1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
8 Changes in 1.8.8 (since 1.8.7)
12 ** Fix possible buffer overruns when parsing numbers
13 ** Fix random number generator on 64-bit platforms
15 Previously the `scm_c_random' function would crash or return only 32-bit
16 worth of randomness. In addition, the new `scm_c_random64' function
17 explicitly returns 64 bits of randomness.
19 ** Add missing range checks in `vector-move-left!' and `vector-move-right!'
21 Previously these procedures could write past the end of a vector.
23 ** Avoid clash with system setjmp/longjmp on IA64
24 ** Don't dynamically link an extension that is already registered
25 ** Fix `wrong type arg' exceptions with IPv6 addresses
26 ** Fix typos in `(srfi srfi-19)'
27 ** Have `(srfi srfi-35)' provide named struct vtables
28 ** Fix C documentation extraction ("snarfing") with recent GCC versions
29 ** Fix some Interix build problems
30 ** Fix Tru64 build problems
31 ** Fix GC-related build issues on Solaris 10 x86 with Sun Studio 12
32 ** Several small documentation fixes
35 Changes in 1.8.7 (since 1.8.6)
39 ** Fix compilation with `--disable-deprecated'
40 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
41 ** Fix MinGW build problem caused by HAVE_STRUCT_TIMESPEC confusion
42 ** Fix build problem when scm_t_timespec is different from struct timespec
43 ** Fix build when compiled with -Wundef -Werror
44 ** More build fixes for `alphaev56-dec-osf5.1b' (Tru64)
45 ** Build fixes for `powerpc-ibm-aix5.3.0.0' (AIX 5.3)
46 ** With GCC, always compile with `-mieee' on `alpha*' and `sh*'
47 ** Better diagnose broken `(strftime "%z" ...)' in `time.test' (bug #24130)
48 ** Fix parsing of SRFI-88/postfix keywords longer than 128 characters
49 ** Fix reading of complex numbers where both parts are inexact decimals
51 ** Allow @ macro to work with (ice-9 syncase)
53 Previously, use of the @ macro in a module whose code is being
54 transformed by (ice-9 syncase) would cause an "Invalid syntax" error.
55 Now it works as you would expect (giving the value of the specified
58 ** Have `scm_take_locale_symbol ()' return an interned symbol (bug #25865)
59 ** Fix potential deadlocks when running on multiple threads
61 ** Fix problems building with the i586-mingw32msvc cross-compiler
63 It's now possible to build Guile for Windows by using the
64 i586-mingw32msvc cross-compiler on GNU/Linux. This kind of build
65 produces DLLs and a main program that can be copied to and used on a
66 Windows PC. For how to do this, see the `Cross building Guile'
70 Changes in 1.8.6 (since 1.8.5)
72 * New features (see the manual for details)
74 ** New convenience function `scm_c_symbol_length ()'
76 ** Single stepping through code from Emacs
78 When you use GDS to evaluate Scheme code from Emacs, you can now use
79 `C-u' to indicate that you want to single step through that code. See
80 `Evaluating Scheme Code' in the manual for more details.
82 ** New "guile(1)" man page!
84 * Changes to the distribution
86 ** Automake's `AM_MAINTAINER_MODE' is no longer used
88 Thus, the `--enable-maintainer-mode' configure option is no longer
89 available: Guile is now always configured in "maintainer mode".
91 ** `ChangeLog' files are no longer updated
93 Instead, changes are detailed in the version control system's logs. See
94 the top-level `ChangeLog' files for details.
99 ** `symbol->string' now returns a read-only string, as per R5RS
100 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
101 ** `guile-config link' now prints `-L$libdir' before `-lguile'
102 ** Fix memory corruption involving GOOPS' `class-redefinition'
103 ** Fix possible deadlock in `mutex-lock'
104 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
105 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
106 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
107 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
108 ** Fix misleading output from `(help rationalize)'
109 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
110 ** Fix `gcd' when called with a single, negative argument.
111 ** Fix `Stack overflow' errors seen when building on some platforms
112 ** Fix bug when `scm_with_guile ()' was called several times from the
114 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
115 dynamic environment of the call to `raise'
116 ** Fix potential deadlock in `make-struct'
117 ** Fix compilation problem with libltdl from Libtool 2.2.x
118 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
121 Changes in 1.8.5 (since 1.8.4)
123 * Infrastructure changes
125 ** Guile repository switched from CVS to Git
127 The new repository can be accessed using
128 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
129 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
131 ** Add support for `pkg-config'
133 See "Autoconf Support" in the manual for details.
135 * New modules (see the manual for details)
139 * New features (see the manual for details)
141 ** New `postfix' read option, for SRFI-88 keyword syntax
142 ** Some I/O primitives have been inlined, which improves I/O performance
143 ** New object-based traps infrastructure
145 This is a GOOPS-based infrastructure that builds on Guile's low-level
146 evaluator trap calls and facilitates the development of debugging
147 features like single-stepping, breakpoints, tracing and profiling.
148 See the `Traps' node of the manual for details.
150 ** New support for working on Guile code from within Emacs
152 Guile now incorporates the `GDS' library (previously distributed
153 separately) for working on Guile code from within Emacs. See the
154 `Using Guile In Emacs' node of the manual for details.
158 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
159 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
161 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
162 would trigger an unbound variable error for `match:andmap'.
164 ** `(oop goops describe)' now properly provides the `describe' feature
165 ** Fixed `args-fold' from `(srfi srfi-37)'
167 Previously, parsing short option names of argument-less options would
168 lead to a stack overflow.
170 ** `(srfi srfi-35)' is now visible through `cond-expand'
171 ** Fixed type-checking for the second argument of `eval'
172 ** Fixed type-checking for SRFI-1 `partition'
173 ** Fixed `struct-ref' and `struct-set!' on "light structs"
174 ** Honor struct field access rights in GOOPS
175 ** Changed the storage strategy of source properties, which fixes a deadlock
176 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
177 ** Fixed build issue for GNU/Linux on IA64
178 ** Fixed build issues on NetBSD 1.6
179 ** Fixed build issue on Solaris 2.10 x86_64
180 ** Fixed build issue with DEC/Compaq/HP's compiler
181 ** Fixed `scm_from_complex_double' build issue on FreeBSD
182 ** Fixed `alloca' build issue on FreeBSD 6
183 ** Removed use of non-portable makefile constructs
184 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
185 ** Make sure all tests honor `$TMPDIR'
188 Changes in 1.8.4 (since 1.8.3)
192 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
193 ** Fixed a segmentation fault which occurred when displaying the
194 backtrace of a stack with a promise object (made by `delay') in it.
195 ** Make `accept' leave guile mode while blocking
196 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
197 ** Fixed a build problem on AIX (use of func_data identifier)
198 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
199 called with an associator proc that returns neither a pair nor #f.
200 ** Secondary threads now always return a valid module for (current-module).
201 ** Avoid MacOS build problems caused by incorrect combination of "64"
202 system and library calls.
203 ** `guile-snarf' now honors `$TMPDIR'
204 ** `guile-config compile' now reports CPPFLAGS used at compile-time
205 ** Fixed build with Sun Studio (Solaris 9)
206 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
207 uniform vectors on AIX.
208 ** Fixed a deadlock that occurs upon GC with multiple threads.
209 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
210 ** Fixed autotool-derived build problems on AIX 6.1.
211 ** Fixed NetBSD/alpha support
212 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
214 * New modules (see the manual for details)
218 * Documentation fixes and improvements
220 ** Removed premature breakpoint documentation
222 The features described are not available in the series of 1.8.x
223 releases, so the documentation was misleading and has been removed.
225 ** More about Guile's default *random-state* variable
227 ** GOOPS: more about how to use `next-method'
229 * Changes to the distribution
231 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
233 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
234 General Public License, and the few incorrect files have now been
235 fixed to agree with the rest of the Guile distribution.
237 ** Removed unnecessary extra copies of COPYING*
239 The distribution now contains a single COPYING.LESSER at its top level.
242 Changes in 1.8.3 (since 1.8.2)
244 * New modules (see the manual for details)
251 ** The `(ice-9 slib)' module now works as expected
252 ** Expressions like "(set! 'x #t)" no longer yield a crash
253 ** Warnings about duplicate bindings now go to stderr
254 ** A memory leak in `make-socket-address' was fixed
255 ** Alignment issues (e.g., on SPARC) in network routines were fixed
256 ** A threading issue that showed up at least on NetBSD was fixed
257 ** Build problems on Solaris and IRIX fixed
259 * Implementation improvements
261 ** The reader is now faster, which reduces startup time
262 ** Procedures returned by `record-accessor' and `record-modifier' are faster
265 Changes in 1.8.2 (since 1.8.1):
267 * New procedures (see the manual for details)
269 ** set-program-arguments
272 * Incompatible changes
274 ** The body of a top-level `define' no longer sees the binding being created
276 In a top-level `define', the binding being created is no longer visible
277 from the `define' body. This breaks code like
278 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
279 unbound in the body. However, such code was not R5RS-compliant anyway,
284 ** Fractions were not `equal?' if stored in unreduced form.
285 (A subtle problem, since printing a value reduced it, making it work.)
286 ** srfi-60 `copy-bit' failed on 64-bit systems
287 ** "guile --use-srfi" option at the REPL can replace core functions
288 (Programs run with that option were ok, but in the interactive REPL
289 the core bindings got priority, preventing SRFI replacements or
291 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
292 ** `kill' on mingw throws an error for a PID other than oneself
293 ** Procedure names are attached to procedure-with-setters
294 ** Array read syntax works with negative lower bound
295 ** `array-in-bounds?' fix if an array has different lower bounds on each index
296 ** `*' returns exact 0 for "(* inexact 0)"
297 This follows what it always did for "(* 0 inexact)".
298 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
299 ** SRFI-19: `date->julian-day' did not account for timezone offset
300 ** `ttyname' no longer crashes when passed a non-tty argument
301 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
302 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
303 ** GOOPS: Fixed a bug in `method-more-specific?'
304 ** Build problems on Solaris fixed
305 ** Build problems on HP-UX IA64 fixed
306 ** Build problems on MinGW fixed
309 Changes in 1.8.1 (since 1.8.0):
311 * LFS functions are now used to access 64-bit files on 32-bit systems.
313 * New procedures (see the manual for details)
315 ** primitive-_exit - [Scheme] the-root-module
316 ** scm_primitive__exit - [C]
317 ** make-completion-function - [Scheme] (ice-9 readline)
318 ** scm_c_locale_stringn_to_number - [C]
319 ** scm_srfi1_append_reverse [C]
320 ** scm_srfi1_append_reverse_x [C]
328 ** Build problems have been fixed on MacOS, SunOS, and QNX.
330 ** `strftime' fix sign of %z timezone offset.
332 ** A one-dimensional array can now be 'equal?' to a vector.
334 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
336 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
338 ** `record-accessor' and `record-modifier' now have strict type checks.
340 Record accessor and modifier procedures now throw an error if the
341 record type of the record they're given is not the type expected.
342 (Previously accessors returned #f and modifiers silently did nothing).
344 ** It is now OK to use both autoload and use-modules on a given module.
346 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
348 Previously there was no checking on primatives like make-vector that
349 accept "one or two" arguments. Now there is.
351 ** The srfi-1 assoc function now calls its equality predicate properly.
353 Previously srfi-1 assoc would call the equality predicate with the key
354 last. According to the SRFI, the key should be first.
356 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
358 ** The array-set! procedure no longer segfaults when given a bit vector.
360 ** Bugs in make-shared-array have been fixed.
362 ** string<? and friends now follow char<? etc order on 8-bit chars.
364 ** The format procedure now handles inf and nan values for ~f correctly.
366 ** exact->inexact should no longer overflow when given certain large fractions.
368 ** srfi-9 accessor and modifier procedures now have strict record type checks.
370 This matches the srfi-9 specification.
372 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
374 Previously the (ice-9 ftw) procedures would ignore any file that had
375 the same inode number as a file they had already seen, even if that
376 file was on a different device.
379 Changes in 1.8.0 (changes since the 1.6.x series):
381 * Changes to the distribution
383 ** Guile is now licensed with the GNU Lesser General Public License.
385 ** The manual is now licensed with the GNU Free Documentation License.
387 ** Guile now requires GNU MP (http://swox.com/gmp).
389 Guile now uses the GNU MP library for arbitrary precision arithmetic.
391 ** Guile now has separate private and public configuration headers.
393 That is, things like HAVE_STRING_H no longer leak from Guile's
396 ** Guile now provides and uses an "effective" version number.
398 Guile now provides scm_effective_version and effective-version
399 functions which return the "effective" version number. This is just
400 the normal full version string without the final micro-version number,
401 so the current effective-version is "1.8". The effective version
402 should remain unchanged during a stable series, and should be used for
403 items like the versioned share directory name
404 i.e. /usr/share/guile/1.8.
406 Providing an unchanging version number during a stable release for
407 things like the versioned share directory can be particularly
408 important for Guile "add-on" packages, since it provides a directory
409 that they can install to that won't be changed out from under them
410 with each micro release during a stable series.
412 ** Thread implementation has changed.
414 When you configure "--with-threads=null", you will get the usual
415 threading API (call-with-new-thread, make-mutex, etc), but you can't
416 actually create new threads. Also, "--with-threads=no" is now
417 equivalent to "--with-threads=null". This means that the thread API
418 is always present, although you might not be able to create new
421 When you configure "--with-threads=pthreads" or "--with-threads=yes",
422 you will get threads that are implemented with the portable POSIX
423 threads. These threads can run concurrently (unlike the previous
424 "coop" thread implementation), but need to cooperate for things like
427 The default is "pthreads", unless your platform doesn't have pthreads,
428 in which case "null" threads are used.
430 See the manual for details, nodes "Initialization", "Multi-Threading",
431 "Blocking", and others.
433 ** There is the new notion of 'discouraged' features.
435 This is a milder form of deprecation.
437 Things that are discouraged should not be used in new code, but it is
438 OK to leave them in old code for now. When a discouraged feature is
439 used, no warning message is printed like there is for 'deprecated'
440 features. Also, things that are merely discouraged are nevertheless
441 implemented efficiently, while deprecated features can be very slow.
443 You can omit discouraged features from libguile by configuring it with
444 the '--disable-discouraged' option.
446 ** Deprecation warnings can be controlled at run-time.
448 (debug-enable 'warn-deprecated) switches them on and (debug-disable
449 'warn-deprecated) switches them off.
451 ** Support for SRFI 61, extended cond syntax for multiple values has
454 This SRFI is always available.
456 ** Support for require-extension, SRFI-55, has been added.
458 The SRFI-55 special form `require-extension' has been added. It is
459 available at startup, and provides a portable way to load Scheme
460 extensions. SRFI-55 only requires support for one type of extension,
461 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
464 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
466 The (srfi srfi-26) module is an implementation of SRFI-26 which
467 provides the `cut' and `cute' syntax. These may be used to specialize
468 parameters without currying.
470 ** New module (srfi srfi-31)
472 This is an implementation of SRFI-31 which provides a special form
473 `rec' for recursive evaluation.
475 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
476 been merged with the core, making their functionality always
479 The modules are still available, tho, and you could use them together
480 with a renaming import, for example.
482 ** Guile no longer includes its own version of libltdl.
484 The official version is good enough now.
486 ** The --enable-htmldoc option has been removed from 'configure'.
488 Support for translating the documentation into HTML is now always
489 provided. Use 'make html'.
491 ** New module (ice-9 serialize):
493 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
494 don't trust the thread safety of most of your program, but where you
495 have some section(s) of code which you consider can run in parallel to
496 other sections. See ice-9/serialize.scm for more information.
498 ** The configure option '--disable-arrays' has been removed.
500 Support for arrays and uniform numeric arrays is now always included
503 * Changes to the stand-alone interpreter
505 ** New command line option `-L'.
507 This option adds a directory to the front of the load path.
509 ** New command line option `--no-debug'.
511 Specifying `--no-debug' on the command line will keep the debugging
512 evaluator turned off, even for interactive sessions.
514 ** User-init file ~/.guile is now loaded with the debugging evaluator.
516 Previously, the normal evaluator would have been used. Using the
517 debugging evaluator gives better error messages.
519 ** The '-e' option now 'read's its argument.
521 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
522 be used with '-e'. For example, you can now write a script like
525 exec guile -e '(@ (demo) main)' -s "$0" "$@"
528 (define-module (demo)
532 (format #t "Demo: ~a~%" args))
535 * Changes to Scheme functions and syntax
537 ** Guardians have changed back to their original semantics
539 Guardians now behave like described in the paper by Dybvig et al. In
540 particular, they no longer make guarantees about the order in which
541 they return objects, and they can no longer be greedy.
543 They no longer drop cyclic data structures.
545 The C function scm_make_guardian has been changed incompatibly and no
546 longer takes the 'greedy_p' argument.
548 ** New function hashx-remove!
550 This function completes the set of 'hashx' functions.
552 ** The concept of dynamic roots has been factored into continuation
553 barriers and dynamic states.
555 Each thread has a current dynamic state that carries the values of the
556 fluids. You can create and copy dynamic states and use them as the
557 second argument for 'eval'. See "Fluids and Dynamic States" in the
560 To restrict the influence that captured continuations can have on the
561 control flow, you can errect continuation barriers. See "Continuation
562 Barriers" in the manual.
564 The function call-with-dynamic-root now essentially temporarily
565 installs a new dynamic state and errects a continuation barrier.
567 ** The default load path no longer includes "." at the end.
569 Automatically loading modules from the current directory should not
570 happen by default. If you want to allow it in a more controlled
571 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
574 ** The uniform vector and array support has been overhauled.
576 It now complies with SRFI-4 and the weird prototype based uniform
577 array creation has been deprecated. See the manual for more details.
579 Some non-compatible changes have been made:
580 - characters can no longer be stored into byte arrays.
581 - strings and bit vectors are no longer considered to be uniform numeric
583 - array-rank throws an error for non-arrays instead of returning zero.
584 - array-ref does no longer accept non-arrays when no indices are given.
586 There is the new notion of 'generalized vectors' and corresponding
587 procedures like 'generalized-vector-ref'. Generalized vectors include
588 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
590 Arrays use generalized vectors as their storage, so that you still
591 have arrays of characters, bits, etc. However, uniform-array-read!
592 and uniform-array-write can no longer read/write strings and
595 ** There is now support for copy-on-write substrings, mutation-sharing
596 substrings and read-only strings.
598 Three new procedures are related to this: substring/shared,
599 substring/copy, and substring/read-only. See the manual for more
602 ** Backtraces will now highlight the value that caused the error.
604 By default, these values are enclosed in "{...}", such as in this
613 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
614 <unnamed port>:1:1: Wrong type (expecting pair): a
615 ABORT: (wrong-type-arg)
617 The prefix and suffix used for highlighting can be set via the two new
618 printer options 'highlight-prefix' and 'highlight-suffix'. For
619 example, putting this into ~/.guile will output the bad value in bold
622 (print-set! highlight-prefix "\x1b[1m")
623 (print-set! highlight-suffix "\x1b[22m")
626 ** 'gettext' support for internationalization has been added.
628 See the manual for details.
630 ** New syntax '@' and '@@':
632 You can now directly refer to variables exported from a module by
635 (@ MODULE-NAME VARIABLE-NAME)
637 For example (@ (ice-9 pretty-print) pretty-print) will directly access
638 the pretty-print variable exported from the (ice-9 pretty-print)
639 module. You don't need to 'use' that module first. You can also use
640 '@' as a target of 'set!', as in (set! (@ mod var) val).
642 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
643 but it can also access variables that have not been exported. It is
644 intended only for kluges and temporary fixes and for debugging, not
647 ** Keyword syntax has been made more disciplined.
649 Previously, the name of a keyword was read as a 'token' but printed as
650 a symbol. Now, it is read as a general Scheme datum which must be a
661 ERROR: In expression (a b c):
665 ERROR: Unbound variable: foo
670 ERROR: Wrong type (expecting symbol): 12
674 ERROR: Wrong type (expecting symbol): (a b c)
678 ** The printing of symbols that might look like keywords can be
681 The new printer option 'quote-keywordish-symbols' controls how symbols
682 are printed that have a colon as their first or last character. The
683 default now is to only quote a symbol with #{...}# when the read
684 option 'keywords' is not '#f'. Thus:
686 guile> (define foo (string->symbol ":foo"))
687 guile> (read-set! keywords #f)
690 guile> (read-set! keywords 'prefix)
693 guile> (print-set! quote-keywordish-symbols #f)
697 ** 'while' now provides 'break' and 'continue'
699 break and continue were previously bound in a while loop, but not
700 documented, and continue didn't quite work properly. The undocumented
701 parameter to break which gave a return value for the while has been
704 ** 'call-with-current-continuation' is now also available under the name
707 ** The module system now checks for duplicate bindings.
709 The module system now can check for name conflicts among imported
712 The behavior can be controlled by specifying one or more 'duplicates'
713 handlers. For example, to make Guile return an error for every name
721 The new default behavior of the module system when a name collision
722 has been detected is to
724 1. Give priority to bindings marked as a replacement.
725 2. Issue a warning (different warning if overriding core binding).
726 3. Give priority to the last encountered binding (this corresponds to
729 If you want the old behavior back without replacements or warnings you
732 (default-duplicate-binding-handler 'last)
734 to your .guile init file.
736 ** New define-module option: :replace
738 :replace works as :export, but, in addition, marks the binding as a
741 A typical example is `format' in (ice-9 format) which is a replacement
742 for the core binding `format'.
744 ** Adding prefixes to imported bindings in the module system
746 There is now a new :use-module option :prefix. It can be used to add
747 a prefix to all imported bindings.
750 :use-module ((bar) :prefix bar:))
752 will import all bindings exported from bar, but rename them by adding
755 ** Conflicting generic functions can be automatically merged.
757 When two imported bindings conflict and they are both generic
758 functions, the two functions can now be merged automatically. This is
759 activated with the 'duplicates' handler 'merge-generics'.
761 ** New function: effective-version
763 Returns the "effective" version number. This is just the normal full
764 version string without the final micro-version number. See "Changes
765 to the distribution" above.
767 ** New threading functions: parallel, letpar, par-map, and friends
769 These are convenient ways to run calculations in parallel in new
770 threads. See "Parallel forms" in the manual for details.
772 ** New function 'try-mutex'.
774 This function will attempt to lock a mutex but will return immediately
775 instead of blocking and indicate failure.
777 ** Waiting on a condition variable can have a timeout.
779 The function 'wait-condition-variable' now takes a third, optional
780 argument that specifies the point in time where the waiting should be
783 ** New function 'broadcast-condition-variable'.
785 ** New functions 'all-threads' and 'current-thread'.
787 ** Signals and system asyncs work better with threads.
789 The function 'sigaction' now takes a fourth, optional, argument that
790 specifies the thread that the handler should run in. When the
791 argument is omitted, the handler will run in the thread that called
794 Likewise, 'system-async-mark' takes a second, optional, argument that
795 specifies the thread that the async should run in. When it is
796 omitted, the async will run in the thread that called
799 C code can use the new functions scm_sigaction_for_thread and
800 scm_system_async_mark_for_thread to pass the new thread argument.
802 When a thread blocks on a mutex, a condition variable or is waiting
803 for IO to be possible, it will still execute system asyncs. This can
804 be used to interrupt such a thread by making it execute a 'throw', for
807 ** The function 'system-async' is deprecated.
809 You can now pass any zero-argument procedure to 'system-async-mark'.
810 The function 'system-async' will just return its argument unchanged
813 ** New functions 'call-with-blocked-asyncs' and
814 'call-with-unblocked-asyncs'
816 The expression (call-with-blocked-asyncs PROC) will call PROC and will
817 block execution of system asyncs for the current thread by one level
818 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
819 procedure and will unblock the execution of system asyncs by one
820 level for the current thread.
822 Only system asyncs are affected by these functions.
824 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
826 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
827 instead. Those functions are easier to use correctly and can be
830 ** New function 'unsetenv'.
832 ** New macro 'define-syntax-public'.
834 It works like 'define-syntax' and also exports the defined macro (but
837 ** There is support for Infinity and NaNs.
839 Following PLT Scheme, Guile can now work with infinite numbers, and
842 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
843 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
844 "+nan.0"). These numbers are inexact and have no exact counterpart.
846 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
847 sign of the dividend. The infinities are integers, and they answer #t
848 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
849 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
860 ERROR: Numerical overflow
862 Two new predicates 'inf?' and 'nan?' can be used to test for the
865 ** Inexact zero can have a sign.
867 Guile can now distinguish between plus and minus inexact zero, if your
868 platform supports this, too. The two zeros are equal according to
869 '=', but not according to 'eqv?'. For example
880 ** Guile now has exact rationals.
882 Guile can now represent fractions such as 1/3 exactly. Computing with
883 them is also done exactly, of course:
888 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
891 For example: (floor 2) now returns an exact 2 where in the past it
892 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
894 ** inexact->exact no longer returns only integers.
896 Without exact rationals, the closest exact number was always an
897 integer, but now inexact->exact returns the fraction that is exactly
898 equal to a floating point number. For example:
900 (inexact->exact 1.234)
901 => 694680242521899/562949953421312
903 When you want the old behavior, use 'round' explicitly:
905 (inexact->exact (round 1.234))
908 ** New function 'rationalize'.
910 This function finds a simple fraction that is close to a given real
911 number. For example (and compare with inexact->exact above):
913 (rationalize (inexact->exact 1.234) 1/2000)
916 Note that, as required by R5RS, rationalize returns only then an exact
917 result when both its arguments are exact.
919 ** 'odd?' and 'even?' work also for inexact integers.
921 Previously, (odd? 1.0) would signal an error since only exact integers
922 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
923 returns #f and (odd? 1.5) signals an error.
925 ** Guile now has uninterned symbols.
927 The new function 'make-symbol' will return an uninterned symbol. This
928 is a symbol that is unique and is guaranteed to remain unique.
929 However, uninterned symbols can not yet be read back in.
931 Use the new function 'symbol-interned?' to check whether a symbol is
934 ** pretty-print has more options.
936 The function pretty-print from the (ice-9 pretty-print) module can now
937 also be invoked with keyword arguments that control things like
938 maximum output width. See the manual for details.
940 ** Variables have no longer a special behavior for `equal?'.
942 Previously, comparing two variables with `equal?' would recursivly
943 compare their values. This is no longer done. Variables are now only
944 `equal?' if they are `eq?'.
946 ** `(begin)' is now valid.
948 You can now use an empty `begin' form. It will yield #<unspecified>
949 when evaluated and simply be ignored in a definition context.
951 ** Deprecated: procedure->macro
953 Change your code to use 'define-macro' or r5rs macros. Also, be aware
954 that macro expansion will not be done during evaluation, but prior to
957 ** Soft ports now allow a `char-ready?' procedure
959 The vector argument to `make-soft-port' can now have a length of
960 either 5 or 6. (Previously the length had to be 5.) The optional 6th
961 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
962 that returns the number of characters that can be read immediately
963 without the soft port blocking.
965 ** Deprecated: undefine
967 There is no replacement for undefine.
969 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
970 have been discouraged.
972 They are relics from a time where a keyword like #:foo was used
973 directly as a Tcl option "-foo" and thus keywords were internally
974 stored as a symbol with a starting dash. We now store a symbol
977 Use symbol->keyword and keyword->symbol instead.
979 ** The `cheap' debug option is now obsolete
981 Evaluator trap calls are now unconditionally "cheap" - in other words,
982 they pass a debug object to the trap handler rather than a full
983 continuation. The trap handler code can capture a full continuation
984 by using `call-with-current-continuation' in the usual way, if it so
987 The `cheap' option is retained for now so as not to break existing
988 code which gets or sets it, but setting it now has no effect. It will
989 be removed in the next major Guile release.
