1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006 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
9 Each release reports the NEWS in the following sections:
11 * Changes to the distribution
12 * Changes to the stand-alone interpreter
13 * Changes to Scheme functions and syntax
14 * Changes to the C interface
17 Changes in 1.9.XXXXXXXX:
19 * Changes to the distribution
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 * Changes to the C interface
28 ** Functions for handling scm_option now no longer require an argument
29 indicating length of the scm_t_option array.
32 Changes in 1.8.2 (since 1.8.1):
34 * New procedures (see the manual for details)
36 ** set-program-arguments
40 ** Fractions were not `equal?' if stored in unreduced form.
41 (A subtle problem, since printing a value reduced it, making it work.)
42 ** srfi-60 `copy-bit' failed on 64-bit systems
43 ** "guile --use-srfi" option at the REPL can replace core functions
44 (Programs run with that option were ok, but in the interactive REPL
45 the core bindings got priority, preventing SRFI replacements or
47 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
48 ** `kill' on mingw throws an error for a pid other than oneself
49 ** Procedure names are attached to procedure-with-setters
50 ** Array read syntax works with negative lower bound
51 ** `array-in-bounds?' fix if an array has different lower bounds on each index
52 ** `*' returns exact 0 for "(* inexact 0)"
53 This follows what it always did for "(* 0 inexact)".
54 ** Build problems on Solaris fixed
55 ** Build problems on Mingw fixed
58 Changes in 1.8.1 (since 1.8.0):
60 * LFS functions are now used to access 64-bit files on 32-bit systems.
62 * New procedures (see the manual for details)
64 ** primitive-_exit - [Scheme] the-root-module
65 ** scm_primitive__exit - [C]
66 ** make-completion-function - [Scheme] (ice-9 readline)
67 ** scm_c_locale_stringn_to_number - [C]
68 ** scm_srfi1_append_reverse [C]
69 ** scm_srfi1_append_reverse_x [C]
75 * New `(ice-9 i18n)' module (see the manual for details)
79 ** Build problems have been fixed on MacOS, SunOS, and QNX.
81 ** A one-dimensional array can now be 'equal?' to a vector.
83 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
85 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
87 ** `record-accessor' and `record-modifier' now have strict type checks.
89 Record accessor and modifier procedures now throw an error if the
90 record type of the record they're given is not the type expected.
91 (Previously accessors returned #f and modifiers silently did nothing).
93 ** It is now OK to use both autoload and use-modules on a given module.
95 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
97 Previously there was no checking on primatives like make-vector that
98 accept "one or two" arguments. Now there is.
100 ** The srfi-1 assoc function now calls its equality predicate properly.
102 Previously srfi-1 assoc would call the equality predicate with the key
103 last. According to the SRFI, the key should be first.
105 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
107 ** The array-set! procedure no longer segfaults when given a bit vector.
109 ** Bugs in make-shared-array have been fixed.
111 ** string<? and friends now follow char<? etc order on 8-bit chars.
113 ** The format procedure now handles inf and nan values for ~f correctly.
115 ** exact->inexact should no longer overflow when given certain large fractions.
117 ** srfi-9 accessor and modifier procedures now have strict record type checks.
119 This matches the srfi-9 specification.
121 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
123 Previously the (ice-9 ftw) procedures would ignore any file that had
124 the same inode number as a file they had already seen, even if that
125 file was on a different device.
128 Changes in 1.8.0 (changes since the 1.6.x series):
130 * Changes to the distribution
132 ** Guile is now licensed with the GNU Lesser General Public License.
134 ** The manual is now licensed with the GNU Free Documentation License.
136 ** Guile now requires GNU MP (http://swox.com/gmp).
138 Guile now uses the GNU MP library for arbitrary precision arithmetic.
140 ** Guile now has separate private and public configuration headers.
142 That is, things like HAVE_STRING_H no longer leak from Guile's
145 ** Guile now provides and uses an "effective" version number.
147 Guile now provides scm_effective_version and effective-version
148 functions which return the "effective" version number. This is just
149 the normal full version string without the final micro-version number,
150 so the current effective-version is "1.8". The effective version
151 should remain unchanged during a stable series, and should be used for
152 items like the versioned share directory name
153 i.e. /usr/share/guile/1.8.
155 Providing an unchanging version number during a stable release for
156 things like the versioned share directory can be particularly
157 important for Guile "add-on" packages, since it provides a directory
158 that they can install to that won't be changed out from under them
159 with each micro release during a stable series.
161 ** Thread implementation has changed.
163 When you configure "--with-threads=null", you will get the usual
164 threading API (call-with-new-thread, make-mutex, etc), but you can't
165 actually create new threads. Also, "--with-threads=no" is now
166 equivalent to "--with-threads=null". This means that the thread API
167 is always present, although you might not be able to create new
170 When you configure "--with-threads=pthreads" or "--with-threads=yes",
171 you will get threads that are implemented with the portable POSIX
172 threads. These threads can run concurrently (unlike the previous
173 "coop" thread implementation), but need to cooperate for things like
176 The default is "pthreads", unless your platform doesn't have pthreads,
177 in which case "null" threads are used.
179 See the manual for details, nodes "Initialization", "Multi-Threading",
180 "Blocking", and others.
182 ** There is the new notion of 'discouraged' features.
184 This is a milder form of deprecation.
186 Things that are discouraged should not be used in new code, but it is
187 OK to leave them in old code for now. When a discouraged feature is
188 used, no warning message is printed like there is for 'deprecated'
189 features. Also, things that are merely discouraged are nevertheless
190 implemented efficiently, while deprecated features can be very slow.
192 You can omit discouraged features from libguile by configuring it with
193 the '--disable-discouraged' option.
195 ** Deprecation warnings can be controlled at run-time.
197 (debug-enable 'warn-deprecated) switches them on and (debug-disable
198 'warn-deprecated) switches them off.
200 ** Support for SRFI 61, extended cond syntax for multiple values has
203 This SRFI is always available.
205 ** Support for require-extension, SRFI-55, has been added.
207 The SRFI-55 special form `require-extension' has been added. It is
208 available at startup, and provides a portable way to load Scheme
209 extensions. SRFI-55 only requires support for one type of extension,
210 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
213 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
215 The (srfi srfi-26) module is an implementation of SRFI-26 which
216 provides the `cut' and `cute' syntax. These may be used to specialize
217 parameters without currying.
219 ** New module (srfi srfi-31)
221 This is an implementation of SRFI-31 which provides a special form
222 `rec' for recursive evaluation.
224 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
225 been merged with the core, making their functionality always
228 The modules are still available, tho, and you could use them together
229 with a renaming import, for example.
231 ** Guile no longer includes its own version of libltdl.
233 The official version is good enough now.
235 ** The --enable-htmldoc option has been removed from 'configure'.
237 Support for translating the documentation into HTML is now always
238 provided. Use 'make html'.
240 ** New module (ice-9 serialize):
242 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
243 don't trust the thread safety of most of your program, but where you
244 have some section(s) of code which you consider can run in parallel to
245 other sections. See ice-9/serialize.scm for more information.
247 ** The configure option '--disable-arrays' has been removed.
249 Support for arrays and uniform numeric arrays is now always included
252 * Changes to the stand-alone interpreter
254 ** New command line option `-L'.
256 This option adds a directory to the front of the load path.
258 ** New command line option `--no-debug'.
260 Specifying `--no-debug' on the command line will keep the debugging
261 evaluator turned off, even for interactive sessions.
263 ** User-init file ~/.guile is now loaded with the debugging evaluator.
265 Previously, the normal evaluator would have been used. Using the
266 debugging evaluator gives better error messages.
268 ** The '-e' option now 'read's its argument.
270 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
271 be used with '-e'. For example, you can now write a script like
274 exec guile -e '(@ (demo) main)' -s "$0" "$@"
277 (define-module (demo)
281 (format #t "Demo: ~a~%" args))
284 * Changes to Scheme functions and syntax
286 ** Guardians have changed back to their original semantics
288 Guardians now behave like described in the paper by Dybvig et al. In
289 particular, they no longer make guarantees about the order in which
290 they return objects, and they can no longer be greedy.
292 They no longer drop cyclic data structures.
294 The C function scm_make_guardian has been changed incompatibly and no
295 longer takes the 'greedy_p' argument.
297 ** New function hashx-remove!
299 This function completes the set of 'hashx' functions.
301 ** The concept of dynamic roots has been factored into continuation
302 barriers and dynamic states.
304 Each thread has a current dynamic state that carries the values of the
305 fluids. You can create and copy dynamic states and use them as the
306 second argument for 'eval'. See "Fluids and Dynamic States" in the
309 To restrict the influence that captured continuations can have on the
310 control flow, you can errect continuation barriers. See "Continuation
311 Barriers" in the manual.
313 The function call-with-dynamic-root now essentially temporarily
314 installs a new dynamic state and errects a continuation barrier.
316 ** The default load path no longer includes "." at the end.
318 Automatically loading modules from the current directory should not
319 happen by default. If you want to allow it in a more controlled
320 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
323 ** The uniform vector and array support has been overhauled.
325 It now complies with SRFI-4 and the weird prototype based uniform
326 array creation has been deprecated. See the manual for more details.
328 Some non-compatible changes have been made:
329 - characters can no longer be stored into byte arrays.
330 - strings and bit vectors are no longer considered to be uniform numeric
332 - array-rank throws an error for non-arrays instead of returning zero.
333 - array-ref does no longer accept non-arrays when no indices are given.
335 There is the new notion of 'generalized vectors' and corresponding
336 procedures like 'generalized-vector-ref'. Generalized vectors include
337 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
339 Arrays use generalized vectors as their storage, so that you still
340 have arrays of characters, bits, etc. However, uniform-array-read!
341 and uniform-array-write can no longer read/write strings and
344 ** There is now support for copy-on-write substrings, mutation-sharing
345 substrings and read-only strings.
347 Three new procedures are related to this: substring/shared,
348 substring/copy, and substring/read-only. See the manual for more
351 ** Backtraces will now highlight the value that caused the error.
353 By default, these values are enclosed in "{...}", such as in this
362 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
363 <unnamed port>:1:1: Wrong type (expecting pair): a
364 ABORT: (wrong-type-arg)
366 The prefix and suffix used for highlighting can be set via the two new
367 printer options 'highlight-prefix' and 'highlight-suffix'. For
368 example, putting this into ~/.guile will output the bad value in bold
371 (print-set! highlight-prefix "\x1b[1m")
372 (print-set! highlight-suffix "\x1b[22m")
375 ** 'gettext' support for internationalization has been added.
377 See the manual for details.
379 ** New syntax '@' and '@@':
381 You can now directly refer to variables exported from a module by
384 (@ MODULE-NAME VARIABLE-NAME)
386 For example (@ (ice-9 pretty-print) pretty-print) will directly access
387 the pretty-print variable exported from the (ice-9 pretty-print)
388 module. You don't need to 'use' that module first. You can also use
389 '@' as a target of 'set!', as in (set! (@ mod var) val).
391 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
392 but it can also access variables that have not been exported. It is
393 intended only for kluges and temporary fixes and for debugging, not
396 ** Keyword syntax has been made more disciplined.
398 Previously, the name of a keyword was read as a 'token' but printed as
399 a symbol. Now, it is read as a general Scheme datum which must be a
410 ERROR: In expression (a b c):
414 ERROR: Unbound variable: foo
419 ERROR: Wrong type (expecting symbol): 12
423 ERROR: Wrong type (expecting symbol): (a b c)
427 ** The printing of symbols that might look like keywords can be
430 The new printer option 'quote-keywordish-symbols' controls how symbols
431 are printed that have a colon as their first or last character. The
432 default now is to only quote a symbol with #{...}# when the read
433 option 'keywords' is not '#f'. Thus:
435 guile> (define foo (string->symbol ":foo"))
436 guile> (read-set! keywords #f)
439 guile> (read-set! keywords 'prefix)
442 guile> (print-set! quote-keywordish-symbols #f)
446 ** 'while' now provides 'break' and 'continue'
448 break and continue were previously bound in a while loop, but not
449 documented, and continue didn't quite work properly. The undocumented
450 parameter to break which gave a return value for the while has been
453 ** 'call-with-current-continuation' is now also available under the name
456 ** The module system now checks for duplicate bindings.
458 The module system now can check for name conflicts among imported
461 The behavior can be controlled by specifying one or more 'duplicates'
462 handlers. For example, to make Guile return an error for every name
470 The new default behavior of the module system when a name collision
471 has been detected is to
473 1. Give priority to bindings marked as a replacement.
474 2. Issue a warning (different warning if overriding core binding).
475 3. Give priority to the last encountered binding (this corresponds to
478 If you want the old behavior back without replacements or warnings you
481 (default-duplicate-binding-handler 'last)
483 to your .guile init file.
485 ** New define-module option: :replace
487 :replace works as :export, but, in addition, marks the binding as a
490 A typical example is `format' in (ice-9 format) which is a replacement
491 for the core binding `format'.
493 ** Adding prefixes to imported bindings in the module system
495 There is now a new :use-module option :prefix. It can be used to add
496 a prefix to all imported bindings.
499 :use-module ((bar) :prefix bar:))
501 will import all bindings exported from bar, but rename them by adding
504 ** Conflicting generic functions can be automatically merged.
506 When two imported bindings conflict and they are both generic
507 functions, the two functions can now be merged automatically. This is
508 activated with the 'duplicates' handler 'merge-generics'.
510 ** New function: effective-version
512 Returns the "effective" version number. This is just the normal full
513 version string without the final micro-version number. See "Changes
514 to the distribution" above.
516 ** New threading functions: parallel, letpar, par-map, and friends
518 These are convenient ways to run calculations in parallel in new
519 threads. See "Parallel forms" in the manual for details.
521 ** New function 'try-mutex'.
523 This function will attempt to lock a mutex but will return immediately
524 instead of blocking and indicate failure.
526 ** Waiting on a condition variable can have a timeout.
528 The function 'wait-condition-variable' now takes a third, optional
529 argument that specifies the point in time where the waiting should be
532 ** New function 'broadcast-condition-variable'.
534 ** New functions 'all-threads' and 'current-thread'.
536 ** Signals and system asyncs work better with threads.
538 The function 'sigaction' now takes a fourth, optional, argument that
539 specifies the thread that the handler should run in. When the
540 argument is omitted, the handler will run in the thread that called
543 Likewise, 'system-async-mark' takes a second, optional, argument that
544 specifies the thread that the async should run in. When it is
545 omitted, the async will run in the thread that called
548 C code can use the new functions scm_sigaction_for_thread and
549 scm_system_async_mark_for_thread to pass the new thread argument.
551 When a thread blocks on a mutex, a condition variable or is waiting
552 for IO to be possible, it will still execute system asyncs. This can
553 be used to interrupt such a thread by making it execute a 'throw', for
556 ** The function 'system-async' is deprecated.
558 You can now pass any zero-argument procedure to 'system-async-mark'.
559 The function 'system-async' will just return its argument unchanged
562 ** New functions 'call-with-blocked-asyncs' and
563 'call-with-unblocked-asyncs'
565 The expression (call-with-blocked-asyncs PROC) will call PROC and will
566 block execution of system asyncs for the current thread by one level
567 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
568 procedure and will unblock the execution of system asyncs by one
569 level for the current thread.
571 Only system asyncs are affected by these functions.
573 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
575 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
576 instead. Those functions are easier to use correctly and can be
579 ** New function 'unsetenv'.
581 ** New macro 'define-syntax-public'.
583 It works like 'define-syntax' and also exports the defined macro (but
586 ** There is support for Infinity and NaNs.
588 Following PLT Scheme, Guile can now work with infinite numbers, and
591 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
592 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
593 "+nan.0"). These numbers are inexact and have no exact counterpart.
595 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
596 sign of the dividend. The infinities are integers, and they answer #t
597 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
598 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
609 ERROR: Numerical overflow
611 Two new predicates 'inf?' and 'nan?' can be used to test for the
614 ** Inexact zero can have a sign.
616 Guile can now distinguish between plus and minus inexact zero, if your
617 platform supports this, too. The two zeros are equal according to
618 '=', but not according to 'eqv?'. For example
629 ** Guile now has exact rationals.
631 Guile can now represent fractions such as 1/3 exactly. Computing with
632 them is also done exactly, of course:
637 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
640 For example: (floor 2) now returns an exact 2 where in the past it
641 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
643 ** inexact->exact no longer returns only integers.
645 Without exact rationals, the closest exact number was always an
646 integer, but now inexact->exact returns the fraction that is exactly
647 equal to a floating point number. For example:
649 (inexact->exact 1.234)
650 => 694680242521899/562949953421312
652 When you want the old behavior, use 'round' explicitly:
654 (inexact->exact (round 1.234))
657 ** New function 'rationalize'.
659 This function finds a simple fraction that is close to a given real
660 number. For example (and compare with inexact->exact above):
662 (rationalize (inexact->exact 1.234) 1/2000)
665 Note that, as required by R5RS, rationalize returns only then an exact
666 result when both its arguments are exact.
668 ** 'odd?' and 'even?' work also for inexact integers.
670 Previously, (odd? 1.0) would signal an error since only exact integers
671 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
672 returns #f and (odd? 1.5) signals an error.
674 ** Guile now has uninterned symbols.
676 The new function 'make-symbol' will return an uninterned symbol. This
677 is a symbol that is unique and is guaranteed to remain unique.
678 However, uninterned symbols can not yet be read back in.
680 Use the new function 'symbol-interned?' to check whether a symbol is
683 ** pretty-print has more options.
685 The function pretty-print from the (ice-9 pretty-print) module can now
686 also be invoked with keyword arguments that control things like
687 maximum output width. See the manual for details.
689 ** Variables have no longer a special behavior for `equal?'.
691 Previously, comparing two variables with `equal?' would recursivly
692 compare their values. This is no longer done. Variables are now only
693 `equal?' if they are `eq?'.
695 ** `(begin)' is now valid.
697 You can now use an empty `begin' form. It will yield #<unspecified>
698 when evaluated and simply be ignored in a definition context.
700 ** Deprecated: procedure->macro
702 Change your code to use 'define-macro' or r5rs macros. Also, be aware
703 that macro expansion will not be done during evaluation, but prior to
706 ** Soft ports now allow a `char-ready?' procedure
708 The vector argument to `make-soft-port' can now have a length of
709 either 5 or 6. (Previously the length had to be 5.) The optional 6th
710 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
711 that returns the number of characters that can be read immediately
712 without the soft port blocking.
714 ** Deprecated: undefine
716 There is no replacement for undefine.
718 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
719 have been discouraged.
721 They are relics from a time where a keyword like #:foo was used
722 directly as a Tcl option "-foo" and thus keywords were internally
723 stored as a symbol with a starting dash. We now store a symbol
726 Use symbol->keyword and keyword->symbol instead.
728 ** The `cheap' debug option is now obsolete
730 Evaluator trap calls are now unconditionally "cheap" - in other words,
731 they pass a debug object to the trap handler rather than a full
732 continuation. The trap handler code can capture a full continuation
733 by using `call-with-current-continuation' in the usual way, if it so
736 The `cheap' option is retained for now so as not to break existing
737 code which gets or sets it, but setting it now has no effect. It will
738 be removed in the next major Guile release.
740 ** Evaluator trap calls now support `tweaking'
742 `Tweaking' means that the trap handler code can modify the Scheme
743 expression that is about to be evaluated (in the case of an
744 enter-frame trap) or the value that is being returned (in the case of
745 an exit-frame trap). The trap handler code indicates that it wants to
746 do this by returning a pair whose car is the symbol 'instead and whose
747 cdr is the modified expression or return value.
749 * Changes to the C interface
751 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
752 take a 'delete' function argument.
754 This argument makes no sense since the delete function is used to
755 remove a pair from an alist, and this must not be configurable.
757 This is an incompatible change.
759 ** The GH interface is now subject to the deprecation mechanism
761 The GH interface has been deprecated for quite some time but now it is
762 actually removed from Guile when it is configured with
763 --disable-deprecated.
765 See the manual "Transitioning away from GH" for more information.
767 ** A new family of functions for converting between C values and
768 Scheme values has been added.
770 These functions follow a common naming scheme and are designed to be
771 easier to use, thread-safe and more future-proof than the older
776 These are predicates that return a C boolean: 1 or 0. Instead of
777 SCM_NFALSEP, you can now use scm_is_true, for example.
779 - <type> scm_to_<type> (SCM val, ...)
781 These are functions that convert a Scheme value into an appropriate
782 C value. For example, you can use scm_to_int to safely convert from
785 - SCM scm_from_<type> (<type> val, ...)
787 These functions convert from a C type to a SCM value; for example,
788 scm_from_int for ints.
790 There is a huge number of these functions, for numbers, strings,
791 symbols, vectors, etc. They are documented in the reference manual in
792 the API section together with the types that they apply to.
794 ** New functions for dealing with complex numbers in C have been added.
796 The new functions are scm_c_make_rectangular, scm_c_make_polar,
797 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
798 They work like scm_make_rectangular etc but take or return doubles
801 ** The function scm_make_complex has been discouraged.
803 Use scm_c_make_rectangular instead.
805 ** The INUM macros have been deprecated.
807 A lot of code uses these macros to do general integer conversions,
808 although the macros only work correctly with fixnums. Use the
809 following alternatives.
811 SCM_INUMP -> scm_is_integer or similar
812 SCM_NINUMP -> !scm_is_integer or similar
813 SCM_MAKINUM -> scm_from_int or similar
814 SCM_INUM -> scm_to_int or similar
816 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
817 do the validating for you.
819 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
820 have been discouraged.
822 Use the newer scm_to_<type> and scm_from_<type> functions instead for
823 new code. The functions have been discouraged since they don't fit
826 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
828 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
829 evaluates its argument twice. Use scm_is_true, etc. instead for new
832 ** The macro SCM_EQ_P has been discouraged.
834 Use scm_is_eq for new code, which fits better into the naming
837 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
840 Use the function scm_is_pair or scm_is_null instead.
842 ** The functions scm_round and scm_truncate have been deprecated and
843 are now available as scm_c_round and scm_c_truncate, respectively.
845 These functions occupy the names that scm_round_number and
846 scm_truncate_number should have.
848 ** The functions scm_c_string2str, scm_c_substring2str, and
849 scm_c_symbol2str have been deprecated.
851 Use scm_to_locale_stringbuf or similar instead, maybe together with
854 ** New functions scm_c_make_string, scm_c_string_length,
855 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
856 scm_c_substring_shared, scm_c_substring_copy.
858 These are like scm_make_string, scm_length, etc. but are slightly
859 easier to use from C.
861 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
862 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
864 They export too many assumptions about the implementation of strings
865 and symbols that are no longer true in the presence of
866 mutation-sharing substrings and when Guile switches to some form of
869 When working with strings, it is often best to use the normal string
870 functions provided by Guile, such as scm_c_string_ref,
871 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
872 manual since many more such functions are now provided than
875 When you want to convert a SCM string to a C string, use the
876 scm_to_locale_string function or similar instead. For symbols, use
877 scm_symbol_to_string and then work with that string. Because of the
878 new string representation, scm_symbol_to_string does not need to copy
879 and is thus quite efficient.
881 ** Some string, symbol and keyword functions have been discouraged.
883 They don't fit into the uniform naming scheme and are not explicit
884 about the character encoding.
886 Replace according to the following table:
888 scm_allocate_string -> scm_c_make_string
889 scm_take_str -> scm_take_locale_stringn
890 scm_take0str -> scm_take_locale_string
891 scm_mem2string -> scm_from_locale_stringn
892 scm_str2string -> scm_from_locale_string
893 scm_makfrom0str -> scm_from_locale_string
894 scm_mem2symbol -> scm_from_locale_symboln
895 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
896 scm_str2symbol -> scm_from_locale_symbol
898 SCM_SYMBOL_HASH -> scm_hashq
899 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
901 scm_c_make_keyword -> scm_from_locale_keyword
903 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
904 now also available to C code.
906 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
908 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
909 the latter returns the true name of the keyword, not the 'dash name',
910 as SCM_KEYWORDSYM used to do.
912 ** A new way to access arrays in a thread-safe and efficient way has
915 See the manual, node "Accessing Arrays From C".
917 ** The old uniform vector and bitvector implementations have been
918 unceremoniously removed.
920 This implementation exposed the details of the tagging system of
921 Guile. Use the new C API explained in the manual in node "Uniform
922 Numeric Vectors" and "Bit Vectors", respectively.
924 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
925 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
926 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
927 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
928 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
929 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
932 ** The macros dealing with vectors have been deprecated.
934 Use the new functions scm_is_vector, scm_vector_elements,
935 scm_vector_writable_elements, etc, or scm_is_simple_vector,
936 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
937 manual for more details.
939 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
940 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
942 The following macros have been removed: SCM_VECTOR_BASE,
943 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
944 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
946 ** Some C functions and macros related to arrays have been deprecated.
948 Migrate according to the following table:
950 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
951 scm_make_ra -> scm_make_array
952 scm_shap2ra -> scm_make_array
953 scm_cvref -> scm_c_generalized_vector_ref
954 scm_ra_set_contp -> do not use
955 scm_aind -> scm_array_handle_pos
956 scm_raprin1 -> scm_display or scm_write
958 SCM_ARRAYP -> scm_is_array
959 SCM_ARRAY_NDIM -> scm_c_array_rank
960 SCM_ARRAY_DIMS -> scm_array_handle_dims
961 SCM_ARRAY_CONTP -> do not use
962 SCM_ARRAY_MEM -> do not use
963 SCM_ARRAY_V -> scm_array_handle_elements or similar
964 SCM_ARRAY_BASE -> do not use
966 ** SCM_CELL_WORD_LOC has been deprecated.
968 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
969 to a SCM, as opposed to a pointer to a scm_t_bits.
971 This was done to allow the correct use of pointers into the Scheme
972 heap. Previously, the heap words were of type scm_t_bits and local
973 variables and function arguments were of type SCM, making it
974 non-standards-conformant to have a pointer that can point to both.
976 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
978 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
979 second and third words of double smobs. Likewise for
980 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
982 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
983 used to get and set the 16 exra bits in the zeroth word of a smob.
985 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
986 accesing the first immediate word of a smob as a SCM value, and there
987 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
988 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
990 ** New way to deal with non-local exits and re-entries.
