1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org. Note that you
6 must be subscribed to this list first, in order to successfully send a
12 * New modules (see the manual for details)
14 ** `(srfi srfi-18)', multithreading support
15 ** The `(ice-9 i18n)' module provides internationalization support
17 * Changes to the distribution
19 ** Guile now uses Gnulib as a portability aid
21 * Changes to the stand-alone interpreter
22 * Changes to Scheme functions and syntax
24 ** A new 'memoize-symbol evaluator trap has been added. This trap can
25 be used for efficiently implementing a Scheme code coverage.
27 ** Duplicate bindings among used modules are resolved lazily.
28 This slightly improves program startup times.
30 ** New thread cancellation and thread cleanup API
31 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
33 * Changes to the C interface
35 ** Functions for handling `scm_option' now no longer require an argument
36 indicating length of the `scm_t_option' array.
40 Changes in 1.8.6 (since 1.8.5)
42 * New features (see the manual for details)
44 ** New convenience function `scm_c_symbol_length ()'
48 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
50 This makes these internal functions technically not callable from
53 ** `guile-config link' now prints `-L$libdir' before `-lguile'
54 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
57 Changes in 1.8.5 (since 1.8.4)
59 * Infrastructure changes
61 ** Guile repository switched from CVS to Git
63 The new repository can be accessed using
64 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
65 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
67 ** Add support for `pkg-config'
69 See "Autoconf Support" in the manual for details.
71 * New modules (see the manual for details)
75 * New features (see the manual for details)
77 ** New `postfix' read option, for SRFI-88 keyword syntax
78 ** Some I/O primitives have been inlined, which improves I/O performance
79 ** New object-based traps infrastructure
81 This is a GOOPS-based infrastructure that builds on Guile's low-level
82 evaluator trap calls and facilitates the development of debugging
83 features like single-stepping, breakpoints, tracing and profiling.
84 See the `Traps' node of the manual for details.
86 ** New support for working on Guile code from within Emacs
88 Guile now incorporates the `GDS' library (previously distributed
89 separately) for working on Guile code from within Emacs. See the
90 `Using Guile In Emacs' node of the manual for details.
94 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
95 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
97 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
98 would trigger an unbound variable error for `match:andmap'.
100 ** `(oop goops describe)' now properly provides the `describe' feature
101 ** Fixed `args-fold' from `(srfi srfi-37)'
103 Previously, parsing short option names of argument-less options would
104 lead to a stack overflow.
106 ** `(srfi srfi-35)' is now visible through `cond-expand'
107 ** Fixed type-checking for the second argument of `eval'
108 ** Fixed type-checking for SRFI-1 `partition'
109 ** Fixed `struct-ref' and `struct-set!' on "light structs"
110 ** Honor struct field access rights in GOOPS
111 ** Changed the storage strategy of source properties, which fixes a deadlock
112 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
113 ** Fixed build issue for GNU/Linux on IA64
114 ** Fixed build issues on NetBSD 1.6
115 ** Fixed build issue on Solaris 2.10 x86_64
116 ** Fixed build issue with DEC/Compaq/HP's compiler
117 ** Fixed `scm_from_complex_double' build issue on FreeBSD
118 ** Fixed `alloca' build issue on FreeBSD 6
119 ** Removed use of non-portable makefile constructs
120 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
121 ** Make sure all tests honor `$TMPDIR'
123 * Changes to the distribution
127 We've started collecting Frequently Asked Questions (FAQ), and will
128 distribute these (with answers!) in future Guile releases.
131 Changes in 1.8.4 (since 1.8.3)
135 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
136 ** Fixed a segmentation fault which occurred when displaying the
137 backtrace of a stack with a promise object (made by `delay') in it.
138 ** Make `accept' leave guile mode while blocking
139 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
140 ** Fixed a build problem on AIX (use of func_data identifier)
141 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
142 called with an associator proc that returns neither a pair nor #f.
143 ** Secondary threads now always return a valid module for (current-module).
144 ** Avoid MacOS build problems caused by incorrect combination of "64"
145 system and library calls.
146 ** `guile-snarf' now honors `$TMPDIR'
147 ** `guile-config compile' now reports CPPFLAGS used at compile-time
148 ** Fixed build with Sun Studio (Solaris 9)
149 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
150 uniform vectors on AIX.
151 ** Fixed a deadlock that occurs upon GC with multiple threads.
152 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
153 ** Fixed autotool-derived build problems on AIX 6.1.
154 ** Fixed NetBSD/alpha support
155 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
157 * New modules (see the manual for details)
161 * Documentation fixes and improvements
163 ** Removed premature breakpoint documentation
165 The features described are not available in the series of 1.8.x
166 releases, so the documentation was misleading and has been removed.
168 ** More about Guile's default *random-state* variable
170 ** GOOPS: more about how to use `next-method'
172 * Changes to the distribution
174 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
176 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
177 General Public License, and the few incorrect files have now been
178 fixed to agree with the rest of the Guile distribution.
180 ** Removed unnecessary extra copies of COPYING*
182 The distribution now contains a single COPYING.LESSER at its top level.
185 Changes in 1.8.3 (since 1.8.2)
187 * New modules (see the manual for details)
194 ** The `(ice-9 slib)' module now works as expected
195 ** Expressions like "(set! 'x #t)" no longer yield a crash
196 ** Warnings about duplicate bindings now go to stderr
197 ** A memory leak in `make-socket-address' was fixed
198 ** Alignment issues (e.g., on SPARC) in network routines were fixed
199 ** A threading issue that showed up at least on NetBSD was fixed
200 ** Build problems on Solaris and IRIX fixed
202 * Implementation improvements
204 ** The reader is now faster, which reduces startup time
205 ** Procedures returned by `record-accessor' and `record-modifier' are faster
209 Changes in 1.8.2 (since 1.8.1):
211 * New procedures (see the manual for details)
213 ** set-program-arguments
218 ** Fractions were not `equal?' if stored in unreduced form.
219 (A subtle problem, since printing a value reduced it, making it work.)
220 ** srfi-60 `copy-bit' failed on 64-bit systems
221 ** "guile --use-srfi" option at the REPL can replace core functions
222 (Programs run with that option were ok, but in the interactive REPL
223 the core bindings got priority, preventing SRFI replacements or
225 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
226 ** `kill' on mingw throws an error for a PID other than oneself
227 ** Procedure names are attached to procedure-with-setters
228 ** Array read syntax works with negative lower bound
229 ** `array-in-bounds?' fix if an array has different lower bounds on each index
230 ** `*' returns exact 0 for "(* inexact 0)"
231 This follows what it always did for "(* 0 inexact)".
232 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
233 ** SRFI-19: `date->julian-day' did not account for timezone offset
234 ** `ttyname' no longer crashes when passed a non-tty argument
235 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
236 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
237 ** GOOPS: Fixed a bug in `method-more-specific?'
238 ** Build problems on Solaris fixed
239 ** Build problems on HP-UX IA64 fixed
240 ** Build problems on MinGW fixed
243 Changes in 1.8.1 (since 1.8.0):
245 * LFS functions are now used to access 64-bit files on 32-bit systems.
247 * New procedures (see the manual for details)
249 ** primitive-_exit - [Scheme] the-root-module
250 ** scm_primitive__exit - [C]
251 ** make-completion-function - [Scheme] (ice-9 readline)
252 ** scm_c_locale_stringn_to_number - [C]
253 ** scm_srfi1_append_reverse [C]
254 ** scm_srfi1_append_reverse_x [C]
260 * New `(ice-9 i18n)' module (see the manual for details)
264 ** Build problems have been fixed on MacOS, SunOS, and QNX.
266 ** `strftime' fix sign of %z timezone offset.
268 ** A one-dimensional array can now be 'equal?' to a vector.
270 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
272 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
274 ** `record-accessor' and `record-modifier' now have strict type checks.
276 Record accessor and modifier procedures now throw an error if the
277 record type of the record they're given is not the type expected.
278 (Previously accessors returned #f and modifiers silently did nothing).
280 ** It is now OK to use both autoload and use-modules on a given module.
282 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
284 Previously there was no checking on primatives like make-vector that
285 accept "one or two" arguments. Now there is.
287 ** The srfi-1 assoc function now calls its equality predicate properly.
289 Previously srfi-1 assoc would call the equality predicate with the key
290 last. According to the SRFI, the key should be first.
292 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
294 ** The array-set! procedure no longer segfaults when given a bit vector.
296 ** Bugs in make-shared-array have been fixed.
298 ** string<? and friends now follow char<? etc order on 8-bit chars.
300 ** The format procedure now handles inf and nan values for ~f correctly.
302 ** exact->inexact should no longer overflow when given certain large fractions.
304 ** srfi-9 accessor and modifier procedures now have strict record type checks.
306 This matches the srfi-9 specification.
308 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
310 Previously the (ice-9 ftw) procedures would ignore any file that had
311 the same inode number as a file they had already seen, even if that
312 file was on a different device.
315 Changes in 1.8.0 (changes since the 1.6.x series):
317 * Changes to the distribution
319 ** Guile is now licensed with the GNU Lesser General Public License.
321 ** The manual is now licensed with the GNU Free Documentation License.
323 ** Guile now requires GNU MP (http://swox.com/gmp).
325 Guile now uses the GNU MP library for arbitrary precision arithmetic.
327 ** Guile now has separate private and public configuration headers.
329 That is, things like HAVE_STRING_H no longer leak from Guile's
332 ** Guile now provides and uses an "effective" version number.
334 Guile now provides scm_effective_version and effective-version
335 functions which return the "effective" version number. This is just
336 the normal full version string without the final micro-version number,
337 so the current effective-version is "1.8". The effective version
338 should remain unchanged during a stable series, and should be used for
339 items like the versioned share directory name
340 i.e. /usr/share/guile/1.8.
342 Providing an unchanging version number during a stable release for
343 things like the versioned share directory can be particularly
344 important for Guile "add-on" packages, since it provides a directory
345 that they can install to that won't be changed out from under them
346 with each micro release during a stable series.
348 ** Thread implementation has changed.
350 When you configure "--with-threads=null", you will get the usual
351 threading API (call-with-new-thread, make-mutex, etc), but you can't
352 actually create new threads. Also, "--with-threads=no" is now
353 equivalent to "--with-threads=null". This means that the thread API
354 is always present, although you might not be able to create new
357 When you configure "--with-threads=pthreads" or "--with-threads=yes",
358 you will get threads that are implemented with the portable POSIX
359 threads. These threads can run concurrently (unlike the previous
360 "coop" thread implementation), but need to cooperate for things like
363 The default is "pthreads", unless your platform doesn't have pthreads,
364 in which case "null" threads are used.
366 See the manual for details, nodes "Initialization", "Multi-Threading",
367 "Blocking", and others.
369 ** There is the new notion of 'discouraged' features.
371 This is a milder form of deprecation.
373 Things that are discouraged should not be used in new code, but it is
374 OK to leave them in old code for now. When a discouraged feature is
375 used, no warning message is printed like there is for 'deprecated'
376 features. Also, things that are merely discouraged are nevertheless
377 implemented efficiently, while deprecated features can be very slow.
379 You can omit discouraged features from libguile by configuring it with
380 the '--disable-discouraged' option.
382 ** Deprecation warnings can be controlled at run-time.
384 (debug-enable 'warn-deprecated) switches them on and (debug-disable
385 'warn-deprecated) switches them off.
387 ** Support for SRFI 61, extended cond syntax for multiple values has
390 This SRFI is always available.
392 ** Support for require-extension, SRFI-55, has been added.
394 The SRFI-55 special form `require-extension' has been added. It is
395 available at startup, and provides a portable way to load Scheme
396 extensions. SRFI-55 only requires support for one type of extension,
397 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
400 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
402 The (srfi srfi-26) module is an implementation of SRFI-26 which
403 provides the `cut' and `cute' syntax. These may be used to specialize
404 parameters without currying.
406 ** New module (srfi srfi-31)
408 This is an implementation of SRFI-31 which provides a special form
409 `rec' for recursive evaluation.
411 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
412 been merged with the core, making their functionality always
415 The modules are still available, tho, and you could use them together
416 with a renaming import, for example.
418 ** Guile no longer includes its own version of libltdl.
420 The official version is good enough now.
422 ** The --enable-htmldoc option has been removed from 'configure'.
424 Support for translating the documentation into HTML is now always
425 provided. Use 'make html'.
427 ** New module (ice-9 serialize):
429 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
430 don't trust the thread safety of most of your program, but where you
431 have some section(s) of code which you consider can run in parallel to
432 other sections. See ice-9/serialize.scm for more information.
434 ** The configure option '--disable-arrays' has been removed.
436 Support for arrays and uniform numeric arrays is now always included
439 * Changes to the stand-alone interpreter
441 ** New command line option `-L'.
443 This option adds a directory to the front of the load path.
445 ** New command line option `--no-debug'.
447 Specifying `--no-debug' on the command line will keep the debugging
448 evaluator turned off, even for interactive sessions.
450 ** User-init file ~/.guile is now loaded with the debugging evaluator.
452 Previously, the normal evaluator would have been used. Using the
453 debugging evaluator gives better error messages.
455 ** The '-e' option now 'read's its argument.
457 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
458 be used with '-e'. For example, you can now write a script like
461 exec guile -e '(@ (demo) main)' -s "$0" "$@"
464 (define-module (demo)
468 (format #t "Demo: ~a~%" args))
471 * Changes to Scheme functions and syntax
473 ** Guardians have changed back to their original semantics
475 Guardians now behave like described in the paper by Dybvig et al. In
476 particular, they no longer make guarantees about the order in which
477 they return objects, and they can no longer be greedy.
479 They no longer drop cyclic data structures.
481 The C function scm_make_guardian has been changed incompatibly and no
482 longer takes the 'greedy_p' argument.
484 ** New function hashx-remove!
486 This function completes the set of 'hashx' functions.
488 ** The concept of dynamic roots has been factored into continuation
489 barriers and dynamic states.
491 Each thread has a current dynamic state that carries the values of the
492 fluids. You can create and copy dynamic states and use them as the
493 second argument for 'eval'. See "Fluids and Dynamic States" in the
496 To restrict the influence that captured continuations can have on the
497 control flow, you can errect continuation barriers. See "Continuation
498 Barriers" in the manual.
500 The function call-with-dynamic-root now essentially temporarily
501 installs a new dynamic state and errects a continuation barrier.
503 ** The default load path no longer includes "." at the end.
505 Automatically loading modules from the current directory should not
506 happen by default. If you want to allow it in a more controlled
507 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
510 ** The uniform vector and array support has been overhauled.
512 It now complies with SRFI-4 and the weird prototype based uniform
513 array creation has been deprecated. See the manual for more details.
515 Some non-compatible changes have been made:
516 - characters can no longer be stored into byte arrays.
517 - strings and bit vectors are no longer considered to be uniform numeric
519 - array-rank throws an error for non-arrays instead of returning zero.
520 - array-ref does no longer accept non-arrays when no indices are given.
522 There is the new notion of 'generalized vectors' and corresponding
523 procedures like 'generalized-vector-ref'. Generalized vectors include
524 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
526 Arrays use generalized vectors as their storage, so that you still
527 have arrays of characters, bits, etc. However, uniform-array-read!
528 and uniform-array-write can no longer read/write strings and
531 ** There is now support for copy-on-write substrings, mutation-sharing
532 substrings and read-only strings.
534 Three new procedures are related to this: substring/shared,
535 substring/copy, and substring/read-only. See the manual for more
538 ** Backtraces will now highlight the value that caused the error.
540 By default, these values are enclosed in "{...}", such as in this
549 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
550 <unnamed port>:1:1: Wrong type (expecting pair): a
551 ABORT: (wrong-type-arg)
553 The prefix and suffix used for highlighting can be set via the two new
554 printer options 'highlight-prefix' and 'highlight-suffix'. For
555 example, putting this into ~/.guile will output the bad value in bold
558 (print-set! highlight-prefix "\x1b[1m")
559 (print-set! highlight-suffix "\x1b[22m")
562 ** 'gettext' support for internationalization has been added.
564 See the manual for details.
566 ** New syntax '@' and '@@':
568 You can now directly refer to variables exported from a module by
571 (@ MODULE-NAME VARIABLE-NAME)
573 For example (@ (ice-9 pretty-print) pretty-print) will directly access
574 the pretty-print variable exported from the (ice-9 pretty-print)
575 module. You don't need to 'use' that module first. You can also use
576 '@' as a target of 'set!', as in (set! (@ mod var) val).
578 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
579 but it can also access variables that have not been exported. It is
580 intended only for kluges and temporary fixes and for debugging, not
583 ** Keyword syntax has been made more disciplined.
585 Previously, the name of a keyword was read as a 'token' but printed as
586 a symbol. Now, it is read as a general Scheme datum which must be a
597 ERROR: In expression (a b c):
601 ERROR: Unbound variable: foo
606 ERROR: Wrong type (expecting symbol): 12
610 ERROR: Wrong type (expecting symbol): (a b c)
614 ** The printing of symbols that might look like keywords can be
617 The new printer option 'quote-keywordish-symbols' controls how symbols
618 are printed that have a colon as their first or last character. The
619 default now is to only quote a symbol with #{...}# when the read
620 option 'keywords' is not '#f'. Thus:
622 guile> (define foo (string->symbol ":foo"))
623 guile> (read-set! keywords #f)
626 guile> (read-set! keywords 'prefix)
629 guile> (print-set! quote-keywordish-symbols #f)
633 ** 'while' now provides 'break' and 'continue'
635 break and continue were previously bound in a while loop, but not
636 documented, and continue didn't quite work properly. The undocumented
637 parameter to break which gave a return value for the while has been
640 ** 'call-with-current-continuation' is now also available under the name
643 ** The module system now checks for duplicate bindings.
645 The module system now can check for name conflicts among imported
648 The behavior can be controlled by specifying one or more 'duplicates'
649 handlers. For example, to make Guile return an error for every name
657 The new default behavior of the module system when a name collision
658 has been detected is to
660 1. Give priority to bindings marked as a replacement.
661 2. Issue a warning (different warning if overriding core binding).
662 3. Give priority to the last encountered binding (this corresponds to
665 If you want the old behavior back without replacements or warnings you
668 (default-duplicate-binding-handler 'last)
670 to your .guile init file.
672 ** New define-module option: :replace
674 :replace works as :export, but, in addition, marks the binding as a
677 A typical example is `format' in (ice-9 format) which is a replacement
678 for the core binding `format'.
680 ** Adding prefixes to imported bindings in the module system
682 There is now a new :use-module option :prefix. It can be used to add
683 a prefix to all imported bindings.
686 :use-module ((bar) :prefix bar:))
688 will import all bindings exported from bar, but rename them by adding
691 ** Conflicting generic functions can be automatically merged.
693 When two imported bindings conflict and they are both generic
694 functions, the two functions can now be merged automatically. This is
695 activated with the 'duplicates' handler 'merge-generics'.
697 ** New function: effective-version
699 Returns the "effective" version number. This is just the normal full
700 version string without the final micro-version number. See "Changes
701 to the distribution" above.
703 ** New threading functions: parallel, letpar, par-map, and friends
705 These are convenient ways to run calculations in parallel in new
706 threads. See "Parallel forms" in the manual for details.
708 ** New function 'try-mutex'.
710 This function will attempt to lock a mutex but will return immediately
711 instead of blocking and indicate failure.
713 ** Waiting on a condition variable can have a timeout.
715 The function 'wait-condition-variable' now takes a third, optional
716 argument that specifies the point in time where the waiting should be
719 ** New function 'broadcast-condition-variable'.
721 ** New functions 'all-threads' and 'current-thread'.
723 ** Signals and system asyncs work better with threads.
725 The function 'sigaction' now takes a fourth, optional, argument that
726 specifies the thread that the handler should run in. When the
727 argument is omitted, the handler will run in the thread that called
730 Likewise, 'system-async-mark' takes a second, optional, argument that
731 specifies the thread that the async should run in. When it is
732 omitted, the async will run in the thread that called
735 C code can use the new functions scm_sigaction_for_thread and
736 scm_system_async_mark_for_thread to pass the new thread argument.
738 When a thread blocks on a mutex, a condition variable or is waiting
739 for IO to be possible, it will still execute system asyncs. This can
740 be used to interrupt such a thread by making it execute a 'throw', for
743 ** The function 'system-async' is deprecated.
745 You can now pass any zero-argument procedure to 'system-async-mark'.
746 The function 'system-async' will just return its argument unchanged
749 ** New functions 'call-with-blocked-asyncs' and
750 'call-with-unblocked-asyncs'
752 The expression (call-with-blocked-asyncs PROC) will call PROC and will
753 block execution of system asyncs for the current thread by one level
754 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
755 procedure and will unblock the execution of system asyncs by one
756 level for the current thread.
758 Only system asyncs are affected by these functions.
760 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
762 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
763 instead. Those functions are easier to use correctly and can be
766 ** New function 'unsetenv'.
768 ** New macro 'define-syntax-public'.
770 It works like 'define-syntax' and also exports the defined macro (but
773 ** There is support for Infinity and NaNs.
775 Following PLT Scheme, Guile can now work with infinite numbers, and
778 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
779 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
780 "+nan.0"). These numbers are inexact and have no exact counterpart.
782 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
783 sign of the dividend. The infinities are integers, and they answer #t
784 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
785 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
796 ERROR: Numerical overflow
798 Two new predicates 'inf?' and 'nan?' can be used to test for the
801 ** Inexact zero can have a sign.
803 Guile can now distinguish between plus and minus inexact zero, if your
804 platform supports this, too. The two zeros are equal according to
805 '=', but not according to 'eqv?'. For example
816 ** Guile now has exact rationals.
818 Guile can now represent fractions such as 1/3 exactly. Computing with
819 them is also done exactly, of course:
824 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
827 For example: (floor 2) now returns an exact 2 where in the past it
828 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
830 ** inexact->exact no longer returns only integers.
832 Without exact rationals, the closest exact number was always an
833 integer, but now inexact->exact returns the fraction that is exactly
834 equal to a floating point number. For example:
836 (inexact->exact 1.234)
837 => 694680242521899/562949953421312
839 When you want the old behavior, use 'round' explicitly:
841 (inexact->exact (round 1.234))
844 ** New function 'rationalize'.
846 This function finds a simple fraction that is close to a given real
847 number. For example (and compare with inexact->exact above):
849 (rationalize (inexact->exact 1.234) 1/2000)
852 Note that, as required by R5RS, rationalize returns only then an exact
853 result when both its arguments are exact.
855 ** 'odd?' and 'even?' work also for inexact integers.
857 Previously, (odd? 1.0) would signal an error since only exact integers
858 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
859 returns #f and (odd? 1.5) signals an error.
861 ** Guile now has uninterned symbols.
863 The new function 'make-symbol' will return an uninterned symbol. This
864 is a symbol that is unique and is guaranteed to remain unique.
865 However, uninterned symbols can not yet be read back in.
867 Use the new function 'symbol-interned?' to check whether a symbol is
870 ** pretty-print has more options.
872 The function pretty-print from the (ice-9 pretty-print) module can now
873 also be invoked with keyword arguments that control things like
874 maximum output width. See the manual for details.
876 ** Variables have no longer a special behavior for `equal?'.
878 Previously, comparing two variables with `equal?' would recursivly
879 compare their values. This is no longer done. Variables are now only
880 `equal?' if they are `eq?'.
882 ** `(begin)' is now valid.
884 You can now use an empty `begin' form. It will yield #<unspecified>
885 when evaluated and simply be ignored in a definition context.
887 ** Deprecated: procedure->macro
889 Change your code to use 'define-macro' or r5rs macros. Also, be aware
890 that macro expansion will not be done during evaluation, but prior to
893 ** Soft ports now allow a `char-ready?' procedure
895 The vector argument to `make-soft-port' can now have a length of
896 either 5 or 6. (Previously the length had to be 5.) The optional 6th
897 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
898 that returns the number of characters that can be read immediately
899 without the soft port blocking.
901 ** Deprecated: undefine
903 There is no replacement for undefine.
905 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
906 have been discouraged.
908 They are relics from a time where a keyword like #:foo was used
909 directly as a Tcl option "-foo" and thus keywords were internally
910 stored as a symbol with a starting dash. We now store a symbol
913 Use symbol->keyword and keyword->symbol instead.
915 ** The `cheap' debug option is now obsolete
917 Evaluator trap calls are now unconditionally "cheap" - in other words,
918 they pass a debug object to the trap handler rather than a full
919 continuation. The trap handler code can capture a full continuation
920 by using `call-with-current-continuation' in the usual way, if it so
923 The `cheap' option is retained for now so as not to break existing
924 code which gets or sets it, but setting it now has no effect. It will
925 be removed in the next major Guile release.
927 ** Evaluator trap calls now support `tweaking'
929 `Tweaking' means that the trap handler code can modify the Scheme
930 expression that is about to be evaluated (in the case of an
931 enter-frame trap) or the value that is being returned (in the case of
932 an exit-frame trap). The trap handler code indicates that it wants to
933 do this by returning a pair whose car is the symbol 'instead and whose
934 cdr is the modified expression or return value.
936 * Changes to the C interface
938 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
939 take a 'delete' function argument.
941 This argument makes no sense since the delete function is used to
942 remove a pair from an alist, and this must not be configurable.
944 This is an incompatible change.
946 ** The GH interface is now subject to the deprecation mechanism
948 The GH interface has been deprecated for quite some time but now it is
949 actually removed from Guile when it is configured with
950 --disable-deprecated.
952 See the manual "Transitioning away from GH" for more information.
954 ** A new family of functions for converting between C values and
955 Scheme values has been added.
957 These functions follow a common naming scheme and are designed to be
958 easier to use, thread-safe and more future-proof than the older
963 These are predicates that return a C boolean: 1 or 0. Instead of
964 SCM_NFALSEP, you can now use scm_is_true, for example.
966 - <type> scm_to_<type> (SCM val, ...)
968 These are functions that convert a Scheme value into an appropriate
969 C value. For example, you can use scm_to_int to safely convert from
972 - SCM scm_from_<type> (<type> val, ...)
974 These functions convert from a C type to a SCM value; for example,
975 scm_from_int for ints.
977 There is a huge number of these functions, for numbers, strings,
978 symbols, vectors, etc. They are documented in the reference manual in
979 the API section together with the types that they apply to.
981 ** New functions for dealing with complex numbers in C have been added.
983 The new functions are scm_c_make_rectangular, scm_c_make_polar,
984 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
985 They work like scm_make_rectangular etc but take or return doubles
988 ** The function scm_make_complex has been discouraged.
990 Use scm_c_make_rectangular instead.
992 ** The INUM macros have been deprecated.
994 A lot of code uses these macros to do general integer conversions,
995 although the macros only work correctly with fixnums. Use the
996 following alternatives.
998 SCM_INUMP -> scm_is_integer or similar
999 SCM_NINUMP -> !scm_is_integer or similar
1000 SCM_MAKINUM -> scm_from_int or similar
1001 SCM_INUM -> scm_to_int or similar
1003 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1004 do the validating for you.
