1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org. Note that you
6 must be subscribed to this list first, in order to successfully send a
12 * New modules (see the manual for details)
14 ** The `(ice-9 i18n)' module provides internationalization support
16 * Changes to the distribution
18 ** Guile now uses Gnulib as a portability aid
20 * Changes to the stand-alone interpreter
21 * Changes to Scheme functions and syntax
23 ** A new 'memoize-symbol evaluator trap has been added. This trap can
24 be used for efficiently implementing a Scheme code coverage.
26 ** Duplicate bindings among used modules are resolved lazily.
27 This slightly improves program startup times.
29 * Changes to the C interface
31 ** Functions for handling `scm_option' now no longer require an argument
32 indicating length of the `scm_t_option' array.
35 Changes in 1.8.3 (since 1.8.2)
37 * New modules (see the manual for details)
44 ** Expressions like "(set! 'x #t)" no longer yield a crash
45 ** Build problems on Solaris fixed
47 * Implementation improvements
49 ** The reader is now faster, which reduces startup time
50 ** Procedures returned by `record-accessor' and `record-modifier' are faster
54 Changes in 1.8.2 (since 1.8.1):
56 * New procedures (see the manual for details)
58 ** set-program-arguments
63 ** Fractions were not `equal?' if stored in unreduced form.
64 (A subtle problem, since printing a value reduced it, making it work.)
65 ** srfi-60 `copy-bit' failed on 64-bit systems
66 ** "guile --use-srfi" option at the REPL can replace core functions
67 (Programs run with that option were ok, but in the interactive REPL
68 the core bindings got priority, preventing SRFI replacements or
70 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
71 ** `kill' on mingw throws an error for a PID other than oneself
72 ** Procedure names are attached to procedure-with-setters
73 ** Array read syntax works with negative lower bound
74 ** `array-in-bounds?' fix if an array has different lower bounds on each index
75 ** `*' returns exact 0 for "(* inexact 0)"
76 This follows what it always did for "(* 0 inexact)".
77 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
78 ** SRFI-19: `date->julian-day' did not account for timezone offset
79 ** `ttyname' no longer crashes when passed a non-tty argument
80 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
81 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
82 ** GOOPS: Fixed a bug in `method-more-specific?'
83 ** Build problems on Solaris fixed
84 ** Build problems on HP-UX IA64 fixed
85 ** Build problems on MinGW fixed
88 Changes in 1.8.1 (since 1.8.0):
90 * LFS functions are now used to access 64-bit files on 32-bit systems.
92 * New procedures (see the manual for details)
94 ** primitive-_exit - [Scheme] the-root-module
95 ** scm_primitive__exit - [C]
96 ** make-completion-function - [Scheme] (ice-9 readline)
97 ** scm_c_locale_stringn_to_number - [C]
98 ** scm_srfi1_append_reverse [C]
99 ** scm_srfi1_append_reverse_x [C]
105 * New `(ice-9 i18n)' module (see the manual for details)
109 ** Build problems have been fixed on MacOS, SunOS, and QNX.
111 ** `strftime' fix sign of %z timezone offset.
113 ** A one-dimensional array can now be 'equal?' to a vector.
115 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
117 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
119 ** `record-accessor' and `record-modifier' now have strict type checks.
121 Record accessor and modifier procedures now throw an error if the
122 record type of the record they're given is not the type expected.
123 (Previously accessors returned #f and modifiers silently did nothing).
125 ** It is now OK to use both autoload and use-modules on a given module.
127 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
129 Previously there was no checking on primatives like make-vector that
130 accept "one or two" arguments. Now there is.
132 ** The srfi-1 assoc function now calls its equality predicate properly.
134 Previously srfi-1 assoc would call the equality predicate with the key
135 last. According to the SRFI, the key should be first.
137 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
139 ** The array-set! procedure no longer segfaults when given a bit vector.
141 ** Bugs in make-shared-array have been fixed.
143 ** string<? and friends now follow char<? etc order on 8-bit chars.
145 ** The format procedure now handles inf and nan values for ~f correctly.
147 ** exact->inexact should no longer overflow when given certain large fractions.
149 ** srfi-9 accessor and modifier procedures now have strict record type checks.
151 This matches the srfi-9 specification.
153 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
155 Previously the (ice-9 ftw) procedures would ignore any file that had
156 the same inode number as a file they had already seen, even if that
157 file was on a different device.
160 Changes in 1.8.0 (changes since the 1.6.x series):
162 * Changes to the distribution
164 ** Guile is now licensed with the GNU Lesser General Public License.
166 ** The manual is now licensed with the GNU Free Documentation License.
168 ** Guile now requires GNU MP (http://swox.com/gmp).
170 Guile now uses the GNU MP library for arbitrary precision arithmetic.
172 ** Guile now has separate private and public configuration headers.
174 That is, things like HAVE_STRING_H no longer leak from Guile's
177 ** Guile now provides and uses an "effective" version number.
179 Guile now provides scm_effective_version and effective-version
180 functions which return the "effective" version number. This is just
181 the normal full version string without the final micro-version number,
182 so the current effective-version is "1.8". The effective version
183 should remain unchanged during a stable series, and should be used for
184 items like the versioned share directory name
185 i.e. /usr/share/guile/1.8.
187 Providing an unchanging version number during a stable release for
188 things like the versioned share directory can be particularly
189 important for Guile "add-on" packages, since it provides a directory
190 that they can install to that won't be changed out from under them
191 with each micro release during a stable series.
193 ** Thread implementation has changed.
195 When you configure "--with-threads=null", you will get the usual
196 threading API (call-with-new-thread, make-mutex, etc), but you can't
197 actually create new threads. Also, "--with-threads=no" is now
198 equivalent to "--with-threads=null". This means that the thread API
199 is always present, although you might not be able to create new
202 When you configure "--with-threads=pthreads" or "--with-threads=yes",
203 you will get threads that are implemented with the portable POSIX
204 threads. These threads can run concurrently (unlike the previous
205 "coop" thread implementation), but need to cooperate for things like
208 The default is "pthreads", unless your platform doesn't have pthreads,
209 in which case "null" threads are used.
