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 ** The GH interface (deprecated in version 1.6, 2001) was removed.
37 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
39 This makes these internal functions technically not callable from
42 ** Functions for handling `scm_option' now no longer require an argument
43 indicating length of the `scm_t_option' array.
47 Changes in 1.8.6 (since 1.8.5)
49 * New features (see the manual for details)
51 ** New convenience function `scm_c_symbol_length ()'
53 ** Single stepping through code from Emacs
55 When you use GDS to evaluate Scheme code from Emacs, you can now use
56 `C-u' to indicate that you want to single step through that code. See
57 `Evaluating Scheme Code' in the manual for more details.
59 * Changes to the distribution
61 ** Automake's `AM_MAINTAINER_MODE' is no longer used
63 Thus, the `--enable-maintainer-mode' configure option is no longer
64 available: Guile is now always configured in "maintainer mode".
68 ** `symbol->string' now returns a read-only string, as per R5RS
69 ** Literal strings as returned by `read' are now read-only, as per R5RS
70 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
71 ** `guile-config link' now prints `-L$libdir' before `-lguile'
72 ** Fix memory corruption involving GOOPS' `class-redefinition'
73 ** Fix possible deadlock in `mutex-lock'
74 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
75 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
76 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
77 ** Fix misleading output from `(help rationalize)'
78 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
79 ** Fix `gcd' when called with a single, negative argument.
82 Changes in 1.8.5 (since 1.8.4)
84 * Infrastructure changes
86 ** Guile repository switched from CVS to Git
88 The new repository can be accessed using
89 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
90 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
92 ** Add support for `pkg-config'
94 See "Autoconf Support" in the manual for details.
96 * New modules (see the manual for details)
100 * New features (see the manual for details)
102 ** New `postfix' read option, for SRFI-88 keyword syntax
103 ** Some I/O primitives have been inlined, which improves I/O performance
104 ** New object-based traps infrastructure
106 This is a GOOPS-based infrastructure that builds on Guile's low-level
107 evaluator trap calls and facilitates the development of debugging
108 features like single-stepping, breakpoints, tracing and profiling.
109 See the `Traps' node of the manual for details.
111 ** New support for working on Guile code from within Emacs
113 Guile now incorporates the `GDS' library (previously distributed
114 separately) for working on Guile code from within Emacs. See the
115 `Using Guile In Emacs' node of the manual for details.
119 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
120 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
122 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
123 would trigger an unbound variable error for `match:andmap'.
125 ** `(oop goops describe)' now properly provides the `describe' feature
126 ** Fixed `args-fold' from `(srfi srfi-37)'
128 Previously, parsing short option names of argument-less options would
129 lead to a stack overflow.
131 ** `(srfi srfi-35)' is now visible through `cond-expand'
132 ** Fixed type-checking for the second argument of `eval'
133 ** Fixed type-checking for SRFI-1 `partition'
134 ** Fixed `struct-ref' and `struct-set!' on "light structs"
135 ** Honor struct field access rights in GOOPS
136 ** Changed the storage strategy of source properties, which fixes a deadlock
137 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
138 ** Fixed build issue for GNU/Linux on IA64
139 ** Fixed build issues on NetBSD 1.6
140 ** Fixed build issue on Solaris 2.10 x86_64
141 ** Fixed build issue with DEC/Compaq/HP's compiler
142 ** Fixed `scm_from_complex_double' build issue on FreeBSD
143 ** Fixed `alloca' build issue on FreeBSD 6
144 ** Removed use of non-portable makefile constructs
145 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
146 ** Make sure all tests honor `$TMPDIR'
148 * Changes to the distribution
152 We've started collecting Frequently Asked Questions (FAQ), and will
153 distribute these (with answers!) in future Guile releases.
156 Changes in 1.8.4 (since 1.8.3)
160 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
161 ** Fixed a segmentation fault which occurred when displaying the
162 backtrace of a stack with a promise object (made by `delay') in it.
163 ** Make `accept' leave guile mode while blocking
164 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
165 ** Fixed a build problem on AIX (use of func_data identifier)
166 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
167 called with an associator proc that returns neither a pair nor #f.
168 ** Secondary threads now always return a valid module for (current-module).
169 ** Avoid MacOS build problems caused by incorrect combination of "64"
170 system and library calls.
171 ** `guile-snarf' now honors `$TMPDIR'
172 ** `guile-config compile' now reports CPPFLAGS used at compile-time
173 ** Fixed build with Sun Studio (Solaris 9)
174 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
175 uniform vectors on AIX.
176 ** Fixed a deadlock that occurs upon GC with multiple threads.
177 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
178 ** Fixed autotool-derived build problems on AIX 6.1.
179 ** Fixed NetBSD/alpha support
180 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
182 * New modules (see the manual for details)
186 * Documentation fixes and improvements
188 ** Removed premature breakpoint documentation
190 The features described are not available in the series of 1.8.x
191 releases, so the documentation was misleading and has been removed.
193 ** More about Guile's default *random-state* variable
195 ** GOOPS: more about how to use `next-method'
197 * Changes to the distribution
199 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
201 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
202 General Public License, and the few incorrect files have now been
203 fixed to agree with the rest of the Guile distribution.
205 ** Removed unnecessary extra copies of COPYING*
207 The distribution now contains a single COPYING.LESSER at its top level.
210 Changes in 1.8.3 (since 1.8.2)
212 * New modules (see the manual for details)
219 ** The `(ice-9 slib)' module now works as expected
220 ** Expressions like "(set! 'x #t)" no longer yield a crash
221 ** Warnings about duplicate bindings now go to stderr
222 ** A memory leak in `make-socket-address' was fixed
223 ** Alignment issues (e.g., on SPARC) in network routines were fixed
224 ** A threading issue that showed up at least on NetBSD was fixed
225 ** Build problems on Solaris and IRIX fixed
227 * Implementation improvements
229 ** The reader is now faster, which reduces startup time
230 ** Procedures returned by `record-accessor' and `record-modifier' are faster
234 Changes in 1.8.2 (since 1.8.1):
236 * New procedures (see the manual for details)
238 ** set-program-arguments
243 ** Fractions were not `equal?' if stored in unreduced form.
244 (A subtle problem, since printing a value reduced it, making it work.)
245 ** srfi-60 `copy-bit' failed on 64-bit systems
246 ** "guile --use-srfi" option at the REPL can replace core functions
247 (Programs run with that option were ok, but in the interactive REPL
248 the core bindings got priority, preventing SRFI replacements or
250 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
251 ** `kill' on mingw throws an error for a PID other than oneself
252 ** Procedure names are attached to procedure-with-setters
253 ** Array read syntax works with negative lower bound
254 ** `array-in-bounds?' fix if an array has different lower bounds on each index
255 ** `*' returns exact 0 for "(* inexact 0)"
256 This follows what it always did for "(* 0 inexact)".
257 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
258 ** SRFI-19: `date->julian-day' did not account for timezone offset
259 ** `ttyname' no longer crashes when passed a non-tty argument
260 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
261 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
262 ** GOOPS: Fixed a bug in `method-more-specific?'
263 ** Build problems on Solaris fixed
264 ** Build problems on HP-UX IA64 fixed
265 ** Build problems on MinGW fixed
268 Changes in 1.8.1 (since 1.8.0):
270 * LFS functions are now used to access 64-bit files on 32-bit systems.
272 * New procedures (see the manual for details)
274 ** primitive-_exit - [Scheme] the-root-module
275 ** scm_primitive__exit - [C]
276 ** make-completion-function - [Scheme] (ice-9 readline)
277 ** scm_c_locale_stringn_to_number - [C]
278 ** scm_srfi1_append_reverse [C]
279 ** scm_srfi1_append_reverse_x [C]
285 * New `(ice-9 i18n)' module (see the manual for details)
289 ** Build problems have been fixed on MacOS, SunOS, and QNX.
291 ** `strftime' fix sign of %z timezone offset.
293 ** A one-dimensional array can now be 'equal?' to a vector.
295 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
297 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
299 ** `record-accessor' and `record-modifier' now have strict type checks.
301 Record accessor and modifier procedures now throw an error if the
302 record type of the record they're given is not the type expected.
303 (Previously accessors returned #f and modifiers silently did nothing).
305 ** It is now OK to use both autoload and use-modules on a given module.
307 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
309 Previously there was no checking on primatives like make-vector that
310 accept "one or two" arguments. Now there is.
312 ** The srfi-1 assoc function now calls its equality predicate properly.
314 Previously srfi-1 assoc would call the equality predicate with the key
315 last. According to the SRFI, the key should be first.
317 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
319 ** The array-set! procedure no longer segfaults when given a bit vector.
321 ** Bugs in make-shared-array have been fixed.
323 ** string<? and friends now follow char<? etc order on 8-bit chars.
325 ** The format procedure now handles inf and nan values for ~f correctly.
327 ** exact->inexact should no longer overflow when given certain large fractions.
329 ** srfi-9 accessor and modifier procedures now have strict record type checks.
331 This matches the srfi-9 specification.
333 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
335 Previously the (ice-9 ftw) procedures would ignore any file that had
336 the same inode number as a file they had already seen, even if that
337 file was on a different device.
340 Changes in 1.8.0 (changes since the 1.6.x series):
342 * Changes to the distribution
344 ** Guile is now licensed with the GNU Lesser General Public License.
346 ** The manual is now licensed with the GNU Free Documentation License.
348 ** Guile now requires GNU MP (http://swox.com/gmp).
350 Guile now uses the GNU MP library for arbitrary precision arithmetic.
352 ** Guile now has separate private and public configuration headers.
354 That is, things like HAVE_STRING_H no longer leak from Guile's
357 ** Guile now provides and uses an "effective" version number.
359 Guile now provides scm_effective_version and effective-version
360 functions which return the "effective" version number. This is just
361 the normal full version string without the final micro-version number,
362 so the current effective-version is "1.8". The effective version
363 should remain unchanged during a stable series, and should be used for
364 items like the versioned share directory name
365 i.e. /usr/share/guile/1.8.
367 Providing an unchanging version number during a stable release for
368 things like the versioned share directory can be particularly
369 important for Guile "add-on" packages, since it provides a directory
370 that they can install to that won't be changed out from under them
371 with each micro release during a stable series.
373 ** Thread implementation has changed.
375 When you configure "--with-threads=null", you will get the usual
376 threading API (call-with-new-thread, make-mutex, etc), but you can't
377 actually create new threads. Also, "--with-threads=no" is now
378 equivalent to "--with-threads=null". This means that the thread API
379 is always present, although you might not be able to create new
382 When you configure "--with-threads=pthreads" or "--with-threads=yes",
383 you will get threads that are implemented with the portable POSIX
384 threads. These threads can run concurrently (unlike the previous
385 "coop" thread implementation), but need to cooperate for things like
388 The default is "pthreads", unless your platform doesn't have pthreads,
389 in which case "null" threads are used.
391 See the manual for details, nodes "Initialization", "Multi-Threading",
392 "Blocking", and others.
394 ** There is the new notion of 'discouraged' features.
396 This is a milder form of deprecation.
398 Things that are discouraged should not be used in new code, but it is
399 OK to leave them in old code for now. When a discouraged feature is
400 used, no warning message is printed like there is for 'deprecated'
401 features. Also, things that are merely discouraged are nevertheless
402 implemented efficiently, while deprecated features can be very slow.
404 You can omit discouraged features from libguile by configuring it with
405 the '--disable-discouraged' option.
407 ** Deprecation warnings can be controlled at run-time.
409 (debug-enable 'warn-deprecated) switches them on and (debug-disable
410 'warn-deprecated) switches them off.
412 ** Support for SRFI 61, extended cond syntax for multiple values has
415 This SRFI is always available.
417 ** Support for require-extension, SRFI-55, has been added.
419 The SRFI-55 special form `require-extension' has been added. It is
420 available at startup, and provides a portable way to load Scheme
421 extensions. SRFI-55 only requires support for one type of extension,
422 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
425 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
427 The (srfi srfi-26) module is an implementation of SRFI-26 which
428 provides the `cut' and `cute' syntax. These may be used to specialize
429 parameters without currying.
431 ** New module (srfi srfi-31)
433 This is an implementation of SRFI-31 which provides a special form
434 `rec' for recursive evaluation.
436 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
437 been merged with the core, making their functionality always
440 The modules are still available, tho, and you could use them together
441 with a renaming import, for example.
443 ** Guile no longer includes its own version of libltdl.
445 The official version is good enough now.
447 ** The --enable-htmldoc option has been removed from 'configure'.
449 Support for translating the documentation into HTML is now always
450 provided. Use 'make html'.
452 ** New module (ice-9 serialize):
454 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
455 don't trust the thread safety of most of your program, but where you
456 have some section(s) of code which you consider can run in parallel to
457 other sections. See ice-9/serialize.scm for more information.
459 ** The configure option '--disable-arrays' has been removed.
461 Support for arrays and uniform numeric arrays is now always included
464 * Changes to the stand-alone interpreter
466 ** New command line option `-L'.
468 This option adds a directory to the front of the load path.
470 ** New command line option `--no-debug'.
472 Specifying `--no-debug' on the command line will keep the debugging
473 evaluator turned off, even for interactive sessions.
475 ** User-init file ~/.guile is now loaded with the debugging evaluator.
477 Previously, the normal evaluator would have been used. Using the
478 debugging evaluator gives better error messages.
480 ** The '-e' option now 'read's its argument.
482 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
483 be used with '-e'. For example, you can now write a script like
486 exec guile -e '(@ (demo) main)' -s "$0" "$@"
489 (define-module (demo)
493 (format #t "Demo: ~a~%" args))
496 * Changes to Scheme functions and syntax
498 ** Guardians have changed back to their original semantics
500 Guardians now behave like described in the paper by Dybvig et al. In
501 particular, they no longer make guarantees about the order in which
502 they return objects, and they can no longer be greedy.
504 They no longer drop cyclic data structures.
506 The C function scm_make_guardian has been changed incompatibly and no
507 longer takes the 'greedy_p' argument.
509 ** New function hashx-remove!
511 This function completes the set of 'hashx' functions.
513 ** The concept of dynamic roots has been factored into continuation
514 barriers and dynamic states.
516 Each thread has a current dynamic state that carries the values of the
517 fluids. You can create and copy dynamic states and use them as the
518 second argument for 'eval'. See "Fluids and Dynamic States" in the
521 To restrict the influence that captured continuations can have on the
522 control flow, you can errect continuation barriers. See "Continuation
523 Barriers" in the manual.
525 The function call-with-dynamic-root now essentially temporarily
526 installs a new dynamic state and errects a continuation barrier.
528 ** The default load path no longer includes "." at the end.
530 Automatically loading modules from the current directory should not
531 happen by default. If you want to allow it in a more controlled
532 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
535 ** The uniform vector and array support has been overhauled.
537 It now complies with SRFI-4 and the weird prototype based uniform
538 array creation has been deprecated. See the manual for more details.
540 Some non-compatible changes have been made:
541 - characters can no longer be stored into byte arrays.
542 - strings and bit vectors are no longer considered to be uniform numeric
544 - array-rank throws an error for non-arrays instead of returning zero.
545 - array-ref does no longer accept non-arrays when no indices are given.
547 There is the new notion of 'generalized vectors' and corresponding
548 procedures like 'generalized-vector-ref'. Generalized vectors include
549 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
551 Arrays use generalized vectors as their storage, so that you still
552 have arrays of characters, bits, etc. However, uniform-array-read!
553 and uniform-array-write can no longer read/write strings and
556 ** There is now support for copy-on-write substrings, mutation-sharing
557 substrings and read-only strings.
559 Three new procedures are related to this: substring/shared,
560 substring/copy, and substring/read-only. See the manual for more
563 ** Backtraces will now highlight the value that caused the error.
565 By default, these values are enclosed in "{...}", such as in this
574 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
575 <unnamed port>:1:1: Wrong type (expecting pair): a
576 ABORT: (wrong-type-arg)
578 The prefix and suffix used for highlighting can be set via the two new
579 printer options 'highlight-prefix' and 'highlight-suffix'. For
580 example, putting this into ~/.guile will output the bad value in bold
583 (print-set! highlight-prefix "\x1b[1m")
584 (print-set! highlight-suffix "\x1b[22m")
587 ** 'gettext' support for internationalization has been added.
589 See the manual for details.
591 ** New syntax '@' and '@@':
593 You can now directly refer to variables exported from a module by
596 (@ MODULE-NAME VARIABLE-NAME)
598 For example (@ (ice-9 pretty-print) pretty-print) will directly access
599 the pretty-print variable exported from the (ice-9 pretty-print)
600 module. You don't need to 'use' that module first. You can also use
601 '@' as a target of 'set!', as in (set! (@ mod var) val).
603 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
604 but it can also access variables that have not been exported. It is
605 intended only for kluges and temporary fixes and for debugging, not
608 ** Keyword syntax has been made more disciplined.
610 Previously, the name of a keyword was read as a 'token' but printed as
611 a symbol. Now, it is read as a general Scheme datum which must be a
622 ERROR: In expression (a b c):
626 ERROR: Unbound variable: foo
631 ERROR: Wrong type (expecting symbol): 12
635 ERROR: Wrong type (expecting symbol): (a b c)
639 ** The printing of symbols that might look like keywords can be
642 The new printer option 'quote-keywordish-symbols' controls how symbols
643 are printed that have a colon as their first or last character. The
644 default now is to only quote a symbol with #{...}# when the read
645 option 'keywords' is not '#f'. Thus:
647 guile> (define foo (string->symbol ":foo"))
648 guile> (read-set! keywords #f)
651 guile> (read-set! keywords 'prefix)
654 guile> (print-set! quote-keywordish-symbols #f)
658 ** 'while' now provides 'break' and 'continue'
660 break and continue were previously bound in a while loop, but not
661 documented, and continue didn't quite work properly. The undocumented
662 parameter to break which gave a return value for the while has been
665 ** 'call-with-current-continuation' is now also available under the name
668 ** The module system now checks for duplicate bindings.
670 The module system now can check for name conflicts among imported
673 The behavior can be controlled by specifying one or more 'duplicates'
674 handlers. For example, to make Guile return an error for every name
682 The new default behavior of the module system when a name collision
683 has been detected is to
685 1. Give priority to bindings marked as a replacement.
686 2. Issue a warning (different warning if overriding core binding).
687 3. Give priority to the last encountered binding (this corresponds to
690 If you want the old behavior back without replacements or warnings you
693 (default-duplicate-binding-handler 'last)
695 to your .guile init file.
697 ** New define-module option: :replace
699 :replace works as :export, but, in addition, marks the binding as a
702 A typical example is `format' in (ice-9 format) which is a replacement
703 for the core binding `format'.
705 ** Adding prefixes to imported bindings in the module system
707 There is now a new :use-module option :prefix. It can be used to add
708 a prefix to all imported bindings.
711 :use-module ((bar) :prefix bar:))
713 will import all bindings exported from bar, but rename them by adding
716 ** Conflicting generic functions can be automatically merged.
718 When two imported bindings conflict and they are both generic
719 functions, the two functions can now be merged automatically. This is
720 activated with the 'duplicates' handler 'merge-generics'.
722 ** New function: effective-version
724 Returns the "effective" version number. This is just the normal full
725 version string without the final micro-version number. See "Changes
726 to the distribution" above.
728 ** New threading functions: parallel, letpar, par-map, and friends
730 These are convenient ways to run calculations in parallel in new
731 threads. See "Parallel forms" in the manual for details.
733 ** New function 'try-mutex'.
735 This function will attempt to lock a mutex but will return immediately
736 instead of blocking and indicate failure.
738 ** Waiting on a condition variable can have a timeout.
740 The function 'wait-condition-variable' now takes a third, optional
741 argument that specifies the point in time where the waiting should be
744 ** New function 'broadcast-condition-variable'.
746 ** New functions 'all-threads' and 'current-thread'.
748 ** Signals and system asyncs work better with threads.
750 The function 'sigaction' now takes a fourth, optional, argument that
751 specifies the thread that the handler should run in. When the
752 argument is omitted, the handler will run in the thread that called
755 Likewise, 'system-async-mark' takes a second, optional, argument that
756 specifies the thread that the async should run in. When it is
757 omitted, the async will run in the thread that called
760 C code can use the new functions scm_sigaction_for_thread and
761 scm_system_async_mark_for_thread to pass the new thread argument.
763 When a thread blocks on a mutex, a condition variable or is waiting
764 for IO to be possible, it will still execute system asyncs. This can
765 be used to interrupt such a thread by making it execute a 'throw', for
768 ** The function 'system-async' is deprecated.
770 You can now pass any zero-argument procedure to 'system-async-mark'.
771 The function 'system-async' will just return its argument unchanged
774 ** New functions 'call-with-blocked-asyncs' and
775 'call-with-unblocked-asyncs'
777 The expression (call-with-blocked-asyncs PROC) will call PROC and will
778 block execution of system asyncs for the current thread by one level
779 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
780 procedure and will unblock the execution of system asyncs by one
781 level for the current thread.
783 Only system asyncs are affected by these functions.
785 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
787 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
788 instead. Those functions are easier to use correctly and can be
791 ** New function 'unsetenv'.
793 ** New macro 'define-syntax-public'.
795 It works like 'define-syntax' and also exports the defined macro (but
798 ** There is support for Infinity and NaNs.
800 Following PLT Scheme, Guile can now work with infinite numbers, and
803 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
804 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
805 "+nan.0"). These numbers are inexact and have no exact counterpart.
807 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
808 sign of the dividend. The infinities are integers, and they answer #t
809 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
810 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
821 ERROR: Numerical overflow
823 Two new predicates 'inf?' and 'nan?' can be used to test for the
826 ** Inexact zero can have a sign.
828 Guile can now distinguish between plus and minus inexact zero, if your
829 platform supports this, too. The two zeros are equal according to
830 '=', but not according to 'eqv?'. For example
841 ** Guile now has exact rationals.
843 Guile can now represent fractions such as 1/3 exactly. Computing with
844 them is also done exactly, of course:
849 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
852 For example: (floor 2) now returns an exact 2 where in the past it
853 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
855 ** inexact->exact no longer returns only integers.
857 Without exact rationals, the closest exact number was always an
858 integer, but now inexact->exact returns the fraction that is exactly
859 equal to a floating point number. For example:
861 (inexact->exact 1.234)
862 => 694680242521899/562949953421312
864 When you want the old behavior, use 'round' explicitly:
866 (inexact->exact (round 1.234))
869 ** New function 'rationalize'.
871 This function finds a simple fraction that is close to a given real
872 number. For example (and compare with inexact->exact above):
874 (rationalize (inexact->exact 1.234) 1/2000)
877 Note that, as required by R5RS, rationalize returns only then an exact
878 result when both its arguments are exact.
880 ** 'odd?' and 'even?' work also for inexact integers.
882 Previously, (odd? 1.0) would signal an error since only exact integers
883 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
884 returns #f and (odd? 1.5) signals an error.
886 ** Guile now has uninterned symbols.
888 The new function 'make-symbol' will return an uninterned symbol. This
889 is a symbol that is unique and is guaranteed to remain unique.
890 However, uninterned symbols can not yet be read back in.
892 Use the new function 'symbol-interned?' to check whether a symbol is
895 ** pretty-print has more options.
897 The function pretty-print from the (ice-9 pretty-print) module can now
898 also be invoked with keyword arguments that control things like
899 maximum output width. See the manual for details.
901 ** Variables have no longer a special behavior for `equal?'.
903 Previously, comparing two variables with `equal?' would recursivly
904 compare their values. This is no longer done. Variables are now only
905 `equal?' if they are `eq?'.
907 ** `(begin)' is now valid.
909 You can now use an empty `begin' form. It will yield #<unspecified>
910 when evaluated and simply be ignored in a definition context.
912 ** Deprecated: procedure->macro
914 Change your code to use 'define-macro' or r5rs macros. Also, be aware
915 that macro expansion will not be done during evaluation, but prior to
918 ** Soft ports now allow a `char-ready?' procedure
920 The vector argument to `make-soft-port' can now have a length of
921 either 5 or 6. (Previously the length had to be 5.) The optional 6th
922 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
923 that returns the number of characters that can be read immediately
924 without the soft port blocking.
926 ** Deprecated: undefine
928 There is no replacement for undefine.
930 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
931 have been discouraged.
933 They are relics from a time where a keyword like #:foo was used
934 directly as a Tcl option "-foo" and thus keywords were internally
935 stored as a symbol with a starting dash. We now store a symbol
938 Use symbol->keyword and keyword->symbol instead.
940 ** The `cheap' debug option is now obsolete
942 Evaluator trap calls are now unconditionally "cheap" - in other words,
943 they pass a debug object to the trap handler rather than a full
944 continuation. The trap handler code can capture a full continuation
945 by using `call-with-current-continuation' in the usual way, if it so
948 The `cheap' option is retained for now so as not to break existing
949 code which gets or sets it, but setting it now has no effect. It will
950 be removed in the next major Guile release.
952 ** Evaluator trap calls now support `tweaking'
954 `Tweaking' means that the trap handler code can modify the Scheme
955 expression that is about to be evaluated (in the case of an
956 enter-frame trap) or the value that is being returned (in the case of
957 an exit-frame trap). The trap handler code indicates that it wants to
958 do this by returning a pair whose car is the symbol 'instead and whose
959 cdr is the modified expression or return value.
961 * Changes to the C interface
963 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
964 take a 'delete' function argument.
966 This argument makes no sense since the delete function is used to
967 remove a pair from an alist, and this must not be configurable.
969 This is an incompatible change.
971 ** The GH interface is now subject to the deprecation mechanism
973 The GH interface has been deprecated for quite some time but now it is
974 actually removed from Guile when it is configured with
975 --disable-deprecated.
977 See the manual "Transitioning away from GH" for more information.
979 ** A new family of functions for converting between C values and
980 Scheme values has been added.
982 These functions follow a common naming scheme and are designed to be
983 easier to use, thread-safe and more future-proof than the older
988 These are predicates that return a C boolean: 1 or 0. Instead of
989 SCM_NFALSEP, you can now use scm_is_true, for example.
991 - <type> scm_to_<type> (SCM val, ...)
993 These are functions that convert a Scheme value into an appropriate
994 C value. For example, you can use scm_to_int to safely convert from
997 - SCM scm_from_<type> (<type> val, ...)
999 These functions convert from a C type to a SCM value; for example,
1000 scm_from_int for ints.
1002 There is a huge number of these functions, for numbers, strings,
1003 symbols, vectors, etc. They are documented in the reference manual in
1004 the API section together with the types that they apply to.
1006 ** New functions for dealing with complex numbers in C have been added.
1008 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1009 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1010 They work like scm_make_rectangular etc but take or return doubles
1013 ** The function scm_make_complex has been discouraged.
1015 Use scm_c_make_rectangular instead.
1017 ** The INUM macros have been deprecated.
1019 A lot of code uses these macros to do general integer conversions,
1020 although the macros only work correctly with fixnums. Use the
1021 following alternatives.