1006 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1007 have been discouraged.
1009 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1010 new code. The functions have been discouraged since they don't fit
1013 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1015 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1016 evaluates its argument twice. Use scm_is_true, etc. instead for new
1019 ** The macro SCM_EQ_P has been discouraged.
1021 Use scm_is_eq for new code, which fits better into the naming
1024 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1027 Use the function scm_is_pair or scm_is_null instead.
1029 ** The functions scm_round and scm_truncate have been deprecated and
1030 are now available as scm_c_round and scm_c_truncate, respectively.
1032 These functions occupy the names that scm_round_number and
1033 scm_truncate_number should have.
1035 ** The functions scm_c_string2str, scm_c_substring2str, and
1036 scm_c_symbol2str have been deprecated.
1038 Use scm_to_locale_stringbuf or similar instead, maybe together with
1041 ** New functions scm_c_make_string, scm_c_string_length,
1042 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1043 scm_c_substring_shared, scm_c_substring_copy.
1045 These are like scm_make_string, scm_length, etc. but are slightly
1046 easier to use from C.
1048 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1049 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1051 They export too many assumptions about the implementation of strings
1052 and symbols that are no longer true in the presence of
1053 mutation-sharing substrings and when Guile switches to some form of
1056 When working with strings, it is often best to use the normal string
1057 functions provided by Guile, such as scm_c_string_ref,
1058 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1059 manual since many more such functions are now provided than
1062 When you want to convert a SCM string to a C string, use the
1063 scm_to_locale_string function or similar instead. For symbols, use
1064 scm_symbol_to_string and then work with that string. Because of the
1065 new string representation, scm_symbol_to_string does not need to copy
1066 and is thus quite efficient.
1068 ** Some string, symbol and keyword functions have been discouraged.
1070 They don't fit into the uniform naming scheme and are not explicit
1071 about the character encoding.
1073 Replace according to the following table:
1075 scm_allocate_string -> scm_c_make_string
1076 scm_take_str -> scm_take_locale_stringn
1077 scm_take0str -> scm_take_locale_string
1078 scm_mem2string -> scm_from_locale_stringn
1079 scm_str2string -> scm_from_locale_string
1080 scm_makfrom0str -> scm_from_locale_string
1081 scm_mem2symbol -> scm_from_locale_symboln
1082 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1083 scm_str2symbol -> scm_from_locale_symbol
1085 SCM_SYMBOL_HASH -> scm_hashq
1086 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1088 scm_c_make_keyword -> scm_from_locale_keyword
1090 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1091 now also available to C code.
1093 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1095 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1096 the latter returns the true name of the keyword, not the 'dash name',
1097 as SCM_KEYWORDSYM used to do.
1099 ** A new way to access arrays in a thread-safe and efficient way has
1102 See the manual, node "Accessing Arrays From C".
1104 ** The old uniform vector and bitvector implementations have been
1105 unceremoniously removed.
1107 This implementation exposed the details of the tagging system of
1108 Guile. Use the new C API explained in the manual in node "Uniform
1109 Numeric Vectors" and "Bit Vectors", respectively.
1111 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1112 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1113 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1114 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1115 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1116 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1119 ** The macros dealing with vectors have been deprecated.
1121 Use the new functions scm_is_vector, scm_vector_elements,
1122 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1123 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1124 manual for more details.
1126 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1127 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1129 The following macros have been removed: SCM_VECTOR_BASE,
1130 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1131 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1133 ** Some C functions and macros related to arrays have been deprecated.
1135 Migrate according to the following table:
1137 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1138 scm_make_ra -> scm_make_array
1139 scm_shap2ra -> scm_make_array
1140 scm_cvref -> scm_c_generalized_vector_ref
1141 scm_ra_set_contp -> do not use
1142 scm_aind -> scm_array_handle_pos
1143 scm_raprin1 -> scm_display or scm_write
1145 SCM_ARRAYP -> scm_is_array
1146 SCM_ARRAY_NDIM -> scm_c_array_rank
1147 SCM_ARRAY_DIMS -> scm_array_handle_dims
1148 SCM_ARRAY_CONTP -> do not use
1149 SCM_ARRAY_MEM -> do not use
1150 SCM_ARRAY_V -> scm_array_handle_elements or similar
1151 SCM_ARRAY_BASE -> do not use
1153 ** SCM_CELL_WORD_LOC has been deprecated.
1155 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1156 to a SCM, as opposed to a pointer to a scm_t_bits.
1158 This was done to allow the correct use of pointers into the Scheme
1159 heap. Previously, the heap words were of type scm_t_bits and local
1160 variables and function arguments were of type SCM, making it
1161 non-standards-conformant to have a pointer that can point to both.
1163 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1165 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1166 second and third words of double smobs. Likewise for
1167 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1169 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1170 used to get and set the 16 exra bits in the zeroth word of a smob.
1172 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1173 accesing the first immediate word of a smob as a SCM value, and there
1174 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1175 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1177 ** New way to deal with non-local exits and re-entries.
1179 There is a new set of functions that essentially do what
1180 scm_internal_dynamic_wind does, but in a way that is more convenient
1181 for C code in some situations. Here is a quick example of how to
1182 prevent a potential memory leak:
1189 scm_dynwind_begin (0);
1191 mem = scm_malloc (100);
1192 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1194 /* MEM would leak if BAR throws an error.
1195 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1202 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1203 SCM_DYNWIND_END as well.
1207 For full documentation, see the node "Dynamic Wind" in the manual.
1209 ** New function scm_dynwind_free
1211 This function calls 'free' on a given pointer when a dynwind context
1212 is left. Thus the call to scm_dynwind_unwind_handler above could be
1213 replaced with simply scm_dynwind_free (mem).
1215 ** New functions scm_c_call_with_blocked_asyncs and
1216 scm_c_call_with_unblocked_asyncs
1218 Like scm_call_with_blocked_asyncs etc. but for C functions.
1220 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1222 In addition to scm_c_call_with_blocked_asyncs you can now also use
1223 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1224 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1226 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1227 SCM_REALLOW_INTS have been deprecated.
1229 They do no longer fulfill their original role of blocking signal
1230 delivery. Depending on what you want to achieve, replace a pair of
1231 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1232 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1235 ** The value 'scm_mask_ints' is no longer writable.
1237 Previously, you could set scm_mask_ints directly. This is no longer
1238 possible. Use scm_c_call_with_blocked_asyncs and
1239 scm_c_call_with_unblocked_asyncs instead.
1241 ** New way to temporarily set the current input, output or error ports
1243 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1244 context' (see above). <foo> is one of "input", "output" or "error".
1246 ** New way to temporarily set fluids
1248 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1249 above) to temporarily set the value of a fluid.
1251 ** New types scm_t_intmax and scm_t_uintmax.
1253 On platforms that have them, these types are identical to intmax_t and
1254 uintmax_t, respectively. On other platforms, they are identical to
1255 the largest integer types that Guile knows about.
1257 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1259 You should not have used them.
1261 ** Many public #defines with generic names have been made private.
1263 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1264 private or renamed with a more suitable public name.
1266 ** The macro SCM_TYP16S has been deprecated.
1268 This macro is not intended for public use.
1270 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1272 Use scm_is_true (scm_inexact_p (...)) instead.
1274 ** The macro SCM_SLOPPY_REALP has been deprecated.
1276 Use scm_is_real instead.
1278 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1280 Use scm_is_complex instead.
1282 ** Some preprocessor defines have been deprecated.
1284 These defines indicated whether a certain feature was present in Guile
1285 or not. Going forward, assume that the features are always present.
1287 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1288 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1290 The following macros have been removed completely: MEMOIZE_LOCALS,
1291 SCM_RECKLESS, SCM_CAUTIOUS.
1293 ** The preprocessor define STACK_DIRECTION has been deprecated.
1295 There should be no need to know about the stack direction for ordinary
1298 ** New function: scm_effective_version
1300 Returns the "effective" version number. This is just the normal full
1301 version string without the final micro-version number. See "Changes
1302 to the distribution" above.
1304 ** The function scm_call_with_new_thread has a new prototype.
1306 Instead of taking a list with the thunk and handler, these two
1307 arguments are now passed directly:
1309 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1311 This is an incompatible change.
1313 ** New snarfer macro SCM_DEFINE_PUBLIC.
1315 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1316 function in the init section.
1318 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1320 ** Garbage collector rewrite.
1322 The garbage collector is cleaned up a lot, and now uses lazy
1323 sweeping. This is reflected in the output of (gc-stats); since cells
1324 are being freed when they are allocated, the cells-allocated field
1325 stays roughly constant.
1327 For malloc related triggers, the behavior is changed. It uses the same
1328 heuristic as the cell-triggered collections. It may be tuned with the
1329 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1330 for minimum yield of malloc related triggers. The default is 40.
1331 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1334 Debugging operations for the freelist have been deprecated, along with
1335 the C variables that control garbage collection. The environment
1336 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1337 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1339 For understanding the memory usage of a GUILE program, the routine
1340 gc-live-object-stats returns an alist containing the number of live
1341 objects for every type.
1344 ** The function scm_definedp has been renamed to scm_defined_p
1346 The name scm_definedp is deprecated.
1348 ** The struct scm_cell type has been renamed to scm_t_cell
1350 This is in accordance to Guile's naming scheme for types. Note that
1351 the name scm_cell is now used for a function that allocates and
1352 initializes a new cell (see below).
1354 ** New functions for memory management
1356 A new set of functions for memory management has been added since the
1357 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1358 indeed, Guile itself contained some long standing bugs that could
1359 cause aborts in long running programs.
1361 The new functions are more symmetrical and do not need cooperation
1362 from smob free routines, among other improvements.
1364 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1365 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1366 scm_gc_free, scm_gc_register_collectable_memory, and
1367 scm_gc_unregister_collectable_memory. Refer to the manual for more
1368 details and for upgrading instructions.
1370 The old functions for memory management have been deprecated. They
1371 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1372 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1374 ** Declarations of exported features are marked with SCM_API.
1376 Every declaration of a feature that belongs to the exported Guile API
1377 has been marked by adding the macro "SCM_API" to the start of the
1378 declaration. This macro can expand into different things, the most
1379 common of which is just "extern" for Unix platforms. On Win32, it can
1380 be used to control which symbols are exported from a DLL.
1382 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1383 will expand into "__declspec (dllimport) extern", which is needed for
1384 linking to the Guile DLL in Windows.
1386 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1387 SCM_SRFI4_IMPORT, for the corresponding libraries.
1389 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1391 Use the new functions scm_cell and scm_double_cell instead. The old
1392 macros had problems because with them allocation and initialization
1393 was separated and the GC could sometimes observe half initialized
1394 cells. Only careful coding by the user of SCM_NEWCELL and
1395 SCM_NEWCELL2 could make this safe and efficient.
1397 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1399 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1402 ** SRCBRKP has been deprecated.
1404 Use scm_c_source_property_breakpoint_p instead.
1406 ** Deprecated: scm_makmacro
1408 Change your code to use either scm_makmmacro or to define macros in
1409 Scheme, using 'define-macro'.
1411 ** New function scm_c_port_for_each.
1413 This function is like scm_port_for_each but takes a pointer to a C
1414 function as the callback instead of a SCM value.
1416 ** The names scm_internal_select, scm_thread_sleep, and
1417 scm_thread_usleep have been discouraged.
1419 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1421 ** The GC can no longer be blocked.
1423 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1424 The GC can now run (partially) concurrently with other code and thus
1425 blocking it is not well defined.
1427 ** Many definitions have been removed that were previously deprecated.
1429 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1430 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1431 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1432 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1433 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1434 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1435 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1436 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1437 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1438 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1439 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1440 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1441 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1442 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1443 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1444 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1445 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1446 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1447 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1448 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1449 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1450 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1451 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1452 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1453 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1454 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1455 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1456 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1457 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1458 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1459 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1461 * Changes to bundled modules
1465 Using the (ice-9 debug) module no longer automatically switches Guile
1466 to use the debugging evaluator. If you want to switch to the
1467 debugging evaluator (which is needed for backtrace information if you
1468 hit an error), please add an explicit "(debug-enable 'debug)" to your
1469 code just after the code to use (ice-9 debug).
1472 Changes since Guile 1.4:
1474 * Changes to the distribution
1476 ** A top-level TODO file is included.
1478 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1480 Guile now always uses three numbers to represent the version,
1481 i.e. "1.6.5". The first number, 1, is the major version number, the
1482 second number, 6, is the minor version number, and the third number,
1483 5, is the micro version number. Changes in major version number
1484 indicate major changes in Guile.
1486 Minor version numbers that are even denote stable releases, and odd
1487 minor version numbers denote development versions (which may be
1488 unstable). The micro version number indicates a minor sub-revision of
1489 a given MAJOR.MINOR release.
1491 In keeping with the new scheme, (minor-version) and scm_minor_version
1492 no longer return everything but the major version number. They now
1493 just return the minor version number. Two new functions
1494 (micro-version) and scm_micro_version have been added to report the
1495 micro version number.
1497 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1499 ** New preprocessor definitions are available for checking versions.
1501 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1502 SCM_MICRO_VERSION to the appropriate integer values.
1504 ** Guile now actively warns about deprecated features.
1506 The new configure option `--enable-deprecated=LEVEL' and the
1507 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1508 See INSTALL and README for more information.
1510 ** Guile is much more likely to work on 64-bit architectures.
1512 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1513 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1514 for the use of a test machine, and thanks to Stefan Jahn for ia64
1517 ** New functions: setitimer and getitimer.
1519 These implement a fairly direct interface to the libc functions of the
1522 ** The #. reader extension is now disabled by default.
1524 For safety reasons, #. evaluation is disabled by default. To
1525 re-enable it, set the fluid read-eval? to #t. For example:
1527 (fluid-set! read-eval? #t)
1529 but make sure you realize the potential security risks involved. With
1530 read-eval? enabled, reading a data file from an untrusted source can
1533 ** New SRFI modules have been added:
1535 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1538 (srfi srfi-1) is a library containing many useful pair- and list-processing
1541 (srfi srfi-2) exports and-let*.
1543 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1545 (srfi srfi-6) is a dummy module for now, since guile already provides
1546 all of the srfi-6 procedures by default: open-input-string,
1547 open-output-string, get-output-string.
1549 (srfi srfi-8) exports receive.
1551 (srfi srfi-9) exports define-record-type.
1553 (srfi srfi-10) exports define-reader-ctor and implements the reader
1556 (srfi srfi-11) exports let-values and let*-values.
1558 (srfi srfi-13) implements the SRFI String Library.
1560 (srfi srfi-14) implements the SRFI Character-Set Library.
1562 (srfi srfi-17) implements setter and getter-with-setter and redefines
1563 some accessor procedures as procedures with getters. (such as car,
1564 cdr, vector-ref etc.)
1566 (srfi srfi-19) implements the SRFI Time/Date Library.
1568 ** New scripts / "executable modules"
1570 Subdirectory "scripts" contains Scheme modules that are packaged to
1571 also be executable as scripts. At this time, these scripts are available:
1580 See README there for more info.
1582 These scripts can be invoked from the shell with the new program
1583 "guile-tools", which keeps track of installation directory for you.
1586 $ guile-tools display-commentary srfi/*.scm
1588 guile-tools is copied to the standard $bindir on "make install".
1590 ** New module (ice-9 stack-catch):
1592 stack-catch is like catch, but saves the current state of the stack in
1593 the fluid the-last-stack. This fluid can be useful when using the
1594 debugger and when re-throwing an error.
1596 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1598 This has been done to prevent problems on lesser operating systems
1599 that can't tolerate `*'s in file names. The exported macro continues
1600 to be named `and-let*', of course.
1602 On systems that support it, there is also a compatibility module named
1603 (ice-9 and-let*). It will go away in the next release.
1605 ** New modules (oop goops) etc.:
1608 (oop goops describe)
1610 (oop goops active-slot)
1611 (oop goops composite-slot)
1613 The Guile Object Oriented Programming System (GOOPS) has been
1614 integrated into Guile. For further information, consult the GOOPS
1615 manual and tutorial in the `doc' directory.
1617 ** New module (ice-9 rdelim).
1619 This exports the following procedures which were previously defined
1620 in the default environment:
1622 read-line read-line! read-delimited read-delimited! %read-delimited!
1623 %read-line write-line
1625 For backwards compatibility the definitions are still imported into the
1626 default environment in this version of Guile. However you should add:
1628 (use-modules (ice-9 rdelim))
1630 to any program which uses the definitions, since this may change in
1633 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1634 can be used for similar functionality.
1636 ** New module (ice-9 rw)
1638 This is a subset of the (scsh rw) module from guile-scsh. Currently
1639 it defines two procedures:
1641 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1643 Read characters from a port or file descriptor into a string STR.
1644 A port must have an underlying file descriptor -- a so-called
1645 fport. This procedure is scsh-compatible and can efficiently read
1648 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1650 Write characters from a string STR to a port or file descriptor.
1651 A port must have an underlying file descriptor -- a so-called
1652 fport. This procedure is mostly compatible and can efficiently
1653 write large strings.
1655 ** New module (ice-9 match)
1657 This module includes Andrew K. Wright's pattern matcher. See
1658 ice-9/match.scm for brief description or
1660 http://www.star-lab.com/wright/code.html
1662 for complete documentation.
1664 ** New module (ice-9 buffered-input)
1666 This module provides procedures to construct an input port from an
1667 underlying source of input that reads and returns its input in chunks.
1668 The underlying input source is a Scheme procedure, specified by the
1669 caller, which the port invokes whenever it needs more input.
1671 This is useful when building an input port whose back end is Readline
1672 or a UI element such as the GtkEntry widget.
1676 The reference and tutorial documentation that was previously
1677 distributed separately, as `guile-doc', is now included in the core
1678 Guile distribution. The documentation consists of the following
1681 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1684 - The Guile Reference Manual (guile.texi) contains (or is intended to
1685 contain) reference documentation on all aspects of Guile.
1687 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1688 reference documentation for using GOOPS, Guile's Object Oriented
1691 - The Revised^5 Report on the Algorithmic Language Scheme
1694 See the README file in the `doc' directory for more details.
1696 ** There are a couple of examples in the examples/ directory now.
1698 * Changes to the stand-alone interpreter
1700 ** New command line option `--use-srfi'
1702 Using this option, SRFI modules can be loaded on startup and be
1703 available right from the beginning. This makes programming portable
1704 Scheme programs easier.
1706 The option `--use-srfi' expects a comma-separated list of numbers,
1707 each representing a SRFI number to be loaded into the interpreter
1708 before starting evaluating a script file or the REPL. Additionally,
1709 the feature identifier for the loaded SRFIs is recognized by
1710 `cond-expand' when using this option.
1713 $ guile --use-srfi=8,13
1714 guile> (receive (x z) (values 1 2) (+ 1 2))
1716 guile> (string-pad "bla" 20)
1719 ** Guile now always starts up in the `(guile-user)' module.
1721 Previously, scripts executed via the `-s' option would run in the
1722 `(guile)' module and the repl would run in the `(guile-user)' module.
1723 Now every user action takes place in the `(guile-user)' module by
1726 * Changes to Scheme functions and syntax
1728 ** Character classifiers work for non-ASCII characters.
1730 The predicates `char-alphabetic?', `char-numeric?',
1731 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1732 no longer check whether their arguments are ASCII characters.
1733 Previously, a character would only be considered alphabetic when it
1734 was also ASCII, for example.
1736 ** Previously deprecated Scheme functions have been removed:
1738 tag - no replacement.
1739 fseek - replaced by seek.
1740 list* - replaced by cons*.
1742 ** It's now possible to create modules with controlled environments
1746 (use-modules (ice-9 safe))
1747 (define m (make-safe-module))
1748 ;;; m will now be a module containing only a safe subset of R5RS
1749 (eval '(+ 1 2) m) --> 3
1750 (eval 'load m) --> ERROR: Unbound variable: load
1752 ** Evaluation of "()", the empty list, is now an error.
1754 Previously, the expression "()" evaluated to the empty list. This has
1755 been changed to signal a "missing expression" error. The correct way
1756 to write the empty list as a literal constant is to use quote: "'()".
1758 ** New concept of `Guile Extensions'.
1760 A Guile Extension is just a ordinary shared library that can be linked
1761 at run-time. We found it advantageous to give this simple concept a
1762 dedicated name to distinguish the issues related to shared libraries
1763 from the issues related to the module system.
1765 *** New function: load-extension
1767 Executing (load-extension lib init) is mostly equivalent to
1769 (dynamic-call init (dynamic-link lib))
1771 except when scm_register_extension has been called previously.
1772 Whenever appropriate, you should use `load-extension' instead of
1773 dynamic-link and dynamic-call.
1775 *** New C function: scm_c_register_extension
1777 This function registers a initialization function for use by
1778 `load-extension'. Use it when you don't want specific extensions to
1779 be loaded as shared libraries (for example on platforms that don't
1780 support dynamic linking).
1782 ** Auto-loading of compiled-code modules is deprecated.
1784 Guile used to be able to automatically find and link a shared
1785 library to satisfy requests for a module. For example, the module
1786 `(foo bar)' could be implemented by placing a shared library named
1787 "foo/libbar.so" (or with a different extension) in a directory on the
1790 This has been found to be too tricky, and is no longer supported. The
1791 shared libraries are now called "extensions". You should now write a
1792 small Scheme file that calls `load-extension' to load the shared
1793 library and initialize it explicitly.
1795 The shared libraries themselves should be installed in the usual
1796 places for shared libraries, with names like "libguile-foo-bar".
1798 For example, place this into a file "foo/bar.scm"
1800 (define-module (foo bar))
1802 (load-extension "libguile-foo-bar" "foobar_init")
1804 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1806 `eval' is now R5RS, that is it takes two arguments.
1807 The second argument is an environment specifier, i.e. either
1809 (scheme-report-environment 5)
1810 (null-environment 5)
1811 (interaction-environment)
1817 ** The module system has been made more disciplined.
1819 The function `eval' will save and restore the current module around
1820 the evaluation of the specified expression. While this expression is
1821 evaluated, `(current-module)' will now return the right module, which
1822 is the module specified as the second argument to `eval'.
1824 A consequence of this change is that `eval' is not particularly
1825 useful when you want allow the evaluated code to change what module is
1826 designated as the current module and have this change persist from one
1827 call to `eval' to the next. The read-eval-print-loop is an example
1828 where `eval' is now inadequate. To compensate, there is a new
1829 function `primitive-eval' that does not take a module specifier and
1830 that does not save/restore the current module. You should use this
1831 function together with `set-current-module', `current-module', etc
1832 when you want to have more control over the state that is carried from
1833 one eval to the next.
1835 Additionally, it has been made sure that forms that are evaluated at
1836 the top level are always evaluated with respect to the current module.
1837 Previously, subforms of top-level forms such as `begin', `case',
1838 etc. did not respect changes to the current module although these
1839 subforms are at the top-level as well.
1841 To prevent strange behavior, the forms `define-module',
1842 `use-modules', `use-syntax', and `export' have been restricted to only
1843 work on the top level. The forms `define-public' and
1844 `defmacro-public' only export the new binding on the top level. They
1845 behave just like `define' and `defmacro', respectively, when they are
1846 used in a lexical environment.
1848 Also, `export' will no longer silently re-export bindings imported
1849 from a used module. It will emit a `deprecation' warning and will
1850 cease to perform any re-export in the next version. If you actually
1851 want to re-export bindings, use the new `re-export' in place of
1852 `export'. The new `re-export' will not make copies of variables when
1853 rexporting them, as `export' did wrongly.
1855 ** Module system now allows selection and renaming of imported bindings
1857 Previously, when using `use-modules' or the `#:use-module' clause in
1858 the `define-module' form, all the bindings (association of symbols to
1859 values) for imported modules were added to the "current module" on an
1860 as-is basis. This has been changed to allow finer control through two
1861 new facilities: selection and renaming.
1863 You can now select which of the imported module's bindings are to be
1864 visible in the current module by using the `:select' clause. This
1865 clause also can be used to rename individual bindings. For example:
1867 ;; import all bindings no questions asked
1868 (use-modules (ice-9 common-list))
1870 ;; import four bindings, renaming two of them;
1871 ;; the current module sees: every some zonk-y zonk-n
1872 (use-modules ((ice-9 common-list)
1874 (remove-if . zonk-y)
1875 (remove-if-not . zonk-n))))
1877 You can also programmatically rename all selected bindings using the
1878 `:renamer' clause, which specifies a proc that takes a symbol and
1879 returns another symbol. Because it is common practice to use a prefix,
1880 we now provide the convenience procedure `symbol-prefix-proc'. For
1883 ;; import four bindings, renaming two of them specifically,
1884 ;; and all four w/ prefix "CL:";
1885 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1886 (use-modules ((ice-9 common-list)
1888 (remove-if . zonk-y)
1889 (remove-if-not . zonk-n))
1890 :renamer (symbol-prefix-proc 'CL:)))
1892 ;; import four bindings, renaming two of them specifically,
1893 ;; and all four by upcasing.
1894 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1895 (define (upcase-symbol sym)
1896 (string->symbol (string-upcase (symbol->string sym))))
1898 (use-modules ((ice-9 common-list)
1900 (remove-if . zonk-y)
1901 (remove-if-not . zonk-n))
1902 :renamer upcase-symbol))
1904 Note that programmatic renaming is done *after* individual renaming.
1905 Also, the above examples show `use-modules', but the same facilities are
1906 available for the `#:use-module' clause of `define-module'.
1908 See manual for more info.
1910 ** The semantics of guardians have changed.
1912 The changes are for the most part compatible. An important criterion
1913 was to keep the typical usage of guardians as simple as before, but to
1914 make the semantics safer and (as a result) more useful.
1916 *** All objects returned from guardians are now properly alive.
1918 It is now guaranteed that any object referenced by an object returned
1919 from a guardian is alive. It's now impossible for a guardian to
1920 return a "contained" object before its "containing" object.
1922 One incompatible (but probably not very important) change resulting
1923 from this is that it is no longer possible to guard objects that
1924 indirectly reference themselves (i.e. are parts of cycles). If you do
1925 so accidentally, you'll get a warning.
1927 *** There are now two types of guardians: greedy and sharing.
1929 If you call (make-guardian #t) or just (make-guardian), you'll get a
1930 greedy guardian, and for (make-guardian #f) a sharing guardian.
1932 Greedy guardians are the default because they are more "defensive".
1933 You can only greedily guard an object once. If you guard an object
1934 more than once, once in a greedy guardian and the rest of times in
1935 sharing guardians, then it is guaranteed that the object won't be
1936 returned from sharing guardians as long as it is greedily guarded
1939 Guardians returned by calls to `make-guardian' can now take one more
1940 optional parameter, which says whether to throw an error in case an
1941 attempt is made to greedily guard an object that is already greedily
1942 guarded. The default is true, i.e. throw an error. If the parameter
1943 is false, the guardian invocation returns #t if guarding was
1944 successful and #f if it wasn't.
1946 Also, since greedy guarding is, in effect, a side-effecting operation
1947 on objects, a new function is introduced: `destroy-guardian!'.
