact integer and N2 is a positive
     exact integer.

     For example:

          > (integer-nth-root 100 3)
          4


9.4 Bitwise-operations on exact integers
========================================

The procedures defined in this section are compatible with the
withdrawn "Integer Bitwise-operation Library SRFI" (SRFI 33).  Note
that some of the procedures specified in SRFI 33 are not provided.

   Most procedures in this section are specified in terms of the binary
representation of exact integers.  The two's complement representation
is assumed where an integer is composed of an infinite number of bits.
The upper section of an integer (the most significant bits) are either
an infinite sequence of ones when the integer is negative, or they are
an infinite sequence of zeros when the integer is nonnegative.

 -- procedure: arithmetic-shift N1 N2
     This procedure returns N1 shifted to the left by N2 bits, that is
     `(floor (* N1 (expt 2 N2)))'.  Both N1 and N2 must be exact
     integers.

     For example:

          > (arithmetic-shift 1000 7)  ; n1=...0000001111101000
          128000
          > (arithmetic-shift 1000 -6) ; n1=...0000001111101000
          15
          > (arithmetic-shift -23 -3)  ; n1=...1111111111101001
          -3


 -- procedure: bitwise-merge N1 N2 N3
     This procedure returns an exact integer whose bits combine the bits
     from N2 and N3 depending on N1.  The bit at index I of the result
     depends only on the bits at index I in N1, N2 and N3: it is equal
     to the bit in N2 when the bit in N1 is 0 and it is equal to the
     bit in N3 when the bit in N1 is 1.  All arguments must be exact
     integers.

     For example:

          > (bitwise-merge -4 -11 10) ; ...11111100 ...11110101 ...00001010
          9
          > (bitwise-merge 12 -11 10) ; ...00001100 ...11110101 ...00001010
          -7


 -- procedure: bitwise-and N...
     This procedure returns the bitwise "and" of the exact integers
     N....  The value -1 is returned when there are no arguments.

     For example:

          > (bitwise-and 6 12)  ; ...00000110 ...00001100
          4
          > (bitwise-and 6 -4)  ; ...00000110 ...11111100
          4
          > (bitwise-and -6 -4) ; ...11111010 ...11111100
          -8
          > (bitwise-and)
          -1


 -- procedure: bitwise-ior N...
     This procedure returns the bitwise "inclusive-or" of the exact
     integers N....  The value 0 is returned when there are no
     arguments.

     For example:

          > (bitwise-ior 6 12)  ; ...00000110 ...00001100
          14
          > (bitwise-ior 6 -4)  ; ...00000110 ...11111100
          -2
          > (bitwise-ior -6 -4) ; ...11111010 ...11111100
          -2
          > (bitwise-ior)
          0


 -- procedure: bitwise-xor N...
     This procedure returns the bitwise "exclusive-or" of the exact
     integers N....  The value 0 is returned when there are no
     arguments.

     For example:

          > (bitwise-xor 6 12)  ; ...00000110 ...00001100
          10
          > (bitwise-xor 6 -4)  ; ...00000110 ...11111100
          -6
          > (bitwise-xor -6 -4) ; ...11111010 ...11111100
          6
          > (bitwise-xor)
          0


 -- procedure: bitwise-not N
     This procedure returns the bitwise complement of the exact integer
     N.

     For example:

          > (bitwise-not 3)  ; ...00000011
          -4
          > (bitwise-not -1) ; ...11111111
          0


 -- procedure: bit-count N
     This procedure returns the bit count of the exact integer N.  If N
     is nonnegative, the bit count is the number of 1 bits in the two's
     complement representation of N.  If N is negative, the bit count
     is the number of 0 bits in the two's complement representation of
     N.

     For example:

          > (bit-count 0)   ; ...00000000
          0
          > (bit-count 1)   ; ...00000001
          1
          > (bit-count 2)   ; ...00000010
          1
          > (bit-count 3)   ; ...00000011
          2
          > (bit-count 4)   ; ...00000100
          1
          > (bit-count -23) ; ...11101001
          3


 -- procedure: integer-length N
     This procedure returns the bit length of the exact integer N.  If
     N is a positive integer the bit length is one more than the index
     of the highest 1 bit (the least significant bit is at index 0).
     If N is a negative integer the bit length is one more than the
     index of the highest 0 bit.  If N is zero, the bit length is 0.

     For example:

          > (integer-length 0)   ; ...00000000
          0
          > (integer-length 1)   ; ...00000001
          1
          > (integer-length 2)   ; ...00000010
          2
          > (integer-length 3)   ; ...00000011
          2
          > (integer-length 4)   ; ...00000100
          3
          > (integer-length -23) ; ...11101001
          5


 -- procedure: bit-set? N1 N2
     This procedure returns a boolean indicating if the bit at index N1
     of N2 is set (i.e. equal to 1) or not.  Both N1 and N2 must be
     exact integers, and N1 must be nonnegative.

     For example:

          > (map (lambda (i) (bit-set? i -23)) ; ...11101001
                 '(7 6 5 4 3 2 1 0))
          (#t #t #t #f #t #f #f #t)


 -- procedure: any-bits-set? N1 N2
     This procedure returns a boolean indicating if the bitwise and of
     N1 and N2 is different from zero or not.  This procedure is
     implemented more efficiently than the naive definition:

          (define (any-bits-set? n1 n2) (not (zero? (bitwise-and n1 n2))))

     For example:

          > (any-bits-set? 5 10)   ; ...00000101 ...00001010
          #f
          > (any-bits-set? -23 32) ; ...11101001 ...00100000
          #t


 -- procedure: all-bits-set? N1 N2
     This procedure returns a boolean indicating if the bitwise and of
     N1 and N2 is equal to N1 or not.  This procedure is implemented
     more efficiently than the naive definition:

          (define (all-bits-set? n1 n2) (= n1 (bitwise-and n1 n2)))

     For example:

          > (all-bits-set? 1 3) ; ...00000001 ...00000011
          #t
          > (all-bits-set? 7 3) ; ...00000111 ...00000011
          #f


 -- procedure: first-bit-set N
     This procedure returns the bit index of the least significant bit
     of N equal to 1 (which is also the number of 0 bits that are below
     the least significant 1 bit).  This procedure returns `-1' when N
     is zero.

     For example:

          > (first-bit-set 24) ; ...00011000
          3
          > (first-bit-set 0)  ; ...00000000
          -1


 -- procedure: extract-bit-field N1 N2 N3
 -- procedure: test-bit-field? N1 N2 N3
 -- procedure: clear-bit-field N1 N2 N3
 -- procedure: replace-bit-field N1 N2 N3 N4
 -- procedure: copy-bit-field N1 N2 N3 N4
     These procedures operate on a bit-field which is N1 bits wide
     starting at bit index N2.  All arguments must be exact integers
     and N1 and N2 must be nonnegative.

     The procedure `extract-bit-field' returns the bit-field of N3
     shifted to the right so that the least significant bit of the
     bit-field is the least significant bit of the result.

     The procedure `test-bit-field?' returns `#t' if any bit in the
     bit-field of N3 is equal to 1, otherwise `#f' is returned.

     The procedure `clear-bit-field' returns N3 with all bits in the
     bit-field replaced with 0.

     The procedure `replace-bit-field' returns N4 with the bit-field
     replaced with the least-significant N1 bits of N3.

     The procedure `copy-bit-field' returns N4 with the bit-field
     replaced with the (same index and size) bit-field in N3.

     For example:

          > (extract-bit-field 5 2 -37)    ; ...11011011
          22
          > (test-bit-field? 5 2 -37)      ; ...11011011
          #t
          > (test-bit-field? 1 2 -37)      ; ...11011011
          #f
          > (clear-bit-field 5 2 -37)      ; ...11011011
          -125
          > (replace-bit-field 5 2 -6 -37) ; ...11111010 ...11011011
          -21
          > (copy-bit-field 5 2 -6 -37)    ; ...11111010 ...11011011
          -5


9.5 Fixnum specific operations
==============================

 -- procedure: fixnum? OBJ

 -- procedure: fx* N1...

 -- procedure: fx+ N1...

 -- procedure: fx- N1 N2...

 -- procedure: fx< N1...

 -- procedure: fx<= N1...

 -- procedure: fx= N1...

 -- procedure: fx> N1...

 -- procedure: fx>= N1...

 -- procedure: fxand N1...

 -- procedure: fxarithmetic-shift N1 N2

 -- procedure: fxarithmetic-shift-left N1 N2

 -- procedure: fxarithmetic-shift-right N1 N2

 -- procedure: fxbit-count N

 -- procedure: fxbit-set? N1 N2

 -- procedure: fxeven? N

 -- procedure: fxfirst-bit-set N

 -- procedure: fxif N1 N2 N3

 -- procedure: fxior N1...

 -- procedure: fxlength N

 -- procedure: fxmax N1 N2...

 -- procedure: fxmin N1 N2...

 -- procedure: fxmodulo N1 N2

 -- procedure: fxnegative? N

 -- procedure: fxnot N

 -- procedure: fxodd? N

 -- procedure: fxpositive? N

 -- procedure: fxquotient N1 N2

 -- procedure: fxremainder N1 N2

 -- procedure: fxwrap* N1...

 -- procedure: fxwrap+ N1...

 -- procedure: fxwrap- N1 N2...

 -- procedure: fxwraparithmetic-shift N1 N2

 -- procedure: fxwraparithmetic-shift-left N1 N2

 -- procedure: fxwraplogical-shift-right N1 N2

 -- procedure: fxwrapquotient N1 N2

 -- procedure: fxxor N1...

 -- procedure: fxzero? N

 -- procedure: fixnum-overflow-exception? OBJ
 -- procedure: fixnum-overflow-exception-procedure EXC
 -- procedure: fixnum-overflow-exception-arguments EXC
     Fixnum-overflow-exception objects are raised by some of the fixnum
     specific procedures when the result is larger than can fit in a
     fixnum.  The parameter EXC must be a fixnum-overflow-exception
     object.

     The procedure `fixnum-overflow-exception?' returns `#t' when OBJ
     is a fixnum-overflow-exception object and `#f' otherwise.

     The procedure `fixnum-overflow-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `fixnum-overflow-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (fixnum-overflow-exception? exc)
                  (list (fixnum-overflow-exception-procedure exc)
                        (fixnum-overflow-exception-arguments exc))
                  'not-fixnum-overflow-exception))
          > (with-exception-catcher
              handler
              (lambda () (fx* 100000 100000)))
          (#<procedure #2 fx*> (100000 100000))


9.6 Flonum specific operations
==============================

 -- procedure: flonum? OBJ

 -- procedure: fixnum->flonum N

 -- procedure: fl* X1...

 -- procedure: fl+ X1...

 -- procedure: fl- X1 X2...

 -- procedure: fl/ X1 X2

 -- procedure: fl< X1...

 -- procedure: fl<= X1...

 -- procedure: fl= X1...

 -- procedure: fl> X1...

 -- procedure: fl>= X1...

 -- procedure: flabs X

 -- procedure: flacos X

 -- procedure: flasin X

 -- procedure: flatan X
 -- procedure: flatan Y X

 -- procedure: flceiling X

 -- procedure: flcos X

 -- procedure: fldenominator X

 -- procedure: fleven? X

 -- procedure: flexp X

 -- procedure: flexpt X Y

 -- procedure: flfinite? X

 -- procedure: flfloor X

 -- procedure: flinfinite? X

 -- procedure: flinteger? X

 -- procedure: fllog X

 -- procedure: flmax X1 X2...

 -- procedure: flmin X1 X2...

 -- procedure: flnan? X

 -- procedure: flnegative? X

 -- procedure: flnumerator X

 -- procedure: flodd? X

 -- procedure: flpositive? X

 -- procedure: flround X

 -- procedure: flsin X

 -- procedure: flsqrt X

 -- procedure: fltan X

 -- procedure: fltruncate X

 -- procedure: flzero? X

9.7 Pseudo random numbers
=========================

The procedures and variables defined in this section are compatible
with the "Sources of Random Bits SRFI" (SRFI 27).  The implementation
is based on Pierre L'Ecuyer's MRG32k3a pseudo random number generator.
At the heart of SRFI 27's interface is the random source type which
encapsulates the state of a pseudo random number generator.  The state
of a random source object changes every time a pseudo random number is
generated from this random source object.

 -- variable: default-random-source
     The global variable `default-random-source' is bound to the random
     source object which is used by the `random-integer' and
     `random-real' procedures.


 -- procedure: random-integer N
     This procedure returns a pseudo random exact integer in the range
     0 to N-1.  The random source object in the global variable
     `default-random-source' is used to generate this number.  The
     parameter N must be a positive exact integer.

     For example:

          > (random-integer 100)
          24
          > (random-integer 100)
          2
          > (random-integer 10000000000000000000000000000000000000000)
          6143360270902284438072426748425263488507


 -- procedure: random-real
     This procedure returns a pseudo random inexact real between, but
     not including, 0 and 1.  The random source object in the global
     variable `default-random-source' is used to generate this number.

     For example:

          > (random-real)
          .24230672079133753
          > (random-real)
          .02317001922506932


 -- procedure: make-random-source
     This procedure returns a new random source object initialized to a
     predetermined state (to initialize to a pseudo random state the
     procedure `random-source-randomize!' should be called).

     For example:

          > (define rs (make-random-source))
          > ((random-source-make-integers rs) 10000000)
          8583952


 -- procedure: random-source? OBJ
     This procedure returns `#t' when OBJ is a random source object and
     `#f' otherwise.

     For example:

          > (random-source? default-random-source)
          #t
          > (random-source? 123)
          #f


 -- procedure: random-source-state-ref RANDOM-SOURCE
 -- procedure: random-source-state-set! RANDOM-SOURCE STATE
     The procedure `random-source-state-ref' extracts the state of the
     random source object RANDOM-SOURCE and returns a vector containing
     the state.

     The procedure `random-source-state-set!' restores the state of the
     random source object RANDOM-SOURCE to STATE which must be a vector
     returned from a call to the procedure `random-source-state-ref'.

     For example:

          > (define s (random-source-state-ref default-random-source))
          > (random-integer 10000000000000000000000000000000000000000)
          7583880188903074396261960585615270693321
          > (random-source-state-set! default-random-source s)
          > (random-integer 10000000000000000000000000000000000000000)
          7583880188903074396261960585615270693321


 -- procedure: random-source-randomize! RANDOM-SOURCE
 -- procedure: random-source-pseudo-randomize! RANDOM-SOURCE I J
     These procedures change the state of the random source object
     RANDOM-SOURCE.  The procedure `random-source-randomize!' sets the
     random source object to a state that depends on the current time
     (which for typical uses can be considered to randomly initialize
     the state).  The procedure `random-source-pseudo-randomize!' sets
     the random source object to a state that is determined only by the
     current state and the nonnegative exact integers I and J.  For
     both procedures the value returned is unspecified.

     For example:

          > (define s (random-source-state-ref default-random-source))
          > (random-source-pseudo-randomize! default-random-source 5 99)
          > (random-integer 10000000000000000000000000000000000000000)
          9816755163910623041601722050112674079767
          > (random-source-state-set! default-random-source s)
          > (random-source-pseudo-randomize! default-random-source 5 99)
          > (random-integer 10000000000000000000000000000000000000000)
          9816755163910623041601722050112674079767
          > (random-source-pseudo-randomize! default-random-source 5 99)
          > (random-integer 10000000000000000000000000000000000000000)
          9816755163910623041601722050112674079767
          > (random-source-state-set! default-random-source s)
          > (random-source-randomize! default-random-source)
          > (random-integer 10000000000000000000000000000000000000000)
          2271441220851914333384493143687768110622
          > (random-source-state-set! default-random-source s)
          > (random-source-randomize! default-random-source)
          > (random-integer 10000000000000000000000000000000000000000)
          6247966138948323029033944059178072366895


 -- procedure: random-source-make-integers RANDOM-SOURCE
     This procedure returns a procedure for generating pseudo random
     exact integers using the random source object RANDOM-SOURCE.  The
     returned procedure accepts a single parameter N, a positive exact
     integer, and returns a pseudo random exact integer in the range 0
     to N-1.

     For example:

          > (define rs (make-random-source))
          > (define ri (random-source-make-integers rs))
          > (ri 10000000)
          8583952
          > (ri 10000000)
          2879793


 -- procedure: random-source-make-reals RANDOM-SOURCE
     This procedure returns a procedure for generating pseudo random
     inexact reals using the random source object RANDOM-SOURCE.  The
     returned procedure accepts no parameters and returns a pseudo
     random inexact real between, but not including, 0 and 1.

     For example:

          > (define rs (make-random-source))
          > (define rr (random-source-make-reals rs))
          > (rr)
          .857402537562821
          > (rr)
          .2876463473845367


10 Homogeneous vectors
**********************

Homogeneous vectors are vectors containing raw numbers of the same type
(signed or unsigned exact integers or inexact reals).  There are 10
types of homogeneous vectors: `s8vector' (vector of exact integers in
the range -2^7 to 2^7-1), `u8vector' (vector of exact integers in the
range 0 to 2^8-1), `s16vector' (vector of exact integers in the range
-2^15 to 2^15-1), `u16vector' (vector of exact integers in the range 0
to 2^16-1), `s32vector' (vector of exact integers in the range -2^31 to
2^31-1), `u32vector' (vector of exact integers in the range 0 to
2^32-1), `s64vector' (vector of exact integers in the range -2^63 to
2^63-1), `u64vector' (vector of exact integers in the range 0 to
2^64-1), `f32vector' (vector of 32 bit floating point numbers), and
`f64vector' (vector of 64 bit floating point numbers).

   The lexical syntax of homogeneous vectors is specified in *Note
Homogeneous vector syntax::.

   The procedures available for homogeneous vectors, listed below, are
the analog of the normal vector/string procedures for each of the
homogeneous vector types.

 -- procedure: s8vector? OBJ
 -- procedure: make-s8vector K [FILL]
 -- procedure: s8vector EXACT-INT8...
 -- procedure: s8vector-length S8VECTOR
 -- procedure: s8vector-ref S8VECTOR K
 -- procedure: s8vector-set! S8VECTOR K EXACT-INT8
 -- procedure: s8vector->list S8VECTOR
 -- procedure: list->s8vector LIST-OF-EXACT-INT8
 -- procedure: s8vector-fill! S8VECTOR FILL
 -- procedure: s8vector-copy S8VECTOR
 -- procedure: s8vector-append S8VECTOR...
 -- procedure: subs8vector S8VECTOR START END

 -- procedure: u8vector? OBJ
 -- procedure: make-u8vector K [FILL]
 -- procedure: u8vector EXACT-INT8...
 -- procedure: u8vector-length U8VECTOR
 -- procedure: u8vector-ref U8VECTOR K
 -- procedure: u8vector-set! U8VECTOR K EXACT-INT8
 -- procedure: u8vector->list U8VECTOR
 -- procedure: list->u8vector LIST-OF-EXACT-INT8
 -- procedure: u8vector-fill! U8VECTOR FILL
 -- procedure: u8vector-copy U8VECTOR
 -- procedure: u8vector-append U8VECTOR...
 -- procedure: subu8vector U8VECTOR START END

 -- procedure: s16vector? OBJ
 -- procedure: make-s16vector K [FILL]
 -- procedure: s16vector EXACT-INT16...
 -- procedure: s16vector-length S16VECTOR
 -- procedure: s16vector-ref S16VECTOR K
 -- procedure: s16vector-set! S16VECTOR K EXACT-INT16
 -- procedure: s16vector->list S16VECTOR
 -- procedure: list->s16vector LIST-OF-EXACT-INT16
 -- procedure: s16vector-fill! S16VECTOR FILL
 -- procedure: s16vector-copy S16VECTOR
 -- procedure: s16vector-append S16VECTOR...
 -- procedure: subs16vector S16VECTOR START END

 -- procedure: u16vector? OBJ
 -- procedure: make-u16vector K [FILL]
 -- procedure: u16vector EXACT-INT16...
 -- procedure: u16vector-length U16VECTOR
 -- procedure: u16vector-ref U16VECTOR K
 -- procedure: u16vector-set! U16VECTOR K EXACT-INT16
 -- procedure: u16vector->list U16VECTOR
 -- procedure: list->u16vector LIST-OF-EXACT-INT16
 -- procedure: u16vector-fill! U16VECTOR FILL
 -- procedure: u16vector-copy U16VECTOR
 -- procedure: u16vector-append U16VECTOR...
 -- procedure: subu16vector U16VECTOR START END

 -- procedure: s32vector? OBJ
 -- procedure: make-s32vector K [FILL]
 -- procedure: s32vector EXACT-INT32...
 -- procedure: s32vector-length S32VECTOR
 -- procedure: s32vector-ref S32VECTOR K
 -- procedure: s32vector-set! S32VECTOR K EXACT-INT32
 -- procedure: s32vector->list S32VECTOR
 -- procedure: list->s32vector LIST-OF-EXACT-INT32
 -- procedure: s32vector-fill! S32VECTOR FILL
 -- procedure: s32vector-copy S32VECTOR
 -- procedure: s32vector-append S32VECTOR...
 -- procedure: subs32vector S32VECTOR START END

 -- procedure: u32vector? OBJ
 -- procedure: make-u32vector K [FILL]
 -- procedure: u32vector EXACT-INT32...
 -- procedure: u32vector-length U32VECTOR
 -- procedure: u32vector-ref U32VECTOR K
 -- procedure: u32vector-set! U32VECTOR K EXACT-INT32
 -- procedure: u32vector->list U32VECTOR
 -- procedure: list->u32vector LIST-OF-EXACT-INT32
 -- procedure: u32vector-fill! U32VECTOR FILL
 -- procedure: u32vector-copy U32VECTOR
 -- procedure: u32vector-append U32VECTOR...
 -- procedure: subu32vector U32VECTOR START END

 -- procedure: s64vector? OBJ
 -- procedure: make-s64vector K [FILL]
 -- procedure: s64vector EXACT-INT64...
 -- procedure: s64vector-length S64VECTOR
 -- procedure: s64vector-ref S64VECTOR K
 -- procedure: s64vector-set! S64VECTOR K EXACT-INT64
 -- procedure: s64vector->list S64VECTOR
 -- procedure: list->s64vector LIST-OF-EXACT-INT64
 -- procedure: s64vector-fill! S64VECTOR FILL
 -- procedure: s64vector-copy S64VECTOR
 -- procedure: s64vector-append S64VECTOR...
 -- procedure: subs64vector S64VECTOR START END

 -- procedure: u64vector? OBJ
 -- procedure: make-u64vector K [FILL]
 -- procedure: u64vector EXACT-INT64...
 -- procedure: u64vector-length U64VECTOR
 -- procedure: u64vector-ref U64VECTOR K
 -- procedure: u64vector-set! U64VECTOR K EXACT-INT64
 -- procedure: u64vector->list U64VECTOR
 -- procedure: list->u64vector LIST-OF-EXACT-INT64
 -- procedure: u64vector-fill! U64VECTOR FILL
 -- procedure: u64vector-copy U64VECTOR
 -- procedure: u64vector-append U64VECTOR...
 -- procedure: subu64vector U64VECTOR START END

 -- procedure: f32vector? OBJ
 -- procedure: make-f32vector K [FILL]
 -- procedure: f32vector INEXACT-REAL...
 -- procedure: f32vector-length F32VECTOR
 -- procedure: f32vector-ref F32VECTOR K
 -- procedure: f32vector-set! F32VECTOR K INEXACT-REAL
 -- procedure: f32vector->list F32VECTOR
 -- procedure: list->f32vector LIST-OF-INEXACT-REAL
 -- procedure: f32vector-fill! F32VECTOR FILL
 -- procedure: f32vector-copy F32VECTOR
 -- procedure: f32vector-append F32VECTOR...
 -- procedure: subf32vector F32VECTOR START END

 -- procedure: f64vector? OBJ
 -- procedure: make-f64vector K [FILL]
 -- procedure: f64vector INEXACT-REAL...
 -- procedure: f64vector-length F64VECTOR
 -- procedure: f64vector-ref F64VECTOR K
 -- procedure: f64vector-set! F64VECTOR K INEXACT-REAL
 -- procedure: f64vector->list F64VECTOR
 -- procedure: list->f64vector LIST-OF-INEXACT-REAL
 -- procedure: f64vector-fill! F64VECTOR FILL
 -- procedure: f64vector-copy F64VECTOR
 -- procedure: f64vector-append F64VECTOR...
 -- procedure: subf64vector F64VECTOR START END

   For example:

     > (define v (u8vector 10 255 13))
     > (u8vector-set! v 2 99)
     > v
     #u8(10 255 99)
     > (u8vector-ref v 1)
     255
     > (u8vector->list v)
     (10 255 99)

 -- procedure: object->u8vector OBJ [ENCODER]
 -- procedure: u8vector->object U8VECTOR [DECODER]
     The procedure `object->u8vector' returns a u8vector that contains
     the sequence of bytes that encodes the object OBJ.  The procedure
     `u8vector->object' decodes the sequence of bytes contained in the
     u8vector U8VECTOR, which was produced by the procedure
     `object->u8vector', and reconstructs an object structurally equal
     to the original object.  In other words the procedures
     `object->u8vector' and `u8vector->object' respectively perform
     serialization and deserialization of Scheme objects.  Note that
     some objects are non-serializable (e.g. threads, wills, some types
     of ports, and any object containing a non-serializable object).

     The optional ENCODER and DECODER parameters are single parameter
     procedures which default to the identity function.  The ENCODER
     procedure is called during serialization.  As the serializer walks
     through OBJ, it calls the ENCODER procedure on each sub-object X
     that is encountered.  The ENCODER transforms the object X into an
     object Y that will be serialized instead of X.  Similarly the
     DECODER procedure is called during deserialization.  When an
     object Y is encountered, the DECODER procedure is called to
     transform it into the object X that is the result of
     deserialization.

     The ENCODER and DECODER procedures are useful to customize the
     serialized representation of objects.  In particular, it can be
     used to define the semantics of serializing objects, such as
     threads and ports, that would otherwise not be serializable.  The
     DECODER procedure is typically the inverse of the ENCODER
     procedure, i.e. `(DECODER (ENCODER X))' = `X'.

     For example:

          > (define (make-adder x) (lambda (y) (+ x y)))
          > (define f (make-adder 10))
          > (define a (object->u8vector f))
          > (define b (u8vector->object a))
          > (u8vector-length a)
          1639
          > (f 5)
          15
          > (b 5)
          15
          > (pp b)
          (lambda (y) (+ x y))


11 Hashing and weak references
******************************

11.1 Hashing
============

 -- procedure: object->serial-number OBJ
 -- procedure: serial-number->object N [DEFAULT]
     All Scheme objects are uniquely identified with a serial number
     which is a nonnegative exact integer.  The `object->serial-number'
     procedure returns the serial number of object OBJ.  This serial
     number is only allocated the first time the `object->serial-number'
     procedure is called on that object.  Objects which do not have an
     external textual representation that can be read by the `read'
     procedure, use an external textual representation that includes a
     serial number of the form `#N'.  Consequently, the procedures
     `write', `pretty-print', etc will call the `object->serial-number'
     procedure to get the serial number, and this may cause the serial
     number to be allocated.

     The `serial-number->object' procedure takes an exact integer
     argument N and returns the object whose serial number is N.  If no
     object currently exists with that serial number, DEFAULT is
     returned if it is specified, otherwise an
     unbound-serial-number-exception object is raised.  The reader
     defines the following abbreviation for calling
     `serial-number->object': the syntax `#N', where N is a sequence of
     decimal digits and it is not followed by ``='' or ``#'', is
     equivalent to the list `(serial-number->object N)'.

     For example:

          > (define z (list (lambda (x) (* x x)) (lambda (y) (/ 1 y))))
          > z
          (#<procedure #2> #<procedure #3>)
          > (#3 10)
          1/10
          > '(#3 10)
          ((serial-number->object 3) 10)
          > car
          #<procedure #4 car>
          > (#4 z)
          #<procedure #2>


 -- procedure: unbound-serial-number-exception? OBJ
 -- procedure: unbound-serial-number-exception-procedure EXC
 -- procedure: unbound-serial-number-exception-arguments EXC
     Unbound-serial-number-exception objects are raised by the procedure
     `serial-number->object' when no object currently exists with that
     serial number.  The parameter EXC must be an
     unbound-serial-number-exception object.

     The procedure `unbound-serial-number-exception?' returns `#t' when
     OBJ is a unbound-serial-number-exception object and `#f' otherwise.

     The procedure `unbound-serial-number-exception-procedure' returns
     the procedure that raised EXC.

     The procedure `unbound-serial-number-exception-arguments' returns
     the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (unbound-serial-number-exception? exc)
                  (list (unbound-serial-number-exception-procedure exc)
                        (unbound-serial-number-exception-arguments exc))
                  'not-unbound-serial-number-exception))
          > (with-exception-catcher
              handler
              (lambda () (serial-number->object 1000)))
          (#<procedure #2 serial-number->object> (1000))


 -- procedure: symbol-hash SYMBOL
     The `symbol-hash' procedure returns the hash number of the symbol
     SYMBOL.  The hash number is a small exact integer (fixnum).  When
     SYMBOL is an interned symbol the value returned is the same as
     `(string=?-hash (symbol->string SYMBOL))'.

     For example:

          > (symbol-hash 'car)
          444471047


 -- procedure: keyword-hash KEYWORD
     The `keyword-hash' procedure returns the hash number of the
     keyword KEYWORD.  The hash number is a small exact integer
     (fixnum).  When KEYWORD is an interned keyword the value returned
     is the same as `(string=?-hash (keyword->string KEYWORD))'.

     For example:

          > (keyword-hash car:)
          444471047


 -- procedure: string=?-hash STRING
     The `string=?-hash' procedure returns the hash number of the
     string STRING.  The hash number is a small exact integer (fixnum).
     For any two strings S1 and S2, `(string=?  S1 S2)' implies `(=
     (string=?-hash S1) (string=?-hash S2))'.

     For example:

          > (string=?-hash "car")
          444471047


 -- procedure: string-ci=?-hash STRING
     The `string-ci=?-hash' procedure returns the hash number of the
     string STRING.  The hash number is a small exact integer (fixnum).
     For any two strings S1 and S2, `(string-ci=?  S1 S2)' implies `(=
     (string-ci=?-hash S1) (string-ci=?-hash S2))'.

     For example:

          > (string-ci=?-hash "CaR")
          444471047


 -- procedure: eq?-hash OBJ
     The `eq?-hash' procedure returns the hash number of the object
     OBJ.  The hash number is a small exact integer (fixnum).  For any
     two objects O1 and O2, `(eq? O1 O2)' implies `(= (eq?-hash O1)
     (eq?-hash O2))'.

     For example:

          > (eq?-hash #t)
          536870910


 -- procedure: eqv?-hash OBJ
     The `eqv?-hash' procedure returns the hash number of the object
     OBJ.  The hash number is a small exact integer (fixnum).  For any
     two objects O1 and O2, `(eqv? O1 O2)' implies `(= (eqv?-hash O1)
     (eqv?-hash O2))'.

     For example:

          > (eqv?-hash 1.5)
          496387656


 -- procedure: equal?-hash OBJ
     The `equal?-hash' procedure returns the hash number of the object
     OBJ.  The hash number is a small exact integer (fixnum).  For any
     two objects O1 and O2, `(equal? O1 O2)' implies `(= (equal?-hash
     O1) (equal?-hash O2))'.

     For example:

          > (equal?-hash (list 1 2 3))
          442438567


11.2 Weak references
====================

The garbage collector is responsible for reclaiming objects that are no
longer needed by the program.  This is done by analyzing the
reachability graph of all objects from the roots (i.e. the global
variables, the runnable threads, permanently allocated objects such as
procedures defined in a compiled file, nonexecutable wills, etc).  If a
root or a reachable object X contains a reference to an object Y then Y
is reachable.  As a general rule, unreachable objects are reclaimed by
the garbage collector.

   There are two types of references: strong references and weak
references.  Most objects, including pairs, vectors, records and
closures, contain strong references.  An object X is "strongly
reachable" if there is a path from the roots to X that traverses only
strong references.  Weak references only occur in wills and tables.
There are two types of weak references: will-weak references and
table-weak references.  If all paths from the roots to an object Y
traverse at least one table-weak reference, then Y will be reclaimed by
the garbage collector.  The will-weak references are used for
finalization and are explained in the next section.

11.2.1 Wills
------------

The following procedures implement the "will" data type.  Will objects
provide support for finalization.  A will is an object that contains a
will-weak reference to a TESTATOR object (the object attached to the
will), and a strong reference to an ACTION procedure which is a one
parameter procedure which is called when the will is executed.

 -- procedure: make-will TESTATOR ACTION
 -- procedure: will? OBJ
 -- procedure: will-testator WILL
 -- procedure: will-execute! WILL
     The `make-will' procedure creates a will object with the given
     TESTATOR object and ACTION procedure.  The `will?' procedure tests
     if OBJ is a will object.  The `will-testator' procedure gets the
     testator object attached to the WILL.  The `will-execute!'
     procedure executes WILL.

     A will becomes "executable" when its TESTATOR object is not
     strongly reachable (i.e. the TESTATOR object is either unreachable
     or only reachable using paths from the roots that traverse at
     least one weak reference).  Some objects, including symbols, small
     exact integers (fixnums), booleans and characters, are considered
     to be always strongly reachable.

     When the runtime system detects that a will has become executable
     the current computation is interrupted, the will's testator is set
     to `#f' and the will's action procedure is called with the will's
     testator as the sole argument.  Currently only the garbage
     collector detects when wills become executable but this may change
     in future versions of Gambit (for example the compiler could
     perform an analysis to infer will executability at compile time).
     The garbage collector builds a list of all executable wills.
     Shortly after a garbage collection, the action procedures of these
     wills will be called.  The link from the will to the action
     procedure is severed when the action procedure is called.

     Note that the testator object will not be reclaimed during the
     garbage collection that determined executability of the will.  It
     is only when an object is not reachable from the roots that it is
     reclaimed by the garbage collector.

     A remarkable feature of wills is that an action procedure can
     "resurrect" an object.  An action procedure could for example
     assign the testator object to a global variable or create a new
     will with the same testator object.

     For example:

          > (define a (list 123))
          > (set-cdr! a a) ; create a circular list
          > (define b (vector a))
          > (define c #f)
          > (define w
              (let ((obj a))
                (make-will obj
                           (lambda (x) ; x will be eq? to obj
                             (display "executing action procedure")
                             (newline)
                             (set! c x)))))
          > (will? w)
          #t
          > (car (will-testator w))
          123
          > (##gc)
          > (set! a #f)
          > (##gc)
          > (set! b #f)
          > (##gc)
          executing action procedure
          > (will-testator w)
          #f
          > (car c)
          123


11.2.2 Tables
-------------

The following procedures implement the "table" data type.  Tables are
heterogenous structures whose elements are indexed by keys which are
arbitrary objects.  Tables are similar to association lists but are
abstract and the access time for large tables is typically smaller.
Each key contained in the table is bound to a value.  The length of the
table is the number of key/value bindings it contains.  New key/value
bindings can be added to a table, the value bound to a key can be
changed, and existing key/value bindings can be removed.

   The references to the keys can either be all strong or all table-weak
and the references to the values can either be all strong or all
table-weak.  The garbage collector removes key/value bindings from a
table when 1) the key is a table-weak reference and the key is
unreachable or only reachable using paths from the roots that traverse
at least one table-weak reference, or 2) the value is a table-weak
reference and the value is unreachable or only reachable using paths
from the roots that traverse at least one table-weak reference.
Key/value bindings that are removed by the garbage collector are
reclaimed immediately.

   Although there are several possible ways of implementing tables, the
current implementation uses hashing with open-addressing.  This is
space efficient and provides constant-time access.  Hash tables are
automatically resized to maintain the load within specified bounds.
The load is the number of active entries (the length of the table)
divided by the total number of entries in the hash table.

   Tables are parameterized with a key comparison procedure.  By default
the `equal?' procedure is used, but `eq?', `eqv?', `string=?',
`string-ci=?', or a user defined procedure can also be used.  To
support arbitrary key comparison procedures, tables are also
parameterized with a hashing procedure accepting a key as its single
parameter and returning a fixnum result.  The hashing procedure HASH
must be consistent with the key comparison procedure TEST, that is, for
any two keys K1 and K2 in the table, `(TEST K1 K2)' implies `(= (HASH
K1) (HASH K2))'.  A default hashing procedure consistent with the key
comparison procedure is provided by the system.  The default hashing
procedure generally gives good performance when the key comparison
procedure is `eq?', `eqv?', `equal?', `string=?', and `string-ci=?'.
However, for user defined key comparison procedures, the default
hashing procedure always returns 0.  This degrades the performance of
the table to a linear search.

   Tables can be compared for equality using the `equal?' procedure.
Two tables `X' and `Y' are considered equal by `equal?' when they have
the same weakness attributes, the same key comparison procedure, the
same hashing procedure, the same length, and for all the keys `K' in
`X', `(equal?  (table-ref X K) (table-ref Y K))'.

 -- procedure: make-table [`size:' SIZE] [`init:' INIT] [`weak-keys:'
          WEAK-KEYS] [`weak-values:' WEAK-VALUES] [`test:' TEST]
          [`hash:' HASH] [`min-load:' MIN-LOAD] [`max-load:' MAX-LOAD]
     The procedure `make-table' returns a new table.  The optional
     keyword parameters specify various parameters of the table.

     The SIZE parameter is a nonnegative exact integer indicating the
     expected length of the table.  The system uses SIZE to choose an
     appropriate initial size of the hash table so that it does not
     need to be resized too often.

     The INIT parameter indicates a value that is associated to keys
     that are not in the table.  When INIT is not specified, no value
     is associated to keys that are not in the table.

     The WEAK-KEYS and WEAK-VALUES parameters are extended booleans
     indicating respectively whether the keys and values are table-weak
     references (true) or strong references (false).  By default the
     keys and values are strong references.

     The TEST parameter indicates the key comparison procedure.  The
     default key comparison procedure is `equal?'.  The key comparison
     procedures `eq?', `eqv?', `equal?', `string=?', and `string-ci=?'
     are special because the system will use a reasonably good hash
     procedure when none is specified.

     The HASH parameter indicates the hash procedure.  This procedure
     must accept a single key parameter, return a fixnum, and be
     consistent with the key comparison procedure.  When HASH is not
     specified, a default hash procedure is used.  The default hash
     procedure is reasonably good when the key comparison procedure is
     `eq?', `eqv?', `equal?', `string=?', or `string-ci=?'.

     The MIN-LOAD and MAX-LOAD parameters are real numbers that
     indicate the minimum and maximum load of the table respectively.
     The table is resized when adding or deleting a key/value binding
     would bring the table's load outside of this range.  The MIN-LOAD
     parameter must be no less than 0.05 and the MAX-LOAD parameter
     must be no greater than 0.95.  Moreover the difference between
     MIN-LOAD and MAX-LOAD must be at least 0.20.  When MIN-LOAD is not
     specified, the value 0.45 is used.  When MAX-LOAD is not
     specified, the value 0.90 is used.

     For example:

          > (define t (make-table))
          > (table? t)
          #t
          > (table-length t)
          0
          > (table-set! t (list 1 2) 3)
          > (table-set! t (list 4 5) 6)
          > (table-ref t (list 1 2))
          3
          > (table-length t)
          2


 -- procedure: table? OBJ
     The procedure `table?' returns `#t' when OBJ is a table and `#f'
     otherwise.

     For example:

          > (table? (make-table))
          #t
          > (table? 123)
          #f


 -- procedure: table-length TABLE
     The procedure `table-length' returns the number of key/value
     bindings contained in the table TABLE.

     For example:

          > (define t (make-table weak-keys: #t))
          > (define x (list 1 2))
          > (define y (list 3 4))
          > (table-set! t x 111)
          > (table-set! t y 222)
          > (table-length t)
          2
          > (table-set! t x)
          > (table-length t)
          1
          > (##gc)
          > (table-length t)
          1
          > (set! y #f)
          > (##gc)
          > (table-length t)
          0


 -- procedure: table-ref TABLE KEY [DEFAULT]
     The procedure `table-ref' returns the value bound to the object
     KEY in the table TABLE.  When KEY is not bound and DEFAULT is
     specified, DEFAULT is returned.  When DEFAULT is not specified but
     an INIT parameter was specified when TABLE was created, INIT is
     returned.  Otherwise an unbound-table-key-exception object is
     raised.

     For example:

          > (define t1 (make-table init: 999))
          > (table-set! t1 (list 1 2) 3)
          > (table-ref t1 (list 1 2))
          3
          > (table-ref t1 (list 4 5))
          999
          > (table-ref t1 (list 4 5) #f)
          #f
          > (define t2 (make-table))
          > (table-ref t2 (list 4 5))
          *** ERROR IN (console)@7.1 -- Unbound table key
          (table-ref '#<table #2> '(4 5))


 -- procedure: table-set! TABLE KEY [VALUE]
     The procedure `table-set!' binds the object KEY to VALUE in the
     table TABLE.  When VALUE is not specified, if TABLE contains a
     binding for KEY then the binding is removed from TABLE.  The
     procedure `table-set!' returns an unspecified value.

     For example:

          > (define t (make-table))
          > (table-set! t (list 1 2) 3)
          > (table-set! t (list 4 5) 6)
          > (table-set! t (list 4 5))
          > (table-set! t (list 7 8))
          > (table-ref t (list 1 2))
          3
          > (table-ref t (list 4 5))
          *** ERROR IN (console)@7.1 -- Unbound table key
          (table-ref '#<table #2> '(4 5))


 -- procedure: table-search PROC TABLE
     The procedure `table-search' searches the table TABLE for a
     key/value binding for which the two argument procedure PROC
     returns a non false result.  For each key/value binding visited by
     `table-search' the procedure PROC is called with the key as the
     first argument and the value as the second argument.  The
     procedure `table-search' returns the first non false value
     returned by PROC, or `#f' if PROC returned `#f' for all key/value
     bindings in TABLE.

     The order in which the key/value bindings are visited is
     unspecified and may vary from one call of `table-search' to the
     next.  While a call to `table-search' is being performed on TABLE,
     it is an error to call any of the following procedures on TABLE:
     `table-ref', `table-set!', `table-search', `table-for-each', and
     `table->list'.  It is also an error to compare with `equal?'
     (directly or indirectly with `member', `assoc', `table-ref', etc.)
     an object that contains TABLE.  All these procedures may cause
     TABLE to be reordered and resized.  This restriction allows a more
     efficient iteration over the key/value bindings.

     For example:

          > (define square (make-table))
          > (table-set! square 2 4)
          > (table-set! square 3 9)
          > (table-search (lambda (k v) (and (odd? k) v)) square)
          9


 -- procedure: table-for-each PROC TABLE
     The procedure `table-for-each' calls the two argument procedure
     PROC for each key/value binding in the table TABLE.  The procedure
     PROC is called with the key as the first argument and the value as
     the second argument.  The procedure `table-for-each' returns an
     unspecified value.

     The order in which the key/value bindings are visited is
     unspecified and may vary from one call of `table-for-each' to the
     next.  While a call to `table-for-each' is being performed on
     TABLE, it is an error to call any of the following procedures on
     TABLE: `table-ref', `table-set!', `table-search',
     `table-for-each', and `table->list'.  It is also an error to
     compare with `equal?' (directly or indirectly with `member',
     `assoc', `table-ref', etc.) an object that contains TABLE.  All
     these procedures may cause TABLE to be reordered and resized.
     This restriction allows a more efficient iteration over the
     key/value bindings.

     For example:

          > (define square (make-table))
          > (table-set! square 2 4)
          > (table-set! square 3 9)
          > (table-for-each (lambda (k v) (write (list k v)) (newline)) square)
          (2 4)
          (3 9)


 -- procedure: table->list TABLE
     The procedure `table->list' returns an association list containing
     the key/value bindings in the table TABLE.  Each key/value binding
     yields a pair whose car field is the key and whose cdr field is
     the value bound to that key.  The order of the bindings in the
     list is unspecified.

     For example:

          > (define square (make-table))
          > (table-set! square 2 4)
          > (table-set! square 3 9)
          > (table->list square)
          ((3 . 9) (2 . 4))


 -- procedure: list->table LIST [`size:' SIZE] [`init:' INIT]
          [`weak-keys:' WEAK-KEYS] [`weak-values:' WEAK-VALUES]
          [`test:' TEST] [`hash:' HASH] [`min-load:' MIN-LOAD]
          [`max-load:' MAX-LOAD]
     The procedure `list->table' returns a new table containing the
     key/value bindings in the association list LIST.  The optional
     keyword parameters specify various parameters of the table and have
     the same meaning as for the `make-table' procedure.

     Each element of LIST is a pair whose car field is a key and whose
     cdr field is the value bound to that key.  If a key appears more
     than once in LIST (tested using the table's key comparison
     procedure) it is the first key/value binding in LIST that has
     precedence.

     For example:

          > (define t (list->table '((b . 2) (a . 1) (c . 3) (a . 4))))
          > (table->list t)
          ((a . 1) (b . 2) (c . 3))


 -- procedure: unbound-table-key-exception? OBJ
 -- procedure: unbound-table-key-exception-procedure EXC
 -- procedure: unbound-table-key-exception-arguments EXC
     Unbound-table-key-exception objects are raised by the procedure
     `table-ref' when the key does not have a binding in the table.
     The parameter EXC must be an unbound-table-key-exception object.

     The procedure `unbound-table-key-exception?' returns `#t' when OBJ
     is a unbound-table-key-exception object and `#f' otherwise.

     The procedure `unbound-table-key-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `unbound-table-key-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define t (make-table))
          > (define (handler exc)
              (if (unbound-table-key-exception? exc)
                  (list (unbound-table-key-exception-procedure exc)
                        (unbound-table-key-exception-arguments exc))
                  'not-unbound-table-key-exception))
          > (with-exception-catcher
              handler
              (lambda () (table-ref t '(1 2))))
          (#<procedure #2 table-ref> (#<table #3> (1 2)))


 -- procedure: table-copy TABLE
     The procedure `table-copy' returns a new table containing the same
     key/value bindings as TABLE and the same table parameters (i.e.
     hash procedure, key comparison procedure, key and value weakness,
     etc).