1023 SCM_INUMP -> scm_is_integer or similar
1024 SCM_NINUMP -> !scm_is_integer or similar
1025 SCM_MAKINUM -> scm_from_int or similar
1026 SCM_INUM -> scm_to_int or similar
1028 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1029 do the validating for you.
1031 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1032 have been discouraged.
1034 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1035 new code. The functions have been discouraged since they don't fit
1038 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1040 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1041 evaluates its argument twice. Use scm_is_true, etc. instead for new
1044 ** The macro SCM_EQ_P has been discouraged.
1046 Use scm_is_eq for new code, which fits better into the naming
1049 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1052 Use the function scm_is_pair or scm_is_null instead.
1054 ** The functions scm_round and scm_truncate have been deprecated and
1055 are now available as scm_c_round and scm_c_truncate, respectively.
1057 These functions occupy the names that scm_round_number and
1058 scm_truncate_number should have.
1060 ** The functions scm_c_string2str, scm_c_substring2str, and
1061 scm_c_symbol2str have been deprecated.
1063 Use scm_to_locale_stringbuf or similar instead, maybe together with
1066 ** New functions scm_c_make_string, scm_c_string_length,
1067 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1068 scm_c_substring_shared, scm_c_substring_copy.
1070 These are like scm_make_string, scm_length, etc. but are slightly
1071 easier to use from C.
1073 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1074 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1076 They export too many assumptions about the implementation of strings
1077 and symbols that are no longer true in the presence of
1078 mutation-sharing substrings and when Guile switches to some form of
1081 When working with strings, it is often best to use the normal string
1082 functions provided by Guile, such as scm_c_string_ref,
1083 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1084 manual since many more such functions are now provided than
1087 When you want to convert a SCM string to a C string, use the
1088 scm_to_locale_string function or similar instead. For symbols, use
1089 scm_symbol_to_string and then work with that string. Because of the
1090 new string representation, scm_symbol_to_string does not need to copy
1091 and is thus quite efficient.
1093 ** Some string, symbol and keyword functions have been discouraged.
1095 They don't fit into the uniform naming scheme and are not explicit
1096 about the character encoding.
1098 Replace according to the following table:
1100 scm_allocate_string -> scm_c_make_string
1101 scm_take_str -> scm_take_locale_stringn
1102 scm_take0str -> scm_take_locale_string
1103 scm_mem2string -> scm_from_locale_stringn
1104 scm_str2string -> scm_from_locale_string
1105 scm_makfrom0str -> scm_from_locale_string
1106 scm_mem2symbol -> scm_from_locale_symboln
1107 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1108 scm_str2symbol -> scm_from_locale_symbol
1110 SCM_SYMBOL_HASH -> scm_hashq
1111 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1113 scm_c_make_keyword -> scm_from_locale_keyword
1115 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1116 now also available to C code.
1118 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1120 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1121 the latter returns the true name of the keyword, not the 'dash name',
1122 as SCM_KEYWORDSYM used to do.
1124 ** A new way to access arrays in a thread-safe and efficient way has
1127 See the manual, node "Accessing Arrays From C".
1129 ** The old uniform vector and bitvector implementations have been
1130 unceremoniously removed.
1132 This implementation exposed the details of the tagging system of
1133 Guile. Use the new C API explained in the manual in node "Uniform
1134 Numeric Vectors" and "Bit Vectors", respectively.
1136 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1137 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1138 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1139 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1140 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1141 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1144 ** The macros dealing with vectors have been deprecated.
1146 Use the new functions scm_is_vector, scm_vector_elements,
1147 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1148 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1149 manual for more details.
1151 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1152 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1154 The following macros have been removed: SCM_VECTOR_BASE,
1155 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1156 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1158 ** Some C functions and macros related to arrays have been deprecated.
1160 Migrate according to the following table:
1162 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1163 scm_make_ra -> scm_make_array
1164 scm_shap2ra -> scm_make_array
1165 scm_cvref -> scm_c_generalized_vector_ref
1166 scm_ra_set_contp -> do not use
1167 scm_aind -> scm_array_handle_pos
1168 scm_raprin1 -> scm_display or scm_write
1170 SCM_ARRAYP -> scm_is_array
1171 SCM_ARRAY_NDIM -> scm_c_array_rank
1172 SCM_ARRAY_DIMS -> scm_array_handle_dims
1173 SCM_ARRAY_CONTP -> do not use
1174 SCM_ARRAY_MEM -> do not use
1175 SCM_ARRAY_V -> scm_array_handle_elements or similar
1176 SCM_ARRAY_BASE -> do not use
1178 ** SCM_CELL_WORD_LOC has been deprecated.
1180 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1181 to a SCM, as opposed to a pointer to a scm_t_bits.
1183 This was done to allow the correct use of pointers into the Scheme
1184 heap. Previously, the heap words were of type scm_t_bits and local
1185 variables and function arguments were of type SCM, making it
1186 non-standards-conformant to have a pointer that can point to both.
1188 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1190 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1191 second and third words of double smobs. Likewise for
1192 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1194 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1195 used to get and set the 16 exra bits in the zeroth word of a smob.
1197 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1198 accesing the first immediate word of a smob as a SCM value, and there
1199 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1200 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1202 ** New way to deal with non-local exits and re-entries.
1204 There is a new set of functions that essentially do what
1205 scm_internal_dynamic_wind does, but in a way that is more convenient
1206 for C code in some situations. Here is a quick example of how to
1207 prevent a potential memory leak:
1214 scm_dynwind_begin (0);
1216 mem = scm_malloc (100);
1217 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1219 /* MEM would leak if BAR throws an error.
1220 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1227 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1228 SCM_DYNWIND_END as well.
1232 For full documentation, see the node "Dynamic Wind" in the manual.
1234 ** New function scm_dynwind_free
1236 This function calls 'free' on a given pointer when a dynwind context
1237 is left. Thus the call to scm_dynwind_unwind_handler above could be
1238 replaced with simply scm_dynwind_free (mem).
1240 ** New functions scm_c_call_with_blocked_asyncs and
1241 scm_c_call_with_unblocked_asyncs
1243 Like scm_call_with_blocked_asyncs etc. but for C functions.
1245 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1247 In addition to scm_c_call_with_blocked_asyncs you can now also use
1248 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1249 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1251 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1252 SCM_REALLOW_INTS have been deprecated.
1254 They do no longer fulfill their original role of blocking signal
1255 delivery. Depending on what you want to achieve, replace a pair of
1256 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1257 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1260 ** The value 'scm_mask_ints' is no longer writable.
1262 Previously, you could set scm_mask_ints directly. This is no longer
1263 possible. Use scm_c_call_with_blocked_asyncs and
1264 scm_c_call_with_unblocked_asyncs instead.
1266 ** New way to temporarily set the current input, output or error ports
1268 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1269 context' (see above). <foo> is one of "input", "output" or "error".
1271 ** New way to temporarily set fluids
1273 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1274 above) to temporarily set the value of a fluid.
1276 ** New types scm_t_intmax and scm_t_uintmax.
1278 On platforms that have them, these types are identical to intmax_t and
1279 uintmax_t, respectively. On other platforms, they are identical to
1280 the largest integer types that Guile knows about.
1282 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1284 You should not have used them.
1286 ** Many public #defines with generic names have been made private.
1288 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1289 private or renamed with a more suitable public name.
1291 ** The macro SCM_TYP16S has been deprecated.
1293 This macro is not intended for public use.
1295 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1297 Use scm_is_true (scm_inexact_p (...)) instead.
1299 ** The macro SCM_SLOPPY_REALP has been deprecated.
1301 Use scm_is_real instead.
1303 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1305 Use scm_is_complex instead.
1307 ** Some preprocessor defines have been deprecated.
1309 These defines indicated whether a certain feature was present in Guile
1310 or not. Going forward, assume that the features are always present.
1312 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1313 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1315 The following macros have been removed completely: MEMOIZE_LOCALS,
1316 SCM_RECKLESS, SCM_CAUTIOUS.
1318 ** The preprocessor define STACK_DIRECTION has been deprecated.
1320 There should be no need to know about the stack direction for ordinary
1323 ** New function: scm_effective_version
1325 Returns the "effective" version number. This is just the normal full
1326 version string without the final micro-version number. See "Changes
1327 to the distribution" above.
1329 ** The function scm_call_with_new_thread has a new prototype.
1331 Instead of taking a list with the thunk and handler, these two
1332 arguments are now passed directly:
1334 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1336 This is an incompatible change.
1338 ** New snarfer macro SCM_DEFINE_PUBLIC.
1340 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1341 function in the init section.
1343 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1345 ** Garbage collector rewrite.
1347 The garbage collector is cleaned up a lot, and now uses lazy
1348 sweeping. This is reflected in the output of (gc-stats); since cells
1349 are being freed when they are allocated, the cells-allocated field
1350 stays roughly constant.
1352 For malloc related triggers, the behavior is changed. It uses the same
1353 heuristic as the cell-triggered collections. It may be tuned with the
1354 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1355 for minimum yield of malloc related triggers. The default is 40.
1356 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1359 Debugging operations for the freelist have been deprecated, along with
1360 the C variables that control garbage collection. The environment
1361 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1362 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1364 For understanding the memory usage of a GUILE program, the routine
1365 gc-live-object-stats returns an alist containing the number of live
1366 objects for every type.
1369 ** The function scm_definedp has been renamed to scm_defined_p
1371 The name scm_definedp is deprecated.
1373 ** The struct scm_cell type has been renamed to scm_t_cell
1375 This is in accordance to Guile's naming scheme for types. Note that
1376 the name scm_cell is now used for a function that allocates and
1377 initializes a new cell (see below).
1379 ** New functions for memory management
1381 A new set of functions for memory management has been added since the
1382 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1383 indeed, Guile itself contained some long standing bugs that could
1384 cause aborts in long running programs.
1386 The new functions are more symmetrical and do not need cooperation
1387 from smob free routines, among other improvements.
1389 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1390 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1391 scm_gc_free, scm_gc_register_collectable_memory, and
1392 scm_gc_unregister_collectable_memory. Refer to the manual for more
1393 details and for upgrading instructions.
1395 The old functions for memory management have been deprecated. They
1396 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1397 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1399 ** Declarations of exported features are marked with SCM_API.
1401 Every declaration of a feature that belongs to the exported Guile API
1402 has been marked by adding the macro "SCM_API" to the start of the
1403 declaration. This macro can expand into different things, the most
1404 common of which is just "extern" for Unix platforms. On Win32, it can
1405 be used to control which symbols are exported from a DLL.
1407 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1408 will expand into "__declspec (dllimport) extern", which is needed for
1409 linking to the Guile DLL in Windows.
1411 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1412 SCM_SRFI4_IMPORT, for the corresponding libraries.
1414 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1416 Use the new functions scm_cell and scm_double_cell instead. The old
1417 macros had problems because with them allocation and initialization
1418 was separated and the GC could sometimes observe half initialized
1419 cells. Only careful coding by the user of SCM_NEWCELL and
1420 SCM_NEWCELL2 could make this safe and efficient.
1422 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1424 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1427 ** SRCBRKP has been deprecated.
1429 Use scm_c_source_property_breakpoint_p instead.
1431 ** Deprecated: scm_makmacro
1433 Change your code to use either scm_makmmacro or to define macros in
1434 Scheme, using 'define-macro'.
1436 ** New function scm_c_port_for_each.
1438 This function is like scm_port_for_each but takes a pointer to a C
1439 function as the callback instead of a SCM value.
1441 ** The names scm_internal_select, scm_thread_sleep, and
1442 scm_thread_usleep have been discouraged.
1444 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1446 ** The GC can no longer be blocked.
1448 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1449 The GC can now run (partially) concurrently with other code and thus
1450 blocking it is not well defined.
1452 ** Many definitions have been removed that were previously deprecated.
1454 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1455 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1456 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1457 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1458 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1459 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1460 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1461 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1462 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1463 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1464 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1465 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1466 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1467 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1468 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1469 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1470 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1471 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1472 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1473 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1474 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1475 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1476 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1477 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1478 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1479 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1480 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1481 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1482 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1483 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1484 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1486 * Changes to bundled modules
1490 Using the (ice-9 debug) module no longer automatically switches Guile
1491 to use the debugging evaluator. If you want to switch to the
1492 debugging evaluator (which is needed for backtrace information if you
1493 hit an error), please add an explicit "(debug-enable 'debug)" to your
1494 code just after the code to use (ice-9 debug).
1497 Changes since Guile 1.4:
1499 * Changes to the distribution
1501 ** A top-level TODO file is included.
1503 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1505 Guile now always uses three numbers to represent the version,
1506 i.e. "1.6.5". The first number, 1, is the major version number, the
1507 second number, 6, is the minor version number, and the third number,
1508 5, is the micro version number. Changes in major version number
1509 indicate major changes in Guile.
1511 Minor version numbers that are even denote stable releases, and odd
1512 minor version numbers denote development versions (which may be
1513 unstable). The micro version number indicates a minor sub-revision of
1514 a given MAJOR.MINOR release.
1516 In keeping with the new scheme, (minor-version) and scm_minor_version
1517 no longer return everything but the major version number. They now
1518 just return the minor version number. Two new functions
1519 (micro-version) and scm_micro_version have been added to report the
1520 micro version number.
1522 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1524 ** New preprocessor definitions are available for checking versions.
1526 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1527 SCM_MICRO_VERSION to the appropriate integer values.
1529 ** Guile now actively warns about deprecated features.
1531 The new configure option `--enable-deprecated=LEVEL' and the
1532 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1533 See INSTALL and README for more information.
1535 ** Guile is much more likely to work on 64-bit architectures.
1537 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1538 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1539 for the use of a test machine, and thanks to Stefan Jahn for ia64
1542 ** New functions: setitimer and getitimer.
1544 These implement a fairly direct interface to the libc functions of the
1547 ** The #. reader extension is now disabled by default.
1549 For safety reasons, #. evaluation is disabled by default. To
1550 re-enable it, set the fluid read-eval? to #t. For example:
1552 (fluid-set! read-eval? #t)
1554 but make sure you realize the potential security risks involved. With
1555 read-eval? enabled, reading a data file from an untrusted source can
1558 ** New SRFI modules have been added:
1560 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1563 (srfi srfi-1) is a library containing many useful pair- and list-processing
1566 (srfi srfi-2) exports and-let*.
1568 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1570 (srfi srfi-6) is a dummy module for now, since guile already provides
1571 all of the srfi-6 procedures by default: open-input-string,
1572 open-output-string, get-output-string.
1574 (srfi srfi-8) exports receive.
1576 (srfi srfi-9) exports define-record-type.
1578 (srfi srfi-10) exports define-reader-ctor and implements the reader
1581 (srfi srfi-11) exports let-values and let*-values.
1583 (srfi srfi-13) implements the SRFI String Library.
1585 (srfi srfi-14) implements the SRFI Character-Set Library.
1587 (srfi srfi-17) implements setter and getter-with-setter and redefines
1588 some accessor procedures as procedures with getters. (such as car,
1589 cdr, vector-ref etc.)
1591 (srfi srfi-19) implements the SRFI Time/Date Library.
1593 ** New scripts / "executable modules"
1595 Subdirectory "scripts" contains Scheme modules that are packaged to
1596 also be executable as scripts. At this time, these scripts are available:
1605 See README there for more info.
1607 These scripts can be invoked from the shell with the new program
1608 "guile-tools", which keeps track of installation directory for you.
1611 $ guile-tools display-commentary srfi/*.scm
1613 guile-tools is copied to the standard $bindir on "make install".
1615 ** New module (ice-9 stack-catch):
1617 stack-catch is like catch, but saves the current state of the stack in
1618 the fluid the-last-stack. This fluid can be useful when using the
1619 debugger and when re-throwing an error.
1621 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1623 This has been done to prevent problems on lesser operating systems
1624 that can't tolerate `*'s in file names. The exported macro continues
1625 to be named `and-let*', of course.
1627 On systems that support it, there is also a compatibility module named
1628 (ice-9 and-let*). It will go away in the next release.
1630 ** New modules (oop goops) etc.:
1633 (oop goops describe)
1635 (oop goops active-slot)
1636 (oop goops composite-slot)
1638 The Guile Object Oriented Programming System (GOOPS) has been
1639 integrated into Guile. For further information, consult the GOOPS
1640 manual and tutorial in the `doc' directory.
1642 ** New module (ice-9 rdelim).
1644 This exports the following procedures which were previously defined
1645 in the default environment:
1647 read-line read-line! read-delimited read-delimited! %read-delimited!
1648 %read-line write-line
1650 For backwards compatibility the definitions are still imported into the
1651 default environment in this version of Guile. However you should add:
1653 (use-modules (ice-9 rdelim))
1655 to any program which uses the definitions, since this may change in
1658 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1659 can be used for similar functionality.
1661 ** New module (ice-9 rw)
1663 This is a subset of the (scsh rw) module from guile-scsh. Currently
1664 it defines two procedures:
1666 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1668 Read characters from a port or file descriptor into a string STR.
1669 A port must have an underlying file descriptor -- a so-called
1670 fport. This procedure is scsh-compatible and can efficiently read
1673 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1675 Write characters from a string STR to a port or file descriptor.
1676 A port must have an underlying file descriptor -- a so-called
1677 fport. This procedure is mostly compatible and can efficiently
1678 write large strings.
1680 ** New module (ice-9 match)
1682 This module includes Andrew K. Wright's pattern matcher. See
1683 ice-9/match.scm for brief description or
1685 http://www.star-lab.com/wright/code.html
1687 for complete documentation.
1689 ** New module (ice-9 buffered-input)
1691 This module provides procedures to construct an input port from an
1692 underlying source of input that reads and returns its input in chunks.
1693 The underlying input source is a Scheme procedure, specified by the
1694 caller, which the port invokes whenever it needs more input.
1696 This is useful when building an input port whose back end is Readline
1697 or a UI element such as the GtkEntry widget.
1701 The reference and tutorial documentation that was previously
1702 distributed separately, as `guile-doc', is now included in the core
1703 Guile distribution. The documentation consists of the following
1706 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1709 - The Guile Reference Manual (guile.texi) contains (or is intended to
1710 contain) reference documentation on all aspects of Guile.
1712 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1713 reference documentation for using GOOPS, Guile's Object Oriented
1716 - The Revised^5 Report on the Algorithmic Language Scheme
1719 See the README file in the `doc' directory for more details.
1721 ** There are a couple of examples in the examples/ directory now.
1723 * Changes to the stand-alone interpreter
1725 ** New command line option `--use-srfi'
1727 Using this option, SRFI modules can be loaded on startup and be
1728 available right from the beginning. This makes programming portable
1729 Scheme programs easier.
1731 The option `--use-srfi' expects a comma-separated list of numbers,
1732 each representing a SRFI number to be loaded into the interpreter
1733 before starting evaluating a script file or the REPL. Additionally,
1734 the feature identifier for the loaded SRFIs is recognized by
1735 `cond-expand' when using this option.
1738 $ guile --use-srfi=8,13
1739 guile> (receive (x z) (values 1 2) (+ 1 2))
1741 guile> (string-pad "bla" 20)
1744 ** Guile now always starts up in the `(guile-user)' module.
1746 Previously, scripts executed via the `-s' option would run in the
1747 `(guile)' module and the repl would run in the `(guile-user)' module.
1748 Now every user action takes place in the `(guile-user)' module by
1751 * Changes to Scheme functions and syntax
1753 ** Character classifiers work for non-ASCII characters.
1755 The predicates `char-alphabetic?', `char-numeric?',
1756 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1757 no longer check whether their arguments are ASCII characters.
1758 Previously, a character would only be considered alphabetic when it
1759 was also ASCII, for example.
1761 ** Previously deprecated Scheme functions have been removed:
1763 tag - no replacement.
1764 fseek - replaced by seek.
1765 list* - replaced by cons*.
1767 ** It's now possible to create modules with controlled environments
1771 (use-modules (ice-9 safe))
1772 (define m (make-safe-module))
1773 ;;; m will now be a module containing only a safe subset of R5RS
1774 (eval '(+ 1 2) m) --> 3
1775 (eval 'load m) --> ERROR: Unbound variable: load
1777 ** Evaluation of "()", the empty list, is now an error.
1779 Previously, the expression "()" evaluated to the empty list. This has
1780 been changed to signal a "missing expression" error. The correct way
1781 to write the empty list as a literal constant is to use quote: "'()".
1783 ** New concept of `Guile Extensions'.
1785 A Guile Extension is just a ordinary shared library that can be linked
1786 at run-time. We found it advantageous to give this simple concept a
1787 dedicated name to distinguish the issues related to shared libraries
1788 from the issues related to the module system.
1790 *** New function: load-extension
1792 Executing (load-extension lib init) is mostly equivalent to
1794 (dynamic-call init (dynamic-link lib))
1796 except when scm_register_extension has been called previously.
1797 Whenever appropriate, you should use `load-extension' instead of
1798 dynamic-link and dynamic-call.
1800 *** New C function: scm_c_register_extension
1802 This function registers a initialization function for use by
1803 `load-extension'. Use it when you don't want specific extensions to
1804 be loaded as shared libraries (for example on platforms that don't
1805 support dynamic linking).
1807 ** Auto-loading of compiled-code modules is deprecated.
1809 Guile used to be able to automatically find and link a shared
1810 library to satisfy requests for a module. For example, the module
1811 `(foo bar)' could be implemented by placing a shared library named
1812 "foo/libbar.so" (or with a different extension) in a directory on the
1815 This has been found to be too tricky, and is no longer supported. The
1816 shared libraries are now called "extensions". You should now write a
1817 small Scheme file that calls `load-extension' to load the shared
1818 library and initialize it explicitly.
1820 The shared libraries themselves should be installed in the usual
1821 places for shared libraries, with names like "libguile-foo-bar".
1823 For example, place this into a file "foo/bar.scm"
1825 (define-module (foo bar))
1827 (load-extension "libguile-foo-bar" "foobar_init")
1829 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1831 `eval' is now R5RS, that is it takes two arguments.
1832 The second argument is an environment specifier, i.e. either
1834 (scheme-report-environment 5)
1835 (null-environment 5)
1836 (interaction-environment)
1842 ** The module system has been made more disciplined.
1844 The function `eval' will save and restore the current module around
1845 the evaluation of the specified expression. While this expression is
1846 evaluated, `(current-module)' will now return the right module, which
1847 is the module specified as the second argument to `eval'.
1849 A consequence of this change is that `eval' is not particularly
1850 useful when you want allow the evaluated code to change what module is
1851 designated as the current module and have this change persist from one
1852 call to `eval' to the next. The read-eval-print-loop is an example
1853 where `eval' is now inadequate. To compensate, there is a new
1854 function `primitive-eval' that does not take a module specifier and
1855 that does not save/restore the current module. You should use this
1856 function together with `set-current-module', `current-module', etc
1857 when you want to have more control over the state that is carried from
1858 one eval to the next.
1860 Additionally, it has been made sure that forms that are evaluated at
1861 the top level are always evaluated with respect to the current module.
1862 Previously, subforms of top-level forms such as `begin', `case',
1863 etc. did not respect changes to the current module although these
1864 subforms are at the top-level as well.
1866 To prevent strange behavior, the forms `define-module',
1867 `use-modules', `use-syntax', and `export' have been restricted to only
1868 work on the top level. The forms `define-public' and
1869 `defmacro-public' only export the new binding on the top level. They
1870 behave just like `define' and `defmacro', respectively, when they are
1871 used in a lexical environment.
1873 Also, `export' will no longer silently re-export bindings imported
1874 from a used module. It will emit a `deprecation' warning and will
1875 cease to perform any re-export in the next version. If you actually
1876 want to re-export bindings, use the new `re-export' in place of
1877 `export'. The new `re-export' will not make copies of variables when
1878 rexporting them, as `export' did wrongly.
1880 ** Module system now allows selection and renaming of imported bindings
1882 Previously, when using `use-modules' or the `#:use-module' clause in
1883 the `define-module' form, all the bindings (association of symbols to
1884 values) for imported modules were added to the "current module" on an
1885 as-is basis. This has been changed to allow finer control through two
1886 new facilities: selection and renaming.
1888 You can now select which of the imported module's bindings are to be
1889 visible in the current module by using the `:select' clause. This
1890 clause also can be used to rename individual bindings. For example:
1892 ;; import all bindings no questions asked
1893 (use-modules (ice-9 common-list))
1895 ;; import four bindings, renaming two of them;
1896 ;; the current module sees: every some zonk-y zonk-n
1897 (use-modules ((ice-9 common-list)
1899 (remove-if . zonk-y)
1900 (remove-if-not . zonk-n))))
1902 You can also programmatically rename all selected bindings using the
1903 `:renamer' clause, which specifies a proc that takes a symbol and
1904 returns another symbol. Because it is common practice to use a prefix,
1905 we now provide the convenience procedure `symbol-prefix-proc'. For
1908 ;; import four bindings, renaming two of them specifically,
1909 ;; and all four w/ prefix "CL:";
1910 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1911 (use-modules ((ice-9 common-list)
1913 (remove-if . zonk-y)
1914 (remove-if-not . zonk-n))
1915 :renamer (symbol-prefix-proc 'CL:)))
1917 ;; import four bindings, renaming two of them specifically,
1918 ;; and all four by upcasing.
1919 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1920 (define (upcase-symbol sym)
1921 (string->symbol (string-upcase (symbol->string sym))))
1923 (use-modules ((ice-9 common-list)
1925 (remove-if . zonk-y)
1926 (remove-if-not . zonk-n))
1927 :renamer upcase-symbol))
1929 Note that programmatic renaming is done *after* individual renaming.
1930 Also, the above examples show `use-modules', but the same facilities are
1931 available for the `#:use-module' clause of `define-module'.
1933 See manual for more info.
1935 ** The semantics of guardians have changed.
1937 The changes are for the most part compatible. An important criterion
1938 was to keep the typical usage of guardians as simple as before, but to
1939 make the semantics safer and (as a result) more useful.
1941 *** All objects returned from guardians are now properly alive.
1943 It is now guaranteed that any object referenced by an object returned
1944 from a guardian is alive. It's now impossible for a guardian to
1945 return a "contained" object before its "containing" object.
1947 One incompatible (but probably not very important) change resulting
1948 from this is that it is no longer possible to guard objects that
1949 indirectly reference themselves (i.e. are parts of cycles). If you do
1950 so accidentally, you'll get a warning.
1952 *** There are now two types of guardians: greedy and sharing.
1954 If you call (make-guardian #t) or just (make-guardian), you'll get a
1955 greedy guardian, and for (make-guardian #f) a sharing guardian.
1957 Greedy guardians are the default because they are more "defensive".
1958 You can only greedily guard an object once. If you guard an object
1959 more than once, once in a greedy guardian and the rest of times in
1960 sharing guardians, then it is guaranteed that the object won't be
1961 returned from sharing guardians as long as it is greedily guarded
1964 Guardians returned by calls to `make-guardian' can now take one more
1965 optional parameter, which says whether to throw an error in case an
1966 attempt is made to greedily guard an object that is already greedily
1967 guarded. The default is true, i.e. throw an error. If the parameter
1968 is false, the guardian invocation returns #t if guarding was
1969 successful and #f if it wasn't.