992 There is a new set of functions that essentially do what
993 scm_internal_dynamic_wind does, but in a way that is more convenient
994 for C code in some situations. Here is a quick example of how to
995 prevent a potential memory leak:
1002 scm_dynwind_begin (0);
1004 mem = scm_malloc (100);
1005 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1007 /* MEM would leak if BAR throws an error.
1008 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1015 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1016 SCM_DYNWIND_END as well.
1020 For full documentation, see the node "Dynamic Wind" in the manual.
1022 ** New function scm_dynwind_free
1024 This function calls 'free' on a given pointer when a dynwind context
1025 is left. Thus the call to scm_dynwind_unwind_handler above could be
1026 replaced with simply scm_dynwind_free (mem).
1028 ** New functions scm_c_call_with_blocked_asyncs and
1029 scm_c_call_with_unblocked_asyncs
1031 Like scm_call_with_blocked_asyncs etc. but for C functions.
1033 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1035 In addition to scm_c_call_with_blocked_asyncs you can now also use
1036 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1037 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1039 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1040 SCM_REALLOW_INTS have been deprecated.
1042 They do no longer fulfill their original role of blocking signal
1043 delivery. Depending on what you want to achieve, replace a pair of
1044 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1045 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1048 ** The value 'scm_mask_ints' is no longer writable.
1050 Previously, you could set scm_mask_ints directly. This is no longer
1051 possible. Use scm_c_call_with_blocked_asyncs and
1052 scm_c_call_with_unblocked_asyncs instead.
1054 ** New way to temporarily set the current input, output or error ports
1056 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1057 context' (see above). <foo> is one of "input", "output" or "error".
1059 ** New way to temporarily set fluids
1061 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1062 above) to temporarily set the value of a fluid.
1064 ** New types scm_t_intmax and scm_t_uintmax.
1066 On platforms that have them, these types are identical to intmax_t and
1067 uintmax_t, respectively. On other platforms, they are identical to
1068 the largest integer types that Guile knows about.
1070 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1072 You should not have used them.
1074 ** Many public #defines with generic names have been made private.
1076 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1077 private or renamed with a more suitable public name.
1079 ** The macro SCM_TYP16S has been deprecated.
1081 This macro is not intended for public use.
1083 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1085 Use scm_is_true (scm_inexact_p (...)) instead.
1087 ** The macro SCM_SLOPPY_REALP has been deprecated.
1089 Use scm_is_real instead.
1091 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1093 Use scm_is_complex instead.
1095 ** Some preprocessor defines have been deprecated.
1097 These defines indicated whether a certain feature was present in Guile
1098 or not. Going forward, assume that the features are always present.
1100 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1101 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1103 The following macros have been removed completely: MEMOIZE_LOCALS,
1104 SCM_RECKLESS, SCM_CAUTIOUS.
1106 ** The preprocessor define STACK_DIRECTION has been deprecated.
1108 There should be no need to know about the stack direction for ordinary
1111 ** New function: scm_effective_version
1113 Returns the "effective" version number. This is just the normal full
1114 version string without the final micro-version number. See "Changes
1115 to the distribution" above.
1117 ** The function scm_call_with_new_thread has a new prototype.
1119 Instead of taking a list with the thunk and handler, these two
1120 arguments are now passed directly:
1122 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1124 This is an incompatible change.
1126 ** New snarfer macro SCM_DEFINE_PUBLIC.
1128 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1129 function in the init section.
1131 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1133 ** Garbage collector rewrite.
1135 The garbage collector is cleaned up a lot, and now uses lazy
1136 sweeping. This is reflected in the output of (gc-stats); since cells
1137 are being freed when they are allocated, the cells-allocated field
1138 stays roughly constant.
1140 For malloc related triggers, the behavior is changed. It uses the same
1141 heuristic as the cell-triggered collections. It may be tuned with the
1142 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1143 for minimum yield of malloc related triggers. The default is 40.
1144 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1147 Debugging operations for the freelist have been deprecated, along with
1148 the C variables that control garbage collection. The environment
1149 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1150 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1152 For understanding the memory usage of a GUILE program, the routine
1153 gc-live-object-stats returns an alist containing the number of live
1154 objects for every type.
1157 ** The function scm_definedp has been renamed to scm_defined_p
1159 The name scm_definedp is deprecated.
1161 ** The struct scm_cell type has been renamed to scm_t_cell
1163 This is in accordance to Guile's naming scheme for types. Note that
1164 the name scm_cell is now used for a function that allocates and
1165 initializes a new cell (see below).
1167 ** New functions for memory management
1169 A new set of functions for memory management has been added since the
1170 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1171 indeed, Guile itself contained some long standing bugs that could
1172 cause aborts in long running programs.
1174 The new functions are more symmetrical and do not need cooperation
1175 from smob free routines, among other improvements.
1177 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1178 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1179 scm_gc_free, scm_gc_register_collectable_memory, and
1180 scm_gc_unregister_collectable_memory. Refer to the manual for more
1181 details and for upgrading instructions.
1183 The old functions for memory management have been deprecated. They
1184 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1185 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1187 ** Declarations of exported features are marked with SCM_API.
1189 Every declaration of a feature that belongs to the exported Guile API
1190 has been marked by adding the macro "SCM_API" to the start of the
1191 declaration. This macro can expand into different things, the most
1192 common of which is just "extern" for Unix platforms. On Win32, it can
1193 be used to control which symbols are exported from a DLL.
1195 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1196 will expand into "__declspec (dllimport) extern", which is needed for
1197 linking to the Guile DLL in Windows.
1199 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1200 SCM_SRFI4_IMPORT, for the corresponding libraries.
1202 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1204 Use the new functions scm_cell and scm_double_cell instead. The old
1205 macros had problems because with them allocation and initialization
1206 was separated and the GC could sometimes observe half initialized
1207 cells. Only careful coding by the user of SCM_NEWCELL and
1208 SCM_NEWCELL2 could make this safe and efficient.
1210 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1212 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1215 ** SRCBRKP has been deprecated.
1217 Use scm_c_source_property_breakpoint_p instead.
1219 ** Deprecated: scm_makmacro
1221 Change your code to use either scm_makmmacro or to define macros in
1222 Scheme, using 'define-macro'.
1224 ** New function scm_c_port_for_each.
1226 This function is like scm_port_for_each but takes a pointer to a C
1227 function as the callback instead of a SCM value.
1229 ** The names scm_internal_select, scm_thread_sleep, and
1230 scm_thread_usleep have been discouraged.
1232 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1234 ** The GC can no longer be blocked.
1236 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1237 The GC can now run (partially) concurrently with other code and thus
1238 blocking it is not well defined.
1240 ** Many definitions have been removed that were previously deprecated.
1242 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1243 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1244 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1245 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1246 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1247 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1248 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1249 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1250 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1251 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1252 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1253 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1254 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1255 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1256 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1257 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1258 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1259 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1260 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1261 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1262 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1263 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1264 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1265 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1266 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1267 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1268 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1269 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1270 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1271 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1272 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1274 * Changes to bundled modules
1278 Using the (ice-9 debug) module no longer automatically switches Guile
1279 to use the debugging evaluator. If you want to switch to the
1280 debugging evaluator (which is needed for backtrace information if you
1281 hit an error), please add an explicit "(debug-enable 'debug)" to your
1282 code just after the code to use (ice-9 debug).
1285 Changes since Guile 1.4:
1287 * Changes to the distribution
1289 ** A top-level TODO file is included.
1291 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1293 Guile now always uses three numbers to represent the version,
1294 i.e. "1.6.5". The first number, 1, is the major version number, the
1295 second number, 6, is the minor version number, and the third number,
1296 5, is the micro version number. Changes in major version number
1297 indicate major changes in Guile.
1299 Minor version numbers that are even denote stable releases, and odd
1300 minor version numbers denote development versions (which may be
1301 unstable). The micro version number indicates a minor sub-revision of
1302 a given MAJOR.MINOR release.
1304 In keeping with the new scheme, (minor-version) and scm_minor_version
1305 no longer return everything but the major version number. They now
1306 just return the minor version number. Two new functions
1307 (micro-version) and scm_micro_version have been added to report the
1308 micro version number.
1310 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1312 ** New preprocessor definitions are available for checking versions.
1314 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1315 SCM_MICRO_VERSION to the appropriate integer values.
1317 ** Guile now actively warns about deprecated features.
1319 The new configure option `--enable-deprecated=LEVEL' and the
1320 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1321 See INSTALL and README for more information.
1323 ** Guile is much more likely to work on 64-bit architectures.
1325 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1326 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1327 for the use of a test machine, and thanks to Stefan Jahn for ia64
1330 ** New functions: setitimer and getitimer.
1332 These implement a fairly direct interface to the libc functions of the
1335 ** The #. reader extension is now disabled by default.
1337 For safety reasons, #. evaluation is disabled by default. To
1338 re-enable it, set the fluid read-eval? to #t. For example:
1340 (fluid-set! read-eval? #t)
1342 but make sure you realize the potential security risks involved. With
1343 read-eval? enabled, reading a data file from an untrusted source can
1346 ** New SRFI modules have been added:
1348 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1351 (srfi srfi-1) is a library containing many useful pair- and list-processing
1354 (srfi srfi-2) exports and-let*.
1356 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1358 (srfi srfi-6) is a dummy module for now, since guile already provides
1359 all of the srfi-6 procedures by default: open-input-string,
1360 open-output-string, get-output-string.
1362 (srfi srfi-8) exports receive.
1364 (srfi srfi-9) exports define-record-type.
1366 (srfi srfi-10) exports define-reader-ctor and implements the reader
1369 (srfi srfi-11) exports let-values and let*-values.
1371 (srfi srfi-13) implements the SRFI String Library.
1373 (srfi srfi-14) implements the SRFI Character-Set Library.
1375 (srfi srfi-17) implements setter and getter-with-setter and redefines
1376 some accessor procedures as procedures with getters. (such as car,
1377 cdr, vector-ref etc.)
1379 (srfi srfi-19) implements the SRFI Time/Date Library.
1381 ** New scripts / "executable modules"
1383 Subdirectory "scripts" contains Scheme modules that are packaged to
1384 also be executable as scripts. At this time, these scripts are available:
1393 See README there for more info.
1395 These scripts can be invoked from the shell with the new program
1396 "guile-tools", which keeps track of installation directory for you.
1399 $ guile-tools display-commentary srfi/*.scm
1401 guile-tools is copied to the standard $bindir on "make install".
1403 ** New module (ice-9 stack-catch):
1405 stack-catch is like catch, but saves the current state of the stack in
1406 the fluid the-last-stack. This fluid can be useful when using the
1407 debugger and when re-throwing an error.
1409 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1411 This has been done to prevent problems on lesser operating systems
1412 that can't tolerate `*'s in file names. The exported macro continues
1413 to be named `and-let*', of course.
1415 On systems that support it, there is also a compatibility module named
1416 (ice-9 and-let*). It will go away in the next release.
1418 ** New modules (oop goops) etc.:
1421 (oop goops describe)
1423 (oop goops active-slot)
1424 (oop goops composite-slot)
1426 The Guile Object Oriented Programming System (GOOPS) has been
1427 integrated into Guile. For further information, consult the GOOPS
1428 manual and tutorial in the `doc' directory.
1430 ** New module (ice-9 rdelim).
1432 This exports the following procedures which were previously defined
1433 in the default environment:
1435 read-line read-line! read-delimited read-delimited! %read-delimited!
1436 %read-line write-line
1438 For backwards compatibility the definitions are still imported into the
1439 default environment in this version of Guile. However you should add:
1441 (use-modules (ice-9 rdelim))
1443 to any program which uses the definitions, since this may change in
1446 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1447 can be used for similar functionality.
1449 ** New module (ice-9 rw)
1451 This is a subset of the (scsh rw) module from guile-scsh. Currently
1452 it defines two procedures:
1454 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1456 Read characters from a port or file descriptor into a string STR.
1457 A port must have an underlying file descriptor -- a so-called
1458 fport. This procedure is scsh-compatible and can efficiently read
1461 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1463 Write characters from a string STR to a port or file descriptor.
1464 A port must have an underlying file descriptor -- a so-called
1465 fport. This procedure is mostly compatible and can efficiently
1466 write large strings.
1468 ** New module (ice-9 match)
1470 This module includes Andrew K. Wright's pattern matcher. See
1471 ice-9/match.scm for brief description or
1473 http://www.star-lab.com/wright/code.html
1475 for complete documentation.
1477 ** New module (ice-9 buffered-input)
1479 This module provides procedures to construct an input port from an
1480 underlying source of input that reads and returns its input in chunks.
1481 The underlying input source is a Scheme procedure, specified by the
1482 caller, which the port invokes whenever it needs more input.
1484 This is useful when building an input port whose back end is Readline
1485 or a UI element such as the GtkEntry widget.
1489 The reference and tutorial documentation that was previously
1490 distributed separately, as `guile-doc', is now included in the core
1491 Guile distribution. The documentation consists of the following
1494 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1497 - The Guile Reference Manual (guile.texi) contains (or is intended to
1498 contain) reference documentation on all aspects of Guile.
1500 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1501 reference documentation for using GOOPS, Guile's Object Oriented
1504 - The Revised^5 Report on the Algorithmic Language Scheme
1507 See the README file in the `doc' directory for more details.
1509 ** There are a couple of examples in the examples/ directory now.
1511 * Changes to the stand-alone interpreter
1513 ** New command line option `--use-srfi'
1515 Using this option, SRFI modules can be loaded on startup and be
1516 available right from the beginning. This makes programming portable
1517 Scheme programs easier.
1519 The option `--use-srfi' expects a comma-separated list of numbers,
1520 each representing a SRFI number to be loaded into the interpreter
1521 before starting evaluating a script file or the REPL. Additionally,
1522 the feature identifier for the loaded SRFIs is recognized by
1523 `cond-expand' when using this option.
1526 $ guile --use-srfi=8,13
1527 guile> (receive (x z) (values 1 2) (+ 1 2))
1529 guile> (string-pad "bla" 20)
1532 ** Guile now always starts up in the `(guile-user)' module.
1534 Previously, scripts executed via the `-s' option would run in the
1535 `(guile)' module and the repl would run in the `(guile-user)' module.
1536 Now every user action takes place in the `(guile-user)' module by
1539 * Changes to Scheme functions and syntax
1541 ** Character classifiers work for non-ASCII characters.
1543 The predicates `char-alphabetic?', `char-numeric?',
1544 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1545 no longer check whether their arguments are ASCII characters.
1546 Previously, a character would only be considered alphabetic when it
1547 was also ASCII, for example.
1549 ** Previously deprecated Scheme functions have been removed:
1551 tag - no replacement.
1552 fseek - replaced by seek.
1553 list* - replaced by cons*.
1555 ** It's now possible to create modules with controlled environments
1559 (use-modules (ice-9 safe))
1560 (define m (make-safe-module))
1561 ;;; m will now be a module containing only a safe subset of R5RS
1562 (eval '(+ 1 2) m) --> 3
1563 (eval 'load m) --> ERROR: Unbound variable: load
1565 ** Evaluation of "()", the empty list, is now an error.
1567 Previously, the expression "()" evaluated to the empty list. This has
1568 been changed to signal a "missing expression" error. The correct way
1569 to write the empty list as a literal constant is to use quote: "'()".
1571 ** New concept of `Guile Extensions'.
1573 A Guile Extension is just a ordinary shared library that can be linked
1574 at run-time. We found it advantageous to give this simple concept a
1575 dedicated name to distinguish the issues related to shared libraries
1576 from the issues related to the module system.
1578 *** New function: load-extension
1580 Executing (load-extension lib init) is mostly equivalent to
1582 (dynamic-call init (dynamic-link lib))
1584 except when scm_register_extension has been called previously.
1585 Whenever appropriate, you should use `load-extension' instead of
1586 dynamic-link and dynamic-call.
1588 *** New C function: scm_c_register_extension
1590 This function registers a initialization function for use by
1591 `load-extension'. Use it when you don't want specific extensions to
1592 be loaded as shared libraries (for example on platforms that don't
1593 support dynamic linking).
1595 ** Auto-loading of compiled-code modules is deprecated.
1597 Guile used to be able to automatically find and link a shared
1598 library to satisfy requests for a module. For example, the module
1599 `(foo bar)' could be implemented by placing a shared library named
1600 "foo/libbar.so" (or with a different extension) in a directory on the
1603 This has been found to be too tricky, and is no longer supported. The
1604 shared libraries are now called "extensions". You should now write a
1605 small Scheme file that calls `load-extension' to load the shared
1606 library and initialize it explicitly.
1608 The shared libraries themselves should be installed in the usual
1609 places for shared libraries, with names like "libguile-foo-bar".
1611 For example, place this into a file "foo/bar.scm"
1613 (define-module (foo bar))
1615 (load-extension "libguile-foo-bar" "foobar_init")
1617 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1619 `eval' is now R5RS, that is it takes two arguments.
1620 The second argument is an environment specifier, i.e. either
1622 (scheme-report-environment 5)
1623 (null-environment 5)
1624 (interaction-environment)
1630 ** The module system has been made more disciplined.
1632 The function `eval' will save and restore the current module around
1633 the evaluation of the specified expression. While this expression is
1634 evaluated, `(current-module)' will now return the right module, which
1635 is the module specified as the second argument to `eval'.
1637 A consequence of this change is that `eval' is not particularly
1638 useful when you want allow the evaluated code to change what module is
1639 designated as the current module and have this change persist from one
1640 call to `eval' to the next. The read-eval-print-loop is an example
1641 where `eval' is now inadequate. To compensate, there is a new
1642 function `primitive-eval' that does not take a module specifier and
1643 that does not save/restore the current module. You should use this
1644 function together with `set-current-module', `current-module', etc
1645 when you want to have more control over the state that is carried from
1646 one eval to the next.
1648 Additionally, it has been made sure that forms that are evaluated at
1649 the top level are always evaluated with respect to the current module.
1650 Previously, subforms of top-level forms such as `begin', `case',
1651 etc. did not respect changes to the current module although these
1652 subforms are at the top-level as well.
1654 To prevent strange behavior, the forms `define-module',
1655 `use-modules', `use-syntax', and `export' have been restricted to only
1656 work on the top level. The forms `define-public' and
1657 `defmacro-public' only export the new binding on the top level. They
1658 behave just like `define' and `defmacro', respectively, when they are
1659 used in a lexical environment.
1661 Also, `export' will no longer silently re-export bindings imported
1662 from a used module. It will emit a `deprecation' warning and will
1663 cease to perform any re-export in the next version. If you actually
1664 want to re-export bindings, use the new `re-export' in place of
1665 `export'. The new `re-export' will not make copies of variables when
1666 rexporting them, as `export' did wrongly.
1668 ** Module system now allows selection and renaming of imported bindings
1670 Previously, when using `use-modules' or the `#:use-module' clause in
1671 the `define-module' form, all the bindings (association of symbols to
1672 values) for imported modules were added to the "current module" on an
1673 as-is basis. This has been changed to allow finer control through two
1674 new facilities: selection and renaming.
1676 You can now select which of the imported module's bindings are to be
1677 visible in the current module by using the `:select' clause. This
1678 clause also can be used to rename individual bindings. For example:
1680 ;; import all bindings no questions asked
1681 (use-modules (ice-9 common-list))
1683 ;; import four bindings, renaming two of them;
1684 ;; the current module sees: every some zonk-y zonk-n
1685 (use-modules ((ice-9 common-list)
1687 (remove-if . zonk-y)
1688 (remove-if-not . zonk-n))))
1690 You can also programmatically rename all selected bindings using the
1691 `:renamer' clause, which specifies a proc that takes a symbol and
1692 returns another symbol. Because it is common practice to use a prefix,
1693 we now provide the convenience procedure `symbol-prefix-proc'. For
1696 ;; import four bindings, renaming two of them specifically,
1697 ;; and all four w/ prefix "CL:";
1698 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1699 (use-modules ((ice-9 common-list)
1701 (remove-if . zonk-y)
1702 (remove-if-not . zonk-n))
1703 :renamer (symbol-prefix-proc 'CL:)))
1705 ;; import four bindings, renaming two of them specifically,
1706 ;; and all four by upcasing.
1707 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1708 (define (upcase-symbol sym)
1709 (string->symbol (string-upcase (symbol->string sym))))
1711 (use-modules ((ice-9 common-list)
1713 (remove-if . zonk-y)
1714 (remove-if-not . zonk-n))
1715 :renamer upcase-symbol))
1717 Note that programmatic renaming is done *after* individual renaming.
1718 Also, the above examples show `use-modules', but the same facilities are
1719 available for the `#:use-module' clause of `define-module'.
1721 See manual for more info.
1723 ** The semantics of guardians have changed.
1725 The changes are for the most part compatible. An important criterion
1726 was to keep the typical usage of guardians as simple as before, but to
1727 make the semantics safer and (as a result) more useful.
1729 *** All objects returned from guardians are now properly alive.
1731 It is now guaranteed that any object referenced by an object returned
1732 from a guardian is alive. It's now impossible for a guardian to
1733 return a "contained" object before its "containing" object.
1735 One incompatible (but probably not very important) change resulting
1736 from this is that it is no longer possible to guard objects that
1737 indirectly reference themselves (i.e. are parts of cycles). If you do
1738 so accidentally, you'll get a warning.
1740 *** There are now two types of guardians: greedy and sharing.
1742 If you call (make-guardian #t) or just (make-guardian), you'll get a
1743 greedy guardian, and for (make-guardian #f) a sharing guardian.
1745 Greedy guardians are the default because they are more "defensive".
1746 You can only greedily guard an object once. If you guard an object
1747 more than once, once in a greedy guardian and the rest of times in
1748 sharing guardians, then it is guaranteed that the object won't be
1749 returned from sharing guardians as long as it is greedily guarded
1752 Guardians returned by calls to `make-guardian' can now take one more
1753 optional parameter, which says whether to throw an error in case an
1754 attempt is made to greedily guard an object that is already greedily
1755 guarded. The default is true, i.e. throw an error. If the parameter
1756 is false, the guardian invocation returns #t if guarding was
1757 successful and #f if it wasn't.
1759 Also, since greedy guarding is, in effect, a side-effecting operation
1760 on objects, a new function is introduced: `destroy-guardian!'.
1761 Invoking this function on a guardian renders it unoperative and, if
1762 the guardian is greedy, clears the "greedily guarded" property of the
1763 objects that were guarded by it, thus undoing the side effect.
1765 Note that all this hair is hardly very important, since guardian
1766 objects are usually permanent.
1768 ** Continuations created by call-with-current-continuation now accept
1769 any number of arguments, as required by R5RS.
1771 ** New function `issue-deprecation-warning'
1773 This function is used to display the deprecation messages that are
1774 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1777 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1781 ;; `id' is deprecated. Use `identity' instead.
1786 ** New syntax `begin-deprecated'
1788 When deprecated features are included (as determined by the configure
1789 option --enable-deprecated), `begin-deprecated' is identical to
1790 `begin'. When deprecated features are excluded, it always evaluates
1791 to `#f', ignoring the body forms.
1793 ** New function `make-object-property'
1795 This function returns a new `procedure with setter' P that can be used
1796 to attach a property to objects. When calling P as
1800 where `obj' is any kind of object, it attaches `val' to `obj' in such
1801 a way that it can be retrieved by calling P as
1805 This function will replace procedure properties, symbol properties and
1806 source properties eventually.
1808 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1810 Instead of #&optional, #&key, etc you should now use #:optional,
1811 #:key, etc. Since #:optional is a keyword, you can write it as just
1812 :optional when (read-set! keywords 'prefix) is active.
1814 The old reader syntax `#&' is still supported, but deprecated. It
1815 will be removed in the next release.
1817 ** New define-module option: pure
1819 Tells the module system not to include any bindings from the root
1824 (define-module (totally-empty-module)
1827 ** New define-module option: export NAME1 ...
1829 Export names NAME1 ...
1831 This option is required if you want to be able to export bindings from
1832 a module which doesn't import one of `define-public' or `export'.
1836 (define-module (foo)
1838 :use-module (ice-9 r5rs)
1841 ;;; Note that we're pure R5RS below this point!
1846 ** New function: object->string OBJ
1848 Return a Scheme string obtained by printing a given object.
1850 ** New function: port? X
1852 Returns a boolean indicating whether X is a port. Equivalent to
1853 `(or (input-port? X) (output-port? X))'.
1855 ** New function: file-port?
1857 Determines whether a given object is a port that is related to a file.
1859 ** New function: port-for-each proc
1861 Apply PROC to each port in the Guile port table in turn. The return
1862 value is unspecified. More specifically, PROC is applied exactly once
1863 to every port that exists in the system at the time PORT-FOR-EACH is
1864 invoked. Changes to the port table while PORT-FOR-EACH is running
1865 have no effect as far as PORT-FOR-EACH is concerned.
1867 ** New function: dup2 oldfd newfd
1869 A simple wrapper for the `dup2' system call. Copies the file
1870 descriptor OLDFD to descriptor number NEWFD, replacing the
1871 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1872 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1873 to move away ports which are using NEWFD. The return value is
1876 ** New function: close-fdes fd
1878 A simple wrapper for the `close' system call. Close file
1879 descriptor FD, which must be an integer. Unlike close (*note
1880 close: Ports and File Descriptors.), the file descriptor will be
1881 closed even if a port is using it. The return value is
1884 ** New function: crypt password salt
1886 Encrypts `password' using the standard unix password encryption
1889 ** New function: chroot path
1891 Change the root directory of the running process to `path'.
1893 ** New functions: getlogin, cuserid
1895 Return the login name or the user name of the current effective user
1898 ** New functions: getpriority which who, setpriority which who prio
1900 Get or set the priority of the running process.
1902 ** New function: getpass prompt
1904 Read a password from the terminal, first displaying `prompt' and
1907 ** New function: flock file operation
1909 Set/remove an advisory shared or exclusive lock on `file'.
1911 ** New functions: sethostname name, gethostname
1913 Set or get the hostname of the machine the current process is running
1916 ** New function: mkstemp! tmpl
1918 mkstemp creates a new unique file in the file system and returns a
1919 new buffered port open for reading and writing to the file. TMPL
1920 is a string specifying where the file should be created: it must
1921 end with `XXXXXX' and will be changed in place to return the name
1922 of the temporary file.
1924 ** New function: open-input-string string
1926 Return an input string port which delivers the characters from
1927 `string'. This procedure, together with `open-output-string' and
1928 `get-output-string' implements SRFI-6.