991 ** Evaluator trap calls now support `tweaking'
993 `Tweaking' means that the trap handler code can modify the Scheme
994 expression that is about to be evaluated (in the case of an
995 enter-frame trap) or the value that is being returned (in the case of
996 an exit-frame trap). The trap handler code indicates that it wants to
997 do this by returning a pair whose car is the symbol 'instead and whose
998 cdr is the modified expression or return value.
1000 * Changes to the C interface
1002 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
1003 take a 'delete' function argument.
1005 This argument makes no sense since the delete function is used to
1006 remove a pair from an alist, and this must not be configurable.
1008 This is an incompatible change.
1010 ** The GH interface is now subject to the deprecation mechanism
1012 The GH interface has been deprecated for quite some time but now it is
1013 actually removed from Guile when it is configured with
1014 --disable-deprecated.
1016 See the manual "Transitioning away from GH" for more information.
1018 ** A new family of functions for converting between C values and
1019 Scheme values has been added.
1021 These functions follow a common naming scheme and are designed to be
1022 easier to use, thread-safe and more future-proof than the older
1025 - int scm_is_* (...)
1027 These are predicates that return a C boolean: 1 or 0. Instead of
1028 SCM_NFALSEP, you can now use scm_is_true, for example.
1030 - <type> scm_to_<type> (SCM val, ...)
1032 These are functions that convert a Scheme value into an appropriate
1033 C value. For example, you can use scm_to_int to safely convert from
1036 - SCM scm_from_<type> (<type> val, ...)
1038 These functions convert from a C type to a SCM value; for example,
1039 scm_from_int for ints.
1041 There is a huge number of these functions, for numbers, strings,
1042 symbols, vectors, etc. They are documented in the reference manual in
1043 the API section together with the types that they apply to.
1045 ** New functions for dealing with complex numbers in C have been added.
1047 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1048 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1049 They work like scm_make_rectangular etc but take or return doubles
1052 ** The function scm_make_complex has been discouraged.
1054 Use scm_c_make_rectangular instead.
1056 ** The INUM macros have been deprecated.
1058 A lot of code uses these macros to do general integer conversions,
1059 although the macros only work correctly with fixnums. Use the
1060 following alternatives.
1062 SCM_INUMP -> scm_is_integer or similar
1063 SCM_NINUMP -> !scm_is_integer or similar
1064 SCM_MAKINUM -> scm_from_int or similar
1065 SCM_INUM -> scm_to_int or similar
1067 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1068 do the validating for you.
1070 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1071 have been discouraged.
1073 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1074 new code. The functions have been discouraged since they don't fit
1077 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1079 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1080 evaluates its argument twice. Use scm_is_true, etc. instead for new
1083 ** The macro SCM_EQ_P has been discouraged.
1085 Use scm_is_eq for new code, which fits better into the naming
1088 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1091 Use the function scm_is_pair or scm_is_null instead.
1093 ** The functions scm_round and scm_truncate have been deprecated and
1094 are now available as scm_c_round and scm_c_truncate, respectively.
1096 These functions occupy the names that scm_round_number and
1097 scm_truncate_number should have.
1099 ** The functions scm_c_string2str, scm_c_substring2str, and
1100 scm_c_symbol2str have been deprecated.
1102 Use scm_to_locale_stringbuf or similar instead, maybe together with
1105 ** New functions scm_c_make_string, scm_c_string_length,
1106 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1107 scm_c_substring_shared, scm_c_substring_copy.
1109 These are like scm_make_string, scm_length, etc. but are slightly
1110 easier to use from C.
1112 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1113 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1115 They export too many assumptions about the implementation of strings
1116 and symbols that are no longer true in the presence of
1117 mutation-sharing substrings and when Guile switches to some form of
1120 When working with strings, it is often best to use the normal string
1121 functions provided by Guile, such as scm_c_string_ref,
1122 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1123 manual since many more such functions are now provided than
1126 When you want to convert a SCM string to a C string, use the
1127 scm_to_locale_string function or similar instead. For symbols, use
1128 scm_symbol_to_string and then work with that string. Because of the
1129 new string representation, scm_symbol_to_string does not need to copy
1130 and is thus quite efficient.
1132 ** Some string, symbol and keyword functions have been discouraged.
1134 They don't fit into the uniform naming scheme and are not explicit
1135 about the character encoding.
1137 Replace according to the following table:
1139 scm_allocate_string -> scm_c_make_string
1140 scm_take_str -> scm_take_locale_stringn
1141 scm_take0str -> scm_take_locale_string
1142 scm_mem2string -> scm_from_locale_stringn
1143 scm_str2string -> scm_from_locale_string
1144 scm_makfrom0str -> scm_from_locale_string
1145 scm_mem2symbol -> scm_from_locale_symboln
1146 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1147 scm_str2symbol -> scm_from_locale_symbol
1149 SCM_SYMBOL_HASH -> scm_hashq
1150 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1152 scm_c_make_keyword -> scm_from_locale_keyword
1154 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1155 now also available to C code.
1157 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1159 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1160 the latter returns the true name of the keyword, not the 'dash name',
1161 as SCM_KEYWORDSYM used to do.
1163 ** A new way to access arrays in a thread-safe and efficient way has
1166 See the manual, node "Accessing Arrays From C".
1168 ** The old uniform vector and bitvector implementations have been
1169 unceremoniously removed.
1171 This implementation exposed the details of the tagging system of
1172 Guile. Use the new C API explained in the manual in node "Uniform
1173 Numeric Vectors" and "Bit Vectors", respectively.
1175 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1176 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1177 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1178 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1179 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1180 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1183 ** The macros dealing with vectors have been deprecated.
1185 Use the new functions scm_is_vector, scm_vector_elements,
1186 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1187 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1188 manual for more details.
1190 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1191 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1193 The following macros have been removed: SCM_VECTOR_BASE,
1194 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1195 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1197 ** Some C functions and macros related to arrays have been deprecated.
1199 Migrate according to the following table:
1201 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1202 scm_make_ra -> scm_make_array
1203 scm_shap2ra -> scm_make_array
1204 scm_cvref -> scm_c_generalized_vector_ref
1205 scm_ra_set_contp -> do not use
1206 scm_aind -> scm_array_handle_pos
1207 scm_raprin1 -> scm_display or scm_write
1209 SCM_ARRAYP -> scm_is_array
1210 SCM_ARRAY_NDIM -> scm_c_array_rank
1211 SCM_ARRAY_DIMS -> scm_array_handle_dims
1212 SCM_ARRAY_CONTP -> do not use
1213 SCM_ARRAY_MEM -> do not use
1214 SCM_ARRAY_V -> scm_array_handle_elements or similar
1215 SCM_ARRAY_BASE -> do not use
1217 ** SCM_CELL_WORD_LOC has been deprecated.
1219 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1220 to a SCM, as opposed to a pointer to a scm_t_bits.
1222 This was done to allow the correct use of pointers into the Scheme
1223 heap. Previously, the heap words were of type scm_t_bits and local
1224 variables and function arguments were of type SCM, making it
1225 non-standards-conformant to have a pointer that can point to both.
1227 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1229 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1230 second and third words of double smobs. Likewise for
1231 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1233 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1234 used to get and set the 16 exra bits in the zeroth word of a smob.
1236 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1237 accesing the first immediate word of a smob as a SCM value, and there
1238 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1239 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1241 ** New way to deal with non-local exits and re-entries.
1243 There is a new set of functions that essentially do what
1244 scm_internal_dynamic_wind does, but in a way that is more convenient
1245 for C code in some situations. Here is a quick example of how to
1246 prevent a potential memory leak:
1253 scm_dynwind_begin (0);
1255 mem = scm_malloc (100);
1256 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1258 /* MEM would leak if BAR throws an error.
1259 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1266 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1267 SCM_DYNWIND_END as well.
1271 For full documentation, see the node "Dynamic Wind" in the manual.
1273 ** New function scm_dynwind_free
1275 This function calls 'free' on a given pointer when a dynwind context
1276 is left. Thus the call to scm_dynwind_unwind_handler above could be
1277 replaced with simply scm_dynwind_free (mem).
1279 ** New functions scm_c_call_with_blocked_asyncs and
1280 scm_c_call_with_unblocked_asyncs
1282 Like scm_call_with_blocked_asyncs etc. but for C functions.
1284 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1286 In addition to scm_c_call_with_blocked_asyncs you can now also use
1287 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1288 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1290 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1291 SCM_REALLOW_INTS have been deprecated.
1293 They do no longer fulfill their original role of blocking signal
1294 delivery. Depending on what you want to achieve, replace a pair of
1295 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1296 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1299 ** The value 'scm_mask_ints' is no longer writable.
1301 Previously, you could set scm_mask_ints directly. This is no longer
1302 possible. Use scm_c_call_with_blocked_asyncs and
1303 scm_c_call_with_unblocked_asyncs instead.
1305 ** New way to temporarily set the current input, output or error ports
1307 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1308 context' (see above). <foo> is one of "input", "output" or "error".
1310 ** New way to temporarily set fluids
1312 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1313 above) to temporarily set the value of a fluid.
1315 ** New types scm_t_intmax and scm_t_uintmax.
1317 On platforms that have them, these types are identical to intmax_t and
1318 uintmax_t, respectively. On other platforms, they are identical to
1319 the largest integer types that Guile knows about.
1321 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1323 You should not have used them.
1325 ** Many public #defines with generic names have been made private.
1327 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1328 private or renamed with a more suitable public name.
1330 ** The macro SCM_TYP16S has been deprecated.
1332 This macro is not intended for public use.
1334 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1336 Use scm_is_true (scm_inexact_p (...)) instead.
1338 ** The macro SCM_SLOPPY_REALP has been deprecated.
1340 Use scm_is_real instead.
1342 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1344 Use scm_is_complex instead.
1346 ** Some preprocessor defines have been deprecated.
1348 These defines indicated whether a certain feature was present in Guile
1349 or not. Going forward, assume that the features are always present.
1351 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1352 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1354 The following macros have been removed completely: MEMOIZE_LOCALS,
1355 SCM_RECKLESS, SCM_CAUTIOUS.
1357 ** The preprocessor define STACK_DIRECTION has been deprecated.
1359 There should be no need to know about the stack direction for ordinary
1362 ** New function: scm_effective_version
1364 Returns the "effective" version number. This is just the normal full
1365 version string without the final micro-version number. See "Changes
1366 to the distribution" above.
1368 ** The function scm_call_with_new_thread has a new prototype.
1370 Instead of taking a list with the thunk and handler, these two
1371 arguments are now passed directly:
1373 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1375 This is an incompatible change.
1377 ** New snarfer macro SCM_DEFINE_PUBLIC.
1379 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1380 function in the init section.
1382 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1384 ** Garbage collector rewrite.
1386 The garbage collector is cleaned up a lot, and now uses lazy
1387 sweeping. This is reflected in the output of (gc-stats); since cells
1388 are being freed when they are allocated, the cells-allocated field
1389 stays roughly constant.
1391 For malloc related triggers, the behavior is changed. It uses the same
1392 heuristic as the cell-triggered collections. It may be tuned with the
1393 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1394 for minimum yield of malloc related triggers. The default is 40.
1395 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1398 Debugging operations for the freelist have been deprecated, along with
1399 the C variables that control garbage collection. The environment
1400 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1401 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1403 For understanding the memory usage of a GUILE program, the routine
1404 gc-live-object-stats returns an alist containing the number of live
1405 objects for every type.
1408 ** The function scm_definedp has been renamed to scm_defined_p
1410 The name scm_definedp is deprecated.
1412 ** The struct scm_cell type has been renamed to scm_t_cell
1414 This is in accordance to Guile's naming scheme for types. Note that
1415 the name scm_cell is now used for a function that allocates and
1416 initializes a new cell (see below).
1418 ** New functions for memory management
1420 A new set of functions for memory management has been added since the
1421 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1422 indeed, Guile itself contained some long standing bugs that could
1423 cause aborts in long running programs.
1425 The new functions are more symmetrical and do not need cooperation
1426 from smob free routines, among other improvements.
1428 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1429 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1430 scm_gc_free, scm_gc_register_collectable_memory, and
1431 scm_gc_unregister_collectable_memory. Refer to the manual for more
1432 details and for upgrading instructions.
1434 The old functions for memory management have been deprecated. They
1435 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1436 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1438 ** Declarations of exported features are marked with SCM_API.
1440 Every declaration of a feature that belongs to the exported Guile API
1441 has been marked by adding the macro "SCM_API" to the start of the
1442 declaration. This macro can expand into different things, the most
1443 common of which is just "extern" for Unix platforms. On Win32, it can
1444 be used to control which symbols are exported from a DLL.
1446 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1447 will expand into "__declspec (dllimport) extern", which is needed for
1448 linking to the Guile DLL in Windows.
1450 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1451 SCM_SRFI4_IMPORT, for the corresponding libraries.
1453 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1455 Use the new functions scm_cell and scm_double_cell instead. The old
1456 macros had problems because with them allocation and initialization
1457 was separated and the GC could sometimes observe half initialized
1458 cells. Only careful coding by the user of SCM_NEWCELL and
1459 SCM_NEWCELL2 could make this safe and efficient.
1461 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1463 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1466 ** SRCBRKP has been deprecated.
1468 Use scm_c_source_property_breakpoint_p instead.
1470 ** Deprecated: scm_makmacro
1472 Change your code to use either scm_makmmacro or to define macros in
1473 Scheme, using 'define-macro'.
1475 ** New function scm_c_port_for_each.
1477 This function is like scm_port_for_each but takes a pointer to a C
1478 function as the callback instead of a SCM value.
1480 ** The names scm_internal_select, scm_thread_sleep, and
1481 scm_thread_usleep have been discouraged.
1483 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1485 ** The GC can no longer be blocked.
1487 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1488 The GC can now run (partially) concurrently with other code and thus
1489 blocking it is not well defined.
1491 ** Many definitions have been removed that were previously deprecated.
1493 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1494 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1495 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1496 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1497 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1498 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1499 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1500 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1501 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1502 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1503 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1504 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1505 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1506 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1507 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1508 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1509 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1510 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1511 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1512 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1513 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1514 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1515 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1516 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1517 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1518 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1519 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1520 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1521 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1522 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1523 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1525 * Changes to bundled modules
1529 Using the (ice-9 debug) module no longer automatically switches Guile
1530 to use the debugging evaluator. If you want to switch to the
1531 debugging evaluator (which is needed for backtrace information if you
1532 hit an error), please add an explicit "(debug-enable 'debug)" to your
1533 code just after the code to use (ice-9 debug).
1536 Changes since Guile 1.4:
1538 * Changes to the distribution
1540 ** A top-level TODO file is included.
1542 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1544 Guile now always uses three numbers to represent the version,
1545 i.e. "1.6.5". The first number, 1, is the major version number, the
1546 second number, 6, is the minor version number, and the third number,
1547 5, is the micro version number. Changes in major version number
1548 indicate major changes in Guile.
1550 Minor version numbers that are even denote stable releases, and odd
1551 minor version numbers denote development versions (which may be
1552 unstable). The micro version number indicates a minor sub-revision of
1553 a given MAJOR.MINOR release.
1555 In keeping with the new scheme, (minor-version) and scm_minor_version
1556 no longer return everything but the major version number. They now
1557 just return the minor version number. Two new functions
1558 (micro-version) and scm_micro_version have been added to report the
1559 micro version number.
1561 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1563 ** New preprocessor definitions are available for checking versions.
1565 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1566 SCM_MICRO_VERSION to the appropriate integer values.
1568 ** Guile now actively warns about deprecated features.
1570 The new configure option `--enable-deprecated=LEVEL' and the
1571 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1572 See INSTALL and README for more information.
1574 ** Guile is much more likely to work on 64-bit architectures.
1576 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1577 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1578 for the use of a test machine, and thanks to Stefan Jahn for ia64
1581 ** New functions: setitimer and getitimer.
1583 These implement a fairly direct interface to the libc functions of the
1586 ** The #. reader extension is now disabled by default.
1588 For safety reasons, #. evaluation is disabled by default. To
1589 re-enable it, set the fluid read-eval? to #t. For example:
1591 (fluid-set! read-eval? #t)
1593 but make sure you realize the potential security risks involved. With
1594 read-eval? enabled, reading a data file from an untrusted source can
1597 ** New SRFI modules have been added:
1599 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1602 (srfi srfi-1) is a library containing many useful pair- and list-processing
1605 (srfi srfi-2) exports and-let*.
1607 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1609 (srfi srfi-6) is a dummy module for now, since guile already provides
1610 all of the srfi-6 procedures by default: open-input-string,
1611 open-output-string, get-output-string.
1613 (srfi srfi-8) exports receive.
1615 (srfi srfi-9) exports define-record-type.
1617 (srfi srfi-10) exports define-reader-ctor and implements the reader
1620 (srfi srfi-11) exports let-values and let*-values.
1622 (srfi srfi-13) implements the SRFI String Library.
1624 (srfi srfi-14) implements the SRFI Character-Set Library.
1626 (srfi srfi-17) implements setter and getter-with-setter and redefines
1627 some accessor procedures as procedures with getters. (such as car,
1628 cdr, vector-ref etc.)
1630 (srfi srfi-19) implements the SRFI Time/Date Library.
1632 ** New scripts / "executable modules"
1634 Subdirectory "scripts" contains Scheme modules that are packaged to
1635 also be executable as scripts. At this time, these scripts are available:
1644 See README there for more info.
1646 These scripts can be invoked from the shell with the new program
1647 "guile-tools", which keeps track of installation directory for you.
1650 $ guile-tools display-commentary srfi/*.scm
1652 guile-tools is copied to the standard $bindir on "make install".
1654 ** New module (ice-9 stack-catch):
1656 stack-catch is like catch, but saves the current state of the stack in
1657 the fluid the-last-stack. This fluid can be useful when using the
1658 debugger and when re-throwing an error.
1660 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1662 This has been done to prevent problems on lesser operating systems
1663 that can't tolerate `*'s in file names. The exported macro continues
1664 to be named `and-let*', of course.
1666 On systems that support it, there is also a compatibility module named
1667 (ice-9 and-let*). It will go away in the next release.
1669 ** New modules (oop goops) etc.:
1672 (oop goops describe)
1674 (oop goops active-slot)
1675 (oop goops composite-slot)
1677 The Guile Object Oriented Programming System (GOOPS) has been
1678 integrated into Guile. For further information, consult the GOOPS
1679 manual and tutorial in the `doc' directory.
1681 ** New module (ice-9 rdelim).
1683 This exports the following procedures which were previously defined
1684 in the default environment:
1686 read-line read-line! read-delimited read-delimited! %read-delimited!
1687 %read-line write-line
1689 For backwards compatibility the definitions are still imported into the
1690 default environment in this version of Guile. However you should add:
1692 (use-modules (ice-9 rdelim))
1694 to any program which uses the definitions, since this may change in
1697 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1698 can be used for similar functionality.
1700 ** New module (ice-9 rw)
1702 This is a subset of the (scsh rw) module from guile-scsh. Currently
1703 it defines two procedures:
1705 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1707 Read characters from a port or file descriptor into a string STR.
1708 A port must have an underlying file descriptor -- a so-called
1709 fport. This procedure is scsh-compatible and can efficiently read
1712 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1714 Write characters from a string STR to a port or file descriptor.
1715 A port must have an underlying file descriptor -- a so-called
1716 fport. This procedure is mostly compatible and can efficiently
1717 write large strings.
1719 ** New module (ice-9 match)
1721 This module includes Andrew K. Wright's pattern matcher. See
1722 ice-9/match.scm for brief description or
1724 http://www.star-lab.com/wright/code.html
1726 for complete documentation.
1728 ** New module (ice-9 buffered-input)
1730 This module provides procedures to construct an input port from an
1731 underlying source of input that reads and returns its input in chunks.
1732 The underlying input source is a Scheme procedure, specified by the
1733 caller, which the port invokes whenever it needs more input.
1735 This is useful when building an input port whose back end is Readline
1736 or a UI element such as the GtkEntry widget.
1740 The reference and tutorial documentation that was previously
1741 distributed separately, as `guile-doc', is now included in the core
1742 Guile distribution. The documentation consists of the following
1745 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1748 - The Guile Reference Manual (guile.texi) contains (or is intended to
1749 contain) reference documentation on all aspects of Guile.
1751 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1752 reference documentation for using GOOPS, Guile's Object Oriented
1755 - The Revised^5 Report on the Algorithmic Language Scheme
1758 See the README file in the `doc' directory for more details.
1760 ** There are a couple of examples in the examples/ directory now.
1762 * Changes to the stand-alone interpreter
1764 ** New command line option `--use-srfi'
1766 Using this option, SRFI modules can be loaded on startup and be
1767 available right from the beginning. This makes programming portable
1768 Scheme programs easier.
1770 The option `--use-srfi' expects a comma-separated list of numbers,
1771 each representing a SRFI number to be loaded into the interpreter
1772 before starting evaluating a script file or the REPL. Additionally,
1773 the feature identifier for the loaded SRFIs is recognized by
1774 `cond-expand' when using this option.
1777 $ guile --use-srfi=8,13
1778 guile> (receive (x z) (values 1 2) (+ 1 2))
1780 guile> (string-pad "bla" 20)
1783 ** Guile now always starts up in the `(guile-user)' module.
1785 Previously, scripts executed via the `-s' option would run in the
1786 `(guile)' module and the repl would run in the `(guile-user)' module.
1787 Now every user action takes place in the `(guile-user)' module by
1790 * Changes to Scheme functions and syntax
1792 ** Character classifiers work for non-ASCII characters.
1794 The predicates `char-alphabetic?', `char-numeric?',
1795 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1796 no longer check whether their arguments are ASCII characters.
1797 Previously, a character would only be considered alphabetic when it
1798 was also ASCII, for example.
1800 ** Previously deprecated Scheme functions have been removed:
1802 tag - no replacement.
1803 fseek - replaced by seek.
1804 list* - replaced by cons*.
1806 ** It's now possible to create modules with controlled environments
1810 (use-modules (ice-9 safe))
1811 (define m (make-safe-module))
1812 ;;; m will now be a module containing only a safe subset of R5RS
1813 (eval '(+ 1 2) m) --> 3
1814 (eval 'load m) --> ERROR: Unbound variable: load
1816 ** Evaluation of "()", the empty list, is now an error.
1818 Previously, the expression "()" evaluated to the empty list. This has
1819 been changed to signal a "missing expression" error. The correct way
1820 to write the empty list as a literal constant is to use quote: "'()".
1822 ** New concept of `Guile Extensions'.
1824 A Guile Extension is just a ordinary shared library that can be linked
1825 at run-time. We found it advantageous to give this simple concept a
1826 dedicated name to distinguish the issues related to shared libraries
1827 from the issues related to the module system.
1829 *** New function: load-extension
1831 Executing (load-extension lib init) is mostly equivalent to
1833 (dynamic-call init (dynamic-link lib))
1835 except when scm_register_extension has been called previously.
1836 Whenever appropriate, you should use `load-extension' instead of
1837 dynamic-link and dynamic-call.
1839 *** New C function: scm_c_register_extension
1841 This function registers a initialization function for use by
1842 `load-extension'. Use it when you don't want specific extensions to
1843 be loaded as shared libraries (for example on platforms that don't
1844 support dynamic linking).
1846 ** Auto-loading of compiled-code modules is deprecated.
1848 Guile used to be able to automatically find and link a shared
1849 library to satisfy requests for a module. For example, the module
1850 `(foo bar)' could be implemented by placing a shared library named
1851 "foo/libbar.so" (or with a different extension) in a directory on the
1854 This has been found to be too tricky, and is no longer supported. The
1855 shared libraries are now called "extensions". You should now write a
1856 small Scheme file that calls `load-extension' to load the shared
1857 library and initialize it explicitly.
1859 The shared libraries themselves should be installed in the usual
1860 places for shared libraries, with names like "libguile-foo-bar".
1862 For example, place this into a file "foo/bar.scm"
1864 (define-module (foo bar))
1866 (load-extension "libguile-foo-bar" "foobar_init")
1868 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1870 `eval' is now R5RS, that is it takes two arguments.
1871 The second argument is an environment specifier, i.e. either
1873 (scheme-report-environment 5)
1874 (null-environment 5)
1875 (interaction-environment)
1881 ** The module system has been made more disciplined.
1883 The function `eval' will save and restore the current module around
1884 the evaluation of the specified expression. While this expression is
1885 evaluated, `(current-module)' will now return the right module, which
1886 is the module specified as the second argument to `eval'.
1888 A consequence of this change is that `eval' is not particularly
1889 useful when you want allow the evaluated code to change what module is
1890 designated as the current module and have this change persist from one
1891 call to `eval' to the next. The read-eval-print-loop is an example
1892 where `eval' is now inadequate. To compensate, there is a new
1893 function `primitive-eval' that does not take a module specifier and
1894 that does not save/restore the current module. You should use this
1895 function together with `set-current-module', `current-module', etc
1896 when you want to have more control over the state that is carried from
1897 one eval to the next.
1899 Additionally, it has been made sure that forms that are evaluated at
1900 the top level are always evaluated with respect to the current module.
1901 Previously, subforms of top-level forms such as `begin', `case',
1902 etc. did not respect changes to the current module although these
1903 subforms are at the top-level as well.
1905 To prevent strange behavior, the forms `define-module',
1906 `use-modules', `use-syntax', and `export' have been restricted to only
1907 work on the top level. The forms `define-public' and
1908 `defmacro-public' only export the new binding on the top level. They
1909 behave just like `define' and `defmacro', respectively, when they are
1910 used in a lexical environment.
1912 Also, `export' will no longer silently re-export bindings imported
1913 from a used module. It will emit a `deprecation' warning and will
1914 cease to perform any re-export in the next version. If you actually
1915 want to re-export bindings, use the new `re-export' in place of
1916 `export'. The new `re-export' will not make copies of variables when
1917 rexporting them, as `export' did wrongly.
1919 ** Module system now allows selection and renaming of imported bindings
1921 Previously, when using `use-modules' or the `#:use-module' clause in
1922 the `define-module' form, all the bindings (association of symbols to
1923 values) for imported modules were added to the "current module" on an
1924 as-is basis. This has been changed to allow finer control through two
1925 new facilities: selection and renaming.
1927 You can now select which of the imported module's bindings are to be
1928 visible in the current module by using the `:select' clause. This
1929 clause also can be used to rename individual bindings. For example:
1931 ;; import all bindings no questions asked
1932 (use-modules (ice-9 common-list))
1934 ;; import four bindings, renaming two of them;
1935 ;; the current module sees: every some zonk-y zonk-n
1936 (use-modules ((ice-9 common-list)
1938 (remove-if . zonk-y)
1939 (remove-if-not . zonk-n))))
1941 You can also programmatically rename all selected bindings using the
1942 `:renamer' clause, which specifies a proc that takes a symbol and
1943 returns another symbol. Because it is common practice to use a prefix,
1944 we now provide the convenience procedure `symbol-prefix-proc'. For
1947 ;; import four bindings, renaming two of them specifically,
1948 ;; and all four w/ prefix "CL:";
1949 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1950 (use-modules ((ice-9 common-list)
1952 (remove-if . zonk-y)
1953 (remove-if-not . zonk-n))
1954 :renamer (symbol-prefix-proc 'CL:)))
1956 ;; import four bindings, renaming two of them specifically,
1957 ;; and all four by upcasing.