1971 Also, since greedy guarding is, in effect, a side-effecting operation
1972 on objects, a new function is introduced: `destroy-guardian!'.
1973 Invoking this function on a guardian renders it unoperative and, if
1974 the guardian is greedy, clears the "greedily guarded" property of the
1975 objects that were guarded by it, thus undoing the side effect.
1977 Note that all this hair is hardly very important, since guardian
1978 objects are usually permanent.
1980 ** Continuations created by call-with-current-continuation now accept
1981 any number of arguments, as required by R5RS.
1983 ** New function `issue-deprecation-warning'
1985 This function is used to display the deprecation messages that are
1986 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1989 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1993 ;; `id' is deprecated. Use `identity' instead.
1998 ** New syntax `begin-deprecated'
2000 When deprecated features are included (as determined by the configure
2001 option --enable-deprecated), `begin-deprecated' is identical to
2002 `begin'. When deprecated features are excluded, it always evaluates
2003 to `#f', ignoring the body forms.
2005 ** New function `make-object-property'
2007 This function returns a new `procedure with setter' P that can be used
2008 to attach a property to objects. When calling P as
2012 where `obj' is any kind of object, it attaches `val' to `obj' in such
2013 a way that it can be retrieved by calling P as
2017 This function will replace procedure properties, symbol properties and
2018 source properties eventually.
2020 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2022 Instead of #&optional, #&key, etc you should now use #:optional,
2023 #:key, etc. Since #:optional is a keyword, you can write it as just
2024 :optional when (read-set! keywords 'prefix) is active.
2026 The old reader syntax `#&' is still supported, but deprecated. It
2027 will be removed in the next release.
2029 ** New define-module option: pure
2031 Tells the module system not to include any bindings from the root
2036 (define-module (totally-empty-module)
2039 ** New define-module option: export NAME1 ...
2041 Export names NAME1 ...
2043 This option is required if you want to be able to export bindings from
2044 a module which doesn't import one of `define-public' or `export'.
2048 (define-module (foo)
2050 :use-module (ice-9 r5rs)
2053 ;;; Note that we're pure R5RS below this point!
2058 ** New function: object->string OBJ
2060 Return a Scheme string obtained by printing a given object.
2062 ** New function: port? X
2064 Returns a boolean indicating whether X is a port. Equivalent to
2065 `(or (input-port? X) (output-port? X))'.
2067 ** New function: file-port?
2069 Determines whether a given object is a port that is related to a file.
2071 ** New function: port-for-each proc
2073 Apply PROC to each port in the Guile port table in turn. The return
2074 value is unspecified. More specifically, PROC is applied exactly once
2075 to every port that exists in the system at the time PORT-FOR-EACH is
2076 invoked. Changes to the port table while PORT-FOR-EACH is running
2077 have no effect as far as PORT-FOR-EACH is concerned.
2079 ** New function: dup2 oldfd newfd
2081 A simple wrapper for the `dup2' system call. Copies the file
2082 descriptor OLDFD to descriptor number NEWFD, replacing the
2083 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2084 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2085 to move away ports which are using NEWFD. The return value is
2088 ** New function: close-fdes fd
2090 A simple wrapper for the `close' system call. Close file
2091 descriptor FD, which must be an integer. Unlike close (*note
2092 close: Ports and File Descriptors.), the file descriptor will be
2093 closed even if a port is using it. The return value is
2096 ** New function: crypt password salt
2098 Encrypts `password' using the standard unix password encryption
2101 ** New function: chroot path
2103 Change the root directory of the running process to `path'.
2105 ** New functions: getlogin, cuserid
2107 Return the login name or the user name of the current effective user
2110 ** New functions: getpriority which who, setpriority which who prio
2112 Get or set the priority of the running process.
2114 ** New function: getpass prompt
2116 Read a password from the terminal, first displaying `prompt' and
2119 ** New function: flock file operation
2121 Set/remove an advisory shared or exclusive lock on `file'.
2123 ** New functions: sethostname name, gethostname
2125 Set or get the hostname of the machine the current process is running
2128 ** New function: mkstemp! tmpl
2130 mkstemp creates a new unique file in the file system and returns a
2131 new buffered port open for reading and writing to the file. TMPL
2132 is a string specifying where the file should be created: it must
2133 end with `XXXXXX' and will be changed in place to return the name
2134 of the temporary file.
2136 ** New function: open-input-string string
2138 Return an input string port which delivers the characters from
2139 `string'. This procedure, together with `open-output-string' and
2140 `get-output-string' implements SRFI-6.
2142 ** New function: open-output-string
2144 Return an output string port which collects all data written to it.
2145 The data can then be retrieved by `get-output-string'.
2147 ** New function: get-output-string
2149 Return the contents of an output string port.
2151 ** New function: identity
2153 Return the argument.
2155 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2156 are represented in Scheme as integers with normal host byte ordering.
2158 ** New function: inet-pton family address
2160 Convert a printable string network address into an integer. Note that
2161 unlike the C version of this function, the result is an integer with
2162 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2165 (inet-pton AF_INET "127.0.0.1") => 2130706433
2166 (inet-pton AF_INET6 "::1") => 1
2168 ** New function: inet-ntop family address
2170 Convert an integer network address into a printable string. Note that
2171 unlike the C version of this function, the input is an integer with
2172 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2175 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2176 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2177 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2181 Use `identity' instead.
2187 ** Deprecated: return-it
2191 ** Deprecated: string-character-length
2193 Use `string-length' instead.
2195 ** Deprecated: flags
2197 Use `logior' instead.
2199 ** Deprecated: close-all-ports-except.
2201 This was intended for closing ports in a child process after a fork,
2202 but it has the undesirable side effect of flushing buffers.
2203 port-for-each is more flexible.
2205 ** The (ice-9 popen) module now attempts to set up file descriptors in
2206 the child process from the current Scheme ports, instead of using the
2207 current values of file descriptors 0, 1, and 2 in the parent process.
2209 ** Removed function: builtin-weak-bindings
2211 There is no such concept as a weak binding any more.
2213 ** Removed constants: bignum-radix, scm-line-incrementors
2215 ** define-method: New syntax mandatory.
2217 The new method syntax is now mandatory:
2219 (define-method (NAME ARG-SPEC ...) BODY ...)
2220 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2222 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2223 REST-ARG ::= ARG-NAME
2225 If you have old code using the old syntax, import
2226 (oop goops old-define-method) before (oop goops) as in:
2228 (use-modules (oop goops old-define-method) (oop goops))
2230 ** Deprecated function: builtin-variable
2231 Removed function: builtin-bindings
2233 There is no longer a distinction between builtin or other variables.
2234 Use module system operations for all variables.
2236 ** Lazy-catch handlers are no longer allowed to return.
2238 That is, a call to `throw', `error', etc is now guaranteed to not
2241 ** Bugfixes for (ice-9 getopt-long)
2243 This module is now tested using test-suite/tests/getopt-long.test.
2244 The following bugs have been fixed:
2246 *** Parsing for options that are specified to have `optional' args now checks
2247 if the next element is an option instead of unconditionally taking it as the
2250 *** An error is now thrown for `--opt=val' when the option description
2251 does not specify `(value #t)' or `(value optional)'. This condition used to
2252 be accepted w/o error, contrary to the documentation.
2254 *** The error message for unrecognized options is now more informative.
2255 It used to be "not a record", an artifact of the implementation.
2257 *** The error message for `--opt' terminating the arg list (no value), when
2258 `(value #t)' is specified, is now more informative. It used to be "not enough
2261 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2262 The expansion used to be like so:
2264 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2266 Note that the "5d" is dropped. Now it is like so:
2268 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2270 This enables single-char options to have adjoining arguments as long as their
2271 constituent characters are not potential single-char options.
2273 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2275 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2276 property `arglist', which can be retrieved by `arity'. The result is that
2277 `arity' can give more detailed information than before:
2281 guile> (use-modules (ice-9 optargs))
2282 guile> (define* (foo #:optional a b c) a)
2284 0 or more arguments in `lambda*:G0'.
2289 3 optional arguments: `a', `b' and `c'.
2290 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2292 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2293 and `d', other keywords allowed.
2294 guile> (define* (baz a b #:optional c #:rest r) a)
2296 2 required arguments: `a' and `b', 1 optional argument: `c',
2299 * Changes to the C interface
2301 ** Types have been renamed from scm_*_t to scm_t_*.
2303 This has been done for POSIX sake. It reserves identifiers ending
2304 with "_t". What a concept.
2306 The old names are still available with status `deprecated'.
2308 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2310 ** Deprecated features have been removed.
2314 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2315 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2317 *** C Functions removed
2319 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2320 scm_fseek - replaced by scm_seek.
2321 gc-thunk - replaced by after-gc-hook.
2322 gh_int2scmb - replaced by gh_bool2scm.
2323 scm_tc_dblr - replaced by scm_tc16_real.
2324 scm_tc_dblc - replaced by scm_tc16_complex.
2325 scm_list_star - replaced by scm_cons_star.
2327 ** Deprecated: scm_makfromstr
2329 Use scm_mem2string instead.
2331 ** Deprecated: scm_make_shared_substring
2333 Explicit shared substrings will disappear from Guile.
2335 Instead, "normal" strings will be implemented using sharing
2336 internally, combined with a copy-on-write strategy.
2338 ** Deprecated: scm_read_only_string_p
2340 The concept of read-only strings will disappear in next release of
2343 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2345 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2347 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2349 Call a procedure with the indicated number of arguments. See "Fly
2350 Evaluation" in the manual.
2352 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2354 Call a procedure with the indicated number of arguments and a list of
2355 further arguments. See "Fly Evaluation" in the manual.
2357 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2359 Create a list of the given number of elements. See "List
2360 Constructors" in the manual.
2362 ** Renamed function: scm_listify has been replaced by scm_list_n.
2364 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2365 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2367 Use functions scm_list_N instead.
2369 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2371 Used by an application to read arbitrary number of bytes from a port.
2372 Same semantics as libc read, except that scm_c_read only returns less
2373 than SIZE bytes if at end-of-file.
2375 Warning: Doesn't update port line and column counts!
2377 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2379 Used by an application to write arbitrary number of bytes to an SCM
2380 port. Similar semantics as libc write. However, unlike libc
2381 write, scm_c_write writes the requested number of bytes and has no
2384 Warning: Doesn't update port line and column counts!
2386 ** New function: scm_init_guile ()
2388 In contrast to scm_boot_guile, scm_init_guile will return normally
2389 after initializing Guile. It is not available on all systems, tho.
2391 ** New functions: scm_str2symbol, scm_mem2symbol
2393 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2394 field of characters and creates a scheme symbol object from that C string.
2395 The function scm_mem2symbol takes a const char* and a number of characters and
2396 creates a symbol from the characters in that memory area.
2398 ** New functions: scm_primitive_make_property
2399 scm_primitive_property_ref
2400 scm_primitive_property_set_x
2401 scm_primitive_property_del_x
2403 These functions implement a new way to deal with object properties.
2404 See libguile/properties.c for their documentation.
2406 ** New function: scm_done_free (long size)
2408 This function is the inverse of scm_done_malloc. Use it to report the
2409 amount of smob memory you free. The previous method, which involved
2410 calling scm_done_malloc with negative argument, was somewhat
2411 unintuitive (and is still available, of course).
2413 ** New function: scm_c_memq (SCM obj, SCM list)
2415 This function provides a fast C level alternative for scm_memq for the case
2416 that the list parameter is known to be a proper list. The function is a
2417 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2418 list input parameter, since for anything else but a proper list the function's
2419 behaviour is undefined - it may even crash or loop endlessly. Further, for
2420 the case that the object is not found in the list, scm_c_memq returns #f which
2421 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2423 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2424 scm_remember_upto_here
2426 These functions replace the function scm_remember.
2428 ** Deprecated function: scm_remember
2430 Use one of the new functions scm_remember_upto_here_1,
2431 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2433 ** New function: scm_allocate_string
2435 This function replaces the function scm_makstr.
2437 ** Deprecated function: scm_makstr
2439 Use the new function scm_allocate_string instead.
2441 ** New global variable scm_gc_running_p introduced.
2443 Use this variable to find out if garbage collection is being executed. Up to
2444 now applications have used scm_gc_heap_lock to test if garbage collection was
2445 running, which also works because of the fact that up to know only the garbage
2446 collector has set this variable. But, this is an implementation detail that
2447 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2448 of this variable is (and has been) not fully safe anyway.
2450 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2452 Use these instead of SCM_LENGTH_MAX.
2454 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2455 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2456 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2458 Use these instead of SCM_LENGTH.
2460 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2461 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2462 SCM_SET_BITVECTOR_LENGTH
2464 Use these instead of SCM_SETLENGTH
2466 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2467 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2470 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2473 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2474 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2477 Use these instead of SCM_SETCHARS.
2479 ** New macro: SCM_BITVECTOR_P
2481 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2483 Use instead of SCM_COERCE_SUBSTR.
2485 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2487 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2489 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2490 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2491 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2492 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2493 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2494 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2495 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2496 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2497 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2498 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2499 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2500 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2501 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2502 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2503 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2505 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2506 Use scm_memory_error instead of SCM_NALLOC.
2507 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2508 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2509 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2510 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2511 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2512 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2513 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2514 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2515 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2516 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2517 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2518 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2519 Use a type specific setter macro instead of SCM_SETLENGTH.
2520 Use a type specific setter macro instead of SCM_SETCHARS.
2521 Use a type specific length macro instead of SCM_LENGTH_MAX.
2522 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2523 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2524 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2525 Use SCM_TYP16 instead of SCM_GCTYP16.
2526 Use SCM_CDR instead of SCM_GCCDR.
2527 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2528 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2529 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2530 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2531 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2532 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2533 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2535 ** Removed function: scm_struct_init
2537 ** Removed variable: scm_symhash_dim
2539 ** Renamed function: scm_make_cont has been replaced by
2540 scm_make_continuation, which has a different interface.
2542 ** Deprecated function: scm_call_catching_errors
2544 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2546 ** Deprecated function: scm_strhash
2548 Use scm_string_hash instead.
2550 ** Deprecated function: scm_vector_set_length_x
2552 Instead, create a fresh vector of the desired size and copy the contents.
2554 ** scm_gensym has changed prototype
2556 scm_gensym now only takes one argument.
2558 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2561 There is now only a single symbol type scm_tc7_symbol.
2562 The tag scm_tc7_lvector was not used anyway.
2564 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2566 Use scm_make_smob_type and scm_set_smob_XXX instead.
2568 ** New function scm_set_smob_apply.
2570 This can be used to set an apply function to a smob type.
2572 ** Deprecated function: scm_strprint_obj
2574 Use scm_object_to_string instead.
2576 ** Deprecated function: scm_wta
2578 Use scm_wrong_type_arg, or another appropriate error signalling function
2581 ** Explicit support for obarrays has been deprecated.
2583 Use `scm_str2symbol' and the generic hashtable functions instead.
2585 ** The concept of `vcells' has been deprecated.
2587 The data type `variable' is now used exclusively. `Vcells' have been
2588 a low-level concept so you are likely not affected by this change.
2590 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2591 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2593 Use scm_c_define or scm_c_lookup instead, as appropriate.
2595 *** New functions: scm_c_module_lookup, scm_c_lookup,
2596 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2597 scm_module_define, scm_define.
2599 These functions work with variables instead of with vcells.
2601 ** New functions for creating and defining `subr's and `gsubr's.
2603 The new functions more clearly distinguish between creating a subr (or
2604 gsubr) object and adding it to the current module.
2606 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2607 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2608 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2609 scm_c_define_gsubr_with_generic.
2611 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2612 scm_make_subr_with_generic, scm_make_gsubr,
2613 scm_make_gsubr_with_generic.
2615 Use the new ones from above instead.
2617 ** C interface to the module system has changed.
2619 While we suggest that you avoid as many explicit module system
2620 operations from C as possible for the time being, the C interface has
2621 been made more similar to the high-level Scheme module system.
2623 *** New functions: scm_c_define_module, scm_c_use_module,
2624 scm_c_export, scm_c_resolve_module.
2626 They mostly work like their Scheme namesakes. scm_c_define_module
2627 takes a function that is called a context where the new module is
2630 *** Deprecated functions: scm_the_root_module, scm_make_module,
2631 scm_ensure_user_module, scm_load_scheme_module.
2633 Use the new functions instead.
2635 ** Renamed function: scm_internal_with_fluids becomes
2638 scm_internal_with_fluids is available as a deprecated function.
2640 ** New function: scm_c_with_fluid.
2642 Just like scm_c_with_fluids, but takes one fluid and one value instead
2645 ** Deprecated typedefs: long_long, ulong_long.
2647 They are of questionable utility and they pollute the global
2650 ** Deprecated typedef: scm_sizet
2652 It is of questionable utility now that Guile requires ANSI C, and is
2655 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2656 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2657 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2659 Made more compliant with the naming policy by adding a _t at the end.
2661 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2662 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2664 With the exception of the mysterious scm_2ulong2big, they are still
2665 available under new names (scm_i_mkbig etc). These functions are not
2666 intended to be used in user code. You should avoid dealing with
2667 bignums directly, and should deal with numbers in general (which can
2670 ** Change in behavior: scm_num2long, scm_num2ulong
2672 The scm_num2[u]long functions don't any longer accept an inexact
2673 argument. This change in behavior is motivated by concordance with
2674 R5RS: It is more common that a primitive doesn't want to accept an
2675 inexact for an exact.
2677 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2678 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2679 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2682 These are conversion functions between the various ANSI C integral
2683 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2684 accept an inexact argument.
2686 ** New functions: scm_float2num, scm_double2num,
2687 scm_num2float, scm_num2double.
2689 These are conversion functions between the two ANSI C float types and
2692 ** New number validation macros:
2693 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2697 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2699 These are just nicer-named old scm_protect_object and
2700 scm_unprotect_object.
2702 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2704 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2706 These functions can be used to register pointers to locations that
2709 ** Deprecated function: scm_create_hook.
2711 Its sins are: misleading name, non-modularity and lack of general
2715 Changes since Guile 1.3.4:
2717 * Changes to the distribution
2719 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2721 We've changed the way we handle generated files in the Guile source
2722 repository. As a result, the procedure for building trees obtained
2723 from the nightly FTP snapshots or via CVS has changed:
2724 - You must have appropriate versions of autoconf, automake, and
2725 libtool installed on your system. See README for info on how to
2726 obtain these programs.
2727 - Before configuring the tree, you must first run the script
2728 `autogen.sh' at the top of the source tree.
2730 The Guile repository used to contain not only source files, written by
2731 humans, but also some generated files, like configure scripts and
2732 Makefile.in files. Even though the contents of these files could be
2733 derived mechanically from other files present, we thought it would
2734 make the tree easier to build if we checked them into CVS.
2736 However, this approach means that minor differences between
2737 developer's installed tools and habits affected the whole team.
2738 So we have removed the generated files from the repository, and
2739 added the autogen.sh script, which will reconstruct them
2743 ** configure now has experimental options to remove support for certain
2746 --disable-arrays omit array and uniform array support
2747 --disable-posix omit posix interfaces
2748 --disable-networking omit networking interfaces
2749 --disable-regex omit regular expression interfaces
2751 These are likely to become separate modules some day.
2753 ** New configure option --enable-debug-freelist
2755 This enables a debugging version of SCM_NEWCELL(), and also registers
2756 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2758 Configure with the --enable-debug-freelist option to enable
2759 the gc-set-debug-check-freelist! primitive, and then use:
2761 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2762 (gc-set-debug-check-freelist! #f) # turn off checking
2764 Checking of the freelist forces a traversal of the freelist and
2765 a garbage collection before each allocation of a cell. This can
2766 slow down the interpreter dramatically, so the setter should be used to
2767 turn on this extra processing only when necessary.
2769 ** New configure option --enable-debug-malloc
2771 Include code for debugging of calls to scm_must_malloc/realloc/free.
2775 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2776 2. objects reallocated by scm_must_realloc has been allocated by
2778 3. reallocated objects are reallocated with the same what string
2780 But, most importantly, it records the number of allocated objects of
2781 each kind. This is useful when searching for memory leaks.
2783 A Guile compiled with this option provides the primitive
2784 `malloc-stats' which returns an alist with pairs of kind and the
2785 number of objects of that kind.
2787 ** All includes are now referenced relative to the root directory
2789 Since some users have had problems with mixups between Guile and
2790 system headers, we have decided to always refer to Guile headers via
2791 their parent directories. This essentially creates a "private name
2792 space" for Guile headers. This means that the compiler only is given
2793 -I options for the root build and root source directory.
2795 ** Header files kw.h and genio.h have been removed.
2797 ** The module (ice-9 getopt-gnu-style) has been removed.
2799 ** New module (ice-9 documentation)
2801 Implements the interface to documentation strings associated with
2804 ** New module (ice-9 time)
2806 Provides a macro `time', which displays execution time of a given form.
2808 ** New module (ice-9 history)
2810 Loading this module enables value history in the repl.
2812 * Changes to the stand-alone interpreter
2814 ** New command line option --debug
2816 Start Guile with debugging evaluator and backtraces enabled.
2818 This is useful when debugging your .guile init file or scripts.
2820 ** New help facility
2822 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2823 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2824 (help 'NAME) gives documentation for NAME, even if it is not an object
2825 (help ,EXPR) gives documentation for object returned by EXPR
2826 (help (my module)) gives module commentary for `(my module)'
2827 (help) gives this text
2829 `help' searches among bindings exported from loaded modules, while
2830 `apropos' searches among bindings visible from the "current" module.
2832 Examples: (help help)
2834 (help "output-string")
2836 ** `help' and `apropos' now prints full module names
2838 ** Dynamic linking now uses libltdl from the libtool package.
2840 The old system dependent code for doing dynamic linking has been
2841 replaced with calls to the libltdl functions which do all the hairy
2844 The major improvement is that you can now directly pass libtool
2845 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2846 will be able to do the best shared library job you can get, via
2849 The way dynamic libraries are found has changed and is not really
2850 portable across platforms, probably. It is therefore recommended to
2851 use absolute filenames when possible.
2853 If you pass a filename without an extension to `dynamic-link', it will
2854 try a few appropriate ones. Thus, the most platform ignorant way is
2855 to specify a name like "libfoo", without any directories and
2858 ** Guile COOP threads are now compatible with LinuxThreads
2860 Previously, COOP threading wasn't possible in applications linked with
2861 Linux POSIX threads due to their use of the stack pointer to find the
2862 thread context. This has now been fixed with a workaround which uses
2863 the pthreads to allocate the stack.
2865 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2867 ** Positions of erring expression in scripts
2869 With version 1.3.4, the location of the erring expression in Guile
2870 scipts is no longer automatically reported. (This should have been
2871 documented before the 1.3.4 release.)
2873 You can get this information by enabling recording of positions of
2874 source expressions and running the debugging evaluator. Put this at
2875 the top of your script (or in your "site" file):
2877 (read-enable 'positions)
2878 (debug-enable 'debug)
2880 ** Backtraces in scripts
2882 It is now possible to get backtraces in scripts.
2886 (debug-enable 'debug 'backtrace)
2888 at the top of the script.
2890 (The first options enables the debugging evaluator.
2891 The second enables backtraces.)
2893 ** Part of module system symbol lookup now implemented in C
2895 The eval closure of most modules is now implemented in C. Since this
2896 was one of the bottlenecks for loading speed, Guile now loads code
2897 substantially faster than before.
2899 ** Attempting to get the value of an unbound variable now produces
2900 an exception with a key of 'unbound-variable instead of 'misc-error.
2902 ** The initial default output port is now unbuffered if it's using a
2903 tty device. Previously in this situation it was line-buffered.
2905 ** New hook: after-gc-hook
2907 after-gc-hook takes over the role of gc-thunk. This hook is run at
2908 the first SCM_TICK after a GC. (Thus, the code is run at the same
2909 point during evaluation as signal handlers.)
2911 Note that this hook should be used only for diagnostic and debugging
2912 purposes. It is not certain that it will continue to be well-defined
2913 when this hook is run in the future.
2915 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2916 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2918 ** Improvements to garbage collector
2920 Guile 1.4 has a new policy for triggering heap allocation and
2921 determining the sizes of heap segments. It fixes a number of problems
2924 1. The new policy can handle two separate pools of cells
2925 (2-word/4-word) better. (The old policy would run wild, allocating
2926 more and more memory for certain programs.)
2928 2. The old code would sometimes allocate far too much heap so that the
2929 Guile process became gigantic. The new code avoids this.
2931 3. The old code would sometimes allocate too little so that few cells
2932 were freed at GC so that, in turn, too much time was spent in GC.
2934 4. The old code would often trigger heap allocation several times in a
2935 row. (The new scheme predicts how large the segments needs to be
2936 in order not to need further allocation.)
2938 All in all, the new GC policy will make larger applications more
2941 The new GC scheme also is prepared for POSIX threading. Threads can
2942 allocate private pools of cells ("clusters") with just a single
2943 function call. Allocation of single cells from such a cluster can
2944 then proceed without any need of inter-thread synchronization.
2946 ** New environment variables controlling GC parameters
2948 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2951 Allocation of 2-word cell heaps:
2953 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2956 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2957 GC in percent of total heap size
2960 Allocation of 4-word cell heaps
2961 (used for real numbers and misc other objects):
2963 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2965 (See entry "Way for application to customize GC parameters" under
2966 section "Changes to the scm_ interface" below.)
2968 ** Guile now implements reals using 4-word cells
2970 This speeds up computation with reals. (They were earlier allocated
2971 with `malloc'.) There is still some room for optimizations, however.
2973 ** Some further steps toward POSIX thread support have been taken
2975 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2976 don't have much effect any longer, and many of them will be removed in
2980 are only handled at the top of the evaluator loop, immediately after
2981 I/O, and in scm_equalp.
2983 *** The GC can allocate thread private pools of pairs.
2985 * Changes to Scheme functions and syntax
2987 ** close-input-port and close-output-port are now R5RS
2989 These procedures have been turned into primitives and have R5RS behaviour.
2991 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2993 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2994 extended by the more sophisticated version in (ice-9 format)
2996 (simple-format port message . args)
2997 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2998 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2999 the escapes are replaced with corresponding members of ARGS:
3000 ~A formats using `display' and ~S formats using `write'.
3001 If DESTINATION is #t, then use the `current-output-port',
3002 if DESTINATION is #f, then return a string containing the formatted text.
3003 Does not add a trailing newline."
3005 ** string-ref: the second argument is no longer optional.
3007 ** string, list->string: no longer accept strings in their arguments,
3008 only characters, for compatibility with R5RS.
3010 ** New procedure: port-closed? PORT
3011 Returns #t if PORT is closed or #f if it is open.
3013 ** Deprecated: list*
3015 The list* functionality is now provided by cons* (SRFI-1 compliant)
3017 ** New procedure: cons* ARG1 ARG2 ... ARGn
3019 Like `list', but the last arg provides the tail of the constructed list,
3020 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3022 Requires at least one argument. If given one argument, that argument
3023 is returned as result.