1948 Invoking this function on a guardian renders it unoperative and, if
1949 the guardian is greedy, clears the "greedily guarded" property of the
1950 objects that were guarded by it, thus undoing the side effect.
1952 Note that all this hair is hardly very important, since guardian
1953 objects are usually permanent.
1955 ** Continuations created by call-with-current-continuation now accept
1956 any number of arguments, as required by R5RS.
1958 ** New function `issue-deprecation-warning'
1960 This function is used to display the deprecation messages that are
1961 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1964 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1968 ;; `id' is deprecated. Use `identity' instead.
1973 ** New syntax `begin-deprecated'
1975 When deprecated features are included (as determined by the configure
1976 option --enable-deprecated), `begin-deprecated' is identical to
1977 `begin'. When deprecated features are excluded, it always evaluates
1978 to `#f', ignoring the body forms.
1980 ** New function `make-object-property'
1982 This function returns a new `procedure with setter' P that can be used
1983 to attach a property to objects. When calling P as
1987 where `obj' is any kind of object, it attaches `val' to `obj' in such
1988 a way that it can be retrieved by calling P as
1992 This function will replace procedure properties, symbol properties and
1993 source properties eventually.
1995 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1997 Instead of #&optional, #&key, etc you should now use #:optional,
1998 #:key, etc. Since #:optional is a keyword, you can write it as just
1999 :optional when (read-set! keywords 'prefix) is active.
2001 The old reader syntax `#&' is still supported, but deprecated. It
2002 will be removed in the next release.
2004 ** New define-module option: pure
2006 Tells the module system not to include any bindings from the root
2011 (define-module (totally-empty-module)
2014 ** New define-module option: export NAME1 ...
2016 Export names NAME1 ...
2018 This option is required if you want to be able to export bindings from
2019 a module which doesn't import one of `define-public' or `export'.
2023 (define-module (foo)
2025 :use-module (ice-9 r5rs)
2028 ;;; Note that we're pure R5RS below this point!
2033 ** New function: object->string OBJ
2035 Return a Scheme string obtained by printing a given object.
2037 ** New function: port? X
2039 Returns a boolean indicating whether X is a port. Equivalent to
2040 `(or (input-port? X) (output-port? X))'.
2042 ** New function: file-port?
2044 Determines whether a given object is a port that is related to a file.
2046 ** New function: port-for-each proc
2048 Apply PROC to each port in the Guile port table in turn. The return
2049 value is unspecified. More specifically, PROC is applied exactly once
2050 to every port that exists in the system at the time PORT-FOR-EACH is
2051 invoked. Changes to the port table while PORT-FOR-EACH is running
2052 have no effect as far as PORT-FOR-EACH is concerned.
2054 ** New function: dup2 oldfd newfd
2056 A simple wrapper for the `dup2' system call. Copies the file
2057 descriptor OLDFD to descriptor number NEWFD, replacing the
2058 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2059 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2060 to move away ports which are using NEWFD. The return value is
2063 ** New function: close-fdes fd
2065 A simple wrapper for the `close' system call. Close file
2066 descriptor FD, which must be an integer. Unlike close (*note
2067 close: Ports and File Descriptors.), the file descriptor will be
2068 closed even if a port is using it. The return value is
2071 ** New function: crypt password salt
2073 Encrypts `password' using the standard unix password encryption
2076 ** New function: chroot path
2078 Change the root directory of the running process to `path'.
2080 ** New functions: getlogin, cuserid
2082 Return the login name or the user name of the current effective user
2085 ** New functions: getpriority which who, setpriority which who prio
2087 Get or set the priority of the running process.
2089 ** New function: getpass prompt
2091 Read a password from the terminal, first displaying `prompt' and
2094 ** New function: flock file operation
2096 Set/remove an advisory shared or exclusive lock on `file'.
2098 ** New functions: sethostname name, gethostname
2100 Set or get the hostname of the machine the current process is running
2103 ** New function: mkstemp! tmpl
2105 mkstemp creates a new unique file in the file system and returns a
2106 new buffered port open for reading and writing to the file. TMPL
2107 is a string specifying where the file should be created: it must
2108 end with `XXXXXX' and will be changed in place to return the name
2109 of the temporary file.
2111 ** New function: open-input-string string
2113 Return an input string port which delivers the characters from
2114 `string'. This procedure, together with `open-output-string' and
2115 `get-output-string' implements SRFI-6.
2117 ** New function: open-output-string
2119 Return an output string port which collects all data written to it.
2120 The data can then be retrieved by `get-output-string'.
2122 ** New function: get-output-string
2124 Return the contents of an output string port.
2126 ** New function: identity
2128 Return the argument.
2130 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2131 are represented in Scheme as integers with normal host byte ordering.
2133 ** New function: inet-pton family address
2135 Convert a printable string network address into an integer. Note that
2136 unlike the C version of this function, the result is an integer with
2137 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2140 (inet-pton AF_INET "127.0.0.1") => 2130706433
2141 (inet-pton AF_INET6 "::1") => 1
2143 ** New function: inet-ntop family address
2145 Convert an integer network address into a printable string. Note that
2146 unlike the C version of this function, the input is an integer with
2147 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2150 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2151 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2152 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2156 Use `identity' instead.
2162 ** Deprecated: return-it
2166 ** Deprecated: string-character-length
2168 Use `string-length' instead.
2170 ** Deprecated: flags
2172 Use `logior' instead.
2174 ** Deprecated: close-all-ports-except.
2176 This was intended for closing ports in a child process after a fork,
2177 but it has the undesirable side effect of flushing buffers.
2178 port-for-each is more flexible.
2180 ** The (ice-9 popen) module now attempts to set up file descriptors in
2181 the child process from the current Scheme ports, instead of using the
2182 current values of file descriptors 0, 1, and 2 in the parent process.
2184 ** Removed function: builtin-weak-bindings
2186 There is no such concept as a weak binding any more.
2188 ** Removed constants: bignum-radix, scm-line-incrementors
2190 ** define-method: New syntax mandatory.
2192 The new method syntax is now mandatory:
2194 (define-method (NAME ARG-SPEC ...) BODY ...)
2195 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2197 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2198 REST-ARG ::= ARG-NAME
2200 If you have old code using the old syntax, import
2201 (oop goops old-define-method) before (oop goops) as in:
2203 (use-modules (oop goops old-define-method) (oop goops))
2205 ** Deprecated function: builtin-variable
2206 Removed function: builtin-bindings
2208 There is no longer a distinction between builtin or other variables.
2209 Use module system operations for all variables.
2211 ** Lazy-catch handlers are no longer allowed to return.
2213 That is, a call to `throw', `error', etc is now guaranteed to not
2216 ** Bugfixes for (ice-9 getopt-long)
2218 This module is now tested using test-suite/tests/getopt-long.test.
2219 The following bugs have been fixed:
2221 *** Parsing for options that are specified to have `optional' args now checks
2222 if the next element is an option instead of unconditionally taking it as the
2225 *** An error is now thrown for `--opt=val' when the option description
2226 does not specify `(value #t)' or `(value optional)'. This condition used to
2227 be accepted w/o error, contrary to the documentation.
2229 *** The error message for unrecognized options is now more informative.
2230 It used to be "not a record", an artifact of the implementation.
2232 *** The error message for `--opt' terminating the arg list (no value), when
2233 `(value #t)' is specified, is now more informative. It used to be "not enough
2236 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2237 The expansion used to be like so:
2239 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2241 Note that the "5d" is dropped. Now it is like so:
2243 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2245 This enables single-char options to have adjoining arguments as long as their
2246 constituent characters are not potential single-char options.
2248 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2250 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2251 property `arglist', which can be retrieved by `arity'. The result is that
2252 `arity' can give more detailed information than before:
2256 guile> (use-modules (ice-9 optargs))
2257 guile> (define* (foo #:optional a b c) a)
2259 0 or more arguments in `lambda*:G0'.
2264 3 optional arguments: `a', `b' and `c'.
2265 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2267 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2268 and `d', other keywords allowed.
2269 guile> (define* (baz a b #:optional c #:rest r) a)
2271 2 required arguments: `a' and `b', 1 optional argument: `c',
2274 * Changes to the C interface
2276 ** Types have been renamed from scm_*_t to scm_t_*.
2278 This has been done for POSIX sake. It reserves identifiers ending
2279 with "_t". What a concept.
2281 The old names are still available with status `deprecated'.
2283 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2285 ** Deprecated features have been removed.
2289 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2290 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2292 *** C Functions removed
2294 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2295 scm_fseek - replaced by scm_seek.
2296 gc-thunk - replaced by after-gc-hook.
2297 gh_int2scmb - replaced by gh_bool2scm.
2298 scm_tc_dblr - replaced by scm_tc16_real.
2299 scm_tc_dblc - replaced by scm_tc16_complex.
2300 scm_list_star - replaced by scm_cons_star.
2302 ** Deprecated: scm_makfromstr
2304 Use scm_mem2string instead.
2306 ** Deprecated: scm_make_shared_substring
2308 Explicit shared substrings will disappear from Guile.
2310 Instead, "normal" strings will be implemented using sharing
2311 internally, combined with a copy-on-write strategy.
2313 ** Deprecated: scm_read_only_string_p
2315 The concept of read-only strings will disappear in next release of
2318 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2320 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2322 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2324 Call a procedure with the indicated number of arguments. See "Fly
2325 Evaluation" in the manual.
2327 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2329 Call a procedure with the indicated number of arguments and a list of
2330 further arguments. See "Fly Evaluation" in the manual.
2332 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2334 Create a list of the given number of elements. See "List
2335 Constructors" in the manual.
2337 ** Renamed function: scm_listify has been replaced by scm_list_n.
2339 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2340 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2342 Use functions scm_list_N instead.
2344 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2346 Used by an application to read arbitrary number of bytes from a port.
2347 Same semantics as libc read, except that scm_c_read only returns less
2348 than SIZE bytes if at end-of-file.
2350 Warning: Doesn't update port line and column counts!
2352 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2354 Used by an application to write arbitrary number of bytes to an SCM
2355 port. Similar semantics as libc write. However, unlike libc
2356 write, scm_c_write writes the requested number of bytes and has no
2359 Warning: Doesn't update port line and column counts!
2361 ** New function: scm_init_guile ()
2363 In contrast to scm_boot_guile, scm_init_guile will return normally
2364 after initializing Guile. It is not available on all systems, tho.
2366 ** New functions: scm_str2symbol, scm_mem2symbol
2368 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2369 field of characters and creates a scheme symbol object from that C string.
2370 The function scm_mem2symbol takes a const char* and a number of characters and
2371 creates a symbol from the characters in that memory area.
2373 ** New functions: scm_primitive_make_property
2374 scm_primitive_property_ref
2375 scm_primitive_property_set_x
2376 scm_primitive_property_del_x
2378 These functions implement a new way to deal with object properties.
2379 See libguile/properties.c for their documentation.
2381 ** New function: scm_done_free (long size)
2383 This function is the inverse of scm_done_malloc. Use it to report the
2384 amount of smob memory you free. The previous method, which involved
2385 calling scm_done_malloc with negative argument, was somewhat
2386 unintuitive (and is still available, of course).
2388 ** New function: scm_c_memq (SCM obj, SCM list)
2390 This function provides a fast C level alternative for scm_memq for the case
2391 that the list parameter is known to be a proper list. The function is a
2392 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2393 list input parameter, since for anything else but a proper list the function's
2394 behaviour is undefined - it may even crash or loop endlessly. Further, for
2395 the case that the object is not found in the list, scm_c_memq returns #f which
2396 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2398 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2399 scm_remember_upto_here
2401 These functions replace the function scm_remember.
2403 ** Deprecated function: scm_remember
2405 Use one of the new functions scm_remember_upto_here_1,
2406 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2408 ** New function: scm_allocate_string
2410 This function replaces the function scm_makstr.
2412 ** Deprecated function: scm_makstr
2414 Use the new function scm_allocate_string instead.
2416 ** New global variable scm_gc_running_p introduced.
2418 Use this variable to find out if garbage collection is being executed. Up to
2419 now applications have used scm_gc_heap_lock to test if garbage collection was
2420 running, which also works because of the fact that up to know only the garbage
2421 collector has set this variable. But, this is an implementation detail that
2422 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2423 of this variable is (and has been) not fully safe anyway.
2425 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2427 Use these instead of SCM_LENGTH_MAX.
2429 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2430 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2431 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2433 Use these instead of SCM_LENGTH.
2435 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2436 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2437 SCM_SET_BITVECTOR_LENGTH
2439 Use these instead of SCM_SETLENGTH
2441 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2442 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2445 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2448 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2449 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2452 Use these instead of SCM_SETCHARS.
2454 ** New macro: SCM_BITVECTOR_P
2456 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2458 Use instead of SCM_COERCE_SUBSTR.
2460 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2462 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2464 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2465 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2466 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2467 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2468 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2469 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2470 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2471 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2472 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2473 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2474 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2475 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2476 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2477 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2478 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2480 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2481 Use scm_memory_error instead of SCM_NALLOC.
2482 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2483 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2484 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2485 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2486 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2487 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2488 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2489 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2490 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2491 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2492 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2493 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2494 Use a type specific setter macro instead of SCM_SETLENGTH.
2495 Use a type specific setter macro instead of SCM_SETCHARS.
2496 Use a type specific length macro instead of SCM_LENGTH_MAX.
2497 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2498 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2499 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2500 Use SCM_TYP16 instead of SCM_GCTYP16.
2501 Use SCM_CDR instead of SCM_GCCDR.
2502 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2503 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2504 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2505 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2506 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2507 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2508 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2510 ** Removed function: scm_struct_init
2512 ** Removed variable: scm_symhash_dim
2514 ** Renamed function: scm_make_cont has been replaced by
2515 scm_make_continuation, which has a different interface.
2517 ** Deprecated function: scm_call_catching_errors
2519 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2521 ** Deprecated function: scm_strhash
2523 Use scm_string_hash instead.
2525 ** Deprecated function: scm_vector_set_length_x
2527 Instead, create a fresh vector of the desired size and copy the contents.
2529 ** scm_gensym has changed prototype
2531 scm_gensym now only takes one argument.
2533 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2536 There is now only a single symbol type scm_tc7_symbol.
2537 The tag scm_tc7_lvector was not used anyway.
2539 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2541 Use scm_make_smob_type and scm_set_smob_XXX instead.
2543 ** New function scm_set_smob_apply.
2545 This can be used to set an apply function to a smob type.
2547 ** Deprecated function: scm_strprint_obj
2549 Use scm_object_to_string instead.
2551 ** Deprecated function: scm_wta
2553 Use scm_wrong_type_arg, or another appropriate error signalling function
2556 ** Explicit support for obarrays has been deprecated.
2558 Use `scm_str2symbol' and the generic hashtable functions instead.
2560 ** The concept of `vcells' has been deprecated.
2562 The data type `variable' is now used exclusively. `Vcells' have been
2563 a low-level concept so you are likely not affected by this change.
2565 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2566 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2568 Use scm_c_define or scm_c_lookup instead, as appropriate.
2570 *** New functions: scm_c_module_lookup, scm_c_lookup,
2571 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2572 scm_module_define, scm_define.
2574 These functions work with variables instead of with vcells.
2576 ** New functions for creating and defining `subr's and `gsubr's.
2578 The new functions more clearly distinguish between creating a subr (or
2579 gsubr) object and adding it to the current module.
2581 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2582 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2583 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2584 scm_c_define_gsubr_with_generic.
2586 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2587 scm_make_subr_with_generic, scm_make_gsubr,
2588 scm_make_gsubr_with_generic.
2590 Use the new ones from above instead.
2592 ** C interface to the module system has changed.
2594 While we suggest that you avoid as many explicit module system
2595 operations from C as possible for the time being, the C interface has
2596 been made more similar to the high-level Scheme module system.
2598 *** New functions: scm_c_define_module, scm_c_use_module,
2599 scm_c_export, scm_c_resolve_module.
2601 They mostly work like their Scheme namesakes. scm_c_define_module
2602 takes a function that is called a context where the new module is
2605 *** Deprecated functions: scm_the_root_module, scm_make_module,
2606 scm_ensure_user_module, scm_load_scheme_module.
2608 Use the new functions instead.
2610 ** Renamed function: scm_internal_with_fluids becomes
2613 scm_internal_with_fluids is available as a deprecated function.
2615 ** New function: scm_c_with_fluid.
2617 Just like scm_c_with_fluids, but takes one fluid and one value instead
2620 ** Deprecated typedefs: long_long, ulong_long.
2622 They are of questionable utility and they pollute the global
2625 ** Deprecated typedef: scm_sizet
2627 It is of questionable utility now that Guile requires ANSI C, and is
2630 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2631 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2632 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2634 Made more compliant with the naming policy by adding a _t at the end.
2636 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2637 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2639 With the exception of the mysterious scm_2ulong2big, they are still
2640 available under new names (scm_i_mkbig etc). These functions are not
2641 intended to be used in user code. You should avoid dealing with
2642 bignums directly, and should deal with numbers in general (which can
2645 ** Change in behavior: scm_num2long, scm_num2ulong
2647 The scm_num2[u]long functions don't any longer accept an inexact
2648 argument. This change in behavior is motivated by concordance with
2649 R5RS: It is more common that a primitive doesn't want to accept an
2650 inexact for an exact.
2652 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2653 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2654 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2657 These are conversion functions between the various ANSI C integral
2658 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2659 accept an inexact argument.
2661 ** New functions: scm_float2num, scm_double2num,
2662 scm_num2float, scm_num2double.
2664 These are conversion functions between the two ANSI C float types and
2667 ** New number validation macros:
2668 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2672 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2674 These are just nicer-named old scm_protect_object and
2675 scm_unprotect_object.
2677 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2679 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2681 These functions can be used to register pointers to locations that
2684 ** Deprecated function: scm_create_hook.
2686 Its sins are: misleading name, non-modularity and lack of general
2690 Changes since Guile 1.3.4:
2692 * Changes to the distribution
2694 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2696 We've changed the way we handle generated files in the Guile source
2697 repository. As a result, the procedure for building trees obtained
2698 from the nightly FTP snapshots or via CVS has changed:
2699 - You must have appropriate versions of autoconf, automake, and
2700 libtool installed on your system. See README for info on how to
2701 obtain these programs.
2702 - Before configuring the tree, you must first run the script
2703 `autogen.sh' at the top of the source tree.
2705 The Guile repository used to contain not only source files, written by
2706 humans, but also some generated files, like configure scripts and
2707 Makefile.in files. Even though the contents of these files could be
2708 derived mechanically from other files present, we thought it would
2709 make the tree easier to build if we checked them into CVS.
2711 However, this approach means that minor differences between
2712 developer's installed tools and habits affected the whole team.
2713 So we have removed the generated files from the repository, and
2714 added the autogen.sh script, which will reconstruct them
2718 ** configure now has experimental options to remove support for certain
2721 --disable-arrays omit array and uniform array support
2722 --disable-posix omit posix interfaces
2723 --disable-networking omit networking interfaces
2724 --disable-regex omit regular expression interfaces
2726 These are likely to become separate modules some day.
2728 ** New configure option --enable-debug-freelist
2730 This enables a debugging version of SCM_NEWCELL(), and also registers
2731 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2733 Configure with the --enable-debug-freelist option to enable
2734 the gc-set-debug-check-freelist! primitive, and then use:
2736 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2737 (gc-set-debug-check-freelist! #f) # turn off checking
2739 Checking of the freelist forces a traversal of the freelist and
2740 a garbage collection before each allocation of a cell. This can
2741 slow down the interpreter dramatically, so the setter should be used to
2742 turn on this extra processing only when necessary.
2744 ** New configure option --enable-debug-malloc
2746 Include code for debugging of calls to scm_must_malloc/realloc/free.
2750 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2751 2. objects reallocated by scm_must_realloc has been allocated by
2753 3. reallocated objects are reallocated with the same what string
2755 But, most importantly, it records the number of allocated objects of
2756 each kind. This is useful when searching for memory leaks.
2758 A Guile compiled with this option provides the primitive
2759 `malloc-stats' which returns an alist with pairs of kind and the
2760 number of objects of that kind.
2762 ** All includes are now referenced relative to the root directory
2764 Since some users have had problems with mixups between Guile and
2765 system headers, we have decided to always refer to Guile headers via
2766 their parent directories. This essentially creates a "private name
2767 space" for Guile headers. This means that the compiler only is given
2768 -I options for the root build and root source directory.
2770 ** Header files kw.h and genio.h have been removed.
2772 ** The module (ice-9 getopt-gnu-style) has been removed.
2774 ** New module (ice-9 documentation)
2776 Implements the interface to documentation strings associated with
2779 ** New module (ice-9 time)
2781 Provides a macro `time', which displays execution time of a given form.
2783 ** New module (ice-9 history)
2785 Loading this module enables value history in the repl.
2787 * Changes to the stand-alone interpreter
2789 ** New command line option --debug
2791 Start Guile with debugging evaluator and backtraces enabled.
2793 This is useful when debugging your .guile init file or scripts.
2795 ** New help facility
2797 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2798 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2799 (help 'NAME) gives documentation for NAME, even if it is not an object
2800 (help ,EXPR) gives documentation for object returned by EXPR
2801 (help (my module)) gives module commentary for `(my module)'
2802 (help) gives this text
2804 `help' searches among bindings exported from loaded modules, while
2805 `apropos' searches among bindings visible from the "current" module.
2807 Examples: (help help)
2809 (help "output-string")
2811 ** `help' and `apropos' now prints full module names
2813 ** Dynamic linking now uses libltdl from the libtool package.
2815 The old system dependent code for doing dynamic linking has been
2816 replaced with calls to the libltdl functions which do all the hairy
2819 The major improvement is that you can now directly pass libtool
2820 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2821 will be able to do the best shared library job you can get, via
2824 The way dynamic libraries are found has changed and is not really
2825 portable across platforms, probably. It is therefore recommended to
2826 use absolute filenames when possible.
2828 If you pass a filename without an extension to `dynamic-link', it will
2829 try a few appropriate ones. Thus, the most platform ignorant way is
2830 to specify a name like "libfoo", without any directories and
2833 ** Guile COOP threads are now compatible with LinuxThreads
2835 Previously, COOP threading wasn't possible in applications linked with
2836 Linux POSIX threads due to their use of the stack pointer to find the
2837 thread context. This has now been fixed with a workaround which uses
2838 the pthreads to allocate the stack.
2840 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2842 ** Positions of erring expression in scripts
2844 With version 1.3.4, the location of the erring expression in Guile
2845 scipts is no longer automatically reported. (This should have been
2846 documented before the 1.3.4 release.)
2848 You can get this information by enabling recording of positions of
2849 source expressions and running the debugging evaluator. Put this at
2850 the top of your script (or in your "site" file):
2852 (read-enable 'positions)
2853 (debug-enable 'debug)
2855 ** Backtraces in scripts
2857 It is now possible to get backtraces in scripts.
2861 (debug-enable 'debug 'backtrace)
2863 at the top of the script.
2865 (The first options enables the debugging evaluator.
2866 The second enables backtraces.)
2868 ** Part of module system symbol lookup now implemented in C
2870 The eval closure of most modules is now implemented in C. Since this
2871 was one of the bottlenecks for loading speed, Guile now loads code
2872 substantially faster than before.
2874 ** Attempting to get the value of an unbound variable now produces
2875 an exception with a key of 'unbound-variable instead of 'misc-error.
2877 ** The initial default output port is now unbuffered if it's using a
2878 tty device. Previously in this situation it was line-buffered.
2880 ** New hook: after-gc-hook
2882 after-gc-hook takes over the role of gc-thunk. This hook is run at
2883 the first SCM_TICK after a GC. (Thus, the code is run at the same
2884 point during evaluation as signal handlers.)
2886 Note that this hook should be used only for diagnostic and debugging
2887 purposes. It is not certain that it will continue to be well-defined
2888 when this hook is run in the future.
2890 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2891 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2893 ** Improvements to garbage collector
2895 Guile 1.4 has a new policy for triggering heap allocation and
2896 determining the sizes of heap segments. It fixes a number of problems
2899 1. The new policy can handle two separate pools of cells
2900 (2-word/4-word) better. (The old policy would run wild, allocating
2901 more and more memory for certain programs.)
2903 2. The old code would sometimes allocate far too much heap so that the
2904 Guile process became gigantic. The new code avoids this.
2906 3. The old code would sometimes allocate too little so that few cells
2907 were freed at GC so that, in turn, too much time was spent in GC.
2909 4. The old code would often trigger heap allocation several times in a
2910 row. (The new scheme predicts how large the segments needs to be
2911 in order not to need further allocation.)
2913 All in all, the new GC policy will make larger applications more
2916 The new GC scheme also is prepared for POSIX threading. Threads can
2917 allocate private pools of cells ("clusters") with just a single
2918 function call. Allocation of single cells from such a cluster can
2919 then proceed without any need of inter-thread synchronization.
2921 ** New environment variables controlling GC parameters
2923 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2926 Allocation of 2-word cell heaps:
2928 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2931 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2932 GC in percent of total heap size
2935 Allocation of 4-word cell heaps
2936 (used for real numbers and misc other objects):
2938 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2940 (See entry "Way for application to customize GC parameters" under
2941 section "Changes to the scm_ interface" below.)
2943 ** Guile now implements reals using 4-word cells
2945 This speeds up computation with reals. (They were earlier allocated
2946 with `malloc'.) There is still some room for optimizations, however.
2948 ** Some further steps toward POSIX thread support have been taken
2950 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2951 don't have much effect any longer, and many of them will be removed in
2955 are only handled at the top of the evaluator loop, immediately after
2956 I/O, and in scm_equalp.
2958 *** The GC can allocate thread private pools of pairs.
2960 * Changes to Scheme functions and syntax
2962 ** close-input-port and close-output-port are now R5RS
2964 These procedures have been turned into primitives and have R5RS behaviour.
2966 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2968 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2969 extended by the more sophisticated version in (ice-9 format)
2971 (simple-format port message . args)
2972 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2973 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2974 the escapes are replaced with corresponding members of ARGS:
2975 ~A formats using `display' and ~S formats using `write'.
2976 If DESTINATION is #t, then use the `current-output-port',
2977 if DESTINATION is #f, then return a string containing the formatted text.
2978 Does not add a trailing newline."
2980 ** string-ref: the second argument is no longer optional.
2982 ** string, list->string: no longer accept strings in their arguments,
2983 only characters, for compatibility with R5RS.
2985 ** New procedure: port-closed? PORT
2986 Returns #t if PORT is closed or #f if it is open.
2988 ** Deprecated: list*
2990 The list* functionality is now provided by cons* (SRFI-1 compliant)
2992 ** New procedure: cons* ARG1 ARG2 ... ARGn
2994 Like `list', but the last arg provides the tail of the constructed list,
2995 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2997 Requires at least one argument. If given one argument, that argument
2998 is returned as result.