1958 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1959 (define (upcase-symbol sym)
1960 (string->symbol (string-upcase (symbol->string sym))))
1962 (use-modules ((ice-9 common-list)
1964 (remove-if . zonk-y)
1965 (remove-if-not . zonk-n))
1966 :renamer upcase-symbol))
1968 Note that programmatic renaming is done *after* individual renaming.
1969 Also, the above examples show `use-modules', but the same facilities are
1970 available for the `#:use-module' clause of `define-module'.
1972 See manual for more info.
1974 ** The semantics of guardians have changed.
1976 The changes are for the most part compatible. An important criterion
1977 was to keep the typical usage of guardians as simple as before, but to
1978 make the semantics safer and (as a result) more useful.
1980 *** All objects returned from guardians are now properly alive.
1982 It is now guaranteed that any object referenced by an object returned
1983 from a guardian is alive. It's now impossible for a guardian to
1984 return a "contained" object before its "containing" object.
1986 One incompatible (but probably not very important) change resulting
1987 from this is that it is no longer possible to guard objects that
1988 indirectly reference themselves (i.e. are parts of cycles). If you do
1989 so accidentally, you'll get a warning.
1991 *** There are now two types of guardians: greedy and sharing.
1993 If you call (make-guardian #t) or just (make-guardian), you'll get a
1994 greedy guardian, and for (make-guardian #f) a sharing guardian.
1996 Greedy guardians are the default because they are more "defensive".
1997 You can only greedily guard an object once. If you guard an object
1998 more than once, once in a greedy guardian and the rest of times in
1999 sharing guardians, then it is guaranteed that the object won't be
2000 returned from sharing guardians as long as it is greedily guarded
2003 Guardians returned by calls to `make-guardian' can now take one more
2004 optional parameter, which says whether to throw an error in case an
2005 attempt is made to greedily guard an object that is already greedily
2006 guarded. The default is true, i.e. throw an error. If the parameter
2007 is false, the guardian invocation returns #t if guarding was
2008 successful and #f if it wasn't.
2010 Also, since greedy guarding is, in effect, a side-effecting operation
2011 on objects, a new function is introduced: `destroy-guardian!'.
2012 Invoking this function on a guardian renders it unoperative and, if
2013 the guardian is greedy, clears the "greedily guarded" property of the
2014 objects that were guarded by it, thus undoing the side effect.
2016 Note that all this hair is hardly very important, since guardian
2017 objects are usually permanent.
2019 ** Continuations created by call-with-current-continuation now accept
2020 any number of arguments, as required by R5RS.
2022 ** New function `issue-deprecation-warning'
2024 This function is used to display the deprecation messages that are
2025 controlled by GUILE_WARN_DEPRECATION as explained in the README.
2028 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
2032 ;; `id' is deprecated. Use `identity' instead.
2037 ** New syntax `begin-deprecated'
2039 When deprecated features are included (as determined by the configure
2040 option --enable-deprecated), `begin-deprecated' is identical to
2041 `begin'. When deprecated features are excluded, it always evaluates
2042 to `#f', ignoring the body forms.
2044 ** New function `make-object-property'
2046 This function returns a new `procedure with setter' P that can be used
2047 to attach a property to objects. When calling P as
2051 where `obj' is any kind of object, it attaches `val' to `obj' in such
2052 a way that it can be retrieved by calling P as
2056 This function will replace procedure properties, symbol properties and
2057 source properties eventually.
2059 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2061 Instead of #&optional, #&key, etc you should now use #:optional,
2062 #:key, etc. Since #:optional is a keyword, you can write it as just
2063 :optional when (read-set! keywords 'prefix) is active.
2065 The old reader syntax `#&' is still supported, but deprecated. It
2066 will be removed in the next release.
2068 ** New define-module option: pure
2070 Tells the module system not to include any bindings from the root
2075 (define-module (totally-empty-module)
2078 ** New define-module option: export NAME1 ...
2080 Export names NAME1 ...
2082 This option is required if you want to be able to export bindings from
2083 a module which doesn't import one of `define-public' or `export'.
2087 (define-module (foo)
2089 :use-module (ice-9 r5rs)
2092 ;;; Note that we're pure R5RS below this point!
2097 ** New function: object->string OBJ
2099 Return a Scheme string obtained by printing a given object.
2101 ** New function: port? X
2103 Returns a boolean indicating whether X is a port. Equivalent to
2104 `(or (input-port? X) (output-port? X))'.
2106 ** New function: file-port?
2108 Determines whether a given object is a port that is related to a file.
2110 ** New function: port-for-each proc
2112 Apply PROC to each port in the Guile port table in turn. The return
2113 value is unspecified. More specifically, PROC is applied exactly once
2114 to every port that exists in the system at the time PORT-FOR-EACH is
2115 invoked. Changes to the port table while PORT-FOR-EACH is running
2116 have no effect as far as PORT-FOR-EACH is concerned.
2118 ** New function: dup2 oldfd newfd
2120 A simple wrapper for the `dup2' system call. Copies the file
2121 descriptor OLDFD to descriptor number NEWFD, replacing the
2122 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2123 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2124 to move away ports which are using NEWFD. The return value is
2127 ** New function: close-fdes fd
2129 A simple wrapper for the `close' system call. Close file
2130 descriptor FD, which must be an integer. Unlike close (*note
2131 close: Ports and File Descriptors.), the file descriptor will be
2132 closed even if a port is using it. The return value is
2135 ** New function: crypt password salt
2137 Encrypts `password' using the standard unix password encryption
2140 ** New function: chroot path
2142 Change the root directory of the running process to `path'.
2144 ** New functions: getlogin, cuserid
2146 Return the login name or the user name of the current effective user
2149 ** New functions: getpriority which who, setpriority which who prio
2151 Get or set the priority of the running process.
2153 ** New function: getpass prompt
2155 Read a password from the terminal, first displaying `prompt' and
2158 ** New function: flock file operation
2160 Set/remove an advisory shared or exclusive lock on `file'.
2162 ** New functions: sethostname name, gethostname
2164 Set or get the hostname of the machine the current process is running
2167 ** New function: mkstemp! tmpl
2169 mkstemp creates a new unique file in the file system and returns a
2170 new buffered port open for reading and writing to the file. TMPL
2171 is a string specifying where the file should be created: it must
2172 end with `XXXXXX' and will be changed in place to return the name
2173 of the temporary file.
2175 ** New function: open-input-string string
2177 Return an input string port which delivers the characters from
2178 `string'. This procedure, together with `open-output-string' and
2179 `get-output-string' implements SRFI-6.
2181 ** New function: open-output-string
2183 Return an output string port which collects all data written to it.
2184 The data can then be retrieved by `get-output-string'.
2186 ** New function: get-output-string
2188 Return the contents of an output string port.
2190 ** New function: identity
2192 Return the argument.
2194 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2195 are represented in Scheme as integers with normal host byte ordering.
2197 ** New function: inet-pton family address
2199 Convert a printable string network address into an integer. Note that
2200 unlike the C version of this function, the result is an integer with
2201 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2204 (inet-pton AF_INET "127.0.0.1") => 2130706433
2205 (inet-pton AF_INET6 "::1") => 1
2207 ** New function: inet-ntop family address
2209 Convert an integer network address into a printable string. Note that
2210 unlike the C version of this function, the input is an integer with
2211 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2214 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2215 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2216 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2220 Use `identity' instead.
2226 ** Deprecated: return-it
2230 ** Deprecated: string-character-length
2232 Use `string-length' instead.
2234 ** Deprecated: flags
2236 Use `logior' instead.
2238 ** Deprecated: close-all-ports-except.
2240 This was intended for closing ports in a child process after a fork,
2241 but it has the undesirable side effect of flushing buffers.
2242 port-for-each is more flexible.
2244 ** The (ice-9 popen) module now attempts to set up file descriptors in
2245 the child process from the current Scheme ports, instead of using the
2246 current values of file descriptors 0, 1, and 2 in the parent process.
2248 ** Removed function: builtin-weak-bindings
2250 There is no such concept as a weak binding any more.
2252 ** Removed constants: bignum-radix, scm-line-incrementors
2254 ** define-method: New syntax mandatory.
2256 The new method syntax is now mandatory:
2258 (define-method (NAME ARG-SPEC ...) BODY ...)
2259 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2261 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2262 REST-ARG ::= ARG-NAME
2264 If you have old code using the old syntax, import
2265 (oop goops old-define-method) before (oop goops) as in:
2267 (use-modules (oop goops old-define-method) (oop goops))
2269 ** Deprecated function: builtin-variable
2270 Removed function: builtin-bindings
2272 There is no longer a distinction between builtin or other variables.
2273 Use module system operations for all variables.
2275 ** Lazy-catch handlers are no longer allowed to return.
2277 That is, a call to `throw', `error', etc is now guaranteed to not
2280 ** Bugfixes for (ice-9 getopt-long)
2282 This module is now tested using test-suite/tests/getopt-long.test.
2283 The following bugs have been fixed:
2285 *** Parsing for options that are specified to have `optional' args now checks
2286 if the next element is an option instead of unconditionally taking it as the
2289 *** An error is now thrown for `--opt=val' when the option description
2290 does not specify `(value #t)' or `(value optional)'. This condition used to
2291 be accepted w/o error, contrary to the documentation.
2293 *** The error message for unrecognized options is now more informative.
2294 It used to be "not a record", an artifact of the implementation.
2296 *** The error message for `--opt' terminating the arg list (no value), when
2297 `(value #t)' is specified, is now more informative. It used to be "not enough
2300 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2301 The expansion used to be like so:
2303 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2305 Note that the "5d" is dropped. Now it is like so:
2307 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2309 This enables single-char options to have adjoining arguments as long as their
2310 constituent characters are not potential single-char options.
2312 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2314 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2315 property `arglist', which can be retrieved by `arity'. The result is that
2316 `arity' can give more detailed information than before:
2320 guile> (use-modules (ice-9 optargs))
2321 guile> (define* (foo #:optional a b c) a)
2323 0 or more arguments in `lambda*:G0'.
2328 3 optional arguments: `a', `b' and `c'.
2329 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2331 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2332 and `d', other keywords allowed.
2333 guile> (define* (baz a b #:optional c #:rest r) a)
2335 2 required arguments: `a' and `b', 1 optional argument: `c',
2338 * Changes to the C interface
2340 ** Types have been renamed from scm_*_t to scm_t_*.
2342 This has been done for POSIX sake. It reserves identifiers ending
2343 with "_t". What a concept.
2345 The old names are still available with status `deprecated'.
2347 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2349 ** Deprecated features have been removed.
2353 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2354 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2356 *** C Functions removed
2358 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2359 scm_fseek - replaced by scm_seek.
2360 gc-thunk - replaced by after-gc-hook.
2361 gh_int2scmb - replaced by gh_bool2scm.
2362 scm_tc_dblr - replaced by scm_tc16_real.
2363 scm_tc_dblc - replaced by scm_tc16_complex.
2364 scm_list_star - replaced by scm_cons_star.
2366 ** Deprecated: scm_makfromstr
2368 Use scm_mem2string instead.
2370 ** Deprecated: scm_make_shared_substring
2372 Explicit shared substrings will disappear from Guile.
2374 Instead, "normal" strings will be implemented using sharing
2375 internally, combined with a copy-on-write strategy.
2377 ** Deprecated: scm_read_only_string_p
2379 The concept of read-only strings will disappear in next release of
2382 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2384 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2386 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2388 Call a procedure with the indicated number of arguments. See "Fly
2389 Evaluation" in the manual.
2391 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2393 Call a procedure with the indicated number of arguments and a list of
2394 further arguments. See "Fly Evaluation" in the manual.
2396 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2398 Create a list of the given number of elements. See "List
2399 Constructors" in the manual.
2401 ** Renamed function: scm_listify has been replaced by scm_list_n.
2403 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2404 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2406 Use functions scm_list_N instead.
2408 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2410 Used by an application to read arbitrary number of bytes from a port.
2411 Same semantics as libc read, except that scm_c_read only returns less
2412 than SIZE bytes if at end-of-file.
2414 Warning: Doesn't update port line and column counts!
2416 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2418 Used by an application to write arbitrary number of bytes to an SCM
2419 port. Similar semantics as libc write. However, unlike libc
2420 write, scm_c_write writes the requested number of bytes and has no
2423 Warning: Doesn't update port line and column counts!
2425 ** New function: scm_init_guile ()
2427 In contrast to scm_boot_guile, scm_init_guile will return normally
2428 after initializing Guile. It is not available on all systems, tho.
2430 ** New functions: scm_str2symbol, scm_mem2symbol
2432 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2433 field of characters and creates a scheme symbol object from that C string.
2434 The function scm_mem2symbol takes a const char* and a number of characters and
2435 creates a symbol from the characters in that memory area.
2437 ** New functions: scm_primitive_make_property
2438 scm_primitive_property_ref
2439 scm_primitive_property_set_x
2440 scm_primitive_property_del_x
2442 These functions implement a new way to deal with object properties.
2443 See libguile/properties.c for their documentation.
2445 ** New function: scm_done_free (long size)
2447 This function is the inverse of scm_done_malloc. Use it to report the
2448 amount of smob memory you free. The previous method, which involved
2449 calling scm_done_malloc with negative argument, was somewhat
2450 unintuitive (and is still available, of course).
2452 ** New function: scm_c_memq (SCM obj, SCM list)
2454 This function provides a fast C level alternative for scm_memq for the case
2455 that the list parameter is known to be a proper list. The function is a
2456 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2457 list input parameter, since for anything else but a proper list the function's
2458 behaviour is undefined - it may even crash or loop endlessly. Further, for
2459 the case that the object is not found in the list, scm_c_memq returns #f which
2460 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2462 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2463 scm_remember_upto_here
2465 These functions replace the function scm_remember.
2467 ** Deprecated function: scm_remember
2469 Use one of the new functions scm_remember_upto_here_1,
2470 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2472 ** New function: scm_allocate_string
2474 This function replaces the function scm_makstr.
2476 ** Deprecated function: scm_makstr
2478 Use the new function scm_allocate_string instead.
2480 ** New global variable scm_gc_running_p introduced.
2482 Use this variable to find out if garbage collection is being executed. Up to
2483 now applications have used scm_gc_heap_lock to test if garbage collection was
2484 running, which also works because of the fact that up to know only the garbage
2485 collector has set this variable. But, this is an implementation detail that
2486 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2487 of this variable is (and has been) not fully safe anyway.
2489 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2491 Use these instead of SCM_LENGTH_MAX.
2493 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2494 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2495 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2497 Use these instead of SCM_LENGTH.
2499 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2500 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2501 SCM_SET_BITVECTOR_LENGTH
2503 Use these instead of SCM_SETLENGTH
2505 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2506 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2509 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2512 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2513 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2516 Use these instead of SCM_SETCHARS.
2518 ** New macro: SCM_BITVECTOR_P
2520 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2522 Use instead of SCM_COERCE_SUBSTR.
2524 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2526 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2528 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2529 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2530 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2531 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2532 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2533 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2534 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2535 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2536 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2537 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2538 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2539 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2540 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2541 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2542 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2544 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2545 Use scm_memory_error instead of SCM_NALLOC.
2546 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2547 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2548 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2549 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2550 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2551 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2552 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2553 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2554 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2555 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2556 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2557 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2558 Use a type specific setter macro instead of SCM_SETLENGTH.
2559 Use a type specific setter macro instead of SCM_SETCHARS.
2560 Use a type specific length macro instead of SCM_LENGTH_MAX.
2561 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2562 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2563 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2564 Use SCM_TYP16 instead of SCM_GCTYP16.
2565 Use SCM_CDR instead of SCM_GCCDR.
2566 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2567 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2568 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2569 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2570 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2571 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2572 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2574 ** Removed function: scm_struct_init
2576 ** Removed variable: scm_symhash_dim
2578 ** Renamed function: scm_make_cont has been replaced by
2579 scm_make_continuation, which has a different interface.
2581 ** Deprecated function: scm_call_catching_errors
2583 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2585 ** Deprecated function: scm_strhash
2587 Use scm_string_hash instead.
2589 ** Deprecated function: scm_vector_set_length_x
2591 Instead, create a fresh vector of the desired size and copy the contents.
2593 ** scm_gensym has changed prototype
2595 scm_gensym now only takes one argument.
2597 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2600 There is now only a single symbol type scm_tc7_symbol.
2601 The tag scm_tc7_lvector was not used anyway.
2603 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2605 Use scm_make_smob_type and scm_set_smob_XXX instead.
2607 ** New function scm_set_smob_apply.
2609 This can be used to set an apply function to a smob type.
2611 ** Deprecated function: scm_strprint_obj
2613 Use scm_object_to_string instead.
2615 ** Deprecated function: scm_wta
2617 Use scm_wrong_type_arg, or another appropriate error signalling function
2620 ** Explicit support for obarrays has been deprecated.
2622 Use `scm_str2symbol' and the generic hashtable functions instead.
2624 ** The concept of `vcells' has been deprecated.
2626 The data type `variable' is now used exclusively. `Vcells' have been
2627 a low-level concept so you are likely not affected by this change.
2629 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2630 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2632 Use scm_c_define or scm_c_lookup instead, as appropriate.
2634 *** New functions: scm_c_module_lookup, scm_c_lookup,
2635 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2636 scm_module_define, scm_define.
2638 These functions work with variables instead of with vcells.
2640 ** New functions for creating and defining `subr's and `gsubr's.
2642 The new functions more clearly distinguish between creating a subr (or
2643 gsubr) object and adding it to the current module.
2645 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2646 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2647 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2648 scm_c_define_gsubr_with_generic.
2650 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2651 scm_make_subr_with_generic, scm_make_gsubr,
2652 scm_make_gsubr_with_generic.
2654 Use the new ones from above instead.
2656 ** C interface to the module system has changed.
2658 While we suggest that you avoid as many explicit module system
2659 operations from C as possible for the time being, the C interface has
2660 been made more similar to the high-level Scheme module system.
2662 *** New functions: scm_c_define_module, scm_c_use_module,
2663 scm_c_export, scm_c_resolve_module.
2665 They mostly work like their Scheme namesakes. scm_c_define_module
2666 takes a function that is called a context where the new module is
2669 *** Deprecated functions: scm_the_root_module, scm_make_module,
2670 scm_ensure_user_module, scm_load_scheme_module.
2672 Use the new functions instead.
2674 ** Renamed function: scm_internal_with_fluids becomes
2677 scm_internal_with_fluids is available as a deprecated function.
2679 ** New function: scm_c_with_fluid.
2681 Just like scm_c_with_fluids, but takes one fluid and one value instead
2684 ** Deprecated typedefs: long_long, ulong_long.
2686 They are of questionable utility and they pollute the global
2689 ** Deprecated typedef: scm_sizet
2691 It is of questionable utility now that Guile requires ANSI C, and is
2694 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2695 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2696 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2698 Made more compliant with the naming policy by adding a _t at the end.
2700 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2701 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2703 With the exception of the mysterious scm_2ulong2big, they are still
2704 available under new names (scm_i_mkbig etc). These functions are not
2705 intended to be used in user code. You should avoid dealing with
2706 bignums directly, and should deal with numbers in general (which can
2709 ** Change in behavior: scm_num2long, scm_num2ulong
2711 The scm_num2[u]long functions don't any longer accept an inexact
2712 argument. This change in behavior is motivated by concordance with
2713 R5RS: It is more common that a primitive doesn't want to accept an
2714 inexact for an exact.
2716 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2717 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2718 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2721 These are conversion functions between the various ANSI C integral
2722 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2723 accept an inexact argument.
2725 ** New functions: scm_float2num, scm_double2num,
2726 scm_num2float, scm_num2double.
2728 These are conversion functions between the two ANSI C float types and
2731 ** New number validation macros:
2732 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2736 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2738 These are just nicer-named old scm_protect_object and
2739 scm_unprotect_object.
2741 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2743 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2745 These functions can be used to register pointers to locations that
2748 ** Deprecated function: scm_create_hook.
2750 Its sins are: misleading name, non-modularity and lack of general
2754 Changes since Guile 1.3.4:
2756 * Changes to the distribution
2758 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2760 We've changed the way we handle generated files in the Guile source
2761 repository. As a result, the procedure for building trees obtained
2762 from the nightly FTP snapshots or via CVS has changed:
2763 - You must have appropriate versions of autoconf, automake, and
2764 libtool installed on your system. See README for info on how to
2765 obtain these programs.
2766 - Before configuring the tree, you must first run the script
2767 `autogen.sh' at the top of the source tree.
2769 The Guile repository used to contain not only source files, written by
2770 humans, but also some generated files, like configure scripts and
2771 Makefile.in files. Even though the contents of these files could be
2772 derived mechanically from other files present, we thought it would
2773 make the tree easier to build if we checked them into CVS.
2775 However, this approach means that minor differences between
2776 developer's installed tools and habits affected the whole team.
2777 So we have removed the generated files from the repository, and
2778 added the autogen.sh script, which will reconstruct them
2782 ** configure now has experimental options to remove support for certain
2785 --disable-arrays omit array and uniform array support
2786 --disable-posix omit posix interfaces
2787 --disable-networking omit networking interfaces
2788 --disable-regex omit regular expression interfaces
2790 These are likely to become separate modules some day.
2792 ** New configure option --enable-debug-freelist
2794 This enables a debugging version of SCM_NEWCELL(), and also registers
2795 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2797 Configure with the --enable-debug-freelist option to enable
2798 the gc-set-debug-check-freelist! primitive, and then use:
2800 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2801 (gc-set-debug-check-freelist! #f) # turn off checking
2803 Checking of the freelist forces a traversal of the freelist and
2804 a garbage collection before each allocation of a cell. This can
2805 slow down the interpreter dramatically, so the setter should be used to
2806 turn on this extra processing only when necessary.
2808 ** New configure option --enable-debug-malloc
2810 Include code for debugging of calls to scm_must_malloc/realloc/free.
2814 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2815 2. objects reallocated by scm_must_realloc has been allocated by
2817 3. reallocated objects are reallocated with the same what string
2819 But, most importantly, it records the number of allocated objects of
2820 each kind. This is useful when searching for memory leaks.
2822 A Guile compiled with this option provides the primitive
2823 `malloc-stats' which returns an alist with pairs of kind and the
2824 number of objects of that kind.
2826 ** All includes are now referenced relative to the root directory
2828 Since some users have had problems with mixups between Guile and
2829 system headers, we have decided to always refer to Guile headers via
2830 their parent directories. This essentially creates a "private name
2831 space" for Guile headers. This means that the compiler only is given
2832 -I options for the root build and root source directory.
2834 ** Header files kw.h and genio.h have been removed.
2836 ** The module (ice-9 getopt-gnu-style) has been removed.
2838 ** New module (ice-9 documentation)
2840 Implements the interface to documentation strings associated with
2843 ** New module (ice-9 time)
2845 Provides a macro `time', which displays execution time of a given form.
2847 ** New module (ice-9 history)
2849 Loading this module enables value history in the repl.
2851 * Changes to the stand-alone interpreter
2853 ** New command line option --debug
2855 Start Guile with debugging evaluator and backtraces enabled.
2857 This is useful when debugging your .guile init file or scripts.
2859 ** New help facility
2861 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2862 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2863 (help 'NAME) gives documentation for NAME, even if it is not an object
2864 (help ,EXPR) gives documentation for object returned by EXPR
2865 (help (my module)) gives module commentary for `(my module)'
2866 (help) gives this text
2868 `help' searches among bindings exported from loaded modules, while
2869 `apropos' searches among bindings visible from the "current" module.
2871 Examples: (help help)
2873 (help "output-string")
2875 ** `help' and `apropos' now prints full module names
2877 ** Dynamic linking now uses libltdl from the libtool package.
2879 The old system dependent code for doing dynamic linking has been
2880 replaced with calls to the libltdl functions which do all the hairy
2883 The major improvement is that you can now directly pass libtool
2884 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2885 will be able to do the best shared library job you can get, via
2888 The way dynamic libraries are found has changed and is not really
2889 portable across platforms, probably. It is therefore recommended to
2890 use absolute filenames when possible.
2892 If you pass a filename without an extension to `dynamic-link', it will
2893 try a few appropriate ones. Thus, the most platform ignorant way is
2894 to specify a name like "libfoo", without any directories and
2897 ** Guile COOP threads are now compatible with LinuxThreads
2899 Previously, COOP threading wasn't possible in applications linked with
2900 Linux POSIX threads due to their use of the stack pointer to find the
2901 thread context. This has now been fixed with a workaround which uses
2902 the pthreads to allocate the stack.
2904 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2906 ** Positions of erring expression in scripts
2908 With version 1.3.4, the location of the erring expression in Guile
2909 scipts is no longer automatically reported. (This should have been
2910 documented before the 1.3.4 release.)
2912 You can get this information by enabling recording of positions of
2913 source expressions and running the debugging evaluator. Put this at
2914 the top of your script (or in your "site" file):
2916 (read-enable 'positions)
2917 (debug-enable 'debug)
2919 ** Backtraces in scripts
2921 It is now possible to get backtraces in scripts.
2925 (debug-enable 'debug 'backtrace)
2927 at the top of the script.
2929 (The first options enables the debugging evaluator.
2930 The second enables backtraces.)
2932 ** Part of module system symbol lookup now implemented in C
2934 The eval closure of most modules is now implemented in C. Since this
2935 was one of the bottlenecks for loading speed, Guile now loads code
2936 substantially faster than before.
2938 ** Attempting to get the value of an unbound variable now produces
2939 an exception with a key of 'unbound-variable instead of 'misc-error.
2941 ** The initial default output port is now unbuffered if it's using a
2942 tty device. Previously in this situation it was line-buffered.
2944 ** New hook: after-gc-hook
2946 after-gc-hook takes over the role of gc-thunk. This hook is run at
2947 the first SCM_TICK after a GC. (Thus, the code is run at the same
2948 point during evaluation as signal handlers.)
2950 Note that this hook should be used only for diagnostic and debugging
2951 purposes. It is not certain that it will continue to be well-defined
2952 when this hook is run in the future.
2954 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2955 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2957 ** Improvements to garbage collector
2959 Guile 1.4 has a new policy for triggering heap allocation and
2960 determining the sizes of heap segments. It fixes a number of problems
2963 1. The new policy can handle two separate pools of cells
2964 (2-word/4-word) better. (The old policy would run wild, allocating
2965 more and more memory for certain programs.)
2967 2. The old code would sometimes allocate far too much heap so that the
2968 Guile process became gigantic. The new code avoids this.
2970 3. The old code would sometimes allocate too little so that few cells
2971 were freed at GC so that, in turn, too much time was spent in GC.
2973 4. The old code would often trigger heap allocation several times in a
2974 row. (The new scheme predicts how large the segments needs to be
2975 in order not to need further allocation.)
2977 All in all, the new GC policy will make larger applications more
2980 The new GC scheme also is prepared for POSIX threading. Threads can
2981 allocate private pools of cells ("clusters") with just a single
2982 function call. Allocation of single cells from such a cluster can
2983 then proceed without any need of inter-thread synchronization.
2985 ** New environment variables controlling GC parameters
2987 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2990 Allocation of 2-word cell heaps:
2992 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2995 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2996 GC in percent of total heap size
2999 Allocation of 4-word cell heaps
3000 (used for real numbers and misc other objects):
3002 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
3004 (See entry "Way for application to customize GC parameters" under
3005 section "Changes to the scm_ interface" below.)