3025 This function is called `list*' in some other Schemes and in Common LISP.
3027 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3029 ** New procedure: object-documentation OBJECT
3031 Returns the documentation string associated with OBJECT. The
3032 procedure uses a caching mechanism so that subsequent lookups are
3035 Exported by (ice-9 documentation).
3037 ** module-name now returns full names of modules
3039 Previously, only the last part of the name was returned (`session' for
3040 `(ice-9 session)'). Ex: `(ice-9 session)'.
3042 * Changes to the gh_ interface
3044 ** Deprecated: gh_int2scmb
3046 Use gh_bool2scm instead.
3048 * Changes to the scm_ interface
3050 ** Guile primitives now carry docstrings!
3052 Thanks to Greg Badros!
3054 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3056 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3057 macros and must contain a docstring that is extracted into foo.doc using a new
3058 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3060 However, a major overhaul of these macros is scheduled for the next release of
3063 ** Guile primitives use a new technique for validation of arguments
3065 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3066 the readability of argument checking.
3068 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3070 ** New macros: SCM_PACK, SCM_UNPACK
3072 Compose/decompose an SCM value.
3074 The SCM type is now treated as an abstract data type and may be defined as a
3075 long, a void* or as a struct, depending on the architecture and compile time
3076 options. This makes it easier to find several types of bugs, for example when
3077 SCM values are treated as integers without conversion. Values of the SCM type
3078 should be treated as "atomic" values. These macros are used when
3079 composing/decomposing an SCM value, either because you want to access
3080 individual bits, or because you want to treat it as an integer value.
3082 E.g., in order to set bit 7 in an SCM value x, use the expression
3084 SCM_PACK (SCM_UNPACK (x) | 0x80)
3086 ** The name property of hooks is deprecated.
3087 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3089 You can emulate this feature by using object properties.
3091 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3092 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3095 These macros will be removed in a future release of Guile.
3097 ** The following types, functions and macros from numbers.h are deprecated:
3098 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3099 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3101 ** Port internals: the rw_random variable in the scm_port structure
3102 must be set to non-zero in any random access port. In recent Guile
3103 releases it was only set for bidirectional random-access ports.
3105 ** Port internals: the seek ptob procedure is now responsible for
3106 resetting the buffers if required. The change was made so that in the
3107 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3108 the fport and strport ptobs can avoid resetting the buffers,
3109 in particular to avoid discarding unread chars. An existing port
3110 type can be fixed by adding something like the following to the
3111 beginning of the ptob seek procedure:
3113 if (pt->rw_active == SCM_PORT_READ)
3114 scm_end_input (object);
3115 else if (pt->rw_active == SCM_PORT_WRITE)
3116 ptob->flush (object);
3118 although to actually avoid resetting the buffers and discard unread
3119 chars requires further hacking that depends on the characteristics
3122 ** Deprecated functions: scm_fseek, scm_tag
3124 These functions are no longer used and will be removed in a future version.
3126 ** The scm_sysmissing procedure is no longer used in libguile.
3127 Unless it turns out to be unexpectedly useful to somebody, it will be
3128 removed in a future version.
3130 ** The format of error message strings has changed
3132 The two C procedures: scm_display_error and scm_error, as well as the
3133 primitive `scm-error', now use scm_simple_format to do their work.
3134 This means that the message strings of all code must be updated to use
3135 ~A where %s was used before, and ~S where %S was used before.
3137 During the period when there still are a lot of old Guiles out there,
3138 you might want to support both old and new versions of Guile.
3140 There are basically two methods to achieve this. Both methods use
3143 AC_CHECK_FUNCS(scm_simple_format)
3145 in your configure.in.
3147 Method 1: Use the string concatenation features of ANSI C's
3152 #ifdef HAVE_SCM_SIMPLE_FORMAT
3158 Then represent each of your error messages using a preprocessor macro:
3160 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3164 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3165 (define make-message string-append)
3167 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3169 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3173 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3178 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3182 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3184 Don't use the functions coop_mutex_init and
3185 coop_condition_variable_init. They will change.
3187 Use scm_mutex_init and scm_cond_init instead.
3189 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3190 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3191 COND, as `scm_cond_wait' does, but it also bounds the duration
3192 of the wait. If COND has not been signaled before time ABSTIME,
3193 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3194 returns the error code `ETIMEDOUT'.
3196 The ABSTIME parameter specifies an absolute time, with the same
3197 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3198 to 00:00:00 GMT, January 1, 1970.
3200 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3201 `scm_cond_broadcast' restarts all the threads that are waiting
3202 on the condition variable COND. Nothing happens if no threads are
3205 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3206 `scm_key_create' allocates a new TSD key. The key is stored in
3207 the location pointed to by KEY. There is no limit on the number
3208 of keys allocated at a given time. The value initially associated
3209 with the returned key is `NULL' in all currently executing threads.
3211 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3212 function associated with the key. When a thread terminates,
3213 DESTR_FUNCTION is called on the value associated with the key in
3214 that thread. The DESTR_FUNCTION is not called if a key is deleted
3215 with `scm_key_delete' or a value is changed with
3216 `scm_setspecific'. The order in which destructor functions are
3217 called at thread termination time is unspecified.
3219 Destructors are not yet implemented.
3221 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3222 `scm_setspecific' changes the value associated with KEY in the
3223 calling thread, storing the given POINTER instead.
3225 ** New function: scm_getspecific (scm_key_t KEY)
3226 `scm_getspecific' returns the value currently associated with
3227 KEY in the calling thread.
3229 ** New function: scm_key_delete (scm_key_t KEY)
3230 `scm_key_delete' deallocates a TSD key. It does not check
3231 whether non-`NULL' values are associated with that key in the
3232 currently executing threads, nor call the destructor function
3233 associated with the key.
3235 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3237 Initialize a C level hook HOOK with associated HOOK_DATA and type
3238 TYPE. (See scm_c_hook_run ().)
3240 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3242 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3243 is true, add it last, otherwise first. The same FUNC can be added
3244 multiple times if FUNC_DATA differ and vice versa.
3246 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3248 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3249 function is only removed if both FUNC and FUNC_DATA matches.
3251 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3253 Run hook HOOK passing DATA to the hook functions.
3255 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3256 returned is undefined.
3258 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3259 returns a non-NULL value. This value is returned as the result of
3260 scm_c_hook_run. If all functions return NULL, NULL is returned.
3262 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3263 returns a NULL value, and NULL is returned. If all functions returns
3264 a non-NULL value, the last value is returned.
3266 ** New C level GC hooks
3268 Five new C level hooks has been added to the garbage collector.
3270 scm_before_gc_c_hook
3273 are run before locking and after unlocking the heap. The system is
3274 thus in a mode where evaluation can take place. (Except that
3275 scm_before_gc_c_hook must not allocate new cells.)
3277 scm_before_mark_c_hook
3278 scm_before_sweep_c_hook
3279 scm_after_sweep_c_hook
3281 are run when the heap is locked. These are intended for extension of
3282 the GC in a modular fashion. Examples are the weaks and guardians
3285 ** Way for application to customize GC parameters
3287 The application can set up other default values for the GC heap
3288 allocation parameters
3290 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3291 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3292 GUILE_MAX_SEGMENT_SIZE,
3296 scm_default_init_heap_size_1, scm_default_min_yield_1,
3297 scm_default_init_heap_size_2, scm_default_min_yield_2,
3298 scm_default_max_segment_size
3300 respectively before callong scm_boot_guile.
3302 (See entry "New environment variables ..." in section
3303 "Changes to the stand-alone interpreter" above.)
3305 ** scm_protect_object/scm_unprotect_object now nest
3307 This means that you can call scm_protect_object multiple times on an
3308 object and count on the object being protected until
3309 scm_unprotect_object has been call the same number of times.
3311 The functions also have better time complexity.
3313 Still, it is usually possible to structure the application in a way
3314 that you don't need to use these functions. For example, if you use a
3315 protected standard Guile list to keep track of live objects rather
3316 than some custom data type, objects will die a natural death when they
3317 are no longer needed.
3319 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3321 Guile does not provide the float representation for inexact real numbers any
3322 more. Now, only doubles are used to represent inexact real numbers. Further,
3323 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3324 and scm_tc16_complex, respectively.
3326 ** Removed deprecated type scm_smobfuns
3328 ** Removed deprecated function scm_newsmob
3330 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3332 There is an ongoing discussion among the developers whether to
3333 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3334 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3335 until this issue has been settled.
3337 ** Removed deprecated type tag scm_tc16_kw
3339 ** Added type tag scm_tc16_keyword
3341 (This was introduced already in release 1.3.4 but was not documented
3344 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3346 * Changes to system call interfaces:
3348 ** The "select" procedure now tests port buffers for the ability to
3349 provide input or accept output. Previously only the underlying file
3350 descriptors were checked.
3352 ** New variable PIPE_BUF: the maximum number of bytes that can be
3353 atomically written to a pipe.
3355 ** If a facility is not available on the system when Guile is
3356 compiled, the corresponding primitive procedure will not be defined.
3357 Previously it would have been defined but would throw a system-error
3358 exception if called. Exception handlers which catch this case may
3359 need minor modification: an error will be thrown with key
3360 'unbound-variable instead of 'system-error. Alternatively it's
3361 now possible to use `defined?' to check whether the facility is
3364 ** Procedures which depend on the timezone should now give the correct
3365 result on systems which cache the TZ environment variable, even if TZ
3366 is changed without calling tzset.
3368 * Changes to the networking interfaces:
3370 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3371 long integers between network and host format. For now, it's not
3372 particularly convenient to do this kind of thing, but consider:
3374 (define write-network-long
3375 (lambda (value port)
3376 (let ((v (make-uniform-vector 1 1 0)))
3377 (uniform-vector-set! v 0 (htonl value))
3378 (uniform-vector-write v port))))
3380 (define read-network-long
3382 (let ((v (make-uniform-vector 1 1 0)))
3383 (uniform-vector-read! v port)
3384 (ntohl (uniform-vector-ref v 0)))))
3386 ** If inet-aton fails, it now throws an error with key 'misc-error
3387 instead of 'system-error, since errno is not relevant.
3389 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3390 specific keys instead of 'system-error. The latter is inappropriate
3391 since errno will not have been set. The keys are:
3392 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3394 ** sethostent, setnetent, setprotoent, setservent: now take an
3395 optional argument STAYOPEN, which specifies whether the database
3396 remains open after a database entry is accessed randomly (e.g., using
3397 gethostbyname for the hosts database.) The default is #f. Previously
3401 Changes since Guile 1.3.2:
3403 * Changes to the stand-alone interpreter
3407 An initial version of the Guile debugger written by Chris Hanson has
3408 been added. The debugger is still under development but is included
3409 in the distribution anyway since it is already quite useful.
3415 after an error to enter the debugger. Type `help' inside the debugger
3416 for a description of available commands.
3418 If you prefer to have stack frames numbered and printed in
3419 anti-chronological order and prefer up in the stack to be down on the
3420 screen as is the case in gdb, you can put
3422 (debug-enable 'backwards)
3424 in your .guile startup file. (However, this means that Guile can't
3425 use indentation to indicate stack level.)
3427 The debugger is autoloaded into Guile at the first use.
3429 ** Further enhancements to backtraces
3431 There is a new debug option `width' which controls the maximum width
3432 on the screen of printed stack frames. Fancy printing parameters
3433 ("level" and "length" as in Common LISP) are adaptively adjusted for
3434 each stack frame to give maximum information while still fitting
3435 within the bounds. If the stack frame can't be made to fit by
3436 adjusting parameters, it is simply cut off at the end. This is marked
3439 ** Some modules are now only loaded when the repl is started
3441 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3442 regex) are now loaded into (guile-user) only if the repl has been
3443 started. The effect is that the startup time for scripts has been
3444 reduced to 30% of what it was previously.
3446 Correctly written scripts load the modules they require at the top of
3447 the file and should not be affected by this change.
3449 ** Hooks are now represented as smobs
3451 * Changes to Scheme functions and syntax
3453 ** Readline support has changed again.
3455 The old (readline-activator) module is gone. Use (ice-9 readline)
3456 instead, which now contains all readline functionality. So the code
3457 to activate readline is now
3459 (use-modules (ice-9 readline))
3462 This should work at any time, including from the guile prompt.
3464 To avoid confusion about the terms of Guile's license, please only
3465 enable readline for your personal use; please don't make it the
3466 default for others. Here is why we make this rather odd-sounding
3469 Guile is normally licensed under a weakened form of the GNU General
3470 Public License, which allows you to link code with Guile without
3471 placing that code under the GPL. This exception is important to some
3474 However, since readline is distributed under the GNU General Public
3475 License, when you link Guile with readline, either statically or
3476 dynamically, you effectively change Guile's license to the strict GPL.
3477 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3478 which are normally permitted become forbidden. This is a rather
3479 non-obvious consequence of the licensing terms.
3481 So, to make sure things remain clear, please let people choose for
3482 themselves whether to link GPL'd libraries like readline with Guile.
3484 ** regexp-substitute/global has changed slightly, but incompatibly.
3486 If you include a function in the item list, the string of the match
3487 object it receives is the same string passed to
3488 regexp-substitute/global, not some suffix of that string.
3489 Correspondingly, the match's positions are relative to the entire
3490 string, not the suffix.
3492 If the regexp can match the empty string, the way matches are chosen
3493 from the string has changed. regexp-substitute/global recognizes the
3494 same set of matches that list-matches does; see below.
3496 ** New function: list-matches REGEXP STRING [FLAGS]
3498 Return a list of match objects, one for every non-overlapping, maximal
3499 match of REGEXP in STRING. The matches appear in left-to-right order.
3500 list-matches only reports matches of the empty string if there are no
3501 other matches which begin on, end at, or include the empty match's
3504 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3506 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3508 For each match of REGEXP in STRING, apply PROC to the match object,
3509 and the last value PROC returned, or INIT for the first call. Return
3510 the last value returned by PROC. We apply PROC to the matches as they
3511 appear from left to right.
3513 This function recognizes matches according to the same criteria as
3516 Thus, you could define list-matches like this:
3518 (define (list-matches regexp string . flags)
3519 (reverse! (apply fold-matches regexp string '() cons flags)))
3521 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3525 *** New function: hook? OBJ
3527 Return #t if OBJ is a hook, otherwise #f.
3529 *** New function: make-hook-with-name NAME [ARITY]
3531 Return a hook with name NAME and arity ARITY. The default value for
3532 ARITY is 0. The only effect of NAME is that it will appear when the
3533 hook object is printed to ease debugging.
3535 *** New function: hook-empty? HOOK
3537 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3539 *** New function: hook->list HOOK
3541 Return a list of the procedures that are called when run-hook is
3544 ** `map' signals an error if its argument lists are not all the same length.
3546 This is the behavior required by R5RS, so this change is really a bug
3547 fix. But it seems to affect a lot of people's code, so we're
3548 mentioning it here anyway.
3550 ** Print-state handling has been made more transparent
3552 Under certain circumstances, ports are represented as a port with an
3553 associated print state. Earlier, this pair was represented as a pair
3554 (see "Some magic has been added to the printer" below). It is now
3555 indistinguishable (almost; see `get-print-state') from a port on the
3558 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3560 Return a new port with the associated print state PRINT-STATE.
3562 *** New function: get-print-state OUTPUT-PORT
3564 Return the print state associated with this port if it exists,
3565 otherwise return #f.
3567 *** New function: directory-stream? OBJECT
3569 Returns true iff OBJECT is a directory stream --- the sort of object
3570 returned by `opendir'.
3572 ** New function: using-readline?
3574 Return #t if readline is in use in the current repl.
3576 ** structs will be removed in 1.4
3578 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3579 and use GOOPS objects as the fundamental record type.
3581 * Changes to the scm_ interface
3583 ** structs will be removed in 1.4
3585 The entire current struct interface (struct.c, struct.h) will be
3586 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3587 GOOPS objects as the fundamental record type.
3589 ** The internal representation of subr's has changed
3591 Instead of giving a hint to the subr name, the CAR field of the subr
3592 now contains an index to a subr entry in scm_subr_table.
3594 *** New variable: scm_subr_table
3596 An array of subr entries. A subr entry contains the name, properties
3597 and documentation associated with the subr. The properties and
3598 documentation slots are not yet used.
3600 ** A new scheme for "forwarding" calls to a builtin to a generic function
3602 It is now possible to extend the functionality of some Guile
3603 primitives by letting them defer a call to a GOOPS generic function on
3604 argument mismatch. This means that there is no loss of efficiency in
3609 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3610 (define-method + ((x <string>) (y <string>))
3611 (string-append x y))
3613 + will still be as efficient as usual in numerical calculations, but
3614 can also be used for concatenating strings.
3616 Who will be the first one to extend Guile's numerical tower to
3617 rationals? :) [OK, there a few other things to fix before this can
3618 be made in a clean way.]
3620 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3622 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3624 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3626 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3627 a variable GENERIC which can be used by the dispatch macros below.
3629 [This is experimental code which may change soon.]
3631 *** New macros for forwarding control to a generic on arg type error
3633 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3635 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3637 These correspond to the scm_wta function call, and have the same
3638 behaviour until the user has called the GOOPS primitive
3639 `enable-primitive-generic!'. After that, these macros will apply the
3640 generic function GENERIC to the argument(s) instead of calling
3643 [This is experimental code which may change soon.]
3645 *** New macros for argument testing with generic dispatch
3647 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3649 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3651 These correspond to the SCM_ASSERT macro, but will defer control to
3652 GENERIC on error after `enable-primitive-generic!' has been called.
3654 [This is experimental code which may change soon.]
3656 ** New function: SCM scm_eval_body (SCM body, SCM env)
3658 Evaluates the body of a special form.
3660 ** The internal representation of struct's has changed
3662 Previously, four slots were allocated for the procedure(s) of entities
3663 and operators. The motivation for this representation had to do with
3664 the structure of the evaluator, the wish to support tail-recursive
3665 generic functions, and efficiency. Since the generic function
3666 dispatch mechanism has changed, there is no longer a need for such an
3667 expensive representation, and the representation has been simplified.
3669 This should not make any difference for most users.
3671 ** GOOPS support has been cleaned up.
3673 Some code has been moved from eval.c to objects.c and code in both of
3674 these compilation units has been cleaned up and better structured.
3676 *** New functions for applying generic functions
3678 New function: SCM scm_apply_generic (GENERIC, ARGS)
3679 New function: SCM scm_call_generic_0 (GENERIC)
3680 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3681 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3682 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3684 ** Deprecated function: scm_make_named_hook
3686 It is now replaced by:
3688 ** New function: SCM scm_create_hook (const char *name, int arity)
3690 Creates a hook in the same way as make-hook above but also
3691 binds a variable named NAME to it.
3693 This is the typical way of creating a hook from C code.
3695 Currently, the variable is created in the "current" module.
3696 This might change when we get the new module system.
3698 [The behaviour is identical to scm_make_named_hook.]
3702 Changes since Guile 1.3:
3704 * Changes to mailing lists
3706 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3708 See the README file to find current addresses for all the Guile
3711 * Changes to the distribution
3713 ** Readline support is no longer included with Guile by default.
3715 Based on the different license terms of Guile and Readline, we
3716 concluded that Guile should not *by default* cause the linking of
3717 Readline into an application program. Readline support is now offered
3718 as a separate module, which is linked into an application only when
3719 you explicitly specify it.
3721 Although Guile is GNU software, its distribution terms add a special
3722 exception to the usual GNU General Public License (GPL). Guile's
3723 license includes a clause that allows you to link Guile with non-free
3724 programs. We add this exception so as not to put Guile at a
3725 disadvantage vis-a-vis other extensibility packages that support other
3728 In contrast, the GNU Readline library is distributed under the GNU
3729 General Public License pure and simple. This means that you may not
3730 link Readline, even dynamically, into an application unless it is
3731 distributed under a free software license that is compatible the GPL.
3733 Because of this difference in distribution terms, an application that
3734 can use Guile may not be able to use Readline. Now users will be
3735 explicitly offered two independent decisions about the use of these
3738 You can activate the readline support by issuing
3740 (use-modules (readline-activator))
3743 from your ".guile" file, for example.
3745 * Changes to the stand-alone interpreter
3747 ** All builtins now print as primitives.
3748 Previously builtin procedures not belonging to the fundamental subr
3749 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3750 Now, they print as #<primitive-procedure NAME>.
3752 ** Backtraces slightly more intelligible.
3753 gsubr-apply and macro transformer application frames no longer appear
3756 * Changes to Scheme functions and syntax
3758 ** Guile now correctly handles internal defines by rewriting them into
3759 their equivalent letrec. Previously, internal defines would
3760 incrementally add to the innermost environment, without checking
3761 whether the restrictions specified in RnRS were met. This lead to the
3762 correct behaviour when these restriction actually were met, but didn't
3763 catch all illegal uses. Such an illegal use could lead to crashes of
3764 the Guile interpreter or or other unwanted results. An example of
3765 incorrect internal defines that made Guile behave erratically:
3777 The problem with this example is that the definition of `c' uses the
3778 value of `b' directly. This confuses the meoization machine of Guile
3779 so that the second call of `b' (this time in a larger environment that
3780 also contains bindings for `c' and `d') refers to the binding of `c'
3781 instead of `a'. You could also make Guile crash with a variation on
3786 (define (b flag) (if flag a 1))
3787 (define c (1+ (b flag)))
3795 From now on, Guile will issue an `Unbound variable: b' error message
3800 A hook contains a list of functions which should be called on
3801 particular occasions in an existing program. Hooks are used for
3804 A window manager might have a hook before-window-map-hook. The window
3805 manager uses the function run-hooks to call all functions stored in
3806 before-window-map-hook each time a window is mapped. The user can
3807 store functions in the hook using add-hook!.
3809 In Guile, hooks are first class objects.
3811 *** New function: make-hook [N_ARGS]
3813 Return a hook for hook functions which can take N_ARGS arguments.
3814 The default value for N_ARGS is 0.
3816 (See also scm_make_named_hook below.)
3818 *** New function: add-hook! HOOK PROC [APPEND_P]
3820 Put PROC at the beginning of the list of functions stored in HOOK.
3821 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3823 PROC must be able to take the number of arguments specified when the
3826 If PROC already exists in HOOK, then remove it first.
3828 *** New function: remove-hook! HOOK PROC
3830 Remove PROC from the list of functions in HOOK.
3832 *** New function: reset-hook! HOOK
3834 Clear the list of hook functions stored in HOOK.
3836 *** New function: run-hook HOOK ARG1 ...
3838 Run all hook functions stored in HOOK with arguments ARG1 ... .
3839 The number of arguments supplied must correspond to the number given
3840 when the hook was created.
3842 ** The function `dynamic-link' now takes optional keyword arguments.
3843 The only keyword argument that is currently defined is `:global
3844 BOOL'. With it, you can control whether the shared library will be
3845 linked in global mode or not. In global mode, the symbols from the
3846 linked library can be used to resolve references from other
3847 dynamically linked libraries. In non-global mode, the linked
3848 library is essentially invisible and can only be accessed via
3849 `dynamic-func', etc. The default is now to link in global mode.
3850 Previously, the default has been non-global mode.
3852 The `#:global' keyword is only effective on platforms that support
3853 the dlopen family of functions.
3855 ** New function `provided?'
3857 - Function: provided? FEATURE
3858 Return true iff FEATURE is supported by this installation of
3859 Guile. FEATURE must be a symbol naming a feature; the global
3860 variable `*features*' is a list of available features.
3862 ** Changes to the module (ice-9 expect):
3864 *** The expect-strings macro now matches `$' in a regular expression
3865 only at a line-break or end-of-file by default. Previously it would
3866 match the end of the string accumulated so far. The old behaviour
3867 can be obtained by setting the variable `expect-strings-exec-flags'
3870 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3871 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3872 in a regular expression will still match before a line-break or
3873 end-of-file. The default is `regexp/noteol'.
3875 *** The expect-strings macro now uses a variable
3876 `expect-strings-compile-flags' for the flags to be supplied to
3877 `make-regexp'. The default is `regexp/newline', which was previously
3880 *** The expect macro now supplies two arguments to a match procedure:
3881 the current accumulated string and a flag to indicate whether
3882 end-of-file has been reached. Previously only the string was supplied.
3883 If end-of-file is reached, the match procedure will be called an
3884 additional time with the same accumulated string as the previous call
3885 but with the flag set.
3887 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3889 This code, and the documentation for it that appears here, was
3890 borrowed from SLIB, with minor adaptations for Guile.
3892 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3893 An almost complete implementation of Common LISP format description
3894 according to the CL reference book `Common LISP' from Guy L.
3895 Steele, Digital Press. Backward compatible to most of the
3896 available Scheme format implementations.
3898 Returns `#t', `#f' or a string; has side effect of printing
3899 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3900 to the current output port and `#t' is returned. If DESTINATION
3901 is `#f', a formatted string is returned as the result of the call.
3902 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3903 format string; FORMAT-STRING is then the first argument and the
3904 output is returned as a string. If DESTINATION is a number, the
3905 output is to the current error port if available by the
3906 implementation. Otherwise DESTINATION must be an output port and
3909 FORMAT-STRING must be a string. In case of a formatting error
3910 format returns `#f' and prints a message on the current output or
3911 error port. Characters are output as if the string were output by
3912 the `display' function with the exception of those prefixed by a
3913 tilde (~). For a detailed description of the FORMAT-STRING syntax
3914 please consult a Common LISP format reference manual. For a test
3915 suite to verify this format implementation load `formatst.scm'.
3916 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3918 Note: `format' is not reentrant, i.e. only one `format'-call may
3919 be executed at a time.
3922 *** Format Specification (Format version 3.0)
3924 Please consult a Common LISP format reference manual for a detailed
3925 description of the format string syntax. For a demonstration of the
3926 implemented directives see `formatst.scm'.
3928 This implementation supports directive parameters and modifiers (`:'
3929 and `@' characters). Multiple parameters must be separated by a comma
3930 (`,'). Parameters can be numerical parameters (positive or negative),
3931 character parameters (prefixed by a quote character (`''), variable
3932 parameters (`v'), number of rest arguments parameter (`#'), empty and
3933 default parameters. Directive characters are case independent. The
3934 general form of a directive is:
3936 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3938 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3940 *** Implemented CL Format Control Directives
3942 Documentation syntax: Uppercase characters represent the
3943 corresponding control directive characters. Lowercase characters
3944 represent control directive parameter descriptions.
3947 Any (print as `display' does).
3951 `~MINCOL,COLINC,MINPAD,PADCHARA'
3955 S-expression (print as `write' does).
3959 `~MINCOL,COLINC,MINPAD,PADCHARS'
3965 print number sign always.
3968 print comma separated.
3970 `~MINCOL,PADCHAR,COMMACHARD'
3976 print number sign always.
3979 print comma separated.
3981 `~MINCOL,PADCHAR,COMMACHARX'
3987 print number sign always.
3990 print comma separated.
3992 `~MINCOL,PADCHAR,COMMACHARO'
3998 print number sign always.
4001 print comma separated.
4003 `~MINCOL,PADCHAR,COMMACHARB'
4008 `~N,MINCOL,PADCHAR,COMMACHARR'
4012 print a number as a Roman numeral.