     For example:

          > (define t (list->table '((b . 2) (a . 1) (c . 3))))
          > (define x (table-copy t))
          > (table-set! t 'b 99)
          > (table->list t)
          ((a . 1) (b . 99) (c . 3))
          > (table->list x)
          ((a . 1) (b . 2) (c . 3))


12 Records
**********

 -- special form: define-structure name field...
     Record data types similar to Pascal records and C `struct' types
     can be defined using the `define-structure' special form.  The
     identifier name specifies the name of the new data type.  The
     structure name is followed by K identifiers naming each field of
     the record.  The `define-structure' expands into a set of
     definitions of the following procedures:

        * `make-name' - A K argument procedure which constructs a new
          record from the value of its K fields.

        * `name?' - A procedure which tests if its single argument is
          of the given record type.

        * `name-field' - For each field, a procedure taking as its
          single argument a value of the given record type and returning
          the content of the corresponding field of the record.

        * `name-field-set!' - For each field, a two argument procedure
          taking as its first argument a value of the given record
          type.  The second argument gets assigned to the corresponding
          field of the record and the void object is returned.


     Record data types have a printed representation that includes the
     name of the type and the name and value of each field.  Record
     data types can not be read by the `read' procedure.

     For example:

          > (define-structure point x y color)
          > (define p (make-point 3 5 'red))
          > p
          #<point #2 x: 3 y: 5 color: red>
          > (point-x p)
          3
          > (point-color p)
          red
          > (point-color-set! p 'black)
          > p
          #<point #2 x: 3 y: 5 color: black>


13 Threads
**********

Gambit supports the execution of multiple Scheme threads.  These
threads are managed entirely by Gambit's runtime and are not related to
the host operating system's threads.  Gambit's runtime does not
currently take advantage of multiprocessors (i.e. at most one thread is
running).

13.1 Introduction
=================

Multithreading is a paradigm that is well suited for building complex
systems such as: servers, GUIs, and high-level operating systems.
Gambit's thread system offers mechanisms for creating threads of
execution and for synchronizing them.  The thread system also supports
features which are useful in a real-time context, such as priorities,
priority inheritance and timeouts.

   The thread system provides the following data types:

   * Thread (a virtual processor which shares object space with all
     other threads)

   * Mutex (a mutual exclusion device, also known as a lock and binary
     semaphore)

   * Condition variable (a set of blocked threads)


13.2 Thread objects
===================

A "running thread" is a thread that is currently executing.  A
"runnable thread" is a thread that is ready to execute or running.  A
thread is "blocked" if it is waiting for a mutex to become unlocked, an
I/O operation to become possible, the end of a "sleep" period, etc.  A
"new thread" is a thread that has been allocated but has not yet been
initialized.  An "initialized thread" is a thread that can be made
runnable.  A new thread becomes runnable when it is started by calling
`thread-start!'.  A "terminated thread" is a thread that can no longer
become runnable (but "deadlocked threads" are not considered
terminated).  The only valid transitions between the thread states are
from new to initialized, from initialized to runnable, between runnable
and blocked, and from any state except new to terminated as indicated in
the following diagram:

                                                 unblock
                               start            <-------
     NEW -------> INITIALIZED -------> RUNNABLE -------> BLOCKED
                            \             |      block  /
                             \            v            /
                              +-----> TERMINATED <----+

   Each thread has a "base priority", which is a real number (where a
higher numerical value means a higher priority), a "priority boost",
which is a nonnegative real number representing the priority increase
applied to a thread when it blocks, and a "quantum", which is a
nonnegative real number representing a duration in seconds.

   Each thread has a "specific field" which can be used in an
application specific way to associate data with the thread (some thread
systems call this "thread local storage").

   Each thread has a "mailbox" which is used for inter-thread
communication.

13.3 Mutex objects
==================

A mutex can be in one of four states: "locked" (either "owned" or "not
owned") and "unlocked" (either "abandoned" or "not abandoned").

   An attempt to lock a mutex only succeeds if the mutex is in an
unlocked state, otherwise the current thread will wait.  A mutex in the
locked/owned state has an associated "owner thread", which by
convention is the thread that is responsible for unlocking the mutex
(this case is typical of critical sections implemented as "lock mutex,
perform operation, unlock mutex").  A mutex in the locked/not-owned
state is not linked to a particular thread.

   A mutex becomes locked when a thread locks it using the
`mutex-lock!' primitive.  A mutex becomes unlocked/abandoned when the
owner of a locked/owned mutex terminates.  A mutex becomes
unlocked/not-abandoned when a thread unlocks it using the
`mutex-unlock!' primitive.

   The mutex primitives do not implement "recursive mutex semantics".
An attempt to lock a mutex that is locked implies that the current
thread waits even if the mutex is owned by the current thread (this can
lead to a deadlock if no other thread unlocks the mutex).

   Each mutex has a "specific field" which can be used in an
application specific way to associate data with the mutex.

13.4 Condition variable objects
===============================

A condition variable represents a set of blocked threads.  These blocked
threads are waiting for a certain condition to become true.  When a
thread modifies some program state that might make the condition true,
the thread unblocks some number of threads (one or all depending on the
primitive used) so they can check if the condition is now true.  This
allows complex forms of interthread synchronization to be expressed more
conveniently than with mutexes alone.

   Each condition variable has a "specific field" which can be used in
an application specific way to associate data with the condition
variable.

13.5 Fairness
=============

In various situations the scheduler must select one thread from a set of
threads (e.g. which thread to run when a running thread blocks or
expires its quantum, which thread to unblock when a mutex becomes
unlocked or a condition variable is signaled).  The constraints on the
selection process determine the scheduler's "fairness".  The selection
depends on the order in which threads become runnable or blocked and on
the "priority" attached to the threads.

   The definition of fairness requires the notion of time ordering,
i.e. "event A occured before event B".  For the purpose of establishing
time ordering, the scheduler uses a clock with a discrete, usually
variable, resolution (a "tick").  Events occuring in a given tick can
be considered to be simultaneous (i.e. if event A occured before event
B in real time, then the scheduler will claim that event A occured
before event B unless both events fall within the same tick, in which
case the scheduler arbitrarily chooses a time ordering).

   Each thread T has three priorities which affect fairness; the "base
priority", the "boosted priority", and the "effective priority".

   * The "base priority" is the value contained in T's "base priority"
     field (which is set with the `thread-base-priority-set!'
     primitive).

   * T's "boosted flag" field contains a boolean that affects T's
     "boosted priority".  When the boosted flag field is false, the
     boosted priority is equal to the base priority, otherwise the
     boosted priority is equal to the base priority plus the value
     contained in T's "priority boost" field (which is set with the
     `thread-priority-boost-set!' primitive).  The boosted flag field is
     set to false when a thread is created, when its quantum expires,
     and when "thread-yield!" is called.  The boosted flag field is set
     to true when a thread blocks.  By carefully choosing the base
     priority and priority boost, relatively to the other threads, it
     is possible to set up an interactive thread so that it has good
     I/O response time without being a CPU hog when it performs long
     computations.

   * The "effective priority" is equal to the maximum of T's boosted
     priority and the effective priority of all the threads that are
     blocked on a mutex owned by T.  This "priority inheritance" avoids
     priority inversion problems that would prevent a high priority
     thread blocked at the entry of a critical section to progress
     because a low priority thread inside the critical section is
     preempted for an arbitrary long time by a medium priority thread.


   Let P(T) be the effective priority of thread T and let R(T) be the
most recent time when one of the following events occurred for thread
T, thus making it runnable: T was started by calling `thread-start!', T
called `thread-yield!', T expired its quantum, or T became unblocked.
Let the relation NL(T1,T2), "T1 no later than T2", be true if
P(T1)<P(T2) or P(T1)=P(T2) and R(T1)>R(T2), and false otherwise.  The
scheduler will schedule the execution of threads in such a way that
whenever there is at least one runnable thread, 1) within a finite time
at least one thread will be running, and 2) there is never a pair of
runnable threads T1 and T2 for which NL(T1,T2) is true and T1 is not
running and T2 is running.

   A thread T expires its quantum when an amount of time equal to T's
quantum has elapsed since T entered the running state and T did not
block, terminate or call `thread-yield!'.  At that point T exits the
running state to allow other threads to run.  A thread's quantum is
thus an indication of the rate of progress of the thread relative to
the other threads of the same priority.  Moreover, the resolution of
the timer measuring the running time may cause a certain deviation from
the quantum, so a thread's quantum should only be viewed as an
approximation of the time it can run before yielding to another thread.

   Threads blocked on a given mutex or condition variable will unblock
in an order which is consistent with decreasing priority and increasing
blocking time (i.e. the highest priority thread unblocks first, and
among equal priority threads the one that blocked first unblocks first).

13.6 Memory coherency
=====================

Read and write operations on the store (such as reading and writing a
variable, an element of a vector or a string) are not atomic.  It is an
error for a thread to write a location in the store while some other
thread reads or writes that same location.  It is the responsibility of
the application to avoid write/read and write/write races through
appropriate uses of the synchronization primitives.

   Concurrent reads and writes to ports are allowed.  It is the
responsibility of the implementation to serialize accesses to a given
port using the appropriate synchronization primitives.

13.7 Timeouts
=============

All synchronization primitives which take a timeout parameter accept
three types of values as a timeout, with the following meaning:

   * a time object represents an absolute point in time

   * an exact or inexact real number represents a relative time in
     seconds from the moment the primitive was called

   * `#f' means that there is no timeout


   When a timeout denotes the current time or a time in the past, the
synchronization primitive claims that the timeout has been reached only
after the other synchronization conditions have been checked.  Moreover
the thread remains running (it does not enter the blocked state).  For
example, `(mutex-lock! m 0)' will lock mutex `m' and return `#t' if `m'
is currently unlocked, otherwise `#f' is returned because the timeout
is reached.

13.8 Primordial thread
======================

The execution of a program is initially under the control of a single
thread known as the "primordial thread".  The primordial thread has an
unspecified base priority, priority boost, boosted flag, quantum, name,
specific field, dynamic environment, `dynamic-wind' stack, and
exception-handler.  All threads are terminated when the primordial
thread terminates (normally or not).

13.9 Procedures
===============

 -- procedure: current-thread
     This procedure returns the current thread.  For example:

          > (current-thread)
          #<thread #1 primordial>
          > (eq? (current-thread) (current-thread))
          #t


 -- procedure: thread? OBJ
     This procedure returns `#t' when OBJ is a thread object and `#f'
     otherwise.

     For example:

          > (thread? (current-thread))
          #t
          > (thread? 'foo)
          #f


 -- procedure: make-thread THUNK [NAME [THREAD-GROUP]]
     This procedure creates and returns an initialized thread.  This
     thread is not automatically made runnable (the procedure
     `thread-start!' must be used for this).  A thread has the
     following fields: base priority, priority boost, boosted flag,
     quantum, name, specific, end-result, end-exception, and a list of
     locked/owned mutexes it owns.  The thread's execution consists of
     a call to THUNK with the "initial continuation".  This
     continuation causes the (then) current thread to store the result
     in its end-result field, abandon all mutexes it owns, and finally
     terminate.  The `dynamic-wind' stack of the initial continuation
     is empty.  The optional NAME is an arbitrary Scheme object which
     identifies the thread (useful for debugging); it defaults to an
     unspecified value.  The specific field is set to an unspecified
     value.  The optional THREAD-GROUP indicates which thread group
     this thread belongs to; it defaults to the thread group of the
     current thread.  The base priority, priority boost, and quantum of
     the thread are set to the same value as the current thread and the
     boosted flag is set to false.  The thread's mailbox is initially
     empty.  The thread inherits the dynamic environment from the
     current thread. Moreover, in this dynamic environment the
     exception-handler is bound to the "initial exception-handler"
     which is a unary procedure which causes the (then) current thread
     to store in its end-exception field an uncaught-exception object
     whose "reason" is the argument of the handler, abandon all mutexes
     it owns, and finally terminate.

     For example:

          > (make-thread (lambda () (write 'hello)))
          #<thread #2>
          > (make-thread (lambda () (write 'world)) 'a-name)
          #<thread #3 a-name>


 -- procedure: thread-name THREAD
     This procedure returns the name of the THREAD.  For example:

          > (thread-name (make-thread (lambda () #f) 'foo))
          foo


 -- procedure: thread-specific THREAD
 -- procedure: thread-specific-set! THREAD OBJ
     The `thread-specific' procedure returns the content of the
     THREAD's specific field.

     The `thread-specific-set!' procedure stores OBJ into the THREAD's
     specific field and returns an unspecified value.

     For example:

          > (thread-specific-set! (current-thread) "hello")
          > (thread-specific (current-thread))
          "hello"


 -- procedure: thread-base-priority THREAD
 -- procedure: thread-base-priority-set! THREAD PRIORITY
     The procedure `thread-base-priority' returns a real number which
     corresponds to the base priority of the THREAD.

     The procedure `thread-base-priority-set!' changes the base
     priority of the THREAD to PRIORITY and returns an unspecified
     value.  The PRIORITY must be a real number.

     For example:

          > (thread-base-priority-set! (current-thread) 12.3)
          > (thread-base-priority (current-thread))
          12.3


 -- procedure: thread-priority-boost THREAD
 -- procedure: thread-priority-boost-set! THREAD PRIORITY-BOOST
     The procedure `thread-priority-boost' returns a real number which
     corresponds to the priority boost of the THREAD.

     The procedure `thread-priority-boost-set!' changes the priority
     boost of the THREAD to PRIORITY-BOOST and returns an unspecified
     value.  The PRIORITY-BOOST must be a nonnegative real.

     For example:

          > (thread-priority-boost-set! (current-thread) 2.5)
          > (thread-priority-boost (current-thread))
          2.5


 -- procedure: thread-quantum THREAD
 -- procedure: thread-quantum-set! THREAD QUANTUM
     The procedure `thread-quantum' returns a real number which
     corresponds to the quantum of the THREAD.

     The procedure `thread-quantum-set!' changes the quantum of the
     THREAD to QUANTUM and returns an unspecified value.  The QUANTUM
     must be a nonnegative real.  A value of zero selects the smallest
     quantum supported by the implementation.

     For example:

          > (thread-quantum-set! (current-thread) 1.5)
          > (thread-quantum (current-thread))
          1.5
          > (thread-quantum-set! (current-thread) 0)
          > (thread-quantum (current-thread))
          0.


 -- procedure: thread-start! THREAD
     This procedure makes THREAD runnable and returns the THREAD.  The
     THREAD must be an initialized thread.

     For example:

          > (let ((t (thread-start! (make-thread (lambda () (write 'a))))))
              (write 'b)
              (thread-join! t))
          ab> or ba>

     NOTE: It is useful to separate thread creation and thread
     activation to avoid the race condition that would occur if the
     created thread tries to examine a table in which the current
     thread stores the created thread.  See the last example of the
     `thread-terminate!' procedure which contains mutually recursive
     threads.


 -- procedure: thread-yield!
     This procedure causes the current thread to exit the running state
     as if its quantum had expired and returns an unspecified value.

     For example:

          ; a busy loop that avoids being too wasteful of the CPU

          (let loop ()
            (if (mutex-lock! m 0) ; try to lock m but don't block
                (begin
                  (display "locked mutex m")
                  (mutex-unlock! m))
                (begin
                  (do-something-else)
                  (thread-yield!) ; relinquish rest of quantum
                  (loop))))


 -- procedure: thread-sleep! TIMEOUT
     This procedure causes the current thread to wait until the timeout
     is reached and returns an unspecified value.  This blocks the
     thread only if TIMEOUT represents a point in the future.  It is an
     error for TIMEOUT to be `#f'.

     For example:

          ; a clock with a gradual drift:

          (let loop ((x 1))
            (thread-sleep! 1)
            (write x)
            (loop (+ x 1)))

          ; a clock with no drift:

          (let ((start (time->seconds (current-time)))
            (let loop ((x 1))
              (thread-sleep! (seconds->time (+ x start)))
              (write x)
              (loop (+ x 1))))


 -- procedure: thread-terminate! THREAD
     This procedure causes an abnormal termination of the THREAD.  If
     the THREAD is not already terminated, all mutexes owned by the
     THREAD become unlocked/abandoned and a terminated-thread-exception
     object is stored in the THREAD's end-exception field.  If THREAD
     is the current thread, `thread-terminate!' does not return.
     Otherwise `thread-terminate!' returns an unspecified value; the
     termination of the THREAD will occur at some point between the
     calling of `thread-terminate!'  and a finite time in the future
     (an explicit thread synchronization is needed to detect
     termination, see `thread-join!').

     For example:

          (define (amb thunk1 thunk2)
            (let ((result #f)
                  (result-mutex (make-mutex))
                  (done-mutex (make-mutex)))
              (letrec ((child1
                        (make-thread
                          (lambda ()
                            (let ((x (thunk1)))
                              (mutex-lock! result-mutex #f #f)
                              (set! result x)
                              (thread-terminate! child2)
                              (mutex-unlock! done-mutex)))))
                       (child2
                        (make-thread
                          (lambda ()
                            (let ((x (thunk2)))
                              (mutex-lock! result-mutex #f #f)
                              (set! result x)
                              (thread-terminate! child1)
                              (mutex-unlock! done-mutex))))))
                (mutex-lock! done-mutex #f #f)
                (thread-start! child1)
                (thread-start! child2)
                (mutex-lock! done-mutex #f #f)
                result)))

     NOTE: This operation must be used carefully because it terminates a
     thread abruptly and it is impossible for that thread to perform any
     kind of cleanup.  This may be a problem if the thread is in the
     middle of a critical section where some structure has been put in
     an inconsistent state.  However, another thread attempting to
     enter this critical section will raise an
     abandoned-mutex-exception object because the mutex is
     unlocked/abandoned.  This helps avoid observing an inconsistent
     state.  Clean termination can be obtained by polling, as shown in
     the example below.

     For example:

          (define (spawn thunk)
            (let ((t (make-thread thunk)))
              (thread-specific-set! t #t)
              (thread-start! t)
              t))

          (define (stop! thread)
            (thread-specific-set! thread #f)
            (thread-join! thread))

          (define (keep-going?)
            (thread-specific (current-thread)))

          (define count!
            (let ((m (make-mutex))
                  (i 0))
              (lambda ()
                (mutex-lock! m)
                (let ((x (+ i 1)))
                  (set! i x)
                  (mutex-unlock! m)
                  x))))

          (define (increment-forever!)
            (let loop () (count!) (if (keep-going?) (loop))))

          (let ((t1 (spawn increment-forever!))
                (t2 (spawn increment-forever!)))
            (thread-sleep! 1)
            (stop! t1)
            (stop! t2)
            (count!))  ==>  377290


 -- procedure: thread-join! thread [TIMEOUT [TIMEOUT-VAL]]
     This procedure causes the current thread to wait until the THREAD
     terminates (normally or not) or until the timeout is reached if
     TIMEOUT is supplied.  If the timeout is reached, THREAD-JOIN!
     returns TIMEOUT-VAL if it is supplied, otherwise a
     join-timeout-exception object is raised.  If the THREAD terminated
     normally, the content of the end-result field is returned,
     otherwise the content of the end-exception field is raised.

     For example:

          (let ((t (thread-start! (make-thread (lambda () (expt 2 100))))))
            (do-something-else)
            (thread-join! t))  ==>  1267650600228229401496703205376

          (let ((t (thread-start! (make-thread (lambda () (raise 123))))))
            (do-something-else)
            (with-exception-handler
              (lambda (exc)
                (if (uncaught-exception? exc)
                    (* 10 (uncaught-exception-reason exc))
                    99999))
              (lambda ()
                (+ 1 (thread-join! t)))))  ==>  1231

          (define thread-alive?
            (let ((unique (list 'unique)))
              (lambda (thread)
                ; Note: this procedure raises an exception if
                ; the thread terminated abnormally.
                (eq? (thread-join! thread 0 unique) unique))))

          (define (wait-for-termination! thread)
            (let ((eh (current-exception-handler)))
              (with-exception-handler
                (lambda (exc)
                  (if (not (or (terminated-thread-exception? exc)
                               (uncaught-exception? exc)))
                      (eh exc))) ; unexpected exceptions are handled by eh
                (lambda ()
                  ; The following call to thread-join! will wait until the
                  ; thread terminates.  If the thread terminated normally
                  ; thread-join! will return normally.  If the thread
                  ; terminated abnormally then one of these two exception
                  ; objects is raised by thread-join!:
                  ;   - terminated-thread-exception object
                  ;   - uncaught-exception object
                  (thread-join! thread)
                  #f)))) ; ignore result of thread-join!


 -- procedure: thread-send THREAD MSG
     Each thread has a mailbox which stores messages delivered to the
     thread in the order delivered.

     The procedure `thread-send' adds the message MSG at the end of the
     mailbox of thread THREAD and returns an unspecified value.

     For example:

          > (thread-send (current-thread) 111)
          > (thread-send (current-thread) 222)
          > (thread-receive)
          111
          > (thread-receive)
          222


 -- procedure: thread-receive [TIMEOUT [DEFAULT]]
 -- procedure: thread-mailbox-next [TIMEOUT [DEFAULT]]
 -- procedure: thread-mailbox-rewind
 -- procedure: thread-mailbox-extract-and-rewind
     To allow a thread to examine the messages in its mailbox without
     removing them from the mailbox, each thread has a "mailbox cursor"
     which normally points to the last message accessed in the mailbox.
     When a mailbox cursor is rewound using the procedure
     `thread-mailbox-rewind' or `thread-mailbox-extract-and-rewind' or
     `thread-receive', the cursor does not point to a message, but the
     next call to `thread-receive' and `thread-mailbox-next' will set
     the cursor to the oldest message in the mailbox.

     The procedure `thread-receive' advances the mailbox cursor of the
     current thread to the next message, removes that message from the
     mailbox, rewinds the mailbox cursor, and returns the message.  When
     TIMEOUT is not specified, the current thread will wait until a
     message is available in the mailbox.  When TIMEOUT is specified
     and DEFAULT is not specified, a mailbox-receive-timeout-exception
     object is raised if the timeout is reached before a message is
     available.  When TIMEOUT is specified and DEFAULT is specified,
     DEFAULT is returned if the timeout is reached before a message is
     available.

     The procedure `thread-mailbox-next' behaves like `thread-receive'
     except that the message remains in the mailbox and the mailbox
     cursor is not rewound.

     The procedures `thread-mailbox-rewind' or
     `thread-mailbox-extract-and-rewind' rewind the mailbox cursor of
     the current thread so that the next call to `thread-mailbox-next'
     and `thread-receive' will access the oldest message in the
     mailbox.  Additionally the procedure
     `thread-mailbox-extract-and-rewind' will remove from the mailbox
     the message most recently accessed by a call to
     `thread-mailbox-next'.  When `thread-mailbox-next' has not been
     called since the last call to `thread-receive' or
     `thread-mailbox-rewind' or `thread-mailbox-extract-and-rewind', a
     call to `thread-mailbox-extract-and-rewind' only resets the mailbox
     cursor (no message is removed).

     For example:

          > (thread-send (current-thread) 111)
          > (thread-receive 1 999)
          111
          > (thread-send (current-thread) 222)
          > (thread-send (current-thread) 333)
          > (thread-mailbox-next 1 999)
          222
          > (thread-mailbox-next 1 999)
          333
          > (thread-mailbox-next 1 999)
          999
          > (thread-mailbox-extract-and-rewind)
          > (thread-receive 1 999)
          222
          > (thread-receive 1 999)
          999


 -- procedure: mailbox-receive-timeout-exception? OBJ
 -- procedure: mailbox-receive-timeout-exception-procedure EXC
 -- procedure: mailbox-receive-timeout-exception-arguments EXC
     Mailbox-receive-timeout-exception objects are raised by the
     procedures `thread-receive' and `thread-mailbox-next' when a
     timeout expires before a message is available and no default value
     is specified.  The parameter EXC must be a
     mailbox-receive-timeout-exception object.

     The procedure `mailbox-receive-timeout-exception?' returns `#t'
     when OBJ is a mailbox-receive-timeout-exception object and `#f'
     otherwise.

     The procedure `mailbox-receive-timeout-exception-procedure'
     returns the procedure that raised EXC.

     The procedure `mailbox-receive-timeout-exception-arguments'
     returns the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (mailbox-receive-timeout-exception? exc)
                  (list (mailbox-receive-timeout-exception-procedure exc)
                        (mailbox-receive-timeout-exception-arguments exc))
                  'not-mailbox-receive-timeout-exception))
          > (with-exception-catcher
              handler
              (lambda () (thread-receive 1)))
          (#<procedure #2 thread-receive> (1))


 -- procedure: mutex? OBJ
     This procedure returns `#t' when OBJ is a mutex object and `#f'
     otherwise.

     For example:

          > (mutex? (make-mutex))
          #t
          > (mutex? 'foo)
          #f


 -- procedure: make-mutex [NAME]
     This procedure returns a new mutex in the unlocked/not-abandoned
     state.  The optional NAME is an arbitrary Scheme object which
     identifies the mutex (useful for debugging); it defaults to an
     unspecified value.  The mutex's specific field is set to an
     unspecified value.

     For example:

          > (make-mutex)
          #<mutex #2>
          > (make-mutex 'foo)
          #<mutex #3 foo>


 -- procedure: mutex-name MUTEX
     Returns the name of the MUTEX.  For example:

          > (mutex-name (make-mutex 'foo))
          foo


 -- procedure: mutex-specific MUTEX
 -- procedure: mutex-specific-set! MUTEX OBJ
     The `mutex-specific' procedure returns the content of the MUTEX's
     specific field.

     The `mutex-specific-set!' procedure stores OBJ into the MUTEX's
     specific field and returns an unspecified value.

     For example:

          > (define m (make-mutex))
          > (mutex-specific-set! m "hello")
          > (mutex-specific m)
          "hello"
          > (define (mutex-lock-recursively! mutex)
              (if (eq? (mutex-state mutex) (current-thread))
                  (let ((n (mutex-specific mutex)))
                    (mutex-specific-set! mutex (+ n 1)))
                  (begin
                    (mutex-lock! mutex)
                    (mutex-specific-set! mutex 0))))
          > (define (mutex-unlock-recursively! mutex)
              (let ((n (mutex-specific mutex)))
                (if (= n 0)
                    (mutex-unlock! mutex)
                    (mutex-specific-set! mutex (- n 1)))))
          > (mutex-lock-recursively! m)
          > (mutex-lock-recursively! m)
          > (mutex-lock-recursively! m)
          > (mutex-specific m)
          2


 -- procedure: mutex-state MUTEX
     Thos procedure returns information about the state of the MUTEX.
     The possible results are:

        * thread T: the MUTEX is in the locked/owned state and thread T
          is the owner of the MUTEX

        * symbol `not-owned': the MUTEX is in the locked/not-owned state

        * symbol `abandoned': the MUTEX is in the unlocked/abandoned
          state

        * symbol `not-abandoned': the MUTEX is in the
          unlocked/not-abandoned state


     For example:

          (mutex-state (make-mutex))  ==>  not-abandoned

          (define (thread-alive? thread)
            (let ((mutex (make-mutex)))
              (mutex-lock! mutex #f thread)
              (let ((state (mutex-state mutex)))
                (mutex-unlock! mutex) ; avoid space leak
                (eq? state thread))))


 -- procedure: mutex-lock! MUTEX [TIMEOUT [THREAD]]
     This procedure locks MUTEX.  If the MUTEX is currently locked, the
     current thread waits until the MUTEX is unlocked, or until the
     timeout is reached if TIMEOUT is supplied.  If the timeout is
     reached, `mutex-lock!' returns `#f'.  Otherwise, the state of the
     MUTEX is changed as follows:

        * if THREAD is `#f' the MUTEX becomes locked/not-owned,

        * otherwise, let T be THREAD (or the current thread if THREAD
          is not supplied),

             * if T is terminated the MUTEX becomes unlocked/abandoned,

             * otherwise MUTEX becomes locked/owned with T as the owner.



     After changing the state of the MUTEX, an
     abandoned-mutex-exception object is raised if the MUTEX was
     unlocked/abandoned before the state change, otherwise
     `mutex-lock!' returns `#t'.  It is not an error if the MUTEX is
     owned by the current thread (but the current thread will have to
     wait).

     For example:

          ; an implementation of a mailbox object of depth one; this
          ; implementation does not behave well in the presence of forced
          ; thread terminations using thread-terminate! (deadlock can occur
          ; if a thread is terminated in the middle of a put! or get! operation)

          (define (make-empty-mailbox)
            (let ((put-mutex (make-mutex)) ; allow put! operation
                  (get-mutex (make-mutex))
                  (cell #f))

              (define (put! obj)
                (mutex-lock! put-mutex #f #f) ; prevent put! operation
                (set! cell obj)
                (mutex-unlock! get-mutex)) ; allow get! operation

              (define (get!)
                (mutex-lock! get-mutex #f #f) ; wait until object in mailbox
                (let ((result cell))
                  (set! cell #f) ; prevent space leaks
                  (mutex-unlock! put-mutex) ; allow put! operation
                  result))

              (mutex-lock! get-mutex #f #f) ; prevent get! operation

              (lambda (msg)
                (case msg
                  ((put!) put!)
                  ((get!) get!)
                  (else (error "unknown message"))))))

          (define (mailbox-put! m obj) ((m 'put!) obj))
          (define (mailbox-get! m) ((m 'get!)))

          ; an alternate implementation of thread-sleep!

          (define (sleep! timeout)
            (let ((m (make-mutex)))
              (mutex-lock! m #f #f)
              (mutex-lock! m timeout #f)))

          ; a procedure that waits for one of two mutexes to unlock

          (define (lock-one-of! mutex1 mutex2)
            ; this procedure assumes that neither mutex1 or mutex2
            ; are owned by the current thread
            (let ((ct (current-thread))
                  (done-mutex (make-mutex)))
              (mutex-lock! done-mutex #f #f)
              (let ((t1 (thread-start!
                         (make-thread
                          (lambda ()
                            (mutex-lock! mutex1 #f ct)
                            (mutex-unlock! done-mutex)))))
                    (t2 (thread-start!
                         (make-thread
                          (lambda ()
                            (mutex-lock! mutex2 #f ct)
                            (mutex-unlock! done-mutex))))))
                (mutex-lock! done-mutex #f #f)
                (thread-terminate! t1)
                (thread-terminate! t2)
                (if (eq? (mutex-state mutex1) ct)
                    (begin
                      (if (eq? (mutex-state mutex2) ct)
                          (mutex-unlock! mutex2)) ; don't lock both
                      mutex1)
                    mutex2))))


 -- procedure: mutex-unlock! MUTEX [CONDITION-VARIABLE [TIMEOUT]]
     This procedure unlocks the MUTEX by making it
     unlocked/not-abandoned.  It is not an error to unlock an unlocked
     mutex and a mutex that is owned by any thread.  If
     CONDITION-VARIABLE is supplied, the current thread is blocked and
     added to the CONDITION-VARIABLE before unlocking MUTEX; the thread
     can unblock at any time but no later than when an appropriate call
     to `condition-variable-signal!' or `condition-variable-broadcast!'
     is performed (see below), and no later than the timeout (if
     TIMEOUT is supplied).  If there are threads waiting to lock this
     MUTEX, the scheduler selects a thread, the mutex becomes
     locked/owned or locked/not-owned, and the thread is unblocked.
     `mutex-unlock!' returns `#f' when the timeout is reached,
     otherwise it returns `#t'.

     NOTE: The reason the thread can unblock at any time (when
     CONDITION-VARIABLE is supplied) is that the scheduler, when it
     detects a serious problem such as a deadlock, must interrupt one of
     the blocked threads (such as the primordial thread) so that it can
     perform some appropriate action.  After a thread blocked on a
     condition-variable has handled such an interrupt it would be wrong
     for the scheduler to return the thread to the blocked state,
     because any calls to `condition-variable-broadcast!' during the
     interrupt will have gone unnoticed.  It is necessary for the
     thread to remain runnable and return from the call to
     `mutex-unlock!' with a result of `#t'.

     NOTE: `mutex-unlock!' is related to the "wait" operation on
     condition variables available in other thread systems.  The main
     difference is that "wait" automatically locks MUTEX just after the
     thread is unblocked.  This operation is not performed by
     `mutex-unlock!' and so must be done by an explicit call to
     `mutex-lock!'.  This has the advantages that a different timeout
     and exception-handler can be specified on the `mutex-lock!' and
     `mutex-unlock!' and the location of all the mutex operations is
     clearly apparent.

     For example:

          (let loop ()
            (mutex-lock! m)
            (if (condition-is-true?)
                (begin
                  (do-something-when-condition-is-true)
                  (mutex-unlock! m))
                (begin
                  (mutex-unlock! m cv)
                  (loop))))


 -- procedure: condition-variable? OBJ
     This procedure returns `#t' when OBJ is a condition-variable
     object and `#f' otherwise.

     For example:

          > (condition-variable? (make-condition-variable))
          #t
          > (condition-variable? 'foo)
          #f


 -- procedure: make-condition-variable [NAME]
     This procedure returns a new empty condition variable.  The
     optional NAME is an arbitrary Scheme object which identifies the
     condition variable (useful for debugging); it defaults to an
     unspecified value.  The condition variable's specific field is set
     to an unspecified value.

     For example:

          > (make-condition-variable)
          #<condition-variable #2>


 -- procedure: condition-variable-name CONDITION-VARIABLE
     This procedure returns the name of the CONDITION-VARIABLE.  For
     example:

          > (condition-variable-name (make-condition-variable 'foo))
          foo


 -- procedure: condition-variable-specific CONDITION-VARIABLE
 -- procedure: condition-variable-specific-set! CONDITION-VARIABLE OBJ
     The `condition-variable-specific' procedure returns the content of
     the CONDITION-VARIABLE's specific field.

     The `condition-variable-specific-set!' procedure stores OBJ into
     the CONDITION-VARIABLE's specific field and returns an unspecified
     value.

     For example:

          > (define cv (make-condition-variable))
          > (condition-variable-specific-set! cv "hello")
          > (condition-variable-specific cv)
          "hello"


 -- procedure: condition-variable-signal! CONDITION-VARIABLE
     This procedure unblocks a thread blocked on the CONDITION-VARIABLE
     (if there is at least one) and returns an unspecified value.

     For example:

          ; an implementation of a mailbox object of depth one; this
          ; implementation behaves gracefully when threads are forcibly
          ; terminated using thread-terminate! (an abandoned-mutex-exception
          ; object will be raised when a put! or get! operation is attempted
          ; after a thread is terminated in the middle of a put! or get!
          ; operation)

          (define (make-empty-mailbox)
            (let ((mutex (make-mutex))
                  (put-condvar (make-condition-variable))
                  (get-condvar (make-condition-variable))
                  (full? #f)
                  (cell #f))

              (define (put! obj)
                (mutex-lock! mutex)
                (if full?
                    (begin
                      (mutex-unlock! mutex put-condvar)
                      (put! obj))
                    (begin
                      (set! cell obj)
                      (set! full? #t)
                      (condition-variable-signal! get-condvar)
                      (mutex-unlock! mutex))))

              (define (get!)
                (mutex-lock! mutex)
                (if (not full?)
                    (begin
                      (mutex-unlock! mutex get-condvar)
                      (get!))
                    (let ((result cell))
                      (set! cell #f) ; avoid space leaks
                      (set! full? #f)
                      (condition-variable-signal! put-condvar)
                      (mutex-unlock! mutex))))

              (lambda (msg)
                (case msg
                  ((put!) put!)
                  ((get!) get!)
                  (else (error "unknown message"))))))

          (define (mailbox-put! m obj) ((m 'put!) obj))
          (define (mailbox-get! m) ((m 'get!)))


 -- procedure: condition-variable-broadcast! CONDITION-VARIABLE
     This procedure unblocks all the thread blocked on the
     CONDITION-VARIABLE and returns an unspecified value.

     For example:

          (define (make-semaphore n)
            (vector n (make-mutex) (make-condition-variable)))

          (define (semaphore-wait! sema)
            (mutex-lock! (vector-ref sema 1))
            (let ((n (vector-ref sema 0)))
              (if (> n 0)
                  (begin
                    (vector-set! sema 0 (- n 1))
                    (mutex-unlock! (vector-ref sema 1)))
                  (begin
                    (mutex-unlock! (vector-ref sema 1) (vector-ref sema 2))
                    (semaphore-wait! sema))))

          (define (semaphore-signal-by! sema increment)
            (mutex-lock! (vector-ref sema 1))
            (let ((n (+ (vector-ref sema 0) increment)))
              (vector-set! sema 0 n)
              (if (> n 0)
                  (condition-variable-broadcast! (vector-ref sema 2)))
              (mutex-unlock! (vector-ref sema 1))))


14 Dynamic environment
**********************

The "dynamic environment" is the structure which allows the system to
find the value returned by the standard procedures `current-input-port'
and `current-output-port'.  The standard procedures
`with-input-from-file' and `with-output-to-file' extend the dynamic
environment to produce a new dynamic environment which is in effect for
the dynamic extent of the call to the thunk passed as their last
argument.  These procedures are essentially special purpose dynamic
binding operations on hidden dynamic variables (one for
`current-input-port' and one for `current-output-port').  Gambit
generalizes this dynamic binding mechanism to allow the user to
introduce new dynamic variables, called "parameter objects", and
dynamically bind them.  The parameter objects implemented by Gambit are
compatible with the specification of the "Parameter objects SRFI" (SRFI
39).

   One important issue is the relationship between the dynamic
environments of the parent and child threads when a thread is created.
Each thread has its own dynamic environment that is accessed when
looking up the value bound to a parameter object by that thread.  When
a thread's dynamic environment is extended it does not affect the
dynamic environment of other threads.  When a thread is created it is
given a dynamic environment whose bindings are inherited from the
parent thread.  In this inherited dynamic environment the parameter
objects are bound to the same cells as the parent's dynamic environment
(in other words an assignment of a new value to a parameter object is
visible in the other thread).

   Another important issue is the interaction between the
`dynamic-wind' procedure and dynamic environments.  When a thread
creates a continuation, the thread's dynamic environment and the
`dynamic-wind' stack are saved within the continuation (an alternate
but equivalent point of view is that the `dynamic-wind' stack is part
of the dynamic environment).  When this continuation is invoked the
required `dynamic-wind' before and after thunks are called and the
saved dynamic environment is reinstated as the dynamic environment of
the current thread.  During the call to each required `dynamic-wind'
before and after thunk, the dynamic environment and the `dynamic-wind'
stack in effect when the corresponding `dynamic-wind' was executed are
reinstated.  Note that this specification precisely defines the
semantics of calling `call-with-current-continuation' or invoking a
continuation within a before or after thunk.  The semantics are well
defined even when a continuation created by another thread is invoked.
Below is an example exercising the subtleties of this semantics.

     (with-output-to-file
      "foo"
      (lambda ()
        (let ((k (call-with-current-continuation
                  (lambda (exit)
                    (with-output-to-file
                     "bar"
                     (lambda ()
                       (dynamic-wind
                        (lambda ()
                          (write '(b1))
                          (force-output))
                        (lambda ()
                          (let ((x (call-with-current-continuation
                                    (lambda (cont) (exit cont)))))
                            (write '(t1))
                            (force-output)
                            x))
                        (lambda ()
                          (write '(a1))
                          (force-output)))))))))
          (if k
              (dynamic-wind
               (lambda ()
                 (write '(b2))
                 (force-output))
               (lambda ()
                 (with-output-to-file
                  "baz"
                  (lambda ()
                    (write '(t2))
                    (force-output)
                    ; go back inside (with-output-to-file "bar" ...)
                    (k #f))))
               (lambda ()
                 (write '(a2))
                 (force-output)))))))

   The following actions will occur when this code is executed:
`(b1)(a1)' is written to "bar", `(b2)' is then written to "foo", `(t2)'
is then written to "baz", `(a2)' is then written to "foo", and finally
`(b1)(t1)(a1)' is written to "bar".

 -- procedure: make-parameter OBJ [FILTER]
     The dynamic environment is composed of two parts: the "local
     dynamic environment" and the "global dynamic environment".  There
     is a single global dynamic environment, and it is used to lookup
     parameter objects that can't be found in the local dynamic
     environment.

     The `make-parameter' procedure returns a new "parameter object".
     The FILTER argument is a one argument conversion procedure.  If it
     is not specified, FILTER defaults to the identity function.

     The global dynamic environment is updated to associate the
     parameter object to a new cell.  The initial content of the cell
     is the result of applying the conversion procedure to OBJ.

     A parameter object is a procedure which accepts zero or one
     argument.  The cell bound to a particular parameter object in the
     dynamic environment is accessed by calling the parameter object.
     When no argument is passed, the content of the cell is returned.
     When one argument is passed the content of the cell is updated
     with the result of applying the parameter object's conversion
     procedure to the argument.  Note that the conversion procedure can
     be used for guaranteeing the type of the parameter object's
     binding and/or to perform some conversion of the value.

     For example:

          > (define radix (make-parameter 10))
          > (radix)
          10
          > (radix 2)
          > (radix)
          2
          > (define prompt
              (make-parameter
                123
                (lambda (x)
                  (if (string? x)
                      x
                      (object->string x)))))
          > (prompt)
          "123"
          > (prompt "$")
          > (prompt)
          "$"
          > (define write-shared
              (make-parameter
                #f
                (lambda (x)
                  (if (boolean? x)
                      x
                      (error "only booleans are accepted by write-shared")))))
          > (write-shared 123)
          *** ERROR IN ##make-parameter -- only booleans are accepted by write-shared


 -- special form: parameterize ((procedure value)...) body
     The `parameterize' form, evaluates all procedure and value
     expressions in an unspecified order.  All the procedure
     expressions must evaluate to procedures, either parameter objects
     or procedures accepting zero and one argument.  Then, for each
     procedure p and in an unspecified order:

        * If p is a parameter object a new cell is created,
          initialized, and bound to the parameter object in the local
          dynamic environment.  The value contained in the cell is the
          result of applying the parameter object's conversion
          procedure to value.  The resulting dynamic environment is
          then used for processing the remaining bindings (or the
          evaluation of body if there are no other bindings).

        * Otherwise p will be used according to the following protocol:
          we say that the call `(p)' "gets p's value" and that the call
          `(p x)' "sets p's value to x".  First, the `parameterize'
          form gets p's value and saves it in a local variable.  It
          then sets p's value to value.  It then processes the
          remaining bindings (or evaluates body if there are no other
          bindings).  Then it sets p's value to the saved value.  These
          steps are performed in a `dynamic-wind' so that it is
          possible to use continuations to jump into and out of the body
          (i.e. the `dynamic-wind''s before thunk sets p's value to
          value and the after thunk sets p's value to the saved value).


     The result(s) of the `parameterize' form are the result(s) of the
     body.

     Note that using procedures instead of parameter objects may lead to
     unexpected results in multithreaded programs because the before and
     after thunks of the `dynamic-wind' are not called when control
     switches between threads.

     For example:

          > (define radix (make-parameter 2))
          > (define prompt
              (make-parameter
                123
                (lambda (x)
                  (if (string? x)
                      x
                      (object->string x)))))
          > (radix)
          2
          > (parameterize ((radix 16)) (radix))
          16
          > (radix)
          2
          > (define (f n) (number->string n (radix)))
          > (f 10)
          "1010"
          > (parameterize ((radix 8)) (f 10))
          "12"
          > (parameterize ((radix 8) (prompt (f 10))) (prompt))
          "1010"
          > (define p
              (let ((x 1))
                (lambda args
                  (if (null? args) x (set! x (car args))))))
          > (let* ((a (p))
                   (b (parameterize ((p 2)) (list (p))))
                   (c (p)))
              (list a b c))
          (1 (2) 1)


15 Exceptions
*************

15.1 Exception-handling
=======================

Gambit's exception-handling model is inspired from the withdrawn
"Exception Handling SRFI" (SRFI 12), the "Multithreading support SRFI"
(SRFI 18), and the "Exception Handling for Programs SRFI" (SRFI 34).
The two fundamental operations are the dynamic binding of an exception
handler (i.e. the procedure `with-exception-handler') and the
invocation of the exception handler (i.e. the procedure `raise').

   All predefined procedures which check for errors (including type
errors, memory allocation errors, host operating-system errors, etc)
report these errors using the exception-handling system (i.e. they
"raise" an exception that can be handled in a user-defined exception
handler).  When an exception is raised and the exception is not handled
by a user-defined exception handler, the predefined exception handler
will display an error message (if the primordial thread raised the
exception) or the thread will silently terminate with no error message
(if it is not the primordial thread that raised the exception).  This
default behavior can be changed through the `-:d' runtime option (*note
Runtime options::).

   Predefined procedures normally raise exceptions by performing a
tail-call to the exception handler (the exceptions are "complex"
procedures such as `eval', `compile-file', `read', `write', etc).  This
means that the continuation of the exception handler and of the REPL
that may be started due to this is normally the continuation of the
predefined procedure that raised the exception.  By exiting the REPL
with the `,(c EXPRESSION)' command it is thus possible to resume the
program as though the call to the predefined procedure returned the
value of EXPRESSION.  For example:

     > (define (f x) (+ (car x) 1))
     > (f 2) ; typo... we meant to say (f '(2))
     *** ERROR IN f, (console)@1.18 -- (Argument 1) PAIR expected
     (car 2)
     1> ,(c 2)
     3

 -- procedure: current-exception-handler [NEW-EXCEPTION-HANDLER]
     The parameter object `current-exception-handler' is bound to the
     current exception-handler.  Calling this procedure with no argument
     returns the current exception-handler and calling this procedure
     with one argument sets the current exception-handler to
     NEW-EXCEPTION-HANDLER.

     For example:

          > (current-exception-handler)
          #<procedure #2 primordial-exception-handler>
          > (current-exception-handler (lambda (exc) (pp exc) 999))
          > (/ 1 0)
          #<divide-by-zero-exception #3>
          999


 -- procedure: with-exception-handler HANDLER THUNK
     Returns the result(s) of calling THUNK with no arguments.  The
     HANDLER, which must be a procedure, is installed as the current
     exception-handler in the dynamic environment in effect during the
     call to THUNK.  Note that the dynamic environment in effect during
     the call to HANDLER has HANDLER as the exception-handler.
     Consequently, an exception raised during the call to HANDLER may
     lead to an infinite loop.

     For example:

          > (with-exception-handler
              (lambda (e) (write e) 5)
              (lambda () (+ 1 (* 2 3) 4)))
          11
          > (with-exception-handler
              (lambda (e) (write e) 5)
              (lambda () (+ 1 (* 'foo 3) 4)))
          #<type-exception #2>10
          > (with-exception-handler
              (lambda (e) (write e 9))
              (lambda () (+ 1 (* 'foo 3) 4)))
          INFINITE LOOP


 -- procedure: with-exception-catcher HANDLER THUNK
     Returns the result(s) of calling THUNK with no arguments.  A new
     exception-handler is installed as the current exception-handler in
     the dynamic environment in effect during the call to THUNK.  This
     new exception-handler will call the HANDLER, which must be a
     procedure, with the exception object as an argument and with the
     same continuation as the call to `with-exception-catcher'.  This
     implies that the dynamic environment in effect during the call to
     HANDLER is the same as the one in effect at the call to
     `with-exception-catcher'.  Consequently, an exception raised
     during the call to HANDLER will not lead to an infinite loop.

