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 ** The `(ice-9 i18n)' module provides internationalization support
16 * Changes to the distribution
18 ** Guile now uses Gnulib as a portability aid
20 * Changes to the stand-alone interpreter
21 * Changes to Scheme functions and syntax
23 ** A new 'memoize-symbol evaluator trap has been added. This trap can
24 be used for efficiently implementing a Scheme code coverage.
26 ** Duplicate bindings among used modules are resolved lazily.
27 This slightly improves program startup times.
29 ** New thread cancellation and thread cleanup API
30 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
32 * Changes to the C interface
34 ** Functions for handling `scm_option' now no longer require an argument
35 indicating length of the `scm_t_option' array.
39 Changes in 1.8.5 (since 1.8.4)
43 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
44 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
46 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
47 would trigger an unbound variable error for `match:andmap'.
49 ** Fixed type-checking for the second argument of `eval'
50 ** Fixed build issue for GNU/Linux on IA64
51 ** Fixed build issues on NetBSD 1.6
52 ** Fixed build issue on Solaris 2.10 x86_64
53 ** Fixed build issue with DEC/Compaq/HP's compiler
55 * Changes to the distribution
59 We've started collecting Frequently Asked Questions (FAQ), and will
60 distribute these (with answers!) in future Guile releases.
63 Changes in 1.8.4 (since 1.8.3)
67 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
68 ** Fixed a segmentation fault which occurred when displaying the
69 backtrace of a stack with a promise object (made by `delay') in it.
70 ** Make `accept' leave guile mode while blocking
71 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
72 ** Fixed a build problem on AIX (use of func_data identifier)
73 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
74 called with an associator proc that returns neither a pair nor #f.
75 ** Secondary threads now always return a valid module for (current-module).
76 ** Avoid MacOS build problems caused by incorrect combination of "64"
77 system and library calls.
78 ** `guile-snarf' now honors `$TMPDIR'
79 ** `guile-config compile' now reports CPPFLAGS used at compile-time
80 ** Fixed build with Sun Studio (Solaris 9)
81 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
82 uniform vectors on AIX.
83 ** Fixed a deadlock that occurs upon GC with multiple threads.
84 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
85 ** Fixed autotool-derived build problems on AIX 6.1.
86 ** Fixed NetBSD/alpha support
87 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
89 * New modules (see the manual for details)
93 * Documentation fixes and improvements
95 ** Removed premature breakpoint documentation
97 The features described are not available in the series of 1.8.x
98 releases, so the documentation was misleading and has been removed.
100 ** More about Guile's default *random-state* variable
102 ** GOOPS: more about how to use `next-method'
104 * Changes to the distribution
106 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
108 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
109 General Public License, and the few incorrect files have now been
110 fixed to agree with the rest of the Guile distribution.
112 ** Removed unnecessary extra copies of COPYING*
114 The distribution now contains a single COPYING.LESSER at its top level.
117 Changes in 1.8.3 (since 1.8.2)
119 * New modules (see the manual for details)
126 ** The `(ice-9 slib)' module now works as expected
127 ** Expressions like "(set! 'x #t)" no longer yield a crash
128 ** Warnings about duplicate bindings now go to stderr
129 ** A memory leak in `make-socket-address' was fixed
130 ** Alignment issues (e.g., on SPARC) in network routines were fixed
131 ** A threading issue that showed up at least on NetBSD was fixed
132 ** Build problems on Solaris and IRIX fixed
134 * Implementation improvements
136 ** The reader is now faster, which reduces startup time
137 ** Procedures returned by `record-accessor' and `record-modifier' are faster
141 Changes in 1.8.2 (since 1.8.1):
143 * New procedures (see the manual for details)
145 ** set-program-arguments
150 ** Fractions were not `equal?' if stored in unreduced form.
151 (A subtle problem, since printing a value reduced it, making it work.)
152 ** srfi-60 `copy-bit' failed on 64-bit systems
153 ** "guile --use-srfi" option at the REPL can replace core functions
154 (Programs run with that option were ok, but in the interactive REPL
155 the core bindings got priority, preventing SRFI replacements or
157 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
158 ** `kill' on mingw throws an error for a PID other than oneself
159 ** Procedure names are attached to procedure-with-setters
160 ** Array read syntax works with negative lower bound
161 ** `array-in-bounds?' fix if an array has different lower bounds on each index
162 ** `*' returns exact 0 for "(* inexact 0)"
163 This follows what it always did for "(* 0 inexact)".
164 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
165 ** SRFI-19: `date->julian-day' did not account for timezone offset
166 ** `ttyname' no longer crashes when passed a non-tty argument
167 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
168 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
169 ** GOOPS: Fixed a bug in `method-more-specific?'
170 ** Build problems on Solaris fixed
171 ** Build problems on HP-UX IA64 fixed
172 ** Build problems on MinGW fixed
175 Changes in 1.8.1 (since 1.8.0):
177 * LFS functions are now used to access 64-bit files on 32-bit systems.
179 * New procedures (see the manual for details)
181 ** primitive-_exit - [Scheme] the-root-module
182 ** scm_primitive__exit - [C]
183 ** make-completion-function - [Scheme] (ice-9 readline)
184 ** scm_c_locale_stringn_to_number - [C]
185 ** scm_srfi1_append_reverse [C]
186 ** scm_srfi1_append_reverse_x [C]
192 * New `(ice-9 i18n)' module (see the manual for details)
196 ** Build problems have been fixed on MacOS, SunOS, and QNX.
198 ** `strftime' fix sign of %z timezone offset.
200 ** A one-dimensional array can now be 'equal?' to a vector.
202 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
204 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
206 ** `record-accessor' and `record-modifier' now have strict type checks.
208 Record accessor and modifier procedures now throw an error if the
209 record type of the record they're given is not the type expected.
210 (Previously accessors returned #f and modifiers silently did nothing).
212 ** It is now OK to use both autoload and use-modules on a given module.
214 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
216 Previously there was no checking on primatives like make-vector that
217 accept "one or two" arguments. Now there is.
219 ** The srfi-1 assoc function now calls its equality predicate properly.
221 Previously srfi-1 assoc would call the equality predicate with the key
222 last. According to the SRFI, the key should be first.
224 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
226 ** The array-set! procedure no longer segfaults when given a bit vector.
228 ** Bugs in make-shared-array have been fixed.
230 ** string<? and friends now follow char<? etc order on 8-bit chars.
232 ** The format procedure now handles inf and nan values for ~f correctly.
234 ** exact->inexact should no longer overflow when given certain large fractions.
236 ** srfi-9 accessor and modifier procedures now have strict record type checks.
238 This matches the srfi-9 specification.
240 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
242 Previously the (ice-9 ftw) procedures would ignore any file that had
243 the same inode number as a file they had already seen, even if that
244 file was on a different device.
247 Changes in 1.8.0 (changes since the 1.6.x series):
249 * Changes to the distribution
251 ** Guile is now licensed with the GNU Lesser General Public License.
253 ** The manual is now licensed with the GNU Free Documentation License.
255 ** Guile now requires GNU MP (http://swox.com/gmp).
257 Guile now uses the GNU MP library for arbitrary precision arithmetic.
259 ** Guile now has separate private and public configuration headers.
261 That is, things like HAVE_STRING_H no longer leak from Guile's
264 ** Guile now provides and uses an "effective" version number.
266 Guile now provides scm_effective_version and effective-version
267 functions which return the "effective" version number. This is just
268 the normal full version string without the final micro-version number,
269 so the current effective-version is "1.8". The effective version
270 should remain unchanged during a stable series, and should be used for
271 items like the versioned share directory name
272 i.e. /usr/share/guile/1.8.
274 Providing an unchanging version number during a stable release for
275 things like the versioned share directory can be particularly
276 important for Guile "add-on" packages, since it provides a directory
277 that they can install to that won't be changed out from under them
278 with each micro release during a stable series.
280 ** Thread implementation has changed.
282 When you configure "--with-threads=null", you will get the usual
283 threading API (call-with-new-thread, make-mutex, etc), but you can't
284 actually create new threads. Also, "--with-threads=no" is now
285 equivalent to "--with-threads=null". This means that the thread API
286 is always present, although you might not be able to create new
289 When you configure "--with-threads=pthreads" or "--with-threads=yes",
290 you will get threads that are implemented with the portable POSIX
291 threads. These threads can run concurrently (unlike the previous
292 "coop" thread implementation), but need to cooperate for things like
295 The default is "pthreads", unless your platform doesn't have pthreads,
296 in which case "null" threads are used.
298 See the manual for details, nodes "Initialization", "Multi-Threading",
299 "Blocking", and others.
301 ** There is the new notion of 'discouraged' features.
303 This is a milder form of deprecation.
305 Things that are discouraged should not be used in new code, but it is
306 OK to leave them in old code for now. When a discouraged feature is
307 used, no warning message is printed like there is for 'deprecated'
308 features. Also, things that are merely discouraged are nevertheless
309 implemented efficiently, while deprecated features can be very slow.
311 You can omit discouraged features from libguile by configuring it with
312 the '--disable-discouraged' option.
314 ** Deprecation warnings can be controlled at run-time.
316 (debug-enable 'warn-deprecated) switches them on and (debug-disable
317 'warn-deprecated) switches them off.
319 ** Support for SRFI 61, extended cond syntax for multiple values has
322 This SRFI is always available.
324 ** Support for require-extension, SRFI-55, has been added.
326 The SRFI-55 special form `require-extension' has been added. It is
327 available at startup, and provides a portable way to load Scheme
328 extensions. SRFI-55 only requires support for one type of extension,
329 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
332 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
334 The (srfi srfi-26) module is an implementation of SRFI-26 which
335 provides the `cut' and `cute' syntax. These may be used to specialize
336 parameters without currying.
338 ** New module (srfi srfi-31)
340 This is an implementation of SRFI-31 which provides a special form
341 `rec' for recursive evaluation.
343 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
344 been merged with the core, making their functionality always
347 The modules are still available, tho, and you could use them together
348 with a renaming import, for example.
350 ** Guile no longer includes its own version of libltdl.
352 The official version is good enough now.
354 ** The --enable-htmldoc option has been removed from 'configure'.
356 Support for translating the documentation into HTML is now always
357 provided. Use 'make html'.
359 ** New module (ice-9 serialize):
361 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
362 don't trust the thread safety of most of your program, but where you
363 have some section(s) of code which you consider can run in parallel to
364 other sections. See ice-9/serialize.scm for more information.
366 ** The configure option '--disable-arrays' has been removed.
368 Support for arrays and uniform numeric arrays is now always included
371 * Changes to the stand-alone interpreter
373 ** New command line option `-L'.
375 This option adds a directory to the front of the load path.
377 ** New command line option `--no-debug'.
379 Specifying `--no-debug' on the command line will keep the debugging
380 evaluator turned off, even for interactive sessions.
382 ** User-init file ~/.guile is now loaded with the debugging evaluator.
384 Previously, the normal evaluator would have been used. Using the
385 debugging evaluator gives better error messages.
387 ** The '-e' option now 'read's its argument.
389 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
390 be used with '-e'. For example, you can now write a script like
393 exec guile -e '(@ (demo) main)' -s "$0" "$@"
396 (define-module (demo)
400 (format #t "Demo: ~a~%" args))
403 * Changes to Scheme functions and syntax
405 ** Guardians have changed back to their original semantics
407 Guardians now behave like described in the paper by Dybvig et al. In
408 particular, they no longer make guarantees about the order in which
409 they return objects, and they can no longer be greedy.
411 They no longer drop cyclic data structures.
413 The C function scm_make_guardian has been changed incompatibly and no
414 longer takes the 'greedy_p' argument.
416 ** New function hashx-remove!
418 This function completes the set of 'hashx' functions.
420 ** The concept of dynamic roots has been factored into continuation
421 barriers and dynamic states.
423 Each thread has a current dynamic state that carries the values of the
424 fluids. You can create and copy dynamic states and use them as the
425 second argument for 'eval'. See "Fluids and Dynamic States" in the
428 To restrict the influence that captured continuations can have on the
429 control flow, you can errect continuation barriers. See "Continuation
430 Barriers" in the manual.
432 The function call-with-dynamic-root now essentially temporarily
433 installs a new dynamic state and errects a continuation barrier.
435 ** The default load path no longer includes "." at the end.
437 Automatically loading modules from the current directory should not
438 happen by default. If you want to allow it in a more controlled
439 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
442 ** The uniform vector and array support has been overhauled.
444 It now complies with SRFI-4 and the weird prototype based uniform
445 array creation has been deprecated. See the manual for more details.
447 Some non-compatible changes have been made:
448 - characters can no longer be stored into byte arrays.
449 - strings and bit vectors are no longer considered to be uniform numeric
451 - array-rank throws an error for non-arrays instead of returning zero.
452 - array-ref does no longer accept non-arrays when no indices are given.
454 There is the new notion of 'generalized vectors' and corresponding
455 procedures like 'generalized-vector-ref'. Generalized vectors include
456 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
458 Arrays use generalized vectors as their storage, so that you still
459 have arrays of characters, bits, etc. However, uniform-array-read!
460 and uniform-array-write can no longer read/write strings and
463 ** There is now support for copy-on-write substrings, mutation-sharing
464 substrings and read-only strings.
466 Three new procedures are related to this: substring/shared,
467 substring/copy, and substring/read-only. See the manual for more
470 ** Backtraces will now highlight the value that caused the error.
472 By default, these values are enclosed in "{...}", such as in this
481 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
482 <unnamed port>:1:1: Wrong type (expecting pair): a
483 ABORT: (wrong-type-arg)
485 The prefix and suffix used for highlighting can be set via the two new
486 printer options 'highlight-prefix' and 'highlight-suffix'. For
487 example, putting this into ~/.guile will output the bad value in bold
490 (print-set! highlight-prefix "\x1b[1m")
491 (print-set! highlight-suffix "\x1b[22m")
494 ** 'gettext' support for internationalization has been added.
496 See the manual for details.
498 ** New syntax '@' and '@@':
500 You can now directly refer to variables exported from a module by
503 (@ MODULE-NAME VARIABLE-NAME)
505 For example (@ (ice-9 pretty-print) pretty-print) will directly access
506 the pretty-print variable exported from the (ice-9 pretty-print)
507 module. You don't need to 'use' that module first. You can also use
508 '@' as a target of 'set!', as in (set! (@ mod var) val).
510 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
511 but it can also access variables that have not been exported. It is
512 intended only for kluges and temporary fixes and for debugging, not
515 ** Keyword syntax has been made more disciplined.
517 Previously, the name of a keyword was read as a 'token' but printed as
518 a symbol. Now, it is read as a general Scheme datum which must be a
529 ERROR: In expression (a b c):
533 ERROR: Unbound variable: foo
538 ERROR: Wrong type (expecting symbol): 12
542 ERROR: Wrong type (expecting symbol): (a b c)
546 ** The printing of symbols that might look like keywords can be
549 The new printer option 'quote-keywordish-symbols' controls how symbols
550 are printed that have a colon as their first or last character. The
551 default now is to only quote a symbol with #{...}# when the read
552 option 'keywords' is not '#f'. Thus:
554 guile> (define foo (string->symbol ":foo"))
555 guile> (read-set! keywords #f)
558 guile> (read-set! keywords 'prefix)
561 guile> (print-set! quote-keywordish-symbols #f)
565 ** 'while' now provides 'break' and 'continue'
567 break and continue were previously bound in a while loop, but not
568 documented, and continue didn't quite work properly. The undocumented
569 parameter to break which gave a return value for the while has been
572 ** 'call-with-current-continuation' is now also available under the name
575 ** The module system now checks for duplicate bindings.
577 The module system now can check for name conflicts among imported
580 The behavior can be controlled by specifying one or more 'duplicates'
581 handlers. For example, to make Guile return an error for every name
589 The new default behavior of the module system when a name collision
590 has been detected is to
592 1. Give priority to bindings marked as a replacement.
593 2. Issue a warning (different warning if overriding core binding).
594 3. Give priority to the last encountered binding (this corresponds to
597 If you want the old behavior back without replacements or warnings you
600 (default-duplicate-binding-handler 'last)
602 to your .guile init file.
604 ** New define-module option: :replace
606 :replace works as :export, but, in addition, marks the binding as a
609 A typical example is `format' in (ice-9 format) which is a replacement
610 for the core binding `format'.
612 ** Adding prefixes to imported bindings in the module system
614 There is now a new :use-module option :prefix. It can be used to add
615 a prefix to all imported bindings.
618 :use-module ((bar) :prefix bar:))
620 will import all bindings exported from bar, but rename them by adding
623 ** Conflicting generic functions can be automatically merged.
625 When two imported bindings conflict and they are both generic
626 functions, the two functions can now be merged automatically. This is
627 activated with the 'duplicates' handler 'merge-generics'.
629 ** New function: effective-version
631 Returns the "effective" version number. This is just the normal full
632 version string without the final micro-version number. See "Changes
633 to the distribution" above.
635 ** New threading functions: parallel, letpar, par-map, and friends
637 These are convenient ways to run calculations in parallel in new
638 threads. See "Parallel forms" in the manual for details.
640 ** New function 'try-mutex'.
642 This function will attempt to lock a mutex but will return immediately
643 instead of blocking and indicate failure.
645 ** Waiting on a condition variable can have a timeout.
647 The function 'wait-condition-variable' now takes a third, optional
648 argument that specifies the point in time where the waiting should be
651 ** New function 'broadcast-condition-variable'.
653 ** New functions 'all-threads' and 'current-thread'.
655 ** Signals and system asyncs work better with threads.
657 The function 'sigaction' now takes a fourth, optional, argument that
658 specifies the thread that the handler should run in. When the
659 argument is omitted, the handler will run in the thread that called
662 Likewise, 'system-async-mark' takes a second, optional, argument that
663 specifies the thread that the async should run in. When it is
664 omitted, the async will run in the thread that called
667 C code can use the new functions scm_sigaction_for_thread and
668 scm_system_async_mark_for_thread to pass the new thread argument.
670 When a thread blocks on a mutex, a condition variable or is waiting
671 for IO to be possible, it will still execute system asyncs. This can
672 be used to interrupt such a thread by making it execute a 'throw', for
675 ** The function 'system-async' is deprecated.
677 You can now pass any zero-argument procedure to 'system-async-mark'.
678 The function 'system-async' will just return its argument unchanged
681 ** New functions 'call-with-blocked-asyncs' and
682 'call-with-unblocked-asyncs'
684 The expression (call-with-blocked-asyncs PROC) will call PROC and will
685 block execution of system asyncs for the current thread by one level
686 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
687 procedure and will unblock the execution of system asyncs by one
688 level for the current thread.
690 Only system asyncs are affected by these functions.
692 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
694 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
695 instead. Those functions are easier to use correctly and can be
698 ** New function 'unsetenv'.
700 ** New macro 'define-syntax-public'.
702 It works like 'define-syntax' and also exports the defined macro (but
705 ** There is support for Infinity and NaNs.
707 Following PLT Scheme, Guile can now work with infinite numbers, and
710 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
711 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
712 "+nan.0"). These numbers are inexact and have no exact counterpart.
714 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
715 sign of the dividend. The infinities are integers, and they answer #t
716 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
717 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
728 ERROR: Numerical overflow
730 Two new predicates 'inf?' and 'nan?' can be used to test for the
733 ** Inexact zero can have a sign.
735 Guile can now distinguish between plus and minus inexact zero, if your
736 platform supports this, too. The two zeros are equal according to
737 '=', but not according to 'eqv?'. For example
748 ** Guile now has exact rationals.
750 Guile can now represent fractions such as 1/3 exactly. Computing with
751 them is also done exactly, of course:
756 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
759 For example: (floor 2) now returns an exact 2 where in the past it
760 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
762 ** inexact->exact no longer returns only integers.
764 Without exact rationals, the closest exact number was always an
765 integer, but now inexact->exact returns the fraction that is exactly
766 equal to a floating point number. For example:
768 (inexact->exact 1.234)
769 => 694680242521899/562949953421312
771 When you want the old behavior, use 'round' explicitly:
773 (inexact->exact (round 1.234))
776 ** New function 'rationalize'.
778 This function finds a simple fraction that is close to a given real
779 number. For example (and compare with inexact->exact above):
781 (rationalize (inexact->exact 1.234) 1/2000)
784 Note that, as required by R5RS, rationalize returns only then an exact
785 result when both its arguments are exact.
787 ** 'odd?' and 'even?' work also for inexact integers.
789 Previously, (odd? 1.0) would signal an error since only exact integers
790 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
791 returns #f and (odd? 1.5) signals an error.
793 ** Guile now has uninterned symbols.
795 The new function 'make-symbol' will return an uninterned symbol. This
796 is a symbol that is unique and is guaranteed to remain unique.
797 However, uninterned symbols can not yet be read back in.
799 Use the new function 'symbol-interned?' to check whether a symbol is
802 ** pretty-print has more options.
804 The function pretty-print from the (ice-9 pretty-print) module can now
805 also be invoked with keyword arguments that control things like
806 maximum output width. See the manual for details.
808 ** Variables have no longer a special behavior for `equal?'.
810 Previously, comparing two variables with `equal?' would recursivly
811 compare their values. This is no longer done. Variables are now only
812 `equal?' if they are `eq?'.
814 ** `(begin)' is now valid.
816 You can now use an empty `begin' form. It will yield #<unspecified>
817 when evaluated and simply be ignored in a definition context.
819 ** Deprecated: procedure->macro
821 Change your code to use 'define-macro' or r5rs macros. Also, be aware
822 that macro expansion will not be done during evaluation, but prior to
825 ** Soft ports now allow a `char-ready?' procedure
827 The vector argument to `make-soft-port' can now have a length of
828 either 5 or 6. (Previously the length had to be 5.) The optional 6th
829 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
830 that returns the number of characters that can be read immediately
831 without the soft port blocking.
833 ** Deprecated: undefine
835 There is no replacement for undefine.
837 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
838 have been discouraged.
840 They are relics from a time where a keyword like #:foo was used
841 directly as a Tcl option "-foo" and thus keywords were internally
842 stored as a symbol with a starting dash. We now store a symbol
845 Use symbol->keyword and keyword->symbol instead.
847 ** The `cheap' debug option is now obsolete
849 Evaluator trap calls are now unconditionally "cheap" - in other words,
850 they pass a debug object to the trap handler rather than a full
851 continuation. The trap handler code can capture a full continuation
852 by using `call-with-current-continuation' in the usual way, if it so
855 The `cheap' option is retained for now so as not to break existing
856 code which gets or sets it, but setting it now has no effect. It will
857 be removed in the next major Guile release.
859 ** Evaluator trap calls now support `tweaking'
861 `Tweaking' means that the trap handler code can modify the Scheme
862 expression that is about to be evaluated (in the case of an
863 enter-frame trap) or the value that is being returned (in the case of
864 an exit-frame trap). The trap handler code indicates that it wants to
865 do this by returning a pair whose car is the symbol 'instead and whose
866 cdr is the modified expression or return value.
868 * Changes to the C interface
870 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
871 take a 'delete' function argument.
873 This argument makes no sense since the delete function is used to
874 remove a pair from an alist, and this must not be configurable.
876 This is an incompatible change.
878 ** The GH interface is now subject to the deprecation mechanism
880 The GH interface has been deprecated for quite some time but now it is
881 actually removed from Guile when it is configured with
882 --disable-deprecated.
884 See the manual "Transitioning away from GH" for more information.
886 ** A new family of functions for converting between C values and
887 Scheme values has been added.
889 These functions follow a common naming scheme and are designed to be
890 easier to use, thread-safe and more future-proof than the older
895 These are predicates that return a C boolean: 1 or 0. Instead of
896 SCM_NFALSEP, you can now use scm_is_true, for example.
898 - <type> scm_to_<type> (SCM val, ...)
900 These are functions that convert a Scheme value into an appropriate
901 C value. For example, you can use scm_to_int to safely convert from
904 - SCM scm_from_<type> (<type> val, ...)
906 These functions convert from a C type to a SCM value; for example,
907 scm_from_int for ints.
909 There is a huge number of these functions, for numbers, strings,
910 symbols, vectors, etc. They are documented in the reference manual in
911 the API section together with the types that they apply to.
913 ** New functions for dealing with complex numbers in C have been added.
915 The new functions are scm_c_make_rectangular, scm_c_make_polar,
916 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
917 They work like scm_make_rectangular etc but take or return doubles
920 ** The function scm_make_complex has been discouraged.
922 Use scm_c_make_rectangular instead.
924 ** The INUM macros have been deprecated.
926 A lot of code uses these macros to do general integer conversions,
927 although the macros only work correctly with fixnums. Use the
928 following alternatives.
930 SCM_INUMP -> scm_is_integer or similar
931 SCM_NINUMP -> !scm_is_integer or similar
932 SCM_MAKINUM -> scm_from_int or similar
933 SCM_INUM -> scm_to_int or similar
935 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
936 do the validating for you.
938 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
939 have been discouraged.
941 Use the newer scm_to_<type> and scm_from_<type> functions instead for
942 new code. The functions have been discouraged since they don't fit
945 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
947 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
948 evaluates its argument twice. Use scm_is_true, etc. instead for new
951 ** The macro SCM_EQ_P has been discouraged.
953 Use scm_is_eq for new code, which fits better into the naming
956 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
959 Use the function scm_is_pair or scm_is_null instead.
961 ** The functions scm_round and scm_truncate have been deprecated and
962 are now available as scm_c_round and scm_c_truncate, respectively.
964 These functions occupy the names that scm_round_number and
965 scm_truncate_number should have.
967 ** The functions scm_c_string2str, scm_c_substring2str, and
968 scm_c_symbol2str have been deprecated.
970 Use scm_to_locale_stringbuf or similar instead, maybe together with
973 ** New functions scm_c_make_string, scm_c_string_length,
974 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
975 scm_c_substring_shared, scm_c_substring_copy.
977 These are like scm_make_string, scm_length, etc. but are slightly
978 easier to use from C.
980 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
981 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
983 They export too many assumptions about the implementation of strings
984 and symbols that are no longer true in the presence of
985 mutation-sharing substrings and when Guile switches to some form of
988 When working with strings, it is often best to use the normal string
989 functions provided by Guile, such as scm_c_string_ref,
990 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
991 manual since many more such functions are now provided than
994 When you want to convert a SCM string to a C string, use the
995 scm_to_locale_string function or similar instead. For symbols, use
996 scm_symbol_to_string and then work with that string. Because of the
997 new string representation, scm_symbol_to_string does not need to copy
998 and is thus quite efficient.
1000 ** Some string, symbol and keyword functions have been discouraged.
1002 They don't fit into the uniform naming scheme and are not explicit
1003 about the character encoding.