4015 print a number as an "old fashioned" Roman numeral.
4018 print a number as an ordinal English number.
4021 print a number as a cardinal English number.
4026 prints `y' and `ies'.
4029 as `~P but jumps 1 argument backward.'
4032 as `~@P but jumps 1 argument backward.'
4037 prints a character as the reader can understand it (i.e. `#\'
4041 prints a character as emacs does (eg. `^C' for ASCII 03).
4044 Fixed-format floating-point (prints a flonum like MMM.NNN).
4045 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4047 If the number is positive a plus sign is printed.
4050 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4051 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4053 If the number is positive a plus sign is printed.
4056 General floating-point (prints a flonum either fixed or
4058 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4060 If the number is positive a plus sign is printed.
4063 Dollars floating-point (prints a flonum in fixed with signs
4065 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4067 If the number is positive a plus sign is printed.
4070 A sign is always printed and appears before the padding.
4073 The sign appears before the padding.
4081 print newline if not at the beginning of the output line.
4083 prints `~&' and then N-1 newlines.
4088 print N page separators.
4098 newline is ignored, white space left.
4101 newline is left, white space ignored.
4106 relative tabulation.
4112 Indirection (expects indirect arguments as a list).
4114 extracts indirect arguments from format arguments.
4117 Case conversion (converts by `string-downcase').
4119 converts by `string-capitalize'.
4122 converts by `string-capitalize-first'.
4125 converts by `string-upcase'.
4128 Argument Jumping (jumps 1 argument forward).
4130 jumps N arguments forward.
4133 jumps 1 argument backward.
4136 jumps N arguments backward.
4139 jumps to the 0th argument.
4142 jumps to the Nth argument (beginning from 0)
4144 `~[STR0~;STR1~;...~;STRN~]'
4145 Conditional Expression (numerical clause conditional).
4147 take argument from N.
4150 true test conditional.
4153 if-else-then conditional.
4159 default clause follows.
4162 Iteration (args come from the next argument (a list)).
4164 at most N iterations.
4167 args from next arg (a list of lists).
4170 args from the rest of arguments.
4173 args from the rest args (lists).
4184 aborts if N <= M <= K
4186 *** Not Implemented CL Format Control Directives
4189 print `#f' as an empty list (see below).
4192 print `#f' as an empty list (see below).
4198 (sorry I don't understand its semantics completely)
4200 *** Extended, Replaced and Additional Control Directives
4202 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4203 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4204 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4205 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4206 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4207 COMMAWIDTH is the number of characters between two comma
4211 print a R4RS complex number as `~F~@Fi' with passed parameters for
4215 Pretty print formatting of an argument for scheme code lists.
4221 Flushes the output if format DESTINATION is a port.
4224 Print a `#\space' character
4226 print N `#\space' characters.
4229 Print a `#\tab' character
4231 print N `#\tab' characters.
4234 Takes N as an integer representation for a character. No arguments
4235 are consumed. N is converted to a character by `integer->char'. N
4236 must be a positive decimal number.
4239 Print out readproof. Prints out internal objects represented as
4240 `#<...>' as strings `"#<...>"' so that the format output can always
4241 be processed by `read'.
4244 Print out readproof. Prints out internal objects represented as
4245 `#<...>' as strings `"#<...>"' so that the format output can always
4246 be processed by `read'.
4249 Prints information and a copyright notice on the format
4252 prints format version.
4255 may also print number strings, i.e. passing a number as a string
4256 and format it accordingly.
4258 *** Configuration Variables
4260 The format module exports some configuration variables to suit the
4261 systems and users needs. There should be no modification necessary for
4262 the configuration that comes with Guile. Format detects automatically
4263 if the running scheme system implements floating point numbers and
4266 format:symbol-case-conv
4267 Symbols are converted by `symbol->string' so the case type of the
4268 printed symbols is implementation dependent.
4269 `format:symbol-case-conv' is a one arg closure which is either
4270 `#f' (no conversion), `string-upcase', `string-downcase' or
4271 `string-capitalize'. (default `#f')
4273 format:iobj-case-conv
4274 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4275 implementation internal objects. (default `#f')
4278 The character prefixing the exponent value in `~E' printing.
4281 *** Compatibility With Other Format Implementations
4287 Downward compatible except for padding support and `~A', `~S',
4288 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4289 `printf' padding support which is completely replaced by the CL
4290 `format' padding style.
4293 Downward compatible except for `~', which is not documented
4294 (ignores all characters inside the format string up to a newline
4295 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4296 numerical and variable parameters and `:/@' modifiers in the CL
4300 Downward compatible except for `~A' and `~S' which print in
4301 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4302 directive parameters or modifiers)).
4305 Downward compatible except for an optional destination parameter:
4306 S2C accepts a format call without a destination which returns a
4307 formatted string. This is equivalent to a #f destination in S2C.
4308 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4309 parameters or modifiers)).
4312 ** Changes to string-handling functions.
4314 These functions were added to support the (ice-9 format) module, above.
4316 *** New function: string-upcase STRING
4317 *** New function: string-downcase STRING
4319 These are non-destructive versions of the existing string-upcase! and
4320 string-downcase! functions.
4322 *** New function: string-capitalize! STRING
4323 *** New function: string-capitalize STRING
4325 These functions convert the first letter of each word in the string to
4328 (string-capitalize "howdy there")
4331 As with the other functions, string-capitalize! modifies the string in
4332 place, while string-capitalize returns a modified copy of its argument.
4334 *** New function: string-ci->symbol STRING
4336 Return a symbol whose name is STRING, but having the same case as if
4337 the symbol had be read by `read'.
4339 Guile can be configured to be sensitive or insensitive to case
4340 differences in Scheme identifiers. If Guile is case-insensitive, all
4341 symbols are converted to lower case on input. The `string-ci->symbol'
4342 function returns a symbol whose name in STRING, transformed as Guile
4343 would if STRING were input.
4345 *** New function: substring-move! STRING1 START END STRING2 START
4347 Copy the substring of STRING1 from START (inclusive) to END
4348 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4349 string, and the source and destination areas may overlap; in all
4350 cases, the function behaves as if all the characters were copied
4353 *** Extended functions: substring-move-left! substring-move-right!
4355 These functions now correctly copy arbitrarily overlapping substrings;
4356 they are both synonyms for substring-move!.
4359 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4361 getopt-long is a function for parsing command-line arguments in a
4362 manner consistent with other GNU programs.
4364 (getopt-long ARGS GRAMMAR)
4365 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4367 ARGS should be a list of strings. Its first element should be the
4368 name of the program; subsequent elements should be the arguments
4369 that were passed to the program on the command line. The
4370 `program-arguments' procedure returns a list of this form.
4372 GRAMMAR is a list of the form:
4373 ((OPTION (PROPERTY VALUE) ...) ...)
4375 Each OPTION should be a symbol. `getopt-long' will accept a
4376 command-line option named `--OPTION'.
4377 Each option can have the following (PROPERTY VALUE) pairs:
4379 (single-char CHAR) --- Accept `-CHAR' as a single-character
4380 equivalent to `--OPTION'. This is how to specify traditional
4382 (required? BOOL) --- If BOOL is true, the option is required.
4383 getopt-long will raise an error if it is not found in ARGS.
4384 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4385 it is #f, it does not; and if it is the symbol
4386 `optional', the option may appear in ARGS with or
4388 (predicate FUNC) --- If the option accepts a value (i.e. you
4389 specified `(value #t)' for this option), then getopt
4390 will apply FUNC to the value, and throw an exception
4391 if it returns #f. FUNC should be a procedure which
4392 accepts a string and returns a boolean value; you may
4393 need to use quasiquotes to get it into GRAMMAR.
4395 The (PROPERTY VALUE) pairs may occur in any order, but each
4396 property may occur only once. By default, options do not have
4397 single-character equivalents, are not required, and do not take
4400 In ARGS, single-character options may be combined, in the usual
4401 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4402 accepts values, then it must be the last option in the
4403 combination; the value is the next argument. So, for example, using
4404 the following grammar:
4405 ((apples (single-char #\a))
4406 (blimps (single-char #\b) (value #t))
4407 (catalexis (single-char #\c) (value #t)))
4408 the following argument lists would be acceptable:
4409 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4410 for "blimps" and "catalexis")
4411 ("-ab" "bang" "-c" "couth") (same)
4412 ("-ac" "couth" "-b" "bang") (same)
4413 ("-abc" "couth" "bang") (an error, since `-b' is not the
4414 last option in its combination)
4416 If an option's value is optional, then `getopt-long' decides
4417 whether it has a value by looking at what follows it in ARGS. If
4418 the next element is a string, and it does not appear to be an
4419 option itself, then that string is the option's value.
4421 The value of a long option can appear as the next element in ARGS,
4422 or it can follow the option name, separated by an `=' character.
4423 Thus, using the same grammar as above, the following argument lists
4425 ("--apples" "Braeburn" "--blimps" "Goodyear")
4426 ("--apples=Braeburn" "--blimps" "Goodyear")
4427 ("--blimps" "Goodyear" "--apples=Braeburn")
4429 If the option "--" appears in ARGS, argument parsing stops there;
4430 subsequent arguments are returned as ordinary arguments, even if
4431 they resemble options. So, in the argument list:
4432 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4433 `getopt-long' will recognize the `apples' option as having the
4434 value "Granny Smith", but it will not recognize the `blimp'
4435 option; it will return the strings "--blimp" and "Goodyear" as
4436 ordinary argument strings.
4438 The `getopt-long' function returns the parsed argument list as an
4439 assocation list, mapping option names --- the symbols from GRAMMAR
4440 --- onto their values, or #t if the option does not accept a value.
4441 Unused options do not appear in the alist.
4443 All arguments that are not the value of any option are returned
4444 as a list, associated with the empty list.
4446 `getopt-long' throws an exception if:
4447 - it finds an unrecognized option in ARGS
4448 - a required option is omitted
4449 - an option that requires an argument doesn't get one
4450 - an option that doesn't accept an argument does get one (this can
4451 only happen using the long option `--opt=value' syntax)
4452 - an option predicate fails
4457 `((lockfile-dir (required? #t)
4460 (predicate ,file-is-directory?))
4461 (verbose (required? #f)
4464 (x-includes (single-char #\x))
4465 (rnet-server (single-char #\y)
4466 (predicate ,string?))))
4468 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4469 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4471 => ((() "foo1" "-fred" "foo2" "foo3")
4472 (rnet-server . "lamprod")
4473 (x-includes . "/usr/include")
4474 (lockfile-dir . "/tmp")
4477 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4479 It will be removed in a few releases.
4481 ** New syntax: lambda*
4482 ** New syntax: define*
4483 ** New syntax: define*-public
4484 ** New syntax: defmacro*
4485 ** New syntax: defmacro*-public
4486 Guile now supports optional arguments.
4488 `lambda*', `define*', `define*-public', `defmacro*' and
4489 `defmacro*-public' are identical to the non-* versions except that
4490 they use an extended type of parameter list that has the following BNF
4491 syntax (parentheses are literal, square brackets indicate grouping,
4492 and `*', `+' and `?' have the usual meaning):
4494 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4495 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4496 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4498 ext-var-decl ::= identifier | ( identifier expression )
4500 The semantics are best illustrated with the following documentation
4501 and examples for `lambda*':
4504 lambda extended for optional and keyword arguments
4506 lambda* creates a procedure that takes optional arguments. These
4507 are specified by putting them inside brackets at the end of the
4508 paramater list, but before any dotted rest argument. For example,
4509 (lambda* (a b #&optional c d . e) '())
4510 creates a procedure with fixed arguments a and b, optional arguments c
4511 and d, and rest argument e. If the optional arguments are omitted
4512 in a call, the variables for them are unbound in the procedure. This
4513 can be checked with the bound? macro.
4515 lambda* can also take keyword arguments. For example, a procedure
4517 (lambda* (#&key xyzzy larch) '())
4518 can be called with any of the argument lists (#:xyzzy 11)
4519 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4520 are given as keywords are bound to values.
4522 Optional and keyword arguments can also be given default values
4523 which they take on when they are not present in a call, by giving a
4524 two-item list in place of an optional argument, for example in:
4525 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4526 foo is a fixed argument, bar is an optional argument with default
4527 value 42, and baz is a keyword argument with default value 73.
4528 Default value expressions are not evaluated unless they are needed
4529 and until the procedure is called.
4531 lambda* now supports two more special parameter list keywords.
4533 lambda*-defined procedures now throw an error by default if a
4534 keyword other than one of those specified is found in the actual
4535 passed arguments. However, specifying #&allow-other-keys
4536 immediately after the kyword argument declarations restores the
4537 previous behavior of ignoring unknown keywords. lambda* also now
4538 guarantees that if the same keyword is passed more than once, the
4539 last one passed is the one that takes effect. For example,
4540 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4541 #:heads 37 #:tails 42 #:heads 99)
4542 would result in (99 47) being displayed.
4544 #&rest is also now provided as a synonym for the dotted syntax rest
4545 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4546 all respects to lambda*. This is provided for more similarity to DSSSL,
4547 MIT-Scheme and Kawa among others, as well as for refugees from other
4550 Further documentation may be found in the optargs.scm file itself.
4552 The optional argument module also exports the macros `let-optional',
4553 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4554 are not documented here because they may be removed in the future, but
4555 full documentation is still available in optargs.scm.
4557 ** New syntax: and-let*
4558 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4560 Syntax: (land* (<clause> ...) <body> ...)
4561 Each <clause> should have one of the following forms:
4562 (<variable> <expression>)
4565 Each <variable> or <bound-variable> should be an identifier. Each
4566 <expression> should be a valid expression. The <body> should be a
4567 possibly empty sequence of expressions, like the <body> of a
4570 Semantics: A LAND* expression is evaluated by evaluating the
4571 <expression> or <bound-variable> of each of the <clause>s from
4572 left to right. The value of the first <expression> or
4573 <bound-variable> that evaluates to a false value is returned; the
4574 remaining <expression>s and <bound-variable>s are not evaluated.
4575 The <body> forms are evaluated iff all the <expression>s and
4576 <bound-variable>s evaluate to true values.
4578 The <expression>s and the <body> are evaluated in an environment
4579 binding each <variable> of the preceding (<variable> <expression>)
4580 clauses to the value of the <expression>. Later bindings
4581 shadow earlier bindings.
4583 Guile's and-let* macro was contributed by Michael Livshin.
4585 ** New sorting functions
4587 *** New function: sorted? SEQUENCE LESS?
4588 Returns `#t' when the sequence argument is in non-decreasing order
4589 according to LESS? (that is, there is no adjacent pair `... x y
4590 ...' for which `(less? y x)').
4592 Returns `#f' when the sequence contains at least one out-of-order
4593 pair. It is an error if the sequence is neither a list nor a
4596 *** New function: merge LIST1 LIST2 LESS?
4597 LIST1 and LIST2 are sorted lists.
4598 Returns the sorted list of all elements in LIST1 and LIST2.
4600 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4601 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4602 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4603 (Here "<" should read "comes before".)
4605 *** New procedure: merge! LIST1 LIST2 LESS?
4606 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4607 the result. If the code is compiled, and LESS? constructs no new
4608 pairs, no pairs at all will be allocated. The first pair of the
4609 result will be either the first pair of LIST1 or the first pair of
4612 *** New function: sort SEQUENCE LESS?
4613 Accepts either a list or a vector, and returns a new sequence
4614 which is sorted. The new sequence is the same type as the input.
4615 Always `(sorted? (sort sequence less?) less?)'. The original
4616 sequence is not altered in any way. The new sequence shares its
4617 elements with the old one; no elements are copied.
4619 *** New procedure: sort! SEQUENCE LESS
4620 Returns its sorted result in the original boxes. No new storage is
4621 allocated at all. Proper usage: (set! slist (sort! slist <))
4623 *** New function: stable-sort SEQUENCE LESS?
4624 Similar to `sort' but stable. That is, if "equal" elements are
4625 ordered a < b in the original sequence, they will have the same order
4628 *** New function: stable-sort! SEQUENCE LESS?
4629 Similar to `sort!' but stable.
4630 Uses temporary storage when sorting vectors.
4632 *** New functions: sort-list, sort-list!
4633 Added for compatibility with scsh.
4635 ** New built-in random number support
4637 *** New function: random N [STATE]
4638 Accepts a positive integer or real N and returns a number of the
4639 same type between zero (inclusive) and N (exclusive). The values
4640 returned have a uniform distribution.
4642 The optional argument STATE must be of the type produced by
4643 `copy-random-state' or `seed->random-state'. It defaults to the value
4644 of the variable `*random-state*'. This object is used to maintain the
4645 state of the pseudo-random-number generator and is altered as a side
4646 effect of the `random' operation.
4648 *** New variable: *random-state*
4649 Holds a data structure that encodes the internal state of the
4650 random-number generator that `random' uses by default. The nature
4651 of this data structure is implementation-dependent. It may be
4652 printed out and successfully read back in, but may or may not
4653 function correctly as a random-number state object in another
4656 *** New function: copy-random-state [STATE]
4657 Returns a new object of type suitable for use as the value of the
4658 variable `*random-state*' and as a second argument to `random'.
4659 If argument STATE is given, a copy of it is returned. Otherwise a
4660 copy of `*random-state*' is returned.
4662 *** New function: seed->random-state SEED
4663 Returns a new object of type suitable for use as the value of the
4664 variable `*random-state*' and as a second argument to `random'.
4665 SEED is a string or a number. A new state is generated and
4666 initialized using SEED.
4668 *** New function: random:uniform [STATE]
4669 Returns an uniformly distributed inexact real random number in the
4670 range between 0 and 1.
4672 *** New procedure: random:solid-sphere! VECT [STATE]
4673 Fills VECT with inexact real random numbers the sum of whose
4674 squares is less than 1.0. Thinking of VECT as coordinates in
4675 space of dimension N = `(vector-length VECT)', the coordinates are
4676 uniformly distributed within the unit N-shere. The sum of the
4677 squares of the numbers is returned. VECT can be either a vector
4678 or a uniform vector of doubles.
4680 *** New procedure: random:hollow-sphere! VECT [STATE]
4681 Fills VECT with inexact real random numbers the sum of whose squares
4682 is equal to 1.0. Thinking of VECT as coordinates in space of
4683 dimension n = `(vector-length VECT)', the coordinates are uniformly
4684 distributed over the surface of the unit n-shere. VECT can be either
4685 a vector or a uniform vector of doubles.
4687 *** New function: random:normal [STATE]
4688 Returns an inexact real in a normal distribution with mean 0 and
4689 standard deviation 1. For a normal distribution with mean M and
4690 standard deviation D use `(+ M (* D (random:normal)))'.
4692 *** New procedure: random:normal-vector! VECT [STATE]
4693 Fills VECT with inexact real random numbers which are independent and
4694 standard normally distributed (i.e., with mean 0 and variance 1).
4695 VECT can be either a vector or a uniform vector of doubles.
4697 *** New function: random:exp STATE
4698 Returns an inexact real in an exponential distribution with mean 1.
4699 For an exponential distribution with mean U use (* U (random:exp)).
4701 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4703 These functions now operate on numbers in the range of a C unsigned
4706 These functions used to operate on numbers in the range of a C signed
4707 long; however, this seems inappropriate, because Guile integers don't
4710 ** New function: make-guardian
4711 This is an implementation of guardians as described in
4712 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4713 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4714 Programming Language Design and Implementation, June 1993
4715 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4717 ** New functions: delq1!, delv1!, delete1!
4718 These procedures behave similar to delq! and friends but delete only
4719 one object if at all.
4721 ** New function: unread-string STRING PORT
4722 Unread STRING to PORT, that is, push it back onto the port so that
4723 next read operation will work on the pushed back characters.
4725 ** unread-char can now be called multiple times
4726 If unread-char is called multiple times, the unread characters will be
4727 read again in last-in first-out order.
4729 ** the procedures uniform-array-read! and uniform-array-write! now
4730 work on any kind of port, not just ports which are open on a file.
4732 ** Now 'l' in a port mode requests line buffering.
4734 ** The procedure truncate-file now works on string ports as well
4735 as file ports. If the size argument is omitted, the current
4736 file position is used.
4738 ** new procedure: seek PORT/FDES OFFSET WHENCE
4739 The arguments are the same as for the old fseek procedure, but it
4740 works on string ports as well as random-access file ports.
4742 ** the fseek procedure now works on string ports, since it has been
4743 redefined using seek.
4745 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4746 size is not supplied.
4748 ** the newline procedure no longer flushes the port if it's not
4749 line-buffered: previously it did if it was the current output port.
4751 ** open-pipe and close-pipe are no longer primitive procedures, but
4752 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4754 ** the freopen procedure has been removed.
4756 ** new procedure: drain-input PORT
4757 Drains PORT's read buffers (including any pushed-back characters)
4758 and returns the contents as a single string.
4760 ** New function: map-in-order PROC LIST1 LIST2 ...
4761 Version of `map' which guarantees that the procedure is applied to the
4762 lists in serial order.
4764 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4765 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4766 now obsolete and will go away in release 1.5.
4768 ** New syntax: collect BODY1 ...
4769 Version of `begin' which returns a list of the results of the body
4770 forms instead of the result of the last body form. In contrast to
4771 `begin', `collect' allows an empty body.
4773 ** New functions: read-history FILENAME, write-history FILENAME
4774 Read/write command line history from/to file. Returns #t on success
4775 and #f if an error occured.
4777 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4779 These procedures return a list of definitions available in the specified
4780 argument, a relative module reference. In the case of no argument,
4781 `(current-module)' is now consulted for definitions to return, instead
4782 of simply returning #f, the former behavior.
4784 ** The #/ syntax for lists is no longer supported.
4786 Earlier versions of Scheme accepted this syntax, but printed a
4789 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4791 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4794 * Changes to the gh_ interface
4798 Now takes a second argument which is the result array. If this
4799 pointer is NULL, a new array is malloced (the old behaviour).
4801 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4802 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4806 * Changes to the scm_ interface
4808 ** Function: scm_make_named_hook (char* name, int n_args)
4810 Creates a hook in the same way as make-hook above but also
4811 binds a variable named NAME to it.
4813 This is the typical way of creating a hook from C code.
4815 Currently, the variable is created in the "current" module. This
4816 might change when we get the new module system.
4818 ** The smob interface
4820 The interface for creating smobs has changed. For documentation, see
4821 data-rep.info (made from guile-core/doc/data-rep.texi).
4823 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4825 >>> This function will be removed in 1.3.4. <<<
4829 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4830 This function adds a new smob type, named NAME, with instance size
4831 SIZE to the system. The return value is a tag that is used in
4832 creating instances of the type. If SIZE is 0, then no memory will
4833 be allocated when instances of the smob are created, and nothing
4834 will be freed by the default free function.
4836 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4837 This function sets the smob marking procedure for the smob type
4838 specified by the tag TC. TC is the tag returned by
4839 `scm_make_smob_type'.
4841 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4842 This function sets the smob freeing procedure for the smob type
4843 specified by the tag TC. TC is the tag returned by
4844 `scm_make_smob_type'.
4846 *** Function: void scm_set_smob_print (tc, print)
4848 - Function: void scm_set_smob_print (long tc,
4849 scm_sizet (*print) (SCM,
4853 This function sets the smob printing procedure for the smob type
4854 specified by the tag TC. TC is the tag returned by
4855 `scm_make_smob_type'.
4857 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4858 This function sets the smob equality-testing predicate for the
4859 smob type specified by the tag TC. TC is the tag returned by
4860 `scm_make_smob_type'.
4862 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4863 Make VALUE contain a smob instance of the type with type code TC and
4864 smob data DATA. VALUE must be previously declared as C type `SCM'.
4866 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4867 This macro expands to a block of code that creates a smob instance
4868 of the type with type code TC and smob data DATA, and returns that
4869 `SCM' value. It should be the last piece of code in a block.
4871 ** The interfaces for using I/O ports and implementing port types
4872 (ptobs) have changed significantly. The new interface is based on
4873 shared access to buffers and a new set of ptob procedures.
4875 *** scm_newptob has been removed
4879 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4881 - Function: SCM scm_make_port_type (char *type_name,
4882 int (*fill_buffer) (SCM port),
4883 void (*write_flush) (SCM port));
4885 Similarly to the new smob interface, there is a set of function
4886 setters by which the user can customize the behaviour of his port
4887 type. See ports.h (scm_set_port_XXX).
4889 ** scm_strport_to_string: New function: creates a new string from
4890 a string port's buffer.
4892 ** Plug in interface for random number generators
4893 The variable `scm_the_rng' in random.c contains a value and three
4894 function pointers which together define the current random number
4895 generator being used by the Scheme level interface and the random
4896 number library functions.
4898 The user is free to replace the default generator with the generator
4901 *** Variable: size_t scm_the_rng.rstate_size
4902 The size of the random state type used by the current RNG
4905 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4906 Given the random STATE, return 32 random bits.
4908 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4909 Seed random state STATE using string S of length N.
4911 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4912 Given random state STATE, return a malloced copy.
4915 The default RNG is the MWC (Multiply With Carry) random number
4916 generator described by George Marsaglia at the Department of
4917 Statistics and Supercomputer Computations Research Institute, The
4918 Florida State University (http://stat.fsu.edu/~geo).
4920 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4921 passes all tests in the DIEHARD test suite
4922 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4923 costs one multiply and one add on platforms which either supports long
4924 longs (gcc does this on most systems) or have 64 bit longs. The cost
4925 is four multiply on other systems but this can be optimized by writing
4926 scm_i_uniform32 in assembler.
4928 These functions are provided through the scm_the_rng interface for use
4929 by libguile and the application.
4931 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4932 Given the random STATE, return 32 random bits.
4933 Don't use this function directly. Instead go through the plugin
4934 interface (see "Plug in interface" above).
4936 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4937 Initialize STATE using SEED of length N.
4939 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4940 Return a malloc:ed copy of STATE. This function can easily be re-used
4941 in the interfaces to other RNGs.
4943 ** Random number library functions
4944 These functions use the current RNG through the scm_the_rng interface.
4945 It might be a good idea to use these functions from your C code so
4946 that only one random generator is used by all code in your program.
4948 The default random state is stored in:
4950 *** Variable: SCM scm_var_random_state
4951 Contains the vcell of the Scheme variable "*random-state*" which is
4952 used as default state by all random number functions in the Scheme
4957 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4959 *** Function: scm_rstate *scm_c_default_rstate (void)
4960 This is a convenience function which returns the value of
4961 scm_var_random_state. An error message is generated if this value
4962 isn't a random state.
4964 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4965 Make a new random state from the string SEED of length LENGTH.
4967 It is generally not a good idea to use multiple random states in a
4968 program. While subsequent random numbers generated from one random
4969 state are guaranteed to be reasonably independent, there is no such
4970 guarantee for numbers generated from different random states.
4972 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4973 Return 32 random bits.
4975 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4976 Return a sample from the uniform(0,1) distribution.
4978 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4979 Return a sample from the normal(0,1) distribution.
4981 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4982 Return a sample from the exp(1) distribution.