3000 This function is called `list*' in some other Schemes and in Common LISP.
3002 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3004 ** New procedure: object-documentation OBJECT
3006 Returns the documentation string associated with OBJECT. The
3007 procedure uses a caching mechanism so that subsequent lookups are
3010 Exported by (ice-9 documentation).
3012 ** module-name now returns full names of modules
3014 Previously, only the last part of the name was returned (`session' for
3015 `(ice-9 session)'). Ex: `(ice-9 session)'.
3017 * Changes to the gh_ interface
3019 ** Deprecated: gh_int2scmb
3021 Use gh_bool2scm instead.
3023 * Changes to the scm_ interface
3025 ** Guile primitives now carry docstrings!
3027 Thanks to Greg Badros!
3029 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3031 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3032 macros and must contain a docstring that is extracted into foo.doc using a new
3033 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3035 However, a major overhaul of these macros is scheduled for the next release of
3038 ** Guile primitives use a new technique for validation of arguments
3040 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3041 the readability of argument checking.
3043 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3045 ** New macros: SCM_PACK, SCM_UNPACK
3047 Compose/decompose an SCM value.
3049 The SCM type is now treated as an abstract data type and may be defined as a
3050 long, a void* or as a struct, depending on the architecture and compile time
3051 options. This makes it easier to find several types of bugs, for example when
3052 SCM values are treated as integers without conversion. Values of the SCM type
3053 should be treated as "atomic" values. These macros are used when
3054 composing/decomposing an SCM value, either because you want to access
3055 individual bits, or because you want to treat it as an integer value.
3057 E.g., in order to set bit 7 in an SCM value x, use the expression
3059 SCM_PACK (SCM_UNPACK (x) | 0x80)
3061 ** The name property of hooks is deprecated.
3062 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3064 You can emulate this feature by using object properties.
3066 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3067 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3070 These macros will be removed in a future release of Guile.
3072 ** The following types, functions and macros from numbers.h are deprecated:
3073 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3074 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3076 ** Port internals: the rw_random variable in the scm_port structure
3077 must be set to non-zero in any random access port. In recent Guile
3078 releases it was only set for bidirectional random-access ports.
3080 ** Port internals: the seek ptob procedure is now responsible for
3081 resetting the buffers if required. The change was made so that in the
3082 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3083 the fport and strport ptobs can avoid resetting the buffers,
3084 in particular to avoid discarding unread chars. An existing port
3085 type can be fixed by adding something like the following to the
3086 beginning of the ptob seek procedure:
3088 if (pt->rw_active == SCM_PORT_READ)
3089 scm_end_input (object);
3090 else if (pt->rw_active == SCM_PORT_WRITE)
3091 ptob->flush (object);
3093 although to actually avoid resetting the buffers and discard unread
3094 chars requires further hacking that depends on the characteristics
3097 ** Deprecated functions: scm_fseek, scm_tag
3099 These functions are no longer used and will be removed in a future version.
3101 ** The scm_sysmissing procedure is no longer used in libguile.
3102 Unless it turns out to be unexpectedly useful to somebody, it will be
3103 removed in a future version.
3105 ** The format of error message strings has changed
3107 The two C procedures: scm_display_error and scm_error, as well as the
3108 primitive `scm-error', now use scm_simple_format to do their work.
3109 This means that the message strings of all code must be updated to use
3110 ~A where %s was used before, and ~S where %S was used before.
3112 During the period when there still are a lot of old Guiles out there,
3113 you might want to support both old and new versions of Guile.
3115 There are basically two methods to achieve this. Both methods use
3118 AC_CHECK_FUNCS(scm_simple_format)
3120 in your configure.in.
3122 Method 1: Use the string concatenation features of ANSI C's
3127 #ifdef HAVE_SCM_SIMPLE_FORMAT
3133 Then represent each of your error messages using a preprocessor macro:
3135 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3139 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3140 (define make-message string-append)
3142 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3144 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3148 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3153 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3157 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3159 Don't use the functions coop_mutex_init and
3160 coop_condition_variable_init. They will change.
3162 Use scm_mutex_init and scm_cond_init instead.
3164 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3165 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3166 COND, as `scm_cond_wait' does, but it also bounds the duration
3167 of the wait. If COND has not been signaled before time ABSTIME,
3168 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3169 returns the error code `ETIMEDOUT'.
3171 The ABSTIME parameter specifies an absolute time, with the same
3172 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3173 to 00:00:00 GMT, January 1, 1970.
3175 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3176 `scm_cond_broadcast' restarts all the threads that are waiting
3177 on the condition variable COND. Nothing happens if no threads are
3180 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3181 `scm_key_create' allocates a new TSD key. The key is stored in
3182 the location pointed to by KEY. There is no limit on the number
3183 of keys allocated at a given time. The value initially associated
3184 with the returned key is `NULL' in all currently executing threads.
3186 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3187 function associated with the key. When a thread terminates,
3188 DESTR_FUNCTION is called on the value associated with the key in
3189 that thread. The DESTR_FUNCTION is not called if a key is deleted
3190 with `scm_key_delete' or a value is changed with
3191 `scm_setspecific'. The order in which destructor functions are
3192 called at thread termination time is unspecified.
3194 Destructors are not yet implemented.
3196 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3197 `scm_setspecific' changes the value associated with KEY in the
3198 calling thread, storing the given POINTER instead.
3200 ** New function: scm_getspecific (scm_key_t KEY)
3201 `scm_getspecific' returns the value currently associated with
3202 KEY in the calling thread.
3204 ** New function: scm_key_delete (scm_key_t KEY)
3205 `scm_key_delete' deallocates a TSD key. It does not check
3206 whether non-`NULL' values are associated with that key in the
3207 currently executing threads, nor call the destructor function
3208 associated with the key.
3210 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3212 Initialize a C level hook HOOK with associated HOOK_DATA and type
3213 TYPE. (See scm_c_hook_run ().)
3215 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3217 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3218 is true, add it last, otherwise first. The same FUNC can be added
3219 multiple times if FUNC_DATA differ and vice versa.
3221 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3223 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3224 function is only removed if both FUNC and FUNC_DATA matches.
3226 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3228 Run hook HOOK passing DATA to the hook functions.
3230 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3231 returned is undefined.
3233 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3234 returns a non-NULL value. This value is returned as the result of
3235 scm_c_hook_run. If all functions return NULL, NULL is returned.
3237 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3238 returns a NULL value, and NULL is returned. If all functions returns
3239 a non-NULL value, the last value is returned.
3241 ** New C level GC hooks
3243 Five new C level hooks has been added to the garbage collector.
3245 scm_before_gc_c_hook
3248 are run before locking and after unlocking the heap. The system is
3249 thus in a mode where evaluation can take place. (Except that
3250 scm_before_gc_c_hook must not allocate new cells.)
3252 scm_before_mark_c_hook
3253 scm_before_sweep_c_hook
3254 scm_after_sweep_c_hook
3256 are run when the heap is locked. These are intended for extension of
3257 the GC in a modular fashion. Examples are the weaks and guardians
3260 ** Way for application to customize GC parameters
3262 The application can set up other default values for the GC heap
3263 allocation parameters
3265 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3266 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3267 GUILE_MAX_SEGMENT_SIZE,
3271 scm_default_init_heap_size_1, scm_default_min_yield_1,
3272 scm_default_init_heap_size_2, scm_default_min_yield_2,
3273 scm_default_max_segment_size
3275 respectively before callong scm_boot_guile.
3277 (See entry "New environment variables ..." in section
3278 "Changes to the stand-alone interpreter" above.)
3280 ** scm_protect_object/scm_unprotect_object now nest
3282 This means that you can call scm_protect_object multiple times on an
3283 object and count on the object being protected until
3284 scm_unprotect_object has been call the same number of times.
3286 The functions also have better time complexity.
3288 Still, it is usually possible to structure the application in a way
3289 that you don't need to use these functions. For example, if you use a
3290 protected standard Guile list to keep track of live objects rather
3291 than some custom data type, objects will die a natural death when they
3292 are no longer needed.
3294 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3296 Guile does not provide the float representation for inexact real numbers any
3297 more. Now, only doubles are used to represent inexact real numbers. Further,
3298 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3299 and scm_tc16_complex, respectively.
3301 ** Removed deprecated type scm_smobfuns
3303 ** Removed deprecated function scm_newsmob
3305 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3307 There is an ongoing discussion among the developers whether to
3308 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3309 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3310 until this issue has been settled.
3312 ** Removed deprecated type tag scm_tc16_kw
3314 ** Added type tag scm_tc16_keyword
3316 (This was introduced already in release 1.3.4 but was not documented
3319 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3321 * Changes to system call interfaces:
3323 ** The "select" procedure now tests port buffers for the ability to
3324 provide input or accept output. Previously only the underlying file
3325 descriptors were checked.
3327 ** New variable PIPE_BUF: the maximum number of bytes that can be
3328 atomically written to a pipe.
3330 ** If a facility is not available on the system when Guile is
3331 compiled, the corresponding primitive procedure will not be defined.
3332 Previously it would have been defined but would throw a system-error
3333 exception if called. Exception handlers which catch this case may
3334 need minor modification: an error will be thrown with key
3335 'unbound-variable instead of 'system-error. Alternatively it's
3336 now possible to use `defined?' to check whether the facility is
3339 ** Procedures which depend on the timezone should now give the correct
3340 result on systems which cache the TZ environment variable, even if TZ
3341 is changed without calling tzset.
3343 * Changes to the networking interfaces:
3345 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3346 long integers between network and host format. For now, it's not
3347 particularly convenient to do this kind of thing, but consider:
3349 (define write-network-long
3350 (lambda (value port)
3351 (let ((v (make-uniform-vector 1 1 0)))
3352 (uniform-vector-set! v 0 (htonl value))
3353 (uniform-vector-write v port))))
3355 (define read-network-long
3357 (let ((v (make-uniform-vector 1 1 0)))
3358 (uniform-vector-read! v port)
3359 (ntohl (uniform-vector-ref v 0)))))
3361 ** If inet-aton fails, it now throws an error with key 'misc-error
3362 instead of 'system-error, since errno is not relevant.
3364 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3365 specific keys instead of 'system-error. The latter is inappropriate
3366 since errno will not have been set. The keys are:
3367 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3369 ** sethostent, setnetent, setprotoent, setservent: now take an
3370 optional argument STAYOPEN, which specifies whether the database
3371 remains open after a database entry is accessed randomly (e.g., using
3372 gethostbyname for the hosts database.) The default is #f. Previously
3376 Changes since Guile 1.3.2:
3378 * Changes to the stand-alone interpreter
3382 An initial version of the Guile debugger written by Chris Hanson has
3383 been added. The debugger is still under development but is included
3384 in the distribution anyway since it is already quite useful.
3390 after an error to enter the debugger. Type `help' inside the debugger
3391 for a description of available commands.
3393 If you prefer to have stack frames numbered and printed in
3394 anti-chronological order and prefer up in the stack to be down on the
3395 screen as is the case in gdb, you can put
3397 (debug-enable 'backwards)
3399 in your .guile startup file. (However, this means that Guile can't
3400 use indentation to indicate stack level.)
3402 The debugger is autoloaded into Guile at the first use.
3404 ** Further enhancements to backtraces
3406 There is a new debug option `width' which controls the maximum width
3407 on the screen of printed stack frames. Fancy printing parameters
3408 ("level" and "length" as in Common LISP) are adaptively adjusted for
3409 each stack frame to give maximum information while still fitting
3410 within the bounds. If the stack frame can't be made to fit by
3411 adjusting parameters, it is simply cut off at the end. This is marked
3414 ** Some modules are now only loaded when the repl is started
3416 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3417 regex) are now loaded into (guile-user) only if the repl has been
3418 started. The effect is that the startup time for scripts has been
3419 reduced to 30% of what it was previously.
3421 Correctly written scripts load the modules they require at the top of
3422 the file and should not be affected by this change.
3424 ** Hooks are now represented as smobs
3426 * Changes to Scheme functions and syntax
3428 ** Readline support has changed again.
3430 The old (readline-activator) module is gone. Use (ice-9 readline)
3431 instead, which now contains all readline functionality. So the code
3432 to activate readline is now
3434 (use-modules (ice-9 readline))
3437 This should work at any time, including from the guile prompt.
3439 To avoid confusion about the terms of Guile's license, please only
3440 enable readline for your personal use; please don't make it the
3441 default for others. Here is why we make this rather odd-sounding
3444 Guile is normally licensed under a weakened form of the GNU General
3445 Public License, which allows you to link code with Guile without
3446 placing that code under the GPL. This exception is important to some
3449 However, since readline is distributed under the GNU General Public
3450 License, when you link Guile with readline, either statically or
3451 dynamically, you effectively change Guile's license to the strict GPL.
3452 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3453 which are normally permitted become forbidden. This is a rather
3454 non-obvious consequence of the licensing terms.
3456 So, to make sure things remain clear, please let people choose for
3457 themselves whether to link GPL'd libraries like readline with Guile.
3459 ** regexp-substitute/global has changed slightly, but incompatibly.
3461 If you include a function in the item list, the string of the match
3462 object it receives is the same string passed to
3463 regexp-substitute/global, not some suffix of that string.
3464 Correspondingly, the match's positions are relative to the entire
3465 string, not the suffix.
3467 If the regexp can match the empty string, the way matches are chosen
3468 from the string has changed. regexp-substitute/global recognizes the
3469 same set of matches that list-matches does; see below.
3471 ** New function: list-matches REGEXP STRING [FLAGS]
3473 Return a list of match objects, one for every non-overlapping, maximal
3474 match of REGEXP in STRING. The matches appear in left-to-right order.
3475 list-matches only reports matches of the empty string if there are no
3476 other matches which begin on, end at, or include the empty match's
3479 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3481 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3483 For each match of REGEXP in STRING, apply PROC to the match object,
3484 and the last value PROC returned, or INIT for the first call. Return
3485 the last value returned by PROC. We apply PROC to the matches as they
3486 appear from left to right.
3488 This function recognizes matches according to the same criteria as
3491 Thus, you could define list-matches like this:
3493 (define (list-matches regexp string . flags)
3494 (reverse! (apply fold-matches regexp string '() cons flags)))
3496 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3500 *** New function: hook? OBJ
3502 Return #t if OBJ is a hook, otherwise #f.
3504 *** New function: make-hook-with-name NAME [ARITY]
3506 Return a hook with name NAME and arity ARITY. The default value for
3507 ARITY is 0. The only effect of NAME is that it will appear when the
3508 hook object is printed to ease debugging.
3510 *** New function: hook-empty? HOOK
3512 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3514 *** New function: hook->list HOOK
3516 Return a list of the procedures that are called when run-hook is
3519 ** `map' signals an error if its argument lists are not all the same length.
3521 This is the behavior required by R5RS, so this change is really a bug
3522 fix. But it seems to affect a lot of people's code, so we're
3523 mentioning it here anyway.
3525 ** Print-state handling has been made more transparent
3527 Under certain circumstances, ports are represented as a port with an
3528 associated print state. Earlier, this pair was represented as a pair
3529 (see "Some magic has been added to the printer" below). It is now
3530 indistinguishable (almost; see `get-print-state') from a port on the
3533 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3535 Return a new port with the associated print state PRINT-STATE.
3537 *** New function: get-print-state OUTPUT-PORT
3539 Return the print state associated with this port if it exists,
3540 otherwise return #f.
3542 *** New function: directory-stream? OBJECT
3544 Returns true iff OBJECT is a directory stream --- the sort of object
3545 returned by `opendir'.
3547 ** New function: using-readline?
3549 Return #t if readline is in use in the current repl.
3551 ** structs will be removed in 1.4
3553 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3554 and use GOOPS objects as the fundamental record type.
3556 * Changes to the scm_ interface
3558 ** structs will be removed in 1.4
3560 The entire current struct interface (struct.c, struct.h) will be
3561 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3562 GOOPS objects as the fundamental record type.
3564 ** The internal representation of subr's has changed
3566 Instead of giving a hint to the subr name, the CAR field of the subr
3567 now contains an index to a subr entry in scm_subr_table.
3569 *** New variable: scm_subr_table
3571 An array of subr entries. A subr entry contains the name, properties
3572 and documentation associated with the subr. The properties and
3573 documentation slots are not yet used.
3575 ** A new scheme for "forwarding" calls to a builtin to a generic function
3577 It is now possible to extend the functionality of some Guile
3578 primitives by letting them defer a call to a GOOPS generic function on
3579 argument mismatch. This means that there is no loss of efficiency in
3584 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3585 (define-method + ((x <string>) (y <string>))
3586 (string-append x y))
3588 + will still be as efficient as usual in numerical calculations, but
3589 can also be used for concatenating strings.
3591 Who will be the first one to extend Guile's numerical tower to
3592 rationals? :) [OK, there a few other things to fix before this can
3593 be made in a clean way.]
3595 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3597 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3599 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3601 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3602 a variable GENERIC which can be used by the dispatch macros below.
3604 [This is experimental code which may change soon.]
3606 *** New macros for forwarding control to a generic on arg type error
3608 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3610 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3612 These correspond to the scm_wta function call, and have the same
3613 behaviour until the user has called the GOOPS primitive
3614 `enable-primitive-generic!'. After that, these macros will apply the
3615 generic function GENERIC to the argument(s) instead of calling
3618 [This is experimental code which may change soon.]
3620 *** New macros for argument testing with generic dispatch
3622 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3624 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3626 These correspond to the SCM_ASSERT macro, but will defer control to
3627 GENERIC on error after `enable-primitive-generic!' has been called.
3629 [This is experimental code which may change soon.]
3631 ** New function: SCM scm_eval_body (SCM body, SCM env)
3633 Evaluates the body of a special form.
3635 ** The internal representation of struct's has changed
3637 Previously, four slots were allocated for the procedure(s) of entities
3638 and operators. The motivation for this representation had to do with
3639 the structure of the evaluator, the wish to support tail-recursive
3640 generic functions, and efficiency. Since the generic function
3641 dispatch mechanism has changed, there is no longer a need for such an
3642 expensive representation, and the representation has been simplified.
3644 This should not make any difference for most users.
3646 ** GOOPS support has been cleaned up.
3648 Some code has been moved from eval.c to objects.c and code in both of
3649 these compilation units has been cleaned up and better structured.
3651 *** New functions for applying generic functions
3653 New function: SCM scm_apply_generic (GENERIC, ARGS)
3654 New function: SCM scm_call_generic_0 (GENERIC)
3655 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3656 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3657 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3659 ** Deprecated function: scm_make_named_hook
3661 It is now replaced by:
3663 ** New function: SCM scm_create_hook (const char *name, int arity)
3665 Creates a hook in the same way as make-hook above but also
3666 binds a variable named NAME to it.
3668 This is the typical way of creating a hook from C code.
3670 Currently, the variable is created in the "current" module.
3671 This might change when we get the new module system.
3673 [The behaviour is identical to scm_make_named_hook.]
3677 Changes since Guile 1.3:
3679 * Changes to mailing lists
3681 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3683 See the README file to find current addresses for all the Guile
3686 * Changes to the distribution
3688 ** Readline support is no longer included with Guile by default.
3690 Based on the different license terms of Guile and Readline, we
3691 concluded that Guile should not *by default* cause the linking of
3692 Readline into an application program. Readline support is now offered
3693 as a separate module, which is linked into an application only when
3694 you explicitly specify it.
3696 Although Guile is GNU software, its distribution terms add a special
3697 exception to the usual GNU General Public License (GPL). Guile's
3698 license includes a clause that allows you to link Guile with non-free
3699 programs. We add this exception so as not to put Guile at a
3700 disadvantage vis-a-vis other extensibility packages that support other
3703 In contrast, the GNU Readline library is distributed under the GNU
3704 General Public License pure and simple. This means that you may not
3705 link Readline, even dynamically, into an application unless it is
3706 distributed under a free software license that is compatible the GPL.
3708 Because of this difference in distribution terms, an application that
3709 can use Guile may not be able to use Readline. Now users will be
3710 explicitly offered two independent decisions about the use of these
3713 You can activate the readline support by issuing
3715 (use-modules (readline-activator))
3718 from your ".guile" file, for example.
3720 * Changes to the stand-alone interpreter
3722 ** All builtins now print as primitives.
3723 Previously builtin procedures not belonging to the fundamental subr
3724 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3725 Now, they print as #<primitive-procedure NAME>.
3727 ** Backtraces slightly more intelligible.
3728 gsubr-apply and macro transformer application frames no longer appear
3731 * Changes to Scheme functions and syntax
3733 ** Guile now correctly handles internal defines by rewriting them into
3734 their equivalent letrec. Previously, internal defines would
3735 incrementally add to the innermost environment, without checking
3736 whether the restrictions specified in RnRS were met. This lead to the
3737 correct behaviour when these restriction actually were met, but didn't
3738 catch all illegal uses. Such an illegal use could lead to crashes of
3739 the Guile interpreter or or other unwanted results. An example of
3740 incorrect internal defines that made Guile behave erratically:
3752 The problem with this example is that the definition of `c' uses the
3753 value of `b' directly. This confuses the meoization machine of Guile
3754 so that the second call of `b' (this time in a larger environment that
3755 also contains bindings for `c' and `d') refers to the binding of `c'
3756 instead of `a'. You could also make Guile crash with a variation on
3761 (define (b flag) (if flag a 1))
3762 (define c (1+ (b flag)))
3770 From now on, Guile will issue an `Unbound variable: b' error message
3775 A hook contains a list of functions which should be called on
3776 particular occasions in an existing program. Hooks are used for
3779 A window manager might have a hook before-window-map-hook. The window
3780 manager uses the function run-hooks to call all functions stored in
3781 before-window-map-hook each time a window is mapped. The user can
3782 store functions in the hook using add-hook!.
3784 In Guile, hooks are first class objects.
3786 *** New function: make-hook [N_ARGS]
3788 Return a hook for hook functions which can take N_ARGS arguments.
3789 The default value for N_ARGS is 0.
3791 (See also scm_make_named_hook below.)
3793 *** New function: add-hook! HOOK PROC [APPEND_P]
3795 Put PROC at the beginning of the list of functions stored in HOOK.
3796 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3798 PROC must be able to take the number of arguments specified when the
3801 If PROC already exists in HOOK, then remove it first.
3803 *** New function: remove-hook! HOOK PROC
3805 Remove PROC from the list of functions in HOOK.
3807 *** New function: reset-hook! HOOK
3809 Clear the list of hook functions stored in HOOK.
3811 *** New function: run-hook HOOK ARG1 ...
3813 Run all hook functions stored in HOOK with arguments ARG1 ... .
3814 The number of arguments supplied must correspond to the number given
3815 when the hook was created.
3817 ** The function `dynamic-link' now takes optional keyword arguments.
3818 The only keyword argument that is currently defined is `:global
3819 BOOL'. With it, you can control whether the shared library will be
3820 linked in global mode or not. In global mode, the symbols from the
3821 linked library can be used to resolve references from other
3822 dynamically linked libraries. In non-global mode, the linked
3823 library is essentially invisible and can only be accessed via
3824 `dynamic-func', etc. The default is now to link in global mode.
3825 Previously, the default has been non-global mode.
3827 The `#:global' keyword is only effective on platforms that support
3828 the dlopen family of functions.
3830 ** New function `provided?'
3832 - Function: provided? FEATURE
3833 Return true iff FEATURE is supported by this installation of
3834 Guile. FEATURE must be a symbol naming a feature; the global
3835 variable `*features*' is a list of available features.
3837 ** Changes to the module (ice-9 expect):
3839 *** The expect-strings macro now matches `$' in a regular expression
3840 only at a line-break or end-of-file by default. Previously it would
3841 match the end of the string accumulated so far. The old behaviour
3842 can be obtained by setting the variable `expect-strings-exec-flags'
3845 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3846 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3847 in a regular expression will still match before a line-break or
3848 end-of-file. The default is `regexp/noteol'.
3850 *** The expect-strings macro now uses a variable
3851 `expect-strings-compile-flags' for the flags to be supplied to
3852 `make-regexp'. The default is `regexp/newline', which was previously
3855 *** The expect macro now supplies two arguments to a match procedure:
3856 the current accumulated string and a flag to indicate whether
3857 end-of-file has been reached. Previously only the string was supplied.
3858 If end-of-file is reached, the match procedure will be called an
3859 additional time with the same accumulated string as the previous call
3860 but with the flag set.
3862 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3864 This code, and the documentation for it that appears here, was
3865 borrowed from SLIB, with minor adaptations for Guile.
3867 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3868 An almost complete implementation of Common LISP format description
3869 according to the CL reference book `Common LISP' from Guy L.
3870 Steele, Digital Press. Backward compatible to most of the
3871 available Scheme format implementations.
3873 Returns `#t', `#f' or a string; has side effect of printing
3874 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3875 to the current output port and `#t' is returned. If DESTINATION
3876 is `#f', a formatted string is returned as the result of the call.
3877 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3878 format string; FORMAT-STRING is then the first argument and the
3879 output is returned as a string. If DESTINATION is a number, the
3880 output is to the current error port if available by the
3881 implementation. Otherwise DESTINATION must be an output port and
3884 FORMAT-STRING must be a string. In case of a formatting error
3885 format returns `#f' and prints a message on the current output or
3886 error port. Characters are output as if the string were output by
3887 the `display' function with the exception of those prefixed by a
3888 tilde (~). For a detailed description of the FORMAT-STRING syntax
3889 please consult a Common LISP format reference manual. For a test
3890 suite to verify this format implementation load `formatst.scm'.
3891 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3893 Note: `format' is not reentrant, i.e. only one `format'-call may
3894 be executed at a time.
3897 *** Format Specification (Format version 3.0)
3899 Please consult a Common LISP format reference manual for a detailed
3900 description of the format string syntax. For a demonstration of the
3901 implemented directives see `formatst.scm'.
3903 This implementation supports directive parameters and modifiers (`:'
3904 and `@' characters). Multiple parameters must be separated by a comma
3905 (`,'). Parameters can be numerical parameters (positive or negative),
3906 character parameters (prefixed by a quote character (`''), variable
3907 parameters (`v'), number of rest arguments parameter (`#'), empty and
3908 default parameters. Directive characters are case independent. The
3909 general form of a directive is:
3911 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3913 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3915 *** Implemented CL Format Control Directives
3917 Documentation syntax: Uppercase characters represent the
3918 corresponding control directive characters. Lowercase characters
3919 represent control directive parameter descriptions.
3922 Any (print as `display' does).
3926 `~MINCOL,COLINC,MINPAD,PADCHARA'
3930 S-expression (print as `write' does).
3934 `~MINCOL,COLINC,MINPAD,PADCHARS'
3940 print number sign always.
3943 print comma separated.
3945 `~MINCOL,PADCHAR,COMMACHARD'
3951 print number sign always.
3954 print comma separated.
3956 `~MINCOL,PADCHAR,COMMACHARX'
3962 print number sign always.
3965 print comma separated.
3967 `~MINCOL,PADCHAR,COMMACHARO'
3973 print number sign always.
3976 print comma separated.
3978 `~MINCOL,PADCHAR,COMMACHARB'
3983 `~N,MINCOL,PADCHAR,COMMACHARR'
3987 print a number as a Roman numeral.