1005 Replace according to the following table:
1007 scm_allocate_string -> scm_c_make_string
1008 scm_take_str -> scm_take_locale_stringn
1009 scm_take0str -> scm_take_locale_string
1010 scm_mem2string -> scm_from_locale_stringn
1011 scm_str2string -> scm_from_locale_string
1012 scm_makfrom0str -> scm_from_locale_string
1013 scm_mem2symbol -> scm_from_locale_symboln
1014 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1015 scm_str2symbol -> scm_from_locale_symbol
1017 SCM_SYMBOL_HASH -> scm_hashq
1018 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1020 scm_c_make_keyword -> scm_from_locale_keyword
1022 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1023 now also available to C code.
1025 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1027 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1028 the latter returns the true name of the keyword, not the 'dash name',
1029 as SCM_KEYWORDSYM used to do.
1031 ** A new way to access arrays in a thread-safe and efficient way has
1034 See the manual, node "Accessing Arrays From C".
1036 ** The old uniform vector and bitvector implementations have been
1037 unceremoniously removed.
1039 This implementation exposed the details of the tagging system of
1040 Guile. Use the new C API explained in the manual in node "Uniform
1041 Numeric Vectors" and "Bit Vectors", respectively.
1043 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1044 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1045 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1046 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1047 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1048 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1051 ** The macros dealing with vectors have been deprecated.
1053 Use the new functions scm_is_vector, scm_vector_elements,
1054 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1055 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1056 manual for more details.
1058 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1059 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1061 The following macros have been removed: SCM_VECTOR_BASE,
1062 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1063 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1065 ** Some C functions and macros related to arrays have been deprecated.
1067 Migrate according to the following table:
1069 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1070 scm_make_ra -> scm_make_array
1071 scm_shap2ra -> scm_make_array
1072 scm_cvref -> scm_c_generalized_vector_ref
1073 scm_ra_set_contp -> do not use
1074 scm_aind -> scm_array_handle_pos
1075 scm_raprin1 -> scm_display or scm_write
1077 SCM_ARRAYP -> scm_is_array
1078 SCM_ARRAY_NDIM -> scm_c_array_rank
1079 SCM_ARRAY_DIMS -> scm_array_handle_dims
1080 SCM_ARRAY_CONTP -> do not use
1081 SCM_ARRAY_MEM -> do not use
1082 SCM_ARRAY_V -> scm_array_handle_elements or similar
1083 SCM_ARRAY_BASE -> do not use
1085 ** SCM_CELL_WORD_LOC has been deprecated.
1087 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1088 to a SCM, as opposed to a pointer to a scm_t_bits.
1090 This was done to allow the correct use of pointers into the Scheme
1091 heap. Previously, the heap words were of type scm_t_bits and local
1092 variables and function arguments were of type SCM, making it
1093 non-standards-conformant to have a pointer that can point to both.
1095 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1097 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1098 second and third words of double smobs. Likewise for
1099 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1101 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1102 used to get and set the 16 exra bits in the zeroth word of a smob.
1104 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1105 accesing the first immediate word of a smob as a SCM value, and there
1106 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1107 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1109 ** New way to deal with non-local exits and re-entries.
1111 There is a new set of functions that essentially do what
1112 scm_internal_dynamic_wind does, but in a way that is more convenient
1113 for C code in some situations. Here is a quick example of how to
1114 prevent a potential memory leak:
1121 scm_dynwind_begin (0);
1123 mem = scm_malloc (100);
1124 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1126 /* MEM would leak if BAR throws an error.
1127 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1134 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1135 SCM_DYNWIND_END as well.
1139 For full documentation, see the node "Dynamic Wind" in the manual.
1141 ** New function scm_dynwind_free
1143 This function calls 'free' on a given pointer when a dynwind context
1144 is left. Thus the call to scm_dynwind_unwind_handler above could be
1145 replaced with simply scm_dynwind_free (mem).
1147 ** New functions scm_c_call_with_blocked_asyncs and
1148 scm_c_call_with_unblocked_asyncs
1150 Like scm_call_with_blocked_asyncs etc. but for C functions.
1152 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1154 In addition to scm_c_call_with_blocked_asyncs you can now also use
1155 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1156 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1158 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1159 SCM_REALLOW_INTS have been deprecated.
1161 They do no longer fulfill their original role of blocking signal
1162 delivery. Depending on what you want to achieve, replace a pair of
1163 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1164 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1167 ** The value 'scm_mask_ints' is no longer writable.
1169 Previously, you could set scm_mask_ints directly. This is no longer
1170 possible. Use scm_c_call_with_blocked_asyncs and
1171 scm_c_call_with_unblocked_asyncs instead.
1173 ** New way to temporarily set the current input, output or error ports
1175 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1176 context' (see above). <foo> is one of "input", "output" or "error".
1178 ** New way to temporarily set fluids
1180 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1181 above) to temporarily set the value of a fluid.
1183 ** New types scm_t_intmax and scm_t_uintmax.
1185 On platforms that have them, these types are identical to intmax_t and
1186 uintmax_t, respectively. On other platforms, they are identical to
1187 the largest integer types that Guile knows about.
1189 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1191 You should not have used them.
1193 ** Many public #defines with generic names have been made private.
1195 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1196 private or renamed with a more suitable public name.
1198 ** The macro SCM_TYP16S has been deprecated.
1200 This macro is not intended for public use.
1202 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1204 Use scm_is_true (scm_inexact_p (...)) instead.
1206 ** The macro SCM_SLOPPY_REALP has been deprecated.
1208 Use scm_is_real instead.
1210 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1212 Use scm_is_complex instead.
1214 ** Some preprocessor defines have been deprecated.
1216 These defines indicated whether a certain feature was present in Guile
1217 or not. Going forward, assume that the features are always present.
1219 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1220 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1222 The following macros have been removed completely: MEMOIZE_LOCALS,
1223 SCM_RECKLESS, SCM_CAUTIOUS.
1225 ** The preprocessor define STACK_DIRECTION has been deprecated.
1227 There should be no need to know about the stack direction for ordinary
1230 ** New function: scm_effective_version
1232 Returns the "effective" version number. This is just the normal full
1233 version string without the final micro-version number. See "Changes
1234 to the distribution" above.
1236 ** The function scm_call_with_new_thread has a new prototype.
1238 Instead of taking a list with the thunk and handler, these two
1239 arguments are now passed directly:
1241 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1243 This is an incompatible change.
1245 ** New snarfer macro SCM_DEFINE_PUBLIC.
1247 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1248 function in the init section.
1250 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1252 ** Garbage collector rewrite.
1254 The garbage collector is cleaned up a lot, and now uses lazy
1255 sweeping. This is reflected in the output of (gc-stats); since cells
1256 are being freed when they are allocated, the cells-allocated field
1257 stays roughly constant.
1259 For malloc related triggers, the behavior is changed. It uses the same
1260 heuristic as the cell-triggered collections. It may be tuned with the
1261 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1262 for minimum yield of malloc related triggers. The default is 40.
1263 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1266 Debugging operations for the freelist have been deprecated, along with
1267 the C variables that control garbage collection. The environment
1268 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1269 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1271 For understanding the memory usage of a GUILE program, the routine
1272 gc-live-object-stats returns an alist containing the number of live
1273 objects for every type.
1276 ** The function scm_definedp has been renamed to scm_defined_p
1278 The name scm_definedp is deprecated.
1280 ** The struct scm_cell type has been renamed to scm_t_cell
1282 This is in accordance to Guile's naming scheme for types. Note that
1283 the name scm_cell is now used for a function that allocates and
1284 initializes a new cell (see below).
1286 ** New functions for memory management
1288 A new set of functions for memory management has been added since the
1289 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1290 indeed, Guile itself contained some long standing bugs that could
1291 cause aborts in long running programs.
1293 The new functions are more symmetrical and do not need cooperation
1294 from smob free routines, among other improvements.
1296 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1297 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1298 scm_gc_free, scm_gc_register_collectable_memory, and
1299 scm_gc_unregister_collectable_memory. Refer to the manual for more
1300 details and for upgrading instructions.
1302 The old functions for memory management have been deprecated. They
1303 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1304 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1306 ** Declarations of exported features are marked with SCM_API.
1308 Every declaration of a feature that belongs to the exported Guile API
1309 has been marked by adding the macro "SCM_API" to the start of the
1310 declaration. This macro can expand into different things, the most
1311 common of which is just "extern" for Unix platforms. On Win32, it can
1312 be used to control which symbols are exported from a DLL.
1314 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1315 will expand into "__declspec (dllimport) extern", which is needed for
1316 linking to the Guile DLL in Windows.
1318 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1319 SCM_SRFI4_IMPORT, for the corresponding libraries.
1321 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1323 Use the new functions scm_cell and scm_double_cell instead. The old
1324 macros had problems because with them allocation and initialization
1325 was separated and the GC could sometimes observe half initialized
1326 cells. Only careful coding by the user of SCM_NEWCELL and
1327 SCM_NEWCELL2 could make this safe and efficient.
1329 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1331 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1334 ** SRCBRKP has been deprecated.
1336 Use scm_c_source_property_breakpoint_p instead.
1338 ** Deprecated: scm_makmacro
1340 Change your code to use either scm_makmmacro or to define macros in
1341 Scheme, using 'define-macro'.
1343 ** New function scm_c_port_for_each.
1345 This function is like scm_port_for_each but takes a pointer to a C
1346 function as the callback instead of a SCM value.
1348 ** The names scm_internal_select, scm_thread_sleep, and
1349 scm_thread_usleep have been discouraged.
1351 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1353 ** The GC can no longer be blocked.
1355 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1356 The GC can now run (partially) concurrently with other code and thus
1357 blocking it is not well defined.
1359 ** Many definitions have been removed that were previously deprecated.
1361 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1362 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1363 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1364 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1365 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1366 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1367 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1368 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1369 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1370 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1371 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1372 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1373 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1374 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1375 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1376 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1377 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1378 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1379 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1380 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1381 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1382 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1383 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1384 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1385 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1386 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1387 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1388 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1389 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1390 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1391 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1393 * Changes to bundled modules
1397 Using the (ice-9 debug) module no longer automatically switches Guile
1398 to use the debugging evaluator. If you want to switch to the
1399 debugging evaluator (which is needed for backtrace information if you
1400 hit an error), please add an explicit "(debug-enable 'debug)" to your
1401 code just after the code to use (ice-9 debug).
1404 Changes since Guile 1.4:
1406 * Changes to the distribution
1408 ** A top-level TODO file is included.
1410 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1412 Guile now always uses three numbers to represent the version,
1413 i.e. "1.6.5". The first number, 1, is the major version number, the
1414 second number, 6, is the minor version number, and the third number,
1415 5, is the micro version number. Changes in major version number
1416 indicate major changes in Guile.
1418 Minor version numbers that are even denote stable releases, and odd
1419 minor version numbers denote development versions (which may be
1420 unstable). The micro version number indicates a minor sub-revision of
1421 a given MAJOR.MINOR release.
1423 In keeping with the new scheme, (minor-version) and scm_minor_version
1424 no longer return everything but the major version number. They now
1425 just return the minor version number. Two new functions
1426 (micro-version) and scm_micro_version have been added to report the
1427 micro version number.
1429 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1431 ** New preprocessor definitions are available for checking versions.
1433 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1434 SCM_MICRO_VERSION to the appropriate integer values.
1436 ** Guile now actively warns about deprecated features.
1438 The new configure option `--enable-deprecated=LEVEL' and the
1439 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1440 See INSTALL and README for more information.
1442 ** Guile is much more likely to work on 64-bit architectures.
1444 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1445 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1446 for the use of a test machine, and thanks to Stefan Jahn for ia64
1449 ** New functions: setitimer and getitimer.
1451 These implement a fairly direct interface to the libc functions of the
1454 ** The #. reader extension is now disabled by default.
1456 For safety reasons, #. evaluation is disabled by default. To
1457 re-enable it, set the fluid read-eval? to #t. For example:
1459 (fluid-set! read-eval? #t)
1461 but make sure you realize the potential security risks involved. With
1462 read-eval? enabled, reading a data file from an untrusted source can
1465 ** New SRFI modules have been added:
1467 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1470 (srfi srfi-1) is a library containing many useful pair- and list-processing
1473 (srfi srfi-2) exports and-let*.
1475 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1477 (srfi srfi-6) is a dummy module for now, since guile already provides
1478 all of the srfi-6 procedures by default: open-input-string,
1479 open-output-string, get-output-string.
1481 (srfi srfi-8) exports receive.
1483 (srfi srfi-9) exports define-record-type.
1485 (srfi srfi-10) exports define-reader-ctor and implements the reader
1488 (srfi srfi-11) exports let-values and let*-values.
1490 (srfi srfi-13) implements the SRFI String Library.
1492 (srfi srfi-14) implements the SRFI Character-Set Library.
1494 (srfi srfi-17) implements setter and getter-with-setter and redefines
1495 some accessor procedures as procedures with getters. (such as car,
1496 cdr, vector-ref etc.)
1498 (srfi srfi-19) implements the SRFI Time/Date Library.
1500 ** New scripts / "executable modules"
1502 Subdirectory "scripts" contains Scheme modules that are packaged to
1503 also be executable as scripts. At this time, these scripts are available:
1512 See README there for more info.
1514 These scripts can be invoked from the shell with the new program
1515 "guile-tools", which keeps track of installation directory for you.
1518 $ guile-tools display-commentary srfi/*.scm
1520 guile-tools is copied to the standard $bindir on "make install".
1522 ** New module (ice-9 stack-catch):
1524 stack-catch is like catch, but saves the current state of the stack in
1525 the fluid the-last-stack. This fluid can be useful when using the
1526 debugger and when re-throwing an error.
1528 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1530 This has been done to prevent problems on lesser operating systems
1531 that can't tolerate `*'s in file names. The exported macro continues
1532 to be named `and-let*', of course.
1534 On systems that support it, there is also a compatibility module named
1535 (ice-9 and-let*). It will go away in the next release.
1537 ** New modules (oop goops) etc.:
1540 (oop goops describe)
1542 (oop goops active-slot)
1543 (oop goops composite-slot)
1545 The Guile Object Oriented Programming System (GOOPS) has been
1546 integrated into Guile. For further information, consult the GOOPS
1547 manual and tutorial in the `doc' directory.
1549 ** New module (ice-9 rdelim).
1551 This exports the following procedures which were previously defined
1552 in the default environment:
1554 read-line read-line! read-delimited read-delimited! %read-delimited!
1555 %read-line write-line
1557 For backwards compatibility the definitions are still imported into the
1558 default environment in this version of Guile. However you should add:
1560 (use-modules (ice-9 rdelim))
1562 to any program which uses the definitions, since this may change in
1565 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1566 can be used for similar functionality.
1568 ** New module (ice-9 rw)
1570 This is a subset of the (scsh rw) module from guile-scsh. Currently
1571 it defines two procedures:
1573 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1575 Read characters from a port or file descriptor into a string STR.
1576 A port must have an underlying file descriptor -- a so-called
1577 fport. This procedure is scsh-compatible and can efficiently read
1580 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1582 Write characters from a string STR to a port or file descriptor.
1583 A port must have an underlying file descriptor -- a so-called
1584 fport. This procedure is mostly compatible and can efficiently
1585 write large strings.
1587 ** New module (ice-9 match)
1589 This module includes Andrew K. Wright's pattern matcher. See
1590 ice-9/match.scm for brief description or
1592 http://www.star-lab.com/wright/code.html
1594 for complete documentation.
1596 ** New module (ice-9 buffered-input)
1598 This module provides procedures to construct an input port from an
1599 underlying source of input that reads and returns its input in chunks.
1600 The underlying input source is a Scheme procedure, specified by the
1601 caller, which the port invokes whenever it needs more input.
1603 This is useful when building an input port whose back end is Readline
1604 or a UI element such as the GtkEntry widget.
1608 The reference and tutorial documentation that was previously
1609 distributed separately, as `guile-doc', is now included in the core
1610 Guile distribution. The documentation consists of the following
1613 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1616 - The Guile Reference Manual (guile.texi) contains (or is intended to
1617 contain) reference documentation on all aspects of Guile.
1619 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1620 reference documentation for using GOOPS, Guile's Object Oriented
1623 - The Revised^5 Report on the Algorithmic Language Scheme
1626 See the README file in the `doc' directory for more details.
1628 ** There are a couple of examples in the examples/ directory now.
1630 * Changes to the stand-alone interpreter
1632 ** New command line option `--use-srfi'
1634 Using this option, SRFI modules can be loaded on startup and be
1635 available right from the beginning. This makes programming portable
1636 Scheme programs easier.
1638 The option `--use-srfi' expects a comma-separated list of numbers,
1639 each representing a SRFI number to be loaded into the interpreter
1640 before starting evaluating a script file or the REPL. Additionally,
1641 the feature identifier for the loaded SRFIs is recognized by
1642 `cond-expand' when using this option.
1645 $ guile --use-srfi=8,13
1646 guile> (receive (x z) (values 1 2) (+ 1 2))
1648 guile> (string-pad "bla" 20)
1651 ** Guile now always starts up in the `(guile-user)' module.
1653 Previously, scripts executed via the `-s' option would run in the
1654 `(guile)' module and the repl would run in the `(guile-user)' module.
1655 Now every user action takes place in the `(guile-user)' module by
1658 * Changes to Scheme functions and syntax
1660 ** Character classifiers work for non-ASCII characters.
1662 The predicates `char-alphabetic?', `char-numeric?',
1663 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1664 no longer check whether their arguments are ASCII characters.
1665 Previously, a character would only be considered alphabetic when it
1666 was also ASCII, for example.
1668 ** Previously deprecated Scheme functions have been removed:
1670 tag - no replacement.
1671 fseek - replaced by seek.
1672 list* - replaced by cons*.
1674 ** It's now possible to create modules with controlled environments
1678 (use-modules (ice-9 safe))
1679 (define m (make-safe-module))
1680 ;;; m will now be a module containing only a safe subset of R5RS
1681 (eval '(+ 1 2) m) --> 3
1682 (eval 'load m) --> ERROR: Unbound variable: load
1684 ** Evaluation of "()", the empty list, is now an error.
1686 Previously, the expression "()" evaluated to the empty list. This has
1687 been changed to signal a "missing expression" error. The correct way
1688 to write the empty list as a literal constant is to use quote: "'()".
1690 ** New concept of `Guile Extensions'.
1692 A Guile Extension is just a ordinary shared library that can be linked
1693 at run-time. We found it advantageous to give this simple concept a
1694 dedicated name to distinguish the issues related to shared libraries
1695 from the issues related to the module system.
1697 *** New function: load-extension
1699 Executing (load-extension lib init) is mostly equivalent to
1701 (dynamic-call init (dynamic-link lib))
1703 except when scm_register_extension has been called previously.
1704 Whenever appropriate, you should use `load-extension' instead of
1705 dynamic-link and dynamic-call.
1707 *** New C function: scm_c_register_extension
1709 This function registers a initialization function for use by
1710 `load-extension'. Use it when you don't want specific extensions to
1711 be loaded as shared libraries (for example on platforms that don't
1712 support dynamic linking).
1714 ** Auto-loading of compiled-code modules is deprecated.
1716 Guile used to be able to automatically find and link a shared
1717 library to satisfy requests for a module. For example, the module
1718 `(foo bar)' could be implemented by placing a shared library named
1719 "foo/libbar.so" (or with a different extension) in a directory on the
1722 This has been found to be too tricky, and is no longer supported. The
1723 shared libraries are now called "extensions". You should now write a
1724 small Scheme file that calls `load-extension' to load the shared
1725 library and initialize it explicitly.
1727 The shared libraries themselves should be installed in the usual
1728 places for shared libraries, with names like "libguile-foo-bar".
1730 For example, place this into a file "foo/bar.scm"
1732 (define-module (foo bar))
1734 (load-extension "libguile-foo-bar" "foobar_init")
1736 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1738 `eval' is now R5RS, that is it takes two arguments.
1739 The second argument is an environment specifier, i.e. either
1741 (scheme-report-environment 5)
1742 (null-environment 5)
1743 (interaction-environment)
1749 ** The module system has been made more disciplined.
1751 The function `eval' will save and restore the current module around
1752 the evaluation of the specified expression. While this expression is
1753 evaluated, `(current-module)' will now return the right module, which
1754 is the module specified as the second argument to `eval'.
1756 A consequence of this change is that `eval' is not particularly
1757 useful when you want allow the evaluated code to change what module is
1758 designated as the current module and have this change persist from one
1759 call to `eval' to the next. The read-eval-print-loop is an example
1760 where `eval' is now inadequate. To compensate, there is a new
1761 function `primitive-eval' that does not take a module specifier and
1762 that does not save/restore the current module. You should use this
1763 function together with `set-current-module', `current-module', etc
1764 when you want to have more control over the state that is carried from
1765 one eval to the next.
1767 Additionally, it has been made sure that forms that are evaluated at
1768 the top level are always evaluated with respect to the current module.
1769 Previously, subforms of top-level forms such as `begin', `case',
1770 etc. did not respect changes to the current module although these
1771 subforms are at the top-level as well.
1773 To prevent strange behavior, the forms `define-module',
1774 `use-modules', `use-syntax', and `export' have been restricted to only
1775 work on the top level. The forms `define-public' and
1776 `defmacro-public' only export the new binding on the top level. They
1777 behave just like `define' and `defmacro', respectively, when they are
1778 used in a lexical environment.
1780 Also, `export' will no longer silently re-export bindings imported
1781 from a used module. It will emit a `deprecation' warning and will
1782 cease to perform any re-export in the next version. If you actually
1783 want to re-export bindings, use the new `re-export' in place of
1784 `export'. The new `re-export' will not make copies of variables when
1785 rexporting them, as `export' did wrongly.
1787 ** Module system now allows selection and renaming of imported bindings
1789 Previously, when using `use-modules' or the `#:use-module' clause in
1790 the `define-module' form, all the bindings (association of symbols to
1791 values) for imported modules were added to the "current module" on an
1792 as-is basis. This has been changed to allow finer control through two
1793 new facilities: selection and renaming.
1795 You can now select which of the imported module's bindings are to be
1796 visible in the current module by using the `:select' clause. This
1797 clause also can be used to rename individual bindings. For example:
1799 ;; import all bindings no questions asked
1800 (use-modules (ice-9 common-list))
1802 ;; import four bindings, renaming two of them;
1803 ;; the current module sees: every some zonk-y zonk-n
1804 (use-modules ((ice-9 common-list)
1806 (remove-if . zonk-y)
1807 (remove-if-not . zonk-n))))
1809 You can also programmatically rename all selected bindings using the
1810 `:renamer' clause, which specifies a proc that takes a symbol and
1811 returns another symbol. Because it is common practice to use a prefix,
1812 we now provide the convenience procedure `symbol-prefix-proc'. For
1815 ;; import four bindings, renaming two of them specifically,
1816 ;; and all four w/ prefix "CL:";
1817 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1818 (use-modules ((ice-9 common-list)
1820 (remove-if . zonk-y)
1821 (remove-if-not . zonk-n))
1822 :renamer (symbol-prefix-proc 'CL:)))
1824 ;; import four bindings, renaming two of them specifically,
1825 ;; and all four by upcasing.
1826 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1827 (define (upcase-symbol sym)
1828 (string->symbol (string-upcase (symbol->string sym))))
1830 (use-modules ((ice-9 common-list)
1832 (remove-if . zonk-y)
1833 (remove-if-not . zonk-n))
1834 :renamer upcase-symbol))
1836 Note that programmatic renaming is done *after* individual renaming.
1837 Also, the above examples show `use-modules', but the same facilities are
1838 available for the `#:use-module' clause of `define-module'.
1840 See manual for more info.
1842 ** The semantics of guardians have changed.
1844 The changes are for the most part compatible. An important criterion
1845 was to keep the typical usage of guardians as simple as before, but to
1846 make the semantics safer and (as a result) more useful.
1848 *** All objects returned from guardians are now properly alive.
1850 It is now guaranteed that any object referenced by an object returned
1851 from a guardian is alive. It's now impossible for a guardian to
1852 return a "contained" object before its "containing" object.
1854 One incompatible (but probably not very important) change resulting
1855 from this is that it is no longer possible to guard objects that
1856 indirectly reference themselves (i.e. are parts of cycles). If you do
1857 so accidentally, you'll get a warning.
1859 *** There are now two types of guardians: greedy and sharing.
1861 If you call (make-guardian #t) or just (make-guardian), you'll get a
1862 greedy guardian, and for (make-guardian #f) a sharing guardian.
1864 Greedy guardians are the default because they are more "defensive".
1865 You can only greedily guard an object once. If you guard an object
1866 more than once, once in a greedy guardian and the rest of times in
1867 sharing guardians, then it is guaranteed that the object won't be
1868 returned from sharing guardians as long as it is greedily guarded
1871 Guardians returned by calls to `make-guardian' can now take one more
1872 optional parameter, which says whether to throw an error in case an
1873 attempt is made to greedily guard an object that is already greedily
1874 guarded. The default is true, i.e. throw an error. If the parameter
1875 is false, the guardian invocation returns #t if guarding was
1876 successful and #f if it wasn't.
1878 Also, since greedy guarding is, in effect, a side-effecting operation
1879 on objects, a new function is introduced: `destroy-guardian!'.
1880 Invoking this function on a guardian renders it unoperative and, if
1881 the guardian is greedy, clears the "greedily guarded" property of the
1882 objects that were guarded by it, thus undoing the side effect.
1884 Note that all this hair is hardly very important, since guardian
1885 objects are usually permanent.
1887 ** Continuations created by call-with-current-continuation now accept
1888 any number of arguments, as required by R5RS.
1890 ** New function `issue-deprecation-warning'
1892 This function is used to display the deprecation messages that are
1893 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1896 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1900 ;; `id' is deprecated. Use `identity' instead.
1905 ** New syntax `begin-deprecated'
1907 When deprecated features are included (as determined by the configure
1908 option --enable-deprecated), `begin-deprecated' is identical to
1909 `begin'. When deprecated features are excluded, it always evaluates
1910 to `#f', ignoring the body forms.
1912 ** New function `make-object-property'
1914 This function returns a new `procedure with setter' P that can be used
1915 to attach a property to objects. When calling P as
1919 where `obj' is any kind of object, it attaches `val' to `obj' in such
1920 a way that it can be retrieved by calling P as
1924 This function will replace procedure properties, symbol properties and
1925 source properties eventually.
1927 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1929 Instead of #&optional, #&key, etc you should now use #:optional,
1930 #:key, etc. Since #:optional is a keyword, you can write it as just
1931 :optional when (read-set! keywords 'prefix) is active.
1933 The old reader syntax `#&' is still supported, but deprecated. It
1934 will be removed in the next release.
1936 ** New define-module option: pure
1938 Tells the module system not to include any bindings from the root
1943 (define-module (totally-empty-module)
1946 ** New define-module option: export NAME1 ...