3007 ** Guile now implements reals using 4-word cells
3009 This speeds up computation with reals. (They were earlier allocated
3010 with `malloc'.) There is still some room for optimizations, however.
3012 ** Some further steps toward POSIX thread support have been taken
3014 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
3015 don't have much effect any longer, and many of them will be removed in
3019 are only handled at the top of the evaluator loop, immediately after
3020 I/O, and in scm_equalp.
3022 *** The GC can allocate thread private pools of pairs.
3024 * Changes to Scheme functions and syntax
3026 ** close-input-port and close-output-port are now R5RS
3028 These procedures have been turned into primitives and have R5RS behaviour.
3030 ** New procedure: simple-format PORT MESSAGE ARG1 ...
3032 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
3033 extended by the more sophisticated version in (ice-9 format)
3035 (simple-format port message . args)
3036 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
3037 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
3038 the escapes are replaced with corresponding members of ARGS:
3039 ~A formats using `display' and ~S formats using `write'.
3040 If DESTINATION is #t, then use the `current-output-port',
3041 if DESTINATION is #f, then return a string containing the formatted text.
3042 Does not add a trailing newline."
3044 ** string-ref: the second argument is no longer optional.
3046 ** string, list->string: no longer accept strings in their arguments,
3047 only characters, for compatibility with R5RS.
3049 ** New procedure: port-closed? PORT
3050 Returns #t if PORT is closed or #f if it is open.
3052 ** Deprecated: list*
3054 The list* functionality is now provided by cons* (SRFI-1 compliant)
3056 ** New procedure: cons* ARG1 ARG2 ... ARGn
3058 Like `list', but the last arg provides the tail of the constructed list,
3059 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3061 Requires at least one argument. If given one argument, that argument
3062 is returned as result.
3064 This function is called `list*' in some other Schemes and in Common LISP.
3066 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3068 ** New procedure: object-documentation OBJECT
3070 Returns the documentation string associated with OBJECT. The
3071 procedure uses a caching mechanism so that subsequent lookups are
3074 Exported by (ice-9 documentation).
3076 ** module-name now returns full names of modules
3078 Previously, only the last part of the name was returned (`session' for
3079 `(ice-9 session)'). Ex: `(ice-9 session)'.
3081 * Changes to the gh_ interface
3083 ** Deprecated: gh_int2scmb
3085 Use gh_bool2scm instead.
3087 * Changes to the scm_ interface
3089 ** Guile primitives now carry docstrings!
3091 Thanks to Greg Badros!
3093 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3095 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3096 macros and must contain a docstring that is extracted into foo.doc using a new
3097 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3099 However, a major overhaul of these macros is scheduled for the next release of
3102 ** Guile primitives use a new technique for validation of arguments
3104 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3105 the readability of argument checking.
3107 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3109 ** New macros: SCM_PACK, SCM_UNPACK
3111 Compose/decompose an SCM value.
3113 The SCM type is now treated as an abstract data type and may be defined as a
3114 long, a void* or as a struct, depending on the architecture and compile time
3115 options. This makes it easier to find several types of bugs, for example when
3116 SCM values are treated as integers without conversion. Values of the SCM type
3117 should be treated as "atomic" values. These macros are used when
3118 composing/decomposing an SCM value, either because you want to access
3119 individual bits, or because you want to treat it as an integer value.
3121 E.g., in order to set bit 7 in an SCM value x, use the expression
3123 SCM_PACK (SCM_UNPACK (x) | 0x80)
3125 ** The name property of hooks is deprecated.
3126 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3128 You can emulate this feature by using object properties.
3130 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3131 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3134 These macros will be removed in a future release of Guile.
3136 ** The following types, functions and macros from numbers.h are deprecated:
3137 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3138 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3140 ** Port internals: the rw_random variable in the scm_port structure
3141 must be set to non-zero in any random access port. In recent Guile
3142 releases it was only set for bidirectional random-access ports.
3144 ** Port internals: the seek ptob procedure is now responsible for
3145 resetting the buffers if required. The change was made so that in the
3146 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3147 the fport and strport ptobs can avoid resetting the buffers,
3148 in particular to avoid discarding unread chars. An existing port
3149 type can be fixed by adding something like the following to the
3150 beginning of the ptob seek procedure:
3152 if (pt->rw_active == SCM_PORT_READ)
3153 scm_end_input (object);
3154 else if (pt->rw_active == SCM_PORT_WRITE)
3155 ptob->flush (object);
3157 although to actually avoid resetting the buffers and discard unread
3158 chars requires further hacking that depends on the characteristics
3161 ** Deprecated functions: scm_fseek, scm_tag
3163 These functions are no longer used and will be removed in a future version.
3165 ** The scm_sysmissing procedure is no longer used in libguile.
3166 Unless it turns out to be unexpectedly useful to somebody, it will be
3167 removed in a future version.
3169 ** The format of error message strings has changed
3171 The two C procedures: scm_display_error and scm_error, as well as the
3172 primitive `scm-error', now use scm_simple_format to do their work.
3173 This means that the message strings of all code must be updated to use
3174 ~A where %s was used before, and ~S where %S was used before.
3176 During the period when there still are a lot of old Guiles out there,
3177 you might want to support both old and new versions of Guile.
3179 There are basically two methods to achieve this. Both methods use
3182 AC_CHECK_FUNCS(scm_simple_format)
3184 in your configure.in.
3186 Method 1: Use the string concatenation features of ANSI C's
3191 #ifdef HAVE_SCM_SIMPLE_FORMAT
3197 Then represent each of your error messages using a preprocessor macro:
3199 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3203 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3204 (define make-message string-append)
3206 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3208 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3212 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3217 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3221 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3223 Don't use the functions coop_mutex_init and
3224 coop_condition_variable_init. They will change.
3226 Use scm_mutex_init and scm_cond_init instead.
3228 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3229 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3230 COND, as `scm_cond_wait' does, but it also bounds the duration
3231 of the wait. If COND has not been signaled before time ABSTIME,
3232 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3233 returns the error code `ETIMEDOUT'.
3235 The ABSTIME parameter specifies an absolute time, with the same
3236 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3237 to 00:00:00 GMT, January 1, 1970.
3239 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3240 `scm_cond_broadcast' restarts all the threads that are waiting
3241 on the condition variable COND. Nothing happens if no threads are
3244 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3245 `scm_key_create' allocates a new TSD key. The key is stored in
3246 the location pointed to by KEY. There is no limit on the number
3247 of keys allocated at a given time. The value initially associated
3248 with the returned key is `NULL' in all currently executing threads.
3250 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3251 function associated with the key. When a thread terminates,
3252 DESTR_FUNCTION is called on the value associated with the key in
3253 that thread. The DESTR_FUNCTION is not called if a key is deleted
3254 with `scm_key_delete' or a value is changed with
3255 `scm_setspecific'. The order in which destructor functions are
3256 called at thread termination time is unspecified.
3258 Destructors are not yet implemented.
3260 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3261 `scm_setspecific' changes the value associated with KEY in the
3262 calling thread, storing the given POINTER instead.
3264 ** New function: scm_getspecific (scm_key_t KEY)
3265 `scm_getspecific' returns the value currently associated with
3266 KEY in the calling thread.
3268 ** New function: scm_key_delete (scm_key_t KEY)
3269 `scm_key_delete' deallocates a TSD key. It does not check
3270 whether non-`NULL' values are associated with that key in the
3271 currently executing threads, nor call the destructor function
3272 associated with the key.
3274 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3276 Initialize a C level hook HOOK with associated HOOK_DATA and type
3277 TYPE. (See scm_c_hook_run ().)
3279 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3281 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3282 is true, add it last, otherwise first. The same FUNC can be added
3283 multiple times if FUNC_DATA differ and vice versa.
3285 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3287 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3288 function is only removed if both FUNC and FUNC_DATA matches.
3290 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3292 Run hook HOOK passing DATA to the hook functions.
3294 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3295 returned is undefined.
3297 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3298 returns a non-NULL value. This value is returned as the result of
3299 scm_c_hook_run. If all functions return NULL, NULL is returned.
3301 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3302 returns a NULL value, and NULL is returned. If all functions returns
3303 a non-NULL value, the last value is returned.
3305 ** New C level GC hooks
3307 Five new C level hooks has been added to the garbage collector.
3309 scm_before_gc_c_hook
3312 are run before locking and after unlocking the heap. The system is
3313 thus in a mode where evaluation can take place. (Except that
3314 scm_before_gc_c_hook must not allocate new cells.)
3316 scm_before_mark_c_hook
3317 scm_before_sweep_c_hook
3318 scm_after_sweep_c_hook
3320 are run when the heap is locked. These are intended for extension of
3321 the GC in a modular fashion. Examples are the weaks and guardians
3324 ** Way for application to customize GC parameters
3326 The application can set up other default values for the GC heap
3327 allocation parameters
3329 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3330 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3331 GUILE_MAX_SEGMENT_SIZE,
3335 scm_default_init_heap_size_1, scm_default_min_yield_1,
3336 scm_default_init_heap_size_2, scm_default_min_yield_2,
3337 scm_default_max_segment_size
3339 respectively before callong scm_boot_guile.
3341 (See entry "New environment variables ..." in section
3342 "Changes to the stand-alone interpreter" above.)
3344 ** scm_protect_object/scm_unprotect_object now nest
3346 This means that you can call scm_protect_object multiple times on an
3347 object and count on the object being protected until
3348 scm_unprotect_object has been call the same number of times.
3350 The functions also have better time complexity.
3352 Still, it is usually possible to structure the application in a way
3353 that you don't need to use these functions. For example, if you use a
3354 protected standard Guile list to keep track of live objects rather
3355 than some custom data type, objects will die a natural death when they
3356 are no longer needed.
3358 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3360 Guile does not provide the float representation for inexact real numbers any
3361 more. Now, only doubles are used to represent inexact real numbers. Further,
3362 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3363 and scm_tc16_complex, respectively.
3365 ** Removed deprecated type scm_smobfuns
3367 ** Removed deprecated function scm_newsmob
3369 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3371 There is an ongoing discussion among the developers whether to
3372 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3373 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3374 until this issue has been settled.
3376 ** Removed deprecated type tag scm_tc16_kw
3378 ** Added type tag scm_tc16_keyword
3380 (This was introduced already in release 1.3.4 but was not documented
3383 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3385 * Changes to system call interfaces:
3387 ** The "select" procedure now tests port buffers for the ability to
3388 provide input or accept output. Previously only the underlying file
3389 descriptors were checked.
3391 ** New variable PIPE_BUF: the maximum number of bytes that can be
3392 atomically written to a pipe.
3394 ** If a facility is not available on the system when Guile is
3395 compiled, the corresponding primitive procedure will not be defined.
3396 Previously it would have been defined but would throw a system-error
3397 exception if called. Exception handlers which catch this case may
3398 need minor modification: an error will be thrown with key
3399 'unbound-variable instead of 'system-error. Alternatively it's
3400 now possible to use `defined?' to check whether the facility is
3403 ** Procedures which depend on the timezone should now give the correct
3404 result on systems which cache the TZ environment variable, even if TZ
3405 is changed without calling tzset.
3407 * Changes to the networking interfaces:
3409 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3410 long integers between network and host format. For now, it's not
3411 particularly convenient to do this kind of thing, but consider:
3413 (define write-network-long
3414 (lambda (value port)
3415 (let ((v (make-uniform-vector 1 1 0)))
3416 (uniform-vector-set! v 0 (htonl value))
3417 (uniform-vector-write v port))))
3419 (define read-network-long
3421 (let ((v (make-uniform-vector 1 1 0)))
3422 (uniform-vector-read! v port)
3423 (ntohl (uniform-vector-ref v 0)))))
3425 ** If inet-aton fails, it now throws an error with key 'misc-error
3426 instead of 'system-error, since errno is not relevant.
3428 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3429 specific keys instead of 'system-error. The latter is inappropriate
3430 since errno will not have been set. The keys are:
3431 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3433 ** sethostent, setnetent, setprotoent, setservent: now take an
3434 optional argument STAYOPEN, which specifies whether the database
3435 remains open after a database entry is accessed randomly (e.g., using
3436 gethostbyname for the hosts database.) The default is #f. Previously
3440 Changes since Guile 1.3.2:
3442 * Changes to the stand-alone interpreter
3446 An initial version of the Guile debugger written by Chris Hanson has
3447 been added. The debugger is still under development but is included
3448 in the distribution anyway since it is already quite useful.
3454 after an error to enter the debugger. Type `help' inside the debugger
3455 for a description of available commands.
3457 If you prefer to have stack frames numbered and printed in
3458 anti-chronological order and prefer up in the stack to be down on the
3459 screen as is the case in gdb, you can put
3461 (debug-enable 'backwards)
3463 in your .guile startup file. (However, this means that Guile can't
3464 use indentation to indicate stack level.)
3466 The debugger is autoloaded into Guile at the first use.
3468 ** Further enhancements to backtraces
3470 There is a new debug option `width' which controls the maximum width
3471 on the screen of printed stack frames. Fancy printing parameters
3472 ("level" and "length" as in Common LISP) are adaptively adjusted for
3473 each stack frame to give maximum information while still fitting
3474 within the bounds. If the stack frame can't be made to fit by
3475 adjusting parameters, it is simply cut off at the end. This is marked
3478 ** Some modules are now only loaded when the repl is started
3480 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3481 regex) are now loaded into (guile-user) only if the repl has been
3482 started. The effect is that the startup time for scripts has been
3483 reduced to 30% of what it was previously.
3485 Correctly written scripts load the modules they require at the top of
3486 the file and should not be affected by this change.
3488 ** Hooks are now represented as smobs
3490 * Changes to Scheme functions and syntax
3492 ** Readline support has changed again.
3494 The old (readline-activator) module is gone. Use (ice-9 readline)
3495 instead, which now contains all readline functionality. So the code
3496 to activate readline is now
3498 (use-modules (ice-9 readline))
3501 This should work at any time, including from the guile prompt.
3503 To avoid confusion about the terms of Guile's license, please only
3504 enable readline for your personal use; please don't make it the
3505 default for others. Here is why we make this rather odd-sounding
3508 Guile is normally licensed under a weakened form of the GNU General
3509 Public License, which allows you to link code with Guile without
3510 placing that code under the GPL. This exception is important to some
3513 However, since readline is distributed under the GNU General Public
3514 License, when you link Guile with readline, either statically or
3515 dynamically, you effectively change Guile's license to the strict GPL.
3516 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3517 which are normally permitted become forbidden. This is a rather
3518 non-obvious consequence of the licensing terms.
3520 So, to make sure things remain clear, please let people choose for
3521 themselves whether to link GPL'd libraries like readline with Guile.
3523 ** regexp-substitute/global has changed slightly, but incompatibly.
3525 If you include a function in the item list, the string of the match
3526 object it receives is the same string passed to
3527 regexp-substitute/global, not some suffix of that string.
3528 Correspondingly, the match's positions are relative to the entire
3529 string, not the suffix.
3531 If the regexp can match the empty string, the way matches are chosen
3532 from the string has changed. regexp-substitute/global recognizes the
3533 same set of matches that list-matches does; see below.
3535 ** New function: list-matches REGEXP STRING [FLAGS]
3537 Return a list of match objects, one for every non-overlapping, maximal
3538 match of REGEXP in STRING. The matches appear in left-to-right order.
3539 list-matches only reports matches of the empty string if there are no
3540 other matches which begin on, end at, or include the empty match's
3543 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3545 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3547 For each match of REGEXP in STRING, apply PROC to the match object,
3548 and the last value PROC returned, or INIT for the first call. Return
3549 the last value returned by PROC. We apply PROC to the matches as they
3550 appear from left to right.
3552 This function recognizes matches according to the same criteria as
3555 Thus, you could define list-matches like this:
3557 (define (list-matches regexp string . flags)
3558 (reverse! (apply fold-matches regexp string '() cons flags)))
3560 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3564 *** New function: hook? OBJ
3566 Return #t if OBJ is a hook, otherwise #f.
3568 *** New function: make-hook-with-name NAME [ARITY]
3570 Return a hook with name NAME and arity ARITY. The default value for
3571 ARITY is 0. The only effect of NAME is that it will appear when the
3572 hook object is printed to ease debugging.
3574 *** New function: hook-empty? HOOK
3576 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3578 *** New function: hook->list HOOK
3580 Return a list of the procedures that are called when run-hook is
3583 ** `map' signals an error if its argument lists are not all the same length.
3585 This is the behavior required by R5RS, so this change is really a bug
3586 fix. But it seems to affect a lot of people's code, so we're
3587 mentioning it here anyway.
3589 ** Print-state handling has been made more transparent
3591 Under certain circumstances, ports are represented as a port with an
3592 associated print state. Earlier, this pair was represented as a pair
3593 (see "Some magic has been added to the printer" below). It is now
3594 indistinguishable (almost; see `get-print-state') from a port on the
3597 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3599 Return a new port with the associated print state PRINT-STATE.
3601 *** New function: get-print-state OUTPUT-PORT
3603 Return the print state associated with this port if it exists,
3604 otherwise return #f.
3606 *** New function: directory-stream? OBJECT
3608 Returns true iff OBJECT is a directory stream --- the sort of object
3609 returned by `opendir'.
3611 ** New function: using-readline?
3613 Return #t if readline is in use in the current repl.
3615 ** structs will be removed in 1.4
3617 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3618 and use GOOPS objects as the fundamental record type.
3620 * Changes to the scm_ interface
3622 ** structs will be removed in 1.4
3624 The entire current struct interface (struct.c, struct.h) will be
3625 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3626 GOOPS objects as the fundamental record type.
3628 ** The internal representation of subr's has changed
3630 Instead of giving a hint to the subr name, the CAR field of the subr
3631 now contains an index to a subr entry in scm_subr_table.
3633 *** New variable: scm_subr_table
3635 An array of subr entries. A subr entry contains the name, properties
3636 and documentation associated with the subr. The properties and
3637 documentation slots are not yet used.
3639 ** A new scheme for "forwarding" calls to a builtin to a generic function
3641 It is now possible to extend the functionality of some Guile
3642 primitives by letting them defer a call to a GOOPS generic function on
3643 argument mismatch. This means that there is no loss of efficiency in
3648 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3649 (define-method + ((x <string>) (y <string>))
3650 (string-append x y))
3652 + will still be as efficient as usual in numerical calculations, but
3653 can also be used for concatenating strings.
3655 Who will be the first one to extend Guile's numerical tower to
3656 rationals? :) [OK, there a few other things to fix before this can
3657 be made in a clean way.]
3659 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3661 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3663 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3665 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3666 a variable GENERIC which can be used by the dispatch macros below.
3668 [This is experimental code which may change soon.]
3670 *** New macros for forwarding control to a generic on arg type error
3672 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3674 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3676 These correspond to the scm_wta function call, and have the same
3677 behaviour until the user has called the GOOPS primitive
3678 `enable-primitive-generic!'. After that, these macros will apply the
3679 generic function GENERIC to the argument(s) instead of calling
3682 [This is experimental code which may change soon.]
3684 *** New macros for argument testing with generic dispatch
3686 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3688 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3690 These correspond to the SCM_ASSERT macro, but will defer control to
3691 GENERIC on error after `enable-primitive-generic!' has been called.
3693 [This is experimental code which may change soon.]
3695 ** New function: SCM scm_eval_body (SCM body, SCM env)
3697 Evaluates the body of a special form.
3699 ** The internal representation of struct's has changed
3701 Previously, four slots were allocated for the procedure(s) of entities
3702 and operators. The motivation for this representation had to do with
3703 the structure of the evaluator, the wish to support tail-recursive
3704 generic functions, and efficiency. Since the generic function
3705 dispatch mechanism has changed, there is no longer a need for such an
3706 expensive representation, and the representation has been simplified.
3708 This should not make any difference for most users.
3710 ** GOOPS support has been cleaned up.
3712 Some code has been moved from eval.c to objects.c and code in both of
3713 these compilation units has been cleaned up and better structured.
3715 *** New functions for applying generic functions
3717 New function: SCM scm_apply_generic (GENERIC, ARGS)
3718 New function: SCM scm_call_generic_0 (GENERIC)
3719 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3720 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3721 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3723 ** Deprecated function: scm_make_named_hook
3725 It is now replaced by:
3727 ** New function: SCM scm_create_hook (const char *name, int arity)
3729 Creates a hook in the same way as make-hook above but also
3730 binds a variable named NAME to it.
3732 This is the typical way of creating a hook from C code.
3734 Currently, the variable is created in the "current" module.
3735 This might change when we get the new module system.
3737 [The behaviour is identical to scm_make_named_hook.]
3741 Changes since Guile 1.3:
3743 * Changes to mailing lists
3745 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3747 See the README file to find current addresses for all the Guile
3750 * Changes to the distribution
3752 ** Readline support is no longer included with Guile by default.
3754 Based on the different license terms of Guile and Readline, we
3755 concluded that Guile should not *by default* cause the linking of
3756 Readline into an application program. Readline support is now offered
3757 as a separate module, which is linked into an application only when
3758 you explicitly specify it.
3760 Although Guile is GNU software, its distribution terms add a special
3761 exception to the usual GNU General Public License (GPL). Guile's
3762 license includes a clause that allows you to link Guile with non-free
3763 programs. We add this exception so as not to put Guile at a
3764 disadvantage vis-a-vis other extensibility packages that support other
3767 In contrast, the GNU Readline library is distributed under the GNU
3768 General Public License pure and simple. This means that you may not
3769 link Readline, even dynamically, into an application unless it is
3770 distributed under a free software license that is compatible the GPL.
3772 Because of this difference in distribution terms, an application that
3773 can use Guile may not be able to use Readline. Now users will be
3774 explicitly offered two independent decisions about the use of these
3777 You can activate the readline support by issuing
3779 (use-modules (readline-activator))
3782 from your ".guile" file, for example.
3784 * Changes to the stand-alone interpreter
3786 ** All builtins now print as primitives.
3787 Previously builtin procedures not belonging to the fundamental subr
3788 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3789 Now, they print as #<primitive-procedure NAME>.
3791 ** Backtraces slightly more intelligible.
3792 gsubr-apply and macro transformer application frames no longer appear
3795 * Changes to Scheme functions and syntax
3797 ** Guile now correctly handles internal defines by rewriting them into
3798 their equivalent letrec. Previously, internal defines would
3799 incrementally add to the innermost environment, without checking
3800 whether the restrictions specified in RnRS were met. This lead to the
3801 correct behaviour when these restriction actually were met, but didn't
3802 catch all illegal uses. Such an illegal use could lead to crashes of
3803 the Guile interpreter or or other unwanted results. An example of
3804 incorrect internal defines that made Guile behave erratically:
3816 The problem with this example is that the definition of `c' uses the
3817 value of `b' directly. This confuses the meoization machine of Guile
3818 so that the second call of `b' (this time in a larger environment that
3819 also contains bindings for `c' and `d') refers to the binding of `c'
3820 instead of `a'. You could also make Guile crash with a variation on
3825 (define (b flag) (if flag a 1))
3826 (define c (1+ (b flag)))
3834 From now on, Guile will issue an `Unbound variable: b' error message
3839 A hook contains a list of functions which should be called on
3840 particular occasions in an existing program. Hooks are used for
3843 A window manager might have a hook before-window-map-hook. The window
3844 manager uses the function run-hooks to call all functions stored in
3845 before-window-map-hook each time a window is mapped. The user can
3846 store functions in the hook using add-hook!.
3848 In Guile, hooks are first class objects.
3850 *** New function: make-hook [N_ARGS]
3852 Return a hook for hook functions which can take N_ARGS arguments.
3853 The default value for N_ARGS is 0.
3855 (See also scm_make_named_hook below.)
3857 *** New function: add-hook! HOOK PROC [APPEND_P]
3859 Put PROC at the beginning of the list of functions stored in HOOK.
3860 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3862 PROC must be able to take the number of arguments specified when the
3865 If PROC already exists in HOOK, then remove it first.
3867 *** New function: remove-hook! HOOK PROC
3869 Remove PROC from the list of functions in HOOK.
3871 *** New function: reset-hook! HOOK
3873 Clear the list of hook functions stored in HOOK.
3875 *** New function: run-hook HOOK ARG1 ...
3877 Run all hook functions stored in HOOK with arguments ARG1 ... .
3878 The number of arguments supplied must correspond to the number given
3879 when the hook was created.
3881 ** The function `dynamic-link' now takes optional keyword arguments.
3882 The only keyword argument that is currently defined is `:global
3883 BOOL'. With it, you can control whether the shared library will be
3884 linked in global mode or not. In global mode, the symbols from the
3885 linked library can be used to resolve references from other
3886 dynamically linked libraries. In non-global mode, the linked
3887 library is essentially invisible and can only be accessed via
3888 `dynamic-func', etc. The default is now to link in global mode.
3889 Previously, the default has been non-global mode.
3891 The `#:global' keyword is only effective on platforms that support
3892 the dlopen family of functions.
3894 ** New function `provided?'
3896 - Function: provided? FEATURE
3897 Return true iff FEATURE is supported by this installation of
3898 Guile. FEATURE must be a symbol naming a feature; the global
3899 variable `*features*' is a list of available features.
3901 ** Changes to the module (ice-9 expect):
3903 *** The expect-strings macro now matches `$' in a regular expression
3904 only at a line-break or end-of-file by default. Previously it would
3905 match the end of the string accumulated so far. The old behaviour
3906 can be obtained by setting the variable `expect-strings-exec-flags'
3909 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3910 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3911 in a regular expression will still match before a line-break or
3912 end-of-file. The default is `regexp/noteol'.
3914 *** The expect-strings macro now uses a variable
3915 `expect-strings-compile-flags' for the flags to be supplied to
3916 `make-regexp'. The default is `regexp/newline', which was previously
3919 *** The expect macro now supplies two arguments to a match procedure:
3920 the current accumulated string and a flag to indicate whether
3921 end-of-file has been reached. Previously only the string was supplied.
3922 If end-of-file is reached, the match procedure will be called an
3923 additional time with the same accumulated string as the previous call
3924 but with the flag set.
3926 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3928 This code, and the documentation for it that appears here, was
3929 borrowed from SLIB, with minor adaptations for Guile.
3931 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3932 An almost complete implementation of Common LISP format description
3933 according to the CL reference book `Common LISP' from Guy L.
3934 Steele, Digital Press. Backward compatible to most of the
3935 available Scheme format implementations.
3937 Returns `#t', `#f' or a string; has side effect of printing
3938 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3939 to the current output port and `#t' is returned. If DESTINATION
3940 is `#f', a formatted string is returned as the result of the call.
3941 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3942 format string; FORMAT-STRING is then the first argument and the
3943 output is returned as a string. If DESTINATION is a number, the
3944 output is to the current error port if available by the
3945 implementation. Otherwise DESTINATION must be an output port and
3948 FORMAT-STRING must be a string. In case of a formatting error
3949 format returns `#f' and prints a message on the current output or
3950 error port. Characters are output as if the string were output by
3951 the `display' function with the exception of those prefixed by a
3952 tilde (~). For a detailed description of the FORMAT-STRING syntax
3953 please consult a Common LISP format reference manual. For a test
3954 suite to verify this format implementation load `formatst.scm'.
3955 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3957 Note: `format' is not reentrant, i.e. only one `format'-call may
3958 be executed at a time.