1930 ** New function: open-output-string
1932 Return an output string port which collects all data written to it.
1933 The data can then be retrieved by `get-output-string'.
1935 ** New function: get-output-string
1937 Return the contents of an output string port.
1939 ** New function: identity
1941 Return the argument.
1943 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1944 are represented in Scheme as integers with normal host byte ordering.
1946 ** New function: inet-pton family address
1948 Convert a printable string network address into an integer. Note that
1949 unlike the C version of this function, the result is an integer with
1950 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1953 (inet-pton AF_INET "127.0.0.1") => 2130706433
1954 (inet-pton AF_INET6 "::1") => 1
1956 ** New function: inet-ntop family address
1958 Convert an integer network address into a printable string. Note that
1959 unlike the C version of this function, the input is an integer with
1960 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1963 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1964 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1965 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1969 Use `identity' instead.
1975 ** Deprecated: return-it
1979 ** Deprecated: string-character-length
1981 Use `string-length' instead.
1983 ** Deprecated: flags
1985 Use `logior' instead.
1987 ** Deprecated: close-all-ports-except.
1989 This was intended for closing ports in a child process after a fork,
1990 but it has the undesirable side effect of flushing buffers.
1991 port-for-each is more flexible.
1993 ** The (ice-9 popen) module now attempts to set up file descriptors in
1994 the child process from the current Scheme ports, instead of using the
1995 current values of file descriptors 0, 1, and 2 in the parent process.
1997 ** Removed function: builtin-weak-bindings
1999 There is no such concept as a weak binding any more.
2001 ** Removed constants: bignum-radix, scm-line-incrementors
2003 ** define-method: New syntax mandatory.
2005 The new method syntax is now mandatory:
2007 (define-method (NAME ARG-SPEC ...) BODY ...)
2008 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2010 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2011 REST-ARG ::= ARG-NAME
2013 If you have old code using the old syntax, import
2014 (oop goops old-define-method) before (oop goops) as in:
2016 (use-modules (oop goops old-define-method) (oop goops))
2018 ** Deprecated function: builtin-variable
2019 Removed function: builtin-bindings
2021 There is no longer a distinction between builtin or other variables.
2022 Use module system operations for all variables.
2024 ** Lazy-catch handlers are no longer allowed to return.
2026 That is, a call to `throw', `error', etc is now guaranteed to not
2029 ** Bugfixes for (ice-9 getopt-long)
2031 This module is now tested using test-suite/tests/getopt-long.test.
2032 The following bugs have been fixed:
2034 *** Parsing for options that are specified to have `optional' args now checks
2035 if the next element is an option instead of unconditionally taking it as the
2038 *** An error is now thrown for `--opt=val' when the option description
2039 does not specify `(value #t)' or `(value optional)'. This condition used to
2040 be accepted w/o error, contrary to the documentation.
2042 *** The error message for unrecognized options is now more informative.
2043 It used to be "not a record", an artifact of the implementation.
2045 *** The error message for `--opt' terminating the arg list (no value), when
2046 `(value #t)' is specified, is now more informative. It used to be "not enough
2049 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2050 The expansion used to be like so:
2052 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2054 Note that the "5d" is dropped. Now it is like so:
2056 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2058 This enables single-char options to have adjoining arguments as long as their
2059 constituent characters are not potential single-char options.
2061 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2063 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2064 property `arglist', which can be retrieved by `arity'. The result is that
2065 `arity' can give more detailed information than before:
2069 guile> (use-modules (ice-9 optargs))
2070 guile> (define* (foo #:optional a b c) a)
2072 0 or more arguments in `lambda*:G0'.
2077 3 optional arguments: `a', `b' and `c'.
2078 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2080 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2081 and `d', other keywords allowed.
2082 guile> (define* (baz a b #:optional c #:rest r) a)
2084 2 required arguments: `a' and `b', 1 optional argument: `c',
2087 * Changes to the C interface
2089 ** Types have been renamed from scm_*_t to scm_t_*.
2091 This has been done for POSIX sake. It reserves identifiers ending
2092 with "_t". What a concept.
2094 The old names are still available with status `deprecated'.
2096 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2098 ** Deprecated features have been removed.
2102 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2103 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2105 *** C Functions removed
2107 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2108 scm_fseek - replaced by scm_seek.
2109 gc-thunk - replaced by after-gc-hook.
2110 gh_int2scmb - replaced by gh_bool2scm.
2111 scm_tc_dblr - replaced by scm_tc16_real.
2112 scm_tc_dblc - replaced by scm_tc16_complex.
2113 scm_list_star - replaced by scm_cons_star.
2115 ** Deprecated: scm_makfromstr
2117 Use scm_mem2string instead.
2119 ** Deprecated: scm_make_shared_substring
2121 Explicit shared substrings will disappear from Guile.
2123 Instead, "normal" strings will be implemented using sharing
2124 internally, combined with a copy-on-write strategy.
2126 ** Deprecated: scm_read_only_string_p
2128 The concept of read-only strings will disappear in next release of
2131 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2133 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2135 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2137 Call a procedure with the indicated number of arguments. See "Fly
2138 Evaluation" in the manual.
2140 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2142 Call a procedure with the indicated number of arguments and a list of
2143 further arguments. See "Fly Evaluation" in the manual.
2145 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2147 Create a list of the given number of elements. See "List
2148 Constructors" in the manual.
2150 ** Renamed function: scm_listify has been replaced by scm_list_n.
2152 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2153 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2155 Use functions scm_list_N instead.
2157 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2159 Used by an application to read arbitrary number of bytes from a port.
2160 Same semantics as libc read, except that scm_c_read only returns less
2161 than SIZE bytes if at end-of-file.
2163 Warning: Doesn't update port line and column counts!
2165 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2167 Used by an application to write arbitrary number of bytes to an SCM
2168 port. Similar semantics as libc write. However, unlike libc
2169 write, scm_c_write writes the requested number of bytes and has no
2172 Warning: Doesn't update port line and column counts!
2174 ** New function: scm_init_guile ()
2176 In contrast to scm_boot_guile, scm_init_guile will return normally
2177 after initializing Guile. It is not available on all systems, tho.
2179 ** New functions: scm_str2symbol, scm_mem2symbol
2181 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2182 field of characters and creates a scheme symbol object from that C string.
2183 The function scm_mem2symbol takes a const char* and a number of characters and
2184 creates a symbol from the characters in that memory area.
2186 ** New functions: scm_primitive_make_property
2187 scm_primitive_property_ref
2188 scm_primitive_property_set_x
2189 scm_primitive_property_del_x
2191 These functions implement a new way to deal with object properties.
2192 See libguile/properties.c for their documentation.
2194 ** New function: scm_done_free (long size)
2196 This function is the inverse of scm_done_malloc. Use it to report the
2197 amount of smob memory you free. The previous method, which involved
2198 calling scm_done_malloc with negative argument, was somewhat
2199 unintuitive (and is still available, of course).
2201 ** New function: scm_c_memq (SCM obj, SCM list)
2203 This function provides a fast C level alternative for scm_memq for the case
2204 that the list parameter is known to be a proper list. The function is a
2205 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2206 list input parameter, since for anything else but a proper list the function's
2207 behaviour is undefined - it may even crash or loop endlessly. Further, for
2208 the case that the object is not found in the list, scm_c_memq returns #f which
2209 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2211 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2212 scm_remember_upto_here
2214 These functions replace the function scm_remember.
2216 ** Deprecated function: scm_remember
2218 Use one of the new functions scm_remember_upto_here_1,
2219 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2221 ** New function: scm_allocate_string
2223 This function replaces the function scm_makstr.
2225 ** Deprecated function: scm_makstr
2227 Use the new function scm_allocate_string instead.
2229 ** New global variable scm_gc_running_p introduced.
2231 Use this variable to find out if garbage collection is being executed. Up to
2232 now applications have used scm_gc_heap_lock to test if garbage collection was
2233 running, which also works because of the fact that up to know only the garbage
2234 collector has set this variable. But, this is an implementation detail that
2235 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2236 of this variable is (and has been) not fully safe anyway.
2238 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2240 Use these instead of SCM_LENGTH_MAX.
2242 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2243 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2244 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2246 Use these instead of SCM_LENGTH.
2248 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2249 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2250 SCM_SET_BITVECTOR_LENGTH
2252 Use these instead of SCM_SETLENGTH
2254 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2255 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2258 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2261 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2262 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2265 Use these instead of SCM_SETCHARS.
2267 ** New macro: SCM_BITVECTOR_P
2269 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2271 Use instead of SCM_COERCE_SUBSTR.
2273 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2275 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2277 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2278 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2279 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2280 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2281 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2282 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2283 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2284 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2285 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2286 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2287 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2288 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2289 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2290 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2291 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2293 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2294 Use scm_memory_error instead of SCM_NALLOC.
2295 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2296 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2297 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2298 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2299 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2300 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2301 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2302 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2303 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2304 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2305 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2306 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2307 Use a type specific setter macro instead of SCM_SETLENGTH.
2308 Use a type specific setter macro instead of SCM_SETCHARS.
2309 Use a type specific length macro instead of SCM_LENGTH_MAX.
2310 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2311 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2312 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2313 Use SCM_TYP16 instead of SCM_GCTYP16.
2314 Use SCM_CDR instead of SCM_GCCDR.
2315 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2316 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2317 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2318 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2319 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2320 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2321 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2323 ** Removed function: scm_struct_init
2325 ** Removed variable: scm_symhash_dim
2327 ** Renamed function: scm_make_cont has been replaced by
2328 scm_make_continuation, which has a different interface.
2330 ** Deprecated function: scm_call_catching_errors
2332 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2334 ** Deprecated function: scm_strhash
2336 Use scm_string_hash instead.
2338 ** Deprecated function: scm_vector_set_length_x
2340 Instead, create a fresh vector of the desired size and copy the contents.
2342 ** scm_gensym has changed prototype
2344 scm_gensym now only takes one argument.
2346 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2349 There is now only a single symbol type scm_tc7_symbol.
2350 The tag scm_tc7_lvector was not used anyway.
2352 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2354 Use scm_make_smob_type and scm_set_smob_XXX instead.
2356 ** New function scm_set_smob_apply.
2358 This can be used to set an apply function to a smob type.
2360 ** Deprecated function: scm_strprint_obj
2362 Use scm_object_to_string instead.
2364 ** Deprecated function: scm_wta
2366 Use scm_wrong_type_arg, or another appropriate error signalling function
2369 ** Explicit support for obarrays has been deprecated.
2371 Use `scm_str2symbol' and the generic hashtable functions instead.
2373 ** The concept of `vcells' has been deprecated.
2375 The data type `variable' is now used exclusively. `Vcells' have been
2376 a low-level concept so you are likely not affected by this change.
2378 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2379 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2381 Use scm_c_define or scm_c_lookup instead, as appropriate.
2383 *** New functions: scm_c_module_lookup, scm_c_lookup,
2384 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2385 scm_module_define, scm_define.
2387 These functions work with variables instead of with vcells.
2389 ** New functions for creating and defining `subr's and `gsubr's.
2391 The new functions more clearly distinguish between creating a subr (or
2392 gsubr) object and adding it to the current module.
2394 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2395 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2396 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2397 scm_c_define_gsubr_with_generic.
2399 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2400 scm_make_subr_with_generic, scm_make_gsubr,
2401 scm_make_gsubr_with_generic.
2403 Use the new ones from above instead.
2405 ** C interface to the module system has changed.
2407 While we suggest that you avoid as many explicit module system
2408 operations from C as possible for the time being, the C interface has
2409 been made more similar to the high-level Scheme module system.
2411 *** New functions: scm_c_define_module, scm_c_use_module,
2412 scm_c_export, scm_c_resolve_module.
2414 They mostly work like their Scheme namesakes. scm_c_define_module
2415 takes a function that is called a context where the new module is
2418 *** Deprecated functions: scm_the_root_module, scm_make_module,
2419 scm_ensure_user_module, scm_load_scheme_module.
2421 Use the new functions instead.
2423 ** Renamed function: scm_internal_with_fluids becomes
2426 scm_internal_with_fluids is available as a deprecated function.
2428 ** New function: scm_c_with_fluid.
2430 Just like scm_c_with_fluids, but takes one fluid and one value instead
2433 ** Deprecated typedefs: long_long, ulong_long.
2435 They are of questionable utility and they pollute the global
2438 ** Deprecated typedef: scm_sizet
2440 It is of questionable utility now that Guile requires ANSI C, and is
2443 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2444 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2445 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2447 Made more compliant with the naming policy by adding a _t at the end.
2449 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2450 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2452 With the exception of the mysterious scm_2ulong2big, they are still
2453 available under new names (scm_i_mkbig etc). These functions are not
2454 intended to be used in user code. You should avoid dealing with
2455 bignums directly, and should deal with numbers in general (which can
2458 ** Change in behavior: scm_num2long, scm_num2ulong
2460 The scm_num2[u]long functions don't any longer accept an inexact
2461 argument. This change in behavior is motivated by concordance with
2462 R5RS: It is more common that a primitive doesn't want to accept an
2463 inexact for an exact.
2465 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2466 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2467 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2470 These are conversion functions between the various ANSI C integral
2471 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2472 accept an inexact argument.
2474 ** New functions: scm_float2num, scm_double2num,
2475 scm_num2float, scm_num2double.
2477 These are conversion functions between the two ANSI C float types and
2480 ** New number validation macros:
2481 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2485 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2487 These are just nicer-named old scm_protect_object and
2488 scm_unprotect_object.
2490 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2492 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2494 These functions can be used to register pointers to locations that
2497 ** Deprecated function: scm_create_hook.
2499 Its sins are: misleading name, non-modularity and lack of general
2503 Changes since Guile 1.3.4:
2505 * Changes to the distribution
2507 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2509 We've changed the way we handle generated files in the Guile source
2510 repository. As a result, the procedure for building trees obtained
2511 from the nightly FTP snapshots or via CVS has changed:
2512 - You must have appropriate versions of autoconf, automake, and
2513 libtool installed on your system. See README for info on how to
2514 obtain these programs.
2515 - Before configuring the tree, you must first run the script
2516 `autogen.sh' at the top of the source tree.
2518 The Guile repository used to contain not only source files, written by
2519 humans, but also some generated files, like configure scripts and
2520 Makefile.in files. Even though the contents of these files could be
2521 derived mechanically from other files present, we thought it would
2522 make the tree easier to build if we checked them into CVS.
2524 However, this approach means that minor differences between
2525 developer's installed tools and habits affected the whole team.
2526 So we have removed the generated files from the repository, and
2527 added the autogen.sh script, which will reconstruct them
2531 ** configure now has experimental options to remove support for certain
2534 --disable-arrays omit array and uniform array support
2535 --disable-posix omit posix interfaces
2536 --disable-networking omit networking interfaces
2537 --disable-regex omit regular expression interfaces
2539 These are likely to become separate modules some day.
2541 ** New configure option --enable-debug-freelist
2543 This enables a debugging version of SCM_NEWCELL(), and also registers
2544 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2546 Configure with the --enable-debug-freelist option to enable
2547 the gc-set-debug-check-freelist! primitive, and then use:
2549 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2550 (gc-set-debug-check-freelist! #f) # turn off checking
2552 Checking of the freelist forces a traversal of the freelist and
2553 a garbage collection before each allocation of a cell. This can
2554 slow down the interpreter dramatically, so the setter should be used to
2555 turn on this extra processing only when necessary.
2557 ** New configure option --enable-debug-malloc
2559 Include code for debugging of calls to scm_must_malloc/realloc/free.
2563 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2564 2. objects reallocated by scm_must_realloc has been allocated by
2566 3. reallocated objects are reallocated with the same what string
2568 But, most importantly, it records the number of allocated objects of
2569 each kind. This is useful when searching for memory leaks.
2571 A Guile compiled with this option provides the primitive
2572 `malloc-stats' which returns an alist with pairs of kind and the
2573 number of objects of that kind.
2575 ** All includes are now referenced relative to the root directory
2577 Since some users have had problems with mixups between Guile and
2578 system headers, we have decided to always refer to Guile headers via
2579 their parent directories. This essentially creates a "private name
2580 space" for Guile headers. This means that the compiler only is given
2581 -I options for the root build and root source directory.
2583 ** Header files kw.h and genio.h have been removed.
2585 ** The module (ice-9 getopt-gnu-style) has been removed.
2587 ** New module (ice-9 documentation)
2589 Implements the interface to documentation strings associated with
2592 ** New module (ice-9 time)
2594 Provides a macro `time', which displays execution time of a given form.
2596 ** New module (ice-9 history)
2598 Loading this module enables value history in the repl.
2600 * Changes to the stand-alone interpreter
2602 ** New command line option --debug
2604 Start Guile with debugging evaluator and backtraces enabled.
2606 This is useful when debugging your .guile init file or scripts.
2608 ** New help facility
2610 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2611 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2612 (help 'NAME) gives documentation for NAME, even if it is not an object
2613 (help ,EXPR) gives documentation for object returned by EXPR
2614 (help (my module)) gives module commentary for `(my module)'
2615 (help) gives this text
2617 `help' searches among bindings exported from loaded modules, while
2618 `apropos' searches among bindings visible from the "current" module.
2620 Examples: (help help)
2622 (help "output-string")
2624 ** `help' and `apropos' now prints full module names
2626 ** Dynamic linking now uses libltdl from the libtool package.
2628 The old system dependent code for doing dynamic linking has been
2629 replaced with calls to the libltdl functions which do all the hairy
2632 The major improvement is that you can now directly pass libtool
2633 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2634 will be able to do the best shared library job you can get, via
2637 The way dynamic libraries are found has changed and is not really
2638 portable across platforms, probably. It is therefore recommended to
2639 use absolute filenames when possible.
2641 If you pass a filename without an extension to `dynamic-link', it will
2642 try a few appropriate ones. Thus, the most platform ignorant way is
2643 to specify a name like "libfoo", without any directories and
2646 ** Guile COOP threads are now compatible with LinuxThreads
2648 Previously, COOP threading wasn't possible in applications linked with
2649 Linux POSIX threads due to their use of the stack pointer to find the
2650 thread context. This has now been fixed with a workaround which uses
2651 the pthreads to allocate the stack.
2653 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2655 ** Positions of erring expression in scripts
2657 With version 1.3.4, the location of the erring expression in Guile
2658 scipts is no longer automatically reported. (This should have been
2659 documented before the 1.3.4 release.)
2661 You can get this information by enabling recording of positions of
2662 source expressions and running the debugging evaluator. Put this at
2663 the top of your script (or in your "site" file):
2665 (read-enable 'positions)
2666 (debug-enable 'debug)
2668 ** Backtraces in scripts
2670 It is now possible to get backtraces in scripts.
2674 (debug-enable 'debug 'backtrace)
2676 at the top of the script.
2678 (The first options enables the debugging evaluator.
2679 The second enables backtraces.)
2681 ** Part of module system symbol lookup now implemented in C
2683 The eval closure of most modules is now implemented in C. Since this
2684 was one of the bottlenecks for loading speed, Guile now loads code
2685 substantially faster than before.
2687 ** Attempting to get the value of an unbound variable now produces
2688 an exception with a key of 'unbound-variable instead of 'misc-error.
2690 ** The initial default output port is now unbuffered if it's using a
2691 tty device. Previously in this situation it was line-buffered.
2693 ** New hook: after-gc-hook
2695 after-gc-hook takes over the role of gc-thunk. This hook is run at
2696 the first SCM_TICK after a GC. (Thus, the code is run at the same
2697 point during evaluation as signal handlers.)
2699 Note that this hook should be used only for diagnostic and debugging
2700 purposes. It is not certain that it will continue to be well-defined
2701 when this hook is run in the future.
2703 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2704 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2706 ** Improvements to garbage collector
2708 Guile 1.4 has a new policy for triggering heap allocation and
2709 determining the sizes of heap segments. It fixes a number of problems
2712 1. The new policy can handle two separate pools of cells
2713 (2-word/4-word) better. (The old policy would run wild, allocating
2714 more and more memory for certain programs.)
2716 2. The old code would sometimes allocate far too much heap so that the
2717 Guile process became gigantic. The new code avoids this.
2719 3. The old code would sometimes allocate too little so that few cells
2720 were freed at GC so that, in turn, too much time was spent in GC.
2722 4. The old code would often trigger heap allocation several times in a
2723 row. (The new scheme predicts how large the segments needs to be
2724 in order not to need further allocation.)
2726 All in all, the new GC policy will make larger applications more
2729 The new GC scheme also is prepared for POSIX threading. Threads can
2730 allocate private pools of cells ("clusters") with just a single
2731 function call. Allocation of single cells from such a cluster can
2732 then proceed without any need of inter-thread synchronization.
2734 ** New environment variables controlling GC parameters
2736 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2739 Allocation of 2-word cell heaps:
2741 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2744 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2745 GC in percent of total heap size
2748 Allocation of 4-word cell heaps
2749 (used for real numbers and misc other objects):
2751 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2753 (See entry "Way for application to customize GC parameters" under
2754 section "Changes to the scm_ interface" below.)
2756 ** Guile now implements reals using 4-word cells
2758 This speeds up computation with reals. (They were earlier allocated
2759 with `malloc'.) There is still some room for optimizations, however.
2761 ** Some further steps toward POSIX thread support have been taken
2763 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2764 don't have much effect any longer, and many of them will be removed in
2768 are only handled at the top of the evaluator loop, immediately after
2769 I/O, and in scm_equalp.
2771 *** The GC can allocate thread private pools of pairs.
2773 * Changes to Scheme functions and syntax
2775 ** close-input-port and close-output-port are now R5RS
2777 These procedures have been turned into primitives and have R5RS behaviour.
2779 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2781 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2782 extended by the more sophisticated version in (ice-9 format)
2784 (simple-format port message . args)
2785 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2786 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2787 the escapes are replaced with corresponding members of ARGS:
2788 ~A formats using `display' and ~S formats using `write'.
2789 If DESTINATION is #t, then use the `current-output-port',
2790 if DESTINATION is #f, then return a string containing the formatted text.
2791 Does not add a trailing newline."
2793 ** string-ref: the second argument is no longer optional.
2795 ** string, list->string: no longer accept strings in their arguments,
2796 only characters, for compatibility with R5RS.
2798 ** New procedure: port-closed? PORT
2799 Returns #t if PORT is closed or #f if it is open.
2801 ** Deprecated: list*
2803 The list* functionality is now provided by cons* (SRFI-1 compliant)
2805 ** New procedure: cons* ARG1 ARG2 ... ARGn
2807 Like `list', but the last arg provides the tail of the constructed list,
2808 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2810 Requires at least one argument. If given one argument, that argument
2811 is returned as result.
2813 This function is called `list*' in some other Schemes and in Common LISP.
2815 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2817 ** New procedure: object-documentation OBJECT
2819 Returns the documentation string associated with OBJECT. The
2820 procedure uses a caching mechanism so that subsequent lookups are
2823 Exported by (ice-9 documentation).
2825 ** module-name now returns full names of modules
2827 Previously, only the last part of the name was returned (`session' for
2828 `(ice-9 session)'). Ex: `(ice-9 session)'.
2830 * Changes to the gh_ interface
2832 ** Deprecated: gh_int2scmb
2834 Use gh_bool2scm instead.
2836 * Changes to the scm_ interface
2838 ** Guile primitives now carry docstrings!
2840 Thanks to Greg Badros!
2842 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2844 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2845 macros and must contain a docstring that is extracted into foo.doc using a new
2846 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2848 However, a major overhaul of these macros is scheduled for the next release of
2851 ** Guile primitives use a new technique for validation of arguments
2853 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2854 the readability of argument checking.
2856 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2858 ** New macros: SCM_PACK, SCM_UNPACK
2860 Compose/decompose an SCM value.
2862 The SCM type is now treated as an abstract data type and may be defined as a
2863 long, a void* or as a struct, depending on the architecture and compile time
2864 options. This makes it easier to find several types of bugs, for example when
2865 SCM values are treated as integers without conversion. Values of the SCM type
2866 should be treated as "atomic" values. These macros are used when
2867 composing/decomposing an SCM value, either because you want to access
2868 individual bits, or because you want to treat it as an integer value.
2870 E.g., in order to set bit 7 in an SCM value x, use the expression
2872 SCM_PACK (SCM_UNPACK (x) | 0x80)
2874 ** The name property of hooks is deprecated.
2875 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2877 You can emulate this feature by using object properties.
2879 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2880 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2883 These macros will be removed in a future release of Guile.
2885 ** The following types, functions and macros from numbers.h are deprecated:
2886 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2887 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2889 ** Port internals: the rw_random variable in the scm_port structure
2890 must be set to non-zero in any random access port. In recent Guile
2891 releases it was only set for bidirectional random-access ports.
2893 ** Port internals: the seek ptob procedure is now responsible for
2894 resetting the buffers if required. The change was made so that in the
2895 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2896 the fport and strport ptobs can avoid resetting the buffers,
2897 in particular to avoid discarding unread chars. An existing port
2898 type can be fixed by adding something like the following to the
2899 beginning of the ptob seek procedure:
2901 if (pt->rw_active == SCM_PORT_READ)
2902 scm_end_input (object);
2903 else if (pt->rw_active == SCM_PORT_WRITE)
2904 ptob->flush (object);
2906 although to actually avoid resetting the buffers and discard unread
2907 chars requires further hacking that depends on the characteristics
2910 ** Deprecated functions: scm_fseek, scm_tag
2912 These functions are no longer used and will be removed in a future version.
2914 ** The scm_sysmissing procedure is no longer used in libguile.
2915 Unless it turns out to be unexpectedly useful to somebody, it will be
2916 removed in a future version.
2918 ** The format of error message strings has changed
2920 The two C procedures: scm_display_error and scm_error, as well as the
2921 primitive `scm-error', now use scm_simple_format to do their work.
2922 This means that the message strings of all code must be updated to use
2923 ~A where %s was used before, and ~S where %S was used before.
2925 During the period when there still are a lot of old Guiles out there,
2926 you might want to support both old and new versions of Guile.
2928 There are basically two methods to achieve this. Both methods use
2931 AC_CHECK_FUNCS(scm_simple_format)
2933 in your configure.in.
