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
22 * Changes to the C interface
25 Changes in 1.8.1 (since 1.8.0):
27 * LFS functions are now used to access 64-bit files on 32-bit systems.
29 * New procedures (see the manual for details)
31 ** primitive-_exit - [Scheme] the-root-module
32 ** scm_primitive__exit - [C]
33 ** make-completion-function - [Scheme] (ice-9 readline)
34 ** scm_c_locale_stringn_to_number - [C]
35 ** scm_srfi1_append_reverse [C]
36 ** scm_srfi1_append_reverse_x [C]
42 * New `(ice-9 i18n)' module (see the manual for details)
46 ** Build problems have been fixed on MacOS, SunOS, and QNX.
48 ** A one-dimensional array can now be 'equal?' to a vector.
50 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
52 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
54 ** `record-accessor' and `record-modifier' now have strict type checks.
56 Record accessor and modifier procedures now throw an error if the
57 record type of the record they're given is not the type expected.
58 (Previously accessors returned #f and modifiers silently did nothing).
60 ** It is now OK to use both autoload and use-modules on a given module.
62 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
64 Previously there was no checking on primatives like make-vector that
65 accept "one or two" arguments. Now there is.
67 ** The srfi-1 assoc function now calls its equality predicate properly.
69 Previously srfi-1 assoc would call the equality predicate with the key
70 last. According to the SRFI, the key should be first.
72 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
74 ** The array-set! procedure no longer segfaults when given a bit vector.
76 ** Bugs in make-shared-array have been fixed.
78 ** string<? and friends now follow char<? etc order on 8-bit chars.
80 ** The format procedure now handles inf and nan values for ~f correctly.
82 ** exact->inexact should no longer overflow when given certain large fractions.
84 ** srfi-9 accessor and modifier procedures now have strict record type checks.
86 This matches the srfi-9 specification.
88 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
90 Previously the (ice-9 ftw) procedures would ignore any file that had
91 the same inode number as a file they had already seen, even if that
92 file was on a different device.
95 Changes in 1.8.0 (changes since the 1.6.x series):
97 * Changes to the distribution
99 ** Guile is now licensed with the GNU Lesser General Public License.
101 ** The manual is now licensed with the GNU Free Documentation License.
103 ** Guile now requires GNU MP (http://swox.com/gmp).
105 Guile now uses the GNU MP library for arbitrary precision arithmetic.
107 ** Guile now has separate private and public configuration headers.
109 That is, things like HAVE_STRING_H no longer leak from Guile's
112 ** Guile now provides and uses an "effective" version number.
114 Guile now provides scm_effective_version and effective-version
115 functions which return the "effective" version number. This is just
116 the normal full version string without the final micro-version number,
117 so the current effective-version is "1.8". The effective version
118 should remain unchanged during a stable series, and should be used for
119 items like the versioned share directory name
120 i.e. /usr/share/guile/1.8.
122 Providing an unchanging version number during a stable release for
123 things like the versioned share directory can be particularly
124 important for Guile "add-on" packages, since it provides a directory
125 that they can install to that won't be changed out from under them
126 with each micro release during a stable series.
128 ** Thread implementation has changed.
130 When you configure "--with-threads=null", you will get the usual
131 threading API (call-with-new-thread, make-mutex, etc), but you can't
132 actually create new threads. Also, "--with-threads=no" is now
133 equivalent to "--with-threads=null". This means that the thread API
134 is always present, although you might not be able to create new
137 When you configure "--with-threads=pthreads" or "--with-threads=yes",
138 you will get threads that are implemented with the portable POSIX
139 threads. These threads can run concurrently (unlike the previous
140 "coop" thread implementation), but need to cooperate for things like
143 The default is "pthreads", unless your platform doesn't have pthreads,
144 in which case "null" threads are used.
146 See the manual for details, nodes "Initialization", "Multi-Threading",
147 "Blocking", and others.
149 ** There is the new notion of 'discouraged' features.
151 This is a milder form of deprecation.
153 Things that are discouraged should not be used in new code, but it is
154 OK to leave them in old code for now. When a discouraged feature is
155 used, no warning message is printed like there is for 'deprecated'
156 features. Also, things that are merely discouraged are nevertheless
157 implemented efficiently, while deprecated features can be very slow.
159 You can omit discouraged features from libguile by configuring it with
160 the '--disable-discouraged' option.
162 ** Deprecation warnings can be controlled at run-time.
164 (debug-enable 'warn-deprecated) switches them on and (debug-disable
165 'warn-deprecated) switches them off.
167 ** Support for SRFI 61, extended cond syntax for multiple values has
170 This SRFI is always available.
172 ** Support for require-extension, SRFI-55, has been added.
174 The SRFI-55 special form `require-extension' has been added. It is
175 available at startup, and provides a portable way to load Scheme
176 extensions. SRFI-55 only requires support for one type of extension,
177 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
180 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
182 The (srfi srfi-26) module is an implementation of SRFI-26 which
183 provides the `cut' and `cute' syntax. These may be used to specialize
184 parameters without currying.
186 ** New module (srfi srfi-31)
188 This is an implementation of SRFI-31 which provides a special form
189 `rec' for recursive evaluation.
191 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
192 been merged with the core, making their functionality always
195 The modules are still available, tho, and you could use them together
196 with a renaming import, for example.
198 ** Guile no longer includes its own version of libltdl.
200 The official version is good enough now.
202 ** The --enable-htmldoc option has been removed from 'configure'.
204 Support for translating the documentation into HTML is now always
205 provided. Use 'make html'.
207 ** New module (ice-9 serialize):
209 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
210 don't trust the thread safety of most of your program, but where you
211 have some section(s) of code which you consider can run in parallel to
212 other sections. See ice-9/serialize.scm for more information.
214 ** The configure option '--disable-arrays' has been removed.
216 Support for arrays and uniform numeric arrays is now always included
219 * Changes to the stand-alone interpreter
221 ** New command line option `-L'.
223 This option adds a directory to the front of the load path.
225 ** New command line option `--no-debug'.
227 Specifying `--no-debug' on the command line will keep the debugging
228 evaluator turned off, even for interactive sessions.
230 ** User-init file ~/.guile is now loaded with the debugging evaluator.
232 Previously, the normal evaluator would have been used. Using the
233 debugging evaluator gives better error messages.
235 ** The '-e' option now 'read's its argument.
237 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
238 be used with '-e'. For example, you can now write a script like
241 exec guile -e '(@ (demo) main)' -s "$0" "$@"
244 (define-module (demo)
248 (format #t "Demo: ~a~%" args))
251 * Changes to Scheme functions and syntax
253 ** Guardians have changed back to their original semantics
255 Guardians now behave like described in the paper by Dybvig et al. In
256 particular, they no longer make guarantees about the order in which
257 they return objects, and they can no longer be greedy.
259 They no longer drop cyclic data structures.
261 The C function scm_make_guardian has been changed incompatibly and no
262 longer takes the 'greedy_p' argument.
264 ** New function hashx-remove!
266 This function completes the set of 'hashx' functions.
268 ** The concept of dynamic roots has been factored into continuation
269 barriers and dynamic states.
271 Each thread has a current dynamic state that carries the values of the
272 fluids. You can create and copy dynamic states and use them as the
273 second argument for 'eval'. See "Fluids and Dynamic States" in the
276 To restrict the influence that captured continuations can have on the
277 control flow, you can errect continuation barriers. See "Continuation
278 Barriers" in the manual.
280 The function call-with-dynamic-root now essentially temporarily
281 installs a new dynamic state and errects a continuation barrier.
283 ** The default load path no longer includes "." at the end.
285 Automatically loading modules from the current directory should not
286 happen by default. If you want to allow it in a more controlled
287 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
290 ** The uniform vector and array support has been overhauled.
292 It now complies with SRFI-4 and the weird prototype based uniform
293 array creation has been deprecated. See the manual for more details.
295 Some non-compatible changes have been made:
296 - characters can no longer be stored into byte arrays.
297 - strings and bit vectors are no longer considered to be uniform numeric
299 - array-rank throws an error for non-arrays instead of returning zero.
300 - array-ref does no longer accept non-arrays when no indices are given.
302 There is the new notion of 'generalized vectors' and corresponding
303 procedures like 'generalized-vector-ref'. Generalized vectors include
304 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
306 Arrays use generalized vectors as their storage, so that you still
307 have arrays of characters, bits, etc. However, uniform-array-read!
308 and uniform-array-write can no longer read/write strings and
311 ** There is now support for copy-on-write substrings, mutation-sharing
312 substrings and read-only strings.
314 Three new procedures are related to this: substring/shared,
315 substring/copy, and substring/read-only. See the manual for more
318 ** Backtraces will now highlight the value that caused the error.
320 By default, these values are enclosed in "{...}", such as in this
329 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
330 <unnamed port>:1:1: Wrong type (expecting pair): a
331 ABORT: (wrong-type-arg)
333 The prefix and suffix used for highlighting can be set via the two new
334 printer options 'highlight-prefix' and 'highlight-suffix'. For
335 example, putting this into ~/.guile will output the bad value in bold
338 (print-set! highlight-prefix "\x1b[1m")
339 (print-set! highlight-suffix "\x1b[22m")
342 ** 'gettext' support for internationalization has been added.
344 See the manual for details.
346 ** New syntax '@' and '@@':
348 You can now directly refer to variables exported from a module by
351 (@ MODULE-NAME VARIABLE-NAME)
353 For example (@ (ice-9 pretty-print) pretty-print) will directly access
354 the pretty-print variable exported from the (ice-9 pretty-print)
355 module. You don't need to 'use' that module first. You can also use
356 '@' as a target of 'set!', as in (set! (@ mod var) val).
358 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
359 but it can also access variables that have not been exported. It is
360 intended only for kluges and temporary fixes and for debugging, not
363 ** Keyword syntax has been made more disciplined.
365 Previously, the name of a keyword was read as a 'token' but printed as
366 a symbol. Now, it is read as a general Scheme datum which must be a
377 ERROR: In expression (a b c):
381 ERROR: Unbound variable: foo
386 ERROR: Wrong type (expecting symbol): 12
390 ERROR: Wrong type (expecting symbol): (a b c)
394 ** The printing of symbols that might look like keywords can be
397 The new printer option 'quote-keywordish-symbols' controls how symbols
398 are printed that have a colon as their first or last character. The
399 default now is to only quote a symbol with #{...}# when the read
400 option 'keywords' is not '#f'. Thus:
402 guile> (define foo (string->symbol ":foo"))
403 guile> (read-set! keywords #f)
406 guile> (read-set! keywords 'prefix)
409 guile> (print-set! quote-keywordish-symbols #f)
413 ** 'while' now provides 'break' and 'continue'
415 break and continue were previously bound in a while loop, but not
416 documented, and continue didn't quite work properly. The undocumented
417 parameter to break which gave a return value for the while has been
420 ** 'call-with-current-continuation' is now also available under the name
423 ** The module system now checks for duplicate bindings.
425 The module system now can check for name conflicts among imported
428 The behavior can be controlled by specifying one or more 'duplicates'
429 handlers. For example, to make Guile return an error for every name
437 The new default behavior of the module system when a name collision
438 has been detected is to
440 1. Give priority to bindings marked as a replacement.
441 2. Issue a warning (different warning if overriding core binding).
442 3. Give priority to the last encountered binding (this corresponds to
445 If you want the old behavior back without replacements or warnings you
448 (default-duplicate-binding-handler 'last)
450 to your .guile init file.
452 ** New define-module option: :replace
454 :replace works as :export, but, in addition, marks the binding as a
457 A typical example is `format' in (ice-9 format) which is a replacement
458 for the core binding `format'.
460 ** Adding prefixes to imported bindings in the module system
462 There is now a new :use-module option :prefix. It can be used to add
463 a prefix to all imported bindings.
466 :use-module ((bar) :prefix bar:))
468 will import all bindings exported from bar, but rename them by adding
471 ** Conflicting generic functions can be automatically merged.
473 When two imported bindings conflict and they are both generic
474 functions, the two functions can now be merged automatically. This is
475 activated with the 'duplicates' handler 'merge-generics'.
477 ** New function: effective-version
479 Returns the "effective" version number. This is just the normal full
480 version string without the final micro-version number. See "Changes
481 to the distribution" above.
483 ** New threading functions: parallel, letpar, par-map, and friends
485 These are convenient ways to run calculations in parallel in new
486 threads. See "Parallel forms" in the manual for details.
488 ** New function 'try-mutex'.
490 This function will attempt to lock a mutex but will return immediately
491 instead of blocking and indicate failure.
493 ** Waiting on a condition variable can have a timeout.
495 The function 'wait-condition-variable' now takes a third, optional
496 argument that specifies the point in time where the waiting should be
499 ** New function 'broadcast-condition-variable'.
501 ** New functions 'all-threads' and 'current-thread'.
503 ** Signals and system asyncs work better with threads.
505 The function 'sigaction' now takes a fourth, optional, argument that
506 specifies the thread that the handler should run in. When the
507 argument is omitted, the handler will run in the thread that called
510 Likewise, 'system-async-mark' takes a second, optional, argument that
511 specifies the thread that the async should run in. When it is
512 omitted, the async will run in the thread that called
515 C code can use the new functions scm_sigaction_for_thread and
516 scm_system_async_mark_for_thread to pass the new thread argument.
518 When a thread blocks on a mutex, a condition variable or is waiting
519 for IO to be possible, it will still execute system asyncs. This can
520 be used to interrupt such a thread by making it execute a 'throw', for
523 ** The function 'system-async' is deprecated.
525 You can now pass any zero-argument procedure to 'system-async-mark'.
526 The function 'system-async' will just return its argument unchanged
529 ** New functions 'call-with-blocked-asyncs' and
530 'call-with-unblocked-asyncs'
532 The expression (call-with-blocked-asyncs PROC) will call PROC and will
533 block execution of system asyncs for the current thread by one level
534 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
535 procedure and will unblock the execution of system asyncs by one
536 level for the current thread.
538 Only system asyncs are affected by these functions.
540 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
542 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
543 instead. Those functions are easier to use correctly and can be
546 ** New function 'unsetenv'.
548 ** New macro 'define-syntax-public'.
550 It works like 'define-syntax' and also exports the defined macro (but
553 ** There is support for Infinity and NaNs.
555 Following PLT Scheme, Guile can now work with infinite numbers, and
558 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
559 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
560 "+nan.0"). These numbers are inexact and have no exact counterpart.
562 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
563 sign of the dividend. The infinities are integers, and they answer #t
564 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
565 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
576 ERROR: Numerical overflow
578 Two new predicates 'inf?' and 'nan?' can be used to test for the
581 ** Inexact zero can have a sign.
583 Guile can now distinguish between plus and minus inexact zero, if your
584 platform supports this, too. The two zeros are equal according to
585 '=', but not according to 'eqv?'. For example
596 ** Guile now has exact rationals.
598 Guile can now represent fractions such as 1/3 exactly. Computing with
599 them is also done exactly, of course:
604 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
607 For example: (floor 2) now returns an exact 2 where in the past it
608 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
610 ** inexact->exact no longer returns only integers.
612 Without exact rationals, the closest exact number was always an
613 integer, but now inexact->exact returns the fraction that is exactly
614 equal to a floating point number. For example:
616 (inexact->exact 1.234)
617 => 694680242521899/562949953421312
619 When you want the old behavior, use 'round' explicitly:
621 (inexact->exact (round 1.234))
624 ** New function 'rationalize'.
626 This function finds a simple fraction that is close to a given real
627 number. For example (and compare with inexact->exact above):
629 (rationalize (inexact->exact 1.234) 1/2000)
632 Note that, as required by R5RS, rationalize returns only then an exact
633 result when both its arguments are exact.
635 ** 'odd?' and 'even?' work also for inexact integers.
637 Previously, (odd? 1.0) would signal an error since only exact integers
638 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
639 returns #f and (odd? 1.5) signals an error.
641 ** Guile now has uninterned symbols.
643 The new function 'make-symbol' will return an uninterned symbol. This
644 is a symbol that is unique and is guaranteed to remain unique.
645 However, uninterned symbols can not yet be read back in.
647 Use the new function 'symbol-interned?' to check whether a symbol is
650 ** pretty-print has more options.
652 The function pretty-print from the (ice-9 pretty-print) module can now
653 also be invoked with keyword arguments that control things like
654 maximum output width. See the manual for details.
656 ** Variables have no longer a special behavior for `equal?'.
658 Previously, comparing two variables with `equal?' would recursivly
659 compare their values. This is no longer done. Variables are now only
660 `equal?' if they are `eq?'.
662 ** `(begin)' is now valid.
664 You can now use an empty `begin' form. It will yield #<unspecified>
665 when evaluated and simply be ignored in a definition context.
667 ** Deprecated: procedure->macro
669 Change your code to use 'define-macro' or r5rs macros. Also, be aware
670 that macro expansion will not be done during evaluation, but prior to
673 ** Soft ports now allow a `char-ready?' procedure
675 The vector argument to `make-soft-port' can now have a length of
676 either 5 or 6. (Previously the length had to be 5.) The optional 6th
677 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
678 that returns the number of characters that can be read immediately
679 without the soft port blocking.
681 ** Deprecated: undefine
683 There is no replacement for undefine.
685 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
686 have been discouraged.
688 They are relics from a time where a keyword like #:foo was used
689 directly as a Tcl option "-foo" and thus keywords were internally
690 stored as a symbol with a starting dash. We now store a symbol
693 Use symbol->keyword and keyword->symbol instead.
695 ** The `cheap' debug option is now obsolete
697 Evaluator trap calls are now unconditionally "cheap" - in other words,
698 they pass a debug object to the trap handler rather than a full
699 continuation. The trap handler code can capture a full continuation
700 by using `call-with-current-continuation' in the usual way, if it so
703 The `cheap' option is retained for now so as not to break existing
704 code which gets or sets it, but setting it now has no effect. It will
705 be removed in the next major Guile release.
707 ** Evaluator trap calls now support `tweaking'
709 `Tweaking' means that the trap handler code can modify the Scheme
710 expression that is about to be evaluated (in the case of an
711 enter-frame trap) or the value that is being returned (in the case of
712 an exit-frame trap). The trap handler code indicates that it wants to
713 do this by returning a pair whose car is the symbol 'instead and whose
714 cdr is the modified expression or return value.
716 * Changes to the C interface
718 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
719 take a 'delete' function argument.
721 This argument makes no sense since the delete function is used to
722 remove a pair from an alist, and this must not be configurable.
724 This is an incompatible change.
726 ** The GH interface is now subject to the deprecation mechanism
728 The GH interface has been deprecated for quite some time but now it is
729 actually removed from Guile when it is configured with
730 --disable-deprecated.
732 See the manual "Transitioning away from GH" for more information.
734 ** A new family of functions for converting between C values and
735 Scheme values has been added.
737 These functions follow a common naming scheme and are designed to be
738 easier to use, thread-safe and more future-proof than the older
743 These are predicates that return a C boolean: 1 or 0. Instead of
744 SCM_NFALSEP, you can now use scm_is_true, for example.
746 - <type> scm_to_<type> (SCM val, ...)
748 These are functions that convert a Scheme value into an appropriate
749 C value. For example, you can use scm_to_int to safely convert from
752 - SCM scm_from_<type> (<type> val, ...)
754 These functions convert from a C type to a SCM value; for example,
755 scm_from_int for ints.
757 There is a huge number of these functions, for numbers, strings,
758 symbols, vectors, etc. They are documented in the reference manual in
759 the API section together with the types that they apply to.
761 ** New functions for dealing with complex numbers in C have been added.
763 The new functions are scm_c_make_rectangular, scm_c_make_polar,
764 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
765 They work like scm_make_rectangular etc but take or return doubles
768 ** The function scm_make_complex has been discouraged.
770 Use scm_c_make_rectangular instead.
772 ** The INUM macros have been deprecated.
774 A lot of code uses these macros to do general integer conversions,
775 although the macros only work correctly with fixnums. Use the
776 following alternatives.
778 SCM_INUMP -> scm_is_integer or similar
779 SCM_NINUMP -> !scm_is_integer or similar
780 SCM_MAKINUM -> scm_from_int or similar
781 SCM_INUM -> scm_to_int or similar
783 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
784 do the validating for you.
786 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
787 have been discouraged.
789 Use the newer scm_to_<type> and scm_from_<type> functions instead for
790 new code. The functions have been discouraged since they don't fit
793 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
795 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
796 evaluates its argument twice. Use scm_is_true, etc. instead for new
799 ** The macro SCM_EQ_P has been discouraged.
801 Use scm_is_eq for new code, which fits better into the naming
804 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
807 Use the function scm_is_pair or scm_is_null instead.
809 ** The functions scm_round and scm_truncate have been deprecated and
810 are now available as scm_c_round and scm_c_truncate, respectively.
812 These functions occupy the names that scm_round_number and
813 scm_truncate_number should have.
815 ** The functions scm_c_string2str, scm_c_substring2str, and
816 scm_c_symbol2str have been deprecated.
818 Use scm_to_locale_stringbuf or similar instead, maybe together with
821 ** New functions scm_c_make_string, scm_c_string_length,
822 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
823 scm_c_substring_shared, scm_c_substring_copy.
825 These are like scm_make_string, scm_length, etc. but are slightly
826 easier to use from C.
828 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
829 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
831 They export too many assumptions about the implementation of strings
832 and symbols that are no longer true in the presence of
833 mutation-sharing substrings and when Guile switches to some form of
836 When working with strings, it is often best to use the normal string
837 functions provided by Guile, such as scm_c_string_ref,
838 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
839 manual since many more such functions are now provided than
842 When you want to convert a SCM string to a C string, use the
843 scm_to_locale_string function or similar instead. For symbols, use
844 scm_symbol_to_string and then work with that string. Because of the
845 new string representation, scm_symbol_to_string does not need to copy
846 and is thus quite efficient.
848 ** Some string, symbol and keyword functions have been discouraged.
850 They don't fit into the uniform naming scheme and are not explicit
851 about the character encoding.
853 Replace according to the following table:
855 scm_allocate_string -> scm_c_make_string
856 scm_take_str -> scm_take_locale_stringn
857 scm_take0str -> scm_take_locale_string
858 scm_mem2string -> scm_from_locale_stringn
859 scm_str2string -> scm_from_locale_string
860 scm_makfrom0str -> scm_from_locale_string
861 scm_mem2symbol -> scm_from_locale_symboln
862 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
863 scm_str2symbol -> scm_from_locale_symbol
865 SCM_SYMBOL_HASH -> scm_hashq
866 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
868 scm_c_make_keyword -> scm_from_locale_keyword
870 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
871 now also available to C code.
873 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
875 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
876 the latter returns the true name of the keyword, not the 'dash name',
877 as SCM_KEYWORDSYM used to do.
879 ** A new way to access arrays in a thread-safe and efficient way has
882 See the manual, node "Accessing Arrays From C".
884 ** The old uniform vector and bitvector implementations have been
885 unceremoniously removed.
887 This implementation exposed the details of the tagging system of
888 Guile. Use the new C API explained in the manual in node "Uniform
889 Numeric Vectors" and "Bit Vectors", respectively.
891 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
892 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
893 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
894 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
895 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
896 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
899 ** The macros dealing with vectors have been deprecated.
901 Use the new functions scm_is_vector, scm_vector_elements,
902 scm_vector_writable_elements, etc, or scm_is_simple_vector,
903 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
904 manual for more details.
906 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
907 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
909 The following macros have been removed: SCM_VECTOR_BASE,
910 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
911 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
913 ** Some C functions and macros related to arrays have been deprecated.
915 Migrate according to the following table:
917 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
918 scm_make_ra -> scm_make_array
919 scm_shap2ra -> scm_make_array
920 scm_cvref -> scm_c_generalized_vector_ref
921 scm_ra_set_contp -> do not use
922 scm_aind -> scm_array_handle_pos
923 scm_raprin1 -> scm_display or scm_write
925 SCM_ARRAYP -> scm_is_array
926 SCM_ARRAY_NDIM -> scm_c_array_rank
927 SCM_ARRAY_DIMS -> scm_array_handle_dims
928 SCM_ARRAY_CONTP -> do not use
929 SCM_ARRAY_MEM -> do not use
930 SCM_ARRAY_V -> scm_array_handle_elements or similar
931 SCM_ARRAY_BASE -> do not use
933 ** SCM_CELL_WORD_LOC has been deprecated.
935 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
936 to a SCM, as opposed to a pointer to a scm_t_bits.
938 This was done to allow the correct use of pointers into the Scheme
939 heap. Previously, the heap words were of type scm_t_bits and local
940 variables and function arguments were of type SCM, making it
941 non-standards-conformant to have a pointer that can point to both.
943 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
945 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
946 second and third words of double smobs. Likewise for
947 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
949 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
950 used to get and set the 16 exra bits in the zeroth word of a smob.
952 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
953 accesing the first immediate word of a smob as a SCM value, and there
954 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
955 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
957 ** New way to deal with non-local exits and re-entries.
959 There is a new set of functions that essentially do what
960 scm_internal_dynamic_wind does, but in a way that is more convenient
961 for C code in some situations. Here is a quick example of how to
962 prevent a potential memory leak:
969 scm_dynwind_begin (0);
971 mem = scm_malloc (100);
972 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
974 /* MEM would leak if BAR throws an error.
975 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
982 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
983 SCM_DYNWIND_END as well.
987 For full documentation, see the node "Dynamic Wind" in the manual.
989 ** New function scm_dynwind_free
991 This function calls 'free' on a given pointer when a dynwind context
992 is left. Thus the call to scm_dynwind_unwind_handler above could be
993 replaced with simply scm_dynwind_free (mem).
995 ** New functions scm_c_call_with_blocked_asyncs and
996 scm_c_call_with_unblocked_asyncs
998 Like scm_call_with_blocked_asyncs etc. but for C functions.
1000 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1002 In addition to scm_c_call_with_blocked_asyncs you can now also use
1003 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1004 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1006 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1007 SCM_REALLOW_INTS have been deprecated.
1009 They do no longer fulfill their original role of blocking signal
1010 delivery. Depending on what you want to achieve, replace a pair of
1011 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1012 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1015 ** The value 'scm_mask_ints' is no longer writable.