211 See the manual for details, nodes "Initialization", "Multi-Threading",
212 "Blocking", and others.
214 ** There is the new notion of 'discouraged' features.
216 This is a milder form of deprecation.
218 Things that are discouraged should not be used in new code, but it is
219 OK to leave them in old code for now. When a discouraged feature is
220 used, no warning message is printed like there is for 'deprecated'
221 features. Also, things that are merely discouraged are nevertheless
222 implemented efficiently, while deprecated features can be very slow.
224 You can omit discouraged features from libguile by configuring it with
225 the '--disable-discouraged' option.
227 ** Deprecation warnings can be controlled at run-time.
229 (debug-enable 'warn-deprecated) switches them on and (debug-disable
230 'warn-deprecated) switches them off.
232 ** Support for SRFI 61, extended cond syntax for multiple values has
235 This SRFI is always available.
237 ** Support for require-extension, SRFI-55, has been added.
239 The SRFI-55 special form `require-extension' has been added. It is
240 available at startup, and provides a portable way to load Scheme
241 extensions. SRFI-55 only requires support for one type of extension,
242 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
245 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
247 The (srfi srfi-26) module is an implementation of SRFI-26 which
248 provides the `cut' and `cute' syntax. These may be used to specialize
249 parameters without currying.
251 ** New module (srfi srfi-31)
253 This is an implementation of SRFI-31 which provides a special form
254 `rec' for recursive evaluation.
256 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
257 been merged with the core, making their functionality always
260 The modules are still available, tho, and you could use them together
261 with a renaming import, for example.
263 ** Guile no longer includes its own version of libltdl.
265 The official version is good enough now.
267 ** The --enable-htmldoc option has been removed from 'configure'.
269 Support for translating the documentation into HTML is now always
270 provided. Use 'make html'.
272 ** New module (ice-9 serialize):
274 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
275 don't trust the thread safety of most of your program, but where you
276 have some section(s) of code which you consider can run in parallel to
277 other sections. See ice-9/serialize.scm for more information.
279 ** The configure option '--disable-arrays' has been removed.
281 Support for arrays and uniform numeric arrays is now always included
284 * Changes to the stand-alone interpreter
286 ** New command line option `-L'.
288 This option adds a directory to the front of the load path.
290 ** New command line option `--no-debug'.
292 Specifying `--no-debug' on the command line will keep the debugging
293 evaluator turned off, even for interactive sessions.
295 ** User-init file ~/.guile is now loaded with the debugging evaluator.
297 Previously, the normal evaluator would have been used. Using the
298 debugging evaluator gives better error messages.
300 ** The '-e' option now 'read's its argument.
302 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
303 be used with '-e'. For example, you can now write a script like
306 exec guile -e '(@ (demo) main)' -s "$0" "$@"
309 (define-module (demo)
313 (format #t "Demo: ~a~%" args))
316 * Changes to Scheme functions and syntax
318 ** Guardians have changed back to their original semantics
320 Guardians now behave like described in the paper by Dybvig et al. In
321 particular, they no longer make guarantees about the order in which
322 they return objects, and they can no longer be greedy.
324 They no longer drop cyclic data structures.
326 The C function scm_make_guardian has been changed incompatibly and no
327 longer takes the 'greedy_p' argument.
329 ** New function hashx-remove!
331 This function completes the set of 'hashx' functions.
333 ** The concept of dynamic roots has been factored into continuation
334 barriers and dynamic states.
336 Each thread has a current dynamic state that carries the values of the
337 fluids. You can create and copy dynamic states and use them as the
338 second argument for 'eval'. See "Fluids and Dynamic States" in the
341 To restrict the influence that captured continuations can have on the
342 control flow, you can errect continuation barriers. See "Continuation
343 Barriers" in the manual.
345 The function call-with-dynamic-root now essentially temporarily
346 installs a new dynamic state and errects a continuation barrier.
348 ** The default load path no longer includes "." at the end.
350 Automatically loading modules from the current directory should not
351 happen by default. If you want to allow it in a more controlled
352 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
355 ** The uniform vector and array support has been overhauled.
357 It now complies with SRFI-4 and the weird prototype based uniform
358 array creation has been deprecated. See the manual for more details.
360 Some non-compatible changes have been made:
361 - characters can no longer be stored into byte arrays.
362 - strings and bit vectors are no longer considered to be uniform numeric
364 - array-rank throws an error for non-arrays instead of returning zero.
365 - array-ref does no longer accept non-arrays when no indices are given.
367 There is the new notion of 'generalized vectors' and corresponding
368 procedures like 'generalized-vector-ref'. Generalized vectors include
369 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
371 Arrays use generalized vectors as their storage, so that you still
372 have arrays of characters, bits, etc. However, uniform-array-read!
373 and uniform-array-write can no longer read/write strings and
376 ** There is now support for copy-on-write substrings, mutation-sharing
377 substrings and read-only strings.
379 Three new procedures are related to this: substring/shared,
380 substring/copy, and substring/read-only. See the manual for more
383 ** Backtraces will now highlight the value that caused the error.
385 By default, these values are enclosed in "{...}", such as in this
394 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
395 <unnamed port>:1:1: Wrong type (expecting pair): a
396 ABORT: (wrong-type-arg)
398 The prefix and suffix used for highlighting can be set via the two new
399 printer options 'highlight-prefix' and 'highlight-suffix'. For
400 example, putting this into ~/.guile will output the bad value in bold
403 (print-set! highlight-prefix "\x1b[1m")
404 (print-set! highlight-suffix "\x1b[22m")
407 ** 'gettext' support for internationalization has been added.