2935 Method 1: Use the string concatenation features of ANSI C's
2940 #ifdef HAVE_SCM_SIMPLE_FORMAT
2946 Then represent each of your error messages using a preprocessor macro:
2948 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2952 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2953 (define make-message string-append)
2955 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2957 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2961 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2966 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2970 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2972 Don't use the functions coop_mutex_init and
2973 coop_condition_variable_init. They will change.
2975 Use scm_mutex_init and scm_cond_init instead.
2977 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2978 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2979 COND, as `scm_cond_wait' does, but it also bounds the duration
2980 of the wait. If COND has not been signaled before time ABSTIME,
2981 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2982 returns the error code `ETIMEDOUT'.
2984 The ABSTIME parameter specifies an absolute time, with the same
2985 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2986 to 00:00:00 GMT, January 1, 1970.
2988 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2989 `scm_cond_broadcast' restarts all the threads that are waiting
2990 on the condition variable COND. Nothing happens if no threads are
2993 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2994 `scm_key_create' allocates a new TSD key. The key is stored in
2995 the location pointed to by KEY. There is no limit on the number
2996 of keys allocated at a given time. The value initially associated
2997 with the returned key is `NULL' in all currently executing threads.
2999 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3000 function associated with the key. When a thread terminates,
3001 DESTR_FUNCTION is called on the value associated with the key in
3002 that thread. The DESTR_FUNCTION is not called if a key is deleted
3003 with `scm_key_delete' or a value is changed with
3004 `scm_setspecific'. The order in which destructor functions are
3005 called at thread termination time is unspecified.
3007 Destructors are not yet implemented.
3009 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3010 `scm_setspecific' changes the value associated with KEY in the
3011 calling thread, storing the given POINTER instead.
3013 ** New function: scm_getspecific (scm_key_t KEY)
3014 `scm_getspecific' returns the value currently associated with
3015 KEY in the calling thread.
3017 ** New function: scm_key_delete (scm_key_t KEY)
3018 `scm_key_delete' deallocates a TSD key. It does not check
3019 whether non-`NULL' values are associated with that key in the
3020 currently executing threads, nor call the destructor function
3021 associated with the key.
3023 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3025 Initialize a C level hook HOOK with associated HOOK_DATA and type
3026 TYPE. (See scm_c_hook_run ().)
3028 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3030 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3031 is true, add it last, otherwise first. The same FUNC can be added
3032 multiple times if FUNC_DATA differ and vice versa.
3034 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3036 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3037 function is only removed if both FUNC and FUNC_DATA matches.
3039 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3041 Run hook HOOK passing DATA to the hook functions.
3043 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3044 returned is undefined.
3046 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3047 returns a non-NULL value. This value is returned as the result of
3048 scm_c_hook_run. If all functions return NULL, NULL is returned.
3050 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3051 returns a NULL value, and NULL is returned. If all functions returns
3052 a non-NULL value, the last value is returned.
3054 ** New C level GC hooks
3056 Five new C level hooks has been added to the garbage collector.
3058 scm_before_gc_c_hook
3061 are run before locking and after unlocking the heap. The system is
3062 thus in a mode where evaluation can take place. (Except that
3063 scm_before_gc_c_hook must not allocate new cells.)
3065 scm_before_mark_c_hook
3066 scm_before_sweep_c_hook
3067 scm_after_sweep_c_hook
3069 are run when the heap is locked. These are intended for extension of
3070 the GC in a modular fashion. Examples are the weaks and guardians
3073 ** Way for application to customize GC parameters
3075 The application can set up other default values for the GC heap
3076 allocation parameters
3078 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3079 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3080 GUILE_MAX_SEGMENT_SIZE,
3084 scm_default_init_heap_size_1, scm_default_min_yield_1,
3085 scm_default_init_heap_size_2, scm_default_min_yield_2,
3086 scm_default_max_segment_size
3088 respectively before callong scm_boot_guile.
3090 (See entry "New environment variables ..." in section
3091 "Changes to the stand-alone interpreter" above.)
3093 ** scm_protect_object/scm_unprotect_object now nest
3095 This means that you can call scm_protect_object multiple times on an
3096 object and count on the object being protected until
3097 scm_unprotect_object has been call the same number of times.
3099 The functions also have better time complexity.
3101 Still, it is usually possible to structure the application in a way
3102 that you don't need to use these functions. For example, if you use a
3103 protected standard Guile list to keep track of live objects rather
3104 than some custom data type, objects will die a natural death when they
3105 are no longer needed.
3107 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3109 Guile does not provide the float representation for inexact real numbers any
3110 more. Now, only doubles are used to represent inexact real numbers. Further,
3111 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3112 and scm_tc16_complex, respectively.
3114 ** Removed deprecated type scm_smobfuns
3116 ** Removed deprecated function scm_newsmob
3118 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3120 There is an ongoing discussion among the developers whether to
3121 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3122 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3123 until this issue has been settled.
3125 ** Removed deprecated type tag scm_tc16_kw
3127 ** Added type tag scm_tc16_keyword
3129 (This was introduced already in release 1.3.4 but was not documented
3132 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3134 * Changes to system call interfaces:
3136 ** The "select" procedure now tests port buffers for the ability to
3137 provide input or accept output. Previously only the underlying file
3138 descriptors were checked.
3140 ** New variable PIPE_BUF: the maximum number of bytes that can be
3141 atomically written to a pipe.
3143 ** If a facility is not available on the system when Guile is
3144 compiled, the corresponding primitive procedure will not be defined.
3145 Previously it would have been defined but would throw a system-error
3146 exception if called. Exception handlers which catch this case may
3147 need minor modification: an error will be thrown with key
3148 'unbound-variable instead of 'system-error. Alternatively it's
3149 now possible to use `defined?' to check whether the facility is
3152 ** Procedures which depend on the timezone should now give the correct
3153 result on systems which cache the TZ environment variable, even if TZ
3154 is changed without calling tzset.
3156 * Changes to the networking interfaces:
3158 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3159 long integers between network and host format. For now, it's not
3160 particularly convenient to do this kind of thing, but consider:
3162 (define write-network-long
3163 (lambda (value port)
3164 (let ((v (make-uniform-vector 1 1 0)))
3165 (uniform-vector-set! v 0 (htonl value))
3166 (uniform-vector-write v port))))
3168 (define read-network-long
3170 (let ((v (make-uniform-vector 1 1 0)))
3171 (uniform-vector-read! v port)
3172 (ntohl (uniform-vector-ref v 0)))))
3174 ** If inet-aton fails, it now throws an error with key 'misc-error
3175 instead of 'system-error, since errno is not relevant.
3177 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3178 specific keys instead of 'system-error. The latter is inappropriate
3179 since errno will not have been set. The keys are:
3180 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3182 ** sethostent, setnetent, setprotoent, setservent: now take an
3183 optional argument STAYOPEN, which specifies whether the database
3184 remains open after a database entry is accessed randomly (e.g., using
3185 gethostbyname for the hosts database.) The default is #f. Previously
3189 Changes since Guile 1.3.2:
3191 * Changes to the stand-alone interpreter
3195 An initial version of the Guile debugger written by Chris Hanson has
3196 been added. The debugger is still under development but is included
3197 in the distribution anyway since it is already quite useful.
3203 after an error to enter the debugger. Type `help' inside the debugger
3204 for a description of available commands.
3206 If you prefer to have stack frames numbered and printed in
3207 anti-chronological order and prefer up in the stack to be down on the
3208 screen as is the case in gdb, you can put
3210 (debug-enable 'backwards)
3212 in your .guile startup file. (However, this means that Guile can't
3213 use indentation to indicate stack level.)
3215 The debugger is autoloaded into Guile at the first use.
3217 ** Further enhancements to backtraces
3219 There is a new debug option `width' which controls the maximum width
3220 on the screen of printed stack frames. Fancy printing parameters
3221 ("level" and "length" as in Common LISP) are adaptively adjusted for
3222 each stack frame to give maximum information while still fitting
3223 within the bounds. If the stack frame can't be made to fit by
3224 adjusting parameters, it is simply cut off at the end. This is marked
3227 ** Some modules are now only loaded when the repl is started
3229 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3230 regex) are now loaded into (guile-user) only if the repl has been
3231 started. The effect is that the startup time for scripts has been
3232 reduced to 30% of what it was previously.
3234 Correctly written scripts load the modules they require at the top of
3235 the file and should not be affected by this change.
3237 ** Hooks are now represented as smobs
3239 * Changes to Scheme functions and syntax
3241 ** Readline support has changed again.
3243 The old (readline-activator) module is gone. Use (ice-9 readline)
3244 instead, which now contains all readline functionality. So the code
3245 to activate readline is now
3247 (use-modules (ice-9 readline))
3250 This should work at any time, including from the guile prompt.
3252 To avoid confusion about the terms of Guile's license, please only
3253 enable readline for your personal use; please don't make it the
3254 default for others. Here is why we make this rather odd-sounding
3257 Guile is normally licensed under a weakened form of the GNU General
3258 Public License, which allows you to link code with Guile without
3259 placing that code under the GPL. This exception is important to some
3262 However, since readline is distributed under the GNU General Public
3263 License, when you link Guile with readline, either statically or
3264 dynamically, you effectively change Guile's license to the strict GPL.
3265 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3266 which are normally permitted become forbidden. This is a rather
3267 non-obvious consequence of the licensing terms.
3269 So, to make sure things remain clear, please let people choose for
3270 themselves whether to link GPL'd libraries like readline with Guile.
3272 ** regexp-substitute/global has changed slightly, but incompatibly.
3274 If you include a function in the item list, the string of the match
3275 object it receives is the same string passed to
3276 regexp-substitute/global, not some suffix of that string.
3277 Correspondingly, the match's positions are relative to the entire
3278 string, not the suffix.
3280 If the regexp can match the empty string, the way matches are chosen
3281 from the string has changed. regexp-substitute/global recognizes the
3282 same set of matches that list-matches does; see below.
3284 ** New function: list-matches REGEXP STRING [FLAGS]
3286 Return a list of match objects, one for every non-overlapping, maximal
3287 match of REGEXP in STRING. The matches appear in left-to-right order.
3288 list-matches only reports matches of the empty string if there are no
3289 other matches which begin on, end at, or include the empty match's
3292 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3294 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3296 For each match of REGEXP in STRING, apply PROC to the match object,
3297 and the last value PROC returned, or INIT for the first call. Return
3298 the last value returned by PROC. We apply PROC to the matches as they
3299 appear from left to right.
3301 This function recognizes matches according to the same criteria as
3304 Thus, you could define list-matches like this:
3306 (define (list-matches regexp string . flags)
3307 (reverse! (apply fold-matches regexp string '() cons flags)))
3309 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3313 *** New function: hook? OBJ
3315 Return #t if OBJ is a hook, otherwise #f.
3317 *** New function: make-hook-with-name NAME [ARITY]
3319 Return a hook with name NAME and arity ARITY. The default value for
3320 ARITY is 0. The only effect of NAME is that it will appear when the
3321 hook object is printed to ease debugging.
3323 *** New function: hook-empty? HOOK
3325 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3327 *** New function: hook->list HOOK
3329 Return a list of the procedures that are called when run-hook is
3332 ** `map' signals an error if its argument lists are not all the same length.
3334 This is the behavior required by R5RS, so this change is really a bug
3335 fix. But it seems to affect a lot of people's code, so we're
3336 mentioning it here anyway.
3338 ** Print-state handling has been made more transparent
3340 Under certain circumstances, ports are represented as a port with an
3341 associated print state. Earlier, this pair was represented as a pair
3342 (see "Some magic has been added to the printer" below). It is now
3343 indistinguishable (almost; see `get-print-state') from a port on the
3346 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3348 Return a new port with the associated print state PRINT-STATE.
3350 *** New function: get-print-state OUTPUT-PORT
3352 Return the print state associated with this port if it exists,
3353 otherwise return #f.
3355 *** New function: directory-stream? OBJECT
3357 Returns true iff OBJECT is a directory stream --- the sort of object
3358 returned by `opendir'.
3360 ** New function: using-readline?
3362 Return #t if readline is in use in the current repl.
3364 ** structs will be removed in 1.4
3366 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3367 and use GOOPS objects as the fundamental record type.
3369 * Changes to the scm_ interface
3371 ** structs will be removed in 1.4
3373 The entire current struct interface (struct.c, struct.h) will be
3374 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3375 GOOPS objects as the fundamental record type.
3377 ** The internal representation of subr's has changed
3379 Instead of giving a hint to the subr name, the CAR field of the subr
3380 now contains an index to a subr entry in scm_subr_table.
3382 *** New variable: scm_subr_table
3384 An array of subr entries. A subr entry contains the name, properties
3385 and documentation associated with the subr. The properties and
3386 documentation slots are not yet used.
3388 ** A new scheme for "forwarding" calls to a builtin to a generic function
3390 It is now possible to extend the functionality of some Guile
3391 primitives by letting them defer a call to a GOOPS generic function on
3392 argument mismatch. This means that there is no loss of efficiency in
3397 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3398 (define-method + ((x <string>) (y <string>))
3399 (string-append x y))
3401 + will still be as efficient as usual in numerical calculations, but
3402 can also be used for concatenating strings.
3404 Who will be the first one to extend Guile's numerical tower to
3405 rationals? :) [OK, there a few other things to fix before this can
3406 be made in a clean way.]
3408 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3410 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3412 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3414 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3415 a variable GENERIC which can be used by the dispatch macros below.
3417 [This is experimental code which may change soon.]
3419 *** New macros for forwarding control to a generic on arg type error
3421 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3423 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3425 These correspond to the scm_wta function call, and have the same
3426 behaviour until the user has called the GOOPS primitive
3427 `enable-primitive-generic!'. After that, these macros will apply the
3428 generic function GENERIC to the argument(s) instead of calling
3431 [This is experimental code which may change soon.]
3433 *** New macros for argument testing with generic dispatch
3435 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3437 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3439 These correspond to the SCM_ASSERT macro, but will defer control to
3440 GENERIC on error after `enable-primitive-generic!' has been called.
3442 [This is experimental code which may change soon.]
3444 ** New function: SCM scm_eval_body (SCM body, SCM env)
3446 Evaluates the body of a special form.
3448 ** The internal representation of struct's has changed
3450 Previously, four slots were allocated for the procedure(s) of entities
3451 and operators. The motivation for this representation had to do with
3452 the structure of the evaluator, the wish to support tail-recursive
3453 generic functions, and efficiency. Since the generic function
3454 dispatch mechanism has changed, there is no longer a need for such an
3455 expensive representation, and the representation has been simplified.
3457 This should not make any difference for most users.
3459 ** GOOPS support has been cleaned up.
3461 Some code has been moved from eval.c to objects.c and code in both of
3462 these compilation units has been cleaned up and better structured.
3464 *** New functions for applying generic functions
3466 New function: SCM scm_apply_generic (GENERIC, ARGS)
3467 New function: SCM scm_call_generic_0 (GENERIC)
3468 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3469 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3470 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3472 ** Deprecated function: scm_make_named_hook
3474 It is now replaced by:
3476 ** New function: SCM scm_create_hook (const char *name, int arity)
3478 Creates a hook in the same way as make-hook above but also
3479 binds a variable named NAME to it.
3481 This is the typical way of creating a hook from C code.
3483 Currently, the variable is created in the "current" module.
3484 This might change when we get the new module system.
3486 [The behaviour is identical to scm_make_named_hook.]
3490 Changes since Guile 1.3:
3492 * Changes to mailing lists
3494 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3496 See the README file to find current addresses for all the Guile
3499 * Changes to the distribution
3501 ** Readline support is no longer included with Guile by default.
3503 Based on the different license terms of Guile and Readline, we
3504 concluded that Guile should not *by default* cause the linking of
3505 Readline into an application program. Readline support is now offered
3506 as a separate module, which is linked into an application only when
3507 you explicitly specify it.
3509 Although Guile is GNU software, its distribution terms add a special
3510 exception to the usual GNU General Public License (GPL). Guile's
3511 license includes a clause that allows you to link Guile with non-free
3512 programs. We add this exception so as not to put Guile at a
3513 disadvantage vis-a-vis other extensibility packages that support other
3516 In contrast, the GNU Readline library is distributed under the GNU
3517 General Public License pure and simple. This means that you may not
3518 link Readline, even dynamically, into an application unless it is
3519 distributed under a free software license that is compatible the GPL.
3521 Because of this difference in distribution terms, an application that
3522 can use Guile may not be able to use Readline. Now users will be
3523 explicitly offered two independent decisions about the use of these
3526 You can activate the readline support by issuing
3528 (use-modules (readline-activator))
3531 from your ".guile" file, for example.
3533 * Changes to the stand-alone interpreter
3535 ** All builtins now print as primitives.
3536 Previously builtin procedures not belonging to the fundamental subr
3537 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3538 Now, they print as #<primitive-procedure NAME>.
3540 ** Backtraces slightly more intelligible.
3541 gsubr-apply and macro transformer application frames no longer appear
3544 * Changes to Scheme functions and syntax
3546 ** Guile now correctly handles internal defines by rewriting them into
3547 their equivalent letrec. Previously, internal defines would
3548 incrementally add to the innermost environment, without checking
3549 whether the restrictions specified in RnRS were met. This lead to the
3550 correct behaviour when these restriction actually were met, but didn't
3551 catch all illegal uses. Such an illegal use could lead to crashes of
3552 the Guile interpreter or or other unwanted results. An example of
3553 incorrect internal defines that made Guile behave erratically:
3565 The problem with this example is that the definition of `c' uses the
3566 value of `b' directly. This confuses the meoization machine of Guile
3567 so that the second call of `b' (this time in a larger environment that
3568 also contains bindings for `c' and `d') refers to the binding of `c'
3569 instead of `a'. You could also make Guile crash with a variation on
3574 (define (b flag) (if flag a 1))
3575 (define c (1+ (b flag)))
3583 From now on, Guile will issue an `Unbound variable: b' error message
3588 A hook contains a list of functions which should be called on
3589 particular occasions in an existing program. Hooks are used for
3592 A window manager might have a hook before-window-map-hook. The window
3593 manager uses the function run-hooks to call all functions stored in
3594 before-window-map-hook each time a window is mapped. The user can
3595 store functions in the hook using add-hook!.
3597 In Guile, hooks are first class objects.
3599 *** New function: make-hook [N_ARGS]
3601 Return a hook for hook functions which can take N_ARGS arguments.
3602 The default value for N_ARGS is 0.
3604 (See also scm_make_named_hook below.)
3606 *** New function: add-hook! HOOK PROC [APPEND_P]
3608 Put PROC at the beginning of the list of functions stored in HOOK.
3609 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3611 PROC must be able to take the number of arguments specified when the
3614 If PROC already exists in HOOK, then remove it first.
3616 *** New function: remove-hook! HOOK PROC
3618 Remove PROC from the list of functions in HOOK.
3620 *** New function: reset-hook! HOOK
3622 Clear the list of hook functions stored in HOOK.
3624 *** New function: run-hook HOOK ARG1 ...
3626 Run all hook functions stored in HOOK with arguments ARG1 ... .
3627 The number of arguments supplied must correspond to the number given
3628 when the hook was created.
3630 ** The function `dynamic-link' now takes optional keyword arguments.
3631 The only keyword argument that is currently defined is `:global
3632 BOOL'. With it, you can control whether the shared library will be
3633 linked in global mode or not. In global mode, the symbols from the
3634 linked library can be used to resolve references from other
3635 dynamically linked libraries. In non-global mode, the linked
3636 library is essentially invisible and can only be accessed via
3637 `dynamic-func', etc. The default is now to link in global mode.
3638 Previously, the default has been non-global mode.
3640 The `#:global' keyword is only effective on platforms that support
3641 the dlopen family of functions.
3643 ** New function `provided?'
3645 - Function: provided? FEATURE
3646 Return true iff FEATURE is supported by this installation of
3647 Guile. FEATURE must be a symbol naming a feature; the global
3648 variable `*features*' is a list of available features.
3650 ** Changes to the module (ice-9 expect):
3652 *** The expect-strings macro now matches `$' in a regular expression
3653 only at a line-break or end-of-file by default. Previously it would
3654 match the end of the string accumulated so far. The old behaviour
3655 can be obtained by setting the variable `expect-strings-exec-flags'
3658 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3659 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3660 in a regular expression will still match before a line-break or
3661 end-of-file. The default is `regexp/noteol'.
3663 *** The expect-strings macro now uses a variable
3664 `expect-strings-compile-flags' for the flags to be supplied to
3665 `make-regexp'. The default is `regexp/newline', which was previously
3668 *** The expect macro now supplies two arguments to a match procedure:
3669 the current accumulated string and a flag to indicate whether
3670 end-of-file has been reached. Previously only the string was supplied.
3671 If end-of-file is reached, the match procedure will be called an
3672 additional time with the same accumulated string as the previous call
3673 but with the flag set.
3675 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3677 This code, and the documentation for it that appears here, was
3678 borrowed from SLIB, with minor adaptations for Guile.
3680 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3681 An almost complete implementation of Common LISP format description
3682 according to the CL reference book `Common LISP' from Guy L.
3683 Steele, Digital Press. Backward compatible to most of the
3684 available Scheme format implementations.
3686 Returns `#t', `#f' or a string; has side effect of printing
3687 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3688 to the current output port and `#t' is returned. If DESTINATION
3689 is `#f', a formatted string is returned as the result of the call.
3690 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3691 format string; FORMAT-STRING is then the first argument and the
3692 output is returned as a string. If DESTINATION is a number, the
3693 output is to the current error port if available by the
3694 implementation. Otherwise DESTINATION must be an output port and
3697 FORMAT-STRING must be a string. In case of a formatting error
3698 format returns `#f' and prints a message on the current output or
3699 error port. Characters are output as if the string were output by
3700 the `display' function with the exception of those prefixed by a
3701 tilde (~). For a detailed description of the FORMAT-STRING syntax
3702 please consult a Common LISP format reference manual. For a test
3703 suite to verify this format implementation load `formatst.scm'.
3704 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3706 Note: `format' is not reentrant, i.e. only one `format'-call may
3707 be executed at a time.
3710 *** Format Specification (Format version 3.0)
3712 Please consult a Common LISP format reference manual for a detailed
3713 description of the format string syntax. For a demonstration of the
3714 implemented directives see `formatst.scm'.
3716 This implementation supports directive parameters and modifiers (`:'
3717 and `@' characters). Multiple parameters must be separated by a comma
3718 (`,'). Parameters can be numerical parameters (positive or negative),
3719 character parameters (prefixed by a quote character (`''), variable
3720 parameters (`v'), number of rest arguments parameter (`#'), empty and
3721 default parameters. Directive characters are case independent. The
3722 general form of a directive is:
3724 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3726 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3728 *** Implemented CL Format Control Directives
3730 Documentation syntax: Uppercase characters represent the
3731 corresponding control directive characters. Lowercase characters
3732 represent control directive parameter descriptions.
3735 Any (print as `display' does).
3739 `~MINCOL,COLINC,MINPAD,PADCHARA'
3743 S-expression (print as `write' does).
3747 `~MINCOL,COLINC,MINPAD,PADCHARS'
3753 print number sign always.
3756 print comma separated.
3758 `~MINCOL,PADCHAR,COMMACHARD'
3764 print number sign always.
3767 print comma separated.
3769 `~MINCOL,PADCHAR,COMMACHARX'
3775 print number sign always.
3778 print comma separated.
3780 `~MINCOL,PADCHAR,COMMACHARO'
3786 print number sign always.
3789 print comma separated.
3791 `~MINCOL,PADCHAR,COMMACHARB'
3796 `~N,MINCOL,PADCHAR,COMMACHARR'
3800 print a number as a Roman numeral.
3803 print a number as an "old fashioned" Roman numeral.
3806 print a number as an ordinal English number.
3809 print a number as a cardinal English number.
3814 prints `y' and `ies'.
3817 as `~P but jumps 1 argument backward.'
3820 as `~@P but jumps 1 argument backward.'
3825 prints a character as the reader can understand it (i.e. `#\'
3829 prints a character as emacs does (eg. `^C' for ASCII 03).
3832 Fixed-format floating-point (prints a flonum like MMM.NNN).
3833 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3835 If the number is positive a plus sign is printed.
3838 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3839 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3841 If the number is positive a plus sign is printed.
3844 General floating-point (prints a flonum either fixed or
3846 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3848 If the number is positive a plus sign is printed.
3851 Dollars floating-point (prints a flonum in fixed with signs
3853 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3855 If the number is positive a plus sign is printed.
3858 A sign is always printed and appears before the padding.
3861 The sign appears before the padding.
3869 print newline if not at the beginning of the output line.
3871 prints `~&' and then N-1 newlines.
3876 print N page separators.
3886 newline is ignored, white space left.
3889 newline is left, white space ignored.
3894 relative tabulation.
3900 Indirection (expects indirect arguments as a list).
3902 extracts indirect arguments from format arguments.
3905 Case conversion (converts by `string-downcase').
3907 converts by `string-capitalize'.
3910 converts by `string-capitalize-first'.
3913 converts by `string-upcase'.
3916 Argument Jumping (jumps 1 argument forward).
3918 jumps N arguments forward.
3921 jumps 1 argument backward.
3924 jumps N arguments backward.
3927 jumps to the 0th argument.
3930 jumps to the Nth argument (beginning from 0)
3932 `~[STR0~;STR1~;...~;STRN~]'
3933 Conditional Expression (numerical clause conditional).
3935 take argument from N.
3938 true test conditional.
3941 if-else-then conditional.
3947 default clause follows.
3950 Iteration (args come from the next argument (a list)).
3952 at most N iterations.
3955 args from next arg (a list of lists).
3958 args from the rest of arguments.
3961 args from the rest args (lists).
3972 aborts if N <= M <= K
3974 *** Not Implemented CL Format Control Directives
3977 print `#f' as an empty list (see below).
3980 print `#f' as an empty list (see below).
3986 (sorry I don't understand its semantics completely)
3988 *** Extended, Replaced and Additional Control Directives
3990 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3991 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3992 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3993 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3994 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3995 COMMAWIDTH is the number of characters between two comma
3999 print a R4RS complex number as `~F~@Fi' with passed parameters for
4003 Pretty print formatting of an argument for scheme code lists.
4009 Flushes the output if format DESTINATION is a port.
4012 Print a `#\space' character
4014 print N `#\space' characters.
4017 Print a `#\tab' character
4019 print N `#\tab' characters.
4022 Takes N as an integer representation for a character. No arguments
4023 are consumed. N is converted to a character by `integer->char'. N
4024 must be a positive decimal number.