1017 Previously, you could set scm_mask_ints directly. This is no longer
1018 possible. Use scm_c_call_with_blocked_asyncs and
1019 scm_c_call_with_unblocked_asyncs instead.
1021 ** New way to temporarily set the current input, output or error ports
1023 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1024 context' (see above). <foo> is one of "input", "output" or "error".
1026 ** New way to temporarily set fluids
1028 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1029 above) to temporarily set the value of a fluid.
1031 ** New types scm_t_intmax and scm_t_uintmax.
1033 On platforms that have them, these types are identical to intmax_t and
1034 uintmax_t, respectively. On other platforms, they are identical to
1035 the largest integer types that Guile knows about.
1037 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1039 You should not have used them.
1041 ** Many public #defines with generic names have been made private.
1043 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1044 private or renamed with a more suitable public name.
1046 ** The macro SCM_TYP16S has been deprecated.
1048 This macro is not intended for public use.
1050 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1052 Use scm_is_true (scm_inexact_p (...)) instead.
1054 ** The macro SCM_SLOPPY_REALP has been deprecated.
1056 Use scm_is_real instead.
1058 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1060 Use scm_is_complex instead.
1062 ** Some preprocessor defines have been deprecated.
1064 These defines indicated whether a certain feature was present in Guile
1065 or not. Going forward, assume that the features are always present.
1067 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1068 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1070 The following macros have been removed completely: MEMOIZE_LOCALS,
1071 SCM_RECKLESS, SCM_CAUTIOUS.
1073 ** The preprocessor define STACK_DIRECTION has been deprecated.
1075 There should be no need to know about the stack direction for ordinary
1078 ** New function: scm_effective_version
1080 Returns the "effective" version number. This is just the normal full
1081 version string without the final micro-version number. See "Changes
1082 to the distribution" above.
1084 ** The function scm_call_with_new_thread has a new prototype.
1086 Instead of taking a list with the thunk and handler, these two
1087 arguments are now passed directly:
1089 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1091 This is an incompatible change.
1093 ** New snarfer macro SCM_DEFINE_PUBLIC.
1095 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1096 function in the init section.
1098 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1100 ** Garbage collector rewrite.
1102 The garbage collector is cleaned up a lot, and now uses lazy
1103 sweeping. This is reflected in the output of (gc-stats); since cells
1104 are being freed when they are allocated, the cells-allocated field
1105 stays roughly constant.
1107 For malloc related triggers, the behavior is changed. It uses the same
1108 heuristic as the cell-triggered collections. It may be tuned with the
1109 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1110 for minimum yield of malloc related triggers. The default is 40.
1111 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1114 Debugging operations for the freelist have been deprecated, along with
1115 the C variables that control garbage collection. The environment
1116 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1117 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1119 For understanding the memory usage of a GUILE program, the routine
1120 gc-live-object-stats returns an alist containing the number of live
1121 objects for every type.
1124 ** The function scm_definedp has been renamed to scm_defined_p
1126 The name scm_definedp is deprecated.
1128 ** The struct scm_cell type has been renamed to scm_t_cell
1130 This is in accordance to Guile's naming scheme for types. Note that
1131 the name scm_cell is now used for a function that allocates and
1132 initializes a new cell (see below).
1134 ** New functions for memory management
1136 A new set of functions for memory management has been added since the
1137 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1138 indeed, Guile itself contained some long standing bugs that could
1139 cause aborts in long running programs.
1141 The new functions are more symmetrical and do not need cooperation
1142 from smob free routines, among other improvements.
1144 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1145 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1146 scm_gc_free, scm_gc_register_collectable_memory, and
1147 scm_gc_unregister_collectable_memory. Refer to the manual for more
1148 details and for upgrading instructions.
1150 The old functions for memory management have been deprecated. They
1151 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1152 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1154 ** Declarations of exported features are marked with SCM_API.
1156 Every declaration of a feature that belongs to the exported Guile API
1157 has been marked by adding the macro "SCM_API" to the start of the
1158 declaration. This macro can expand into different things, the most
1159 common of which is just "extern" for Unix platforms. On Win32, it can
1160 be used to control which symbols are exported from a DLL.
1162 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1163 will expand into "__declspec (dllimport) extern", which is needed for
1164 linking to the Guile DLL in Windows.
1166 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1167 SCM_SRFI4_IMPORT, for the corresponding libraries.
1169 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1171 Use the new functions scm_cell and scm_double_cell instead. The old
1172 macros had problems because with them allocation and initialization
1173 was separated and the GC could sometimes observe half initialized
1174 cells. Only careful coding by the user of SCM_NEWCELL and
1175 SCM_NEWCELL2 could make this safe and efficient.
1177 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1179 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1182 ** SRCBRKP has been deprecated.
1184 Use scm_c_source_property_breakpoint_p instead.
1186 ** Deprecated: scm_makmacro
1188 Change your code to use either scm_makmmacro or to define macros in
1189 Scheme, using 'define-macro'.
1191 ** New function scm_c_port_for_each.
1193 This function is like scm_port_for_each but takes a pointer to a C
1194 function as the callback instead of a SCM value.
1196 ** The names scm_internal_select, scm_thread_sleep, and
1197 scm_thread_usleep have been discouraged.
1199 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1201 ** The GC can no longer be blocked.
1203 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1204 The GC can now run (partially) concurrently with other code and thus
1205 blocking it is not well defined.
1207 ** Many definitions have been removed that were previously deprecated.
1209 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1210 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1211 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1212 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1213 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1214 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1215 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1216 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1217 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1218 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1219 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1220 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1221 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1222 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1223 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1224 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1225 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1226 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1227 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1228 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1229 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1230 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1231 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1232 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1233 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1234 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1235 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1236 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1237 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1238 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1239 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1241 * Changes to bundled modules
1245 Using the (ice-9 debug) module no longer automatically switches Guile
1246 to use the debugging evaluator. If you want to switch to the
1247 debugging evaluator (which is needed for backtrace information if you
1248 hit an error), please add an explicit "(debug-enable 'debug)" to your
1249 code just after the code to use (ice-9 debug).
1252 Changes since Guile 1.4:
1254 * Changes to the distribution
1256 ** A top-level TODO file is included.
1258 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1260 Guile now always uses three numbers to represent the version,
1261 i.e. "1.6.5". The first number, 1, is the major version number, the
1262 second number, 6, is the minor version number, and the third number,
1263 5, is the micro version number. Changes in major version number
1264 indicate major changes in Guile.
1266 Minor version numbers that are even denote stable releases, and odd
1267 minor version numbers denote development versions (which may be
1268 unstable). The micro version number indicates a minor sub-revision of
1269 a given MAJOR.MINOR release.
1271 In keeping with the new scheme, (minor-version) and scm_minor_version
1272 no longer return everything but the major version number. They now
1273 just return the minor version number. Two new functions
1274 (micro-version) and scm_micro_version have been added to report the
1275 micro version number.
1277 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1279 ** New preprocessor definitions are available for checking versions.
1281 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1282 SCM_MICRO_VERSION to the appropriate integer values.
1284 ** Guile now actively warns about deprecated features.
1286 The new configure option `--enable-deprecated=LEVEL' and the
1287 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1288 See INSTALL and README for more information.
1290 ** Guile is much more likely to work on 64-bit architectures.
1292 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1293 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1294 for the use of a test machine, and thanks to Stefan Jahn for ia64
1297 ** New functions: setitimer and getitimer.
1299 These implement a fairly direct interface to the libc functions of the
1302 ** The #. reader extension is now disabled by default.
1304 For safety reasons, #. evaluation is disabled by default. To
1305 re-enable it, set the fluid read-eval? to #t. For example:
1307 (fluid-set! read-eval? #t)
1309 but make sure you realize the potential security risks involved. With
1310 read-eval? enabled, reading a data file from an untrusted source can
1313 ** New SRFI modules have been added:
1315 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1318 (srfi srfi-1) is a library containing many useful pair- and list-processing
1321 (srfi srfi-2) exports and-let*.
1323 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1325 (srfi srfi-6) is a dummy module for now, since guile already provides
1326 all of the srfi-6 procedures by default: open-input-string,
1327 open-output-string, get-output-string.
1329 (srfi srfi-8) exports receive.
1331 (srfi srfi-9) exports define-record-type.
1333 (srfi srfi-10) exports define-reader-ctor and implements the reader
1336 (srfi srfi-11) exports let-values and let*-values.
1338 (srfi srfi-13) implements the SRFI String Library.
1340 (srfi srfi-14) implements the SRFI Character-Set Library.
1342 (srfi srfi-17) implements setter and getter-with-setter and redefines
1343 some accessor procedures as procedures with getters. (such as car,
1344 cdr, vector-ref etc.)
1346 (srfi srfi-19) implements the SRFI Time/Date Library.
1348 ** New scripts / "executable modules"
1350 Subdirectory "scripts" contains Scheme modules that are packaged to
1351 also be executable as scripts. At this time, these scripts are available:
1360 See README there for more info.
1362 These scripts can be invoked from the shell with the new program
1363 "guile-tools", which keeps track of installation directory for you.
1366 $ guile-tools display-commentary srfi/*.scm
1368 guile-tools is copied to the standard $bindir on "make install".
1370 ** New module (ice-9 stack-catch):
1372 stack-catch is like catch, but saves the current state of the stack in
1373 the fluid the-last-stack. This fluid can be useful when using the
1374 debugger and when re-throwing an error.
1376 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1378 This has been done to prevent problems on lesser operating systems
1379 that can't tolerate `*'s in file names. The exported macro continues
1380 to be named `and-let*', of course.
1382 On systems that support it, there is also a compatibility module named
1383 (ice-9 and-let*). It will go away in the next release.
1385 ** New modules (oop goops) etc.:
1388 (oop goops describe)
1390 (oop goops active-slot)
1391 (oop goops composite-slot)
1393 The Guile Object Oriented Programming System (GOOPS) has been
1394 integrated into Guile. For further information, consult the GOOPS
1395 manual and tutorial in the `doc' directory.
1397 ** New module (ice-9 rdelim).
1399 This exports the following procedures which were previously defined
1400 in the default environment:
1402 read-line read-line! read-delimited read-delimited! %read-delimited!
1403 %read-line write-line
1405 For backwards compatibility the definitions are still imported into the
1406 default environment in this version of Guile. However you should add:
1408 (use-modules (ice-9 rdelim))
1410 to any program which uses the definitions, since this may change in
1413 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1414 can be used for similar functionality.
1416 ** New module (ice-9 rw)
1418 This is a subset of the (scsh rw) module from guile-scsh. Currently
1419 it defines two procedures:
1421 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1423 Read characters from a port or file descriptor into a string STR.
1424 A port must have an underlying file descriptor -- a so-called
1425 fport. This procedure is scsh-compatible and can efficiently read
1428 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1430 Write characters from a string STR to a port or file descriptor.
1431 A port must have an underlying file descriptor -- a so-called
1432 fport. This procedure is mostly compatible and can efficiently
1433 write large strings.
1435 ** New module (ice-9 match)
1437 This module includes Andrew K. Wright's pattern matcher. See
1438 ice-9/match.scm for brief description or
1440 http://www.star-lab.com/wright/code.html
1442 for complete documentation.
1444 ** New module (ice-9 buffered-input)
1446 This module provides procedures to construct an input port from an
1447 underlying source of input that reads and returns its input in chunks.
1448 The underlying input source is a Scheme procedure, specified by the
1449 caller, which the port invokes whenever it needs more input.
1451 This is useful when building an input port whose back end is Readline
1452 or a UI element such as the GtkEntry widget.
1456 The reference and tutorial documentation that was previously
1457 distributed separately, as `guile-doc', is now included in the core
1458 Guile distribution. The documentation consists of the following
1461 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1464 - The Guile Reference Manual (guile.texi) contains (or is intended to
1465 contain) reference documentation on all aspects of Guile.
1467 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1468 reference documentation for using GOOPS, Guile's Object Oriented
1471 - The Revised^5 Report on the Algorithmic Language Scheme
1474 See the README file in the `doc' directory for more details.
1476 ** There are a couple of examples in the examples/ directory now.
1478 * Changes to the stand-alone interpreter
1480 ** New command line option `--use-srfi'
1482 Using this option, SRFI modules can be loaded on startup and be
1483 available right from the beginning. This makes programming portable
1484 Scheme programs easier.
1486 The option `--use-srfi' expects a comma-separated list of numbers,
1487 each representing a SRFI number to be loaded into the interpreter
1488 before starting evaluating a script file or the REPL. Additionally,
1489 the feature identifier for the loaded SRFIs is recognized by
1490 `cond-expand' when using this option.
1493 $ guile --use-srfi=8,13
1494 guile> (receive (x z) (values 1 2) (+ 1 2))
1496 guile> (string-pad "bla" 20)
1499 ** Guile now always starts up in the `(guile-user)' module.
1501 Previously, scripts executed via the `-s' option would run in the
1502 `(guile)' module and the repl would run in the `(guile-user)' module.
1503 Now every user action takes place in the `(guile-user)' module by
1506 * Changes to Scheme functions and syntax
1508 ** Character classifiers work for non-ASCII characters.
1510 The predicates `char-alphabetic?', `char-numeric?',
1511 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1512 no longer check whether their arguments are ASCII characters.
1513 Previously, a character would only be considered alphabetic when it
1514 was also ASCII, for example.
1516 ** Previously deprecated Scheme functions have been removed:
1518 tag - no replacement.
1519 fseek - replaced by seek.
1520 list* - replaced by cons*.
1522 ** It's now possible to create modules with controlled environments
1526 (use-modules (ice-9 safe))
1527 (define m (make-safe-module))
1528 ;;; m will now be a module containing only a safe subset of R5RS
1529 (eval '(+ 1 2) m) --> 3
1530 (eval 'load m) --> ERROR: Unbound variable: load
1532 ** Evaluation of "()", the empty list, is now an error.
1534 Previously, the expression "()" evaluated to the empty list. This has
1535 been changed to signal a "missing expression" error. The correct way
1536 to write the empty list as a literal constant is to use quote: "'()".
1538 ** New concept of `Guile Extensions'.
1540 A Guile Extension is just a ordinary shared library that can be linked
1541 at run-time. We found it advantageous to give this simple concept a
1542 dedicated name to distinguish the issues related to shared libraries
1543 from the issues related to the module system.
1545 *** New function: load-extension
1547 Executing (load-extension lib init) is mostly equivalent to
1549 (dynamic-call init (dynamic-link lib))
1551 except when scm_register_extension has been called previously.
1552 Whenever appropriate, you should use `load-extension' instead of
1553 dynamic-link and dynamic-call.
1555 *** New C function: scm_c_register_extension
1557 This function registers a initialization function for use by
1558 `load-extension'. Use it when you don't want specific extensions to
1559 be loaded as shared libraries (for example on platforms that don't
1560 support dynamic linking).
1562 ** Auto-loading of compiled-code modules is deprecated.
1564 Guile used to be able to automatically find and link a shared
1565 library to satisfy requests for a module. For example, the module
1566 `(foo bar)' could be implemented by placing a shared library named
1567 "foo/libbar.so" (or with a different extension) in a directory on the
1570 This has been found to be too tricky, and is no longer supported. The
1571 shared libraries are now called "extensions". You should now write a
1572 small Scheme file that calls `load-extension' to load the shared
1573 library and initialize it explicitly.
1575 The shared libraries themselves should be installed in the usual
1576 places for shared libraries, with names like "libguile-foo-bar".
1578 For example, place this into a file "foo/bar.scm"
1580 (define-module (foo bar))
1582 (load-extension "libguile-foo-bar" "foobar_init")
1584 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1586 `eval' is now R5RS, that is it takes two arguments.
1587 The second argument is an environment specifier, i.e. either
1589 (scheme-report-environment 5)
1590 (null-environment 5)
1591 (interaction-environment)
1597 ** The module system has been made more disciplined.
1599 The function `eval' will save and restore the current module around
1600 the evaluation of the specified expression. While this expression is
1601 evaluated, `(current-module)' will now return the right module, which
1602 is the module specified as the second argument to `eval'.
1604 A consequence of this change is that `eval' is not particularly
1605 useful when you want allow the evaluated code to change what module is
1606 designated as the current module and have this change persist from one
1607 call to `eval' to the next. The read-eval-print-loop is an example
1608 where `eval' is now inadequate. To compensate, there is a new
1609 function `primitive-eval' that does not take a module specifier and
1610 that does not save/restore the current module. You should use this
1611 function together with `set-current-module', `current-module', etc
1612 when you want to have more control over the state that is carried from
1613 one eval to the next.
1615 Additionally, it has been made sure that forms that are evaluated at
1616 the top level are always evaluated with respect to the current module.
1617 Previously, subforms of top-level forms such as `begin', `case',
1618 etc. did not respect changes to the current module although these
1619 subforms are at the top-level as well.
1621 To prevent strange behavior, the forms `define-module',
1622 `use-modules', `use-syntax', and `export' have been restricted to only
1623 work on the top level. The forms `define-public' and
1624 `defmacro-public' only export the new binding on the top level. They
1625 behave just like `define' and `defmacro', respectively, when they are
1626 used in a lexical environment.
1628 Also, `export' will no longer silently re-export bindings imported
1629 from a used module. It will emit a `deprecation' warning and will
1630 cease to perform any re-export in the next version. If you actually
1631 want to re-export bindings, use the new `re-export' in place of
1632 `export'. The new `re-export' will not make copies of variables when
1633 rexporting them, as `export' did wrongly.
1635 ** Module system now allows selection and renaming of imported bindings
1637 Previously, when using `use-modules' or the `#:use-module' clause in
1638 the `define-module' form, all the bindings (association of symbols to
1639 values) for imported modules were added to the "current module" on an
1640 as-is basis. This has been changed to allow finer control through two
1641 new facilities: selection and renaming.
1643 You can now select which of the imported module's bindings are to be
1644 visible in the current module by using the `:select' clause. This
1645 clause also can be used to rename individual bindings. For example:
1647 ;; import all bindings no questions asked
1648 (use-modules (ice-9 common-list))
1650 ;; import four bindings, renaming two of them;
1651 ;; the current module sees: every some zonk-y zonk-n
1652 (use-modules ((ice-9 common-list)
1654 (remove-if . zonk-y)
1655 (remove-if-not . zonk-n))))
1657 You can also programmatically rename all selected bindings using the
1658 `:renamer' clause, which specifies a proc that takes a symbol and
1659 returns another symbol. Because it is common practice to use a prefix,
1660 we now provide the convenience procedure `symbol-prefix-proc'. For
1663 ;; import four bindings, renaming two of them specifically,
1664 ;; and all four w/ prefix "CL:";
1665 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1666 (use-modules ((ice-9 common-list)
1668 (remove-if . zonk-y)
1669 (remove-if-not . zonk-n))
1670 :renamer (symbol-prefix-proc 'CL:)))
1672 ;; import four bindings, renaming two of them specifically,
1673 ;; and all four by upcasing.
1674 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1675 (define (upcase-symbol sym)
1676 (string->symbol (string-upcase (symbol->string sym))))
1678 (use-modules ((ice-9 common-list)
1680 (remove-if . zonk-y)
1681 (remove-if-not . zonk-n))
1682 :renamer upcase-symbol))
1684 Note that programmatic renaming is done *after* individual renaming.
1685 Also, the above examples show `use-modules', but the same facilities are
1686 available for the `#:use-module' clause of `define-module'.
1688 See manual for more info.
1690 ** The semantics of guardians have changed.
1692 The changes are for the most part compatible. An important criterion
1693 was to keep the typical usage of guardians as simple as before, but to
1694 make the semantics safer and (as a result) more useful.
1696 *** All objects returned from guardians are now properly alive.
1698 It is now guaranteed that any object referenced by an object returned
1699 from a guardian is alive. It's now impossible for a guardian to
1700 return a "contained" object before its "containing" object.
1702 One incompatible (but probably not very important) change resulting
1703 from this is that it is no longer possible to guard objects that
1704 indirectly reference themselves (i.e. are parts of cycles). If you do
1705 so accidentally, you'll get a warning.
1707 *** There are now two types of guardians: greedy and sharing.
1709 If you call (make-guardian #t) or just (make-guardian), you'll get a
1710 greedy guardian, and for (make-guardian #f) a sharing guardian.
1712 Greedy guardians are the default because they are more "defensive".
1713 You can only greedily guard an object once. If you guard an object
1714 more than once, once in a greedy guardian and the rest of times in
1715 sharing guardians, then it is guaranteed that the object won't be
1716 returned from sharing guardians as long as it is greedily guarded
1719 Guardians returned by calls to `make-guardian' can now take one more
1720 optional parameter, which says whether to throw an error in case an
1721 attempt is made to greedily guard an object that is already greedily
1722 guarded. The default is true, i.e. throw an error. If the parameter
1723 is false, the guardian invocation returns #t if guarding was
1724 successful and #f if it wasn't.
1726 Also, since greedy guarding is, in effect, a side-effecting operation
1727 on objects, a new function is introduced: `destroy-guardian!'.
1728 Invoking this function on a guardian renders it unoperative and, if
1729 the guardian is greedy, clears the "greedily guarded" property of the
1730 objects that were guarded by it, thus undoing the side effect.
1732 Note that all this hair is hardly very important, since guardian
1733 objects are usually permanent.
1735 ** Continuations created by call-with-current-continuation now accept
1736 any number of arguments, as required by R5RS.
1738 ** New function `issue-deprecation-warning'
1740 This function is used to display the deprecation messages that are
1741 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1744 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1748 ;; `id' is deprecated. Use `identity' instead.
1753 ** New syntax `begin-deprecated'
1755 When deprecated features are included (as determined by the configure
1756 option --enable-deprecated), `begin-deprecated' is identical to
1757 `begin'. When deprecated features are excluded, it always evaluates
1758 to `#f', ignoring the body forms.
1760 ** New function `make-object-property'
1762 This function returns a new `procedure with setter' P that can be used
1763 to attach a property to objects. When calling P as
1767 where `obj' is any kind of object, it attaches `val' to `obj' in such
1768 a way that it can be retrieved by calling P as
1772 This function will replace procedure properties, symbol properties and
1773 source properties eventually.
1775 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1777 Instead of #&optional, #&key, etc you should now use #:optional,
1778 #:key, etc. Since #:optional is a keyword, you can write it as just
1779 :optional when (read-set! keywords 'prefix) is active.
1781 The old reader syntax `#&' is still supported, but deprecated. It
1782 will be removed in the next release.
1784 ** New define-module option: pure
1786 Tells the module system not to include any bindings from the root
1791 (define-module (totally-empty-module)
1794 ** New define-module option: export NAME1 ...
1796 Export names NAME1 ...
1798 This option is required if you want to be able to export bindings from
1799 a module which doesn't import one of `define-public' or `export'.
1803 (define-module (foo)
1805 :use-module (ice-9 r5rs)
1808 ;;; Note that we're pure R5RS below this point!
1813 ** New function: object->string OBJ
1815 Return a Scheme string obtained by printing a given object.
1817 ** New function: port? X
1819 Returns a boolean indicating whether X is a port. Equivalent to
1820 `(or (input-port? X) (output-port? X))'.
1822 ** New function: file-port?
1824 Determines whether a given object is a port that is related to a file.
1826 ** New function: port-for-each proc
1828 Apply PROC to each port in the Guile port table in turn. The return
1829 value is unspecified. More specifically, PROC is applied exactly once
1830 to every port that exists in the system at the time PORT-FOR-EACH is
1831 invoked. Changes to the port table while PORT-FOR-EACH is running
1832 have no effect as far as PORT-FOR-EACH is concerned.
1834 ** New function: dup2 oldfd newfd
1836 A simple wrapper for the `dup2' system call. Copies the file
1837 descriptor OLDFD to descriptor number NEWFD, replacing the
1838 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1839 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1840 to move away ports which are using NEWFD. The return value is
1843 ** New function: close-fdes fd
1845 A simple wrapper for the `close' system call. Close file
1846 descriptor FD, which must be an integer. Unlike close (*note
1847 close: Ports and File Descriptors.), the file descriptor will be
1848 closed even if a port is using it. The return value is
1851 ** New function: crypt password salt
1853 Encrypts `password' using the standard unix password encryption
1856 ** New function: chroot path
1858 Change the root directory of the running process to `path'.
1860 ** New functions: getlogin, cuserid
1862 Return the login name or the user name of the current effective user
1865 ** New functions: getpriority which who, setpriority which who prio
1867 Get or set the priority of the running process.
1869 ** New function: getpass prompt
1871 Read a password from the terminal, first displaying `prompt' and
1874 ** New function: flock file operation
1876 Set/remove an advisory shared or exclusive lock on `file'.
1878 ** New functions: sethostname name, gethostname
1880 Set or get the hostname of the machine the current process is running
1883 ** New function: mkstemp! tmpl
1885 mkstemp creates a new unique file in the file system and returns a
1886 new buffered port open for reading and writing to the file. TMPL
1887 is a string specifying where the file should be created: it must
1888 end with `XXXXXX' and will be changed in place to return the name
1889 of the temporary file.
1891 ** New function: open-input-string string
1893 Return an input string port which delivers the characters from
1894 `string'. This procedure, together with `open-output-string' and
1895 `get-output-string' implements SRFI-6.
1897 ** New function: open-output-string
1899 Return an output string port which collects all data written to it.
1900 The data can then be retrieved by `get-output-string'.
1902 ** New function: get-output-string
1904 Return the contents of an output string port.
1906 ** New function: identity
1908 Return the argument.
1910 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1911 are represented in Scheme as integers with normal host byte ordering.
1913 ** New function: inet-pton family address
1915 Convert a printable string network address into an integer. Note that
1916 unlike the C version of this function, the result is an integer with
1917 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1920 (inet-pton AF_INET "127.0.0.1") => 2130706433
1921 (inet-pton AF_INET6 "::1") => 1
1923 ** New function: inet-ntop family address
1925 Convert an integer network address into a printable string. Note that
1926 unlike the C version of this function, the input is an integer with
1927 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1930 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1931 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1932 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1936 Use `identity' instead.
1942 ** Deprecated: return-it
1946 ** Deprecated: string-character-length
1948 Use `string-length' instead.
1950 ** Deprecated: flags
1952 Use `logior' instead.
1954 ** Deprecated: close-all-ports-except.