409 See the manual for details.
411 ** New syntax '@' and '@@':
413 You can now directly refer to variables exported from a module by
416 (@ MODULE-NAME VARIABLE-NAME)
418 For example (@ (ice-9 pretty-print) pretty-print) will directly access
419 the pretty-print variable exported from the (ice-9 pretty-print)
420 module. You don't need to 'use' that module first. You can also use
421 '@' as a target of 'set!', as in (set! (@ mod var) val).
423 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
424 but it can also access variables that have not been exported. It is
425 intended only for kluges and temporary fixes and for debugging, not
428 ** Keyword syntax has been made more disciplined.
430 Previously, the name of a keyword was read as a 'token' but printed as
431 a symbol. Now, it is read as a general Scheme datum which must be a
442 ERROR: In expression (a b c):
446 ERROR: Unbound variable: foo
451 ERROR: Wrong type (expecting symbol): 12
455 ERROR: Wrong type (expecting symbol): (a b c)
459 ** The printing of symbols that might look like keywords can be
462 The new printer option 'quote-keywordish-symbols' controls how symbols
463 are printed that have a colon as their first or last character. The
464 default now is to only quote a symbol with #{...}# when the read
465 option 'keywords' is not '#f'. Thus:
467 guile> (define foo (string->symbol ":foo"))
468 guile> (read-set! keywords #f)
471 guile> (read-set! keywords 'prefix)
474 guile> (print-set! quote-keywordish-symbols #f)
478 ** 'while' now provides 'break' and 'continue'
480 break and continue were previously bound in a while loop, but not
481 documented, and continue didn't quite work properly. The undocumented
482 parameter to break which gave a return value for the while has been
485 ** 'call-with-current-continuation' is now also available under the name
488 ** The module system now checks for duplicate bindings.
490 The module system now can check for name conflicts among imported
493 The behavior can be controlled by specifying one or more 'duplicates'
494 handlers. For example, to make Guile return an error for every name
502 The new default behavior of the module system when a name collision
503 has been detected is to
505 1. Give priority to bindings marked as a replacement.
506 2. Issue a warning (different warning if overriding core binding).
507 3. Give priority to the last encountered binding (this corresponds to
510 If you want the old behavior back without replacements or warnings you
513 (default-duplicate-binding-handler 'last)
515 to your .guile init file.
517 ** New define-module option: :replace
519 :replace works as :export, but, in addition, marks the binding as a
522 A typical example is `format' in (ice-9 format) which is a replacement
523 for the core binding `format'.
525 ** Adding prefixes to imported bindings in the module system
527 There is now a new :use-module option :prefix. It can be used to add
528 a prefix to all imported bindings.
531 :use-module ((bar) :prefix bar:))
533 will import all bindings exported from bar, but rename them by adding
536 ** Conflicting generic functions can be automatically merged.
538 When two imported bindings conflict and they are both generic
539 functions, the two functions can now be merged automatically. This is
540 activated with the 'duplicates' handler 'merge-generics'.
542 ** New function: effective-version
544 Returns the "effective" version number. This is just the normal full
545 version string without the final micro-version number. See "Changes
546 to the distribution" above.
548 ** New threading functions: parallel, letpar, par-map, and friends
550 These are convenient ways to run calculations in parallel in new
551 threads. See "Parallel forms" in the manual for details.
553 ** New function 'try-mutex'.
555 This function will attempt to lock a mutex but will return immediately
556 instead of blocking and indicate failure.
558 ** Waiting on a condition variable can have a timeout.
560 The function 'wait-condition-variable' now takes a third, optional
561 argument that specifies the point in time where the waiting should be
564 ** New function 'broadcast-condition-variable'.
566 ** New functions 'all-threads' and 'current-thread'.
568 ** Signals and system asyncs work better with threads.
570 The function 'sigaction' now takes a fourth, optional, argument that
571 specifies the thread that the handler should run in. When the
572 argument is omitted, the handler will run in the thread that called
575 Likewise, 'system-async-mark' takes a second, optional, argument that
576 specifies the thread that the async should run in. When it is
577 omitted, the async will run in the thread that called
580 C code can use the new functions scm_sigaction_for_thread and
581 scm_system_async_mark_for_thread to pass the new thread argument.
583 When a thread blocks on a mutex, a condition variable or is waiting
584 for IO to be possible, it will still execute system asyncs. This can
585 be used to interrupt such a thread by making it execute a 'throw', for
588 ** The function 'system-async' is deprecated.
590 You can now pass any zero-argument procedure to 'system-async-mark'.
591 The function 'system-async' will just return its argument unchanged
594 ** New functions 'call-with-blocked-asyncs' and
595 'call-with-unblocked-asyncs'
597 The expression (call-with-blocked-asyncs PROC) will call PROC and will
598 block execution of system asyncs for the current thread by one level
599 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
600 procedure and will unblock the execution of system asyncs by one
601 level for the current thread.
603 Only system asyncs are affected by these functions.
605 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
607 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
608 instead. Those functions are easier to use correctly and can be
611 ** New function 'unsetenv'.
613 ** New macro 'define-syntax-public'.
615 It works like 'define-syntax' and also exports the defined macro (but
618 ** There is support for Infinity and NaNs.
620 Following PLT Scheme, Guile can now work with infinite numbers, and
623 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
624 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
625 "+nan.0"). These numbers are inexact and have no exact counterpart.
627 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
628 sign of the dividend. The infinities are integers, and they answer #t
629 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
630 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
641 ERROR: Numerical overflow
643 Two new predicates 'inf?' and 'nan?' can be used to test for the
646 ** Inexact zero can have a sign.
648 Guile can now distinguish between plus and minus inexact zero, if your
649 platform supports this, too. The two zeros are equal according to
650 '=', but not according to 'eqv?'. For example
661 ** Guile now has exact rationals.
663 Guile can now represent fractions such as 1/3 exactly. Computing with
664 them is also done exactly, of course:
669 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
672 For example: (floor 2) now returns an exact 2 where in the past it
673 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
675 ** inexact->exact no longer returns only integers.