     For example:

          > (with-exception-catcher
              (lambda (e) (write e) 5)
              (lambda () (+ 1 (* 2 3) 4)))
          11
          > (with-exception-catcher
              (lambda (e) (write e) 5)
              (lambda () (+ 1 (* 'foo 3) 4)))
          #<type-exception #2>5
          > (with-exception-catcher
              (lambda (e) (write e 9))
              (lambda () (+ 1 (* 'foo 3) 4)))
          *** ERROR IN (console)@7.1 -- (Argument 2) OUTPUT PORT expected
          (write '#<type-exception #3> 9)


 -- procedure: raise OBJ
     This procedure tail-calls the current exception-handler with OBJ
     as the sole argument.  If the exception-handler returns, the
     continuation of the call to `raise' is invoked.

     For example:

          > (with-exception-handler
              (lambda (exc)
                (pp exc)
                100)
              (lambda ()
                (+ 1 (raise "hello"))))
          "hello"
          101


 -- procedure: abort OBJ
 -- procedure: noncontinuable-exception? OBJ
 -- procedure: noncontinuable-exception-reason EXC
     The procedure `abort' calls the current exception-handler with OBJ
     as the sole argument.  If the exception-handler returns, the
     procedure `abort' will be tail-called with a
     noncontinuable-exception object, whose reason field is OBJ, as
     sole argument.

     Noncontinuable-exception objects are raised by the `abort'
     procedure when the exception-handler returns.  The parameter EXC
     must be a noncontinuable-exception object.

     The procedure `noncontinuable-exception?' returns `#t' when OBJ is
     a noncontinuable-exception object and `#f' otherwise.

     The procedure `noncontinuable-exception-reason' returns the
     argument of the call to `abort' that raised EXC.

     For example:

          > (call-with-current-continuation
              (lambda (k)
                (with-exception-handler
                  (lambda (exc)
                    (pp exc)
                    (if (noncontinuable-exception? exc)
                        (k (list (noncontinuable-exception-reason exc)))
                        100))
                  (lambda ()
                    (+ 1 (abort "hello"))))))
          "hello"
          #<noncontinuable-exception #2>
          ("hello")


15.2 Exception objects related to memory management
===================================================

 -- procedure: heap-overflow-exception? OBJ
     Heap-overflow-exception objects are raised when the allocation of
     an object would cause the heap to use more memory space than is
     available.

     The procedure `heap-overflow-exception?' returns `#t' when OBJ is
     a heap-overflow-exception object and `#f' otherwise.

     For example:

          > (define (handler exc)
              (if (heap-overflow-exception? exc)
                  exc
                  'not-heap-overflow-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (define (f x) (f (cons 1 x)))
                (f '())))
          #<heap-overflow-exception #2>


 -- procedure: stack-overflow-exception? OBJ
     Stack-overflow-exception objects are raised when the allocation of
     a continuation frame would cause the heap to use more memory space
     than is available.

     The procedure `stack-overflow-exception?' returns `#t' when OBJ is
     a stack-overflow-exception object and `#f' otherwise.

     For example:

          > (define (handler exc)
              (if (stack-overflow-exception? exc)
                  exc
                  'not-stack-overflow-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (define (f) (+ 1 (f)))
                (f)))
          #<stack-overflow-exception #2>


15.3 Exception objects related to the host environment
======================================================

 -- procedure: os-exception? OBJ
 -- procedure: os-exception-procedure EXC
 -- procedure: os-exception-arguments EXC
 -- procedure: os-exception-code EXC
 -- procedure: os-exception-message EXC
     Os-exception objects are raised by procedures which access the host
     operating-system's services when the requested operation fails.
     The parameter EXC must be a os-exception object.

     The procedure `os-exception?' returns `#t' when OBJ is a
     os-exception object and `#f' otherwise.

     The procedure `os-exception-procedure' returns the procedure that
     raised EXC.

     The procedure `os-exception-arguments' returns the list of
     arguments of the procedure that raised EXC.

     The procedure `os-exception-code' returns an exact integer error
     code that can be converted to a string by the `err-code->string'
     procedure.  Note that the error code is operating-system dependent.

     The procedure `os-exception-message' returns `#f' or a string
     giving details of the exception in a human-readable form.

     For example:

          > (define (handler exc)
              (if (os-exception? exc)
                  (list (os-exception-procedure exc)
                        (os-exception-arguments exc)
                        (err-code->string (os-exception-code exc))
                        (os-exception-message exc))
                  'not-os-exception))
          > (with-exception-catcher
              handler
              (lambda () (host-info "x.y.z")))
          (#<procedure #2 host-info> ("x.y.z") "Unknown host" #f)


 -- procedure: no-such-file-or-directory-exception? OBJ
 -- procedure: no-such-file-or-directory-exception-procedure EXC
 -- procedure: no-such-file-or-directory-exception-arguments EXC
     No-such-file-or-directory-exception objects are raised by
     procedures which access the filesystem (such as `open-input-file'
     and `directory-files') when the path specified can't be found on
     the filesystem.  The parameter EXC must be a
     no-such-file-or-directory-exception object.

     The procedure `no-such-file-or-directory-exception?' returns `#t'
     when OBJ is a no-such-file-or-directory-exception object and `#f'
     otherwise.

     The procedure `no-such-file-or-directory-exception-procedure'
     returns the procedure that raised EXC.

     The procedure `no-such-file-or-directory-exception-arguments'
     returns the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (no-such-file-or-directory-exception? exc)
                  (list (no-such-file-or-directory-exception-procedure exc)
                        (no-such-file-or-directory-exception-arguments exc))
                  'not-no-such-file-or-directory-exception))
          > (with-exception-catcher
              handler
              (lambda () (with-input-from-file "nofile" read)))
          (#<procedure #2 with-input-from-file> ("nofile" #<procedure #3 read>))


 -- procedure: unbound-os-environment-variable-exception? OBJ
 -- procedure: unbound-os-environment-variable-exception-procedure EXC
 -- procedure: unbound-os-environment-variable-exception-arguments EXC
     Unbound-os-environment-variable-exception objects are raised when
     an unbound operating-system environment variable is accessed by the
     procedures `getenv' and `setenv'.  The parameter EXC must be an
     unbound-os-environment-variable-exception object.

     The procedure `unbound-os-environment-variable-exception?' returns
     `#t' when OBJ is an unbound-os-environment-variable-exception
     object and `#f' otherwise.

     The procedure `unbound-os-environment-variable-exception-procedure'
     returns the procedure that raised EXC.

     The procedure `unbound-os-environment-variable-exception-arguments'
     returns the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (unbound-os-environment-variable-exception? exc)
                  (list (unbound-os-environment-variable-exception-procedure exc)
                        (unbound-os-environment-variable-exception-arguments exc))
                  'not-unbound-os-environment-variable-exception))
          > (with-exception-catcher
              handler
              (lambda () (getenv "DOES_NOT_EXIST")))
          (#<procedure #2 getenv> ("DOES_NOT_EXIST"))


15.4 Exception objects related to threads
=========================================

 -- procedure: scheduler-exception? OBJ
 -- procedure: scheduler-exception-reason EXC
     Scheduler-exception objects are raised by the scheduler when some
     operation requested from the host operating system failed (e.g.
     checking the status of the devices in order to wake up threads
     waiting to perform I/O on these devices).  The parameter EXC must
     be a scheduler-exception object.

     The procedure `scheduler-exception?' returns `#t' when OBJ is a
     scheduler-exception object and `#f' otherwise.

     The procedure `scheduler-exception-reason' returns the
     os-exception object that describes the failure detected by the
     scheduler.


 -- procedure: deadlock-exception? OBJ
     Deadlock-exception objects are raised when the scheduler discovers
     that all threads are blocked and can make no further progress.  In
     that case the scheduler unblocks the primordial-thread and forces
     it to raise a deadlock-exception object.

     The procedure `deadlock-exception?' returns `#t' when OBJ is a
     deadlock-exception object and `#f' otherwise.

     For example:

          > (define (handler exc)
              (if (deadlock-exception? exc)
                  exc
                  'not-deadlock-exception))
          > (with-exception-catcher
              handler
              (lambda () (read (open-vector))))
          #<deadlock-exception #2>


 -- procedure: abandoned-mutex-exception? OBJ
     Abandoned-mutex-exception objects are raised when the current
     thread locks a mutex that was owned by a thread which terminated
     (see `mutex-lock!').

     The procedure `abandoned-mutex-exception?' returns `#t' when OBJ
     is a abandoned-mutex-exception object and `#f' otherwise.

     For example:

          > (define (handler exc)
              (if (abandoned-mutex-exception? exc)
                  exc
                  'not-abandoned-mutex-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (let ((m (make-mutex)))
                  (thread-join!
                    (thread-start!
                      (make-thread
                        (lambda () (mutex-lock! m)))))
                  (mutex-lock! m))))
          #<abandoned-mutex-exception #2>


 -- procedure: join-timeout-exception? OBJ
 -- procedure: join-timeout-exception-procedure EXC
 -- procedure: join-timeout-exception-arguments EXC
     Join-timeout-exception objects are raised when a call to the
     `thread-join!' procedure reaches its timeout before the target
     thread terminates and a timeout-value parameter is not specified.
     The parameter EXC must be a join-timeout-exception object.

     The procedure `join-timeout-exception?' returns `#t' when OBJ is a
     join-timeout-exception object and `#f' otherwise.

     The procedure `join-timeout-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `join-timeout-exception-arguments' returns the list
     of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (join-timeout-exception? exc)
                  (list (join-timeout-exception-procedure exc)
                        (join-timeout-exception-arguments exc))
                  'not-join-timeout-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (thread-join!
                  (thread-start!
                    (make-thread
                      (lambda () (thread-sleep! 10))))
                  5)))
          (#<procedure #2 thread-join!> (#<thread #3> 5))


 -- procedure: started-thread-exception? OBJ
 -- procedure: started-thread-exception-procedure EXC
 -- procedure: started-thread-exception-arguments EXC
     Started-thread-exception objects are raised when the target thread
     of a call to the procedure `thread-start!' is already started.  The
     parameter EXC must be a started-thread-exception object.

     The procedure `started-thread-exception?' returns `#t' when OBJ is
     a started-thread-exception object and `#f' otherwise.

     The procedure `started-thread-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `started-thread-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (started-thread-exception? exc)
                  (list (started-thread-exception-procedure exc)
                        (started-thread-exception-arguments exc))
                  'not-started-thread-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (let ((t (make-thread (lambda () (expt 2 1000)))))
                  (thread-start! t)
                  (thread-start! t))))
          (#<procedure #2 thread-start!> (#<thread #3>))


 -- procedure: terminated-thread-exception? OBJ
 -- procedure: terminated-thread-exception-procedure EXC
 -- procedure: terminated-thread-exception-arguments EXC
     Terminated-thread-exception objects are raised when the
     `thread-join!' procedure is called and the target thread has
     terminated as a result of a call to the `thread-terminate!'
     procedure.  The parameter EXC must be a
     terminated-thread-exception object.

     The procedure `terminated-thread-exception?' returns `#t' when OBJ
     is a terminated-thread-exception object and `#f' otherwise.

     The procedure `terminated-thread-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `terminated-thread-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (terminated-thread-exception? exc)
                  (list (terminated-thread-exception-procedure exc)
                        (terminated-thread-exception-arguments exc))
                  'not-terminated-thread-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (thread-join!
                  (thread-start!
                    (make-thread
                      (lambda () (thread-terminate! (current-thread))))))))
          (#<procedure #2 thread-join!> (#<thread #3>))


 -- procedure: uncaught-exception? OBJ
 -- procedure: uncaught-exception-procedure EXC
 -- procedure: uncaught-exception-arguments EXC
 -- procedure: uncaught-exception-reason EXC
     Uncaught-exception objects are raised when an object is raised in a
     thread and that thread does not handle it (i.e. the thread
     terminated because it did not catch an exception it raised).  The
     parameter EXC must be an uncaught-exception object.

     The procedure `uncaught-exception?' returns `#t' when OBJ is an
     uncaught-exception object and `#f' otherwise.

     The procedure `uncaught-exception-procedure' returns the procedure
     that raised EXC.

     The procedure `uncaught-exception-arguments' returns the list of
     arguments of the procedure that raised EXC.

     The procedure `uncaught-exception-reason' returns the object that
     was raised by the thread and not handled by that thread.

     For example:

          > (define (handler exc)
              (if (uncaught-exception? exc)
                  (list (uncaught-exception-procedure exc)
                        (uncaught-exception-arguments exc)
                        (uncaught-exception-reason exc))
                  'not-uncaught-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (thread-join!
                  (thread-start!
                    (make-thread
                      (lambda () (open-input-file "data" 99)))))))
          (#<procedure #2 thread-join!>
           (#<thread #3>)
           #<wrong-number-of-arguments-exception #4>)


15.5 Exception objects related to C-interface
=============================================

 -- procedure: cfun-conversion-exception? OBJ
 -- procedure: cfun-conversion-exception-procedure EXC
 -- procedure: cfun-conversion-exception-arguments EXC
 -- procedure: cfun-conversion-exception-code EXC
 -- procedure: cfun-conversion-exception-message EXC
     Cfun-conversion-exception objects are raised by the C-interface
     when converting between the Scheme representation and the C
     representation of a value during a call from Scheme to C.  The
     parameter EXC must be a cfun-conversion-exception object.

     The procedure `cfun-conversion-exception?' returns `#t' when OBJ
     is a cfun-conversion-exception object and `#f' otherwise.

     The procedure `cfun-conversion-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `cfun-conversion-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     The procedure `cfun-conversion-exception-code' returns an exact
     integer error code that can be converted to a string by the
     `err-code->string' procedure.

     The procedure `cfun-conversion-exception-message' returns `#f' or
     a string giving details of the exception in a human-readable form.

     For example:

          $ cat test1.scm
          (define weird
            (c-lambda (char-string) nonnull-char-string
              "___result = ___arg1;"))
          $ gsc test1.scm
          $ gsi
          Gambit Version 4.0 beta 20

          > (load "test1")
          "/Users/feeley/gambit/doc/test1.o1"
          > (weird "hello")
          "hello"
          > (define (handler exc)
              (if (cfun-conversion-exception? exc)
                  (list (cfun-conversion-exception-procedure exc)
                        (cfun-conversion-exception-arguments exc)
                        (err-code->string (cfun-conversion-exception-code exc))
                        (cfun-conversion-exception-message exc))
                  'not-cfun-conversion-exception))
          > (with-exception-catcher
              handler
              (lambda () (weird 'not-a-string)))
          (#<procedure #2 weird>
           (not-a-string)
           "(Argument 1) Can't convert to C char-string"
           #f)
          > (with-exception-catcher
              handler
              (lambda () (weird #f)))
          (#<procedure #2 weird>
           (#f)
           "Can't convert result from C nonnull-char-string"
           #f)


 -- procedure: sfun-conversion-exception? OBJ
 -- procedure: sfun-conversion-exception-procedure EXC
 -- procedure: sfun-conversion-exception-arguments EXC
 -- procedure: sfun-conversion-exception-code EXC
 -- procedure: sfun-conversion-exception-message EXC
     Sfun-conversion-exception objects are raised by the C-interface
     when converting between the Scheme representation and the C
     representation of a value during a call from C to Scheme.  The
     parameter EXC must be a sfun-conversion-exception object.

     The procedure `sfun-conversion-exception?' returns `#t' when OBJ
     is a sfun-conversion-exception object and `#f' otherwise.

     The procedure `sfun-conversion-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `sfun-conversion-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     The procedure `sfun-conversion-exception-code' returns an exact
     integer error code that can be converted to a string by the
     `err-code->string' procedure.

     The procedure `sfun-conversion-exception-message' returns `#f' or
     a string giving details of the exception in a human-readable form.

     For example:

          $ cat test2.scm
          (c-define (f str) (nonnull-char-string) int "f" ""
            (string->number str))
          (define t1 (c-lambda () int "___result = f (\"123\");"))
          (define t2 (c-lambda () int "___result = f (0);"))
          (define t3 (c-lambda () int "___result = f (\"1.5\");"))
          $ gsc test2.scm
          $ gsi
          Gambit Version 4.0 beta 20

          > (load "test2")
          "/u/feeley/test2.o1"
          > (t1)
          123
          > (define (handler exc)
              (if (sfun-conversion-exception? exc)
                  (list (sfun-conversion-exception-procedure exc)
                        (sfun-conversion-exception-arguments exc)
                        (err-code->string (sfun-conversion-exception-code exc))
                        (sfun-conversion-exception-message exc))
                  'not-sfun-conversion-exception))
          > (with-exception-catcher handler t2)
          (#<procedure #2 f>
           ()
           "(Argument 1) Can't convert from C nonnull-char-string"
           #f)
          > (with-exception-catcher handler t3)
          (#<procedure #2 f> () "Can't convert result to C int" #f)


 -- procedure: multiple-c-return-exception? OBJ
     Multiple-c-return-exception objects are raised by the C-interface
     when a C to Scheme procedure call returns and that call's stack
     frame is no longer on the C stack because the call has already
     returned, or has been removed from the C stack by a `longjump'.

     The procedure `multiple-c-return-exception?' returns `#t' when OBJ
     is a multiple-c-return-exception object and `#f' otherwise.

     For example:

          $ cat test3.scm
          (c-define (f str) (char-string) scheme-object "f" ""
            (pp (list 'entry 'str= str))
            (let ((k (call-with-current-continuation (lambda (k) k))))
              (pp (list 'exit 'k= k))
              k))
          (define scheme-to-c-to-scheme-and-back
            (c-lambda (char-string) scheme-object
              "___result = f (___arg1);"))
          $ gsc test3.scm
          $ gsi
          Gambit Version 4.0 beta 20

          > (load "test3")
          "/Users/feeley/gambit/doc/test3.o1"
          > (define (handler exc)
              (if (multiple-c-return-exception? exc)
                  exc
                  'not-multiple-c-return-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (let ((c (scheme-to-c-to-scheme-and-back "hello")))
                  (pp c)
                  (c 999))))
          (entry str= "hello")
          (exit k= #<procedure #2>)
          #<procedure #2>
          (exit k= 999)
          #<multiple-c-return-exception #3>


15.6 Exception objects related to the reader
============================================

 -- procedure: datum-parsing-exception? OBJ
 -- procedure: datum-parsing-exception-kind EXC
 -- procedure: datum-parsing-exception-parameters EXC
 -- procedure: datum-parsing-exception-readenv EXC
     Datum-parsing-exception objects are raised by the reader (i.e. the
     `read' procedure) when the input does not conform to the grammar
     for datum.  The parameter EXC must be a datum-parsing-exception
     object.

     The procedure `datum-parsing-exception?' returns `#t' when OBJ is
     a datum-parsing-exception object and `#f' otherwise.

     The procedure `datum-parsing-exception-kind' returns a symbol
     denoting the kind of parsing error that was encountered by the
     reader when it raised EXC.  Here is a table of the possible return
     values:

     `datum-or-eof-expected'            Datum or EOF expected
     `datum-expected'                   Datum expected
     `improperly-placed-dot'            Improperly placed dot
     `incomplete-form-eof-reached'      Incomplete form, EOF reached
     `incomplete-form'                  Incomplete form
     `character-out-of-range'           Character out of range
     `invalid-character-name'           Invalid '#\' name
     `illegal-character'                Illegal character
     `s8-expected'                      Signed 8 bit exact integer
                                        expected
     `u8-expected'                      Unsigned 8 bit exact integer
                                        expected
     `s16-expected'                     Signed 16 bit exact integer
                                        expected
     `u16-expected'                     Unsigned 16 bit exact integer
                                        expected
     `s32-expected'                     Signed 32 bit exact integer
                                        expected
     `u32-expected'                     Unsigned 32 bit exact integer
                                        expected
     `s64-expected'                     Signed 64 bit exact integer
                                        expected
     `u64-expected'                     Unsigned 64 bit exact integer
                                        expected
     `inexact-real-expected'            Inexact real expected
     `invalid-hex-escape'               Invalid hexadecimal escape
     `invalid-escaped-character'        Invalid escaped character
     `open-paren-expected'              '(' expected
     `invalid-token'                    Invalid token
     `invalid-sharp-bang-name'          Invalid '#!' name
     `duplicate-label-definition'       Duplicate definition for label
     `missing-label-definition'         Missing definition for label
     `illegal-label-definition'         Illegal definition of label
     `invalid-infix-syntax-character'   Invalid infix syntax character
     `invalid-infix-syntax-number'      Invalid infix syntax number
     `invalid-infix-syntax'             Invalid infix syntax

     The procedure `datum-parsing-exception-parameters' returns a list
     of the parameters associated with the parsing error that was
     encountered by the reader when it raised EXC.

     For example:

          > (define (handler exc)
              (if (datum-parsing-exception? exc)
                  (list (datum-parsing-exception-kind exc)
                        (datum-parsing-exception-parameters exc))
                  'not-datum-parsing-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (with-input-from-string "(s #\\pace)" read)))
          (invalid-character-name ("pace"))


15.7 Exception objects related to evaluation and compilation
============================================================

 -- procedure: expression-parsing-exception? OBJ
 -- procedure: expression-parsing-exception-kind EXC
 -- procedure: expression-parsing-exception-parameters EXC
 -- procedure: expression-parsing-exception-source EXC
     Expression-parsing-exception objects are raised by the evaluator
     and compiler (i.e. the procedures `eval', `compile-file', etc)
     when the input does not conform to the grammar for expression.  The
     parameter EXC must be a expression-parsing-exception object.

     The procedure `expression-parsing-exception?' returns `#t' when
     OBJ is a expression-parsing-exception object and `#f' otherwise.

     The procedure `expression-parsing-exception-kind' returns a symbol
     denoting the kind of parsing error that was encountered by the
     evaluator or compiler when it raised EXC.  Here is a table of the
     possible return values:

     `id-expected'                      Identifier expected
     `ill-formed-namespace'             Ill-formed namespace
     `ill-formed-namespace-prefix'      Ill-formed namespace prefix
     `namespace-prefix-must-be-string'  Namespace prefix must be a string
     `macro-used-as-variable'           Macro name can't be used as a
                                        variable
     `ill-formed-macro-transformer'     Macro transformer must be a
                                        lambda expression
     `reserved-used-as-variable'        Reserved identifier can't be used
                                        as a variable
     `ill-formed-special-form'          Ill-formed special form
     `cannot-open-file'                 Can't open file
     `filename-expected'                Filename expected
     `ill-placed-define'                Ill-placed 'define'
     `ill-placed-**include'             Ill-placed '##include'
     `ill-placed-**define-macro'        Ill-placed '##define-macro'
     `ill-placed-**declare'             Ill-placed '##declare'
     `ill-placed-**namespace'           Ill-placed '##namespace'
     `ill-formed-expression'            Ill-formed expression
     `unsupported-special-form'         Interpreter does not support
     `ill-placed-unquote'               Ill-placed 'unquote'
     `ill-placed-unquote-splicing'      Ill-placed 'unquote-splicing'
     `parameter-must-be-id'             Parameter must be an identifier
     `parameter-must-be-id-or-default'  Parameter must be an identifier
                                        or default binding
     `duplicate-parameter'              Duplicate parameter in parameter
                                        list
     `ill-placed-dotted-rest-parameter' Ill-placed dotted rest parameter
     `parameter-expected-after-rest'    #!rest must be followed by a
                                        parameter
     `ill-formed-default'               Ill-formed default binding
     `ill-placed-optional'              Ill-placed #!optional
     `ill-placed-rest'                  Ill-placed #!rest
     `ill-placed-key'                   Ill-placed #!key
     `key-expected-after-rest'          #!key expected after rest
                                        parameter
     `ill-placed-default'               Ill-placed default binding
     `duplicate-variable-definition'    Duplicate definition of a variable
     `empty-body'                       Body must contain at least one
                                        expression
     `variable-must-be-id'              Defined variable must be an
                                        identifier
     `else-clause-not-last'             Else clause must be last
     `ill-formed-selector-list'         Ill-formed selector list
     `duplicate-variable-binding'       Duplicate variable in bindings
     `ill-formed-binding-list'          Ill-formed binding list
     `ill-formed-call'                  Ill-formed procedure call
     `ill-formed-cond-expand'           Ill-formed 'cond-expand'
     `unfulfilled-cond-expand'          Unfulfilled 'cond-expand'

     The procedure `expression-parsing-exception-parameters' returns a
     list of the parameters associated with the parsing error that was
     encountered by the evaluator or compiler when it raised EXC.

     For example:

          > (define (handler exc)
              (if (expression-parsing-exception? exc)
                  (list (expression-parsing-exception-kind exc)
                        (expression-parsing-exception-parameters exc))
                  'not-expression-parsing-exception))
          > (with-exception-catcher
              handler
              (lambda ()
                (eval '(+ do 1))))
          (reserved-used-as-variable (do))


 -- procedure: unbound-global-exception? OBJ
 -- procedure: unbound-global-exception-variable EXC
 -- procedure: unbound-global-exception-code EXC
 -- procedure: unbound-global-exception-rte EXC
     Unbound-global-exception objects are raised when an unbound global
     variable is accessed.  The parameter EXC must be an
     unbound-global-exception object.

     The procedure `unbound-global-exception?' returns `#t' when OBJ is
     an unbound-global-exception object and `#f' otherwise.

     The procedure `unbound-global-exception-variable' returns a symbol
     identifying the unbound global variable.

     For example:

          > (define (handler exc)
              (if (unbound-global-exception? exc)
                  (list 'variable= (unbound-global-exception-variable exc))
                  'not-unbound-global-exception))
          > (with-exception-catcher
              handler
              (lambda () foo))
          (variable= foo)


15.8 Exception objects related to type checking
===============================================

 -- procedure: type-exception? OBJ
 -- procedure: type-exception-procedure EXC
 -- procedure: type-exception-arguments EXC
 -- procedure: type-exception-arg-num EXC
 -- procedure: type-exception-type-id EXC
     Type-exception objects are raised when a primitive procedure is
     called with an argument of incorrect type (i.e. when a run time
     type-check fails).  The parameter EXC must be a type-exception
     object.

     The procedure `type-exception?' returns `#t' when OBJ is a
     type-exception object and `#f' otherwise.

     The procedure `type-exception-procedure' returns the procedure
     that raised EXC.

     The procedure `type-exception-arguments' returns the list of
     arguments of the procedure that raised EXC.

     The procedure `type-exception-arg-num' returns the position of the
     argument whose type is incorrect.  Position 1 is the first
     argument.

     The procedure `type-exception-type-id' returns an identifier of
     the type expected.  The type-id can be a symbol, such as `number'
     and `string-or-nonnegative-fixnum', or a record type descriptor.

     For example:

          > (define (handler exc)
              (if (type-exception? exc)
                  (list (type-exception-procedure exc)
                        (type-exception-arguments exc)
                        (type-exception-arg-num exc)
                        (type-exception-type-id exc))
                  'not-type-exception))
          > (with-exception-catcher
              handler
              (lambda () (vector-ref '#(a b c) 'foo)))
          (#<procedure #2 vector-ref> (#(a b c) foo) 2 exact-integer)
          > (with-exception-catcher
              handler
              (lambda () (time->seconds 'foo)))
          (#<procedure #3 time->seconds> (foo) 1 #<type #4 time>)


 -- procedure: range-exception? OBJ
 -- procedure: range-exception-procedure EXC
 -- procedure: range-exception-arguments EXC
 -- procedure: range-exception-arg-num EXC
     Range-exception objects are raised when a numeric parameter is not
     in the allowed range.  The parameter EXC must be a range-exception
     object.

     The procedure `range-exception?' returns `#t' when OBJ is a
     range-exception object and `#f' otherwise.

     The procedure `range-exception-procedure' returns the procedure
     that raised EXC.

     The procedure `range-exception-arguments' returns the list of
     arguments of the procedure that raised EXC.

     The procedure `range-exception-arg-num' returns the position of
     the argument which is not in the allowed range.  Position 1 is the
     first argument.

     For example:

          > (define (handler exc)
              (if (range-exception? exc)
                  (list (range-exception-procedure exc)
                        (range-exception-arguments exc)
                        (range-exception-arg-num exc))
                  'not-range-exception))
          > (with-exception-catcher
              handler
              (lambda () (string-ref "abcde" 10)))
          (#<procedure #2 string-ref> ("abcde" 10) 2)


 -- procedure: divide-by-zero-exception? OBJ
 -- procedure: divide-by-zero-exception-procedure EXC
 -- procedure: divide-by-zero-exception-arguments EXC
     Divide-by-zero-exception objects are raised when a division by
     zero is attempted.  The parameter EXC must be a
     divide-by-zero-exception object.

     The procedure `divide-by-zero-exception?' returns `#t' when OBJ is
     a divide-by-zero-exception object and `#f' otherwise.

     The procedure `divide-by-zero-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `divide-by-zero-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (divide-by-zero-exception? exc)
                  (list (divide-by-zero-exception-procedure exc)
                        (divide-by-zero-exception-arguments exc))
                  'not-divide-by-zero-exception))
          > (with-exception-catcher
              handler
              (lambda () (/ 5 0 7)))
          (#<procedure #2 /> (5 0 7))


 -- procedure: improper-length-list-exception? OBJ
 -- procedure: improper-length-list-exception-procedure EXC
 -- procedure: improper-length-list-exception-arguments EXC
 -- procedure: improper-length-list-exception-arg-num EXC
     Improper-length-list-exception objects are raised by the `map' and
     `for-each' procedures when they are called with two or more list
     arguments and the lists are not of the same length.  The parameter
     EXC must be an improper-length-list-exception object.

     The procedure `improper-length-list-exception?' returns `#t' when
     OBJ is an improper-length-list-exception object and `#f' otherwise.

     The procedure `improper-length-list-exception-procedure' returns
     the procedure that raised EXC.

     The procedure `improper-length-list-exception-arguments' returns
     the list of arguments of the procedure that raised EXC.

     The procedure `improper-length-list-exception-arg-num' returns the
     position of the argument whose length is the shortest.  Position 1
     is the first argument.

     For example:

          > (define (handler exc)
              (if (improper-length-list-exception? exc)
                  (list (improper-length-list-exception-procedure exc)
                        (improper-length-list-exception-arguments exc)
                        (improper-length-list-exception-arg-num exc))
                  'not-improper-length-list-exception))
          > (with-exception-catcher
              handler
              (lambda () (map + '(1 2) '(3) '(4 5))))
          (#<procedure #2 map> (#<procedure #3 +> (1 2) (3) (4 5)) 3)


15.9 Exception objects related to procedure call
================================================

 -- procedure: wrong-number-of-arguments-exception? OBJ
 -- procedure: wrong-number-of-arguments-exception-procedure EXC
 -- procedure: wrong-number-of-arguments-exception-arguments EXC
     Wrong-number-of-arguments-exception objects are raised when a
     procedure is called with the wrong number of arguments.  The
     parameter EXC must be a wrong-number-of-arguments-exception object.

     The procedure `wrong-number-of-arguments-exception?' returns `#t'
     when OBJ is a wrong-number-of-arguments-exception object and `#f'
     otherwise.

     The procedure `wrong-number-of-arguments-exception-procedure'
     returns the procedure that raised EXC.

     The procedure `wrong-number-of-arguments-exception-arguments'
     returns the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (wrong-number-of-arguments-exception? exc)
                  (list (wrong-number-of-arguments-exception-procedure exc)
                        (wrong-number-of-arguments-exception-arguments exc))
                  'not-wrong-number-of-arguments-exception))
          > (with-exception-catcher
              handler
              (lambda () (open-input-file "data" 99)))
          (#<procedure #2 open-input-file> ("data" 99))


 -- procedure: number-of-arguments-limit-exception? OBJ
 -- procedure: number-of-arguments-limit-exception-procedure EXC
 -- procedure: number-of-arguments-limit-exception-arguments EXC
     Number-of-arguments-limit-exception objects are raised by the
     `apply' procedure when the procedure being called is passed more
     than 8192 arguments.  The parameter EXC must be a
     number-of-arguments-limit-exception object.

     The procedure `number-of-arguments-limit-exception?' returns `#t'
     when OBJ is a number-of-arguments-limit-exception object and `#f'
     otherwise.

     The procedure `number-of-arguments-limit-exception-procedure'
     returns the target procedure of the call to `apply' that raised
     EXC.

     The procedure `number-of-arguments-limit-exception-arguments'
     returns the list of arguments of the target procedure of the call
     to `apply' that raised EXC.

     For example:

          > (define (iota n) (if (= n 0) '() (cons n (iota (- n 1)))))
          > (define (handler exc)
              (if (number-of-arguments-limit-exception? exc)
                  (list (number-of-arguments-limit-exception-procedure exc)
                        (length (number-of-arguments-limit-exception-arguments exc)))
                  'not-number-of-arguments-limit-exception))
          > (with-exception-catcher
              handler
              (lambda () (apply + 1 2 3 (iota 8190))))
          (#<procedure #2 +> 8193)


 -- procedure: nonprocedure-operator-exception? OBJ
 -- procedure: nonprocedure-operator-exception-operator EXC
 -- procedure: nonprocedure-operator-exception-arguments EXC
 -- procedure: nonprocedure-operator-exception-code EXC
 -- procedure: nonprocedure-operator-exception-rte EXC
     Nonprocedure-operator-exception objects are raised when a procedure
     call is executed and the operator position is not a procedure.  The
     parameter EXC must be an nonprocedure-operator-exception object.

     The procedure `nonprocedure-operator-exception?' returns `#t' when
     OBJ is an nonprocedure-operator-exception object and `#f'
     otherwise.

     The procedure `nonprocedure-operator-exception-operator' returns
     the value in operator position of the procedure call that raised
     EXC.

     The procedure `nonprocedure-operator-exception-arguments' returns
     the list of arguments of the procedure call that raised EXC.

     For example:

          > (define (handler exc)
              (if (nonprocedure-operator-exception? exc)
                  (list (nonprocedure-operator-exception-operator exc)
                        (nonprocedure-operator-exception-arguments exc))
                  'not-nonprocedure-operator-exception))
          > (with-exception-catcher
              handler
              (lambda () (11 22 33)))
          (11 (22 33))


 -- procedure: unknown-keyword-argument-exception? OBJ
 -- procedure: unknown-keyword-argument-exception-procedure EXC
 -- procedure: unknown-keyword-argument-exception-arguments EXC
     Unknown-keyword-argument-exception objects are raised when a
     procedure accepting keyword arguments is called and one of the
     keywords supplied is not among those that are expected.  The
     parameter EXC must be an unknown-keyword-argument-exception object.

     The procedure `unknown-keyword-argument-exception?' returns `#t'
     when OBJ is an unknown-keyword-argument-exception object and `#f'
     otherwise.

     The procedure `unknown-keyword-argument-exception-procedure'
     returns the procedure that raised EXC.

     The procedure `unknown-keyword-argument-exception-arguments'
     returns the list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (unknown-keyword-argument-exception? exc)
                  (list (unknown-keyword-argument-exception-procedure exc)
                        (unknown-keyword-argument-exception-arguments exc))
                  'not-unknown-keyword-argument-exception))
          > (with-exception-catcher
              handler
              (lambda () ((lambda (#!key (foo 5)) foo) bar: 11)))
          (#<procedure #2> (bar: 11))


 -- procedure: keyword-expected-exception? OBJ
 -- procedure: keyword-expected-exception-procedure EXC
 -- procedure: keyword-expected-exception-arguments EXC
     Keyword-expected-exception objects are raised when a procedure
     accepting keyword arguments is called and a nonkeyword was supplied
     where a keyword was expected.  The parameter EXC must be an
     keyword-expected-exception object.

     The procedure `keyword-expected-exception?' returns `#t' when OBJ
     is an keyword-expected-exception object and `#f' otherwise.

     The procedure `keyword-expected-exception-procedure' returns the
     procedure that raised EXC.

     The procedure `keyword-expected-exception-arguments' returns the
     list of arguments of the procedure that raised EXC.

     For example:

          > (define (handler exc)
              (if (keyword-expected-exception? exc)
                  (list (keyword-expected-exception-procedure exc)
                        (keyword-expected-exception-arguments exc))
                  'not-keyword-expected-exception))
          > (with-exception-catcher
              handler
              (lambda () ((lambda (#!key (foo 5)) foo) 11 22)))
          (#<procedure #2> (11 22))


15.10 Other exception objects
=============================

 -- procedure: error-exception? OBJ
 -- procedure: error-exception-message EXC
 -- procedure: error-exception-parameters EXC
 -- procedure: error MESSAGE OBJ...
     Error-exception objects are raised when the `error' procedure is
     called.  The parameter EXC must be an error-exception object.

     The procedure `error-exception?' returns `#t' when OBJ is an
     error-exception object and `#f' otherwise.

     The procedure `error-exception-message' returns the first argument
     of the call to `error' that raised EXC.

     The procedure `error-exception-parameters' returns the list of
     arguments, starting with the second argument, of the call to
     `error' that raised EXC.

     The `error' procedure raises an error-exception object whose
     message field is MESSAGE and parameters field is the list of
     values OBJ....

     For example:

          > (define (handler exc)
              (if (error-exception? exc)
                  (list (error-exception-message exc)
                        (error-exception-parameters exc))
                  'not-error-exception))
          > (with-exception-catcher
              handler
              (lambda () (error "unexpected object:" 123)))
          ("unexpected object:" (123))


16 Host environment
*******************

The host environment is the set of resources, such as the filesystem,
network and processes, that are managed by the operating system within
which the Scheme program is executing.  This chapter specifies how the
host environment can be accessed from within the Scheme program.

   In this chapter we say that the Scheme program being executed is a
process, even though the concept of process does not exist in some
operating systems supported by Gambit (e.g. MSDOS).

16.1 Handling of file names
===========================

Gambit uses a naming convention for files that is compatible with the
one used by the host environment but extended to allow referring to the
"home directory" of the current user or some specific user and the
"Gambit installation directory".

   A "path" is a string that denotes a file, for example
`"src/readme.txt"'.  Each component of a path is separated by a `/'
under UNIX and Mac OS X and by a `/' or `\' under MSDOS and Microsoft
Windows.  A leading separator indicates an absolute path under UNIX,
Mac OS X, MSDOS and Microsoft Windows.  A path which does not contain a
path separator is relative to the "current working directory" on all
operating systems.  A volume specifier such as `C:' may prefix a file
name under MSDOS and Microsoft Windows.

   The rest of this section uses `/' to represent the path separator.

   A path which starts with the characters `~~/' denotes a file in the
Gambit installation directory.  This directory is normally
`/usr/local/Gambit-C/version' under UNIX and Mac OS X and
`C:/Gambit-C/version' under MSDOS and Microsoft Windows.  To override
this binding under UNIX, Mac OS X, MSDOS and Microsoft Windows, use the
`-:=<DIR>' runtime option or define the `GAMBCOPT' environment variable.

   A path which starts with the characters `~/' denotes a file in the
user's home directory.  The user's home directory is contained in the
`HOME' environment variable under UNIX, Mac OS X, MSDOS and Microsoft
Windows.  Under MSDOS and Microsoft Windows, if the `HOME' environment
variable is not defined, the environment variables `HOMEDRIVE' and
`HOMEPATH' are concatenated if they are defined.  If this fails to
yield a home directory, the Gambit installation directory is used
instead.

   A path which starts with the characters `~USERNAME/' denotes a file
in the home directory of the given user.  Under UNIX and Mac OS X this
is found using the password file.  There is no equivalent under MSDOS
and Microsoft Windows.

 -- procedure: current-directory [NEW-CURRENT-DIRECTORY]
     The parameter object `current-directory' is bound to the current
     working directory.  Calling this procedure with no argument returns
     the absolute "normalized path" of the directory and calling this
     procedure with one argument sets the directory to
     NEW-CURRENT-DIRECTORY.  The initial binding of this parameter
     object is the current working directory of the current process.
     The path returned by `current-directory' always contains a trailing
     directory separator.  Modifications of the parameter object do not
     change the current working directory of the current process (i.e.
     that is accessible with the UNIX `getcwd()' function and the
     Microsoft Windows `GetCurrentDirectory' function).  It is an error
     to mutate the string returned by `current-directory'.

     For example under UNIX:

          > (current-directory)
          "/Users/feeley/gambit/doc/"
          > (current-directory "..")
          > (current-directory)
          "/Users/feeley/gambit/"
          > (path-expand "foo" "~~")
          "/usr/local/Gambit-C/4.0b20/foo"
          > (parameterize ((current-directory "~~")) (path-expand "foo"))
          "/usr/local/Gambit-C/4.0b20/foo"


 -- procedure: path-expand PATH [ORIGIN-DIRECTORY]
     The procedure `path-expand' takes the path of a file or directory
     and returns an expanded path, which is an absolute path when PATH
     or ORIGIN-DIRECTORY are absolute paths.  The optional
     ORIGIN-DIRECTORY parameter, which defaults to the current working
     directory, is the directory used to resolve relative paths.
     Components of the paths PATH and ORIGIN-DIRECTORY need not exist.

     For example under UNIX:

          > (path-expand "foo")
          "/Users/feeley/gambit/doc/foo"
          > (path-expand "~/foo")
          "/Users/feeley/foo"
          > (path-expand "~~/foo")
          "/usr/local/Gambit-C/4.0b20/foo"
          > (path-expand "../foo")
          "/Users/feeley/gambit/doc/../foo"
          > (path-expand "foo" "")
          "foo"
          > (path-expand "foo" "/tmp")
          "/tmp/foo"
          > (path-expand "this/file/does/not/exist")
          "/Users/feeley/gambit/doc/this/file/does/not/exist"
          > (path-expand "")
          "/Users/feeley/gambit/doc/"


 -- procedure: path-normalize PATH [ALLOW-RELATIVE? [ORIGIN-DIRECTORY]]
     The procedure `path-normalize' takes a path of a file or directory
     and returns its normalized path.  The optional ORIGIN-DIRECTORY
     parameter, which defaults to the current working directory, is the
     directory used to resolve relative paths.  All components of the
     paths PATH and ORIGIN-DIRECTORY must exist, except possibly the
     last component of PATH.  A normalized path is a path containing no
     redundant parts and which is consistent with the current structure
     of the filesystem.  A normalized path of a directory will always
     end with a path separator (i.e. `/', `\', or `:' depending on the
     operating system).  The optional ALLOW-RELATIVE? parameter, which
     defaults to `#f', indicates if the path returned can be expressed
     relatively to ORIGIN-DIRECTORY: a `#f' requests an absolute path,
     the symbol `shortest' requests the shortest of the absolute and
     relative paths, and any other value requests the relative path.
     The shortest path is useful for interaction with the user because
     short relative paths are typically easier to read than long
     absolute paths.

     For example under UNIX:

          > (path-expand "../foo")
          "/Users/feeley/gambit/doc/../foo"
          > (path-normalize "../foo")
          "/Users/feeley/gambit/foo"
          > (path-normalize "this/file/does/not/exist")
          *** ERROR IN (console)@3.1 -- No such file or directory
          (path-normalize "this/file/does/not/exist")


 -- procedure: path-extension PATH
 -- procedure: path-strip-extension PATH
 -- procedure: path-directory PATH
 -- procedure: path-strip-directory PATH
 -- procedure: path-strip-trailing-directory-separator PATH
 -- procedure: path-volume PATH
 -- procedure: path-strip-volume PATH
     These procedures extract various parts of a path, which need not
     be a normalized path.  The procedure `path-extension' returns the
     file extension (including the period) or the empty string if there
     is no extension.  The procedure `path-strip-extension' returns the
     path with the extension stripped off.  The procedure
     `path-directory' returns the file's directory (including the last
     path separator) or the empty string if no directory is specified
     in the path.  The procedure `path-strip-directory' returns the
     path with the directory stripped off.  The procedure
     `path-strip-trailing-directory-separator' returns the path with
     the directory separator stripped off if one is at the end of the
     path.  The procedure `path-volume' returns the file's volume
     (including the last path separator) or the empty string if no
     volume is specified in the path.  The procedure
     `path-strip-volume' returns the path with the volume stripped off.

     For example under UNIX:

          > (path-extension "/tmp/foo")
          ""
          > (path-extension "/tmp/foo.txt")
          ".txt"
          > (path-strip-extension "/tmp/foo.txt")
          "/tmp/foo"
          > (path-directory "/tmp/foo.txt")
          "/tmp/"
          > (path-strip-directory "/tmp/foo.txt")
          "foo.txt"
          > (path-strip-trailing-directory-separator "/usr/local/bin/")
          "/usr/local/bin"
          > (path-strip-trailing-directory-separator "/usr/local/bin")
          "/usr/local/bin"
          > (path-volume "/tmp/foo.txt")
          ""
          > (path-volume "C:/tmp/foo.txt")
          "" ; result is "C:" under Microsoft Windows
          > (path-strip-volume "C:/tmp/foo.txt")
          "C:/tmp/foo.txt" ; result is "/tmp/foo.txt" under Microsoft Windows


16.2 Filesystem operations
==========================

 -- procedure: create-directory PATH-OR-SETTINGS
     This procedure creates a directory.  The argument PATH-OR-SETTINGS
     is either a string denoting a filesystem path or a list of port
     settings which must contain a `path:' setting.  Here are the
     settings allowed:

        * `path:' STRING

          This setting indicates the location of the directory to
          create in the filesystem.  There is no default value for this
          setting.

        * `permissions:' 12-BIT-EXACT-INTEGER

          This setting controls the UNIX permissions that will be
          attached to the file if it is created.  The default value of
          this setting is `#o777'.


     For example:

          > (create-directory "newdir")
          > (create-directory "newdir")
          *** ERROR IN (console)@2.1 -- File exists
          (create-directory "newdir")


 -- procedure: create-fifo PATH-OR-SETTINGS
     This procedure creates a FIFO.  The argument PATH-OR-SETTINGS is
     either a string denoting a filesystem path or a list of port
     settings which must contain a `path:' setting.  Here are the
     settings allowed:

        * `path:' STRING

          This setting indicates the location of the FIFO to create in
          the filesystem.  There is no default value for this setting.