1956 This was intended for closing ports in a child process after a fork,
1957 but it has the undesirable side effect of flushing buffers.
1958 port-for-each is more flexible.
1960 ** The (ice-9 popen) module now attempts to set up file descriptors in
1961 the child process from the current Scheme ports, instead of using the
1962 current values of file descriptors 0, 1, and 2 in the parent process.
1964 ** Removed function: builtin-weak-bindings
1966 There is no such concept as a weak binding any more.
1968 ** Removed constants: bignum-radix, scm-line-incrementors
1970 ** define-method: New syntax mandatory.
1972 The new method syntax is now mandatory:
1974 (define-method (NAME ARG-SPEC ...) BODY ...)
1975 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1977 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1978 REST-ARG ::= ARG-NAME
1980 If you have old code using the old syntax, import
1981 (oop goops old-define-method) before (oop goops) as in:
1983 (use-modules (oop goops old-define-method) (oop goops))
1985 ** Deprecated function: builtin-variable
1986 Removed function: builtin-bindings
1988 There is no longer a distinction between builtin or other variables.
1989 Use module system operations for all variables.
1991 ** Lazy-catch handlers are no longer allowed to return.
1993 That is, a call to `throw', `error', etc is now guaranteed to not
1996 ** Bugfixes for (ice-9 getopt-long)
1998 This module is now tested using test-suite/tests/getopt-long.test.
1999 The following bugs have been fixed:
2001 *** Parsing for options that are specified to have `optional' args now checks
2002 if the next element is an option instead of unconditionally taking it as the
2005 *** An error is now thrown for `--opt=val' when the option description
2006 does not specify `(value #t)' or `(value optional)'. This condition used to
2007 be accepted w/o error, contrary to the documentation.
2009 *** The error message for unrecognized options is now more informative.
2010 It used to be "not a record", an artifact of the implementation.
2012 *** The error message for `--opt' terminating the arg list (no value), when
2013 `(value #t)' is specified, is now more informative. It used to be "not enough
2016 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2017 The expansion used to be like so:
2019 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2021 Note that the "5d" is dropped. Now it is like so:
2023 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2025 This enables single-char options to have adjoining arguments as long as their
2026 constituent characters are not potential single-char options.
2028 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2030 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2031 property `arglist', which can be retrieved by `arity'. The result is that
2032 `arity' can give more detailed information than before:
2036 guile> (use-modules (ice-9 optargs))
2037 guile> (define* (foo #:optional a b c) a)
2039 0 or more arguments in `lambda*:G0'.
2044 3 optional arguments: `a', `b' and `c'.
2045 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2047 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2048 and `d', other keywords allowed.
2049 guile> (define* (baz a b #:optional c #:rest r) a)
2051 2 required arguments: `a' and `b', 1 optional argument: `c',
2054 * Changes to the C interface
2056 ** Types have been renamed from scm_*_t to scm_t_*.
2058 This has been done for POSIX sake. It reserves identifiers ending
2059 with "_t". What a concept.
2061 The old names are still available with status `deprecated'.
2063 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2065 ** Deprecated features have been removed.
2069 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2070 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2072 *** C Functions removed
2074 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2075 scm_fseek - replaced by scm_seek.
2076 gc-thunk - replaced by after-gc-hook.
2077 gh_int2scmb - replaced by gh_bool2scm.
2078 scm_tc_dblr - replaced by scm_tc16_real.
2079 scm_tc_dblc - replaced by scm_tc16_complex.
2080 scm_list_star - replaced by scm_cons_star.
2082 ** Deprecated: scm_makfromstr
2084 Use scm_mem2string instead.
2086 ** Deprecated: scm_make_shared_substring
2088 Explicit shared substrings will disappear from Guile.
2090 Instead, "normal" strings will be implemented using sharing
2091 internally, combined with a copy-on-write strategy.
2093 ** Deprecated: scm_read_only_string_p
2095 The concept of read-only strings will disappear in next release of
2098 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2100 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2102 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2104 Call a procedure with the indicated number of arguments. See "Fly
2105 Evaluation" in the manual.
2107 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2109 Call a procedure with the indicated number of arguments and a list of
2110 further arguments. See "Fly Evaluation" in the manual.
2112 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2114 Create a list of the given number of elements. See "List
2115 Constructors" in the manual.
2117 ** Renamed function: scm_listify has been replaced by scm_list_n.
2119 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2120 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2122 Use functions scm_list_N instead.
2124 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2126 Used by an application to read arbitrary number of bytes from a port.
2127 Same semantics as libc read, except that scm_c_read only returns less
2128 than SIZE bytes if at end-of-file.
2130 Warning: Doesn't update port line and column counts!
2132 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2134 Used by an application to write arbitrary number of bytes to an SCM
2135 port. Similar semantics as libc write. However, unlike libc
2136 write, scm_c_write writes the requested number of bytes and has no
2139 Warning: Doesn't update port line and column counts!
2141 ** New function: scm_init_guile ()
2143 In contrast to scm_boot_guile, scm_init_guile will return normally
2144 after initializing Guile. It is not available on all systems, tho.
2146 ** New functions: scm_str2symbol, scm_mem2symbol
2148 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2149 field of characters and creates a scheme symbol object from that C string.
2150 The function scm_mem2symbol takes a const char* and a number of characters and
2151 creates a symbol from the characters in that memory area.
2153 ** New functions: scm_primitive_make_property
2154 scm_primitive_property_ref
2155 scm_primitive_property_set_x
2156 scm_primitive_property_del_x
2158 These functions implement a new way to deal with object properties.
2159 See libguile/properties.c for their documentation.
2161 ** New function: scm_done_free (long size)
2163 This function is the inverse of scm_done_malloc. Use it to report the
2164 amount of smob memory you free. The previous method, which involved
2165 calling scm_done_malloc with negative argument, was somewhat
2166 unintuitive (and is still available, of course).
2168 ** New function: scm_c_memq (SCM obj, SCM list)
2170 This function provides a fast C level alternative for scm_memq for the case
2171 that the list parameter is known to be a proper list. The function is a
2172 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2173 list input parameter, since for anything else but a proper list the function's
2174 behaviour is undefined - it may even crash or loop endlessly. Further, for
2175 the case that the object is not found in the list, scm_c_memq returns #f which
2176 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2178 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2179 scm_remember_upto_here
2181 These functions replace the function scm_remember.
2183 ** Deprecated function: scm_remember
2185 Use one of the new functions scm_remember_upto_here_1,
2186 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2188 ** New function: scm_allocate_string
2190 This function replaces the function scm_makstr.
2192 ** Deprecated function: scm_makstr
2194 Use the new function scm_allocate_string instead.
2196 ** New global variable scm_gc_running_p introduced.
2198 Use this variable to find out if garbage collection is being executed. Up to
2199 now applications have used scm_gc_heap_lock to test if garbage collection was
2200 running, which also works because of the fact that up to know only the garbage
2201 collector has set this variable. But, this is an implementation detail that
2202 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2203 of this variable is (and has been) not fully safe anyway.
2205 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2207 Use these instead of SCM_LENGTH_MAX.
2209 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2210 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2211 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2213 Use these instead of SCM_LENGTH.
2215 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2216 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2217 SCM_SET_BITVECTOR_LENGTH
2219 Use these instead of SCM_SETLENGTH
2221 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2222 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2225 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2228 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2229 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2232 Use these instead of SCM_SETCHARS.
2234 ** New macro: SCM_BITVECTOR_P
2236 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2238 Use instead of SCM_COERCE_SUBSTR.
2240 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2242 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2244 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2245 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2246 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2247 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2248 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2249 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2250 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2251 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2252 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2253 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2254 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2255 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2256 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2257 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2258 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2260 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2261 Use scm_memory_error instead of SCM_NALLOC.
2262 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2263 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2264 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2265 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2266 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2267 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2268 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2269 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2270 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2271 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2272 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2273 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2274 Use a type specific setter macro instead of SCM_SETLENGTH.
2275 Use a type specific setter macro instead of SCM_SETCHARS.
2276 Use a type specific length macro instead of SCM_LENGTH_MAX.
2277 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2278 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2279 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2280 Use SCM_TYP16 instead of SCM_GCTYP16.
2281 Use SCM_CDR instead of SCM_GCCDR.
2282 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2283 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2284 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2285 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2286 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2287 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2288 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2290 ** Removed function: scm_struct_init
2292 ** Removed variable: scm_symhash_dim
2294 ** Renamed function: scm_make_cont has been replaced by
2295 scm_make_continuation, which has a different interface.
2297 ** Deprecated function: scm_call_catching_errors
2299 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2301 ** Deprecated function: scm_strhash
2303 Use scm_string_hash instead.
2305 ** Deprecated function: scm_vector_set_length_x
2307 Instead, create a fresh vector of the desired size and copy the contents.
2309 ** scm_gensym has changed prototype
2311 scm_gensym now only takes one argument.
2313 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2316 There is now only a single symbol type scm_tc7_symbol.
2317 The tag scm_tc7_lvector was not used anyway.
2319 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2321 Use scm_make_smob_type and scm_set_smob_XXX instead.
2323 ** New function scm_set_smob_apply.
2325 This can be used to set an apply function to a smob type.
2327 ** Deprecated function: scm_strprint_obj
2329 Use scm_object_to_string instead.
2331 ** Deprecated function: scm_wta
2333 Use scm_wrong_type_arg, or another appropriate error signalling function
2336 ** Explicit support for obarrays has been deprecated.
2338 Use `scm_str2symbol' and the generic hashtable functions instead.
2340 ** The concept of `vcells' has been deprecated.
2342 The data type `variable' is now used exclusively. `Vcells' have been
2343 a low-level concept so you are likely not affected by this change.
2345 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2346 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2348 Use scm_c_define or scm_c_lookup instead, as appropriate.
2350 *** New functions: scm_c_module_lookup, scm_c_lookup,
2351 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2352 scm_module_define, scm_define.
2354 These functions work with variables instead of with vcells.
2356 ** New functions for creating and defining `subr's and `gsubr's.
2358 The new functions more clearly distinguish between creating a subr (or
2359 gsubr) object and adding it to the current module.
2361 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2362 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2363 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2364 scm_c_define_gsubr_with_generic.
2366 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2367 scm_make_subr_with_generic, scm_make_gsubr,
2368 scm_make_gsubr_with_generic.
2370 Use the new ones from above instead.
2372 ** C interface to the module system has changed.
2374 While we suggest that you avoid as many explicit module system
2375 operations from C as possible for the time being, the C interface has
2376 been made more similar to the high-level Scheme module system.
2378 *** New functions: scm_c_define_module, scm_c_use_module,
2379 scm_c_export, scm_c_resolve_module.
2381 They mostly work like their Scheme namesakes. scm_c_define_module
2382 takes a function that is called a context where the new module is
2385 *** Deprecated functions: scm_the_root_module, scm_make_module,
2386 scm_ensure_user_module, scm_load_scheme_module.
2388 Use the new functions instead.
2390 ** Renamed function: scm_internal_with_fluids becomes
2393 scm_internal_with_fluids is available as a deprecated function.
2395 ** New function: scm_c_with_fluid.
2397 Just like scm_c_with_fluids, but takes one fluid and one value instead
2400 ** Deprecated typedefs: long_long, ulong_long.
2402 They are of questionable utility and they pollute the global
2405 ** Deprecated typedef: scm_sizet
2407 It is of questionable utility now that Guile requires ANSI C, and is
2410 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2411 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2412 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2414 Made more compliant with the naming policy by adding a _t at the end.
2416 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2417 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2419 With the exception of the mysterious scm_2ulong2big, they are still
2420 available under new names (scm_i_mkbig etc). These functions are not
2421 intended to be used in user code. You should avoid dealing with
2422 bignums directly, and should deal with numbers in general (which can
2425 ** Change in behavior: scm_num2long, scm_num2ulong
2427 The scm_num2[u]long functions don't any longer accept an inexact
2428 argument. This change in behavior is motivated by concordance with
2429 R5RS: It is more common that a primitive doesn't want to accept an
2430 inexact for an exact.
2432 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2433 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2434 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2437 These are conversion functions between the various ANSI C integral
2438 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2439 accept an inexact argument.
2441 ** New functions: scm_float2num, scm_double2num,
2442 scm_num2float, scm_num2double.
2444 These are conversion functions between the two ANSI C float types and
2447 ** New number validation macros:
2448 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2452 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2454 These are just nicer-named old scm_protect_object and
2455 scm_unprotect_object.
2457 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2459 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2461 These functions can be used to register pointers to locations that
2464 ** Deprecated function: scm_create_hook.
2466 Its sins are: misleading name, non-modularity and lack of general
2470 Changes since Guile 1.3.4:
2472 * Changes to the distribution
2474 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2476 We've changed the way we handle generated files in the Guile source
2477 repository. As a result, the procedure for building trees obtained
2478 from the nightly FTP snapshots or via CVS has changed:
2479 - You must have appropriate versions of autoconf, automake, and
2480 libtool installed on your system. See README for info on how to
2481 obtain these programs.
2482 - Before configuring the tree, you must first run the script
2483 `autogen.sh' at the top of the source tree.
2485 The Guile repository used to contain not only source files, written by
2486 humans, but also some generated files, like configure scripts and
2487 Makefile.in files. Even though the contents of these files could be
2488 derived mechanically from other files present, we thought it would
2489 make the tree easier to build if we checked them into CVS.
2491 However, this approach means that minor differences between
2492 developer's installed tools and habits affected the whole team.
2493 So we have removed the generated files from the repository, and
2494 added the autogen.sh script, which will reconstruct them
2498 ** configure now has experimental options to remove support for certain
2501 --disable-arrays omit array and uniform array support
2502 --disable-posix omit posix interfaces
2503 --disable-networking omit networking interfaces
2504 --disable-regex omit regular expression interfaces
2506 These are likely to become separate modules some day.
2508 ** New configure option --enable-debug-freelist
2510 This enables a debugging version of SCM_NEWCELL(), and also registers
2511 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2513 Configure with the --enable-debug-freelist option to enable
2514 the gc-set-debug-check-freelist! primitive, and then use:
2516 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2517 (gc-set-debug-check-freelist! #f) # turn off checking
2519 Checking of the freelist forces a traversal of the freelist and
2520 a garbage collection before each allocation of a cell. This can
2521 slow down the interpreter dramatically, so the setter should be used to
2522 turn on this extra processing only when necessary.
2524 ** New configure option --enable-debug-malloc
2526 Include code for debugging of calls to scm_must_malloc/realloc/free.
2530 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2531 2. objects reallocated by scm_must_realloc has been allocated by
2533 3. reallocated objects are reallocated with the same what string
2535 But, most importantly, it records the number of allocated objects of
2536 each kind. This is useful when searching for memory leaks.
2538 A Guile compiled with this option provides the primitive
2539 `malloc-stats' which returns an alist with pairs of kind and the
2540 number of objects of that kind.
2542 ** All includes are now referenced relative to the root directory
2544 Since some users have had problems with mixups between Guile and
2545 system headers, we have decided to always refer to Guile headers via
2546 their parent directories. This essentially creates a "private name
2547 space" for Guile headers. This means that the compiler only is given
2548 -I options for the root build and root source directory.
2550 ** Header files kw.h and genio.h have been removed.
2552 ** The module (ice-9 getopt-gnu-style) has been removed.
2554 ** New module (ice-9 documentation)
2556 Implements the interface to documentation strings associated with
2559 ** New module (ice-9 time)
2561 Provides a macro `time', which displays execution time of a given form.
2563 ** New module (ice-9 history)
2565 Loading this module enables value history in the repl.
2567 * Changes to the stand-alone interpreter
2569 ** New command line option --debug
2571 Start Guile with debugging evaluator and backtraces enabled.
2573 This is useful when debugging your .guile init file or scripts.
2575 ** New help facility
2577 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2578 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2579 (help 'NAME) gives documentation for NAME, even if it is not an object
2580 (help ,EXPR) gives documentation for object returned by EXPR
2581 (help (my module)) gives module commentary for `(my module)'
2582 (help) gives this text
2584 `help' searches among bindings exported from loaded modules, while
2585 `apropos' searches among bindings visible from the "current" module.
2587 Examples: (help help)
2589 (help "output-string")
2591 ** `help' and `apropos' now prints full module names
2593 ** Dynamic linking now uses libltdl from the libtool package.
2595 The old system dependent code for doing dynamic linking has been
2596 replaced with calls to the libltdl functions which do all the hairy
2599 The major improvement is that you can now directly pass libtool
2600 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2601 will be able to do the best shared library job you can get, via
2604 The way dynamic libraries are found has changed and is not really
2605 portable across platforms, probably. It is therefore recommended to
2606 use absolute filenames when possible.
2608 If you pass a filename without an extension to `dynamic-link', it will
2609 try a few appropriate ones. Thus, the most platform ignorant way is
2610 to specify a name like "libfoo", without any directories and
2613 ** Guile COOP threads are now compatible with LinuxThreads
2615 Previously, COOP threading wasn't possible in applications linked with
2616 Linux POSIX threads due to their use of the stack pointer to find the
2617 thread context. This has now been fixed with a workaround which uses
2618 the pthreads to allocate the stack.
2620 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2622 ** Positions of erring expression in scripts
2624 With version 1.3.4, the location of the erring expression in Guile
2625 scipts is no longer automatically reported. (This should have been
2626 documented before the 1.3.4 release.)
2628 You can get this information by enabling recording of positions of
2629 source expressions and running the debugging evaluator. Put this at
2630 the top of your script (or in your "site" file):
2632 (read-enable 'positions)
2633 (debug-enable 'debug)
2635 ** Backtraces in scripts
2637 It is now possible to get backtraces in scripts.
2641 (debug-enable 'debug 'backtrace)
2643 at the top of the script.
2645 (The first options enables the debugging evaluator.
2646 The second enables backtraces.)
2648 ** Part of module system symbol lookup now implemented in C
2650 The eval closure of most modules is now implemented in C. Since this
2651 was one of the bottlenecks for loading speed, Guile now loads code
2652 substantially faster than before.
2654 ** Attempting to get the value of an unbound variable now produces
2655 an exception with a key of 'unbound-variable instead of 'misc-error.
2657 ** The initial default output port is now unbuffered if it's using a
2658 tty device. Previously in this situation it was line-buffered.
2660 ** New hook: after-gc-hook
2662 after-gc-hook takes over the role of gc-thunk. This hook is run at
2663 the first SCM_TICK after a GC. (Thus, the code is run at the same
2664 point during evaluation as signal handlers.)
2666 Note that this hook should be used only for diagnostic and debugging
2667 purposes. It is not certain that it will continue to be well-defined
2668 when this hook is run in the future.
2670 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2671 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2673 ** Improvements to garbage collector
2675 Guile 1.4 has a new policy for triggering heap allocation and
2676 determining the sizes of heap segments. It fixes a number of problems
2679 1. The new policy can handle two separate pools of cells
2680 (2-word/4-word) better. (The old policy would run wild, allocating
2681 more and more memory for certain programs.)
2683 2. The old code would sometimes allocate far too much heap so that the
2684 Guile process became gigantic. The new code avoids this.
2686 3. The old code would sometimes allocate too little so that few cells
2687 were freed at GC so that, in turn, too much time was spent in GC.
2689 4. The old code would often trigger heap allocation several times in a
2690 row. (The new scheme predicts how large the segments needs to be
2691 in order not to need further allocation.)
2693 All in all, the new GC policy will make larger applications more
2696 The new GC scheme also is prepared for POSIX threading. Threads can
2697 allocate private pools of cells ("clusters") with just a single
2698 function call. Allocation of single cells from such a cluster can
2699 then proceed without any need of inter-thread synchronization.
2701 ** New environment variables controlling GC parameters
2703 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2706 Allocation of 2-word cell heaps:
2708 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2711 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2712 GC in percent of total heap size
2715 Allocation of 4-word cell heaps
2716 (used for real numbers and misc other objects):
2718 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2720 (See entry "Way for application to customize GC parameters" under
2721 section "Changes to the scm_ interface" below.)
2723 ** Guile now implements reals using 4-word cells
2725 This speeds up computation with reals. (They were earlier allocated
2726 with `malloc'.) There is still some room for optimizations, however.
2728 ** Some further steps toward POSIX thread support have been taken
2730 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2731 don't have much effect any longer, and many of them will be removed in
2735 are only handled at the top of the evaluator loop, immediately after
2736 I/O, and in scm_equalp.
2738 *** The GC can allocate thread private pools of pairs.
2740 * Changes to Scheme functions and syntax
2742 ** close-input-port and close-output-port are now R5RS
2744 These procedures have been turned into primitives and have R5RS behaviour.
2746 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2748 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2749 extended by the more sophisticated version in (ice-9 format)
2751 (simple-format port message . args)
2752 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2753 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2754 the escapes are replaced with corresponding members of ARGS:
2755 ~A formats using `display' and ~S formats using `write'.
2756 If DESTINATION is #t, then use the `current-output-port',
2757 if DESTINATION is #f, then return a string containing the formatted text.
2758 Does not add a trailing newline."
2760 ** string-ref: the second argument is no longer optional.
2762 ** string, list->string: no longer accept strings in their arguments,
2763 only characters, for compatibility with R5RS.
2765 ** New procedure: port-closed? PORT
2766 Returns #t if PORT is closed or #f if it is open.
2768 ** Deprecated: list*
2770 The list* functionality is now provided by cons* (SRFI-1 compliant)
2772 ** New procedure: cons* ARG1 ARG2 ... ARGn
2774 Like `list', but the last arg provides the tail of the constructed list,
2775 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2777 Requires at least one argument. If given one argument, that argument
2778 is returned as result.
2780 This function is called `list*' in some other Schemes and in Common LISP.
2782 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2784 ** New procedure: object-documentation OBJECT
2786 Returns the documentation string associated with OBJECT. The
2787 procedure uses a caching mechanism so that subsequent lookups are
2790 Exported by (ice-9 documentation).
2792 ** module-name now returns full names of modules
2794 Previously, only the last part of the name was returned (`session' for
2795 `(ice-9 session)'). Ex: `(ice-9 session)'.
2797 * Changes to the gh_ interface
2799 ** Deprecated: gh_int2scmb
2801 Use gh_bool2scm instead.
2803 * Changes to the scm_ interface
2805 ** Guile primitives now carry docstrings!
2807 Thanks to Greg Badros!
2809 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2811 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2812 macros and must contain a docstring that is extracted into foo.doc using a new
2813 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2815 However, a major overhaul of these macros is scheduled for the next release of
2818 ** Guile primitives use a new technique for validation of arguments
2820 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2821 the readability of argument checking.
2823 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2825 ** New macros: SCM_PACK, SCM_UNPACK
2827 Compose/decompose an SCM value.
2829 The SCM type is now treated as an abstract data type and may be defined as a
2830 long, a void* or as a struct, depending on the architecture and compile time
2831 options. This makes it easier to find several types of bugs, for example when
2832 SCM values are treated as integers without conversion. Values of the SCM type
2833 should be treated as "atomic" values. These macros are used when
2834 composing/decomposing an SCM value, either because you want to access
2835 individual bits, or because you want to treat it as an integer value.
2837 E.g., in order to set bit 7 in an SCM value x, use the expression
2839 SCM_PACK (SCM_UNPACK (x) | 0x80)
2841 ** The name property of hooks is deprecated.
2842 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2844 You can emulate this feature by using object properties.
2846 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2847 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2850 These macros will be removed in a future release of Guile.
2852 ** The following types, functions and macros from numbers.h are deprecated:
2853 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2854 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2856 ** Port internals: the rw_random variable in the scm_port structure
2857 must be set to non-zero in any random access port. In recent Guile
2858 releases it was only set for bidirectional random-access ports.
2860 ** Port internals: the seek ptob procedure is now responsible for
2861 resetting the buffers if required. The change was made so that in the
2862 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2863 the fport and strport ptobs can avoid resetting the buffers,
2864 in particular to avoid discarding unread chars. An existing port
2865 type can be fixed by adding something like the following to the
2866 beginning of the ptob seek procedure:
2868 if (pt->rw_active == SCM_PORT_READ)
2869 scm_end_input (object);
2870 else if (pt->rw_active == SCM_PORT_WRITE)
2871 ptob->flush (object);
2873 although to actually avoid resetting the buffers and discard unread
2874 chars requires further hacking that depends on the characteristics
2877 ** Deprecated functions: scm_fseek, scm_tag
2879 These functions are no longer used and will be removed in a future version.
2881 ** The scm_sysmissing procedure is no longer used in libguile.
2882 Unless it turns out to be unexpectedly useful to somebody, it will be
2883 removed in a future version.
2885 ** The format of error message strings has changed
2887 The two C procedures: scm_display_error and scm_error, as well as the
2888 primitive `scm-error', now use scm_simple_format to do their work.
2889 This means that the message strings of all code must be updated to use
2890 ~A where %s was used before, and ~S where %S was used before.
2892 During the period when there still are a lot of old Guiles out there,
2893 you might want to support both old and new versions of Guile.
2895 There are basically two methods to achieve this. Both methods use
2898 AC_CHECK_FUNCS(scm_simple_format)
2900 in your configure.in.
2902 Method 1: Use the string concatenation features of ANSI C's
2907 #ifdef HAVE_SCM_SIMPLE_FORMAT
2913 Then represent each of your error messages using a preprocessor macro:
2915 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2919 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2920 (define make-message string-append)
2922 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2924 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2928 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2933 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2937 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2939 Don't use the functions coop_mutex_init and
2940 coop_condition_variable_init. They will change.
2942 Use scm_mutex_init and scm_cond_init instead.
2944 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2945 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2946 COND, as `scm_cond_wait' does, but it also bounds the duration
2947 of the wait. If COND has not been signaled before time ABSTIME,
2948 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2949 returns the error code `ETIMEDOUT'.
2951 The ABSTIME parameter specifies an absolute time, with the same
2952 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2953 to 00:00:00 GMT, January 1, 1970.
2955 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2956 `scm_cond_broadcast' restarts all the threads that are waiting
2957 on the condition variable COND. Nothing happens if no threads are
2960 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2961 `scm_key_create' allocates a new TSD key. The key is stored in
2962 the location pointed to by KEY. There is no limit on the number
2963 of keys allocated at a given time. The value initially associated
2964 with the returned key is `NULL' in all currently executing threads.