677 Without exact rationals, the closest exact number was always an
678 integer, but now inexact->exact returns the fraction that is exactly
679 equal to a floating point number. For example:
681 (inexact->exact 1.234)
682 => 694680242521899/562949953421312
684 When you want the old behavior, use 'round' explicitly:
686 (inexact->exact (round 1.234))
689 ** New function 'rationalize'.
691 This function finds a simple fraction that is close to a given real
692 number. For example (and compare with inexact->exact above):
694 (rationalize (inexact->exact 1.234) 1/2000)
697 Note that, as required by R5RS, rationalize returns only then an exact
698 result when both its arguments are exact.
700 ** 'odd?' and 'even?' work also for inexact integers.
702 Previously, (odd? 1.0) would signal an error since only exact integers
703 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
704 returns #f and (odd? 1.5) signals an error.
706 ** Guile now has uninterned symbols.
708 The new function 'make-symbol' will return an uninterned symbol. This
709 is a symbol that is unique and is guaranteed to remain unique.
710 However, uninterned symbols can not yet be read back in.
712 Use the new function 'symbol-interned?' to check whether a symbol is
715 ** pretty-print has more options.
717 The function pretty-print from the (ice-9 pretty-print) module can now
718 also be invoked with keyword arguments that control things like
719 maximum output width. See the manual for details.
721 ** Variables have no longer a special behavior for `equal?'.
723 Previously, comparing two variables with `equal?' would recursivly
724 compare their values. This is no longer done. Variables are now only
725 `equal?' if they are `eq?'.
727 ** `(begin)' is now valid.
729 You can now use an empty `begin' form. It will yield #<unspecified>
730 when evaluated and simply be ignored in a definition context.
732 ** Deprecated: procedure->macro
734 Change your code to use 'define-macro' or r5rs macros. Also, be aware
735 that macro expansion will not be done during evaluation, but prior to
738 ** Soft ports now allow a `char-ready?' procedure
740 The vector argument to `make-soft-port' can now have a length of
741 either 5 or 6. (Previously the length had to be 5.) The optional 6th
742 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
743 that returns the number of characters that can be read immediately
744 without the soft port blocking.
746 ** Deprecated: undefine
748 There is no replacement for undefine.
750 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
751 have been discouraged.
753 They are relics from a time where a keyword like #:foo was used
754 directly as a Tcl option "-foo" and thus keywords were internally
755 stored as a symbol with a starting dash. We now store a symbol
758 Use symbol->keyword and keyword->symbol instead.
760 ** The `cheap' debug option is now obsolete
762 Evaluator trap calls are now unconditionally "cheap" - in other words,
763 they pass a debug object to the trap handler rather than a full
764 continuation. The trap handler code can capture a full continuation
765 by using `call-with-current-continuation' in the usual way, if it so
768 The `cheap' option is retained for now so as not to break existing
769 code which gets or sets it, but setting it now has no effect. It will
770 be removed in the next major Guile release.
772 ** Evaluator trap calls now support `tweaking'
774 `Tweaking' means that the trap handler code can modify the Scheme
775 expression that is about to be evaluated (in the case of an
776 enter-frame trap) or the value that is being returned (in the case of
777 an exit-frame trap). The trap handler code indicates that it wants to
778 do this by returning a pair whose car is the symbol 'instead and whose
779 cdr is the modified expression or return value.
781 * Changes to the C interface
783 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
784 take a 'delete' function argument.
786 This argument makes no sense since the delete function is used to
787 remove a pair from an alist, and this must not be configurable.
789 This is an incompatible change.
791 ** The GH interface is now subject to the deprecation mechanism
793 The GH interface has been deprecated for quite some time but now it is
794 actually removed from Guile when it is configured with
795 --disable-deprecated.
797 See the manual "Transitioning away from GH" for more information.
799 ** A new family of functions for converting between C values and
800 Scheme values has been added.
802 These functions follow a common naming scheme and are designed to be
803 easier to use, thread-safe and more future-proof than the older
808 These are predicates that return a C boolean: 1 or 0. Instead of
809 SCM_NFALSEP, you can now use scm_is_true, for example.
811 - <type> scm_to_<type> (SCM val, ...)
813 These are functions that convert a Scheme value into an appropriate
814 C value. For example, you can use scm_to_int to safely convert from
817 - SCM scm_from_<type> (<type> val, ...)
819 These functions convert from a C type to a SCM value; for example,
820 scm_from_int for ints.
822 There is a huge number of these functions, for numbers, strings,
823 symbols, vectors, etc. They are documented in the reference manual in
824 the API section together with the types that they apply to.
826 ** New functions for dealing with complex numbers in C have been added.
828 The new functions are scm_c_make_rectangular, scm_c_make_polar,
829 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
830 They work like scm_make_rectangular etc but take or return doubles
833 ** The function scm_make_complex has been discouraged.
835 Use scm_c_make_rectangular instead.
837 ** The INUM macros have been deprecated.
839 A lot of code uses these macros to do general integer conversions,
840 although the macros only work correctly with fixnums. Use the
841 following alternatives.
843 SCM_INUMP -> scm_is_integer or similar
844 SCM_NINUMP -> !scm_is_integer or similar
845 SCM_MAKINUM -> scm_from_int or similar
846 SCM_INUM -> scm_to_int or similar
848 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
849 do the validating for you.
851 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
852 have been discouraged.
854 Use the newer scm_to_<type> and scm_from_<type> functions instead for
855 new code. The functions have been discouraged since they don't fit
858 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
860 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
861 evaluates its argument twice. Use scm_is_true, etc. instead for new
864 ** The macro SCM_EQ_P has been discouraged.
866 Use scm_is_eq for new code, which fits better into the naming
869 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
872 Use the function scm_is_pair or scm_is_null instead.
874 ** The functions scm_round and scm_truncate have been deprecated and
875 are now available as scm_c_round and scm_c_truncate, respectively.
877 These functions occupy the names that scm_round_number and
878 scm_truncate_number should have.
880 ** The functions scm_c_string2str, scm_c_substring2str, and
881 scm_c_symbol2str have been deprecated.