        * `permissions:' 12-BIT-EXACT-INTEGER

          This setting controls the UNIX permissions that will be
          attached to the file if it is created.  The default value of
          this setting is `#o666'.


     For example:

          > (create-fifo "fifo")
          > (define a (open-input-file "fifo"))
          > (define b (open-output-file "fifo"))
          > (display "1 22 333" b)
          > (force-output b)
          > (read a)
          1
          > (read a)
          22


 -- procedure: create-link SOURCE-PATH DESTINATION-PATH
     This procedure creates a hard link between SOURCE-PATH and
     DESTINATION-PATH.  The argument SOURCE-PATH must be a string
     denoting the path of an existing file.  The argument
     DESTINATION-PATH must be a string denoting the path of the link to
     create.


 -- procedure: create-symbolic-link SOURCE-PATH DESTINATION-PATH
     This procedure creates a symbolic link between SOURCE-PATH and
     DESTINATION-PATH.  The argument SOURCE-PATH must be a string
     denoting the path of an existing file.  The argument
     DESTINATION-PATH must be a string denoting the path of the
     symbolic link to create.


 -- procedure: rename-file SOURCE-PATH DESTINATION-PATH
     This procedure renames the file SOURCE-PATH to DESTINATION-PATH.
     The argument SOURCE-PATH must be a string denoting the path of an
     existing file.  The argument DESTINATION-PATH must be a string
     denoting the new path of the file.


 -- procedure: copy-file SOURCE-PATH DESTINATION-PATH
     This procedure copies the file SOURCE-PATH to DESTINATION-PATH.
     The argument SOURCE-PATH must be a string denoting the path of an
     existing file.  The argument DESTINATION-PATH must be a string
     denoting the path of the file to create.


 -- procedure: delete-file PATH
     This procedure deletes the file PATH.  The argument PATH must be a
     string denoting the path of an existing file.


 -- procedure: delete-directory PATH
     This procedure deletes the directory PATH.  The argument PATH must
     be a string denoting the path of an existing directory.


 -- procedure: directory-files [PATH-OR-SETTINGS]
     This procedure returns the list of the files in a directory.  The
     argument PATH-OR-SETTINGS is either a string denoting a filesystem
     path to a directory or a list of settings which must contain a
     `path:' setting.  If it is not specified, PATH-OR-SETTINGS
     defaults to the current directory (the value bound to the
     `current-directory' parameter object).  Here are the settings
     allowed:

        * `path:' STRING

          This setting indicates the location of the directory in the
          filesystem.  There is no default value for this setting.

        * `ignore-hidden:' ( `#f' | `#t' | `dot-and-dot-dot' )

          This setting controls whether hidden-files will be returned.
          Under UNIX and Mac OS X hidden-files are those that start
          with a period (such as `.', `..', and `.profile').  Under
          Microsoft Windows hidden files are the `.' and `..' entries
          and the files whose "hidden file" attribute is set.  A
          setting of `#f' will enumerate all the files.  A setting of
          `#t' will only enumerate the files that are not hidden.  A
          setting of `dot-and-dot-dot' will enumerate all the files
          except for the `.' and `..' hidden files.  The default value
          of this setting is `#t'.


     For example:

          > (directory-files)
          ("complex" "README" "simple")
          > (directory-files "../include")
          ("config.h" "config.h.in" "gambit.h" "makefile" "makefile.in")
          > (directory-files (list path: "../include" ignore-hidden: #f))
          ("." ".." "config.h" "config.h.in" "gambit.h" "makefile" "makefile.in")


16.3 Shell command execution
============================

 -- procedure: shell-command COMMAND
     The procedure `shell-command' calls up the shell to execute
     COMMAND which must be a string.  This procedure returns the exit
     status of the shell in the form that the C library's `system'
     routine returns.

     For example under UNIX:

          > (shell-command "ls -sk f*.scm")
          4 fact.scm   4 fib.scm
          0


16.4 Process termination
========================

 -- procedure: exit [STATUS]
     The procedure `exit' causes the process to terminate with the
     status STATUS which must be an exact integer in the range 0 to
     255.  If it is not specified, STATUS defaults to 0.

     For example under UNIX:

          $ gsi
          Gambit Version 4.0 beta 20

          > (exit 42)
          $ echo $?
          42


16.5 Command line arguments
===========================

 -- procedure: command-line
     This procedure returns a list of strings corresponding to the
     command line arguments, including the program file name as the
     first element of the list.  When the interpreter executes a Scheme
     script, the list returned by `command-line' contains the script's
     absolute path followed by the remaining command line arguments.

     For example under UNIX:

          $ gsi -:d -e "(pretty-print (command-line))"
          ("gsi" "-e" "(pretty-print (command-line))")
          $ cat foo
          #!/usr/local/Gambit-C/current/bin/gsi-script
          (pretty-print (command-line))
          $ ./foo 1 2 "3 4"
          ("/u/feeley/./foo" "1" "2" "3 4")


16.6 Environment variables
==========================

 -- procedure: getenv NAME [DEFAULT]
 -- procedure: setenv NAME [NEW-VALUE]
     The procedure `getenv' returns the value of the environment
     variable NAME of the current process.  Variable names are denoted
     with strings.  A string is returned if the environment variable is
     bound, otherwise DEFAULT is returned if it is specified, otherwise
     an exception is raised.

     The procedure `setenv' changes the binding of the environment
     variable NAME to NEW-VALUE which must be a string.  If NEW-VALUE
     is not specified the binding is removed.

     For example under UNIX:

          > (getenv "HOME")
          "/Users/feeley"
          > (getenv "DOES_NOT_EXIST" #f)
          #f
          > (setenv "DOES_NOT_EXIST" "it does now")
          > (getenv "DOES_NOT_EXIST" #f)
          "it does now"
          > (setenv "DOES_NOT_EXIST")
          > (getenv "DOES_NOT_EXIST" #f)
          #f
          > (getenv "DOES_NOT_EXIST")
          *** ERROR IN (console)@7.1 -- Unbound OS environment variable
          (getenv "DOES_NOT_EXIST")


16.7 Measuring time
===================

Procedures are available for measuring real time (aka "wall" time) and
cpu time (the amount of time the cpu has been executing the process).
The resolution of the real time and cpu time clock is operating system
dependent.  Typically the resolution of the cpu time clock is rather
coarse (measured in "ticks" of 1/60th or 1/100th of a second).  Real
time is internally computed relative to some arbitrary point in time
using floating point numbers, which means that there is a gradual loss
of resolution as time elapses.  Moreover, some operating systems report
time in number of ticks using a 32 bit integer so the value returned by
the time related procedures may wraparound much before any significant
loss of resolution occurs (for example 2.7 years if ticks are 1/50th of
a second).

 -- procedure: current-time
 -- procedure: time? OBJ
 -- procedure: time->seconds TIME
 -- procedure: seconds->time X
     The procedure `current-time' returns a "time object" representing
     the current point in real time.

     The procedure `time?' returns `#t' when OBJ is a time object and
     `#f' otherwise.

     The procedure `time->seconds' converts the time object TIME into
     an inexact real number representing the number of seconds elapsed
     since the "epoch" (which is 00:00:00 Coordinated Universal Time
     01-01-1970).

     The procedure `seconds->time' converts the real number X
     representing the number of seconds elapsed since the "epoch" into a
     time object.

     For example:

          > (current-time)
          #<time #2>
          > (time? (current-time))
          #t
          > (time? 123)
          #f
          > (time->seconds (current-time))
          1083118758.63973
          > (time->seconds (current-time))
          1083118759.909163
          > (seconds->time (+ 10 (time->seconds (current-time))
          #<time #3>  ; a time object representing 10 seconds in the future


 -- procedure: process-times
 -- procedure: cpu-time
 -- procedure: real-time
     The procedure `process-times' returns a three element f64vector
     containing the cpu time that has been used by the program and the
     real time that has elapsed since it was started.  The first element
     corresponds to "user" time in seconds, the second element
     corresponds to "system" time in seconds and the third element is
     the elapsed real time in seconds.  On operating systems that can't
     differentiate user and system time, the system time is zero.  On
     operating systems that can't measure cpu time, the user time is
     equal to the elapsed real time and the system time is zero.

     The procedure `cpu-time' returns the cpu time in seconds that has
     been used by the program (user time plus system time).

     The procedure `real-time' returns the real time that has elapsed
     since the program was started.

     For example:

          > (process-times)
          #f64(.02794 .021754 .159926176071167)
          > (cpu-time)
          .051223
          > (real-time)
          .40660619735717773


 -- special form: time expr
     The `time' special form evaluates expr and returns the result.  As
     a side effect it displays a message on the interaction channel
     which indicates how long the evaluation took (in real time and cpu
     time), how much time was spent in the garbage collector, how much
     memory was allocated during the evaluation and how many minor and
     major page faults occured (0 is reported if not running under
     UNIX).

     For example:

          > (define (f x)
              (let loop ((x x) (lst '()))
                (if (= x 0)
                    lst
                    (loop (- x 1) (cons x lst)))))
          > (length (time (f 100000)))
          (time (f 100000))
              683 ms real time
              558 ms cpu time (535 user, 23 system)
              8 collections accounting for 102 ms real time (70 user, 5 system)
              6400160 bytes allocated
              no minor faults
              no major faults
          100000


16.8 File information
=====================

 -- procedure: file-exists? PATH [CHASE?]
     The PATH argument must be a string.  This procedure returns `#t'
     when a file by that name exists, and returns `#f' otherwise.

     When CHASE? is present and `#f', symbolic links will not be
     chased, in other words if PATH refers to a symbolic link,
     `file-exists?' will return `#t' whether or not it points to an
     existing file.

     For example:

          > (file-exists? "nofile")
          #f


 -- procedure: file-info PATH [CHASE?]
     This procedure accesses the filesystem to get information about the
     file whose location is given by the string PATH.  A
     file-information record is returned that contains the file's type,
     the device number, the inode number, the mode (permission bits),
     the number of links, the file's user id, the file's group id, the
     file's size in bytes, the times of last-access, last-modification
     and last-change, the attributes, and the creation time.

     When CHASE? is present and `#f', symbolic links will not be
     chased, in other words if PATH refers to a symbolic link the
     `file-info' procedure will return information about the link
     rather than the file it links to.

     For example:

          > (file-info "/dev/tty")
          #<file-info #2
             type: character-special
             device: 19513156
             inode: 20728196
             mode: 438
             number-of-links: 1
             owner: 0
             group: 0
             size: 0
             last-access-time: #<time #3>
             last-modification-time: #<time #4>
             last-change-time: #<time #5>
             attributes: 128
             creation-time: #<time #6>>


 -- procedure: file-info? OBJ
     This procedure returns `#t' when OBJ is a file-information record
     and `#f' otherwise.

     For example:

          > (file-info? (file-info "/dev/tty"))
          #t
          > (file-info? 123)
          #f


 -- procedure: file-info-type FILE-INFO
     Returns the type field of the file-information record FILE-INFO.
     The type is denoted by a symbol.  The following types are possible:

    `regular'
          Regular file

    `directory'
          Directory

    `character-special'
          Character special device

    `block-special'
          Block special device

    `fifo'
          FIFO

    `symbolic-link'
          Symbolic link

    `socket'
          Socket

    `unknown'
          File is of an unknown type

     For example:

          > (file-info-type (file-info "/dev/tty"))
          character-special
          > (file-info-type (file-info "/dev"))
          directory


 -- procedure: file-info-device FILE-INFO
     Returns the device field of the file-information record FILE-INFO.

     For example:

          > (file-info-device (file-info "/dev/tty"))
          19513156


 -- procedure: file-info-inode FILE-INFO
     Returns the inode field of the file-information record FILE-INFO.

     For example:

          > (file-info-inode (file-info "/dev/tty"))
          20728196


 -- procedure: file-info-mode FILE-INFO
     Returns the mode field of the file-information record FILE-INFO.

     For example:

          > (file-info-mode (file-info "/dev/tty"))
          438


 -- procedure: file-info-number-of-links FILE-INFO
     Returns the number-of-links field of the file-information record
     FILE-INFO.

     For example:

          > (file-info-number-of-links (file-info "/dev/tty"))
          1


 -- procedure: file-info-owner FILE-INFO
     Returns the owner field of the file-information record FILE-INFO.

     For example:

          > (file-info-owner (file-info "/dev/tty"))
          0


 -- procedure: file-info-group FILE-INFO
     Returns the group field of the file-information record FILE-INFO.

     For example:

          > (file-info-group (file-info "/dev/tty"))
          0


 -- procedure: file-info-size FILE-INFO
     Returns the size field of the file-information record FILE-INFO.

     For example:

          > (file-info-size (file-info "/dev/tty"))
          0


 -- procedure: file-info-last-access-time FILE-INFO
     Returns the last-access-time field of the file-information record
     FILE-INFO.

     For example:

          > (file-info-last-access-time (file-info "/dev/tty"))
          #<time #2>


 -- procedure: file-info-last-modification-time FILE-INFO
     Returns the last-modification-time field of the file-information
     record FILE-INFO.

     For example:

          > (file-info-last-modification-time (file-info "/dev/tty"))
          #<time #2>


 -- procedure: file-info-last-change-time FILE-INFO
     Returns the last-change-time field of the file-information record
     FILE-INFO.

     For example:

          > (file-info-last-change-time (file-info "/dev/tty"))
          #<time #2>


 -- procedure: file-info-attributes FILE-INFO
     Returns the attributes field of the file-information record
     FILE-INFO.

     For example:

          > (file-info-attributes (file-info "/dev/tty"))
          128


 -- procedure: file-info-creation-time FILE-INFO
     Returns the creation-time field of the file-information record
     FILE-INFO.

     For example:

          > (file-info-creation-time (file-info "/dev/tty"))
          #<time #2>


 -- procedure: file-type PATH
 -- procedure: file-device PATH
 -- procedure: file-inode PATH
 -- procedure: file-mode PATH
 -- procedure: file-number-of-links PATH
 -- procedure: file-owner PATH
 -- procedure: file-group PATH
 -- procedure: file-size PATH
 -- procedure: file-last-access-time PATH
 -- procedure: file-last-modification-time PATH
 -- procedure: file-last-change-time PATH
 -- procedure: file-attributes PATH
 -- procedure: file-creation-time PATH
     These procedures combine a call to the `file-info' procedure and a
     call to a file-information record field accessor.  For instance
     `(file-type PATH)' is equivalent to `(file-info-type (file-info
     PATH))'.


16.9 Group information
======================

 -- procedure: group-info GROUP-NAME-OR-ID
     This procedure accesses the group database to get information
     about the group identified by GROUP-NAME-OR-ID, which is the
     group's symbolic name (string) or the group's GID (exact integer).
     A group-information record is returned that contains the group's
     symbolic name, the group's id (GID), and the group's members (list
     of symbolic user names).

     For example:

          > (group-info "staff")
          #<group-info #2 name: "staff" gid: 20 members: ("root")>
          > (group-info 29)
          #<group-info #3
             name: "certusers"
             gid: 29
             members: ("root" "jabber" "postfix" "cyrusimap")>
          > (group-info 5000)
          *** ERROR IN (console)@3.1 -- Resource temporarily unavailable
          (group-info 5000)


 -- procedure: group-info? OBJ
     This procedure returns `#t' when OBJ is a group-information record
     and `#f' otherwise.

     For example:

          > (group-info? (group-info "daemon"))
          #t
          > (group-info? 123)
          #f


 -- procedure: group-info-name GROUP-INFO
     Returns the symbolic name field of the group-information record
     GROUP-INFO.

     For example:

          > (group-info-name (group-info 29))
          "certusers"


 -- procedure: group-info-gid GROUP-INFO
     Returns the group id field of the group-information record
     GROUP-INFO.

     For example:

          > (group-info-gid (group-info "staff"))
          20


 -- procedure: group-info-members GROUP-INFO
     Returns the members field of the group-information record
     GROUP-INFO.

     For example:

          > (group-info-members (group-info "staff"))
          ("root")


16.10 User information
======================

 -- procedure: user-name
     This procedure returns the user's name as a string.

     For example:

          > (user-name)
          "feeley"


 -- procedure: user-info USER-NAME-OR-ID
     This procedure accesses the user database to get information about
     the user identified by USER-NAME-OR-ID, which is the user's
     symbolic name (string) or the user's UID (exact integer).  A
     user-information record is returned that contains the user's
     symbolic name, the user's id (UID), the user's group id (GID), the
     path to the user's home directory, and the user's login shell.

     For example:

          > (user-info "feeley")
          #<user-info #2
             name: "feeley"
             uid: 506
             gid: 506
             home: "/Users/feeley"
             shell: "/bin/bash">
          > (user-info 0)
          #<user-info #3 name: "root" uid: 0 gid: 0 home: "/var/root" shell: "/bin/sh">
          > (user-info 5000)
          *** ERROR IN (console)@3.1 -- Resource temporarily unavailable
          (user-info 5000)


 -- procedure: user-info? OBJ
     This procedure returns `#t' when OBJ is a user-information record
     and `#f' otherwise.

     For example:

          > (user-info? (user-info "feeley"))
          #t
          > (user-info? 123)
          #f


 -- procedure: user-info-name USER-INFO
     Returns the symbolic name field of the user-information record
     USER-INFO.

     For example:

          > (user-info-name (user-info 0))
          "root"


 -- procedure: user-info-uid USER-INFO
     Returns the user id field of the user-information record USER-INFO.

     For example:

          > (user-info-uid (user-info "feeley"))
          506


 -- procedure: user-info-gid USER-INFO
     Returns the group id field of the user-information record
     USER-INFO.

     For example:

          > (user-info-gid (user-info "feeley"))
          506


 -- procedure: user-info-home USER-INFO
     Returns the home directory field of the user-information record
     USER-INFO.

     For example:

          > (user-info-home (user-info 0))
          "/var/root"


 -- procedure: user-info-shell USER-INFO
     Returns the shell field of the user-information record USER-INFO.

     For example:

          > (user-info-shell (user-info 0))
          "/bin/sh"


16.11 Host information
======================

 -- procedure: host-name
     This procedure returns the machine's host name as a string.

     For example:

          > (host-name)
          "mega.iro.umontreal.ca"


 -- procedure: host-info HOST-NAME
     This procedure accesses the internet host database to get
     information about the machine whose name is denoted by the string
     HOST-NAME.  A host-information record is returned that contains
     the official name of the machine, a list of aliases (alternative
     names), and a non-empty list of IP addresses for this machine.  An
     exception is raised when HOST-NAME does not appear in the database.

     For example:

          > (host-info "www.google.com")
          #<host-info #2
             name: "www.l.google.com"
             aliases: ("www.google.com")
             addresses: (#u8(66 249 85 99) #u8(66 249 85 104))>
          > (host-info "unknown.domain")
          *** ERROR IN (console)@2.1 -- Unknown host
          (host-info "unknown.domain")


 -- procedure: host-info? OBJ
     This procedure returns `#t' when OBJ is a host-information record
     and `#f' otherwise.

     For example:

          > (host-info? (host-info "www.google.com"))
          #t
          > (host-info? 123)
          #f


 -- procedure: host-info-name HOST-INFO
     Returns the official name field of the host-information record
     HOST-INFO.

     For example:

          > (host-info-name (host-info "www.google.com"))
          "www.l.google.com"


 -- procedure: host-info-aliases HOST-INFO
     Returns the aliases field of the host-information record
     HOST-INFO.  This field is a possibly empty list of strings.

     For example:

          > (host-info-aliases (host-info "www.google.com"))
          ("www.google.com")


 -- procedure: host-info-addresses HOST-INFO
     Returns the addresses field of the host-information record
     HOST-INFO.  This field is a non-empty list of u8vectors denoting
     IP addresses.

     For example:

          > (host-info-addresses (host-info "www.google.com"))
          (#u8(66 249 85 99) #u8(66 249 85 104))


16.12 Service information
=========================

 -- procedure: service-info SERVICE-NAME-OR-ID
     This procedure accesses the service database to get information
     about the service identified by SERVICE-NAME-OR-ID, which is the
     service's symbolic name (string) or the service's port number
     (exact integer).  A service-information record is returned that
     contains the service's symbolic name, a list of aliases
     (alternative names), the port number (exact integer), and the
     protocol name (string).  An exception is raised when
     SERVICE-NAME-OR-ID does not appear in the database.

     For example:

          > (service-info "time")
          #<service-info #2
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (service-info 127)
          #<service-info #3
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (service-info 0)
          *** ERROR IN (console)@3.1 -- Resource temporarily unavailable
          (service-info 0)


 -- procedure: service-info? OBJ
     This procedure returns `#t' when OBJ is a service-information
     record and `#f' otherwise.

     For example:

          > (service-info? (service-info "loopback"))
          #t
          > (service-info? 123)
          #f


 -- procedure: service-info-name SERVICE-INFO
     Returns the symbolic name field of the service-information record
     SERVICE-INFO.

     For example:

          > (service-info-name (service-info 127))
          "loopback"


 -- procedure: service-info-aliases SERVICE-INFO
     Returns the aliases field of the service-information record
     SERVICE-INFO.  This field is a possibly empty list of strings.

     For example:

          > (service-info-aliases (service-info "loopback"))
          ("loopback-net")


 -- procedure: service-info-number SERVICE-INFO
     Returns the service number field of the service-information record
     SERVICE-INFO.

     For example:

          > (service-info-number (service-info "loopback"))
          2130706432


16.13 Protocol information
==========================

 -- procedure: protocol-info PROTOCOL-NAME-OR-ID
     This procedure accesses the protocol database to get information
     about the protocol identified by PROTOCOL-NAME-OR-ID, which is the
     protocol's symbolic name (string) or the protocol's number (exact
     integer).  A protocol-information record is returned that contains
     the protocol's symbolic name, a list of aliases (alternative
     names), and the protocol number (32 bit unsigned exact integer).
     An exception is raised when PROTOCOL-NAME-OR-ID does not appear in
     the database.

     For example:

          > (protocol-info "loopback")
          #<protocol-info #2
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (protocol-info 127)
          #<protocol-info #3
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (protocol-info 0)
          *** ERROR IN (console)@3.1 -- Resource temporarily unavailable
          (protocol-info 0)


 -- procedure: protocol-info? OBJ
     This procedure returns `#t' when OBJ is a protocol-information
     record and `#f' otherwise.

     For example:

          > (protocol-info? (protocol-info "loopback"))
          #t
          > (protocol-info? 123)
          #f


 -- procedure: protocol-info-name PROTOCOL-INFO
     Returns the symbolic name field of the protocol-information record
     PROTOCOL-INFO.

     For example:

          > (protocol-info-name (protocol-info 127))
          "loopback"


 -- procedure: protocol-info-aliases PROTOCOL-INFO
     Returns the aliases field of the protocol-information record
     PROTOCOL-INFO.  This field is a possibly empty list of strings.

     For example:

          > (protocol-info-aliases (protocol-info "loopback"))
          ("loopback-net")


 -- procedure: protocol-info-number PROTOCOL-INFO
     Returns the protocol number field of the protocol-information
     record PROTOCOL-INFO.

     For example:

          > (protocol-info-number (protocol-info "loopback"))
          2130706432


16.14 Network information
=========================

 -- procedure: network-info NETWORK-NAME-OR-ID
     This procedure accesses the network database to get information
     about the network identified by NETWORK-NAME-OR-ID, which is the
     network's symbolic name (string) or the network's number (exact
     integer).  A network-information record is returned that contains
     the network's symbolic name, a list of aliases (alternative
     names), and the network number (32 bit unsigned exact integer).
     An exception is raised when NETWORK-NAME-OR-ID does not appear in
     the database.

     For example:

          > (network-info "loopback")
          #<network-info #2
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (network-info 127)
          #<network-info #3
             name: "loopback"
             aliases: ("loopback-net")
             number: 2130706432>
          > (network-info 0)
          *** ERROR IN (console)@3.1 -- Resource temporarily unavailable
          (network-info 0)


 -- procedure: network-info? OBJ
     This procedure returns `#t' when OBJ is a network-information
     record and `#f' otherwise.

     For example:

          > (network-info? (network-info "loopback"))
          #t
          > (network-info? 123)
          #f


 -- procedure: network-info-name NETWORK-INFO
     Returns the symbolic name field of the network-information record
     NETWORK-INFO.

     For example:

          > (network-info-name (network-info 127))
          "loopback"


 -- procedure: network-info-aliases NETWORK-INFO
     Returns the aliases field of the network-information record
     NETWORK-INFO.  This field is a possibly empty list of strings.

     For example:

          > (network-info-aliases (network-info "loopback"))
          ("loopback-net")


 -- procedure: network-info-number NETWORK-INFO
     Returns the network number field of the network-information record
     NETWORK-INFO.

     For example:

          > (network-info-number (network-info "loopback"))
          2130706432


17 I/O and ports
****************

17.1 Unidirectional and bidirectional ports
===========================================

Unidirectional ports allow communication between a producer of
information and a consumer.  An input-port's producer is typically a
resource managed by the operating system (such as a file, a process or
a network connection) and the consumer is the Scheme program.  The
roles are reversed for an output-port.

   Associated with each port are settings that affect I/O operations on
that port (encoding of characters to bytes, end-of-line encoding, type
of buffering, etc).  Port settings are specified when the port is
created.  Some port settings can be changed after a port is created.

   Bidirectional ports, also called input-output-ports, allow
communication in both directions.  They are best viewed as an object
that groups two separate unidirectional ports (one in each direction).
Each direction has its own port settings and can be closed
independently from the other direction.

17.2 Port classes
=================

The four classes of ports listed below form an inheritance hierarchy.
Operations possible for a certain class of port are also possible for
the subclasses.  Only device-ports are connected to a device managed by
the operating system.  For instance it is possible to create ports that
behave as a FIFO where the Scheme program is both the producer and
consumer of information (possibly one thread is the producer and
another thread is the consumer).

  1. An "object-port" (or simply a port) provides operations to read
     and write Scheme data (i.e. any Scheme object) to/from the port.
     It also provides operations to force output to occur, to change
     the way threads block on the port, and to close the port.  Note
     that the class of objects for which write/read invariance is
     guaranteed depends on the particular class of port.

  2. A "character-port" provides all the operations of an object-port,
     and also operations to read and write individual characters to/from
     the port.  When a Scheme object is written to a character-port, it
     is converted into the sequence of characters that corresponds to
     its external-representation.  When reading a Scheme object, an
     inverse conversion occurs.  Note that some Scheme objects do not
     have an external textual representation that can be read back.

  3. A "byte-port" provides all the operations of a character-port, and
     also operations to read and write individual bytes to/from the
     port.  When a character is written to a byte-port, some encoding
     of that character into a sequence of bytes will occur (for example,
     `#\newline' will be encoded as the 2 bytes CR-LF when using
     ISO-8859-1 character encoding and `cr-lf' end-of-line encoding, and
     a non-ASCII character will generate more than 1 byte when using
     UTF-8 character encoding).  When reading a character, a similar
     decoding occurs.

  4. A "device-port" provides all the operations of a byte-port, and
     also operations to control the operating system managed device
     (file, network connection, terminal, etc) that is connected to the
     port.


17.3 Port settings
==================

Some port settings are only valid for specific port classes whereas
some others are valid for all ports.  Port settings are specified when
a port is created.  The settings that are not specified will default to
some reasonable values.  Keyword objects are used to name the settings
to be set.  As a simple example, a device-port connected to the file
`"foo"' can be created using the call

     (open-input-file "foo")

   This will use default settings for the character encoding, buffering,
etc.  If the UTF-8 character encoding is desired, then the port could
be opened using the call

     (open-input-file (list path: "foo" char-encoding: 'UTF-8))

   Here the argument of the procedure `open-input-file' has been
replaced by a "port settings list" which specifies the value of each
port setting that should not be set to the default value.  Note that
some port settings have no useful default and it is therefore required
to specify a value for them, such as the `path:' in the case of the
file opening procedures.  All port creation procedures (i.e. named
`open-...') take a single argument that can either be a port settings
list or a value of a type that depends on the kind of port being
created (a path string for files, an IP port number for socket servers,
etc).

17.4 Object-ports
=================

17.4.1 Object-port settings
---------------------------

The following is a list of port settings that are valid for all types
of ports.

   * `direction:' ( `input' | `output' | `input-output' )

     This setting controls the direction of the port.  The symbol
     `input' indicates a unidirectional input-port, the symbol `output'
     indicates a unidirectional output-port, and the symbol
     `input-output' indicates a bidirectional port.  The default value
     of this setting depends on the port creation procedure.

   * `buffering:' ( `#f' | `#t' | `line' )

     This setting controls the buffering of the port.  To set each
     direction separately the keywords `input-buffering:' and
     `output-buffering:' must be used instead of `buffering:'.  The
     value `#f' selects unbuffered I/O, the value `#t' selects fully
     buffered I/O, and the symbol `line' selects line buffered I/O (the
     output buffer is drained when a `#\newline' character is written).
     Line buffered I/O only applies to character-ports.  The default
     value of this setting is operating system dependent except
     consoles which are unbuffered.


17.4.2 Object-port operations
-----------------------------

 -- procedure: input-port? OBJ
 -- procedure: output-port? OBJ
 -- procedure: port? OBJ
     The procedure `input-port?' returns `#t' when OBJ is a
     unidirectional input-port or a bidirectional port and `#f'
     otherwise.

     The procedure `output-port?' returns `#t' when OBJ is a
     unidirectional output-port or a bidirectional port and `#f'
     otherwise.

     The procedure `port?' returns `#t' when OBJ is a port (either
     unidirectional or bidirectional) and `#f' otherwise.

     For example:

          > (input-port? (current-input-port))
          #t
          > (call-with-input-string "some text" output-port?)
          #f
          > (port? (current-output-port))
          #t


 -- procedure: read [PORT]
     This procedure reads and returns the next Scheme datum from the
     input-port PORT.  The end-of-file object is returned when the end
     of the stream is reached.  If it is not specified, PORT defaults
     to the current input-port.

     For example:

          > (call-with-input-string "some text" read)
          some
          > (call-with-input-string "" read)
          #!eof


 -- procedure: read-all [PORT [READER]]
     This procedure repeatedly calls the procedure READER with PORT as
     the sole argument and accumulates a list of each value returned up
     to the end-of-file object.  The procedure `read-all' returns the
     accumulated list without the end-of-file object.  If it is not
     specified, PORT defaults to the current input-port.  If it is not
     specified, READER defaults to the procedure `read'.

     For example:

          > (call-with-input-string "3,2,1\ngo!" read-all)
          (3 ,2 ,1 go!)
          > (call-with-input-string "3,2,1\ngo!"
                                    (lambda (p) (read-all p read-char)))
          (#\3 #\, #\2 #\, #\1 #\newline #\g #\o #\!)
          > (call-with-input-string "3,2,1\ngo!"
                                    (lambda (p) (read-all p read-line)))
          ("3,2,1" "go!")


 -- procedure: write OBJ [PORT]
     This procedure writes the Scheme datum OBJ to the output-port PORT
     and the value returned is unspecified.  If it is not specified,
     PORT defaults to the current output-port.

     For example:

          > (write (list 'compare (list 'quote '@x) 'and (list 'unquote '@x)))
          (compare '@x and , @x)>


 -- procedure: newline [PORT]
     This procedure writes an "object separator" to the output-port
     PORT and the value returned is unspecified.  The separator ensures
     that the next Scheme datum written with the `write' procedure will
     not be confused with the latest datum that was written.  On
     character-ports this is done by writing the character `#\newline'.
     On ports where successive objects are implicitly distinct (such
     as "vector ports") this procedure does nothing.

     Regardless of the class of a port P and assuming that the external
     textual representation of the object X is readable, the expression
     `(begin (write X P) (newline P))' will write to P a representation
     of X that can be read back with the procedure `read'.  If it is
     not specified, PORT defaults to the current output-port.

     For example:

          > (begin (write 123) (newline) (write 456) (newline))
          123
          456


 -- procedure: force-output [PORT]
     The procedure `force-output' causes the output buffers of the
     output-port PORT to be drained (i.e. the data is sent to its
     destination).  If PORT is not specified, the current output-port
     is used.

     For example:

          > (define p (open-tcp-client
                        (list server-address: "www.iro.umontreal.ca"
                              port-number: 80)))
          > (display "GET /\n" p)
          > (force-output p)
          > (read-line p)
          "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\""


 -- procedure: close-input-port PORT
 -- procedure: close-output-port PORT
 -- procedure: close-port PORT
     The PORT argument of these procedures must be a unidirectional or
     a bidirectional port.  For all three procedures the value returned
     is unspecified.

     The procedure `close-input-port' closes the input-port side of
     PORT, which must not be a unidirectional output-port.

     The procedure `close-output-port' closes the output-port side of
     PORT, which must not be a unidirectional input-port.  The ouput
     buffers are drained before PORT is closed.

     The procedure `close-port' closes all sides of the PORT.  Unless
     PORT is a unidirectional input-port, the output buffers are
     drained before PORT is closed.

     For example:

          > (define p (open-tcp-client
                        (list server-address: "www.iro.umontreal.ca"
                              port-number: 80)))
          > (display "GET /\n" p)
          > (close-output-port p)
          > (read-line p)
          "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\""


 -- procedure: input-port-timeout-set! PORT TIMEOUT [THUNK]
 -- procedure: output-port-timeout-set! PORT TIMEOUT [THUNK]
     When a thread tries to perform an I/O operation on a port, the
     requested operation may not be immediately possible and the thread
     must wait.  For example, the thread may be trying to read a line of
     text from the console and the user has not typed anything yet, or
     the thread may be trying to write to a network connection faster
     than the network can handle.  In such situations the thread
     normally blocks until the operation becomes possible.

     It is sometimes necessary to guarantee that the thread will not
     block too long.  For this purpose, to each input-port and
     output-port is attached a "timeout" and "timeout-thunk".  The
     timeout indicates the point in time beyond which the thread should
     stop waiting on an input and output operation respectively.  When
     the timeout is reached, the thread calls the port's timeout-thunk.
     If the timeout-thunk returns `#f' the thread abandons trying to
     perform the operation (in the case of an input operation an
     end-of-file is read and in the case of an output operation an
     exception is raised).  Otherwise, the thread will block again
     waiting for the operation to become possible (note that if the
     port's timeout has not changed the thread will immediately call
     the timeout-thunk again).

     The procedure `input-port-timeout-set!' sets the timeout of the
     input-port PORT to TIMEOUT and the timeout-thunk to THUNK.  The
     procedure `output-port-timeout-set!' sets the timeout of the
     output-port PORT to TIMEOUT and the timeout-thunk to THUNK.  If it
     is not specified, the THUNK defaults to a thunk that returns `#f'.
     The TIMEOUT is either a time object indicating an absolute point
     in time, or it is a real number which indicates the number of
     seconds relative to the moment the procedure is called.  For both
     procedures the value returned is unspecified.

     When a port is created the timeout is set to infinity (`+inf.0').
     This causes the thread to wait as long as needed for the operation
     to become possible.  Setting the timeout to a point in the past
     (`-inf.0') will cause the thread to attempt the I/O operation and
     never block (i.e. the timeout-thunk is called if the operation is
     not immediately possible).

     The following example shows how to cause the REPL to terminate
     when the user does not enter an expression within the next 60
     seconds.

          > (input-port-timeout-set! (repl-input-port) 60)
          >
          *** EOF again to exit


17.5 Character-ports
====================

17.5.1 Character-port settings
------------------------------

The following is a list of port settings that are valid for
character-ports.

   * `readtable:' READTABLE

     This setting determines the readtable attached to the
     character-port.  To set each direction separately the keywords
     `input-readtable:' and `output-readtable:' must be used instead of
     `readtable:'.  Readtables control the external textual
     representation of Scheme objects, that is the encoding of Scheme
     objects using characters.  The behavior of the `read' procedure
     depends on the port's input-readtable and the behavior of the
     procedures `write', `pretty-print', and related procedures is
     affected by the port's output-readtable.  The default value of this
     setting is the value bound to the parameter object
     `current-readtable'.

   * `output-width:' POSITIVE-INTEGER

     This setting indicates the width of the character output-port in
     number of characters.  This information is used by the
     pretty-printer.  The default value of this setting is 80.


17.5.2 Character-port operations
--------------------------------

 -- procedure: input-port-line PORT
 -- procedure: input-port-column PORT
 -- procedure: output-port-line PORT
 -- procedure: output-port-column PORT
     The current character location of a character input-port is the
     location of the next character to read.  The current character
     location of a character output-port is the location of the next
     character to write.  Location is denoted by a line number (the
     first line is line 1) and a column number, that is the location on
     the current line (the first column is column 1).  The procedures
     `input-port-line' and `input-port-column' return the line location
     and the column location respectively of the character input-port
     PORT.  The procedures `output-port-line' and `output-port-column'
     return the line location and the column location respectively of
     the character output-port PORT.

     For example:

          > (call-with-output-string
              '()
              (lambda (p)
                (display "abc\n123def" p)
                (write (list (output-port-line p) (output-port-column p))
                       p)))
          "abc\n123def(2 7)"


 -- procedure: output-port-width PORT
     This procedure returns the width, in characters, of the character
     output-port PORT.  The value returned is the port's output-width
     setting.

     For example:

          > (output-port-width (repl-output-port))
          80


 -- procedure: read-char [PORT]
     This procedure reads the character input-port PORT and returns the
     character at the current character location and advances the
     current character location to the next character, unless the PORT
     is already at end-of-file in which case `read-char' returns the
     end-of-file object.  If it is not specified, PORT defaults to the
     current input-port.

     For example:

          > (call-with-input-string
              "some text"
              (lambda (p)
                (let ((a (read-char p))) (list a (read-char p)))))
          (#\s #\o)
          > (call-with-input-string "" read-char)
          #!eof


 -- procedure: peek-char [PORT]
     This procedure returns the same result as `read-char' but it does
     not advance the current character location of the input-port PORT.
     If it is not specified, PORT defaults to the current input-port.

     For example:

          > (call-with-input-string
              "some text"
              (lambda (p)
                (let ((a (peek-char p))) (list a (read-char p)))))
          (#\s #\s)
          > (call-with-input-string "" peek-char)
          #!eof


 -- procedure: write-char CHAR [PORT]
     This procedure writes the character CHAR to the character
     output-port PORT and advances the current character location of
     that output-port.  The value returned is unspecified.  If it is not
     specified, PORT defaults to the current output-port.

     For example:

          > (write-char #\=)
          =>


 -- procedure: read-line [PORT [SEPARATOR [INCLUDE-SEPARATOR?]]]
     This procedure reads characters from the character input-port PORT
     until a specific SEPARATOR or the end-of-file is encountered and
     returns a string containing the sequence of characters read.  The
     SEPARATOR is included at the end of the string only if it was the
     last character read and INCLUDE-SEPARATOR? is not `#f'.  The
     SEPARATOR must be a character or `#f' (in which case all the
     characters until the end-of-file are read).  If it is not
     specified, PORT defaults to the current input-port.  If it is not
     specified, SEPARATOR defaults to `#\newline'.  If it is not
     specified, INCLUDE-SEPARATOR? defaults to `#f'.

     For example:

          > (define (split sep)
              (lambda (str)
                (call-with-input-string
                  str
                  (lambda (p)
                    (read-all p (lambda (p) (read-line p sep)))))))
          > ((split #\,) "a,b,c")
          ("a" "b" "c")
          > (map (split #\,)
                 (call-with-input-string "1,2,3\n4,5"
                                         (lambda (p) (read-all p read-line))))
          (("1" "2" "3") ("4" "5"))


 -- procedure: read-substring STRING START END [PORT]
 -- procedure: write-substring STRING START END [PORT]
     These procedures support bulk character I/O.  The part of the
     string STRING starting at index START and ending just before index
     END is used as a character buffer that will be the target of
     `read-substring' or the source of the `write-substring'.  Up to
     END-START characters will be transferred.  The number of
     characters transferred, possibly zero, is returned by these
     procedures.  Fewer characters will be read by `read-substring' if
     an end-of-file is read, or a timeout occurs before all the
     requested characters are transferred and the timeout thunk returns
     `#f' (see the procedure `input-port-timeout-set!').  Fewer
     characters will be written by `write-substring' if a timeout
     occurs before all the requested characters are transferred and the
     timeout thunk returns `#f' (see the procedure
     `output-port-timeout-set!').  If it is not specified, PORT
     defaults to the current input-port and current output-port
     respectively.

     For example:

          > (define s (make-string 10 #\x))
          > (read-substring s 2 5)123456789
          3
          > 456789
          > s
          "xx123xxxxx"


 -- procedure: input-port-readtable PORT
 -- procedure: output-port-readtable PORT
     These procedures return the readtable attached to the
     character-port PORT.  The PORT parameter of `input-port-readtable'
     must be an input-port.  The PORT parameter of
     `output-port-readtable' must be an output-port.


 -- procedure: input-port-readtable-set! PORT READTABLE
 -- procedure: output-port-readtable-set! PORT READTABLE
     These procedures change the readtable attached to the
     character-port PORT to the readtable READTABLE.  The PORT parameter
     of `input-port-readtable-set!'  must be an input-port.  The PORT
     parameter of `output-port-readtable-set!' must be an output-port.
     The value returned is unspecified.


17.6 Byte-ports
===============

17.6.1 Byte-port settings
-------------------------

The following is a list of port settings that are valid for byte-ports.

   * `char-encoding:' ENCODING

     This setting controls the character encoding of the byte-port.  For
     bidirectional byte-ports, the character encoding for input and
     output is set.  To set each direction separately the keywords
     `input-char-encoding:' and `output-char-encoding:' must be used
     instead of `char-encoding:'.  The default value of this setting is
     operating system dependent, but this can be overriden through the
     runtime options (*note Runtime options::).  The following
     encodings are supported:

    `ISO-8859-1'
          ISO-8859-1 character encoding.  Each character is encoded by
          a single byte.  Only Unicode characters with a code in the
          range 0 to 255 are allowed.

    `ASCII'
          ASCII character encoding.  Each character is encoded by a
          single byte.  In principle only Unicode characters with a
          code in the range 0 to 127 are allowed but most types of
          ports treat this exactly like `ISO-8859-1'.

    `UTF-8'
          UTF-8 character encoding.  Each character is encoded by a
          sequence of one to four bytes.  The minimum length UTF-8
          encoding is used.  If a BOM is needed at the beginning of the
          stream then it must be explicitly written.

    `UTF-16'
          UTF-16 character encoding.  Each character is encoded by one
          or two 16 bit integers (2 or 4 bytes).  The 16 bit integers
          may be encoded using little-endian encoding or big-endian
          encoding.  If the port is an input-port and the first two
          bytes read are a BOM ("Byte Order Mark" character with
          hexadecimal code FEFF) then the BOM will be discarded and the
          endianness will be set accordingly, otherwise the endianness
          depends on the operating system and how the Gambit runtime was
          compiled.  If the port is an output-port then a BOM will be
          output at the beginning of the stream and the endianness
          depends on the operating system and how the Gambit runtime
          was compiled.

    `UTF-16LE'
          UTF-16 character encoding with little-endian endianness.  It
          is like `UTF-16' except the endianness is set to
          little-endian and there is no BOM processing.  If a BOM is
          needed at the beginning of the stream then it must be
          explicitly written.

    `UTF-16BE'
          UTF-16 character encoding with big-endian endianness.  It is
          like `UTF-16LE' except the endianness is set to big-endian.

    `UCS-2'
          UCS-2 character encoding.  Each character is encoded by a 16
          bit integer (2 bytes).  The 16 bit integers may be encoded
          using little-endian encoding or big-endian encoding.  If the
          port is an input-port and the first two bytes read are a BOM
          ("Byte Order Mark" character with hexadecimal code FEFF) then
          the BOM will be discarded and the endianness will be set
          accordingly, otherwise the endianness depends on the
          operating system and how the Gambit runtime was compiled.  If
          the port is an output-port then a BOM will be output at the
          beginning of the stream and the endianness depends on the
          operating system and how the Gambit runtime was compiled.

    `UCS-2LE'
          UCS-2 character encoding with little-endian endianness.  It
          is like `UCS-2' except the endianness is set to little-endian
          and there is no BOM processing.  If a BOM is needed at the
          beginning of the stream then it must be explicitly written.

    `UCS-2BE'
          UCS-2 character encoding with big-endian endianness.  It is
          like `UCS-2LE' except the endianness is set to big-endian.

    `UCS-4'
          UCS-4 character encoding.  Each character is encoded by a 32
          integer (4 bytes).  The 32 bit integers may be encoded using
          little-endian encoding or big-endian encoding.  If the port
          is an input-port and the first four bytes read are a BOM
          ("Byte Order Mark" character with hexadecimal code 0000FEFF)
          then the BOM will be discarded and the endianness will be set
          accordingly, otherwise the endianness depends on the
          operating system and how the Gambit runtime was compiled.  If
          the port is an output-port then a BOM will be output at the
          beginning of the stream and the endianness depends on the
          operating system and how the Gambit runtime was compiled.

    `UCS-4LE'
          UCS-4 character encoding with little-endian endianness.  It
          is like `UCS-4' except the endianness is set to little-endian
          and there is no BOM processing.  If a BOM is needed at the
          beginning of the stream then it must be explicitly written.

    `UCS-4BE'
          UCS-4 character encoding with big-endian endianness.  It is
          like `UCS-4LE' except the endianness is set to big-endian.


   * `eol-encoding:' ENCODING

     This setting controls the end-of-line encoding of the byte-port.
     To set each direction separately the keywords `input-eol-encoding:'
     and `output-eol-encoding:' must be used instead of
     `eol-encoding:'.  The default value of this setting is operating
     system dependent, but this can be overriden through the runtime
     options (*note Runtime options::).  Note that for output-ports the
     end-of-line encoding is applied before the character encoding, and
     for input-ports it is applied after.  The following encodings are
     supported:

    `lf'
          For an output-port, writing a `#\newline' character outputs a
          `#\linefeed' character to the stream (Unicode character code
          10).  For an input-port, a `#\newline' character is read when
          a `#\linefeed' character is encountered on the stream.  Note
          that `#\linefeed' and `#\newline' are two names for the same
          character, so this end-of-line encoding is actually the
          identity function.  Text files created by UNIX applications
          typically use this end-of-line encoding.

    `cr'
          For an output-port, writing a `#\newline' character outputs a
          `#\return' character to the stream (Unicode character code
          10).  For an input-port, a `#\newline' character is read when
          a `#\linefeed' character or a `#\return' character is
          encountered on the stream.  Text files created by Classic Mac
          OS applications typically use this end-of-line encoding.