4027 Print out readproof. Prints out internal objects represented as
4028 `#<...>' as strings `"#<...>"' so that the format output can always
4029 be processed by `read'.
4032 Print out readproof. Prints out internal objects represented as
4033 `#<...>' as strings `"#<...>"' so that the format output can always
4034 be processed by `read'.
4037 Prints information and a copyright notice on the format
4040 prints format version.
4043 may also print number strings, i.e. passing a number as a string
4044 and format it accordingly.
4046 *** Configuration Variables
4048 The format module exports some configuration variables to suit the
4049 systems and users needs. There should be no modification necessary for
4050 the configuration that comes with Guile. Format detects automatically
4051 if the running scheme system implements floating point numbers and
4054 format:symbol-case-conv
4055 Symbols are converted by `symbol->string' so the case type of the
4056 printed symbols is implementation dependent.
4057 `format:symbol-case-conv' is a one arg closure which is either
4058 `#f' (no conversion), `string-upcase', `string-downcase' or
4059 `string-capitalize'. (default `#f')
4061 format:iobj-case-conv
4062 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4063 implementation internal objects. (default `#f')
4066 The character prefixing the exponent value in `~E' printing.
4069 *** Compatibility With Other Format Implementations
4075 Downward compatible except for padding support and `~A', `~S',
4076 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4077 `printf' padding support which is completely replaced by the CL
4078 `format' padding style.
4081 Downward compatible except for `~', which is not documented
4082 (ignores all characters inside the format string up to a newline
4083 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4084 numerical and variable parameters and `:/@' modifiers in the CL
4088 Downward compatible except for `~A' and `~S' which print in
4089 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4090 directive parameters or modifiers)).
4093 Downward compatible except for an optional destination parameter:
4094 S2C accepts a format call without a destination which returns a
4095 formatted string. This is equivalent to a #f destination in S2C.
4096 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4097 parameters or modifiers)).
4100 ** Changes to string-handling functions.
4102 These functions were added to support the (ice-9 format) module, above.
4104 *** New function: string-upcase STRING
4105 *** New function: string-downcase STRING
4107 These are non-destructive versions of the existing string-upcase! and
4108 string-downcase! functions.
4110 *** New function: string-capitalize! STRING
4111 *** New function: string-capitalize STRING
4113 These functions convert the first letter of each word in the string to
4116 (string-capitalize "howdy there")
4119 As with the other functions, string-capitalize! modifies the string in
4120 place, while string-capitalize returns a modified copy of its argument.
4122 *** New function: string-ci->symbol STRING
4124 Return a symbol whose name is STRING, but having the same case as if
4125 the symbol had be read by `read'.
4127 Guile can be configured to be sensitive or insensitive to case
4128 differences in Scheme identifiers. If Guile is case-insensitive, all
4129 symbols are converted to lower case on input. The `string-ci->symbol'
4130 function returns a symbol whose name in STRING, transformed as Guile
4131 would if STRING were input.
4133 *** New function: substring-move! STRING1 START END STRING2 START
4135 Copy the substring of STRING1 from START (inclusive) to END
4136 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4137 string, and the source and destination areas may overlap; in all
4138 cases, the function behaves as if all the characters were copied
4141 *** Extended functions: substring-move-left! substring-move-right!
4143 These functions now correctly copy arbitrarily overlapping substrings;
4144 they are both synonyms for substring-move!.
4147 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4149 getopt-long is a function for parsing command-line arguments in a
4150 manner consistent with other GNU programs.
4152 (getopt-long ARGS GRAMMAR)
4153 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4155 ARGS should be a list of strings. Its first element should be the
4156 name of the program; subsequent elements should be the arguments
4157 that were passed to the program on the command line. The
4158 `program-arguments' procedure returns a list of this form.
4160 GRAMMAR is a list of the form:
4161 ((OPTION (PROPERTY VALUE) ...) ...)
4163 Each OPTION should be a symbol. `getopt-long' will accept a
4164 command-line option named `--OPTION'.
4165 Each option can have the following (PROPERTY VALUE) pairs:
4167 (single-char CHAR) --- Accept `-CHAR' as a single-character
4168 equivalent to `--OPTION'. This is how to specify traditional
4170 (required? BOOL) --- If BOOL is true, the option is required.
4171 getopt-long will raise an error if it is not found in ARGS.
4172 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4173 it is #f, it does not; and if it is the symbol
4174 `optional', the option may appear in ARGS with or
4176 (predicate FUNC) --- If the option accepts a value (i.e. you
4177 specified `(value #t)' for this option), then getopt
4178 will apply FUNC to the value, and throw an exception
4179 if it returns #f. FUNC should be a procedure which
4180 accepts a string and returns a boolean value; you may
4181 need to use quasiquotes to get it into GRAMMAR.
4183 The (PROPERTY VALUE) pairs may occur in any order, but each
4184 property may occur only once. By default, options do not have
4185 single-character equivalents, are not required, and do not take
4188 In ARGS, single-character options may be combined, in the usual
4189 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4190 accepts values, then it must be the last option in the
4191 combination; the value is the next argument. So, for example, using
4192 the following grammar:
4193 ((apples (single-char #\a))
4194 (blimps (single-char #\b) (value #t))
4195 (catalexis (single-char #\c) (value #t)))
4196 the following argument lists would be acceptable:
4197 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4198 for "blimps" and "catalexis")
4199 ("-ab" "bang" "-c" "couth") (same)
4200 ("-ac" "couth" "-b" "bang") (same)
4201 ("-abc" "couth" "bang") (an error, since `-b' is not the
4202 last option in its combination)
4204 If an option's value is optional, then `getopt-long' decides
4205 whether it has a value by looking at what follows it in ARGS. If
4206 the next element is a string, and it does not appear to be an
4207 option itself, then that string is the option's value.
4209 The value of a long option can appear as the next element in ARGS,
4210 or it can follow the option name, separated by an `=' character.
4211 Thus, using the same grammar as above, the following argument lists
4213 ("--apples" "Braeburn" "--blimps" "Goodyear")
4214 ("--apples=Braeburn" "--blimps" "Goodyear")
4215 ("--blimps" "Goodyear" "--apples=Braeburn")
4217 If the option "--" appears in ARGS, argument parsing stops there;
4218 subsequent arguments are returned as ordinary arguments, even if
4219 they resemble options. So, in the argument list:
4220 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4221 `getopt-long' will recognize the `apples' option as having the
4222 value "Granny Smith", but it will not recognize the `blimp'
4223 option; it will return the strings "--blimp" and "Goodyear" as
4224 ordinary argument strings.
4226 The `getopt-long' function returns the parsed argument list as an
4227 assocation list, mapping option names --- the symbols from GRAMMAR
4228 --- onto their values, or #t if the option does not accept a value.
4229 Unused options do not appear in the alist.
4231 All arguments that are not the value of any option are returned
4232 as a list, associated with the empty list.
4234 `getopt-long' throws an exception if:
4235 - it finds an unrecognized option in ARGS
4236 - a required option is omitted
4237 - an option that requires an argument doesn't get one
4238 - an option that doesn't accept an argument does get one (this can
4239 only happen using the long option `--opt=value' syntax)
4240 - an option predicate fails
4245 `((lockfile-dir (required? #t)
4248 (predicate ,file-is-directory?))
4249 (verbose (required? #f)
4252 (x-includes (single-char #\x))
4253 (rnet-server (single-char #\y)
4254 (predicate ,string?))))
4256 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4257 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4259 => ((() "foo1" "-fred" "foo2" "foo3")
4260 (rnet-server . "lamprod")
4261 (x-includes . "/usr/include")
4262 (lockfile-dir . "/tmp")
4265 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4267 It will be removed in a few releases.
4269 ** New syntax: lambda*
4270 ** New syntax: define*
4271 ** New syntax: define*-public
4272 ** New syntax: defmacro*
4273 ** New syntax: defmacro*-public
4274 Guile now supports optional arguments.
4276 `lambda*', `define*', `define*-public', `defmacro*' and
4277 `defmacro*-public' are identical to the non-* versions except that
4278 they use an extended type of parameter list that has the following BNF
4279 syntax (parentheses are literal, square brackets indicate grouping,
4280 and `*', `+' and `?' have the usual meaning):
4282 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4283 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4284 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4286 ext-var-decl ::= identifier | ( identifier expression )
4288 The semantics are best illustrated with the following documentation
4289 and examples for `lambda*':
4292 lambda extended for optional and keyword arguments
4294 lambda* creates a procedure that takes optional arguments. These
4295 are specified by putting them inside brackets at the end of the
4296 paramater list, but before any dotted rest argument. For example,
4297 (lambda* (a b #&optional c d . e) '())
4298 creates a procedure with fixed arguments a and b, optional arguments c
4299 and d, and rest argument e. If the optional arguments are omitted
4300 in a call, the variables for them are unbound in the procedure. This
4301 can be checked with the bound? macro.
4303 lambda* can also take keyword arguments. For example, a procedure
4305 (lambda* (#&key xyzzy larch) '())
4306 can be called with any of the argument lists (#:xyzzy 11)
4307 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4308 are given as keywords are bound to values.
4310 Optional and keyword arguments can also be given default values
4311 which they take on when they are not present in a call, by giving a
4312 two-item list in place of an optional argument, for example in:
4313 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4314 foo is a fixed argument, bar is an optional argument with default
4315 value 42, and baz is a keyword argument with default value 73.
4316 Default value expressions are not evaluated unless they are needed
4317 and until the procedure is called.
4319 lambda* now supports two more special parameter list keywords.
4321 lambda*-defined procedures now throw an error by default if a
4322 keyword other than one of those specified is found in the actual
4323 passed arguments. However, specifying #&allow-other-keys
4324 immediately after the kyword argument declarations restores the
4325 previous behavior of ignoring unknown keywords. lambda* also now
4326 guarantees that if the same keyword is passed more than once, the
4327 last one passed is the one that takes effect. For example,
4328 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4329 #:heads 37 #:tails 42 #:heads 99)
4330 would result in (99 47) being displayed.
4332 #&rest is also now provided as a synonym for the dotted syntax rest
4333 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4334 all respects to lambda*. This is provided for more similarity to DSSSL,
4335 MIT-Scheme and Kawa among others, as well as for refugees from other
4338 Further documentation may be found in the optargs.scm file itself.
4340 The optional argument module also exports the macros `let-optional',
4341 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4342 are not documented here because they may be removed in the future, but
4343 full documentation is still available in optargs.scm.
4345 ** New syntax: and-let*
4346 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4348 Syntax: (land* (<clause> ...) <body> ...)
4349 Each <clause> should have one of the following forms:
4350 (<variable> <expression>)
4353 Each <variable> or <bound-variable> should be an identifier. Each
4354 <expression> should be a valid expression. The <body> should be a
4355 possibly empty sequence of expressions, like the <body> of a
4358 Semantics: A LAND* expression is evaluated by evaluating the
4359 <expression> or <bound-variable> of each of the <clause>s from
4360 left to right. The value of the first <expression> or
4361 <bound-variable> that evaluates to a false value is returned; the
4362 remaining <expression>s and <bound-variable>s are not evaluated.
4363 The <body> forms are evaluated iff all the <expression>s and
4364 <bound-variable>s evaluate to true values.
4366 The <expression>s and the <body> are evaluated in an environment
4367 binding each <variable> of the preceding (<variable> <expression>)
4368 clauses to the value of the <expression>. Later bindings
4369 shadow earlier bindings.
4371 Guile's and-let* macro was contributed by Michael Livshin.
4373 ** New sorting functions
4375 *** New function: sorted? SEQUENCE LESS?
4376 Returns `#t' when the sequence argument is in non-decreasing order
4377 according to LESS? (that is, there is no adjacent pair `... x y
4378 ...' for which `(less? y x)').
4380 Returns `#f' when the sequence contains at least one out-of-order
4381 pair. It is an error if the sequence is neither a list nor a
4384 *** New function: merge LIST1 LIST2 LESS?
4385 LIST1 and LIST2 are sorted lists.
4386 Returns the sorted list of all elements in LIST1 and LIST2.
4388 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4389 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4390 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4391 (Here "<" should read "comes before".)
4393 *** New procedure: merge! LIST1 LIST2 LESS?
4394 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4395 the result. If the code is compiled, and LESS? constructs no new
4396 pairs, no pairs at all will be allocated. The first pair of the
4397 result will be either the first pair of LIST1 or the first pair of
4400 *** New function: sort SEQUENCE LESS?
4401 Accepts either a list or a vector, and returns a new sequence
4402 which is sorted. The new sequence is the same type as the input.
4403 Always `(sorted? (sort sequence less?) less?)'. The original
4404 sequence is not altered in any way. The new sequence shares its
4405 elements with the old one; no elements are copied.
4407 *** New procedure: sort! SEQUENCE LESS
4408 Returns its sorted result in the original boxes. No new storage is
4409 allocated at all. Proper usage: (set! slist (sort! slist <))
4411 *** New function: stable-sort SEQUENCE LESS?
4412 Similar to `sort' but stable. That is, if "equal" elements are
4413 ordered a < b in the original sequence, they will have the same order
4416 *** New function: stable-sort! SEQUENCE LESS?
4417 Similar to `sort!' but stable.
4418 Uses temporary storage when sorting vectors.
4420 *** New functions: sort-list, sort-list!
4421 Added for compatibility with scsh.
4423 ** New built-in random number support
4425 *** New function: random N [STATE]
4426 Accepts a positive integer or real N and returns a number of the
4427 same type between zero (inclusive) and N (exclusive). The values
4428 returned have a uniform distribution.
4430 The optional argument STATE must be of the type produced by
4431 `copy-random-state' or `seed->random-state'. It defaults to the value
4432 of the variable `*random-state*'. This object is used to maintain the
4433 state of the pseudo-random-number generator and is altered as a side
4434 effect of the `random' operation.
4436 *** New variable: *random-state*
4437 Holds a data structure that encodes the internal state of the
4438 random-number generator that `random' uses by default. The nature
4439 of this data structure is implementation-dependent. It may be
4440 printed out and successfully read back in, but may or may not
4441 function correctly as a random-number state object in another
4444 *** New function: copy-random-state [STATE]
4445 Returns a new object of type suitable for use as the value of the
4446 variable `*random-state*' and as a second argument to `random'.
4447 If argument STATE is given, a copy of it is returned. Otherwise a
4448 copy of `*random-state*' is returned.
4450 *** New function: seed->random-state SEED
4451 Returns a new object of type suitable for use as the value of the
4452 variable `*random-state*' and as a second argument to `random'.
4453 SEED is a string or a number. A new state is generated and
4454 initialized using SEED.
4456 *** New function: random:uniform [STATE]
4457 Returns an uniformly distributed inexact real random number in the
4458 range between 0 and 1.
4460 *** New procedure: random:solid-sphere! VECT [STATE]
4461 Fills VECT with inexact real random numbers the sum of whose
4462 squares is less than 1.0. Thinking of VECT as coordinates in
4463 space of dimension N = `(vector-length VECT)', the coordinates are
4464 uniformly distributed within the unit N-shere. The sum of the
4465 squares of the numbers is returned. VECT can be either a vector
4466 or a uniform vector of doubles.
4468 *** New procedure: random:hollow-sphere! VECT [STATE]
4469 Fills VECT with inexact real random numbers the sum of whose squares
4470 is equal to 1.0. Thinking of VECT as coordinates in space of
4471 dimension n = `(vector-length VECT)', the coordinates are uniformly
4472 distributed over the surface of the unit n-shere. VECT can be either
4473 a vector or a uniform vector of doubles.
4475 *** New function: random:normal [STATE]
4476 Returns an inexact real in a normal distribution with mean 0 and
4477 standard deviation 1. For a normal distribution with mean M and
4478 standard deviation D use `(+ M (* D (random:normal)))'.
4480 *** New procedure: random:normal-vector! VECT [STATE]
4481 Fills VECT with inexact real random numbers which are independent and
4482 standard normally distributed (i.e., with mean 0 and variance 1).
4483 VECT can be either a vector or a uniform vector of doubles.
4485 *** New function: random:exp STATE
4486 Returns an inexact real in an exponential distribution with mean 1.
4487 For an exponential distribution with mean U use (* U (random:exp)).
4489 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4491 These functions now operate on numbers in the range of a C unsigned
4494 These functions used to operate on numbers in the range of a C signed
4495 long; however, this seems inappropriate, because Guile integers don't
4498 ** New function: make-guardian
4499 This is an implementation of guardians as described in
4500 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4501 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4502 Programming Language Design and Implementation, June 1993
4503 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4505 ** New functions: delq1!, delv1!, delete1!
4506 These procedures behave similar to delq! and friends but delete only
4507 one object if at all.
4509 ** New function: unread-string STRING PORT
4510 Unread STRING to PORT, that is, push it back onto the port so that
4511 next read operation will work on the pushed back characters.
4513 ** unread-char can now be called multiple times
4514 If unread-char is called multiple times, the unread characters will be
4515 read again in last-in first-out order.
4517 ** the procedures uniform-array-read! and uniform-array-write! now
4518 work on any kind of port, not just ports which are open on a file.
4520 ** Now 'l' in a port mode requests line buffering.
4522 ** The procedure truncate-file now works on string ports as well
4523 as file ports. If the size argument is omitted, the current
4524 file position is used.
4526 ** new procedure: seek PORT/FDES OFFSET WHENCE
4527 The arguments are the same as for the old fseek procedure, but it
4528 works on string ports as well as random-access file ports.
4530 ** the fseek procedure now works on string ports, since it has been
4531 redefined using seek.
4533 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4534 size is not supplied.
4536 ** the newline procedure no longer flushes the port if it's not
4537 line-buffered: previously it did if it was the current output port.
4539 ** open-pipe and close-pipe are no longer primitive procedures, but
4540 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4542 ** the freopen procedure has been removed.
4544 ** new procedure: drain-input PORT
4545 Drains PORT's read buffers (including any pushed-back characters)
4546 and returns the contents as a single string.
4548 ** New function: map-in-order PROC LIST1 LIST2 ...
4549 Version of `map' which guarantees that the procedure is applied to the
4550 lists in serial order.
4552 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4553 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4554 now obsolete and will go away in release 1.5.
4556 ** New syntax: collect BODY1 ...
4557 Version of `begin' which returns a list of the results of the body
4558 forms instead of the result of the last body form. In contrast to
4559 `begin', `collect' allows an empty body.
4561 ** New functions: read-history FILENAME, write-history FILENAME
4562 Read/write command line history from/to file. Returns #t on success
4563 and #f if an error occured.
4565 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4567 These procedures return a list of definitions available in the specified
4568 argument, a relative module reference. In the case of no argument,
4569 `(current-module)' is now consulted for definitions to return, instead
4570 of simply returning #f, the former behavior.
4572 ** The #/ syntax for lists is no longer supported.
4574 Earlier versions of Scheme accepted this syntax, but printed a
4577 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4579 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4582 * Changes to the gh_ interface
4586 Now takes a second argument which is the result array. If this
4587 pointer is NULL, a new array is malloced (the old behaviour).
4589 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4590 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4594 * Changes to the scm_ interface
4596 ** Function: scm_make_named_hook (char* name, int n_args)
4598 Creates a hook in the same way as make-hook above but also
4599 binds a variable named NAME to it.
4601 This is the typical way of creating a hook from C code.
4603 Currently, the variable is created in the "current" module. This
4604 might change when we get the new module system.
4606 ** The smob interface
4608 The interface for creating smobs has changed. For documentation, see
4609 data-rep.info (made from guile-core/doc/data-rep.texi).
4611 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4613 >>> This function will be removed in 1.3.4. <<<
4617 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4618 This function adds a new smob type, named NAME, with instance size
4619 SIZE to the system. The return value is a tag that is used in
4620 creating instances of the type. If SIZE is 0, then no memory will
4621 be allocated when instances of the smob are created, and nothing
4622 will be freed by the default free function.
4624 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4625 This function sets the smob marking procedure for the smob type
4626 specified by the tag TC. TC is the tag returned by
4627 `scm_make_smob_type'.
4629 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4630 This function sets the smob freeing procedure for the smob type
4631 specified by the tag TC. TC is the tag returned by
4632 `scm_make_smob_type'.
4634 *** Function: void scm_set_smob_print (tc, print)
4636 - Function: void scm_set_smob_print (long tc,
4637 scm_sizet (*print) (SCM,
4641 This function sets the smob printing procedure for the smob type
4642 specified by the tag TC. TC is the tag returned by
4643 `scm_make_smob_type'.
4645 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4646 This function sets the smob equality-testing predicate for the
4647 smob type specified by the tag TC. TC is the tag returned by
4648 `scm_make_smob_type'.
4650 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4651 Make VALUE contain a smob instance of the type with type code TC and
4652 smob data DATA. VALUE must be previously declared as C type `SCM'.
4654 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4655 This macro expands to a block of code that creates a smob instance
4656 of the type with type code TC and smob data DATA, and returns that
4657 `SCM' value. It should be the last piece of code in a block.
4659 ** The interfaces for using I/O ports and implementing port types
4660 (ptobs) have changed significantly. The new interface is based on
4661 shared access to buffers and a new set of ptob procedures.
4663 *** scm_newptob has been removed
4667 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4669 - Function: SCM scm_make_port_type (char *type_name,
4670 int (*fill_buffer) (SCM port),
4671 void (*write_flush) (SCM port));
4673 Similarly to the new smob interface, there is a set of function
4674 setters by which the user can customize the behaviour of his port
4675 type. See ports.h (scm_set_port_XXX).
4677 ** scm_strport_to_string: New function: creates a new string from
4678 a string port's buffer.
4680 ** Plug in interface for random number generators
4681 The variable `scm_the_rng' in random.c contains a value and three
4682 function pointers which together define the current random number
4683 generator being used by the Scheme level interface and the random
4684 number library functions.
4686 The user is free to replace the default generator with the generator
4689 *** Variable: size_t scm_the_rng.rstate_size
4690 The size of the random state type used by the current RNG
4693 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4694 Given the random STATE, return 32 random bits.
4696 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4697 Seed random state STATE using string S of length N.
4699 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4700 Given random state STATE, return a malloced copy.
4703 The default RNG is the MWC (Multiply With Carry) random number
4704 generator described by George Marsaglia at the Department of
4705 Statistics and Supercomputer Computations Research Institute, The
4706 Florida State University (http://stat.fsu.edu/~geo).
4708 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4709 passes all tests in the DIEHARD test suite
4710 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4711 costs one multiply and one add on platforms which either supports long
4712 longs (gcc does this on most systems) or have 64 bit longs. The cost
4713 is four multiply on other systems but this can be optimized by writing
4714 scm_i_uniform32 in assembler.
4716 These functions are provided through the scm_the_rng interface for use
4717 by libguile and the application.
4719 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4720 Given the random STATE, return 32 random bits.
4721 Don't use this function directly. Instead go through the plugin
4722 interface (see "Plug in interface" above).
4724 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4725 Initialize STATE using SEED of length N.
4727 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4728 Return a malloc:ed copy of STATE. This function can easily be re-used
4729 in the interfaces to other RNGs.
4731 ** Random number library functions
4732 These functions use the current RNG through the scm_the_rng interface.
4733 It might be a good idea to use these functions from your C code so
4734 that only one random generator is used by all code in your program.
4736 The default random state is stored in:
4738 *** Variable: SCM scm_var_random_state
4739 Contains the vcell of the Scheme variable "*random-state*" which is
4740 used as default state by all random number functions in the Scheme
4745 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4747 *** Function: scm_rstate *scm_c_default_rstate (void)
4748 This is a convenience function which returns the value of
4749 scm_var_random_state. An error message is generated if this value
4750 isn't a random state.
4752 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4753 Make a new random state from the string SEED of length LENGTH.
4755 It is generally not a good idea to use multiple random states in a
4756 program. While subsequent random numbers generated from one random
4757 state are guaranteed to be reasonably independent, there is no such
4758 guarantee for numbers generated from different random states.
4760 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4761 Return 32 random bits.
4763 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4764 Return a sample from the uniform(0,1) distribution.
4766 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4767 Return a sample from the normal(0,1) distribution.
4769 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4770 Return a sample from the exp(1) distribution.
4772 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4773 Return a sample from the discrete uniform(0,M) distribution.
4775 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4776 Return a sample from the discrete uniform(0,M) distribution.
4777 M must be a bignum object. The returned value may be an INUM.
4781 Changes in Guile 1.3 (released Monday, October 19, 1998):
4783 * Changes to the distribution
4785 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4786 To avoid conflicts, programs should name environment variables after
4787 themselves, except when there's a common practice establishing some
4790 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4791 giving the former precedence, and printing a warning message if the
4792 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4794 ** The header files related to multi-byte characters have been removed.
4795 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4796 which referred to these explicitly will probably need to be rewritten,
4797 since the support for the variant string types has been removed; see
4800 ** The header files append.h and sequences.h have been removed. These
4801 files implemented non-R4RS operations which would encourage
4802 non-portable programming style and less easy-to-read code.
4804 * Changes to the stand-alone interpreter
4806 ** New procedures have been added to implement a "batch mode":
4808 *** Function: batch-mode?
4810 Returns a boolean indicating whether the interpreter is in batch
4813 *** Function: set-batch-mode?! ARG
4815 If ARG is true, switches the interpreter to batch mode. The `#f'
4816 case has not been implemented.
4818 ** Guile now provides full command-line editing, when run interactively.
4819 To use this feature, you must have the readline library installed.
4820 The Guile build process will notice it, and automatically include
4823 The readline library is available via anonymous FTP from any GNU
4824 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4826 ** the-last-stack is now a fluid.
4828 * Changes to the procedure for linking libguile with your programs
4830 ** You can now use the `guile-config' utility to build programs that use Guile.
4832 Guile now includes a command-line utility called `guile-config', which
4833 can provide information about how to compile and link programs that
4836 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4837 You should include this command's output on the command line you use
4838 to compile C or C++ code that #includes the Guile header files. It's
4839 usually just a `-I' flag to help the compiler find the Guile headers.
4842 *** `guile-config link' prints any linker flags necessary to link with Guile.
4844 This command writes to its standard output a list of flags which you
4845 must pass to the linker to link your code against the Guile library.
4846 The flags include '-lguile' itself, any other libraries the Guile
4847 library depends upon, and any `-L' flags needed to help the linker
4848 find those libraries.