883 Use scm_to_locale_stringbuf or similar instead, maybe together with
886 ** New functions scm_c_make_string, scm_c_string_length,
887 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
888 scm_c_substring_shared, scm_c_substring_copy.
890 These are like scm_make_string, scm_length, etc. but are slightly
891 easier to use from C.
893 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
894 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
896 They export too many assumptions about the implementation of strings
897 and symbols that are no longer true in the presence of
898 mutation-sharing substrings and when Guile switches to some form of
901 When working with strings, it is often best to use the normal string
902 functions provided by Guile, such as scm_c_string_ref,
903 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
904 manual since many more such functions are now provided than
907 When you want to convert a SCM string to a C string, use the
908 scm_to_locale_string function or similar instead. For symbols, use
909 scm_symbol_to_string and then work with that string. Because of the
910 new string representation, scm_symbol_to_string does not need to copy
911 and is thus quite efficient.
913 ** Some string, symbol and keyword functions have been discouraged.
915 They don't fit into the uniform naming scheme and are not explicit
916 about the character encoding.
918 Replace according to the following table:
920 scm_allocate_string -> scm_c_make_string
921 scm_take_str -> scm_take_locale_stringn
922 scm_take0str -> scm_take_locale_string
923 scm_mem2string -> scm_from_locale_stringn
924 scm_str2string -> scm_from_locale_string
925 scm_makfrom0str -> scm_from_locale_string
926 scm_mem2symbol -> scm_from_locale_symboln
927 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
928 scm_str2symbol -> scm_from_locale_symbol
930 SCM_SYMBOL_HASH -> scm_hashq
931 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
933 scm_c_make_keyword -> scm_from_locale_keyword
935 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
936 now also available to C code.
938 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
940 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
941 the latter returns the true name of the keyword, not the 'dash name',
942 as SCM_KEYWORDSYM used to do.
944 ** A new way to access arrays in a thread-safe and efficient way has
947 See the manual, node "Accessing Arrays From C".
949 ** The old uniform vector and bitvector implementations have been
950 unceremoniously removed.
952 This implementation exposed the details of the tagging system of
953 Guile. Use the new C API explained in the manual in node "Uniform
954 Numeric Vectors" and "Bit Vectors", respectively.
956 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
957 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
958 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
959 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
960 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
961 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
964 ** The macros dealing with vectors have been deprecated.
966 Use the new functions scm_is_vector, scm_vector_elements,
967 scm_vector_writable_elements, etc, or scm_is_simple_vector,
968 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
969 manual for more details.
971 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
972 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
974 The following macros have been removed: SCM_VECTOR_BASE,
975 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
976 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
978 ** Some C functions and macros related to arrays have been deprecated.
980 Migrate according to the following table:
982 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
983 scm_make_ra -> scm_make_array
984 scm_shap2ra -> scm_make_array
985 scm_cvref -> scm_c_generalized_vector_ref
986 scm_ra_set_contp -> do not use
987 scm_aind -> scm_array_handle_pos
988 scm_raprin1 -> scm_display or scm_write
990 SCM_ARRAYP -> scm_is_array
991 SCM_ARRAY_NDIM -> scm_c_array_rank
992 SCM_ARRAY_DIMS -> scm_array_handle_dims
993 SCM_ARRAY_CONTP -> do not use
994 SCM_ARRAY_MEM -> do not use
995 SCM_ARRAY_V -> scm_array_handle_elements or similar
996 SCM_ARRAY_BASE -> do not use
998 ** SCM_CELL_WORD_LOC has been deprecated.
1000 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1001 to a SCM, as opposed to a pointer to a scm_t_bits.
1003 This was done to allow the correct use of pointers into the Scheme
1004 heap. Previously, the heap words were of type scm_t_bits and local
1005 variables and function arguments were of type SCM, making it
1006 non-standards-conformant to have a pointer that can point to both.
1008 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1010 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1011 second and third words of double smobs. Likewise for
1012 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1014 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1015 used to get and set the 16 exra bits in the zeroth word of a smob.
1017 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1018 accesing the first immediate word of a smob as a SCM value, and there
1019 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1020 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1022 ** New way to deal with non-local exits and re-entries.
1024 There is a new set of functions that essentially do what
1025 scm_internal_dynamic_wind does, but in a way that is more convenient
1026 for C code in some situations. Here is a quick example of how to
1027 prevent a potential memory leak:
1034 scm_dynwind_begin (0);
1036 mem = scm_malloc (100);
1037 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1039 /* MEM would leak if BAR throws an error.
1040 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1047 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1048 SCM_DYNWIND_END as well.
1052 For full documentation, see the node "Dynamic Wind" in the manual.
1054 ** New function scm_dynwind_free
1056 This function calls 'free' on a given pointer when a dynwind context
1057 is left. Thus the call to scm_dynwind_unwind_handler above could be
1058 replaced with simply scm_dynwind_free (mem).
1060 ** New functions scm_c_call_with_blocked_asyncs and
1061 scm_c_call_with_unblocked_asyncs
1063 Like scm_call_with_blocked_asyncs etc. but for C functions.
1065 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1067 In addition to scm_c_call_with_blocked_asyncs you can now also use
1068 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1069 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1071 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1072 SCM_REALLOW_INTS have been deprecated.
1074 They do no longer fulfill their original role of blocking signal
1075 delivery. Depending on what you want to achieve, replace a pair of
1076 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1077 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1080 ** The value 'scm_mask_ints' is no longer writable.
1082 Previously, you could set scm_mask_ints directly. This is no longer
1083 possible. Use scm_c_call_with_blocked_asyncs and
1084 scm_c_call_with_unblocked_asyncs instead.
1086 ** New way to temporarily set the current input, output or error ports
1088 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1089 context' (see above). <foo> is one of "input", "output" or "error".
1091 ** New way to temporarily set fluids
1093 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1094 above) to temporarily set the value of a fluid.
1096 ** New types scm_t_intmax and scm_t_uintmax.