    `cr-lf'
          For an output-port, writing a `#\newline' character outputs to
          the stream a `#\return' character followed by a `#\linefeed'
          character.  For an input-port, a `#\newline' character is read
          when a `#\linefeed' character or a `#\return' character is
          encountered on the stream.  Moreover, if this character is
          immediately followed by the opposite character (`#\linefeed'
          followed by `#\return' or `#\return' followed by
          `#\linefeed') then the second character is ignored.  In other
          words, all four possible end-of-line encodings are read as a
          single `#\newline' character.  Text files created by DOS and
          Microsoft Windows applications typically use this end-of-line
          encoding.



17.6.2 Byte-port operations
---------------------------

 -- procedure: read-u8 [PORT]
     This procedure reads the byte input-port PORT and returns the byte
     at the current byte location and advances the current byte
     location to the next byte, unless the PORT is already at
     end-of-file in which case `read-u8' returns the end-of-file
     object.  If it is not specified, PORT defaults to the current
     input-port.

     This procedure must be called when the port's input character
     buffer is empty otherwise the character-stream and byte-stream may
     be out of sync due to buffering.  The input character buffer is
     used for bulk decoding of encoded characters (i.e. to translate
     the byte-stream into a character-stream).  The input character
     buffer is initially empty.  It is only when characters are read
     that it is filled with characters obtained by decoding the
     byte-stream.

     One way to ensure that the port's input character buffer is empty
     is to call `read-u8' strictly before any use of the port in a
     character input operation (i.e. a call to the procedures `read',
     `read-char', `peek-char', etc).  Alternatively
     `input-port-characters-buffered' can be used to get the number of
     characters in the port's input character buffer, and to empty the
     buffer with calls to `read-char' or `read-substring'.

     For example:

          > (call-with-input-u8vector
              '#u8(11 22 33 44)
              (lambda (p)
                (let ((a (read-u8 p))) (list a (read-u8 p)))))
          (11 22)
          > (call-with-input-u8vector '#u8() read-u8)
          #!eof


 -- procedure: write-u8 N [PORT]
     This procedure writes the byte N to the byte output-port PORT and
     advances the current byte location of that output-port.  The value
     returned is unspecified.  If it is not specified, PORT defaults to
     the current output-port.

     For example:

          > (call-with-output-u8vector '() (lambda (p) (write-u8 33 p)))
          #u8(33)


 -- procedure: read-subu8vector U8VECTOR START END [PORT]
 -- procedure: write-subu8vector U8VECTOR START END [PORT]
     These procedures support bulk byte I/O.  The part of the u8vector
     U8VECTOR starting at index START and ending just before index END
     is used as a byte buffer that will be the target of
     `read-subu8vector' or the source of the `write-subu8vector'.  Up
     to END-START bytes will be transferred.  The number of bytes
     transferred, possibly zero, is returned by these procedures.
     Fewer bytes will be read by `read-subu8vector' if an end-of-file
     is read, or a timeout occurs before all the requested bytes are
     transferred and the timeout thunk returns `#f' (see the procedure
     `input-port-timeout-set!').  Fewer bytes will be written by
     `write-subu8vector' if a timeout occurs before all the requested
     bytes are transferred and the timeout thunk returns `#f' (see the
     procedure `output-port-timeout-set!').  If it is not specified,
     PORT defaults to the current input-port and current output-port
     respectively.

     The procedure `read-subu8vector' must be called before any use of
     the port in a character input operation (i.e. a call to the
     procedures `read', `read-char', `peek-char', etc) because
     otherwise the character-stream and byte-stream may be out of sync
     due to the port buffering.

     For example:

          > (define v (make-u8vector 10))
          > (read-subu8vector v 2 5)123456789
          3
          > 456789
          > v
          #u8(0 0 49 50 51 0 0 0 0 0)


17.7 Device-ports
=================

17.7.1 Filesystem devices
-------------------------

 -- procedure: open-file PATH-OR-SETTINGS
 -- procedure: open-input-file PATH-OR-SETTINGS
 -- procedure: open-output-file PATH-OR-SETTINGS
 -- procedure: call-with-input-file PATH-OR-SETTINGS PROC
 -- procedure: call-with-output-file PATH-OR-SETTINGS PROC
 -- procedure: with-input-from-file PATH-OR-SETTINGS THUNK
 -- procedure: with-output-to-file PATH-OR-SETTINGS THUNK
     All of these procedures create a port to interface to a byte-stream
     device (such as a file, console, serial port, named pipe, etc)
     whose name is given by a path of the filesystem.  The `direction:'
     setting will default to the value `input' for the procedures
     `open-input-file', `call-with-input-file' and
     `with-input-from-file', to the value `output' for the procedures
     `open-output-file', `call-with-output-file' and
     `with-output-to-file', and to the value `input-output' for the
     procedure `open-file'.  The procedures `open-file',
     `open-input-file' and `open-output-file' return the port that is
     created.  The procedures `call-with-input-file' and
     `call-with-output-file' call the procedure PROC with the port as
     single argument, and then return the value(s) of this call after
     closing the port.  The procedures `with-input-from-file' and
     `with-output-to-file' dynamically bind the current input-port and
     current output-port respectively to the port created for the
     duration of a call to the procedure THUNK with no argument.  The
     value(s) of the call to THUNK are returned after closing the port.

     The first argument of these procedures is either a string denoting
     a filesystem path or a list of port settings which must contain a
     `path:' setting.  Here are the settings allowed in addition to the
     generic settings of byte-ports:

        * `path:' STRING

          This setting indicates the location of the file in the
          filesystem.  There is no default value for this setting.

        * `append:' ( `#f' | `#t' )

          This setting controls whether output will be added to the end
          of the file.  This is useful for writing to log files that
          might be open by more than one process.  The default value of
          this setting is `#f'.

        * `create:' ( `#f' | `#t' | `maybe' )

          This setting controls whether the file will be created when
          it is opened.  A setting of `#f' requires that the file exist
          (otherwise an exception is raised).  A setting of `#t'
          requires that the file does not exist (otherwise an exception
          is raised).  A setting of `maybe' will create the file if it
          does not exist.  The default value of this setting is `maybe'
          for output-ports and `#f' for input-ports and bidirectional
          ports.

        * `permissions:' 12-BIT-EXACT-INTEGER

          This setting controls the UNIX permissions that will be
          attached to the file if it is created.  The default value of
          this setting is `#o666'.

        * `truncate:' ( `#f' | `#t' )

          This setting controls whether the file will be truncated when
          it is opened.  For input-ports, the default value of this
          setting is `#f'.  For output-ports, the default value of this
          setting is `#t' when the `append:' setting is `#f', and `#f'
          otherwise.


     For example:

          > (with-output-to-file
              (list path: "nofile"
                    create: #f)
              (lambda ()
                (display "hello world!\n")))
          *** ERROR IN (console)@1.1 -- No such file or directory
          (with-output-to-file '(path: "nofile" create: #f) '#<procedure #2>)


 -- procedure: input-port-byte-position PORT [POSITION [WHENCE]]
 -- procedure: output-port-byte-position PORT [POSITION [WHENCE]]
     When called with a single argument these procedures return the byte
     position where the next I/O operation would take place in the file
     attached to the given PORT (relative to the beginning of the
     file).  When called with two or three arguments, the byte position
     for subsequent I/O operations on the given PORT is changed to
     POSITION, which must be an exact integer.  When WHENCE is omitted
     or is 0, the POSITION is relative to the beginning of the file.
     When WHENCE is 1, the POSITION is relative to the current byte
     position of the file.  When WHENCE is 2, the POSITION is relative
     to the end of the file.  The return value is the new byte
     position.  On most operating systems the byte position for reading
     and writing of a given bidirectional port are the same.

     When `input-port-byte-position' is called to change the byte
     position of an input-port, all input buffers will be flushed so
     that the next byte read will be the one at the given position.

     When `output-port-byte-position' is called to change the byte
     position of an output-port, there is an implicit call to
     `force-output' before the position is changed.

     For example:

          > (define p  ; p is an input-output-port
              (open-file '(path: "test" char-encoding: ISO-8859-1 create: maybe)))
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (0 0)
          > (display "abcdefghij\n" p)
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (0 0)
          > (force-output p)
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (11 11)
          > (input-port-byte-position p 2)
          2
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (2 2)
          > (peek-char p)
          #\c
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (11 11)
          > (output-port-byte-position p -7 2)
          4
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (4 4)
          > (write-char #\! p)
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (4 4)
          > (force-output p)
          > (list (input-port-byte-position p) (output-port-byte-position p))
          (5 5)
          > (input-port-byte-position p 1)
          1
          > (read p)
          bcd!fghij


17.7.2 Process devices
----------------------

 -- procedure: open-process PATH-OR-SETTINGS
     This procedure starts a new process and returns a port that allows
     communication with that process on its standard input and standard
     output.  The default value of the `direction:' setting is
     `input-output', i.e. the Scheme program can write to the process'
     standard input and can read from the process' standard output.

     The first argument of this procedure is either a string denoting a
     filesystem path of an executable program or a list of port settings
     which must contain a `path:' setting.  Here are the settings
     allowed in addition to the generic settings of byte-ports:

        * `path:' STRING

          This setting indicates the location of the executable program
          in the filesystem.  There is no default value for this
          setting.

        * `arguments:' LIST-OF-STRINGS

          This setting indicates the string arguments that are passed
          to the program.  The default value of this setting is the
          empty list (i.e. no arguments).

        * `environment:' LIST-OF-STRINGS

          This setting indicates the set of environment variable
          bindings that the process receives.  Each element of the list
          is a string of the form "`VAR=VALUE'", where `VAR' is the
          name of the variable and `VALUE' is its binding.  When
          LIST-OF-STRINGS is `#f', the process inherits the environment
          variable bindings of the Scheme program.  The default value
          of this setting is `#f'.

        * `directory:' DIR

          This setting indicates the current working directory of the
          process.  When DIR is `#f', the process uses the value of
          `(current-directory)'.  The default value of this setting is
          `#f'.

        * `stderr-redirection:' ( `#f' | `#t' )

          This setting indicates how the standard error of the process
          is redirected.  A setting of `#t' will redirect the standard
          error to the standard output (i.e. all output to standard
          error can be read from the process-port).  A setting of `#f'
          will leave the standard error as-is, which typically results
          in error messages being output to the console.  The default
          value of this setting is `#f'.

        * `pseudo-terminal:' ( `#f' | `#t' )

          This setting indicates what type of device will be bound to
          the process' standard input and standard output.  A setting
          of `#t' will use a pseudo-terminal device (this is a device
          that behaves like a tty device even though there is no real
          terminal or user directly involved).  A setting of `#f' will
          use a pair of pipes.  The difference is important for
          programs which behave differently when they are used
          interactively, for example shells.  The default value of this
          setting is `#f'.


     For example:

          > (define p (open-process (list path: "ls"
                                          arguments: '("../examples"))))
          > (read-line p)
          "README"
          > (read-line p)
          "Xlib-simple"
          > (close-port p)
          > (define p (open-process "/usr/bin/dc"))
          > (display "2 100 ^ p\n" p)
          > (force-output p)
          > (read-line p)
          "1267650600228229401496703205376"


17.7.3 Network devices
----------------------

 -- procedure: open-tcp-client SETTINGS
     This procedure opens a network connection to a socket server and
     returns a tcp-client-port (a subtype of device-port) that
     represents this connection and allows communication with that
     server.  The default value of the `direction:' setting is
     `input-output', i.e. the Scheme program can send information to
     the server and receive information from the server.  The sending
     direction can be "shutdown" using the `close-output-port'
     procedure and the receiving direction can be "shutdown" using the
     `close-input-port' procedure.  The `close-port' procedure closes
     both directions of the connection.

     The first argument of this procedure is a list of port settings
     which must contain a `server-address:' setting and a
     `port-number:' setting.  Here are the settings allowed in addition
     to the generic settings of byte-ports:

        * `server-address:' STRING-OR-IP-ADDRESS

          This setting indicates the internet address of the server.
          It can be a string denoting a host name, which will be
          translated to an IP address by the `host-info' procedure, or
          a 4 element u8vector which contains the 32-bit IPv4 address
          or an 8 element u16vector which contains the 128-bit IPv6
          address.  There is no default value for this setting.

        * `port-number:' 16-BIT-EXACT-INTEGER

          This setting indicates the IP port-number of the server to
          connect to (e.g. 80 for the standard HTTP server, 23 for the
          standard telnet server).  There is no default value for this
          setting.

        * `keep-alive:' ( `#f' | `#t' )

          This setting controls the use of the "keep alive" option on
          the connection.  The "keep alive" option will periodically
          send control packets on otherwise idle network connections to
          ensure that the server host is active and reachable.  The
          default value of this setting is `#f'.

        * `coalesce:' ( `#f' | `#t' )

          This setting controls the use of TCP's "Nagle algorithm" which
          reduces the number of small packets by delaying their
          transmission and coalescing them into larger packets.  A
          setting of `#t' will coalesce small packets into larger ones.
          A setting of `#f' will transmit packets as soon as possible.
          The default value of this setting is `#t'.  Note that this
          setting does not affect the buffering of the port.


     Here is an example of the client-side code that opens a connection
     to an HTTP server on port 8080 on the same computer (for the
     server-side code see the example for the procedure
     `open-tcp-server'):

          > (define p (open-tcp-client (list server-address: '#u8(127 0 0 1)
                                             port-number: 8080
                                             eol-encoding: 'cr-lf)))
          > p
          #<input-output-port #2 (tcp-client #u8(127 0 0 1) 8080)>
          > (display "GET / HTTP/1.1\n" p)
          > (force-output p)
          > (read-line p)
          "<HTML>"


 -- procedure: open-tcp-server PORT-NUMBER-OR-SETTINGS
     This procedure sets up a socket to accept network connection
     requests from clients and returns a tcp-server-port from which
     network connections to clients are obtained.  Tcp-server-ports are
     a direct subtype of object-ports (i.e. they are not
     character-ports) and are input-ports.  Reading from a
     tcp-server-port with the `read' procedure will block until a
     network connection request is received from a client.  The `read'
     procedure will then return a tcp-client-port (a subtype of
     device-port) that represents this connection and allows
     communication with that client.  Closing a tcp-server-port with
     either the `close-input-port' or `close-port' procedures will
     cause the network subsystem to stop accepting connections on that
     socket.

     The first argument of this procedure is an IP port-number (16-bit
     nonnegative exact integer) or a list of port settings which must
     contain a `port-number:' setting.  Below is a list of the settings
     allowed in addition to the settings `keep-alive:' and `coalesce:'
     allowed by the `open-tcp-client' procedure and the generic
     settings of byte-ports.  The settings which are not listed below
     apply to the tcp-client-port that is returned by `read' when a
     connection is accepted and have the same meaning as if they were
     used in a call to the `open-tcp-client' procedure.

        * `server-address:' STRING-OR-IP-ADDRESS

          This setting indicates the internet address of the network
          interface on which connections requests are accepted.  When
          this parameter is not specified or is `#f', the connection
          requests are accepted on all network interfaces (i.e. address
          INADDR_ANY).  The parameter can be a string denoting a host
          name, which will be translated to an IP address by the
          `host-info' procedure, or a 4 element u8vector which contains
          the 32-bit IPv4 address or an 8 element u16vector which
          contains the 128-bit IPv6 address.

        * `port-number:' 16-BIT-EXACT-INTEGER

          This setting indicates the IP port-number assigned to the
          socket which accepts connection requests from clients.  So
          called "well-known ports", which are reserved for standard
          services, have a port-number below 1024 and can only be
          assigned to a socket by a process with superuser priviledges
          (e.g. 80 for the HTTP service, 23 for the telnet service).
          No special priviledges are needed to assign higher
          port-numbers to a socket.  There is no default value for this
          setting.

        * `backlog:' POSITIVE-EXACT-INTEGER

          This setting indicates the maximum number of connection
          requests that can be waiting to be accepted by a call to
          `read' (technically it is the value passed as the second
          argument of the UNIX `listen()' function).  The default value
          of this setting is 128.

        * `reuse-address:' ( `#f' | `#t' )

          This setting controls whether it is possible to assign a
          port-number that is currently active.  Note that when a
          server process terminates, the socket it was using to accept
          connection requests does not become inactive immediately.
          Instead it remains active for a few minutes to ensure clean
          termination of the connections.  A setting of `#f' will cause
          an exception to be raised in that case.  A setting of `#t'
          will allow a port-number to be used even if it is active.
          The default value of this setting is `#t'.


     Here is an example of the server-side code that accepts
     connections on port 8080 (for the client-side code see the example
     for the procedure `open-tcp-client'):

          > (define s (open-tcp-server (list port-number: 8080
                                             eol-encoding: 'cr-lf)))
          > (define p (read s))  ; blocks until client connects
          > p
          #<input-output-port #2 (tcp-client 8080)>
          > (read-line p)
          "GET / HTTP/1.1"
          > (display "<HTML>\n" p)
          > (force-output p)


17.8 Directory-ports
====================

 -- procedure: open-directory PATH-OR-SETTINGS
     This procedure opens a directory of the filesystem for reading its
     entries and returns a directory-port from which the entries can be
     enumerated.  Directory-ports are a direct subtype of object-ports
     (i.e. they are not character-ports) and are input-ports.  Reading
     from a directory-port with the `read' procedure returns the next
     file name in the directory as a string.  The end-of-file object is
     returned when all the file names have been enumerated.  Another
     way to get the list of all files in a directory is the
     `directory-files' procedure which returns a list of the files in
     the directory.  The advantage of using directory-ports is that it
     allows iterating over the files in a directory in constant space,
     which is interesting when the number of files in the directory is
     not known in advance and may be large.  Note that the order in
     which the names are returned is operating-system dependent.

     The first argument of this procedure is either a string denoting a
     filesystem path to a directory or a list of port settings which
     must contain a `path:' setting.  Here are the settings allowed in
     addition to the generic settings of object-ports:

        * `path:' STRING

          This setting indicates the location of the directory in the
          filesystem.  There is no default value for this setting.

        * `ignore-hidden:' ( `#f' | `#t' | `dot-and-dot-dot' )

          This setting controls whether hidden-files will be returned.
          Under UNIX and Mac OS X hidden-files are those that start
          with a period (such as `.', `..', and `.profile').  Under
          Microsoft Windows hidden files are the `.' and `..' entries
          and the files whose "hidden file" attribute is set.  A
          setting of `#f' will enumerate all the files.  A setting of
          `#t' will only enumerate the files that are not hidden.  A
          setting of `dot-and-dot-dot' will enumerate all the files
          except for the `.' and `..' hidden files.  The default value
          of this setting is `#t'.


     For example:

          > (let ((p (open-directory (list path: "../examples"
                                           ignore-hidden: #f))))
              (let loop ()
                (let ((fn (read p)))
                  (if (string? fn)
                      (begin
                        (pp (path-expand fn))
                        (loop)))))
              (close-input-port p))
          "/u/feeley/examples/."
          "/u/feeley/examples/.."
          "/u/feeley/examples/complex"
          "/u/feeley/examples/README"
          "/u/feeley/examples/simple"
          > (define x (open-directory "../examples"))
          > (read-all x)
          ("complex" "README" "simple")


17.9 Vector-ports
=================

 -- procedure: open-vector [VECTOR-OR-SETTINGS]
 -- procedure: open-input-vector [VECTOR-OR-SETTINGS]
 -- procedure: open-output-vector [VECTOR-OR-SETTINGS]
 -- procedure: call-with-input-vector VECTOR-OR-SETTINGS PROC
 -- procedure: call-with-output-vector VECTOR-OR-SETTINGS PROC
 -- procedure: with-input-from-vector VECTOR-OR-SETTINGS THUNK
 -- procedure: with-output-to-vector VECTOR-OR-SETTINGS THUNK
     Vector-ports represent streams of Scheme objects.  They are a
     direct subtype of object-ports (i.e. they are not
     character-ports).  All of these procedures create vector-ports
     that are either unidirectional or bidirectional.  The `direction:'
     setting will default to the value `input' for the procedures
     `open-input-vector', `call-with-input-vector' and
     `with-input-from-vector', to the value `output' for the procedures
     `open-output-vector', `call-with-output-vector' and
     `with-output-to-vector', and to the value `input-output' for the
     procedure `open-vector'.  Bidirectional vector-ports behave like
     FIFOs: data written to the port is added to the end of the stream
     that is read.  It is only when a bidirectional vector-port's
     output-side is closed with a call to the `close-output-port'
     procedure that the stream's end is known (when the stream's end is
     reached, reading the port returns the end-of-file object).

     The procedures `open-vector', `open-input-vector' and
     `open-output-vector' return the port that is created.  The
     procedures `call-with-input-vector' and `call-with-output-vector'
     call the procedure PROC with the port as single argument, and then
     return the value(s) of this call after closing the port.  The
     procedures `with-input-from-vector' and `with-output-to-vector'
     dynamically bind the current input-port and current output-port
     respectively to the port created for the duration of a call to the
     procedure THUNK with no argument.  The value(s) of the call to
     THUNK are returned after closing the port.

     The first argument of these procedures is either a vector of the
     elements used to initialize the stream or a list of port settings.
     If it is not specified, the argument of the `open-vector',
     `open-input-vector', and `open-output-vector' procedures defaults
     to an empty list of port settings.  Here are the settings allowed
     in addition to the generic settings of object-ports:

        * `init:' VECTOR

          This setting indicates the initial content of the stream.
          The default value of this setting is an empty vector.

        * `permanent-close:' ( `#f' | `#t' )

          This setting controls whether a call to the procedures
          `close-output-port' will close the output-side of a
          bidirectional vector-port permanently or not.  A permanently
          closed bidirectional vector-port whose end-of-file has been
          reached on the input-side will return the end-of-file object
          for all subsequent calls to the `read' procedure.  A
          non-permanently closed bidirectional vector-port will return
          to its opened state when its end-of-file is read.  The
          default value of this setting is `#t'.


     For example:

          > (define p (open-vector))
          > (write 1 p)
          > (write 2 p)
          > (write 3 p)
          > (read p)
          1
          > (read p)
          2
          > (close-output-port p)
          > (read p)
          3
          > (read p)
          #!eof


 -- procedure: open-vector-pipe [VECTOR-OR-SETTINGS1
          [VECTOR-OR-SETTINGS2]]
     The procedure `open-vector-pipe' creates two vector-ports and
     returns these two ports.  The two ports are interrelated as
     follows: the first port's output-side is connected to the second
     port's input-side and the first port's input-side is connected to
     the second port's output-side.  The value VECTOR-OR-SETTINGS1 is
     used to setup the first vector-port and VECTOR-OR-SETTINGS2 is
     used to setup the second vector-port.  The same settings as for
     `open-vector' are allowed.  The default `direction:' setting is
     `input-output' (i.e. a bidirectional port is created).  If it is
     not specified VECTOR-OR-SETTINGS1 defaults to the empty list.  If
     it is not specified VECTOR-OR-SETTINGS2 defaults to
     VECTOR-OR-SETTINGS1 but with the `init:' setting set to the empty
     vector and with the input and output settings exchanged (e.g. if
     the first port is an input-port then the second port is an
     output-port, if the first port's input-side is non-buffered then
     the second port's output-side is non-buffered).

     For example:

          > (define (server op)
              (receive (c s) (open-vector-pipe)  ; client-side and server-side ports
                (thread-start!
                  (make-thread
                    (lambda ()
                      (let loop ()
                        (let ((request (read s)))
                          (if (not (eof-object? request))
                              (begin
                                (write (op request) s)
                                (newline s)
                                (force-output s)
                                (loop))))))))
                c))
          > (define a (server (lambda (x) (expt 2 x))))
          > (define b (server (lambda (x) (expt 10 x))))
          > (write 100 a)
          > (write 30 b)
          > (read a)
          1267650600228229401496703205376
          > (read b)
          1000000000000000000000000000000


 -- procedure: get-output-vector VECTOR-PORT
     The procedure `get-output-vector' takes an output vector-port or a
     bidirectional vector-port as argument and removes all the objects
     currently on the output-side, returning them in a vector.  The port
     remains open and subsequent output to the port and calls to the
     procedure `get-output-vector' are possible.

     For example:

          > (define p (open-vector '#(1 2 3)))
          > (write 4 p)
          > (get-output-vector p)
          #(1 2 3 4)
          > (write 5 p)
          > (write 6 p)
          > (get-output-vector p)
          #(5 6)


17.10 String-ports
==================

 -- procedure: open-string [STRING-OR-SETTINGS]
 -- procedure: open-input-string [STRING-OR-SETTINGS]
 -- procedure: open-output-string [STRING-OR-SETTINGS]
 -- procedure: call-with-input-string STRING-OR-SETTINGS PROC
 -- procedure: call-with-output-string STRING-OR-SETTINGS PROC
 -- procedure: with-input-from-string STRING-OR-SETTINGS THUNK
 -- procedure: with-output-to-string STRING-OR-SETTINGS THUNK
 -- procedure: open-string-pipe [STRING-OR-SETTINGS1
          [STRING-OR-SETTINGS2]]
 -- procedure: get-output-string STRING-PORT
     String-ports represent streams of characters.  They are a direct
     subtype of character-ports.  These procedures are the string-port
     analog of the procedures specified in the vector-ports section.
     Note that these procedures are a superset of the procedures
     specified in the "Basic String Ports SRFI" (SRFI 6).


 -- procedure: object->string OBJ [N]
     This procedure converts the object OBJ to its external
     representation and returns it in a string.  The parameter N
     specifies the maximal width of the resulting string.  If the
     external representation is wider than N, the resulting string will
     be truncated to N characters and the last 3 characters will be set
     to periods.  Note that the current readtable is used.


17.11 U8vector-ports
====================

 -- procedure: open-u8vector [U8VECTOR-OR-SETTINGS]
 -- procedure: open-input-u8vector [U8VECTOR-OR-SETTINGS]
 -- procedure: open-output-u8vector [U8VECTOR-OR-SETTINGS]
 -- procedure: call-with-input-u8vector U8VECTOR-OR-SETTINGS PROC
 -- procedure: call-with-output-u8vector U8VECTOR-OR-SETTINGS PROC
 -- procedure: with-input-from-u8vector U8VECTOR-OR-SETTINGS THUNK
 -- procedure: with-output-to-u8vector U8VECTOR-OR-SETTINGS THUNK
 -- procedure: open-u8vector-pipe [U8VECTOR-OR-SETTINGS1
          [U8VECTOR-OR-SETTINGS2]]
 -- procedure: get-output-u8vector U8VECTOR-PORT
     U8vector-ports represent streams of bytes.  They are a direct
     subtype of byte-ports.  These procedures are the u8vector-port
     analog of the procedures specified in the vector-ports section.


17.12 Parameter objects related to I/O
======================================

 -- procedure: current-input-port [NEW-VALUE]
 -- procedure: current-output-port [NEW-VALUE]
 -- procedure: current-error-port [NEW-VALUE]
 -- procedure: current-readtable [NEW-VALUE]
     These procedures are parameter objects which represent
     respectively: the current input-port, the current output-port, the
     current error-port, and the current readtable.


18 Lexical syntax and readtables
********************************

18.1 Readtables
===============

Readtables control the external textual representation of Scheme
objects, that is the encoding of Scheme objects using characters.
Readtables affect the behavior of the reader (i.e. the `read' procedure
and the parser used by the `load' procedure and the interpreter and
compiler) and the printer (i.e. the procedures `write', `display',
`pretty-print', and `pp', and the procedure used by the REPL to print
results).  To preserve write/read invariance the printer and reader
must be using compatible readtables.  For example a symbol which
contains upper case letters will be printed with special escapes if the
readtable indicates that the reader is case-insensitive.

   Readtables are immutable records whose fields specify various textual
representation aspects.  There are accessor procedures to retrieve the
content of specific fields.  There are also functional update
procedures that create a copy of a readtable, with a specific field set
to a new value.

 -- procedure: readtable? OBJ
     This procedure returns `#t' when OBJ is a readtable and `#f'
     otherwise.

     For example:

          > (readtable? (current-readtable))
          #t
          > (readtable? 123)
          #f


 -- procedure: readtable-case-conversion? READTABLE
 -- procedure: readtable-case-conversion?-set READTABLE NEW-VALUE
     The procedure `readtable-case-conversion?' returns the content of
     the `case-conversion?' field of READTABLE.  When the content of
     this field is `#f', the reader preserves the case of symbols and
     keyword objects that are read (i.e. `Ice' and `ice' are distinct
     symbols).  When the content of this field is the symbol `upcase',
     the reader converts lowercase letters to uppercase when reading
     symbols and keywords (i.e. `Ice' is read as the symbol
     `(string->symbol "ICE")').  Otherwise the reader converts
     uppercase letters to lowercase when reading symbols and keywords
     (i.e. `Ice' is read as the symbol `(string->symbol "ice")').

     The procedure `readtable-case-conversion?-set' returns a copy of
     READTABLE where only the `case-conversion?' field has been changed
     to NEW-VALUE.

     For example:

          > (output-port-readtable-set!
              (repl-output-port)
              (readtable-case-conversion?-set
                (output-port-readtable (repl-output-port))
                #f))
          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-case-conversion?-set
                (input-port-readtable (repl-input-port))
                #f))
          > 'Ice
          Ice
          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-case-conversion?-set
                (input-port-readtable (repl-input-port))
                #t))
          > 'Ice
          ice
          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-case-conversion?-set
                (input-port-readtable (repl-input-port))
                'upcase))
          > 'Ice
          ICE


 -- procedure: readtable-keywords-allowed? READTABLE
 -- procedure: readtable-keywords-allowed?-set READTABLE NEW-VALUE
     The procedure `readtable-keywords-allowed?' returns the content of
     the `keywords-allowed?' field of READTABLE.  When the content of
     this field is `#f', the reader does not recognize keyword objects
     (i.e. `:foo' and `foo:' are read as the symbols `(string->symbol
     ":foo")' and `(string->symbol "foo:")' respectively).  When the
     content of this field is the symbol `prefix', the reader
     recognizes keyword objects that start with a colon, as in Common
     Lisp (i.e. `:foo' is read as the keyword `(string->keyword
     "foo")').  Otherwise the reader recognizes keyword objects that
     end with a colon, as in DSSSL (i.e. `foo:' is read as the symbol
     `(string->symbol "foo")').

     The procedure `readtable-keywords-allowed?-set' returns a copy of
     READTABLE where only the `keywords-allowed?' field has been
     changed to NEW-VALUE.

     For example:

          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-keywords-allowed?-set
                (input-port-readtable (repl-input-port))
                #f))
          > (map keyword? '(foo :foo foo:))
          (#f #f #f)
          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-keywords-allowed?-set
                (input-port-readtable (repl-input-port))
                #t))
          > (map keyword? '(foo :foo foo:))
          (#f #f #t)
          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-keywords-allowed?-set
                (input-port-readtable (repl-input-port))
                'prefix))
          > (map keyword? '(foo :foo foo:))
          (#f #t #f)


 -- procedure: readtable-sharing-allowed? READTABLE
 -- procedure: readtable-sharing-allowed?-set READTABLE NEW-VALUE
     The procedure `readtable-sharing-allowed?' returns the content of
     the `sharing-allowed?' field of READTABLE.  The reader recognizes
     the `#N#' and `#N=DATUM' notation for circular structures and the
     printer uses this notation if and only if the content of the
     `sharing-allowed?' field is not `#f'.  Moreover when the content
     of the `sharing-allowed?' field is the symbol `serialize', the
     printer uses a special external representation that the reader
     understands and that extends write/read invariance to the
     following types: records, procedures and continuations.  Note that
     an object can be serialized and deserialized if and only if all of
     its components are serializable.

     The procedure `readtable-sharing-allowed?-set' returns a copy of
     READTABLE where only the `sharing-allowed?' field has been changed
     to NEW-VALUE.

     Here is a simple example:

          > (define (wr obj allow?)
              (call-with-output-string
                '()
                (lambda (p)
                  (output-port-readtable-set!
                    p
                    (readtable-sharing-allowed?-set
                      (output-port-readtable p)
                      allow?))
                  (write obj p))))
          > (define (rd str allow?)
              (call-with-input-string
                str
                (lambda (p)
                  (input-port-readtable-set!
                    p
                    (readtable-sharing-allowed?-set
                      (input-port-readtable p)
                      allow?))
                  (read p))))
          > (define x (list 1 2 3))
          > (set-car! (cdr x) (cddr x))
          > (wr x #f)
          "(1 (3) 3)"
          > (wr x #t)
          "(1 #0=(3) . #0#)"
          > (define y (rd (wr x #t) #t))
          > y
          (1 (3) 3)
          > (eq? (cadr y) (cddr y))
          #t
          > (define f #f)
          > (let ((free (expt 2 10)))
            (set! f (lambda (x) (+ x free))))
          > (define s (wr f 'serialize))
          > (string-length s)
          4196
          > (define g (rd s 'serialize))
          > (eq? f g)
          #f
          > (g 4)
          1028

     Continuations are tricky to serialize because they contain a
     dynamic environment and this dynamic environment may contain
     non-serializable objects, in particular ports attached to
     operating-system streams such as files, the console or standard
     input/output.  Indeed, all dynamic environments contain a binding
     for the `current-input-port' and `current-output-port'.  Moreover,
     any thread that has started a REPL has a continuation which refers
     to the "repl-context" object in its dynamic environment.  A
     repl-context object contains the interaction channel, which is
     typically connected to a non-serializable port, such as the
     console.  Another problem is that the `parameterize' form saves
     the old binding of the parameter in the continuation, so it is not
     possible to eliminate the references to these ports in the
     continuation by using the `parameterize' form alone.

     Serialization of continuations can be achieved dependably by taking
     advantage of string-ports, which are serializable objects (unless
     there is a blocked thread), and the following features of threads:
     they inherit the dynamic environment of the parent thread and they
     start with an initial continuation that contains only serializable
     objects.  So a thread created in a dynamic environment where
     `current-input-port' and `current-output-port' are bound to a
     dummy string-port has a serializable continuation.

     Here is an example where continuations are serialized:

          > (define (wr obj)
              (call-with-output-string
               '()
               (lambda (p)
                 (output-port-readtable-set!
                  p
                  (readtable-sharing-allowed?-set
                   (output-port-readtable p)
                   'serialize))
                 (write obj p))))
          > (define (rd str)
              (call-with-input-string
               str
               (lambda (p)
                 (input-port-readtable-set!
                  p
                  (readtable-sharing-allowed?-set
                   (input-port-readtable p)
                   'serialize))
                 (read p))))
          > (define fifo (open-vector))
          > (define (suspend-and-die!)
              (call-with-current-continuation
               (lambda (k)
                 (write (wr k) fifo)
                 (newline fifo)
                 (force-output fifo)
                 (thread-terminate! (current-thread)))))
          > (let ((dummy-port (open-string)))
              (parameterize ((current-input-port dummy-port)
                             (current-output-port dummy-port))
                (thread-start!
                 (make-thread
                  (lambda ()
                    (* 100
                       (suspend-and-die!)))))))
          #<thread #2>
          > (define s (read fifo))
          > (thread-join!
              (thread-start!
                (make-thread
                  (lambda ()
                    ((rd s) 111)))))
          11100
          > (thread-join!
              (thread-start!
                (make-thread
                  (lambda ()
                    ((rd s) 222)))))
          22200
          > (string-length s)
          13114


 -- procedure: readtable-eval-allowed? READTABLE
 -- procedure: readtable-eval-allowed?-set READTABLE NEW-VALUE
     The procedure `readtable-eval-allowed?' returns the content of the
     `eval-allowed?' field of READTABLE.  The reader recognizes the
     `#.EXPRESSION' notation for read-time evaluation if and only if
     the content of the `eval-allowed?' field is not `#f'.

     The procedure `readtable-eval-allowed?-set' returns a copy of
     READTABLE where only the `eval-allowed?' field has been changed to
     NEW-VALUE.

     For example:

          > (input-port-readtable-set!
              (repl-input-port)
              (readtable-eval-allowed?-set
                (input-port-readtable (repl-input-port))
                #t))
          > '(5 plus 7 is #.(+ 5 7))
          (5 plus 7 is 12)
          > '(buf = #.(make-u8vector 25))
          (buf = #u8(0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0))


 -- procedure: readtable-max-write-level READTABLE
 -- procedure: readtable-max-write-level-set READTABLE NEW-VALUE
     The procedure `readtable-max-write-level' returns the content of
     the `max-write-level' field of READTABLE.  The printer will
     display an ellipsis for the elements of lists and vectors that are
     nested deeper than this level.

     The procedure `readtable-max-write-level-set' returns a copy of
     READTABLE where only the `max-write-level' field has been changed
     to NEW-VALUE, which must be an nonnegative fixnum.

     For example:

          > (define (wr obj n)
              (call-with-output-string
                '()
                (lambda (p)
                  (output-port-readtable-set!
                    p
                    (readtable-max-write-level-set
                      (output-port-readtable p)
                      n))
                  (write obj p))))
          > (wr '(a #(b (c c) #u8(9 9 9) b) a) 3)
          "(a #(b (c c) #u8(9 9 9) b) a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) a) 2)
          "(a #(b (...) #u8(...) b) a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) a) 1)
          "(a #(...) a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) a) 0)
          "(...)"
          > (wr 'hello 0)
          "hello"


 -- procedure: readtable-max-write-length READTABLE
 -- procedure: readtable-max-write-length-set READTABLE NEW-VALUE
     The procedure `readtable-max-write-length' returns the content of
     the `max-write-length' field of READTABLE.  The printer will
     display an ellipsis for the elements of lists and vectors that are
     at an index beyond that length.

     The procedure `readtable-max-write-length-set' returns a copy of
     READTABLE where only the `max-write-length' field has been changed
     to NEW-VALUE, which must be an nonnegative fixnum.

     For example:

          > (define (wr obj n)
              (call-with-output-string
                '()
                (lambda (p)
                  (output-port-readtable-set!
                    p
                    (readtable-max-write-length-set
                      (output-port-readtable p)
                      n))
                  (write obj p))))
          > (wr '(a #(b (c c) #u8(9 9 9) b) . a) 4)
          "(a #(b (c c) #u8(9 9 9) b) . a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) . a) 3)
          "(a #(b (c c) #u8(9 9 9) ...) . a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) . a) 2)
          "(a #(b (c c) ...) . a)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) . a) 1)
          "(a ...)"
          > (wr '(a #(b (c c) #u8(9 9 9) b) . a) 0)
          "(...)"


 -- procedure: readtable-start-syntax READTABLE
 -- procedure: readtable-start-syntax-set READTABLE NEW-VALUE
     The procedure `readtable-start-syntax' returns the content of the
     `start-syntax' field of READTABLE.  The reader uses this field to
     determine in which syntax to start parsing the input.  When the
     content of this field is the symbol `six', the reader starts in
     the infix syntax.  Otherwise the reader starts in the prefix
     syntax.

     The procedure `readtable-start-syntax-set' returns a copy of
     READTABLE where only the `start-syntax' field has been changed to
     NEW-VALUE.

     For example:

          > (+ 2 3)
          5
          > (input-port-readtable-set!
             (repl-input-port)
             (readtable-start-syntax-set
               (input-port-readtable (repl-input-port))
               'six))
          > 2+3;
          5
          > exit();


18.2 Boolean syntax
===================

Booleans are required to be followed by a delimiter (i.e. `#f64()' is
not the boolean `#f' followed by the number `64' and the empty list).

18.3 Character syntax
=====================

Characters are required to be followed by a delimiter (i.e.
`#\spaceballs' is not the character `#\space' followed by the symbol
`balls').  The lexical syntax of characters is extended to allow the
following:

`#\nul'
     Unicode character 0

`#\alarm'
     Unicode character 7

`#\backspace'
     Unicode character 8

`#\tab'
     Unicode character 9

`#\newline'
     Unicode character 10 (newline character)

`#\linefeed'
     Unicode character 10

`#\vtab'
     Unicode character 11

`#\page'
     Unicode character 12

`#\return'
     Unicode character 13

`#\space'
     Unicode character 32 (space character)

`#\rubout'
     Unicode character 127

`#\xHH'
     character encoded in hexadecimal (>= 1 hexadecimal digit)

`#\uHHHH'
     character encoded in hexadecimal (exactly 4 hexadecimal digits)

`#\UHHHHHHHH'
     character encoded in hexadecimal (exactly 8 hexadecimal digits)

18.4 String syntax
==================

The lexical syntax of quoted strings is extended to allow the following
escape codes:

`\a'
     Unicode character 7

`\b'
     Unicode character 8

`\t'
     Unicode character 9

`\n'
     Unicode character 10 (newline character)

`\v'
     Unicode character 11

`\f'
     Unicode character 12

`\r'
     Unicode character 13

`\"'
     `"'

`\\'
     `\'

`\|'
     `|'

`\?'
     `?'

`\OOO'
     character encoded in octal (1 to 3 octal digits, first digit must
     be less than 4 when there are 3 octal digits)

`\xHH'
     character encoded in hexadecimal (>= 1 hexadecimal digit)

`\uHHHH'
     character encoded in hexadecimal (exactly 4 hexadecimal digits)

`\UHHHHHHHH'
     character encoded in hexadecimal (exactly 8 hexadecimal digits)

`\<space>'
     Unicode character 32 (space character)

`\<newline><whitespace-except-newline>*'
     This sequence expands to nothing (it is useful for splitting a long
     string literal on multiple lines while respecting proper
     indentation of the source code)

   Gambit also supports a "here string" syntax that is similar to shell
"here documents".  For example:

     > (pretty-print #<<THE-END
     hello
     world
     THE-END
     )
     "hello\nworld"

   The here string starts with the sequence `#<<'.  The part of the
line after the `#<<' up to and including the newline character is the
key. The first line afterward that matches the key marks the end of the
here string.  The string contains all the characters between the start
key and the end key, with the exception of the newline character before
the end key.

18.5 Symbol syntax
==================

The lexical syntax of symbols is extended to allow a leading and
trailing vertical bar (e.g. `|a\|b"c:|').  The symbol's name
corresponds verbatim to the characters between the vertical bars except
for escaped characters.  The same escape sequences as for strings are
permitted except that `"' does not need to be escaped and `|' needs to
be escaped.

   For example:

     > (symbol->string '|a\|b"c:|)
     "a|b\"c:"

18.6 Keyword syntax
===================

The lexical syntax of keywords is like symbols, but with a colon at the
end (note that this can be changed to a leading colon by setting the
`keywords-allowed?' field of the readtable to the symbol `prefix').  A
colon by itself is not a keyword, it is a symbol.  Vertical bars can be
used like symbols but the colon must be outside the vertical bars.
Note that the string returned by the `keyword->string' procedure does
not include the colon.

   For example:

     > (keyword->string 'foo:)
     "foo"
     > (map keyword? '(|ab()cd:| |ab()cd|: : ||:))
     (#f #t #f #t)

18.7 Box syntax
===============

The lexical syntax of boxes is `#&OBJ' where OBJ is the content of the
box.

   For example:

     > (list '#&"hello" '#&123)
     (#&"hello" #&123)
     > (box (box (+ 10 20)))
     #&#&30

18.8 Number syntax
==================

The lexical syntax of the special inexact real numbers is as follows:

`+inf.0'
     positive infinity

`-inf.0'
     negative infinity

`+nan.0'
     "not a number"

`-0.'
     negative zero (`0.' is the positive zero)

18.9 Homogeneous vector syntax
==============================

Homogeneous vectors are vectors containing raw numbers of the same type
(signed or unsigned exact integers or inexact reals).  There are 10
types of homogeneous vectors: `s8vector' (vector of 8 bit signed
integers), `u8vector' (vector of 8 bit unsigned integers), `s16vector'
(vector of 16 bit signed integers), `u16vector' (vector of 16 bit
unsigned integers), `s32vector' (vector of 32 bit signed integers),
`u32vector' (vector of 32 bit unsigned integers), `s64vector' (vector
of 64 bit signed integers), `u64vector' (vector of 64 bit unsigned
integers), `f32vector' (vector of 32 bit floating point numbers), and
`f64vector' (vector of 64 bit floating point numbers).

   The external representation of homogeneous vectors is similar to
normal vectors but with the `#(' prefix replaced respectively with
`#s8(', `#u8(', `#s16(', `#u16(', `#s32(', `#u32(', `#s64(', `#u64(',
`#f32(', and `#f64('.

   The elements of the integer homogeneous vectors must be exact
integers fitting in the given precision.  The elements of the floating
point homogeneous vectors must be inexact reals.

18.10 Special `#!' syntax
=========================

The lexical syntax of the special `#!' objects is as follows:

`#!eof'
     end-of-file object

`#!void'
     void object

`#!optional'
     optional object

`#!rest'
     rest object

`#!key'
     key object

18.11 Multiline comment syntax
==============================

Multiline comments are delimited by the tokens `#|' and `|#'.  These
comments can be nested.

18.12 Scheme infix syntax extension
===================================

The reader supports an infix syntax extension which is called SIX
(Scheme Infix eXtension).  This extension is both supported by the
`read' procedure and in program source code.

   The backslash character is a delimiter that marks the beginning of a
single datum expressed in the infix syntax (the details are given
below).  One way to think about it is that the backslash character
escapes the prefix syntax temporarily to use the infix syntax.  For
example a three element list could be written as `(X \Y Z)', the
elements X and Z are expressed using the normal prefix syntax and Y is
expressed using the infix syntax.

   When the reader encounters an infix datum, it constructs a syntax
tree for that particular datum.  Each node of this tree is represented
with a list whose first element is a symbol indicating the type of node.
For example, `(six.identifier abc)' is the representation of the infix
identifier `abc' and `(six.index (six.identifier abc) (six.identifier
i))' is the representation of the infix datum `abc[i];'.

18.12.1 SIX grammar
-------------------

The SIX grammar is given below.  On the left hand side are the
production rules.  On the right hand side is the datum that is
constructed by the reader.  The notation $I denotes the datum that is
constructed by the reader for the Ith part of the production rule.