1948 Export names NAME1 ...
1950 This option is required if you want to be able to export bindings from
1951 a module which doesn't import one of `define-public' or `export'.
1955 (define-module (foo)
1957 :use-module (ice-9 r5rs)
1960 ;;; Note that we're pure R5RS below this point!
1965 ** New function: object->string OBJ
1967 Return a Scheme string obtained by printing a given object.
1969 ** New function: port? X
1971 Returns a boolean indicating whether X is a port. Equivalent to
1972 `(or (input-port? X) (output-port? X))'.
1974 ** New function: file-port?
1976 Determines whether a given object is a port that is related to a file.
1978 ** New function: port-for-each proc
1980 Apply PROC to each port in the Guile port table in turn. The return
1981 value is unspecified. More specifically, PROC is applied exactly once
1982 to every port that exists in the system at the time PORT-FOR-EACH is
1983 invoked. Changes to the port table while PORT-FOR-EACH is running
1984 have no effect as far as PORT-FOR-EACH is concerned.
1986 ** New function: dup2 oldfd newfd
1988 A simple wrapper for the `dup2' system call. Copies the file
1989 descriptor OLDFD to descriptor number NEWFD, replacing the
1990 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1991 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1992 to move away ports which are using NEWFD. The return value is
1995 ** New function: close-fdes fd
1997 A simple wrapper for the `close' system call. Close file
1998 descriptor FD, which must be an integer. Unlike close (*note
1999 close: Ports and File Descriptors.), the file descriptor will be
2000 closed even if a port is using it. The return value is
2003 ** New function: crypt password salt
2005 Encrypts `password' using the standard unix password encryption
2008 ** New function: chroot path
2010 Change the root directory of the running process to `path'.
2012 ** New functions: getlogin, cuserid
2014 Return the login name or the user name of the current effective user
2017 ** New functions: getpriority which who, setpriority which who prio
2019 Get or set the priority of the running process.
2021 ** New function: getpass prompt
2023 Read a password from the terminal, first displaying `prompt' and
2026 ** New function: flock file operation
2028 Set/remove an advisory shared or exclusive lock on `file'.
2030 ** New functions: sethostname name, gethostname
2032 Set or get the hostname of the machine the current process is running
2035 ** New function: mkstemp! tmpl
2037 mkstemp creates a new unique file in the file system and returns a
2038 new buffered port open for reading and writing to the file. TMPL
2039 is a string specifying where the file should be created: it must
2040 end with `XXXXXX' and will be changed in place to return the name
2041 of the temporary file.
2043 ** New function: open-input-string string
2045 Return an input string port which delivers the characters from
2046 `string'. This procedure, together with `open-output-string' and
2047 `get-output-string' implements SRFI-6.
2049 ** New function: open-output-string
2051 Return an output string port which collects all data written to it.
2052 The data can then be retrieved by `get-output-string'.
2054 ** New function: get-output-string
2056 Return the contents of an output string port.
2058 ** New function: identity
2060 Return the argument.
2062 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2063 are represented in Scheme as integers with normal host byte ordering.
2065 ** New function: inet-pton family address
2067 Convert a printable string network address into an integer. Note that
2068 unlike the C version of this function, the result is an integer with
2069 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2072 (inet-pton AF_INET "127.0.0.1") => 2130706433
2073 (inet-pton AF_INET6 "::1") => 1
2075 ** New function: inet-ntop family address
2077 Convert an integer network address into a printable string. Note that
2078 unlike the C version of this function, the input is an integer with
2079 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2082 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2083 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2084 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2088 Use `identity' instead.
2094 ** Deprecated: return-it
2098 ** Deprecated: string-character-length
2100 Use `string-length' instead.
2102 ** Deprecated: flags
2104 Use `logior' instead.
2106 ** Deprecated: close-all-ports-except.
2108 This was intended for closing ports in a child process after a fork,
2109 but it has the undesirable side effect of flushing buffers.
2110 port-for-each is more flexible.
2112 ** The (ice-9 popen) module now attempts to set up file descriptors in
2113 the child process from the current Scheme ports, instead of using the
2114 current values of file descriptors 0, 1, and 2 in the parent process.
2116 ** Removed function: builtin-weak-bindings
2118 There is no such concept as a weak binding any more.
2120 ** Removed constants: bignum-radix, scm-line-incrementors
2122 ** define-method: New syntax mandatory.
2124 The new method syntax is now mandatory:
2126 (define-method (NAME ARG-SPEC ...) BODY ...)
2127 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2129 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2130 REST-ARG ::= ARG-NAME
2132 If you have old code using the old syntax, import
2133 (oop goops old-define-method) before (oop goops) as in:
2135 (use-modules (oop goops old-define-method) (oop goops))
2137 ** Deprecated function: builtin-variable
2138 Removed function: builtin-bindings
2140 There is no longer a distinction between builtin or other variables.
2141 Use module system operations for all variables.
2143 ** Lazy-catch handlers are no longer allowed to return.
2145 That is, a call to `throw', `error', etc is now guaranteed to not
2148 ** Bugfixes for (ice-9 getopt-long)
2150 This module is now tested using test-suite/tests/getopt-long.test.
2151 The following bugs have been fixed:
2153 *** Parsing for options that are specified to have `optional' args now checks
2154 if the next element is an option instead of unconditionally taking it as the
2157 *** An error is now thrown for `--opt=val' when the option description
2158 does not specify `(value #t)' or `(value optional)'. This condition used to
2159 be accepted w/o error, contrary to the documentation.
2161 *** The error message for unrecognized options is now more informative.
2162 It used to be "not a record", an artifact of the implementation.
2164 *** The error message for `--opt' terminating the arg list (no value), when
2165 `(value #t)' is specified, is now more informative. It used to be "not enough
2168 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2169 The expansion used to be like so:
2171 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2173 Note that the "5d" is dropped. Now it is like so:
2175 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2177 This enables single-char options to have adjoining arguments as long as their
2178 constituent characters are not potential single-char options.
2180 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2182 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2183 property `arglist', which can be retrieved by `arity'. The result is that
2184 `arity' can give more detailed information than before:
2188 guile> (use-modules (ice-9 optargs))
2189 guile> (define* (foo #:optional a b c) a)
2191 0 or more arguments in `lambda*:G0'.
2196 3 optional arguments: `a', `b' and `c'.
2197 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2199 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2200 and `d', other keywords allowed.
2201 guile> (define* (baz a b #:optional c #:rest r) a)
2203 2 required arguments: `a' and `b', 1 optional argument: `c',
2206 * Changes to the C interface
2208 ** Types have been renamed from scm_*_t to scm_t_*.
2210 This has been done for POSIX sake. It reserves identifiers ending
2211 with "_t". What a concept.
2213 The old names are still available with status `deprecated'.
2215 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2217 ** Deprecated features have been removed.
2221 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2222 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2224 *** C Functions removed
2226 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2227 scm_fseek - replaced by scm_seek.
2228 gc-thunk - replaced by after-gc-hook.
2229 gh_int2scmb - replaced by gh_bool2scm.
2230 scm_tc_dblr - replaced by scm_tc16_real.
2231 scm_tc_dblc - replaced by scm_tc16_complex.
2232 scm_list_star - replaced by scm_cons_star.
2234 ** Deprecated: scm_makfromstr
2236 Use scm_mem2string instead.
2238 ** Deprecated: scm_make_shared_substring
2240 Explicit shared substrings will disappear from Guile.
2242 Instead, "normal" strings will be implemented using sharing
2243 internally, combined with a copy-on-write strategy.
2245 ** Deprecated: scm_read_only_string_p
2247 The concept of read-only strings will disappear in next release of
2250 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2252 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2254 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2256 Call a procedure with the indicated number of arguments. See "Fly
2257 Evaluation" in the manual.
2259 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2261 Call a procedure with the indicated number of arguments and a list of
2262 further arguments. See "Fly Evaluation" in the manual.
2264 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2266 Create a list of the given number of elements. See "List
2267 Constructors" in the manual.
2269 ** Renamed function: scm_listify has been replaced by scm_list_n.
2271 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2272 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2274 Use functions scm_list_N instead.
2276 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2278 Used by an application to read arbitrary number of bytes from a port.
2279 Same semantics as libc read, except that scm_c_read only returns less
2280 than SIZE bytes if at end-of-file.
2282 Warning: Doesn't update port line and column counts!
2284 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2286 Used by an application to write arbitrary number of bytes to an SCM
2287 port. Similar semantics as libc write. However, unlike libc
2288 write, scm_c_write writes the requested number of bytes and has no
2291 Warning: Doesn't update port line and column counts!
2293 ** New function: scm_init_guile ()
2295 In contrast to scm_boot_guile, scm_init_guile will return normally
2296 after initializing Guile. It is not available on all systems, tho.
2298 ** New functions: scm_str2symbol, scm_mem2symbol
2300 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2301 field of characters and creates a scheme symbol object from that C string.
2302 The function scm_mem2symbol takes a const char* and a number of characters and
2303 creates a symbol from the characters in that memory area.
2305 ** New functions: scm_primitive_make_property
2306 scm_primitive_property_ref
2307 scm_primitive_property_set_x
2308 scm_primitive_property_del_x
2310 These functions implement a new way to deal with object properties.
2311 See libguile/properties.c for their documentation.
2313 ** New function: scm_done_free (long size)
2315 This function is the inverse of scm_done_malloc. Use it to report the
2316 amount of smob memory you free. The previous method, which involved
2317 calling scm_done_malloc with negative argument, was somewhat
2318 unintuitive (and is still available, of course).
2320 ** New function: scm_c_memq (SCM obj, SCM list)
2322 This function provides a fast C level alternative for scm_memq for the case
2323 that the list parameter is known to be a proper list. The function is a
2324 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2325 list input parameter, since for anything else but a proper list the function's
2326 behaviour is undefined - it may even crash or loop endlessly. Further, for
2327 the case that the object is not found in the list, scm_c_memq returns #f which
2328 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2330 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2331 scm_remember_upto_here
2333 These functions replace the function scm_remember.
2335 ** Deprecated function: scm_remember
2337 Use one of the new functions scm_remember_upto_here_1,
2338 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2340 ** New function: scm_allocate_string
2342 This function replaces the function scm_makstr.
2344 ** Deprecated function: scm_makstr
2346 Use the new function scm_allocate_string instead.
2348 ** New global variable scm_gc_running_p introduced.
2350 Use this variable to find out if garbage collection is being executed. Up to
2351 now applications have used scm_gc_heap_lock to test if garbage collection was
2352 running, which also works because of the fact that up to know only the garbage
2353 collector has set this variable. But, this is an implementation detail that
2354 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2355 of this variable is (and has been) not fully safe anyway.
2357 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2359 Use these instead of SCM_LENGTH_MAX.
2361 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2362 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2363 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2365 Use these instead of SCM_LENGTH.
2367 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2368 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2369 SCM_SET_BITVECTOR_LENGTH
2371 Use these instead of SCM_SETLENGTH
2373 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2374 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2377 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2380 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2381 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2384 Use these instead of SCM_SETCHARS.
2386 ** New macro: SCM_BITVECTOR_P
2388 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2390 Use instead of SCM_COERCE_SUBSTR.
2392 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2394 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2396 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2397 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2398 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2399 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2400 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2401 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2402 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2403 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2404 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2405 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2406 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2407 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2408 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2409 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2410 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2412 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2413 Use scm_memory_error instead of SCM_NALLOC.
2414 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2415 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2416 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2417 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2418 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2419 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2420 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2421 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2422 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2423 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2424 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2425 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2426 Use a type specific setter macro instead of SCM_SETLENGTH.
2427 Use a type specific setter macro instead of SCM_SETCHARS.
2428 Use a type specific length macro instead of SCM_LENGTH_MAX.
2429 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2430 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2431 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2432 Use SCM_TYP16 instead of SCM_GCTYP16.
2433 Use SCM_CDR instead of SCM_GCCDR.
2434 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2435 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2436 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2437 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2438 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2439 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2440 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2442 ** Removed function: scm_struct_init
2444 ** Removed variable: scm_symhash_dim
2446 ** Renamed function: scm_make_cont has been replaced by
2447 scm_make_continuation, which has a different interface.
2449 ** Deprecated function: scm_call_catching_errors
2451 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2453 ** Deprecated function: scm_strhash
2455 Use scm_string_hash instead.
2457 ** Deprecated function: scm_vector_set_length_x
2459 Instead, create a fresh vector of the desired size and copy the contents.
2461 ** scm_gensym has changed prototype
2463 scm_gensym now only takes one argument.
2465 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2468 There is now only a single symbol type scm_tc7_symbol.
2469 The tag scm_tc7_lvector was not used anyway.
2471 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2473 Use scm_make_smob_type and scm_set_smob_XXX instead.
2475 ** New function scm_set_smob_apply.
2477 This can be used to set an apply function to a smob type.
2479 ** Deprecated function: scm_strprint_obj
2481 Use scm_object_to_string instead.
2483 ** Deprecated function: scm_wta
2485 Use scm_wrong_type_arg, or another appropriate error signalling function
2488 ** Explicit support for obarrays has been deprecated.
2490 Use `scm_str2symbol' and the generic hashtable functions instead.
2492 ** The concept of `vcells' has been deprecated.
2494 The data type `variable' is now used exclusively. `Vcells' have been
2495 a low-level concept so you are likely not affected by this change.
2497 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2498 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2500 Use scm_c_define or scm_c_lookup instead, as appropriate.
2502 *** New functions: scm_c_module_lookup, scm_c_lookup,
2503 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2504 scm_module_define, scm_define.
2506 These functions work with variables instead of with vcells.
2508 ** New functions for creating and defining `subr's and `gsubr's.
2510 The new functions more clearly distinguish between creating a subr (or
2511 gsubr) object and adding it to the current module.
2513 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2514 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2515 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2516 scm_c_define_gsubr_with_generic.
2518 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2519 scm_make_subr_with_generic, scm_make_gsubr,
2520 scm_make_gsubr_with_generic.
2522 Use the new ones from above instead.
2524 ** C interface to the module system has changed.
2526 While we suggest that you avoid as many explicit module system
2527 operations from C as possible for the time being, the C interface has
2528 been made more similar to the high-level Scheme module system.
2530 *** New functions: scm_c_define_module, scm_c_use_module,
2531 scm_c_export, scm_c_resolve_module.
2533 They mostly work like their Scheme namesakes. scm_c_define_module
2534 takes a function that is called a context where the new module is
2537 *** Deprecated functions: scm_the_root_module, scm_make_module,
2538 scm_ensure_user_module, scm_load_scheme_module.
2540 Use the new functions instead.
2542 ** Renamed function: scm_internal_with_fluids becomes
2545 scm_internal_with_fluids is available as a deprecated function.
2547 ** New function: scm_c_with_fluid.
2549 Just like scm_c_with_fluids, but takes one fluid and one value instead
2552 ** Deprecated typedefs: long_long, ulong_long.
2554 They are of questionable utility and they pollute the global
2557 ** Deprecated typedef: scm_sizet
2559 It is of questionable utility now that Guile requires ANSI C, and is
2562 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2563 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2564 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2566 Made more compliant with the naming policy by adding a _t at the end.
2568 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2569 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2571 With the exception of the mysterious scm_2ulong2big, they are still
2572 available under new names (scm_i_mkbig etc). These functions are not
2573 intended to be used in user code. You should avoid dealing with
2574 bignums directly, and should deal with numbers in general (which can
2577 ** Change in behavior: scm_num2long, scm_num2ulong
2579 The scm_num2[u]long functions don't any longer accept an inexact
2580 argument. This change in behavior is motivated by concordance with
2581 R5RS: It is more common that a primitive doesn't want to accept an
2582 inexact for an exact.
2584 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2585 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2586 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2589 These are conversion functions between the various ANSI C integral
2590 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2591 accept an inexact argument.
2593 ** New functions: scm_float2num, scm_double2num,
2594 scm_num2float, scm_num2double.
2596 These are conversion functions between the two ANSI C float types and
2599 ** New number validation macros:
2600 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2604 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2606 These are just nicer-named old scm_protect_object and
2607 scm_unprotect_object.
2609 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2611 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2613 These functions can be used to register pointers to locations that
2616 ** Deprecated function: scm_create_hook.
2618 Its sins are: misleading name, non-modularity and lack of general
2622 Changes since Guile 1.3.4:
2624 * Changes to the distribution
2626 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2628 We've changed the way we handle generated files in the Guile source
2629 repository. As a result, the procedure for building trees obtained
2630 from the nightly FTP snapshots or via CVS has changed:
2631 - You must have appropriate versions of autoconf, automake, and
2632 libtool installed on your system. See README for info on how to
2633 obtain these programs.
2634 - Before configuring the tree, you must first run the script
2635 `autogen.sh' at the top of the source tree.
2637 The Guile repository used to contain not only source files, written by
2638 humans, but also some generated files, like configure scripts and
2639 Makefile.in files. Even though the contents of these files could be
2640 derived mechanically from other files present, we thought it would
2641 make the tree easier to build if we checked them into CVS.
2643 However, this approach means that minor differences between
2644 developer's installed tools and habits affected the whole team.
2645 So we have removed the generated files from the repository, and
2646 added the autogen.sh script, which will reconstruct them
2650 ** configure now has experimental options to remove support for certain
2653 --disable-arrays omit array and uniform array support
2654 --disable-posix omit posix interfaces
2655 --disable-networking omit networking interfaces
2656 --disable-regex omit regular expression interfaces
2658 These are likely to become separate modules some day.
2660 ** New configure option --enable-debug-freelist
2662 This enables a debugging version of SCM_NEWCELL(), and also registers
2663 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2665 Configure with the --enable-debug-freelist option to enable
2666 the gc-set-debug-check-freelist! primitive, and then use:
2668 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2669 (gc-set-debug-check-freelist! #f) # turn off checking
2671 Checking of the freelist forces a traversal of the freelist and
2672 a garbage collection before each allocation of a cell. This can
2673 slow down the interpreter dramatically, so the setter should be used to
2674 turn on this extra processing only when necessary.
2676 ** New configure option --enable-debug-malloc
2678 Include code for debugging of calls to scm_must_malloc/realloc/free.
2682 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2683 2. objects reallocated by scm_must_realloc has been allocated by
2685 3. reallocated objects are reallocated with the same what string
2687 But, most importantly, it records the number of allocated objects of
2688 each kind. This is useful when searching for memory leaks.
2690 A Guile compiled with this option provides the primitive
2691 `malloc-stats' which returns an alist with pairs of kind and the
2692 number of objects of that kind.
2694 ** All includes are now referenced relative to the root directory
2696 Since some users have had problems with mixups between Guile and
2697 system headers, we have decided to always refer to Guile headers via
2698 their parent directories. This essentially creates a "private name
2699 space" for Guile headers. This means that the compiler only is given
2700 -I options for the root build and root source directory.
2702 ** Header files kw.h and genio.h have been removed.
2704 ** The module (ice-9 getopt-gnu-style) has been removed.
2706 ** New module (ice-9 documentation)
2708 Implements the interface to documentation strings associated with
2711 ** New module (ice-9 time)
2713 Provides a macro `time', which displays execution time of a given form.
2715 ** New module (ice-9 history)
2717 Loading this module enables value history in the repl.
2719 * Changes to the stand-alone interpreter
2721 ** New command line option --debug
2723 Start Guile with debugging evaluator and backtraces enabled.
2725 This is useful when debugging your .guile init file or scripts.
2727 ** New help facility
2729 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2730 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2731 (help 'NAME) gives documentation for NAME, even if it is not an object
2732 (help ,EXPR) gives documentation for object returned by EXPR
2733 (help (my module)) gives module commentary for `(my module)'
2734 (help) gives this text
2736 `help' searches among bindings exported from loaded modules, while
2737 `apropos' searches among bindings visible from the "current" module.
2739 Examples: (help help)
2741 (help "output-string")
2743 ** `help' and `apropos' now prints full module names
2745 ** Dynamic linking now uses libltdl from the libtool package.
2747 The old system dependent code for doing dynamic linking has been
2748 replaced with calls to the libltdl functions which do all the hairy
2751 The major improvement is that you can now directly pass libtool
2752 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2753 will be able to do the best shared library job you can get, via
2756 The way dynamic libraries are found has changed and is not really
2757 portable across platforms, probably. It is therefore recommended to
2758 use absolute filenames when possible.
2760 If you pass a filename without an extension to `dynamic-link', it will
2761 try a few appropriate ones. Thus, the most platform ignorant way is
2762 to specify a name like "libfoo", without any directories and
2765 ** Guile COOP threads are now compatible with LinuxThreads
2767 Previously, COOP threading wasn't possible in applications linked with
2768 Linux POSIX threads due to their use of the stack pointer to find the
2769 thread context. This has now been fixed with a workaround which uses
2770 the pthreads to allocate the stack.
2772 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2774 ** Positions of erring expression in scripts
2776 With version 1.3.4, the location of the erring expression in Guile
2777 scipts is no longer automatically reported. (This should have been
2778 documented before the 1.3.4 release.)
2780 You can get this information by enabling recording of positions of
2781 source expressions and running the debugging evaluator. Put this at
2782 the top of your script (or in your "site" file):
2784 (read-enable 'positions)
2785 (debug-enable 'debug)
2787 ** Backtraces in scripts
2789 It is now possible to get backtraces in scripts.
2793 (debug-enable 'debug 'backtrace)
2795 at the top of the script.
2797 (The first options enables the debugging evaluator.
2798 The second enables backtraces.)
2800 ** Part of module system symbol lookup now implemented in C
2802 The eval closure of most modules is now implemented in C. Since this
2803 was one of the bottlenecks for loading speed, Guile now loads code
2804 substantially faster than before.
2806 ** Attempting to get the value of an unbound variable now produces
2807 an exception with a key of 'unbound-variable instead of 'misc-error.
2809 ** The initial default output port is now unbuffered if it's using a
2810 tty device. Previously in this situation it was line-buffered.
2812 ** New hook: after-gc-hook
2814 after-gc-hook takes over the role of gc-thunk. This hook is run at
2815 the first SCM_TICK after a GC. (Thus, the code is run at the same
2816 point during evaluation as signal handlers.)
2818 Note that this hook should be used only for diagnostic and debugging
2819 purposes. It is not certain that it will continue to be well-defined
2820 when this hook is run in the future.
2822 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2823 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2825 ** Improvements to garbage collector
2827 Guile 1.4 has a new policy for triggering heap allocation and
2828 determining the sizes of heap segments. It fixes a number of problems
2831 1. The new policy can handle two separate pools of cells
2832 (2-word/4-word) better. (The old policy would run wild, allocating
2833 more and more memory for certain programs.)
2835 2. The old code would sometimes allocate far too much heap so that the
2836 Guile process became gigantic. The new code avoids this.
2838 3. The old code would sometimes allocate too little so that few cells
2839 were freed at GC so that, in turn, too much time was spent in GC.
2841 4. The old code would often trigger heap allocation several times in a
2842 row. (The new scheme predicts how large the segments needs to be
2843 in order not to need further allocation.)
2845 All in all, the new GC policy will make larger applications more
2848 The new GC scheme also is prepared for POSIX threading. Threads can
2849 allocate private pools of cells ("clusters") with just a single
2850 function call. Allocation of single cells from such a cluster can
2851 then proceed without any need of inter-thread synchronization.
2853 ** New environment variables controlling GC parameters
2855 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2858 Allocation of 2-word cell heaps:
2860 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2863 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2864 GC in percent of total heap size
2867 Allocation of 4-word cell heaps
2868 (used for real numbers and misc other objects):
2870 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2872 (See entry "Way for application to customize GC parameters" under
2873 section "Changes to the scm_ interface" below.)
2875 ** Guile now implements reals using 4-word cells
2877 This speeds up computation with reals. (They were earlier allocated
2878 with `malloc'.) There is still some room for optimizations, however.
2880 ** Some further steps toward POSIX thread support have been taken
2882 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2883 don't have much effect any longer, and many of them will be removed in
2887 are only handled at the top of the evaluator loop, immediately after
2888 I/O, and in scm_equalp.
2890 *** The GC can allocate thread private pools of pairs.
2892 * Changes to Scheme functions and syntax
2894 ** close-input-port and close-output-port are now R5RS
2896 These procedures have been turned into primitives and have R5RS behaviour.
2898 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2900 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2901 extended by the more sophisticated version in (ice-9 format)
2903 (simple-format port message . args)
2904 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2905 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2906 the escapes are replaced with corresponding members of ARGS:
2907 ~A formats using `display' and ~S formats using `write'.
2908 If DESTINATION is #t, then use the `current-output-port',
2909 if DESTINATION is #f, then return a string containing the formatted text.
2910 Does not add a trailing newline."
2912 ** string-ref: the second argument is no longer optional.
2914 ** string, list->string: no longer accept strings in their arguments,
2915 only characters, for compatibility with R5RS.
2917 ** New procedure: port-closed? PORT
2918 Returns #t if PORT is closed or #f if it is open.
2920 ** Deprecated: list*
2922 The list* functionality is now provided by cons* (SRFI-1 compliant)
2924 ** New procedure: cons* ARG1 ARG2 ... ARGn
2926 Like `list', but the last arg provides the tail of the constructed list,
2927 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2929 Requires at least one argument. If given one argument, that argument
2930 is returned as result.
2932 This function is called `list*' in some other Schemes and in Common LISP.
2934 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2936 ** New procedure: object-documentation OBJECT
2938 Returns the documentation string associated with OBJECT. The
2939 procedure uses a caching mechanism so that subsequent lookups are
2942 Exported by (ice-9 documentation).
2944 ** module-name now returns full names of modules
2946 Previously, only the last part of the name was returned (`session' for
2947 `(ice-9 session)'). Ex: `(ice-9 session)'.
2949 * Changes to the gh_ interface
2951 ** Deprecated: gh_int2scmb
2953 Use gh_bool2scm instead.
2955 * Changes to the scm_ interface
2957 ** Guile primitives now carry docstrings!
2959 Thanks to Greg Badros!
2961 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2963 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2964 macros and must contain a docstring that is extracted into foo.doc using a new
2965 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2967 However, a major overhaul of these macros is scheduled for the next release of
2970 ** Guile primitives use a new technique for validation of arguments
2972 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2973 the readability of argument checking.
2975 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2977 ** New macros: SCM_PACK, SCM_UNPACK
2979 Compose/decompose an SCM value.
2981 The SCM type is now treated as an abstract data type and may be defined as a
2982 long, a void* or as a struct, depending on the architecture and compile time
2983 options. This makes it easier to find several types of bugs, for example when
2984 SCM values are treated as integers without conversion. Values of the SCM type
2985 should be treated as "atomic" values. These macros are used when
2986 composing/decomposing an SCM value, either because you want to access
2987 individual bits, or because you want to treat it as an integer value.