3990 print a number as an "old fashioned" Roman numeral.
3993 print a number as an ordinal English number.
3996 print a number as a cardinal English number.
4001 prints `y' and `ies'.
4004 as `~P but jumps 1 argument backward.'
4007 as `~@P but jumps 1 argument backward.'
4012 prints a character as the reader can understand it (i.e. `#\'
4016 prints a character as emacs does (eg. `^C' for ASCII 03).
4019 Fixed-format floating-point (prints a flonum like MMM.NNN).
4020 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4022 If the number is positive a plus sign is printed.
4025 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4026 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4028 If the number is positive a plus sign is printed.
4031 General floating-point (prints a flonum either fixed or
4033 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4035 If the number is positive a plus sign is printed.
4038 Dollars floating-point (prints a flonum in fixed with signs
4040 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4042 If the number is positive a plus sign is printed.
4045 A sign is always printed and appears before the padding.
4048 The sign appears before the padding.
4056 print newline if not at the beginning of the output line.
4058 prints `~&' and then N-1 newlines.
4063 print N page separators.
4073 newline is ignored, white space left.
4076 newline is left, white space ignored.
4081 relative tabulation.
4087 Indirection (expects indirect arguments as a list).
4089 extracts indirect arguments from format arguments.
4092 Case conversion (converts by `string-downcase').
4094 converts by `string-capitalize'.
4097 converts by `string-capitalize-first'.
4100 converts by `string-upcase'.
4103 Argument Jumping (jumps 1 argument forward).
4105 jumps N arguments forward.
4108 jumps 1 argument backward.
4111 jumps N arguments backward.
4114 jumps to the 0th argument.
4117 jumps to the Nth argument (beginning from 0)
4119 `~[STR0~;STR1~;...~;STRN~]'
4120 Conditional Expression (numerical clause conditional).
4122 take argument from N.
4125 true test conditional.
4128 if-else-then conditional.
4134 default clause follows.
4137 Iteration (args come from the next argument (a list)).
4139 at most N iterations.
4142 args from next arg (a list of lists).
4145 args from the rest of arguments.
4148 args from the rest args (lists).
4159 aborts if N <= M <= K
4161 *** Not Implemented CL Format Control Directives
4164 print `#f' as an empty list (see below).
4167 print `#f' as an empty list (see below).
4173 (sorry I don't understand its semantics completely)
4175 *** Extended, Replaced and Additional Control Directives
4177 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4178 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4179 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4180 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4181 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4182 COMMAWIDTH is the number of characters between two comma
4186 print a R4RS complex number as `~F~@Fi' with passed parameters for
4190 Pretty print formatting of an argument for scheme code lists.
4196 Flushes the output if format DESTINATION is a port.
4199 Print a `#\space' character
4201 print N `#\space' characters.
4204 Print a `#\tab' character
4206 print N `#\tab' characters.
4209 Takes N as an integer representation for a character. No arguments
4210 are consumed. N is converted to a character by `integer->char'. N
4211 must be a positive decimal number.
4214 Print out readproof. Prints out internal objects represented as
4215 `#<...>' as strings `"#<...>"' so that the format output can always
4216 be processed by `read'.
4219 Print out readproof. Prints out internal objects represented as
4220 `#<...>' as strings `"#<...>"' so that the format output can always
4221 be processed by `read'.
4224 Prints information and a copyright notice on the format
4227 prints format version.
4230 may also print number strings, i.e. passing a number as a string
4231 and format it accordingly.
4233 *** Configuration Variables
4235 The format module exports some configuration variables to suit the
4236 systems and users needs. There should be no modification necessary for
4237 the configuration that comes with Guile. Format detects automatically
4238 if the running scheme system implements floating point numbers and
4241 format:symbol-case-conv
4242 Symbols are converted by `symbol->string' so the case type of the
4243 printed symbols is implementation dependent.
4244 `format:symbol-case-conv' is a one arg closure which is either
4245 `#f' (no conversion), `string-upcase', `string-downcase' or
4246 `string-capitalize'. (default `#f')
4248 format:iobj-case-conv
4249 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4250 implementation internal objects. (default `#f')
4253 The character prefixing the exponent value in `~E' printing.
4256 *** Compatibility With Other Format Implementations
4262 Downward compatible except for padding support and `~A', `~S',
4263 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4264 `printf' padding support which is completely replaced by the CL
4265 `format' padding style.
4268 Downward compatible except for `~', which is not documented
4269 (ignores all characters inside the format string up to a newline
4270 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4271 numerical and variable parameters and `:/@' modifiers in the CL
4275 Downward compatible except for `~A' and `~S' which print in
4276 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4277 directive parameters or modifiers)).
4280 Downward compatible except for an optional destination parameter:
4281 S2C accepts a format call without a destination which returns a
4282 formatted string. This is equivalent to a #f destination in S2C.
4283 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4284 parameters or modifiers)).
4287 ** Changes to string-handling functions.
4289 These functions were added to support the (ice-9 format) module, above.
4291 *** New function: string-upcase STRING
4292 *** New function: string-downcase STRING
4294 These are non-destructive versions of the existing string-upcase! and
4295 string-downcase! functions.
4297 *** New function: string-capitalize! STRING
4298 *** New function: string-capitalize STRING
4300 These functions convert the first letter of each word in the string to
4303 (string-capitalize "howdy there")
4306 As with the other functions, string-capitalize! modifies the string in
4307 place, while string-capitalize returns a modified copy of its argument.
4309 *** New function: string-ci->symbol STRING
4311 Return a symbol whose name is STRING, but having the same case as if
4312 the symbol had be read by `read'.
4314 Guile can be configured to be sensitive or insensitive to case
4315 differences in Scheme identifiers. If Guile is case-insensitive, all
4316 symbols are converted to lower case on input. The `string-ci->symbol'
4317 function returns a symbol whose name in STRING, transformed as Guile
4318 would if STRING were input.
4320 *** New function: substring-move! STRING1 START END STRING2 START
4322 Copy the substring of STRING1 from START (inclusive) to END
4323 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4324 string, and the source and destination areas may overlap; in all
4325 cases, the function behaves as if all the characters were copied
4328 *** Extended functions: substring-move-left! substring-move-right!
4330 These functions now correctly copy arbitrarily overlapping substrings;
4331 they are both synonyms for substring-move!.
4334 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4336 getopt-long is a function for parsing command-line arguments in a
4337 manner consistent with other GNU programs.
4339 (getopt-long ARGS GRAMMAR)
4340 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4342 ARGS should be a list of strings. Its first element should be the
4343 name of the program; subsequent elements should be the arguments
4344 that were passed to the program on the command line. The
4345 `program-arguments' procedure returns a list of this form.
4347 GRAMMAR is a list of the form:
4348 ((OPTION (PROPERTY VALUE) ...) ...)
4350 Each OPTION should be a symbol. `getopt-long' will accept a
4351 command-line option named `--OPTION'.
4352 Each option can have the following (PROPERTY VALUE) pairs:
4354 (single-char CHAR) --- Accept `-CHAR' as a single-character
4355 equivalent to `--OPTION'. This is how to specify traditional
4357 (required? BOOL) --- If BOOL is true, the option is required.
4358 getopt-long will raise an error if it is not found in ARGS.
4359 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4360 it is #f, it does not; and if it is the symbol
4361 `optional', the option may appear in ARGS with or
4363 (predicate FUNC) --- If the option accepts a value (i.e. you
4364 specified `(value #t)' for this option), then getopt
4365 will apply FUNC to the value, and throw an exception
4366 if it returns #f. FUNC should be a procedure which
4367 accepts a string and returns a boolean value; you may
4368 need to use quasiquotes to get it into GRAMMAR.
4370 The (PROPERTY VALUE) pairs may occur in any order, but each
4371 property may occur only once. By default, options do not have
4372 single-character equivalents, are not required, and do not take
4375 In ARGS, single-character options may be combined, in the usual
4376 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4377 accepts values, then it must be the last option in the
4378 combination; the value is the next argument. So, for example, using
4379 the following grammar:
4380 ((apples (single-char #\a))
4381 (blimps (single-char #\b) (value #t))
4382 (catalexis (single-char #\c) (value #t)))
4383 the following argument lists would be acceptable:
4384 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4385 for "blimps" and "catalexis")
4386 ("-ab" "bang" "-c" "couth") (same)
4387 ("-ac" "couth" "-b" "bang") (same)
4388 ("-abc" "couth" "bang") (an error, since `-b' is not the
4389 last option in its combination)
4391 If an option's value is optional, then `getopt-long' decides
4392 whether it has a value by looking at what follows it in ARGS. If
4393 the next element is a string, and it does not appear to be an
4394 option itself, then that string is the option's value.
4396 The value of a long option can appear as the next element in ARGS,
4397 or it can follow the option name, separated by an `=' character.
4398 Thus, using the same grammar as above, the following argument lists
4400 ("--apples" "Braeburn" "--blimps" "Goodyear")
4401 ("--apples=Braeburn" "--blimps" "Goodyear")
4402 ("--blimps" "Goodyear" "--apples=Braeburn")
4404 If the option "--" appears in ARGS, argument parsing stops there;
4405 subsequent arguments are returned as ordinary arguments, even if
4406 they resemble options. So, in the argument list:
4407 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4408 `getopt-long' will recognize the `apples' option as having the
4409 value "Granny Smith", but it will not recognize the `blimp'
4410 option; it will return the strings "--blimp" and "Goodyear" as
4411 ordinary argument strings.
4413 The `getopt-long' function returns the parsed argument list as an
4414 assocation list, mapping option names --- the symbols from GRAMMAR
4415 --- onto their values, or #t if the option does not accept a value.
4416 Unused options do not appear in the alist.
4418 All arguments that are not the value of any option are returned
4419 as a list, associated with the empty list.
4421 `getopt-long' throws an exception if:
4422 - it finds an unrecognized option in ARGS
4423 - a required option is omitted
4424 - an option that requires an argument doesn't get one
4425 - an option that doesn't accept an argument does get one (this can
4426 only happen using the long option `--opt=value' syntax)
4427 - an option predicate fails
4432 `((lockfile-dir (required? #t)
4435 (predicate ,file-is-directory?))
4436 (verbose (required? #f)
4439 (x-includes (single-char #\x))
4440 (rnet-server (single-char #\y)
4441 (predicate ,string?))))
4443 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4444 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4446 => ((() "foo1" "-fred" "foo2" "foo3")
4447 (rnet-server . "lamprod")
4448 (x-includes . "/usr/include")
4449 (lockfile-dir . "/tmp")
4452 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4454 It will be removed in a few releases.
4456 ** New syntax: lambda*
4457 ** New syntax: define*
4458 ** New syntax: define*-public
4459 ** New syntax: defmacro*
4460 ** New syntax: defmacro*-public
4461 Guile now supports optional arguments.
4463 `lambda*', `define*', `define*-public', `defmacro*' and
4464 `defmacro*-public' are identical to the non-* versions except that
4465 they use an extended type of parameter list that has the following BNF
4466 syntax (parentheses are literal, square brackets indicate grouping,
4467 and `*', `+' and `?' have the usual meaning):
4469 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4470 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4471 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4473 ext-var-decl ::= identifier | ( identifier expression )
4475 The semantics are best illustrated with the following documentation
4476 and examples for `lambda*':
4479 lambda extended for optional and keyword arguments
4481 lambda* creates a procedure that takes optional arguments. These
4482 are specified by putting them inside brackets at the end of the
4483 paramater list, but before any dotted rest argument. For example,
4484 (lambda* (a b #&optional c d . e) '())
4485 creates a procedure with fixed arguments a and b, optional arguments c
4486 and d, and rest argument e. If the optional arguments are omitted
4487 in a call, the variables for them are unbound in the procedure. This
4488 can be checked with the bound? macro.
4490 lambda* can also take keyword arguments. For example, a procedure
4492 (lambda* (#&key xyzzy larch) '())
4493 can be called with any of the argument lists (#:xyzzy 11)
4494 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4495 are given as keywords are bound to values.
4497 Optional and keyword arguments can also be given default values
4498 which they take on when they are not present in a call, by giving a
4499 two-item list in place of an optional argument, for example in:
4500 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4501 foo is a fixed argument, bar is an optional argument with default
4502 value 42, and baz is a keyword argument with default value 73.
4503 Default value expressions are not evaluated unless they are needed
4504 and until the procedure is called.
4506 lambda* now supports two more special parameter list keywords.
4508 lambda*-defined procedures now throw an error by default if a
4509 keyword other than one of those specified is found in the actual
4510 passed arguments. However, specifying #&allow-other-keys
4511 immediately after the kyword argument declarations restores the
4512 previous behavior of ignoring unknown keywords. lambda* also now
4513 guarantees that if the same keyword is passed more than once, the
4514 last one passed is the one that takes effect. For example,
4515 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4516 #:heads 37 #:tails 42 #:heads 99)
4517 would result in (99 47) being displayed.
4519 #&rest is also now provided as a synonym for the dotted syntax rest
4520 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4521 all respects to lambda*. This is provided for more similarity to DSSSL,
4522 MIT-Scheme and Kawa among others, as well as for refugees from other
4525 Further documentation may be found in the optargs.scm file itself.
4527 The optional argument module also exports the macros `let-optional',
4528 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4529 are not documented here because they may be removed in the future, but
4530 full documentation is still available in optargs.scm.
4532 ** New syntax: and-let*
4533 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4535 Syntax: (land* (<clause> ...) <body> ...)
4536 Each <clause> should have one of the following forms:
4537 (<variable> <expression>)
4540 Each <variable> or <bound-variable> should be an identifier. Each
4541 <expression> should be a valid expression. The <body> should be a
4542 possibly empty sequence of expressions, like the <body> of a
4545 Semantics: A LAND* expression is evaluated by evaluating the
4546 <expression> or <bound-variable> of each of the <clause>s from
4547 left to right. The value of the first <expression> or
4548 <bound-variable> that evaluates to a false value is returned; the
4549 remaining <expression>s and <bound-variable>s are not evaluated.
4550 The <body> forms are evaluated iff all the <expression>s and
4551 <bound-variable>s evaluate to true values.
4553 The <expression>s and the <body> are evaluated in an environment
4554 binding each <variable> of the preceding (<variable> <expression>)
4555 clauses to the value of the <expression>. Later bindings
4556 shadow earlier bindings.
4558 Guile's and-let* macro was contributed by Michael Livshin.
4560 ** New sorting functions
4562 *** New function: sorted? SEQUENCE LESS?
4563 Returns `#t' when the sequence argument is in non-decreasing order
4564 according to LESS? (that is, there is no adjacent pair `... x y
4565 ...' for which `(less? y x)').
4567 Returns `#f' when the sequence contains at least one out-of-order
4568 pair. It is an error if the sequence is neither a list nor a
4571 *** New function: merge LIST1 LIST2 LESS?
4572 LIST1 and LIST2 are sorted lists.
4573 Returns the sorted list of all elements in LIST1 and LIST2.
4575 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4576 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4577 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4578 (Here "<" should read "comes before".)
4580 *** New procedure: merge! LIST1 LIST2 LESS?
4581 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4582 the result. If the code is compiled, and LESS? constructs no new
4583 pairs, no pairs at all will be allocated. The first pair of the
4584 result will be either the first pair of LIST1 or the first pair of
4587 *** New function: sort SEQUENCE LESS?
4588 Accepts either a list or a vector, and returns a new sequence
4589 which is sorted. The new sequence is the same type as the input.
4590 Always `(sorted? (sort sequence less?) less?)'. The original
4591 sequence is not altered in any way. The new sequence shares its
4592 elements with the old one; no elements are copied.
4594 *** New procedure: sort! SEQUENCE LESS
4595 Returns its sorted result in the original boxes. No new storage is
4596 allocated at all. Proper usage: (set! slist (sort! slist <))
4598 *** New function: stable-sort SEQUENCE LESS?
4599 Similar to `sort' but stable. That is, if "equal" elements are
4600 ordered a < b in the original sequence, they will have the same order
4603 *** New function: stable-sort! SEQUENCE LESS?
4604 Similar to `sort!' but stable.
4605 Uses temporary storage when sorting vectors.
4607 *** New functions: sort-list, sort-list!
4608 Added for compatibility with scsh.
4610 ** New built-in random number support
4612 *** New function: random N [STATE]
4613 Accepts a positive integer or real N and returns a number of the
4614 same type between zero (inclusive) and N (exclusive). The values
4615 returned have a uniform distribution.
4617 The optional argument STATE must be of the type produced by
4618 `copy-random-state' or `seed->random-state'. It defaults to the value
4619 of the variable `*random-state*'. This object is used to maintain the
4620 state of the pseudo-random-number generator and is altered as a side
4621 effect of the `random' operation.
4623 *** New variable: *random-state*
4624 Holds a data structure that encodes the internal state of the
4625 random-number generator that `random' uses by default. The nature
4626 of this data structure is implementation-dependent. It may be
4627 printed out and successfully read back in, but may or may not
4628 function correctly as a random-number state object in another
4631 *** New function: copy-random-state [STATE]
4632 Returns a new object of type suitable for use as the value of the
4633 variable `*random-state*' and as a second argument to `random'.
4634 If argument STATE is given, a copy of it is returned. Otherwise a
4635 copy of `*random-state*' is returned.
4637 *** New function: seed->random-state SEED
4638 Returns a new object of type suitable for use as the value of the
4639 variable `*random-state*' and as a second argument to `random'.
4640 SEED is a string or a number. A new state is generated and
4641 initialized using SEED.
4643 *** New function: random:uniform [STATE]
4644 Returns an uniformly distributed inexact real random number in the
4645 range between 0 and 1.
4647 *** New procedure: random:solid-sphere! VECT [STATE]
4648 Fills VECT with inexact real random numbers the sum of whose
4649 squares is less than 1.0. Thinking of VECT as coordinates in
4650 space of dimension N = `(vector-length VECT)', the coordinates are
4651 uniformly distributed within the unit N-shere. The sum of the
4652 squares of the numbers is returned. VECT can be either a vector
4653 or a uniform vector of doubles.
4655 *** New procedure: random:hollow-sphere! VECT [STATE]
4656 Fills VECT with inexact real random numbers the sum of whose squares
4657 is equal to 1.0. Thinking of VECT as coordinates in space of
4658 dimension n = `(vector-length VECT)', the coordinates are uniformly
4659 distributed over the surface of the unit n-shere. VECT can be either
4660 a vector or a uniform vector of doubles.
4662 *** New function: random:normal [STATE]
4663 Returns an inexact real in a normal distribution with mean 0 and
4664 standard deviation 1. For a normal distribution with mean M and
4665 standard deviation D use `(+ M (* D (random:normal)))'.
4667 *** New procedure: random:normal-vector! VECT [STATE]
4668 Fills VECT with inexact real random numbers which are independent and
4669 standard normally distributed (i.e., with mean 0 and variance 1).
4670 VECT can be either a vector or a uniform vector of doubles.
4672 *** New function: random:exp STATE
4673 Returns an inexact real in an exponential distribution with mean 1.
4674 For an exponential distribution with mean U use (* U (random:exp)).
4676 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4678 These functions now operate on numbers in the range of a C unsigned
4681 These functions used to operate on numbers in the range of a C signed
4682 long; however, this seems inappropriate, because Guile integers don't
4685 ** New function: make-guardian
4686 This is an implementation of guardians as described in
4687 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4688 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4689 Programming Language Design and Implementation, June 1993
4690 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4692 ** New functions: delq1!, delv1!, delete1!
4693 These procedures behave similar to delq! and friends but delete only
4694 one object if at all.
4696 ** New function: unread-string STRING PORT
4697 Unread STRING to PORT, that is, push it back onto the port so that
4698 next read operation will work on the pushed back characters.
4700 ** unread-char can now be called multiple times
4701 If unread-char is called multiple times, the unread characters will be
4702 read again in last-in first-out order.
4704 ** the procedures uniform-array-read! and uniform-array-write! now
4705 work on any kind of port, not just ports which are open on a file.
4707 ** Now 'l' in a port mode requests line buffering.
4709 ** The procedure truncate-file now works on string ports as well
4710 as file ports. If the size argument is omitted, the current
4711 file position is used.
4713 ** new procedure: seek PORT/FDES OFFSET WHENCE
4714 The arguments are the same as for the old fseek procedure, but it
4715 works on string ports as well as random-access file ports.
4717 ** the fseek procedure now works on string ports, since it has been
4718 redefined using seek.
4720 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4721 size is not supplied.
4723 ** the newline procedure no longer flushes the port if it's not
4724 line-buffered: previously it did if it was the current output port.
4726 ** open-pipe and close-pipe are no longer primitive procedures, but
4727 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4729 ** the freopen procedure has been removed.
4731 ** new procedure: drain-input PORT
4732 Drains PORT's read buffers (including any pushed-back characters)
4733 and returns the contents as a single string.
4735 ** New function: map-in-order PROC LIST1 LIST2 ...
4736 Version of `map' which guarantees that the procedure is applied to the
4737 lists in serial order.
4739 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4740 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4741 now obsolete and will go away in release 1.5.
4743 ** New syntax: collect BODY1 ...
4744 Version of `begin' which returns a list of the results of the body
4745 forms instead of the result of the last body form. In contrast to
4746 `begin', `collect' allows an empty body.
4748 ** New functions: read-history FILENAME, write-history FILENAME
4749 Read/write command line history from/to file. Returns #t on success
4750 and #f if an error occured.
4752 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4754 These procedures return a list of definitions available in the specified
4755 argument, a relative module reference. In the case of no argument,
4756 `(current-module)' is now consulted for definitions to return, instead
4757 of simply returning #f, the former behavior.
4759 ** The #/ syntax for lists is no longer supported.
4761 Earlier versions of Scheme accepted this syntax, but printed a
4764 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4766 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4769 * Changes to the gh_ interface
4773 Now takes a second argument which is the result array. If this
4774 pointer is NULL, a new array is malloced (the old behaviour).
4776 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4777 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4781 * Changes to the scm_ interface
4783 ** Function: scm_make_named_hook (char* name, int n_args)
4785 Creates a hook in the same way as make-hook above but also
4786 binds a variable named NAME to it.
4788 This is the typical way of creating a hook from C code.
4790 Currently, the variable is created in the "current" module. This
4791 might change when we get the new module system.
4793 ** The smob interface
4795 The interface for creating smobs has changed. For documentation, see
4796 data-rep.info (made from guile-core/doc/data-rep.texi).
4798 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4800 >>> This function will be removed in 1.3.4. <<<
4804 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4805 This function adds a new smob type, named NAME, with instance size
4806 SIZE to the system. The return value is a tag that is used in
4807 creating instances of the type. If SIZE is 0, then no memory will
4808 be allocated when instances of the smob are created, and nothing
4809 will be freed by the default free function.
4811 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4812 This function sets the smob marking procedure for the smob type
4813 specified by the tag TC. TC is the tag returned by
4814 `scm_make_smob_type'.
4816 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4817 This function sets the smob freeing procedure for the smob type
4818 specified by the tag TC. TC is the tag returned by
4819 `scm_make_smob_type'.
4821 *** Function: void scm_set_smob_print (tc, print)
4823 - Function: void scm_set_smob_print (long tc,
4824 scm_sizet (*print) (SCM,
4828 This function sets the smob printing procedure for the smob type
4829 specified by the tag TC. TC is the tag returned by
4830 `scm_make_smob_type'.
4832 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4833 This function sets the smob equality-testing predicate for the
4834 smob type specified by the tag TC. TC is the tag returned by
4835 `scm_make_smob_type'.
4837 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4838 Make VALUE contain a smob instance of the type with type code TC and
4839 smob data DATA. VALUE must be previously declared as C type `SCM'.
4841 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4842 This macro expands to a block of code that creates a smob instance
4843 of the type with type code TC and smob data DATA, and returns that
4844 `SCM' value. It should be the last piece of code in a block.
4846 ** The interfaces for using I/O ports and implementing port types
4847 (ptobs) have changed significantly. The new interface is based on
4848 shared access to buffers and a new set of ptob procedures.
4850 *** scm_newptob has been removed
4854 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4856 - Function: SCM scm_make_port_type (char *type_name,
4857 int (*fill_buffer) (SCM port),
4858 void (*write_flush) (SCM port));
4860 Similarly to the new smob interface, there is a set of function
4861 setters by which the user can customize the behaviour of his port
4862 type. See ports.h (scm_set_port_XXX).
4864 ** scm_strport_to_string: New function: creates a new string from
4865 a string port's buffer.
4867 ** Plug in interface for random number generators
4868 The variable `scm_the_rng' in random.c contains a value and three
4869 function pointers which together define the current random number
4870 generator being used by the Scheme level interface and the random
4871 number library functions.
4873 The user is free to replace the default generator with the generator
4876 *** Variable: size_t scm_the_rng.rstate_size
4877 The size of the random state type used by the current RNG
4880 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4881 Given the random STATE, return 32 random bits.
4883 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4884 Seed random state STATE using string S of length N.
4886 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4887 Given random state STATE, return a malloced copy.
4890 The default RNG is the MWC (Multiply With Carry) random number
4891 generator described by George Marsaglia at the Department of
4892 Statistics and Supercomputer Computations Research Institute, The
4893 Florida State University (http://stat.fsu.edu/~geo).
4895 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4896 passes all tests in the DIEHARD test suite
4897 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4898 costs one multiply and one add on platforms which either supports long
4899 longs (gcc does this on most systems) or have 64 bit longs. The cost
4900 is four multiply on other systems but this can be optimized by writing
4901 scm_i_uniform32 in assembler.
4903 These functions are provided through the scm_the_rng interface for use
4904 by libguile and the application.
4906 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4907 Given the random STATE, return 32 random bits.
4908 Don't use this function directly. Instead go through the plugin
4909 interface (see "Plug in interface" above).
4911 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4912 Initialize STATE using SEED of length N.
4914 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4915 Return a malloc:ed copy of STATE. This function can easily be re-used
4916 in the interfaces to other RNGs.
4918 ** Random number library functions
4919 These functions use the current RNG through the scm_the_rng interface.
4920 It might be a good idea to use these functions from your C code so
4921 that only one random generator is used by all code in your program.
4923 The default random state is stored in:
4925 *** Variable: SCM scm_var_random_state
4926 Contains the vcell of the Scheme variable "*random-state*" which is
4927 used as default state by all random number functions in the Scheme
4932 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4934 *** Function: scm_rstate *scm_c_default_rstate (void)
4935 This is a convenience function which returns the value of
4936 scm_var_random_state. An error message is generated if this value
4937 isn't a random state.
4939 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4940 Make a new random state from the string SEED of length LENGTH.
4942 It is generally not a good idea to use multiple random states in a
4943 program. While subsequent random numbers generated from one random
4944 state are guaranteed to be reasonably independent, there is no such
4945 guarantee for numbers generated from different random states.
4947 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4948 Return 32 random bits.
4950 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4951 Return a sample from the uniform(0,1) distribution.
4953 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4954 Return a sample from the normal(0,1) distribution.
4956 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4957 Return a sample from the exp(1) distribution.
4959 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4960 Return a sample from the discrete uniform(0,M) distribution.