4984 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4985 Return a sample from the discrete uniform(0,M) distribution.
4987 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4988 Return a sample from the discrete uniform(0,M) distribution.
4989 M must be a bignum object. The returned value may be an INUM.
4993 Changes in Guile 1.3 (released Monday, October 19, 1998):
4995 * Changes to the distribution
4997 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4998 To avoid conflicts, programs should name environment variables after
4999 themselves, except when there's a common practice establishing some
5002 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5003 giving the former precedence, and printing a warning message if the
5004 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5006 ** The header files related to multi-byte characters have been removed.
5007 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5008 which referred to these explicitly will probably need to be rewritten,
5009 since the support for the variant string types has been removed; see
5012 ** The header files append.h and sequences.h have been removed. These
5013 files implemented non-R4RS operations which would encourage
5014 non-portable programming style and less easy-to-read code.
5016 * Changes to the stand-alone interpreter
5018 ** New procedures have been added to implement a "batch mode":
5020 *** Function: batch-mode?
5022 Returns a boolean indicating whether the interpreter is in batch
5025 *** Function: set-batch-mode?! ARG
5027 If ARG is true, switches the interpreter to batch mode. The `#f'
5028 case has not been implemented.
5030 ** Guile now provides full command-line editing, when run interactively.
5031 To use this feature, you must have the readline library installed.
5032 The Guile build process will notice it, and automatically include
5035 The readline library is available via anonymous FTP from any GNU
5036 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5038 ** the-last-stack is now a fluid.
5040 * Changes to the procedure for linking libguile with your programs
5042 ** You can now use the `guile-config' utility to build programs that use Guile.
5044 Guile now includes a command-line utility called `guile-config', which
5045 can provide information about how to compile and link programs that
5048 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5049 You should include this command's output on the command line you use
5050 to compile C or C++ code that #includes the Guile header files. It's
5051 usually just a `-I' flag to help the compiler find the Guile headers.
5054 *** `guile-config link' prints any linker flags necessary to link with Guile.
5056 This command writes to its standard output a list of flags which you
5057 must pass to the linker to link your code against the Guile library.
5058 The flags include '-lguile' itself, any other libraries the Guile
5059 library depends upon, and any `-L' flags needed to help the linker
5060 find those libraries.
5062 For example, here is a Makefile rule that builds a program named 'foo'
5063 from the object files ${FOO_OBJECTS}, and links them against Guile:
5066 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5068 Previous Guile releases recommended that you use autoconf to detect
5069 which of a predefined set of libraries were present on your system.
5070 It is more robust to use `guile-config', since it records exactly which
5071 libraries the installed Guile library requires.
5073 This was originally called `build-guile', but was renamed to
5074 `guile-config' before Guile 1.3 was released, to be consistent with
5075 the analogous script for the GTK+ GUI toolkit, which is called
5079 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5081 If you are using the GNU autoconf package to configure your program,
5082 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5083 (described above) and gather the necessary values for use in your
5086 The GUILE_FLAGS macro expands to configure script code which runs the
5087 `guile-config' script, to find out where Guile's header files and
5088 libraries are installed. It sets two variables, marked for
5089 substitution, as by AC_SUBST.
5091 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5092 code that uses Guile header files. This is almost always just a
5095 GUILE_LDFLAGS --- flags to pass to the linker to link a
5096 program against Guile. This includes `-lguile' for the Guile
5097 library itself, any libraries that Guile itself requires (like
5098 -lqthreads), and so on. It may also include a -L flag to tell the
5099 compiler where to find the libraries.
5101 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5102 directory of the Guile distribution. You can copy it into your
5103 package's aclocal.m4 file, and then use it in your configure.in file.
5105 If you are using the `aclocal' program, distributed with GNU automake,
5106 to maintain your aclocal.m4 file, the Guile installation process
5107 installs guile.m4 where aclocal will find it. All you need to do is
5108 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5109 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5113 * Changes to Scheme functions and syntax
5115 ** Multi-byte strings have been removed, as have multi-byte and wide
5116 ports. We felt that these were the wrong approach to
5117 internationalization support.
5119 ** New function: readline [PROMPT]
5120 Read a line from the terminal, and allow the user to edit it,
5121 prompting with PROMPT. READLINE provides a large set of Emacs-like
5122 editing commands, lets the user recall previously typed lines, and
5123 works on almost every kind of terminal, including dumb terminals.
5125 READLINE assumes that the cursor is at the beginning of the line when
5126 it is invoked. Thus, you can't print a prompt yourself, and then call
5127 READLINE; you need to package up your prompt as a string, pass it to
5128 the function, and let READLINE print the prompt itself. This is
5129 because READLINE needs to know the prompt's screen width.
5131 For Guile to provide this function, you must have the readline
5132 library, version 2.1 or later, installed on your system. Readline is
5133 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5134 any GNU mirror site.
5136 See also ADD-HISTORY function.
5138 ** New function: add-history STRING
5139 Add STRING as the most recent line in the history used by the READLINE
5140 command. READLINE does not add lines to the history itself; you must
5141 call ADD-HISTORY to make previous input available to the user.
5143 ** The behavior of the read-line function has changed.
5145 This function now uses standard C library functions to read the line,
5146 for speed. This means that it doesn not respect the value of
5147 scm-line-incrementors; it assumes that lines are delimited with
5150 (Note that this is read-line, the function that reads a line of text
5151 from a port, not readline, the function that reads a line from a
5152 terminal, providing full editing capabilities.)
5154 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5156 This module provides some simple argument parsing. It exports one
5159 Function: getopt-gnu-style ARG-LS
5160 Parse a list of program arguments into an alist of option
5163 Each item in the list of program arguments is examined to see if
5164 it meets the syntax of a GNU long-named option. An argument like
5165 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5166 returned alist, where MUMBLE is a keyword object with the same
5167 name as the argument. An argument like `--MUMBLE=FROB' produces
5168 an element of the form (MUMBLE . FROB), where FROB is a string.
5170 As a special case, the returned alist also contains a pair whose
5171 car is the symbol `rest'. The cdr of this pair is a list
5172 containing all the items in the argument list that are not options
5173 of the form mentioned above.
5175 The argument `--' is treated specially: all items in the argument
5176 list appearing after such an argument are not examined, and are
5177 returned in the special `rest' list.
5179 This function does not parse normal single-character switches.
5180 You will need to parse them out of the `rest' list yourself.
5182 ** The read syntax for byte vectors and short vectors has changed.
5184 Instead of #bytes(...), write #y(...).
5186 Instead of #short(...), write #h(...).
5188 This may seem nutty, but, like the other uniform vectors, byte vectors
5189 and short vectors want to have the same print and read syntax (and,
5190 more basic, want to have read syntax!). Changing the read syntax to
5191 use multiple characters after the hash sign breaks with the
5192 conventions used in R5RS and the conventions used for the other
5193 uniform vectors. It also introduces complexity in the current reader,
5194 both on the C and Scheme levels. (The Right solution is probably to
5195 change the syntax and prototypes for uniform vectors entirely.)
5198 ** The new module (ice-9 session) provides useful interactive functions.
5200 *** New procedure: (apropos REGEXP OPTION ...)
5202 Display a list of top-level variables whose names match REGEXP, and
5203 the modules they are imported from. Each OPTION should be one of the
5206 value --- Show the value of each matching variable.
5207 shadow --- Show bindings shadowed by subsequently imported modules.
5208 full --- Same as both `shadow' and `value'.
5212 guile> (apropos "trace" 'full)
5213 debug: trace #<procedure trace args>
5214 debug: untrace #<procedure untrace args>
5215 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5216 the-scm-module: before-backtrace-hook ()
5217 the-scm-module: backtrace #<primitive-procedure backtrace>
5218 the-scm-module: after-backtrace-hook ()
5219 the-scm-module: has-shown-backtrace-hint? #f
5222 ** There are new functions and syntax for working with macros.
5224 Guile implements macros as a special object type. Any variable whose
5225 top-level binding is a macro object acts as a macro. The macro object
5226 specifies how the expression should be transformed before evaluation.
5228 *** Macro objects now print in a reasonable way, resembling procedures.
5230 *** New function: (macro? OBJ)
5231 True iff OBJ is a macro object.
5233 *** New function: (primitive-macro? OBJ)
5234 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5235 macro transformers, implemented in eval.c rather than Scheme code.
5237 Why do we have this function?
5238 - For symmetry with procedure? and primitive-procedure?,
5239 - to allow custom print procedures to tell whether a macro is
5240 primitive, and display it differently, and
5241 - to allow compilers and user-written evaluators to distinguish
5242 builtin special forms from user-defined ones, which could be
5245 *** New function: (macro-type OBJ)
5246 Return a value indicating what kind of macro OBJ is. Possible return
5249 The symbol `syntax' --- a macro created by procedure->syntax.
5250 The symbol `macro' --- a macro created by procedure->macro.
5251 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5252 The boolean #f --- if OBJ is not a macro object.
5254 *** New function: (macro-name MACRO)
5255 Return the name of the macro object MACRO's procedure, as returned by
5258 *** New function: (macro-transformer MACRO)
5259 Return the transformer procedure for MACRO.
5261 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5263 Specify a new macro expander to use in the current module. Each
5264 MODULE is a module name, with the same meaning as in the `use-modules'
5265 form; each named module's exported bindings are added to the current
5266 top-level environment. TRANSFORMER is an expression evaluated in the
5267 resulting environment which must yield a procedure to use as the
5268 module's eval transformer: every expression evaluated in this module
5269 is passed to this function, and the result passed to the Guile
5272 *** macro-eval! is removed. Use local-eval instead.
5274 ** Some magic has been added to the printer to better handle user
5275 written printing routines (like record printers, closure printers).
5277 The problem is that these user written routines must have access to
5278 the current `print-state' to be able to handle fancy things like
5279 detection of circular references. These print-states have to be
5280 passed to the builtin printing routines (display, write, etc) to
5281 properly continue the print chain.
5283 We didn't want to change all existing print code so that it
5284 explicitly passes thru a print state in addition to a port. Instead,
5285 we extented the possible values that the builtin printing routines
5286 accept as a `port'. In addition to a normal port, they now also take
5287 a pair of a normal port and a print-state. Printing will go to the
5288 port and the print-state will be used to control the detection of
5289 circular references, etc. If the builtin function does not care for a
5290 print-state, it is simply ignored.
5292 User written callbacks are now called with such a pair as their
5293 `port', but because every function now accepts this pair as a PORT
5294 argument, you don't have to worry about that. In fact, it is probably
5295 safest to not check for these pairs.
5297 However, it is sometimes necessary to continue a print chain on a
5298 different port, for example to get a intermediate string
5299 representation of the printed value, mangle that string somehow, and
5300 then to finally print the mangled string. Use the new function
5302 inherit-print-state OLD-PORT NEW-PORT
5304 for this. It constructs a new `port' that prints to NEW-PORT but
5305 inherits the print-state of OLD-PORT.
5307 ** struct-vtable-offset renamed to vtable-offset-user
5309 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5311 ** There is now a third optional argument to make-vtable-vtable
5312 (and fourth to make-struct) when constructing new types (vtables).
5313 This argument initializes field vtable-index-printer of the vtable.
5315 ** The detection of circular references has been extended to structs.
5316 That is, a structure that -- in the process of being printed -- prints
5317 itself does not lead to infinite recursion.
5319 ** There is now some basic support for fluids. Please read
5320 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5321 the following functions and macros:
5323 Function: make-fluid
5325 Create a new fluid object. Fluids are not special variables or
5326 some other extension to the semantics of Scheme, but rather
5327 ordinary Scheme objects. You can store them into variables (that
5328 are still lexically scoped, of course) or into any other place you
5329 like. Every fluid has a initial value of `#f'.
5331 Function: fluid? OBJ
5333 Test whether OBJ is a fluid.
5335 Function: fluid-ref FLUID
5336 Function: fluid-set! FLUID VAL
5338 Access/modify the fluid FLUID. Modifications are only visible
5339 within the current dynamic root (that includes threads).
5341 Function: with-fluids* FLUIDS VALUES THUNK
5343 FLUIDS is a list of fluids and VALUES a corresponding list of
5344 values for these fluids. Before THUNK gets called the values are
5345 installed in the fluids and the old values of the fluids are
5346 saved in the VALUES list. When the flow of control leaves THUNK
5347 or reenters it, the values get swapped again. You might think of
5348 this as a `safe-fluid-excursion'. Note that the VALUES list is
5349 modified by `with-fluids*'.
5351 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5353 The same as `with-fluids*' but with a different syntax. It looks
5354 just like `let', but both FLUID and VALUE are evaluated. Remember,
5355 fluids are not special variables but ordinary objects. FLUID
5356 should evaluate to a fluid.
5358 ** Changes to system call interfaces:
5360 *** close-port, close-input-port and close-output-port now return a
5361 boolean instead of an `unspecified' object. #t means that the port
5362 was successfully closed, while #f means it was already closed. It is
5363 also now possible for these procedures to raise an exception if an
5364 error occurs (some errors from write can be delayed until close.)
5366 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5369 *** the third argument to fcntl is now optional.
5371 *** the first argument to chown can now be a file descriptor or a port.
5373 *** the argument to stat can now be a port.
5375 *** The following new procedures have been added (most use scsh
5378 *** procedure: close PORT/FD
5379 Similar to close-port (*note close-port: Closing Ports.), but also
5380 works on file descriptors. A side effect of closing a file
5381 descriptor is that any ports using that file descriptor are moved
5382 to a different file descriptor and have their revealed counts set
5385 *** procedure: port->fdes PORT
5386 Returns the integer file descriptor underlying PORT. As a side
5387 effect the revealed count of PORT is incremented.
5389 *** procedure: fdes->ports FDES
5390 Returns a list of existing ports which have FDES as an underlying
5391 file descriptor, without changing their revealed counts.
5393 *** procedure: fdes->inport FDES
5394 Returns an existing input port which has FDES as its underlying
5395 file descriptor, if one exists, and increments its revealed count.
5396 Otherwise, returns a new input port with a revealed count of 1.
5398 *** procedure: fdes->outport FDES
5399 Returns an existing output port which has FDES as its underlying
5400 file descriptor, if one exists, and increments its revealed count.
5401 Otherwise, returns a new output port with a revealed count of 1.
5403 The next group of procedures perform a `dup2' system call, if NEWFD
5404 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5405 duplicated can be supplied as an integer or contained in a port. The
5406 type of value returned varies depending on which procedure is used.
5408 All procedures also have the side effect when performing `dup2' that
5409 any ports using NEWFD are moved to a different file descriptor and have
5410 their revealed counts set to zero.
5412 *** procedure: dup->fdes PORT/FD [NEWFD]
5413 Returns an integer file descriptor.
5415 *** procedure: dup->inport PORT/FD [NEWFD]
5416 Returns a new input port using the new file descriptor.
5418 *** procedure: dup->outport PORT/FD [NEWFD]
5419 Returns a new output port using the new file descriptor.
5421 *** procedure: dup PORT/FD [NEWFD]
5422 Returns a new port if PORT/FD is a port, with the same mode as the
5423 supplied port, otherwise returns an integer file descriptor.
5425 *** procedure: dup->port PORT/FD MODE [NEWFD]
5426 Returns a new port using the new file descriptor. MODE supplies a
5427 mode string for the port (*note open-file: File Ports.).
5429 *** procedure: setenv NAME VALUE
5430 Modifies the environment of the current process, which is also the
5431 default environment inherited by child processes.
5433 If VALUE is `#f', then NAME is removed from the environment.
5434 Otherwise, the string NAME=VALUE is added to the environment,
5435 replacing any existing string with name matching NAME.
5437 The return value is unspecified.
5439 *** procedure: truncate-file OBJ SIZE
5440 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5441 can be a string containing a file name or an integer file
5442 descriptor or port open for output on the file. The underlying
5443 system calls are `truncate' and `ftruncate'.
5445 The return value is unspecified.
5447 *** procedure: setvbuf PORT MODE [SIZE]
5448 Set the buffering mode for PORT. MODE can be:
5456 block buffered, using a newly allocated buffer of SIZE bytes.
5457 However if SIZE is zero or unspecified, the port will be made
5460 This procedure should not be used after I/O has been performed with
5463 Ports are usually block buffered by default, with a default buffer
5464 size. Procedures e.g., *Note open-file: File Ports, which accept a
5465 mode string allow `0' to be added to request an unbuffered port.
5467 *** procedure: fsync PORT/FD
5468 Copies any unwritten data for the specified output file descriptor
5469 to disk. If PORT/FD is a port, its buffer is flushed before the
5470 underlying file descriptor is fsync'd. The return value is
5473 *** procedure: open-fdes PATH FLAGS [MODES]
5474 Similar to `open' but returns a file descriptor instead of a port.
5476 *** procedure: execle PATH ENV [ARG] ...
5477 Similar to `execl', but the environment of the new process is
5478 specified by ENV, which must be a list of strings as returned by
5479 the `environ' procedure.
5481 This procedure is currently implemented using the `execve' system
5482 call, but we call it `execle' because of its Scheme calling
5485 *** procedure: strerror ERRNO
5486 Returns the Unix error message corresponding to ERRNO, an integer.
5488 *** procedure: primitive-exit [STATUS]
5489 Terminate the current process without unwinding the Scheme stack.
5490 This is would typically be useful after a fork. The exit status
5491 is STATUS if supplied, otherwise zero.
5493 *** procedure: times
5494 Returns an object with information about real and processor time.
5495 The following procedures accept such an object as an argument and
5496 return a selected component:
5499 The current real time, expressed as time units relative to an
5503 The CPU time units used by the calling process.
5506 The CPU time units used by the system on behalf of the
5510 The CPU time units used by terminated child processes of the
5511 calling process, whose status has been collected (e.g., using
5515 Similarly, the CPU times units used by the system on behalf of
5516 terminated child processes.
5518 ** Removed: list-length
5519 ** Removed: list-append, list-append!
5520 ** Removed: list-reverse, list-reverse!
5522 ** array-map renamed to array-map!
5524 ** serial-array-map renamed to serial-array-map!
5526 ** catch doesn't take #f as first argument any longer
5528 Previously, it was possible to pass #f instead of a key to `catch'.
5529 That would cause `catch' to pass a jump buffer object to the procedure
5530 passed as second argument. The procedure could then use this jump
5531 buffer objekt as an argument to throw.
5533 This mechanism has been removed since its utility doesn't motivate the
5534 extra complexity it introduces.
5536 ** The `#/' notation for lists now provokes a warning message from Guile.
5537 This syntax will be removed from Guile in the near future.
5539 To disable the warning message, set the GUILE_HUSH environment
5540 variable to any non-empty value.
5542 ** The newline character now prints as `#\newline', following the
5543 normal Scheme notation, not `#\nl'.
5545 * Changes to the gh_ interface
5547 ** The gh_enter function now takes care of loading the Guile startup files.
5548 gh_enter works by calling scm_boot_guile; see the remarks below.
5550 ** Function: void gh_write (SCM x)
5552 Write the printed representation of the scheme object x to the current
5553 output port. Corresponds to the scheme level `write'.
5555 ** gh_list_length renamed to gh_length.
5557 ** vector handling routines
5559 Several major changes. In particular, gh_vector() now resembles
5560 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5561 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5562 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5563 vector-related gh_ functions have been implemented.
5565 ** pair and list routines
5567 Implemented several of the R4RS pair and list functions that were
5570 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5572 New function. Converts double arrays back and forth between Scheme
5575 * Changes to the scm_ interface
5577 ** The function scm_boot_guile now takes care of loading the startup files.
5579 Guile's primary initialization function, scm_boot_guile, now takes
5580 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5581 Guile, define the module system, and put together some standard
5582 bindings. It also loads `init.scm', which is intended to hold
5583 site-specific initialization code.
5585 Since Guile cannot operate properly until boot-9.scm is loaded, there
5586 is no reason to separate loading boot-9.scm from Guile's other
5587 initialization processes.
5589 This job used to be done by scm_compile_shell_switches, which didn't
5590 make much sense; in particular, it meant that people using Guile for
5591 non-shell-like applications had to jump through hoops to get Guile
5592 initialized properly.
5594 ** The function scm_compile_shell_switches no longer loads the startup files.
5595 Now, Guile always loads the startup files, whenever it is initialized;
5596 see the notes above for scm_boot_guile and scm_load_startup_files.
5598 ** Function: scm_load_startup_files
5599 This new function takes care of loading Guile's initialization file
5600 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5601 this is always called by the Guile initialization process, it's
5602 probably not too useful to call this yourself, but it's there anyway.
5604 ** The semantics of smob marking have changed slightly.
5606 The smob marking function (the `mark' member of the scm_smobfuns
5607 structure) is no longer responsible for setting the mark bit on the
5608 smob. The generic smob handling code in the garbage collector will
5609 set this bit. The mark function need only ensure that any other
5610 objects the smob refers to get marked.
5612 Note that this change means that the smob's GC8MARK bit is typically
5613 already set upon entry to the mark function. Thus, marking functions
5614 which look like this:
5617 if (SCM_GC8MARKP (ptr))
5619 SCM_SETGC8MARK (ptr);
5620 ... mark objects to which the smob refers ...
5623 are now incorrect, since they will return early, and fail to mark any
5624 other objects the smob refers to. Some code in the Guile library used
5627 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5629 If you have implemented your own I/O port type, by writing the
5630 functions required by the scm_ptobfuns and then calling scm_newptob,
5631 you will need to change your functions slightly.
5633 The functions in a scm_ptobfuns structure now expect the port itself
5634 as their argument; they used to expect the `stream' member of the
5635 port's scm_port_table structure. This allows functions in an
5636 scm_ptobfuns structure to easily access the port's cell (and any flags
5637 it its CAR), and the port's scm_port_table structure.
5639 Guile now passes the I/O port itself as the `port' argument in the
5640 following scm_ptobfuns functions:
5642 int (*free) (SCM port);
5643 int (*fputc) (int, SCM port);
5644 int (*fputs) (char *, SCM port);
5645 scm_sizet (*fwrite) SCM_P ((char *ptr,
5649 int (*fflush) (SCM port);
5650 int (*fgetc) (SCM port);
5651 int (*fclose) (SCM port);
5653 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5656 If you have existing code which defines its own port types, it is easy
5657 to convert your code to the new interface; simply apply SCM_STREAM to
5658 the port argument to yield the value you code used to expect.
5660 Note that since both the port and the stream have the same type in the
5661 C code --- they are both SCM values --- the C compiler will not remind
5662 you if you forget to update your scm_ptobfuns functions.
5665 ** Function: int scm_internal_select (int fds,
5669 struct timeval *timeout);
5671 This is a replacement for the `select' function provided by the OS.
5672 It enables I/O blocking and sleeping to happen for one cooperative
5673 thread without blocking other threads. It also avoids busy-loops in
5674 these situations. It is intended that all I/O blocking and sleeping
5675 will finally go through this function. Currently, this function is
5676 only available on systems providing `gettimeofday' and `select'.
5678 ** Function: SCM scm_internal_stack_catch (SCM tag,
5679 scm_catch_body_t body,
5681 scm_catch_handler_t handler,
5684 A new sibling to the other two C level `catch' functions
5685 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5686 the stack to be saved automatically into the variable `the-last-stack'
5687 (scm_the_last_stack_var) on error. This is necessary if you want to
5688 use advanced error reporting, such as calling scm_display_error and
5689 scm_display_backtrace. (They both take a stack object as argument.)
5691 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5693 scm_catch_handler_t handler,
5696 Spawns a new thread. It does a job similar to
5697 scm_call_with_new_thread but takes arguments more suitable when
5698 spawning threads from application C code.
5700 ** The hook scm_error_callback has been removed. It was originally
5701 intended as a way for the user to install his own error handler. But
5702 that method works badly since it intervenes between throw and catch,
5703 thereby changing the semantics of expressions like (catch #t ...).
5704 The correct way to do it is to use one of the C level catch functions
5705 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5707 ** Removed functions:
5709 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5710 scm_list_reverse, scm_list_reverse_x
5712 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5714 These can be used for pretty list creation from C. The idea is taken
5715 from Erick Gallesio's STk.
5717 ** scm_array_map renamed to scm_array_map_x
5719 ** mbstrings are now removed
5721 This means that the type codes scm_tc7_mb_string and
5722 scm_tc7_mb_substring has been removed.
5724 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5726 Since we no longer support multi-byte strings, these I/O functions
5727 have been simplified, and renamed. Here are their old names, and
5728 their new names and arguments:
5730 scm_gen_putc -> void scm_putc (int c, SCM port);
5731 scm_gen_puts -> void scm_puts (char *s, SCM port);
5732 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5733 scm_gen_getc -> void scm_getc (SCM port);
5736 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5738 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5740 SCM_TYP7S now masks away the bit which distinguishes substrings from
5743 ** scm_catch_body_t: Backward incompatible change!
5745 Body functions to scm_internal_catch and friends do not any longer
5746 take a second argument. This is because it is no longer possible to
5747 pass a #f arg to catch.
5749 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5751 The function scm_protect_object protects its argument from being freed
5752 by the garbage collector. scm_unprotect_object removes that
5755 These functions now nest properly. That is, for every object O, there
5756 is a counter which scm_protect_object(O) increments and
5757 scm_unprotect_object(O) decrements, if the counter is greater than
5758 zero. Every object's counter is zero when it is first created. If an
5759 object's counter is greater than zero, the garbage collector will not
5760 reclaim its storage.
5762 This allows you to use scm_protect_object in your code without
5763 worrying that some other function you call will call
5764 scm_unprotect_object, and allow it to be freed. Assuming that the
5765 functions you call are well-behaved, and unprotect only those objects
5766 they protect, you can follow the same rule and have confidence that
5767 objects will be freed only at appropriate times.
5770 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5772 * Changes to the distribution
5774 ** Nightly snapshots are now available from ftp.red-bean.com.
5775 The old server, ftp.cyclic.com, has been relinquished to its rightful
5778 Nightly snapshots of the Guile development sources are now available via
5779 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5781 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5782 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5784 ** To run Guile without installing it, the procedure has changed a bit.
5786 If you used a separate build directory to compile Guile, you'll need
5787 to include the build directory in SCHEME_LOAD_PATH, as well as the
5788 source directory. See the `INSTALL' file for examples.
5790 * Changes to the procedure for linking libguile with your programs
5792 ** The standard Guile load path for Scheme code now includes
5793 $(datadir)/guile (usually /usr/local/share/guile). This means that
5794 you can install your own Scheme files there, and Guile will find them.
5795 (Previous versions of Guile only checked a directory whose name
5796 contained the Guile version number, so you had to re-install or move
5797 your Scheme sources each time you installed a fresh version of Guile.)
5799 The load path also includes $(datadir)/guile/site; we recommend
5800 putting individual Scheme files there. If you want to install a
5801 package with multiple source files, create a directory for them under
5804 ** Guile 1.2 will now use the Rx regular expression library, if it is
5805 installed on your system. When you are linking libguile into your own
5806 programs, this means you will have to link against -lguile, -lqt (if
5807 you configured Guile with thread support), and -lrx.