1098 On platforms that have them, these types are identical to intmax_t and
1099 uintmax_t, respectively. On other platforms, they are identical to
1100 the largest integer types that Guile knows about.
1102 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1104 You should not have used them.
1106 ** Many public #defines with generic names have been made private.
1108 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1109 private or renamed with a more suitable public name.
1111 ** The macro SCM_TYP16S has been deprecated.
1113 This macro is not intended for public use.
1115 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1117 Use scm_is_true (scm_inexact_p (...)) instead.
1119 ** The macro SCM_SLOPPY_REALP has been deprecated.
1121 Use scm_is_real instead.
1123 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1125 Use scm_is_complex instead.
1127 ** Some preprocessor defines have been deprecated.
1129 These defines indicated whether a certain feature was present in Guile
1130 or not. Going forward, assume that the features are always present.
1132 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1133 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1135 The following macros have been removed completely: MEMOIZE_LOCALS,
1136 SCM_RECKLESS, SCM_CAUTIOUS.
1138 ** The preprocessor define STACK_DIRECTION has been deprecated.
1140 There should be no need to know about the stack direction for ordinary
1143 ** New function: scm_effective_version
1145 Returns the "effective" version number. This is just the normal full
1146 version string without the final micro-version number. See "Changes
1147 to the distribution" above.
1149 ** The function scm_call_with_new_thread has a new prototype.
1151 Instead of taking a list with the thunk and handler, these two
1152 arguments are now passed directly:
1154 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1156 This is an incompatible change.
1158 ** New snarfer macro SCM_DEFINE_PUBLIC.
1160 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1161 function in the init section.
1163 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1165 ** Garbage collector rewrite.
1167 The garbage collector is cleaned up a lot, and now uses lazy
1168 sweeping. This is reflected in the output of (gc-stats); since cells
1169 are being freed when they are allocated, the cells-allocated field
1170 stays roughly constant.
1172 For malloc related triggers, the behavior is changed. It uses the same
1173 heuristic as the cell-triggered collections. It may be tuned with the
1174 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1175 for minimum yield of malloc related triggers. The default is 40.
1176 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1179 Debugging operations for the freelist have been deprecated, along with
1180 the C variables that control garbage collection. The environment
1181 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1182 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1184 For understanding the memory usage of a GUILE program, the routine
1185 gc-live-object-stats returns an alist containing the number of live
1186 objects for every type.
1189 ** The function scm_definedp has been renamed to scm_defined_p
1191 The name scm_definedp is deprecated.
1193 ** The struct scm_cell type has been renamed to scm_t_cell
1195 This is in accordance to Guile's naming scheme for types. Note that
1196 the name scm_cell is now used for a function that allocates and
1197 initializes a new cell (see below).
1199 ** New functions for memory management
1201 A new set of functions for memory management has been added since the
1202 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1203 indeed, Guile itself contained some long standing bugs that could
1204 cause aborts in long running programs.
1206 The new functions are more symmetrical and do not need cooperation
1207 from smob free routines, among other improvements.
1209 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1210 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1211 scm_gc_free, scm_gc_register_collectable_memory, and
1212 scm_gc_unregister_collectable_memory. Refer to the manual for more
1213 details and for upgrading instructions.
1215 The old functions for memory management have been deprecated. They
1216 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1217 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1219 ** Declarations of exported features are marked with SCM_API.
1221 Every declaration of a feature that belongs to the exported Guile API
1222 has been marked by adding the macro "SCM_API" to the start of the
1223 declaration. This macro can expand into different things, the most
1224 common of which is just "extern" for Unix platforms. On Win32, it can
1225 be used to control which symbols are exported from a DLL.
1227 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1228 will expand into "__declspec (dllimport) extern", which is needed for
1229 linking to the Guile DLL in Windows.
1231 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1232 SCM_SRFI4_IMPORT, for the corresponding libraries.
1234 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1236 Use the new functions scm_cell and scm_double_cell instead. The old
1237 macros had problems because with them allocation and initialization
1238 was separated and the GC could sometimes observe half initialized
1239 cells. Only careful coding by the user of SCM_NEWCELL and
1240 SCM_NEWCELL2 could make this safe and efficient.
1242 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1244 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1247 ** SRCBRKP has been deprecated.
1249 Use scm_c_source_property_breakpoint_p instead.
1251 ** Deprecated: scm_makmacro
1253 Change your code to use either scm_makmmacro or to define macros in
1254 Scheme, using 'define-macro'.
1256 ** New function scm_c_port_for_each.
1258 This function is like scm_port_for_each but takes a pointer to a C
1259 function as the callback instead of a SCM value.
1261 ** The names scm_internal_select, scm_thread_sleep, and
1262 scm_thread_usleep have been discouraged.
1264 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1266 ** The GC can no longer be blocked.
1268 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1269 The GC can now run (partially) concurrently with other code and thus
1270 blocking it is not well defined.
1272 ** Many definitions have been removed that were previously deprecated.
1274 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1275 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1276 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1277 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1278 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1279 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1280 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1281 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1282 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1283 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1284 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1285 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1286 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1287 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1288 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1289 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1290 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1291 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1292 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1293 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1294 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1295 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1296 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1297 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1298 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1299 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1300 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1301 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1302 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1303 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1304 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1306 * Changes to bundled modules
1310 Using the (ice-9 debug) module no longer automatically switches Guile
1311 to use the debugging evaluator. If you want to switch to the
1312 debugging evaluator (which is needed for backtrace information if you
1313 hit an error), please add an explicit "(debug-enable 'debug)" to your
1314 code just after the code to use (ice-9 debug).
1317 Changes since Guile 1.4:
1319 * Changes to the distribution
1321 ** A top-level TODO file is included.
1323 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1325 Guile now always uses three numbers to represent the version,
1326 i.e. "1.6.5". The first number, 1, is the major version number, the
1327 second number, 6, is the minor version number, and the third number,
1328 5, is the micro version number. Changes in major version number
1329 indicate major changes in Guile.