2989 E.g., in order to set bit 7 in an SCM value x, use the expression
2991 SCM_PACK (SCM_UNPACK (x) | 0x80)
2993 ** The name property of hooks is deprecated.
2994 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2996 You can emulate this feature by using object properties.
2998 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2999 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3002 These macros will be removed in a future release of Guile.
3004 ** The following types, functions and macros from numbers.h are deprecated:
3005 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3006 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3008 ** Port internals: the rw_random variable in the scm_port structure
3009 must be set to non-zero in any random access port. In recent Guile
3010 releases it was only set for bidirectional random-access ports.
3012 ** Port internals: the seek ptob procedure is now responsible for
3013 resetting the buffers if required. The change was made so that in the
3014 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3015 the fport and strport ptobs can avoid resetting the buffers,
3016 in particular to avoid discarding unread chars. An existing port
3017 type can be fixed by adding something like the following to the
3018 beginning of the ptob seek procedure:
3020 if (pt->rw_active == SCM_PORT_READ)
3021 scm_end_input (object);
3022 else if (pt->rw_active == SCM_PORT_WRITE)
3023 ptob->flush (object);
3025 although to actually avoid resetting the buffers and discard unread
3026 chars requires further hacking that depends on the characteristics
3029 ** Deprecated functions: scm_fseek, scm_tag
3031 These functions are no longer used and will be removed in a future version.
3033 ** The scm_sysmissing procedure is no longer used in libguile.
3034 Unless it turns out to be unexpectedly useful to somebody, it will be
3035 removed in a future version.
3037 ** The format of error message strings has changed
3039 The two C procedures: scm_display_error and scm_error, as well as the
3040 primitive `scm-error', now use scm_simple_format to do their work.
3041 This means that the message strings of all code must be updated to use
3042 ~A where %s was used before, and ~S where %S was used before.
3044 During the period when there still are a lot of old Guiles out there,
3045 you might want to support both old and new versions of Guile.
3047 There are basically two methods to achieve this. Both methods use
3050 AC_CHECK_FUNCS(scm_simple_format)
3052 in your configure.in.
3054 Method 1: Use the string concatenation features of ANSI C's
3059 #ifdef HAVE_SCM_SIMPLE_FORMAT
3065 Then represent each of your error messages using a preprocessor macro:
3067 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3071 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3072 (define make-message string-append)
3074 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3076 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3080 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3085 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3089 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3091 Don't use the functions coop_mutex_init and
3092 coop_condition_variable_init. They will change.
3094 Use scm_mutex_init and scm_cond_init instead.
3096 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3097 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3098 COND, as `scm_cond_wait' does, but it also bounds the duration
3099 of the wait. If COND has not been signaled before time ABSTIME,
3100 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3101 returns the error code `ETIMEDOUT'.
3103 The ABSTIME parameter specifies an absolute time, with the same
3104 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3105 to 00:00:00 GMT, January 1, 1970.
3107 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3108 `scm_cond_broadcast' restarts all the threads that are waiting
3109 on the condition variable COND. Nothing happens if no threads are
3112 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3113 `scm_key_create' allocates a new TSD key. The key is stored in
3114 the location pointed to by KEY. There is no limit on the number
3115 of keys allocated at a given time. The value initially associated
3116 with the returned key is `NULL' in all currently executing threads.
3118 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3119 function associated with the key. When a thread terminates,
3120 DESTR_FUNCTION is called on the value associated with the key in
3121 that thread. The DESTR_FUNCTION is not called if a key is deleted
3122 with `scm_key_delete' or a value is changed with
3123 `scm_setspecific'. The order in which destructor functions are
3124 called at thread termination time is unspecified.
3126 Destructors are not yet implemented.
3128 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3129 `scm_setspecific' changes the value associated with KEY in the
3130 calling thread, storing the given POINTER instead.
3132 ** New function: scm_getspecific (scm_key_t KEY)
3133 `scm_getspecific' returns the value currently associated with
3134 KEY in the calling thread.
3136 ** New function: scm_key_delete (scm_key_t KEY)
3137 `scm_key_delete' deallocates a TSD key. It does not check
3138 whether non-`NULL' values are associated with that key in the
3139 currently executing threads, nor call the destructor function
3140 associated with the key.
3142 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3144 Initialize a C level hook HOOK with associated HOOK_DATA and type
3145 TYPE. (See scm_c_hook_run ().)
3147 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3149 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3150 is true, add it last, otherwise first. The same FUNC can be added
3151 multiple times if FUNC_DATA differ and vice versa.
3153 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3155 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3156 function is only removed if both FUNC and FUNC_DATA matches.
3158 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3160 Run hook HOOK passing DATA to the hook functions.
3162 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3163 returned is undefined.
3165 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3166 returns a non-NULL value. This value is returned as the result of
3167 scm_c_hook_run. If all functions return NULL, NULL is returned.
3169 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3170 returns a NULL value, and NULL is returned. If all functions returns
3171 a non-NULL value, the last value is returned.
3173 ** New C level GC hooks
3175 Five new C level hooks has been added to the garbage collector.
3177 scm_before_gc_c_hook
3180 are run before locking and after unlocking the heap. The system is
3181 thus in a mode where evaluation can take place. (Except that
3182 scm_before_gc_c_hook must not allocate new cells.)
3184 scm_before_mark_c_hook
3185 scm_before_sweep_c_hook
3186 scm_after_sweep_c_hook
3188 are run when the heap is locked. These are intended for extension of
3189 the GC in a modular fashion. Examples are the weaks and guardians
3192 ** Way for application to customize GC parameters
3194 The application can set up other default values for the GC heap
3195 allocation parameters
3197 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3198 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3199 GUILE_MAX_SEGMENT_SIZE,
3203 scm_default_init_heap_size_1, scm_default_min_yield_1,
3204 scm_default_init_heap_size_2, scm_default_min_yield_2,
3205 scm_default_max_segment_size
3207 respectively before callong scm_boot_guile.
3209 (See entry "New environment variables ..." in section
3210 "Changes to the stand-alone interpreter" above.)
3212 ** scm_protect_object/scm_unprotect_object now nest
3214 This means that you can call scm_protect_object multiple times on an
3215 object and count on the object being protected until
3216 scm_unprotect_object has been call the same number of times.
3218 The functions also have better time complexity.
3220 Still, it is usually possible to structure the application in a way
3221 that you don't need to use these functions. For example, if you use a
3222 protected standard Guile list to keep track of live objects rather
3223 than some custom data type, objects will die a natural death when they
3224 are no longer needed.
3226 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3228 Guile does not provide the float representation for inexact real numbers any
3229 more. Now, only doubles are used to represent inexact real numbers. Further,
3230 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3231 and scm_tc16_complex, respectively.
3233 ** Removed deprecated type scm_smobfuns
3235 ** Removed deprecated function scm_newsmob
3237 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3239 There is an ongoing discussion among the developers whether to
3240 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3241 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3242 until this issue has been settled.
3244 ** Removed deprecated type tag scm_tc16_kw
3246 ** Added type tag scm_tc16_keyword
3248 (This was introduced already in release 1.3.4 but was not documented
3251 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3253 * Changes to system call interfaces:
3255 ** The "select" procedure now tests port buffers for the ability to
3256 provide input or accept output. Previously only the underlying file
3257 descriptors were checked.
3259 ** New variable PIPE_BUF: the maximum number of bytes that can be
3260 atomically written to a pipe.
3262 ** If a facility is not available on the system when Guile is
3263 compiled, the corresponding primitive procedure will not be defined.
3264 Previously it would have been defined but would throw a system-error
3265 exception if called. Exception handlers which catch this case may
3266 need minor modification: an error will be thrown with key
3267 'unbound-variable instead of 'system-error. Alternatively it's
3268 now possible to use `defined?' to check whether the facility is
3271 ** Procedures which depend on the timezone should now give the correct
3272 result on systems which cache the TZ environment variable, even if TZ
3273 is changed without calling tzset.
3275 * Changes to the networking interfaces:
3277 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3278 long integers between network and host format. For now, it's not
3279 particularly convenient to do this kind of thing, but consider:
3281 (define write-network-long
3282 (lambda (value port)
3283 (let ((v (make-uniform-vector 1 1 0)))
3284 (uniform-vector-set! v 0 (htonl value))
3285 (uniform-vector-write v port))))
3287 (define read-network-long
3289 (let ((v (make-uniform-vector 1 1 0)))
3290 (uniform-vector-read! v port)
3291 (ntohl (uniform-vector-ref v 0)))))
3293 ** If inet-aton fails, it now throws an error with key 'misc-error
3294 instead of 'system-error, since errno is not relevant.
3296 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3297 specific keys instead of 'system-error. The latter is inappropriate
3298 since errno will not have been set. The keys are:
3299 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3301 ** sethostent, setnetent, setprotoent, setservent: now take an
3302 optional argument STAYOPEN, which specifies whether the database
3303 remains open after a database entry is accessed randomly (e.g., using
3304 gethostbyname for the hosts database.) The default is #f. Previously
3308 Changes since Guile 1.3.2:
3310 * Changes to the stand-alone interpreter
3314 An initial version of the Guile debugger written by Chris Hanson has
3315 been added. The debugger is still under development but is included
3316 in the distribution anyway since it is already quite useful.
3322 after an error to enter the debugger. Type `help' inside the debugger
3323 for a description of available commands.
3325 If you prefer to have stack frames numbered and printed in
3326 anti-chronological order and prefer up in the stack to be down on the
3327 screen as is the case in gdb, you can put
3329 (debug-enable 'backwards)
3331 in your .guile startup file. (However, this means that Guile can't
3332 use indentation to indicate stack level.)
3334 The debugger is autoloaded into Guile at the first use.
3336 ** Further enhancements to backtraces
3338 There is a new debug option `width' which controls the maximum width
3339 on the screen of printed stack frames. Fancy printing parameters
3340 ("level" and "length" as in Common LISP) are adaptively adjusted for
3341 each stack frame to give maximum information while still fitting
3342 within the bounds. If the stack frame can't be made to fit by
3343 adjusting parameters, it is simply cut off at the end. This is marked
3346 ** Some modules are now only loaded when the repl is started
3348 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3349 regex) are now loaded into (guile-user) only if the repl has been
3350 started. The effect is that the startup time for scripts has been
3351 reduced to 30% of what it was previously.
3353 Correctly written scripts load the modules they require at the top of
3354 the file and should not be affected by this change.
3356 ** Hooks are now represented as smobs
3358 * Changes to Scheme functions and syntax
3360 ** Readline support has changed again.
3362 The old (readline-activator) module is gone. Use (ice-9 readline)
3363 instead, which now contains all readline functionality. So the code
3364 to activate readline is now
3366 (use-modules (ice-9 readline))
3369 This should work at any time, including from the guile prompt.
3371 To avoid confusion about the terms of Guile's license, please only
3372 enable readline for your personal use; please don't make it the
3373 default for others. Here is why we make this rather odd-sounding
3376 Guile is normally licensed under a weakened form of the GNU General
3377 Public License, which allows you to link code with Guile without
3378 placing that code under the GPL. This exception is important to some
3381 However, since readline is distributed under the GNU General Public
3382 License, when you link Guile with readline, either statically or
3383 dynamically, you effectively change Guile's license to the strict GPL.
3384 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3385 which are normally permitted become forbidden. This is a rather
3386 non-obvious consequence of the licensing terms.
3388 So, to make sure things remain clear, please let people choose for
3389 themselves whether to link GPL'd libraries like readline with Guile.
3391 ** regexp-substitute/global has changed slightly, but incompatibly.
3393 If you include a function in the item list, the string of the match
3394 object it receives is the same string passed to
3395 regexp-substitute/global, not some suffix of that string.
3396 Correspondingly, the match's positions are relative to the entire
3397 string, not the suffix.
3399 If the regexp can match the empty string, the way matches are chosen
3400 from the string has changed. regexp-substitute/global recognizes the
3401 same set of matches that list-matches does; see below.
3403 ** New function: list-matches REGEXP STRING [FLAGS]
3405 Return a list of match objects, one for every non-overlapping, maximal
3406 match of REGEXP in STRING. The matches appear in left-to-right order.
3407 list-matches only reports matches of the empty string if there are no
3408 other matches which begin on, end at, or include the empty match's
3411 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3413 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3415 For each match of REGEXP in STRING, apply PROC to the match object,
3416 and the last value PROC returned, or INIT for the first call. Return
3417 the last value returned by PROC. We apply PROC to the matches as they
3418 appear from left to right.
3420 This function recognizes matches according to the same criteria as
3423 Thus, you could define list-matches like this:
3425 (define (list-matches regexp string . flags)
3426 (reverse! (apply fold-matches regexp string '() cons flags)))
3428 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3432 *** New function: hook? OBJ
3434 Return #t if OBJ is a hook, otherwise #f.
3436 *** New function: make-hook-with-name NAME [ARITY]
3438 Return a hook with name NAME and arity ARITY. The default value for
3439 ARITY is 0. The only effect of NAME is that it will appear when the
3440 hook object is printed to ease debugging.
3442 *** New function: hook-empty? HOOK
3444 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3446 *** New function: hook->list HOOK
3448 Return a list of the procedures that are called when run-hook is
3451 ** `map' signals an error if its argument lists are not all the same length.
3453 This is the behavior required by R5RS, so this change is really a bug
3454 fix. But it seems to affect a lot of people's code, so we're
3455 mentioning it here anyway.
3457 ** Print-state handling has been made more transparent
3459 Under certain circumstances, ports are represented as a port with an
3460 associated print state. Earlier, this pair was represented as a pair
3461 (see "Some magic has been added to the printer" below). It is now
3462 indistinguishable (almost; see `get-print-state') from a port on the
3465 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3467 Return a new port with the associated print state PRINT-STATE.
3469 *** New function: get-print-state OUTPUT-PORT
3471 Return the print state associated with this port if it exists,
3472 otherwise return #f.
3474 *** New function: directory-stream? OBJECT
3476 Returns true iff OBJECT is a directory stream --- the sort of object
3477 returned by `opendir'.
3479 ** New function: using-readline?
3481 Return #t if readline is in use in the current repl.
3483 ** structs will be removed in 1.4
3485 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3486 and use GOOPS objects as the fundamental record type.
3488 * Changes to the scm_ interface
3490 ** structs will be removed in 1.4
3492 The entire current struct interface (struct.c, struct.h) will be
3493 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3494 GOOPS objects as the fundamental record type.
3496 ** The internal representation of subr's has changed
3498 Instead of giving a hint to the subr name, the CAR field of the subr
3499 now contains an index to a subr entry in scm_subr_table.
3501 *** New variable: scm_subr_table
3503 An array of subr entries. A subr entry contains the name, properties
3504 and documentation associated with the subr. The properties and
3505 documentation slots are not yet used.
3507 ** A new scheme for "forwarding" calls to a builtin to a generic function
3509 It is now possible to extend the functionality of some Guile
3510 primitives by letting them defer a call to a GOOPS generic function on
3511 argument mismatch. This means that there is no loss of efficiency in
3516 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3517 (define-method + ((x <string>) (y <string>))
3518 (string-append x y))
3520 + will still be as efficient as usual in numerical calculations, but
3521 can also be used for concatenating strings.
3523 Who will be the first one to extend Guile's numerical tower to
3524 rationals? :) [OK, there a few other things to fix before this can
3525 be made in a clean way.]
3527 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3529 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3531 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3533 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3534 a variable GENERIC which can be used by the dispatch macros below.
3536 [This is experimental code which may change soon.]
3538 *** New macros for forwarding control to a generic on arg type error
3540 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3542 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3544 These correspond to the scm_wta function call, and have the same
3545 behaviour until the user has called the GOOPS primitive
3546 `enable-primitive-generic!'. After that, these macros will apply the
3547 generic function GENERIC to the argument(s) instead of calling
3550 [This is experimental code which may change soon.]
3552 *** New macros for argument testing with generic dispatch
3554 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3556 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3558 These correspond to the SCM_ASSERT macro, but will defer control to
3559 GENERIC on error after `enable-primitive-generic!' has been called.
3561 [This is experimental code which may change soon.]
3563 ** New function: SCM scm_eval_body (SCM body, SCM env)
3565 Evaluates the body of a special form.
3567 ** The internal representation of struct's has changed
3569 Previously, four slots were allocated for the procedure(s) of entities
3570 and operators. The motivation for this representation had to do with
3571 the structure of the evaluator, the wish to support tail-recursive
3572 generic functions, and efficiency. Since the generic function
3573 dispatch mechanism has changed, there is no longer a need for such an
3574 expensive representation, and the representation has been simplified.
3576 This should not make any difference for most users.
3578 ** GOOPS support has been cleaned up.
3580 Some code has been moved from eval.c to objects.c and code in both of
3581 these compilation units has been cleaned up and better structured.
3583 *** New functions for applying generic functions
3585 New function: SCM scm_apply_generic (GENERIC, ARGS)
3586 New function: SCM scm_call_generic_0 (GENERIC)
3587 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3588 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3589 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3591 ** Deprecated function: scm_make_named_hook
3593 It is now replaced by:
3595 ** New function: SCM scm_create_hook (const char *name, int arity)
3597 Creates a hook in the same way as make-hook above but also
3598 binds a variable named NAME to it.
3600 This is the typical way of creating a hook from C code.
3602 Currently, the variable is created in the "current" module.
3603 This might change when we get the new module system.
3605 [The behaviour is identical to scm_make_named_hook.]
3609 Changes since Guile 1.3:
3611 * Changes to mailing lists
3613 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3615 See the README file to find current addresses for all the Guile
3618 * Changes to the distribution
3620 ** Readline support is no longer included with Guile by default.
3622 Based on the different license terms of Guile and Readline, we
3623 concluded that Guile should not *by default* cause the linking of
3624 Readline into an application program. Readline support is now offered
3625 as a separate module, which is linked into an application only when
3626 you explicitly specify it.
3628 Although Guile is GNU software, its distribution terms add a special
3629 exception to the usual GNU General Public License (GPL). Guile's
3630 license includes a clause that allows you to link Guile with non-free
3631 programs. We add this exception so as not to put Guile at a
3632 disadvantage vis-a-vis other extensibility packages that support other
3635 In contrast, the GNU Readline library is distributed under the GNU
3636 General Public License pure and simple. This means that you may not
3637 link Readline, even dynamically, into an application unless it is
3638 distributed under a free software license that is compatible the GPL.
3640 Because of this difference in distribution terms, an application that
3641 can use Guile may not be able to use Readline. Now users will be
3642 explicitly offered two independent decisions about the use of these
3645 You can activate the readline support by issuing
3647 (use-modules (readline-activator))
3650 from your ".guile" file, for example.
3652 * Changes to the stand-alone interpreter
3654 ** All builtins now print as primitives.
3655 Previously builtin procedures not belonging to the fundamental subr
3656 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3657 Now, they print as #<primitive-procedure NAME>.
3659 ** Backtraces slightly more intelligible.
3660 gsubr-apply and macro transformer application frames no longer appear
3663 * Changes to Scheme functions and syntax
3665 ** Guile now correctly handles internal defines by rewriting them into
3666 their equivalent letrec. Previously, internal defines would
3667 incrementally add to the innermost environment, without checking
3668 whether the restrictions specified in RnRS were met. This lead to the
3669 correct behaviour when these restriction actually were met, but didn't
3670 catch all illegal uses. Such an illegal use could lead to crashes of
3671 the Guile interpreter or or other unwanted results. An example of
3672 incorrect internal defines that made Guile behave erratically:
3684 The problem with this example is that the definition of `c' uses the
3685 value of `b' directly. This confuses the meoization machine of Guile
3686 so that the second call of `b' (this time in a larger environment that
3687 also contains bindings for `c' and `d') refers to the binding of `c'
3688 instead of `a'. You could also make Guile crash with a variation on
3693 (define (b flag) (if flag a 1))
3694 (define c (1+ (b flag)))
3702 From now on, Guile will issue an `Unbound variable: b' error message
3707 A hook contains a list of functions which should be called on
3708 particular occasions in an existing program. Hooks are used for
3711 A window manager might have a hook before-window-map-hook. The window
3712 manager uses the function run-hooks to call all functions stored in
3713 before-window-map-hook each time a window is mapped. The user can
3714 store functions in the hook using add-hook!.
3716 In Guile, hooks are first class objects.
3718 *** New function: make-hook [N_ARGS]
3720 Return a hook for hook functions which can take N_ARGS arguments.
3721 The default value for N_ARGS is 0.
3723 (See also scm_make_named_hook below.)
3725 *** New function: add-hook! HOOK PROC [APPEND_P]
3727 Put PROC at the beginning of the list of functions stored in HOOK.
3728 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3730 PROC must be able to take the number of arguments specified when the
3733 If PROC already exists in HOOK, then remove it first.
3735 *** New function: remove-hook! HOOK PROC
3737 Remove PROC from the list of functions in HOOK.
3739 *** New function: reset-hook! HOOK
3741 Clear the list of hook functions stored in HOOK.
3743 *** New function: run-hook HOOK ARG1 ...
3745 Run all hook functions stored in HOOK with arguments ARG1 ... .
3746 The number of arguments supplied must correspond to the number given
3747 when the hook was created.
3749 ** The function `dynamic-link' now takes optional keyword arguments.
3750 The only keyword argument that is currently defined is `:global
3751 BOOL'. With it, you can control whether the shared library will be
3752 linked in global mode or not. In global mode, the symbols from the
3753 linked library can be used to resolve references from other
3754 dynamically linked libraries. In non-global mode, the linked
3755 library is essentially invisible and can only be accessed via
3756 `dynamic-func', etc. The default is now to link in global mode.
3757 Previously, the default has been non-global mode.
3759 The `#:global' keyword is only effective on platforms that support
3760 the dlopen family of functions.
3762 ** New function `provided?'
3764 - Function: provided? FEATURE
3765 Return true iff FEATURE is supported by this installation of
3766 Guile. FEATURE must be a symbol naming a feature; the global
3767 variable `*features*' is a list of available features.
3769 ** Changes to the module (ice-9 expect):
3771 *** The expect-strings macro now matches `$' in a regular expression
3772 only at a line-break or end-of-file by default. Previously it would
3773 match the end of the string accumulated so far. The old behaviour
3774 can be obtained by setting the variable `expect-strings-exec-flags'
3777 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3778 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3779 in a regular expression will still match before a line-break or
3780 end-of-file. The default is `regexp/noteol'.
3782 *** The expect-strings macro now uses a variable
3783 `expect-strings-compile-flags' for the flags to be supplied to
3784 `make-regexp'. The default is `regexp/newline', which was previously
3787 *** The expect macro now supplies two arguments to a match procedure:
3788 the current accumulated string and a flag to indicate whether
3789 end-of-file has been reached. Previously only the string was supplied.
3790 If end-of-file is reached, the match procedure will be called an
3791 additional time with the same accumulated string as the previous call
3792 but with the flag set.
3794 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3796 This code, and the documentation for it that appears here, was
3797 borrowed from SLIB, with minor adaptations for Guile.
3799 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3800 An almost complete implementation of Common LISP format description
3801 according to the CL reference book `Common LISP' from Guy L.
3802 Steele, Digital Press. Backward compatible to most of the
3803 available Scheme format implementations.
3805 Returns `#t', `#f' or a string; has side effect of printing
3806 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3807 to the current output port and `#t' is returned. If DESTINATION
3808 is `#f', a formatted string is returned as the result of the call.
3809 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3810 format string; FORMAT-STRING is then the first argument and the
3811 output is returned as a string. If DESTINATION is a number, the
3812 output is to the current error port if available by the
3813 implementation. Otherwise DESTINATION must be an output port and
3816 FORMAT-STRING must be a string. In case of a formatting error
3817 format returns `#f' and prints a message on the current output or
3818 error port. Characters are output as if the string were output by
3819 the `display' function with the exception of those prefixed by a
3820 tilde (~). For a detailed description of the FORMAT-STRING syntax
3821 please consult a Common LISP format reference manual. For a test
3822 suite to verify this format implementation load `formatst.scm'.
3823 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3825 Note: `format' is not reentrant, i.e. only one `format'-call may
3826 be executed at a time.
3829 *** Format Specification (Format version 3.0)
3831 Please consult a Common LISP format reference manual for a detailed
3832 description of the format string syntax. For a demonstration of the
3833 implemented directives see `formatst.scm'.
3835 This implementation supports directive parameters and modifiers (`:'
3836 and `@' characters). Multiple parameters must be separated by a comma
3837 (`,'). Parameters can be numerical parameters (positive or negative),
3838 character parameters (prefixed by a quote character (`''), variable
3839 parameters (`v'), number of rest arguments parameter (`#'), empty and
3840 default parameters. Directive characters are case independent. The
3841 general form of a directive is:
3843 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3845 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3847 *** Implemented CL Format Control Directives
3849 Documentation syntax: Uppercase characters represent the
3850 corresponding control directive characters. Lowercase characters
3851 represent control directive parameter descriptions.
3854 Any (print as `display' does).
3858 `~MINCOL,COLINC,MINPAD,PADCHARA'
3862 S-expression (print as `write' does).
3866 `~MINCOL,COLINC,MINPAD,PADCHARS'
3872 print number sign always.
3875 print comma separated.
3877 `~MINCOL,PADCHAR,COMMACHARD'
3883 print number sign always.
3886 print comma separated.
3888 `~MINCOL,PADCHAR,COMMACHARX'
3894 print number sign always.
3897 print comma separated.
3899 `~MINCOL,PADCHAR,COMMACHARO'
3905 print number sign always.
3908 print comma separated.
3910 `~MINCOL,PADCHAR,COMMACHARB'
3915 `~N,MINCOL,PADCHAR,COMMACHARR'
3919 print a number as a Roman numeral.
3922 print a number as an "old fashioned" Roman numeral.
3925 print a number as an ordinal English number.
3928 print a number as a cardinal English number.
3933 prints `y' and `ies'.
3936 as `~P but jumps 1 argument backward.'
3939 as `~@P but jumps 1 argument backward.'
3944 prints a character as the reader can understand it (i.e. `#\'
3948 prints a character as emacs does (eg. `^C' for ASCII 03).
3951 Fixed-format floating-point (prints a flonum like MMM.NNN).
3952 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3954 If the number is positive a plus sign is printed.
3957 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3958 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3960 If the number is positive a plus sign is printed.
3963 General floating-point (prints a flonum either fixed or
3965 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3967 If the number is positive a plus sign is printed.
3970 Dollars floating-point (prints a flonum in fixed with signs
3972 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3974 If the number is positive a plus sign is printed.
3977 A sign is always printed and appears before the padding.
3980 The sign appears before the padding.
3988 print newline if not at the beginning of the output line.
3990 prints `~&' and then N-1 newlines.
3995 print N page separators.
4005 newline is ignored, white space left.
4008 newline is left, white space ignored.