2966 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2967 function associated with the key. When a thread terminates,
2968 DESTR_FUNCTION is called on the value associated with the key in
2969 that thread. The DESTR_FUNCTION is not called if a key is deleted
2970 with `scm_key_delete' or a value is changed with
2971 `scm_setspecific'. The order in which destructor functions are
2972 called at thread termination time is unspecified.
2974 Destructors are not yet implemented.
2976 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2977 `scm_setspecific' changes the value associated with KEY in the
2978 calling thread, storing the given POINTER instead.
2980 ** New function: scm_getspecific (scm_key_t KEY)
2981 `scm_getspecific' returns the value currently associated with
2982 KEY in the calling thread.
2984 ** New function: scm_key_delete (scm_key_t KEY)
2985 `scm_key_delete' deallocates a TSD key. It does not check
2986 whether non-`NULL' values are associated with that key in the
2987 currently executing threads, nor call the destructor function
2988 associated with the key.
2990 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2992 Initialize a C level hook HOOK with associated HOOK_DATA and type
2993 TYPE. (See scm_c_hook_run ().)
2995 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2997 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2998 is true, add it last, otherwise first. The same FUNC can be added
2999 multiple times if FUNC_DATA differ and vice versa.
3001 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3003 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3004 function is only removed if both FUNC and FUNC_DATA matches.
3006 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3008 Run hook HOOK passing DATA to the hook functions.
3010 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3011 returned is undefined.
3013 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3014 returns a non-NULL value. This value is returned as the result of
3015 scm_c_hook_run. If all functions return NULL, NULL is returned.
3017 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3018 returns a NULL value, and NULL is returned. If all functions returns
3019 a non-NULL value, the last value is returned.
3021 ** New C level GC hooks
3023 Five new C level hooks has been added to the garbage collector.
3025 scm_before_gc_c_hook
3028 are run before locking and after unlocking the heap. The system is
3029 thus in a mode where evaluation can take place. (Except that
3030 scm_before_gc_c_hook must not allocate new cells.)
3032 scm_before_mark_c_hook
3033 scm_before_sweep_c_hook
3034 scm_after_sweep_c_hook
3036 are run when the heap is locked. These are intended for extension of
3037 the GC in a modular fashion. Examples are the weaks and guardians
3040 ** Way for application to customize GC parameters
3042 The application can set up other default values for the GC heap
3043 allocation parameters
3045 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3046 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3047 GUILE_MAX_SEGMENT_SIZE,
3051 scm_default_init_heap_size_1, scm_default_min_yield_1,
3052 scm_default_init_heap_size_2, scm_default_min_yield_2,
3053 scm_default_max_segment_size
3055 respectively before callong scm_boot_guile.
3057 (See entry "New environment variables ..." in section
3058 "Changes to the stand-alone interpreter" above.)
3060 ** scm_protect_object/scm_unprotect_object now nest
3062 This means that you can call scm_protect_object multiple times on an
3063 object and count on the object being protected until
3064 scm_unprotect_object has been call the same number of times.
3066 The functions also have better time complexity.
3068 Still, it is usually possible to structure the application in a way
3069 that you don't need to use these functions. For example, if you use a
3070 protected standard Guile list to keep track of live objects rather
3071 than some custom data type, objects will die a natural death when they
3072 are no longer needed.
3074 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3076 Guile does not provide the float representation for inexact real numbers any
3077 more. Now, only doubles are used to represent inexact real numbers. Further,
3078 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3079 and scm_tc16_complex, respectively.
3081 ** Removed deprecated type scm_smobfuns
3083 ** Removed deprecated function scm_newsmob
3085 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3087 There is an ongoing discussion among the developers whether to
3088 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3089 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3090 until this issue has been settled.
3092 ** Removed deprecated type tag scm_tc16_kw
3094 ** Added type tag scm_tc16_keyword
3096 (This was introduced already in release 1.3.4 but was not documented
3099 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3101 * Changes to system call interfaces:
3103 ** The "select" procedure now tests port buffers for the ability to
3104 provide input or accept output. Previously only the underlying file
3105 descriptors were checked.
3107 ** New variable PIPE_BUF: the maximum number of bytes that can be
3108 atomically written to a pipe.
3110 ** If a facility is not available on the system when Guile is
3111 compiled, the corresponding primitive procedure will not be defined.
3112 Previously it would have been defined but would throw a system-error
3113 exception if called. Exception handlers which catch this case may
3114 need minor modification: an error will be thrown with key
3115 'unbound-variable instead of 'system-error. Alternatively it's
3116 now possible to use `defined?' to check whether the facility is
3119 ** Procedures which depend on the timezone should now give the correct
3120 result on systems which cache the TZ environment variable, even if TZ
3121 is changed without calling tzset.
3123 * Changes to the networking interfaces:
3125 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3126 long integers between network and host format. For now, it's not
3127 particularly convenient to do this kind of thing, but consider:
3129 (define write-network-long
3130 (lambda (value port)
3131 (let ((v (make-uniform-vector 1 1 0)))
3132 (uniform-vector-set! v 0 (htonl value))
3133 (uniform-vector-write v port))))
3135 (define read-network-long
3137 (let ((v (make-uniform-vector 1 1 0)))
3138 (uniform-vector-read! v port)
3139 (ntohl (uniform-vector-ref v 0)))))
3141 ** If inet-aton fails, it now throws an error with key 'misc-error
3142 instead of 'system-error, since errno is not relevant.
3144 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3145 specific keys instead of 'system-error. The latter is inappropriate
3146 since errno will not have been set. The keys are:
3147 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3149 ** sethostent, setnetent, setprotoent, setservent: now take an
3150 optional argument STAYOPEN, which specifies whether the database
3151 remains open after a database entry is accessed randomly (e.g., using
3152 gethostbyname for the hosts database.) The default is #f. Previously
3156 Changes since Guile 1.3.2:
3158 * Changes to the stand-alone interpreter
3162 An initial version of the Guile debugger written by Chris Hanson has
3163 been added. The debugger is still under development but is included
3164 in the distribution anyway since it is already quite useful.
3170 after an error to enter the debugger. Type `help' inside the debugger
3171 for a description of available commands.
3173 If you prefer to have stack frames numbered and printed in
3174 anti-chronological order and prefer up in the stack to be down on the
3175 screen as is the case in gdb, you can put
3177 (debug-enable 'backwards)
3179 in your .guile startup file. (However, this means that Guile can't
3180 use indentation to indicate stack level.)
3182 The debugger is autoloaded into Guile at the first use.
3184 ** Further enhancements to backtraces
3186 There is a new debug option `width' which controls the maximum width
3187 on the screen of printed stack frames. Fancy printing parameters
3188 ("level" and "length" as in Common LISP) are adaptively adjusted for
3189 each stack frame to give maximum information while still fitting
3190 within the bounds. If the stack frame can't be made to fit by
3191 adjusting parameters, it is simply cut off at the end. This is marked
3194 ** Some modules are now only loaded when the repl is started
3196 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3197 regex) are now loaded into (guile-user) only if the repl has been
3198 started. The effect is that the startup time for scripts has been
3199 reduced to 30% of what it was previously.
3201 Correctly written scripts load the modules they require at the top of
3202 the file and should not be affected by this change.
3204 ** Hooks are now represented as smobs
3206 * Changes to Scheme functions and syntax
3208 ** Readline support has changed again.
3210 The old (readline-activator) module is gone. Use (ice-9 readline)
3211 instead, which now contains all readline functionality. So the code
3212 to activate readline is now
3214 (use-modules (ice-9 readline))
3217 This should work at any time, including from the guile prompt.
3219 To avoid confusion about the terms of Guile's license, please only
3220 enable readline for your personal use; please don't make it the
3221 default for others. Here is why we make this rather odd-sounding
3224 Guile is normally licensed under a weakened form of the GNU General
3225 Public License, which allows you to link code with Guile without
3226 placing that code under the GPL. This exception is important to some
3229 However, since readline is distributed under the GNU General Public
3230 License, when you link Guile with readline, either statically or
3231 dynamically, you effectively change Guile's license to the strict GPL.
3232 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3233 which are normally permitted become forbidden. This is a rather
3234 non-obvious consequence of the licensing terms.
3236 So, to make sure things remain clear, please let people choose for
3237 themselves whether to link GPL'd libraries like readline with Guile.
3239 ** regexp-substitute/global has changed slightly, but incompatibly.
3241 If you include a function in the item list, the string of the match
3242 object it receives is the same string passed to
3243 regexp-substitute/global, not some suffix of that string.
3244 Correspondingly, the match's positions are relative to the entire
3245 string, not the suffix.
3247 If the regexp can match the empty string, the way matches are chosen
3248 from the string has changed. regexp-substitute/global recognizes the
3249 same set of matches that list-matches does; see below.
3251 ** New function: list-matches REGEXP STRING [FLAGS]
3253 Return a list of match objects, one for every non-overlapping, maximal
3254 match of REGEXP in STRING. The matches appear in left-to-right order.
3255 list-matches only reports matches of the empty string if there are no
3256 other matches which begin on, end at, or include the empty match's
3259 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3261 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3263 For each match of REGEXP in STRING, apply PROC to the match object,
3264 and the last value PROC returned, or INIT for the first call. Return
3265 the last value returned by PROC. We apply PROC to the matches as they
3266 appear from left to right.
3268 This function recognizes matches according to the same criteria as
3271 Thus, you could define list-matches like this:
3273 (define (list-matches regexp string . flags)
3274 (reverse! (apply fold-matches regexp string '() cons flags)))
3276 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3280 *** New function: hook? OBJ
3282 Return #t if OBJ is a hook, otherwise #f.
3284 *** New function: make-hook-with-name NAME [ARITY]
3286 Return a hook with name NAME and arity ARITY. The default value for
3287 ARITY is 0. The only effect of NAME is that it will appear when the
3288 hook object is printed to ease debugging.
3290 *** New function: hook-empty? HOOK
3292 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3294 *** New function: hook->list HOOK
3296 Return a list of the procedures that are called when run-hook is
3299 ** `map' signals an error if its argument lists are not all the same length.
3301 This is the behavior required by R5RS, so this change is really a bug
3302 fix. But it seems to affect a lot of people's code, so we're
3303 mentioning it here anyway.
3305 ** Print-state handling has been made more transparent
3307 Under certain circumstances, ports are represented as a port with an
3308 associated print state. Earlier, this pair was represented as a pair
3309 (see "Some magic has been added to the printer" below). It is now
3310 indistinguishable (almost; see `get-print-state') from a port on the
3313 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3315 Return a new port with the associated print state PRINT-STATE.
3317 *** New function: get-print-state OUTPUT-PORT
3319 Return the print state associated with this port if it exists,
3320 otherwise return #f.
3322 *** New function: directory-stream? OBJECT
3324 Returns true iff OBJECT is a directory stream --- the sort of object
3325 returned by `opendir'.
3327 ** New function: using-readline?
3329 Return #t if readline is in use in the current repl.
3331 ** structs will be removed in 1.4
3333 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3334 and use GOOPS objects as the fundamental record type.
3336 * Changes to the scm_ interface
3338 ** structs will be removed in 1.4
3340 The entire current struct interface (struct.c, struct.h) will be
3341 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3342 GOOPS objects as the fundamental record type.
3344 ** The internal representation of subr's has changed
3346 Instead of giving a hint to the subr name, the CAR field of the subr
3347 now contains an index to a subr entry in scm_subr_table.
3349 *** New variable: scm_subr_table
3351 An array of subr entries. A subr entry contains the name, properties
3352 and documentation associated with the subr. The properties and
3353 documentation slots are not yet used.
3355 ** A new scheme for "forwarding" calls to a builtin to a generic function
3357 It is now possible to extend the functionality of some Guile
3358 primitives by letting them defer a call to a GOOPS generic function on
3359 argument mismatch. This means that there is no loss of efficiency in
3364 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3365 (define-method + ((x <string>) (y <string>))
3366 (string-append x y))
3368 + will still be as efficient as usual in numerical calculations, but
3369 can also be used for concatenating strings.
3371 Who will be the first one to extend Guile's numerical tower to
3372 rationals? :) [OK, there a few other things to fix before this can
3373 be made in a clean way.]
3375 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3377 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3379 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3381 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3382 a variable GENERIC which can be used by the dispatch macros below.
3384 [This is experimental code which may change soon.]
3386 *** New macros for forwarding control to a generic on arg type error
3388 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3390 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3392 These correspond to the scm_wta function call, and have the same
3393 behaviour until the user has called the GOOPS primitive
3394 `enable-primitive-generic!'. After that, these macros will apply the
3395 generic function GENERIC to the argument(s) instead of calling
3398 [This is experimental code which may change soon.]
3400 *** New macros for argument testing with generic dispatch
3402 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3404 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3406 These correspond to the SCM_ASSERT macro, but will defer control to
3407 GENERIC on error after `enable-primitive-generic!' has been called.
3409 [This is experimental code which may change soon.]
3411 ** New function: SCM scm_eval_body (SCM body, SCM env)
3413 Evaluates the body of a special form.
3415 ** The internal representation of struct's has changed
3417 Previously, four slots were allocated for the procedure(s) of entities
3418 and operators. The motivation for this representation had to do with
3419 the structure of the evaluator, the wish to support tail-recursive
3420 generic functions, and efficiency. Since the generic function
3421 dispatch mechanism has changed, there is no longer a need for such an
3422 expensive representation, and the representation has been simplified.
3424 This should not make any difference for most users.
3426 ** GOOPS support has been cleaned up.
3428 Some code has been moved from eval.c to objects.c and code in both of
3429 these compilation units has been cleaned up and better structured.
3431 *** New functions for applying generic functions
3433 New function: SCM scm_apply_generic (GENERIC, ARGS)
3434 New function: SCM scm_call_generic_0 (GENERIC)
3435 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3436 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3437 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3439 ** Deprecated function: scm_make_named_hook
3441 It is now replaced by:
3443 ** New function: SCM scm_create_hook (const char *name, int arity)
3445 Creates a hook in the same way as make-hook above but also
3446 binds a variable named NAME to it.
3448 This is the typical way of creating a hook from C code.
3450 Currently, the variable is created in the "current" module.
3451 This might change when we get the new module system.
3453 [The behaviour is identical to scm_make_named_hook.]
3457 Changes since Guile 1.3:
3459 * Changes to mailing lists
3461 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3463 See the README file to find current addresses for all the Guile
3466 * Changes to the distribution
3468 ** Readline support is no longer included with Guile by default.
3470 Based on the different license terms of Guile and Readline, we
3471 concluded that Guile should not *by default* cause the linking of
3472 Readline into an application program. Readline support is now offered
3473 as a separate module, which is linked into an application only when
3474 you explicitly specify it.
3476 Although Guile is GNU software, its distribution terms add a special
3477 exception to the usual GNU General Public License (GPL). Guile's
3478 license includes a clause that allows you to link Guile with non-free
3479 programs. We add this exception so as not to put Guile at a
3480 disadvantage vis-a-vis other extensibility packages that support other
3483 In contrast, the GNU Readline library is distributed under the GNU
3484 General Public License pure and simple. This means that you may not
3485 link Readline, even dynamically, into an application unless it is
3486 distributed under a free software license that is compatible the GPL.
3488 Because of this difference in distribution terms, an application that
3489 can use Guile may not be able to use Readline. Now users will be
3490 explicitly offered two independent decisions about the use of these
3493 You can activate the readline support by issuing
3495 (use-modules (readline-activator))
3498 from your ".guile" file, for example.
3500 * Changes to the stand-alone interpreter
3502 ** All builtins now print as primitives.
3503 Previously builtin procedures not belonging to the fundamental subr
3504 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3505 Now, they print as #<primitive-procedure NAME>.
3507 ** Backtraces slightly more intelligible.
3508 gsubr-apply and macro transformer application frames no longer appear
3511 * Changes to Scheme functions and syntax
3513 ** Guile now correctly handles internal defines by rewriting them into
3514 their equivalent letrec. Previously, internal defines would
3515 incrementally add to the innermost environment, without checking
3516 whether the restrictions specified in RnRS were met. This lead to the
3517 correct behaviour when these restriction actually were met, but didn't
3518 catch all illegal uses. Such an illegal use could lead to crashes of
3519 the Guile interpreter or or other unwanted results. An example of
3520 incorrect internal defines that made Guile behave erratically:
3532 The problem with this example is that the definition of `c' uses the
3533 value of `b' directly. This confuses the meoization machine of Guile
3534 so that the second call of `b' (this time in a larger environment that
3535 also contains bindings for `c' and `d') refers to the binding of `c'
3536 instead of `a'. You could also make Guile crash with a variation on
3541 (define (b flag) (if flag a 1))
3542 (define c (1+ (b flag)))
3550 From now on, Guile will issue an `Unbound variable: b' error message
3555 A hook contains a list of functions which should be called on
3556 particular occasions in an existing program. Hooks are used for
3559 A window manager might have a hook before-window-map-hook. The window
3560 manager uses the function run-hooks to call all functions stored in
3561 before-window-map-hook each time a window is mapped. The user can
3562 store functions in the hook using add-hook!.
3564 In Guile, hooks are first class objects.
3566 *** New function: make-hook [N_ARGS]
3568 Return a hook for hook functions which can take N_ARGS arguments.
3569 The default value for N_ARGS is 0.
3571 (See also scm_make_named_hook below.)
3573 *** New function: add-hook! HOOK PROC [APPEND_P]
3575 Put PROC at the beginning of the list of functions stored in HOOK.
3576 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3578 PROC must be able to take the number of arguments specified when the
3581 If PROC already exists in HOOK, then remove it first.
3583 *** New function: remove-hook! HOOK PROC
3585 Remove PROC from the list of functions in HOOK.
3587 *** New function: reset-hook! HOOK
3589 Clear the list of hook functions stored in HOOK.
3591 *** New function: run-hook HOOK ARG1 ...
3593 Run all hook functions stored in HOOK with arguments ARG1 ... .
3594 The number of arguments supplied must correspond to the number given
3595 when the hook was created.
3597 ** The function `dynamic-link' now takes optional keyword arguments.
3598 The only keyword argument that is currently defined is `:global
3599 BOOL'. With it, you can control whether the shared library will be
3600 linked in global mode or not. In global mode, the symbols from the
3601 linked library can be used to resolve references from other
3602 dynamically linked libraries. In non-global mode, the linked
3603 library is essentially invisible and can only be accessed via
3604 `dynamic-func', etc. The default is now to link in global mode.
3605 Previously, the default has been non-global mode.
3607 The `#:global' keyword is only effective on platforms that support
3608 the dlopen family of functions.
3610 ** New function `provided?'
3612 - Function: provided? FEATURE
3613 Return true iff FEATURE is supported by this installation of
3614 Guile. FEATURE must be a symbol naming a feature; the global
3615 variable `*features*' is a list of available features.
3617 ** Changes to the module (ice-9 expect):
3619 *** The expect-strings macro now matches `$' in a regular expression
3620 only at a line-break or end-of-file by default. Previously it would
3621 match the end of the string accumulated so far. The old behaviour
3622 can be obtained by setting the variable `expect-strings-exec-flags'
3625 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3626 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3627 in a regular expression will still match before a line-break or
3628 end-of-file. The default is `regexp/noteol'.
3630 *** The expect-strings macro now uses a variable
3631 `expect-strings-compile-flags' for the flags to be supplied to
3632 `make-regexp'. The default is `regexp/newline', which was previously
3635 *** The expect macro now supplies two arguments to a match procedure:
3636 the current accumulated string and a flag to indicate whether
3637 end-of-file has been reached. Previously only the string was supplied.
3638 If end-of-file is reached, the match procedure will be called an
3639 additional time with the same accumulated string as the previous call
3640 but with the flag set.
3642 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3644 This code, and the documentation for it that appears here, was
3645 borrowed from SLIB, with minor adaptations for Guile.
3647 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3648 An almost complete implementation of Common LISP format description
3649 according to the CL reference book `Common LISP' from Guy L.
3650 Steele, Digital Press. Backward compatible to most of the
3651 available Scheme format implementations.
3653 Returns `#t', `#f' or a string; has side effect of printing
3654 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3655 to the current output port and `#t' is returned. If DESTINATION
3656 is `#f', a formatted string is returned as the result of the call.
3657 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3658 format string; FORMAT-STRING is then the first argument and the
3659 output is returned as a string. If DESTINATION is a number, the
3660 output is to the current error port if available by the
3661 implementation. Otherwise DESTINATION must be an output port and
3664 FORMAT-STRING must be a string. In case of a formatting error
3665 format returns `#f' and prints a message on the current output or
3666 error port. Characters are output as if the string were output by
3667 the `display' function with the exception of those prefixed by a
3668 tilde (~). For a detailed description of the FORMAT-STRING syntax
3669 please consult a Common LISP format reference manual. For a test
3670 suite to verify this format implementation load `formatst.scm'.
3671 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3673 Note: `format' is not reentrant, i.e. only one `format'-call may
3674 be executed at a time.
3677 *** Format Specification (Format version 3.0)
3679 Please consult a Common LISP format reference manual for a detailed
3680 description of the format string syntax. For a demonstration of the
3681 implemented directives see `formatst.scm'.
3683 This implementation supports directive parameters and modifiers (`:'
3684 and `@' characters). Multiple parameters must be separated by a comma
3685 (`,'). Parameters can be numerical parameters (positive or negative),
3686 character parameters (prefixed by a quote character (`''), variable
3687 parameters (`v'), number of rest arguments parameter (`#'), empty and
3688 default parameters. Directive characters are case independent. The
3689 general form of a directive is:
3691 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3693 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3695 *** Implemented CL Format Control Directives
3697 Documentation syntax: Uppercase characters represent the
3698 corresponding control directive characters. Lowercase characters
3699 represent control directive parameter descriptions.
3702 Any (print as `display' does).
3706 `~MINCOL,COLINC,MINPAD,PADCHARA'
3710 S-expression (print as `write' does).
3714 `~MINCOL,COLINC,MINPAD,PADCHARS'
3720 print number sign always.
3723 print comma separated.
3725 `~MINCOL,PADCHAR,COMMACHARD'
3731 print number sign always.
3734 print comma separated.
3736 `~MINCOL,PADCHAR,COMMACHARX'
3742 print number sign always.
3745 print comma separated.
3747 `~MINCOL,PADCHAR,COMMACHARO'
3753 print number sign always.
3756 print comma separated.
3758 `~MINCOL,PADCHAR,COMMACHARB'
3763 `~N,MINCOL,PADCHAR,COMMACHARR'
3767 print a number as a Roman numeral.
3770 print a number as an "old fashioned" Roman numeral.
3773 print a number as an ordinal English number.
3776 print a number as a cardinal English number.
3781 prints `y' and `ies'.
3784 as `~P but jumps 1 argument backward.'
3787 as `~@P but jumps 1 argument backward.'
3792 prints a character as the reader can understand it (i.e. `#\'
3796 prints a character as emacs does (eg. `^C' for ASCII 03).
3799 Fixed-format floating-point (prints a flonum like MMM.NNN).
3800 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3802 If the number is positive a plus sign is printed.
3805 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3806 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3808 If the number is positive a plus sign is printed.
3811 General floating-point (prints a flonum either fixed or
3813 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3815 If the number is positive a plus sign is printed.
3818 Dollars floating-point (prints a flonum in fixed with signs
3820 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3822 If the number is positive a plus sign is printed.
3825 A sign is always printed and appears before the padding.
3828 The sign appears before the padding.
3836 print newline if not at the beginning of the output line.
3838 prints `~&' and then N-1 newlines.
3843 print N page separators.
3853 newline is ignored, white space left.
3856 newline is left, white space ignored.
3861 relative tabulation.
3867 Indirection (expects indirect arguments as a list).
3869 extracts indirect arguments from format arguments.
3872 Case conversion (converts by `string-downcase').
3874 converts by `string-capitalize'.
3877 converts by `string-capitalize-first'.
3880 converts by `string-upcase'.
3883 Argument Jumping (jumps 1 argument forward).
3885 jumps N arguments forward.
3888 jumps 1 argument backward.
3891 jumps N arguments backward.
3894 jumps to the 0th argument.
3897 jumps to the Nth argument (beginning from 0)
3899 `~[STR0~;STR1~;...~;STRN~]'
3900 Conditional Expression (numerical clause conditional).
3902 take argument from N.
3905 true test conditional.
3908 if-else-then conditional.
3914 default clause follows.
3917 Iteration (args come from the next argument (a list)).
3919 at most N iterations.
3922 args from next arg (a list of lists).
3925 args from the rest of arguments.
3928 args from the rest args (lists).
3939 aborts if N <= M <= K
3941 *** Not Implemented CL Format Control Directives
3944 print `#f' as an empty list (see below).
3947 print `#f' as an empty list (see below).
3953 (sorry I don't understand its semantics completely)
3955 *** Extended, Replaced and Additional Control Directives
3957 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3958 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3959 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3960 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3961 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3962 COMMAWIDTH is the number of characters between two comma
3966 print a R4RS complex number as `~F~@Fi' with passed parameters for
3970 Pretty print formatting of an argument for scheme code lists.
3976 Flushes the output if format DESTINATION is a port.
3979 Print a `#\space' character
3981 print N `#\space' characters.
3984 Print a `#\tab' character
3986 print N `#\tab' characters.
3989 Takes N as an integer representation for a character. No arguments
3990 are consumed. N is converted to a character by `integer->char'. N
3991 must be a positive decimal number.
3994 Print out readproof. Prints out internal objects represented as
3995 `#<...>' as strings `"#<...>"' so that the format output can always
3996 be processed by `read'.
3999 Print out readproof. Prints out internal objects represented as
4000 `#<...>' as strings `"#<...>"' so that the format output can always
4001 be processed by `read'.
4004 Prints information and a copyright notice on the format
4007 prints format version.
4010 may also print number strings, i.e. passing a number as a string
4011 and format it accordingly.