4850 For example, here is a Makefile rule that builds a program named 'foo'
4851 from the object files ${FOO_OBJECTS}, and links them against Guile:
4854 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4856 Previous Guile releases recommended that you use autoconf to detect
4857 which of a predefined set of libraries were present on your system.
4858 It is more robust to use `guile-config', since it records exactly which
4859 libraries the installed Guile library requires.
4861 This was originally called `build-guile', but was renamed to
4862 `guile-config' before Guile 1.3 was released, to be consistent with
4863 the analogous script for the GTK+ GUI toolkit, which is called
4867 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4869 If you are using the GNU autoconf package to configure your program,
4870 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4871 (described above) and gather the necessary values for use in your
4874 The GUILE_FLAGS macro expands to configure script code which runs the
4875 `guile-config' script, to find out where Guile's header files and
4876 libraries are installed. It sets two variables, marked for
4877 substitution, as by AC_SUBST.
4879 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4880 code that uses Guile header files. This is almost always just a
4883 GUILE_LDFLAGS --- flags to pass to the linker to link a
4884 program against Guile. This includes `-lguile' for the Guile
4885 library itself, any libraries that Guile itself requires (like
4886 -lqthreads), and so on. It may also include a -L flag to tell the
4887 compiler where to find the libraries.
4889 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4890 directory of the Guile distribution. You can copy it into your
4891 package's aclocal.m4 file, and then use it in your configure.in file.
4893 If you are using the `aclocal' program, distributed with GNU automake,
4894 to maintain your aclocal.m4 file, the Guile installation process
4895 installs guile.m4 where aclocal will find it. All you need to do is
4896 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4897 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4901 * Changes to Scheme functions and syntax
4903 ** Multi-byte strings have been removed, as have multi-byte and wide
4904 ports. We felt that these were the wrong approach to
4905 internationalization support.
4907 ** New function: readline [PROMPT]
4908 Read a line from the terminal, and allow the user to edit it,
4909 prompting with PROMPT. READLINE provides a large set of Emacs-like
4910 editing commands, lets the user recall previously typed lines, and
4911 works on almost every kind of terminal, including dumb terminals.
4913 READLINE assumes that the cursor is at the beginning of the line when
4914 it is invoked. Thus, you can't print a prompt yourself, and then call
4915 READLINE; you need to package up your prompt as a string, pass it to
4916 the function, and let READLINE print the prompt itself. This is
4917 because READLINE needs to know the prompt's screen width.
4919 For Guile to provide this function, you must have the readline
4920 library, version 2.1 or later, installed on your system. Readline is
4921 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4922 any GNU mirror site.
4924 See also ADD-HISTORY function.
4926 ** New function: add-history STRING
4927 Add STRING as the most recent line in the history used by the READLINE
4928 command. READLINE does not add lines to the history itself; you must
4929 call ADD-HISTORY to make previous input available to the user.
4931 ** The behavior of the read-line function has changed.
4933 This function now uses standard C library functions to read the line,
4934 for speed. This means that it doesn not respect the value of
4935 scm-line-incrementors; it assumes that lines are delimited with
4938 (Note that this is read-line, the function that reads a line of text
4939 from a port, not readline, the function that reads a line from a
4940 terminal, providing full editing capabilities.)
4942 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4944 This module provides some simple argument parsing. It exports one
4947 Function: getopt-gnu-style ARG-LS
4948 Parse a list of program arguments into an alist of option
4951 Each item in the list of program arguments is examined to see if
4952 it meets the syntax of a GNU long-named option. An argument like
4953 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4954 returned alist, where MUMBLE is a keyword object with the same
4955 name as the argument. An argument like `--MUMBLE=FROB' produces
4956 an element of the form (MUMBLE . FROB), where FROB is a string.
4958 As a special case, the returned alist also contains a pair whose
4959 car is the symbol `rest'. The cdr of this pair is a list
4960 containing all the items in the argument list that are not options
4961 of the form mentioned above.
4963 The argument `--' is treated specially: all items in the argument
4964 list appearing after such an argument are not examined, and are
4965 returned in the special `rest' list.
4967 This function does not parse normal single-character switches.
4968 You will need to parse them out of the `rest' list yourself.
4970 ** The read syntax for byte vectors and short vectors has changed.
4972 Instead of #bytes(...), write #y(...).
4974 Instead of #short(...), write #h(...).
4976 This may seem nutty, but, like the other uniform vectors, byte vectors
4977 and short vectors want to have the same print and read syntax (and,
4978 more basic, want to have read syntax!). Changing the read syntax to
4979 use multiple characters after the hash sign breaks with the
4980 conventions used in R5RS and the conventions used for the other
4981 uniform vectors. It also introduces complexity in the current reader,
4982 both on the C and Scheme levels. (The Right solution is probably to
4983 change the syntax and prototypes for uniform vectors entirely.)
4986 ** The new module (ice-9 session) provides useful interactive functions.
4988 *** New procedure: (apropos REGEXP OPTION ...)
4990 Display a list of top-level variables whose names match REGEXP, and
4991 the modules they are imported from. Each OPTION should be one of the
4994 value --- Show the value of each matching variable.
4995 shadow --- Show bindings shadowed by subsequently imported modules.
4996 full --- Same as both `shadow' and `value'.
5000 guile> (apropos "trace" 'full)
5001 debug: trace #<procedure trace args>
5002 debug: untrace #<procedure untrace args>
5003 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5004 the-scm-module: before-backtrace-hook ()
5005 the-scm-module: backtrace #<primitive-procedure backtrace>
5006 the-scm-module: after-backtrace-hook ()
5007 the-scm-module: has-shown-backtrace-hint? #f
5010 ** There are new functions and syntax for working with macros.
5012 Guile implements macros as a special object type. Any variable whose
5013 top-level binding is a macro object acts as a macro. The macro object
5014 specifies how the expression should be transformed before evaluation.
5016 *** Macro objects now print in a reasonable way, resembling procedures.
5018 *** New function: (macro? OBJ)
5019 True iff OBJ is a macro object.
5021 *** New function: (primitive-macro? OBJ)
5022 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5023 macro transformers, implemented in eval.c rather than Scheme code.
5025 Why do we have this function?
5026 - For symmetry with procedure? and primitive-procedure?,
5027 - to allow custom print procedures to tell whether a macro is
5028 primitive, and display it differently, and
5029 - to allow compilers and user-written evaluators to distinguish
5030 builtin special forms from user-defined ones, which could be
5033 *** New function: (macro-type OBJ)
5034 Return a value indicating what kind of macro OBJ is. Possible return
5037 The symbol `syntax' --- a macro created by procedure->syntax.
5038 The symbol `macro' --- a macro created by procedure->macro.
5039 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5040 The boolean #f --- if OBJ is not a macro object.
5042 *** New function: (macro-name MACRO)
5043 Return the name of the macro object MACRO's procedure, as returned by
5046 *** New function: (macro-transformer MACRO)
5047 Return the transformer procedure for MACRO.
5049 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5051 Specify a new macro expander to use in the current module. Each
5052 MODULE is a module name, with the same meaning as in the `use-modules'
5053 form; each named module's exported bindings are added to the current
5054 top-level environment. TRANSFORMER is an expression evaluated in the
5055 resulting environment which must yield a procedure to use as the
5056 module's eval transformer: every expression evaluated in this module
5057 is passed to this function, and the result passed to the Guile
5060 *** macro-eval! is removed. Use local-eval instead.
5062 ** Some magic has been added to the printer to better handle user
5063 written printing routines (like record printers, closure printers).
5065 The problem is that these user written routines must have access to
5066 the current `print-state' to be able to handle fancy things like
5067 detection of circular references. These print-states have to be
5068 passed to the builtin printing routines (display, write, etc) to
5069 properly continue the print chain.
5071 We didn't want to change all existing print code so that it
5072 explicitly passes thru a print state in addition to a port. Instead,
5073 we extented the possible values that the builtin printing routines
5074 accept as a `port'. In addition to a normal port, they now also take
5075 a pair of a normal port and a print-state. Printing will go to the
5076 port and the print-state will be used to control the detection of
5077 circular references, etc. If the builtin function does not care for a
5078 print-state, it is simply ignored.
5080 User written callbacks are now called with such a pair as their
5081 `port', but because every function now accepts this pair as a PORT
5082 argument, you don't have to worry about that. In fact, it is probably
5083 safest to not check for these pairs.
5085 However, it is sometimes necessary to continue a print chain on a
5086 different port, for example to get a intermediate string
5087 representation of the printed value, mangle that string somehow, and
5088 then to finally print the mangled string. Use the new function
5090 inherit-print-state OLD-PORT NEW-PORT
5092 for this. It constructs a new `port' that prints to NEW-PORT but
5093 inherits the print-state of OLD-PORT.
5095 ** struct-vtable-offset renamed to vtable-offset-user
5097 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5099 ** There is now a third optional argument to make-vtable-vtable
5100 (and fourth to make-struct) when constructing new types (vtables).
5101 This argument initializes field vtable-index-printer of the vtable.
5103 ** The detection of circular references has been extended to structs.
5104 That is, a structure that -- in the process of being printed -- prints
5105 itself does not lead to infinite recursion.
5107 ** There is now some basic support for fluids. Please read
5108 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5109 the following functions and macros:
5111 Function: make-fluid
5113 Create a new fluid object. Fluids are not special variables or
5114 some other extension to the semantics of Scheme, but rather
5115 ordinary Scheme objects. You can store them into variables (that
5116 are still lexically scoped, of course) or into any other place you
5117 like. Every fluid has a initial value of `#f'.
5119 Function: fluid? OBJ
5121 Test whether OBJ is a fluid.
5123 Function: fluid-ref FLUID
5124 Function: fluid-set! FLUID VAL
5126 Access/modify the fluid FLUID. Modifications are only visible
5127 within the current dynamic root (that includes threads).
5129 Function: with-fluids* FLUIDS VALUES THUNK
5131 FLUIDS is a list of fluids and VALUES a corresponding list of
5132 values for these fluids. Before THUNK gets called the values are
5133 installed in the fluids and the old values of the fluids are
5134 saved in the VALUES list. When the flow of control leaves THUNK
5135 or reenters it, the values get swapped again. You might think of
5136 this as a `safe-fluid-excursion'. Note that the VALUES list is
5137 modified by `with-fluids*'.
5139 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5141 The same as `with-fluids*' but with a different syntax. It looks
5142 just like `let', but both FLUID and VALUE are evaluated. Remember,
5143 fluids are not special variables but ordinary objects. FLUID
5144 should evaluate to a fluid.
5146 ** Changes to system call interfaces:
5148 *** close-port, close-input-port and close-output-port now return a
5149 boolean instead of an `unspecified' object. #t means that the port
5150 was successfully closed, while #f means it was already closed. It is
5151 also now possible for these procedures to raise an exception if an
5152 error occurs (some errors from write can be delayed until close.)
5154 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5157 *** the third argument to fcntl is now optional.
5159 *** the first argument to chown can now be a file descriptor or a port.
5161 *** the argument to stat can now be a port.
5163 *** The following new procedures have been added (most use scsh
5166 *** procedure: close PORT/FD
5167 Similar to close-port (*note close-port: Closing Ports.), but also
5168 works on file descriptors. A side effect of closing a file
5169 descriptor is that any ports using that file descriptor are moved
5170 to a different file descriptor and have their revealed counts set
5173 *** procedure: port->fdes PORT
5174 Returns the integer file descriptor underlying PORT. As a side
5175 effect the revealed count of PORT is incremented.
5177 *** procedure: fdes->ports FDES
5178 Returns a list of existing ports which have FDES as an underlying
5179 file descriptor, without changing their revealed counts.
5181 *** procedure: fdes->inport FDES
5182 Returns an existing input port which has FDES as its underlying
5183 file descriptor, if one exists, and increments its revealed count.
5184 Otherwise, returns a new input port with a revealed count of 1.
5186 *** procedure: fdes->outport FDES
5187 Returns an existing output port which has FDES as its underlying
5188 file descriptor, if one exists, and increments its revealed count.
5189 Otherwise, returns a new output port with a revealed count of 1.
5191 The next group of procedures perform a `dup2' system call, if NEWFD
5192 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5193 duplicated can be supplied as an integer or contained in a port. The
5194 type of value returned varies depending on which procedure is used.
5196 All procedures also have the side effect when performing `dup2' that
5197 any ports using NEWFD are moved to a different file descriptor and have
5198 their revealed counts set to zero.
5200 *** procedure: dup->fdes PORT/FD [NEWFD]
5201 Returns an integer file descriptor.
5203 *** procedure: dup->inport PORT/FD [NEWFD]
5204 Returns a new input port using the new file descriptor.
5206 *** procedure: dup->outport PORT/FD [NEWFD]
5207 Returns a new output port using the new file descriptor.
5209 *** procedure: dup PORT/FD [NEWFD]
5210 Returns a new port if PORT/FD is a port, with the same mode as the
5211 supplied port, otherwise returns an integer file descriptor.
5213 *** procedure: dup->port PORT/FD MODE [NEWFD]
5214 Returns a new port using the new file descriptor. MODE supplies a
5215 mode string for the port (*note open-file: File Ports.).
5217 *** procedure: setenv NAME VALUE
5218 Modifies the environment of the current process, which is also the
5219 default environment inherited by child processes.
5221 If VALUE is `#f', then NAME is removed from the environment.
5222 Otherwise, the string NAME=VALUE is added to the environment,
5223 replacing any existing string with name matching NAME.
5225 The return value is unspecified.
5227 *** procedure: truncate-file OBJ SIZE
5228 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5229 can be a string containing a file name or an integer file
5230 descriptor or port open for output on the file. The underlying
5231 system calls are `truncate' and `ftruncate'.
5233 The return value is unspecified.
5235 *** procedure: setvbuf PORT MODE [SIZE]
5236 Set the buffering mode for PORT. MODE can be:
5244 block buffered, using a newly allocated buffer of SIZE bytes.
5245 However if SIZE is zero or unspecified, the port will be made
5248 This procedure should not be used after I/O has been performed with
5251 Ports are usually block buffered by default, with a default buffer
5252 size. Procedures e.g., *Note open-file: File Ports, which accept a
5253 mode string allow `0' to be added to request an unbuffered port.
5255 *** procedure: fsync PORT/FD
5256 Copies any unwritten data for the specified output file descriptor
5257 to disk. If PORT/FD is a port, its buffer is flushed before the
5258 underlying file descriptor is fsync'd. The return value is
5261 *** procedure: open-fdes PATH FLAGS [MODES]
5262 Similar to `open' but returns a file descriptor instead of a port.
5264 *** procedure: execle PATH ENV [ARG] ...
5265 Similar to `execl', but the environment of the new process is
5266 specified by ENV, which must be a list of strings as returned by
5267 the `environ' procedure.
5269 This procedure is currently implemented using the `execve' system
5270 call, but we call it `execle' because of its Scheme calling
5273 *** procedure: strerror ERRNO
5274 Returns the Unix error message corresponding to ERRNO, an integer.
5276 *** procedure: primitive-exit [STATUS]
5277 Terminate the current process without unwinding the Scheme stack.
5278 This is would typically be useful after a fork. The exit status
5279 is STATUS if supplied, otherwise zero.
5281 *** procedure: times
5282 Returns an object with information about real and processor time.
5283 The following procedures accept such an object as an argument and
5284 return a selected component:
5287 The current real time, expressed as time units relative to an
5291 The CPU time units used by the calling process.
5294 The CPU time units used by the system on behalf of the
5298 The CPU time units used by terminated child processes of the
5299 calling process, whose status has been collected (e.g., using
5303 Similarly, the CPU times units used by the system on behalf of
5304 terminated child processes.
5306 ** Removed: list-length
5307 ** Removed: list-append, list-append!
5308 ** Removed: list-reverse, list-reverse!
5310 ** array-map renamed to array-map!
5312 ** serial-array-map renamed to serial-array-map!
5314 ** catch doesn't take #f as first argument any longer
5316 Previously, it was possible to pass #f instead of a key to `catch'.
5317 That would cause `catch' to pass a jump buffer object to the procedure
5318 passed as second argument. The procedure could then use this jump
5319 buffer objekt as an argument to throw.
5321 This mechanism has been removed since its utility doesn't motivate the
5322 extra complexity it introduces.
5324 ** The `#/' notation for lists now provokes a warning message from Guile.
5325 This syntax will be removed from Guile in the near future.
5327 To disable the warning message, set the GUILE_HUSH environment
5328 variable to any non-empty value.
5330 ** The newline character now prints as `#\newline', following the
5331 normal Scheme notation, not `#\nl'.
5333 * Changes to the gh_ interface
5335 ** The gh_enter function now takes care of loading the Guile startup files.
5336 gh_enter works by calling scm_boot_guile; see the remarks below.
5338 ** Function: void gh_write (SCM x)
5340 Write the printed representation of the scheme object x to the current
5341 output port. Corresponds to the scheme level `write'.
5343 ** gh_list_length renamed to gh_length.
5345 ** vector handling routines
5347 Several major changes. In particular, gh_vector() now resembles
5348 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5349 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5350 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5351 vector-related gh_ functions have been implemented.
5353 ** pair and list routines
5355 Implemented several of the R4RS pair and list functions that were
5358 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5360 New function. Converts double arrays back and forth between Scheme
5363 * Changes to the scm_ interface
5365 ** The function scm_boot_guile now takes care of loading the startup files.
5367 Guile's primary initialization function, scm_boot_guile, now takes
5368 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5369 Guile, define the module system, and put together some standard
5370 bindings. It also loads `init.scm', which is intended to hold
5371 site-specific initialization code.
5373 Since Guile cannot operate properly until boot-9.scm is loaded, there
5374 is no reason to separate loading boot-9.scm from Guile's other
5375 initialization processes.
5377 This job used to be done by scm_compile_shell_switches, which didn't
5378 make much sense; in particular, it meant that people using Guile for
5379 non-shell-like applications had to jump through hoops to get Guile
5380 initialized properly.
5382 ** The function scm_compile_shell_switches no longer loads the startup files.
5383 Now, Guile always loads the startup files, whenever it is initialized;
5384 see the notes above for scm_boot_guile and scm_load_startup_files.
5386 ** Function: scm_load_startup_files
5387 This new function takes care of loading Guile's initialization file
5388 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5389 this is always called by the Guile initialization process, it's
5390 probably not too useful to call this yourself, but it's there anyway.
5392 ** The semantics of smob marking have changed slightly.
5394 The smob marking function (the `mark' member of the scm_smobfuns
5395 structure) is no longer responsible for setting the mark bit on the
5396 smob. The generic smob handling code in the garbage collector will
5397 set this bit. The mark function need only ensure that any other
5398 objects the smob refers to get marked.
5400 Note that this change means that the smob's GC8MARK bit is typically
5401 already set upon entry to the mark function. Thus, marking functions
5402 which look like this:
5405 if (SCM_GC8MARKP (ptr))
5407 SCM_SETGC8MARK (ptr);
5408 ... mark objects to which the smob refers ...
5411 are now incorrect, since they will return early, and fail to mark any
5412 other objects the smob refers to. Some code in the Guile library used
5415 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5417 If you have implemented your own I/O port type, by writing the
5418 functions required by the scm_ptobfuns and then calling scm_newptob,
5419 you will need to change your functions slightly.
5421 The functions in a scm_ptobfuns structure now expect the port itself
5422 as their argument; they used to expect the `stream' member of the
5423 port's scm_port_table structure. This allows functions in an
5424 scm_ptobfuns structure to easily access the port's cell (and any flags
5425 it its CAR), and the port's scm_port_table structure.
5427 Guile now passes the I/O port itself as the `port' argument in the
5428 following scm_ptobfuns functions:
5430 int (*free) (SCM port);
5431 int (*fputc) (int, SCM port);
5432 int (*fputs) (char *, SCM port);
5433 scm_sizet (*fwrite) SCM_P ((char *ptr,
5437 int (*fflush) (SCM port);
5438 int (*fgetc) (SCM port);
5439 int (*fclose) (SCM port);
5441 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5444 If you have existing code which defines its own port types, it is easy
5445 to convert your code to the new interface; simply apply SCM_STREAM to
5446 the port argument to yield the value you code used to expect.
5448 Note that since both the port and the stream have the same type in the
5449 C code --- they are both SCM values --- the C compiler will not remind
5450 you if you forget to update your scm_ptobfuns functions.
5453 ** Function: int scm_internal_select (int fds,
5457 struct timeval *timeout);
5459 This is a replacement for the `select' function provided by the OS.
5460 It enables I/O blocking and sleeping to happen for one cooperative
5461 thread without blocking other threads. It also avoids busy-loops in
5462 these situations. It is intended that all I/O blocking and sleeping
5463 will finally go through this function. Currently, this function is
5464 only available on systems providing `gettimeofday' and `select'.
5466 ** Function: SCM scm_internal_stack_catch (SCM tag,
5467 scm_catch_body_t body,
5469 scm_catch_handler_t handler,
5472 A new sibling to the other two C level `catch' functions
5473 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5474 the stack to be saved automatically into the variable `the-last-stack'
5475 (scm_the_last_stack_var) on error. This is necessary if you want to
5476 use advanced error reporting, such as calling scm_display_error and
5477 scm_display_backtrace. (They both take a stack object as argument.)
5479 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5481 scm_catch_handler_t handler,
5484 Spawns a new thread. It does a job similar to
5485 scm_call_with_new_thread but takes arguments more suitable when
5486 spawning threads from application C code.
5488 ** The hook scm_error_callback has been removed. It was originally
5489 intended as a way for the user to install his own error handler. But
5490 that method works badly since it intervenes between throw and catch,
5491 thereby changing the semantics of expressions like (catch #t ...).
5492 The correct way to do it is to use one of the C level catch functions
5493 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5495 ** Removed functions:
5497 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5498 scm_list_reverse, scm_list_reverse_x
5500 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5502 These can be used for pretty list creation from C. The idea is taken
5503 from Erick Gallesio's STk.
5505 ** scm_array_map renamed to scm_array_map_x
5507 ** mbstrings are now removed
5509 This means that the type codes scm_tc7_mb_string and
5510 scm_tc7_mb_substring has been removed.
5512 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5514 Since we no longer support multi-byte strings, these I/O functions
5515 have been simplified, and renamed. Here are their old names, and
5516 their new names and arguments:
5518 scm_gen_putc -> void scm_putc (int c, SCM port);
5519 scm_gen_puts -> void scm_puts (char *s, SCM port);
5520 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5521 scm_gen_getc -> void scm_getc (SCM port);
5524 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5526 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5528 SCM_TYP7S now masks away the bit which distinguishes substrings from
5531 ** scm_catch_body_t: Backward incompatible change!
5533 Body functions to scm_internal_catch and friends do not any longer
5534 take a second argument. This is because it is no longer possible to
5535 pass a #f arg to catch.
5537 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5539 The function scm_protect_object protects its argument from being freed
5540 by the garbage collector. scm_unprotect_object removes that
5543 These functions now nest properly. That is, for every object O, there
5544 is a counter which scm_protect_object(O) increments and
5545 scm_unprotect_object(O) decrements, if the counter is greater than
5546 zero. Every object's counter is zero when it is first created. If an
5547 object's counter is greater than zero, the garbage collector will not
5548 reclaim its storage.
5550 This allows you to use scm_protect_object in your code without
5551 worrying that some other function you call will call
5552 scm_unprotect_object, and allow it to be freed. Assuming that the
5553 functions you call are well-behaved, and unprotect only those objects
5554 they protect, you can follow the same rule and have confidence that
5555 objects will be freed only at appropriate times.
5558 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5560 * Changes to the distribution
5562 ** Nightly snapshots are now available from ftp.red-bean.com.
5563 The old server, ftp.cyclic.com, has been relinquished to its rightful
5566 Nightly snapshots of the Guile development sources are now available via
5567 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5569 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5570 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5572 ** To run Guile without installing it, the procedure has changed a bit.
5574 If you used a separate build directory to compile Guile, you'll need
5575 to include the build directory in SCHEME_LOAD_PATH, as well as the
5576 source directory. See the `INSTALL' file for examples.
5578 * Changes to the procedure for linking libguile with your programs
5580 ** The standard Guile load path for Scheme code now includes
5581 $(datadir)/guile (usually /usr/local/share/guile). This means that
5582 you can install your own Scheme files there, and Guile will find them.
5583 (Previous versions of Guile only checked a directory whose name
5584 contained the Guile version number, so you had to re-install or move
5585 your Scheme sources each time you installed a fresh version of Guile.)
5587 The load path also includes $(datadir)/guile/site; we recommend
5588 putting individual Scheme files there. If you want to install a
5589 package with multiple source files, create a directory for them under
5592 ** Guile 1.2 will now use the Rx regular expression library, if it is
5593 installed on your system. When you are linking libguile into your own
5594 programs, this means you will have to link against -lguile, -lqt (if
5595 you configured Guile with thread support), and -lrx.
5597 If you are using autoconf to generate configuration scripts for your
5598 application, the following lines should suffice to add the appropriate
5599 libraries to your link command:
5601 ### Find Rx, quickthreads and libguile.
5602 AC_CHECK_LIB(rx, main)
5603 AC_CHECK_LIB(qt, main)
5604 AC_CHECK_LIB(guile, scm_shell)
5606 The Guile 1.2 distribution does not contain sources for the Rx
5607 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5608 retrieve it from a GNU FTP site and install it separately.
5610 * Changes to Scheme functions and syntax
5612 ** The dynamic linking features of Guile are now enabled by default.
5613 You can disable them by giving the `--disable-dynamic-linking' option
5616 (dynamic-link FILENAME)
5618 Find the object file denoted by FILENAME (a string) and link it
5619 into the running Guile application. When everything works out,
5620 return a Scheme object suitable for representing the linked object
5621 file. Otherwise an error is thrown. How object files are
5622 searched is system dependent.
5624 (dynamic-object? VAL)
5626 Determine whether VAL represents a dynamically linked object file.
5628 (dynamic-unlink DYNOBJ)
5630 Unlink the indicated object file from the application. DYNOBJ
5631 should be one of the values returned by `dynamic-link'.