4013 relative tabulation.
4019 Indirection (expects indirect arguments as a list).
4021 extracts indirect arguments from format arguments.
4024 Case conversion (converts by `string-downcase').
4026 converts by `string-capitalize'.
4029 converts by `string-capitalize-first'.
4032 converts by `string-upcase'.
4035 Argument Jumping (jumps 1 argument forward).
4037 jumps N arguments forward.
4040 jumps 1 argument backward.
4043 jumps N arguments backward.
4046 jumps to the 0th argument.
4049 jumps to the Nth argument (beginning from 0)
4051 `~[STR0~;STR1~;...~;STRN~]'
4052 Conditional Expression (numerical clause conditional).
4054 take argument from N.
4057 true test conditional.
4060 if-else-then conditional.
4066 default clause follows.
4069 Iteration (args come from the next argument (a list)).
4071 at most N iterations.
4074 args from next arg (a list of lists).
4077 args from the rest of arguments.
4080 args from the rest args (lists).
4091 aborts if N <= M <= K
4093 *** Not Implemented CL Format Control Directives
4096 print `#f' as an empty list (see below).
4099 print `#f' as an empty list (see below).
4105 (sorry I don't understand its semantics completely)
4107 *** Extended, Replaced and Additional Control Directives
4109 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4110 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4111 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4112 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4113 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4114 COMMAWIDTH is the number of characters between two comma
4118 print a R4RS complex number as `~F~@Fi' with passed parameters for
4122 Pretty print formatting of an argument for scheme code lists.
4128 Flushes the output if format DESTINATION is a port.
4131 Print a `#\space' character
4133 print N `#\space' characters.
4136 Print a `#\tab' character
4138 print N `#\tab' characters.
4141 Takes N as an integer representation for a character. No arguments
4142 are consumed. N is converted to a character by `integer->char'. N
4143 must be a positive decimal number.
4146 Print out readproof. Prints out internal objects represented as
4147 `#<...>' as strings `"#<...>"' so that the format output can always
4148 be processed by `read'.
4151 Print out readproof. Prints out internal objects represented as
4152 `#<...>' as strings `"#<...>"' so that the format output can always
4153 be processed by `read'.
4156 Prints information and a copyright notice on the format
4159 prints format version.
4162 may also print number strings, i.e. passing a number as a string
4163 and format it accordingly.
4165 *** Configuration Variables
4167 The format module exports some configuration variables to suit the
4168 systems and users needs. There should be no modification necessary for
4169 the configuration that comes with Guile. Format detects automatically
4170 if the running scheme system implements floating point numbers and
4173 format:symbol-case-conv
4174 Symbols are converted by `symbol->string' so the case type of the
4175 printed symbols is implementation dependent.
4176 `format:symbol-case-conv' is a one arg closure which is either
4177 `#f' (no conversion), `string-upcase', `string-downcase' or
4178 `string-capitalize'. (default `#f')
4180 format:iobj-case-conv
4181 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4182 implementation internal objects. (default `#f')
4185 The character prefixing the exponent value in `~E' printing.
4188 *** Compatibility With Other Format Implementations
4194 Downward compatible except for padding support and `~A', `~S',
4195 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4196 `printf' padding support which is completely replaced by the CL
4197 `format' padding style.
4200 Downward compatible except for `~', which is not documented
4201 (ignores all characters inside the format string up to a newline
4202 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4203 numerical and variable parameters and `:/@' modifiers in the CL
4207 Downward compatible except for `~A' and `~S' which print in
4208 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4209 directive parameters or modifiers)).
4212 Downward compatible except for an optional destination parameter:
4213 S2C accepts a format call without a destination which returns a
4214 formatted string. This is equivalent to a #f destination in S2C.
4215 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4216 parameters or modifiers)).
4219 ** Changes to string-handling functions.
4221 These functions were added to support the (ice-9 format) module, above.
4223 *** New function: string-upcase STRING
4224 *** New function: string-downcase STRING
4226 These are non-destructive versions of the existing string-upcase! and
4227 string-downcase! functions.
4229 *** New function: string-capitalize! STRING
4230 *** New function: string-capitalize STRING
4232 These functions convert the first letter of each word in the string to
4235 (string-capitalize "howdy there")
4238 As with the other functions, string-capitalize! modifies the string in
4239 place, while string-capitalize returns a modified copy of its argument.
4241 *** New function: string-ci->symbol STRING
4243 Return a symbol whose name is STRING, but having the same case as if
4244 the symbol had be read by `read'.
4246 Guile can be configured to be sensitive or insensitive to case
4247 differences in Scheme identifiers. If Guile is case-insensitive, all
4248 symbols are converted to lower case on input. The `string-ci->symbol'
4249 function returns a symbol whose name in STRING, transformed as Guile
4250 would if STRING were input.
4252 *** New function: substring-move! STRING1 START END STRING2 START
4254 Copy the substring of STRING1 from START (inclusive) to END
4255 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4256 string, and the source and destination areas may overlap; in all
4257 cases, the function behaves as if all the characters were copied
4260 *** Extended functions: substring-move-left! substring-move-right!
4262 These functions now correctly copy arbitrarily overlapping substrings;
4263 they are both synonyms for substring-move!.
4266 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4268 getopt-long is a function for parsing command-line arguments in a
4269 manner consistent with other GNU programs.
4271 (getopt-long ARGS GRAMMAR)
4272 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4274 ARGS should be a list of strings. Its first element should be the
4275 name of the program; subsequent elements should be the arguments
4276 that were passed to the program on the command line. The
4277 `program-arguments' procedure returns a list of this form.
4279 GRAMMAR is a list of the form:
4280 ((OPTION (PROPERTY VALUE) ...) ...)
4282 Each OPTION should be a symbol. `getopt-long' will accept a
4283 command-line option named `--OPTION'.
4284 Each option can have the following (PROPERTY VALUE) pairs:
4286 (single-char CHAR) --- Accept `-CHAR' as a single-character
4287 equivalent to `--OPTION'. This is how to specify traditional
4289 (required? BOOL) --- If BOOL is true, the option is required.
4290 getopt-long will raise an error if it is not found in ARGS.
4291 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4292 it is #f, it does not; and if it is the symbol
4293 `optional', the option may appear in ARGS with or
4295 (predicate FUNC) --- If the option accepts a value (i.e. you
4296 specified `(value #t)' for this option), then getopt
4297 will apply FUNC to the value, and throw an exception
4298 if it returns #f. FUNC should be a procedure which
4299 accepts a string and returns a boolean value; you may
4300 need to use quasiquotes to get it into GRAMMAR.
4302 The (PROPERTY VALUE) pairs may occur in any order, but each
4303 property may occur only once. By default, options do not have
4304 single-character equivalents, are not required, and do not take
4307 In ARGS, single-character options may be combined, in the usual
4308 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4309 accepts values, then it must be the last option in the
4310 combination; the value is the next argument. So, for example, using
4311 the following grammar:
4312 ((apples (single-char #\a))
4313 (blimps (single-char #\b) (value #t))
4314 (catalexis (single-char #\c) (value #t)))
4315 the following argument lists would be acceptable:
4316 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4317 for "blimps" and "catalexis")
4318 ("-ab" "bang" "-c" "couth") (same)
4319 ("-ac" "couth" "-b" "bang") (same)
4320 ("-abc" "couth" "bang") (an error, since `-b' is not the
4321 last option in its combination)
4323 If an option's value is optional, then `getopt-long' decides
4324 whether it has a value by looking at what follows it in ARGS. If
4325 the next element is a string, and it does not appear to be an
4326 option itself, then that string is the option's value.
4328 The value of a long option can appear as the next element in ARGS,
4329 or it can follow the option name, separated by an `=' character.
4330 Thus, using the same grammar as above, the following argument lists
4332 ("--apples" "Braeburn" "--blimps" "Goodyear")
4333 ("--apples=Braeburn" "--blimps" "Goodyear")
4334 ("--blimps" "Goodyear" "--apples=Braeburn")
4336 If the option "--" appears in ARGS, argument parsing stops there;
4337 subsequent arguments are returned as ordinary arguments, even if
4338 they resemble options. So, in the argument list:
4339 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4340 `getopt-long' will recognize the `apples' option as having the
4341 value "Granny Smith", but it will not recognize the `blimp'
4342 option; it will return the strings "--blimp" and "Goodyear" as
4343 ordinary argument strings.
4345 The `getopt-long' function returns the parsed argument list as an
4346 assocation list, mapping option names --- the symbols from GRAMMAR
4347 --- onto their values, or #t if the option does not accept a value.
4348 Unused options do not appear in the alist.
4350 All arguments that are not the value of any option are returned
4351 as a list, associated with the empty list.
4353 `getopt-long' throws an exception if:
4354 - it finds an unrecognized option in ARGS
4355 - a required option is omitted
4356 - an option that requires an argument doesn't get one
4357 - an option that doesn't accept an argument does get one (this can
4358 only happen using the long option `--opt=value' syntax)
4359 - an option predicate fails
4364 `((lockfile-dir (required? #t)
4367 (predicate ,file-is-directory?))
4368 (verbose (required? #f)
4371 (x-includes (single-char #\x))
4372 (rnet-server (single-char #\y)
4373 (predicate ,string?))))
4375 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4376 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4378 => ((() "foo1" "-fred" "foo2" "foo3")
4379 (rnet-server . "lamprod")
4380 (x-includes . "/usr/include")
4381 (lockfile-dir . "/tmp")
4384 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4386 It will be removed in a few releases.
4388 ** New syntax: lambda*
4389 ** New syntax: define*
4390 ** New syntax: define*-public
4391 ** New syntax: defmacro*
4392 ** New syntax: defmacro*-public
4393 Guile now supports optional arguments.
4395 `lambda*', `define*', `define*-public', `defmacro*' and
4396 `defmacro*-public' are identical to the non-* versions except that
4397 they use an extended type of parameter list that has the following BNF
4398 syntax (parentheses are literal, square brackets indicate grouping,
4399 and `*', `+' and `?' have the usual meaning):
4401 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4402 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4403 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4405 ext-var-decl ::= identifier | ( identifier expression )
4407 The semantics are best illustrated with the following documentation
4408 and examples for `lambda*':
4411 lambda extended for optional and keyword arguments
4413 lambda* creates a procedure that takes optional arguments. These
4414 are specified by putting them inside brackets at the end of the
4415 paramater list, but before any dotted rest argument. For example,
4416 (lambda* (a b #&optional c d . e) '())
4417 creates a procedure with fixed arguments a and b, optional arguments c
4418 and d, and rest argument e. If the optional arguments are omitted
4419 in a call, the variables for them are unbound in the procedure. This
4420 can be checked with the bound? macro.
4422 lambda* can also take keyword arguments. For example, a procedure
4424 (lambda* (#&key xyzzy larch) '())
4425 can be called with any of the argument lists (#:xyzzy 11)
4426 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4427 are given as keywords are bound to values.
4429 Optional and keyword arguments can also be given default values
4430 which they take on when they are not present in a call, by giving a
4431 two-item list in place of an optional argument, for example in:
4432 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4433 foo is a fixed argument, bar is an optional argument with default
4434 value 42, and baz is a keyword argument with default value 73.
4435 Default value expressions are not evaluated unless they are needed
4436 and until the procedure is called.
4438 lambda* now supports two more special parameter list keywords.
4440 lambda*-defined procedures now throw an error by default if a
4441 keyword other than one of those specified is found in the actual
4442 passed arguments. However, specifying #&allow-other-keys
4443 immediately after the kyword argument declarations restores the
4444 previous behavior of ignoring unknown keywords. lambda* also now
4445 guarantees that if the same keyword is passed more than once, the
4446 last one passed is the one that takes effect. For example,
4447 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4448 #:heads 37 #:tails 42 #:heads 99)
4449 would result in (99 47) being displayed.
4451 #&rest is also now provided as a synonym for the dotted syntax rest
4452 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4453 all respects to lambda*. This is provided for more similarity to DSSSL,
4454 MIT-Scheme and Kawa among others, as well as for refugees from other
4457 Further documentation may be found in the optargs.scm file itself.
4459 The optional argument module also exports the macros `let-optional',
4460 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4461 are not documented here because they may be removed in the future, but
4462 full documentation is still available in optargs.scm.
4464 ** New syntax: and-let*
4465 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4467 Syntax: (land* (<clause> ...) <body> ...)
4468 Each <clause> should have one of the following forms:
4469 (<variable> <expression>)
4472 Each <variable> or <bound-variable> should be an identifier. Each
4473 <expression> should be a valid expression. The <body> should be a
4474 possibly empty sequence of expressions, like the <body> of a
4477 Semantics: A LAND* expression is evaluated by evaluating the
4478 <expression> or <bound-variable> of each of the <clause>s from
4479 left to right. The value of the first <expression> or
4480 <bound-variable> that evaluates to a false value is returned; the
4481 remaining <expression>s and <bound-variable>s are not evaluated.
4482 The <body> forms are evaluated iff all the <expression>s and
4483 <bound-variable>s evaluate to true values.
4485 The <expression>s and the <body> are evaluated in an environment
4486 binding each <variable> of the preceding (<variable> <expression>)
4487 clauses to the value of the <expression>. Later bindings
4488 shadow earlier bindings.
4490 Guile's and-let* macro was contributed by Michael Livshin.
4492 ** New sorting functions
4494 *** New function: sorted? SEQUENCE LESS?
4495 Returns `#t' when the sequence argument is in non-decreasing order
4496 according to LESS? (that is, there is no adjacent pair `... x y
4497 ...' for which `(less? y x)').
4499 Returns `#f' when the sequence contains at least one out-of-order
4500 pair. It is an error if the sequence is neither a list nor a
4503 *** New function: merge LIST1 LIST2 LESS?
4504 LIST1 and LIST2 are sorted lists.
4505 Returns the sorted list of all elements in LIST1 and LIST2.
4507 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4508 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4509 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4510 (Here "<" should read "comes before".)
4512 *** New procedure: merge! LIST1 LIST2 LESS?
4513 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4514 the result. If the code is compiled, and LESS? constructs no new
4515 pairs, no pairs at all will be allocated. The first pair of the
4516 result will be either the first pair of LIST1 or the first pair of
4519 *** New function: sort SEQUENCE LESS?
4520 Accepts either a list or a vector, and returns a new sequence
4521 which is sorted. The new sequence is the same type as the input.
4522 Always `(sorted? (sort sequence less?) less?)'. The original
4523 sequence is not altered in any way. The new sequence shares its
4524 elements with the old one; no elements are copied.
4526 *** New procedure: sort! SEQUENCE LESS
4527 Returns its sorted result in the original boxes. No new storage is
4528 allocated at all. Proper usage: (set! slist (sort! slist <))
4530 *** New function: stable-sort SEQUENCE LESS?
4531 Similar to `sort' but stable. That is, if "equal" elements are
4532 ordered a < b in the original sequence, they will have the same order
4535 *** New function: stable-sort! SEQUENCE LESS?
4536 Similar to `sort!' but stable.
4537 Uses temporary storage when sorting vectors.
4539 *** New functions: sort-list, sort-list!
4540 Added for compatibility with scsh.
4542 ** New built-in random number support
4544 *** New function: random N [STATE]
4545 Accepts a positive integer or real N and returns a number of the
4546 same type between zero (inclusive) and N (exclusive). The values
4547 returned have a uniform distribution.
4549 The optional argument STATE must be of the type produced by
4550 `copy-random-state' or `seed->random-state'. It defaults to the value
4551 of the variable `*random-state*'. This object is used to maintain the
4552 state of the pseudo-random-number generator and is altered as a side
4553 effect of the `random' operation.
4555 *** New variable: *random-state*
4556 Holds a data structure that encodes the internal state of the
4557 random-number generator that `random' uses by default. The nature
4558 of this data structure is implementation-dependent. It may be
4559 printed out and successfully read back in, but may or may not
4560 function correctly as a random-number state object in another
4563 *** New function: copy-random-state [STATE]
4564 Returns a new object of type suitable for use as the value of the
4565 variable `*random-state*' and as a second argument to `random'.
4566 If argument STATE is given, a copy of it is returned. Otherwise a
4567 copy of `*random-state*' is returned.
4569 *** New function: seed->random-state SEED
4570 Returns a new object of type suitable for use as the value of the
4571 variable `*random-state*' and as a second argument to `random'.
4572 SEED is a string or a number. A new state is generated and
4573 initialized using SEED.
4575 *** New function: random:uniform [STATE]
4576 Returns an uniformly distributed inexact real random number in the
4577 range between 0 and 1.
4579 *** New procedure: random:solid-sphere! VECT [STATE]
4580 Fills VECT with inexact real random numbers the sum of whose
4581 squares is less than 1.0. Thinking of VECT as coordinates in
4582 space of dimension N = `(vector-length VECT)', the coordinates are
4583 uniformly distributed within the unit N-shere. The sum of the
4584 squares of the numbers is returned. VECT can be either a vector
4585 or a uniform vector of doubles.
4587 *** New procedure: random:hollow-sphere! VECT [STATE]
4588 Fills VECT with inexact real random numbers the sum of whose squares
4589 is equal to 1.0. Thinking of VECT as coordinates in space of
4590 dimension n = `(vector-length VECT)', the coordinates are uniformly
4591 distributed over the surface of the unit n-shere. VECT can be either
4592 a vector or a uniform vector of doubles.
4594 *** New function: random:normal [STATE]
4595 Returns an inexact real in a normal distribution with mean 0 and
4596 standard deviation 1. For a normal distribution with mean M and
4597 standard deviation D use `(+ M (* D (random:normal)))'.
4599 *** New procedure: random:normal-vector! VECT [STATE]
4600 Fills VECT with inexact real random numbers which are independent and
4601 standard normally distributed (i.e., with mean 0 and variance 1).
4602 VECT can be either a vector or a uniform vector of doubles.
4604 *** New function: random:exp STATE
4605 Returns an inexact real in an exponential distribution with mean 1.
4606 For an exponential distribution with mean U use (* U (random:exp)).
4608 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4610 These functions now operate on numbers in the range of a C unsigned
4613 These functions used to operate on numbers in the range of a C signed
4614 long; however, this seems inappropriate, because Guile integers don't
4617 ** New function: make-guardian
4618 This is an implementation of guardians as described in
4619 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4620 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4621 Programming Language Design and Implementation, June 1993
4622 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4624 ** New functions: delq1!, delv1!, delete1!
4625 These procedures behave similar to delq! and friends but delete only
4626 one object if at all.
4628 ** New function: unread-string STRING PORT
4629 Unread STRING to PORT, that is, push it back onto the port so that
4630 next read operation will work on the pushed back characters.
4632 ** unread-char can now be called multiple times
4633 If unread-char is called multiple times, the unread characters will be
4634 read again in last-in first-out order.
4636 ** the procedures uniform-array-read! and uniform-array-write! now
4637 work on any kind of port, not just ports which are open on a file.
4639 ** Now 'l' in a port mode requests line buffering.
4641 ** The procedure truncate-file now works on string ports as well
4642 as file ports. If the size argument is omitted, the current
4643 file position is used.
4645 ** new procedure: seek PORT/FDES OFFSET WHENCE
4646 The arguments are the same as for the old fseek procedure, but it
4647 works on string ports as well as random-access file ports.
4649 ** the fseek procedure now works on string ports, since it has been
4650 redefined using seek.
4652 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4653 size is not supplied.
4655 ** the newline procedure no longer flushes the port if it's not
4656 line-buffered: previously it did if it was the current output port.
4658 ** open-pipe and close-pipe are no longer primitive procedures, but
4659 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4661 ** the freopen procedure has been removed.
4663 ** new procedure: drain-input PORT
4664 Drains PORT's read buffers (including any pushed-back characters)
4665 and returns the contents as a single string.
4667 ** New function: map-in-order PROC LIST1 LIST2 ...
4668 Version of `map' which guarantees that the procedure is applied to the
4669 lists in serial order.
4671 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4672 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4673 now obsolete and will go away in release 1.5.
4675 ** New syntax: collect BODY1 ...
4676 Version of `begin' which returns a list of the results of the body
4677 forms instead of the result of the last body form. In contrast to
4678 `begin', `collect' allows an empty body.
4680 ** New functions: read-history FILENAME, write-history FILENAME
4681 Read/write command line history from/to file. Returns #t on success
4682 and #f if an error occured.
4684 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4686 These procedures return a list of definitions available in the specified
4687 argument, a relative module reference. In the case of no argument,
4688 `(current-module)' is now consulted for definitions to return, instead
4689 of simply returning #f, the former behavior.
4691 ** The #/ syntax for lists is no longer supported.
4693 Earlier versions of Scheme accepted this syntax, but printed a
4696 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4698 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4701 * Changes to the gh_ interface
4705 Now takes a second argument which is the result array. If this
4706 pointer is NULL, a new array is malloced (the old behaviour).
4708 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4709 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4713 * Changes to the scm_ interface
4715 ** Function: scm_make_named_hook (char* name, int n_args)
4717 Creates a hook in the same way as make-hook above but also
4718 binds a variable named NAME to it.
4720 This is the typical way of creating a hook from C code.
4722 Currently, the variable is created in the "current" module. This
4723 might change when we get the new module system.
4725 ** The smob interface
4727 The interface for creating smobs has changed. For documentation, see
4728 data-rep.info (made from guile-core/doc/data-rep.texi).
4730 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4732 >>> This function will be removed in 1.3.4. <<<
4736 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4737 This function adds a new smob type, named NAME, with instance size
4738 SIZE to the system. The return value is a tag that is used in
4739 creating instances of the type. If SIZE is 0, then no memory will
4740 be allocated when instances of the smob are created, and nothing
4741 will be freed by the default free function.
4743 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4744 This function sets the smob marking procedure for the smob type
4745 specified by the tag TC. TC is the tag returned by
4746 `scm_make_smob_type'.
4748 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4749 This function sets the smob freeing procedure for the smob type
4750 specified by the tag TC. TC is the tag returned by
4751 `scm_make_smob_type'.
4753 *** Function: void scm_set_smob_print (tc, print)
4755 - Function: void scm_set_smob_print (long tc,
4756 scm_sizet (*print) (SCM,
4760 This function sets the smob printing procedure for the smob type
4761 specified by the tag TC. TC is the tag returned by
4762 `scm_make_smob_type'.
4764 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4765 This function sets the smob equality-testing predicate for the
4766 smob type specified by the tag TC. TC is the tag returned by
4767 `scm_make_smob_type'.
4769 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4770 Make VALUE contain a smob instance of the type with type code TC and
4771 smob data DATA. VALUE must be previously declared as C type `SCM'.
4773 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4774 This macro expands to a block of code that creates a smob instance
4775 of the type with type code TC and smob data DATA, and returns that
4776 `SCM' value. It should be the last piece of code in a block.
4778 ** The interfaces for using I/O ports and implementing port types
4779 (ptobs) have changed significantly. The new interface is based on
4780 shared access to buffers and a new set of ptob procedures.
4782 *** scm_newptob has been removed
4786 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4788 - Function: SCM scm_make_port_type (char *type_name,
4789 int (*fill_buffer) (SCM port),
4790 void (*write_flush) (SCM port));
4792 Similarly to the new smob interface, there is a set of function
4793 setters by which the user can customize the behaviour of his port
4794 type. See ports.h (scm_set_port_XXX).
4796 ** scm_strport_to_string: New function: creates a new string from
4797 a string port's buffer.
4799 ** Plug in interface for random number generators
4800 The variable `scm_the_rng' in random.c contains a value and three
4801 function pointers which together define the current random number
4802 generator being used by the Scheme level interface and the random
4803 number library functions.
4805 The user is free to replace the default generator with the generator
4808 *** Variable: size_t scm_the_rng.rstate_size
4809 The size of the random state type used by the current RNG
4812 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4813 Given the random STATE, return 32 random bits.
4815 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4816 Seed random state STATE using string S of length N.
4818 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4819 Given random state STATE, return a malloced copy.
4822 The default RNG is the MWC (Multiply With Carry) random number
4823 generator described by George Marsaglia at the Department of
4824 Statistics and Supercomputer Computations Research Institute, The
4825 Florida State University (http://stat.fsu.edu/~geo).
4827 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4828 passes all tests in the DIEHARD test suite
4829 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4830 costs one multiply and one add on platforms which either supports long
4831 longs (gcc does this on most systems) or have 64 bit longs. The cost
4832 is four multiply on other systems but this can be optimized by writing
4833 scm_i_uniform32 in assembler.
4835 These functions are provided through the scm_the_rng interface for use
4836 by libguile and the application.
4838 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4839 Given the random STATE, return 32 random bits.
4840 Don't use this function directly. Instead go through the plugin
4841 interface (see "Plug in interface" above).
4843 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4844 Initialize STATE using SEED of length N.
4846 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4847 Return a malloc:ed copy of STATE. This function can easily be re-used
4848 in the interfaces to other RNGs.
4850 ** Random number library functions
4851 These functions use the current RNG through the scm_the_rng interface.
4852 It might be a good idea to use these functions from your C code so
4853 that only one random generator is used by all code in your program.
4855 The default random state is stored in:
4857 *** Variable: SCM scm_var_random_state
4858 Contains the vcell of the Scheme variable "*random-state*" which is
4859 used as default state by all random number functions in the Scheme
4864 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4866 *** Function: scm_rstate *scm_c_default_rstate (void)
4867 This is a convenience function which returns the value of
4868 scm_var_random_state. An error message is generated if this value
4869 isn't a random state.
4871 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4872 Make a new random state from the string SEED of length LENGTH.
4874 It is generally not a good idea to use multiple random states in a
4875 program. While subsequent random numbers generated from one random
4876 state are guaranteed to be reasonably independent, there is no such
4877 guarantee for numbers generated from different random states.
4879 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4880 Return 32 random bits.
4882 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4883 Return a sample from the uniform(0,1) distribution.
4885 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4886 Return a sample from the normal(0,1) distribution.
4888 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4889 Return a sample from the exp(1) distribution.
4891 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4892 Return a sample from the discrete uniform(0,M) distribution.
4894 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4895 Return a sample from the discrete uniform(0,M) distribution.
4896 M must be a bignum object. The returned value may be an INUM.
4900 Changes in Guile 1.3 (released Monday, October 19, 1998):
4902 * Changes to the distribution
4904 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4905 To avoid conflicts, programs should name environment variables after
4906 themselves, except when there's a common practice establishing some
4909 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4910 giving the former precedence, and printing a warning message if the
4911 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4913 ** The header files related to multi-byte characters have been removed.
4914 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4915 which referred to these explicitly will probably need to be rewritten,
4916 since the support for the variant string types has been removed; see
4919 ** The header files append.h and sequences.h have been removed. These
4920 files implemented non-R4RS operations which would encourage
4921 non-portable programming style and less easy-to-read code.