4013 *** Configuration Variables
4015 The format module exports some configuration variables to suit the
4016 systems and users needs. There should be no modification necessary for
4017 the configuration that comes with Guile. Format detects automatically
4018 if the running scheme system implements floating point numbers and
4021 format:symbol-case-conv
4022 Symbols are converted by `symbol->string' so the case type of the
4023 printed symbols is implementation dependent.
4024 `format:symbol-case-conv' is a one arg closure which is either
4025 `#f' (no conversion), `string-upcase', `string-downcase' or
4026 `string-capitalize'. (default `#f')
4028 format:iobj-case-conv
4029 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4030 implementation internal objects. (default `#f')
4033 The character prefixing the exponent value in `~E' printing.
4036 *** Compatibility With Other Format Implementations
4042 Downward compatible except for padding support and `~A', `~S',
4043 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4044 `printf' padding support which is completely replaced by the CL
4045 `format' padding style.
4048 Downward compatible except for `~', which is not documented
4049 (ignores all characters inside the format string up to a newline
4050 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4051 numerical and variable parameters and `:/@' modifiers in the CL
4055 Downward compatible except for `~A' and `~S' which print in
4056 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4057 directive parameters or modifiers)).
4060 Downward compatible except for an optional destination parameter:
4061 S2C accepts a format call without a destination which returns a
4062 formatted string. This is equivalent to a #f destination in S2C.
4063 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4064 parameters or modifiers)).
4067 ** Changes to string-handling functions.
4069 These functions were added to support the (ice-9 format) module, above.
4071 *** New function: string-upcase STRING
4072 *** New function: string-downcase STRING
4074 These are non-destructive versions of the existing string-upcase! and
4075 string-downcase! functions.
4077 *** New function: string-capitalize! STRING
4078 *** New function: string-capitalize STRING
4080 These functions convert the first letter of each word in the string to
4083 (string-capitalize "howdy there")
4086 As with the other functions, string-capitalize! modifies the string in
4087 place, while string-capitalize returns a modified copy of its argument.
4089 *** New function: string-ci->symbol STRING
4091 Return a symbol whose name is STRING, but having the same case as if
4092 the symbol had be read by `read'.
4094 Guile can be configured to be sensitive or insensitive to case
4095 differences in Scheme identifiers. If Guile is case-insensitive, all
4096 symbols are converted to lower case on input. The `string-ci->symbol'
4097 function returns a symbol whose name in STRING, transformed as Guile
4098 would if STRING were input.
4100 *** New function: substring-move! STRING1 START END STRING2 START
4102 Copy the substring of STRING1 from START (inclusive) to END
4103 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4104 string, and the source and destination areas may overlap; in all
4105 cases, the function behaves as if all the characters were copied
4108 *** Extended functions: substring-move-left! substring-move-right!
4110 These functions now correctly copy arbitrarily overlapping substrings;
4111 they are both synonyms for substring-move!.
4114 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4116 getopt-long is a function for parsing command-line arguments in a
4117 manner consistent with other GNU programs.
4119 (getopt-long ARGS GRAMMAR)
4120 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4122 ARGS should be a list of strings. Its first element should be the
4123 name of the program; subsequent elements should be the arguments
4124 that were passed to the program on the command line. The
4125 `program-arguments' procedure returns a list of this form.
4127 GRAMMAR is a list of the form:
4128 ((OPTION (PROPERTY VALUE) ...) ...)
4130 Each OPTION should be a symbol. `getopt-long' will accept a
4131 command-line option named `--OPTION'.
4132 Each option can have the following (PROPERTY VALUE) pairs:
4134 (single-char CHAR) --- Accept `-CHAR' as a single-character
4135 equivalent to `--OPTION'. This is how to specify traditional
4137 (required? BOOL) --- If BOOL is true, the option is required.
4138 getopt-long will raise an error if it is not found in ARGS.
4139 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4140 it is #f, it does not; and if it is the symbol
4141 `optional', the option may appear in ARGS with or
4143 (predicate FUNC) --- If the option accepts a value (i.e. you
4144 specified `(value #t)' for this option), then getopt
4145 will apply FUNC to the value, and throw an exception
4146 if it returns #f. FUNC should be a procedure which
4147 accepts a string and returns a boolean value; you may
4148 need to use quasiquotes to get it into GRAMMAR.
4150 The (PROPERTY VALUE) pairs may occur in any order, but each
4151 property may occur only once. By default, options do not have
4152 single-character equivalents, are not required, and do not take
4155 In ARGS, single-character options may be combined, in the usual
4156 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4157 accepts values, then it must be the last option in the
4158 combination; the value is the next argument. So, for example, using
4159 the following grammar:
4160 ((apples (single-char #\a))
4161 (blimps (single-char #\b) (value #t))
4162 (catalexis (single-char #\c) (value #t)))
4163 the following argument lists would be acceptable:
4164 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4165 for "blimps" and "catalexis")
4166 ("-ab" "bang" "-c" "couth") (same)
4167 ("-ac" "couth" "-b" "bang") (same)
4168 ("-abc" "couth" "bang") (an error, since `-b' is not the
4169 last option in its combination)
4171 If an option's value is optional, then `getopt-long' decides
4172 whether it has a value by looking at what follows it in ARGS. If
4173 the next element is a string, and it does not appear to be an
4174 option itself, then that string is the option's value.
4176 The value of a long option can appear as the next element in ARGS,
4177 or it can follow the option name, separated by an `=' character.
4178 Thus, using the same grammar as above, the following argument lists
4180 ("--apples" "Braeburn" "--blimps" "Goodyear")
4181 ("--apples=Braeburn" "--blimps" "Goodyear")
4182 ("--blimps" "Goodyear" "--apples=Braeburn")
4184 If the option "--" appears in ARGS, argument parsing stops there;
4185 subsequent arguments are returned as ordinary arguments, even if
4186 they resemble options. So, in the argument list:
4187 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4188 `getopt-long' will recognize the `apples' option as having the
4189 value "Granny Smith", but it will not recognize the `blimp'
4190 option; it will return the strings "--blimp" and "Goodyear" as
4191 ordinary argument strings.
4193 The `getopt-long' function returns the parsed argument list as an
4194 assocation list, mapping option names --- the symbols from GRAMMAR
4195 --- onto their values, or #t if the option does not accept a value.
4196 Unused options do not appear in the alist.
4198 All arguments that are not the value of any option are returned
4199 as a list, associated with the empty list.
4201 `getopt-long' throws an exception if:
4202 - it finds an unrecognized option in ARGS
4203 - a required option is omitted
4204 - an option that requires an argument doesn't get one
4205 - an option that doesn't accept an argument does get one (this can
4206 only happen using the long option `--opt=value' syntax)
4207 - an option predicate fails
4212 `((lockfile-dir (required? #t)
4215 (predicate ,file-is-directory?))
4216 (verbose (required? #f)
4219 (x-includes (single-char #\x))
4220 (rnet-server (single-char #\y)
4221 (predicate ,string?))))
4223 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4224 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4226 => ((() "foo1" "-fred" "foo2" "foo3")
4227 (rnet-server . "lamprod")
4228 (x-includes . "/usr/include")
4229 (lockfile-dir . "/tmp")
4232 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4234 It will be removed in a few releases.
4236 ** New syntax: lambda*
4237 ** New syntax: define*
4238 ** New syntax: define*-public
4239 ** New syntax: defmacro*
4240 ** New syntax: defmacro*-public
4241 Guile now supports optional arguments.
4243 `lambda*', `define*', `define*-public', `defmacro*' and
4244 `defmacro*-public' are identical to the non-* versions except that
4245 they use an extended type of parameter list that has the following BNF
4246 syntax (parentheses are literal, square brackets indicate grouping,
4247 and `*', `+' and `?' have the usual meaning):
4249 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4250 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4251 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4253 ext-var-decl ::= identifier | ( identifier expression )
4255 The semantics are best illustrated with the following documentation
4256 and examples for `lambda*':
4259 lambda extended for optional and keyword arguments
4261 lambda* creates a procedure that takes optional arguments. These
4262 are specified by putting them inside brackets at the end of the
4263 paramater list, but before any dotted rest argument. For example,
4264 (lambda* (a b #&optional c d . e) '())
4265 creates a procedure with fixed arguments a and b, optional arguments c
4266 and d, and rest argument e. If the optional arguments are omitted
4267 in a call, the variables for them are unbound in the procedure. This
4268 can be checked with the bound? macro.
4270 lambda* can also take keyword arguments. For example, a procedure
4272 (lambda* (#&key xyzzy larch) '())
4273 can be called with any of the argument lists (#:xyzzy 11)
4274 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4275 are given as keywords are bound to values.
4277 Optional and keyword arguments can also be given default values
4278 which they take on when they are not present in a call, by giving a
4279 two-item list in place of an optional argument, for example in:
4280 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4281 foo is a fixed argument, bar is an optional argument with default
4282 value 42, and baz is a keyword argument with default value 73.
4283 Default value expressions are not evaluated unless they are needed
4284 and until the procedure is called.
4286 lambda* now supports two more special parameter list keywords.
4288 lambda*-defined procedures now throw an error by default if a
4289 keyword other than one of those specified is found in the actual
4290 passed arguments. However, specifying #&allow-other-keys
4291 immediately after the kyword argument declarations restores the
4292 previous behavior of ignoring unknown keywords. lambda* also now
4293 guarantees that if the same keyword is passed more than once, the
4294 last one passed is the one that takes effect. For example,
4295 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4296 #:heads 37 #:tails 42 #:heads 99)
4297 would result in (99 47) being displayed.
4299 #&rest is also now provided as a synonym for the dotted syntax rest
4300 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4301 all respects to lambda*. This is provided for more similarity to DSSSL,
4302 MIT-Scheme and Kawa among others, as well as for refugees from other
4305 Further documentation may be found in the optargs.scm file itself.
4307 The optional argument module also exports the macros `let-optional',
4308 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4309 are not documented here because they may be removed in the future, but
4310 full documentation is still available in optargs.scm.
4312 ** New syntax: and-let*
4313 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4315 Syntax: (land* (<clause> ...) <body> ...)
4316 Each <clause> should have one of the following forms:
4317 (<variable> <expression>)
4320 Each <variable> or <bound-variable> should be an identifier. Each
4321 <expression> should be a valid expression. The <body> should be a
4322 possibly empty sequence of expressions, like the <body> of a
4325 Semantics: A LAND* expression is evaluated by evaluating the
4326 <expression> or <bound-variable> of each of the <clause>s from
4327 left to right. The value of the first <expression> or
4328 <bound-variable> that evaluates to a false value is returned; the
4329 remaining <expression>s and <bound-variable>s are not evaluated.
4330 The <body> forms are evaluated iff all the <expression>s and
4331 <bound-variable>s evaluate to true values.
4333 The <expression>s and the <body> are evaluated in an environment
4334 binding each <variable> of the preceding (<variable> <expression>)
4335 clauses to the value of the <expression>. Later bindings
4336 shadow earlier bindings.
4338 Guile's and-let* macro was contributed by Michael Livshin.
4340 ** New sorting functions
4342 *** New function: sorted? SEQUENCE LESS?
4343 Returns `#t' when the sequence argument is in non-decreasing order
4344 according to LESS? (that is, there is no adjacent pair `... x y
4345 ...' for which `(less? y x)').
4347 Returns `#f' when the sequence contains at least one out-of-order
4348 pair. It is an error if the sequence is neither a list nor a
4351 *** New function: merge LIST1 LIST2 LESS?
4352 LIST1 and LIST2 are sorted lists.
4353 Returns the sorted list of all elements in LIST1 and LIST2.
4355 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4356 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4357 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4358 (Here "<" should read "comes before".)
4360 *** New procedure: merge! LIST1 LIST2 LESS?
4361 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4362 the result. If the code is compiled, and LESS? constructs no new
4363 pairs, no pairs at all will be allocated. The first pair of the
4364 result will be either the first pair of LIST1 or the first pair of
4367 *** New function: sort SEQUENCE LESS?
4368 Accepts either a list or a vector, and returns a new sequence
4369 which is sorted. The new sequence is the same type as the input.
4370 Always `(sorted? (sort sequence less?) less?)'. The original
4371 sequence is not altered in any way. The new sequence shares its
4372 elements with the old one; no elements are copied.
4374 *** New procedure: sort! SEQUENCE LESS
4375 Returns its sorted result in the original boxes. No new storage is
4376 allocated at all. Proper usage: (set! slist (sort! slist <))
4378 *** New function: stable-sort SEQUENCE LESS?
4379 Similar to `sort' but stable. That is, if "equal" elements are
4380 ordered a < b in the original sequence, they will have the same order
4383 *** New function: stable-sort! SEQUENCE LESS?
4384 Similar to `sort!' but stable.
4385 Uses temporary storage when sorting vectors.
4387 *** New functions: sort-list, sort-list!
4388 Added for compatibility with scsh.
4390 ** New built-in random number support
4392 *** New function: random N [STATE]
4393 Accepts a positive integer or real N and returns a number of the
4394 same type between zero (inclusive) and N (exclusive). The values
4395 returned have a uniform distribution.
4397 The optional argument STATE must be of the type produced by
4398 `copy-random-state' or `seed->random-state'. It defaults to the value
4399 of the variable `*random-state*'. This object is used to maintain the
4400 state of the pseudo-random-number generator and is altered as a side
4401 effect of the `random' operation.
4403 *** New variable: *random-state*
4404 Holds a data structure that encodes the internal state of the
4405 random-number generator that `random' uses by default. The nature
4406 of this data structure is implementation-dependent. It may be
4407 printed out and successfully read back in, but may or may not
4408 function correctly as a random-number state object in another
4411 *** New function: copy-random-state [STATE]
4412 Returns a new object of type suitable for use as the value of the
4413 variable `*random-state*' and as a second argument to `random'.
4414 If argument STATE is given, a copy of it is returned. Otherwise a
4415 copy of `*random-state*' is returned.
4417 *** New function: seed->random-state SEED
4418 Returns a new object of type suitable for use as the value of the
4419 variable `*random-state*' and as a second argument to `random'.
4420 SEED is a string or a number. A new state is generated and
4421 initialized using SEED.
4423 *** New function: random:uniform [STATE]
4424 Returns an uniformly distributed inexact real random number in the
4425 range between 0 and 1.
4427 *** New procedure: random:solid-sphere! VECT [STATE]
4428 Fills VECT with inexact real random numbers the sum of whose
4429 squares is less than 1.0. Thinking of VECT as coordinates in
4430 space of dimension N = `(vector-length VECT)', the coordinates are
4431 uniformly distributed within the unit N-shere. The sum of the
4432 squares of the numbers is returned. VECT can be either a vector
4433 or a uniform vector of doubles.
4435 *** New procedure: random:hollow-sphere! VECT [STATE]
4436 Fills VECT with inexact real random numbers the sum of whose squares
4437 is equal to 1.0. Thinking of VECT as coordinates in space of
4438 dimension n = `(vector-length VECT)', the coordinates are uniformly
4439 distributed over the surface of the unit n-shere. VECT can be either
4440 a vector or a uniform vector of doubles.
4442 *** New function: random:normal [STATE]
4443 Returns an inexact real in a normal distribution with mean 0 and
4444 standard deviation 1. For a normal distribution with mean M and
4445 standard deviation D use `(+ M (* D (random:normal)))'.
4447 *** New procedure: random:normal-vector! VECT [STATE]
4448 Fills VECT with inexact real random numbers which are independent and
4449 standard normally distributed (i.e., with mean 0 and variance 1).
4450 VECT can be either a vector or a uniform vector of doubles.
4452 *** New function: random:exp STATE
4453 Returns an inexact real in an exponential distribution with mean 1.
4454 For an exponential distribution with mean U use (* U (random:exp)).
4456 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4458 These functions now operate on numbers in the range of a C unsigned
4461 These functions used to operate on numbers in the range of a C signed
4462 long; however, this seems inappropriate, because Guile integers don't
4465 ** New function: make-guardian
4466 This is an implementation of guardians as described in
4467 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4468 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4469 Programming Language Design and Implementation, June 1993
4470 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4472 ** New functions: delq1!, delv1!, delete1!
4473 These procedures behave similar to delq! and friends but delete only
4474 one object if at all.
4476 ** New function: unread-string STRING PORT
4477 Unread STRING to PORT, that is, push it back onto the port so that
4478 next read operation will work on the pushed back characters.
4480 ** unread-char can now be called multiple times
4481 If unread-char is called multiple times, the unread characters will be
4482 read again in last-in first-out order.
4484 ** the procedures uniform-array-read! and uniform-array-write! now
4485 work on any kind of port, not just ports which are open on a file.
4487 ** Now 'l' in a port mode requests line buffering.
4489 ** The procedure truncate-file now works on string ports as well
4490 as file ports. If the size argument is omitted, the current
4491 file position is used.
4493 ** new procedure: seek PORT/FDES OFFSET WHENCE
4494 The arguments are the same as for the old fseek procedure, but it
4495 works on string ports as well as random-access file ports.
4497 ** the fseek procedure now works on string ports, since it has been
4498 redefined using seek.
4500 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4501 size is not supplied.
4503 ** the newline procedure no longer flushes the port if it's not
4504 line-buffered: previously it did if it was the current output port.
4506 ** open-pipe and close-pipe are no longer primitive procedures, but
4507 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4509 ** the freopen procedure has been removed.
4511 ** new procedure: drain-input PORT
4512 Drains PORT's read buffers (including any pushed-back characters)
4513 and returns the contents as a single string.
4515 ** New function: map-in-order PROC LIST1 LIST2 ...
4516 Version of `map' which guarantees that the procedure is applied to the
4517 lists in serial order.
4519 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4520 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4521 now obsolete and will go away in release 1.5.
4523 ** New syntax: collect BODY1 ...
4524 Version of `begin' which returns a list of the results of the body
4525 forms instead of the result of the last body form. In contrast to
4526 `begin', `collect' allows an empty body.
4528 ** New functions: read-history FILENAME, write-history FILENAME
4529 Read/write command line history from/to file. Returns #t on success
4530 and #f if an error occured.
4532 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4534 These procedures return a list of definitions available in the specified
4535 argument, a relative module reference. In the case of no argument,
4536 `(current-module)' is now consulted for definitions to return, instead
4537 of simply returning #f, the former behavior.
4539 ** The #/ syntax for lists is no longer supported.
4541 Earlier versions of Scheme accepted this syntax, but printed a
4544 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4546 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4549 * Changes to the gh_ interface
4553 Now takes a second argument which is the result array. If this
4554 pointer is NULL, a new array is malloced (the old behaviour).
4556 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4557 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4561 * Changes to the scm_ interface
4563 ** Function: scm_make_named_hook (char* name, int n_args)
4565 Creates a hook in the same way as make-hook above but also
4566 binds a variable named NAME to it.
4568 This is the typical way of creating a hook from C code.
4570 Currently, the variable is created in the "current" module. This
4571 might change when we get the new module system.
4573 ** The smob interface
4575 The interface for creating smobs has changed. For documentation, see
4576 data-rep.info (made from guile-core/doc/data-rep.texi).
4578 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4580 >>> This function will be removed in 1.3.4. <<<
4584 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4585 This function adds a new smob type, named NAME, with instance size
4586 SIZE to the system. The return value is a tag that is used in
4587 creating instances of the type. If SIZE is 0, then no memory will
4588 be allocated when instances of the smob are created, and nothing
4589 will be freed by the default free function.
4591 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4592 This function sets the smob marking procedure for the smob type
4593 specified by the tag TC. TC is the tag returned by
4594 `scm_make_smob_type'.
4596 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4597 This function sets the smob freeing procedure for the smob type
4598 specified by the tag TC. TC is the tag returned by
4599 `scm_make_smob_type'.
4601 *** Function: void scm_set_smob_print (tc, print)
4603 - Function: void scm_set_smob_print (long tc,
4604 scm_sizet (*print) (SCM,
4608 This function sets the smob printing procedure for the smob type
4609 specified by the tag TC. TC is the tag returned by
4610 `scm_make_smob_type'.
4612 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4613 This function sets the smob equality-testing predicate for the
4614 smob type specified by the tag TC. TC is the tag returned by
4615 `scm_make_smob_type'.
4617 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4618 Make VALUE contain a smob instance of the type with type code TC and
4619 smob data DATA. VALUE must be previously declared as C type `SCM'.
4621 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4622 This macro expands to a block of code that creates a smob instance
4623 of the type with type code TC and smob data DATA, and returns that
4624 `SCM' value. It should be the last piece of code in a block.
4626 ** The interfaces for using I/O ports and implementing port types
4627 (ptobs) have changed significantly. The new interface is based on
4628 shared access to buffers and a new set of ptob procedures.
4630 *** scm_newptob has been removed
4634 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4636 - Function: SCM scm_make_port_type (char *type_name,
4637 int (*fill_buffer) (SCM port),
4638 void (*write_flush) (SCM port));
4640 Similarly to the new smob interface, there is a set of function
4641 setters by which the user can customize the behaviour of his port
4642 type. See ports.h (scm_set_port_XXX).
4644 ** scm_strport_to_string: New function: creates a new string from
4645 a string port's buffer.
4647 ** Plug in interface for random number generators
4648 The variable `scm_the_rng' in random.c contains a value and three
4649 function pointers which together define the current random number
4650 generator being used by the Scheme level interface and the random
4651 number library functions.
4653 The user is free to replace the default generator with the generator
4656 *** Variable: size_t scm_the_rng.rstate_size
4657 The size of the random state type used by the current RNG
4660 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4661 Given the random STATE, return 32 random bits.
4663 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4664 Seed random state STATE using string S of length N.
4666 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4667 Given random state STATE, return a malloced copy.
4670 The default RNG is the MWC (Multiply With Carry) random number
4671 generator described by George Marsaglia at the Department of
4672 Statistics and Supercomputer Computations Research Institute, The
4673 Florida State University (http://stat.fsu.edu/~geo).
4675 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4676 passes all tests in the DIEHARD test suite
4677 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4678 costs one multiply and one add on platforms which either supports long
4679 longs (gcc does this on most systems) or have 64 bit longs. The cost
4680 is four multiply on other systems but this can be optimized by writing
4681 scm_i_uniform32 in assembler.
4683 These functions are provided through the scm_the_rng interface for use
4684 by libguile and the application.
4686 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4687 Given the random STATE, return 32 random bits.
4688 Don't use this function directly. Instead go through the plugin
4689 interface (see "Plug in interface" above).
4691 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4692 Initialize STATE using SEED of length N.
4694 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4695 Return a malloc:ed copy of STATE. This function can easily be re-used
4696 in the interfaces to other RNGs.
4698 ** Random number library functions
4699 These functions use the current RNG through the scm_the_rng interface.
4700 It might be a good idea to use these functions from your C code so
4701 that only one random generator is used by all code in your program.
4703 The default random state is stored in:
4705 *** Variable: SCM scm_var_random_state
4706 Contains the vcell of the Scheme variable "*random-state*" which is
4707 used as default state by all random number functions in the Scheme
4712 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4714 *** Function: scm_rstate *scm_c_default_rstate (void)
4715 This is a convenience function which returns the value of
4716 scm_var_random_state. An error message is generated if this value
4717 isn't a random state.
4719 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4720 Make a new random state from the string SEED of length LENGTH.
4722 It is generally not a good idea to use multiple random states in a
4723 program. While subsequent random numbers generated from one random
4724 state are guaranteed to be reasonably independent, there is no such
4725 guarantee for numbers generated from different random states.
4727 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4728 Return 32 random bits.
4730 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4731 Return a sample from the uniform(0,1) distribution.
4733 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4734 Return a sample from the normal(0,1) distribution.
4736 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4737 Return a sample from the exp(1) distribution.
4739 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4740 Return a sample from the discrete uniform(0,M) distribution.
4742 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4743 Return a sample from the discrete uniform(0,M) distribution.
4744 M must be a bignum object. The returned value may be an INUM.
4748 Changes in Guile 1.3 (released Monday, October 19, 1998):
4750 * Changes to the distribution
4752 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4753 To avoid conflicts, programs should name environment variables after
4754 themselves, except when there's a common practice establishing some
4757 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4758 giving the former precedence, and printing a warning message if the
4759 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4761 ** The header files related to multi-byte characters have been removed.
4762 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4763 which referred to these explicitly will probably need to be rewritten,
4764 since the support for the variant string types has been removed; see
4767 ** The header files append.h and sequences.h have been removed. These
4768 files implemented non-R4RS operations which would encourage
4769 non-portable programming style and less easy-to-read code.
4771 * Changes to the stand-alone interpreter
4773 ** New procedures have been added to implement a "batch mode":
4775 *** Function: batch-mode?
4777 Returns a boolean indicating whether the interpreter is in batch
4780 *** Function: set-batch-mode?! ARG
4782 If ARG is true, switches the interpreter to batch mode. The `#f'
4783 case has not been implemented.
4785 ** Guile now provides full command-line editing, when run interactively.
4786 To use this feature, you must have the readline library installed.
4787 The Guile build process will notice it, and automatically include
4790 The readline library is available via anonymous FTP from any GNU
4791 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4793 ** the-last-stack is now a fluid.
4795 * Changes to the procedure for linking libguile with your programs
4797 ** You can now use the `guile-config' utility to build programs that use Guile.
4799 Guile now includes a command-line utility called `guile-config', which
4800 can provide information about how to compile and link programs that
4803 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4804 You should include this command's output on the command line you use
4805 to compile C or C++ code that #includes the Guile header files. It's
4806 usually just a `-I' flag to help the compiler find the Guile headers.
4809 *** `guile-config link' prints any linker flags necessary to link with Guile.
4811 This command writes to its standard output a list of flags which you
4812 must pass to the linker to link your code against the Guile library.
4813 The flags include '-lguile' itself, any other libraries the Guile
4814 library depends upon, and any `-L' flags needed to help the linker
4815 find those libraries.
4817 For example, here is a Makefile rule that builds a program named 'foo'
4818 from the object files ${FOO_OBJECTS}, and links them against Guile:
4821 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4823 Previous Guile releases recommended that you use autoconf to detect
4824 which of a predefined set of libraries were present on your system.
4825 It is more robust to use `guile-config', since it records exactly which
4826 libraries the installed Guile library requires.