3961 *** Format Specification (Format version 3.0)
3963 Please consult a Common LISP format reference manual for a detailed
3964 description of the format string syntax. For a demonstration of the
3965 implemented directives see `formatst.scm'.
3967 This implementation supports directive parameters and modifiers (`:'
3968 and `@' characters). Multiple parameters must be separated by a comma
3969 (`,'). Parameters can be numerical parameters (positive or negative),
3970 character parameters (prefixed by a quote character (`''), variable
3971 parameters (`v'), number of rest arguments parameter (`#'), empty and
3972 default parameters. Directive characters are case independent. The
3973 general form of a directive is:
3975 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3977 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3979 *** Implemented CL Format Control Directives
3981 Documentation syntax: Uppercase characters represent the
3982 corresponding control directive characters. Lowercase characters
3983 represent control directive parameter descriptions.
3986 Any (print as `display' does).
3990 `~MINCOL,COLINC,MINPAD,PADCHARA'
3994 S-expression (print as `write' does).
3998 `~MINCOL,COLINC,MINPAD,PADCHARS'
4004 print number sign always.
4007 print comma separated.
4009 `~MINCOL,PADCHAR,COMMACHARD'
4015 print number sign always.
4018 print comma separated.
4020 `~MINCOL,PADCHAR,COMMACHARX'
4026 print number sign always.
4029 print comma separated.
4031 `~MINCOL,PADCHAR,COMMACHARO'
4037 print number sign always.
4040 print comma separated.
4042 `~MINCOL,PADCHAR,COMMACHARB'
4047 `~N,MINCOL,PADCHAR,COMMACHARR'
4051 print a number as a Roman numeral.
4054 print a number as an "old fashioned" Roman numeral.
4057 print a number as an ordinal English number.
4060 print a number as a cardinal English number.
4065 prints `y' and `ies'.
4068 as `~P but jumps 1 argument backward.'
4071 as `~@P but jumps 1 argument backward.'
4076 prints a character as the reader can understand it (i.e. `#\'
4080 prints a character as emacs does (eg. `^C' for ASCII 03).
4083 Fixed-format floating-point (prints a flonum like MMM.NNN).
4084 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4086 If the number is positive a plus sign is printed.
4089 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4090 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4092 If the number is positive a plus sign is printed.
4095 General floating-point (prints a flonum either fixed or
4097 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4099 If the number is positive a plus sign is printed.
4102 Dollars floating-point (prints a flonum in fixed with signs
4104 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4106 If the number is positive a plus sign is printed.
4109 A sign is always printed and appears before the padding.
4112 The sign appears before the padding.
4120 print newline if not at the beginning of the output line.
4122 prints `~&' and then N-1 newlines.
4127 print N page separators.
4137 newline is ignored, white space left.
4140 newline is left, white space ignored.
4145 relative tabulation.
4151 Indirection (expects indirect arguments as a list).
4153 extracts indirect arguments from format arguments.
4156 Case conversion (converts by `string-downcase').
4158 converts by `string-capitalize'.
4161 converts by `string-capitalize-first'.
4164 converts by `string-upcase'.
4167 Argument Jumping (jumps 1 argument forward).
4169 jumps N arguments forward.
4172 jumps 1 argument backward.
4175 jumps N arguments backward.
4178 jumps to the 0th argument.
4181 jumps to the Nth argument (beginning from 0)
4183 `~[STR0~;STR1~;...~;STRN~]'
4184 Conditional Expression (numerical clause conditional).
4186 take argument from N.
4189 true test conditional.
4192 if-else-then conditional.
4198 default clause follows.
4201 Iteration (args come from the next argument (a list)).
4203 at most N iterations.
4206 args from next arg (a list of lists).
4209 args from the rest of arguments.
4212 args from the rest args (lists).
4223 aborts if N <= M <= K
4225 *** Not Implemented CL Format Control Directives
4228 print `#f' as an empty list (see below).
4231 print `#f' as an empty list (see below).
4237 (sorry I don't understand its semantics completely)
4239 *** Extended, Replaced and Additional Control Directives
4241 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4242 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4243 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4244 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4245 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4246 COMMAWIDTH is the number of characters between two comma
4250 print a R4RS complex number as `~F~@Fi' with passed parameters for
4254 Pretty print formatting of an argument for scheme code lists.
4260 Flushes the output if format DESTINATION is a port.
4263 Print a `#\space' character
4265 print N `#\space' characters.
4268 Print a `#\tab' character
4270 print N `#\tab' characters.
4273 Takes N as an integer representation for a character. No arguments
4274 are consumed. N is converted to a character by `integer->char'. N
4275 must be a positive decimal number.
4278 Print out readproof. Prints out internal objects represented as
4279 `#<...>' as strings `"#<...>"' so that the format output can always
4280 be processed by `read'.
4283 Print out readproof. Prints out internal objects represented as
4284 `#<...>' as strings `"#<...>"' so that the format output can always
4285 be processed by `read'.
4288 Prints information and a copyright notice on the format
4291 prints format version.
4294 may also print number strings, i.e. passing a number as a string
4295 and format it accordingly.
4297 *** Configuration Variables
4299 The format module exports some configuration variables to suit the
4300 systems and users needs. There should be no modification necessary for
4301 the configuration that comes with Guile. Format detects automatically
4302 if the running scheme system implements floating point numbers and
4305 format:symbol-case-conv
4306 Symbols are converted by `symbol->string' so the case type of the
4307 printed symbols is implementation dependent.
4308 `format:symbol-case-conv' is a one arg closure which is either
4309 `#f' (no conversion), `string-upcase', `string-downcase' or
4310 `string-capitalize'. (default `#f')
4312 format:iobj-case-conv
4313 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4314 implementation internal objects. (default `#f')
4317 The character prefixing the exponent value in `~E' printing.
4320 *** Compatibility With Other Format Implementations
4326 Downward compatible except for padding support and `~A', `~S',
4327 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4328 `printf' padding support which is completely replaced by the CL
4329 `format' padding style.
4332 Downward compatible except for `~', which is not documented
4333 (ignores all characters inside the format string up to a newline
4334 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4335 numerical and variable parameters and `:/@' modifiers in the CL
4339 Downward compatible except for `~A' and `~S' which print in
4340 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4341 directive parameters or modifiers)).
4344 Downward compatible except for an optional destination parameter:
4345 S2C accepts a format call without a destination which returns a
4346 formatted string. This is equivalent to a #f destination in S2C.
4347 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4348 parameters or modifiers)).
4351 ** Changes to string-handling functions.
4353 These functions were added to support the (ice-9 format) module, above.
4355 *** New function: string-upcase STRING
4356 *** New function: string-downcase STRING
4358 These are non-destructive versions of the existing string-upcase! and
4359 string-downcase! functions.
4361 *** New function: string-capitalize! STRING
4362 *** New function: string-capitalize STRING
4364 These functions convert the first letter of each word in the string to
4367 (string-capitalize "howdy there")
4370 As with the other functions, string-capitalize! modifies the string in
4371 place, while string-capitalize returns a modified copy of its argument.
4373 *** New function: string-ci->symbol STRING
4375 Return a symbol whose name is STRING, but having the same case as if
4376 the symbol had be read by `read'.
4378 Guile can be configured to be sensitive or insensitive to case
4379 differences in Scheme identifiers. If Guile is case-insensitive, all
4380 symbols are converted to lower case on input. The `string-ci->symbol'
4381 function returns a symbol whose name in STRING, transformed as Guile
4382 would if STRING were input.
4384 *** New function: substring-move! STRING1 START END STRING2 START
4386 Copy the substring of STRING1 from START (inclusive) to END
4387 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4388 string, and the source and destination areas may overlap; in all
4389 cases, the function behaves as if all the characters were copied
4392 *** Extended functions: substring-move-left! substring-move-right!
4394 These functions now correctly copy arbitrarily overlapping substrings;
4395 they are both synonyms for substring-move!.
4398 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4400 getopt-long is a function for parsing command-line arguments in a
4401 manner consistent with other GNU programs.
4403 (getopt-long ARGS GRAMMAR)
4404 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4406 ARGS should be a list of strings. Its first element should be the
4407 name of the program; subsequent elements should be the arguments
4408 that were passed to the program on the command line. The
4409 `program-arguments' procedure returns a list of this form.
4411 GRAMMAR is a list of the form:
4412 ((OPTION (PROPERTY VALUE) ...) ...)
4414 Each OPTION should be a symbol. `getopt-long' will accept a
4415 command-line option named `--OPTION'.
4416 Each option can have the following (PROPERTY VALUE) pairs:
4418 (single-char CHAR) --- Accept `-CHAR' as a single-character
4419 equivalent to `--OPTION'. This is how to specify traditional
4421 (required? BOOL) --- If BOOL is true, the option is required.
4422 getopt-long will raise an error if it is not found in ARGS.
4423 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4424 it is #f, it does not; and if it is the symbol
4425 `optional', the option may appear in ARGS with or
4427 (predicate FUNC) --- If the option accepts a value (i.e. you
4428 specified `(value #t)' for this option), then getopt
4429 will apply FUNC to the value, and throw an exception
4430 if it returns #f. FUNC should be a procedure which
4431 accepts a string and returns a boolean value; you may
4432 need to use quasiquotes to get it into GRAMMAR.
4434 The (PROPERTY VALUE) pairs may occur in any order, but each
4435 property may occur only once. By default, options do not have
4436 single-character equivalents, are not required, and do not take
4439 In ARGS, single-character options may be combined, in the usual
4440 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4441 accepts values, then it must be the last option in the
4442 combination; the value is the next argument. So, for example, using
4443 the following grammar:
4444 ((apples (single-char #\a))
4445 (blimps (single-char #\b) (value #t))
4446 (catalexis (single-char #\c) (value #t)))
4447 the following argument lists would be acceptable:
4448 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4449 for "blimps" and "catalexis")
4450 ("-ab" "bang" "-c" "couth") (same)
4451 ("-ac" "couth" "-b" "bang") (same)
4452 ("-abc" "couth" "bang") (an error, since `-b' is not the
4453 last option in its combination)
4455 If an option's value is optional, then `getopt-long' decides
4456 whether it has a value by looking at what follows it in ARGS. If
4457 the next element is a string, and it does not appear to be an
4458 option itself, then that string is the option's value.
4460 The value of a long option can appear as the next element in ARGS,
4461 or it can follow the option name, separated by an `=' character.
4462 Thus, using the same grammar as above, the following argument lists
4464 ("--apples" "Braeburn" "--blimps" "Goodyear")
4465 ("--apples=Braeburn" "--blimps" "Goodyear")
4466 ("--blimps" "Goodyear" "--apples=Braeburn")
4468 If the option "--" appears in ARGS, argument parsing stops there;
4469 subsequent arguments are returned as ordinary arguments, even if
4470 they resemble options. So, in the argument list:
4471 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4472 `getopt-long' will recognize the `apples' option as having the
4473 value "Granny Smith", but it will not recognize the `blimp'
4474 option; it will return the strings "--blimp" and "Goodyear" as
4475 ordinary argument strings.
4477 The `getopt-long' function returns the parsed argument list as an
4478 assocation list, mapping option names --- the symbols from GRAMMAR
4479 --- onto their values, or #t if the option does not accept a value.
4480 Unused options do not appear in the alist.
4482 All arguments that are not the value of any option are returned
4483 as a list, associated with the empty list.
4485 `getopt-long' throws an exception if:
4486 - it finds an unrecognized option in ARGS
4487 - a required option is omitted
4488 - an option that requires an argument doesn't get one
4489 - an option that doesn't accept an argument does get one (this can
4490 only happen using the long option `--opt=value' syntax)
4491 - an option predicate fails
4496 `((lockfile-dir (required? #t)
4499 (predicate ,file-is-directory?))
4500 (verbose (required? #f)
4503 (x-includes (single-char #\x))
4504 (rnet-server (single-char #\y)
4505 (predicate ,string?))))
4507 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4508 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4510 => ((() "foo1" "-fred" "foo2" "foo3")
4511 (rnet-server . "lamprod")
4512 (x-includes . "/usr/include")
4513 (lockfile-dir . "/tmp")
4516 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4518 It will be removed in a few releases.
4520 ** New syntax: lambda*
4521 ** New syntax: define*
4522 ** New syntax: define*-public
4523 ** New syntax: defmacro*
4524 ** New syntax: defmacro*-public
4525 Guile now supports optional arguments.
4527 `lambda*', `define*', `define*-public', `defmacro*' and
4528 `defmacro*-public' are identical to the non-* versions except that
4529 they use an extended type of parameter list that has the following BNF
4530 syntax (parentheses are literal, square brackets indicate grouping,
4531 and `*', `+' and `?' have the usual meaning):
4533 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4534 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4535 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4537 ext-var-decl ::= identifier | ( identifier expression )
4539 The semantics are best illustrated with the following documentation
4540 and examples for `lambda*':
4543 lambda extended for optional and keyword arguments
4545 lambda* creates a procedure that takes optional arguments. These
4546 are specified by putting them inside brackets at the end of the
4547 paramater list, but before any dotted rest argument. For example,
4548 (lambda* (a b #&optional c d . e) '())
4549 creates a procedure with fixed arguments a and b, optional arguments c
4550 and d, and rest argument e. If the optional arguments are omitted
4551 in a call, the variables for them are unbound in the procedure. This
4552 can be checked with the bound? macro.
4554 lambda* can also take keyword arguments. For example, a procedure
4556 (lambda* (#&key xyzzy larch) '())
4557 can be called with any of the argument lists (#:xyzzy 11)
4558 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4559 are given as keywords are bound to values.
4561 Optional and keyword arguments can also be given default values
4562 which they take on when they are not present in a call, by giving a
4563 two-item list in place of an optional argument, for example in:
4564 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4565 foo is a fixed argument, bar is an optional argument with default
4566 value 42, and baz is a keyword argument with default value 73.
4567 Default value expressions are not evaluated unless they are needed
4568 and until the procedure is called.
4570 lambda* now supports two more special parameter list keywords.
4572 lambda*-defined procedures now throw an error by default if a
4573 keyword other than one of those specified is found in the actual
4574 passed arguments. However, specifying #&allow-other-keys
4575 immediately after the kyword argument declarations restores the
4576 previous behavior of ignoring unknown keywords. lambda* also now
4577 guarantees that if the same keyword is passed more than once, the
4578 last one passed is the one that takes effect. For example,
4579 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4580 #:heads 37 #:tails 42 #:heads 99)
4581 would result in (99 47) being displayed.
4583 #&rest is also now provided as a synonym for the dotted syntax rest
4584 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4585 all respects to lambda*. This is provided for more similarity to DSSSL,
4586 MIT-Scheme and Kawa among others, as well as for refugees from other
4589 Further documentation may be found in the optargs.scm file itself.
4591 The optional argument module also exports the macros `let-optional',
4592 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4593 are not documented here because they may be removed in the future, but
4594 full documentation is still available in optargs.scm.
4596 ** New syntax: and-let*
4597 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4599 Syntax: (land* (<clause> ...) <body> ...)
4600 Each <clause> should have one of the following forms:
4601 (<variable> <expression>)
4604 Each <variable> or <bound-variable> should be an identifier. Each
4605 <expression> should be a valid expression. The <body> should be a
4606 possibly empty sequence of expressions, like the <body> of a
4609 Semantics: A LAND* expression is evaluated by evaluating the
4610 <expression> or <bound-variable> of each of the <clause>s from
4611 left to right. The value of the first <expression> or
4612 <bound-variable> that evaluates to a false value is returned; the
4613 remaining <expression>s and <bound-variable>s are not evaluated.
4614 The <body> forms are evaluated iff all the <expression>s and
4615 <bound-variable>s evaluate to true values.
4617 The <expression>s and the <body> are evaluated in an environment
4618 binding each <variable> of the preceding (<variable> <expression>)
4619 clauses to the value of the <expression>. Later bindings
4620 shadow earlier bindings.
4622 Guile's and-let* macro was contributed by Michael Livshin.
4624 ** New sorting functions
4626 *** New function: sorted? SEQUENCE LESS?
4627 Returns `#t' when the sequence argument is in non-decreasing order
4628 according to LESS? (that is, there is no adjacent pair `... x y
4629 ...' for which `(less? y x)').
4631 Returns `#f' when the sequence contains at least one out-of-order
4632 pair. It is an error if the sequence is neither a list nor a
4635 *** New function: merge LIST1 LIST2 LESS?
4636 LIST1 and LIST2 are sorted lists.
4637 Returns the sorted list of all elements in LIST1 and LIST2.
4639 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4640 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4641 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4642 (Here "<" should read "comes before".)
4644 *** New procedure: merge! LIST1 LIST2 LESS?
4645 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4646 the result. If the code is compiled, and LESS? constructs no new
4647 pairs, no pairs at all will be allocated. The first pair of the
4648 result will be either the first pair of LIST1 or the first pair of
4651 *** New function: sort SEQUENCE LESS?
4652 Accepts either a list or a vector, and returns a new sequence
4653 which is sorted. The new sequence is the same type as the input.
4654 Always `(sorted? (sort sequence less?) less?)'. The original
4655 sequence is not altered in any way. The new sequence shares its
4656 elements with the old one; no elements are copied.
4658 *** New procedure: sort! SEQUENCE LESS
4659 Returns its sorted result in the original boxes. No new storage is
4660 allocated at all. Proper usage: (set! slist (sort! slist <))
4662 *** New function: stable-sort SEQUENCE LESS?
4663 Similar to `sort' but stable. That is, if "equal" elements are
4664 ordered a < b in the original sequence, they will have the same order
4667 *** New function: stable-sort! SEQUENCE LESS?
4668 Similar to `sort!' but stable.
4669 Uses temporary storage when sorting vectors.
4671 *** New functions: sort-list, sort-list!
4672 Added for compatibility with scsh.
4674 ** New built-in random number support
4676 *** New function: random N [STATE]
4677 Accepts a positive integer or real N and returns a number of the
4678 same type between zero (inclusive) and N (exclusive). The values
4679 returned have a uniform distribution.
4681 The optional argument STATE must be of the type produced by
4682 `copy-random-state' or `seed->random-state'. It defaults to the value
4683 of the variable `*random-state*'. This object is used to maintain the
4684 state of the pseudo-random-number generator and is altered as a side
4685 effect of the `random' operation.
4687 *** New variable: *random-state*
4688 Holds a data structure that encodes the internal state of the
4689 random-number generator that `random' uses by default. The nature
4690 of this data structure is implementation-dependent. It may be
4691 printed out and successfully read back in, but may or may not
4692 function correctly as a random-number state object in another
4695 *** New function: copy-random-state [STATE]
4696 Returns a new object of type suitable for use as the value of the
4697 variable `*random-state*' and as a second argument to `random'.
4698 If argument STATE is given, a copy of it is returned. Otherwise a
4699 copy of `*random-state*' is returned.
4701 *** New function: seed->random-state SEED
4702 Returns a new object of type suitable for use as the value of the
4703 variable `*random-state*' and as a second argument to `random'.
4704 SEED is a string or a number. A new state is generated and
4705 initialized using SEED.
4707 *** New function: random:uniform [STATE]
4708 Returns an uniformly distributed inexact real random number in the
4709 range between 0 and 1.
4711 *** New procedure: random:solid-sphere! VECT [STATE]
4712 Fills VECT with inexact real random numbers the sum of whose
4713 squares is less than 1.0. Thinking of VECT as coordinates in
4714 space of dimension N = `(vector-length VECT)', the coordinates are
4715 uniformly distributed within the unit N-shere. The sum of the
4716 squares of the numbers is returned. VECT can be either a vector
4717 or a uniform vector of doubles.
4719 *** New procedure: random:hollow-sphere! VECT [STATE]
4720 Fills VECT with inexact real random numbers the sum of whose squares
4721 is equal to 1.0. Thinking of VECT as coordinates in space of
4722 dimension n = `(vector-length VECT)', the coordinates are uniformly
4723 distributed over the surface of the unit n-shere. VECT can be either
4724 a vector or a uniform vector of doubles.
4726 *** New function: random:normal [STATE]
4727 Returns an inexact real in a normal distribution with mean 0 and
4728 standard deviation 1. For a normal distribution with mean M and
4729 standard deviation D use `(+ M (* D (random:normal)))'.
4731 *** New procedure: random:normal-vector! VECT [STATE]
4732 Fills VECT with inexact real random numbers which are independent and
4733 standard normally distributed (i.e., with mean 0 and variance 1).
4734 VECT can be either a vector or a uniform vector of doubles.
4736 *** New function: random:exp STATE
4737 Returns an inexact real in an exponential distribution with mean 1.
4738 For an exponential distribution with mean U use (* U (random:exp)).
4740 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4742 These functions now operate on numbers in the range of a C unsigned
4745 These functions used to operate on numbers in the range of a C signed
4746 long; however, this seems inappropriate, because Guile integers don't
4749 ** New function: make-guardian
4750 This is an implementation of guardians as described in
4751 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4752 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4753 Programming Language Design and Implementation, June 1993
4754 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4756 ** New functions: delq1!, delv1!, delete1!
4757 These procedures behave similar to delq! and friends but delete only
4758 one object if at all.
4760 ** New function: unread-string STRING PORT
4761 Unread STRING to PORT, that is, push it back onto the port so that
4762 next read operation will work on the pushed back characters.
4764 ** unread-char can now be called multiple times
4765 If unread-char is called multiple times, the unread characters will be
4766 read again in last-in first-out order.
4768 ** the procedures uniform-array-read! and uniform-array-write! now
4769 work on any kind of port, not just ports which are open on a file.
4771 ** Now 'l' in a port mode requests line buffering.
4773 ** The procedure truncate-file now works on string ports as well
4774 as file ports. If the size argument is omitted, the current
4775 file position is used.
4777 ** new procedure: seek PORT/FDES OFFSET WHENCE
4778 The arguments are the same as for the old fseek procedure, but it
4779 works on string ports as well as random-access file ports.
4781 ** the fseek procedure now works on string ports, since it has been
4782 redefined using seek.
4784 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4785 size is not supplied.
4787 ** the newline procedure no longer flushes the port if it's not
4788 line-buffered: previously it did if it was the current output port.
4790 ** open-pipe and close-pipe are no longer primitive procedures, but
4791 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4793 ** the freopen procedure has been removed.
4795 ** new procedure: drain-input PORT
4796 Drains PORT's read buffers (including any pushed-back characters)
4797 and returns the contents as a single string.
4799 ** New function: map-in-order PROC LIST1 LIST2 ...
4800 Version of `map' which guarantees that the procedure is applied to the
4801 lists in serial order.
4803 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4804 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4805 now obsolete and will go away in release 1.5.
4807 ** New syntax: collect BODY1 ...
4808 Version of `begin' which returns a list of the results of the body
4809 forms instead of the result of the last body form. In contrast to
4810 `begin', `collect' allows an empty body.
4812 ** New functions: read-history FILENAME, write-history FILENAME
4813 Read/write command line history from/to file. Returns #t on success
4814 and #f if an error occured.
4816 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4818 These procedures return a list of definitions available in the specified
4819 argument, a relative module reference. In the case of no argument,
4820 `(current-module)' is now consulted for definitions to return, instead
4821 of simply returning #f, the former behavior.
4823 ** The #/ syntax for lists is no longer supported.
4825 Earlier versions of Scheme accepted this syntax, but printed a
4828 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4830 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4833 * Changes to the gh_ interface
4837 Now takes a second argument which is the result array. If this
4838 pointer is NULL, a new array is malloced (the old behaviour).
4840 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4841 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4845 * Changes to the scm_ interface
4847 ** Function: scm_make_named_hook (char* name, int n_args)
4849 Creates a hook in the same way as make-hook above but also
4850 binds a variable named NAME to it.
4852 This is the typical way of creating a hook from C code.
4854 Currently, the variable is created in the "current" module. This
4855 might change when we get the new module system.
4857 ** The smob interface
4859 The interface for creating smobs has changed. For documentation, see
4860 data-rep.info (made from guile-core/doc/data-rep.texi).
4862 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4864 >>> This function will be removed in 1.3.4. <<<
4868 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4869 This function adds a new smob type, named NAME, with instance size
4870 SIZE to the system. The return value is a tag that is used in
4871 creating instances of the type. If SIZE is 0, then no memory will
4872 be allocated when instances of the smob are created, and nothing
4873 will be freed by the default free function.
4875 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4876 This function sets the smob marking procedure for the smob type
4877 specified by the tag TC. TC is the tag returned by
4878 `scm_make_smob_type'.
4880 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4881 This function sets the smob freeing procedure for the smob type
4882 specified by the tag TC. TC is the tag returned by
4883 `scm_make_smob_type'.
4885 *** Function: void scm_set_smob_print (tc, print)
4887 - Function: void scm_set_smob_print (long tc,
4888 scm_sizet (*print) (SCM,
4892 This function sets the smob printing procedure for the smob type
4893 specified by the tag TC. TC is the tag returned by
4894 `scm_make_smob_type'.
4896 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4897 This function sets the smob equality-testing predicate for the
4898 smob type specified by the tag TC. TC is the tag returned by
4899 `scm_make_smob_type'.
4901 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4902 Make VALUE contain a smob instance of the type with type code TC and
4903 smob data DATA. VALUE must be previously declared as C type `SCM'.
4905 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4906 This macro expands to a block of code that creates a smob instance
4907 of the type with type code TC and smob data DATA, and returns that
4908 `SCM' value. It should be the last piece of code in a block.
4910 ** The interfaces for using I/O ports and implementing port types
4911 (ptobs) have changed significantly. The new interface is based on
4912 shared access to buffers and a new set of ptob procedures.
4914 *** scm_newptob has been removed
4918 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4920 - Function: SCM scm_make_port_type (char *type_name,
4921 int (*fill_buffer) (SCM port),
4922 void (*write_flush) (SCM port));
4924 Similarly to the new smob interface, there is a set of function
4925 setters by which the user can customize the behaviour of his port
4926 type. See ports.h (scm_set_port_XXX).
4928 ** scm_strport_to_string: New function: creates a new string from
4929 a string port's buffer.
4931 ** Plug in interface for random number generators
4932 The variable `scm_the_rng' in random.c contains a value and three
4933 function pointers which together define the current random number
4934 generator being used by the Scheme level interface and the random
4935 number library functions.
4937 The user is free to replace the default generator with the generator
4940 *** Variable: size_t scm_the_rng.rstate_size
4941 The size of the random state type used by the current RNG
4944 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4945 Given the random STATE, return 32 random bits.
4947 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4948 Seed random state STATE using string S of length N.
4950 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4951 Given random state STATE, return a malloced copy.
4954 The default RNG is the MWC (Multiply With Carry) random number
4955 generator described by George Marsaglia at the Department of
4956 Statistics and Supercomputer Computations Research Institute, The
4957 Florida State University (http://stat.fsu.edu/~geo).
4959 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4960 passes all tests in the DIEHARD test suite
4961 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4962 costs one multiply and one add on platforms which either supports long
4963 longs (gcc does this on most systems) or have 64 bit longs. The cost
4964 is four multiply on other systems but this can be optimized by writing
4965 scm_i_uniform32 in assembler.
4967 These functions are provided through the scm_the_rng interface for use
4968 by libguile and the application.
4970 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4971 Given the random STATE, return 32 random bits.
4972 Don't use this function directly. Instead go through the plugin
4973 interface (see "Plug in interface" above).
4975 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4976 Initialize STATE using SEED of length N.
4978 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4979 Return a malloc:ed copy of STATE. This function can easily be re-used
4980 in the interfaces to other RNGs.