5809 If you are using autoconf to generate configuration scripts for your
5810 application, the following lines should suffice to add the appropriate
5811 libraries to your link command:
5813 ### Find Rx, quickthreads and libguile.
5814 AC_CHECK_LIB(rx, main)
5815 AC_CHECK_LIB(qt, main)
5816 AC_CHECK_LIB(guile, scm_shell)
5818 The Guile 1.2 distribution does not contain sources for the Rx
5819 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5820 retrieve it from a GNU FTP site and install it separately.
5822 * Changes to Scheme functions and syntax
5824 ** The dynamic linking features of Guile are now enabled by default.
5825 You can disable them by giving the `--disable-dynamic-linking' option
5828 (dynamic-link FILENAME)
5830 Find the object file denoted by FILENAME (a string) and link it
5831 into the running Guile application. When everything works out,
5832 return a Scheme object suitable for representing the linked object
5833 file. Otherwise an error is thrown. How object files are
5834 searched is system dependent.
5836 (dynamic-object? VAL)
5838 Determine whether VAL represents a dynamically linked object file.
5840 (dynamic-unlink DYNOBJ)
5842 Unlink the indicated object file from the application. DYNOBJ
5843 should be one of the values returned by `dynamic-link'.
5845 (dynamic-func FUNCTION DYNOBJ)
5847 Search the C function indicated by FUNCTION (a string or symbol)
5848 in DYNOBJ and return some Scheme object that can later be used
5849 with `dynamic-call' to actually call this function. Right now,
5850 these Scheme objects are formed by casting the address of the
5851 function to `long' and converting this number to its Scheme
5854 (dynamic-call FUNCTION DYNOBJ)
5856 Call the C function indicated by FUNCTION and DYNOBJ. The
5857 function is passed no arguments and its return value is ignored.
5858 When FUNCTION is something returned by `dynamic-func', call that
5859 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5860 etc.), look it up in DYNOBJ; this is equivalent to
5862 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5864 Interrupts are deferred while the C function is executing (with
5865 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5867 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5869 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5870 some arguments and return its return value. The C function is
5871 expected to take two arguments and return an `int', just like
5874 int c_func (int argc, char **argv);
5876 ARGS must be a list of strings and is converted into an array of
5877 `char *'. The array is passed in ARGV and its size in ARGC. The
5878 return value is converted to a Scheme number and returned from the
5879 call to `dynamic-args-call'.
5881 When dynamic linking is disabled or not supported on your system,
5882 the above functions throw errors, but they are still available.
5884 Here is a small example that works on GNU/Linux:
5886 (define libc-obj (dynamic-link "libc.so"))
5887 (dynamic-args-call 'rand libc-obj '())
5889 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5891 ** The #/ syntax for module names is depreciated, and will be removed
5892 in a future version of Guile. Instead of
5900 The latter syntax is more consistent with existing Lisp practice.
5902 ** Guile now does fancier printing of structures. Structures are the
5903 underlying implementation for records, which in turn are used to
5904 implement modules, so all of these object now print differently and in
5905 a more informative way.
5907 The Scheme printer will examine the builtin variable *struct-printer*
5908 whenever it needs to print a structure object. When this variable is
5909 not `#f' it is deemed to be a procedure and will be applied to the
5910 structure object and the output port. When *struct-printer* is `#f'
5911 or the procedure return `#f' the structure object will be printed in
5912 the boring #<struct 80458270> form.
5914 This hook is used by some routines in ice-9/boot-9.scm to implement
5915 type specific printing routines. Please read the comments there about
5918 One of the more specific uses of structs are records. The printing
5919 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5920 called. It should behave like a *struct-printer* procedure (described
5923 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5924 token of the form #:NAME, where NAME has the same syntax as a Scheme
5925 symbol, is the external representation of the keyword named NAME.
5926 Keyword objects print using this syntax as well, so values containing
5927 keyword objects can be read back into Guile. When used in an
5928 expression, keywords are self-quoting objects.
5930 Guile suports this read syntax, and uses this print syntax, regardless
5931 of the current setting of the `keyword' read option. The `keyword'
5932 read option only controls whether Guile recognizes the `:NAME' syntax,
5933 which is incompatible with R4RS. (R4RS says such token represent
5936 ** Guile has regular expression support again. Guile 1.0 included
5937 functions for matching regular expressions, based on the Rx library.
5938 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5939 distribution, and thus Guile had no regular expression support. Guile
5940 1.2 again supports the most commonly used functions, and supports all
5941 of SCSH's regular expression functions.
5943 If your system does not include a POSIX regular expression library,
5944 and you have not linked Guile with a third-party regexp library such as
5945 Rx, these functions will not be available. You can tell whether your
5946 Guile installation includes regular expression support by checking
5947 whether the `*features*' list includes the `regex' symbol.
5949 *** regexp functions
5951 By default, Guile supports POSIX extended regular expressions. That
5952 means that the characters `(', `)', `+' and `?' are special, and must
5953 be escaped if you wish to match the literal characters.
5955 This regular expression interface was modeled after that implemented
5956 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5957 with SCSH regular expressions.
5959 **** Function: string-match PATTERN STR [START]
5960 Compile the string PATTERN into a regular expression and compare
5961 it with STR. The optional numeric argument START specifies the
5962 position of STR at which to begin matching.
5964 `string-match' returns a "match structure" which describes what,
5965 if anything, was matched by the regular expression. *Note Match
5966 Structures::. If STR does not match PATTERN at all,
5967 `string-match' returns `#f'.
5969 Each time `string-match' is called, it must compile its PATTERN
5970 argument into a regular expression structure. This operation is
5971 expensive, which makes `string-match' inefficient if the same regular
5972 expression is used several times (for example, in a loop). For better
5973 performance, you can compile a regular expression in advance and then
5974 match strings against the compiled regexp.
5976 **** Function: make-regexp STR [FLAGS]
5977 Compile the regular expression described by STR, and return the
5978 compiled regexp structure. If STR does not describe a legal
5979 regular expression, `make-regexp' throws a
5980 `regular-expression-syntax' error.
5982 FLAGS may be the bitwise-or of one or more of the following:
5984 **** Constant: regexp/extended
5985 Use POSIX Extended Regular Expression syntax when interpreting
5986 STR. If not set, POSIX Basic Regular Expression syntax is used.
5987 If the FLAGS argument is omitted, we assume regexp/extended.
5989 **** Constant: regexp/icase
5990 Do not differentiate case. Subsequent searches using the
5991 returned regular expression will be case insensitive.
5993 **** Constant: regexp/newline
5994 Match-any-character operators don't match a newline.
5996 A non-matching list ([^...]) not containing a newline matches a
5999 Match-beginning-of-line operator (^) matches the empty string
6000 immediately after a newline, regardless of whether the FLAGS
6001 passed to regexp-exec contain regexp/notbol.
6003 Match-end-of-line operator ($) matches the empty string
6004 immediately before a newline, regardless of whether the FLAGS
6005 passed to regexp-exec contain regexp/noteol.
6007 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6008 Match the compiled regular expression REGEXP against `str'. If
6009 the optional integer START argument is provided, begin matching
6010 from that position in the string. Return a match structure
6011 describing the results of the match, or `#f' if no match could be
6014 FLAGS may be the bitwise-or of one or more of the following:
6016 **** Constant: regexp/notbol
6017 The match-beginning-of-line operator always fails to match (but
6018 see the compilation flag regexp/newline above) This flag may be
6019 used when different portions of a string are passed to
6020 regexp-exec and the beginning of the string should not be
6021 interpreted as the beginning of the line.
6023 **** Constant: regexp/noteol
6024 The match-end-of-line operator always fails to match (but see the
6025 compilation flag regexp/newline above)
6027 **** Function: regexp? OBJ
6028 Return `#t' if OBJ is a compiled regular expression, or `#f'
6031 Regular expressions are commonly used to find patterns in one string
6032 and replace them with the contents of another string.
6034 **** Function: regexp-substitute PORT MATCH [ITEM...]
6035 Write to the output port PORT selected contents of the match
6036 structure MATCH. Each ITEM specifies what should be written, and
6037 may be one of the following arguments:
6039 * A string. String arguments are written out verbatim.
6041 * An integer. The submatch with that number is written.
6043 * The symbol `pre'. The portion of the matched string preceding
6044 the regexp match is written.
6046 * The symbol `post'. The portion of the matched string
6047 following the regexp match is written.
6049 PORT may be `#f', in which case nothing is written; instead,
6050 `regexp-substitute' constructs a string from the specified ITEMs
6053 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6054 Similar to `regexp-substitute', but can be used to perform global
6055 substitutions on STR. Instead of taking a match structure as an
6056 argument, `regexp-substitute/global' takes two string arguments: a
6057 REGEXP string describing a regular expression, and a TARGET string
6058 which should be matched against this regular expression.
6060 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6063 * A function may be supplied. When this function is called, it
6064 will be passed one argument: a match structure for a given
6065 regular expression match. It should return a string to be
6066 written out to PORT.
6068 * The `post' symbol causes `regexp-substitute/global' to recurse
6069 on the unmatched portion of STR. This *must* be supplied in
6070 order to perform global search-and-replace on STR; if it is
6071 not present among the ITEMs, then `regexp-substitute/global'
6072 will return after processing a single match.
6074 *** Match Structures
6076 A "match structure" is the object returned by `string-match' and
6077 `regexp-exec'. It describes which portion of a string, if any, matched
6078 the given regular expression. Match structures include: a reference to
6079 the string that was checked for matches; the starting and ending
6080 positions of the regexp match; and, if the regexp included any
6081 parenthesized subexpressions, the starting and ending positions of each
6084 In each of the regexp match functions described below, the `match'
6085 argument must be a match structure returned by a previous call to
6086 `string-match' or `regexp-exec'. Most of these functions return some
6087 information about the original target string that was matched against a
6088 regular expression; we will call that string TARGET for easy reference.
6090 **** Function: regexp-match? OBJ
6091 Return `#t' if OBJ is a match structure returned by a previous
6092 call to `regexp-exec', or `#f' otherwise.
6094 **** Function: match:substring MATCH [N]
6095 Return the portion of TARGET matched by subexpression number N.
6096 Submatch 0 (the default) represents the entire regexp match. If
6097 the regular expression as a whole matched, but the subexpression
6098 number N did not match, return `#f'.
6100 **** Function: match:start MATCH [N]
6101 Return the starting position of submatch number N.
6103 **** Function: match:end MATCH [N]
6104 Return the ending position of submatch number N.
6106 **** Function: match:prefix MATCH
6107 Return the unmatched portion of TARGET preceding the regexp match.
6109 **** Function: match:suffix MATCH
6110 Return the unmatched portion of TARGET following the regexp match.
6112 **** Function: match:count MATCH
6113 Return the number of parenthesized subexpressions from MATCH.
6114 Note that the entire regular expression match itself counts as a
6115 subexpression, and failed submatches are included in the count.
6117 **** Function: match:string MATCH
6118 Return the original TARGET string.
6120 *** Backslash Escapes
6122 Sometimes you will want a regexp to match characters like `*' or `$'
6123 exactly. For example, to check whether a particular string represents
6124 a menu entry from an Info node, it would be useful to match it against
6125 a regexp like `^* [^:]*::'. However, this won't work; because the
6126 asterisk is a metacharacter, it won't match the `*' at the beginning of
6127 the string. In this case, we want to make the first asterisk un-magic.
6129 You can do this by preceding the metacharacter with a backslash
6130 character `\'. (This is also called "quoting" the metacharacter, and
6131 is known as a "backslash escape".) When Guile sees a backslash in a
6132 regular expression, it considers the following glyph to be an ordinary
6133 character, no matter what special meaning it would ordinarily have.
6134 Therefore, we can make the above example work by changing the regexp to
6135 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6136 to match only a single asterisk in the target string.
6138 Since the backslash is itself a metacharacter, you may force a
6139 regexp to match a backslash in the target string by preceding the
6140 backslash with itself. For example, to find variable references in a
6141 TeX program, you might want to find occurrences of the string `\let\'
6142 followed by any number of alphabetic characters. The regular expression
6143 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6144 each match a single backslash in the target string.
6146 **** Function: regexp-quote STR
6147 Quote each special character found in STR with a backslash, and
6148 return the resulting string.
6150 *Very important:* Using backslash escapes in Guile source code (as
6151 in Emacs Lisp or C) can be tricky, because the backslash character has
6152 special meaning for the Guile reader. For example, if Guile encounters
6153 the character sequence `\n' in the middle of a string while processing
6154 Scheme code, it replaces those characters with a newline character.
6155 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6156 Several of these "escape sequences" are processed by the Guile reader
6157 before your code is executed. Unrecognized escape sequences are
6158 ignored: if the characters `\*' appear in a string, they will be
6159 translated to the single character `*'.
6161 This translation is obviously undesirable for regular expressions,
6162 since we want to be able to include backslashes in a string in order to
6163 escape regexp metacharacters. Therefore, to make sure that a backslash
6164 is preserved in a string in your Guile program, you must use *two*
6165 consecutive backslashes:
6167 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6169 The string in this example is preprocessed by the Guile reader before
6170 any code is executed. The resulting argument to `make-regexp' is the
6171 string `^\* [^:]*', which is what we really want.
6173 This also means that in order to write a regular expression that
6174 matches a single backslash character, the regular expression string in
6175 the source code must include *four* backslashes. Each consecutive pair
6176 of backslashes gets translated by the Guile reader to a single
6177 backslash, and the resulting double-backslash is interpreted by the
6178 regexp engine as matching a single backslash character. Hence:
6180 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6182 The reason for the unwieldiness of this syntax is historical. Both
6183 regular expression pattern matchers and Unix string processing systems
6184 have traditionally used backslashes with the special meanings described
6185 above. The POSIX regular expression specification and ANSI C standard
6186 both require these semantics. Attempting to abandon either convention
6187 would cause other kinds of compatibility problems, possibly more severe
6188 ones. Therefore, without extending the Scheme reader to support
6189 strings with different quoting conventions (an ungainly and confusing
6190 extension when implemented in other languages), we must adhere to this
6191 cumbersome escape syntax.
6193 * Changes to the gh_ interface
6195 * Changes to the scm_ interface
6197 * Changes to system call interfaces:
6199 ** The value returned by `raise' is now unspecified. It throws an exception
6202 *** A new procedure `sigaction' can be used to install signal handlers
6204 (sigaction signum [action] [flags])
6206 signum is the signal number, which can be specified using the value
6209 If action is omitted, sigaction returns a pair: the CAR is the current
6210 signal hander, which will be either an integer with the value SIG_DFL
6211 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6212 handles the signal, or #f if a non-Scheme procedure handles the
6213 signal. The CDR contains the current sigaction flags for the handler.
6215 If action is provided, it is installed as the new handler for signum.
6216 action can be a Scheme procedure taking one argument, or the value of
6217 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6218 whatever signal handler was installed before sigaction was first used.
6219 Flags can optionally be specified for the new handler (SA_RESTART is
6220 always used if the system provides it, so need not be specified.) The
6221 return value is a pair with information about the old handler as
6224 This interface does not provide access to the "signal blocking"
6225 facility. Maybe this is not needed, since the thread support may
6226 provide solutions to the problem of consistent access to data
6229 *** A new procedure `flush-all-ports' is equivalent to running
6230 `force-output' on every port open for output.
6232 ** Guile now provides information on how it was built, via the new
6233 global variable, %guile-build-info. This variable records the values
6234 of the standard GNU makefile directory variables as an assocation
6235 list, mapping variable names (symbols) onto directory paths (strings).
6236 For example, to find out where the Guile link libraries were
6237 installed, you can say:
6239 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6242 * Changes to the scm_ interface
6244 ** The new function scm_handle_by_message_noexit is just like the
6245 existing scm_handle_by_message function, except that it doesn't call
6246 exit to terminate the process. Instead, it prints a message and just
6247 returns #f. This might be a more appropriate catch-all handler for
6248 new dynamic roots and threads.
6251 Changes in Guile 1.1 (released Friday, May 16 1997):
6253 * Changes to the distribution.
6255 The Guile 1.0 distribution has been split up into several smaller
6257 guile-core --- the Guile interpreter itself.
6258 guile-tcltk --- the interface between the Guile interpreter and
6259 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6260 is a toolkit for building graphical user interfaces.
6261 guile-rgx-ctax --- the interface between Guile and the Rx regular
6262 expression matcher, and the translator for the Ctax
6263 programming language. These are packaged together because the
6264 Ctax translator uses Rx to parse Ctax source code.
6266 This NEWS file describes the changes made to guile-core since the 1.0
6269 We no longer distribute the documentation, since it was either out of
6270 date, or incomplete. As soon as we have current documentation, we
6275 * Changes to the stand-alone interpreter
6277 ** guile now accepts command-line arguments compatible with SCSH, Olin
6278 Shivers' Scheme Shell.
6280 In general, arguments are evaluated from left to right, but there are
6281 exceptions. The following switches stop argument processing, and
6282 stash all remaining command-line arguments as the value returned by
6283 the (command-line) function.
6284 -s SCRIPT load Scheme source code from FILE, and exit
6285 -c EXPR evalute Scheme expression EXPR, and exit
6286 -- stop scanning arguments; run interactively
6288 The switches below are processed as they are encountered.
6289 -l FILE load Scheme source code from FILE
6290 -e FUNCTION after reading script, apply FUNCTION to
6291 command line arguments
6292 -ds do -s script at this point
6293 --emacs enable Emacs protocol (experimental)
6294 -h, --help display this help and exit
6295 -v, --version display version information and exit
6296 \ read arguments from following script lines
6298 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6299 which re-implements the traditional "echo" command:
6301 #!/usr/local/bin/guile -s
6304 (map (lambda (arg) (display arg) (display " "))
6308 (main (command-line))
6310 Suppose we invoke this script as follows:
6312 ekko a speckled gecko
6314 Through the magic of Unix script processing (triggered by the `#!'
6315 token at the top of the file), /usr/local/bin/guile receives the
6316 following list of command-line arguments:
6318 ("-s" "./ekko" "a" "speckled" "gecko")
6320 Unix inserts the name of the script after the argument specified on
6321 the first line of the file (in this case, "-s"), and then follows that
6322 with the arguments given to the script. Guile loads the script, which
6323 defines the `main' function, and then applies it to the list of
6324 remaining command-line arguments, ("a" "speckled" "gecko").
6326 In Unix, the first line of a script file must take the following form:
6328 #!INTERPRETER ARGUMENT
6330 where INTERPRETER is the absolute filename of the interpreter
6331 executable, and ARGUMENT is a single command-line argument to pass to
6334 You may only pass one argument to the interpreter, and its length is
6335 limited. These restrictions can be annoying to work around, so Guile
6336 provides a general mechanism (borrowed from, and compatible with,
6337 SCSH) for circumventing them.
6339 If the ARGUMENT in a Guile script is a single backslash character,
6340 `\', Guile will open the script file, parse arguments from its second
6341 and subsequent lines, and replace the `\' with them. So, for example,
6342 here is another implementation of the `ekko' script:
6344 #!/usr/local/bin/guile \
6348 (for-each (lambda (arg) (display arg) (display " "))
6352 If the user invokes this script as follows:
6354 ekko a speckled gecko
6356 Unix expands this into
6358 /usr/local/bin/guile \ ekko a speckled gecko
6360 When Guile sees the `\' argument, it replaces it with the arguments
6361 read from the second line of the script, producing:
6363 /usr/local/bin/guile -e main -s ekko a speckled gecko
6365 This tells Guile to load the `ekko' script, and apply the function
6366 `main' to the argument list ("a" "speckled" "gecko").
6368 Here is how Guile parses the command-line arguments:
6369 - Each space character terminates an argument. This means that two
6370 spaces in a row introduce an empty-string argument.
6371 - The tab character is not permitted (unless you quote it with the
6372 backslash character, as described below), to avoid confusion.
6373 - The newline character terminates the sequence of arguments, and will
6374 also terminate a final non-empty argument. (However, a newline
6375 following a space will not introduce a final empty-string argument;
6376 it only terminates the argument list.)
6377 - The backslash character is the escape character. It escapes
6378 backslash, space, tab, and newline. The ANSI C escape sequences
6379 like \n and \t are also supported. These produce argument
6380 constituents; the two-character combination \n doesn't act like a
6381 terminating newline. The escape sequence \NNN for exactly three
6382 octal digits reads as the character whose ASCII code is NNN. As
6383 above, characters produced this way are argument constituents.
6384 Backslash followed by other characters is not allowed.
6386 * Changes to the procedure for linking libguile with your programs
6388 ** Guile now builds and installs a shared guile library, if your
6389 system support shared libraries. (It still builds a static library on
6390 all systems.) Guile automatically detects whether your system
6391 supports shared libraries. To prevent Guile from buildisg shared
6392 libraries, pass the `--disable-shared' flag to the configure script.
6394 Guile takes longer to compile when it builds shared libraries, because
6395 it must compile every file twice --- once to produce position-
6396 independent object code, and once to produce normal object code.
6398 ** The libthreads library has been merged into libguile.
6400 To link a program against Guile, you now need only link against
6401 -lguile and -lqt; -lthreads is no longer needed. If you are using
6402 autoconf to generate configuration scripts for your application, the
6403 following lines should suffice to add the appropriate libraries to
6406 ### Find quickthreads and libguile.
6407 AC_CHECK_LIB(qt, main)
6408 AC_CHECK_LIB(guile, scm_shell)
6410 * Changes to Scheme functions
6412 ** Guile Scheme's special syntax for keyword objects is now optional,
6413 and disabled by default.
6415 The syntax variation from R4RS made it difficult to port some
6416 interesting packages to Guile. The routines which accepted keyword
6417 arguments (mostly in the module system) have been modified to also
6418 accept symbols whose names begin with `:'.
6420 To change the keyword syntax, you must first import the (ice-9 debug)
6422 (use-modules (ice-9 debug))
6424 Then you can enable the keyword syntax as follows:
6425 (read-set! keywords 'prefix)
6427 To disable keyword syntax, do this:
6428 (read-set! keywords #f)
6430 ** Many more primitive functions accept shared substrings as
6431 arguments. In the past, these functions required normal, mutable
6432 strings as arguments, although they never made use of this
6435 ** The uniform array functions now operate on byte vectors. These
6436 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6437 `serial-array-map', `array-map', `array-for-each', and
6440 ** The new functions `trace' and `untrace' implement simple debugging
6441 support for Scheme functions.
6443 The `trace' function accepts any number of procedures as arguments,
6444 and tells the Guile interpreter to display each procedure's name and
6445 arguments each time the procedure is invoked. When invoked with no
6446 arguments, `trace' returns the list of procedures currently being
6449 The `untrace' function accepts any number of procedures as arguments,
6450 and tells the Guile interpreter not to trace them any more. When
6451 invoked with no arguments, `untrace' untraces all curretly traced
6454 The tracing in Guile has an advantage over most other systems: we
6455 don't create new procedure objects, but mark the procedure objects
6456 themselves. This means that anonymous and internal procedures can be
6459 ** The function `assert-repl-prompt' has been renamed to
6460 `set-repl-prompt!'. It takes one argument, PROMPT.
6461 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6462 - If PROMPT is a string, we use it as a prompt.
6463 - If PROMPT is a procedure accepting no arguments, we call it, and
6464 display the result as a prompt.
6465 - Otherwise, we display "> ".
6467 ** The new function `eval-string' reads Scheme expressions from a
6468 string and evaluates them, returning the value of the last expression
6469 in the string. If the string contains no expressions, it returns an
6472 ** The new function `thunk?' returns true iff its argument is a
6473 procedure of zero arguments.
6475 ** `defined?' is now a builtin function, instead of syntax. This
6476 means that its argument should be quoted. It returns #t iff its
6477 argument is bound in the current module.
6479 ** The new syntax `use-modules' allows you to add new modules to your
6480 environment without re-typing a complete `define-module' form. It
6481 accepts any number of module names as arguments, and imports their
6482 public bindings into the current module.
6484 ** The new function (module-defined? NAME MODULE) returns true iff
6485 NAME, a symbol, is defined in MODULE, a module object.
6487 ** The new function `builtin-bindings' creates and returns a hash
6488 table containing copies of all the root module's bindings.
6490 ** The new function `builtin-weak-bindings' does the same as
6491 `builtin-bindings', but creates a doubly-weak hash table.
6493 ** The `equal?' function now considers variable objects to be
6494 equivalent if they have the same name and the same value.
6496 ** The new function `command-line' returns the command-line arguments
6497 given to Guile, as a list of strings.
6499 When using guile as a script interpreter, `command-line' returns the
6500 script's arguments; those processed by the interpreter (like `-s' or
6501 `-c') are omitted. (In other words, you get the normal, expected
6502 behavior.) Any application that uses scm_shell to process its
6503 command-line arguments gets this behavior as well.
6505 ** The new function `load-user-init' looks for a file called `.guile'
6506 in the user's home directory, and loads it if it exists. This is
6507 mostly for use by the code generated by scm_compile_shell_switches,
6508 but we thought it might also be useful in other circumstances.
6510 ** The new function `log10' returns the base-10 logarithm of its
6513 ** Changes to I/O functions
6515 *** The functions `read', `primitive-load', `read-and-eval!', and
6516 `primitive-load-path' no longer take optional arguments controlling
6517 case insensitivity and a `#' parser.
6519 Case sensitivity is now controlled by a read option called
6520 `case-insensitive'. The user can add new `#' syntaxes with the
6521 `read-hash-extend' function (see below).
6523 *** The new function `read-hash-extend' allows the user to change the
6524 syntax of Guile Scheme in a somewhat controlled way.
6526 (read-hash-extend CHAR PROC)
6527 When parsing S-expressions, if we read a `#' character followed by
6528 the character CHAR, use PROC to parse an object from the stream.
6529 If PROC is #f, remove any parsing procedure registered for CHAR.
6531 The reader applies PROC to two arguments: CHAR and an input port.
6533 *** The new functions read-delimited and read-delimited! provide a
6534 general mechanism for doing delimited input on streams.
6536 (read-delimited DELIMS [PORT HANDLE-DELIM])
6537 Read until we encounter one of the characters in DELIMS (a string),
6538 or end-of-file. PORT is the input port to read from; it defaults to
6539 the current input port. The HANDLE-DELIM parameter determines how
6540 the terminating character is handled; it should be one of the
6543 'trim omit delimiter from result
6544 'peek leave delimiter character in input stream
6545 'concat append delimiter character to returned value
6546 'split return a pair: (RESULT . TERMINATOR)
6548 HANDLE-DELIM defaults to 'peek.
6550 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6551 A side-effecting variant of `read-delimited'.
6553 The data is written into the string BUF at the indices in the
6554 half-open interval [START, END); the default interval is the whole
6555 string: START = 0 and END = (string-length BUF). The values of
6556 START and END must specify a well-defined interval in BUF, i.e.
6557 0 <= START <= END <= (string-length BUF).
6559 It returns NBYTES, the number of bytes read. If the buffer filled
6560 up without a delimiter character being found, it returns #f. If the
6561 port is at EOF when the read starts, it returns the EOF object.