4962 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4963 Return a sample from the discrete uniform(0,M) distribution.
4964 M must be a bignum object. The returned value may be an INUM.
4968 Changes in Guile 1.3 (released Monday, October 19, 1998):
4970 * Changes to the distribution
4972 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4973 To avoid conflicts, programs should name environment variables after
4974 themselves, except when there's a common practice establishing some
4977 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4978 giving the former precedence, and printing a warning message if the
4979 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4981 ** The header files related to multi-byte characters have been removed.
4982 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4983 which referred to these explicitly will probably need to be rewritten,
4984 since the support for the variant string types has been removed; see
4987 ** The header files append.h and sequences.h have been removed. These
4988 files implemented non-R4RS operations which would encourage
4989 non-portable programming style and less easy-to-read code.
4991 * Changes to the stand-alone interpreter
4993 ** New procedures have been added to implement a "batch mode":
4995 *** Function: batch-mode?
4997 Returns a boolean indicating whether the interpreter is in batch
5000 *** Function: set-batch-mode?! ARG
5002 If ARG is true, switches the interpreter to batch mode. The `#f'
5003 case has not been implemented.
5005 ** Guile now provides full command-line editing, when run interactively.
5006 To use this feature, you must have the readline library installed.
5007 The Guile build process will notice it, and automatically include
5010 The readline library is available via anonymous FTP from any GNU
5011 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5013 ** the-last-stack is now a fluid.
5015 * Changes to the procedure for linking libguile with your programs
5017 ** You can now use the `guile-config' utility to build programs that use Guile.
5019 Guile now includes a command-line utility called `guile-config', which
5020 can provide information about how to compile and link programs that
5023 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5024 You should include this command's output on the command line you use
5025 to compile C or C++ code that #includes the Guile header files. It's
5026 usually just a `-I' flag to help the compiler find the Guile headers.
5029 *** `guile-config link' prints any linker flags necessary to link with Guile.
5031 This command writes to its standard output a list of flags which you
5032 must pass to the linker to link your code against the Guile library.
5033 The flags include '-lguile' itself, any other libraries the Guile
5034 library depends upon, and any `-L' flags needed to help the linker
5035 find those libraries.
5037 For example, here is a Makefile rule that builds a program named 'foo'
5038 from the object files ${FOO_OBJECTS}, and links them against Guile:
5041 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5043 Previous Guile releases recommended that you use autoconf to detect
5044 which of a predefined set of libraries were present on your system.
5045 It is more robust to use `guile-config', since it records exactly which
5046 libraries the installed Guile library requires.
5048 This was originally called `build-guile', but was renamed to
5049 `guile-config' before Guile 1.3 was released, to be consistent with
5050 the analogous script for the GTK+ GUI toolkit, which is called
5054 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5056 If you are using the GNU autoconf package to configure your program,
5057 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5058 (described above) and gather the necessary values for use in your
5061 The GUILE_FLAGS macro expands to configure script code which runs the
5062 `guile-config' script, to find out where Guile's header files and
5063 libraries are installed. It sets two variables, marked for
5064 substitution, as by AC_SUBST.
5066 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5067 code that uses Guile header files. This is almost always just a
5070 GUILE_LDFLAGS --- flags to pass to the linker to link a
5071 program against Guile. This includes `-lguile' for the Guile
5072 library itself, any libraries that Guile itself requires (like
5073 -lqthreads), and so on. It may also include a -L flag to tell the
5074 compiler where to find the libraries.
5076 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5077 directory of the Guile distribution. You can copy it into your
5078 package's aclocal.m4 file, and then use it in your configure.in file.
5080 If you are using the `aclocal' program, distributed with GNU automake,
5081 to maintain your aclocal.m4 file, the Guile installation process
5082 installs guile.m4 where aclocal will find it. All you need to do is
5083 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5084 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5088 * Changes to Scheme functions and syntax
5090 ** Multi-byte strings have been removed, as have multi-byte and wide
5091 ports. We felt that these were the wrong approach to
5092 internationalization support.
5094 ** New function: readline [PROMPT]
5095 Read a line from the terminal, and allow the user to edit it,
5096 prompting with PROMPT. READLINE provides a large set of Emacs-like
5097 editing commands, lets the user recall previously typed lines, and
5098 works on almost every kind of terminal, including dumb terminals.
5100 READLINE assumes that the cursor is at the beginning of the line when
5101 it is invoked. Thus, you can't print a prompt yourself, and then call
5102 READLINE; you need to package up your prompt as a string, pass it to
5103 the function, and let READLINE print the prompt itself. This is
5104 because READLINE needs to know the prompt's screen width.
5106 For Guile to provide this function, you must have the readline
5107 library, version 2.1 or later, installed on your system. Readline is
5108 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5109 any GNU mirror site.
5111 See also ADD-HISTORY function.
5113 ** New function: add-history STRING
5114 Add STRING as the most recent line in the history used by the READLINE
5115 command. READLINE does not add lines to the history itself; you must
5116 call ADD-HISTORY to make previous input available to the user.
5118 ** The behavior of the read-line function has changed.
5120 This function now uses standard C library functions to read the line,
5121 for speed. This means that it doesn not respect the value of
5122 scm-line-incrementors; it assumes that lines are delimited with
5125 (Note that this is read-line, the function that reads a line of text
5126 from a port, not readline, the function that reads a line from a
5127 terminal, providing full editing capabilities.)
5129 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5131 This module provides some simple argument parsing. It exports one
5134 Function: getopt-gnu-style ARG-LS
5135 Parse a list of program arguments into an alist of option
5138 Each item in the list of program arguments is examined to see if
5139 it meets the syntax of a GNU long-named option. An argument like
5140 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5141 returned alist, where MUMBLE is a keyword object with the same
5142 name as the argument. An argument like `--MUMBLE=FROB' produces
5143 an element of the form (MUMBLE . FROB), where FROB is a string.
5145 As a special case, the returned alist also contains a pair whose
5146 car is the symbol `rest'. The cdr of this pair is a list
5147 containing all the items in the argument list that are not options
5148 of the form mentioned above.
5150 The argument `--' is treated specially: all items in the argument
5151 list appearing after such an argument are not examined, and are
5152 returned in the special `rest' list.
5154 This function does not parse normal single-character switches.
5155 You will need to parse them out of the `rest' list yourself.
5157 ** The read syntax for byte vectors and short vectors has changed.
5159 Instead of #bytes(...), write #y(...).
5161 Instead of #short(...), write #h(...).
5163 This may seem nutty, but, like the other uniform vectors, byte vectors
5164 and short vectors want to have the same print and read syntax (and,
5165 more basic, want to have read syntax!). Changing the read syntax to
5166 use multiple characters after the hash sign breaks with the
5167 conventions used in R5RS and the conventions used for the other
5168 uniform vectors. It also introduces complexity in the current reader,
5169 both on the C and Scheme levels. (The Right solution is probably to
5170 change the syntax and prototypes for uniform vectors entirely.)
5173 ** The new module (ice-9 session) provides useful interactive functions.
5175 *** New procedure: (apropos REGEXP OPTION ...)
5177 Display a list of top-level variables whose names match REGEXP, and
5178 the modules they are imported from. Each OPTION should be one of the
5181 value --- Show the value of each matching variable.
5182 shadow --- Show bindings shadowed by subsequently imported modules.
5183 full --- Same as both `shadow' and `value'.
5187 guile> (apropos "trace" 'full)
5188 debug: trace #<procedure trace args>
5189 debug: untrace #<procedure untrace args>
5190 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5191 the-scm-module: before-backtrace-hook ()
5192 the-scm-module: backtrace #<primitive-procedure backtrace>
5193 the-scm-module: after-backtrace-hook ()
5194 the-scm-module: has-shown-backtrace-hint? #f
5197 ** There are new functions and syntax for working with macros.
5199 Guile implements macros as a special object type. Any variable whose
5200 top-level binding is a macro object acts as a macro. The macro object
5201 specifies how the expression should be transformed before evaluation.
5203 *** Macro objects now print in a reasonable way, resembling procedures.
5205 *** New function: (macro? OBJ)
5206 True iff OBJ is a macro object.
5208 *** New function: (primitive-macro? OBJ)
5209 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5210 macro transformers, implemented in eval.c rather than Scheme code.
5212 Why do we have this function?
5213 - For symmetry with procedure? and primitive-procedure?,
5214 - to allow custom print procedures to tell whether a macro is
5215 primitive, and display it differently, and
5216 - to allow compilers and user-written evaluators to distinguish
5217 builtin special forms from user-defined ones, which could be
5220 *** New function: (macro-type OBJ)
5221 Return a value indicating what kind of macro OBJ is. Possible return
5224 The symbol `syntax' --- a macro created by procedure->syntax.
5225 The symbol `macro' --- a macro created by procedure->macro.
5226 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5227 The boolean #f --- if OBJ is not a macro object.
5229 *** New function: (macro-name MACRO)
5230 Return the name of the macro object MACRO's procedure, as returned by
5233 *** New function: (macro-transformer MACRO)
5234 Return the transformer procedure for MACRO.
5236 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5238 Specify a new macro expander to use in the current module. Each
5239 MODULE is a module name, with the same meaning as in the `use-modules'
5240 form; each named module's exported bindings are added to the current
5241 top-level environment. TRANSFORMER is an expression evaluated in the
5242 resulting environment which must yield a procedure to use as the
5243 module's eval transformer: every expression evaluated in this module
5244 is passed to this function, and the result passed to the Guile
5247 *** macro-eval! is removed. Use local-eval instead.
5249 ** Some magic has been added to the printer to better handle user
5250 written printing routines (like record printers, closure printers).
5252 The problem is that these user written routines must have access to
5253 the current `print-state' to be able to handle fancy things like
5254 detection of circular references. These print-states have to be
5255 passed to the builtin printing routines (display, write, etc) to
5256 properly continue the print chain.
5258 We didn't want to change all existing print code so that it
5259 explicitly passes thru a print state in addition to a port. Instead,
5260 we extented the possible values that the builtin printing routines
5261 accept as a `port'. In addition to a normal port, they now also take
5262 a pair of a normal port and a print-state. Printing will go to the
5263 port and the print-state will be used to control the detection of
5264 circular references, etc. If the builtin function does not care for a
5265 print-state, it is simply ignored.
5267 User written callbacks are now called with such a pair as their
5268 `port', but because every function now accepts this pair as a PORT
5269 argument, you don't have to worry about that. In fact, it is probably
5270 safest to not check for these pairs.
5272 However, it is sometimes necessary to continue a print chain on a
5273 different port, for example to get a intermediate string
5274 representation of the printed value, mangle that string somehow, and
5275 then to finally print the mangled string. Use the new function
5277 inherit-print-state OLD-PORT NEW-PORT
5279 for this. It constructs a new `port' that prints to NEW-PORT but
5280 inherits the print-state of OLD-PORT.
5282 ** struct-vtable-offset renamed to vtable-offset-user
5284 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5286 ** There is now a third optional argument to make-vtable-vtable
5287 (and fourth to make-struct) when constructing new types (vtables).
5288 This argument initializes field vtable-index-printer of the vtable.
5290 ** The detection of circular references has been extended to structs.
5291 That is, a structure that -- in the process of being printed -- prints
5292 itself does not lead to infinite recursion.
5294 ** There is now some basic support for fluids. Please read
5295 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5296 the following functions and macros:
5298 Function: make-fluid
5300 Create a new fluid object. Fluids are not special variables or
5301 some other extension to the semantics of Scheme, but rather
5302 ordinary Scheme objects. You can store them into variables (that
5303 are still lexically scoped, of course) or into any other place you
5304 like. Every fluid has a initial value of `#f'.
5306 Function: fluid? OBJ
5308 Test whether OBJ is a fluid.
5310 Function: fluid-ref FLUID
5311 Function: fluid-set! FLUID VAL
5313 Access/modify the fluid FLUID. Modifications are only visible
5314 within the current dynamic root (that includes threads).
5316 Function: with-fluids* FLUIDS VALUES THUNK
5318 FLUIDS is a list of fluids and VALUES a corresponding list of
5319 values for these fluids. Before THUNK gets called the values are
5320 installed in the fluids and the old values of the fluids are
5321 saved in the VALUES list. When the flow of control leaves THUNK
5322 or reenters it, the values get swapped again. You might think of
5323 this as a `safe-fluid-excursion'. Note that the VALUES list is
5324 modified by `with-fluids*'.
5326 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5328 The same as `with-fluids*' but with a different syntax. It looks
5329 just like `let', but both FLUID and VALUE are evaluated. Remember,
5330 fluids are not special variables but ordinary objects. FLUID
5331 should evaluate to a fluid.
5333 ** Changes to system call interfaces:
5335 *** close-port, close-input-port and close-output-port now return a
5336 boolean instead of an `unspecified' object. #t means that the port
5337 was successfully closed, while #f means it was already closed. It is
5338 also now possible for these procedures to raise an exception if an
5339 error occurs (some errors from write can be delayed until close.)
5341 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5344 *** the third argument to fcntl is now optional.
5346 *** the first argument to chown can now be a file descriptor or a port.
5348 *** the argument to stat can now be a port.
5350 *** The following new procedures have been added (most use scsh
5353 *** procedure: close PORT/FD
5354 Similar to close-port (*note close-port: Closing Ports.), but also
5355 works on file descriptors. A side effect of closing a file
5356 descriptor is that any ports using that file descriptor are moved
5357 to a different file descriptor and have their revealed counts set
5360 *** procedure: port->fdes PORT
5361 Returns the integer file descriptor underlying PORT. As a side
5362 effect the revealed count of PORT is incremented.
5364 *** procedure: fdes->ports FDES
5365 Returns a list of existing ports which have FDES as an underlying
5366 file descriptor, without changing their revealed counts.
5368 *** procedure: fdes->inport FDES
5369 Returns an existing input port which has FDES as its underlying
5370 file descriptor, if one exists, and increments its revealed count.
5371 Otherwise, returns a new input port with a revealed count of 1.
5373 *** procedure: fdes->outport FDES
5374 Returns an existing output port which has FDES as its underlying
5375 file descriptor, if one exists, and increments its revealed count.
5376 Otherwise, returns a new output port with a revealed count of 1.
5378 The next group of procedures perform a `dup2' system call, if NEWFD
5379 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5380 duplicated can be supplied as an integer or contained in a port. The
5381 type of value returned varies depending on which procedure is used.
5383 All procedures also have the side effect when performing `dup2' that
5384 any ports using NEWFD are moved to a different file descriptor and have
5385 their revealed counts set to zero.
5387 *** procedure: dup->fdes PORT/FD [NEWFD]
5388 Returns an integer file descriptor.
5390 *** procedure: dup->inport PORT/FD [NEWFD]
5391 Returns a new input port using the new file descriptor.
5393 *** procedure: dup->outport PORT/FD [NEWFD]
5394 Returns a new output port using the new file descriptor.
5396 *** procedure: dup PORT/FD [NEWFD]
5397 Returns a new port if PORT/FD is a port, with the same mode as the
5398 supplied port, otherwise returns an integer file descriptor.
5400 *** procedure: dup->port PORT/FD MODE [NEWFD]
5401 Returns a new port using the new file descriptor. MODE supplies a
5402 mode string for the port (*note open-file: File Ports.).
5404 *** procedure: setenv NAME VALUE
5405 Modifies the environment of the current process, which is also the
5406 default environment inherited by child processes.
5408 If VALUE is `#f', then NAME is removed from the environment.
5409 Otherwise, the string NAME=VALUE is added to the environment,
5410 replacing any existing string with name matching NAME.
5412 The return value is unspecified.
5414 *** procedure: truncate-file OBJ SIZE
5415 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5416 can be a string containing a file name or an integer file
5417 descriptor or port open for output on the file. The underlying
5418 system calls are `truncate' and `ftruncate'.
5420 The return value is unspecified.
5422 *** procedure: setvbuf PORT MODE [SIZE]
5423 Set the buffering mode for PORT. MODE can be:
5431 block buffered, using a newly allocated buffer of SIZE bytes.
5432 However if SIZE is zero or unspecified, the port will be made
5435 This procedure should not be used after I/O has been performed with
5438 Ports are usually block buffered by default, with a default buffer
5439 size. Procedures e.g., *Note open-file: File Ports, which accept a
5440 mode string allow `0' to be added to request an unbuffered port.
5442 *** procedure: fsync PORT/FD
5443 Copies any unwritten data for the specified output file descriptor
5444 to disk. If PORT/FD is a port, its buffer is flushed before the
5445 underlying file descriptor is fsync'd. The return value is
5448 *** procedure: open-fdes PATH FLAGS [MODES]
5449 Similar to `open' but returns a file descriptor instead of a port.
5451 *** procedure: execle PATH ENV [ARG] ...
5452 Similar to `execl', but the environment of the new process is
5453 specified by ENV, which must be a list of strings as returned by
5454 the `environ' procedure.
5456 This procedure is currently implemented using the `execve' system
5457 call, but we call it `execle' because of its Scheme calling
5460 *** procedure: strerror ERRNO
5461 Returns the Unix error message corresponding to ERRNO, an integer.
5463 *** procedure: primitive-exit [STATUS]
5464 Terminate the current process without unwinding the Scheme stack.
5465 This is would typically be useful after a fork. The exit status
5466 is STATUS if supplied, otherwise zero.
5468 *** procedure: times
5469 Returns an object with information about real and processor time.
5470 The following procedures accept such an object as an argument and
5471 return a selected component:
5474 The current real time, expressed as time units relative to an
5478 The CPU time units used by the calling process.
5481 The CPU time units used by the system on behalf of the
5485 The CPU time units used by terminated child processes of the
5486 calling process, whose status has been collected (e.g., using
5490 Similarly, the CPU times units used by the system on behalf of
5491 terminated child processes.
5493 ** Removed: list-length
5494 ** Removed: list-append, list-append!
5495 ** Removed: list-reverse, list-reverse!
5497 ** array-map renamed to array-map!
5499 ** serial-array-map renamed to serial-array-map!
5501 ** catch doesn't take #f as first argument any longer
5503 Previously, it was possible to pass #f instead of a key to `catch'.
5504 That would cause `catch' to pass a jump buffer object to the procedure
5505 passed as second argument. The procedure could then use this jump
5506 buffer objekt as an argument to throw.
5508 This mechanism has been removed since its utility doesn't motivate the
5509 extra complexity it introduces.
5511 ** The `#/' notation for lists now provokes a warning message from Guile.
5512 This syntax will be removed from Guile in the near future.
5514 To disable the warning message, set the GUILE_HUSH environment
5515 variable to any non-empty value.
5517 ** The newline character now prints as `#\newline', following the
5518 normal Scheme notation, not `#\nl'.
5520 * Changes to the gh_ interface
5522 ** The gh_enter function now takes care of loading the Guile startup files.
5523 gh_enter works by calling scm_boot_guile; see the remarks below.
5525 ** Function: void gh_write (SCM x)
5527 Write the printed representation of the scheme object x to the current
5528 output port. Corresponds to the scheme level `write'.
5530 ** gh_list_length renamed to gh_length.
5532 ** vector handling routines
5534 Several major changes. In particular, gh_vector() now resembles
5535 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5536 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5537 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5538 vector-related gh_ functions have been implemented.
5540 ** pair and list routines
5542 Implemented several of the R4RS pair and list functions that were
5545 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5547 New function. Converts double arrays back and forth between Scheme
5550 * Changes to the scm_ interface
5552 ** The function scm_boot_guile now takes care of loading the startup files.
5554 Guile's primary initialization function, scm_boot_guile, now takes
5555 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5556 Guile, define the module system, and put together some standard
5557 bindings. It also loads `init.scm', which is intended to hold
5558 site-specific initialization code.
5560 Since Guile cannot operate properly until boot-9.scm is loaded, there
5561 is no reason to separate loading boot-9.scm from Guile's other
5562 initialization processes.
5564 This job used to be done by scm_compile_shell_switches, which didn't
5565 make much sense; in particular, it meant that people using Guile for
5566 non-shell-like applications had to jump through hoops to get Guile
5567 initialized properly.
5569 ** The function scm_compile_shell_switches no longer loads the startup files.
5570 Now, Guile always loads the startup files, whenever it is initialized;
5571 see the notes above for scm_boot_guile and scm_load_startup_files.
5573 ** Function: scm_load_startup_files
5574 This new function takes care of loading Guile's initialization file
5575 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5576 this is always called by the Guile initialization process, it's
5577 probably not too useful to call this yourself, but it's there anyway.
5579 ** The semantics of smob marking have changed slightly.
5581 The smob marking function (the `mark' member of the scm_smobfuns
5582 structure) is no longer responsible for setting the mark bit on the
5583 smob. The generic smob handling code in the garbage collector will
5584 set this bit. The mark function need only ensure that any other
5585 objects the smob refers to get marked.
5587 Note that this change means that the smob's GC8MARK bit is typically
5588 already set upon entry to the mark function. Thus, marking functions
5589 which look like this:
5592 if (SCM_GC8MARKP (ptr))
5594 SCM_SETGC8MARK (ptr);
5595 ... mark objects to which the smob refers ...
5598 are now incorrect, since they will return early, and fail to mark any
5599 other objects the smob refers to. Some code in the Guile library used
5602 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5604 If you have implemented your own I/O port type, by writing the
5605 functions required by the scm_ptobfuns and then calling scm_newptob,
5606 you will need to change your functions slightly.
5608 The functions in a scm_ptobfuns structure now expect the port itself
5609 as their argument; they used to expect the `stream' member of the
5610 port's scm_port_table structure. This allows functions in an
5611 scm_ptobfuns structure to easily access the port's cell (and any flags
5612 it its CAR), and the port's scm_port_table structure.
5614 Guile now passes the I/O port itself as the `port' argument in the
5615 following scm_ptobfuns functions:
5617 int (*free) (SCM port);
5618 int (*fputc) (int, SCM port);
5619 int (*fputs) (char *, SCM port);
5620 scm_sizet (*fwrite) SCM_P ((char *ptr,
5624 int (*fflush) (SCM port);
5625 int (*fgetc) (SCM port);
5626 int (*fclose) (SCM port);
5628 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5631 If you have existing code which defines its own port types, it is easy
5632 to convert your code to the new interface; simply apply SCM_STREAM to
5633 the port argument to yield the value you code used to expect.
5635 Note that since both the port and the stream have the same type in the
5636 C code --- they are both SCM values --- the C compiler will not remind
5637 you if you forget to update your scm_ptobfuns functions.
5640 ** Function: int scm_internal_select (int fds,
5644 struct timeval *timeout);
5646 This is a replacement for the `select' function provided by the OS.
5647 It enables I/O blocking and sleeping to happen for one cooperative
5648 thread without blocking other threads. It also avoids busy-loops in
5649 these situations. It is intended that all I/O blocking and sleeping
5650 will finally go through this function. Currently, this function is
5651 only available on systems providing `gettimeofday' and `select'.
5653 ** Function: SCM scm_internal_stack_catch (SCM tag,
5654 scm_catch_body_t body,
5656 scm_catch_handler_t handler,
5659 A new sibling to the other two C level `catch' functions
5660 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5661 the stack to be saved automatically into the variable `the-last-stack'
5662 (scm_the_last_stack_var) on error. This is necessary if you want to
5663 use advanced error reporting, such as calling scm_display_error and
5664 scm_display_backtrace. (They both take a stack object as argument.)
5666 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5668 scm_catch_handler_t handler,
5671 Spawns a new thread. It does a job similar to
5672 scm_call_with_new_thread but takes arguments more suitable when
5673 spawning threads from application C code.
5675 ** The hook scm_error_callback has been removed. It was originally
5676 intended as a way for the user to install his own error handler. But
5677 that method works badly since it intervenes between throw and catch,
5678 thereby changing the semantics of expressions like (catch #t ...).
5679 The correct way to do it is to use one of the C level catch functions
5680 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5682 ** Removed functions:
5684 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5685 scm_list_reverse, scm_list_reverse_x
5687 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5689 These can be used for pretty list creation from C. The idea is taken
5690 from Erick Gallesio's STk.
5692 ** scm_array_map renamed to scm_array_map_x
5694 ** mbstrings are now removed
5696 This means that the type codes scm_tc7_mb_string and
5697 scm_tc7_mb_substring has been removed.
5699 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5701 Since we no longer support multi-byte strings, these I/O functions
5702 have been simplified, and renamed. Here are their old names, and
5703 their new names and arguments:
5705 scm_gen_putc -> void scm_putc (int c, SCM port);
5706 scm_gen_puts -> void scm_puts (char *s, SCM port);
5707 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5708 scm_gen_getc -> void scm_getc (SCM port);
5711 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5713 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5715 SCM_TYP7S now masks away the bit which distinguishes substrings from
5718 ** scm_catch_body_t: Backward incompatible change!
5720 Body functions to scm_internal_catch and friends do not any longer
5721 take a second argument. This is because it is no longer possible to
5722 pass a #f arg to catch.
5724 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5726 The function scm_protect_object protects its argument from being freed
5727 by the garbage collector. scm_unprotect_object removes that
5730 These functions now nest properly. That is, for every object O, there
5731 is a counter which scm_protect_object(O) increments and
5732 scm_unprotect_object(O) decrements, if the counter is greater than
5733 zero. Every object's counter is zero when it is first created. If an
5734 object's counter is greater than zero, the garbage collector will not
5735 reclaim its storage.
5737 This allows you to use scm_protect_object in your code without
5738 worrying that some other function you call will call
5739 scm_unprotect_object, and allow it to be freed. Assuming that the
5740 functions you call are well-behaved, and unprotect only those objects
5741 they protect, you can follow the same rule and have confidence that
5742 objects will be freed only at appropriate times.
5745 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5747 * Changes to the distribution
5749 ** Nightly snapshots are now available from ftp.red-bean.com.
5750 The old server, ftp.cyclic.com, has been relinquished to its rightful
5753 Nightly snapshots of the Guile development sources are now available via
5754 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5756 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5757 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5759 ** To run Guile without installing it, the procedure has changed a bit.
5761 If you used a separate build directory to compile Guile, you'll need
5762 to include the build directory in SCHEME_LOAD_PATH, as well as the
5763 source directory. See the `INSTALL' file for examples.
5765 * Changes to the procedure for linking libguile with your programs
5767 ** The standard Guile load path for Scheme code now includes
5768 $(datadir)/guile (usually /usr/local/share/guile). This means that
5769 you can install your own Scheme files there, and Guile will find them.
5770 (Previous versions of Guile only checked a directory whose name
5771 contained the Guile version number, so you had to re-install or move
5772 your Scheme sources each time you installed a fresh version of Guile.)
5774 The load path also includes $(datadir)/guile/site; we recommend
5775 putting individual Scheme files there. If you want to install a
5776 package with multiple source files, create a directory for them under
5779 ** Guile 1.2 will now use the Rx regular expression library, if it is
5780 installed on your system. When you are linking libguile into your own
5781 programs, this means you will have to link against -lguile, -lqt (if
5782 you configured Guile with thread support), and -lrx.