4982 ** Random number library functions
4983 These functions use the current RNG through the scm_the_rng interface.
4984 It might be a good idea to use these functions from your C code so
4985 that only one random generator is used by all code in your program.
4987 The default random state is stored in:
4989 *** Variable: SCM scm_var_random_state
4990 Contains the vcell of the Scheme variable "*random-state*" which is
4991 used as default state by all random number functions in the Scheme
4996 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4998 *** Function: scm_rstate *scm_c_default_rstate (void)
4999 This is a convenience function which returns the value of
5000 scm_var_random_state. An error message is generated if this value
5001 isn't a random state.
5003 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
5004 Make a new random state from the string SEED of length LENGTH.
5006 It is generally not a good idea to use multiple random states in a
5007 program. While subsequent random numbers generated from one random
5008 state are guaranteed to be reasonably independent, there is no such
5009 guarantee for numbers generated from different random states.
5011 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
5012 Return 32 random bits.
5014 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
5015 Return a sample from the uniform(0,1) distribution.
5017 *** Function: double scm_c_normal01 (scm_rstate *STATE)
5018 Return a sample from the normal(0,1) distribution.
5020 *** Function: double scm_c_exp1 (scm_rstate *STATE)
5021 Return a sample from the exp(1) distribution.
5023 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
5024 Return a sample from the discrete uniform(0,M) distribution.
5026 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
5027 Return a sample from the discrete uniform(0,M) distribution.
5028 M must be a bignum object. The returned value may be an INUM.
5032 Changes in Guile 1.3 (released Monday, October 19, 1998):
5034 * Changes to the distribution
5036 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
5037 To avoid conflicts, programs should name environment variables after
5038 themselves, except when there's a common practice establishing some
5041 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5042 giving the former precedence, and printing a warning message if the
5043 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5045 ** The header files related to multi-byte characters have been removed.
5046 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5047 which referred to these explicitly will probably need to be rewritten,
5048 since the support for the variant string types has been removed; see
5051 ** The header files append.h and sequences.h have been removed. These
5052 files implemented non-R4RS operations which would encourage
5053 non-portable programming style and less easy-to-read code.
5055 * Changes to the stand-alone interpreter
5057 ** New procedures have been added to implement a "batch mode":
5059 *** Function: batch-mode?
5061 Returns a boolean indicating whether the interpreter is in batch
5064 *** Function: set-batch-mode?! ARG
5066 If ARG is true, switches the interpreter to batch mode. The `#f'
5067 case has not been implemented.
5069 ** Guile now provides full command-line editing, when run interactively.
5070 To use this feature, you must have the readline library installed.
5071 The Guile build process will notice it, and automatically include
5074 The readline library is available via anonymous FTP from any GNU
5075 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5077 ** the-last-stack is now a fluid.
5079 * Changes to the procedure for linking libguile with your programs
5081 ** You can now use the `guile-config' utility to build programs that use Guile.
5083 Guile now includes a command-line utility called `guile-config', which
5084 can provide information about how to compile and link programs that
5087 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5088 You should include this command's output on the command line you use
5089 to compile C or C++ code that #includes the Guile header files. It's
5090 usually just a `-I' flag to help the compiler find the Guile headers.
5093 *** `guile-config link' prints any linker flags necessary to link with Guile.
5095 This command writes to its standard output a list of flags which you
5096 must pass to the linker to link your code against the Guile library.
5097 The flags include '-lguile' itself, any other libraries the Guile
5098 library depends upon, and any `-L' flags needed to help the linker
5099 find those libraries.
5101 For example, here is a Makefile rule that builds a program named 'foo'
5102 from the object files ${FOO_OBJECTS}, and links them against Guile:
5105 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5107 Previous Guile releases recommended that you use autoconf to detect
5108 which of a predefined set of libraries were present on your system.
5109 It is more robust to use `guile-config', since it records exactly which
5110 libraries the installed Guile library requires.
5112 This was originally called `build-guile', but was renamed to
5113 `guile-config' before Guile 1.3 was released, to be consistent with
5114 the analogous script for the GTK+ GUI toolkit, which is called
5118 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5120 If you are using the GNU autoconf package to configure your program,
5121 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5122 (described above) and gather the necessary values for use in your
5125 The GUILE_FLAGS macro expands to configure script code which runs the
5126 `guile-config' script, to find out where Guile's header files and
5127 libraries are installed. It sets two variables, marked for
5128 substitution, as by AC_SUBST.
5130 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5131 code that uses Guile header files. This is almost always just a
5134 GUILE_LDFLAGS --- flags to pass to the linker to link a
5135 program against Guile. This includes `-lguile' for the Guile
5136 library itself, any libraries that Guile itself requires (like
5137 -lqthreads), and so on. It may also include a -L flag to tell the
5138 compiler where to find the libraries.
5140 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5141 directory of the Guile distribution. You can copy it into your
5142 package's aclocal.m4 file, and then use it in your configure.in file.
5144 If you are using the `aclocal' program, distributed with GNU automake,
5145 to maintain your aclocal.m4 file, the Guile installation process
5146 installs guile.m4 where aclocal will find it. All you need to do is
5147 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5148 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5152 * Changes to Scheme functions and syntax
5154 ** Multi-byte strings have been removed, as have multi-byte and wide
5155 ports. We felt that these were the wrong approach to
5156 internationalization support.
5158 ** New function: readline [PROMPT]
5159 Read a line from the terminal, and allow the user to edit it,
5160 prompting with PROMPT. READLINE provides a large set of Emacs-like
5161 editing commands, lets the user recall previously typed lines, and
5162 works on almost every kind of terminal, including dumb terminals.
5164 READLINE assumes that the cursor is at the beginning of the line when
5165 it is invoked. Thus, you can't print a prompt yourself, and then call
5166 READLINE; you need to package up your prompt as a string, pass it to
5167 the function, and let READLINE print the prompt itself. This is
5168 because READLINE needs to know the prompt's screen width.
5170 For Guile to provide this function, you must have the readline
5171 library, version 2.1 or later, installed on your system. Readline is
5172 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5173 any GNU mirror site.
5175 See also ADD-HISTORY function.
5177 ** New function: add-history STRING
5178 Add STRING as the most recent line in the history used by the READLINE
5179 command. READLINE does not add lines to the history itself; you must
5180 call ADD-HISTORY to make previous input available to the user.
5182 ** The behavior of the read-line function has changed.
5184 This function now uses standard C library functions to read the line,
5185 for speed. This means that it doesn not respect the value of
5186 scm-line-incrementors; it assumes that lines are delimited with
5189 (Note that this is read-line, the function that reads a line of text
5190 from a port, not readline, the function that reads a line from a
5191 terminal, providing full editing capabilities.)
5193 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5195 This module provides some simple argument parsing. It exports one
5198 Function: getopt-gnu-style ARG-LS
5199 Parse a list of program arguments into an alist of option
5202 Each item in the list of program arguments is examined to see if
5203 it meets the syntax of a GNU long-named option. An argument like
5204 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5205 returned alist, where MUMBLE is a keyword object with the same
5206 name as the argument. An argument like `--MUMBLE=FROB' produces
5207 an element of the form (MUMBLE . FROB), where FROB is a string.
5209 As a special case, the returned alist also contains a pair whose
5210 car is the symbol `rest'. The cdr of this pair is a list
5211 containing all the items in the argument list that are not options
5212 of the form mentioned above.
5214 The argument `--' is treated specially: all items in the argument
5215 list appearing after such an argument are not examined, and are
5216 returned in the special `rest' list.
5218 This function does not parse normal single-character switches.
5219 You will need to parse them out of the `rest' list yourself.
5221 ** The read syntax for byte vectors and short vectors has changed.
5223 Instead of #bytes(...), write #y(...).
5225 Instead of #short(...), write #h(...).
5227 This may seem nutty, but, like the other uniform vectors, byte vectors
5228 and short vectors want to have the same print and read syntax (and,
5229 more basic, want to have read syntax!). Changing the read syntax to
5230 use multiple characters after the hash sign breaks with the
5231 conventions used in R5RS and the conventions used for the other
5232 uniform vectors. It also introduces complexity in the current reader,
5233 both on the C and Scheme levels. (The Right solution is probably to
5234 change the syntax and prototypes for uniform vectors entirely.)
5237 ** The new module (ice-9 session) provides useful interactive functions.
5239 *** New procedure: (apropos REGEXP OPTION ...)
5241 Display a list of top-level variables whose names match REGEXP, and
5242 the modules they are imported from. Each OPTION should be one of the
5245 value --- Show the value of each matching variable.
5246 shadow --- Show bindings shadowed by subsequently imported modules.
5247 full --- Same as both `shadow' and `value'.
5251 guile> (apropos "trace" 'full)
5252 debug: trace #<procedure trace args>
5253 debug: untrace #<procedure untrace args>
5254 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5255 the-scm-module: before-backtrace-hook ()
5256 the-scm-module: backtrace #<primitive-procedure backtrace>
5257 the-scm-module: after-backtrace-hook ()
5258 the-scm-module: has-shown-backtrace-hint? #f
5261 ** There are new functions and syntax for working with macros.
5263 Guile implements macros as a special object type. Any variable whose
5264 top-level binding is a macro object acts as a macro. The macro object
5265 specifies how the expression should be transformed before evaluation.
5267 *** Macro objects now print in a reasonable way, resembling procedures.
5269 *** New function: (macro? OBJ)
5270 True iff OBJ is a macro object.
5272 *** New function: (primitive-macro? OBJ)
5273 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5274 macro transformers, implemented in eval.c rather than Scheme code.
5276 Why do we have this function?
5277 - For symmetry with procedure? and primitive-procedure?,
5278 - to allow custom print procedures to tell whether a macro is
5279 primitive, and display it differently, and
5280 - to allow compilers and user-written evaluators to distinguish
5281 builtin special forms from user-defined ones, which could be
5284 *** New function: (macro-type OBJ)
5285 Return a value indicating what kind of macro OBJ is. Possible return
5288 The symbol `syntax' --- a macro created by procedure->syntax.
5289 The symbol `macro' --- a macro created by procedure->macro.
5290 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5291 The boolean #f --- if OBJ is not a macro object.
5293 *** New function: (macro-name MACRO)
5294 Return the name of the macro object MACRO's procedure, as returned by
5297 *** New function: (macro-transformer MACRO)
5298 Return the transformer procedure for MACRO.
5300 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5302 Specify a new macro expander to use in the current module. Each
5303 MODULE is a module name, with the same meaning as in the `use-modules'
5304 form; each named module's exported bindings are added to the current
5305 top-level environment. TRANSFORMER is an expression evaluated in the
5306 resulting environment which must yield a procedure to use as the
5307 module's eval transformer: every expression evaluated in this module
5308 is passed to this function, and the result passed to the Guile
5311 *** macro-eval! is removed. Use local-eval instead.
5313 ** Some magic has been added to the printer to better handle user
5314 written printing routines (like record printers, closure printers).
5316 The problem is that these user written routines must have access to
5317 the current `print-state' to be able to handle fancy things like
5318 detection of circular references. These print-states have to be
5319 passed to the builtin printing routines (display, write, etc) to
5320 properly continue the print chain.
5322 We didn't want to change all existing print code so that it
5323 explicitly passes thru a print state in addition to a port. Instead,
5324 we extented the possible values that the builtin printing routines
5325 accept as a `port'. In addition to a normal port, they now also take
5326 a pair of a normal port and a print-state. Printing will go to the
5327 port and the print-state will be used to control the detection of
5328 circular references, etc. If the builtin function does not care for a
5329 print-state, it is simply ignored.
5331 User written callbacks are now called with such a pair as their
5332 `port', but because every function now accepts this pair as a PORT
5333 argument, you don't have to worry about that. In fact, it is probably
5334 safest to not check for these pairs.
5336 However, it is sometimes necessary to continue a print chain on a
5337 different port, for example to get a intermediate string
5338 representation of the printed value, mangle that string somehow, and
5339 then to finally print the mangled string. Use the new function
5341 inherit-print-state OLD-PORT NEW-PORT
5343 for this. It constructs a new `port' that prints to NEW-PORT but
5344 inherits the print-state of OLD-PORT.
5346 ** struct-vtable-offset renamed to vtable-offset-user
5348 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5350 ** There is now a third optional argument to make-vtable-vtable
5351 (and fourth to make-struct) when constructing new types (vtables).
5352 This argument initializes field vtable-index-printer of the vtable.
5354 ** The detection of circular references has been extended to structs.
5355 That is, a structure that -- in the process of being printed -- prints
5356 itself does not lead to infinite recursion.
5358 ** There is now some basic support for fluids. Please read
5359 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5360 the following functions and macros:
5362 Function: make-fluid
5364 Create a new fluid object. Fluids are not special variables or
5365 some other extension to the semantics of Scheme, but rather
5366 ordinary Scheme objects. You can store them into variables (that
5367 are still lexically scoped, of course) or into any other place you
5368 like. Every fluid has a initial value of `#f'.
5370 Function: fluid? OBJ
5372 Test whether OBJ is a fluid.
5374 Function: fluid-ref FLUID
5375 Function: fluid-set! FLUID VAL
5377 Access/modify the fluid FLUID. Modifications are only visible
5378 within the current dynamic root (that includes threads).
5380 Function: with-fluids* FLUIDS VALUES THUNK
5382 FLUIDS is a list of fluids and VALUES a corresponding list of
5383 values for these fluids. Before THUNK gets called the values are
5384 installed in the fluids and the old values of the fluids are
5385 saved in the VALUES list. When the flow of control leaves THUNK
5386 or reenters it, the values get swapped again. You might think of
5387 this as a `safe-fluid-excursion'. Note that the VALUES list is
5388 modified by `with-fluids*'.
5390 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5392 The same as `with-fluids*' but with a different syntax. It looks
5393 just like `let', but both FLUID and VALUE are evaluated. Remember,
5394 fluids are not special variables but ordinary objects. FLUID
5395 should evaluate to a fluid.
5397 ** Changes to system call interfaces:
5399 *** close-port, close-input-port and close-output-port now return a
5400 boolean instead of an `unspecified' object. #t means that the port
5401 was successfully closed, while #f means it was already closed. It is
5402 also now possible for these procedures to raise an exception if an
5403 error occurs (some errors from write can be delayed until close.)
5405 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5408 *** the third argument to fcntl is now optional.
5410 *** the first argument to chown can now be a file descriptor or a port.
5412 *** the argument to stat can now be a port.
5414 *** The following new procedures have been added (most use scsh
5417 *** procedure: close PORT/FD
5418 Similar to close-port (*note close-port: Closing Ports.), but also
5419 works on file descriptors. A side effect of closing a file
5420 descriptor is that any ports using that file descriptor are moved
5421 to a different file descriptor and have their revealed counts set
5424 *** procedure: port->fdes PORT
5425 Returns the integer file descriptor underlying PORT. As a side
5426 effect the revealed count of PORT is incremented.
5428 *** procedure: fdes->ports FDES
5429 Returns a list of existing ports which have FDES as an underlying
5430 file descriptor, without changing their revealed counts.
5432 *** procedure: fdes->inport FDES
5433 Returns an existing input port which has FDES as its underlying
5434 file descriptor, if one exists, and increments its revealed count.
5435 Otherwise, returns a new input port with a revealed count of 1.
5437 *** procedure: fdes->outport FDES
5438 Returns an existing output port which has FDES as its underlying
5439 file descriptor, if one exists, and increments its revealed count.
5440 Otherwise, returns a new output port with a revealed count of 1.
5442 The next group of procedures perform a `dup2' system call, if NEWFD
5443 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5444 duplicated can be supplied as an integer or contained in a port. The
5445 type of value returned varies depending on which procedure is used.
5447 All procedures also have the side effect when performing `dup2' that
5448 any ports using NEWFD are moved to a different file descriptor and have
5449 their revealed counts set to zero.
5451 *** procedure: dup->fdes PORT/FD [NEWFD]
5452 Returns an integer file descriptor.
5454 *** procedure: dup->inport PORT/FD [NEWFD]
5455 Returns a new input port using the new file descriptor.
5457 *** procedure: dup->outport PORT/FD [NEWFD]
5458 Returns a new output port using the new file descriptor.
5460 *** procedure: dup PORT/FD [NEWFD]
5461 Returns a new port if PORT/FD is a port, with the same mode as the
5462 supplied port, otherwise returns an integer file descriptor.
5464 *** procedure: dup->port PORT/FD MODE [NEWFD]
5465 Returns a new port using the new file descriptor. MODE supplies a
5466 mode string for the port (*note open-file: File Ports.).
5468 *** procedure: setenv NAME VALUE
5469 Modifies the environment of the current process, which is also the
5470 default environment inherited by child processes.
5472 If VALUE is `#f', then NAME is removed from the environment.
5473 Otherwise, the string NAME=VALUE is added to the environment,
5474 replacing any existing string with name matching NAME.
5476 The return value is unspecified.
5478 *** procedure: truncate-file OBJ SIZE
5479 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5480 can be a string containing a file name or an integer file
5481 descriptor or port open for output on the file. The underlying
5482 system calls are `truncate' and `ftruncate'.
5484 The return value is unspecified.
5486 *** procedure: setvbuf PORT MODE [SIZE]
5487 Set the buffering mode for PORT. MODE can be:
5495 block buffered, using a newly allocated buffer of SIZE bytes.
5496 However if SIZE is zero or unspecified, the port will be made
5499 This procedure should not be used after I/O has been performed with
5502 Ports are usually block buffered by default, with a default buffer
5503 size. Procedures e.g., *Note open-file: File Ports, which accept a
5504 mode string allow `0' to be added to request an unbuffered port.
5506 *** procedure: fsync PORT/FD
5507 Copies any unwritten data for the specified output file descriptor
5508 to disk. If PORT/FD is a port, its buffer is flushed before the
5509 underlying file descriptor is fsync'd. The return value is
5512 *** procedure: open-fdes PATH FLAGS [MODES]
5513 Similar to `open' but returns a file descriptor instead of a port.
5515 *** procedure: execle PATH ENV [ARG] ...
5516 Similar to `execl', but the environment of the new process is
5517 specified by ENV, which must be a list of strings as returned by
5518 the `environ' procedure.
5520 This procedure is currently implemented using the `execve' system
5521 call, but we call it `execle' because of its Scheme calling
5524 *** procedure: strerror ERRNO
5525 Returns the Unix error message corresponding to ERRNO, an integer.
5527 *** procedure: primitive-exit [STATUS]
5528 Terminate the current process without unwinding the Scheme stack.
5529 This is would typically be useful after a fork. The exit status
5530 is STATUS if supplied, otherwise zero.
5532 *** procedure: times
5533 Returns an object with information about real and processor time.
5534 The following procedures accept such an object as an argument and
5535 return a selected component:
5538 The current real time, expressed as time units relative to an
5542 The CPU time units used by the calling process.
5545 The CPU time units used by the system on behalf of the
5549 The CPU time units used by terminated child processes of the
5550 calling process, whose status has been collected (e.g., using
5554 Similarly, the CPU times units used by the system on behalf of
5555 terminated child processes.
5557 ** Removed: list-length
5558 ** Removed: list-append, list-append!
5559 ** Removed: list-reverse, list-reverse!
5561 ** array-map renamed to array-map!
5563 ** serial-array-map renamed to serial-array-map!
5565 ** catch doesn't take #f as first argument any longer
5567 Previously, it was possible to pass #f instead of a key to `catch'.
5568 That would cause `catch' to pass a jump buffer object to the procedure
5569 passed as second argument. The procedure could then use this jump
5570 buffer objekt as an argument to throw.
5572 This mechanism has been removed since its utility doesn't motivate the
5573 extra complexity it introduces.
5575 ** The `#/' notation for lists now provokes a warning message from Guile.
5576 This syntax will be removed from Guile in the near future.
5578 To disable the warning message, set the GUILE_HUSH environment
5579 variable to any non-empty value.
5581 ** The newline character now prints as `#\newline', following the
5582 normal Scheme notation, not `#\nl'.
5584 * Changes to the gh_ interface
5586 ** The gh_enter function now takes care of loading the Guile startup files.
5587 gh_enter works by calling scm_boot_guile; see the remarks below.
5589 ** Function: void gh_write (SCM x)
5591 Write the printed representation of the scheme object x to the current
5592 output port. Corresponds to the scheme level `write'.
5594 ** gh_list_length renamed to gh_length.
5596 ** vector handling routines
5598 Several major changes. In particular, gh_vector() now resembles
5599 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5600 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5601 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5602 vector-related gh_ functions have been implemented.
5604 ** pair and list routines
5606 Implemented several of the R4RS pair and list functions that were
5609 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5611 New function. Converts double arrays back and forth between Scheme
5614 * Changes to the scm_ interface
5616 ** The function scm_boot_guile now takes care of loading the startup files.
5618 Guile's primary initialization function, scm_boot_guile, now takes
5619 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5620 Guile, define the module system, and put together some standard
5621 bindings. It also loads `init.scm', which is intended to hold
5622 site-specific initialization code.
5624 Since Guile cannot operate properly until boot-9.scm is loaded, there
5625 is no reason to separate loading boot-9.scm from Guile's other
5626 initialization processes.
5628 This job used to be done by scm_compile_shell_switches, which didn't
5629 make much sense; in particular, it meant that people using Guile for
5630 non-shell-like applications had to jump through hoops to get Guile
5631 initialized properly.
5633 ** The function scm_compile_shell_switches no longer loads the startup files.
5634 Now, Guile always loads the startup files, whenever it is initialized;
5635 see the notes above for scm_boot_guile and scm_load_startup_files.
5637 ** Function: scm_load_startup_files
5638 This new function takes care of loading Guile's initialization file
5639 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5640 this is always called by the Guile initialization process, it's
5641 probably not too useful to call this yourself, but it's there anyway.
5643 ** The semantics of smob marking have changed slightly.
5645 The smob marking function (the `mark' member of the scm_smobfuns
5646 structure) is no longer responsible for setting the mark bit on the
5647 smob. The generic smob handling code in the garbage collector will
5648 set this bit. The mark function need only ensure that any other
5649 objects the smob refers to get marked.
5651 Note that this change means that the smob's GC8MARK bit is typically
5652 already set upon entry to the mark function. Thus, marking functions
5653 which look like this:
5656 if (SCM_GC8MARKP (ptr))
5658 SCM_SETGC8MARK (ptr);
5659 ... mark objects to which the smob refers ...
5662 are now incorrect, since they will return early, and fail to mark any
5663 other objects the smob refers to. Some code in the Guile library used
5666 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5668 If you have implemented your own I/O port type, by writing the
5669 functions required by the scm_ptobfuns and then calling scm_newptob,
5670 you will need to change your functions slightly.
5672 The functions in a scm_ptobfuns structure now expect the port itself
5673 as their argument; they used to expect the `stream' member of the
5674 port's scm_port_table structure. This allows functions in an
5675 scm_ptobfuns structure to easily access the port's cell (and any flags
5676 it its CAR), and the port's scm_port_table structure.
5678 Guile now passes the I/O port itself as the `port' argument in the
5679 following scm_ptobfuns functions:
5681 int (*free) (SCM port);
5682 int (*fputc) (int, SCM port);
5683 int (*fputs) (char *, SCM port);
5684 scm_sizet (*fwrite) SCM_P ((char *ptr,
5688 int (*fflush) (SCM port);
5689 int (*fgetc) (SCM port);
5690 int (*fclose) (SCM port);
5692 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5695 If you have existing code which defines its own port types, it is easy
5696 to convert your code to the new interface; simply apply SCM_STREAM to
5697 the port argument to yield the value you code used to expect.
5699 Note that since both the port and the stream have the same type in the
5700 C code --- they are both SCM values --- the C compiler will not remind
5701 you if you forget to update your scm_ptobfuns functions.
5704 ** Function: int scm_internal_select (int fds,
5708 struct timeval *timeout);
5710 This is a replacement for the `select' function provided by the OS.
5711 It enables I/O blocking and sleeping to happen for one cooperative
5712 thread without blocking other threads. It also avoids busy-loops in
5713 these situations. It is intended that all I/O blocking and sleeping
5714 will finally go through this function. Currently, this function is
5715 only available on systems providing `gettimeofday' and `select'.
5717 ** Function: SCM scm_internal_stack_catch (SCM tag,
5718 scm_catch_body_t body,
5720 scm_catch_handler_t handler,
5723 A new sibling to the other two C level `catch' functions
5724 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5725 the stack to be saved automatically into the variable `the-last-stack'
5726 (scm_the_last_stack_var) on error. This is necessary if you want to
5727 use advanced error reporting, such as calling scm_display_error and
5728 scm_display_backtrace. (They both take a stack object as argument.)
5730 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5732 scm_catch_handler_t handler,
5735 Spawns a new thread. It does a job similar to
5736 scm_call_with_new_thread but takes arguments more suitable when
5737 spawning threads from application C code.
5739 ** The hook scm_error_callback has been removed. It was originally
5740 intended as a way for the user to install his own error handler. But
5741 that method works badly since it intervenes between throw and catch,
5742 thereby changing the semantics of expressions like (catch #t ...).
5743 The correct way to do it is to use one of the C level catch functions
5744 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5746 ** Removed functions:
5748 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5749 scm_list_reverse, scm_list_reverse_x
5751 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5753 These can be used for pretty list creation from C. The idea is taken
5754 from Erick Gallesio's STk.
5756 ** scm_array_map renamed to scm_array_map_x
5758 ** mbstrings are now removed
5760 This means that the type codes scm_tc7_mb_string and
5761 scm_tc7_mb_substring has been removed.
5763 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5765 Since we no longer support multi-byte strings, these I/O functions
5766 have been simplified, and renamed. Here are their old names, and
5767 their new names and arguments:
5769 scm_gen_putc -> void scm_putc (int c, SCM port);
5770 scm_gen_puts -> void scm_puts (char *s, SCM port);
5771 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5772 scm_gen_getc -> void scm_getc (SCM port);
5775 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5777 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5779 SCM_TYP7S now masks away the bit which distinguishes substrings from
5782 ** scm_catch_body_t: Backward incompatible change!
5784 Body functions to scm_internal_catch and friends do not any longer
5785 take a second argument. This is because it is no longer possible to
5786 pass a #f arg to catch.
5788 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5790 The function scm_protect_object protects its argument from being freed
5791 by the garbage collector. scm_unprotect_object removes that
5794 These functions now nest properly. That is, for every object O, there
5795 is a counter which scm_protect_object(O) increments and
5796 scm_unprotect_object(O) decrements, if the counter is greater than
5797 zero. Every object's counter is zero when it is first created. If an
5798 object's counter is greater than zero, the garbage collector will not
5799 reclaim its storage.
5801 This allows you to use scm_protect_object in your code without
5802 worrying that some other function you call will call
5803 scm_unprotect_object, and allow it to be freed. Assuming that the
5804 functions you call are well-behaved, and unprotect only those objects
5805 they protect, you can follow the same rule and have confidence that
5806 objects will be freed only at appropriate times.
5809 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5811 * Changes to the distribution
5813 ** Nightly snapshots are now available from ftp.red-bean.com.
5814 The old server, ftp.cyclic.com, has been relinquished to its rightful
5817 Nightly snapshots of the Guile development sources are now available via
5818 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5820 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5821 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5823 ** To run Guile without installing it, the procedure has changed a bit.
5825 If you used a separate build directory to compile Guile, you'll need
5826 to include the build directory in SCHEME_LOAD_PATH, as well as the
5827 source directory. See the `INSTALL' file for examples.