5633 (dynamic-func FUNCTION DYNOBJ)
5635 Search the C function indicated by FUNCTION (a string or symbol)
5636 in DYNOBJ and return some Scheme object that can later be used
5637 with `dynamic-call' to actually call this function. Right now,
5638 these Scheme objects are formed by casting the address of the
5639 function to `long' and converting this number to its Scheme
5642 (dynamic-call FUNCTION DYNOBJ)
5644 Call the C function indicated by FUNCTION and DYNOBJ. The
5645 function is passed no arguments and its return value is ignored.
5646 When FUNCTION is something returned by `dynamic-func', call that
5647 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5648 etc.), look it up in DYNOBJ; this is equivalent to
5650 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5652 Interrupts are deferred while the C function is executing (with
5653 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5655 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5657 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5658 some arguments and return its return value. The C function is
5659 expected to take two arguments and return an `int', just like
5662 int c_func (int argc, char **argv);
5664 ARGS must be a list of strings and is converted into an array of
5665 `char *'. The array is passed in ARGV and its size in ARGC. The
5666 return value is converted to a Scheme number and returned from the
5667 call to `dynamic-args-call'.
5669 When dynamic linking is disabled or not supported on your system,
5670 the above functions throw errors, but they are still available.
5672 Here is a small example that works on GNU/Linux:
5674 (define libc-obj (dynamic-link "libc.so"))
5675 (dynamic-args-call 'rand libc-obj '())
5677 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5679 ** The #/ syntax for module names is depreciated, and will be removed
5680 in a future version of Guile. Instead of
5688 The latter syntax is more consistent with existing Lisp practice.
5690 ** Guile now does fancier printing of structures. Structures are the
5691 underlying implementation for records, which in turn are used to
5692 implement modules, so all of these object now print differently and in
5693 a more informative way.
5695 The Scheme printer will examine the builtin variable *struct-printer*
5696 whenever it needs to print a structure object. When this variable is
5697 not `#f' it is deemed to be a procedure and will be applied to the
5698 structure object and the output port. When *struct-printer* is `#f'
5699 or the procedure return `#f' the structure object will be printed in
5700 the boring #<struct 80458270> form.
5702 This hook is used by some routines in ice-9/boot-9.scm to implement
5703 type specific printing routines. Please read the comments there about
5706 One of the more specific uses of structs are records. The printing
5707 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5708 called. It should behave like a *struct-printer* procedure (described
5711 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5712 token of the form #:NAME, where NAME has the same syntax as a Scheme
5713 symbol, is the external representation of the keyword named NAME.
5714 Keyword objects print using this syntax as well, so values containing
5715 keyword objects can be read back into Guile. When used in an
5716 expression, keywords are self-quoting objects.
5718 Guile suports this read syntax, and uses this print syntax, regardless
5719 of the current setting of the `keyword' read option. The `keyword'
5720 read option only controls whether Guile recognizes the `:NAME' syntax,
5721 which is incompatible with R4RS. (R4RS says such token represent
5724 ** Guile has regular expression support again. Guile 1.0 included
5725 functions for matching regular expressions, based on the Rx library.
5726 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5727 distribution, and thus Guile had no regular expression support. Guile
5728 1.2 again supports the most commonly used functions, and supports all
5729 of SCSH's regular expression functions.
5731 If your system does not include a POSIX regular expression library,
5732 and you have not linked Guile with a third-party regexp library such as
5733 Rx, these functions will not be available. You can tell whether your
5734 Guile installation includes regular expression support by checking
5735 whether the `*features*' list includes the `regex' symbol.
5737 *** regexp functions
5739 By default, Guile supports POSIX extended regular expressions. That
5740 means that the characters `(', `)', `+' and `?' are special, and must
5741 be escaped if you wish to match the literal characters.
5743 This regular expression interface was modeled after that implemented
5744 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5745 with SCSH regular expressions.
5747 **** Function: string-match PATTERN STR [START]
5748 Compile the string PATTERN into a regular expression and compare
5749 it with STR. The optional numeric argument START specifies the
5750 position of STR at which to begin matching.
5752 `string-match' returns a "match structure" which describes what,
5753 if anything, was matched by the regular expression. *Note Match
5754 Structures::. If STR does not match PATTERN at all,
5755 `string-match' returns `#f'.
5757 Each time `string-match' is called, it must compile its PATTERN
5758 argument into a regular expression structure. This operation is
5759 expensive, which makes `string-match' inefficient if the same regular
5760 expression is used several times (for example, in a loop). For better
5761 performance, you can compile a regular expression in advance and then
5762 match strings against the compiled regexp.
5764 **** Function: make-regexp STR [FLAGS]
5765 Compile the regular expression described by STR, and return the
5766 compiled regexp structure. If STR does not describe a legal
5767 regular expression, `make-regexp' throws a
5768 `regular-expression-syntax' error.
5770 FLAGS may be the bitwise-or of one or more of the following:
5772 **** Constant: regexp/extended
5773 Use POSIX Extended Regular Expression syntax when interpreting
5774 STR. If not set, POSIX Basic Regular Expression syntax is used.
5775 If the FLAGS argument is omitted, we assume regexp/extended.
5777 **** Constant: regexp/icase
5778 Do not differentiate case. Subsequent searches using the
5779 returned regular expression will be case insensitive.
5781 **** Constant: regexp/newline
5782 Match-any-character operators don't match a newline.
5784 A non-matching list ([^...]) not containing a newline matches a
5787 Match-beginning-of-line operator (^) matches the empty string
5788 immediately after a newline, regardless of whether the FLAGS
5789 passed to regexp-exec contain regexp/notbol.
5791 Match-end-of-line operator ($) matches the empty string
5792 immediately before a newline, regardless of whether the FLAGS
5793 passed to regexp-exec contain regexp/noteol.
5795 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5796 Match the compiled regular expression REGEXP against `str'. If
5797 the optional integer START argument is provided, begin matching
5798 from that position in the string. Return a match structure
5799 describing the results of the match, or `#f' if no match could be
5802 FLAGS may be the bitwise-or of one or more of the following:
5804 **** Constant: regexp/notbol
5805 The match-beginning-of-line operator always fails to match (but
5806 see the compilation flag regexp/newline above) This flag may be
5807 used when different portions of a string are passed to
5808 regexp-exec and the beginning of the string should not be
5809 interpreted as the beginning of the line.
5811 **** Constant: regexp/noteol
5812 The match-end-of-line operator always fails to match (but see the
5813 compilation flag regexp/newline above)
5815 **** Function: regexp? OBJ
5816 Return `#t' if OBJ is a compiled regular expression, or `#f'
5819 Regular expressions are commonly used to find patterns in one string
5820 and replace them with the contents of another string.
5822 **** Function: regexp-substitute PORT MATCH [ITEM...]
5823 Write to the output port PORT selected contents of the match
5824 structure MATCH. Each ITEM specifies what should be written, and
5825 may be one of the following arguments:
5827 * A string. String arguments are written out verbatim.
5829 * An integer. The submatch with that number is written.
5831 * The symbol `pre'. The portion of the matched string preceding
5832 the regexp match is written.
5834 * The symbol `post'. The portion of the matched string
5835 following the regexp match is written.
5837 PORT may be `#f', in which case nothing is written; instead,
5838 `regexp-substitute' constructs a string from the specified ITEMs
5841 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5842 Similar to `regexp-substitute', but can be used to perform global
5843 substitutions on STR. Instead of taking a match structure as an
5844 argument, `regexp-substitute/global' takes two string arguments: a
5845 REGEXP string describing a regular expression, and a TARGET string
5846 which should be matched against this regular expression.
5848 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5851 * A function may be supplied. When this function is called, it
5852 will be passed one argument: a match structure for a given
5853 regular expression match. It should return a string to be
5854 written out to PORT.
5856 * The `post' symbol causes `regexp-substitute/global' to recurse
5857 on the unmatched portion of STR. This *must* be supplied in
5858 order to perform global search-and-replace on STR; if it is
5859 not present among the ITEMs, then `regexp-substitute/global'
5860 will return after processing a single match.
5862 *** Match Structures
5864 A "match structure" is the object returned by `string-match' and
5865 `regexp-exec'. It describes which portion of a string, if any, matched
5866 the given regular expression. Match structures include: a reference to
5867 the string that was checked for matches; the starting and ending
5868 positions of the regexp match; and, if the regexp included any
5869 parenthesized subexpressions, the starting and ending positions of each
5872 In each of the regexp match functions described below, the `match'
5873 argument must be a match structure returned by a previous call to
5874 `string-match' or `regexp-exec'. Most of these functions return some
5875 information about the original target string that was matched against a
5876 regular expression; we will call that string TARGET for easy reference.
5878 **** Function: regexp-match? OBJ
5879 Return `#t' if OBJ is a match structure returned by a previous
5880 call to `regexp-exec', or `#f' otherwise.
5882 **** Function: match:substring MATCH [N]
5883 Return the portion of TARGET matched by subexpression number N.
5884 Submatch 0 (the default) represents the entire regexp match. If
5885 the regular expression as a whole matched, but the subexpression
5886 number N did not match, return `#f'.
5888 **** Function: match:start MATCH [N]
5889 Return the starting position of submatch number N.
5891 **** Function: match:end MATCH [N]
5892 Return the ending position of submatch number N.
5894 **** Function: match:prefix MATCH
5895 Return the unmatched portion of TARGET preceding the regexp match.
5897 **** Function: match:suffix MATCH
5898 Return the unmatched portion of TARGET following the regexp match.
5900 **** Function: match:count MATCH
5901 Return the number of parenthesized subexpressions from MATCH.
5902 Note that the entire regular expression match itself counts as a
5903 subexpression, and failed submatches are included in the count.
5905 **** Function: match:string MATCH
5906 Return the original TARGET string.
5908 *** Backslash Escapes
5910 Sometimes you will want a regexp to match characters like `*' or `$'
5911 exactly. For example, to check whether a particular string represents
5912 a menu entry from an Info node, it would be useful to match it against
5913 a regexp like `^* [^:]*::'. However, this won't work; because the
5914 asterisk is a metacharacter, it won't match the `*' at the beginning of
5915 the string. In this case, we want to make the first asterisk un-magic.
5917 You can do this by preceding the metacharacter with a backslash
5918 character `\'. (This is also called "quoting" the metacharacter, and
5919 is known as a "backslash escape".) When Guile sees a backslash in a
5920 regular expression, it considers the following glyph to be an ordinary
5921 character, no matter what special meaning it would ordinarily have.
5922 Therefore, we can make the above example work by changing the regexp to
5923 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5924 to match only a single asterisk in the target string.
5926 Since the backslash is itself a metacharacter, you may force a
5927 regexp to match a backslash in the target string by preceding the
5928 backslash with itself. For example, to find variable references in a
5929 TeX program, you might want to find occurrences of the string `\let\'
5930 followed by any number of alphabetic characters. The regular expression
5931 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5932 each match a single backslash in the target string.
5934 **** Function: regexp-quote STR
5935 Quote each special character found in STR with a backslash, and
5936 return the resulting string.
5938 *Very important:* Using backslash escapes in Guile source code (as
5939 in Emacs Lisp or C) can be tricky, because the backslash character has
5940 special meaning for the Guile reader. For example, if Guile encounters
5941 the character sequence `\n' in the middle of a string while processing
5942 Scheme code, it replaces those characters with a newline character.
5943 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5944 Several of these "escape sequences" are processed by the Guile reader
5945 before your code is executed. Unrecognized escape sequences are
5946 ignored: if the characters `\*' appear in a string, they will be
5947 translated to the single character `*'.
5949 This translation is obviously undesirable for regular expressions,
5950 since we want to be able to include backslashes in a string in order to
5951 escape regexp metacharacters. Therefore, to make sure that a backslash
5952 is preserved in a string in your Guile program, you must use *two*
5953 consecutive backslashes:
5955 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5957 The string in this example is preprocessed by the Guile reader before
5958 any code is executed. The resulting argument to `make-regexp' is the
5959 string `^\* [^:]*', which is what we really want.
5961 This also means that in order to write a regular expression that
5962 matches a single backslash character, the regular expression string in
5963 the source code must include *four* backslashes. Each consecutive pair
5964 of backslashes gets translated by the Guile reader to a single
5965 backslash, and the resulting double-backslash is interpreted by the
5966 regexp engine as matching a single backslash character. Hence:
5968 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5970 The reason for the unwieldiness of this syntax is historical. Both
5971 regular expression pattern matchers and Unix string processing systems
5972 have traditionally used backslashes with the special meanings described
5973 above. The POSIX regular expression specification and ANSI C standard
5974 both require these semantics. Attempting to abandon either convention
5975 would cause other kinds of compatibility problems, possibly more severe
5976 ones. Therefore, without extending the Scheme reader to support
5977 strings with different quoting conventions (an ungainly and confusing
5978 extension when implemented in other languages), we must adhere to this
5979 cumbersome escape syntax.
5981 * Changes to the gh_ interface
5983 * Changes to the scm_ interface
5985 * Changes to system call interfaces:
5987 ** The value returned by `raise' is now unspecified. It throws an exception
5990 *** A new procedure `sigaction' can be used to install signal handlers
5992 (sigaction signum [action] [flags])
5994 signum is the signal number, which can be specified using the value
5997 If action is omitted, sigaction returns a pair: the CAR is the current
5998 signal hander, which will be either an integer with the value SIG_DFL
5999 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6000 handles the signal, or #f if a non-Scheme procedure handles the
6001 signal. The CDR contains the current sigaction flags for the handler.
6003 If action is provided, it is installed as the new handler for signum.
6004 action can be a Scheme procedure taking one argument, or the value of
6005 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6006 whatever signal handler was installed before sigaction was first used.
6007 Flags can optionally be specified for the new handler (SA_RESTART is
6008 always used if the system provides it, so need not be specified.) The
6009 return value is a pair with information about the old handler as
6012 This interface does not provide access to the "signal blocking"
6013 facility. Maybe this is not needed, since the thread support may
6014 provide solutions to the problem of consistent access to data
6017 *** A new procedure `flush-all-ports' is equivalent to running
6018 `force-output' on every port open for output.
6020 ** Guile now provides information on how it was built, via the new
6021 global variable, %guile-build-info. This variable records the values
6022 of the standard GNU makefile directory variables as an assocation
6023 list, mapping variable names (symbols) onto directory paths (strings).
6024 For example, to find out where the Guile link libraries were
6025 installed, you can say:
6027 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6030 * Changes to the scm_ interface
6032 ** The new function scm_handle_by_message_noexit is just like the
6033 existing scm_handle_by_message function, except that it doesn't call
6034 exit to terminate the process. Instead, it prints a message and just
6035 returns #f. This might be a more appropriate catch-all handler for
6036 new dynamic roots and threads.
6039 Changes in Guile 1.1 (released Friday, May 16 1997):
6041 * Changes to the distribution.
6043 The Guile 1.0 distribution has been split up into several smaller
6045 guile-core --- the Guile interpreter itself.
6046 guile-tcltk --- the interface between the Guile interpreter and
6047 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6048 is a toolkit for building graphical user interfaces.
6049 guile-rgx-ctax --- the interface between Guile and the Rx regular
6050 expression matcher, and the translator for the Ctax
6051 programming language. These are packaged together because the
6052 Ctax translator uses Rx to parse Ctax source code.
6054 This NEWS file describes the changes made to guile-core since the 1.0
6057 We no longer distribute the documentation, since it was either out of
6058 date, or incomplete. As soon as we have current documentation, we
6063 * Changes to the stand-alone interpreter
6065 ** guile now accepts command-line arguments compatible with SCSH, Olin
6066 Shivers' Scheme Shell.
6068 In general, arguments are evaluated from left to right, but there are
6069 exceptions. The following switches stop argument processing, and
6070 stash all remaining command-line arguments as the value returned by
6071 the (command-line) function.
6072 -s SCRIPT load Scheme source code from FILE, and exit
6073 -c EXPR evalute Scheme expression EXPR, and exit
6074 -- stop scanning arguments; run interactively
6076 The switches below are processed as they are encountered.
6077 -l FILE load Scheme source code from FILE
6078 -e FUNCTION after reading script, apply FUNCTION to
6079 command line arguments
6080 -ds do -s script at this point
6081 --emacs enable Emacs protocol (experimental)
6082 -h, --help display this help and exit
6083 -v, --version display version information and exit
6084 \ read arguments from following script lines
6086 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6087 which re-implements the traditional "echo" command:
6089 #!/usr/local/bin/guile -s
6092 (map (lambda (arg) (display arg) (display " "))
6096 (main (command-line))
6098 Suppose we invoke this script as follows:
6100 ekko a speckled gecko
6102 Through the magic of Unix script processing (triggered by the `#!'
6103 token at the top of the file), /usr/local/bin/guile receives the
6104 following list of command-line arguments:
6106 ("-s" "./ekko" "a" "speckled" "gecko")
6108 Unix inserts the name of the script after the argument specified on
6109 the first line of the file (in this case, "-s"), and then follows that
6110 with the arguments given to the script. Guile loads the script, which
6111 defines the `main' function, and then applies it to the list of
6112 remaining command-line arguments, ("a" "speckled" "gecko").
6114 In Unix, the first line of a script file must take the following form:
6116 #!INTERPRETER ARGUMENT
6118 where INTERPRETER is the absolute filename of the interpreter
6119 executable, and ARGUMENT is a single command-line argument to pass to
6122 You may only pass one argument to the interpreter, and its length is
6123 limited. These restrictions can be annoying to work around, so Guile
6124 provides a general mechanism (borrowed from, and compatible with,
6125 SCSH) for circumventing them.
6127 If the ARGUMENT in a Guile script is a single backslash character,
6128 `\', Guile will open the script file, parse arguments from its second
6129 and subsequent lines, and replace the `\' with them. So, for example,
6130 here is another implementation of the `ekko' script:
6132 #!/usr/local/bin/guile \
6136 (for-each (lambda (arg) (display arg) (display " "))
6140 If the user invokes this script as follows:
6142 ekko a speckled gecko
6144 Unix expands this into
6146 /usr/local/bin/guile \ ekko a speckled gecko
6148 When Guile sees the `\' argument, it replaces it with the arguments
6149 read from the second line of the script, producing:
6151 /usr/local/bin/guile -e main -s ekko a speckled gecko
6153 This tells Guile to load the `ekko' script, and apply the function
6154 `main' to the argument list ("a" "speckled" "gecko").
6156 Here is how Guile parses the command-line arguments:
6157 - Each space character terminates an argument. This means that two
6158 spaces in a row introduce an empty-string argument.
6159 - The tab character is not permitted (unless you quote it with the
6160 backslash character, as described below), to avoid confusion.
6161 - The newline character terminates the sequence of arguments, and will
6162 also terminate a final non-empty argument. (However, a newline
6163 following a space will not introduce a final empty-string argument;
6164 it only terminates the argument list.)
6165 - The backslash character is the escape character. It escapes
6166 backslash, space, tab, and newline. The ANSI C escape sequences
6167 like \n and \t are also supported. These produce argument
6168 constituents; the two-character combination \n doesn't act like a
6169 terminating newline. The escape sequence \NNN for exactly three
6170 octal digits reads as the character whose ASCII code is NNN. As
6171 above, characters produced this way are argument constituents.
6172 Backslash followed by other characters is not allowed.
6174 * Changes to the procedure for linking libguile with your programs
6176 ** Guile now builds and installs a shared guile library, if your
6177 system support shared libraries. (It still builds a static library on
6178 all systems.) Guile automatically detects whether your system
6179 supports shared libraries. To prevent Guile from buildisg shared
6180 libraries, pass the `--disable-shared' flag to the configure script.
6182 Guile takes longer to compile when it builds shared libraries, because
6183 it must compile every file twice --- once to produce position-
6184 independent object code, and once to produce normal object code.
6186 ** The libthreads library has been merged into libguile.
6188 To link a program against Guile, you now need only link against
6189 -lguile and -lqt; -lthreads is no longer needed. If you are using
6190 autoconf to generate configuration scripts for your application, the
6191 following lines should suffice to add the appropriate libraries to
6194 ### Find quickthreads and libguile.
6195 AC_CHECK_LIB(qt, main)
6196 AC_CHECK_LIB(guile, scm_shell)
6198 * Changes to Scheme functions
6200 ** Guile Scheme's special syntax for keyword objects is now optional,
6201 and disabled by default.
6203 The syntax variation from R4RS made it difficult to port some
6204 interesting packages to Guile. The routines which accepted keyword
6205 arguments (mostly in the module system) have been modified to also
6206 accept symbols whose names begin with `:'.
6208 To change the keyword syntax, you must first import the (ice-9 debug)
6210 (use-modules (ice-9 debug))
6212 Then you can enable the keyword syntax as follows:
6213 (read-set! keywords 'prefix)
6215 To disable keyword syntax, do this:
6216 (read-set! keywords #f)
6218 ** Many more primitive functions accept shared substrings as
6219 arguments. In the past, these functions required normal, mutable
6220 strings as arguments, although they never made use of this
6223 ** The uniform array functions now operate on byte vectors. These
6224 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6225 `serial-array-map', `array-map', `array-for-each', and
6228 ** The new functions `trace' and `untrace' implement simple debugging
6229 support for Scheme functions.
6231 The `trace' function accepts any number of procedures as arguments,
6232 and tells the Guile interpreter to display each procedure's name and
6233 arguments each time the procedure is invoked. When invoked with no
6234 arguments, `trace' returns the list of procedures currently being
6237 The `untrace' function accepts any number of procedures as arguments,
6238 and tells the Guile interpreter not to trace them any more. When
6239 invoked with no arguments, `untrace' untraces all curretly traced
6242 The tracing in Guile has an advantage over most other systems: we
6243 don't create new procedure objects, but mark the procedure objects
6244 themselves. This means that anonymous and internal procedures can be
6247 ** The function `assert-repl-prompt' has been renamed to
6248 `set-repl-prompt!'. It takes one argument, PROMPT.
6249 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6250 - If PROMPT is a string, we use it as a prompt.
6251 - If PROMPT is a procedure accepting no arguments, we call it, and
6252 display the result as a prompt.
6253 - Otherwise, we display "> ".
6255 ** The new function `eval-string' reads Scheme expressions from a
6256 string and evaluates them, returning the value of the last expression
6257 in the string. If the string contains no expressions, it returns an
6260 ** The new function `thunk?' returns true iff its argument is a
6261 procedure of zero arguments.
6263 ** `defined?' is now a builtin function, instead of syntax. This
6264 means that its argument should be quoted. It returns #t iff its
6265 argument is bound in the current module.
6267 ** The new syntax `use-modules' allows you to add new modules to your
6268 environment without re-typing a complete `define-module' form. It
6269 accepts any number of module names as arguments, and imports their
6270 public bindings into the current module.
6272 ** The new function (module-defined? NAME MODULE) returns true iff
6273 NAME, a symbol, is defined in MODULE, a module object.
6275 ** The new function `builtin-bindings' creates and returns a hash
6276 table containing copies of all the root module's bindings.
6278 ** The new function `builtin-weak-bindings' does the same as
6279 `builtin-bindings', but creates a doubly-weak hash table.
6281 ** The `equal?' function now considers variable objects to be
6282 equivalent if they have the same name and the same value.
6284 ** The new function `command-line' returns the command-line arguments
6285 given to Guile, as a list of strings.
6287 When using guile as a script interpreter, `command-line' returns the
6288 script's arguments; those processed by the interpreter (like `-s' or
6289 `-c') are omitted. (In other words, you get the normal, expected
6290 behavior.) Any application that uses scm_shell to process its
6291 command-line arguments gets this behavior as well.
6293 ** The new function `load-user-init' looks for a file called `.guile'
6294 in the user's home directory, and loads it if it exists. This is
6295 mostly for use by the code generated by scm_compile_shell_switches,
6296 but we thought it might also be useful in other circumstances.
6298 ** The new function `log10' returns the base-10 logarithm of its
6301 ** Changes to I/O functions
6303 *** The functions `read', `primitive-load', `read-and-eval!', and
6304 `primitive-load-path' no longer take optional arguments controlling
6305 case insensitivity and a `#' parser.
6307 Case sensitivity is now controlled by a read option called
6308 `case-insensitive'. The user can add new `#' syntaxes with the
6309 `read-hash-extend' function (see below).
6311 *** The new function `read-hash-extend' allows the user to change the
6312 syntax of Guile Scheme in a somewhat controlled way.
6314 (read-hash-extend CHAR PROC)
6315 When parsing S-expressions, if we read a `#' character followed by
6316 the character CHAR, use PROC to parse an object from the stream.
6317 If PROC is #f, remove any parsing procedure registered for CHAR.
6319 The reader applies PROC to two arguments: CHAR and an input port.
6321 *** The new functions read-delimited and read-delimited! provide a
6322 general mechanism for doing delimited input on streams.
6324 (read-delimited DELIMS [PORT HANDLE-DELIM])
6325 Read until we encounter one of the characters in DELIMS (a string),
6326 or end-of-file. PORT is the input port to read from; it defaults to
6327 the current input port. The HANDLE-DELIM parameter determines how
6328 the terminating character is handled; it should be one of the
6331 'trim omit delimiter from result
6332 'peek leave delimiter character in input stream
6333 'concat append delimiter character to returned value
6334 'split return a pair: (RESULT . TERMINATOR)
6336 HANDLE-DELIM defaults to 'peek.
6338 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6339 A side-effecting variant of `read-delimited'.
6341 The data is written into the string BUF at the indices in the
6342 half-open interval [START, END); the default interval is the whole
6343 string: START = 0 and END = (string-length BUF). The values of
6344 START and END must specify a well-defined interval in BUF, i.e.
6345 0 <= START <= END <= (string-length BUF).
6347 It returns NBYTES, the number of bytes read. If the buffer filled
6348 up without a delimiter character being found, it returns #f. If the
6349 port is at EOF when the read starts, it returns the EOF object.
6351 If an integer is returned (i.e., the read is successfully terminated
6352 by reading a delimiter character), then the HANDLE-DELIM parameter
6353 determines how to handle the terminating character. It is described
6354 above, and defaults to 'peek.
6356 (The descriptions of these functions were borrowed from the SCSH
6357 manual, by Olin Shivers and Brian Carlstrom.)
6359 *** The `%read-delimited!' function is the primitive used to implement
6360 `read-delimited' and `read-delimited!'.
6362 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6364 This returns a pair of values: (TERMINATOR . NUM-READ).
6365 - TERMINATOR describes why the read was terminated. If it is a
6366 character or the eof object, then that is the value that terminated
6367 the read. If it is #f, the function filled the buffer without finding
6368 a delimiting character.