5784 If you are using autoconf to generate configuration scripts for your
5785 application, the following lines should suffice to add the appropriate
5786 libraries to your link command:
5788 ### Find Rx, quickthreads and libguile.
5789 AC_CHECK_LIB(rx, main)
5790 AC_CHECK_LIB(qt, main)
5791 AC_CHECK_LIB(guile, scm_shell)
5793 The Guile 1.2 distribution does not contain sources for the Rx
5794 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5795 retrieve it from a GNU FTP site and install it separately.
5797 * Changes to Scheme functions and syntax
5799 ** The dynamic linking features of Guile are now enabled by default.
5800 You can disable them by giving the `--disable-dynamic-linking' option
5803 (dynamic-link FILENAME)
5805 Find the object file denoted by FILENAME (a string) and link it
5806 into the running Guile application. When everything works out,
5807 return a Scheme object suitable for representing the linked object
5808 file. Otherwise an error is thrown. How object files are
5809 searched is system dependent.
5811 (dynamic-object? VAL)
5813 Determine whether VAL represents a dynamically linked object file.
5815 (dynamic-unlink DYNOBJ)
5817 Unlink the indicated object file from the application. DYNOBJ
5818 should be one of the values returned by `dynamic-link'.
5820 (dynamic-func FUNCTION DYNOBJ)
5822 Search the C function indicated by FUNCTION (a string or symbol)
5823 in DYNOBJ and return some Scheme object that can later be used
5824 with `dynamic-call' to actually call this function. Right now,
5825 these Scheme objects are formed by casting the address of the
5826 function to `long' and converting this number to its Scheme
5829 (dynamic-call FUNCTION DYNOBJ)
5831 Call the C function indicated by FUNCTION and DYNOBJ. The
5832 function is passed no arguments and its return value is ignored.
5833 When FUNCTION is something returned by `dynamic-func', call that
5834 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5835 etc.), look it up in DYNOBJ; this is equivalent to
5837 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5839 Interrupts are deferred while the C function is executing (with
5840 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5842 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5844 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5845 some arguments and return its return value. The C function is
5846 expected to take two arguments and return an `int', just like
5849 int c_func (int argc, char **argv);
5851 ARGS must be a list of strings and is converted into an array of
5852 `char *'. The array is passed in ARGV and its size in ARGC. The
5853 return value is converted to a Scheme number and returned from the
5854 call to `dynamic-args-call'.
5856 When dynamic linking is disabled or not supported on your system,
5857 the above functions throw errors, but they are still available.
5859 Here is a small example that works on GNU/Linux:
5861 (define libc-obj (dynamic-link "libc.so"))
5862 (dynamic-args-call 'rand libc-obj '())
5864 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5866 ** The #/ syntax for module names is depreciated, and will be removed
5867 in a future version of Guile. Instead of
5875 The latter syntax is more consistent with existing Lisp practice.
5877 ** Guile now does fancier printing of structures. Structures are the
5878 underlying implementation for records, which in turn are used to
5879 implement modules, so all of these object now print differently and in
5880 a more informative way.
5882 The Scheme printer will examine the builtin variable *struct-printer*
5883 whenever it needs to print a structure object. When this variable is
5884 not `#f' it is deemed to be a procedure and will be applied to the
5885 structure object and the output port. When *struct-printer* is `#f'
5886 or the procedure return `#f' the structure object will be printed in
5887 the boring #<struct 80458270> form.
5889 This hook is used by some routines in ice-9/boot-9.scm to implement
5890 type specific printing routines. Please read the comments there about
5893 One of the more specific uses of structs are records. The printing
5894 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5895 called. It should behave like a *struct-printer* procedure (described
5898 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5899 token of the form #:NAME, where NAME has the same syntax as a Scheme
5900 symbol, is the external representation of the keyword named NAME.
5901 Keyword objects print using this syntax as well, so values containing
5902 keyword objects can be read back into Guile. When used in an
5903 expression, keywords are self-quoting objects.
5905 Guile suports this read syntax, and uses this print syntax, regardless
5906 of the current setting of the `keyword' read option. The `keyword'
5907 read option only controls whether Guile recognizes the `:NAME' syntax,
5908 which is incompatible with R4RS. (R4RS says such token represent
5911 ** Guile has regular expression support again. Guile 1.0 included
5912 functions for matching regular expressions, based on the Rx library.
5913 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5914 distribution, and thus Guile had no regular expression support. Guile
5915 1.2 again supports the most commonly used functions, and supports all
5916 of SCSH's regular expression functions.
5918 If your system does not include a POSIX regular expression library,
5919 and you have not linked Guile with a third-party regexp library such as
5920 Rx, these functions will not be available. You can tell whether your
5921 Guile installation includes regular expression support by checking
5922 whether the `*features*' list includes the `regex' symbol.
5924 *** regexp functions
5926 By default, Guile supports POSIX extended regular expressions. That
5927 means that the characters `(', `)', `+' and `?' are special, and must
5928 be escaped if you wish to match the literal characters.
5930 This regular expression interface was modeled after that implemented
5931 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5932 with SCSH regular expressions.
5934 **** Function: string-match PATTERN STR [START]
5935 Compile the string PATTERN into a regular expression and compare
5936 it with STR. The optional numeric argument START specifies the
5937 position of STR at which to begin matching.
5939 `string-match' returns a "match structure" which describes what,
5940 if anything, was matched by the regular expression. *Note Match
5941 Structures::. If STR does not match PATTERN at all,
5942 `string-match' returns `#f'.
5944 Each time `string-match' is called, it must compile its PATTERN
5945 argument into a regular expression structure. This operation is
5946 expensive, which makes `string-match' inefficient if the same regular
5947 expression is used several times (for example, in a loop). For better
5948 performance, you can compile a regular expression in advance and then
5949 match strings against the compiled regexp.
5951 **** Function: make-regexp STR [FLAGS]
5952 Compile the regular expression described by STR, and return the
5953 compiled regexp structure. If STR does not describe a legal
5954 regular expression, `make-regexp' throws a
5955 `regular-expression-syntax' error.
5957 FLAGS may be the bitwise-or of one or more of the following:
5959 **** Constant: regexp/extended
5960 Use POSIX Extended Regular Expression syntax when interpreting
5961 STR. If not set, POSIX Basic Regular Expression syntax is used.
5962 If the FLAGS argument is omitted, we assume regexp/extended.
5964 **** Constant: regexp/icase
5965 Do not differentiate case. Subsequent searches using the
5966 returned regular expression will be case insensitive.
5968 **** Constant: regexp/newline
5969 Match-any-character operators don't match a newline.
5971 A non-matching list ([^...]) not containing a newline matches a
5974 Match-beginning-of-line operator (^) matches the empty string
5975 immediately after a newline, regardless of whether the FLAGS
5976 passed to regexp-exec contain regexp/notbol.
5978 Match-end-of-line operator ($) matches the empty string
5979 immediately before a newline, regardless of whether the FLAGS
5980 passed to regexp-exec contain regexp/noteol.
5982 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5983 Match the compiled regular expression REGEXP against `str'. If
5984 the optional integer START argument is provided, begin matching
5985 from that position in the string. Return a match structure
5986 describing the results of the match, or `#f' if no match could be
5989 FLAGS may be the bitwise-or of one or more of the following:
5991 **** Constant: regexp/notbol
5992 The match-beginning-of-line operator always fails to match (but
5993 see the compilation flag regexp/newline above) This flag may be
5994 used when different portions of a string are passed to
5995 regexp-exec and the beginning of the string should not be
5996 interpreted as the beginning of the line.
5998 **** Constant: regexp/noteol
5999 The match-end-of-line operator always fails to match (but see the
6000 compilation flag regexp/newline above)
6002 **** Function: regexp? OBJ
6003 Return `#t' if OBJ is a compiled regular expression, or `#f'
6006 Regular expressions are commonly used to find patterns in one string
6007 and replace them with the contents of another string.
6009 **** Function: regexp-substitute PORT MATCH [ITEM...]
6010 Write to the output port PORT selected contents of the match
6011 structure MATCH. Each ITEM specifies what should be written, and
6012 may be one of the following arguments:
6014 * A string. String arguments are written out verbatim.
6016 * An integer. The submatch with that number is written.
6018 * The symbol `pre'. The portion of the matched string preceding
6019 the regexp match is written.
6021 * The symbol `post'. The portion of the matched string
6022 following the regexp match is written.
6024 PORT may be `#f', in which case nothing is written; instead,
6025 `regexp-substitute' constructs a string from the specified ITEMs
6028 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6029 Similar to `regexp-substitute', but can be used to perform global
6030 substitutions on STR. Instead of taking a match structure as an
6031 argument, `regexp-substitute/global' takes two string arguments: a
6032 REGEXP string describing a regular expression, and a TARGET string
6033 which should be matched against this regular expression.
6035 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6038 * A function may be supplied. When this function is called, it
6039 will be passed one argument: a match structure for a given
6040 regular expression match. It should return a string to be
6041 written out to PORT.
6043 * The `post' symbol causes `regexp-substitute/global' to recurse
6044 on the unmatched portion of STR. This *must* be supplied in
6045 order to perform global search-and-replace on STR; if it is
6046 not present among the ITEMs, then `regexp-substitute/global'
6047 will return after processing a single match.
6049 *** Match Structures
6051 A "match structure" is the object returned by `string-match' and
6052 `regexp-exec'. It describes which portion of a string, if any, matched
6053 the given regular expression. Match structures include: a reference to
6054 the string that was checked for matches; the starting and ending
6055 positions of the regexp match; and, if the regexp included any
6056 parenthesized subexpressions, the starting and ending positions of each
6059 In each of the regexp match functions described below, the `match'
6060 argument must be a match structure returned by a previous call to
6061 `string-match' or `regexp-exec'. Most of these functions return some
6062 information about the original target string that was matched against a
6063 regular expression; we will call that string TARGET for easy reference.
6065 **** Function: regexp-match? OBJ
6066 Return `#t' if OBJ is a match structure returned by a previous
6067 call to `regexp-exec', or `#f' otherwise.
6069 **** Function: match:substring MATCH [N]
6070 Return the portion of TARGET matched by subexpression number N.
6071 Submatch 0 (the default) represents the entire regexp match. If
6072 the regular expression as a whole matched, but the subexpression
6073 number N did not match, return `#f'.
6075 **** Function: match:start MATCH [N]
6076 Return the starting position of submatch number N.
6078 **** Function: match:end MATCH [N]
6079 Return the ending position of submatch number N.
6081 **** Function: match:prefix MATCH
6082 Return the unmatched portion of TARGET preceding the regexp match.
6084 **** Function: match:suffix MATCH
6085 Return the unmatched portion of TARGET following the regexp match.
6087 **** Function: match:count MATCH
6088 Return the number of parenthesized subexpressions from MATCH.
6089 Note that the entire regular expression match itself counts as a
6090 subexpression, and failed submatches are included in the count.
6092 **** Function: match:string MATCH
6093 Return the original TARGET string.
6095 *** Backslash Escapes
6097 Sometimes you will want a regexp to match characters like `*' or `$'
6098 exactly. For example, to check whether a particular string represents
6099 a menu entry from an Info node, it would be useful to match it against
6100 a regexp like `^* [^:]*::'. However, this won't work; because the
6101 asterisk is a metacharacter, it won't match the `*' at the beginning of
6102 the string. In this case, we want to make the first asterisk un-magic.
6104 You can do this by preceding the metacharacter with a backslash
6105 character `\'. (This is also called "quoting" the metacharacter, and
6106 is known as a "backslash escape".) When Guile sees a backslash in a
6107 regular expression, it considers the following glyph to be an ordinary
6108 character, no matter what special meaning it would ordinarily have.
6109 Therefore, we can make the above example work by changing the regexp to
6110 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6111 to match only a single asterisk in the target string.
6113 Since the backslash is itself a metacharacter, you may force a
6114 regexp to match a backslash in the target string by preceding the
6115 backslash with itself. For example, to find variable references in a
6116 TeX program, you might want to find occurrences of the string `\let\'
6117 followed by any number of alphabetic characters. The regular expression
6118 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6119 each match a single backslash in the target string.
6121 **** Function: regexp-quote STR
6122 Quote each special character found in STR with a backslash, and
6123 return the resulting string.
6125 *Very important:* Using backslash escapes in Guile source code (as
6126 in Emacs Lisp or C) can be tricky, because the backslash character has
6127 special meaning for the Guile reader. For example, if Guile encounters
6128 the character sequence `\n' in the middle of a string while processing
6129 Scheme code, it replaces those characters with a newline character.
6130 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6131 Several of these "escape sequences" are processed by the Guile reader
6132 before your code is executed. Unrecognized escape sequences are
6133 ignored: if the characters `\*' appear in a string, they will be
6134 translated to the single character `*'.
6136 This translation is obviously undesirable for regular expressions,
6137 since we want to be able to include backslashes in a string in order to
6138 escape regexp metacharacters. Therefore, to make sure that a backslash
6139 is preserved in a string in your Guile program, you must use *two*
6140 consecutive backslashes:
6142 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6144 The string in this example is preprocessed by the Guile reader before
6145 any code is executed. The resulting argument to `make-regexp' is the
6146 string `^\* [^:]*', which is what we really want.
6148 This also means that in order to write a regular expression that
6149 matches a single backslash character, the regular expression string in
6150 the source code must include *four* backslashes. Each consecutive pair
6151 of backslashes gets translated by the Guile reader to a single
6152 backslash, and the resulting double-backslash is interpreted by the
6153 regexp engine as matching a single backslash character. Hence:
6155 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6157 The reason for the unwieldiness of this syntax is historical. Both
6158 regular expression pattern matchers and Unix string processing systems
6159 have traditionally used backslashes with the special meanings described
6160 above. The POSIX regular expression specification and ANSI C standard
6161 both require these semantics. Attempting to abandon either convention
6162 would cause other kinds of compatibility problems, possibly more severe
6163 ones. Therefore, without extending the Scheme reader to support
6164 strings with different quoting conventions (an ungainly and confusing
6165 extension when implemented in other languages), we must adhere to this
6166 cumbersome escape syntax.
6168 * Changes to the gh_ interface
6170 * Changes to the scm_ interface
6172 * Changes to system call interfaces:
6174 ** The value returned by `raise' is now unspecified. It throws an exception
6177 *** A new procedure `sigaction' can be used to install signal handlers
6179 (sigaction signum [action] [flags])
6181 signum is the signal number, which can be specified using the value
6184 If action is omitted, sigaction returns a pair: the CAR is the current
6185 signal hander, which will be either an integer with the value SIG_DFL
6186 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6187 handles the signal, or #f if a non-Scheme procedure handles the
6188 signal. The CDR contains the current sigaction flags for the handler.
6190 If action is provided, it is installed as the new handler for signum.
6191 action can be a Scheme procedure taking one argument, or the value of
6192 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6193 whatever signal handler was installed before sigaction was first used.
6194 Flags can optionally be specified for the new handler (SA_RESTART is
6195 always used if the system provides it, so need not be specified.) The
6196 return value is a pair with information about the old handler as
6199 This interface does not provide access to the "signal blocking"
6200 facility. Maybe this is not needed, since the thread support may
6201 provide solutions to the problem of consistent access to data
6204 *** A new procedure `flush-all-ports' is equivalent to running
6205 `force-output' on every port open for output.
6207 ** Guile now provides information on how it was built, via the new
6208 global variable, %guile-build-info. This variable records the values
6209 of the standard GNU makefile directory variables as an assocation
6210 list, mapping variable names (symbols) onto directory paths (strings).
6211 For example, to find out where the Guile link libraries were
6212 installed, you can say:
6214 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6217 * Changes to the scm_ interface
6219 ** The new function scm_handle_by_message_noexit is just like the
6220 existing scm_handle_by_message function, except that it doesn't call
6221 exit to terminate the process. Instead, it prints a message and just
6222 returns #f. This might be a more appropriate catch-all handler for
6223 new dynamic roots and threads.
6226 Changes in Guile 1.1 (released Friday, May 16 1997):
6228 * Changes to the distribution.
6230 The Guile 1.0 distribution has been split up into several smaller
6232 guile-core --- the Guile interpreter itself.
6233 guile-tcltk --- the interface between the Guile interpreter and
6234 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6235 is a toolkit for building graphical user interfaces.
6236 guile-rgx-ctax --- the interface between Guile and the Rx regular
6237 expression matcher, and the translator for the Ctax
6238 programming language. These are packaged together because the
6239 Ctax translator uses Rx to parse Ctax source code.
6241 This NEWS file describes the changes made to guile-core since the 1.0
6244 We no longer distribute the documentation, since it was either out of
6245 date, or incomplete. As soon as we have current documentation, we
6250 * Changes to the stand-alone interpreter
6252 ** guile now accepts command-line arguments compatible with SCSH, Olin
6253 Shivers' Scheme Shell.
6255 In general, arguments are evaluated from left to right, but there are
6256 exceptions. The following switches stop argument processing, and
6257 stash all remaining command-line arguments as the value returned by
6258 the (command-line) function.
6259 -s SCRIPT load Scheme source code from FILE, and exit
6260 -c EXPR evalute Scheme expression EXPR, and exit
6261 -- stop scanning arguments; run interactively
6263 The switches below are processed as they are encountered.
6264 -l FILE load Scheme source code from FILE
6265 -e FUNCTION after reading script, apply FUNCTION to
6266 command line arguments
6267 -ds do -s script at this point
6268 --emacs enable Emacs protocol (experimental)
6269 -h, --help display this help and exit
6270 -v, --version display version information and exit
6271 \ read arguments from following script lines
6273 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6274 which re-implements the traditional "echo" command:
6276 #!/usr/local/bin/guile -s
6279 (map (lambda (arg) (display arg) (display " "))
6283 (main (command-line))
6285 Suppose we invoke this script as follows:
6287 ekko a speckled gecko
6289 Through the magic of Unix script processing (triggered by the `#!'
6290 token at the top of the file), /usr/local/bin/guile receives the
6291 following list of command-line arguments:
6293 ("-s" "./ekko" "a" "speckled" "gecko")
6295 Unix inserts the name of the script after the argument specified on
6296 the first line of the file (in this case, "-s"), and then follows that
6297 with the arguments given to the script. Guile loads the script, which
6298 defines the `main' function, and then applies it to the list of
6299 remaining command-line arguments, ("a" "speckled" "gecko").
6301 In Unix, the first line of a script file must take the following form:
6303 #!INTERPRETER ARGUMENT
6305 where INTERPRETER is the absolute filename of the interpreter
6306 executable, and ARGUMENT is a single command-line argument to pass to
6309 You may only pass one argument to the interpreter, and its length is
6310 limited. These restrictions can be annoying to work around, so Guile
6311 provides a general mechanism (borrowed from, and compatible with,
6312 SCSH) for circumventing them.
6314 If the ARGUMENT in a Guile script is a single backslash character,
6315 `\', Guile will open the script file, parse arguments from its second
6316 and subsequent lines, and replace the `\' with them. So, for example,
6317 here is another implementation of the `ekko' script:
6319 #!/usr/local/bin/guile \
6323 (for-each (lambda (arg) (display arg) (display " "))
6327 If the user invokes this script as follows:
6329 ekko a speckled gecko
6331 Unix expands this into
6333 /usr/local/bin/guile \ ekko a speckled gecko
6335 When Guile sees the `\' argument, it replaces it with the arguments
6336 read from the second line of the script, producing:
6338 /usr/local/bin/guile -e main -s ekko a speckled gecko
6340 This tells Guile to load the `ekko' script, and apply the function
6341 `main' to the argument list ("a" "speckled" "gecko").
6343 Here is how Guile parses the command-line arguments:
6344 - Each space character terminates an argument. This means that two
6345 spaces in a row introduce an empty-string argument.
6346 - The tab character is not permitted (unless you quote it with the
6347 backslash character, as described below), to avoid confusion.
6348 - The newline character terminates the sequence of arguments, and will
6349 also terminate a final non-empty argument. (However, a newline
6350 following a space will not introduce a final empty-string argument;
6351 it only terminates the argument list.)
6352 - The backslash character is the escape character. It escapes
6353 backslash, space, tab, and newline. The ANSI C escape sequences
6354 like \n and \t are also supported. These produce argument
6355 constituents; the two-character combination \n doesn't act like a
6356 terminating newline. The escape sequence \NNN for exactly three
6357 octal digits reads as the character whose ASCII code is NNN. As
6358 above, characters produced this way are argument constituents.
6359 Backslash followed by other characters is not allowed.
6361 * Changes to the procedure for linking libguile with your programs
6363 ** Guile now builds and installs a shared guile library, if your
6364 system support shared libraries. (It still builds a static library on
6365 all systems.) Guile automatically detects whether your system
6366 supports shared libraries. To prevent Guile from buildisg shared
6367 libraries, pass the `--disable-shared' flag to the configure script.
6369 Guile takes longer to compile when it builds shared libraries, because
6370 it must compile every file twice --- once to produce position-
6371 independent object code, and once to produce normal object code.
6373 ** The libthreads library has been merged into libguile.
6375 To link a program against Guile, you now need only link against
6376 -lguile and -lqt; -lthreads is no longer needed. If you are using
6377 autoconf to generate configuration scripts for your application, the
6378 following lines should suffice to add the appropriate libraries to
6381 ### Find quickthreads and libguile.
6382 AC_CHECK_LIB(qt, main)
6383 AC_CHECK_LIB(guile, scm_shell)
6385 * Changes to Scheme functions
6387 ** Guile Scheme's special syntax for keyword objects is now optional,
6388 and disabled by default.
6390 The syntax variation from R4RS made it difficult to port some
6391 interesting packages to Guile. The routines which accepted keyword
6392 arguments (mostly in the module system) have been modified to also
6393 accept symbols whose names begin with `:'.
6395 To change the keyword syntax, you must first import the (ice-9 debug)
6397 (use-modules (ice-9 debug))
6399 Then you can enable the keyword syntax as follows:
6400 (read-set! keywords 'prefix)
6402 To disable keyword syntax, do this:
6403 (read-set! keywords #f)
6405 ** Many more primitive functions accept shared substrings as
6406 arguments. In the past, these functions required normal, mutable
6407 strings as arguments, although they never made use of this
6410 ** The uniform array functions now operate on byte vectors. These
6411 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6412 `serial-array-map', `array-map', `array-for-each', and
6415 ** The new functions `trace' and `untrace' implement simple debugging
6416 support for Scheme functions.
6418 The `trace' function accepts any number of procedures as arguments,
6419 and tells the Guile interpreter to display each procedure's name and
6420 arguments each time the procedure is invoked. When invoked with no
6421 arguments, `trace' returns the list of procedures currently being
6424 The `untrace' function accepts any number of procedures as arguments,
6425 and tells the Guile interpreter not to trace them any more. When
6426 invoked with no arguments, `untrace' untraces all curretly traced
6429 The tracing in Guile has an advantage over most other systems: we
6430 don't create new procedure objects, but mark the procedure objects
6431 themselves. This means that anonymous and internal procedures can be
6434 ** The function `assert-repl-prompt' has been renamed to
6435 `set-repl-prompt!'. It takes one argument, PROMPT.
6436 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6437 - If PROMPT is a string, we use it as a prompt.
6438 - If PROMPT is a procedure accepting no arguments, we call it, and
6439 display the result as a prompt.
6440 - Otherwise, we display "> ".
6442 ** The new function `eval-string' reads Scheme expressions from a
6443 string and evaluates them, returning the value of the last expression
6444 in the string. If the string contains no expressions, it returns an
6447 ** The new function `thunk?' returns true iff its argument is a
6448 procedure of zero arguments.
6450 ** `defined?' is now a builtin function, instead of syntax. This
6451 means that its argument should be quoted. It returns #t iff its
6452 argument is bound in the current module.
6454 ** The new syntax `use-modules' allows you to add new modules to your
6455 environment without re-typing a complete `define-module' form. It
6456 accepts any number of module names as arguments, and imports their
6457 public bindings into the current module.
6459 ** The new function (module-defined? NAME MODULE) returns true iff
6460 NAME, a symbol, is defined in MODULE, a module object.
6462 ** The new function `builtin-bindings' creates and returns a hash
6463 table containing copies of all the root module's bindings.
6465 ** The new function `builtin-weak-bindings' does the same as
6466 `builtin-bindings', but creates a doubly-weak hash table.
6468 ** The `equal?' function now considers variable objects to be
6469 equivalent if they have the same name and the same value.
6471 ** The new function `command-line' returns the command-line arguments
6472 given to Guile, as a list of strings.
6474 When using guile as a script interpreter, `command-line' returns the
6475 script's arguments; those processed by the interpreter (like `-s' or
6476 `-c') are omitted. (In other words, you get the normal, expected
6477 behavior.) Any application that uses scm_shell to process its
6478 command-line arguments gets this behavior as well.
6480 ** The new function `load-user-init' looks for a file called `.guile'
6481 in the user's home directory, and loads it if it exists. This is
6482 mostly for use by the code generated by scm_compile_shell_switches,
6483 but we thought it might also be useful in other circumstances.
6485 ** The new function `log10' returns the base-10 logarithm of its
6488 ** Changes to I/O functions
6490 *** The functions `read', `primitive-load', `read-and-eval!', and
6491 `primitive-load-path' no longer take optional arguments controlling
6492 case insensitivity and a `#' parser.
6494 Case sensitivity is now controlled by a read option called
6495 `case-insensitive'. The user can add new `#' syntaxes with the
6496 `read-hash-extend' function (see below).
6498 *** The new function `read-hash-extend' allows the user to change the
6499 syntax of Guile Scheme in a somewhat controlled way.
6501 (read-hash-extend CHAR PROC)
6502 When parsing S-expressions, if we read a `#' character followed by
6503 the character CHAR, use PROC to parse an object from the stream.
6504 If PROC is #f, remove any parsing procedure registered for CHAR.
6506 The reader applies PROC to two arguments: CHAR and an input port.
6508 *** The new functions read-delimited and read-delimited! provide a
6509 general mechanism for doing delimited input on streams.
6511 (read-delimited DELIMS [PORT HANDLE-DELIM])
6512 Read until we encounter one of the characters in DELIMS (a string),
6513 or end-of-file. PORT is the input port to read from; it defaults to
6514 the current input port. The HANDLE-DELIM parameter determines how
6515 the terminating character is handled; it should be one of the
6518 'trim omit delimiter from result
6519 'peek leave delimiter character in input stream
6520 'concat append delimiter character to returned value
6521 'split return a pair: (RESULT . TERMINATOR)
6523 HANDLE-DELIM defaults to 'peek.
6525 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6526 A side-effecting variant of `read-delimited'.
6528 The data is written into the string BUF at the indices in the
6529 half-open interval [START, END); the default interval is the whole
6530 string: START = 0 and END = (string-length BUF). The values of
6531 START and END must specify a well-defined interval in BUF, i.e.
6532 0 <= START <= END <= (string-length BUF).
6534 It returns NBYTES, the number of bytes read. If the buffer filled
6535 up without a delimiter character being found, it returns #f. If the
6536 port is at EOF when the read starts, it returns the EOF object.