5829 * Changes to the procedure for linking libguile with your programs
5831 ** The standard Guile load path for Scheme code now includes
5832 $(datadir)/guile (usually /usr/local/share/guile). This means that
5833 you can install your own Scheme files there, and Guile will find them.
5834 (Previous versions of Guile only checked a directory whose name
5835 contained the Guile version number, so you had to re-install or move
5836 your Scheme sources each time you installed a fresh version of Guile.)
5838 The load path also includes $(datadir)/guile/site; we recommend
5839 putting individual Scheme files there. If you want to install a
5840 package with multiple source files, create a directory for them under
5843 ** Guile 1.2 will now use the Rx regular expression library, if it is
5844 installed on your system. When you are linking libguile into your own
5845 programs, this means you will have to link against -lguile, -lqt (if
5846 you configured Guile with thread support), and -lrx.
5848 If you are using autoconf to generate configuration scripts for your
5849 application, the following lines should suffice to add the appropriate
5850 libraries to your link command:
5852 ### Find Rx, quickthreads and libguile.
5853 AC_CHECK_LIB(rx, main)
5854 AC_CHECK_LIB(qt, main)
5855 AC_CHECK_LIB(guile, scm_shell)
5857 The Guile 1.2 distribution does not contain sources for the Rx
5858 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5859 retrieve it from a GNU FTP site and install it separately.
5861 * Changes to Scheme functions and syntax
5863 ** The dynamic linking features of Guile are now enabled by default.
5864 You can disable them by giving the `--disable-dynamic-linking' option
5867 (dynamic-link FILENAME)
5869 Find the object file denoted by FILENAME (a string) and link it
5870 into the running Guile application. When everything works out,
5871 return a Scheme object suitable for representing the linked object
5872 file. Otherwise an error is thrown. How object files are
5873 searched is system dependent.
5875 (dynamic-object? VAL)
5877 Determine whether VAL represents a dynamically linked object file.
5879 (dynamic-unlink DYNOBJ)
5881 Unlink the indicated object file from the application. DYNOBJ
5882 should be one of the values returned by `dynamic-link'.
5884 (dynamic-func FUNCTION DYNOBJ)
5886 Search the C function indicated by FUNCTION (a string or symbol)
5887 in DYNOBJ and return some Scheme object that can later be used
5888 with `dynamic-call' to actually call this function. Right now,
5889 these Scheme objects are formed by casting the address of the
5890 function to `long' and converting this number to its Scheme
5893 (dynamic-call FUNCTION DYNOBJ)
5895 Call the C function indicated by FUNCTION and DYNOBJ. The
5896 function is passed no arguments and its return value is ignored.
5897 When FUNCTION is something returned by `dynamic-func', call that
5898 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5899 etc.), look it up in DYNOBJ; this is equivalent to
5901 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5903 Interrupts are deferred while the C function is executing (with
5904 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5906 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5908 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5909 some arguments and return its return value. The C function is
5910 expected to take two arguments and return an `int', just like
5913 int c_func (int argc, char **argv);
5915 ARGS must be a list of strings and is converted into an array of
5916 `char *'. The array is passed in ARGV and its size in ARGC. The
5917 return value is converted to a Scheme number and returned from the
5918 call to `dynamic-args-call'.
5920 When dynamic linking is disabled or not supported on your system,
5921 the above functions throw errors, but they are still available.
5923 Here is a small example that works on GNU/Linux:
5925 (define libc-obj (dynamic-link "libc.so"))
5926 (dynamic-args-call 'rand libc-obj '())
5928 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5930 ** The #/ syntax for module names is depreciated, and will be removed
5931 in a future version of Guile. Instead of
5939 The latter syntax is more consistent with existing Lisp practice.
5941 ** Guile now does fancier printing of structures. Structures are the
5942 underlying implementation for records, which in turn are used to
5943 implement modules, so all of these object now print differently and in
5944 a more informative way.
5946 The Scheme printer will examine the builtin variable *struct-printer*
5947 whenever it needs to print a structure object. When this variable is
5948 not `#f' it is deemed to be a procedure and will be applied to the
5949 structure object and the output port. When *struct-printer* is `#f'
5950 or the procedure return `#f' the structure object will be printed in
5951 the boring #<struct 80458270> form.
5953 This hook is used by some routines in ice-9/boot-9.scm to implement
5954 type specific printing routines. Please read the comments there about
5957 One of the more specific uses of structs are records. The printing
5958 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5959 called. It should behave like a *struct-printer* procedure (described
5962 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5963 token of the form #:NAME, where NAME has the same syntax as a Scheme
5964 symbol, is the external representation of the keyword named NAME.
5965 Keyword objects print using this syntax as well, so values containing
5966 keyword objects can be read back into Guile. When used in an
5967 expression, keywords are self-quoting objects.
5969 Guile suports this read syntax, and uses this print syntax, regardless
5970 of the current setting of the `keyword' read option. The `keyword'
5971 read option only controls whether Guile recognizes the `:NAME' syntax,
5972 which is incompatible with R4RS. (R4RS says such token represent
5975 ** Guile has regular expression support again. Guile 1.0 included
5976 functions for matching regular expressions, based on the Rx library.
5977 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5978 distribution, and thus Guile had no regular expression support. Guile
5979 1.2 again supports the most commonly used functions, and supports all
5980 of SCSH's regular expression functions.
5982 If your system does not include a POSIX regular expression library,
5983 and you have not linked Guile with a third-party regexp library such as
5984 Rx, these functions will not be available. You can tell whether your
5985 Guile installation includes regular expression support by checking
5986 whether the `*features*' list includes the `regex' symbol.
5988 *** regexp functions
5990 By default, Guile supports POSIX extended regular expressions. That
5991 means that the characters `(', `)', `+' and `?' are special, and must
5992 be escaped if you wish to match the literal characters.
5994 This regular expression interface was modeled after that implemented
5995 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5996 with SCSH regular expressions.
5998 **** Function: string-match PATTERN STR [START]
5999 Compile the string PATTERN into a regular expression and compare
6000 it with STR. The optional numeric argument START specifies the
6001 position of STR at which to begin matching.
6003 `string-match' returns a "match structure" which describes what,
6004 if anything, was matched by the regular expression. *Note Match
6005 Structures::. If STR does not match PATTERN at all,
6006 `string-match' returns `#f'.
6008 Each time `string-match' is called, it must compile its PATTERN
6009 argument into a regular expression structure. This operation is
6010 expensive, which makes `string-match' inefficient if the same regular
6011 expression is used several times (for example, in a loop). For better
6012 performance, you can compile a regular expression in advance and then
6013 match strings against the compiled regexp.
6015 **** Function: make-regexp STR [FLAGS]
6016 Compile the regular expression described by STR, and return the
6017 compiled regexp structure. If STR does not describe a legal
6018 regular expression, `make-regexp' throws a
6019 `regular-expression-syntax' error.
6021 FLAGS may be the bitwise-or of one or more of the following:
6023 **** Constant: regexp/extended
6024 Use POSIX Extended Regular Expression syntax when interpreting
6025 STR. If not set, POSIX Basic Regular Expression syntax is used.
6026 If the FLAGS argument is omitted, we assume regexp/extended.
6028 **** Constant: regexp/icase
6029 Do not differentiate case. Subsequent searches using the
6030 returned regular expression will be case insensitive.
6032 **** Constant: regexp/newline
6033 Match-any-character operators don't match a newline.
6035 A non-matching list ([^...]) not containing a newline matches a
6038 Match-beginning-of-line operator (^) matches the empty string
6039 immediately after a newline, regardless of whether the FLAGS
6040 passed to regexp-exec contain regexp/notbol.
6042 Match-end-of-line operator ($) matches the empty string
6043 immediately before a newline, regardless of whether the FLAGS
6044 passed to regexp-exec contain regexp/noteol.
6046 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6047 Match the compiled regular expression REGEXP against `str'. If
6048 the optional integer START argument is provided, begin matching
6049 from that position in the string. Return a match structure
6050 describing the results of the match, or `#f' if no match could be
6053 FLAGS may be the bitwise-or of one or more of the following:
6055 **** Constant: regexp/notbol
6056 The match-beginning-of-line operator always fails to match (but
6057 see the compilation flag regexp/newline above) This flag may be
6058 used when different portions of a string are passed to
6059 regexp-exec and the beginning of the string should not be
6060 interpreted as the beginning of the line.
6062 **** Constant: regexp/noteol
6063 The match-end-of-line operator always fails to match (but see the
6064 compilation flag regexp/newline above)
6066 **** Function: regexp? OBJ
6067 Return `#t' if OBJ is a compiled regular expression, or `#f'
6070 Regular expressions are commonly used to find patterns in one string
6071 and replace them with the contents of another string.
6073 **** Function: regexp-substitute PORT MATCH [ITEM...]
6074 Write to the output port PORT selected contents of the match
6075 structure MATCH. Each ITEM specifies what should be written, and
6076 may be one of the following arguments:
6078 * A string. String arguments are written out verbatim.
6080 * An integer. The submatch with that number is written.
6082 * The symbol `pre'. The portion of the matched string preceding
6083 the regexp match is written.
6085 * The symbol `post'. The portion of the matched string
6086 following the regexp match is written.
6088 PORT may be `#f', in which case nothing is written; instead,
6089 `regexp-substitute' constructs a string from the specified ITEMs
6092 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6093 Similar to `regexp-substitute', but can be used to perform global
6094 substitutions on STR. Instead of taking a match structure as an
6095 argument, `regexp-substitute/global' takes two string arguments: a
6096 REGEXP string describing a regular expression, and a TARGET string
6097 which should be matched against this regular expression.
6099 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6102 * A function may be supplied. When this function is called, it
6103 will be passed one argument: a match structure for a given
6104 regular expression match. It should return a string to be
6105 written out to PORT.
6107 * The `post' symbol causes `regexp-substitute/global' to recurse
6108 on the unmatched portion of STR. This *must* be supplied in
6109 order to perform global search-and-replace on STR; if it is
6110 not present among the ITEMs, then `regexp-substitute/global'
6111 will return after processing a single match.
6113 *** Match Structures
6115 A "match structure" is the object returned by `string-match' and
6116 `regexp-exec'. It describes which portion of a string, if any, matched
6117 the given regular expression. Match structures include: a reference to
6118 the string that was checked for matches; the starting and ending
6119 positions of the regexp match; and, if the regexp included any
6120 parenthesized subexpressions, the starting and ending positions of each
6123 In each of the regexp match functions described below, the `match'
6124 argument must be a match structure returned by a previous call to
6125 `string-match' or `regexp-exec'. Most of these functions return some
6126 information about the original target string that was matched against a
6127 regular expression; we will call that string TARGET for easy reference.
6129 **** Function: regexp-match? OBJ
6130 Return `#t' if OBJ is a match structure returned by a previous
6131 call to `regexp-exec', or `#f' otherwise.
6133 **** Function: match:substring MATCH [N]
6134 Return the portion of TARGET matched by subexpression number N.
6135 Submatch 0 (the default) represents the entire regexp match. If
6136 the regular expression as a whole matched, but the subexpression
6137 number N did not match, return `#f'.
6139 **** Function: match:start MATCH [N]
6140 Return the starting position of submatch number N.
6142 **** Function: match:end MATCH [N]
6143 Return the ending position of submatch number N.
6145 **** Function: match:prefix MATCH
6146 Return the unmatched portion of TARGET preceding the regexp match.
6148 **** Function: match:suffix MATCH
6149 Return the unmatched portion of TARGET following the regexp match.
6151 **** Function: match:count MATCH
6152 Return the number of parenthesized subexpressions from MATCH.
6153 Note that the entire regular expression match itself counts as a
6154 subexpression, and failed submatches are included in the count.
6156 **** Function: match:string MATCH
6157 Return the original TARGET string.
6159 *** Backslash Escapes
6161 Sometimes you will want a regexp to match characters like `*' or `$'
6162 exactly. For example, to check whether a particular string represents
6163 a menu entry from an Info node, it would be useful to match it against
6164 a regexp like `^* [^:]*::'. However, this won't work; because the
6165 asterisk is a metacharacter, it won't match the `*' at the beginning of
6166 the string. In this case, we want to make the first asterisk un-magic.
6168 You can do this by preceding the metacharacter with a backslash
6169 character `\'. (This is also called "quoting" the metacharacter, and
6170 is known as a "backslash escape".) When Guile sees a backslash in a
6171 regular expression, it considers the following glyph to be an ordinary
6172 character, no matter what special meaning it would ordinarily have.
6173 Therefore, we can make the above example work by changing the regexp to
6174 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6175 to match only a single asterisk in the target string.
6177 Since the backslash is itself a metacharacter, you may force a
6178 regexp to match a backslash in the target string by preceding the
6179 backslash with itself. For example, to find variable references in a
6180 TeX program, you might want to find occurrences of the string `\let\'
6181 followed by any number of alphabetic characters. The regular expression
6182 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6183 each match a single backslash in the target string.
6185 **** Function: regexp-quote STR
6186 Quote each special character found in STR with a backslash, and
6187 return the resulting string.
6189 *Very important:* Using backslash escapes in Guile source code (as
6190 in Emacs Lisp or C) can be tricky, because the backslash character has
6191 special meaning for the Guile reader. For example, if Guile encounters
6192 the character sequence `\n' in the middle of a string while processing
6193 Scheme code, it replaces those characters with a newline character.
6194 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6195 Several of these "escape sequences" are processed by the Guile reader
6196 before your code is executed. Unrecognized escape sequences are
6197 ignored: if the characters `\*' appear in a string, they will be
6198 translated to the single character `*'.
6200 This translation is obviously undesirable for regular expressions,
6201 since we want to be able to include backslashes in a string in order to
6202 escape regexp metacharacters. Therefore, to make sure that a backslash
6203 is preserved in a string in your Guile program, you must use *two*
6204 consecutive backslashes:
6206 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6208 The string in this example is preprocessed by the Guile reader before
6209 any code is executed. The resulting argument to `make-regexp' is the
6210 string `^\* [^:]*', which is what we really want.
6212 This also means that in order to write a regular expression that
6213 matches a single backslash character, the regular expression string in
6214 the source code must include *four* backslashes. Each consecutive pair
6215 of backslashes gets translated by the Guile reader to a single
6216 backslash, and the resulting double-backslash is interpreted by the
6217 regexp engine as matching a single backslash character. Hence:
6219 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6221 The reason for the unwieldiness of this syntax is historical. Both
6222 regular expression pattern matchers and Unix string processing systems
6223 have traditionally used backslashes with the special meanings described
6224 above. The POSIX regular expression specification and ANSI C standard
6225 both require these semantics. Attempting to abandon either convention
6226 would cause other kinds of compatibility problems, possibly more severe
6227 ones. Therefore, without extending the Scheme reader to support
6228 strings with different quoting conventions (an ungainly and confusing
6229 extension when implemented in other languages), we must adhere to this
6230 cumbersome escape syntax.
6232 * Changes to the gh_ interface
6234 * Changes to the scm_ interface
6236 * Changes to system call interfaces:
6238 ** The value returned by `raise' is now unspecified. It throws an exception
6241 *** A new procedure `sigaction' can be used to install signal handlers
6243 (sigaction signum [action] [flags])
6245 signum is the signal number, which can be specified using the value
6248 If action is omitted, sigaction returns a pair: the CAR is the current
6249 signal hander, which will be either an integer with the value SIG_DFL
6250 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6251 handles the signal, or #f if a non-Scheme procedure handles the
6252 signal. The CDR contains the current sigaction flags for the handler.
6254 If action is provided, it is installed as the new handler for signum.
6255 action can be a Scheme procedure taking one argument, or the value of
6256 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6257 whatever signal handler was installed before sigaction was first used.
6258 Flags can optionally be specified for the new handler (SA_RESTART is
6259 always used if the system provides it, so need not be specified.) The
6260 return value is a pair with information about the old handler as
6263 This interface does not provide access to the "signal blocking"
6264 facility. Maybe this is not needed, since the thread support may
6265 provide solutions to the problem of consistent access to data
6268 *** A new procedure `flush-all-ports' is equivalent to running
6269 `force-output' on every port open for output.
6271 ** Guile now provides information on how it was built, via the new
6272 global variable, %guile-build-info. This variable records the values
6273 of the standard GNU makefile directory variables as an assocation
6274 list, mapping variable names (symbols) onto directory paths (strings).
6275 For example, to find out where the Guile link libraries were
6276 installed, you can say:
6278 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6281 * Changes to the scm_ interface
6283 ** The new function scm_handle_by_message_noexit is just like the
6284 existing scm_handle_by_message function, except that it doesn't call
6285 exit to terminate the process. Instead, it prints a message and just
6286 returns #f. This might be a more appropriate catch-all handler for
6287 new dynamic roots and threads.
6290 Changes in Guile 1.1 (released Friday, May 16 1997):
6292 * Changes to the distribution.
6294 The Guile 1.0 distribution has been split up into several smaller
6296 guile-core --- the Guile interpreter itself.
6297 guile-tcltk --- the interface between the Guile interpreter and
6298 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6299 is a toolkit for building graphical user interfaces.
6300 guile-rgx-ctax --- the interface between Guile and the Rx regular
6301 expression matcher, and the translator for the Ctax
6302 programming language. These are packaged together because the
6303 Ctax translator uses Rx to parse Ctax source code.
6305 This NEWS file describes the changes made to guile-core since the 1.0
6308 We no longer distribute the documentation, since it was either out of
6309 date, or incomplete. As soon as we have current documentation, we
6314 * Changes to the stand-alone interpreter
6316 ** guile now accepts command-line arguments compatible with SCSH, Olin
6317 Shivers' Scheme Shell.
6319 In general, arguments are evaluated from left to right, but there are
6320 exceptions. The following switches stop argument processing, and
6321 stash all remaining command-line arguments as the value returned by
6322 the (command-line) function.
6323 -s SCRIPT load Scheme source code from FILE, and exit
6324 -c EXPR evalute Scheme expression EXPR, and exit
6325 -- stop scanning arguments; run interactively
6327 The switches below are processed as they are encountered.
6328 -l FILE load Scheme source code from FILE
6329 -e FUNCTION after reading script, apply FUNCTION to
6330 command line arguments
6331 -ds do -s script at this point
6332 --emacs enable Emacs protocol (experimental)
6333 -h, --help display this help and exit
6334 -v, --version display version information and exit
6335 \ read arguments from following script lines
6337 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6338 which re-implements the traditional "echo" command:
6340 #!/usr/local/bin/guile -s
6343 (map (lambda (arg) (display arg) (display " "))
6347 (main (command-line))
6349 Suppose we invoke this script as follows:
6351 ekko a speckled gecko
6353 Through the magic of Unix script processing (triggered by the `#!'
6354 token at the top of the file), /usr/local/bin/guile receives the
6355 following list of command-line arguments:
6357 ("-s" "./ekko" "a" "speckled" "gecko")
6359 Unix inserts the name of the script after the argument specified on
6360 the first line of the file (in this case, "-s"), and then follows that
6361 with the arguments given to the script. Guile loads the script, which
6362 defines the `main' function, and then applies it to the list of
6363 remaining command-line arguments, ("a" "speckled" "gecko").
6365 In Unix, the first line of a script file must take the following form:
6367 #!INTERPRETER ARGUMENT
6369 where INTERPRETER is the absolute filename of the interpreter
6370 executable, and ARGUMENT is a single command-line argument to pass to
6373 You may only pass one argument to the interpreter, and its length is
6374 limited. These restrictions can be annoying to work around, so Guile
6375 provides a general mechanism (borrowed from, and compatible with,
6376 SCSH) for circumventing them.
6378 If the ARGUMENT in a Guile script is a single backslash character,
6379 `\', Guile will open the script file, parse arguments from its second
6380 and subsequent lines, and replace the `\' with them. So, for example,
6381 here is another implementation of the `ekko' script:
6383 #!/usr/local/bin/guile \
6387 (for-each (lambda (arg) (display arg) (display " "))
6391 If the user invokes this script as follows:
6393 ekko a speckled gecko
6395 Unix expands this into
6397 /usr/local/bin/guile \ ekko a speckled gecko
6399 When Guile sees the `\' argument, it replaces it with the arguments
6400 read from the second line of the script, producing:
6402 /usr/local/bin/guile -e main -s ekko a speckled gecko
6404 This tells Guile to load the `ekko' script, and apply the function
6405 `main' to the argument list ("a" "speckled" "gecko").
6407 Here is how Guile parses the command-line arguments:
6408 - Each space character terminates an argument. This means that two
6409 spaces in a row introduce an empty-string argument.
6410 - The tab character is not permitted (unless you quote it with the
6411 backslash character, as described below), to avoid confusion.
6412 - The newline character terminates the sequence of arguments, and will
6413 also terminate a final non-empty argument. (However, a newline
6414 following a space will not introduce a final empty-string argument;
6415 it only terminates the argument list.)
6416 - The backslash character is the escape character. It escapes
6417 backslash, space, tab, and newline. The ANSI C escape sequences
6418 like \n and \t are also supported. These produce argument
6419 constituents; the two-character combination \n doesn't act like a
6420 terminating newline. The escape sequence \NNN for exactly three
6421 octal digits reads as the character whose ASCII code is NNN. As
6422 above, characters produced this way are argument constituents.
6423 Backslash followed by other characters is not allowed.
6425 * Changes to the procedure for linking libguile with your programs
6427 ** Guile now builds and installs a shared guile library, if your
6428 system support shared libraries. (It still builds a static library on
6429 all systems.) Guile automatically detects whether your system
6430 supports shared libraries. To prevent Guile from buildisg shared
6431 libraries, pass the `--disable-shared' flag to the configure script.
6433 Guile takes longer to compile when it builds shared libraries, because
6434 it must compile every file twice --- once to produce position-
6435 independent object code, and once to produce normal object code.
6437 ** The libthreads library has been merged into libguile.
6439 To link a program against Guile, you now need only link against
6440 -lguile and -lqt; -lthreads is no longer needed. If you are using
6441 autoconf to generate configuration scripts for your application, the
6442 following lines should suffice to add the appropriate libraries to
6445 ### Find quickthreads and libguile.
6446 AC_CHECK_LIB(qt, main)
6447 AC_CHECK_LIB(guile, scm_shell)
6449 * Changes to Scheme functions
6451 ** Guile Scheme's special syntax for keyword objects is now optional,
6452 and disabled by default.
6454 The syntax variation from R4RS made it difficult to port some
6455 interesting packages to Guile. The routines which accepted keyword
6456 arguments (mostly in the module system) have been modified to also
6457 accept symbols whose names begin with `:'.
6459 To change the keyword syntax, you must first import the (ice-9 debug)
6461 (use-modules (ice-9 debug))
6463 Then you can enable the keyword syntax as follows:
6464 (read-set! keywords 'prefix)
6466 To disable keyword syntax, do this:
6467 (read-set! keywords #f)
6469 ** Many more primitive functions accept shared substrings as
6470 arguments. In the past, these functions required normal, mutable
6471 strings as arguments, although they never made use of this
6474 ** The uniform array functions now operate on byte vectors. These
6475 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6476 `serial-array-map', `array-map', `array-for-each', and
6479 ** The new functions `trace' and `untrace' implement simple debugging
6480 support for Scheme functions.
6482 The `trace' function accepts any number of procedures as arguments,
6483 and tells the Guile interpreter to display each procedure's name and
6484 arguments each time the procedure is invoked. When invoked with no
6485 arguments, `trace' returns the list of procedures currently being
6488 The `untrace' function accepts any number of procedures as arguments,
6489 and tells the Guile interpreter not to trace them any more. When
6490 invoked with no arguments, `untrace' untraces all curretly traced
6493 The tracing in Guile has an advantage over most other systems: we
6494 don't create new procedure objects, but mark the procedure objects
6495 themselves. This means that anonymous and internal procedures can be
6498 ** The function `assert-repl-prompt' has been renamed to
6499 `set-repl-prompt!'. It takes one argument, PROMPT.
6500 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6501 - If PROMPT is a string, we use it as a prompt.
6502 - If PROMPT is a procedure accepting no arguments, we call it, and
6503 display the result as a prompt.
6504 - Otherwise, we display "> ".
6506 ** The new function `eval-string' reads Scheme expressions from a
6507 string and evaluates them, returning the value of the last expression
6508 in the string. If the string contains no expressions, it returns an
6511 ** The new function `thunk?' returns true iff its argument is a
6512 procedure of zero arguments.
6514 ** `defined?' is now a builtin function, instead of syntax. This
6515 means that its argument should be quoted. It returns #t iff its
6516 argument is bound in the current module.
6518 ** The new syntax `use-modules' allows you to add new modules to your
6519 environment without re-typing a complete `define-module' form. It
6520 accepts any number of module names as arguments, and imports their
6521 public bindings into the current module.
6523 ** The new function (module-defined? NAME MODULE) returns true iff
6524 NAME, a symbol, is defined in MODULE, a module object.
6526 ** The new function `builtin-bindings' creates and returns a hash
6527 table containing copies of all the root module's bindings.
6529 ** The new function `builtin-weak-bindings' does the same as
6530 `builtin-bindings', but creates a doubly-weak hash table.
6532 ** The `equal?' function now considers variable objects to be
6533 equivalent if they have the same name and the same value.
6535 ** The new function `command-line' returns the command-line arguments
6536 given to Guile, as a list of strings.
6538 When using guile as a script interpreter, `command-line' returns the
6539 script's arguments; those processed by the interpreter (like `-s' or
6540 `-c') are omitted. (In other words, you get the normal, expected
6541 behavior.) Any application that uses scm_shell to process its
6542 command-line arguments gets this behavior as well.
6544 ** The new function `load-user-init' looks for a file called `.guile'
6545 in the user's home directory, and loads it if it exists. This is
6546 mostly for use by the code generated by scm_compile_shell_switches,
6547 but we thought it might also be useful in other circumstances.
6549 ** The new function `log10' returns the base-10 logarithm of its
6552 ** Changes to I/O functions
6554 *** The functions `read', `primitive-load', `read-and-eval!', and
6555 `primitive-load-path' no longer take optional arguments controlling
6556 case insensitivity and a `#' parser.
6558 Case sensitivity is now controlled by a read option called
6559 `case-insensitive'. The user can add new `#' syntaxes with the
6560 `read-hash-extend' function (see below).
6562 *** The new function `read-hash-extend' allows the user to change the
6563 syntax of Guile Scheme in a somewhat controlled way.
6565 (read-hash-extend CHAR PROC)
6566 When parsing S-expressions, if we read a `#' character followed by
6567 the character CHAR, use PROC to parse an object from the stream.
6568 If PROC is #f, remove any parsing procedure registered for CHAR.
6570 The reader applies PROC to two arguments: CHAR and an input port.
6572 *** The new functions read-delimited and read-delimited! provide a
6573 general mechanism for doing delimited input on streams.
6575 (read-delimited DELIMS [PORT HANDLE-DELIM])
6576 Read until we encounter one of the characters in DELIMS (a string),
6577 or end-of-file. PORT is the input port to read from; it defaults to
6578 the current input port. The HANDLE-DELIM parameter determines how
6579 the terminating character is handled; it should be one of the
6582 'trim omit delimiter from result
6583 'peek leave delimiter character in input stream
6584 'concat append delimiter character to returned value
6585 'split return a pair: (RESULT . TERMINATOR)
6587 HANDLE-DELIM defaults to 'peek.
6589 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6590 A side-effecting variant of `read-delimited'.