<infix datum> ::=                           
<stat>                                      $1
<stat> ::=                                  
<if stat>                                   $1
| <for stat>                                $1
| <while stat>                              $1
| <do stat>                                 $1
| <switch stat>                             $1
| <case stat>                               $1
| <break stat>                              $1
| <continue stat>                           $1
| <label stat>                              $1
| <goto stat>                               $1
| <return stat>                             $1
| <expression stat>                         $1
| <procedure definition>                    $1
| <variable definition> `;'                 $1
| <clause stat>                             $1
| <compound stat>                           $1
| `;'                                       `(six.compound)'
<if stat> ::=                               
`if' `(' <pexpr> `)' <stat>                 `(six.if $3 $5)'
| `if' `(' <pexpr> `)' <stat> `else'        `(six.if $3 $5 $7)'
<stat>                                      
<for stat> ::=                              
`for' `(' <stat> `;' <oexpr> `;' <oexpr>    `(six.for $3 $5 $7 $9)'
`)' <stat>                                  
<while stat> ::=                            
`while' `(' <pexpr> `)' <stat>              `(six.while $3 $5)'
<do stat> ::=                               
`do' <stat> `while' `(' <pexpr> `)' `;'     `(six.do-while $2 $5)'
<switch stat> ::=                           
`switch' `(' <pexpr> `)' <stat>             `(six.switch $3 $5)'
<case stat> ::=                             
`case' <expr> `:' <stat>                    `(six.case $2 $4)'
<break stat> ::=                            
`break' `;'                                 `(six.break)'
<continue stat> ::=                         
`continue' `;'                              `(six.continue)'
<label stat> ::=                            
<identifier> `:' <stat>                     `(six.label $1 $3)'
<goto stat> ::=                             
`goto' <expr> `;'                           `(six.goto $2)'
<return stat> ::=                           
`return' `;'                                `(six.return)'
| `return' <expr> `;'                       `(six.return $2)'
<expression stat> ::=                       
<expr> `;'                                  $1
<clause stat> ::=                           
<expr> `.'                                  `(six.clause $1)'
<pexpr> ::=                                 
<procedure definition>                      $1
| <variable definition>                     $1
| <expr>                                    $1
<procedure definition> ::=                  
<type> <id-or-prefix> `(' <parameters> `)'  `(six.define-procedure $2
<body>                                      (six.procedure $1 $4 $6))'
<variable definition> ::=                   
<type> <id-or-prefix> <dimensions> <iexpr>  `(six.define-variable $2 $1
                                            $3 $4)'
<iexpr> ::=                                 
`=' <expr>                                  $2
|                                           `#f'
<dimensions> ::=                            
| `[' <expr> `]' <dimensions>               `($2 . $4)'
|                                           `()'
<oexpr> ::=                                 
<expr>                                      $1
|                                           `#f'
<expr> ::=                                  
<expr18>                                    $1
<expr18> ::=                                
<expr17> `:-' <expr18>                      `(six.x:-y $1 $3)'
| <expr17>                                  $1
<expr17> ::=                                
<expr17> `,' <expr16>                       `(|six.x,y| $1 $3)'
| <expr16>                                  $1
<expr16> ::=                                
<expr15> `:=' <expr16>                      `(six.x:=y $1 $3)'
| <expr15>                                  $1
<expr15> ::=                                
<expr14> `%=' <expr15>                      `(six.x%=y $1 $3)'
| <expr14> `&=' <expr15>                    `(six.x&=y $1 $3)'
| <expr14> `*=' <expr15>                    `(six.x*=y $1 $3)'
| <expr14> `+=' <expr15>                    `(six.x+=y $1 $3)'
| <expr14> `-=' <expr15>                    `(six.x-=y $1 $3)'
| <expr14> `/=' <expr15>                    `(six.x/=y $1 $3)'
| <expr14> `<<=' <expr15>                   `(six.x<<=y $1 $3)'
| <expr14> `=' <expr15>                     `(six.x=y $1 $3)'
| <expr14> `>>=' <expr15>                   `(six.x>>=y $1 $3)'
| <expr14> `^=' <expr15>                    `(six.x^=y $1 $3)'
| <expr14> `|=' <expr15>                    `(|six.x\|=y| $1 $3)'
| <expr14>                                  $1
<expr14> ::=                                
<expr13> `:' <expr14>                       `(six.x:y $1 $3)'
| <expr13>                                  $1
<expr13> ::=                                
<expr12> `?' <expr> `:' <expr13>            `(six.x?y:z $1 $3 $5)'
| <expr12>                                  $1
<expr12> ::=                                
<expr12> `||' <expr11>                      `(|six.x\|\|y| $1 $3)'
| <expr11>                                  $1
<expr11> ::=                                
<expr11> `&&' <expr10>                      `(six.x&&y $1 $3)'
| <expr10>                                  $1
<expr10> ::=                                
<expr10> `|' <expr9>                        `(|six.x\|y| $1 $3)'
| <expr9>                                   $1
<expr9> ::=                                 
<expr9> `^' <expr8>                         `(six.x^y $1 $3)'
| <expr8>                                   $1
<expr8> ::=                                 
<expr8> `&' <expr7>                         `(six.x&y $1 $3)'
| <expr7>                                   $1
<expr7> ::=                                 
<expr7> `!=' <expr6>                        `(six.x!=y $1 $3)'
| <expr7> `==' <expr6>                      `(six.x==y $1 $3)'
| <expr6>                                   $1
<expr6> ::=                                 
<expr6> `<' <expr5>                         `(six.x<y $1 $3)'
| <expr6> `<=' <expr5>                      `(six.x<=y $1 $3)'
| <expr6> `>' <expr5>                       `(six.x>y $1 $3)'
| <expr6> `>=' <expr5>                      `(six.x>=y $1 $3)'
| <expr5>                                   $1
<expr5> ::=                                 
<expr5> `<<' <expr4>                        `(six.x<<y $1 $3)'
| <expr5> `>>' <expr4>                      `(six.x>>y $1 $3)'
| <expr4>                                   $1
<expr4> ::=                                 
<expr4> `+' <expr3>                         `(six.x+y $1 $3)'
| <expr4> `-' <expr3>                       `(six.x-y $1 $3)'
| <expr3>                                   $1
<expr3> ::=                                 
<expr3> `%' <expr2>                         `(six.x%y $1 $3)'
| <expr3> `*' <expr2>                       `(six.x*y $1 $3)'
| <expr3> `/' <expr2>                       `(six.x/y $1 $3)'
| <expr2>                                   $1
<expr2> ::=                                 
`&' <expr2>                                 `(six.&x $2)'
| `+' <expr2>                               `(six.+x $2)'
| `-' <expr2>                               `(six.-x $2)'
| `*' <expr2>                               `(six.*x $2)'
| `!' <expr2>                               `(six.!x $2)'
| `!'                                       `(six.!)'
| `++' <expr2>                              `(six.++x $2)'
| `--' <expr2>                              `(six.--x $2)'
| `~' <expr2>                               `(six.~x $2)'
| `new' <id-or-prefix> `(' <arguments> `)'  `(six.new $2 . $4)'
| <expr1>                                   $1
<expr1> ::=                                 
<expr1> `++'                                `(six.x++ $1)'
| <expr1> `--'                              `(six.x-- $1)'
| <expr1> `(' <arguments> `)'               `(six.call $1 . $3)'
| <expr1> `[' <expr> `]'                    `(six.index $1 $3)'
| <expr1> `->' <id-or-prefix>               `(six.arrow $1 $3)'
| <expr1> `.' <id-or-prefix>                `(six.dot $1 $3)'
| <expr0>                                   $1
<expr0> ::=                                 
<id-or-prefix>                              $1
| <string>                                  `(six.literal $1)'
| <char>                                    `(six.literal $1)'
| <number>                                  `(six.literal $1)'
| `(' <expr> `)'                            $2
| `(' <block stat> `)'                      $2
| <datum-starting-with-#-or-backquote>      `(six.prefix $1)'
| `[' <elements> `]'                        $2
| <type> `(' <parameters> `)' <body>        `(six.procedure $1 $3 $5)'
<block stat> ::=                            
`{' <stat list> `}'                         `(six.compound . $2)'
<body> ::=                                  
`{' <stat list> `}'                         `(six.procedure-body . $2)'
<stat list> ::=                             
<stat> <stat list>                          `($1 . $2)'
|                                           `()'
<parameters> ::=                            
<nonempty parameters>                       $1
|                                           `()'
<nonempty parameters> ::=                   
<parameter> `,' <nonempty parameters>       `($1 . $3)'
| <parameter>                               `($1)'
<parameter> ::=                             
<type> <id-or-prefix>                       `($2 $1)'
<arguments> ::=                             
<nonempty arguments>                        $1
|                                           `()'
<nonempty arguments> ::=                    
<expr> `,' <nonempty arguments>             `($1 . $3)'
| <expr>                                    `($1)'
<elements> ::=                              
<nonempty elements>                         $1
|                                           `(six.null)'
<nonempty elements> ::=                     
<expr>                                      `(six.list $1 (six.null))'
| <expr> `,' <nonempty elements>            `(six.list $1 $3)'
| <expr> `|' <expr>                         `(six.cons $1 $3)'
<id-or-prefix> ::=                          
<identifier>                                `(six.identifier $1)'
| `\' <datum>                               `(six.prefix $2)'
<type> ::=                                  
`int'                                       `int'
| `char'                                    `char'
| `bool'                                    `bool'
| `void'                                    `void'
| `float'                                   `float'
| `double'                                  `double'
| `obj'                                     `obj'

18.12.2 SIX semantics
---------------------

The semantics of SIX depends on the definition of the `six.XXX'
identifiers (as functions and macros).  Many of these identifiers are
predefined macros which give SIX a semantics that is close to C's.  The
user may override these definitions to change the semantics either
globally or locally.  For example, `six.x^y' is a predefined macro that
expands `(six.x^y x y)' into `(bitwise-xor x y)'.  If the user prefers
the `^' operator to express exponentiation in a specific function, then
in that function `six.x^y' can be redefined as a macro that expands
`(six.x^y x y)' into `(expt x y)'.  Note that the associativity and
precedence of operators cannot be changed as that is a syntactic issue.

   Note that the following identifiers are not predefined, and
consequently they do not have a predefined semantics: `six.label',
`six.goto', `six.switch', `six.case', `six.break', `six.continue',
`six.return', `six.clause', `six.x:-y', and `six.!'.

   The following is an example showing some of the predefined semantics
of SIX:

     > (list (+ 1 2) \3+4; (+ 5 6))
     (3 7 11)
     > \[ 1+2, \(+ 3 4), 5+6 ];
     (3 7 11)
     > (map (lambda (x) \(x*x-1)/log(x+1);) '(1 2 3 4))
     (0 2.730717679880512 5.7707801635558535 9.320024018394177)
     > \obj n = expt(10,5);
     > n
     100000
     > \obj t[3][10] = 88;
     > \t[0][9] = t[2][1] = 11;
     11
     > t
     #(#(88 88 88 88 88 88 88 88 88 11)
       #(88 88 88 88 88 88 88 88 88 88)
       #(88 11 88 88 88 88 88 88 88 88))
     > \obj radix = new parameter (10);
     > \radix(2);
     > \radix();
     2
     > \for (int i=0; i<5; i++) pp(1<<i*8);
     1
     256
     65536
     16777216
     4294967296
     > \obj \make-adder (obj x) { obj (obj y) { x+y; }; }
     > \map (new adder (100), [1,2,3,4]);
     (101 102 103 104)
     > (map (make-adder 100) (list 1 2 3 4))
     (101 102 103 104)

19 C-interface
**************

The Gambit Scheme system offers a mechanism for interfacing Scheme code
and C code called the "C-interface".  A Scheme program indicates which
C functions it needs to have access to and which Scheme procedures can
be called from C, and the C interface automatically constructs the
corresponding Scheme procedures and C functions.  The conversions needed
to transform data from the Scheme representation to the C representation
(and back), are generated automatically in accordance with the argument
and result types of the C function or Scheme procedure.

   The C-interface places some restrictions on the types of data that
can be exchanged between C and Scheme.  The mapping of data types
between C and Scheme is discussed in the next section.  The remaining
sections of this chapter describe each special form of the C-interface.

19.1 The mapping of types between C and Scheme
==============================================

Scheme and C do not provide the same set of built-in data types so it is
important to understand which Scheme type is compatible with which C
type and how values get mapped from one environment to the other.  To
improve compatibility a new type is added to Scheme, the `foreign'
object type, and the following data types are added to C:

`scheme-object'
     denotes the universal type of Scheme objects (type `___SCMOBJ'
     defined in `gambit.h')

`bool'
     denotes the C++ `bool' type or the C `int' type (type `___BOOL'
     defined in `gambit.h')

`int8'
     8 bit signed integer (type `___S8' defined in `gambit.h')

`unsigned-int8'
     8 bit unsigned integer (type `___U8' defined in `gambit.h')

`int16'
     16 bit signed integer (type `___S16' defined in `gambit.h')

`unsigned-int16'
     16 bit unsigned integer (type `___U16' defined in `gambit.h')

`int32'
     32 bit signed integer (type `___S32' defined in `gambit.h')

`unsigned-int32'
     32 bit unsigned integer (type `___U32' defined in `gambit.h')

`int64'
     64 bit signed integer (type `___S64' defined in `gambit.h')

`unsigned-int64'
     64 bit unsigned integer (type `___U64' defined in `gambit.h')

`float32'
     32 bit floating point number (type `___F32' defined in `gambit.h')

`float64'
     64 bit floating point number (type `___F64' defined in `gambit.h')

`ISO-8859-1'
     denotes ISO-8859-1 encoded characters (8 bit unsigned integer,
     type `___ISO_8859_1' defined in `gambit.h')

`UCS-2'
     denotes UCS-2 encoded characters (16 bit unsigned integer, type
     `___UCS_2' defined in `gambit.h')

`UCS-4'
     denotes UCS-4 encoded characters (32 bit unsigned integer, type
     `___UCS_4' defined in `gambit.h')

`char-string'
     denotes the C `char*' type when used as a null terminated string

`nonnull-char-string'
     denotes the nonnull C `char*' type when used as a null terminated
     string

`nonnull-char-string-list'
     denotes an array of nonnull C `char*' terminated with a null
     pointer

`ISO-8859-1-string'
     denotes ISO-8859-1 encoded strings (null terminated string of 8
     bit unsigned integers, i.e. `___ISO_8859_1*')

`nonnull-ISO-8859-1-string'
     denotes nonnull ISO-8859-1 encoded strings (null terminated string
     of 8 bit unsigned integers, i.e. `___ISO_8859_1*')

`nonnull-ISO-8859-1-stringlist'
     denotes an array of nonnull ISO-8859-1 encoded strings terminated
     with a null pointer

`UTF-8-string'
     denotes UTF-8 encoded strings (null terminated string of `char',
     i.e. `char*')

`nonnull-UTF-8-string'
     denotes nonnull UTF-8 encoded strings (null terminated string of
     `char', i.e. `char*')

`nonnull-UTF-8-string-list'
     denotes an array of nonnull UTF-8 encoded strings terminated with
     a null pointer

`UCS-2-string'
     denotes UCS-2 encoded strings (null terminated string of 16 bit
     unsigned integers, i.e. `___UCS_2*')

`nonnull-UCS-2-string'
     denotes nonnull UCS-2 encoded strings (null terminated string of
     16 bit unsigned integers, i.e. `___UCS_2*')

`nonnull-UCS-2-string-list'
     denotes an array of nonnull UCS-2 encoded strings terminated with
     a null pointer

`UCS-4-string'
     denotes UCS-4 encoded strings (null terminated string of 32 bit
     unsigned integers, i.e. `___UCS_4*')

`nonnull-UCS-4-string'
     denotes nonnull UCS-4 encoded strings (null terminated string of
     32 bit unsigned integers, i.e. `___UCS_4*')

`nonnull-UCS-4-string-list'
     denotes an array of nonnull UCS-4 encoded strings terminated with
     a null pointer

`wchar_t-string'
     denotes `wchar_t' encoded strings (null terminated string of
     `wchar_t', i.e. `wchar_t*')

`nonnull-wchar_t-string'
     denotes nonnull `wchar_t' encoded strings (null terminated string
     of `wchar_t', i.e. `wchar_t*')

`nonnull-wchar_t-string-list'
     denotes an array of nonnull `wchar_t' encoded strings terminated
     with a null pointer

   To specify a particular C type inside the `c-lambda', `c-define' and
`c-define-type' forms, the following "Scheme notation" is used:

`Scheme notation'
     C type

`void'
     `void'

`bool'
     `bool'

`char'
     `char'  (may be signed or unsigned depending on the C compiler)

`signed-char'
     `signed char'

`unsigned-char'
     `unsigned char'

`ISO-8859-1'
     `ISO-8859-1'

`UCS-2'
     `UCS-2'

`UCS-4'
     `UCS-4'

`wchar_t'
     `wchar_t'

`short'
     `short'

`unsigned-short'
     `unsigned short'

`int'
     `int'

`unsigned-int'
     `unsigned int'

`long'
     `long'

`unsigned-long'
     `unsigned long'

`long-long'
     `long long'

`unsigned-long-long'
     `unsigned long long'

`float'
     `float'

`double'
     `double'

`int8'
     `int8'

`unsigned-int8'
     `unsigned-int8'

`int16'
     `int16'

`unsigned-int16'
     `unsigned-int16'

`int32'
     `int32'

`unsigned-int32'
     `unsigned-int32'

`int64'
     `int64'

`unsigned-int64'
     `unsigned-int64'

`float32'
     `float32'

`float64'
     `float64'

`(struct "C-STRUCT-ID" [TAG [RELEASE-FUNCTION]])'
     `struct C-STRUCT-ID'  (where C-STRUCT-ID is the name of a C
     structure; see below for the meaning of TAG and RELEASE-FUNCTION)

`(union "C-UNION-ID" [TAG [RELEASE-FUNCTION]])'
     `union C-UNION-ID'  (where C-UNION-ID is the name of a C union;
     see below for the meaning of TAG and RELEASE-FUNCTION)

`(type "C-TYPE-ID" [TAG [RELEASE-FUNCTION]])'
     `C-TYPE-ID'  (where C-TYPE-ID is an identifier naming a C type;
     see below for the meaning of TAG and RELEASE-FUNCTION)

`(pointer TYPE [TAG [RELEASE-FUNCTION]])'
     `T*'  (where T is the C equivalent of TYPE which must be the
     Scheme notation of a C type; see below for the meaning of TAG and
     RELEASE-FUNCTION)

`(nonnull-pointer TYPE [TAG [RELEASE-FUNCTION]])'
     same as `(pointer TYPE [TAG [RELEASE-FUNCTION]])' except the
     `NULL' pointer is not allowed

`(function (TYPE1...) RESULT-TYPE)'
     function with the given argument types and result type

`(nonnull-function (TYPE1...) RESULT-TYPE)'
     same as `(function (TYPE1...) RESULT-TYPE)' except the `NULL'
     pointer is not allowed

`char-string'
     `char-string'

`nonnull-char-string'
     `nonnull-char-string'

`nonnull-char-string-list'
     `nonnull-char-string-list'

`ISO-8859-1-string'
     `ISO-8859-1-string'

`nonnull-ISO-8859-1-string'
     `nonnull-ISO-8859-1-string'

`nonnull-ISO-8859-1-string-list'
     `nonnull-ISO-8859-1-string-list'

`UTF-8-string'
     `UTF-8-string'

`nonnull-UTF-8-string'
     `nonnull-UTF-8-string'

`nonnull-UTF-8-string-list'
     `nonnull-UTF-8-string-list'

`UCS-2-string'
     `UCS-2-string'

`nonnull-UCS-2-string'
     `nonnull-UCS-2-string'

`nonnull-UCS-2-string-list'
     `nonnull-UCS-2-string-list'

`UCS-4-string'
     `UCS-4-string'

`nonnull-UCS-4-string'
     `nonnull-UCS-4-string'

`nonnull-UCS-4-string-list'
     `nonnull-UCS-4-string-list'

`wchar_t-string'
     `wchar_t-string'

`nonnull-wchar_t-string'
     `nonnull-wchar_t-string'

`nonnull-wchar_t-string-list'
     `nonnull-wchar_t-string-list'

`scheme-object'
     `scheme-object'

`NAME'
     appropriate translation of NAME (where NAME is a C type defined
     with `c-define-type')

`"C-TYPE-ID"'
     C-TYPE-ID (this form is equivalent to `(type "C-TYPE-ID")')

   The `struct', `union', `type', `pointer' and `nonnull-pointer' types
are "foreign types" and they are represented on the Scheme side as
"foreign objects".  A foreign object is internally represented as a
pointer.  This internal pointer is identical to the C pointer being
represented in the case of the `pointer' and `nonnull-pointer' types.

   In the case of the `struct', `union' and `type' types, the internal
pointer points to a copy of the C data type being represented.  When an
instance of one of these types is converted from C to Scheme, a block
of memory is allocated from the C heap and initialized with the
instance and then a foreign object is allocated from the Scheme heap
and initialized with the pointer to this copy.  This approach may
appear overly complex, but it allows the conversion of C++ classes that
do not have a zero parameter constructor or an assignment method (i.e.
when compiling with a C++ compiler an instance is copied using `new
TYPE (INSTANCE)', which calls the copy-constructor of TYPE if it is a
class; TYPE's assignment operator is never used).  Conversion from
Scheme to C simply dereferences the internal pointer (no allocation
from the C heap is performed).  Deallocation of the copy on the C heap
is under the control of the release function attached to the foreign
object (see below).

   For type checking on the Scheme side, a TAG can be specified within
a foreign type specification.  The TAG must be `#f' or a symbol.  When
it is not specified the TAG defaults to a symbol whose name, as
returned by `symbol->string', is the C type declaration for that type.
For example the default tag for the type `(pointer (pointer char))' is
the symbol `char**'.  Two foreign types are compatible (i.e. can be
converted from one to the other) if they have identical tags or if at
least one of the tags is `#f'.  For the safest code the `#f' tag should
be used sparingly, as it completely bypasses type checking.  The
external representation of Scheme foreign objects (used by the `write'
procedure) contains the tag if it is not `#f', and the hexadecimal
address denoted by the internal pointer, for example `#<char** #2
0x2AAC535C>'.  Note that the hexadecimal address is in C notation, which
can be easily transferred to a C debugger with a "cut-and-paste".

   A RELEASE-FUNCTION can also be specified within a foreign type
specification.  The RELEASE-FUNCTION must be `#f' or a string naming a
C function with a single parameter of type `void*' (in which the
internal pointer is passed) and with a result of type `___SCMOBJ' (for
returning an error code).  When the RELEASE-FUNCTION is not specified
or is `#f' a default function is constructed by the C-interface.  This
default function does nothing in the case of the `pointer' and
`nonnull-pointer' types (deallocation is not the responsibility of the
C-interface) and returns the fixnum `___FIX(___NO_ERR)' to indicate no
error.  In the case of the `struct', `union' and `type' types, the
default function reclaims the copy on the C heap referenced by the
internal pointer (when using a C++ compiler this is done using `delete
(TYPE*)INTERNAL-POINTER', which calls the destructor of TYPE if it is a
class) and returns `___FIX(___NO_ERR)'.  In many situations the default
RELEASE-FUNCTION will perform the appropriate cleanup for the foreign
type.  However, in certain cases special operations (such as
decrementing a reference count, removing the object from a table, etc)
must be performed.  For such cases a user supplied RELEASE-FUNCTION is
needed.

   The RELEASE-FUNCTION is invoked at most once for any foreign object.
After the RELEASE-FUNCTION is invoked, the foreign object is
considered "released" and can no longer be used in a foreign type
conversion.  When the garbage collector detects that a foreign object
is no longer reachable by the program, it will invoke the
RELEASE-FUNCTION if the foreign object is not yet released.  When there
is a need to release the foreign object promptly, the program can
explicitly call `(foreign-release! OBJ)' which invokes the
RELEASE-FUNCTION if the foreign object is not yet released, and does
nothing otherwise.  The call `(foreign-released? OBJ)' returns a
boolean indicating whether the foreign object OBJ has been released yet
or not.  Finally, the call `(foreign-address OBJ)' returns the address
denoted by the internal pointer of foreign object OBJ or 0 if it has
been released.

   The following table gives the C types to which each Scheme type can
be converted:

Scheme type
     Allowed target C types

boolean `#f'
     `scheme-object'; `bool'; `pointer'; `function'; `char-string';
     `ISO-8859-1-string'; `UTF-8-string'; `UCS-2-string';
     `UCS-4-string'; `wchar_t-string'

boolean `#t'
     `scheme-object'; `bool'

character
     `scheme-object'; `bool'; [`[un]signed'] `char'; `ISO-8859-1';
     `UCS-2'; `UCS-4'; `wchar_t'

exact integer
     `scheme-object'; `bool'; [`unsigned-']
     `int8'/`int16'/`int32'/`int64'; [`unsigned'] `short'/`int'/`long'

inexact real
     `scheme-object'; `bool'; `float'; `double'; `float32'; `float64'

string
     `scheme-object'; `bool'; `char-string'; `nonnull-char-string';
     `ISO-8859-1-string'; `nonnull-ISO-8859-1-string'; `UTF-8-string';
     `nonnull-UTF-8-string'; `UCS-2-string'; `nonnull-UCS-2-string';
     `UCS-4-string'; `nonnull-UCS-4-string'; `wchar_t-string';
     `nonnull-wchar_t-string'

foreign object
     `scheme-object'; `bool';
     `struct'/`union'/`type'/`pointer'/`nonnull-pointer' with the
     appropriate tag

vector
     `scheme-object'; `bool'

symbol
     `scheme-object'; `bool'

procedure
     `scheme-object'; `bool'; `function'; `nonnull-function'

other objects
     `scheme-object'; `bool'

   The following table gives the Scheme types to which each C type will
be converted:

C type
     Resulting Scheme type

scheme-object
     the Scheme object encoded

bool
     boolean

[`[un]signed'] `char'; `ISO-8859-1'; `UCS-2'; `UCS-4'; `wchar_t'
     character

[`unsigned-'] `int8'/`int16'/`int32'/`int64'; [`unsigned'] `short'/`int'/`long'
     exact integer

`float'; `double'; `float32'; `float64'
     inexact real

`char-string'; `ISO-8859-1-string'; `UTF-8-string'; `UCS-2-string'; `UCS-4-string'; `wchar_t-string'
     string or `#f' if it is equal to `NULL'

`nonnull-char-string'; `nonnull-ISO-8859-1-string'; `nonnull-UTF-8-string'; `nonnull-UCS-2-string'; `nonnull-UCS-4-string'; `nonnull-wchar_t-string'
     string

`struct'/`union'/`type'/`pointer'/`nonnull-pointer'
     foreign object with the appropriate tag or `#f' in the case of a
     `pointer' equal to `NULL'

`function'
     procedure or `#f' if it is equal to `NULL'

`nonnull-function'
     procedure

`void'
     void object

   All Scheme types are compatible with the C types `scheme-object' and
`bool'.  Conversion to and from the C type `scheme-object' is the
identity function on the object encoding.  This provides a low-level
mechanism for accessing Scheme's object representation from C (with the
help of the macros in the `gambit.h' header file).  When a C `bool'
type is expected, an extended Scheme boolean can be passed (`#f' is
converted to 0 and all other values are converted to 1).

   The Scheme boolean `#f' can be passed to the C environment where a
`char-string', `ISO-8859-1-string', `UTF-8-string', `UCS-2-string',
`UCS-4-string', `wchar_t-string', `pointer' or `function' type is
expected.  In this case, `#f' is converted to the `NULL' pointer.  C
`bool's are extended booleans so any value different from 0 represents
true.  Thus, a C `bool' passed to the Scheme environment is mapped as
follows: 0 to `#f' and all other values to `#t'.

   A Scheme character passed to the C environment where any C character
type is expected is converted to the corresponding character in the C
environment.  An error is signaled if the Scheme character does not fit
in the C character.  Any C character type passed to Scheme is converted
to the corresponding Scheme character.  An error is signaled if the C
character does not fit in the Scheme character.

   A Scheme exact integer passed to the C environment where a C integer
type (other than `char') is expected is converted to the corresponding
integral value.  An error is signaled if the value falls outside of the
range representable by that integral type.  C integer values passed to
the Scheme environment are mapped to the same Scheme exact integer.  If
the value is outside the fixnum range, a bignum is created.

   A Scheme inexact real passed to the C environment is converted to the
corresponding `float', `double', `float32' or `float64' value.  C
`float', `double', `float32' and `float64' values passed to the Scheme
environment are mapped to the closest Scheme inexact real.

   Scheme's rational numbers and complex numbers are not compatible with
any C numeric type.

   A Scheme string passed to the C environment where any C string type
is expected is converted to a null terminated string using the
appropriate encoding.  The C string is a fresh copy of the Scheme
string.  If the C string was created for an argument of a `c-lambda',
the C string will be reclaimed when the `c-lambda' returns.  If the C
string was created for returning the result of a `c-define' to C, the
caller is responsible for reclaiming the C string with a call to the
`___release_string' function (see below for an example).  Any C string
type passed to the Scheme environment causes the creation of a fresh
Scheme string containing a copy of the C string (unless the C string is
equal to `NULL', in which case it is converted to `#f').

   A foreign type passed to the Scheme environment causes the creation
and initialization of a Scheme foreign object with the appropriate tag
(except for the case of a `pointer' equal to `NULL' which is converted
to `#f').  A Scheme foreign object can be passed where a foreign type
is expected, on the condition that the tags are appropriate (identical
or one is `#f') and the Scheme foreign object is not yet released.  The
value `#f' is also acceptable for a `pointer' type, and is converted to
`NULL'.

   Scheme procedures defined with the `c-define' special form can be
passed where the `function' and `nonnull-function' types are expected.
The value `#f' is also acceptable for a `function' type, and is
converted to `NULL'.  No other Scheme procedures are acceptable.
Conversion from the `function' and `nonnull-function' types to Scheme
procedures is not currently implemented.

19.2 The `c-declare' special form
=================================

Synopsis:

     (c-declare c-declaration)

   Initially, the C file produced by `gsc' contains only an `#include'
of `gambit.h'.  This header file provides a number of macro and
procedure declarations to access the Scheme object representation.  The
special form `c-declare' adds c-declaration (which must be a string
containing the C declarations) to the C file.  This string is copied to
the C file on a new line so it can start with preprocessor directives.
All types of C declarations are allowed (including type declarations,
variable declarations, function declarations, `#include' directives,
`#define's, and so on).  These declarations are visible to subsequent
`c-declare's, `c-initialize's, and `c-lambda's, and `c-define's in the
same module.  The most common use of this special form is to declare
the external functions that are referenced in `c-lambda' special forms.
Such functions must either be declared explicitly or by including a
header file which contains the appropriate C declarations.

   The `c-declare' special form does not return a value.  It can only
appear at top level.

   For example:

     (c-declare #<<c-declare-end

     #include <stdio.h>

     extern char *getlogin ();

     #ifdef sparc
     char *host = "sparc";
     #else
     char *host = "unknown";
     #endif

     FILE *tfile;

     c-declare-end
     )

19.3 The `c-initialize' special form
====================================

Synopsis:

     (c-initialize c-code)

   Just after the program is loaded and before control is passed to the
Scheme code, each C file is initialized by calling its associated
initialization function.  The body of this function is normally empty
but it can be extended by using the `c-initialize' form.  Each
occurence of the `c-initialize' form adds code to the body of the
initialization function in the order of appearance in the source file.
c-code must be a string containing the C code to execute.  This string
is copied to the C file on a new line so it can start with preprocessor
directives.

   The `c-initialize' special form does not return a value.  It can
only appear at top level.

   For example:

     (c-initialize "tfile = tmpfile ();")

19.4 The `c-lambda' special form
================================

Synopsis:

     (c-lambda (type1...) result-type c-name-or-code)

   The `c-lambda' special form makes it possible to create a Scheme
procedure that will act as a representative of some C function or C code
sequence.  The first subform is a list containing the type of each
argument.  The type of the function's result is given next.  Finally,
the last subform is a string that either contains the name of the C
function to call or some sequence of C code to execute.  Variadic C
functions are not supported.  The resulting Scheme procedure takes
exactly the number of arguments specified and delivers them in the same
order to the C function.  When the Scheme procedure is called, the
arguments will be converted to their C representation and then the C
function will be called.  The result returned by the C function will be
converted to its Scheme representation and this value will be returned
from the Scheme procedure call.  An error will be signaled if some
conversion is not possible.  The temporary memory allocated from the C
heap for the conversion of the arguments and result will be reclaimed
whether there is an error or not.

   When c-name-or-code is not a valid C identifier, it is treated as an
arbitrary piece of C code.  Within the C code the variables `___arg1',
`___arg2', etc. can be referenced to access the converted arguments.
Similarly, the result to be returned from the call should be assigned
to the variable `___result' except when the result is of type `struct',
`union', `type', `pointer', `nonnull-pointer', `function' or
`nonnull-function' in which case a pointer must be assigned to the
variable `___result_voidstar' which is of type `void*'.  For results of
type `pointer', `nonnull-pointer', `function' and `nonnull-function',
the value assigned to the variable `___result_voidstar' must be the
pointer or function cast to `void*'.  For results of type `struct',
`union', and `type', the value assigned to the variable
`___result_voidstar' must be a pointer to a memory allocated block
containing a copy of the result.  Note that this block will be
reclaimed by the RELEASE-FUNCTION associated with the type.  If no
result needs to be returned, the result-type should be `void' and no
assignment to the variable `___result' or `___result_voidstar' should
take place.  Note that the C code should not contain `return'
statements as this is meaningless.  Control must always fall off the
end of the C code.  The C code is copied to the C file on a new line so
it can start with preprocessor directives.  Moreover the C code is
always placed at the head of a compound statement whose lifetime
encloses the C to Scheme conversion of the result.  Consequently,
temporary storage (strings in particular) declared at the head of the C
code can be returned by assigning them to `___result' or
`___result_voidstar'.  In the c-name-or-code, the macro `___AT_END' may
be defined as the piece of C code to execute before control is returned
to Scheme but after the result is converted to its Scheme
representation.  This is mainly useful to deallocate temporary storage
contained in the result.

   When passed to the Scheme environment, the C `void' type is
converted to the void object.

   For example:

     (define fopen
       (c-lambda (nonnull-char-string nonnull-char-string)
                 (pointer "FILE")
        "fopen"))

     (define fgetc
       (c-lambda ((pointer "FILE"))
                 int
        "fgetc"))

     (let ((f (fopen "datafile" "r")))
       (if f (write (fgetc f))))

     (define char-code (c-lambda (char) int "___result = ___arg1;"))

     (define host ((c-lambda () nonnull-char-string "___result = host;")))

     (define stdin ((c-lambda () (pointer "FILE") "___result = stdin;")))

     ((c-lambda () void
     #<<c-lambda-end
       printf( "hello\n" );
       printf( "world\n" );
     c-lambda-end
     ))

     (define pack-1-char
       (c-lambda (char)
                 nonnull-char-string
     #<<c-lambda-end
        ___result = malloc (2);
        if (___result != NULL) { ___result[0] = ___arg1; ___result[1] = 0; }
        #define ___AT_END if (___result != NULL) free (___result);
     c-lambda-end
     ))

     (define pack-2-chars
       (c-lambda (char char)
                 nonnull-char-string
     #<<c-lambda-end
        char s[3];
        s[0] = ___arg1;
        s[1] = ___arg2;
        s[2] = 0;
        ___result = s;
     c-lambda-end
     ))

19.5 The `c-define' special form
================================

Synopsis:

     (c-define (variable define-formals) (type1...) result-type c-name scope
       body)

   The `c-define' special form makes it possible to create a C function
that will act as a representative of some Scheme procedure.  A C
function named c-name as well as a Scheme procedure bound to the
variable variable are defined.  The parameters of the Scheme procedure
are define-formals and its body is at the end of the form.  The type of
each argument of the C function, its result type and c-name (which must
be a string) are specified after the parameter specification of the
Scheme procedure.  When the C function c-name is called from C, its
arguments are converted to their Scheme representation and passed to
the Scheme procedure.  The result of the Scheme procedure is then
converted to its C representation and the C function c-name returns it
to its caller.

   The scope of the C function can be changed with the scope parameter,
which must be a string.  This string is placed immediately before the
declaration of the C function.  So if scope is the string `"static"',
the scope of c-name is local to the module it is in, whereas if scope
is the empty string, c-name is visible from other modules.

   The `c-define' special form does not return a value.  It can only
appear at top level.

   For example:

     (c-define (proc x #!optional (y x) #!rest z) (int int char float) int "f" ""
       (write (cons x (cons y z)))
       (newline)
       (+ x y))

     (proc 1 2 #\x 1.5) => 3 and prints (1 2 #\x 1.5)
     (proc 1)           => 2 and prints (1 1)

     ; if f is called from C with the call  f (1, 2, 'x', 1.5)
     ; the value 3 is returned and (1 2 #\x 1.5) is printed.
     ; f has to be called with 4 arguments.

   The `c-define' special form is particularly useful when the driving
part of an application is written in C and Scheme procedures are called
directly from C.  The Scheme part of the application is in a sense a
"server" that is providing services to the C part.  The Scheme
procedures that are to be called from C need to be defined using the
`c-define' special form.  Before it can be used, the Scheme part must
be initialized with a call to the function `___setup'.  Before the
program terminates, it must call the function `___cleanup' so that the
Scheme part may do final cleanup.  A sample application is given in the
file `tests/server.scm'.

19.6 The `c-define-type' special form
=====================================

Synopsis:

     (c-define-type name type [c-to-scheme scheme-to-c [cleanup]])

   This form associates the type identifier name to the C type type.
The name must not clash with predefined types (e.g. `char-string',
`ISO-8859-1', etc.) or with types previously defined with
`c-define-type' in the same file.  The `c-define-type' special form
does not return a value.  It can only appear at top level.

   If only the two parameters name and type are supplied then after
this definition, the use of name in a type specification is synonymous
to type.

   For example:

     (c-define-type FILE "FILE")
     (c-define-type FILE* (pointer FILE))
     (c-define-type time-struct-ptr (pointer (struct "tms")))
     (define fopen (c-lambda (char-string char-string) FILE* "fopen"))
     (define fgetc (c-lambda (FILE*) int "fgetc"))

   Note that identifiers are not case-sensitive in standard Scheme but
it is good programming practice to use a name with the same case as in
C.

   If four or more parameters are supplied, then type must be a string
naming the C type, c-to-scheme and scheme-to-c must be strings
suffixing the C macros that convert data of that type between C and
Scheme.  If cleanup is supplied it must be a boolean indicating whether
it is necessary to perform a cleanup operation (such as freeing memory)
when data of that type is converted from Scheme to C (it defaults to
`#t').  The cleanup information is used when the C stack is unwound due
to a continuation invocation (see *Note continuations::).  Although it
is safe to always specify `#t', it is more efficient in time and space
to specify `#f' because the unwinding mechanism can skip C-interface
frames which only contain conversions of data types requiring no
cleanup.  Two pairs of C macros need to be defined for conversions
performed by `c-lambda' forms and two pairs for conversions performed
by `c-define' forms:

     ___BEGIN_CFUN_scheme-to-c(___SCMOBJ, type, int)
     ___END_CFUN_scheme-to-c(___SCMOBJ, type, int)

     ___BEGIN_CFUN_c-to-scheme(type, ___SCMOBJ)
     ___END_CFUN_c-to-scheme(type, ___SCMOBJ)

     ___BEGIN_SFUN_c-to-scheme(type, ___SCMOBJ, int)
     ___END_SFUN_c-to-scheme(type, ___SCMOBJ, int)

     ___BEGIN_SFUN_scheme-to-c(___SCMOBJ, type)
     ___END_SFUN_scheme-to-c(___SCMOBJ, type)

   The macros prefixed with `___BEGIN' perform the conversion and those
prefixed with `___END' perform any cleanup necessary (such as freeing
memory temporarily allocated for the conversion).  The macro
`___END_CFUN_scheme-to-c' must free the result of the conversion if it
is memory allocated, and `___END_SFUN_scheme-to-c' must not (i.e. it is
the responsibility of the caller to free the result).

   The first parameter of these macros is the C variable that contains
the value to be converted, and the second parameter is the C variable in
which to store the converted value.  The third parameter, when present,
is the index (starting at 1) of the parameter of the `c-lambda' or
`c-define' form that is being converted (this is useful for reporting
precise error information when a conversion is impossible).

   To allow for type checking, the first three `___BEGIN' macros must
expand to an unterminated compound statement prefixed by an `if',
conditional on the absence of type check error:

     if ((___err = CONVERSION_OPERATION) == ___FIX(___NO_ERR)) {

   The last `___BEGIN' macro must expand to an unterminated compound
statement:

     { ___err = CONVERSION_OPERATION;

   If type check errors are impossible then a `___BEGIN' macro can
simply expand to an unterminated compound statement performing the
conversion:

     { CONVERSION_OPERATION;

   The `___END' macros must expand to a statement, or to nothing if no
cleanup is required, followed by a closing brace (to terminate the
compound statement started at the corresponding `___BEGIN' macro).

   The CONVERSION_OPERATION is typically a function call that returns
an error code value of type `___SCMOBJ' (the error codes are defined in
`gambit.h', and the error code `___FIX(___UNKNOWN_ERR)' is available
for generic errors).  CONVERSION_OPERATION can also set the variable
`___errmsg' of type `___SCMOBJ' to a specific Scheme string error
message.

   Below is a simple example showing how to interface to an `EBCDIC'
character type.  Memory allocation is not needed for conversion and type
check errors are impossible when converting EBCDIC to Scheme characters,
but they are possible when converting from Scheme characters to EBCDIC
since Gambit supports Unicode characters.

     (c-declare #<<c-declare-end

     typedef char EBCDIC; /* EBCDIC encoded characters */

     void put_char (EBCDIC c) { ... } /* EBCDIC I/O functions */
     EBCDIC get_char (void) { ... }

     char EBCDIC_to_ISO_8859_1[256] = { ... }; /* conversion tables */
     char ISO_8859_1_to_EBCDIC[256] = { ... };

     ___SCMOBJ SCMOBJ_to_EBCDIC (___SCMOBJ src, EBCDIC *dst)
     {
       int x = ___INT(src); /* convert from Scheme character to int */
       if (x > 255) return ___FIX(___UNKNOWN_ERR);
       *dst = ISO_8859_1_to_EBCDIC[x];
       return ___FIX(___NO_ERR);
     }

     #define ___BEGIN_CFUN_SCMOBJ_to_EBCDIC(src,dst,i) \
     if ((___err = SCMOBJ_to_EBCDIC (src, &dst)) == ___FIX(___NO_ERR)) {
     #define ___END_CFUN_SCMOBJ_to_EBCDIC(src,dst,i) }

     #define ___BEGIN_CFUN_EBCDIC_to_SCMOBJ(src,dst) \
     { dst = ___CHR(EBCDIC_to_ISO_8859_1[src]);
     #define ___END_CFUN_EBCDIC_to_SCMOBJ(src,dst) }

     #define ___BEGIN_SFUN_EBCDIC_to_SCMOBJ(src,dst,i) \
     { dst = ___CHR(EBCDIC_to_ISO_8859_1[src]);
     #define ___END_SFUN_EBCDIC_to_SCMOBJ(src,dst,i) }

     #define ___BEGIN_SFUN_SCMOBJ_to_EBCDIC(src,dst) \
     { ___err = SCMOBJ_to_EBCDIC (src, &dst);
     #define ___END_SFUN_SCMOBJ_to_EBCDIC(src,dst) }

     c-declare-end
     )

     (c-define-type EBCDIC "EBCDIC" "EBCDIC_to_SCMOBJ" "SCMOBJ_to_EBCDIC" #f)

     (define put-char (c-lambda (EBCDIC) void "put_char"))
     (define get-char (c-lambda () EBCDIC "get_char"))

     (c-define (write-EBCDIC c) (EBCDIC) void "write_EBCDIC" ""
       (write-char c))

     (c-define (read-EBCDIC) () EBCDIC "read_EBCDIC" ""
       (read-char))

   Below is a more complex example that requires memory allocation when
converting from C to Scheme.  It is an interface to a 2D `point' type
which is represented in Scheme by a pair of integers.  The conversion
of the `x' and `y' components is done by calls to the conversion macros
for the `int' type (defined in `gambit.h').  Note that no cleanup is
necessary when converting from Scheme to C (i.e. the last parameter of
the `c-define-type' is `#f').

     (c-declare #<<c-declare-end

     typedef struct { int x, y; } point;

     void line_to (point p) { ... }
     point get_mouse (void) { ... }
     point add_points (point p1, point p2) { ... }

     ___SCMOBJ SCMOBJ_to_POINT (___SCMOBJ src, point *dst, int arg_num)
     {
       ___SCMOBJ ___err = ___FIX(___NO_ERR);
       if (!___PAIRP(src))
         ___err = ___FIX(___UNKNOWN_ERR);
       else
         {
           ___SCMOBJ car = ___CAR(src);
           ___SCMOBJ cdr = ___CDR(src);
           ___BEGIN_CFUN_SCMOBJ_TO_INT(car,dst->x,arg_num)
           ___BEGIN_CFUN_SCMOBJ_TO_INT(cdr,dst->y,arg_num)
           ___END_CFUN_SCMOBJ_TO_INT(cdr,dst->y,arg_num)
           ___END_CFUN_SCMOBJ_TO_INT(car,dst->x,arg_num)
         }
       return ___err;
     }

     ___SCMOBJ POINT_to_SCMOBJ (point src, ___SCMOBJ *dst, int arg_num)
     {
       ___SCMOBJ ___err = ___FIX(___NO_ERR);
       ___SCMOBJ x_scmobj;
       ___SCMOBJ y_scmobj;
       ___BEGIN_SFUN_INT_TO_SCMOBJ(src.x,x_scmobj,arg_num)
       ___BEGIN_SFUN_INT_TO_SCMOBJ(src.y,y_scmobj,arg_num)
       *dst = ___EXT(___make_pair) (x_scmobj, y_scmobj, ___STILL);
       if (___FIXNUMP(*dst))
         ___err = *dst; /* return allocation error */
       ___END_SFUN_INT_TO_SCMOBJ(src.y,y_scmobj,arg_num)
       ___END_SFUN_INT_TO_SCMOBJ(src.x,x_scmobj,arg_num)
       return ___err;
     }

     #define ___BEGIN_CFUN_SCMOBJ_to_POINT(src,dst,i) \
     if ((___err = SCMOBJ_to_POINT (src, &dst, i)) == ___FIX(___NO_ERR)) {
     #define ___END_CFUN_SCMOBJ_to_POINT(src,dst,i) }

     #define ___BEGIN_CFUN_POINT_to_SCMOBJ(src,dst) \
     if ((___err = POINT_to_SCMOBJ (src, &dst, ___RETURN_POS)) == ___FIX(___NO_ERR)) {
     #define ___END_CFUN_POINT_to_SCMOBJ(src,dst) \
     ___EXT(___release_scmobj) (dst); }

     #define ___BEGIN_SFUN_POINT_to_SCMOBJ(src,dst,i) \
     if ((___err = POINT_to_SCMOBJ (src, &dst, i)) == ___FIX(___NO_ERR)) {
     #define ___END_SFUN_POINT_to_SCMOBJ(src,dst,i) \
     ___EXT(___release_scmobj) (dst); }

     #define ___BEGIN_SFUN_SCMOBJ_to_POINT(src,dst) \
     { ___err = SCMOBJ_to_POINT (src, &dst, ___RETURN_POS);
     #define ___END_SFUN_SCMOBJ_to_POINT(src,dst) }

     c-declare-end
     )

     (c-define-type point "point" "POINT_to_SCMOBJ" "SCMOBJ_to_POINT" #f)

     (define line-to (c-lambda (point) void "line_to"))
     (define get-mouse (c-lambda () point "get_mouse"))
     (define add-points (c-lambda (point point) point "add_points"))

     (c-define (write-point p) (point) void "write_point" ""
       (write p))

     (c-define (read-point) () point "read_point" ""
       (read))

   An example that requires memory allocation when converting from C to
Scheme and Scheme to C is shown below.  It is an interface to a
"null-terminated array of strings" type which is represented in Scheme
by a list of strings.  Note that some cleanup is necessary when
converting from Scheme to C.