6369 - NUM-READ is the number of characters read into BUF.
6371 If the read is successfully terminated by reading a delimiter
6372 character, then the gobble? parameter determines what to do with the
6373 terminating character. If true, the character is removed from the
6374 input stream; if false, the character is left in the input stream
6375 where a subsequent read operation will retrieve it. In either case,
6376 the character is also the first value returned by the procedure call.
6378 (The descriptions of this function was borrowed from the SCSH manual,
6379 by Olin Shivers and Brian Carlstrom.)
6381 *** The `read-line' and `read-line!' functions have changed; they now
6382 trim the terminator by default; previously they appended it to the
6383 returned string. For the old behavior, use (read-line PORT 'concat).
6385 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6386 take new optional START and END arguments, specifying the region of
6387 the array to read and write.
6389 *** The `ungetc-char-ready?' function has been removed. We feel it's
6390 inappropriate for an interface to expose implementation details this
6393 ** Changes to the Unix library and system call interface
6395 *** The new fcntl function provides access to the Unix `fcntl' system
6398 (fcntl PORT COMMAND VALUE)
6399 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6400 Values for COMMAND are:
6402 F_DUPFD duplicate a file descriptor
6403 F_GETFD read the descriptor's close-on-exec flag
6404 F_SETFD set the descriptor's close-on-exec flag to VALUE
6405 F_GETFL read the descriptor's flags, as set on open
6406 F_SETFL set the descriptor's flags, as set on open to VALUE
6407 F_GETOWN return the process ID of a socket's owner, for SIGIO
6408 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6409 FD_CLOEXEC not sure what this is
6411 For details, see the documentation for the fcntl system call.
6413 *** The arguments to `select' have changed, for compatibility with
6414 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6415 expected behavior. The MILLISECONDS parameter has been changed to
6416 MICROSECONDS, to more closely resemble the underlying system call.
6417 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6418 corresponding return set will be the same.
6420 *** The arguments to the `mknod' system call have changed. They are
6423 (mknod PATH TYPE PERMS DEV)
6424 Create a new file (`node') in the file system. PATH is the name of
6425 the file to create. TYPE is the kind of file to create; it should
6426 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6427 permission bits to give the newly created file. If TYPE is
6428 'block-special or 'char-special, DEV specifies which device the
6429 special file refers to; its interpretation depends on the kind of
6430 special file being created.
6432 *** The `fork' function has been renamed to `primitive-fork', to avoid
6433 clashing with various SCSH forks.
6435 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6436 and `recvfrom!'. They no longer accept a size for a second argument;
6437 you must pass a string to hold the received value. They no longer
6438 return the buffer. Instead, `recv' returns the length of the message
6439 received, and `recvfrom' returns a pair containing the packet's length
6440 and originating address.
6442 *** The file descriptor datatype has been removed, as have the
6443 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6444 We plan to replace these functions with a SCSH-compatible interface.
6446 *** The `create' function has been removed; it's just a special case
6449 *** There are new functions to break down process termination status
6450 values. In the descriptions below, STATUS is a value returned by
6453 (status:exit-val STATUS)
6454 If the child process exited normally, this function returns the exit
6455 code for the child process (i.e., the value passed to exit, or
6456 returned from main). If the child process did not exit normally,
6457 this function returns #f.
6459 (status:stop-sig STATUS)
6460 If the child process was suspended by a signal, this function
6461 returns the signal that suspended the child. Otherwise, it returns
6464 (status:term-sig STATUS)
6465 If the child process terminated abnormally, this function returns
6466 the signal that terminated the child. Otherwise, this function
6469 POSIX promises that exactly one of these functions will return true on
6470 a valid STATUS value.
6472 These functions are compatible with SCSH.
6474 *** There are new accessors and setters for the broken-out time vectors
6475 returned by `localtime', `gmtime', and that ilk. They are:
6477 Component Accessor Setter
6478 ========================= ============ ============
6479 seconds tm:sec set-tm:sec
6480 minutes tm:min set-tm:min
6481 hours tm:hour set-tm:hour
6482 day of the month tm:mday set-tm:mday
6483 month tm:mon set-tm:mon
6484 year tm:year set-tm:year
6485 day of the week tm:wday set-tm:wday
6486 day in the year tm:yday set-tm:yday
6487 daylight saving time tm:isdst set-tm:isdst
6488 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6489 name of time zone tm:zone set-tm:zone
6491 *** There are new accessors for the vectors returned by `uname',
6492 describing the host system:
6495 ============================================== ================
6496 name of the operating system implementation utsname:sysname
6497 network name of this machine utsname:nodename
6498 release level of the operating system utsname:release
6499 version level of the operating system utsname:version
6500 machine hardware platform utsname:machine
6502 *** There are new accessors for the vectors returned by `getpw',
6503 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6504 system's user database:
6507 ====================== =================
6508 user name passwd:name
6509 user password passwd:passwd
6512 real name passwd:gecos
6513 home directory passwd:dir
6514 shell program passwd:shell
6516 *** There are new accessors for the vectors returned by `getgr',
6517 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6518 system's group database:
6521 ======================= ============
6522 group name group:name
6523 group password group:passwd
6525 group members group:mem
6527 *** There are new accessors for the vectors returned by `gethost',
6528 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6532 ========================= ===============
6533 official name of host hostent:name
6534 alias list hostent:aliases
6535 host address type hostent:addrtype
6536 length of address hostent:length
6537 list of addresses hostent:addr-list
6539 *** There are new accessors for the vectors returned by `getnet',
6540 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6544 ========================= ===============
6545 official name of net netent:name
6546 alias list netent:aliases
6547 net number type netent:addrtype
6548 net number netent:net
6550 *** There are new accessors for the vectors returned by `getproto',
6551 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6555 ========================= ===============
6556 official protocol name protoent:name
6557 alias list protoent:aliases
6558 protocol number protoent:proto
6560 *** There are new accessors for the vectors returned by `getserv',
6561 `getservbyname', `getservbyport', and `getservent', describing
6565 ========================= ===============
6566 official service name servent:name
6567 alias list servent:aliases
6568 port number servent:port
6569 protocol to use servent:proto
6571 *** There are new accessors for the sockaddr structures returned by
6572 `accept', `getsockname', `getpeername', `recvfrom!':
6575 ======================================== ===============
6576 address format (`family') sockaddr:fam
6577 path, for file domain addresses sockaddr:path
6578 address, for internet domain addresses sockaddr:addr
6579 TCP or UDP port, for internet sockaddr:port
6581 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6582 `getprotoent', and `getservent' functions now return #f at the end of
6583 the user database. (They used to throw an exception.)
6585 Note that calling MUMBLEent function is equivalent to calling the
6586 corresponding MUMBLE function with no arguments.
6588 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6589 `setprotoent', and `setservent' routines now take no arguments.
6591 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6592 provide more useful information when they throw an exception.
6594 *** The `lnaof' function has been renamed to `inet-lnaof'.
6596 *** Guile now claims to have the `current-time' feature.
6598 *** The `mktime' function now takes an optional second argument ZONE,
6599 giving the time zone to use for the conversion. ZONE should be a
6600 string, in the same format as expected for the "TZ" environment variable.
6602 *** The `strptime' function now returns a pair (TIME . COUNT), where
6603 TIME is the parsed time as a vector, and COUNT is the number of
6604 characters from the string left unparsed. This function used to
6605 return the remaining characters as a string.
6607 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6608 The return value is now (SECONDS . MICROSECONDS); the fractional
6609 component is no longer expressed in "ticks".
6611 *** The `ticks/sec' constant has been removed, in light of the above change.
6613 * Changes to the gh_ interface
6615 ** gh_eval_str() now returns an SCM object which is the result of the
6618 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6621 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6622 and returns the array
6624 ** gh_scm2str0() is gone: there is no need to distinguish
6625 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6626 the user to interpret the data both ways.
6628 * Changes to the scm_ interface
6630 ** The new function scm_symbol_value0 provides an easy way to get a
6631 symbol's value from C code:
6633 SCM scm_symbol_value0 (char *NAME)
6634 Return the value of the symbol named by the null-terminated string
6635 NAME in the current module. If the symbol named NAME is unbound in
6636 the current module, return SCM_UNDEFINED.
6638 ** The new function scm_sysintern0 creates new top-level variables,
6639 without assigning them a value.
6641 SCM scm_sysintern0 (char *NAME)
6642 Create a new Scheme top-level variable named NAME. NAME is a
6643 null-terminated string. Return the variable's value cell.
6645 ** The function scm_internal_catch is the guts of catch. It handles
6646 all the mechanics of setting up a catch target, invoking the catch
6647 body, and perhaps invoking the handler if the body does a throw.
6649 The function is designed to be usable from C code, but is general
6650 enough to implement all the semantics Guile Scheme expects from throw.
6652 TAG is the catch tag. Typically, this is a symbol, but this function
6653 doesn't actually care about that.
6655 BODY is a pointer to a C function which runs the body of the catch;
6656 this is the code you can throw from. We call it like this:
6657 BODY (BODY_DATA, JMPBUF)
6659 BODY_DATA is just the BODY_DATA argument we received; we pass it
6660 through to BODY as its first argument. The caller can make
6661 BODY_DATA point to anything useful that BODY might need.
6662 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6663 which we have just created and initialized.
6665 HANDLER is a pointer to a C function to deal with a throw to TAG,
6666 should one occur. We call it like this:
6667 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6669 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6670 same idea as BODY_DATA above.
6671 THROWN_TAG is the tag that the user threw to; usually this is
6672 TAG, but it could be something else if TAG was #t (i.e., a
6673 catch-all), or the user threw to a jmpbuf.
6674 THROW_ARGS is the list of arguments the user passed to the THROW
6677 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6678 is just a pointer we pass through to HANDLER. We don't actually
6679 use either of those pointers otherwise ourselves. The idea is
6680 that, if our caller wants to communicate something to BODY or
6681 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6682 HANDLER can then use. Think of it as a way to make BODY and
6683 HANDLER closures, not just functions; MUMBLE_DATA points to the
6686 Of course, it's up to the caller to make sure that any data a
6687 MUMBLE_DATA needs is protected from GC. A common way to do this is
6688 to make MUMBLE_DATA a pointer to data stored in an automatic
6689 structure variable; since the collector must scan the stack for
6690 references anyway, this assures that any references in MUMBLE_DATA
6693 ** The new function scm_internal_lazy_catch is exactly like
6694 scm_internal_catch, except:
6696 - It does not unwind the stack (this is the major difference).
6697 - If handler returns, its value is returned from the throw.
6698 - BODY always receives #f as its JMPBUF argument (since there's no
6699 jmpbuf associated with a lazy catch, because we don't unwind the
6702 ** scm_body_thunk is a new body function you can pass to
6703 scm_internal_catch if you want the body to be like Scheme's `catch'
6704 --- a thunk, or a function of one argument if the tag is #f.
6706 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6707 contains the Scheme procedure to invoke as the body, and the tag
6708 we're catching. If the tag is #f, then we pass JMPBUF (created by
6709 scm_internal_catch) to the body procedure; otherwise, the body gets
6712 ** scm_handle_by_proc is a new handler function you can pass to
6713 scm_internal_catch if you want the handler to act like Scheme's catch
6714 --- call a procedure with the tag and the throw arguments.
6716 If the user does a throw to this catch, this function runs a handler
6717 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6718 variable holding the Scheme procedure object to invoke. It ought to
6719 be a pointer to an automatic variable (i.e., one living on the stack),
6720 or the procedure object should be otherwise protected from GC.
6722 ** scm_handle_by_message is a new handler function to use with
6723 `scm_internal_catch' if you want Guile to print a message and die.
6724 It's useful for dealing with throws to uncaught keys at the top level.
6726 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6727 message header to print; if zero, we use "guile" instead. That
6728 text is followed by a colon, then the message described by ARGS.
6730 ** The return type of scm_boot_guile is now void; the function does
6731 not return a value, and indeed, never returns at all.
6733 ** The new function scm_shell makes it easy for user applications to
6734 process command-line arguments in a way that is compatible with the
6735 stand-alone guile interpreter (which is in turn compatible with SCSH,
6738 To use the scm_shell function, first initialize any guile modules
6739 linked into your application, and then call scm_shell with the values
6740 of ARGC and ARGV your `main' function received. scm_shell will add
6741 any SCSH-style meta-arguments from the top of the script file to the
6742 argument vector, and then process the command-line arguments. This
6743 generally means loading a script file or starting up an interactive
6744 command interpreter. For details, see "Changes to the stand-alone
6747 ** The new functions scm_get_meta_args and scm_count_argv help you
6748 implement the SCSH-style meta-argument, `\'.
6750 char **scm_get_meta_args (int ARGC, char **ARGV)
6751 If the second element of ARGV is a string consisting of a single
6752 backslash character (i.e. "\\" in Scheme notation), open the file
6753 named by the following argument, parse arguments from it, and return
6754 the spliced command line. The returned array is terminated by a
6757 For details of argument parsing, see above, under "guile now accepts
6758 command-line arguments compatible with SCSH..."
6760 int scm_count_argv (char **ARGV)
6761 Count the arguments in ARGV, assuming it is terminated by a null
6764 For an example of how these functions might be used, see the source
6765 code for the function scm_shell in libguile/script.c.
6767 You will usually want to use scm_shell instead of calling this
6770 ** The new function scm_compile_shell_switches turns an array of
6771 command-line arguments into Scheme code to carry out the actions they
6772 describe. Given ARGC and ARGV, it returns a Scheme expression to
6773 evaluate, and calls scm_set_program_arguments to make any remaining
6774 command-line arguments available to the Scheme code. For example,
6775 given the following arguments:
6777 -e main -s ekko a speckled gecko
6779 scm_set_program_arguments will return the following expression:
6781 (begin (load "ekko") (main (command-line)) (quit))
6783 You will usually want to use scm_shell instead of calling this
6786 ** The function scm_shell_usage prints a usage message appropriate for
6787 an interpreter that uses scm_compile_shell_switches to handle its
6788 command-line arguments.
6790 void scm_shell_usage (int FATAL, char *MESSAGE)
6791 Print a usage message to the standard error output. If MESSAGE is
6792 non-zero, write it before the usage message, followed by a newline.
6793 If FATAL is non-zero, exit the process, using FATAL as the
6794 termination status. (If you want to be compatible with Guile,
6795 always use 1 as the exit status when terminating due to command-line
6798 You will usually want to use scm_shell instead of calling this
6801 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6802 expressions. It used to return SCM_EOL. Earth-shattering.
6804 ** The macros for declaring scheme objects in C code have been
6805 rearranged slightly. They are now:
6807 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6808 Declare a static SCM variable named C_NAME, and initialize it to
6809 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6810 be a C identifier, and SCHEME_NAME should be a C string.
6812 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6813 Just like SCM_SYMBOL, but make C_NAME globally visible.
6815 SCM_VCELL (C_NAME, SCHEME_NAME)
6816 Create a global variable at the Scheme level named SCHEME_NAME.
6817 Declare a static SCM variable named C_NAME, and initialize it to
6818 point to the Scheme variable's value cell.
6820 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6821 Just like SCM_VCELL, but make C_NAME globally visible.
6823 The `guile-snarf' script writes initialization code for these macros
6824 to its standard output, given C source code as input.
6826 The SCM_GLOBAL macro is gone.
6828 ** The scm_read_line and scm_read_line_x functions have been replaced
6829 by Scheme code based on the %read-delimited! procedure (known to C
6830 code as scm_read_delimited_x). See its description above for more
6833 ** The function scm_sys_open has been renamed to scm_open. It now
6834 returns a port instead of an FD object.
6836 * The dynamic linking support has changed. For more information, see
6837 libguile/DYNAMIC-LINKING.
6842 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6845 * Changes to the 'guile' program:
6847 ** Guile now loads some new files when it starts up. Guile first
6848 searches the load path for init.scm, and loads it if found. Then, if
6849 Guile is not being used to execute a script, and the user's home
6850 directory contains a file named `.guile', Guile loads that.
6852 ** You can now use Guile as a shell script interpreter.
6854 To paraphrase the SCSH manual:
6856 When Unix tries to execute an executable file whose first two
6857 characters are the `#!', it treats the file not as machine code to
6858 be directly executed by the native processor, but as source code
6859 to be executed by some interpreter. The interpreter to use is
6860 specified immediately after the #! sequence on the first line of
6861 the source file. The kernel reads in the name of the interpreter,
6862 and executes that instead. It passes the interpreter the source
6863 filename as its first argument, with the original arguments
6864 following. Consult the Unix man page for the `exec' system call
6865 for more information.
6867 Now you can use Guile as an interpreter, using a mechanism which is a
6868 compatible subset of that provided by SCSH.
6870 Guile now recognizes a '-s' command line switch, whose argument is the
6871 name of a file of Scheme code to load. It also treats the two
6872 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6873 to make a file of Scheme code directly executable by Unix, insert the
6874 following two lines at the top of the file:
6876 #!/usr/local/bin/guile -s
6879 Guile treats the argument of the `-s' command-line switch as the name
6880 of a file of Scheme code to load, and treats the sequence `#!' as the
6881 start of a block comment, terminated by `!#'.
6883 For example, here's a version of 'echo' written in Scheme:
6885 #!/usr/local/bin/guile -s
6887 (let loop ((args (cdr (program-arguments))))
6890 (display (car args))
6891 (if (pair? (cdr args))
6893 (loop (cdr args)))))
6896 Why does `#!' start a block comment terminated by `!#', instead of the
6897 end of the line? That is the notation SCSH uses, and although we
6898 don't yet support the other SCSH features that motivate that choice,
6899 we would like to be backward-compatible with any existing Guile
6900 scripts once we do. Furthermore, if the path to Guile on your system
6901 is too long for your kernel, you can start the script with this
6905 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6908 Note that some very old Unix systems don't support the `#!' syntax.
6911 ** You can now run Guile without installing it.
6913 Previous versions of the interactive Guile interpreter (`guile')
6914 couldn't start up unless Guile's Scheme library had been installed;
6915 they used the value of the environment variable `SCHEME_LOAD_PATH'
6916 later on in the startup process, but not to find the startup code
6917 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6920 To run Guile without installing it, build it in the normal way, and
6921 then set the environment variable `SCHEME_LOAD_PATH' to a
6922 colon-separated list of directories, including the top-level directory
6923 of the Guile sources. For example, if you unpacked Guile so that the
6924 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6927 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6930 ** Guile's read-eval-print loop no longer prints #<unspecified>
6931 results. If the user wants to see this, she can evaluate the
6932 expression (assert-repl-print-unspecified #t), perhaps in her startup
6935 ** Guile no longer shows backtraces by default when an error occurs;
6936 however, it does display a message saying how to get one, and how to
6937 request that they be displayed by default. After an error, evaluate
6939 to see a backtrace, and
6940 (debug-enable 'backtrace)
6941 to see them by default.
6945 * Changes to Guile Scheme:
6947 ** Guile now distinguishes between #f and the empty list.
6949 This is for compatibility with the IEEE standard, the (possibly)
6950 upcoming Revised^5 Report on Scheme, and many extant Scheme
6953 Guile used to have #f and '() denote the same object, to make Scheme's
6954 type system more compatible with Emacs Lisp's. However, the change
6955 caused too much trouble for Scheme programmers, and we found another
6956 way to reconcile Emacs Lisp with Scheme that didn't require this.
6959 ** Guile's delq, delv, delete functions, and their destructive
6960 counterparts, delq!, delv!, and delete!, now remove all matching
6961 elements from the list, not just the first. This matches the behavior
6962 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6963 functions which inspired them.
6965 I recognize that this change may break code in subtle ways, but it
6966 seems best to make the change before the FSF's first Guile release,
6970 ** The compiled-library-path function has been deleted from libguile.
6972 ** The facilities for loading Scheme source files have changed.
6974 *** The variable %load-path now tells Guile which directories to search
6975 for Scheme code. Its value is a list of strings, each of which names
6978 *** The variable %load-extensions now tells Guile which extensions to
6979 try appending to a filename when searching the load path. Its value
6980 is a list of strings. Its default value is ("" ".scm").
6982 *** (%search-load-path FILENAME) searches the directories listed in the
6983 value of the %load-path variable for a Scheme file named FILENAME,
6984 with all the extensions listed in %load-extensions. If it finds a
6985 match, then it returns its full filename. If FILENAME is absolute, it
6986 returns it unchanged. Otherwise, it returns #f.
6988 %search-load-path will not return matches that refer to directories.
6990 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6991 uses %seach-load-path to find a file named FILENAME, and loads it if
6992 it finds it. If it can't read FILENAME for any reason, it throws an
6995 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6998 *** load uses the same searching semantics as primitive-load.
7000 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7001 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7002 path, and load-module-with-path have been deleted. The functions
7003 above should serve their purposes.
7005 *** If the value of the variable %load-hook is a procedure,
7006 `primitive-load' applies its value to the name of the file being
7007 loaded (without the load path directory name prepended). If its value
7008 is #f, it is ignored. Otherwise, an error occurs.
7010 This is mostly useful for printing load notification messages.
7013 ** The function `eval!' is no longer accessible from the scheme level.
7014 We can't allow operations which introduce glocs into the scheme level,
7015 because Guile's type system can't handle these as data. Use `eval' or
7016 `read-and-eval!' (see below) as replacement.
7018 ** The new function read-and-eval! reads an expression from PORT,
7019 evaluates it, and returns the result. This is more efficient than
7020 simply calling `read' and `eval', since it is not necessary to make a
7021 copy of the expression for the evaluator to munge.
7023 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7024 for the `read' function.
7027 ** The function `int?' has been removed; its definition was identical
7028 to that of `integer?'.
7030 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7031 use the R4RS names for these functions.
7033 ** The function object-properties no longer returns the hash handle;
7034 it simply returns the object's property list.
7036 ** Many functions have been changed to throw errors, instead of
7037 returning #f on failure. The point of providing exception handling in
7038 the language is to simplify the logic of user code, but this is less
7039 useful if Guile's primitives don't throw exceptions.
7041 ** The function `fileno' has been renamed from `%fileno'.
7043 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7046 * Changes to Guile's C interface:
7048 ** The library's initialization procedure has been simplified.
7049 scm_boot_guile now has the prototype:
7051 void scm_boot_guile (int ARGC,
7053 void (*main_func) (),
7056 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7057 MAIN_FUNC should do all the work of the program (initializing other
7058 packages, reading user input, etc.) before returning. When MAIN_FUNC
7059 returns, call exit (0); this function never returns. If you want some
7060 other exit value, MAIN_FUNC may call exit itself.
7062 scm_boot_guile arranges for program-arguments to return the strings
7063 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7064 scm_set_program_arguments with the final list, so Scheme code will
7065 know which arguments have been processed.
7067 scm_boot_guile establishes a catch-all catch handler which prints an
7068 error message and exits the process. This means that Guile exits in a
7069 coherent way when system errors occur and the user isn't prepared to
7070 handle it. If the user doesn't like this behavior, they can establish
7071 their own universal catcher in MAIN_FUNC to shadow this one.
7073 Why must the caller do all the real work from MAIN_FUNC? The garbage
7074 collector assumes that all local variables of type SCM will be above
7075 scm_boot_guile's stack frame on the stack. If you try to manipulate
7076 SCM values after this function returns, it's the luck of the draw
7077 whether the GC will be able to find the objects you allocate. So,
7078 scm_boot_guile function exits, rather than returning, to discourage
7079 people from making that mistake.
7081 The IN, OUT, and ERR arguments were removed; there are other
7082 convenient ways to override these when desired.
7084 The RESULT argument was deleted; this function should never return.
7086 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7090 ** Guile's header files should no longer conflict with your system's
7093 In order to compile code which #included <libguile.h>, previous
7094 versions of Guile required you to add a directory containing all the
7095 Guile header files to your #include path. This was a problem, since
7096 Guile's header files have names which conflict with many systems'
7099 Now only <libguile.h> need appear in your #include path; you must
7100 refer to all Guile's other header files as <libguile/mumble.h>.
7101 Guile's installation procedure puts libguile.h in $(includedir), and
7102 the rest in $(includedir)/libguile.
7105 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7106 have been added to the Guile library.
7108 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7109 OBJ will not be freed, even if all other references are dropped,
7110 until someone does scm_unprotect_object (OBJ). Both functions
7113 Note that calls to scm_protect_object do not nest. You can call
7114 scm_protect_object any number of times on a given object, and the
7115 next call to scm_unprotect_object will unprotect it completely.
7117 Basically, scm_protect_object and scm_unprotect_object just
7118 maintain a list of references to things. Since the GC knows about
7119 this list, all objects it mentions stay alive. scm_protect_object
7120 adds its argument to the list; scm_unprotect_object remove its
7121 argument from the list.
7124 ** scm_eval_0str now returns the value of the last expression
7127 ** The new function scm_read_0str reads an s-expression from a
7128 null-terminated string, and returns it.
7130 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7131 to a Scheme port object.
7133 ** The new function `scm_set_program_arguments' allows C code to set
7134 the value returned by the Scheme `program-arguments' function.
7139 * Guile no longer includes sophisticated Tcl/Tk support.
7141 The old Tcl/Tk support was unsatisfying to us, because it required the
7142 user to link against the Tcl library, as well as Tk and Guile. The
7143 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7144 referring to widgets by names, rather than exporting widgets to Scheme
7145 code as a special datatype.
7147 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7148 maintainers described some very interesting changes in progress to the
7149 Tcl/Tk internals, which would facilitate clean interfaces between lone
7150 Tk and other interpreters --- even for garbage-collected languages
7151 like Scheme. They expected the new Tk to be publicly available in the
7154 Since it seems that Guile might soon have a new, cleaner interface to
7155 lone Tk, and that the old Guile/Tk glue code would probably need to be
7156 completely rewritten, we (Jim Blandy and Richard Stallman) have
7157 decided not to support the old code. We'll spend the time instead on
7158 a good interface to the newer Tk, as soon as it is available.
7160 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7163 Copyright information:
7165 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7167 Permission is granted to anyone to make or distribute verbatim copies
7168 of this document as received, in any medium, provided that the
7169 copyright notice and this permission notice are preserved,
7170 thus giving the recipient permission to redistribute in turn.
7172 Permission is granted to distribute modified versions
7173 of this document, or of portions of it,
7174 under the above conditions, provided also that they
7175 carry prominent notices stating who last changed them.
7180 paragraph-separate: "[
\f]*$"