6563 If an integer is returned (i.e., the read is successfully terminated
6564 by reading a delimiter character), then the HANDLE-DELIM parameter
6565 determines how to handle the terminating character. It is described
6566 above, and defaults to 'peek.
6568 (The descriptions of these functions were borrowed from the SCSH
6569 manual, by Olin Shivers and Brian Carlstrom.)
6571 *** The `%read-delimited!' function is the primitive used to implement
6572 `read-delimited' and `read-delimited!'.
6574 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6576 This returns a pair of values: (TERMINATOR . NUM-READ).
6577 - TERMINATOR describes why the read was terminated. If it is a
6578 character or the eof object, then that is the value that terminated
6579 the read. If it is #f, the function filled the buffer without finding
6580 a delimiting character.
6581 - NUM-READ is the number of characters read into BUF.
6583 If the read is successfully terminated by reading a delimiter
6584 character, then the gobble? parameter determines what to do with the
6585 terminating character. If true, the character is removed from the
6586 input stream; if false, the character is left in the input stream
6587 where a subsequent read operation will retrieve it. In either case,
6588 the character is also the first value returned by the procedure call.
6590 (The descriptions of this function was borrowed from the SCSH manual,
6591 by Olin Shivers and Brian Carlstrom.)
6593 *** The `read-line' and `read-line!' functions have changed; they now
6594 trim the terminator by default; previously they appended it to the
6595 returned string. For the old behavior, use (read-line PORT 'concat).
6597 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6598 take new optional START and END arguments, specifying the region of
6599 the array to read and write.
6601 *** The `ungetc-char-ready?' function has been removed. We feel it's
6602 inappropriate for an interface to expose implementation details this
6605 ** Changes to the Unix library and system call interface
6607 *** The new fcntl function provides access to the Unix `fcntl' system
6610 (fcntl PORT COMMAND VALUE)
6611 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6612 Values for COMMAND are:
6614 F_DUPFD duplicate a file descriptor
6615 F_GETFD read the descriptor's close-on-exec flag
6616 F_SETFD set the descriptor's close-on-exec flag to VALUE
6617 F_GETFL read the descriptor's flags, as set on open
6618 F_SETFL set the descriptor's flags, as set on open to VALUE
6619 F_GETOWN return the process ID of a socket's owner, for SIGIO
6620 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6621 FD_CLOEXEC not sure what this is
6623 For details, see the documentation for the fcntl system call.
6625 *** The arguments to `select' have changed, for compatibility with
6626 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6627 expected behavior. The MILLISECONDS parameter has been changed to
6628 MICROSECONDS, to more closely resemble the underlying system call.
6629 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6630 corresponding return set will be the same.
6632 *** The arguments to the `mknod' system call have changed. They are
6635 (mknod PATH TYPE PERMS DEV)
6636 Create a new file (`node') in the file system. PATH is the name of
6637 the file to create. TYPE is the kind of file to create; it should
6638 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6639 permission bits to give the newly created file. If TYPE is
6640 'block-special or 'char-special, DEV specifies which device the
6641 special file refers to; its interpretation depends on the kind of
6642 special file being created.
6644 *** The `fork' function has been renamed to `primitive-fork', to avoid
6645 clashing with various SCSH forks.
6647 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6648 and `recvfrom!'. They no longer accept a size for a second argument;
6649 you must pass a string to hold the received value. They no longer
6650 return the buffer. Instead, `recv' returns the length of the message
6651 received, and `recvfrom' returns a pair containing the packet's length
6652 and originating address.
6654 *** The file descriptor datatype has been removed, as have the
6655 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6656 We plan to replace these functions with a SCSH-compatible interface.
6658 *** The `create' function has been removed; it's just a special case
6661 *** There are new functions to break down process termination status
6662 values. In the descriptions below, STATUS is a value returned by
6665 (status:exit-val STATUS)
6666 If the child process exited normally, this function returns the exit
6667 code for the child process (i.e., the value passed to exit, or
6668 returned from main). If the child process did not exit normally,
6669 this function returns #f.
6671 (status:stop-sig STATUS)
6672 If the child process was suspended by a signal, this function
6673 returns the signal that suspended the child. Otherwise, it returns
6676 (status:term-sig STATUS)
6677 If the child process terminated abnormally, this function returns
6678 the signal that terminated the child. Otherwise, this function
6681 POSIX promises that exactly one of these functions will return true on
6682 a valid STATUS value.
6684 These functions are compatible with SCSH.
6686 *** There are new accessors and setters for the broken-out time vectors
6687 returned by `localtime', `gmtime', and that ilk. They are:
6689 Component Accessor Setter
6690 ========================= ============ ============
6691 seconds tm:sec set-tm:sec
6692 minutes tm:min set-tm:min
6693 hours tm:hour set-tm:hour
6694 day of the month tm:mday set-tm:mday
6695 month tm:mon set-tm:mon
6696 year tm:year set-tm:year
6697 day of the week tm:wday set-tm:wday
6698 day in the year tm:yday set-tm:yday
6699 daylight saving time tm:isdst set-tm:isdst
6700 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6701 name of time zone tm:zone set-tm:zone
6703 *** There are new accessors for the vectors returned by `uname',
6704 describing the host system:
6707 ============================================== ================
6708 name of the operating system implementation utsname:sysname
6709 network name of this machine utsname:nodename
6710 release level of the operating system utsname:release
6711 version level of the operating system utsname:version
6712 machine hardware platform utsname:machine
6714 *** There are new accessors for the vectors returned by `getpw',
6715 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6716 system's user database:
6719 ====================== =================
6720 user name passwd:name
6721 user password passwd:passwd
6724 real name passwd:gecos
6725 home directory passwd:dir
6726 shell program passwd:shell
6728 *** There are new accessors for the vectors returned by `getgr',
6729 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6730 system's group database:
6733 ======================= ============
6734 group name group:name
6735 group password group:passwd
6737 group members group:mem
6739 *** There are new accessors for the vectors returned by `gethost',
6740 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6744 ========================= ===============
6745 official name of host hostent:name
6746 alias list hostent:aliases
6747 host address type hostent:addrtype
6748 length of address hostent:length
6749 list of addresses hostent:addr-list
6751 *** There are new accessors for the vectors returned by `getnet',
6752 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6756 ========================= ===============
6757 official name of net netent:name
6758 alias list netent:aliases
6759 net number type netent:addrtype
6760 net number netent:net
6762 *** There are new accessors for the vectors returned by `getproto',
6763 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6767 ========================= ===============
6768 official protocol name protoent:name
6769 alias list protoent:aliases
6770 protocol number protoent:proto
6772 *** There are new accessors for the vectors returned by `getserv',
6773 `getservbyname', `getservbyport', and `getservent', describing
6777 ========================= ===============
6778 official service name servent:name
6779 alias list servent:aliases
6780 port number servent:port
6781 protocol to use servent:proto
6783 *** There are new accessors for the sockaddr structures returned by
6784 `accept', `getsockname', `getpeername', `recvfrom!':
6787 ======================================== ===============
6788 address format (`family') sockaddr:fam
6789 path, for file domain addresses sockaddr:path
6790 address, for internet domain addresses sockaddr:addr
6791 TCP or UDP port, for internet sockaddr:port
6793 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6794 `getprotoent', and `getservent' functions now return #f at the end of
6795 the user database. (They used to throw an exception.)
6797 Note that calling MUMBLEent function is equivalent to calling the
6798 corresponding MUMBLE function with no arguments.
6800 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6801 `setprotoent', and `setservent' routines now take no arguments.
6803 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6804 provide more useful information when they throw an exception.
6806 *** The `lnaof' function has been renamed to `inet-lnaof'.
6808 *** Guile now claims to have the `current-time' feature.
6810 *** The `mktime' function now takes an optional second argument ZONE,
6811 giving the time zone to use for the conversion. ZONE should be a
6812 string, in the same format as expected for the "TZ" environment variable.
6814 *** The `strptime' function now returns a pair (TIME . COUNT), where
6815 TIME is the parsed time as a vector, and COUNT is the number of
6816 characters from the string left unparsed. This function used to
6817 return the remaining characters as a string.
6819 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6820 The return value is now (SECONDS . MICROSECONDS); the fractional
6821 component is no longer expressed in "ticks".
6823 *** The `ticks/sec' constant has been removed, in light of the above change.
6825 * Changes to the gh_ interface
6827 ** gh_eval_str() now returns an SCM object which is the result of the
6830 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6833 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6834 and returns the array
6836 ** gh_scm2str0() is gone: there is no need to distinguish
6837 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6838 the user to interpret the data both ways.
6840 * Changes to the scm_ interface
6842 ** The new function scm_symbol_value0 provides an easy way to get a
6843 symbol's value from C code:
6845 SCM scm_symbol_value0 (char *NAME)
6846 Return the value of the symbol named by the null-terminated string
6847 NAME in the current module. If the symbol named NAME is unbound in
6848 the current module, return SCM_UNDEFINED.
6850 ** The new function scm_sysintern0 creates new top-level variables,
6851 without assigning them a value.
6853 SCM scm_sysintern0 (char *NAME)
6854 Create a new Scheme top-level variable named NAME. NAME is a
6855 null-terminated string. Return the variable's value cell.
6857 ** The function scm_internal_catch is the guts of catch. It handles
6858 all the mechanics of setting up a catch target, invoking the catch
6859 body, and perhaps invoking the handler if the body does a throw.
6861 The function is designed to be usable from C code, but is general
6862 enough to implement all the semantics Guile Scheme expects from throw.
6864 TAG is the catch tag. Typically, this is a symbol, but this function
6865 doesn't actually care about that.
6867 BODY is a pointer to a C function which runs the body of the catch;
6868 this is the code you can throw from. We call it like this:
6869 BODY (BODY_DATA, JMPBUF)
6871 BODY_DATA is just the BODY_DATA argument we received; we pass it
6872 through to BODY as its first argument. The caller can make
6873 BODY_DATA point to anything useful that BODY might need.
6874 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6875 which we have just created and initialized.
6877 HANDLER is a pointer to a C function to deal with a throw to TAG,
6878 should one occur. We call it like this:
6879 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6881 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6882 same idea as BODY_DATA above.
6883 THROWN_TAG is the tag that the user threw to; usually this is
6884 TAG, but it could be something else if TAG was #t (i.e., a
6885 catch-all), or the user threw to a jmpbuf.
6886 THROW_ARGS is the list of arguments the user passed to the THROW
6889 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6890 is just a pointer we pass through to HANDLER. We don't actually
6891 use either of those pointers otherwise ourselves. The idea is
6892 that, if our caller wants to communicate something to BODY or
6893 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6894 HANDLER can then use. Think of it as a way to make BODY and
6895 HANDLER closures, not just functions; MUMBLE_DATA points to the
6898 Of course, it's up to the caller to make sure that any data a
6899 MUMBLE_DATA needs is protected from GC. A common way to do this is
6900 to make MUMBLE_DATA a pointer to data stored in an automatic
6901 structure variable; since the collector must scan the stack for
6902 references anyway, this assures that any references in MUMBLE_DATA
6905 ** The new function scm_internal_lazy_catch is exactly like
6906 scm_internal_catch, except:
6908 - It does not unwind the stack (this is the major difference).
6909 - If handler returns, its value is returned from the throw.
6910 - BODY always receives #f as its JMPBUF argument (since there's no
6911 jmpbuf associated with a lazy catch, because we don't unwind the
6914 ** scm_body_thunk is a new body function you can pass to
6915 scm_internal_catch if you want the body to be like Scheme's `catch'
6916 --- a thunk, or a function of one argument if the tag is #f.
6918 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6919 contains the Scheme procedure to invoke as the body, and the tag
6920 we're catching. If the tag is #f, then we pass JMPBUF (created by
6921 scm_internal_catch) to the body procedure; otherwise, the body gets
6924 ** scm_handle_by_proc is a new handler function you can pass to
6925 scm_internal_catch if you want the handler to act like Scheme's catch
6926 --- call a procedure with the tag and the throw arguments.
6928 If the user does a throw to this catch, this function runs a handler
6929 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6930 variable holding the Scheme procedure object to invoke. It ought to
6931 be a pointer to an automatic variable (i.e., one living on the stack),
6932 or the procedure object should be otherwise protected from GC.
6934 ** scm_handle_by_message is a new handler function to use with
6935 `scm_internal_catch' if you want Guile to print a message and die.
6936 It's useful for dealing with throws to uncaught keys at the top level.
6938 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6939 message header to print; if zero, we use "guile" instead. That
6940 text is followed by a colon, then the message described by ARGS.
6942 ** The return type of scm_boot_guile is now void; the function does
6943 not return a value, and indeed, never returns at all.
6945 ** The new function scm_shell makes it easy for user applications to
6946 process command-line arguments in a way that is compatible with the
6947 stand-alone guile interpreter (which is in turn compatible with SCSH,
6950 To use the scm_shell function, first initialize any guile modules
6951 linked into your application, and then call scm_shell with the values
6952 of ARGC and ARGV your `main' function received. scm_shell will add
6953 any SCSH-style meta-arguments from the top of the script file to the
6954 argument vector, and then process the command-line arguments. This
6955 generally means loading a script file or starting up an interactive
6956 command interpreter. For details, see "Changes to the stand-alone
6959 ** The new functions scm_get_meta_args and scm_count_argv help you
6960 implement the SCSH-style meta-argument, `\'.
6962 char **scm_get_meta_args (int ARGC, char **ARGV)
6963 If the second element of ARGV is a string consisting of a single
6964 backslash character (i.e. "\\" in Scheme notation), open the file
6965 named by the following argument, parse arguments from it, and return
6966 the spliced command line. The returned array is terminated by a
6969 For details of argument parsing, see above, under "guile now accepts
6970 command-line arguments compatible with SCSH..."
6972 int scm_count_argv (char **ARGV)
6973 Count the arguments in ARGV, assuming it is terminated by a null
6976 For an example of how these functions might be used, see the source
6977 code for the function scm_shell in libguile/script.c.
6979 You will usually want to use scm_shell instead of calling this
6982 ** The new function scm_compile_shell_switches turns an array of
6983 command-line arguments into Scheme code to carry out the actions they
6984 describe. Given ARGC and ARGV, it returns a Scheme expression to
6985 evaluate, and calls scm_set_program_arguments to make any remaining
6986 command-line arguments available to the Scheme code. For example,
6987 given the following arguments:
6989 -e main -s ekko a speckled gecko
6991 scm_set_program_arguments will return the following expression:
6993 (begin (load "ekko") (main (command-line)) (quit))
6995 You will usually want to use scm_shell instead of calling this
6998 ** The function scm_shell_usage prints a usage message appropriate for
6999 an interpreter that uses scm_compile_shell_switches to handle its
7000 command-line arguments.
7002 void scm_shell_usage (int FATAL, char *MESSAGE)
7003 Print a usage message to the standard error output. If MESSAGE is
7004 non-zero, write it before the usage message, followed by a newline.
7005 If FATAL is non-zero, exit the process, using FATAL as the
7006 termination status. (If you want to be compatible with Guile,
7007 always use 1 as the exit status when terminating due to command-line
7010 You will usually want to use scm_shell instead of calling this
7013 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7014 expressions. It used to return SCM_EOL. Earth-shattering.
7016 ** The macros for declaring scheme objects in C code have been
7017 rearranged slightly. They are now:
7019 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7020 Declare a static SCM variable named C_NAME, and initialize it to
7021 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7022 be a C identifier, and SCHEME_NAME should be a C string.
7024 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7025 Just like SCM_SYMBOL, but make C_NAME globally visible.
7027 SCM_VCELL (C_NAME, SCHEME_NAME)
7028 Create a global variable at the Scheme level named SCHEME_NAME.
7029 Declare a static SCM variable named C_NAME, and initialize it to
7030 point to the Scheme variable's value cell.
7032 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7033 Just like SCM_VCELL, but make C_NAME globally visible.
7035 The `guile-snarf' script writes initialization code for these macros
7036 to its standard output, given C source code as input.
7038 The SCM_GLOBAL macro is gone.
7040 ** The scm_read_line and scm_read_line_x functions have been replaced
7041 by Scheme code based on the %read-delimited! procedure (known to C
7042 code as scm_read_delimited_x). See its description above for more
7045 ** The function scm_sys_open has been renamed to scm_open. It now
7046 returns a port instead of an FD object.
7048 * The dynamic linking support has changed. For more information, see
7049 libguile/DYNAMIC-LINKING.
7054 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7057 * Changes to the 'guile' program:
7059 ** Guile now loads some new files when it starts up. Guile first
7060 searches the load path for init.scm, and loads it if found. Then, if
7061 Guile is not being used to execute a script, and the user's home
7062 directory contains a file named `.guile', Guile loads that.
7064 ** You can now use Guile as a shell script interpreter.
7066 To paraphrase the SCSH manual:
7068 When Unix tries to execute an executable file whose first two
7069 characters are the `#!', it treats the file not as machine code to
7070 be directly executed by the native processor, but as source code
7071 to be executed by some interpreter. The interpreter to use is
7072 specified immediately after the #! sequence on the first line of
7073 the source file. The kernel reads in the name of the interpreter,
7074 and executes that instead. It passes the interpreter the source
7075 filename as its first argument, with the original arguments
7076 following. Consult the Unix man page for the `exec' system call
7077 for more information.
7079 Now you can use Guile as an interpreter, using a mechanism which is a
7080 compatible subset of that provided by SCSH.
7082 Guile now recognizes a '-s' command line switch, whose argument is the
7083 name of a file of Scheme code to load. It also treats the two
7084 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7085 to make a file of Scheme code directly executable by Unix, insert the
7086 following two lines at the top of the file:
7088 #!/usr/local/bin/guile -s
7091 Guile treats the argument of the `-s' command-line switch as the name
7092 of a file of Scheme code to load, and treats the sequence `#!' as the
7093 start of a block comment, terminated by `!#'.
7095 For example, here's a version of 'echo' written in Scheme:
7097 #!/usr/local/bin/guile -s
7099 (let loop ((args (cdr (program-arguments))))
7102 (display (car args))
7103 (if (pair? (cdr args))
7105 (loop (cdr args)))))
7108 Why does `#!' start a block comment terminated by `!#', instead of the
7109 end of the line? That is the notation SCSH uses, and although we
7110 don't yet support the other SCSH features that motivate that choice,
7111 we would like to be backward-compatible with any existing Guile
7112 scripts once we do. Furthermore, if the path to Guile on your system
7113 is too long for your kernel, you can start the script with this
7117 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7120 Note that some very old Unix systems don't support the `#!' syntax.
7123 ** You can now run Guile without installing it.
7125 Previous versions of the interactive Guile interpreter (`guile')
7126 couldn't start up unless Guile's Scheme library had been installed;
7127 they used the value of the environment variable `SCHEME_LOAD_PATH'
7128 later on in the startup process, but not to find the startup code
7129 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7132 To run Guile without installing it, build it in the normal way, and
7133 then set the environment variable `SCHEME_LOAD_PATH' to a
7134 colon-separated list of directories, including the top-level directory
7135 of the Guile sources. For example, if you unpacked Guile so that the
7136 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7139 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7142 ** Guile's read-eval-print loop no longer prints #<unspecified>
7143 results. If the user wants to see this, she can evaluate the
7144 expression (assert-repl-print-unspecified #t), perhaps in her startup
7147 ** Guile no longer shows backtraces by default when an error occurs;
7148 however, it does display a message saying how to get one, and how to
7149 request that they be displayed by default. After an error, evaluate
7151 to see a backtrace, and
7152 (debug-enable 'backtrace)
7153 to see them by default.
7157 * Changes to Guile Scheme:
7159 ** Guile now distinguishes between #f and the empty list.
7161 This is for compatibility with the IEEE standard, the (possibly)
7162 upcoming Revised^5 Report on Scheme, and many extant Scheme
7165 Guile used to have #f and '() denote the same object, to make Scheme's
7166 type system more compatible with Emacs Lisp's. However, the change
7167 caused too much trouble for Scheme programmers, and we found another
7168 way to reconcile Emacs Lisp with Scheme that didn't require this.
7171 ** Guile's delq, delv, delete functions, and their destructive
7172 counterparts, delq!, delv!, and delete!, now remove all matching
7173 elements from the list, not just the first. This matches the behavior
7174 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7175 functions which inspired them.
7177 I recognize that this change may break code in subtle ways, but it
7178 seems best to make the change before the FSF's first Guile release,
7182 ** The compiled-library-path function has been deleted from libguile.
7184 ** The facilities for loading Scheme source files have changed.
7186 *** The variable %load-path now tells Guile which directories to search
7187 for Scheme code. Its value is a list of strings, each of which names
7190 *** The variable %load-extensions now tells Guile which extensions to
7191 try appending to a filename when searching the load path. Its value
7192 is a list of strings. Its default value is ("" ".scm").
7194 *** (%search-load-path FILENAME) searches the directories listed in the
7195 value of the %load-path variable for a Scheme file named FILENAME,
7196 with all the extensions listed in %load-extensions. If it finds a
7197 match, then it returns its full filename. If FILENAME is absolute, it
7198 returns it unchanged. Otherwise, it returns #f.
7200 %search-load-path will not return matches that refer to directories.
7202 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7203 uses %seach-load-path to find a file named FILENAME, and loads it if
7204 it finds it. If it can't read FILENAME for any reason, it throws an
7207 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7210 *** load uses the same searching semantics as primitive-load.
7212 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7213 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7214 path, and load-module-with-path have been deleted. The functions
7215 above should serve their purposes.
7217 *** If the value of the variable %load-hook is a procedure,
7218 `primitive-load' applies its value to the name of the file being
7219 loaded (without the load path directory name prepended). If its value
7220 is #f, it is ignored. Otherwise, an error occurs.
7222 This is mostly useful for printing load notification messages.
7225 ** The function `eval!' is no longer accessible from the scheme level.
7226 We can't allow operations which introduce glocs into the scheme level,
7227 because Guile's type system can't handle these as data. Use `eval' or
7228 `read-and-eval!' (see below) as replacement.
7230 ** The new function read-and-eval! reads an expression from PORT,
7231 evaluates it, and returns the result. This is more efficient than
7232 simply calling `read' and `eval', since it is not necessary to make a
7233 copy of the expression for the evaluator to munge.
7235 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7236 for the `read' function.
7239 ** The function `int?' has been removed; its definition was identical
7240 to that of `integer?'.
7242 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7243 use the R4RS names for these functions.
7245 ** The function object-properties no longer returns the hash handle;
7246 it simply returns the object's property list.
7248 ** Many functions have been changed to throw errors, instead of
7249 returning #f on failure. The point of providing exception handling in
7250 the language is to simplify the logic of user code, but this is less
7251 useful if Guile's primitives don't throw exceptions.
7253 ** The function `fileno' has been renamed from `%fileno'.
7255 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7258 * Changes to Guile's C interface:
7260 ** The library's initialization procedure has been simplified.
7261 scm_boot_guile now has the prototype:
7263 void scm_boot_guile (int ARGC,
7265 void (*main_func) (),
7268 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7269 MAIN_FUNC should do all the work of the program (initializing other
7270 packages, reading user input, etc.) before returning. When MAIN_FUNC
7271 returns, call exit (0); this function never returns. If you want some
7272 other exit value, MAIN_FUNC may call exit itself.
7274 scm_boot_guile arranges for program-arguments to return the strings
7275 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7276 scm_set_program_arguments with the final list, so Scheme code will
7277 know which arguments have been processed.
7279 scm_boot_guile establishes a catch-all catch handler which prints an
7280 error message and exits the process. This means that Guile exits in a
7281 coherent way when system errors occur and the user isn't prepared to
7282 handle it. If the user doesn't like this behavior, they can establish
7283 their own universal catcher in MAIN_FUNC to shadow this one.
7285 Why must the caller do all the real work from MAIN_FUNC? The garbage
7286 collector assumes that all local variables of type SCM will be above
7287 scm_boot_guile's stack frame on the stack. If you try to manipulate
7288 SCM values after this function returns, it's the luck of the draw
7289 whether the GC will be able to find the objects you allocate. So,
7290 scm_boot_guile function exits, rather than returning, to discourage
7291 people from making that mistake.
7293 The IN, OUT, and ERR arguments were removed; there are other
7294 convenient ways to override these when desired.
7296 The RESULT argument was deleted; this function should never return.
7298 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7302 ** Guile's header files should no longer conflict with your system's
7305 In order to compile code which #included <libguile.h>, previous
7306 versions of Guile required you to add a directory containing all the
7307 Guile header files to your #include path. This was a problem, since
7308 Guile's header files have names which conflict with many systems'
7311 Now only <libguile.h> need appear in your #include path; you must
7312 refer to all Guile's other header files as <libguile/mumble.h>.
7313 Guile's installation procedure puts libguile.h in $(includedir), and
7314 the rest in $(includedir)/libguile.
7317 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7318 have been added to the Guile library.
7320 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7321 OBJ will not be freed, even if all other references are dropped,
7322 until someone does scm_unprotect_object (OBJ). Both functions
7325 Note that calls to scm_protect_object do not nest. You can call
7326 scm_protect_object any number of times on a given object, and the
7327 next call to scm_unprotect_object will unprotect it completely.
7329 Basically, scm_protect_object and scm_unprotect_object just
7330 maintain a list of references to things. Since the GC knows about
7331 this list, all objects it mentions stay alive. scm_protect_object
7332 adds its argument to the list; scm_unprotect_object remove its
7333 argument from the list.
7336 ** scm_eval_0str now returns the value of the last expression
7339 ** The new function scm_read_0str reads an s-expression from a
7340 null-terminated string, and returns it.
7342 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7343 to a Scheme port object.
7345 ** The new function `scm_set_program_arguments' allows C code to set
7346 the value returned by the Scheme `program-arguments' function.
7351 * Guile no longer includes sophisticated Tcl/Tk support.
7353 The old Tcl/Tk support was unsatisfying to us, because it required the
7354 user to link against the Tcl library, as well as Tk and Guile. The
7355 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7356 referring to widgets by names, rather than exporting widgets to Scheme
7357 code as a special datatype.
7359 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7360 maintainers described some very interesting changes in progress to the
7361 Tcl/Tk internals, which would facilitate clean interfaces between lone
7362 Tk and other interpreters --- even for garbage-collected languages
7363 like Scheme. They expected the new Tk to be publicly available in the
7366 Since it seems that Guile might soon have a new, cleaner interface to
7367 lone Tk, and that the old Guile/Tk glue code would probably need to be
7368 completely rewritten, we (Jim Blandy and Richard Stallman) have
7369 decided not to support the old code. We'll spend the time instead on
7370 a good interface to the newer Tk, as soon as it is available.
7372 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7375 Copyright information:
7377 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7379 Permission is granted to anyone to make or distribute verbatim copies
7380 of this document as received, in any medium, provided that the
7381 copyright notice and this permission notice are preserved,
7382 thus giving the recipient permission to redistribute in turn.
7384 Permission is granted to distribute modified versions
7385 of this document, or of portions of it,
7386 under the above conditions, provided also that they
7387 carry prominent notices stating who last changed them.
7392 paragraph-separate: "[
\f]*$"