4923 * Changes to the stand-alone interpreter
4925 ** New procedures have been added to implement a "batch mode":
4927 *** Function: batch-mode?
4929 Returns a boolean indicating whether the interpreter is in batch
4932 *** Function: set-batch-mode?! ARG
4934 If ARG is true, switches the interpreter to batch mode. The `#f'
4935 case has not been implemented.
4937 ** Guile now provides full command-line editing, when run interactively.
4938 To use this feature, you must have the readline library installed.
4939 The Guile build process will notice it, and automatically include
4942 The readline library is available via anonymous FTP from any GNU
4943 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4945 ** the-last-stack is now a fluid.
4947 * Changes to the procedure for linking libguile with your programs
4949 ** You can now use the `guile-config' utility to build programs that use Guile.
4951 Guile now includes a command-line utility called `guile-config', which
4952 can provide information about how to compile and link programs that
4955 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4956 You should include this command's output on the command line you use
4957 to compile C or C++ code that #includes the Guile header files. It's
4958 usually just a `-I' flag to help the compiler find the Guile headers.
4961 *** `guile-config link' prints any linker flags necessary to link with Guile.
4963 This command writes to its standard output a list of flags which you
4964 must pass to the linker to link your code against the Guile library.
4965 The flags include '-lguile' itself, any other libraries the Guile
4966 library depends upon, and any `-L' flags needed to help the linker
4967 find those libraries.
4969 For example, here is a Makefile rule that builds a program named 'foo'
4970 from the object files ${FOO_OBJECTS}, and links them against Guile:
4973 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4975 Previous Guile releases recommended that you use autoconf to detect
4976 which of a predefined set of libraries were present on your system.
4977 It is more robust to use `guile-config', since it records exactly which
4978 libraries the installed Guile library requires.
4980 This was originally called `build-guile', but was renamed to
4981 `guile-config' before Guile 1.3 was released, to be consistent with
4982 the analogous script for the GTK+ GUI toolkit, which is called
4986 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4988 If you are using the GNU autoconf package to configure your program,
4989 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4990 (described above) and gather the necessary values for use in your
4993 The GUILE_FLAGS macro expands to configure script code which runs the
4994 `guile-config' script, to find out where Guile's header files and
4995 libraries are installed. It sets two variables, marked for
4996 substitution, as by AC_SUBST.
4998 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4999 code that uses Guile header files. This is almost always just a
5002 GUILE_LDFLAGS --- flags to pass to the linker to link a
5003 program against Guile. This includes `-lguile' for the Guile
5004 library itself, any libraries that Guile itself requires (like
5005 -lqthreads), and so on. It may also include a -L flag to tell the
5006 compiler where to find the libraries.
5008 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5009 directory of the Guile distribution. You can copy it into your
5010 package's aclocal.m4 file, and then use it in your configure.in file.
5012 If you are using the `aclocal' program, distributed with GNU automake,
5013 to maintain your aclocal.m4 file, the Guile installation process
5014 installs guile.m4 where aclocal will find it. All you need to do is
5015 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5016 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5020 * Changes to Scheme functions and syntax
5022 ** Multi-byte strings have been removed, as have multi-byte and wide
5023 ports. We felt that these were the wrong approach to
5024 internationalization support.
5026 ** New function: readline [PROMPT]
5027 Read a line from the terminal, and allow the user to edit it,
5028 prompting with PROMPT. READLINE provides a large set of Emacs-like
5029 editing commands, lets the user recall previously typed lines, and
5030 works on almost every kind of terminal, including dumb terminals.
5032 READLINE assumes that the cursor is at the beginning of the line when
5033 it is invoked. Thus, you can't print a prompt yourself, and then call
5034 READLINE; you need to package up your prompt as a string, pass it to
5035 the function, and let READLINE print the prompt itself. This is
5036 because READLINE needs to know the prompt's screen width.
5038 For Guile to provide this function, you must have the readline
5039 library, version 2.1 or later, installed on your system. Readline is
5040 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5041 any GNU mirror site.
5043 See also ADD-HISTORY function.
5045 ** New function: add-history STRING
5046 Add STRING as the most recent line in the history used by the READLINE
5047 command. READLINE does not add lines to the history itself; you must
5048 call ADD-HISTORY to make previous input available to the user.
5050 ** The behavior of the read-line function has changed.
5052 This function now uses standard C library functions to read the line,
5053 for speed. This means that it doesn not respect the value of
5054 scm-line-incrementors; it assumes that lines are delimited with
5057 (Note that this is read-line, the function that reads a line of text
5058 from a port, not readline, the function that reads a line from a
5059 terminal, providing full editing capabilities.)
5061 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5063 This module provides some simple argument parsing. It exports one
5066 Function: getopt-gnu-style ARG-LS
5067 Parse a list of program arguments into an alist of option
5070 Each item in the list of program arguments is examined to see if
5071 it meets the syntax of a GNU long-named option. An argument like
5072 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5073 returned alist, where MUMBLE is a keyword object with the same
5074 name as the argument. An argument like `--MUMBLE=FROB' produces
5075 an element of the form (MUMBLE . FROB), where FROB is a string.
5077 As a special case, the returned alist also contains a pair whose
5078 car is the symbol `rest'. The cdr of this pair is a list
5079 containing all the items in the argument list that are not options
5080 of the form mentioned above.
5082 The argument `--' is treated specially: all items in the argument
5083 list appearing after such an argument are not examined, and are
5084 returned in the special `rest' list.
5086 This function does not parse normal single-character switches.
5087 You will need to parse them out of the `rest' list yourself.
5089 ** The read syntax for byte vectors and short vectors has changed.
5091 Instead of #bytes(...), write #y(...).
5093 Instead of #short(...), write #h(...).
5095 This may seem nutty, but, like the other uniform vectors, byte vectors
5096 and short vectors want to have the same print and read syntax (and,
5097 more basic, want to have read syntax!). Changing the read syntax to
5098 use multiple characters after the hash sign breaks with the
5099 conventions used in R5RS and the conventions used for the other
5100 uniform vectors. It also introduces complexity in the current reader,
5101 both on the C and Scheme levels. (The Right solution is probably to
5102 change the syntax and prototypes for uniform vectors entirely.)
5105 ** The new module (ice-9 session) provides useful interactive functions.
5107 *** New procedure: (apropos REGEXP OPTION ...)
5109 Display a list of top-level variables whose names match REGEXP, and
5110 the modules they are imported from. Each OPTION should be one of the
5113 value --- Show the value of each matching variable.
5114 shadow --- Show bindings shadowed by subsequently imported modules.
5115 full --- Same as both `shadow' and `value'.
5119 guile> (apropos "trace" 'full)
5120 debug: trace #<procedure trace args>
5121 debug: untrace #<procedure untrace args>
5122 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5123 the-scm-module: before-backtrace-hook ()
5124 the-scm-module: backtrace #<primitive-procedure backtrace>
5125 the-scm-module: after-backtrace-hook ()
5126 the-scm-module: has-shown-backtrace-hint? #f
5129 ** There are new functions and syntax for working with macros.
5131 Guile implements macros as a special object type. Any variable whose
5132 top-level binding is a macro object acts as a macro. The macro object
5133 specifies how the expression should be transformed before evaluation.
5135 *** Macro objects now print in a reasonable way, resembling procedures.
5137 *** New function: (macro? OBJ)
5138 True iff OBJ is a macro object.
5140 *** New function: (primitive-macro? OBJ)
5141 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5142 macro transformers, implemented in eval.c rather than Scheme code.
5144 Why do we have this function?
5145 - For symmetry with procedure? and primitive-procedure?,
5146 - to allow custom print procedures to tell whether a macro is
5147 primitive, and display it differently, and
5148 - to allow compilers and user-written evaluators to distinguish
5149 builtin special forms from user-defined ones, which could be
5152 *** New function: (macro-type OBJ)
5153 Return a value indicating what kind of macro OBJ is. Possible return
5156 The symbol `syntax' --- a macro created by procedure->syntax.
5157 The symbol `macro' --- a macro created by procedure->macro.
5158 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5159 The boolean #f --- if OBJ is not a macro object.
5161 *** New function: (macro-name MACRO)
5162 Return the name of the macro object MACRO's procedure, as returned by
5165 *** New function: (macro-transformer MACRO)
5166 Return the transformer procedure for MACRO.
5168 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5170 Specify a new macro expander to use in the current module. Each
5171 MODULE is a module name, with the same meaning as in the `use-modules'
5172 form; each named module's exported bindings are added to the current
5173 top-level environment. TRANSFORMER is an expression evaluated in the
5174 resulting environment which must yield a procedure to use as the
5175 module's eval transformer: every expression evaluated in this module
5176 is passed to this function, and the result passed to the Guile
5179 *** macro-eval! is removed. Use local-eval instead.
5181 ** Some magic has been added to the printer to better handle user
5182 written printing routines (like record printers, closure printers).
5184 The problem is that these user written routines must have access to
5185 the current `print-state' to be able to handle fancy things like
5186 detection of circular references. These print-states have to be
5187 passed to the builtin printing routines (display, write, etc) to
5188 properly continue the print chain.
5190 We didn't want to change all existing print code so that it
5191 explicitly passes thru a print state in addition to a port. Instead,
5192 we extented the possible values that the builtin printing routines
5193 accept as a `port'. In addition to a normal port, they now also take
5194 a pair of a normal port and a print-state. Printing will go to the
5195 port and the print-state will be used to control the detection of
5196 circular references, etc. If the builtin function does not care for a
5197 print-state, it is simply ignored.
5199 User written callbacks are now called with such a pair as their
5200 `port', but because every function now accepts this pair as a PORT
5201 argument, you don't have to worry about that. In fact, it is probably
5202 safest to not check for these pairs.
5204 However, it is sometimes necessary to continue a print chain on a
5205 different port, for example to get a intermediate string
5206 representation of the printed value, mangle that string somehow, and
5207 then to finally print the mangled string. Use the new function
5209 inherit-print-state OLD-PORT NEW-PORT
5211 for this. It constructs a new `port' that prints to NEW-PORT but
5212 inherits the print-state of OLD-PORT.
5214 ** struct-vtable-offset renamed to vtable-offset-user
5216 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5218 ** There is now a third optional argument to make-vtable-vtable
5219 (and fourth to make-struct) when constructing new types (vtables).
5220 This argument initializes field vtable-index-printer of the vtable.
5222 ** The detection of circular references has been extended to structs.
5223 That is, a structure that -- in the process of being printed -- prints
5224 itself does not lead to infinite recursion.
5226 ** There is now some basic support for fluids. Please read
5227 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5228 the following functions and macros:
5230 Function: make-fluid
5232 Create a new fluid object. Fluids are not special variables or
5233 some other extension to the semantics of Scheme, but rather
5234 ordinary Scheme objects. You can store them into variables (that
5235 are still lexically scoped, of course) or into any other place you
5236 like. Every fluid has a initial value of `#f'.
5238 Function: fluid? OBJ
5240 Test whether OBJ is a fluid.
5242 Function: fluid-ref FLUID
5243 Function: fluid-set! FLUID VAL
5245 Access/modify the fluid FLUID. Modifications are only visible
5246 within the current dynamic root (that includes threads).
5248 Function: with-fluids* FLUIDS VALUES THUNK
5250 FLUIDS is a list of fluids and VALUES a corresponding list of
5251 values for these fluids. Before THUNK gets called the values are
5252 installed in the fluids and the old values of the fluids are
5253 saved in the VALUES list. When the flow of control leaves THUNK
5254 or reenters it, the values get swapped again. You might think of
5255 this as a `safe-fluid-excursion'. Note that the VALUES list is
5256 modified by `with-fluids*'.
5258 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5260 The same as `with-fluids*' but with a different syntax. It looks
5261 just like `let', but both FLUID and VALUE are evaluated. Remember,
5262 fluids are not special variables but ordinary objects. FLUID
5263 should evaluate to a fluid.
5265 ** Changes to system call interfaces:
5267 *** close-port, close-input-port and close-output-port now return a
5268 boolean instead of an `unspecified' object. #t means that the port
5269 was successfully closed, while #f means it was already closed. It is
5270 also now possible for these procedures to raise an exception if an
5271 error occurs (some errors from write can be delayed until close.)
5273 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5276 *** the third argument to fcntl is now optional.
5278 *** the first argument to chown can now be a file descriptor or a port.
5280 *** the argument to stat can now be a port.
5282 *** The following new procedures have been added (most use scsh
5285 *** procedure: close PORT/FD
5286 Similar to close-port (*note close-port: Closing Ports.), but also
5287 works on file descriptors. A side effect of closing a file
5288 descriptor is that any ports using that file descriptor are moved
5289 to a different file descriptor and have their revealed counts set
5292 *** procedure: port->fdes PORT
5293 Returns the integer file descriptor underlying PORT. As a side
5294 effect the revealed count of PORT is incremented.
5296 *** procedure: fdes->ports FDES
5297 Returns a list of existing ports which have FDES as an underlying
5298 file descriptor, without changing their revealed counts.
5300 *** procedure: fdes->inport FDES
5301 Returns an existing input port which has FDES as its underlying
5302 file descriptor, if one exists, and increments its revealed count.
5303 Otherwise, returns a new input port with a revealed count of 1.
5305 *** procedure: fdes->outport FDES
5306 Returns an existing output port which has FDES as its underlying
5307 file descriptor, if one exists, and increments its revealed count.
5308 Otherwise, returns a new output port with a revealed count of 1.
5310 The next group of procedures perform a `dup2' system call, if NEWFD
5311 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5312 duplicated can be supplied as an integer or contained in a port. The
5313 type of value returned varies depending on which procedure is used.
5315 All procedures also have the side effect when performing `dup2' that
5316 any ports using NEWFD are moved to a different file descriptor and have
5317 their revealed counts set to zero.
5319 *** procedure: dup->fdes PORT/FD [NEWFD]
5320 Returns an integer file descriptor.
5322 *** procedure: dup->inport PORT/FD [NEWFD]
5323 Returns a new input port using the new file descriptor.
5325 *** procedure: dup->outport PORT/FD [NEWFD]
5326 Returns a new output port using the new file descriptor.
5328 *** procedure: dup PORT/FD [NEWFD]
5329 Returns a new port if PORT/FD is a port, with the same mode as the
5330 supplied port, otherwise returns an integer file descriptor.
5332 *** procedure: dup->port PORT/FD MODE [NEWFD]
5333 Returns a new port using the new file descriptor. MODE supplies a
5334 mode string for the port (*note open-file: File Ports.).
5336 *** procedure: setenv NAME VALUE
5337 Modifies the environment of the current process, which is also the
5338 default environment inherited by child processes.
5340 If VALUE is `#f', then NAME is removed from the environment.
5341 Otherwise, the string NAME=VALUE is added to the environment,
5342 replacing any existing string with name matching NAME.
5344 The return value is unspecified.
5346 *** procedure: truncate-file OBJ SIZE
5347 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5348 can be a string containing a file name or an integer file
5349 descriptor or port open for output on the file. The underlying
5350 system calls are `truncate' and `ftruncate'.
5352 The return value is unspecified.
5354 *** procedure: setvbuf PORT MODE [SIZE]
5355 Set the buffering mode for PORT. MODE can be:
5363 block buffered, using a newly allocated buffer of SIZE bytes.
5364 However if SIZE is zero or unspecified, the port will be made
5367 This procedure should not be used after I/O has been performed with
5370 Ports are usually block buffered by default, with a default buffer
5371 size. Procedures e.g., *Note open-file: File Ports, which accept a
5372 mode string allow `0' to be added to request an unbuffered port.
5374 *** procedure: fsync PORT/FD
5375 Copies any unwritten data for the specified output file descriptor
5376 to disk. If PORT/FD is a port, its buffer is flushed before the
5377 underlying file descriptor is fsync'd. The return value is
5380 *** procedure: open-fdes PATH FLAGS [MODES]
5381 Similar to `open' but returns a file descriptor instead of a port.
5383 *** procedure: execle PATH ENV [ARG] ...
5384 Similar to `execl', but the environment of the new process is
5385 specified by ENV, which must be a list of strings as returned by
5386 the `environ' procedure.
5388 This procedure is currently implemented using the `execve' system
5389 call, but we call it `execle' because of its Scheme calling
5392 *** procedure: strerror ERRNO
5393 Returns the Unix error message corresponding to ERRNO, an integer.
5395 *** procedure: primitive-exit [STATUS]
5396 Terminate the current process without unwinding the Scheme stack.
5397 This is would typically be useful after a fork. The exit status
5398 is STATUS if supplied, otherwise zero.
5400 *** procedure: times
5401 Returns an object with information about real and processor time.
5402 The following procedures accept such an object as an argument and
5403 return a selected component:
5406 The current real time, expressed as time units relative to an
5410 The CPU time units used by the calling process.
5413 The CPU time units used by the system on behalf of the
5417 The CPU time units used by terminated child processes of the
5418 calling process, whose status has been collected (e.g., using
5422 Similarly, the CPU times units used by the system on behalf of
5423 terminated child processes.
5425 ** Removed: list-length
5426 ** Removed: list-append, list-append!
5427 ** Removed: list-reverse, list-reverse!
5429 ** array-map renamed to array-map!
5431 ** serial-array-map renamed to serial-array-map!
5433 ** catch doesn't take #f as first argument any longer
5435 Previously, it was possible to pass #f instead of a key to `catch'.
5436 That would cause `catch' to pass a jump buffer object to the procedure
5437 passed as second argument. The procedure could then use this jump
5438 buffer objekt as an argument to throw.
5440 This mechanism has been removed since its utility doesn't motivate the
5441 extra complexity it introduces.
5443 ** The `#/' notation for lists now provokes a warning message from Guile.
5444 This syntax will be removed from Guile in the near future.
5446 To disable the warning message, set the GUILE_HUSH environment
5447 variable to any non-empty value.
5449 ** The newline character now prints as `#\newline', following the
5450 normal Scheme notation, not `#\nl'.
5452 * Changes to the gh_ interface
5454 ** The gh_enter function now takes care of loading the Guile startup files.
5455 gh_enter works by calling scm_boot_guile; see the remarks below.
5457 ** Function: void gh_write (SCM x)
5459 Write the printed representation of the scheme object x to the current
5460 output port. Corresponds to the scheme level `write'.
5462 ** gh_list_length renamed to gh_length.
5464 ** vector handling routines
5466 Several major changes. In particular, gh_vector() now resembles
5467 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5468 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5469 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5470 vector-related gh_ functions have been implemented.
5472 ** pair and list routines
5474 Implemented several of the R4RS pair and list functions that were
5477 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5479 New function. Converts double arrays back and forth between Scheme
5482 * Changes to the scm_ interface
5484 ** The function scm_boot_guile now takes care of loading the startup files.
5486 Guile's primary initialization function, scm_boot_guile, now takes
5487 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5488 Guile, define the module system, and put together some standard
5489 bindings. It also loads `init.scm', which is intended to hold
5490 site-specific initialization code.
5492 Since Guile cannot operate properly until boot-9.scm is loaded, there
5493 is no reason to separate loading boot-9.scm from Guile's other
5494 initialization processes.
5496 This job used to be done by scm_compile_shell_switches, which didn't
5497 make much sense; in particular, it meant that people using Guile for
5498 non-shell-like applications had to jump through hoops to get Guile
5499 initialized properly.
5501 ** The function scm_compile_shell_switches no longer loads the startup files.
5502 Now, Guile always loads the startup files, whenever it is initialized;
5503 see the notes above for scm_boot_guile and scm_load_startup_files.
5505 ** Function: scm_load_startup_files
5506 This new function takes care of loading Guile's initialization file
5507 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5508 this is always called by the Guile initialization process, it's
5509 probably not too useful to call this yourself, but it's there anyway.
5511 ** The semantics of smob marking have changed slightly.
5513 The smob marking function (the `mark' member of the scm_smobfuns
5514 structure) is no longer responsible for setting the mark bit on the
5515 smob. The generic smob handling code in the garbage collector will
5516 set this bit. The mark function need only ensure that any other
5517 objects the smob refers to get marked.
5519 Note that this change means that the smob's GC8MARK bit is typically
5520 already set upon entry to the mark function. Thus, marking functions
5521 which look like this:
5524 if (SCM_GC8MARKP (ptr))
5526 SCM_SETGC8MARK (ptr);
5527 ... mark objects to which the smob refers ...
5530 are now incorrect, since they will return early, and fail to mark any
5531 other objects the smob refers to. Some code in the Guile library used
5534 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5536 If you have implemented your own I/O port type, by writing the
5537 functions required by the scm_ptobfuns and then calling scm_newptob,
5538 you will need to change your functions slightly.
5540 The functions in a scm_ptobfuns structure now expect the port itself
5541 as their argument; they used to expect the `stream' member of the
5542 port's scm_port_table structure. This allows functions in an
5543 scm_ptobfuns structure to easily access the port's cell (and any flags
5544 it its CAR), and the port's scm_port_table structure.
5546 Guile now passes the I/O port itself as the `port' argument in the
5547 following scm_ptobfuns functions:
5549 int (*free) (SCM port);
5550 int (*fputc) (int, SCM port);
5551 int (*fputs) (char *, SCM port);
5552 scm_sizet (*fwrite) SCM_P ((char *ptr,
5556 int (*fflush) (SCM port);
5557 int (*fgetc) (SCM port);
5558 int (*fclose) (SCM port);
5560 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5563 If you have existing code which defines its own port types, it is easy
5564 to convert your code to the new interface; simply apply SCM_STREAM to
5565 the port argument to yield the value you code used to expect.
5567 Note that since both the port and the stream have the same type in the
5568 C code --- they are both SCM values --- the C compiler will not remind
5569 you if you forget to update your scm_ptobfuns functions.
5572 ** Function: int scm_internal_select (int fds,
5576 struct timeval *timeout);
5578 This is a replacement for the `select' function provided by the OS.
5579 It enables I/O blocking and sleeping to happen for one cooperative
5580 thread without blocking other threads. It also avoids busy-loops in
5581 these situations. It is intended that all I/O blocking and sleeping
5582 will finally go through this function. Currently, this function is
5583 only available on systems providing `gettimeofday' and `select'.
5585 ** Function: SCM scm_internal_stack_catch (SCM tag,
5586 scm_catch_body_t body,
5588 scm_catch_handler_t handler,
5591 A new sibling to the other two C level `catch' functions
5592 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5593 the stack to be saved automatically into the variable `the-last-stack'
5594 (scm_the_last_stack_var) on error. This is necessary if you want to
5595 use advanced error reporting, such as calling scm_display_error and
5596 scm_display_backtrace. (They both take a stack object as argument.)
5598 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5600 scm_catch_handler_t handler,
5603 Spawns a new thread. It does a job similar to
5604 scm_call_with_new_thread but takes arguments more suitable when
5605 spawning threads from application C code.
5607 ** The hook scm_error_callback has been removed. It was originally
5608 intended as a way for the user to install his own error handler. But
5609 that method works badly since it intervenes between throw and catch,
5610 thereby changing the semantics of expressions like (catch #t ...).
5611 The correct way to do it is to use one of the C level catch functions
5612 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5614 ** Removed functions:
5616 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5617 scm_list_reverse, scm_list_reverse_x
5619 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5621 These can be used for pretty list creation from C. The idea is taken
5622 from Erick Gallesio's STk.
5624 ** scm_array_map renamed to scm_array_map_x
5626 ** mbstrings are now removed
5628 This means that the type codes scm_tc7_mb_string and
5629 scm_tc7_mb_substring has been removed.
5631 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5633 Since we no longer support multi-byte strings, these I/O functions
5634 have been simplified, and renamed. Here are their old names, and
5635 their new names and arguments:
5637 scm_gen_putc -> void scm_putc (int c, SCM port);
5638 scm_gen_puts -> void scm_puts (char *s, SCM port);
5639 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5640 scm_gen_getc -> void scm_getc (SCM port);
5643 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5645 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5647 SCM_TYP7S now masks away the bit which distinguishes substrings from
5650 ** scm_catch_body_t: Backward incompatible change!
5652 Body functions to scm_internal_catch and friends do not any longer
5653 take a second argument. This is because it is no longer possible to
5654 pass a #f arg to catch.
5656 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5658 The function scm_protect_object protects its argument from being freed
5659 by the garbage collector. scm_unprotect_object removes that
5662 These functions now nest properly. That is, for every object O, there
5663 is a counter which scm_protect_object(O) increments and
5664 scm_unprotect_object(O) decrements, if the counter is greater than
5665 zero. Every object's counter is zero when it is first created. If an
5666 object's counter is greater than zero, the garbage collector will not
5667 reclaim its storage.
5669 This allows you to use scm_protect_object in your code without
5670 worrying that some other function you call will call
5671 scm_unprotect_object, and allow it to be freed. Assuming that the
5672 functions you call are well-behaved, and unprotect only those objects
5673 they protect, you can follow the same rule and have confidence that
5674 objects will be freed only at appropriate times.
5677 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5679 * Changes to the distribution
5681 ** Nightly snapshots are now available from ftp.red-bean.com.
5682 The old server, ftp.cyclic.com, has been relinquished to its rightful
5685 Nightly snapshots of the Guile development sources are now available via
5686 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5688 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5689 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5691 ** To run Guile without installing it, the procedure has changed a bit.
5693 If you used a separate build directory to compile Guile, you'll need
5694 to include the build directory in SCHEME_LOAD_PATH, as well as the
5695 source directory. See the `INSTALL' file for examples.
5697 * Changes to the procedure for linking libguile with your programs
5699 ** The standard Guile load path for Scheme code now includes
5700 $(datadir)/guile (usually /usr/local/share/guile). This means that
5701 you can install your own Scheme files there, and Guile will find them.
5702 (Previous versions of Guile only checked a directory whose name
5703 contained the Guile version number, so you had to re-install or move
5704 your Scheme sources each time you installed a fresh version of Guile.)
5706 The load path also includes $(datadir)/guile/site; we recommend
5707 putting individual Scheme files there. If you want to install a
5708 package with multiple source files, create a directory for them under
5711 ** Guile 1.2 will now use the Rx regular expression library, if it is
5712 installed on your system. When you are linking libguile into your own
5713 programs, this means you will have to link against -lguile, -lqt (if
5714 you configured Guile with thread support), and -lrx.
5716 If you are using autoconf to generate configuration scripts for your
5717 application, the following lines should suffice to add the appropriate
5718 libraries to your link command:
5720 ### Find Rx, quickthreads and libguile.
5721 AC_CHECK_LIB(rx, main)
5722 AC_CHECK_LIB(qt, main)
5723 AC_CHECK_LIB(guile, scm_shell)
5725 The Guile 1.2 distribution does not contain sources for the Rx
5726 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5727 retrieve it from a GNU FTP site and install it separately.
5729 * Changes to Scheme functions and syntax
5731 ** The dynamic linking features of Guile are now enabled by default.