6592 The data is written into the string BUF at the indices in the
6593 half-open interval [START, END); the default interval is the whole
6594 string: START = 0 and END = (string-length BUF). The values of
6595 START and END must specify a well-defined interval in BUF, i.e.
6596 0 <= START <= END <= (string-length BUF).
6598 It returns NBYTES, the number of bytes read. If the buffer filled
6599 up without a delimiter character being found, it returns #f. If the
6600 port is at EOF when the read starts, it returns the EOF object.
6602 If an integer is returned (i.e., the read is successfully terminated
6603 by reading a delimiter character), then the HANDLE-DELIM parameter
6604 determines how to handle the terminating character. It is described
6605 above, and defaults to 'peek.
6607 (The descriptions of these functions were borrowed from the SCSH
6608 manual, by Olin Shivers and Brian Carlstrom.)
6610 *** The `%read-delimited!' function is the primitive used to implement
6611 `read-delimited' and `read-delimited!'.
6613 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6615 This returns a pair of values: (TERMINATOR . NUM-READ).
6616 - TERMINATOR describes why the read was terminated. If it is a
6617 character or the eof object, then that is the value that terminated
6618 the read. If it is #f, the function filled the buffer without finding
6619 a delimiting character.
6620 - NUM-READ is the number of characters read into BUF.
6622 If the read is successfully terminated by reading a delimiter
6623 character, then the gobble? parameter determines what to do with the
6624 terminating character. If true, the character is removed from the
6625 input stream; if false, the character is left in the input stream
6626 where a subsequent read operation will retrieve it. In either case,
6627 the character is also the first value returned by the procedure call.
6629 (The descriptions of this function was borrowed from the SCSH manual,
6630 by Olin Shivers and Brian Carlstrom.)
6632 *** The `read-line' and `read-line!' functions have changed; they now
6633 trim the terminator by default; previously they appended it to the
6634 returned string. For the old behavior, use (read-line PORT 'concat).
6636 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6637 take new optional START and END arguments, specifying the region of
6638 the array to read and write.
6640 *** The `ungetc-char-ready?' function has been removed. We feel it's
6641 inappropriate for an interface to expose implementation details this
6644 ** Changes to the Unix library and system call interface
6646 *** The new fcntl function provides access to the Unix `fcntl' system
6649 (fcntl PORT COMMAND VALUE)
6650 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6651 Values for COMMAND are:
6653 F_DUPFD duplicate a file descriptor
6654 F_GETFD read the descriptor's close-on-exec flag
6655 F_SETFD set the descriptor's close-on-exec flag to VALUE
6656 F_GETFL read the descriptor's flags, as set on open
6657 F_SETFL set the descriptor's flags, as set on open to VALUE
6658 F_GETOWN return the process ID of a socket's owner, for SIGIO
6659 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6660 FD_CLOEXEC not sure what this is
6662 For details, see the documentation for the fcntl system call.
6664 *** The arguments to `select' have changed, for compatibility with
6665 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6666 expected behavior. The MILLISECONDS parameter has been changed to
6667 MICROSECONDS, to more closely resemble the underlying system call.
6668 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6669 corresponding return set will be the same.
6671 *** The arguments to the `mknod' system call have changed. They are
6674 (mknod PATH TYPE PERMS DEV)
6675 Create a new file (`node') in the file system. PATH is the name of
6676 the file to create. TYPE is the kind of file to create; it should
6677 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6678 permission bits to give the newly created file. If TYPE is
6679 'block-special or 'char-special, DEV specifies which device the
6680 special file refers to; its interpretation depends on the kind of
6681 special file being created.
6683 *** The `fork' function has been renamed to `primitive-fork', to avoid
6684 clashing with various SCSH forks.
6686 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6687 and `recvfrom!'. They no longer accept a size for a second argument;
6688 you must pass a string to hold the received value. They no longer
6689 return the buffer. Instead, `recv' returns the length of the message
6690 received, and `recvfrom' returns a pair containing the packet's length
6691 and originating address.
6693 *** The file descriptor datatype has been removed, as have the
6694 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6695 We plan to replace these functions with a SCSH-compatible interface.
6697 *** The `create' function has been removed; it's just a special case
6700 *** There are new functions to break down process termination status
6701 values. In the descriptions below, STATUS is a value returned by
6704 (status:exit-val STATUS)
6705 If the child process exited normally, this function returns the exit
6706 code for the child process (i.e., the value passed to exit, or
6707 returned from main). If the child process did not exit normally,
6708 this function returns #f.
6710 (status:stop-sig STATUS)
6711 If the child process was suspended by a signal, this function
6712 returns the signal that suspended the child. Otherwise, it returns
6715 (status:term-sig STATUS)
6716 If the child process terminated abnormally, this function returns
6717 the signal that terminated the child. Otherwise, this function
6720 POSIX promises that exactly one of these functions will return true on
6721 a valid STATUS value.
6723 These functions are compatible with SCSH.
6725 *** There are new accessors and setters for the broken-out time vectors
6726 returned by `localtime', `gmtime', and that ilk. They are:
6728 Component Accessor Setter
6729 ========================= ============ ============
6730 seconds tm:sec set-tm:sec
6731 minutes tm:min set-tm:min
6732 hours tm:hour set-tm:hour
6733 day of the month tm:mday set-tm:mday
6734 month tm:mon set-tm:mon
6735 year tm:year set-tm:year
6736 day of the week tm:wday set-tm:wday
6737 day in the year tm:yday set-tm:yday
6738 daylight saving time tm:isdst set-tm:isdst
6739 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6740 name of time zone tm:zone set-tm:zone
6742 *** There are new accessors for the vectors returned by `uname',
6743 describing the host system:
6746 ============================================== ================
6747 name of the operating system implementation utsname:sysname
6748 network name of this machine utsname:nodename
6749 release level of the operating system utsname:release
6750 version level of the operating system utsname:version
6751 machine hardware platform utsname:machine
6753 *** There are new accessors for the vectors returned by `getpw',
6754 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6755 system's user database:
6758 ====================== =================
6759 user name passwd:name
6760 user password passwd:passwd
6763 real name passwd:gecos
6764 home directory passwd:dir
6765 shell program passwd:shell
6767 *** There are new accessors for the vectors returned by `getgr',
6768 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6769 system's group database:
6772 ======================= ============
6773 group name group:name
6774 group password group:passwd
6776 group members group:mem
6778 *** There are new accessors for the vectors returned by `gethost',
6779 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6783 ========================= ===============
6784 official name of host hostent:name
6785 alias list hostent:aliases
6786 host address type hostent:addrtype
6787 length of address hostent:length
6788 list of addresses hostent:addr-list
6790 *** There are new accessors for the vectors returned by `getnet',
6791 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6795 ========================= ===============
6796 official name of net netent:name
6797 alias list netent:aliases
6798 net number type netent:addrtype
6799 net number netent:net
6801 *** There are new accessors for the vectors returned by `getproto',
6802 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6806 ========================= ===============
6807 official protocol name protoent:name
6808 alias list protoent:aliases
6809 protocol number protoent:proto
6811 *** There are new accessors for the vectors returned by `getserv',
6812 `getservbyname', `getservbyport', and `getservent', describing
6816 ========================= ===============
6817 official service name servent:name
6818 alias list servent:aliases
6819 port number servent:port
6820 protocol to use servent:proto
6822 *** There are new accessors for the sockaddr structures returned by
6823 `accept', `getsockname', `getpeername', `recvfrom!':
6826 ======================================== ===============
6827 address format (`family') sockaddr:fam
6828 path, for file domain addresses sockaddr:path
6829 address, for internet domain addresses sockaddr:addr
6830 TCP or UDP port, for internet sockaddr:port
6832 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6833 `getprotoent', and `getservent' functions now return #f at the end of
6834 the user database. (They used to throw an exception.)
6836 Note that calling MUMBLEent function is equivalent to calling the
6837 corresponding MUMBLE function with no arguments.
6839 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6840 `setprotoent', and `setservent' routines now take no arguments.
6842 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6843 provide more useful information when they throw an exception.
6845 *** The `lnaof' function has been renamed to `inet-lnaof'.
6847 *** Guile now claims to have the `current-time' feature.
6849 *** The `mktime' function now takes an optional second argument ZONE,
6850 giving the time zone to use for the conversion. ZONE should be a
6851 string, in the same format as expected for the "TZ" environment variable.
6853 *** The `strptime' function now returns a pair (TIME . COUNT), where
6854 TIME is the parsed time as a vector, and COUNT is the number of
6855 characters from the string left unparsed. This function used to
6856 return the remaining characters as a string.
6858 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6859 The return value is now (SECONDS . MICROSECONDS); the fractional
6860 component is no longer expressed in "ticks".
6862 *** The `ticks/sec' constant has been removed, in light of the above change.
6864 * Changes to the gh_ interface
6866 ** gh_eval_str() now returns an SCM object which is the result of the
6869 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6872 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6873 and returns the array
6875 ** gh_scm2str0() is gone: there is no need to distinguish
6876 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6877 the user to interpret the data both ways.
6879 * Changes to the scm_ interface
6881 ** The new function scm_symbol_value0 provides an easy way to get a
6882 symbol's value from C code:
6884 SCM scm_symbol_value0 (char *NAME)
6885 Return the value of the symbol named by the null-terminated string
6886 NAME in the current module. If the symbol named NAME is unbound in
6887 the current module, return SCM_UNDEFINED.
6889 ** The new function scm_sysintern0 creates new top-level variables,
6890 without assigning them a value.
6892 SCM scm_sysintern0 (char *NAME)
6893 Create a new Scheme top-level variable named NAME. NAME is a
6894 null-terminated string. Return the variable's value cell.
6896 ** The function scm_internal_catch is the guts of catch. It handles
6897 all the mechanics of setting up a catch target, invoking the catch
6898 body, and perhaps invoking the handler if the body does a throw.
6900 The function is designed to be usable from C code, but is general
6901 enough to implement all the semantics Guile Scheme expects from throw.
6903 TAG is the catch tag. Typically, this is a symbol, but this function
6904 doesn't actually care about that.
6906 BODY is a pointer to a C function which runs the body of the catch;
6907 this is the code you can throw from. We call it like this:
6908 BODY (BODY_DATA, JMPBUF)
6910 BODY_DATA is just the BODY_DATA argument we received; we pass it
6911 through to BODY as its first argument. The caller can make
6912 BODY_DATA point to anything useful that BODY might need.
6913 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6914 which we have just created and initialized.
6916 HANDLER is a pointer to a C function to deal with a throw to TAG,
6917 should one occur. We call it like this:
6918 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6920 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6921 same idea as BODY_DATA above.
6922 THROWN_TAG is the tag that the user threw to; usually this is
6923 TAG, but it could be something else if TAG was #t (i.e., a
6924 catch-all), or the user threw to a jmpbuf.
6925 THROW_ARGS is the list of arguments the user passed to the THROW
6928 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6929 is just a pointer we pass through to HANDLER. We don't actually
6930 use either of those pointers otherwise ourselves. The idea is
6931 that, if our caller wants to communicate something to BODY or
6932 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6933 HANDLER can then use. Think of it as a way to make BODY and
6934 HANDLER closures, not just functions; MUMBLE_DATA points to the
6937 Of course, it's up to the caller to make sure that any data a
6938 MUMBLE_DATA needs is protected from GC. A common way to do this is
6939 to make MUMBLE_DATA a pointer to data stored in an automatic
6940 structure variable; since the collector must scan the stack for
6941 references anyway, this assures that any references in MUMBLE_DATA
6944 ** The new function scm_internal_lazy_catch is exactly like
6945 scm_internal_catch, except:
6947 - It does not unwind the stack (this is the major difference).
6948 - If handler returns, its value is returned from the throw.
6949 - BODY always receives #f as its JMPBUF argument (since there's no
6950 jmpbuf associated with a lazy catch, because we don't unwind the
6953 ** scm_body_thunk is a new body function you can pass to
6954 scm_internal_catch if you want the body to be like Scheme's `catch'
6955 --- a thunk, or a function of one argument if the tag is #f.
6957 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6958 contains the Scheme procedure to invoke as the body, and the tag
6959 we're catching. If the tag is #f, then we pass JMPBUF (created by
6960 scm_internal_catch) to the body procedure; otherwise, the body gets
6963 ** scm_handle_by_proc is a new handler function you can pass to
6964 scm_internal_catch if you want the handler to act like Scheme's catch
6965 --- call a procedure with the tag and the throw arguments.
6967 If the user does a throw to this catch, this function runs a handler
6968 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6969 variable holding the Scheme procedure object to invoke. It ought to
6970 be a pointer to an automatic variable (i.e., one living on the stack),
6971 or the procedure object should be otherwise protected from GC.
6973 ** scm_handle_by_message is a new handler function to use with
6974 `scm_internal_catch' if you want Guile to print a message and die.
6975 It's useful for dealing with throws to uncaught keys at the top level.
6977 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6978 message header to print; if zero, we use "guile" instead. That
6979 text is followed by a colon, then the message described by ARGS.
6981 ** The return type of scm_boot_guile is now void; the function does
6982 not return a value, and indeed, never returns at all.
6984 ** The new function scm_shell makes it easy for user applications to
6985 process command-line arguments in a way that is compatible with the
6986 stand-alone guile interpreter (which is in turn compatible with SCSH,
6989 To use the scm_shell function, first initialize any guile modules
6990 linked into your application, and then call scm_shell with the values
6991 of ARGC and ARGV your `main' function received. scm_shell will add
6992 any SCSH-style meta-arguments from the top of the script file to the
6993 argument vector, and then process the command-line arguments. This
6994 generally means loading a script file or starting up an interactive
6995 command interpreter. For details, see "Changes to the stand-alone
6998 ** The new functions scm_get_meta_args and scm_count_argv help you
6999 implement the SCSH-style meta-argument, `\'.
7001 char **scm_get_meta_args (int ARGC, char **ARGV)
7002 If the second element of ARGV is a string consisting of a single
7003 backslash character (i.e. "\\" in Scheme notation), open the file
7004 named by the following argument, parse arguments from it, and return
7005 the spliced command line. The returned array is terminated by a
7008 For details of argument parsing, see above, under "guile now accepts
7009 command-line arguments compatible with SCSH..."
7011 int scm_count_argv (char **ARGV)
7012 Count the arguments in ARGV, assuming it is terminated by a null
7015 For an example of how these functions might be used, see the source
7016 code for the function scm_shell in libguile/script.c.
7018 You will usually want to use scm_shell instead of calling this
7021 ** The new function scm_compile_shell_switches turns an array of
7022 command-line arguments into Scheme code to carry out the actions they
7023 describe. Given ARGC and ARGV, it returns a Scheme expression to
7024 evaluate, and calls scm_set_program_arguments to make any remaining
7025 command-line arguments available to the Scheme code. For example,
7026 given the following arguments:
7028 -e main -s ekko a speckled gecko
7030 scm_set_program_arguments will return the following expression:
7032 (begin (load "ekko") (main (command-line)) (quit))
7034 You will usually want to use scm_shell instead of calling this
7037 ** The function scm_shell_usage prints a usage message appropriate for
7038 an interpreter that uses scm_compile_shell_switches to handle its
7039 command-line arguments.
7041 void scm_shell_usage (int FATAL, char *MESSAGE)
7042 Print a usage message to the standard error output. If MESSAGE is
7043 non-zero, write it before the usage message, followed by a newline.
7044 If FATAL is non-zero, exit the process, using FATAL as the
7045 termination status. (If you want to be compatible with Guile,
7046 always use 1 as the exit status when terminating due to command-line
7049 You will usually want to use scm_shell instead of calling this
7052 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7053 expressions. It used to return SCM_EOL. Earth-shattering.
7055 ** The macros for declaring scheme objects in C code have been
7056 rearranged slightly. They are now:
7058 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7059 Declare a static SCM variable named C_NAME, and initialize it to
7060 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7061 be a C identifier, and SCHEME_NAME should be a C string.
7063 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7064 Just like SCM_SYMBOL, but make C_NAME globally visible.
7066 SCM_VCELL (C_NAME, SCHEME_NAME)
7067 Create a global variable at the Scheme level named SCHEME_NAME.
7068 Declare a static SCM variable named C_NAME, and initialize it to
7069 point to the Scheme variable's value cell.
7071 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7072 Just like SCM_VCELL, but make C_NAME globally visible.
7074 The `guile-snarf' script writes initialization code for these macros
7075 to its standard output, given C source code as input.
7077 The SCM_GLOBAL macro is gone.
7079 ** The scm_read_line and scm_read_line_x functions have been replaced
7080 by Scheme code based on the %read-delimited! procedure (known to C
7081 code as scm_read_delimited_x). See its description above for more
7084 ** The function scm_sys_open has been renamed to scm_open. It now
7085 returns a port instead of an FD object.
7087 * The dynamic linking support has changed. For more information, see
7088 libguile/DYNAMIC-LINKING.
7093 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7096 * Changes to the 'guile' program:
7098 ** Guile now loads some new files when it starts up. Guile first
7099 searches the load path for init.scm, and loads it if found. Then, if
7100 Guile is not being used to execute a script, and the user's home
7101 directory contains a file named `.guile', Guile loads that.
7103 ** You can now use Guile as a shell script interpreter.
7105 To paraphrase the SCSH manual:
7107 When Unix tries to execute an executable file whose first two
7108 characters are the `#!', it treats the file not as machine code to
7109 be directly executed by the native processor, but as source code
7110 to be executed by some interpreter. The interpreter to use is
7111 specified immediately after the #! sequence on the first line of
7112 the source file. The kernel reads in the name of the interpreter,
7113 and executes that instead. It passes the interpreter the source
7114 filename as its first argument, with the original arguments
7115 following. Consult the Unix man page for the `exec' system call
7116 for more information.
7118 Now you can use Guile as an interpreter, using a mechanism which is a
7119 compatible subset of that provided by SCSH.
7121 Guile now recognizes a '-s' command line switch, whose argument is the
7122 name of a file of Scheme code to load. It also treats the two
7123 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7124 to make a file of Scheme code directly executable by Unix, insert the
7125 following two lines at the top of the file:
7127 #!/usr/local/bin/guile -s
7130 Guile treats the argument of the `-s' command-line switch as the name
7131 of a file of Scheme code to load, and treats the sequence `#!' as the
7132 start of a block comment, terminated by `!#'.
7134 For example, here's a version of 'echo' written in Scheme:
7136 #!/usr/local/bin/guile -s
7138 (let loop ((args (cdr (program-arguments))))
7141 (display (car args))
7142 (if (pair? (cdr args))
7144 (loop (cdr args)))))
7147 Why does `#!' start a block comment terminated by `!#', instead of the
7148 end of the line? That is the notation SCSH uses, and although we
7149 don't yet support the other SCSH features that motivate that choice,
7150 we would like to be backward-compatible with any existing Guile
7151 scripts once we do. Furthermore, if the path to Guile on your system
7152 is too long for your kernel, you can start the script with this
7156 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7159 Note that some very old Unix systems don't support the `#!' syntax.
7162 ** You can now run Guile without installing it.
7164 Previous versions of the interactive Guile interpreter (`guile')
7165 couldn't start up unless Guile's Scheme library had been installed;
7166 they used the value of the environment variable `SCHEME_LOAD_PATH'
7167 later on in the startup process, but not to find the startup code
7168 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7171 To run Guile without installing it, build it in the normal way, and
7172 then set the environment variable `SCHEME_LOAD_PATH' to a
7173 colon-separated list of directories, including the top-level directory
7174 of the Guile sources. For example, if you unpacked Guile so that the
7175 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7178 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7181 ** Guile's read-eval-print loop no longer prints #<unspecified>
7182 results. If the user wants to see this, she can evaluate the
7183 expression (assert-repl-print-unspecified #t), perhaps in her startup
7186 ** Guile no longer shows backtraces by default when an error occurs;
7187 however, it does display a message saying how to get one, and how to
7188 request that they be displayed by default. After an error, evaluate
7190 to see a backtrace, and
7191 (debug-enable 'backtrace)
7192 to see them by default.
7196 * Changes to Guile Scheme:
7198 ** Guile now distinguishes between #f and the empty list.
7200 This is for compatibility with the IEEE standard, the (possibly)
7201 upcoming Revised^5 Report on Scheme, and many extant Scheme
7204 Guile used to have #f and '() denote the same object, to make Scheme's
7205 type system more compatible with Emacs Lisp's. However, the change
7206 caused too much trouble for Scheme programmers, and we found another
7207 way to reconcile Emacs Lisp with Scheme that didn't require this.
7210 ** Guile's delq, delv, delete functions, and their destructive
7211 counterparts, delq!, delv!, and delete!, now remove all matching
7212 elements from the list, not just the first. This matches the behavior
7213 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7214 functions which inspired them.
7216 I recognize that this change may break code in subtle ways, but it
7217 seems best to make the change before the FSF's first Guile release,
7221 ** The compiled-library-path function has been deleted from libguile.
7223 ** The facilities for loading Scheme source files have changed.
7225 *** The variable %load-path now tells Guile which directories to search
7226 for Scheme code. Its value is a list of strings, each of which names
7229 *** The variable %load-extensions now tells Guile which extensions to
7230 try appending to a filename when searching the load path. Its value
7231 is a list of strings. Its default value is ("" ".scm").
7233 *** (%search-load-path FILENAME) searches the directories listed in the
7234 value of the %load-path variable for a Scheme file named FILENAME,
7235 with all the extensions listed in %load-extensions. If it finds a
7236 match, then it returns its full filename. If FILENAME is absolute, it
7237 returns it unchanged. Otherwise, it returns #f.
7239 %search-load-path will not return matches that refer to directories.
7241 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7242 uses %seach-load-path to find a file named FILENAME, and loads it if
7243 it finds it. If it can't read FILENAME for any reason, it throws an
7246 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7249 *** load uses the same searching semantics as primitive-load.
7251 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7252 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7253 path, and load-module-with-path have been deleted. The functions
7254 above should serve their purposes.
7256 *** If the value of the variable %load-hook is a procedure,
7257 `primitive-load' applies its value to the name of the file being
7258 loaded (without the load path directory name prepended). If its value
7259 is #f, it is ignored. Otherwise, an error occurs.
7261 This is mostly useful for printing load notification messages.
7264 ** The function `eval!' is no longer accessible from the scheme level.
7265 We can't allow operations which introduce glocs into the scheme level,
7266 because Guile's type system can't handle these as data. Use `eval' or
7267 `read-and-eval!' (see below) as replacement.
7269 ** The new function read-and-eval! reads an expression from PORT,
7270 evaluates it, and returns the result. This is more efficient than
7271 simply calling `read' and `eval', since it is not necessary to make a
7272 copy of the expression for the evaluator to munge.
7274 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7275 for the `read' function.
7278 ** The function `int?' has been removed; its definition was identical
7279 to that of `integer?'.
7281 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7282 use the R4RS names for these functions.
7284 ** The function object-properties no longer returns the hash handle;
7285 it simply returns the object's property list.
7287 ** Many functions have been changed to throw errors, instead of
7288 returning #f on failure. The point of providing exception handling in
7289 the language is to simplify the logic of user code, but this is less
7290 useful if Guile's primitives don't throw exceptions.
7292 ** The function `fileno' has been renamed from `%fileno'.
7294 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7297 * Changes to Guile's C interface:
7299 ** The library's initialization procedure has been simplified.
7300 scm_boot_guile now has the prototype:
7302 void scm_boot_guile (int ARGC,
7304 void (*main_func) (),
7307 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7308 MAIN_FUNC should do all the work of the program (initializing other
7309 packages, reading user input, etc.) before returning. When MAIN_FUNC
7310 returns, call exit (0); this function never returns. If you want some
7311 other exit value, MAIN_FUNC may call exit itself.
7313 scm_boot_guile arranges for program-arguments to return the strings
7314 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7315 scm_set_program_arguments with the final list, so Scheme code will
7316 know which arguments have been processed.
7318 scm_boot_guile establishes a catch-all catch handler which prints an
7319 error message and exits the process. This means that Guile exits in a
7320 coherent way when system errors occur and the user isn't prepared to
7321 handle it. If the user doesn't like this behavior, they can establish
7322 their own universal catcher in MAIN_FUNC to shadow this one.
7324 Why must the caller do all the real work from MAIN_FUNC? The garbage
7325 collector assumes that all local variables of type SCM will be above
7326 scm_boot_guile's stack frame on the stack. If you try to manipulate
7327 SCM values after this function returns, it's the luck of the draw
7328 whether the GC will be able to find the objects you allocate. So,
7329 scm_boot_guile function exits, rather than returning, to discourage
7330 people from making that mistake.
7332 The IN, OUT, and ERR arguments were removed; there are other
7333 convenient ways to override these when desired.
7335 The RESULT argument was deleted; this function should never return.
7337 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7341 ** Guile's header files should no longer conflict with your system's
7344 In order to compile code which #included <libguile.h>, previous
7345 versions of Guile required you to add a directory containing all the
7346 Guile header files to your #include path. This was a problem, since
7347 Guile's header files have names which conflict with many systems'
7350 Now only <libguile.h> need appear in your #include path; you must
7351 refer to all Guile's other header files as <libguile/mumble.h>.
7352 Guile's installation procedure puts libguile.h in $(includedir), and
7353 the rest in $(includedir)/libguile.
7356 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7357 have been added to the Guile library.
7359 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7360 OBJ will not be freed, even if all other references are dropped,
7361 until someone does scm_unprotect_object (OBJ). Both functions
7364 Note that calls to scm_protect_object do not nest. You can call
7365 scm_protect_object any number of times on a given object, and the
7366 next call to scm_unprotect_object will unprotect it completely.
7368 Basically, scm_protect_object and scm_unprotect_object just
7369 maintain a list of references to things. Since the GC knows about
7370 this list, all objects it mentions stay alive. scm_protect_object
7371 adds its argument to the list; scm_unprotect_object remove its
7372 argument from the list.
7375 ** scm_eval_0str now returns the value of the last expression
7378 ** The new function scm_read_0str reads an s-expression from a
7379 null-terminated string, and returns it.
7381 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7382 to a Scheme port object.
7384 ** The new function `scm_set_program_arguments' allows C code to set
7385 the value returned by the Scheme `program-arguments' function.
7390 * Guile no longer includes sophisticated Tcl/Tk support.
7392 The old Tcl/Tk support was unsatisfying to us, because it required the
7393 user to link against the Tcl library, as well as Tk and Guile. The
7394 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7395 referring to widgets by names, rather than exporting widgets to Scheme
7396 code as a special datatype.
7398 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7399 maintainers described some very interesting changes in progress to the
7400 Tcl/Tk internals, which would facilitate clean interfaces between lone
7401 Tk and other interpreters --- even for garbage-collected languages
7402 like Scheme. They expected the new Tk to be publicly available in the
7405 Since it seems that Guile might soon have a new, cleaner interface to
7406 lone Tk, and that the old Guile/Tk glue code would probably need to be
7407 completely rewritten, we (Jim Blandy and Richard Stallman) have
7408 decided not to support the old code. We'll spend the time instead on
7409 a good interface to the newer Tk, as soon as it is available.
7411 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7414 Copyright information:
7416 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7418 Permission is granted to anyone to make or distribute verbatim copies
7419 of this document as received, in any medium, provided that the
7420 copyright notice and this permission notice are preserved,
7421 thus giving the recipient permission to redistribute in turn.
7423 Permission is granted to distribute modified versions
7424 of this document, or of portions of it,
7425 under the above conditions, provided also that they
7426 carry prominent notices stating who last changed them.
7431 paragraph-separate: "[
\f]*$"