1331 Minor version numbers that are even denote stable releases, and odd
1332 minor version numbers denote development versions (which may be
1333 unstable). The micro version number indicates a minor sub-revision of
1334 a given MAJOR.MINOR release.
1336 In keeping with the new scheme, (minor-version) and scm_minor_version
1337 no longer return everything but the major version number. They now
1338 just return the minor version number. Two new functions
1339 (micro-version) and scm_micro_version have been added to report the
1340 micro version number.
1342 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1344 ** New preprocessor definitions are available for checking versions.
1346 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1347 SCM_MICRO_VERSION to the appropriate integer values.
1349 ** Guile now actively warns about deprecated features.
1351 The new configure option `--enable-deprecated=LEVEL' and the
1352 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1353 See INSTALL and README for more information.
1355 ** Guile is much more likely to work on 64-bit architectures.
1357 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1358 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1359 for the use of a test machine, and thanks to Stefan Jahn for ia64
1362 ** New functions: setitimer and getitimer.
1364 These implement a fairly direct interface to the libc functions of the
1367 ** The #. reader extension is now disabled by default.
1369 For safety reasons, #. evaluation is disabled by default. To
1370 re-enable it, set the fluid read-eval? to #t. For example:
1372 (fluid-set! read-eval? #t)
1374 but make sure you realize the potential security risks involved. With
1375 read-eval? enabled, reading a data file from an untrusted source can
1378 ** New SRFI modules have been added:
1380 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1383 (srfi srfi-1) is a library containing many useful pair- and list-processing
1386 (srfi srfi-2) exports and-let*.
1388 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1390 (srfi srfi-6) is a dummy module for now, since guile already provides
1391 all of the srfi-6 procedures by default: open-input-string,
1392 open-output-string, get-output-string.
1394 (srfi srfi-8) exports receive.
1396 (srfi srfi-9) exports define-record-type.
1398 (srfi srfi-10) exports define-reader-ctor and implements the reader
1401 (srfi srfi-11) exports let-values and let*-values.
1403 (srfi srfi-13) implements the SRFI String Library.
1405 (srfi srfi-14) implements the SRFI Character-Set Library.
1407 (srfi srfi-17) implements setter and getter-with-setter and redefines
1408 some accessor procedures as procedures with getters. (such as car,
1409 cdr, vector-ref etc.)
1411 (srfi srfi-19) implements the SRFI Time/Date Library.
1413 ** New scripts / "executable modules"
1415 Subdirectory "scripts" contains Scheme modules that are packaged to
1416 also be executable as scripts. At this time, these scripts are available:
1425 See README there for more info.
1427 These scripts can be invoked from the shell with the new program
1428 "guile-tools", which keeps track of installation directory for you.
1431 $ guile-tools display-commentary srfi/*.scm
1433 guile-tools is copied to the standard $bindir on "make install".
1435 ** New module (ice-9 stack-catch):
1437 stack-catch is like catch, but saves the current state of the stack in
1438 the fluid the-last-stack. This fluid can be useful when using the
1439 debugger and when re-throwing an error.
1441 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1443 This has been done to prevent problems on lesser operating systems
1444 that can't tolerate `*'s in file names. The exported macro continues
1445 to be named `and-let*', of course.
1447 On systems that support it, there is also a compatibility module named
1448 (ice-9 and-let*). It will go away in the next release.
1450 ** New modules (oop goops) etc.:
1453 (oop goops describe)
1455 (oop goops active-slot)
1456 (oop goops composite-slot)
1458 The Guile Object Oriented Programming System (GOOPS) has been
1459 integrated into Guile. For further information, consult the GOOPS
1460 manual and tutorial in the `doc' directory.
1462 ** New module (ice-9 rdelim).
1464 This exports the following procedures which were previously defined
1465 in the default environment:
1467 read-line read-line! read-delimited read-delimited! %read-delimited!
1468 %read-line write-line
1470 For backwards compatibility the definitions are still imported into the
1471 default environment in this version of Guile. However you should add:
1473 (use-modules (ice-9 rdelim))
1475 to any program which uses the definitions, since this may change in
1478 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1479 can be used for similar functionality.
1481 ** New module (ice-9 rw)
1483 This is a subset of the (scsh rw) module from guile-scsh. Currently
1484 it defines two procedures:
1486 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1488 Read characters from a port or file descriptor into a string STR.
1489 A port must have an underlying file descriptor -- a so-called
1490 fport. This procedure is scsh-compatible and can efficiently read
1493 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1495 Write characters from a string STR to a port or file descriptor.
1496 A port must have an underlying file descriptor -- a so-called
1497 fport. This procedure is mostly compatible and can efficiently
1498 write large strings.
1500 ** New module (ice-9 match)
1502 This module includes Andrew K. Wright's pattern matcher. See
1503 ice-9/match.scm for brief description or
1505 http://www.star-lab.com/wright/code.html
1507 for complete documentation.
1509 ** New module (ice-9 buffered-input)
1511 This module provides procedures to construct an input port from an
1512 underlying source of input that reads and returns its input in chunks.
1513 The underlying input source is a Scheme procedure, specified by the
1514 caller, which the port invokes whenever it needs more input.
1516 This is useful when building an input port whose back end is Readline
1517 or a UI element such as the GtkEntry widget.
1521 The reference and tutorial documentation that was previously
1522 distributed separately, as `guile-doc', is now included in the core
1523 Guile distribution. The documentation consists of the following
1526 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1529 - The Guile Reference Manual (guile.texi) contains (or is intended to
1530 contain) reference documentation on all aspects of Guile.
1532 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1533 reference documentation for using GOOPS, Guile's Object Oriented
1536 - The Revised^5 Report on the Algorithmic Language Scheme
1539 See the README file in the `doc' directory for more details.
1541 ** There are a couple of examples in the examples/ directory now.