5732 You can disable them by giving the `--disable-dynamic-linking' option
5735 (dynamic-link FILENAME)
5737 Find the object file denoted by FILENAME (a string) and link it
5738 into the running Guile application. When everything works out,
5739 return a Scheme object suitable for representing the linked object
5740 file. Otherwise an error is thrown. How object files are
5741 searched is system dependent.
5743 (dynamic-object? VAL)
5745 Determine whether VAL represents a dynamically linked object file.
5747 (dynamic-unlink DYNOBJ)
5749 Unlink the indicated object file from the application. DYNOBJ
5750 should be one of the values returned by `dynamic-link'.
5752 (dynamic-func FUNCTION DYNOBJ)
5754 Search the C function indicated by FUNCTION (a string or symbol)
5755 in DYNOBJ and return some Scheme object that can later be used
5756 with `dynamic-call' to actually call this function. Right now,
5757 these Scheme objects are formed by casting the address of the
5758 function to `long' and converting this number to its Scheme
5761 (dynamic-call FUNCTION DYNOBJ)
5763 Call the C function indicated by FUNCTION and DYNOBJ. The
5764 function is passed no arguments and its return value is ignored.
5765 When FUNCTION is something returned by `dynamic-func', call that
5766 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5767 etc.), look it up in DYNOBJ; this is equivalent to
5769 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5771 Interrupts are deferred while the C function is executing (with
5772 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5774 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5776 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5777 some arguments and return its return value. The C function is
5778 expected to take two arguments and return an `int', just like
5781 int c_func (int argc, char **argv);
5783 ARGS must be a list of strings and is converted into an array of
5784 `char *'. The array is passed in ARGV and its size in ARGC. The
5785 return value is converted to a Scheme number and returned from the
5786 call to `dynamic-args-call'.
5788 When dynamic linking is disabled or not supported on your system,
5789 the above functions throw errors, but they are still available.
5791 Here is a small example that works on GNU/Linux:
5793 (define libc-obj (dynamic-link "libc.so"))
5794 (dynamic-args-call 'rand libc-obj '())
5796 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5798 ** The #/ syntax for module names is depreciated, and will be removed
5799 in a future version of Guile. Instead of
5807 The latter syntax is more consistent with existing Lisp practice.
5809 ** Guile now does fancier printing of structures. Structures are the
5810 underlying implementation for records, which in turn are used to
5811 implement modules, so all of these object now print differently and in
5812 a more informative way.
5814 The Scheme printer will examine the builtin variable *struct-printer*
5815 whenever it needs to print a structure object. When this variable is
5816 not `#f' it is deemed to be a procedure and will be applied to the
5817 structure object and the output port. When *struct-printer* is `#f'
5818 or the procedure return `#f' the structure object will be printed in
5819 the boring #<struct 80458270> form.
5821 This hook is used by some routines in ice-9/boot-9.scm to implement
5822 type specific printing routines. Please read the comments there about
5825 One of the more specific uses of structs are records. The printing
5826 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5827 called. It should behave like a *struct-printer* procedure (described
5830 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5831 token of the form #:NAME, where NAME has the same syntax as a Scheme
5832 symbol, is the external representation of the keyword named NAME.
5833 Keyword objects print using this syntax as well, so values containing
5834 keyword objects can be read back into Guile. When used in an
5835 expression, keywords are self-quoting objects.
5837 Guile suports this read syntax, and uses this print syntax, regardless
5838 of the current setting of the `keyword' read option. The `keyword'
5839 read option only controls whether Guile recognizes the `:NAME' syntax,
5840 which is incompatible with R4RS. (R4RS says such token represent
5843 ** Guile has regular expression support again. Guile 1.0 included
5844 functions for matching regular expressions, based on the Rx library.
5845 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5846 distribution, and thus Guile had no regular expression support. Guile
5847 1.2 again supports the most commonly used functions, and supports all
5848 of SCSH's regular expression functions.
5850 If your system does not include a POSIX regular expression library,
5851 and you have not linked Guile with a third-party regexp library such as
5852 Rx, these functions will not be available. You can tell whether your
5853 Guile installation includes regular expression support by checking
5854 whether the `*features*' list includes the `regex' symbol.
5856 *** regexp functions
5858 By default, Guile supports POSIX extended regular expressions. That
5859 means that the characters `(', `)', `+' and `?' are special, and must
5860 be escaped if you wish to match the literal characters.
5862 This regular expression interface was modeled after that implemented
5863 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5864 with SCSH regular expressions.
5866 **** Function: string-match PATTERN STR [START]
5867 Compile the string PATTERN into a regular expression and compare
5868 it with STR. The optional numeric argument START specifies the
5869 position of STR at which to begin matching.
5871 `string-match' returns a "match structure" which describes what,
5872 if anything, was matched by the regular expression. *Note Match
5873 Structures::. If STR does not match PATTERN at all,
5874 `string-match' returns `#f'.
5876 Each time `string-match' is called, it must compile its PATTERN
5877 argument into a regular expression structure. This operation is
5878 expensive, which makes `string-match' inefficient if the same regular
5879 expression is used several times (for example, in a loop). For better
5880 performance, you can compile a regular expression in advance and then
5881 match strings against the compiled regexp.
5883 **** Function: make-regexp STR [FLAGS]
5884 Compile the regular expression described by STR, and return the
5885 compiled regexp structure. If STR does not describe a legal
5886 regular expression, `make-regexp' throws a
5887 `regular-expression-syntax' error.
5889 FLAGS may be the bitwise-or of one or more of the following:
5891 **** Constant: regexp/extended
5892 Use POSIX Extended Regular Expression syntax when interpreting
5893 STR. If not set, POSIX Basic Regular Expression syntax is used.
5894 If the FLAGS argument is omitted, we assume regexp/extended.
5896 **** Constant: regexp/icase
5897 Do not differentiate case. Subsequent searches using the
5898 returned regular expression will be case insensitive.
5900 **** Constant: regexp/newline
5901 Match-any-character operators don't match a newline.
5903 A non-matching list ([^...]) not containing a newline matches a
5906 Match-beginning-of-line operator (^) matches the empty string
5907 immediately after a newline, regardless of whether the FLAGS
5908 passed to regexp-exec contain regexp/notbol.
5910 Match-end-of-line operator ($) matches the empty string
5911 immediately before a newline, regardless of whether the FLAGS
5912 passed to regexp-exec contain regexp/noteol.
5914 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5915 Match the compiled regular expression REGEXP against `str'. If
5916 the optional integer START argument is provided, begin matching
5917 from that position in the string. Return a match structure
5918 describing the results of the match, or `#f' if no match could be
5921 FLAGS may be the bitwise-or of one or more of the following:
5923 **** Constant: regexp/notbol
5924 The match-beginning-of-line operator always fails to match (but
5925 see the compilation flag regexp/newline above) This flag may be
5926 used when different portions of a string are passed to
5927 regexp-exec and the beginning of the string should not be
5928 interpreted as the beginning of the line.
5930 **** Constant: regexp/noteol
5931 The match-end-of-line operator always fails to match (but see the
5932 compilation flag regexp/newline above)
5934 **** Function: regexp? OBJ
5935 Return `#t' if OBJ is a compiled regular expression, or `#f'
5938 Regular expressions are commonly used to find patterns in one string
5939 and replace them with the contents of another string.
5941 **** Function: regexp-substitute PORT MATCH [ITEM...]
5942 Write to the output port PORT selected contents of the match
5943 structure MATCH. Each ITEM specifies what should be written, and
5944 may be one of the following arguments:
5946 * A string. String arguments are written out verbatim.
5948 * An integer. The submatch with that number is written.
5950 * The symbol `pre'. The portion of the matched string preceding
5951 the regexp match is written.
5953 * The symbol `post'. The portion of the matched string
5954 following the regexp match is written.
5956 PORT may be `#f', in which case nothing is written; instead,
5957 `regexp-substitute' constructs a string from the specified ITEMs
5960 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5961 Similar to `regexp-substitute', but can be used to perform global
5962 substitutions on STR. Instead of taking a match structure as an
5963 argument, `regexp-substitute/global' takes two string arguments: a
5964 REGEXP string describing a regular expression, and a TARGET string
5965 which should be matched against this regular expression.
5967 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5970 * A function may be supplied. When this function is called, it
5971 will be passed one argument: a match structure for a given
5972 regular expression match. It should return a string to be
5973 written out to PORT.
5975 * The `post' symbol causes `regexp-substitute/global' to recurse
5976 on the unmatched portion of STR. This *must* be supplied in
5977 order to perform global search-and-replace on STR; if it is
5978 not present among the ITEMs, then `regexp-substitute/global'
5979 will return after processing a single match.
5981 *** Match Structures
5983 A "match structure" is the object returned by `string-match' and
5984 `regexp-exec'. It describes which portion of a string, if any, matched
5985 the given regular expression. Match structures include: a reference to
5986 the string that was checked for matches; the starting and ending
5987 positions of the regexp match; and, if the regexp included any
5988 parenthesized subexpressions, the starting and ending positions of each
5991 In each of the regexp match functions described below, the `match'
5992 argument must be a match structure returned by a previous call to
5993 `string-match' or `regexp-exec'. Most of these functions return some
5994 information about the original target string that was matched against a
5995 regular expression; we will call that string TARGET for easy reference.
5997 **** Function: regexp-match? OBJ
5998 Return `#t' if OBJ is a match structure returned by a previous
5999 call to `regexp-exec', or `#f' otherwise.
6001 **** Function: match:substring MATCH [N]
6002 Return the portion of TARGET matched by subexpression number N.
6003 Submatch 0 (the default) represents the entire regexp match. If
6004 the regular expression as a whole matched, but the subexpression
6005 number N did not match, return `#f'.
6007 **** Function: match:start MATCH [N]
6008 Return the starting position of submatch number N.
6010 **** Function: match:end MATCH [N]
6011 Return the ending position of submatch number N.
6013 **** Function: match:prefix MATCH
6014 Return the unmatched portion of TARGET preceding the regexp match.
6016 **** Function: match:suffix MATCH
6017 Return the unmatched portion of TARGET following the regexp match.
6019 **** Function: match:count MATCH
6020 Return the number of parenthesized subexpressions from MATCH.
6021 Note that the entire regular expression match itself counts as a
6022 subexpression, and failed submatches are included in the count.
6024 **** Function: match:string MATCH
6025 Return the original TARGET string.
6027 *** Backslash Escapes
6029 Sometimes you will want a regexp to match characters like `*' or `$'
6030 exactly. For example, to check whether a particular string represents
6031 a menu entry from an Info node, it would be useful to match it against
6032 a regexp like `^* [^:]*::'. However, this won't work; because the
6033 asterisk is a metacharacter, it won't match the `*' at the beginning of
6034 the string. In this case, we want to make the first asterisk un-magic.
6036 You can do this by preceding the metacharacter with a backslash
6037 character `\'. (This is also called "quoting" the metacharacter, and
6038 is known as a "backslash escape".) When Guile sees a backslash in a
6039 regular expression, it considers the following glyph to be an ordinary
6040 character, no matter what special meaning it would ordinarily have.
6041 Therefore, we can make the above example work by changing the regexp to
6042 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6043 to match only a single asterisk in the target string.
6045 Since the backslash is itself a metacharacter, you may force a
6046 regexp to match a backslash in the target string by preceding the
6047 backslash with itself. For example, to find variable references in a
6048 TeX program, you might want to find occurrences of the string `\let\'
6049 followed by any number of alphabetic characters. The regular expression
6050 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6051 each match a single backslash in the target string.
6053 **** Function: regexp-quote STR
6054 Quote each special character found in STR with a backslash, and
6055 return the resulting string.
6057 *Very important:* Using backslash escapes in Guile source code (as
6058 in Emacs Lisp or C) can be tricky, because the backslash character has
6059 special meaning for the Guile reader. For example, if Guile encounters
6060 the character sequence `\n' in the middle of a string while processing
6061 Scheme code, it replaces those characters with a newline character.
6062 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6063 Several of these "escape sequences" are processed by the Guile reader
6064 before your code is executed. Unrecognized escape sequences are
6065 ignored: if the characters `\*' appear in a string, they will be
6066 translated to the single character `*'.
6068 This translation is obviously undesirable for regular expressions,
6069 since we want to be able to include backslashes in a string in order to
6070 escape regexp metacharacters. Therefore, to make sure that a backslash
6071 is preserved in a string in your Guile program, you must use *two*
6072 consecutive backslashes:
6074 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6076 The string in this example is preprocessed by the Guile reader before
6077 any code is executed. The resulting argument to `make-regexp' is the
6078 string `^\* [^:]*', which is what we really want.
6080 This also means that in order to write a regular expression that
6081 matches a single backslash character, the regular expression string in
6082 the source code must include *four* backslashes. Each consecutive pair
6083 of backslashes gets translated by the Guile reader to a single
6084 backslash, and the resulting double-backslash is interpreted by the
6085 regexp engine as matching a single backslash character. Hence:
6087 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6089 The reason for the unwieldiness of this syntax is historical. Both
6090 regular expression pattern matchers and Unix string processing systems
6091 have traditionally used backslashes with the special meanings described
6092 above. The POSIX regular expression specification and ANSI C standard
6093 both require these semantics. Attempting to abandon either convention
6094 would cause other kinds of compatibility problems, possibly more severe
6095 ones. Therefore, without extending the Scheme reader to support
6096 strings with different quoting conventions (an ungainly and confusing
6097 extension when implemented in other languages), we must adhere to this
6098 cumbersome escape syntax.
6100 * Changes to the gh_ interface
6102 * Changes to the scm_ interface
6104 * Changes to system call interfaces:
6106 ** The value returned by `raise' is now unspecified. It throws an exception
6109 *** A new procedure `sigaction' can be used to install signal handlers
6111 (sigaction signum [action] [flags])
6113 signum is the signal number, which can be specified using the value
6116 If action is omitted, sigaction returns a pair: the CAR is the current
6117 signal hander, which will be either an integer with the value SIG_DFL
6118 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6119 handles the signal, or #f if a non-Scheme procedure handles the
6120 signal. The CDR contains the current sigaction flags for the handler.
6122 If action is provided, it is installed as the new handler for signum.
6123 action can be a Scheme procedure taking one argument, or the value of
6124 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6125 whatever signal handler was installed before sigaction was first used.
6126 Flags can optionally be specified for the new handler (SA_RESTART is
6127 always used if the system provides it, so need not be specified.) The
6128 return value is a pair with information about the old handler as
6131 This interface does not provide access to the "signal blocking"
6132 facility. Maybe this is not needed, since the thread support may
6133 provide solutions to the problem of consistent access to data
6136 *** A new procedure `flush-all-ports' is equivalent to running
6137 `force-output' on every port open for output.
6139 ** Guile now provides information on how it was built, via the new
6140 global variable, %guile-build-info. This variable records the values
6141 of the standard GNU makefile directory variables as an assocation
6142 list, mapping variable names (symbols) onto directory paths (strings).
6143 For example, to find out where the Guile link libraries were
6144 installed, you can say:
6146 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6149 * Changes to the scm_ interface
6151 ** The new function scm_handle_by_message_noexit is just like the
6152 existing scm_handle_by_message function, except that it doesn't call
6153 exit to terminate the process. Instead, it prints a message and just
6154 returns #f. This might be a more appropriate catch-all handler for
6155 new dynamic roots and threads.
6158 Changes in Guile 1.1 (released Friday, May 16 1997):
6160 * Changes to the distribution.
6162 The Guile 1.0 distribution has been split up into several smaller
6164 guile-core --- the Guile interpreter itself.
6165 guile-tcltk --- the interface between the Guile interpreter and
6166 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6167 is a toolkit for building graphical user interfaces.
6168 guile-rgx-ctax --- the interface between Guile and the Rx regular
6169 expression matcher, and the translator for the Ctax
6170 programming language. These are packaged together because the
6171 Ctax translator uses Rx to parse Ctax source code.
6173 This NEWS file describes the changes made to guile-core since the 1.0
6176 We no longer distribute the documentation, since it was either out of
6177 date, or incomplete. As soon as we have current documentation, we
6182 * Changes to the stand-alone interpreter
6184 ** guile now accepts command-line arguments compatible with SCSH, Olin
6185 Shivers' Scheme Shell.
6187 In general, arguments are evaluated from left to right, but there are
6188 exceptions. The following switches stop argument processing, and
6189 stash all remaining command-line arguments as the value returned by
6190 the (command-line) function.
6191 -s SCRIPT load Scheme source code from FILE, and exit
6192 -c EXPR evalute Scheme expression EXPR, and exit
6193 -- stop scanning arguments; run interactively
6195 The switches below are processed as they are encountered.
6196 -l FILE load Scheme source code from FILE
6197 -e FUNCTION after reading script, apply FUNCTION to
6198 command line arguments
6199 -ds do -s script at this point
6200 --emacs enable Emacs protocol (experimental)
6201 -h, --help display this help and exit
6202 -v, --version display version information and exit
6203 \ read arguments from following script lines
6205 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6206 which re-implements the traditional "echo" command:
6208 #!/usr/local/bin/guile -s
6211 (map (lambda (arg) (display arg) (display " "))
6215 (main (command-line))
6217 Suppose we invoke this script as follows:
6219 ekko a speckled gecko
6221 Through the magic of Unix script processing (triggered by the `#!'
6222 token at the top of the file), /usr/local/bin/guile receives the
6223 following list of command-line arguments:
6225 ("-s" "./ekko" "a" "speckled" "gecko")
6227 Unix inserts the name of the script after the argument specified on
6228 the first line of the file (in this case, "-s"), and then follows that
6229 with the arguments given to the script. Guile loads the script, which
6230 defines the `main' function, and then applies it to the list of
6231 remaining command-line arguments, ("a" "speckled" "gecko").
6233 In Unix, the first line of a script file must take the following form:
6235 #!INTERPRETER ARGUMENT
6237 where INTERPRETER is the absolute filename of the interpreter
6238 executable, and ARGUMENT is a single command-line argument to pass to
6241 You may only pass one argument to the interpreter, and its length is
6242 limited. These restrictions can be annoying to work around, so Guile
6243 provides a general mechanism (borrowed from, and compatible with,
6244 SCSH) for circumventing them.
6246 If the ARGUMENT in a Guile script is a single backslash character,
6247 `\', Guile will open the script file, parse arguments from its second
6248 and subsequent lines, and replace the `\' with them. So, for example,
6249 here is another implementation of the `ekko' script:
6251 #!/usr/local/bin/guile \
6255 (for-each (lambda (arg) (display arg) (display " "))
6259 If the user invokes this script as follows:
6261 ekko a speckled gecko
6263 Unix expands this into
6265 /usr/local/bin/guile \ ekko a speckled gecko
6267 When Guile sees the `\' argument, it replaces it with the arguments
6268 read from the second line of the script, producing:
6270 /usr/local/bin/guile -e main -s ekko a speckled gecko
6272 This tells Guile to load the `ekko' script, and apply the function
6273 `main' to the argument list ("a" "speckled" "gecko").
6275 Here is how Guile parses the command-line arguments:
6276 - Each space character terminates an argument. This means that two
6277 spaces in a row introduce an empty-string argument.
6278 - The tab character is not permitted (unless you quote it with the
6279 backslash character, as described below), to avoid confusion.
6280 - The newline character terminates the sequence of arguments, and will
6281 also terminate a final non-empty argument. (However, a newline
6282 following a space will not introduce a final empty-string argument;
6283 it only terminates the argument list.)
6284 - The backslash character is the escape character. It escapes
6285 backslash, space, tab, and newline. The ANSI C escape sequences
6286 like \n and \t are also supported. These produce argument
6287 constituents; the two-character combination \n doesn't act like a
6288 terminating newline. The escape sequence \NNN for exactly three
6289 octal digits reads as the character whose ASCII code is NNN. As
6290 above, characters produced this way are argument constituents.
6291 Backslash followed by other characters is not allowed.
6293 * Changes to the procedure for linking libguile with your programs
6295 ** Guile now builds and installs a shared guile library, if your
6296 system support shared libraries. (It still builds a static library on
6297 all systems.) Guile automatically detects whether your system
6298 supports shared libraries. To prevent Guile from buildisg shared
6299 libraries, pass the `--disable-shared' flag to the configure script.
6301 Guile takes longer to compile when it builds shared libraries, because
6302 it must compile every file twice --- once to produce position-
6303 independent object code, and once to produce normal object code.
6305 ** The libthreads library has been merged into libguile.
6307 To link a program against Guile, you now need only link against
6308 -lguile and -lqt; -lthreads is no longer needed. If you are using
6309 autoconf to generate configuration scripts for your application, the
6310 following lines should suffice to add the appropriate libraries to
6313 ### Find quickthreads and libguile.
6314 AC_CHECK_LIB(qt, main)
6315 AC_CHECK_LIB(guile, scm_shell)
6317 * Changes to Scheme functions
6319 ** Guile Scheme's special syntax for keyword objects is now optional,
6320 and disabled by default.
6322 The syntax variation from R4RS made it difficult to port some
6323 interesting packages to Guile. The routines which accepted keyword
6324 arguments (mostly in the module system) have been modified to also
6325 accept symbols whose names begin with `:'.
6327 To change the keyword syntax, you must first import the (ice-9 debug)
6329 (use-modules (ice-9 debug))
6331 Then you can enable the keyword syntax as follows:
6332 (read-set! keywords 'prefix)
6334 To disable keyword syntax, do this:
6335 (read-set! keywords #f)
6337 ** Many more primitive functions accept shared substrings as
6338 arguments. In the past, these functions required normal, mutable
6339 strings as arguments, although they never made use of this
6342 ** The uniform array functions now operate on byte vectors. These
6343 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6344 `serial-array-map', `array-map', `array-for-each', and
6347 ** The new functions `trace' and `untrace' implement simple debugging
6348 support for Scheme functions.
6350 The `trace' function accepts any number of procedures as arguments,
6351 and tells the Guile interpreter to display each procedure's name and
6352 arguments each time the procedure is invoked. When invoked with no
6353 arguments, `trace' returns the list of procedures currently being
6356 The `untrace' function accepts any number of procedures as arguments,
6357 and tells the Guile interpreter not to trace them any more. When
6358 invoked with no arguments, `untrace' untraces all curretly traced
6361 The tracing in Guile has an advantage over most other systems: we
6362 don't create new procedure objects, but mark the procedure objects
6363 themselves. This means that anonymous and internal procedures can be
6366 ** The function `assert-repl-prompt' has been renamed to
6367 `set-repl-prompt!'. It takes one argument, PROMPT.
6368 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6369 - If PROMPT is a string, we use it as a prompt.
6370 - If PROMPT is a procedure accepting no arguments, we call it, and
6371 display the result as a prompt.
6372 - Otherwise, we display "> ".
6374 ** The new function `eval-string' reads Scheme expressions from a
6375 string and evaluates them, returning the value of the last expression
6376 in the string. If the string contains no expressions, it returns an
6379 ** The new function `thunk?' returns true iff its argument is a
6380 procedure of zero arguments.
6382 ** `defined?' is now a builtin function, instead of syntax. This
6383 means that its argument should be quoted. It returns #t iff its
6384 argument is bound in the current module.
6386 ** The new syntax `use-modules' allows you to add new modules to your
6387 environment without re-typing a complete `define-module' form. It
6388 accepts any number of module names as arguments, and imports their
6389 public bindings into the current module.
6391 ** The new function (module-defined? NAME MODULE) returns true iff
6392 NAME, a symbol, is defined in MODULE, a module object.
6394 ** The new function `builtin-bindings' creates and returns a hash
6395 table containing copies of all the root module's bindings.
6397 ** The new function `builtin-weak-bindings' does the same as
6398 `builtin-bindings', but creates a doubly-weak hash table.
6400 ** The `equal?' function now considers variable objects to be
6401 equivalent if they have the same name and the same value.
6403 ** The new function `command-line' returns the command-line arguments
6404 given to Guile, as a list of strings.
6406 When using guile as a script interpreter, `command-line' returns the
6407 script's arguments; those processed by the interpreter (like `-s' or
6408 `-c') are omitted. (In other words, you get the normal, expected
6409 behavior.) Any application that uses scm_shell to process its
6410 command-line arguments gets this behavior as well.
6412 ** The new function `load-user-init' looks for a file called `.guile'
6413 in the user's home directory, and loads it if it exists. This is
6414 mostly for use by the code generated by scm_compile_shell_switches,
6415 but we thought it might also be useful in other circumstances.
6417 ** The new function `log10' returns the base-10 logarithm of its
6420 ** Changes to I/O functions
6422 *** The functions `read', `primitive-load', `read-and-eval!', and
6423 `primitive-load-path' no longer take optional arguments controlling
6424 case insensitivity and a `#' parser.
6426 Case sensitivity is now controlled by a read option called
6427 `case-insensitive'. The user can add new `#' syntaxes with the
6428 `read-hash-extend' function (see below).
6430 *** The new function `read-hash-extend' allows the user to change the
6431 syntax of Guile Scheme in a somewhat controlled way.
6433 (read-hash-extend CHAR PROC)
6434 When parsing S-expressions, if we read a `#' character followed by
6435 the character CHAR, use PROC to parse an object from the stream.
6436 If PROC is #f, remove any parsing procedure registered for CHAR.
6438 The reader applies PROC to two arguments: CHAR and an input port.
6440 *** The new functions read-delimited and read-delimited! provide a
6441 general mechanism for doing delimited input on streams.
6443 (read-delimited DELIMS [PORT HANDLE-DELIM])
6444 Read until we encounter one of the characters in DELIMS (a string),
6445 or end-of-file. PORT is the input port to read from; it defaults to
6446 the current input port. The HANDLE-DELIM parameter determines how
6447 the terminating character is handled; it should be one of the
6450 'trim omit delimiter from result
6451 'peek leave delimiter character in input stream
6452 'concat append delimiter character to returned value
6453 'split return a pair: (RESULT . TERMINATOR)
6455 HANDLE-DELIM defaults to 'peek.
6457 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6458 A side-effecting variant of `read-delimited'.
6460 The data is written into the string BUF at the indices in the
6461 half-open interval [START, END); the default interval is the whole
6462 string: START = 0 and END = (string-length BUF). The values of
6463 START and END must specify a well-defined interval in BUF, i.e.
6464 0 <= START <= END <= (string-length BUF).
6466 It returns NBYTES, the number of bytes read. If the buffer filled
6467 up without a delimiter character being found, it returns #f. If the
6468 port is at EOF when the read starts, it returns the EOF object.
6470 If an integer is returned (i.e., the read is successfully terminated
6471 by reading a delimiter character), then the HANDLE-DELIM parameter
6472 determines how to handle the terminating character. It is described
6473 above, and defaults to 'peek.
6475 (The descriptions of these functions were borrowed from the SCSH
6476 manual, by Olin Shivers and Brian Carlstrom.)