     (c-declare #<<c-declare-end

     #include <stdlib.h>
     #include <unistd.h>

     extern char **environ;

     char **get_environ (void) { return environ; }

     void free_strings (char **strings)
     {
       char **ptr = strings;
       while (*ptr != NULL)
         {
           ___EXT(___release_string) (*ptr);
           ptr++;
         }
       free (strings);
     }

     ___SCMOBJ SCMOBJ_to_STRINGS (___SCMOBJ src, char ***dst, int arg_num)
     {
       /*
        * Src is a list of Scheme strings.  Dst will be a null terminated
        * array of C strings.
        */

       int i;
       ___SCMOBJ lst = src;
       int len = 4; /* start with a small result array */
       char **result = (char**) malloc (len * sizeof (char*));

       if (result == NULL)
         return ___FIX(___HEAP_OVERFLOW_ERR);

       i = 0;
       result[i] = NULL; /* always keep array null terminated */

       while (___PAIRP(lst))
         {
           ___SCMOBJ scm_str = ___CAR(lst);
           char *c_str;
           ___SCMOBJ ___err;

           if (i >= len-1) /* need to grow the result array? */
             {
               char **new_result;
               int j;

               len = len * 3 / 2;
               new_result = (char**) malloc (len * sizeof (char*));
               if (new_result == NULL)
                 {
                   free_strings (result);
                   return ___FIX(___HEAP_OVERFLOW_ERR);
                 }
               for (j=i; j>=0; j--)
                 new_result[j] = result[j];
               free (result);
               result = new_result;
             }

           ___err = ___EXT(___SCMOBJ_to_CHARSTRING) (scm_str, &c_str, arg_num);

           if (___err != ___FIX(___NO_ERR))
             {
               free_strings (result);
               return ___err;
             }

           result[i++] = c_str;
           result[i] = NULL;
           lst = ___CDR(lst);
         }

       if (!___NULLP(lst))
         {
           free_strings (result);
           return ___FIX(___UNKNOWN_ERR);
         }

       /*
        * Note that the caller is responsible for calling free_strings
        * when it is done with the result.
        */

       *dst = result;
       return ___FIX(___NO_ERR);
     }

     ___SCMOBJ STRINGS_to_SCMOBJ (char **src, ___SCMOBJ *dst, int arg_num)
     {
       ___SCMOBJ ___err = ___FIX(___NO_ERR);
       ___SCMOBJ result = ___NUL; /* start with the empty list */
       int i = 0;

       while (src[i] != NULL)
         i++;

       /* build the list of strings starting at the tail */

       while (--i >= 0)
         {
           ___SCMOBJ scm_str;
           ___SCMOBJ new_result;

           /*
            * Invariant: result is either the empty list or a ___STILL pair
            * with reference count equal to 1.  This is important because
            * it is possible that ___CHARSTRING_to_SCMOBJ and ___make_pair
            * will invoke the garbage collector and we don't want the
            * reference in result to become invalid (which would be the
            * case if result was a ___MOVABLE pair or if it had a zero
            * reference count).
            */

           ___err = ___EXT(___CHARSTRING_to_SCMOBJ) (src[i], &scm_str, arg_num);

           if (___err != ___FIX(___NO_ERR))
             {
               ___EXT(___release_scmobj) (result); /* allow GC to reclaim result */
               return ___FIX(___UNKNOWN_ERR);
             }

           /*
            * Note that scm_str will be a ___STILL object with reference
            * count equal to 1, so there is no risk that it will be
            * reclaimed or moved if ___make_pair invokes the garbage
            * collector.
            */

           new_result = ___EXT(___make_pair) (scm_str, result, ___STILL);

           /*
            * We can zero the reference count of scm_str and result (if
            * not the empty list) because the pair now references these
            * objects and the pair is reachable (it can't be reclaimed
            * or moved by the garbage collector).
            */

           ___EXT(___release_scmobj) (scm_str);
           ___EXT(___release_scmobj) (result);

           result = new_result;

           if (___FIXNUMP(result))
             return result; /* allocation failed */
         }

       /*
        * Note that result is either the empty list or a ___STILL pair
        * with a reference count equal to 1.  There will be a call to
        * ___release_scmobj later on (in ___END_CFUN_STRINGS_to_SCMOBJ
        * or ___END_SFUN_STRINGS_to_SCMOBJ) that will allow the garbage
        * collector to reclaim the whole list of strings when the Scheme
        * world no longer references it.
        */

       *dst = result;
       return ___FIX(___NO_ERR);
     }

     #define ___BEGIN_CFUN_SCMOBJ_to_STRINGS(src,dst,i) \
     if ((___err = SCMOBJ_to_STRINGS (src, &dst, i)) == ___FIX(___NO_ERR)) {
     #define ___END_CFUN_SCMOBJ_to_STRINGS(src,dst,i) \
     free_strings (dst); }

     #define ___BEGIN_CFUN_STRINGS_to_SCMOBJ(src,dst) \
     if ((___err = STRINGS_to_SCMOBJ (src, &dst, ___RETURN_POS)) == ___FIX(___NO_ERR)) {
     #define ___END_CFUN_STRINGS_to_SCMOBJ(src,dst) \
     ___EXT(___release_scmobj) (dst); }

     #define ___BEGIN_SFUN_STRINGS_to_SCMOBJ(src,dst,i) \
     if ((___err = STRINGS_to_SCMOBJ (src, &dst, i)) == ___FIX(___NO_ERR)) {
     #define ___END_SFUN_STRINGS_to_SCMOBJ(src,dst,i) \
     ___EXT(___release_scmobj) (dst); }

     #define ___BEGIN_SFUN_SCMOBJ_to_STRINGS(src,dst) \
     { ___err = SCMOBJ_to_STRINGS (src, &dst, ___RETURN_POS);
     #define ___END_SFUN_SCMOBJ_to_STRINGS(src,dst) }

     c-declare-end
     )

     (c-define-type char** "char**" "STRINGS_to_SCMOBJ" "SCMOBJ_to_STRINGS")

     (define execv (c-lambda (char-string char**) int "execv"))
     (define get-environ (c-lambda () char** "get_environ"))

     (c-define (write-strings x) (char**) void "write_strings" ""
       (write x))

     (c-define (read-strings) () char** "read_strings" ""
       (read))

19.7 Continuations and the C-interface
======================================

The C-interface allows C to Scheme calls to be nested.  This means that
during a call from C to Scheme another call from C to Scheme can be
performed.  This case occurs in the following program:

     (c-declare #<<c-declare-end

     int p (char *); /* forward declarations */
     int q (void);

     int a (char *x) { return 2 * p (x+1); }
     int b (short y) { return y + q (); }

     c-declare-end
     )

     (define a (c-lambda (char-string) int "a"))
     (define b (c-lambda (short) int "b"))

     (c-define (p z) (char-string) int "p" ""
       (+ (b 10) (string-length z)))

     (c-define (q) () int "q" ""
       123)

     (write (a "hello"))

   In this example, the main Scheme program calls the C function `a'
which calls the Scheme procedure `p' which in turn calls the C function
`b' which finally calls the Scheme procedure `q'.

   Gambit-C maintains the Scheme continuation separately from the C
stack, thus allowing the Scheme continuation to be unwound
independently from the C stack.  The C stack frame created for the C
function `f' is only removed from the C stack when control returns from
`f' or when control returns to a C function "above" `f'.  Special care
is required for programs which escape to Scheme (using first-class
continuations) from a Scheme to C (to Scheme) call because the C stack
frame will remain on the stack.  The C stack may overflow if this
happens in a loop with no intervening return to a C function.  To avoid
this problem make sure the C stack gets cleaned up by executing a normal
return from a Scheme to C call.

20 System limitations
*********************

   * On some systems floating point overflows will cause the program to
     terminate with a floating point exception.

   * On some systems floating point operations involving `+nan.0'
     `+inf.0', `-inf.0', or `-0.' do not return the value required by
     the IEEE 754 floating point standard.

   * The compiler will not properly compile files with more than one
     definition (with `define') of the same procedure.  Replace all but
     the first `define' with assignments (`set!').

   * The maximum number of arguments that can be passed to a procedure
     by the `apply' procedure is 8192.


21 Copyright and license
************************

The Gambit-C system is Copyright (C) 1994-2006 by Marc Feeley, all
rights reserved.  The Gambit-C system Version 4.0 beta 20 is licensed
under two licenses: the Apache License, Version 2.0, and the GNU LESSER
GENERAL PUBLIC LICENSE, Version 2.1.  You have the option to choose
which of these two licenses to abide by.  The licenses are copied below.


                                   Apache License
                             Version 2.0, January 2004
                          http://www.apache.org/licenses/

     TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION

     1. Definitions.

        "License" shall mean the terms and conditions for use, reproduction,
        and distribution as defined by Sections 1 through 9 of this document.

        "Licensor" shall mean the copyright owner or entity authorized by
        the copyright owner that is granting the License.

        "Legal Entity" shall mean the union of the acting entity and all
        other entities that control, are controlled by, or are under common
        control with that entity. For the purposes of this definition,
        "control" means (i) the power, direct or indirect, to cause the
        direction or management of such entity, whether by contract or
        otherwise, or (ii) ownership of fifty percent (50%) or more of the
        outstanding shares, or (iii) beneficial ownership of such entity.

        "You" (or "Your") shall mean an individual or Legal Entity
        exercising permissions granted by this License.

        "Source" form shall mean the preferred form for making modifications,
        including but not limited to software source code, documentation
        source, and configuration files.

        "Object" form shall mean any form resulting from mechanical
        transformation or translation of a Source form, including but
        not limited to compiled object code, generated documentation,
        and conversions to other media types.

        "Work" shall mean the work of authorship, whether in Source or
        Object form, made available under the License, as indicated by a
        copyright notice that is included in or attached to the work
        (an example is provided in the Appendix below).

        "Derivative Works" shall mean any work, whether in Source or Object
        form, that is based on (or derived from) the Work and for which the
        editorial revisions, annotations, elaborations, or other modifications
        represent, as a whole, an original work of authorship. For the purposes
        of this License, Derivative Works shall not include works that remain
        separable from, or merely link (or bind by name) to the interfaces of,
        the Work and Derivative Works thereof.

        "Contribution" shall mean any work of authorship, including
        the original version of the Work and any modifications or additions
        to that Work or Derivative Works thereof, that is intentionally
        submitted to Licensor for inclusion in the Work by the copyright owner
        or by an individual or Legal Entity authorized to submit on behalf of
        the copyright owner. For the purposes of this definition, "submitted"
        means any form of electronic, verbal, or written communication sent
        to the Licensor or its representatives, including but not limited to
        communication on electronic mailing lists, source code control systems,
        and issue tracking systems that are managed by, or on behalf of, the
        Licensor for the purpose of discussing and improving the Work, but
        excluding communication that is conspicuously marked or otherwise
        designated in writing by the copyright owner as "Not a Contribution."

        "Contributor" shall mean Licensor and any individual or Legal Entity
        on behalf of whom a Contribution has been received by Licensor and
        subsequently incorporated within the Work.

     2. Grant of Copyright License. Subject to the terms and conditions of
        this License, each Contributor hereby grants to You a perpetual,
        worldwide, non-exclusive, no-charge, royalty-free, irrevocable
        copyright license to reproduce, prepare Derivative Works of,
        publicly display, publicly perform, sublicense, and distribute the
        Work and such Derivative Works in Source or Object form.

     3. Grant of Patent License. Subject to the terms and conditions of
        this License, each Contributor hereby grants to You a perpetual,
        worldwide, non-exclusive, no-charge, royalty-free, irrevocable
        (except as stated in this section) patent license to make, have made,
        use, offer to sell, sell, import, and otherwise transfer the Work,
        where such license applies only to those patent claims licensable
        by such Contributor that are necessarily infringed by their
        Contribution(s) alone or by combination of their Contribution(s)
        with the Work to which such Contribution(s) was submitted. If You
        institute patent litigation against any entity (including a
        cross-claim or counterclaim in a lawsuit) alleging that the Work
        or a Contribution incorporated within the Work constitutes direct
        or contributory patent infringement, then any patent licenses
        granted to You under this License for that Work shall terminate
        as of the date such litigation is filed.

     4. Redistribution. You may reproduce and distribute copies of the
        Work or Derivative Works thereof in any medium, with or without
        modifications, and in Source or Object form, provided that You
        meet the following conditions:

        (a) You must give any other recipients of the Work or
            Derivative Works a copy of this License; and

        (b) You must cause any modified files to carry prominent notices
            stating that You changed the files; and

        (c) You must retain, in the Source form of any Derivative Works
            that You distribute, all copyright, patent, trademark, and
            attribution notices from the Source form of the Work,
            excluding those notices that do not pertain to any part of
            the Derivative Works; and

        (d) If the Work includes a "NOTICE" text file as part of its
            distribution, then any Derivative Works that You distribute must
            include a readable copy of the attribution notices contained
            within such NOTICE file, excluding those notices that do not
            pertain to any part of the Derivative Works, in at least one
            of the following places: within a NOTICE text file distributed
            as part of the Derivative Works; within the Source form or
            documentation, if provided along with the Derivative Works; or,
            within a display generated by the Derivative Works, if and
            wherever such third-party notices normally appear. The contents
            of the NOTICE file are for informational purposes only and
            do not modify the License. You may add Your own attribution
            notices within Derivative Works that You distribute, alongside
            or as an addendum to the NOTICE text from the Work, provided
            that such additional attribution notices cannot be construed
            as modifying the License.

        You may add Your own copyright statement to Your modifications and
        may provide additional or different license terms and conditions
        for use, reproduction, or distribution of Your modifications, or
        for any such Derivative Works as a whole, provided Your use,
        reproduction, and distribution of the Work otherwise complies with
        the conditions stated in this License.

     5. Submission of Contributions. Unless You explicitly state otherwise,
        any Contribution intentionally submitted for inclusion in the Work
        by You to the Licensor shall be under the terms and conditions of
        this License, without any additional terms or conditions.
        Notwithstanding the above, nothing herein shall supersede or modify
        the terms of any separate license agreement you may have executed
        with Licensor regarding such Contributions.

     6. Trademarks. This License does not grant permission to use the trade
        names, trademarks, service marks, or product names of the Licensor,
        except as required for reasonable and customary use in describing the
        origin of the Work and reproducing the content of the NOTICE file.

     7. Disclaimer of Warranty. Unless required by applicable law or
        agreed to in writing, Licensor provides the Work (and each
        Contributor provides its Contributions) on an "AS IS" BASIS,
        WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
        implied, including, without limitation, any warranties or conditions
        of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
        PARTICULAR PURPOSE. You are solely responsible for determining the
        appropriateness of using or redistributing the Work and assume any
        risks associated with Your exercise of permissions under this License.

     8. Limitation of Liability. In no event and under no legal theory,
        whether in tort (including negligence), contract, or otherwise,
        unless required by applicable law (such as deliberate and grossly
        negligent acts) or agreed to in writing, shall any Contributor be
        liable to You for damages, including any direct, indirect, special,
        incidental, or consequential damages of any character arising as a
        result of this License or out of the use or inability to use the
        Work (including but not limited to damages for loss of goodwill,
        work stoppage, computer failure or malfunction, or any and all
        other commercial damages or losses), even if such Contributor
        has been advised of the possibility of such damages.

     9. Accepting Warranty or Additional Liability. While redistributing
        the Work or Derivative Works thereof, You may choose to offer,
        and charge a fee for, acceptance of support, warranty, indemnity,
        or other liability obligations and/or rights consistent with this
        License. However, in accepting such obligations, You may act only
        on Your own behalf and on Your sole responsibility, not on behalf
        of any other Contributor, and only if You agree to indemnify,
        defend, and hold each Contributor harmless for any liability
        incurred by, or claims asserted against, such Contributor by reason
        of your accepting any such warranty or additional liability.

     END OF TERMS AND CONDITIONS

     APPENDIX: How to apply the Apache License to your work.

        To apply the Apache License to your work, attach the following
        boilerplate notice, with the fields enclosed by brackets "[]"
        replaced with your own identifying information. (Don't include
        the brackets!)  The text should be enclosed in the appropriate
        comment syntax for the file format. We also recommend that a
        file or class name and description of purpose be included on the
        same "printed page" as the copyright notice for easier
        identification within third-party archives.

     Copyright [yyyy] [name of copyright owner]

     Licensed under the Apache License, Version 2.0 (the "License");
     you may not use this file except in compliance with the License.
     You may obtain a copy of the License at

         http://www.apache.org/licenses/LICENSE-2.0

     Unless required by applicable law or agreed to in writing, software
     distributed under the License is distributed on an "AS IS" BASIS,
     WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     See the License for the specific language governing permissions and
     limitations under the License.

                       GNU LESSER GENERAL PUBLIC LICENSE
                            Version 2.1, February 1999

      Copyright (C) 1991, 1999 Free Software Foundation, Inc.
          51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
      Everyone is permitted to copy and distribute verbatim copies
      of this license document, but changing it is not allowed.

     [This is the first released version of the Lesser GPL.  It also counts
      as the successor of the GNU Library Public License, version 2, hence
      the version number 2.1.]

                                 Preamble

       The licenses for most software are designed to take away your
     freedom to share and change it.  By contrast, the GNU General Public
     Licenses are intended to guarantee your freedom to share and change
     free software--to make sure the software is free for all its users.

       This license, the Lesser General Public License, applies to some
     specially designated software packages--typically libraries--of the
     Free Software Foundation and other authors who decide to use it.  You
     can use it too, but we suggest you first think carefully about whether
     this license or the ordinary General Public License is the better
     strategy to use in any particular case, based on the explanations below.

       When we speak of free software, we are referring to freedom of use,
     not price.  Our General Public Licenses are designed to make sure that
     you have the freedom to distribute copies of free software (and charge
     for this service if you wish); that you receive source code or can get
     it if you want it; that you can change the software and use pieces of
     it in new free programs; and that you are informed that you can do
     these things.

       To protect your rights, we need to make restrictions that forbid
     distributors to deny you these rights or to ask you to surrender these
     rights.  These restrictions translate to certain responsibilities for
     you if you distribute copies of the library or if you modify it.

       For example, if you distribute copies of the library, whether gratis
     or for a fee, you must give the recipients all the rights that we gave
     you.  You must make sure that they, too, receive or can get the source
     code.  If you link other code with the library, you must provide
     complete object files to the recipients, so that they can relink them
     with the library after making changes to the library and recompiling
     it.  And you must show them these terms so they know their rights.

       We protect your rights with a two-step method: (1) we copyright the
     library, and (2) we offer you this license, which gives you legal
     permission to copy, distribute and/or modify the library.

       To protect each distributor, we want to make it very clear that
     there is no warranty for the free library.  Also, if the library is
     modified by someone else and passed on, the recipients should know
     that what they have is not the original version, so that the original
     author's reputation will not be affected by problems that might be
     introduced by others.

       Finally, software patents pose a constant threat to the existence of
     any free program.  We wish to make sure that a company cannot
     effectively restrict the users of a free program by obtaining a
     restrictive license from a patent holder.  Therefore, we insist that
     any patent license obtained for a version of the library must be
     consistent with the full freedom of use specified in this license.

       Most GNU software, including some libraries, is covered by the
     ordinary GNU General Public License.  This license, the GNU Lesser
     General Public License, applies to certain designated libraries, and
     is quite different from the ordinary General Public License.  We use
     this license for certain libraries in order to permit linking those
     libraries into non-free programs.

       When a program is linked with a library, whether statically or using
     a shared library, the combination of the two is legally speaking a
     combined work, a derivative of the original library.  The ordinary
     General Public License therefore permits such linking only if the
     entire combination fits its criteria of freedom.  The Lesser General
     Public License permits more lax criteria for linking other code with
     the library.

       We call this license the "Lesser" General Public License because it
     does Less to protect the user's freedom than the ordinary General
     Public License.  It also provides other free software developers Less
     of an advantage over competing non-free programs.  These disadvantages
     are the reason we use the ordinary General Public License for many
     libraries.  However, the Lesser license provides advantages in certain
     special circumstances.

       For example, on rare occasions, there may be a special need to
     encourage the widest possible use of a certain library, so that it becomes
     a de-facto standard.  To achieve this, non-free programs must be
     allowed to use the library.  A more frequent case is that a free
     library does the same job as widely used non-free libraries.  In this
     case, there is little to gain by limiting the free library to free
     software only, so we use the Lesser General Public License.

       In other cases, permission to use a particular library in non-free
     programs enables a greater number of people to use a large body of
     free software.  For example, permission to use the GNU C Library in
     non-free programs enables many more people to use the whole GNU
     operating system, as well as its variant, the GNU/Linux operating
     system.

       Although the Lesser General Public License is Less protective of the
     users' freedom, it does ensure that the user of a program that is
     linked with the Library has the freedom and the wherewithal to run
     that program using a modified version of the Library.

       The precise terms and conditions for copying, distribution and
     modification follow.  Pay close attention to the difference between a
     "work based on the library" and a "work that uses the library".  The
     former contains code derived from the library, whereas the latter must
     be combined with the library in order to run.

                       GNU LESSER GENERAL PUBLIC LICENSE
        TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION

       0. This License Agreement applies to any software library or other
     program which contains a notice placed by the copyright holder or
     other authorized party saying it may be distributed under the terms of
     this Lesser General Public License (also called "this License").
     Each licensee is addressed as "you".

       A "library" means a collection of software functions and/or data
     prepared so as to be conveniently linked with application programs
     (which use some of those functions and data) to form executables.

       The "Library", below, refers to any such software library or work
     which has been distributed under these terms.  A "work based on the
     Library" means either the Library or any derivative work under
     copyright law: that is to say, a work containing the Library or a
     portion of it, either verbatim or with modifications and/or translated
     straightforwardly into another language.  (Hereinafter, translation is
     included without limitation in the term "modification".)

       "Source code" for a work means the preferred form of the work for
     making modifications to it.  For a library, complete source code means
     all the source code for all modules it contains, plus any associated
     interface definition files, plus the scripts used to control compilation
     and installation of the library.

       Activities other than copying, distribution and modification are not
     covered by this License; they are outside its scope.  The act of
     running a program using the Library is not restricted, and output from
     such a program is covered only if its contents constitute a work based
     on the Library (independent of the use of the Library in a tool for
     writing it).  Whether that is true depends on what the Library does
     and what the program that uses the Library does.

       1. You may copy and distribute verbatim copies of the Library's
     complete source code as you receive it, in any medium, provided that
     you conspicuously and appropriately publish on each copy an
     appropriate copyright notice and disclaimer of warranty; keep intact
     all the notices that refer to this License and to the absence of any
     warranty; and distribute a copy of this License along with the
     Library.

       You may charge a fee for the physical act of transferring a copy,
     and you may at your option offer warranty protection in exchange for a
     fee.

       2. You may modify your copy or copies of the Library or any portion
     of it, thus forming a work based on the Library, and copy and
     distribute such modifications or work under the terms of Section 1
     above, provided that you also meet all of these conditions:

         a) The modified work must itself be a software library.

         b) You must cause the files modified to carry prominent notices
         stating that you changed the files and the date of any change.

         c) You must cause the whole of the work to be licensed at no
         charge to all third parties under the terms of this License.

         d) If a facility in the modified Library refers to a function or a
         table of data to be supplied by an application program that uses
         the facility, other than as an argument passed when the facility
         is invoked, then you must make a good faith effort to ensure that,
         in the event an application does not supply such function or
         table, the facility still operates, and performs whatever part of
         its purpose remains meaningful.

         (For example, a function in a library to compute square roots has
         a purpose that is entirely well-defined independent of the
         application.  Therefore, Subsection 2d requires that any
         application-supplied function or table used by this function must
         be optional: if the application does not supply it, the square
         root function must still compute square roots.)

     These requirements apply to the modified work as a whole.  If
     identifiable sections of that work are not derived from the Library,
     and can be reasonably considered independent and separate works in
     themselves, then this License, and its terms, do not apply to those
     sections when you distribute them as separate works.  But when you
     distribute the same sections as part of a whole which is a work based
     on the Library, the distribution of the whole must be on the terms of
     this License, whose permissions for other licensees extend to the
     entire whole, and thus to each and every part regardless of who wrote
     it.

     Thus, it is not the intent of this section to claim rights or contest
     your rights to work written entirely by you; rather, the intent is to
     exercise the right to control the distribution of derivative or
     collective works based on the Library.

     In addition, mere aggregation of another work not based on the Library
     with the Library (or with a work based on the Library) on a volume of
     a storage or distribution medium does not bring the other work under
     the scope of this License.

       3. You may opt to apply the terms of the ordinary GNU General Public
     License instead of this License to a given copy of the Library.  To do
     this, you must alter all the notices that refer to this License, so
     that they refer to the ordinary GNU General Public License, version 2,
     instead of to this License.  (If a newer version than version 2 of the
     ordinary GNU General Public License has appeared, then you can specify
     that version instead if you wish.)  Do not make any other change in
     these notices.

       Once this change is made in a given copy, it is irreversible for
     that copy, so the ordinary GNU General Public License applies to all
     subsequent copies and derivative works made from that copy.

       This option is useful when you wish to copy part of the code of
     the Library into a program that is not a library.

       4. You may copy and distribute the Library (or a portion or
     derivative of it, under Section 2) in object code or executable form
     under the terms of Sections 1 and 2 above provided that you accompany
     it with the complete corresponding machine-readable source code, which
     must be distributed under the terms of Sections 1 and 2 above on a
     medium customarily used for software interchange.

       If distribution of object code is made by offering access to copy
     from a designated place, then offering equivalent access to copy the
     source code from the same place satisfies the requirement to
     distribute the source code, even though third parties are not
     compelled to copy the source along with the object code.

       5. A program that contains no derivative of any portion of the
     Library, but is designed to work with the Library by being compiled or
     linked with it, is called a "work that uses the Library".  Such a
     work, in isolation, is not a derivative work of the Library, and
     therefore falls outside the scope of this License.

       However, linking a "work that uses the Library" with the Library
     creates an executable that is a derivative of the Library (because it
     contains portions of the Library), rather than a "work that uses the
     library".  The executable is therefore covered by this License.
     Section 6 states terms for distribution of such executables.

       When a "work that uses the Library" uses material from a header file
     that is part of the Library, the object code for the work may be a
     derivative work of the Library even though the source code is not.
     Whether this is true is especially significant if the work can be
     linked without the Library, or if the work is itself a library.  The
     threshold for this to be true is not precisely defined by law.

       If such an object file uses only numerical parameters, data
     structure layouts and accessors, and small macros and small inline
     functions (ten lines or less in length), then the use of the object
     file is unrestricted, regardless of whether it is legally a derivative
     work.  (Executables containing this object code plus portions of the
     Library will still fall under Section 6.)

       Otherwise, if the work is a derivative of the Library, you may
     distribute the object code for the work under the terms of Section 6.
     Any executables containing that work also fall under Section 6,
     whether or not they are linked directly with the Library itself.

       6. As an exception to the Sections above, you may also combine or
     link a "work that uses the Library" with the Library to produce a
     work containing portions of the Library, and distribute that work
     under terms of your choice, provided that the terms permit
     modification of the work for the customer's own use and reverse
     engineering for debugging such modifications.

       You must give prominent notice with each copy of the work that the
     Library is used in it and that the Library and its use are covered by
     this License.  You must supply a copy of this License.  If the work
     during execution displays copyright notices, you must include the
     copyright notice for the Library among them, as well as a reference
     directing the user to the copy of this License.  Also, you must do one
     of these things:

         a) Accompany the work with the complete corresponding
         machine-readable source code for the Library including whatever
         changes were used in the work (which must be distributed under
         Sections 1 and 2 above); and, if the work is an executable linked
         with the Library, with the complete machine-readable "work that
         uses the Library", as object code and/or source code, so that the
         user can modify the Library and then relink to produce a modified
         executable containing the modified Library.  (It is understood
         that the user who changes the contents of definitions files in the
         Library will not necessarily be able to recompile the application
         to use the modified definitions.)

         b) Use a suitable shared library mechanism for linking with the
         Library.  A suitable mechanism is one that (1) uses at run time a
         copy of the library already present on the user's computer system,
         rather than copying library functions into the executable, and (2)
         will operate properly with a modified version of the library, if
         the user installs one, as long as the modified version is
         interface-compatible with the version that the work was made with.

         c) Accompany the work with a written offer, valid for at
         least three years, to give the same user the materials
         specified in Subsection 6a, above, for a charge no more
         than the cost of performing this distribution.

         d) If distribution of the work is made by offering access to copy
         from a designated place, offer equivalent access to copy the above
         specified materials from the same place.

         e) Verify that the user has already received a copy of these
         materials or that you have already sent this user a copy.

       For an executable, the required form of the "work that uses the
     Library" must include any data and utility programs needed for
     reproducing the executable from it.  However, as a special exception,
     the materials to be distributed need not include anything that is
     normally distributed (in either source or binary form) with the major
     components (compiler, kernel, and so on) of the operating system on
     which the executable runs, unless that component itself accompanies
     the executable.

       It may happen that this requirement contradicts the license
     restrictions of other proprietary libraries that do not normally
     accompany the operating system.  Such a contradiction means you cannot
     use both them and the Library together in an executable that you
     distribute.

       7. You may place library facilities that are a work based on the
     Library side-by-side in a single library together with other library
     facilities not covered by this License, and distribute such a combined
     library, provided that the separate distribution of the work based on
     the Library and of the other library facilities is otherwise
     permitted, and provided that you do these two things:

         a) Accompany the combined library with a copy of the same work
         based on the Library, uncombined with any other library
         facilities.  This must be distributed under the terms of the
         Sections above.

         b) Give prominent notice with the combined library of the fact
         that part of it is a work based on the Library, and explaining
         where to find the accompanying uncombined form of the same work.

       8. You may not copy, modify, sublicense, link with, or distribute
     the Library except as expressly provided under this License.  Any
     attempt otherwise to copy, modify, sublicense, link with, or
     distribute the Library is void, and will automatically terminate your
     rights under this License.  However, parties who have received copies,
     or rights, from you under this License will not have their licenses
     terminated so long as such parties remain in full compliance.

       9. You are not required to accept this License, since you have not
     signed it.  However, nothing else grants you permission to modify or
     distribute the Library or its derivative works.  These actions are
     prohibited by law if you do not accept this License.  Therefore, by
     modifying or distributing the Library (or any work based on the
     Library), you indicate your acceptance of this License to do so, and
     all its terms and conditions for copying, distributing or modifying
     the Library or works based on it.

       10. Each time you redistribute the Library (or any work based on the
     Library), the recipient automatically receives a license from the
     original licensor to copy, distribute, link with or modify the Library
     subject to these terms and conditions.  You may not impose any further
     restrictions on the recipients' exercise of the rights granted herein.
     You are not responsible for enforcing compliance by third parties with
     this License.

       11. If, as a consequence of a court judgment or allegation of patent
     infringement or for any other reason (not limited to patent issues),
     conditions are imposed on you (whether by court order, agreement or
     otherwise) that contradict the conditions of this License, they do not
     excuse you from the conditions of this License.  If you cannot
     distribute so as to satisfy simultaneously your obligations under this
     License and any other pertinent obligations, then as a consequence you
     may not distribute the Library at all.  For example, if a patent
     license would not permit royalty-free redistribution of the Library by
     all those who receive copies directly or indirectly through you, then
     the only way you could satisfy both it and this License would be to
     refrain entirely from distribution of the Library.

     If any portion of this section is held invalid or unenforceable under any
     particular circumstance, the balance of the section is intended to apply,
     and the section as a whole is intended to apply in other circumstances.

     It is not the purpose of this section to induce you to infringe any
     patents or other property right claims or to contest validity of any
     such claims; this section has the sole purpose of protecting the
     integrity of the free software distribution system which is
     implemented by public license practices.  Many people have made
     generous contributions to the wide range of software distributed
     through that system in reliance on consistent application of that
     system; it is up to the author/donor to decide if he or she is willing
     to distribute software through any other system and a licensee cannot
     impose that choice.

     This section is intended to make thoroughly clear what is believed to
     be a consequence of the rest of this License.

       12. If the distribution and/or use of the Library is restricted in
     certain countries either by patents or by copyrighted interfaces, the
     original copyright holder who places the Library under this License may add
     an explicit geographical distribution limitation excluding those countries,
     so that distribution is permitted only in or among countries not thus
     excluded.  In such case, this License incorporates the limitation as if
     written in the body of this License.

       13. The Free Software Foundation may publish revised and/or new
     versions of the Lesser General Public License from time to time.
     Such new versions will be similar in spirit to the present version,
     but may differ in detail to address new problems or concerns.

     Each version is given a distinguishing version number.  If the Library
     specifies a version number of this License which applies to it and
     "any later version", you have the option of following the terms and
     conditions either of that version or of any later version published by
     the Free Software Foundation.  If the Library does not specify a
     license version number, you may choose any version ever published by
     the Free Software Foundation.

       14. If you wish to incorporate parts of the Library into other free
     programs whose distribution conditions are incompatible with these,
     write to the author to ask for permission.  For software which is
     copyrighted by the Free Software Foundation, write to the Free
     Software Foundation; we sometimes make exceptions for this.  Our
     decision will be guided by the two goals of preserving the free status
     of all derivatives of our free software and of promoting the sharing
     and reuse of software generally.

                                 NO WARRANTY

       15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
     WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
     EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
     OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
     KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
     IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
     PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
     LIBRARY IS WITH YOU.  SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
     THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

       16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
     WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
     AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
     FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
     CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
     LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
     RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
     FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
     SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
     DAMAGES.

                          END OF TERMS AND CONDITIONS

                How to Apply These Terms to Your New Libraries

       If you develop a new library, and you want it to be of the greatest
     possible use to the public, we recommend making it free software that
     everyone can redistribute and change.  You can do so by permitting
     redistribution under these terms (or, alternatively, under the terms of the
     ordinary General Public License).

       To apply these terms, attach the following notices to the library.  It is
     safest to attach them to the start of each source file to most effectively
     convey the exclusion of warranty; and each file should have at least the
     "copyright" line and a pointer to where the full notice is found.

         <one line to give the library's name and a brief idea of what it does.>
         Copyright (C) <year>  <name of author>

         This library is free software; you can redistribute it and/or
         modify it under the terms of the GNU Lesser General Public
         License as published by the Free Software Foundation; either
         version 2.1 of the License, or (at your option) any later version.

         This library is distributed in the hope that it will be useful,
         but WITHOUT ANY WARRANTY; without even the implied warranty of
         MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
         Lesser General Public License for more details.

         You should have received a copy of the GNU Lesser General Public
         License along with this library; if not, write to the Free Software
         Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA

     Also add information on how to contact you by electronic and paper mail.

     You should also get your employer (if you work as a programmer) or your
     school, if any, to sign a "copyright disclaimer" for the library, if
     necessary.  Here is a sample; alter the names:

       Yoyodyne, Inc., hereby disclaims all copyright interest in the
       library `Frob' (a library for tweaking knobs) written by James Random Hacker.

       <signature of Ty Coon>, 1 April 1990
       Ty Coon, President of Vice

     That's all there is to it!