4828 This was originally called `build-guile', but was renamed to
4829 `guile-config' before Guile 1.3 was released, to be consistent with
4830 the analogous script for the GTK+ GUI toolkit, which is called
4834 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4836 If you are using the GNU autoconf package to configure your program,
4837 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4838 (described above) and gather the necessary values for use in your
4841 The GUILE_FLAGS macro expands to configure script code which runs the
4842 `guile-config' script, to find out where Guile's header files and
4843 libraries are installed. It sets two variables, marked for
4844 substitution, as by AC_SUBST.
4846 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4847 code that uses Guile header files. This is almost always just a
4850 GUILE_LDFLAGS --- flags to pass to the linker to link a
4851 program against Guile. This includes `-lguile' for the Guile
4852 library itself, any libraries that Guile itself requires (like
4853 -lqthreads), and so on. It may also include a -L flag to tell the
4854 compiler where to find the libraries.
4856 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4857 directory of the Guile distribution. You can copy it into your
4858 package's aclocal.m4 file, and then use it in your configure.in file.
4860 If you are using the `aclocal' program, distributed with GNU automake,
4861 to maintain your aclocal.m4 file, the Guile installation process
4862 installs guile.m4 where aclocal will find it. All you need to do is
4863 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4864 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4868 * Changes to Scheme functions and syntax
4870 ** Multi-byte strings have been removed, as have multi-byte and wide
4871 ports. We felt that these were the wrong approach to
4872 internationalization support.
4874 ** New function: readline [PROMPT]
4875 Read a line from the terminal, and allow the user to edit it,
4876 prompting with PROMPT. READLINE provides a large set of Emacs-like
4877 editing commands, lets the user recall previously typed lines, and
4878 works on almost every kind of terminal, including dumb terminals.
4880 READLINE assumes that the cursor is at the beginning of the line when
4881 it is invoked. Thus, you can't print a prompt yourself, and then call
4882 READLINE; you need to package up your prompt as a string, pass it to
4883 the function, and let READLINE print the prompt itself. This is
4884 because READLINE needs to know the prompt's screen width.
4886 For Guile to provide this function, you must have the readline
4887 library, version 2.1 or later, installed on your system. Readline is
4888 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4889 any GNU mirror site.
4891 See also ADD-HISTORY function.
4893 ** New function: add-history STRING
4894 Add STRING as the most recent line in the history used by the READLINE
4895 command. READLINE does not add lines to the history itself; you must
4896 call ADD-HISTORY to make previous input available to the user.
4898 ** The behavior of the read-line function has changed.
4900 This function now uses standard C library functions to read the line,
4901 for speed. This means that it doesn not respect the value of
4902 scm-line-incrementors; it assumes that lines are delimited with
4905 (Note that this is read-line, the function that reads a line of text
4906 from a port, not readline, the function that reads a line from a
4907 terminal, providing full editing capabilities.)
4909 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4911 This module provides some simple argument parsing. It exports one
4914 Function: getopt-gnu-style ARG-LS
4915 Parse a list of program arguments into an alist of option
4918 Each item in the list of program arguments is examined to see if
4919 it meets the syntax of a GNU long-named option. An argument like
4920 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4921 returned alist, where MUMBLE is a keyword object with the same
4922 name as the argument. An argument like `--MUMBLE=FROB' produces
4923 an element of the form (MUMBLE . FROB), where FROB is a string.
4925 As a special case, the returned alist also contains a pair whose
4926 car is the symbol `rest'. The cdr of this pair is a list
4927 containing all the items in the argument list that are not options
4928 of the form mentioned above.
4930 The argument `--' is treated specially: all items in the argument
4931 list appearing after such an argument are not examined, and are
4932 returned in the special `rest' list.
4934 This function does not parse normal single-character switches.
4935 You will need to parse them out of the `rest' list yourself.
4937 ** The read syntax for byte vectors and short vectors has changed.
4939 Instead of #bytes(...), write #y(...).
4941 Instead of #short(...), write #h(...).
4943 This may seem nutty, but, like the other uniform vectors, byte vectors
4944 and short vectors want to have the same print and read syntax (and,
4945 more basic, want to have read syntax!). Changing the read syntax to
4946 use multiple characters after the hash sign breaks with the
4947 conventions used in R5RS and the conventions used for the other
4948 uniform vectors. It also introduces complexity in the current reader,
4949 both on the C and Scheme levels. (The Right solution is probably to
4950 change the syntax and prototypes for uniform vectors entirely.)
4953 ** The new module (ice-9 session) provides useful interactive functions.
4955 *** New procedure: (apropos REGEXP OPTION ...)
4957 Display a list of top-level variables whose names match REGEXP, and
4958 the modules they are imported from. Each OPTION should be one of the
4961 value --- Show the value of each matching variable.
4962 shadow --- Show bindings shadowed by subsequently imported modules.
4963 full --- Same as both `shadow' and `value'.
4967 guile> (apropos "trace" 'full)
4968 debug: trace #<procedure trace args>
4969 debug: untrace #<procedure untrace args>
4970 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4971 the-scm-module: before-backtrace-hook ()
4972 the-scm-module: backtrace #<primitive-procedure backtrace>
4973 the-scm-module: after-backtrace-hook ()
4974 the-scm-module: has-shown-backtrace-hint? #f
4977 ** There are new functions and syntax for working with macros.
4979 Guile implements macros as a special object type. Any variable whose
4980 top-level binding is a macro object acts as a macro. The macro object
4981 specifies how the expression should be transformed before evaluation.
4983 *** Macro objects now print in a reasonable way, resembling procedures.
4985 *** New function: (macro? OBJ)
4986 True iff OBJ is a macro object.
4988 *** New function: (primitive-macro? OBJ)
4989 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4990 macro transformers, implemented in eval.c rather than Scheme code.
4992 Why do we have this function?
4993 - For symmetry with procedure? and primitive-procedure?,
4994 - to allow custom print procedures to tell whether a macro is
4995 primitive, and display it differently, and
4996 - to allow compilers and user-written evaluators to distinguish
4997 builtin special forms from user-defined ones, which could be
5000 *** New function: (macro-type OBJ)
5001 Return a value indicating what kind of macro OBJ is. Possible return
5004 The symbol `syntax' --- a macro created by procedure->syntax.
5005 The symbol `macro' --- a macro created by procedure->macro.
5006 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5007 The boolean #f --- if OBJ is not a macro object.
5009 *** New function: (macro-name MACRO)
5010 Return the name of the macro object MACRO's procedure, as returned by
5013 *** New function: (macro-transformer MACRO)
5014 Return the transformer procedure for MACRO.
5016 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5018 Specify a new macro expander to use in the current module. Each
5019 MODULE is a module name, with the same meaning as in the `use-modules'
5020 form; each named module's exported bindings are added to the current
5021 top-level environment. TRANSFORMER is an expression evaluated in the
5022 resulting environment which must yield a procedure to use as the
5023 module's eval transformer: every expression evaluated in this module
5024 is passed to this function, and the result passed to the Guile
5027 *** macro-eval! is removed. Use local-eval instead.
5029 ** Some magic has been added to the printer to better handle user
5030 written printing routines (like record printers, closure printers).
5032 The problem is that these user written routines must have access to
5033 the current `print-state' to be able to handle fancy things like
5034 detection of circular references. These print-states have to be
5035 passed to the builtin printing routines (display, write, etc) to
5036 properly continue the print chain.
5038 We didn't want to change all existing print code so that it
5039 explicitly passes thru a print state in addition to a port. Instead,
5040 we extented the possible values that the builtin printing routines
5041 accept as a `port'. In addition to a normal port, they now also take
5042 a pair of a normal port and a print-state. Printing will go to the
5043 port and the print-state will be used to control the detection of
5044 circular references, etc. If the builtin function does not care for a
5045 print-state, it is simply ignored.
5047 User written callbacks are now called with such a pair as their
5048 `port', but because every function now accepts this pair as a PORT
5049 argument, you don't have to worry about that. In fact, it is probably
5050 safest to not check for these pairs.
5052 However, it is sometimes necessary to continue a print chain on a
5053 different port, for example to get a intermediate string
5054 representation of the printed value, mangle that string somehow, and
5055 then to finally print the mangled string. Use the new function
5057 inherit-print-state OLD-PORT NEW-PORT
5059 for this. It constructs a new `port' that prints to NEW-PORT but
5060 inherits the print-state of OLD-PORT.
5062 ** struct-vtable-offset renamed to vtable-offset-user
5064 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5066 ** There is now a third optional argument to make-vtable-vtable
5067 (and fourth to make-struct) when constructing new types (vtables).
5068 This argument initializes field vtable-index-printer of the vtable.
5070 ** The detection of circular references has been extended to structs.
5071 That is, a structure that -- in the process of being printed -- prints
5072 itself does not lead to infinite recursion.
5074 ** There is now some basic support for fluids. Please read
5075 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5076 the following functions and macros:
5078 Function: make-fluid
5080 Create a new fluid object. Fluids are not special variables or
5081 some other extension to the semantics of Scheme, but rather
5082 ordinary Scheme objects. You can store them into variables (that
5083 are still lexically scoped, of course) or into any other place you
5084 like. Every fluid has a initial value of `#f'.
5086 Function: fluid? OBJ
5088 Test whether OBJ is a fluid.
5090 Function: fluid-ref FLUID
5091 Function: fluid-set! FLUID VAL
5093 Access/modify the fluid FLUID. Modifications are only visible
5094 within the current dynamic root (that includes threads).
5096 Function: with-fluids* FLUIDS VALUES THUNK
5098 FLUIDS is a list of fluids and VALUES a corresponding list of
5099 values for these fluids. Before THUNK gets called the values are
5100 installed in the fluids and the old values of the fluids are
5101 saved in the VALUES list. When the flow of control leaves THUNK
5102 or reenters it, the values get swapped again. You might think of
5103 this as a `safe-fluid-excursion'. Note that the VALUES list is
5104 modified by `with-fluids*'.
5106 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5108 The same as `with-fluids*' but with a different syntax. It looks
5109 just like `let', but both FLUID and VALUE are evaluated. Remember,
5110 fluids are not special variables but ordinary objects. FLUID
5111 should evaluate to a fluid.
5113 ** Changes to system call interfaces:
5115 *** close-port, close-input-port and close-output-port now return a
5116 boolean instead of an `unspecified' object. #t means that the port
5117 was successfully closed, while #f means it was already closed. It is
5118 also now possible for these procedures to raise an exception if an
5119 error occurs (some errors from write can be delayed until close.)
5121 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5124 *** the third argument to fcntl is now optional.
5126 *** the first argument to chown can now be a file descriptor or a port.
5128 *** the argument to stat can now be a port.
5130 *** The following new procedures have been added (most use scsh
5133 *** procedure: close PORT/FD
5134 Similar to close-port (*note close-port: Closing Ports.), but also
5135 works on file descriptors. A side effect of closing a file
5136 descriptor is that any ports using that file descriptor are moved
5137 to a different file descriptor and have their revealed counts set
5140 *** procedure: port->fdes PORT
5141 Returns the integer file descriptor underlying PORT. As a side
5142 effect the revealed count of PORT is incremented.
5144 *** procedure: fdes->ports FDES
5145 Returns a list of existing ports which have FDES as an underlying
5146 file descriptor, without changing their revealed counts.
5148 *** procedure: fdes->inport FDES
5149 Returns an existing input port which has FDES as its underlying
5150 file descriptor, if one exists, and increments its revealed count.
5151 Otherwise, returns a new input port with a revealed count of 1.
5153 *** procedure: fdes->outport FDES
5154 Returns an existing output port which has FDES as its underlying
5155 file descriptor, if one exists, and increments its revealed count.
5156 Otherwise, returns a new output port with a revealed count of 1.
5158 The next group of procedures perform a `dup2' system call, if NEWFD
5159 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5160 duplicated can be supplied as an integer or contained in a port. The
5161 type of value returned varies depending on which procedure is used.
5163 All procedures also have the side effect when performing `dup2' that
5164 any ports using NEWFD are moved to a different file descriptor and have
5165 their revealed counts set to zero.
5167 *** procedure: dup->fdes PORT/FD [NEWFD]
5168 Returns an integer file descriptor.
5170 *** procedure: dup->inport PORT/FD [NEWFD]
5171 Returns a new input port using the new file descriptor.
5173 *** procedure: dup->outport PORT/FD [NEWFD]
5174 Returns a new output port using the new file descriptor.
5176 *** procedure: dup PORT/FD [NEWFD]
5177 Returns a new port if PORT/FD is a port, with the same mode as the
5178 supplied port, otherwise returns an integer file descriptor.
5180 *** procedure: dup->port PORT/FD MODE [NEWFD]
5181 Returns a new port using the new file descriptor. MODE supplies a
5182 mode string for the port (*note open-file: File Ports.).
5184 *** procedure: setenv NAME VALUE
5185 Modifies the environment of the current process, which is also the
5186 default environment inherited by child processes.
5188 If VALUE is `#f', then NAME is removed from the environment.
5189 Otherwise, the string NAME=VALUE is added to the environment,
5190 replacing any existing string with name matching NAME.
5192 The return value is unspecified.
5194 *** procedure: truncate-file OBJ SIZE
5195 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5196 can be a string containing a file name or an integer file
5197 descriptor or port open for output on the file. The underlying
5198 system calls are `truncate' and `ftruncate'.
5200 The return value is unspecified.
5202 *** procedure: setvbuf PORT MODE [SIZE]
5203 Set the buffering mode for PORT. MODE can be:
5211 block buffered, using a newly allocated buffer of SIZE bytes.
5212 However if SIZE is zero or unspecified, the port will be made
5215 This procedure should not be used after I/O has been performed with
5218 Ports are usually block buffered by default, with a default buffer
5219 size. Procedures e.g., *Note open-file: File Ports, which accept a
5220 mode string allow `0' to be added to request an unbuffered port.
5222 *** procedure: fsync PORT/FD
5223 Copies any unwritten data for the specified output file descriptor
5224 to disk. If PORT/FD is a port, its buffer is flushed before the
5225 underlying file descriptor is fsync'd. The return value is
5228 *** procedure: open-fdes PATH FLAGS [MODES]
5229 Similar to `open' but returns a file descriptor instead of a port.
5231 *** procedure: execle PATH ENV [ARG] ...
5232 Similar to `execl', but the environment of the new process is
5233 specified by ENV, which must be a list of strings as returned by
5234 the `environ' procedure.
5236 This procedure is currently implemented using the `execve' system
5237 call, but we call it `execle' because of its Scheme calling
5240 *** procedure: strerror ERRNO
5241 Returns the Unix error message corresponding to ERRNO, an integer.
5243 *** procedure: primitive-exit [STATUS]
5244 Terminate the current process without unwinding the Scheme stack.
5245 This is would typically be useful after a fork. The exit status
5246 is STATUS if supplied, otherwise zero.
5248 *** procedure: times
5249 Returns an object with information about real and processor time.
5250 The following procedures accept such an object as an argument and
5251 return a selected component:
5254 The current real time, expressed as time units relative to an
5258 The CPU time units used by the calling process.
5261 The CPU time units used by the system on behalf of the
5265 The CPU time units used by terminated child processes of the
5266 calling process, whose status has been collected (e.g., using
5270 Similarly, the CPU times units used by the system on behalf of
5271 terminated child processes.
5273 ** Removed: list-length
5274 ** Removed: list-append, list-append!
5275 ** Removed: list-reverse, list-reverse!
5277 ** array-map renamed to array-map!
5279 ** serial-array-map renamed to serial-array-map!
5281 ** catch doesn't take #f as first argument any longer
5283 Previously, it was possible to pass #f instead of a key to `catch'.
5284 That would cause `catch' to pass a jump buffer object to the procedure
5285 passed as second argument. The procedure could then use this jump
5286 buffer objekt as an argument to throw.
5288 This mechanism has been removed since its utility doesn't motivate the
5289 extra complexity it introduces.
5291 ** The `#/' notation for lists now provokes a warning message from Guile.
5292 This syntax will be removed from Guile in the near future.
5294 To disable the warning message, set the GUILE_HUSH environment
5295 variable to any non-empty value.
5297 ** The newline character now prints as `#\newline', following the
5298 normal Scheme notation, not `#\nl'.
5300 * Changes to the gh_ interface
5302 ** The gh_enter function now takes care of loading the Guile startup files.
5303 gh_enter works by calling scm_boot_guile; see the remarks below.
5305 ** Function: void gh_write (SCM x)
5307 Write the printed representation of the scheme object x to the current
5308 output port. Corresponds to the scheme level `write'.
5310 ** gh_list_length renamed to gh_length.
5312 ** vector handling routines
5314 Several major changes. In particular, gh_vector() now resembles
5315 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5316 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5317 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5318 vector-related gh_ functions have been implemented.
5320 ** pair and list routines
5322 Implemented several of the R4RS pair and list functions that were
5325 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5327 New function. Converts double arrays back and forth between Scheme
5330 * Changes to the scm_ interface
5332 ** The function scm_boot_guile now takes care of loading the startup files.
5334 Guile's primary initialization function, scm_boot_guile, now takes
5335 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5336 Guile, define the module system, and put together some standard
5337 bindings. It also loads `init.scm', which is intended to hold
5338 site-specific initialization code.
5340 Since Guile cannot operate properly until boot-9.scm is loaded, there
5341 is no reason to separate loading boot-9.scm from Guile's other
5342 initialization processes.
5344 This job used to be done by scm_compile_shell_switches, which didn't
5345 make much sense; in particular, it meant that people using Guile for
5346 non-shell-like applications had to jump through hoops to get Guile
5347 initialized properly.
5349 ** The function scm_compile_shell_switches no longer loads the startup files.
5350 Now, Guile always loads the startup files, whenever it is initialized;
5351 see the notes above for scm_boot_guile and scm_load_startup_files.
5353 ** Function: scm_load_startup_files
5354 This new function takes care of loading Guile's initialization file
5355 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5356 this is always called by the Guile initialization process, it's
5357 probably not too useful to call this yourself, but it's there anyway.
5359 ** The semantics of smob marking have changed slightly.
5361 The smob marking function (the `mark' member of the scm_smobfuns
5362 structure) is no longer responsible for setting the mark bit on the
5363 smob. The generic smob handling code in the garbage collector will
5364 set this bit. The mark function need only ensure that any other
5365 objects the smob refers to get marked.
5367 Note that this change means that the smob's GC8MARK bit is typically
5368 already set upon entry to the mark function. Thus, marking functions
5369 which look like this:
5372 if (SCM_GC8MARKP (ptr))
5374 SCM_SETGC8MARK (ptr);
5375 ... mark objects to which the smob refers ...
5378 are now incorrect, since they will return early, and fail to mark any
5379 other objects the smob refers to. Some code in the Guile library used
5382 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5384 If you have implemented your own I/O port type, by writing the
5385 functions required by the scm_ptobfuns and then calling scm_newptob,
5386 you will need to change your functions slightly.
5388 The functions in a scm_ptobfuns structure now expect the port itself
5389 as their argument; they used to expect the `stream' member of the
5390 port's scm_port_table structure. This allows functions in an
5391 scm_ptobfuns structure to easily access the port's cell (and any flags
5392 it its CAR), and the port's scm_port_table structure.
5394 Guile now passes the I/O port itself as the `port' argument in the
5395 following scm_ptobfuns functions:
5397 int (*free) (SCM port);
5398 int (*fputc) (int, SCM port);
5399 int (*fputs) (char *, SCM port);
5400 scm_sizet (*fwrite) SCM_P ((char *ptr,
5404 int (*fflush) (SCM port);
5405 int (*fgetc) (SCM port);
5406 int (*fclose) (SCM port);
5408 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5411 If you have existing code which defines its own port types, it is easy
5412 to convert your code to the new interface; simply apply SCM_STREAM to
5413 the port argument to yield the value you code used to expect.
5415 Note that since both the port and the stream have the same type in the
5416 C code --- they are both SCM values --- the C compiler will not remind
5417 you if you forget to update your scm_ptobfuns functions.
5420 ** Function: int scm_internal_select (int fds,
5424 struct timeval *timeout);
5426 This is a replacement for the `select' function provided by the OS.
5427 It enables I/O blocking and sleeping to happen for one cooperative
5428 thread without blocking other threads. It also avoids busy-loops in
5429 these situations. It is intended that all I/O blocking and sleeping
5430 will finally go through this function. Currently, this function is
5431 only available on systems providing `gettimeofday' and `select'.
5433 ** Function: SCM scm_internal_stack_catch (SCM tag,
5434 scm_catch_body_t body,
5436 scm_catch_handler_t handler,
5439 A new sibling to the other two C level `catch' functions
5440 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5441 the stack to be saved automatically into the variable `the-last-stack'
5442 (scm_the_last_stack_var) on error. This is necessary if you want to
5443 use advanced error reporting, such as calling scm_display_error and
5444 scm_display_backtrace. (They both take a stack object as argument.)
5446 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5448 scm_catch_handler_t handler,
5451 Spawns a new thread. It does a job similar to
5452 scm_call_with_new_thread but takes arguments more suitable when
5453 spawning threads from application C code.
5455 ** The hook scm_error_callback has been removed. It was originally
5456 intended as a way for the user to install his own error handler. But
5457 that method works badly since it intervenes between throw and catch,
5458 thereby changing the semantics of expressions like (catch #t ...).
5459 The correct way to do it is to use one of the C level catch functions
5460 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5462 ** Removed functions:
5464 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5465 scm_list_reverse, scm_list_reverse_x
5467 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5469 These can be used for pretty list creation from C. The idea is taken
5470 from Erick Gallesio's STk.
5472 ** scm_array_map renamed to scm_array_map_x
5474 ** mbstrings are now removed
5476 This means that the type codes scm_tc7_mb_string and
5477 scm_tc7_mb_substring has been removed.
5479 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5481 Since we no longer support multi-byte strings, these I/O functions
5482 have been simplified, and renamed. Here are their old names, and
5483 their new names and arguments:
5485 scm_gen_putc -> void scm_putc (int c, SCM port);
5486 scm_gen_puts -> void scm_puts (char *s, SCM port);
5487 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5488 scm_gen_getc -> void scm_getc (SCM port);
5491 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5493 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5495 SCM_TYP7S now masks away the bit which distinguishes substrings from
5498 ** scm_catch_body_t: Backward incompatible change!
5500 Body functions to scm_internal_catch and friends do not any longer
5501 take a second argument. This is because it is no longer possible to
5502 pass a #f arg to catch.
5504 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5506 The function scm_protect_object protects its argument from being freed
5507 by the garbage collector. scm_unprotect_object removes that
5510 These functions now nest properly. That is, for every object O, there
5511 is a counter which scm_protect_object(O) increments and
5512 scm_unprotect_object(O) decrements, if the counter is greater than
5513 zero. Every object's counter is zero when it is first created. If an
5514 object's counter is greater than zero, the garbage collector will not
5515 reclaim its storage.
5517 This allows you to use scm_protect_object in your code without
5518 worrying that some other function you call will call
5519 scm_unprotect_object, and allow it to be freed. Assuming that the
5520 functions you call are well-behaved, and unprotect only those objects
5521 they protect, you can follow the same rule and have confidence that
5522 objects will be freed only at appropriate times.
5525 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5527 * Changes to the distribution
5529 ** Nightly snapshots are now available from ftp.red-bean.com.
5530 The old server, ftp.cyclic.com, has been relinquished to its rightful
5533 Nightly snapshots of the Guile development sources are now available via
5534 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5536 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5537 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5539 ** To run Guile without installing it, the procedure has changed a bit.
5541 If you used a separate build directory to compile Guile, you'll need
5542 to include the build directory in SCHEME_LOAD_PATH, as well as the
5543 source directory. See the `INSTALL' file for examples.
5545 * Changes to the procedure for linking libguile with your programs
5547 ** The standard Guile load path for Scheme code now includes
5548 $(datadir)/guile (usually /usr/local/share/guile). This means that
5549 you can install your own Scheme files there, and Guile will find them.
5550 (Previous versions of Guile only checked a directory whose name
5551 contained the Guile version number, so you had to re-install or move
5552 your Scheme sources each time you installed a fresh version of Guile.)
5554 The load path also includes $(datadir)/guile/site; we recommend
5555 putting individual Scheme files there. If you want to install a
5556 package with multiple source files, create a directory for them under
5559 ** Guile 1.2 will now use the Rx regular expression library, if it is
5560 installed on your system. When you are linking libguile into your own
5561 programs, this means you will have to link against -lguile, -lqt (if
5562 you configured Guile with thread support), and -lrx.
5564 If you are using autoconf to generate configuration scripts for your
5565 application, the following lines should suffice to add the appropriate
5566 libraries to your link command:
5568 ### Find Rx, quickthreads and libguile.
5569 AC_CHECK_LIB(rx, main)
5570 AC_CHECK_LIB(qt, main)
5571 AC_CHECK_LIB(guile, scm_shell)
5573 The Guile 1.2 distribution does not contain sources for the Rx
5574 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5575 retrieve it from a GNU FTP site and install it separately.
5577 * Changes to Scheme functions and syntax
5579 ** The dynamic linking features of Guile are now enabled by default.
5580 You can disable them by giving the `--disable-dynamic-linking' option
5583 (dynamic-link FILENAME)
5585 Find the object file denoted by FILENAME (a string) and link it
5586 into the running Guile application. When everything works out,
5587 return a Scheme object suitable for representing the linked object
5588 file. Otherwise an error is thrown. How object files are
5589 searched is system dependent.
5591 (dynamic-object? VAL)
5593 Determine whether VAL represents a dynamically linked object file.
5595 (dynamic-unlink DYNOBJ)
5597 Unlink the indicated object file from the application. DYNOBJ
5598 should be one of the values returned by `dynamic-link'.
5600 (dynamic-func FUNCTION DYNOBJ)
5602 Search the C function indicated by FUNCTION (a string or symbol)
5603 in DYNOBJ and return some Scheme object that can later be used
5604 with `dynamic-call' to actually call this function. Right now,
5605 these Scheme objects are formed by casting the address of the
5606 function to `long' and converting this number to its Scheme
5609 (dynamic-call FUNCTION DYNOBJ)
5611 Call the C function indicated by FUNCTION and DYNOBJ. The
5612 function is passed no arguments and its return value is ignored.