6478 *** The `%read-delimited!' function is the primitive used to implement
6479 `read-delimited' and `read-delimited!'.
6481 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6483 This returns a pair of values: (TERMINATOR . NUM-READ).
6484 - TERMINATOR describes why the read was terminated. If it is a
6485 character or the eof object, then that is the value that terminated
6486 the read. If it is #f, the function filled the buffer without finding
6487 a delimiting character.
6488 - NUM-READ is the number of characters read into BUF.
6490 If the read is successfully terminated by reading a delimiter
6491 character, then the gobble? parameter determines what to do with the
6492 terminating character. If true, the character is removed from the
6493 input stream; if false, the character is left in the input stream
6494 where a subsequent read operation will retrieve it. In either case,
6495 the character is also the first value returned by the procedure call.
6497 (The descriptions of this function was borrowed from the SCSH manual,
6498 by Olin Shivers and Brian Carlstrom.)
6500 *** The `read-line' and `read-line!' functions have changed; they now
6501 trim the terminator by default; previously they appended it to the
6502 returned string. For the old behavior, use (read-line PORT 'concat).
6504 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6505 take new optional START and END arguments, specifying the region of
6506 the array to read and write.
6508 *** The `ungetc-char-ready?' function has been removed. We feel it's
6509 inappropriate for an interface to expose implementation details this
6512 ** Changes to the Unix library and system call interface
6514 *** The new fcntl function provides access to the Unix `fcntl' system
6517 (fcntl PORT COMMAND VALUE)
6518 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6519 Values for COMMAND are:
6521 F_DUPFD duplicate a file descriptor
6522 F_GETFD read the descriptor's close-on-exec flag
6523 F_SETFD set the descriptor's close-on-exec flag to VALUE
6524 F_GETFL read the descriptor's flags, as set on open
6525 F_SETFL set the descriptor's flags, as set on open to VALUE
6526 F_GETOWN return the process ID of a socket's owner, for SIGIO
6527 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6528 FD_CLOEXEC not sure what this is
6530 For details, see the documentation for the fcntl system call.
6532 *** The arguments to `select' have changed, for compatibility with
6533 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6534 expected behavior. The MILLISECONDS parameter has been changed to
6535 MICROSECONDS, to more closely resemble the underlying system call.
6536 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6537 corresponding return set will be the same.
6539 *** The arguments to the `mknod' system call have changed. They are
6542 (mknod PATH TYPE PERMS DEV)
6543 Create a new file (`node') in the file system. PATH is the name of
6544 the file to create. TYPE is the kind of file to create; it should
6545 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6546 permission bits to give the newly created file. If TYPE is
6547 'block-special or 'char-special, DEV specifies which device the
6548 special file refers to; its interpretation depends on the kind of
6549 special file being created.
6551 *** The `fork' function has been renamed to `primitive-fork', to avoid
6552 clashing with various SCSH forks.
6554 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6555 and `recvfrom!'. They no longer accept a size for a second argument;
6556 you must pass a string to hold the received value. They no longer
6557 return the buffer. Instead, `recv' returns the length of the message
6558 received, and `recvfrom' returns a pair containing the packet's length
6559 and originating address.
6561 *** The file descriptor datatype has been removed, as have the
6562 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6563 We plan to replace these functions with a SCSH-compatible interface.
6565 *** The `create' function has been removed; it's just a special case
6568 *** There are new functions to break down process termination status
6569 values. In the descriptions below, STATUS is a value returned by
6572 (status:exit-val STATUS)
6573 If the child process exited normally, this function returns the exit
6574 code for the child process (i.e., the value passed to exit, or
6575 returned from main). If the child process did not exit normally,
6576 this function returns #f.
6578 (status:stop-sig STATUS)
6579 If the child process was suspended by a signal, this function
6580 returns the signal that suspended the child. Otherwise, it returns
6583 (status:term-sig STATUS)
6584 If the child process terminated abnormally, this function returns
6585 the signal that terminated the child. Otherwise, this function
6588 POSIX promises that exactly one of these functions will return true on
6589 a valid STATUS value.
6591 These functions are compatible with SCSH.
6593 *** There are new accessors and setters for the broken-out time vectors
6594 returned by `localtime', `gmtime', and that ilk. They are:
6596 Component Accessor Setter
6597 ========================= ============ ============
6598 seconds tm:sec set-tm:sec
6599 minutes tm:min set-tm:min
6600 hours tm:hour set-tm:hour
6601 day of the month tm:mday set-tm:mday
6602 month tm:mon set-tm:mon
6603 year tm:year set-tm:year
6604 day of the week tm:wday set-tm:wday
6605 day in the year tm:yday set-tm:yday
6606 daylight saving time tm:isdst set-tm:isdst
6607 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6608 name of time zone tm:zone set-tm:zone
6610 *** There are new accessors for the vectors returned by `uname',
6611 describing the host system:
6614 ============================================== ================
6615 name of the operating system implementation utsname:sysname
6616 network name of this machine utsname:nodename
6617 release level of the operating system utsname:release
6618 version level of the operating system utsname:version
6619 machine hardware platform utsname:machine
6621 *** There are new accessors for the vectors returned by `getpw',
6622 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6623 system's user database:
6626 ====================== =================
6627 user name passwd:name
6628 user password passwd:passwd
6631 real name passwd:gecos
6632 home directory passwd:dir
6633 shell program passwd:shell
6635 *** There are new accessors for the vectors returned by `getgr',
6636 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6637 system's group database:
6640 ======================= ============
6641 group name group:name
6642 group password group:passwd
6644 group members group:mem
6646 *** There are new accessors for the vectors returned by `gethost',
6647 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6651 ========================= ===============
6652 official name of host hostent:name
6653 alias list hostent:aliases
6654 host address type hostent:addrtype
6655 length of address hostent:length
6656 list of addresses hostent:addr-list
6658 *** There are new accessors for the vectors returned by `getnet',
6659 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6663 ========================= ===============
6664 official name of net netent:name
6665 alias list netent:aliases
6666 net number type netent:addrtype
6667 net number netent:net
6669 *** There are new accessors for the vectors returned by `getproto',
6670 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6674 ========================= ===============
6675 official protocol name protoent:name
6676 alias list protoent:aliases
6677 protocol number protoent:proto
6679 *** There are new accessors for the vectors returned by `getserv',
6680 `getservbyname', `getservbyport', and `getservent', describing
6684 ========================= ===============
6685 official service name servent:name
6686 alias list servent:aliases
6687 port number servent:port
6688 protocol to use servent:proto
6690 *** There are new accessors for the sockaddr structures returned by
6691 `accept', `getsockname', `getpeername', `recvfrom!':
6694 ======================================== ===============
6695 address format (`family') sockaddr:fam
6696 path, for file domain addresses sockaddr:path
6697 address, for internet domain addresses sockaddr:addr
6698 TCP or UDP port, for internet sockaddr:port
6700 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6701 `getprotoent', and `getservent' functions now return #f at the end of
6702 the user database. (They used to throw an exception.)
6704 Note that calling MUMBLEent function is equivalent to calling the
6705 corresponding MUMBLE function with no arguments.
6707 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6708 `setprotoent', and `setservent' routines now take no arguments.
6710 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6711 provide more useful information when they throw an exception.
6713 *** The `lnaof' function has been renamed to `inet-lnaof'.
6715 *** Guile now claims to have the `current-time' feature.
6717 *** The `mktime' function now takes an optional second argument ZONE,
6718 giving the time zone to use for the conversion. ZONE should be a
6719 string, in the same format as expected for the "TZ" environment variable.
6721 *** The `strptime' function now returns a pair (TIME . COUNT), where
6722 TIME is the parsed time as a vector, and COUNT is the number of
6723 characters from the string left unparsed. This function used to
6724 return the remaining characters as a string.
6726 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6727 The return value is now (SECONDS . MICROSECONDS); the fractional
6728 component is no longer expressed in "ticks".
6730 *** The `ticks/sec' constant has been removed, in light of the above change.
6732 * Changes to the gh_ interface
6734 ** gh_eval_str() now returns an SCM object which is the result of the
6737 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6740 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6741 and returns the array
6743 ** gh_scm2str0() is gone: there is no need to distinguish
6744 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6745 the user to interpret the data both ways.
6747 * Changes to the scm_ interface
6749 ** The new function scm_symbol_value0 provides an easy way to get a
6750 symbol's value from C code:
6752 SCM scm_symbol_value0 (char *NAME)
6753 Return the value of the symbol named by the null-terminated string
6754 NAME in the current module. If the symbol named NAME is unbound in
6755 the current module, return SCM_UNDEFINED.
6757 ** The new function scm_sysintern0 creates new top-level variables,
6758 without assigning them a value.
6760 SCM scm_sysintern0 (char *NAME)
6761 Create a new Scheme top-level variable named NAME. NAME is a
6762 null-terminated string. Return the variable's value cell.
6764 ** The function scm_internal_catch is the guts of catch. It handles
6765 all the mechanics of setting up a catch target, invoking the catch
6766 body, and perhaps invoking the handler if the body does a throw.
6768 The function is designed to be usable from C code, but is general
6769 enough to implement all the semantics Guile Scheme expects from throw.
6771 TAG is the catch tag. Typically, this is a symbol, but this function
6772 doesn't actually care about that.
6774 BODY is a pointer to a C function which runs the body of the catch;
6775 this is the code you can throw from. We call it like this:
6776 BODY (BODY_DATA, JMPBUF)
6778 BODY_DATA is just the BODY_DATA argument we received; we pass it
6779 through to BODY as its first argument. The caller can make
6780 BODY_DATA point to anything useful that BODY might need.
6781 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6782 which we have just created and initialized.
6784 HANDLER is a pointer to a C function to deal with a throw to TAG,
6785 should one occur. We call it like this:
6786 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6788 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6789 same idea as BODY_DATA above.
6790 THROWN_TAG is the tag that the user threw to; usually this is
6791 TAG, but it could be something else if TAG was #t (i.e., a
6792 catch-all), or the user threw to a jmpbuf.
6793 THROW_ARGS is the list of arguments the user passed to the THROW
6796 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6797 is just a pointer we pass through to HANDLER. We don't actually
6798 use either of those pointers otherwise ourselves. The idea is
6799 that, if our caller wants to communicate something to BODY or
6800 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6801 HANDLER can then use. Think of it as a way to make BODY and
6802 HANDLER closures, not just functions; MUMBLE_DATA points to the
6805 Of course, it's up to the caller to make sure that any data a
6806 MUMBLE_DATA needs is protected from GC. A common way to do this is
6807 to make MUMBLE_DATA a pointer to data stored in an automatic
6808 structure variable; since the collector must scan the stack for
6809 references anyway, this assures that any references in MUMBLE_DATA
6812 ** The new function scm_internal_lazy_catch is exactly like
6813 scm_internal_catch, except:
6815 - It does not unwind the stack (this is the major difference).
6816 - If handler returns, its value is returned from the throw.
6817 - BODY always receives #f as its JMPBUF argument (since there's no
6818 jmpbuf associated with a lazy catch, because we don't unwind the
6821 ** scm_body_thunk is a new body function you can pass to
6822 scm_internal_catch if you want the body to be like Scheme's `catch'
6823 --- a thunk, or a function of one argument if the tag is #f.
6825 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6826 contains the Scheme procedure to invoke as the body, and the tag
6827 we're catching. If the tag is #f, then we pass JMPBUF (created by
6828 scm_internal_catch) to the body procedure; otherwise, the body gets
6831 ** scm_handle_by_proc is a new handler function you can pass to
6832 scm_internal_catch if you want the handler to act like Scheme's catch
6833 --- call a procedure with the tag and the throw arguments.
6835 If the user does a throw to this catch, this function runs a handler
6836 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6837 variable holding the Scheme procedure object to invoke. It ought to
6838 be a pointer to an automatic variable (i.e., one living on the stack),
6839 or the procedure object should be otherwise protected from GC.
6841 ** scm_handle_by_message is a new handler function to use with
6842 `scm_internal_catch' if you want Guile to print a message and die.
6843 It's useful for dealing with throws to uncaught keys at the top level.
6845 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6846 message header to print; if zero, we use "guile" instead. That
6847 text is followed by a colon, then the message described by ARGS.
6849 ** The return type of scm_boot_guile is now void; the function does
6850 not return a value, and indeed, never returns at all.
6852 ** The new function scm_shell makes it easy for user applications to
6853 process command-line arguments in a way that is compatible with the
6854 stand-alone guile interpreter (which is in turn compatible with SCSH,
6857 To use the scm_shell function, first initialize any guile modules
6858 linked into your application, and then call scm_shell with the values
6859 of ARGC and ARGV your `main' function received. scm_shell will add
6860 any SCSH-style meta-arguments from the top of the script file to the
6861 argument vector, and then process the command-line arguments. This
6862 generally means loading a script file or starting up an interactive
6863 command interpreter. For details, see "Changes to the stand-alone
6866 ** The new functions scm_get_meta_args and scm_count_argv help you
6867 implement the SCSH-style meta-argument, `\'.
6869 char **scm_get_meta_args (int ARGC, char **ARGV)
6870 If the second element of ARGV is a string consisting of a single
6871 backslash character (i.e. "\\" in Scheme notation), open the file
6872 named by the following argument, parse arguments from it, and return
6873 the spliced command line. The returned array is terminated by a
6876 For details of argument parsing, see above, under "guile now accepts
6877 command-line arguments compatible with SCSH..."
6879 int scm_count_argv (char **ARGV)
6880 Count the arguments in ARGV, assuming it is terminated by a null
6883 For an example of how these functions might be used, see the source
6884 code for the function scm_shell in libguile/script.c.
6886 You will usually want to use scm_shell instead of calling this
6889 ** The new function scm_compile_shell_switches turns an array of
6890 command-line arguments into Scheme code to carry out the actions they
6891 describe. Given ARGC and ARGV, it returns a Scheme expression to
6892 evaluate, and calls scm_set_program_arguments to make any remaining
6893 command-line arguments available to the Scheme code. For example,
6894 given the following arguments:
6896 -e main -s ekko a speckled gecko
6898 scm_set_program_arguments will return the following expression:
6900 (begin (load "ekko") (main (command-line)) (quit))
6902 You will usually want to use scm_shell instead of calling this
6905 ** The function scm_shell_usage prints a usage message appropriate for
6906 an interpreter that uses scm_compile_shell_switches to handle its
6907 command-line arguments.
6909 void scm_shell_usage (int FATAL, char *MESSAGE)
6910 Print a usage message to the standard error output. If MESSAGE is
6911 non-zero, write it before the usage message, followed by a newline.
6912 If FATAL is non-zero, exit the process, using FATAL as the
6913 termination status. (If you want to be compatible with Guile,
6914 always use 1 as the exit status when terminating due to command-line
6917 You will usually want to use scm_shell instead of calling this
6920 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6921 expressions. It used to return SCM_EOL. Earth-shattering.
6923 ** The macros for declaring scheme objects in C code have been
6924 rearranged slightly. They are now:
6926 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6927 Declare a static SCM variable named C_NAME, and initialize it to
6928 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6929 be a C identifier, and SCHEME_NAME should be a C string.
6931 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6932 Just like SCM_SYMBOL, but make C_NAME globally visible.
6934 SCM_VCELL (C_NAME, SCHEME_NAME)
6935 Create a global variable at the Scheme level named SCHEME_NAME.
6936 Declare a static SCM variable named C_NAME, and initialize it to
6937 point to the Scheme variable's value cell.
6939 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6940 Just like SCM_VCELL, but make C_NAME globally visible.
6942 The `guile-snarf' script writes initialization code for these macros
6943 to its standard output, given C source code as input.
6945 The SCM_GLOBAL macro is gone.
6947 ** The scm_read_line and scm_read_line_x functions have been replaced
6948 by Scheme code based on the %read-delimited! procedure (known to C
6949 code as scm_read_delimited_x). See its description above for more
6952 ** The function scm_sys_open has been renamed to scm_open. It now
6953 returns a port instead of an FD object.
6955 * The dynamic linking support has changed. For more information, see
6956 libguile/DYNAMIC-LINKING.
6961 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6964 * Changes to the 'guile' program:
6966 ** Guile now loads some new files when it starts up. Guile first
6967 searches the load path for init.scm, and loads it if found. Then, if
6968 Guile is not being used to execute a script, and the user's home
6969 directory contains a file named `.guile', Guile loads that.
6971 ** You can now use Guile as a shell script interpreter.
6973 To paraphrase the SCSH manual:
6975 When Unix tries to execute an executable file whose first two
6976 characters are the `#!', it treats the file not as machine code to
6977 be directly executed by the native processor, but as source code
6978 to be executed by some interpreter. The interpreter to use is
6979 specified immediately after the #! sequence on the first line of
6980 the source file. The kernel reads in the name of the interpreter,
6981 and executes that instead. It passes the interpreter the source
6982 filename as its first argument, with the original arguments
6983 following. Consult the Unix man page for the `exec' system call
6984 for more information.
6986 Now you can use Guile as an interpreter, using a mechanism which is a
6987 compatible subset of that provided by SCSH.
6989 Guile now recognizes a '-s' command line switch, whose argument is the
6990 name of a file of Scheme code to load. It also treats the two
6991 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6992 to make a file of Scheme code directly executable by Unix, insert the
6993 following two lines at the top of the file:
6995 #!/usr/local/bin/guile -s
6998 Guile treats the argument of the `-s' command-line switch as the name
6999 of a file of Scheme code to load, and treats the sequence `#!' as the
7000 start of a block comment, terminated by `!#'.
7002 For example, here's a version of 'echo' written in Scheme:
7004 #!/usr/local/bin/guile -s
7006 (let loop ((args (cdr (program-arguments))))
7009 (display (car args))
7010 (if (pair? (cdr args))
7012 (loop (cdr args)))))
7015 Why does `#!' start a block comment terminated by `!#', instead of the
7016 end of the line? That is the notation SCSH uses, and although we
7017 don't yet support the other SCSH features that motivate that choice,
7018 we would like to be backward-compatible with any existing Guile
7019 scripts once we do. Furthermore, if the path to Guile on your system
7020 is too long for your kernel, you can start the script with this
7024 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7027 Note that some very old Unix systems don't support the `#!' syntax.
7030 ** You can now run Guile without installing it.
7032 Previous versions of the interactive Guile interpreter (`guile')
7033 couldn't start up unless Guile's Scheme library had been installed;
7034 they used the value of the environment variable `SCHEME_LOAD_PATH'
7035 later on in the startup process, but not to find the startup code
7036 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7039 To run Guile without installing it, build it in the normal way, and
7040 then set the environment variable `SCHEME_LOAD_PATH' to a
7041 colon-separated list of directories, including the top-level directory
7042 of the Guile sources. For example, if you unpacked Guile so that the
7043 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7046 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7049 ** Guile's read-eval-print loop no longer prints #<unspecified>
7050 results. If the user wants to see this, she can evaluate the
7051 expression (assert-repl-print-unspecified #t), perhaps in her startup
7054 ** Guile no longer shows backtraces by default when an error occurs;
7055 however, it does display a message saying how to get one, and how to
7056 request that they be displayed by default. After an error, evaluate
7058 to see a backtrace, and
7059 (debug-enable 'backtrace)
7060 to see them by default.
7064 * Changes to Guile Scheme:
7066 ** Guile now distinguishes between #f and the empty list.
7068 This is for compatibility with the IEEE standard, the (possibly)
7069 upcoming Revised^5 Report on Scheme, and many extant Scheme
7072 Guile used to have #f and '() denote the same object, to make Scheme's
7073 type system more compatible with Emacs Lisp's. However, the change
7074 caused too much trouble for Scheme programmers, and we found another
7075 way to reconcile Emacs Lisp with Scheme that didn't require this.
7078 ** Guile's delq, delv, delete functions, and their destructive
7079 counterparts, delq!, delv!, and delete!, now remove all matching
7080 elements from the list, not just the first. This matches the behavior
7081 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7082 functions which inspired them.
7084 I recognize that this change may break code in subtle ways, but it
7085 seems best to make the change before the FSF's first Guile release,
7089 ** The compiled-library-path function has been deleted from libguile.
7091 ** The facilities for loading Scheme source files have changed.
7093 *** The variable %load-path now tells Guile which directories to search
7094 for Scheme code. Its value is a list of strings, each of which names
7097 *** The variable %load-extensions now tells Guile which extensions to
7098 try appending to a filename when searching the load path. Its value
7099 is a list of strings. Its default value is ("" ".scm").
7101 *** (%search-load-path FILENAME) searches the directories listed in the
7102 value of the %load-path variable for a Scheme file named FILENAME,
7103 with all the extensions listed in %load-extensions. If it finds a
7104 match, then it returns its full filename. If FILENAME is absolute, it
7105 returns it unchanged. Otherwise, it returns #f.
7107 %search-load-path will not return matches that refer to directories.
7109 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7110 uses %seach-load-path to find a file named FILENAME, and loads it if
7111 it finds it. If it can't read FILENAME for any reason, it throws an
7114 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7117 *** load uses the same searching semantics as primitive-load.
7119 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7120 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7121 path, and load-module-with-path have been deleted. The functions
7122 above should serve their purposes.
7124 *** If the value of the variable %load-hook is a procedure,
7125 `primitive-load' applies its value to the name of the file being
7126 loaded (without the load path directory name prepended). If its value
7127 is #f, it is ignored. Otherwise, an error occurs.
7129 This is mostly useful for printing load notification messages.
7132 ** The function `eval!' is no longer accessible from the scheme level.
7133 We can't allow operations which introduce glocs into the scheme level,
7134 because Guile's type system can't handle these as data. Use `eval' or
7135 `read-and-eval!' (see below) as replacement.
7137 ** The new function read-and-eval! reads an expression from PORT,
7138 evaluates it, and returns the result. This is more efficient than
7139 simply calling `read' and `eval', since it is not necessary to make a
7140 copy of the expression for the evaluator to munge.
7142 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7143 for the `read' function.
7146 ** The function `int?' has been removed; its definition was identical
7147 to that of `integer?'.
7149 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7150 use the R4RS names for these functions.
7152 ** The function object-properties no longer returns the hash handle;
7153 it simply returns the object's property list.
7155 ** Many functions have been changed to throw errors, instead of
7156 returning #f on failure. The point of providing exception handling in
7157 the language is to simplify the logic of user code, but this is less
7158 useful if Guile's primitives don't throw exceptions.
7160 ** The function `fileno' has been renamed from `%fileno'.
7162 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7165 * Changes to Guile's C interface:
7167 ** The library's initialization procedure has been simplified.
7168 scm_boot_guile now has the prototype:
7170 void scm_boot_guile (int ARGC,
7172 void (*main_func) (),
7175 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7176 MAIN_FUNC should do all the work of the program (initializing other
7177 packages, reading user input, etc.) before returning. When MAIN_FUNC
7178 returns, call exit (0); this function never returns. If you want some
7179 other exit value, MAIN_FUNC may call exit itself.
7181 scm_boot_guile arranges for program-arguments to return the strings
7182 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7183 scm_set_program_arguments with the final list, so Scheme code will
7184 know which arguments have been processed.
7186 scm_boot_guile establishes a catch-all catch handler which prints an
7187 error message and exits the process. This means that Guile exits in a
7188 coherent way when system errors occur and the user isn't prepared to
7189 handle it. If the user doesn't like this behavior, they can establish
7190 their own universal catcher in MAIN_FUNC to shadow this one.
7192 Why must the caller do all the real work from MAIN_FUNC? The garbage
7193 collector assumes that all local variables of type SCM will be above
7194 scm_boot_guile's stack frame on the stack. If you try to manipulate
7195 SCM values after this function returns, it's the luck of the draw
7196 whether the GC will be able to find the objects you allocate. So,
7197 scm_boot_guile function exits, rather than returning, to discourage
7198 people from making that mistake.
7200 The IN, OUT, and ERR arguments were removed; there are other
7201 convenient ways to override these when desired.
7203 The RESULT argument was deleted; this function should never return.
7205 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7209 ** Guile's header files should no longer conflict with your system's
7212 In order to compile code which #included <libguile.h>, previous
7213 versions of Guile required you to add a directory containing all the
7214 Guile header files to your #include path. This was a problem, since
7215 Guile's header files have names which conflict with many systems'
7218 Now only <libguile.h> need appear in your #include path; you must
7219 refer to all Guile's other header files as <libguile/mumble.h>.
7220 Guile's installation procedure puts libguile.h in $(includedir), and
7221 the rest in $(includedir)/libguile.
7224 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7225 have been added to the Guile library.
7227 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7228 OBJ will not be freed, even if all other references are dropped,
7229 until someone does scm_unprotect_object (OBJ). Both functions
7232 Note that calls to scm_protect_object do not nest. You can call
7233 scm_protect_object any number of times on a given object, and the
7234 next call to scm_unprotect_object will unprotect it completely.
7236 Basically, scm_protect_object and scm_unprotect_object just
7237 maintain a list of references to things. Since the GC knows about
7238 this list, all objects it mentions stay alive. scm_protect_object
7239 adds its argument to the list; scm_unprotect_object remove its
7240 argument from the list.
7243 ** scm_eval_0str now returns the value of the last expression
7246 ** The new function scm_read_0str reads an s-expression from a
7247 null-terminated string, and returns it.
7249 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7250 to a Scheme port object.
7252 ** The new function `scm_set_program_arguments' allows C code to set
7253 the value returned by the Scheme `program-arguments' function.
7258 * Guile no longer includes sophisticated Tcl/Tk support.
7260 The old Tcl/Tk support was unsatisfying to us, because it required the
7261 user to link against the Tcl library, as well as Tk and Guile. The
7262 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7263 referring to widgets by names, rather than exporting widgets to Scheme
7264 code as a special datatype.
7266 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7267 maintainers described some very interesting changes in progress to the
7268 Tcl/Tk internals, which would facilitate clean interfaces between lone
7269 Tk and other interpreters --- even for garbage-collected languages
7270 like Scheme. They expected the new Tk to be publicly available in the
7273 Since it seems that Guile might soon have a new, cleaner interface to
7274 lone Tk, and that the old Guile/Tk glue code would probably need to be
7275 completely rewritten, we (Jim Blandy and Richard Stallman) have
7276 decided not to support the old code. We'll spend the time instead on
7277 a good interface to the newer Tk, as soon as it is available.
7279 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7282 Copyright information:
7284 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7286 Permission is granted to anyone to make or distribute verbatim copies
7287 of this document as received, in any medium, provided that the
7288 copyright notice and this permission notice are preserved,
7289 thus giving the recipient permission to redistribute in turn.
7291 Permission is granted to distribute modified versions
7292 of this document, or of portions of it,
7293 under the above conditions, provided also that they
7294 carry prominent notices stating who last changed them.
7299 paragraph-separate: "[
\f]*$"