6538 If an integer is returned (i.e., the read is successfully terminated
6539 by reading a delimiter character), then the HANDLE-DELIM parameter
6540 determines how to handle the terminating character. It is described
6541 above, and defaults to 'peek.
6543 (The descriptions of these functions were borrowed from the SCSH
6544 manual, by Olin Shivers and Brian Carlstrom.)
6546 *** The `%read-delimited!' function is the primitive used to implement
6547 `read-delimited' and `read-delimited!'.
6549 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6551 This returns a pair of values: (TERMINATOR . NUM-READ).
6552 - TERMINATOR describes why the read was terminated. If it is a
6553 character or the eof object, then that is the value that terminated
6554 the read. If it is #f, the function filled the buffer without finding
6555 a delimiting character.
6556 - NUM-READ is the number of characters read into BUF.
6558 If the read is successfully terminated by reading a delimiter
6559 character, then the gobble? parameter determines what to do with the
6560 terminating character. If true, the character is removed from the
6561 input stream; if false, the character is left in the input stream
6562 where a subsequent read operation will retrieve it. In either case,
6563 the character is also the first value returned by the procedure call.
6565 (The descriptions of this function was borrowed from the SCSH manual,
6566 by Olin Shivers and Brian Carlstrom.)
6568 *** The `read-line' and `read-line!' functions have changed; they now
6569 trim the terminator by default; previously they appended it to the
6570 returned string. For the old behavior, use (read-line PORT 'concat).
6572 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6573 take new optional START and END arguments, specifying the region of
6574 the array to read and write.
6576 *** The `ungetc-char-ready?' function has been removed. We feel it's
6577 inappropriate for an interface to expose implementation details this
6580 ** Changes to the Unix library and system call interface
6582 *** The new fcntl function provides access to the Unix `fcntl' system
6585 (fcntl PORT COMMAND VALUE)
6586 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6587 Values for COMMAND are:
6589 F_DUPFD duplicate a file descriptor
6590 F_GETFD read the descriptor's close-on-exec flag
6591 F_SETFD set the descriptor's close-on-exec flag to VALUE
6592 F_GETFL read the descriptor's flags, as set on open
6593 F_SETFL set the descriptor's flags, as set on open to VALUE
6594 F_GETOWN return the process ID of a socket's owner, for SIGIO
6595 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6596 FD_CLOEXEC not sure what this is
6598 For details, see the documentation for the fcntl system call.
6600 *** The arguments to `select' have changed, for compatibility with
6601 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6602 expected behavior. The MILLISECONDS parameter has been changed to
6603 MICROSECONDS, to more closely resemble the underlying system call.
6604 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6605 corresponding return set will be the same.
6607 *** The arguments to the `mknod' system call have changed. They are
6610 (mknod PATH TYPE PERMS DEV)
6611 Create a new file (`node') in the file system. PATH is the name of
6612 the file to create. TYPE is the kind of file to create; it should
6613 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6614 permission bits to give the newly created file. If TYPE is
6615 'block-special or 'char-special, DEV specifies which device the
6616 special file refers to; its interpretation depends on the kind of
6617 special file being created.
6619 *** The `fork' function has been renamed to `primitive-fork', to avoid
6620 clashing with various SCSH forks.
6622 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6623 and `recvfrom!'. They no longer accept a size for a second argument;
6624 you must pass a string to hold the received value. They no longer
6625 return the buffer. Instead, `recv' returns the length of the message
6626 received, and `recvfrom' returns a pair containing the packet's length
6627 and originating address.
6629 *** The file descriptor datatype has been removed, as have the
6630 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6631 We plan to replace these functions with a SCSH-compatible interface.
6633 *** The `create' function has been removed; it's just a special case
6636 *** There are new functions to break down process termination status
6637 values. In the descriptions below, STATUS is a value returned by
6640 (status:exit-val STATUS)
6641 If the child process exited normally, this function returns the exit
6642 code for the child process (i.e., the value passed to exit, or
6643 returned from main). If the child process did not exit normally,
6644 this function returns #f.
6646 (status:stop-sig STATUS)
6647 If the child process was suspended by a signal, this function
6648 returns the signal that suspended the child. Otherwise, it returns
6651 (status:term-sig STATUS)
6652 If the child process terminated abnormally, this function returns
6653 the signal that terminated the child. Otherwise, this function
6656 POSIX promises that exactly one of these functions will return true on
6657 a valid STATUS value.
6659 These functions are compatible with SCSH.
6661 *** There are new accessors and setters for the broken-out time vectors
6662 returned by `localtime', `gmtime', and that ilk. They are:
6664 Component Accessor Setter
6665 ========================= ============ ============
6666 seconds tm:sec set-tm:sec
6667 minutes tm:min set-tm:min
6668 hours tm:hour set-tm:hour
6669 day of the month tm:mday set-tm:mday
6670 month tm:mon set-tm:mon
6671 year tm:year set-tm:year
6672 day of the week tm:wday set-tm:wday
6673 day in the year tm:yday set-tm:yday
6674 daylight saving time tm:isdst set-tm:isdst
6675 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6676 name of time zone tm:zone set-tm:zone
6678 *** There are new accessors for the vectors returned by `uname',
6679 describing the host system:
6682 ============================================== ================
6683 name of the operating system implementation utsname:sysname
6684 network name of this machine utsname:nodename
6685 release level of the operating system utsname:release
6686 version level of the operating system utsname:version
6687 machine hardware platform utsname:machine
6689 *** There are new accessors for the vectors returned by `getpw',
6690 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6691 system's user database:
6694 ====================== =================
6695 user name passwd:name
6696 user password passwd:passwd
6699 real name passwd:gecos
6700 home directory passwd:dir
6701 shell program passwd:shell
6703 *** There are new accessors for the vectors returned by `getgr',
6704 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6705 system's group database:
6708 ======================= ============
6709 group name group:name
6710 group password group:passwd
6712 group members group:mem
6714 *** There are new accessors for the vectors returned by `gethost',
6715 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6719 ========================= ===============
6720 official name of host hostent:name
6721 alias list hostent:aliases
6722 host address type hostent:addrtype
6723 length of address hostent:length
6724 list of addresses hostent:addr-list
6726 *** There are new accessors for the vectors returned by `getnet',
6727 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6731 ========================= ===============
6732 official name of net netent:name
6733 alias list netent:aliases
6734 net number type netent:addrtype
6735 net number netent:net
6737 *** There are new accessors for the vectors returned by `getproto',
6738 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6742 ========================= ===============
6743 official protocol name protoent:name
6744 alias list protoent:aliases
6745 protocol number protoent:proto
6747 *** There are new accessors for the vectors returned by `getserv',
6748 `getservbyname', `getservbyport', and `getservent', describing
6752 ========================= ===============
6753 official service name servent:name
6754 alias list servent:aliases
6755 port number servent:port
6756 protocol to use servent:proto
6758 *** There are new accessors for the sockaddr structures returned by
6759 `accept', `getsockname', `getpeername', `recvfrom!':
6762 ======================================== ===============
6763 address format (`family') sockaddr:fam
6764 path, for file domain addresses sockaddr:path
6765 address, for internet domain addresses sockaddr:addr
6766 TCP or UDP port, for internet sockaddr:port
6768 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6769 `getprotoent', and `getservent' functions now return #f at the end of
6770 the user database. (They used to throw an exception.)
6772 Note that calling MUMBLEent function is equivalent to calling the
6773 corresponding MUMBLE function with no arguments.
6775 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6776 `setprotoent', and `setservent' routines now take no arguments.
6778 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6779 provide more useful information when they throw an exception.
6781 *** The `lnaof' function has been renamed to `inet-lnaof'.
6783 *** Guile now claims to have the `current-time' feature.
6785 *** The `mktime' function now takes an optional second argument ZONE,
6786 giving the time zone to use for the conversion. ZONE should be a
6787 string, in the same format as expected for the "TZ" environment variable.
6789 *** The `strptime' function now returns a pair (TIME . COUNT), where
6790 TIME is the parsed time as a vector, and COUNT is the number of
6791 characters from the string left unparsed. This function used to
6792 return the remaining characters as a string.
6794 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6795 The return value is now (SECONDS . MICROSECONDS); the fractional
6796 component is no longer expressed in "ticks".
6798 *** The `ticks/sec' constant has been removed, in light of the above change.
6800 * Changes to the gh_ interface
6802 ** gh_eval_str() now returns an SCM object which is the result of the
6805 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6808 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6809 and returns the array
6811 ** gh_scm2str0() is gone: there is no need to distinguish
6812 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6813 the user to interpret the data both ways.
6815 * Changes to the scm_ interface
6817 ** The new function scm_symbol_value0 provides an easy way to get a
6818 symbol's value from C code:
6820 SCM scm_symbol_value0 (char *NAME)
6821 Return the value of the symbol named by the null-terminated string
6822 NAME in the current module. If the symbol named NAME is unbound in
6823 the current module, return SCM_UNDEFINED.
6825 ** The new function scm_sysintern0 creates new top-level variables,
6826 without assigning them a value.
6828 SCM scm_sysintern0 (char *NAME)
6829 Create a new Scheme top-level variable named NAME. NAME is a
6830 null-terminated string. Return the variable's value cell.
6832 ** The function scm_internal_catch is the guts of catch. It handles
6833 all the mechanics of setting up a catch target, invoking the catch
6834 body, and perhaps invoking the handler if the body does a throw.
6836 The function is designed to be usable from C code, but is general
6837 enough to implement all the semantics Guile Scheme expects from throw.
6839 TAG is the catch tag. Typically, this is a symbol, but this function
6840 doesn't actually care about that.
6842 BODY is a pointer to a C function which runs the body of the catch;
6843 this is the code you can throw from. We call it like this:
6844 BODY (BODY_DATA, JMPBUF)
6846 BODY_DATA is just the BODY_DATA argument we received; we pass it
6847 through to BODY as its first argument. The caller can make
6848 BODY_DATA point to anything useful that BODY might need.
6849 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6850 which we have just created and initialized.
6852 HANDLER is a pointer to a C function to deal with a throw to TAG,
6853 should one occur. We call it like this:
6854 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6856 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6857 same idea as BODY_DATA above.
6858 THROWN_TAG is the tag that the user threw to; usually this is
6859 TAG, but it could be something else if TAG was #t (i.e., a
6860 catch-all), or the user threw to a jmpbuf.
6861 THROW_ARGS is the list of arguments the user passed to the THROW
6864 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6865 is just a pointer we pass through to HANDLER. We don't actually
6866 use either of those pointers otherwise ourselves. The idea is
6867 that, if our caller wants to communicate something to BODY or
6868 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6869 HANDLER can then use. Think of it as a way to make BODY and
6870 HANDLER closures, not just functions; MUMBLE_DATA points to the
6873 Of course, it's up to the caller to make sure that any data a
6874 MUMBLE_DATA needs is protected from GC. A common way to do this is
6875 to make MUMBLE_DATA a pointer to data stored in an automatic
6876 structure variable; since the collector must scan the stack for
6877 references anyway, this assures that any references in MUMBLE_DATA
6880 ** The new function scm_internal_lazy_catch is exactly like
6881 scm_internal_catch, except:
6883 - It does not unwind the stack (this is the major difference).
6884 - If handler returns, its value is returned from the throw.
6885 - BODY always receives #f as its JMPBUF argument (since there's no
6886 jmpbuf associated with a lazy catch, because we don't unwind the
6889 ** scm_body_thunk is a new body function you can pass to
6890 scm_internal_catch if you want the body to be like Scheme's `catch'
6891 --- a thunk, or a function of one argument if the tag is #f.
6893 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6894 contains the Scheme procedure to invoke as the body, and the tag
6895 we're catching. If the tag is #f, then we pass JMPBUF (created by
6896 scm_internal_catch) to the body procedure; otherwise, the body gets
6899 ** scm_handle_by_proc is a new handler function you can pass to
6900 scm_internal_catch if you want the handler to act like Scheme's catch
6901 --- call a procedure with the tag and the throw arguments.
6903 If the user does a throw to this catch, this function runs a handler
6904 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6905 variable holding the Scheme procedure object to invoke. It ought to
6906 be a pointer to an automatic variable (i.e., one living on the stack),
6907 or the procedure object should be otherwise protected from GC.
6909 ** scm_handle_by_message is a new handler function to use with
6910 `scm_internal_catch' if you want Guile to print a message and die.
6911 It's useful for dealing with throws to uncaught keys at the top level.
6913 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6914 message header to print; if zero, we use "guile" instead. That
6915 text is followed by a colon, then the message described by ARGS.
6917 ** The return type of scm_boot_guile is now void; the function does
6918 not return a value, and indeed, never returns at all.
6920 ** The new function scm_shell makes it easy for user applications to
6921 process command-line arguments in a way that is compatible with the
6922 stand-alone guile interpreter (which is in turn compatible with SCSH,
6925 To use the scm_shell function, first initialize any guile modules
6926 linked into your application, and then call scm_shell with the values
6927 of ARGC and ARGV your `main' function received. scm_shell will add
6928 any SCSH-style meta-arguments from the top of the script file to the
6929 argument vector, and then process the command-line arguments. This
6930 generally means loading a script file or starting up an interactive
6931 command interpreter. For details, see "Changes to the stand-alone
6934 ** The new functions scm_get_meta_args and scm_count_argv help you
6935 implement the SCSH-style meta-argument, `\'.
6937 char **scm_get_meta_args (int ARGC, char **ARGV)
6938 If the second element of ARGV is a string consisting of a single
6939 backslash character (i.e. "\\" in Scheme notation), open the file
6940 named by the following argument, parse arguments from it, and return
6941 the spliced command line. The returned array is terminated by a
6944 For details of argument parsing, see above, under "guile now accepts
6945 command-line arguments compatible with SCSH..."
6947 int scm_count_argv (char **ARGV)
6948 Count the arguments in ARGV, assuming it is terminated by a null
6951 For an example of how these functions might be used, see the source
6952 code for the function scm_shell in libguile/script.c.
6954 You will usually want to use scm_shell instead of calling this
6957 ** The new function scm_compile_shell_switches turns an array of
6958 command-line arguments into Scheme code to carry out the actions they
6959 describe. Given ARGC and ARGV, it returns a Scheme expression to
6960 evaluate, and calls scm_set_program_arguments to make any remaining
6961 command-line arguments available to the Scheme code. For example,
6962 given the following arguments:
6964 -e main -s ekko a speckled gecko
6966 scm_set_program_arguments will return the following expression:
6968 (begin (load "ekko") (main (command-line)) (quit))
6970 You will usually want to use scm_shell instead of calling this
6973 ** The function scm_shell_usage prints a usage message appropriate for
6974 an interpreter that uses scm_compile_shell_switches to handle its
6975 command-line arguments.
6977 void scm_shell_usage (int FATAL, char *MESSAGE)
6978 Print a usage message to the standard error output. If MESSAGE is
6979 non-zero, write it before the usage message, followed by a newline.
6980 If FATAL is non-zero, exit the process, using FATAL as the
6981 termination status. (If you want to be compatible with Guile,
6982 always use 1 as the exit status when terminating due to command-line
6985 You will usually want to use scm_shell instead of calling this
6988 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6989 expressions. It used to return SCM_EOL. Earth-shattering.
6991 ** The macros for declaring scheme objects in C code have been
6992 rearranged slightly. They are now:
6994 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6995 Declare a static SCM variable named C_NAME, and initialize it to
6996 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6997 be a C identifier, and SCHEME_NAME should be a C string.
6999 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7000 Just like SCM_SYMBOL, but make C_NAME globally visible.
7002 SCM_VCELL (C_NAME, SCHEME_NAME)
7003 Create a global variable at the Scheme level named SCHEME_NAME.
7004 Declare a static SCM variable named C_NAME, and initialize it to
7005 point to the Scheme variable's value cell.
7007 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7008 Just like SCM_VCELL, but make C_NAME globally visible.
7010 The `guile-snarf' script writes initialization code for these macros
7011 to its standard output, given C source code as input.
7013 The SCM_GLOBAL macro is gone.
7015 ** The scm_read_line and scm_read_line_x functions have been replaced
7016 by Scheme code based on the %read-delimited! procedure (known to C
7017 code as scm_read_delimited_x). See its description above for more
7020 ** The function scm_sys_open has been renamed to scm_open. It now
7021 returns a port instead of an FD object.
7023 * The dynamic linking support has changed. For more information, see
7024 libguile/DYNAMIC-LINKING.
7029 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7032 * Changes to the 'guile' program:
7034 ** Guile now loads some new files when it starts up. Guile first
7035 searches the load path for init.scm, and loads it if found. Then, if
7036 Guile is not being used to execute a script, and the user's home
7037 directory contains a file named `.guile', Guile loads that.
7039 ** You can now use Guile as a shell script interpreter.
7041 To paraphrase the SCSH manual:
7043 When Unix tries to execute an executable file whose first two
7044 characters are the `#!', it treats the file not as machine code to
7045 be directly executed by the native processor, but as source code
7046 to be executed by some interpreter. The interpreter to use is
7047 specified immediately after the #! sequence on the first line of
7048 the source file. The kernel reads in the name of the interpreter,
7049 and executes that instead. It passes the interpreter the source
7050 filename as its first argument, with the original arguments
7051 following. Consult the Unix man page for the `exec' system call
7052 for more information.
7054 Now you can use Guile as an interpreter, using a mechanism which is a
7055 compatible subset of that provided by SCSH.
7057 Guile now recognizes a '-s' command line switch, whose argument is the
7058 name of a file of Scheme code to load. It also treats the two
7059 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7060 to make a file of Scheme code directly executable by Unix, insert the
7061 following two lines at the top of the file:
7063 #!/usr/local/bin/guile -s
7066 Guile treats the argument of the `-s' command-line switch as the name
7067 of a file of Scheme code to load, and treats the sequence `#!' as the
7068 start of a block comment, terminated by `!#'.
7070 For example, here's a version of 'echo' written in Scheme:
7072 #!/usr/local/bin/guile -s
7074 (let loop ((args (cdr (program-arguments))))
7077 (display (car args))
7078 (if (pair? (cdr args))
7080 (loop (cdr args)))))
7083 Why does `#!' start a block comment terminated by `!#', instead of the
7084 end of the line? That is the notation SCSH uses, and although we
7085 don't yet support the other SCSH features that motivate that choice,
7086 we would like to be backward-compatible with any existing Guile
7087 scripts once we do. Furthermore, if the path to Guile on your system
7088 is too long for your kernel, you can start the script with this
7092 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7095 Note that some very old Unix systems don't support the `#!' syntax.
7098 ** You can now run Guile without installing it.
7100 Previous versions of the interactive Guile interpreter (`guile')
7101 couldn't start up unless Guile's Scheme library had been installed;
7102 they used the value of the environment variable `SCHEME_LOAD_PATH'
7103 later on in the startup process, but not to find the startup code
7104 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7107 To run Guile without installing it, build it in the normal way, and
7108 then set the environment variable `SCHEME_LOAD_PATH' to a
7109 colon-separated list of directories, including the top-level directory
7110 of the Guile sources. For example, if you unpacked Guile so that the
7111 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7114 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7117 ** Guile's read-eval-print loop no longer prints #<unspecified>
7118 results. If the user wants to see this, she can evaluate the
7119 expression (assert-repl-print-unspecified #t), perhaps in her startup
7122 ** Guile no longer shows backtraces by default when an error occurs;
7123 however, it does display a message saying how to get one, and how to
7124 request that they be displayed by default. After an error, evaluate
7126 to see a backtrace, and
7127 (debug-enable 'backtrace)
7128 to see them by default.
7132 * Changes to Guile Scheme:
7134 ** Guile now distinguishes between #f and the empty list.
7136 This is for compatibility with the IEEE standard, the (possibly)
7137 upcoming Revised^5 Report on Scheme, and many extant Scheme
7140 Guile used to have #f and '() denote the same object, to make Scheme's
7141 type system more compatible with Emacs Lisp's. However, the change
7142 caused too much trouble for Scheme programmers, and we found another
7143 way to reconcile Emacs Lisp with Scheme that didn't require this.
7146 ** Guile's delq, delv, delete functions, and their destructive
7147 counterparts, delq!, delv!, and delete!, now remove all matching
7148 elements from the list, not just the first. This matches the behavior
7149 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7150 functions which inspired them.
7152 I recognize that this change may break code in subtle ways, but it
7153 seems best to make the change before the FSF's first Guile release,
7157 ** The compiled-library-path function has been deleted from libguile.
7159 ** The facilities for loading Scheme source files have changed.
7161 *** The variable %load-path now tells Guile which directories to search
7162 for Scheme code. Its value is a list of strings, each of which names
7165 *** The variable %load-extensions now tells Guile which extensions to
7166 try appending to a filename when searching the load path. Its value
7167 is a list of strings. Its default value is ("" ".scm").
7169 *** (%search-load-path FILENAME) searches the directories listed in the
7170 value of the %load-path variable for a Scheme file named FILENAME,
7171 with all the extensions listed in %load-extensions. If it finds a
7172 match, then it returns its full filename. If FILENAME is absolute, it
7173 returns it unchanged. Otherwise, it returns #f.
7175 %search-load-path will not return matches that refer to directories.
7177 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7178 uses %seach-load-path to find a file named FILENAME, and loads it if
7179 it finds it. If it can't read FILENAME for any reason, it throws an
7182 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7185 *** load uses the same searching semantics as primitive-load.
7187 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7188 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7189 path, and load-module-with-path have been deleted. The functions
7190 above should serve their purposes.
7192 *** If the value of the variable %load-hook is a procedure,
7193 `primitive-load' applies its value to the name of the file being
7194 loaded (without the load path directory name prepended). If its value
7195 is #f, it is ignored. Otherwise, an error occurs.
7197 This is mostly useful for printing load notification messages.
7200 ** The function `eval!' is no longer accessible from the scheme level.
7201 We can't allow operations which introduce glocs into the scheme level,
7202 because Guile's type system can't handle these as data. Use `eval' or
7203 `read-and-eval!' (see below) as replacement.
7205 ** The new function read-and-eval! reads an expression from PORT,
7206 evaluates it, and returns the result. This is more efficient than
7207 simply calling `read' and `eval', since it is not necessary to make a
7208 copy of the expression for the evaluator to munge.
7210 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7211 for the `read' function.
7214 ** The function `int?' has been removed; its definition was identical
7215 to that of `integer?'.
7217 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7218 use the R4RS names for these functions.
7220 ** The function object-properties no longer returns the hash handle;
7221 it simply returns the object's property list.
7223 ** Many functions have been changed to throw errors, instead of
7224 returning #f on failure. The point of providing exception handling in
7225 the language is to simplify the logic of user code, but this is less
7226 useful if Guile's primitives don't throw exceptions.
7228 ** The function `fileno' has been renamed from `%fileno'.
7230 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7233 * Changes to Guile's C interface:
7235 ** The library's initialization procedure has been simplified.
7236 scm_boot_guile now has the prototype:
7238 void scm_boot_guile (int ARGC,
7240 void (*main_func) (),
7243 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7244 MAIN_FUNC should do all the work of the program (initializing other
7245 packages, reading user input, etc.) before returning. When MAIN_FUNC
7246 returns, call exit (0); this function never returns. If you want some
7247 other exit value, MAIN_FUNC may call exit itself.
7249 scm_boot_guile arranges for program-arguments to return the strings
7250 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7251 scm_set_program_arguments with the final list, so Scheme code will
7252 know which arguments have been processed.
7254 scm_boot_guile establishes a catch-all catch handler which prints an
7255 error message and exits the process. This means that Guile exits in a
7256 coherent way when system errors occur and the user isn't prepared to
7257 handle it. If the user doesn't like this behavior, they can establish
7258 their own universal catcher in MAIN_FUNC to shadow this one.
7260 Why must the caller do all the real work from MAIN_FUNC? The garbage
7261 collector assumes that all local variables of type SCM will be above
7262 scm_boot_guile's stack frame on the stack. If you try to manipulate
7263 SCM values after this function returns, it's the luck of the draw
7264 whether the GC will be able to find the objects you allocate. So,
7265 scm_boot_guile function exits, rather than returning, to discourage
7266 people from making that mistake.
7268 The IN, OUT, and ERR arguments were removed; there are other
7269 convenient ways to override these when desired.
7271 The RESULT argument was deleted; this function should never return.
7273 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7277 ** Guile's header files should no longer conflict with your system's
7280 In order to compile code which #included <libguile.h>, previous
7281 versions of Guile required you to add a directory containing all the
7282 Guile header files to your #include path. This was a problem, since
7283 Guile's header files have names which conflict with many systems'
7286 Now only <libguile.h> need appear in your #include path; you must
7287 refer to all Guile's other header files as <libguile/mumble.h>.
7288 Guile's installation procedure puts libguile.h in $(includedir), and
7289 the rest in $(includedir)/libguile.
7292 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7293 have been added to the Guile library.
7295 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7296 OBJ will not be freed, even if all other references are dropped,
7297 until someone does scm_unprotect_object (OBJ). Both functions
7300 Note that calls to scm_protect_object do not nest. You can call
7301 scm_protect_object any number of times on a given object, and the
7302 next call to scm_unprotect_object will unprotect it completely.
7304 Basically, scm_protect_object and scm_unprotect_object just
7305 maintain a list of references to things. Since the GC knows about
7306 this list, all objects it mentions stay alive. scm_protect_object
7307 adds its argument to the list; scm_unprotect_object remove its
7308 argument from the list.
7311 ** scm_eval_0str now returns the value of the last expression
7314 ** The new function scm_read_0str reads an s-expression from a
7315 null-terminated string, and returns it.
7317 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7318 to a Scheme port object.
7320 ** The new function `scm_set_program_arguments' allows C code to set
7321 the value returned by the Scheme `program-arguments' function.
7326 * Guile no longer includes sophisticated Tcl/Tk support.
7328 The old Tcl/Tk support was unsatisfying to us, because it required the
7329 user to link against the Tcl library, as well as Tk and Guile. The
7330 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7331 referring to widgets by names, rather than exporting widgets to Scheme
7332 code as a special datatype.
7334 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7335 maintainers described some very interesting changes in progress to the
7336 Tcl/Tk internals, which would facilitate clean interfaces between lone
7337 Tk and other interpreters --- even for garbage-collected languages
7338 like Scheme. They expected the new Tk to be publicly available in the
7341 Since it seems that Guile might soon have a new, cleaner interface to
7342 lone Tk, and that the old Guile/Tk glue code would probably need to be
7343 completely rewritten, we (Jim Blandy and Richard Stallman) have
7344 decided not to support the old code. We'll spend the time instead on
7345 a good interface to the newer Tk, as soon as it is available.
7347 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7350 Copyright information:
7352 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7354 Permission is granted to anyone to make or distribute verbatim copies
7355 of this document as received, in any medium, provided that the
7356 copyright notice and this permission notice are preserved,
7357 thus giving the recipient permission to redistribute in turn.
7359 Permission is granted to distribute modified versions
7360 of this document, or of portions of it,
7361 under the above conditions, provided also that they
7362 carry prominent notices stating who last changed them.
7367 paragraph-separate: "[
\f]*$"