5613 When FUNCTION is something returned by `dynamic-func', call that
5614 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5615 etc.), look it up in DYNOBJ; this is equivalent to
5617 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5619 Interrupts are deferred while the C function is executing (with
5620 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5622 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5624 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5625 some arguments and return its return value. The C function is
5626 expected to take two arguments and return an `int', just like
5629 int c_func (int argc, char **argv);
5631 ARGS must be a list of strings and is converted into an array of
5632 `char *'. The array is passed in ARGV and its size in ARGC. The
5633 return value is converted to a Scheme number and returned from the
5634 call to `dynamic-args-call'.
5636 When dynamic linking is disabled or not supported on your system,
5637 the above functions throw errors, but they are still available.
5639 Here is a small example that works on GNU/Linux:
5641 (define libc-obj (dynamic-link "libc.so"))
5642 (dynamic-args-call 'rand libc-obj '())
5644 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5646 ** The #/ syntax for module names is depreciated, and will be removed
5647 in a future version of Guile. Instead of
5655 The latter syntax is more consistent with existing Lisp practice.
5657 ** Guile now does fancier printing of structures. Structures are the
5658 underlying implementation for records, which in turn are used to
5659 implement modules, so all of these object now print differently and in
5660 a more informative way.
5662 The Scheme printer will examine the builtin variable *struct-printer*
5663 whenever it needs to print a structure object. When this variable is
5664 not `#f' it is deemed to be a procedure and will be applied to the
5665 structure object and the output port. When *struct-printer* is `#f'
5666 or the procedure return `#f' the structure object will be printed in
5667 the boring #<struct 80458270> form.
5669 This hook is used by some routines in ice-9/boot-9.scm to implement
5670 type specific printing routines. Please read the comments there about
5673 One of the more specific uses of structs are records. The printing
5674 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5675 called. It should behave like a *struct-printer* procedure (described
5678 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5679 token of the form #:NAME, where NAME has the same syntax as a Scheme
5680 symbol, is the external representation of the keyword named NAME.
5681 Keyword objects print using this syntax as well, so values containing
5682 keyword objects can be read back into Guile. When used in an
5683 expression, keywords are self-quoting objects.
5685 Guile suports this read syntax, and uses this print syntax, regardless
5686 of the current setting of the `keyword' read option. The `keyword'
5687 read option only controls whether Guile recognizes the `:NAME' syntax,
5688 which is incompatible with R4RS. (R4RS says such token represent
5691 ** Guile has regular expression support again. Guile 1.0 included
5692 functions for matching regular expressions, based on the Rx library.
5693 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5694 distribution, and thus Guile had no regular expression support. Guile
5695 1.2 again supports the most commonly used functions, and supports all
5696 of SCSH's regular expression functions.
5698 If your system does not include a POSIX regular expression library,
5699 and you have not linked Guile with a third-party regexp library such as
5700 Rx, these functions will not be available. You can tell whether your
5701 Guile installation includes regular expression support by checking
5702 whether the `*features*' list includes the `regex' symbol.
5704 *** regexp functions
5706 By default, Guile supports POSIX extended regular expressions. That
5707 means that the characters `(', `)', `+' and `?' are special, and must
5708 be escaped if you wish to match the literal characters.
5710 This regular expression interface was modeled after that implemented
5711 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5712 with SCSH regular expressions.
5714 **** Function: string-match PATTERN STR [START]
5715 Compile the string PATTERN into a regular expression and compare
5716 it with STR. The optional numeric argument START specifies the
5717 position of STR at which to begin matching.
5719 `string-match' returns a "match structure" which describes what,
5720 if anything, was matched by the regular expression. *Note Match
5721 Structures::. If STR does not match PATTERN at all,
5722 `string-match' returns `#f'.
5724 Each time `string-match' is called, it must compile its PATTERN
5725 argument into a regular expression structure. This operation is
5726 expensive, which makes `string-match' inefficient if the same regular
5727 expression is used several times (for example, in a loop). For better
5728 performance, you can compile a regular expression in advance and then
5729 match strings against the compiled regexp.
5731 **** Function: make-regexp STR [FLAGS]
5732 Compile the regular expression described by STR, and return the
5733 compiled regexp structure. If STR does not describe a legal
5734 regular expression, `make-regexp' throws a
5735 `regular-expression-syntax' error.
5737 FLAGS may be the bitwise-or of one or more of the following:
5739 **** Constant: regexp/extended
5740 Use POSIX Extended Regular Expression syntax when interpreting
5741 STR. If not set, POSIX Basic Regular Expression syntax is used.
5742 If the FLAGS argument is omitted, we assume regexp/extended.
5744 **** Constant: regexp/icase
5745 Do not differentiate case. Subsequent searches using the
5746 returned regular expression will be case insensitive.
5748 **** Constant: regexp/newline
5749 Match-any-character operators don't match a newline.
5751 A non-matching list ([^...]) not containing a newline matches a
5754 Match-beginning-of-line operator (^) matches the empty string
5755 immediately after a newline, regardless of whether the FLAGS
5756 passed to regexp-exec contain regexp/notbol.
5758 Match-end-of-line operator ($) matches the empty string
5759 immediately before a newline, regardless of whether the FLAGS
5760 passed to regexp-exec contain regexp/noteol.
5762 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5763 Match the compiled regular expression REGEXP against `str'. If
5764 the optional integer START argument is provided, begin matching
5765 from that position in the string. Return a match structure
5766 describing the results of the match, or `#f' if no match could be
5769 FLAGS may be the bitwise-or of one or more of the following:
5771 **** Constant: regexp/notbol
5772 The match-beginning-of-line operator always fails to match (but
5773 see the compilation flag regexp/newline above) This flag may be
5774 used when different portions of a string are passed to
5775 regexp-exec and the beginning of the string should not be
5776 interpreted as the beginning of the line.
5778 **** Constant: regexp/noteol
5779 The match-end-of-line operator always fails to match (but see the
5780 compilation flag regexp/newline above)
5782 **** Function: regexp? OBJ
5783 Return `#t' if OBJ is a compiled regular expression, or `#f'
5786 Regular expressions are commonly used to find patterns in one string
5787 and replace them with the contents of another string.
5789 **** Function: regexp-substitute PORT MATCH [ITEM...]
5790 Write to the output port PORT selected contents of the match
5791 structure MATCH. Each ITEM specifies what should be written, and
5792 may be one of the following arguments:
5794 * A string. String arguments are written out verbatim.
5796 * An integer. The submatch with that number is written.
5798 * The symbol `pre'. The portion of the matched string preceding
5799 the regexp match is written.
5801 * The symbol `post'. The portion of the matched string
5802 following the regexp match is written.
5804 PORT may be `#f', in which case nothing is written; instead,
5805 `regexp-substitute' constructs a string from the specified ITEMs
5808 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5809 Similar to `regexp-substitute', but can be used to perform global
5810 substitutions on STR. Instead of taking a match structure as an
5811 argument, `regexp-substitute/global' takes two string arguments: a
5812 REGEXP string describing a regular expression, and a TARGET string
5813 which should be matched against this regular expression.
5815 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5818 * A function may be supplied. When this function is called, it
5819 will be passed one argument: a match structure for a given
5820 regular expression match. It should return a string to be
5821 written out to PORT.
5823 * The `post' symbol causes `regexp-substitute/global' to recurse
5824 on the unmatched portion of STR. This *must* be supplied in
5825 order to perform global search-and-replace on STR; if it is
5826 not present among the ITEMs, then `regexp-substitute/global'
5827 will return after processing a single match.
5829 *** Match Structures
5831 A "match structure" is the object returned by `string-match' and
5832 `regexp-exec'. It describes which portion of a string, if any, matched
5833 the given regular expression. Match structures include: a reference to
5834 the string that was checked for matches; the starting and ending
5835 positions of the regexp match; and, if the regexp included any
5836 parenthesized subexpressions, the starting and ending positions of each
5839 In each of the regexp match functions described below, the `match'
5840 argument must be a match structure returned by a previous call to
5841 `string-match' or `regexp-exec'. Most of these functions return some
5842 information about the original target string that was matched against a
5843 regular expression; we will call that string TARGET for easy reference.
5845 **** Function: regexp-match? OBJ
5846 Return `#t' if OBJ is a match structure returned by a previous
5847 call to `regexp-exec', or `#f' otherwise.
5849 **** Function: match:substring MATCH [N]
5850 Return the portion of TARGET matched by subexpression number N.
5851 Submatch 0 (the default) represents the entire regexp match. If
5852 the regular expression as a whole matched, but the subexpression
5853 number N did not match, return `#f'.
5855 **** Function: match:start MATCH [N]
5856 Return the starting position of submatch number N.
5858 **** Function: match:end MATCH [N]
5859 Return the ending position of submatch number N.
5861 **** Function: match:prefix MATCH
5862 Return the unmatched portion of TARGET preceding the regexp match.
5864 **** Function: match:suffix MATCH
5865 Return the unmatched portion of TARGET following the regexp match.
5867 **** Function: match:count MATCH
5868 Return the number of parenthesized subexpressions from MATCH.
5869 Note that the entire regular expression match itself counts as a
5870 subexpression, and failed submatches are included in the count.
5872 **** Function: match:string MATCH
5873 Return the original TARGET string.
5875 *** Backslash Escapes
5877 Sometimes you will want a regexp to match characters like `*' or `$'
5878 exactly. For example, to check whether a particular string represents
5879 a menu entry from an Info node, it would be useful to match it against
5880 a regexp like `^* [^:]*::'. However, this won't work; because the
5881 asterisk is a metacharacter, it won't match the `*' at the beginning of
5882 the string. In this case, we want to make the first asterisk un-magic.
5884 You can do this by preceding the metacharacter with a backslash
5885 character `\'. (This is also called "quoting" the metacharacter, and
5886 is known as a "backslash escape".) When Guile sees a backslash in a
5887 regular expression, it considers the following glyph to be an ordinary
5888 character, no matter what special meaning it would ordinarily have.
5889 Therefore, we can make the above example work by changing the regexp to
5890 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5891 to match only a single asterisk in the target string.
5893 Since the backslash is itself a metacharacter, you may force a
5894 regexp to match a backslash in the target string by preceding the
5895 backslash with itself. For example, to find variable references in a
5896 TeX program, you might want to find occurrences of the string `\let\'
5897 followed by any number of alphabetic characters. The regular expression
5898 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5899 each match a single backslash in the target string.
5901 **** Function: regexp-quote STR
5902 Quote each special character found in STR with a backslash, and
5903 return the resulting string.
5905 *Very important:* Using backslash escapes in Guile source code (as
5906 in Emacs Lisp or C) can be tricky, because the backslash character has
5907 special meaning for the Guile reader. For example, if Guile encounters
5908 the character sequence `\n' in the middle of a string while processing
5909 Scheme code, it replaces those characters with a newline character.
5910 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5911 Several of these "escape sequences" are processed by the Guile reader
5912 before your code is executed. Unrecognized escape sequences are
5913 ignored: if the characters `\*' appear in a string, they will be
5914 translated to the single character `*'.
5916 This translation is obviously undesirable for regular expressions,
5917 since we want to be able to include backslashes in a string in order to
5918 escape regexp metacharacters. Therefore, to make sure that a backslash
5919 is preserved in a string in your Guile program, you must use *two*
5920 consecutive backslashes:
5922 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5924 The string in this example is preprocessed by the Guile reader before
5925 any code is executed. The resulting argument to `make-regexp' is the
5926 string `^\* [^:]*', which is what we really want.
5928 This also means that in order to write a regular expression that
5929 matches a single backslash character, the regular expression string in
5930 the source code must include *four* backslashes. Each consecutive pair
5931 of backslashes gets translated by the Guile reader to a single
5932 backslash, and the resulting double-backslash is interpreted by the
5933 regexp engine as matching a single backslash character. Hence:
5935 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5937 The reason for the unwieldiness of this syntax is historical. Both
5938 regular expression pattern matchers and Unix string processing systems
5939 have traditionally used backslashes with the special meanings described
5940 above. The POSIX regular expression specification and ANSI C standard
5941 both require these semantics. Attempting to abandon either convention
5942 would cause other kinds of compatibility problems, possibly more severe
5943 ones. Therefore, without extending the Scheme reader to support
5944 strings with different quoting conventions (an ungainly and confusing
5945 extension when implemented in other languages), we must adhere to this
5946 cumbersome escape syntax.
5948 * Changes to the gh_ interface
5950 * Changes to the scm_ interface
5952 * Changes to system call interfaces:
5954 ** The value returned by `raise' is now unspecified. It throws an exception
5957 *** A new procedure `sigaction' can be used to install signal handlers
5959 (sigaction signum [action] [flags])
5961 signum is the signal number, which can be specified using the value
5964 If action is omitted, sigaction returns a pair: the CAR is the current
5965 signal hander, which will be either an integer with the value SIG_DFL
5966 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5967 handles the signal, or #f if a non-Scheme procedure handles the
5968 signal. The CDR contains the current sigaction flags for the handler.
5970 If action is provided, it is installed as the new handler for signum.
5971 action can be a Scheme procedure taking one argument, or the value of
5972 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5973 whatever signal handler was installed before sigaction was first used.
5974 Flags can optionally be specified for the new handler (SA_RESTART is
5975 always used if the system provides it, so need not be specified.) The
5976 return value is a pair with information about the old handler as
5979 This interface does not provide access to the "signal blocking"
5980 facility. Maybe this is not needed, since the thread support may
5981 provide solutions to the problem of consistent access to data
5984 *** A new procedure `flush-all-ports' is equivalent to running
5985 `force-output' on every port open for output.
5987 ** Guile now provides information on how it was built, via the new
5988 global variable, %guile-build-info. This variable records the values
5989 of the standard GNU makefile directory variables as an assocation
5990 list, mapping variable names (symbols) onto directory paths (strings).
5991 For example, to find out where the Guile link libraries were
5992 installed, you can say:
5994 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5997 * Changes to the scm_ interface
5999 ** The new function scm_handle_by_message_noexit is just like the
6000 existing scm_handle_by_message function, except that it doesn't call
6001 exit to terminate the process. Instead, it prints a message and just
6002 returns #f. This might be a more appropriate catch-all handler for
6003 new dynamic roots and threads.
6006 Changes in Guile 1.1 (released Friday, May 16 1997):
6008 * Changes to the distribution.
6010 The Guile 1.0 distribution has been split up into several smaller
6012 guile-core --- the Guile interpreter itself.
6013 guile-tcltk --- the interface between the Guile interpreter and
6014 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6015 is a toolkit for building graphical user interfaces.
6016 guile-rgx-ctax --- the interface between Guile and the Rx regular
6017 expression matcher, and the translator for the Ctax
6018 programming language. These are packaged together because the
6019 Ctax translator uses Rx to parse Ctax source code.
6021 This NEWS file describes the changes made to guile-core since the 1.0
6024 We no longer distribute the documentation, since it was either out of
6025 date, or incomplete. As soon as we have current documentation, we
6030 * Changes to the stand-alone interpreter
6032 ** guile now accepts command-line arguments compatible with SCSH, Olin
6033 Shivers' Scheme Shell.
6035 In general, arguments are evaluated from left to right, but there are
6036 exceptions. The following switches stop argument processing, and
6037 stash all remaining command-line arguments as the value returned by
6038 the (command-line) function.
6039 -s SCRIPT load Scheme source code from FILE, and exit
6040 -c EXPR evalute Scheme expression EXPR, and exit
6041 -- stop scanning arguments; run interactively
6043 The switches below are processed as they are encountered.
6044 -l FILE load Scheme source code from FILE
6045 -e FUNCTION after reading script, apply FUNCTION to
6046 command line arguments
6047 -ds do -s script at this point
6048 --emacs enable Emacs protocol (experimental)
6049 -h, --help display this help and exit
6050 -v, --version display version information and exit
6051 \ read arguments from following script lines
6053 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6054 which re-implements the traditional "echo" command:
6056 #!/usr/local/bin/guile -s
6059 (map (lambda (arg) (display arg) (display " "))
6063 (main (command-line))
6065 Suppose we invoke this script as follows:
6067 ekko a speckled gecko
6069 Through the magic of Unix script processing (triggered by the `#!'
6070 token at the top of the file), /usr/local/bin/guile receives the
6071 following list of command-line arguments:
6073 ("-s" "./ekko" "a" "speckled" "gecko")
6075 Unix inserts the name of the script after the argument specified on
6076 the first line of the file (in this case, "-s"), and then follows that
6077 with the arguments given to the script. Guile loads the script, which
6078 defines the `main' function, and then applies it to the list of
6079 remaining command-line arguments, ("a" "speckled" "gecko").
6081 In Unix, the first line of a script file must take the following form:
6083 #!INTERPRETER ARGUMENT
6085 where INTERPRETER is the absolute filename of the interpreter
6086 executable, and ARGUMENT is a single command-line argument to pass to
6089 You may only pass one argument to the interpreter, and its length is
6090 limited. These restrictions can be annoying to work around, so Guile
6091 provides a general mechanism (borrowed from, and compatible with,
6092 SCSH) for circumventing them.
6094 If the ARGUMENT in a Guile script is a single backslash character,
6095 `\', Guile will open the script file, parse arguments from its second
6096 and subsequent lines, and replace the `\' with them. So, for example,
6097 here is another implementation of the `ekko' script:
6099 #!/usr/local/bin/guile \
6103 (for-each (lambda (arg) (display arg) (display " "))
6107 If the user invokes this script as follows:
6109 ekko a speckled gecko
6111 Unix expands this into
6113 /usr/local/bin/guile \ ekko a speckled gecko
6115 When Guile sees the `\' argument, it replaces it with the arguments
6116 read from the second line of the script, producing:
6118 /usr/local/bin/guile -e main -s ekko a speckled gecko
6120 This tells Guile to load the `ekko' script, and apply the function
6121 `main' to the argument list ("a" "speckled" "gecko").
6123 Here is how Guile parses the command-line arguments:
6124 - Each space character terminates an argument. This means that two
6125 spaces in a row introduce an empty-string argument.
6126 - The tab character is not permitted (unless you quote it with the
6127 backslash character, as described below), to avoid confusion.
6128 - The newline character terminates the sequence of arguments, and will
6129 also terminate a final non-empty argument. (However, a newline
6130 following a space will not introduce a final empty-string argument;
6131 it only terminates the argument list.)
6132 - The backslash character is the escape character. It escapes
6133 backslash, space, tab, and newline. The ANSI C escape sequences
6134 like \n and \t are also supported. These produce argument
6135 constituents; the two-character combination \n doesn't act like a
6136 terminating newline. The escape sequence \NNN for exactly three
6137 octal digits reads as the character whose ASCII code is NNN. As
6138 above, characters produced this way are argument constituents.
6139 Backslash followed by other characters is not allowed.
6141 * Changes to the procedure for linking libguile with your programs
6143 ** Guile now builds and installs a shared guile library, if your
6144 system support shared libraries. (It still builds a static library on
6145 all systems.) Guile automatically detects whether your system
6146 supports shared libraries. To prevent Guile from buildisg shared
6147 libraries, pass the `--disable-shared' flag to the configure script.
6149 Guile takes longer to compile when it builds shared libraries, because
6150 it must compile every file twice --- once to produce position-
6151 independent object code, and once to produce normal object code.
6153 ** The libthreads library has been merged into libguile.
6155 To link a program against Guile, you now need only link against
6156 -lguile and -lqt; -lthreads is no longer needed. If you are using
6157 autoconf to generate configuration scripts for your application, the
6158 following lines should suffice to add the appropriate libraries to
6161 ### Find quickthreads and libguile.
6162 AC_CHECK_LIB(qt, main)
6163 AC_CHECK_LIB(guile, scm_shell)
6165 * Changes to Scheme functions
6167 ** Guile Scheme's special syntax for keyword objects is now optional,
6168 and disabled by default.
6170 The syntax variation from R4RS made it difficult to port some
6171 interesting packages to Guile. The routines which accepted keyword
6172 arguments (mostly in the module system) have been modified to also
6173 accept symbols whose names begin with `:'.
6175 To change the keyword syntax, you must first import the (ice-9 debug)
6177 (use-modules (ice-9 debug))
6179 Then you can enable the keyword syntax as follows:
6180 (read-set! keywords 'prefix)
6182 To disable keyword syntax, do this:
6183 (read-set! keywords #f)
6185 ** Many more primitive functions accept shared substrings as
6186 arguments. In the past, these functions required normal, mutable
6187 strings as arguments, although they never made use of this
6190 ** The uniform array functions now operate on byte vectors. These
6191 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6192 `serial-array-map', `array-map', `array-for-each', and
6195 ** The new functions `trace' and `untrace' implement simple debugging
6196 support for Scheme functions.
6198 The `trace' function accepts any number of procedures as arguments,
6199 and tells the Guile interpreter to display each procedure's name and
6200 arguments each time the procedure is invoked. When invoked with no
6201 arguments, `trace' returns the list of procedures currently being
6204 The `untrace' function accepts any number of procedures as arguments,
6205 and tells the Guile interpreter not to trace them any more. When
6206 invoked with no arguments, `untrace' untraces all curretly traced
6209 The tracing in Guile has an advantage over most other systems: we
6210 don't create new procedure objects, but mark the procedure objects
6211 themselves. This means that anonymous and internal procedures can be
6214 ** The function `assert-repl-prompt' has been renamed to
6215 `set-repl-prompt!'. It takes one argument, PROMPT.
6216 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6217 - If PROMPT is a string, we use it as a prompt.
6218 - If PROMPT is a procedure accepting no arguments, we call it, and
6219 display the result as a prompt.
6220 - Otherwise, we display "> ".
6222 ** The new function `eval-string' reads Scheme expressions from a
6223 string and evaluates them, returning the value of the last expression
6224 in the string. If the string contains no expressions, it returns an
6227 ** The new function `thunk?' returns true iff its argument is a
6228 procedure of zero arguments.
6230 ** `defined?' is now a builtin function, instead of syntax. This
6231 means that its argument should be quoted. It returns #t iff its
6232 argument is bound in the current module.
6234 ** The new syntax `use-modules' allows you to add new modules to your
6235 environment without re-typing a complete `define-module' form. It
6236 accepts any number of module names as arguments, and imports their
6237 public bindings into the current module.
6239 ** The new function (module-defined? NAME MODULE) returns true iff
6240 NAME, a symbol, is defined in MODULE, a module object.
6242 ** The new function `builtin-bindings' creates and returns a hash
6243 table containing copies of all the root module's bindings.
6245 ** The new function `builtin-weak-bindings' does the same as
6246 `builtin-bindings', but creates a doubly-weak hash table.
6248 ** The `equal?' function now considers variable objects to be
6249 equivalent if they have the same name and the same value.
6251 ** The new function `command-line' returns the command-line arguments
6252 given to Guile, as a list of strings.
6254 When using guile as a script interpreter, `command-line' returns the
6255 script's arguments; those processed by the interpreter (like `-s' or
6256 `-c') are omitted. (In other words, you get the normal, expected
6257 behavior.) Any application that uses scm_shell to process its
6258 command-line arguments gets this behavior as well.
6260 ** The new function `load-user-init' looks for a file called `.guile'
6261 in the user's home directory, and loads it if it exists. This is
6262 mostly for use by the code generated by scm_compile_shell_switches,
6263 but we thought it might also be useful in other circumstances.
6265 ** The new function `log10' returns the base-10 logarithm of its
6268 ** Changes to I/O functions
6270 *** The functions `read', `primitive-load', `read-and-eval!', and
6271 `primitive-load-path' no longer take optional arguments controlling
6272 case insensitivity and a `#' parser.
6274 Case sensitivity is now controlled by a read option called
6275 `case-insensitive'. The user can add new `#' syntaxes with the
6276 `read-hash-extend' function (see below).
6278 *** The new function `read-hash-extend' allows the user to change the
6279 syntax of Guile Scheme in a somewhat controlled way.
6281 (read-hash-extend CHAR PROC)
6282 When parsing S-expressions, if we read a `#' character followed by
6283 the character CHAR, use PROC to parse an object from the stream.
6284 If PROC is #f, remove any parsing procedure registered for CHAR.
6286 The reader applies PROC to two arguments: CHAR and an input port.
6288 *** The new functions read-delimited and read-delimited! provide a
6289 general mechanism for doing delimited input on streams.
6291 (read-delimited DELIMS [PORT HANDLE-DELIM])
6292 Read until we encounter one of the characters in DELIMS (a string),
6293 or end-of-file. PORT is the input port to read from; it defaults to
6294 the current input port. The HANDLE-DELIM parameter determines how
6295 the terminating character is handled; it should be one of the
6298 'trim omit delimiter from result
6299 'peek leave delimiter character in input stream
6300 'concat append delimiter character to returned value
6301 'split return a pair: (RESULT . TERMINATOR)
6303 HANDLE-DELIM defaults to 'peek.
6305 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6306 A side-effecting variant of `read-delimited'.
6308 The data is written into the string BUF at the indices in the
6309 half-open interval [START, END); the default interval is the whole
6310 string: START = 0 and END = (string-length BUF). The values of
6311 START and END must specify a well-defined interval in BUF, i.e.
6312 0 <= START <= END <= (string-length BUF).
6314 It returns NBYTES, the number of bytes read. If the buffer filled
6315 up without a delimiter character being found, it returns #f. If the
6316 port is at EOF when the read starts, it returns the EOF object.
6318 If an integer is returned (i.e., the read is successfully terminated
6319 by reading a delimiter character), then the HANDLE-DELIM parameter
6320 determines how to handle the terminating character. It is described
6321 above, and defaults to 'peek.
6323 (The descriptions of these functions were borrowed from the SCSH
6324 manual, by Olin Shivers and Brian Carlstrom.)
6326 *** The `%read-delimited!' function is the primitive used to implement
6327 `read-delimited' and `read-delimited!'.
6329 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6331 This returns a pair of values: (TERMINATOR . NUM-READ).
6332 - TERMINATOR describes why the read was terminated. If it is a
6333 character or the eof object, then that is the value that terminated
6334 the read. If it is #f, the function filled the buffer without finding
6335 a delimiting character.
6336 - NUM-READ is the number of characters read into BUF.