General index
*************

+z:                                            See 3.4.4.   (line  1003)
,(c EXPR):                                     See 5.2.     (line  1411)
,+:                                            See 5.2.     (line  1460)
,-:                                            See 5.2.     (line  1466)
,?:                                            See 5.2.     (line  1396)
,b:                                            See 5.2.     (line  1488)
,c:                                            See 5.2.     (line  1428)
,d:                                            See 5.2.     (line  1407)
,e:                                            See 5.2.     (line  1503)
,i:                                            See 5.2.     (line  1497)
,l:                                            See 5.2.     (line  1448)
,N:                                            See 5.2.     (line  1455)
,q:                                            See 5.2.     (line  1399)
,s:                                            See 5.2.     (line  1434)
,t:                                            See 5.2.     (line  1404)
,y:                                            See 5.2.     (line  1472)
-:                                             See 3.3.     (line   691)
-:+:                                           See 4.       (line  1253)
-:-:                                           See 4.       (line  1256)
-:=:                                           See 4.       (line  1250)
-:d:                                           See 4.       (line  1206)
-:d-:                                          See 4.       (line  1235)
-:da:                                          See 4.       (line  1214)
-:dc:                                          See 4.       (line  1232)
-:di:                                          See 4.       (line  1228)
-:dLEVEL:                                      See 4.       (line  1239)
-:dp:                                          See 4.       (line  1210)
-:dq:                                          See 4.       (line  1224)
-:dr:                                          See 4.       (line  1217)
-:ds:                                          See 4.       (line  1220)
-:f:                                           See 4.       (line  1256)
-:h:                                           See 4.       (line  1187)
-:l:                                           See 4.       (line  1193)
-:m:                                           See 4.       (line  1182)
-:S:                                           See 4.       (line  1200)
-:s:                                           See 4.       (line  1200)
-:t:                                           See 4.       (line  1256)
-c:                                            See 3.3.     (line   668)
-call_shared:                                  See 3.4.4.   (line  1003)
-cc-options:                                   See 3.3.     (line   613)
-D___DYNAMIC:                                  See 3.4.2.   (line   818)
-D___LIBRARY:                                  See 3.4.3.   (line   950)
-D___PRIMAL:                                   See 3.4.3.   (line   950)
-D___SHARED:                                   See 3.4.3.   (line   950)
-D___SINGLE_HOST:                              See 3.4.4.   (line   982)
-debug:                                        See 3.3.     (line   651)
-dynamic:                                      See 3.3.     (line   668)
-e:                                            See 3.3.     (line   693)
-expansion:                                    See 3.3.     (line   645)
-flat:                                         See 3.3.     (line   678)
-fpic:                                         See 3.4.4.   (line  1003)
-fPIC:                                         See 3.4.4.   (line  1003)
-G:                                            See 3.4.4.   (line  1003)
-gvm:                                          See 3.3.     (line   648)
-i:                                            See 3.3.     (line   595)
-I/usr/local/Gambit-C/current/include:         See 3.4.4.   (line   994)
-Kpic:                                         See 3.4.4.   (line  1003)
-KPIC:                                         See 3.4.4.   (line  1003)
-l BASE:                                       See 3.3.     (line   684)
-L/usr/local/Gambit-C/current/lib:             See 3.4.4.   (line   994)
-ld-options:                                   See 3.3.     (line   623)
-link:                                         See 3.3.     (line   519)
-O:                                            See 3.4.4.   (line   982)
-o OUTPUT:                                     See 3.3.     (line   662)
-pic:                                          See 3.4.4.   (line  1003)
-postlude:                                     See 3.3.     (line   606)
-prelude:                                      See 3.3.     (line   598)
-rdynamic:                                     See 3.4.4.   (line  1003)
-report:                                       See 3.3.     (line   640)
-shared:                                       See 3.4.4.   (line  1003)
-track-scheme:                                 See 3.3.     (line   658)
-verbose:                                      See 3.3.     (line   637)
-warnings:                                     See 3.3.     (line   634)
.c:                                            See 3.3.     (line   498)
.scm:                                          See 3.3.     (line   498)
.six:                                          See 3.3.     (line   498)
<:                                             See 9.1.     (line  2812)
<=:                                            See 9.1.     (line  2814)
=:                                             See 9.1.     (line  2811)
>:                                             See 9.1.     (line  2813)
>=:                                            See 9.1.     (line  2815)
^C:                                            See 5.1.     (line  1333)
^D:                                            See 5.1.     (line  1359)
___cleanup:                                    See 19.5.    (line 11120)
___setup:                                      See 19.5.    (line 11120)
abandoned-mutex-exception?:                    See 15.4.    (line  5947)
abort:                                         See 15.1.    (line  5718)
absolute path:                                 See 16.1.    (line  6898)
all-bits-set?:                                 See 9.4.     (line  3053)
any-bits-set?:                                 See 9.4.     (line  3038)
arithmetic-shift:                              See 9.4.     (line  2887)
bit-count:                                     See 9.4.     (line  2980)
bit-set?:                                      See 9.4.     (line  3026)
bitwise-and:                                   See 9.4.     (line  2918)
bitwise-ior:                                   See 9.4.     (line  2934)
bitwise-merge:                                 See 9.4.     (line  2902)
bitwise-not:                                   See 9.4.     (line  2968)
bitwise-xor:                                   See 9.4.     (line  2951)
block:                                         See 6.3.     (line  2491)
box:                                           See 6.3.     (line  2270)
box?:                                          See 6.3.     (line  2271)
boxes:                                         See 6.3.     (line  2273)
break:                                         See 5.3.     (line  1739)
c-declare:                                     See 19.2.    (line 10903)
c-define:                                      See 19.5.    (line 11079)
c-define-type:                                 See 19.6.    (line 11134)
c-initialize:                                  See 19.3.    (line 10947)
c-lambda:                                      See 19.4.    (line 10971)
call-with-current-continuation:                See 6.1.     (line  2061)
call-with-input-file:                          See 17.7.1.  (line  8776)
call-with-input-string:                        See 17.10.   (line  9364)
call-with-input-u8vector:                      See 17.11.   (line  9393)
call-with-input-vector:                        See 17.9.    (line  9221)
call-with-output-file:                         See 17.7.1.  (line  8777)
call-with-output-string:                       See 17.10.   (line  9365)
call-with-output-u8vector:                     See 17.11.   (line  9394)
call-with-output-vector:                       See 17.9.    (line  9222)
call/cc:                                       See 6.1.     (line  2062)
cfun-conversion-exception-arguments:           See 15.5.    (line  6121)
cfun-conversion-exception-code:                See 15.5.    (line  6122)
cfun-conversion-exception-message:             See 15.5.    (line  6123)
cfun-conversion-exception-procedure:           See 15.5.    (line  6120)
cfun-conversion-exception?:                    See 15.5.    (line  6119)
char->integer:                                 See 8.1.     (line  2750)
char-ci<=?:                                    See 8.1.     (line  2776)
char-ci<?:                                     See 8.1.     (line  2774)
char-ci=?:                                     See 8.1.     (line  2773)
char-ci>=?:                                    See 8.1.     (line  2777)
char-ci>?:                                     See 8.1.     (line  2775)
char<=?:                                       See 8.1.     (line  2771)
char<?:                                        See 8.1.     (line  2769)
char=?:                                        See 8.1.     (line  2768)
char>=?:                                       See 8.1.     (line  2772)
char>?:                                        See 8.1.     (line  2770)
clear-bit-field:                               See 9.4.     (line  3084)
close-input-port:                              See 17.4.2.  (line  8257)
close-output-port:                             See 17.4.2.  (line  8258)
close-port:                                    See 17.4.2.  (line  8259)
command-line <1>:                              See 16.5.    (line  7179)
command-line:                                  See 2.5.     (line   346)
compile-file:                                  See 3.5.     (line  1030)
compile-file-to-c:                             See 3.5.     (line  1016)
compiler:                                      See 3.       (line   470)
compiler options:                              See 3.3.     (line   527)
condition-variable-broadcast!:                 See 13.9.    (line  5343)
condition-variable-name:                       See 13.9.    (line  5264)
condition-variable-signal!:                    See 13.9.    (line  5289)
condition-variable-specific:                   See 13.9.    (line  5272)
condition-variable-specific-set!:              See 13.9.    (line  5273)
condition-variable?:                           See 13.9.    (line  5239)
console-port:                                  See 6.4.     (line  2661)
constant-fold:                                 See 6.3.     (line  2530)
continuations:                                 See 19.7.    (line 11568)
copy-bit-field:                                See 9.4.     (line  3086)
copy-file:                                     See 16.2.    (line  7087)
cpu-time:                                      See 16.7.    (line  7281)
create-directory:                              See 16.2.    (line  7006)
create-fifo:                                   See 16.2.    (line  7033)
create-link:                                   See 16.2.    (line  7064)
create-symbolic-link:                          See 16.2.    (line  7072)
current exception-handler:                     See 15.1.    (line  5630)
current working directory:                     See 16.1.    (line  6833)
current-directory:                             See 16.1.    (line  6871)
current-error-port:                            See 17.12.   (line  9410)
current-exception-handler:                     See 15.1.    (line  5630)
current-input-port:                            See 17.12.   (line  9408)
current-output-port:                           See 17.12.   (line  9409)
current-readtable:                             See 17.12.   (line  9411)
current-thread:                                See 13.9.    (line  4558)
current-time:                                  See 16.7.    (line  7245)
current-user-interrupt-handler:                See 6.4.     (line  2663)
datum-parsing-exception-kind:                  See 15.6.    (line  6287)
datum-parsing-exception-parameters:            See 15.6.    (line  6288)
datum-parsing-exception-readenv:               See 15.6.    (line  6289)
datum-parsing-exception?:                      See 15.6.    (line  6286)
deadlock-exception?:                           See 15.4.    (line  5926)
declare:                                       See 6.3.     (line  2480)
default-random-source:                         See 9.7.     (line  3325)
define <1>:                                    See 20.      (line 11620)
define:                                        See 6.2.     (line  2071)
define-macro:                                  See 6.3.     (line  2422)
define-structure:                              See 12.      (line  4279)
define-syntax:                                 See 6.3.     (line  2454)
define-type-of-thread:                         See 6.4.     (line  2623)
delete-directory:                              See 16.2.    (line  7099)
delete-file:                                   See 16.2.    (line  7094)
deserialization <1>:                           See 10.      (line  3636)
deserialization:                               See 18.1.    (line  9543)
directory-files:                               See 16.2.    (line  7104)
display-environment-set!:                      See 5.3.     (line  1826)
divide-by-zero-exception-arguments:            See 15.8.    (line  6558)
divide-by-zero-exception-procedure:            See 15.8.    (line  6557)
divide-by-zero-exception?:                     See 15.8.    (line  6556)
Emacs:                                         See 5.5.     (line  1958)
eq?-hash:                                      See 11.1.    (line  3805)
equal?-hash:                                   See 11.1.    (line  3829)
eqv?-hash:                                     See 11.1.    (line  3817)
err-code->string:                              See 6.4.     (line  2667)
error:                                         See 15.10.   (line  6788)
error-exception-message:                       See 15.10.   (line  6786)
error-exception-parameters:                    See 15.10.   (line  6787)
error-exception?:                              See 15.10.   (line  6785)
eval:                                          See 6.3.     (line  2389)
exit:                                          See 16.4.    (line  7161)
expression-parsing-exception-kind:             See 15.7.    (line  6362)
expression-parsing-exception-parameters:       See 15.7.    (line  6363)
expression-parsing-exception-source:           See 15.7.    (line  6364)
expression-parsing-exception?:                 See 15.7.    (line  6361)
extended-bindings:                             See 6.3.     (line  2538)
extract-bit-field:                             See 9.4.     (line  3082)
f32vector:                                     See 10.      (line  3600)
f32vector->list:                               See 10.      (line  3604)
f32vector-append:                              See 10.      (line  3608)
f32vector-copy:                                See 10.      (line  3607)
f32vector-fill!:                               See 10.      (line  3606)
f32vector-length:                              See 10.      (line  3601)
f32vector-ref:                                 See 10.      (line  3602)
f32vector-set!:                                See 10.      (line  3603)
f32vector?:                                    See 10.      (line  3598)
f64vector:                                     See 10.      (line  3613)
f64vector->list:                               See 10.      (line  3617)
f64vector-append:                              See 10.      (line  3621)
f64vector-copy:                                See 10.      (line  3620)
f64vector-fill!:                               See 10.      (line  3619)
f64vector-length:                              See 10.      (line  3614)
f64vector-ref:                                 See 10.      (line  3615)
f64vector-set!:                                See 10.      (line  3616)
f64vector?:                                    See 10.      (line  3611)
FFI:                                           See 19.      (line 10344)
file names:                                    See 16.1.    (line  6833)
file-attributes:                               See 16.8.    (line  7560)
file-creation-time:                            See 16.8.    (line  7561)
file-device:                                   See 16.8.    (line  7550)
file-exists?:                                  See 16.8.    (line  7339)
file-group:                                    See 16.8.    (line  7555)
file-info:                                     See 16.8.    (line  7354)
file-info-attributes:                          See 16.8.    (line  7529)
file-info-creation-time:                       See 16.8.    (line  7539)
file-info-device:                              See 16.8.    (line  7435)
file-info-group:                               See 16.8.    (line  7481)
file-info-inode:                               See 16.8.    (line  7444)
file-info-last-access-time:                    See 16.8.    (line  7499)
file-info-last-change-time:                    See 16.8.    (line  7519)
file-info-last-modification-time:              See 16.8.    (line  7509)
file-info-mode:                                See 16.8.    (line  7453)
file-info-number-of-links:                     See 16.8.    (line  7462)
file-info-owner:                               See 16.8.    (line  7472)
file-info-size:                                See 16.8.    (line  7490)
file-info-type:                                See 16.8.    (line  7399)
file-info?:                                    See 16.8.    (line  7387)
file-inode:                                    See 16.8.    (line  7551)
file-last-access-time:                         See 16.8.    (line  7557)
file-last-change-time:                         See 16.8.    (line  7559)
file-last-modification-time:                   See 16.8.    (line  7558)
file-mode:                                     See 16.8.    (line  7552)
file-number-of-links:                          See 16.8.    (line  7553)
file-owner:                                    See 16.8.    (line  7554)
file-size:                                     See 16.8.    (line  7556)
file-type:                                     See 16.8.    (line  7549)
FILE.c:                                        See 3.3.     (line   498)
FILE.scm:                                      See 3.3.     (line   498)
FILE.six:                                      See 3.3.     (line   498)
first-bit-set:                                 See 9.4.     (line  3068)
fixnum:                                        See 6.3.     (line  2563)
fixnum->flonum:                                See 9.6.     (line  3237)
fixnum-overflow-exception-arguments:           See 9.5.     (line  3204)
fixnum-overflow-exception-procedure:           See 9.5.     (line  3203)
fixnum-overflow-exception?:                    See 9.5.     (line  3202)
fixnum?:                                       See 9.5.     (line  3126)
fl*:                                           See 9.6.     (line  3239)
fl+:                                           See 9.6.     (line  3241)
fl-:                                           See 9.6.     (line  3243)
fl/:                                           See 9.6.     (line  3245)
fl<:                                           See 9.6.     (line  3247)
fl<=:                                          See 9.6.     (line  3249)
fl=:                                           See 9.6.     (line  3251)
fl>:                                           See 9.6.     (line  3253)
fl>=:                                          See 9.6.     (line  3255)
flabs:                                         See 9.6.     (line  3257)
flacos:                                        See 9.6.     (line  3259)
flasin:                                        See 9.6.     (line  3261)
flatan:                                        See 9.6.     (line  3264)
flceiling:                                     See 9.6.     (line  3266)
flcos:                                         See 9.6.     (line  3268)
fldenominator:                                 See 9.6.     (line  3270)
fleven?:                                       See 9.6.     (line  3272)
flexp:                                         See 9.6.     (line  3274)
flexpt:                                        See 9.6.     (line  3276)
flfinite?:                                     See 9.6.     (line  3278)
flfloor:                                       See 9.6.     (line  3280)
flinfinite?:                                   See 9.6.     (line  3282)
flinteger?:                                    See 9.6.     (line  3284)
fllog:                                         See 9.6.     (line  3286)
flmax:                                         See 9.6.     (line  3288)
flmin:                                         See 9.6.     (line  3290)
flnan?:                                        See 9.6.     (line  3292)
flnegative?:                                   See 9.6.     (line  3294)
flnumerator:                                   See 9.6.     (line  3296)
floating point overflow:                       See 20.      (line 11613)
flodd?:                                        See 9.6.     (line  3298)
flonum:                                        See 6.3.     (line  2563)
flonum?:                                       See 9.6.     (line  3235)
flpositive?:                                   See 9.6.     (line  3300)
flround:                                       See 9.6.     (line  3302)
flsin:                                         See 9.6.     (line  3304)
flsqrt:                                        See 9.6.     (line  3306)
fltan:                                         See 9.6.     (line  3308)
fltruncate:                                    See 9.6.     (line  3310)
flzero?:                                       See 9.6.     (line  3312)
force-output:                                  See 17.4.2.  (line  8240)
foreign function interface:                    See 19.      (line 10344)
foreign-address:                               See 6.4.     (line  2669)
foreign-release!:                              See 6.4.     (line  2670)
foreign-released?:                             See 6.4.     (line  2671)
fx*:                                           See 9.5.     (line  3128)
fx+:                                           See 9.5.     (line  3130)
fx-:                                           See 9.5.     (line  3132)
fx<:                                           See 9.5.     (line  3134)
fx<=:                                          See 9.5.     (line  3136)
fx=:                                           See 9.5.     (line  3138)
fx>:                                           See 9.5.     (line  3140)
fx>=:                                          See 9.5.     (line  3142)
fxand:                                         See 9.5.     (line  3144)
fxarithmetic-shift:                            See 9.5.     (line  3146)
fxarithmetic-shift-left:                       See 9.5.     (line  3148)
fxarithmetic-shift-right:                      See 9.5.     (line  3150)
fxbit-count:                                   See 9.5.     (line  3152)
fxbit-set?:                                    See 9.5.     (line  3154)
fxeven?:                                       See 9.5.     (line  3156)
fxfirst-bit-set:                               See 9.5.     (line  3158)
fxif:                                          See 9.5.     (line  3160)
fxior:                                         See 9.5.     (line  3162)
fxlength:                                      See 9.5.     (line  3164)
fxmax:                                         See 9.5.     (line  3166)
fxmin:                                         See 9.5.     (line  3168)
fxmodulo:                                      See 9.5.     (line  3170)
fxnegative?:                                   See 9.5.     (line  3172)
fxnot:                                         See 9.5.     (line  3174)
fxodd?:                                        See 9.5.     (line  3176)
fxpositive?:                                   See 9.5.     (line  3178)
fxquotient:                                    See 9.5.     (line  3180)
fxremainder:                                   See 9.5.     (line  3182)
fxwrap*:                                       See 9.5.     (line  3184)
fxwrap+:                                       See 9.5.     (line  3186)
fxwrap-:                                       See 9.5.     (line  3188)
fxwraparithmetic-shift:                        See 9.5.     (line  3190)
fxwraparithmetic-shift-left:                   See 9.5.     (line  3192)
fxwraplogical-shift-right:                     See 9.5.     (line  3194)
fxwrapquotient:                                See 9.5.     (line  3196)
fxxor:                                         See 9.5.     (line  3198)
fxzero?:                                       See 9.5.     (line  3200)
GAMBCOPT, environment variable:                See 4.       (line  1305)
Gambit:                                        See 1.       (line    33)
Gambit installation directory:                 See 16.1.    (line  6833)
Gambit-C <1>:
          See ``Gambit-C''.                                 (line    28)
Gambit-C:                                      See 1.       (line    33)
gambit.el:                                     See 5.5.     (line  1958)
GC:                                            See 5.3.     (line  1862)
gc-report-set!:                                See 5.3.     (line  1862)
generate-proper-tail-calls:                    See 5.3.     (line  1789)
generic:                                       See 6.3.     (line  2563)
gensym:                                        See 6.3.     (line  2323)
get-output-string:                             See 17.10.   (line  9370)
get-output-u8vector:                           See 17.11.   (line  9399)
get-output-vector:                             See 17.9.    (line  9339)
getenv:                                        See 16.6.    (line  7200)
group-info:                                    See 16.9.    (line  7571)
group-info-gid:                                See 16.9.    (line  7615)
group-info-members:                            See 16.9.    (line  7625)
group-info-name:                               See 16.9.    (line  7605)
group-info?:                                   See 16.9.    (line  7593)
gsc <1>:                                       See 4.       (line  1143)
gsc <2>:                                       See 3.5.     (line  1030)
gsc <3>:                                       See 1.       (line    33)
gsc <4>:                                       See 3.3.     (line   498)
gsc:                                           See 3.5.     (line  1086)
gsc-script:                                    See 2.5.     (line   329)
gsi <1>:                                       See 4.       (line  1143)
gsi <2>:                                       See 2.       (line   107)
gsi:                                           See 1.       (line    33)
gsi-script:                                    See 2.5.     (line   325)
hashing:                                       See 11.      (line  3684)
heap-overflow-exception?:                      See 15.2.    (line  5757)
home directory:                                See 16.1.    (line  6833)
homogeneous vectors <1>:                       See 10.      (line  3473)
homogeneous vectors:                           See 18.9.    (line 10004)
host-info:                                     See 16.11.   (line  7743)
host-info-addresses:                           See 16.11.   (line  7795)
host-info-aliases:                             See 16.11.   (line  7785)
host-info-name:                                See 16.11.   (line  7775)
host-info?:                                    See 16.11.   (line  7763)
host-name:                                     See 16.11.   (line  7734)
ieee-scheme:                                   See 6.3.     (line  2487)
improper-length-list-exception-arg-num:        See 15.8.    (line  6588)
improper-length-list-exception-arguments:      See 15.8.    (line  6587)
improper-length-list-exception-procedure:      See 15.8.    (line  6586)
improper-length-list-exception?:               See 15.8.    (line  6585)
include:                                       See 6.3.     (line  2422)
initialized-thread-exception-arguments:        See 6.4.     (line  2629)
initialized-thread-exception-procedure:        See 6.4.     (line  2628)
initialized-thread-exception?:                 See 6.4.     (line  2627)
inline:                                        See 6.3.     (line  2498)
inline-primitives:                             See 6.3.     (line  2501)
inlining-limit:                                See 6.3.     (line  2505)
input-port-byte-position:                      See 17.7.1.  (line  8853)
input-port-bytes-buffered:                     See 6.4.     (line  2649)
input-port-char-position:                      See 6.4.     (line  2723)
input-port-characters-buffered:                See 6.4.     (line  2651)
input-port-column:                             See 17.5.2.  (line  8370)
input-port-line:                               See 17.5.2.  (line  8369)
input-port-readtable:                          See 17.5.2.  (line  8511)
input-port-readtable-set!:                     See 17.5.2.  (line  8519)
input-port-timeout-set!:                       See 17.4.2.  (line  8286)
input-port?:                                   See 17.4.2.  (line  8149)
integer->char:                                 See 8.1.     (line  2751)
integer-length:                                See 9.4.     (line  3003)
integer-nth-root:                              See 9.3.     (line  2862)
integer-sqrt:                                  See 9.3.     (line  2852)
interpreter <1>:                               See 3.       (line   470)
interpreter:                                   See 2.       (line   103)
interrupts-enabled:                            See 6.3.     (line  2557)
invalid-hash-number-exception-arguments:       See 6.4.     (line  2675)
invalid-hash-number-exception-procedure:       See 6.4.     (line  2674)
invalid-hash-number-exception?:                See 6.4.     (line  2673)
join-timeout-exception-arguments:              See 15.4.    (line  5975)
join-timeout-exception-procedure:              See 15.4.    (line  5974)
join-timeout-exception?:                       See 15.4.    (line  5973)
keyword->string:                               See 6.3.     (line  2299)
keyword-expected-exception-arguments:          See 15.9.    (line  6754)
keyword-expected-exception-procedure:          See 15.9.    (line  6753)
keyword-expected-exception?:                   See 15.9.    (line  6752)
keyword-hash:                                  See 11.1.    (line  3769)
keyword?:                                      See 6.3.     (line  2298)
keywords:                                      See 6.3.     (line  2300)
lambda:                                        See 6.2.     (line  2070)
lambda-lift:                                   See 6.3.     (line  2527)
LAST_.c:                                       See 3.3.     (line   668)
limitations:                                   See 20.      (line 11613)
link-flat:                                     See 3.5.     (line  1086)
link-incremental:                              See 3.5.     (line  1050)
list->f32vector:                               See 10.      (line  3605)
list->f64vector:                               See 10.      (line  3618)
list->s16vector:                               See 10.      (line  3527)
list->s32vector:                               See 10.      (line  3553)
list->s64vector:                               See 10.      (line  3579)
list->s8vector:                                See 10.      (line  3501)
list->table:                                   See 11.2.2.  (line  4210)
list->u16vector:                               See 10.      (line  3540)
list->u32vector:                               See 10.      (line  3566)
list->u64vector:                               See 10.      (line  3592)
list->u8vector:                                See 10.      (line  3514)
load:                                          See 3.5.     (line  1030)
mailbox-receive-timeout-exception-arguments:   See 13.9.    (line  4968)
mailbox-receive-timeout-exception-procedure:   See 13.9.    (line  4967)
mailbox-receive-timeout-exception?:            See 13.9.    (line  4966)
main:                                          See 6.4.     (line  2728)
make-condition-variable:                       See 13.9.    (line  5251)
make-f32vector:                                See 10.      (line  3599)
make-f64vector:                                See 10.      (line  3612)
make-mutex:                                    See 13.9.    (line  5010)
make-parameter:                                See 14.      (line  5462)
make-random-source:                            See 9.7.     (line  3360)
make-s16vector:                                See 10.      (line  3521)
make-s32vector:                                See 10.      (line  3547)
make-s64vector:                                See 10.      (line  3573)
make-s8vector:                                 See 10.      (line  3495)
make-table:                                    See 11.2.2.  (line  3995)
make-thread:                                   See 13.9.    (line  4579)
make-thread-group:                             See 6.4.     (line  2610)
make-u16vector:                                See 10.      (line  3534)
make-u32vector:                                See 10.      (line  3560)
make-u64vector:                                See 10.      (line  3586)
make-u8vector:                                 See 10.      (line  3508)
make-uninterned-keyword:                       See 6.3.     (line  2363)
make-uninterned-symbol:                        See 6.3.     (line  2340)
make-will:                                     See 11.2.1.  (line  3873)
mostly-fixnum:                                 See 6.3.     (line  2569)
mostly-fixnum-flonum:                          See 6.3.     (line  2569)
mostly-flonum:                                 See 6.3.     (line  2569)
mostly-flonum-fixnum:                          See 6.3.     (line  2569)
mostly-generic:                                See 6.3.     (line  2569)
multiple-c-return-exception?:                  See 15.5.    (line  6240)
mutex-lock!:                                   See 13.9.    (line  5093)
mutex-name:                                    See 13.9.    (line  5025)
mutex-specific:                                See 13.9.    (line  5032)
mutex-specific-set!:                           See 13.9.    (line  5033)
mutex-state:                                   See 13.9.    (line  5065)
mutex-unlock!:                                 See 13.9.    (line  5189)
mutex?:                                        See 13.9.    (line  4998)
network-info <1>:                              See 16.14.   (line  7953)
network-info:                                  See 6.4.     (line  2677)
network-info-aliases <1>:                      See 6.4.     (line  2681)
network-info-aliases:                          See 16.14.   (line  8002)
network-info-name <1>:                         See 6.4.     (line  2679)
network-info-name:                             See 16.14.   (line  7992)
network-info-net:                              See 6.4.     (line  2680)
network-info-number:                           See 16.14.   (line  8012)
network-info? <1>:                             See 16.14.   (line  7980)
network-info?:                                 See 6.4.     (line  2678)
newline:                                       See 17.4.2.  (line  8218)
no-such-file-or-directory-exception-arguments: See 15.3.    (line  5846)
no-such-file-or-directory-exception-procedure: See 15.3.    (line  5845)
no-such-file-or-directory-exception?:          See 15.3.    (line  5844)
noncontinuable-exception-reason:               See 15.1.    (line  5720)
noncontinuable-exception?:                     See 15.1.    (line  5719)
nonempty-input-port-character-buffer-exception-arguments:See 6.4.
                                                            (line  2655)
nonempty-input-port-character-buffer-exception-procedure:See 6.4.
                                                            (line  2657)
nonempty-input-port-character-buffer-exception?:See 6.4.    (line  2653)
nonprocedure-operator-exception-arguments:     See 15.9.    (line  6690)
nonprocedure-operator-exception-code:          See 15.9.    (line  6691)
nonprocedure-operator-exception-operator:      See 15.9.    (line  6689)
nonprocedure-operator-exception-rte:           See 15.9.    (line  6692)
nonprocedure-operator-exception?:              See 15.9.    (line  6688)
normalized path:                               See 16.1.    (line  6926)
number-of-arguments-limit-exception-arguments: See 15.9.    (line  6656)
number-of-arguments-limit-exception-procedure: See 15.9.    (line  6655)
number-of-arguments-limit-exception?:          See 15.9.    (line  6654)
object file:                                   See 3.5.     (line  1030)
object->serial-number:                         See 11.1.    (line  3688)
object->string:                                See 17.10.   (line  9378)
object->u8vector:                              See 10.      (line  3635)
open-directory:                                See 17.8.    (line  9154)
open-dummy:                                    See 6.4.     (line  2645)
open-event-queue:                              See 6.4.     (line  2726)
open-file:                                     See 17.7.1.  (line  8773)
open-input-file:                               See 17.7.1.  (line  8774)
open-input-string:                             See 17.10.   (line  9362)
open-input-u8vector:                           See 17.11.   (line  9391)
open-input-vector:                             See 17.9.    (line  9219)
open-output-file:                              See 17.7.1.  (line  8775)
open-output-string:                            See 17.10.   (line  9363)
open-output-u8vector:                          See 17.11.   (line  9392)
open-output-vector:                            See 17.9.    (line  9220)
open-process:                                  See 17.7.2.  (line  8915)
open-string:                                   See 17.10.   (line  9361)
open-string-pipe:                              See 17.10.   (line  9369)
open-tcp-client:                               See 17.7.3.  (line  8998)
open-tcp-server:                               See 17.7.3.  (line  9066)
open-u8vector:                                 See 17.11.   (line  9390)
open-u8vector-pipe:                            See 17.11.   (line  9398)
open-vector:                                   See 17.9.    (line  9218)
open-vector-pipe:                              See 17.9.    (line  9295)
options, compiler:                             See 3.3.     (line   527)
options, runtime:                              See 4.       (line  1143)
os-exception-arguments:                        See 15.3.    (line  5806)
os-exception-code:                             See 15.3.    (line  5807)
os-exception-message:                          See 15.3.    (line  5808)
os-exception-procedure:                        See 15.3.    (line  5805)
os-exception?:                                 See 15.3.    (line  5804)
output-port-byte-position:                     See 17.7.1.  (line  8854)
output-port-char-position:                     See 6.4.     (line  2724)
output-port-column:                            See 17.5.2.  (line  8372)
output-port-line:                              See 17.5.2.  (line  8371)
output-port-readtable:                         See 17.5.2.  (line  8512)
output-port-readtable-set!:                    See 17.5.2.  (line  8520)
output-port-timeout-set!:                      See 17.4.2.  (line  8287)
output-port-width:                             See 17.5.2.  (line  8396)
output-port?:                                  See 17.4.2.  (line  8150)
overflow, floating point:                      See 20.      (line 11613)
parameterize:                                  See 14.      (line  5518)
path-directory:                                See 16.1.    (line  6958)
path-expand:                                   See 16.1.    (line  6898)
path-extension:                                See 16.1.    (line  6956)
path-normalize:                                See 16.1.    (line  6926)
path-strip-directory:                          See 16.1.    (line  6959)
path-strip-extension:                          See 16.1.    (line  6957)
path-strip-trailing-directory-separator:       See 16.1.    (line  6960)
path-strip-volume:                             See 16.1.    (line  6962)
path-volume:                                   See 16.1.    (line  6961)
peek-char:                                     See 17.5.2.  (line  8426)
port-settings-set!:                            See 6.4.     (line  2647)
port?:                                         See 17.4.2.  (line  8151)
pp:                                            See 5.3.     (line  1846)
pretty-print:                                  See 5.3.     (line  1833)
primordial-exception-handler:                  See 6.4.     (line  2665)
print:                                         See 6.4.     (line  2607)
println:                                       See 6.4.     (line  2608)
process-pid:                                   See 6.4.     (line  2635)
process-status:                                See 6.4.     (line  2637)
process-times:                                 See 16.7.    (line  7280)
proper tail-calls:                             See 5.3.     (line  1789)
protocol-info <1>:                             See 16.13.   (line  7881)
protocol-info:                                 See 6.4.     (line  2683)
protocol-info-aliases <1>:                     See 16.13.   (line  7930)
protocol-info-aliases:                         See 6.4.     (line  2687)
protocol-info-name <1>:                        See 6.4.     (line  2685)
protocol-info-name:                            See 16.13.   (line  7920)
protocol-info-number <1>:                      See 16.13.   (line  7940)
protocol-info-number:                          See 6.4.     (line  2686)
protocol-info? <1>:                            See 6.4.     (line  2684)
protocol-info?:                                See 16.13.   (line  7908)
r4rs-scheme:                                   See 6.3.     (line  2487)
raise:                                         See 15.1.    (line  5701)
random-integer:                                See 9.7.     (line  3331)
random-real:                                   See 9.7.     (line  3347)
random-source-make-integers:                   See 9.7.     (line  3438)
random-source-make-reals:                      See 9.7.     (line  3455)
random-source-pseudo-randomize!:               See 9.7.     (line  3405)
random-source-randomize!:                      See 9.7.     (line  3404)
random-source-state-ref:                       See 9.7.     (line  3384)
random-source-state-set!:                      See 9.7.     (line  3385)
random-source?:                                See 9.7.     (line  3372)
range-exception-arg-num:                       See 15.8.    (line  6524)
range-exception-arguments:                     See 15.8.    (line  6523)
range-exception-procedure:                     See 15.8.    (line  6522)
range-exception?:                              See 15.8.    (line  6521)
read:                                          See 17.4.2.  (line  8173)
read-all:                                      See 17.4.2.  (line  8187)
read-char:                                     See 17.5.2.  (line  8407)
read-line:                                     See 17.5.2.  (line  8454)
read-substring:                                See 17.5.2.  (line  8482)
read-subu8vector:                              See 17.6.2.  (line  8733)
read-u8:                                       See 17.6.2.  (line  8685)
readtable-case-conversion?:                    See 18.1.    (line  9452)
readtable-case-conversion?-set:                See 18.1.    (line  9453)
readtable-eval-allowed?:                       See 18.1.    (line  9685)
readtable-eval-allowed?-set:                   See 18.1.    (line  9686)
readtable-keywords-allowed?:                   See 18.1.    (line  9499)
readtable-keywords-allowed?-set:               See 18.1.    (line  9500)
readtable-max-write-length:                    See 18.1.    (line  9744)
readtable-max-write-length-set:                See 18.1.    (line  9745)
readtable-max-write-level:                     See 18.1.    (line  9709)
readtable-max-write-level-set:                 See 18.1.    (line  9710)
readtable-sharing-allowed?:                    See 18.1.    (line  9542)
readtable-sharing-allowed?-set:                See 18.1.    (line  9543)
readtable-start-syntax:                        See 18.1.    (line  9779)
readtable-start-syntax-set:                    See 18.1.    (line  9780)
readtable?:                                    See 18.1.    (line  9440)
real-time:                                     See 16.7.    (line  7282)
relative path:                                 See 16.1.    (line  6833)
rename-file:                                   See 16.2.    (line  7080)
repl-input-port:                               See 6.4.     (line  2659)
repl-output-port:                              See 6.4.     (line  2660)
replace-bit-field:                             See 9.4.     (line  3085)
run-time-bindings:                             See 6.3.     (line  2543)
runtime options:                               See 4.       (line  1143)
s16vector:                                     See 10.      (line  3522)
s16vector->list:                               See 10.      (line  3526)
s16vector-append:                              See 10.      (line  3530)
s16vector-copy:                                See 10.      (line  3529)
s16vector-fill!:                               See 10.      (line  3528)
s16vector-length:                              See 10.      (line  3523)
s16vector-ref:                                 See 10.      (line  3524)
s16vector-set!:                                See 10.      (line  3525)
s16vector?:                                    See 10.      (line  3520)
s32vector:                                     See 10.      (line  3548)
s32vector->list:                               See 10.      (line  3552)
s32vector-append:                              See 10.      (line  3556)
s32vector-copy:                                See 10.      (line  3555)
s32vector-fill!:                               See 10.      (line  3554)
s32vector-length:                              See 10.      (line  3549)
s32vector-ref:                                 See 10.      (line  3550)
s32vector-set!:                                See 10.      (line  3551)
s32vector?:                                    See 10.      (line  3546)
s64vector:                                     See 10.      (line  3574)
s64vector->list:                               See 10.      (line  3578)
s64vector-append:                              See 10.      (line  3582)
s64vector-copy:                                See 10.      (line  3581)
s64vector-fill!:                               See 10.      (line  3580)
s64vector-length:                              See 10.      (line  3575)
s64vector-ref:                                 See 10.      (line  3576)
s64vector-set!:                                See 10.      (line  3577)
s64vector?:                                    See 10.      (line  3572)
s8vector:                                      See 10.      (line  3496)
s8vector->list:                                See 10.      (line  3500)
s8vector-append:                               See 10.      (line  3504)
s8vector-copy:                                 See 10.      (line  3503)
s8vector-fill!:                                See 10.      (line  3502)
s8vector-length:                               See 10.      (line  3497)
s8vector-ref:                                  See 10.      (line  3498)
s8vector-set!:                                 See 10.      (line  3499)
s8vector?:                                     See 10.      (line  3494)
safe:                                          See 6.3.     (line  2549)
scheduler-exception-reason:                    See 15.4.    (line  5911)
scheduler-exception?:                          See 15.4.    (line  5910)
Scheme:                                        See 1.       (line    33)
Scheme, implementation of:
          See ``Gambit-C''.                                 (line    28)
scheme-ieee-1178-1990:                         See 2.5.     (line   316)
scheme-r4rs:                                   See 2.5.     (line   308)
scheme-r5rs:                                   See 2.5.     (line   312)
scheme-srfi-0:                                 See 2.5.     (line   320)
seconds->time:                                 See 16.7.    (line  7248)
separate:                                      See 6.3.     (line  2491)
serial-number->object:                         See 11.1.    (line  3689)
serialization <1>:                             See 10.      (line  3636)
serialization:                                 See 18.1.    (line  9543)
service-info <1>:                              See 16.12.   (line  7809)
service-info:                                  See 6.4.     (line  2689)
service-info-aliases <1>:                      See 6.4.     (line  2694)
service-info-aliases:                          See 16.12.   (line  7858)
service-info-name <1>:                         See 16.12.   (line  7848)
service-info-name:                             See 6.4.     (line  2691)
service-info-number:                           See 16.12.   (line  7868)
service-info-port:                             See 6.4.     (line  2692)
service-info-protocol:                         See 6.4.     (line  2693)
service-info? <1>:                             See 16.12.   (line  7836)
service-info?:                                 See 6.4.     (line  2690)
set!:                                          See 20.      (line 11620)
set-box!:                                      See 6.3.     (line  2273)
setenv:                                        See 16.6.    (line  7201)
sfun-conversion-exception-arguments:           See 15.5.    (line  6184)
sfun-conversion-exception-code:                See 15.5.    (line  6185)
sfun-conversion-exception-message:             See 15.5.    (line  6186)
sfun-conversion-exception-procedure:           See 15.5.    (line  6183)
sfun-conversion-exception?:                    See 15.5.    (line  6182)
shell-command:                                 See 16.3.    (line  7145)
six-script:                                    See 2.5.     (line   333)
six.make-array:                                See 6.4.     (line  2704)
socket-info-address:                           See 6.4.     (line  2700)
socket-info-family:                            See 6.4.     (line  2701)
socket-info-port-number:                       See 6.4.     (line  2702)
socket-info?:                                  See 6.4.     (line  2699)
stack-overflow-exception?:                     See 15.2.    (line  5779)
standard-bindings:                             See 6.3.     (line  2533)
started-thread-exception-arguments:            See 15.4.    (line  6010)
started-thread-exception-procedure:            See 15.4.    (line  6009)
started-thread-exception?:                     See 15.4.    (line  6008)
step:                                          See 5.3.     (line  1662)
step-level-set!:                               See 5.3.     (line  1663)
string->keyword:                               See 6.3.     (line  2300)
string-ci<=?:                                  See 8.2.     (line  2796)
string-ci<?:                                   See 8.2.     (line  2794)
string-ci=?:                                   See 8.2.     (line  2793)
string-ci=?-hash:                              See 11.1.    (line  3793)
string-ci>=?:                                  See 8.2.     (line  2797)
string-ci>?:                                   See 8.2.     (line  2795)
string<=?:                                     See 8.2.     (line  2791)
string<?:                                      See 8.2.     (line  2789)
string=?:                                      See 8.2.     (line  2788)
string=?-hash:                                 See 11.1.    (line  3781)
string>=?:                                     See 8.2.     (line  2792)
string>?:                                      See 8.2.     (line  2790)
subf32vector:                                  See 10.      (line  3609)
subf64vector:                                  See 10.      (line  3622)
subs16vector:                                  See 10.      (line  3531)
subs32vector:                                  See 10.      (line  3557)
subs64vector:                                  See 10.      (line  3583)
subs8vector:                                   See 10.      (line  3505)
subu16vector:                                  See 10.      (line  3544)
subu32vector:                                  See 10.      (line  3570)
subu64vector:                                  See 10.      (line  3596)
subu8vector:                                   See 10.      (line  3518)
subvector:                                     See 6.3.     (line  2263)
symbol-hash:                                   See 11.1.    (line  3757)
syntax-case:                                   See 6.3.     (line  2454)
syntax-rules:                                  See 6.3.     (line  2454)
system-version:                                See 6.4.     (line  2706)
system-version-string:                         See 6.4.     (line  2707)
table->list:                                   See 11.2.2.  (line  4191)
table-copy:                                    See 11.2.2.  (line  4259)
table-for-each:                                See 11.2.2.  (line  4162)
table-length:                                  See 11.2.2.  (line  4063)
table-ref:                                     See 11.2.2.  (line  4088)
table-search:                                  See 11.2.2.  (line  4132)
table-set!:                                    See 11.2.2.  (line  4112)
table?:                                        See 11.2.2.  (line  4051)
tables:                                        See 11.      (line  3684)
tail-calls:                                    See 5.3.     (line  1789)
tcp-client-peer-socket-info:                   See 6.4.     (line  2696)
tcp-client-self-socket-info:                   See 6.4.     (line  2697)
terminated-thread-exception-arguments:         See 15.4.    (line  6042)
terminated-thread-exception-procedure:         See 15.4.    (line  6041)
terminated-thread-exception?:                  See 15.4.    (line  6040)
test-bit-field?:                               See 9.4.     (line  3083)
thread-base-priority:                          See 13.9.    (line  4636)
thread-base-priority-set!:                     See 13.9.    (line  4637)
thread-group-name:                             See 6.4.     (line  2612)
thread-group-parent:                           See 6.4.     (line  2613)
thread-group-resume!:                          See 6.4.     (line  2614)
thread-group-suspend!:                         See 6.4.     (line  2615)
thread-group-terminate!:                       See 6.4.     (line  2616)
thread-group?:                                 See 6.4.     (line  2611)
thread-init!:                                  See 6.4.     (line  2625)
thread-join!:                                  See 13.9.    (line  4837)
thread-mailbox-extract-and-rewind:             See 13.9.    (line  4908)
thread-mailbox-next:                           See 13.9.    (line  4906)
thread-mailbox-rewind:                         See 13.9.    (line  4907)
thread-name:                                   See 13.9.    (line  4614)
thread-priority-boost:                         See 13.9.    (line  4652)
thread-priority-boost-set!:                    See 13.9.    (line  4653)
thread-quantum:                                See 13.9.    (line  4668)
thread-quantum-set!:                           See 13.9.    (line  4669)
thread-receive:                                See 13.9.    (line  4905)
thread-resume!:                                See 6.4.     (line  2619)
thread-send:                                   See 13.9.    (line  4888)
thread-sleep!:                                 See 13.9.    (line  4726)
thread-specific:                               See 13.9.    (line  4621)
thread-specific-set!:                          See 13.9.    (line  4622)
thread-start!:                                 See 13.9.    (line  4688)
thread-suspend!:                               See 6.4.     (line  2618)
thread-terminate!:                             See 13.9.    (line  4750)
thread-thread-group:                           See 6.4.     (line  2621)
thread-yield!:                                 See 13.9.    (line  4707)
thread?:                                       See 13.9.    (line  4567)
threads:                                       See 13.      (line  4324)
time:                                          See 16.7.    (line  7309)
time->seconds:                                 See 16.7.    (line  7247)
time?:                                         See 16.7.    (line  7246)
timeout->time:                                 See 6.4.     (line  2643)
touch:                                         See 6.4.     (line  2709)
trace:                                         See 5.3.     (line  1589)
transcript-off:                                See 6.1.     (line  2057)
transcript-on:                                 See 6.1.     (line  2056)
tty-history:                                   See 6.4.     (line  2712)
tty-history-set!:                              See 6.4.     (line  2713)
tty-max-history-length-set!:                   See 6.4.     (line  2714)
tty-mode-set!:                                 See 6.4.     (line  2717)
tty-paren-balance-duration-set!:               See 6.4.     (line  2715)
tty-text-attributes-set!:                      See 6.4.     (line  2716)
tty-type-set!:                                 See 6.4.     (line  2718)
tty?:                                          See 6.4.     (line  2711)
type-exception-arg-num:                        See 15.8.    (line  6478)
type-exception-arguments:                      See 15.8.    (line  6477)
type-exception-procedure:                      See 15.8.    (line  6476)
type-exception-type-id:                        See 15.8.    (line  6479)
type-exception?:                               See 15.8.    (line  6475)
u16vector:                                     See 10.      (line  3535)
u16vector->list:                               See 10.      (line  3539)
u16vector-append:                              See 10.      (line  3543)
u16vector-copy:                                See 10.      (line  3542)
u16vector-fill!:                               See 10.      (line  3541)
u16vector-length:                              See 10.      (line  3536)
u16vector-ref:                                 See 10.      (line  3537)
u16vector-set!:                                See 10.      (line  3538)
u16vector?:                                    See 10.      (line  3533)
u32vector:                                     See 10.      (line  3561)
u32vector->list:                               See 10.      (line  3565)
u32vector-append:                              See 10.      (line  3569)
u32vector-copy:                                See 10.      (line  3568)
u32vector-fill!:                               See 10.      (line  3567)
u32vector-length:                              See 10.      (line  3562)
u32vector-ref:                                 See 10.      (line  3563)
u32vector-set!:                                See 10.      (line  3564)
u32vector?:                                    See 10.      (line  3559)
u64vector:                                     See 10.      (line  3587)
u64vector->list:                               See 10.      (line  3591)
u64vector-append:                              See 10.      (line  3595)
u64vector-copy:                                See 10.      (line  3594)
u64vector-fill!:                               See 10.      (line  3593)
u64vector-length:                              See 10.      (line  3588)
u64vector-ref:                                 See 10.      (line  3589)
u64vector-set!:                                See 10.      (line  3590)
u64vector?:                                    See 10.      (line  3585)
u8vector:                                      See 10.      (line  3509)
u8vector->list:                                See 10.      (line  3513)
u8vector->object:                              See 10.      (line  3636)
u8vector-append:                               See 10.      (line  3517)
u8vector-copy:                                 See 10.      (line  3516)
u8vector-fill!:                                See 10.      (line  3515)
u8vector-length:                               See 10.      (line  3510)
u8vector-ref:                                  See 10.      (line  3511)
u8vector-set!:                                 See 10.      (line  3512)
u8vector?:                                     See 10.      (line  3507)
unbound-global-exception-code:                 See 15.7.    (line  6448)
unbound-global-exception-rte:                  See 15.7.    (line  6449)
unbound-global-exception-variable:             See 15.7.    (line  6447)
unbound-global-exception?:                     See 15.7.    (line  6446)
unbound-os-environment-variable-exception-arguments:See 15.3.
                                                            (line  5878)
unbound-os-environment-variable-exception-procedure:See 15.3.
                                                            (line  5877)
unbound-os-environment-variable-exception?:    See 15.3.    (line  5876)
unbound-serial-number-exception-arguments:     See 11.1.    (line  3729)
unbound-serial-number-exception-procedure:     See 11.1.    (line  3728)
unbound-serial-number-exception?:              See 11.1.    (line  3727)
unbound-table-key-exception-arguments:         See 11.2.2.  (line  4231)
unbound-table-key-exception-procedure:         See 11.2.2.  (line  4230)
unbound-table-key-exception?:                  See 11.2.2.  (line  4229)
unbox:                                         See 6.3.     (line  2272)
unbreak:                                       See 5.3.     (line  1740)
uncaught-exception-arguments:                  See 15.4.    (line  6077)
uncaught-exception-procedure:                  See 15.4.    (line  6076)
uncaught-exception-reason:                     See 15.4.    (line  6078)
uncaught-exception?:                           See 15.4.    (line  6075)
uninitialized-thread-exception-arguments:      See 6.4.     (line  2633)
uninitialized-thread-exception-procedure:      See 6.4.     (line  2632)
uninitialized-thread-exception?:               See 6.4.     (line  2631)
uninterned-keyword?:                           See 6.3.     (line  2364)
uninterned-symbol?:                            See 6.3.     (line  2341)
unknown-keyword-argument-exception-arguments:  See 15.9.    (line  6723)
unknown-keyword-argument-exception-procedure:  See 15.9.    (line  6722)
unknown-keyword-argument-exception?:           See 15.9.    (line  6721)
unterminated-process-exception-arguments:      See 6.4.     (line  2641)
unterminated-process-exception-procedure:      See 6.4.     (line  2640)
unterminated-process-exception?:               See 6.4.     (line  2639)
untrace:                                       See 5.3.     (line  1590)
user-info:                                     See 16.10.   (line  7647)
user-info-gid:                                 See 16.10.   (line  7702)
user-info-home:                                See 16.10.   (line  7712)
user-info-name:                                See 16.10.   (line  7683)
user-info-shell:                               See 16.10.   (line  7722)
user-info-uid:                                 See 16.10.   (line  7693)
user-info?:                                    See 16.10.   (line  7671)
user-name:                                     See 16.10.   (line  7638)
vector-append:                                 See 6.3.     (line  2257)
vector-copy:                                   See 6.3.     (line  2251)
void:                                          See 6.3.     (line  2384)
weak references:                               See 11.      (line  3684)
will-execute!:                                 See 11.2.1.  (line  3876)
will-testator:                                 See 11.2.1.  (line  3875)
will?:                                         See 11.2.1.  (line  3874)
with-exception-catcher:                        See 15.1.    (line  5672)
with-exception-handler:                        See 15.1.    (line  5647)
with-input-from-file:                          See 17.7.1.  (line  8778)
with-input-from-port:                          See 6.4.     (line  2720)
with-input-from-string:                        See 17.10.   (line  9366)
with-input-from-u8vector:                      See 17.11.   (line  9395)
with-input-from-vector:                        See 17.9.    (line  9223)
with-output-to-file:                           See 17.7.1.  (line  8779)
with-output-to-port:                           See 6.4.     (line  2721)
with-output-to-string:                         See 17.10.   (line  9367)
with-output-to-u8vector:                       See 17.11.   (line  9396)
with-output-to-vector:                         See 17.9.    (line  9224)
write:                                         See 17.4.2.  (line  8207)
write-char:                                    See 17.5.2.  (line  8442)
write-substring:                               See 17.5.2.  (line  8483)
write-subu8vector:                             See 17.6.2.  (line  8734)
write-u8:                                      See 17.6.2.  (line  8721)
wrong-number-of-arguments-exception-arguments: See 15.9.    (line  6626)
wrong-number-of-arguments-exception-procedure: See 15.9.    (line  6625)
wrong-number-of-arguments-exception?:          See 15.9.    (line  6624)
~/:                                            See 16.1.    (line  6856)
~username/:                                    See 16.1.    (line  6865)
~~/:                                           See 16.1.    (line  6849)
Table of Contents
*****************

Gambit-C
1 The Gambit-C system
  1.1 Accessing the system files
2 The Gambit Scheme interpreter
  2.1 Interactive mode
  2.2 Batch mode
  2.3 Customization
  2.4 Process exit status
  2.5 Scheme scripts
    2.5.1 Scripts under UNIX and Mac OS X
    2.5.2 Scripts under Microsoft Windows
    2.5.3 Compiling scripts
3 The Gambit Scheme compiler
  3.1 Interactive mode
  3.2 Customization
  3.3 Batch mode
  3.4 Link files
    3.4.1 Building an executable program
    3.4.2 Building a loadable library
    3.4.3 Building a shared-library
    3.4.4 Other compilation options
  3.5 Procedures specific to compiler
4 Runtime options for all programs
5 Debugging
  5.1 Debugging model
  5.2 Debugging commands
  5.3 Procedures related to debugging
  5.4 Console line-editing
  5.5 Emacs interface
  5.6 GUIDE
6 Scheme extensions
  6.1 Extensions to standard procedures
  6.2 Extensions to standard special forms
  6.3 Miscellaneous extensions
  6.4 Undocumented extensions
7 Modules
8 Characters and strings
  8.1 Extensions to character procedures
  8.2 Extensions to string procedures
9 Numbers
  9.1 Extensions to numeric procedures
  9.2 IEEE floating point arithmetic
  9.3 Integer square root and nth root
  9.4 Bitwise-operations on exact integers
  9.5 Fixnum specific operations
  9.6 Flonum specific operations
  9.7 Pseudo random numbers
10 Homogeneous vectors
11 Hashing and weak references
  11.1 Hashing
  11.2 Weak references
    11.2.1 Wills
    11.2.2 Tables
12 Records
13 Threads
  13.1 Introduction
  13.2 Thread objects
  13.3 Mutex objects
  13.4 Condition variable objects
  13.5 Fairness
  13.6 Memory coherency
  13.7 Timeouts
  13.8 Primordial thread
  13.9 Procedures
14 Dynamic environment
15 Exceptions
  15.1 Exception-handling
  15.2 Exception objects related to memory management
  15.3 Exception objects related to the host environment
  15.4 Exception objects related to threads
  15.5 Exception objects related to C-interface
  15.6 Exception objects related to the reader
  15.7 Exception objects related to evaluation and compilation
  15.8 Exception objects related to type checking
  15.9 Exception objects related to procedure call
  15.10 Other exception objects
16 Host environment
  16.1 Handling of file names
  16.2 Filesystem operations
  16.3 Shell command execution
  16.4 Process termination
  16.5 Command line arguments
  16.6 Environment variables
  16.7 Measuring time
  16.8 File information
  16.9 Group information
  16.10 User information
  16.11 Host information
  16.12 Service information
  16.13 Protocol information
  16.14 Network information
17 I/O and ports
  17.1 Unidirectional and bidirectional ports
  17.2 Port classes
  17.3 Port settings
  17.4 Object-ports
    17.4.1 Object-port settings
    17.4.2 Object-port operations
  17.5 Character-ports
    17.5.1 Character-port settings
    17.5.2 Character-port operations
  17.6 Byte-ports
    17.6.1 Byte-port settings
    17.6.2 Byte-port operations
  17.7 Device-ports
    17.7.1 Filesystem devices
    17.7.2 Process devices
    17.7.3 Network devices
  17.8 Directory-ports
  17.9 Vector-ports
  17.10 String-ports
  17.11 U8vector-ports
  17.12 Parameter objects related to I/O
18 Lexical syntax and readtables
  18.1 Readtables
  18.2 Boolean syntax
  18.3 Character syntax
  18.4 String syntax
  18.5 Symbol syntax
  18.6 Keyword syntax
  18.7 Box syntax
  18.8 Number syntax
  18.9 Homogeneous vector syntax
  18.10 Special `#!' syntax
  18.11 Multiline comment syntax
  18.12 Scheme infix syntax extension
    18.12.1 SIX grammar
    18.12.2 SIX semantics
19 C-interface
  19.1 The mapping of types between C and Scheme
  19.2 The `c-declare' special form
  19.3 The `c-initialize' special form
  19.4 The `c-lambda' special form
  19.5 The `c-define' special form
  19.6 The `c-define-type' special form
  19.7 Continuations and the C-interface
20 System limitations
21 Copyright and license
General index