6338 If the read is successfully terminated by reading a delimiter
6339 character, then the gobble? parameter determines what to do with the
6340 terminating character. If true, the character is removed from the
6341 input stream; if false, the character is left in the input stream
6342 where a subsequent read operation will retrieve it. In either case,
6343 the character is also the first value returned by the procedure call.
6345 (The descriptions of this function was borrowed from the SCSH manual,
6346 by Olin Shivers and Brian Carlstrom.)
6348 *** The `read-line' and `read-line!' functions have changed; they now
6349 trim the terminator by default; previously they appended it to the
6350 returned string. For the old behavior, use (read-line PORT 'concat).
6352 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6353 take new optional START and END arguments, specifying the region of
6354 the array to read and write.
6356 *** The `ungetc-char-ready?' function has been removed. We feel it's
6357 inappropriate for an interface to expose implementation details this
6360 ** Changes to the Unix library and system call interface
6362 *** The new fcntl function provides access to the Unix `fcntl' system
6365 (fcntl PORT COMMAND VALUE)
6366 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6367 Values for COMMAND are:
6369 F_DUPFD duplicate a file descriptor
6370 F_GETFD read the descriptor's close-on-exec flag
6371 F_SETFD set the descriptor's close-on-exec flag to VALUE
6372 F_GETFL read the descriptor's flags, as set on open
6373 F_SETFL set the descriptor's flags, as set on open to VALUE
6374 F_GETOWN return the process ID of a socket's owner, for SIGIO
6375 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6376 FD_CLOEXEC not sure what this is
6378 For details, see the documentation for the fcntl system call.
6380 *** The arguments to `select' have changed, for compatibility with
6381 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6382 expected behavior. The MILLISECONDS parameter has been changed to
6383 MICROSECONDS, to more closely resemble the underlying system call.
6384 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6385 corresponding return set will be the same.
6387 *** The arguments to the `mknod' system call have changed. They are
6390 (mknod PATH TYPE PERMS DEV)
6391 Create a new file (`node') in the file system. PATH is the name of
6392 the file to create. TYPE is the kind of file to create; it should
6393 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6394 permission bits to give the newly created file. If TYPE is
6395 'block-special or 'char-special, DEV specifies which device the
6396 special file refers to; its interpretation depends on the kind of
6397 special file being created.
6399 *** The `fork' function has been renamed to `primitive-fork', to avoid
6400 clashing with various SCSH forks.
6402 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6403 and `recvfrom!'. They no longer accept a size for a second argument;
6404 you must pass a string to hold the received value. They no longer
6405 return the buffer. Instead, `recv' returns the length of the message
6406 received, and `recvfrom' returns a pair containing the packet's length
6407 and originating address.
6409 *** The file descriptor datatype has been removed, as have the
6410 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6411 We plan to replace these functions with a SCSH-compatible interface.
6413 *** The `create' function has been removed; it's just a special case
6416 *** There are new functions to break down process termination status
6417 values. In the descriptions below, STATUS is a value returned by
6420 (status:exit-val STATUS)
6421 If the child process exited normally, this function returns the exit
6422 code for the child process (i.e., the value passed to exit, or
6423 returned from main). If the child process did not exit normally,
6424 this function returns #f.
6426 (status:stop-sig STATUS)
6427 If the child process was suspended by a signal, this function
6428 returns the signal that suspended the child. Otherwise, it returns
6431 (status:term-sig STATUS)
6432 If the child process terminated abnormally, this function returns
6433 the signal that terminated the child. Otherwise, this function
6436 POSIX promises that exactly one of these functions will return true on
6437 a valid STATUS value.
6439 These functions are compatible with SCSH.
6441 *** There are new accessors and setters for the broken-out time vectors
6442 returned by `localtime', `gmtime', and that ilk. They are:
6444 Component Accessor Setter
6445 ========================= ============ ============
6446 seconds tm:sec set-tm:sec
6447 minutes tm:min set-tm:min
6448 hours tm:hour set-tm:hour
6449 day of the month tm:mday set-tm:mday
6450 month tm:mon set-tm:mon
6451 year tm:year set-tm:year
6452 day of the week tm:wday set-tm:wday
6453 day in the year tm:yday set-tm:yday
6454 daylight saving time tm:isdst set-tm:isdst
6455 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6456 name of time zone tm:zone set-tm:zone
6458 *** There are new accessors for the vectors returned by `uname',
6459 describing the host system:
6462 ============================================== ================
6463 name of the operating system implementation utsname:sysname
6464 network name of this machine utsname:nodename
6465 release level of the operating system utsname:release
6466 version level of the operating system utsname:version
6467 machine hardware platform utsname:machine
6469 *** There are new accessors for the vectors returned by `getpw',
6470 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6471 system's user database:
6474 ====================== =================
6475 user name passwd:name
6476 user password passwd:passwd
6479 real name passwd:gecos
6480 home directory passwd:dir
6481 shell program passwd:shell
6483 *** There are new accessors for the vectors returned by `getgr',
6484 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6485 system's group database:
6488 ======================= ============
6489 group name group:name
6490 group password group:passwd
6492 group members group:mem
6494 *** There are new accessors for the vectors returned by `gethost',
6495 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6499 ========================= ===============
6500 official name of host hostent:name
6501 alias list hostent:aliases
6502 host address type hostent:addrtype
6503 length of address hostent:length
6504 list of addresses hostent:addr-list
6506 *** There are new accessors for the vectors returned by `getnet',
6507 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6511 ========================= ===============
6512 official name of net netent:name
6513 alias list netent:aliases
6514 net number type netent:addrtype
6515 net number netent:net
6517 *** There are new accessors for the vectors returned by `getproto',
6518 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6522 ========================= ===============
6523 official protocol name protoent:name
6524 alias list protoent:aliases
6525 protocol number protoent:proto
6527 *** There are new accessors for the vectors returned by `getserv',
6528 `getservbyname', `getservbyport', and `getservent', describing
6532 ========================= ===============
6533 official service name servent:name
6534 alias list servent:aliases
6535 port number servent:port
6536 protocol to use servent:proto
6538 *** There are new accessors for the sockaddr structures returned by
6539 `accept', `getsockname', `getpeername', `recvfrom!':
6542 ======================================== ===============
6543 address format (`family') sockaddr:fam
6544 path, for file domain addresses sockaddr:path
6545 address, for internet domain addresses sockaddr:addr
6546 TCP or UDP port, for internet sockaddr:port
6548 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6549 `getprotoent', and `getservent' functions now return #f at the end of
6550 the user database. (They used to throw an exception.)
6552 Note that calling MUMBLEent function is equivalent to calling the
6553 corresponding MUMBLE function with no arguments.
6555 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6556 `setprotoent', and `setservent' routines now take no arguments.
6558 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6559 provide more useful information when they throw an exception.
6561 *** The `lnaof' function has been renamed to `inet-lnaof'.
6563 *** Guile now claims to have the `current-time' feature.
6565 *** The `mktime' function now takes an optional second argument ZONE,
6566 giving the time zone to use for the conversion. ZONE should be a
6567 string, in the same format as expected for the "TZ" environment variable.
6569 *** The `strptime' function now returns a pair (TIME . COUNT), where
6570 TIME is the parsed time as a vector, and COUNT is the number of
6571 characters from the string left unparsed. This function used to
6572 return the remaining characters as a string.
6574 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6575 The return value is now (SECONDS . MICROSECONDS); the fractional
6576 component is no longer expressed in "ticks".
6578 *** The `ticks/sec' constant has been removed, in light of the above change.
6580 * Changes to the gh_ interface
6582 ** gh_eval_str() now returns an SCM object which is the result of the
6585 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6588 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6589 and returns the array
6591 ** gh_scm2str0() is gone: there is no need to distinguish
6592 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6593 the user to interpret the data both ways.
6595 * Changes to the scm_ interface
6597 ** The new function scm_symbol_value0 provides an easy way to get a
6598 symbol's value from C code:
6600 SCM scm_symbol_value0 (char *NAME)
6601 Return the value of the symbol named by the null-terminated string
6602 NAME in the current module. If the symbol named NAME is unbound in
6603 the current module, return SCM_UNDEFINED.
6605 ** The new function scm_sysintern0 creates new top-level variables,
6606 without assigning them a value.
6608 SCM scm_sysintern0 (char *NAME)
6609 Create a new Scheme top-level variable named NAME. NAME is a
6610 null-terminated string. Return the variable's value cell.
6612 ** The function scm_internal_catch is the guts of catch. It handles
6613 all the mechanics of setting up a catch target, invoking the catch
6614 body, and perhaps invoking the handler if the body does a throw.
6616 The function is designed to be usable from C code, but is general
6617 enough to implement all the semantics Guile Scheme expects from throw.
6619 TAG is the catch tag. Typically, this is a symbol, but this function
6620 doesn't actually care about that.
6622 BODY is a pointer to a C function which runs the body of the catch;
6623 this is the code you can throw from. We call it like this:
6624 BODY (BODY_DATA, JMPBUF)
6626 BODY_DATA is just the BODY_DATA argument we received; we pass it
6627 through to BODY as its first argument. The caller can make
6628 BODY_DATA point to anything useful that BODY might need.
6629 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6630 which we have just created and initialized.
6632 HANDLER is a pointer to a C function to deal with a throw to TAG,
6633 should one occur. We call it like this:
6634 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6636 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6637 same idea as BODY_DATA above.
6638 THROWN_TAG is the tag that the user threw to; usually this is
6639 TAG, but it could be something else if TAG was #t (i.e., a
6640 catch-all), or the user threw to a jmpbuf.
6641 THROW_ARGS is the list of arguments the user passed to the THROW
6644 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6645 is just a pointer we pass through to HANDLER. We don't actually
6646 use either of those pointers otherwise ourselves. The idea is
6647 that, if our caller wants to communicate something to BODY or
6648 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6649 HANDLER can then use. Think of it as a way to make BODY and
6650 HANDLER closures, not just functions; MUMBLE_DATA points to the
6653 Of course, it's up to the caller to make sure that any data a
6654 MUMBLE_DATA needs is protected from GC. A common way to do this is
6655 to make MUMBLE_DATA a pointer to data stored in an automatic
6656 structure variable; since the collector must scan the stack for
6657 references anyway, this assures that any references in MUMBLE_DATA
6660 ** The new function scm_internal_lazy_catch is exactly like
6661 scm_internal_catch, except:
6663 - It does not unwind the stack (this is the major difference).
6664 - If handler returns, its value is returned from the throw.
6665 - BODY always receives #f as its JMPBUF argument (since there's no
6666 jmpbuf associated with a lazy catch, because we don't unwind the
6669 ** scm_body_thunk is a new body function you can pass to
6670 scm_internal_catch if you want the body to be like Scheme's `catch'
6671 --- a thunk, or a function of one argument if the tag is #f.
6673 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6674 contains the Scheme procedure to invoke as the body, and the tag
6675 we're catching. If the tag is #f, then we pass JMPBUF (created by
6676 scm_internal_catch) to the body procedure; otherwise, the body gets
6679 ** scm_handle_by_proc is a new handler function you can pass to
6680 scm_internal_catch if you want the handler to act like Scheme's catch
6681 --- call a procedure with the tag and the throw arguments.
6683 If the user does a throw to this catch, this function runs a handler
6684 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6685 variable holding the Scheme procedure object to invoke. It ought to
6686 be a pointer to an automatic variable (i.e., one living on the stack),
6687 or the procedure object should be otherwise protected from GC.
6689 ** scm_handle_by_message is a new handler function to use with
6690 `scm_internal_catch' if you want Guile to print a message and die.
6691 It's useful for dealing with throws to uncaught keys at the top level.
6693 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6694 message header to print; if zero, we use "guile" instead. That
6695 text is followed by a colon, then the message described by ARGS.
6697 ** The return type of scm_boot_guile is now void; the function does
6698 not return a value, and indeed, never returns at all.
6700 ** The new function scm_shell makes it easy for user applications to
6701 process command-line arguments in a way that is compatible with the
6702 stand-alone guile interpreter (which is in turn compatible with SCSH,
6705 To use the scm_shell function, first initialize any guile modules
6706 linked into your application, and then call scm_shell with the values
6707 of ARGC and ARGV your `main' function received. scm_shell will add
6708 any SCSH-style meta-arguments from the top of the script file to the
6709 argument vector, and then process the command-line arguments. This
6710 generally means loading a script file or starting up an interactive
6711 command interpreter. For details, see "Changes to the stand-alone
6714 ** The new functions scm_get_meta_args and scm_count_argv help you
6715 implement the SCSH-style meta-argument, `\'.
6717 char **scm_get_meta_args (int ARGC, char **ARGV)
6718 If the second element of ARGV is a string consisting of a single
6719 backslash character (i.e. "\\" in Scheme notation), open the file
6720 named by the following argument, parse arguments from it, and return
6721 the spliced command line. The returned array is terminated by a
6724 For details of argument parsing, see above, under "guile now accepts
6725 command-line arguments compatible with SCSH..."
6727 int scm_count_argv (char **ARGV)
6728 Count the arguments in ARGV, assuming it is terminated by a null
6731 For an example of how these functions might be used, see the source
6732 code for the function scm_shell in libguile/script.c.
6734 You will usually want to use scm_shell instead of calling this
6737 ** The new function scm_compile_shell_switches turns an array of
6738 command-line arguments into Scheme code to carry out the actions they
6739 describe. Given ARGC and ARGV, it returns a Scheme expression to
6740 evaluate, and calls scm_set_program_arguments to make any remaining
6741 command-line arguments available to the Scheme code. For example,
6742 given the following arguments:
6744 -e main -s ekko a speckled gecko
6746 scm_set_program_arguments will return the following expression:
6748 (begin (load "ekko") (main (command-line)) (quit))
6750 You will usually want to use scm_shell instead of calling this
6753 ** The function scm_shell_usage prints a usage message appropriate for
6754 an interpreter that uses scm_compile_shell_switches to handle its
6755 command-line arguments.
6757 void scm_shell_usage (int FATAL, char *MESSAGE)
6758 Print a usage message to the standard error output. If MESSAGE is
6759 non-zero, write it before the usage message, followed by a newline.
6760 If FATAL is non-zero, exit the process, using FATAL as the
6761 termination status. (If you want to be compatible with Guile,
6762 always use 1 as the exit status when terminating due to command-line
6765 You will usually want to use scm_shell instead of calling this
6768 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6769 expressions. It used to return SCM_EOL. Earth-shattering.
6771 ** The macros for declaring scheme objects in C code have been
6772 rearranged slightly. They are now:
6774 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6775 Declare a static SCM variable named C_NAME, and initialize it to
6776 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6777 be a C identifier, and SCHEME_NAME should be a C string.
6779 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6780 Just like SCM_SYMBOL, but make C_NAME globally visible.
6782 SCM_VCELL (C_NAME, SCHEME_NAME)
6783 Create a global variable at the Scheme level named SCHEME_NAME.
6784 Declare a static SCM variable named C_NAME, and initialize it to
6785 point to the Scheme variable's value cell.
6787 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6788 Just like SCM_VCELL, but make C_NAME globally visible.
6790 The `guile-snarf' script writes initialization code for these macros
6791 to its standard output, given C source code as input.
6793 The SCM_GLOBAL macro is gone.
6795 ** The scm_read_line and scm_read_line_x functions have been replaced
6796 by Scheme code based on the %read-delimited! procedure (known to C
6797 code as scm_read_delimited_x). See its description above for more
6800 ** The function scm_sys_open has been renamed to scm_open. It now
6801 returns a port instead of an FD object.
6803 * The dynamic linking support has changed. For more information, see
6804 libguile/DYNAMIC-LINKING.
6809 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6812 * Changes to the 'guile' program:
6814 ** Guile now loads some new files when it starts up. Guile first
6815 searches the load path for init.scm, and loads it if found. Then, if
6816 Guile is not being used to execute a script, and the user's home
6817 directory contains a file named `.guile', Guile loads that.
6819 ** You can now use Guile as a shell script interpreter.
6821 To paraphrase the SCSH manual:
6823 When Unix tries to execute an executable file whose first two
6824 characters are the `#!', it treats the file not as machine code to
6825 be directly executed by the native processor, but as source code
6826 to be executed by some interpreter. The interpreter to use is
6827 specified immediately after the #! sequence on the first line of
6828 the source file. The kernel reads in the name of the interpreter,
6829 and executes that instead. It passes the interpreter the source
6830 filename as its first argument, with the original arguments
6831 following. Consult the Unix man page for the `exec' system call
6832 for more information.
6834 Now you can use Guile as an interpreter, using a mechanism which is a
6835 compatible subset of that provided by SCSH.
6837 Guile now recognizes a '-s' command line switch, whose argument is the
6838 name of a file of Scheme code to load. It also treats the two
6839 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6840 to make a file of Scheme code directly executable by Unix, insert the
6841 following two lines at the top of the file:
6843 #!/usr/local/bin/guile -s
6846 Guile treats the argument of the `-s' command-line switch as the name
6847 of a file of Scheme code to load, and treats the sequence `#!' as the
6848 start of a block comment, terminated by `!#'.
6850 For example, here's a version of 'echo' written in Scheme:
6852 #!/usr/local/bin/guile -s
6854 (let loop ((args (cdr (program-arguments))))
6857 (display (car args))
6858 (if (pair? (cdr args))
6860 (loop (cdr args)))))
6863 Why does `#!' start a block comment terminated by `!#', instead of the
6864 end of the line? That is the notation SCSH uses, and although we
6865 don't yet support the other SCSH features that motivate that choice,
6866 we would like to be backward-compatible with any existing Guile
6867 scripts once we do. Furthermore, if the path to Guile on your system
6868 is too long for your kernel, you can start the script with this
6872 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6875 Note that some very old Unix systems don't support the `#!' syntax.
6878 ** You can now run Guile without installing it.
6880 Previous versions of the interactive Guile interpreter (`guile')
6881 couldn't start up unless Guile's Scheme library had been installed;
6882 they used the value of the environment variable `SCHEME_LOAD_PATH'
6883 later on in the startup process, but not to find the startup code
6884 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6887 To run Guile without installing it, build it in the normal way, and
6888 then set the environment variable `SCHEME_LOAD_PATH' to a
6889 colon-separated list of directories, including the top-level directory
6890 of the Guile sources. For example, if you unpacked Guile so that the
6891 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6894 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6897 ** Guile's read-eval-print loop no longer prints #<unspecified>
6898 results. If the user wants to see this, she can evaluate the
6899 expression (assert-repl-print-unspecified #t), perhaps in her startup
6902 ** Guile no longer shows backtraces by default when an error occurs;
6903 however, it does display a message saying how to get one, and how to
6904 request that they be displayed by default. After an error, evaluate
6906 to see a backtrace, and
6907 (debug-enable 'backtrace)
6908 to see them by default.
6912 * Changes to Guile Scheme:
6914 ** Guile now distinguishes between #f and the empty list.
6916 This is for compatibility with the IEEE standard, the (possibly)
6917 upcoming Revised^5 Report on Scheme, and many extant Scheme
6920 Guile used to have #f and '() denote the same object, to make Scheme's
6921 type system more compatible with Emacs Lisp's. However, the change
6922 caused too much trouble for Scheme programmers, and we found another
6923 way to reconcile Emacs Lisp with Scheme that didn't require this.
6926 ** Guile's delq, delv, delete functions, and their destructive
6927 counterparts, delq!, delv!, and delete!, now remove all matching
6928 elements from the list, not just the first. This matches the behavior
6929 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6930 functions which inspired them.
6932 I recognize that this change may break code in subtle ways, but it
6933 seems best to make the change before the FSF's first Guile release,
6937 ** The compiled-library-path function has been deleted from libguile.
6939 ** The facilities for loading Scheme source files have changed.
6941 *** The variable %load-path now tells Guile which directories to search
6942 for Scheme code. Its value is a list of strings, each of which names
6945 *** The variable %load-extensions now tells Guile which extensions to
6946 try appending to a filename when searching the load path. Its value
6947 is a list of strings. Its default value is ("" ".scm").
6949 *** (%search-load-path FILENAME) searches the directories listed in the
6950 value of the %load-path variable for a Scheme file named FILENAME,
6951 with all the extensions listed in %load-extensions. If it finds a
6952 match, then it returns its full filename. If FILENAME is absolute, it
6953 returns it unchanged. Otherwise, it returns #f.
6955 %search-load-path will not return matches that refer to directories.
6957 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6958 uses %seach-load-path to find a file named FILENAME, and loads it if
6959 it finds it. If it can't read FILENAME for any reason, it throws an
6962 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6965 *** load uses the same searching semantics as primitive-load.
6967 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6968 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6969 path, and load-module-with-path have been deleted. The functions
6970 above should serve their purposes.
6972 *** If the value of the variable %load-hook is a procedure,
6973 `primitive-load' applies its value to the name of the file being
6974 loaded (without the load path directory name prepended). If its value
6975 is #f, it is ignored. Otherwise, an error occurs.
6977 This is mostly useful for printing load notification messages.
6980 ** The function `eval!' is no longer accessible from the scheme level.
6981 We can't allow operations which introduce glocs into the scheme level,
6982 because Guile's type system can't handle these as data. Use `eval' or
6983 `read-and-eval!' (see below) as replacement.
6985 ** The new function read-and-eval! reads an expression from PORT,
6986 evaluates it, and returns the result. This is more efficient than
6987 simply calling `read' and `eval', since it is not necessary to make a
6988 copy of the expression for the evaluator to munge.
6990 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6991 for the `read' function.
6994 ** The function `int?' has been removed; its definition was identical
6995 to that of `integer?'.
6997 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6998 use the R4RS names for these functions.
7000 ** The function object-properties no longer returns the hash handle;
7001 it simply returns the object's property list.
7003 ** Many functions have been changed to throw errors, instead of
7004 returning #f on failure. The point of providing exception handling in
7005 the language is to simplify the logic of user code, but this is less
7006 useful if Guile's primitives don't throw exceptions.
7008 ** The function `fileno' has been renamed from `%fileno'.
7010 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7013 * Changes to Guile's C interface:
7015 ** The library's initialization procedure has been simplified.
7016 scm_boot_guile now has the prototype:
7018 void scm_boot_guile (int ARGC,
7020 void (*main_func) (),
7023 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7024 MAIN_FUNC should do all the work of the program (initializing other
7025 packages, reading user input, etc.) before returning. When MAIN_FUNC
7026 returns, call exit (0); this function never returns. If you want some
7027 other exit value, MAIN_FUNC may call exit itself.
7029 scm_boot_guile arranges for program-arguments to return the strings
7030 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7031 scm_set_program_arguments with the final list, so Scheme code will
7032 know which arguments have been processed.
7034 scm_boot_guile establishes a catch-all catch handler which prints an
7035 error message and exits the process. This means that Guile exits in a
7036 coherent way when system errors occur and the user isn't prepared to
7037 handle it. If the user doesn't like this behavior, they can establish
7038 their own universal catcher in MAIN_FUNC to shadow this one.
7040 Why must the caller do all the real work from MAIN_FUNC? The garbage
7041 collector assumes that all local variables of type SCM will be above
7042 scm_boot_guile's stack frame on the stack. If you try to manipulate
7043 SCM values after this function returns, it's the luck of the draw
7044 whether the GC will be able to find the objects you allocate. So,
7045 scm_boot_guile function exits, rather than returning, to discourage
7046 people from making that mistake.
7048 The IN, OUT, and ERR arguments were removed; there are other
7049 convenient ways to override these when desired.
7051 The RESULT argument was deleted; this function should never return.
7053 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7057 ** Guile's header files should no longer conflict with your system's
7060 In order to compile code which #included <libguile.h>, previous
7061 versions of Guile required you to add a directory containing all the
7062 Guile header files to your #include path. This was a problem, since
7063 Guile's header files have names which conflict with many systems'
7066 Now only <libguile.h> need appear in your #include path; you must
7067 refer to all Guile's other header files as <libguile/mumble.h>.
7068 Guile's installation procedure puts libguile.h in $(includedir), and
7069 the rest in $(includedir)/libguile.
7072 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7073 have been added to the Guile library.
7075 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7076 OBJ will not be freed, even if all other references are dropped,
7077 until someone does scm_unprotect_object (OBJ). Both functions
7080 Note that calls to scm_protect_object do not nest. You can call
7081 scm_protect_object any number of times on a given object, and the
7082 next call to scm_unprotect_object will unprotect it completely.
7084 Basically, scm_protect_object and scm_unprotect_object just
7085 maintain a list of references to things. Since the GC knows about
7086 this list, all objects it mentions stay alive. scm_protect_object
7087 adds its argument to the list; scm_unprotect_object remove its
7088 argument from the list.
7091 ** scm_eval_0str now returns the value of the last expression
7094 ** The new function scm_read_0str reads an s-expression from a
7095 null-terminated string, and returns it.
7097 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7098 to a Scheme port object.
7100 ** The new function `scm_set_program_arguments' allows C code to set
7101 the value returned by the Scheme `program-arguments' function.
7106 * Guile no longer includes sophisticated Tcl/Tk support.
7108 The old Tcl/Tk support was unsatisfying to us, because it required the
7109 user to link against the Tcl library, as well as Tk and Guile. The
7110 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7111 referring to widgets by names, rather than exporting widgets to Scheme
7112 code as a special datatype.
7114 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7115 maintainers described some very interesting changes in progress to the
7116 Tcl/Tk internals, which would facilitate clean interfaces between lone
7117 Tk and other interpreters --- even for garbage-collected languages
7118 like Scheme. They expected the new Tk to be publicly available in the
7121 Since it seems that Guile might soon have a new, cleaner interface to
7122 lone Tk, and that the old Guile/Tk glue code would probably need to be
7123 completely rewritten, we (Jim Blandy and Richard Stallman) have
7124 decided not to support the old code. We'll spend the time instead on
7125 a good interface to the newer Tk, as soon as it is available.
7127 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7130 Copyright information:
7132 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7134 Permission is granted to anyone to make or distribute verbatim copies
7135 of this document as received, in any medium, provided that the
7136 copyright notice and this permission notice are preserved,
7137 thus giving the recipient permission to redistribute in turn.
7139 Permission is granted to distribute modified versions
7140 of this document, or of portions of it,
7141 under the above conditions, provided also that they
7142 carry prominent notices stating who last changed them.
7147 paragraph-separate: "[
\f]*$"