1543 * Changes to the stand-alone interpreter
1545 ** New command line option `--use-srfi'
1547 Using this option, SRFI modules can be loaded on startup and be
1548 available right from the beginning. This makes programming portable
1549 Scheme programs easier.
1551 The option `--use-srfi' expects a comma-separated list of numbers,
1552 each representing a SRFI number to be loaded into the interpreter
1553 before starting evaluating a script file or the REPL. Additionally,
1554 the feature identifier for the loaded SRFIs is recognized by
1555 `cond-expand' when using this option.
1558 $ guile --use-srfi=8,13
1559 guile> (receive (x z) (values 1 2) (+ 1 2))
1561 guile> (string-pad "bla" 20)
1564 ** Guile now always starts up in the `(guile-user)' module.
1566 Previously, scripts executed via the `-s' option would run in the
1567 `(guile)' module and the repl would run in the `(guile-user)' module.
1568 Now every user action takes place in the `(guile-user)' module by
1571 * Changes to Scheme functions and syntax
1573 ** Character classifiers work for non-ASCII characters.
1575 The predicates `char-alphabetic?', `char-numeric?',
1576 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1577 no longer check whether their arguments are ASCII characters.
1578 Previously, a character would only be considered alphabetic when it
1579 was also ASCII, for example.
1581 ** Previously deprecated Scheme functions have been removed:
1583 tag - no replacement.
1584 fseek - replaced by seek.
1585 list* - replaced by cons*.
1587 ** It's now possible to create modules with controlled environments
1591 (use-modules (ice-9 safe))
1592 (define m (make-safe-module))
1593 ;;; m will now be a module containing only a safe subset of R5RS
1594 (eval '(+ 1 2) m) --> 3
1595 (eval 'load m) --> ERROR: Unbound variable: load
1597 ** Evaluation of "()", the empty list, is now an error.
1599 Previously, the expression "()" evaluated to the empty list. This has
1600 been changed to signal a "missing expression" error. The correct way
1601 to write the empty list as a literal constant is to use quote: "'()".
1603 ** New concept of `Guile Extensions'.
1605 A Guile Extension is just a ordinary shared library that can be linked
1606 at run-time. We found it advantageous to give this simple concept a
1607 dedicated name to distinguish the issues related to shared libraries
1608 from the issues related to the module system.
1610 *** New function: load-extension
1612 Executing (load-extension lib init) is mostly equivalent to
1614 (dynamic-call init (dynamic-link lib))
1616 except when scm_register_extension has been called previously.
1617 Whenever appropriate, you should use `load-extension' instead of
1618 dynamic-link and dynamic-call.
1620 *** New C function: scm_c_register_extension
1622 This function registers a initialization function for use by
1623 `load-extension'. Use it when you don't want specific extensions to
1624 be loaded as shared libraries (for example on platforms that don't
1625 support dynamic linking).
1627 ** Auto-loading of compiled-code modules is deprecated.
1629 Guile used to be able to automatically find and link a shared
1630 library to satisfy requests for a module. For example, the module
1631 `(foo bar)' could be implemented by placing a shared library named
1632 "foo/libbar.so" (or with a different extension) in a directory on the
1635 This has been found to be too tricky, and is no longer supported. The
1636 shared libraries are now called "extensions". You should now write a
1637 small Scheme file that calls `load-extension' to load the shared
1638 library and initialize it explicitly.
1640 The shared libraries themselves should be installed in the usual
1641 places for shared libraries, with names like "libguile-foo-bar".
1643 For example, place this into a file "foo/bar.scm"
1645 (define-module (foo bar))
1647 (load-extension "libguile-foo-bar" "foobar_init")
1649 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1651 `eval' is now R5RS, that is it takes two arguments.
1652 The second argument is an environment specifier, i.e. either
1654 (scheme-report-environment 5)
1655 (null-environment 5)
1656 (interaction-environment)
1662 ** The module system has been made more disciplined.
1664 The function `eval' will save and restore the current module around
1665 the evaluation of the specified expression. While this expression is
1666 evaluated, `(current-module)' will now return the right module, which
1667 is the module specified as the second argument to `eval'.
1669 A consequence of this change is that `eval' is not particularly
1670 useful when you want allow the evaluated code to change what module is
1671 designated as the current module and have this change persist from one
1672 call to `eval' to the next. The read-eval-print-loop is an example
1673 where `eval' is now inadequate. To compensate, there is a new
1674 function `primitive-eval' that does not take a module specifier and
1675 that does not save/restore the current module. You should use this
1676 function together with `set-current-module', `current-module', etc
1677 when you want to have more control over the state that is carried from
1678 one eval to the next.
1680 Additionally, it has been made sure that forms that are evaluated at
1681 the top level are always evaluated with respect to the current module.
1682 Previously, subforms of top-level forms such as `begin', `case',
1683 etc. did not respect changes to the current module although these
1684 subforms are at the top-level as well.
1686 To prevent strange behavior, the forms `define-module',
1687 `use-modules', `use-syntax', and `export' have been restricted to only
1688 work on the top level. The forms `define-public' and
1689 `defmacro-public' only export the new binding on the top level. They
1690 behave just like `define' and `defmacro', respectively, when they are
1691 used in a lexical environment.
1693 Also, `export' will no longer silently re-export bindings imported
1694 from a used module. It will emit a `deprecation' warning and will
1695 cease to perform any re-export in the next version. If you actually
1696 want to re-export bindings, use the new `re-export' in place of
1697 `export'. The new `re-export' will not make copies of variables when
1698 rexporting them, as `export' did wrongly.
1700 ** Module system now allows selection and renaming of imported bindings
1702 Previously, when using `use-modules' or the `#:use-module' clause in
1703 the `define-module' form, all the bindings (association of symbols to
1704 values) for imported modules were added to the "current module" on an
1705 as-is basis. This has been changed to allow finer control through two
1706 new facilities: selection and renaming.
1708 You can now select which of the imported module's bindings are to be
1709 visible in the current module by using the `:select' clause. This
1710 clause also can be used to rename individual bindings. For example:
1712 ;; import all bindings no questions asked
1713 (use-modules (ice-9 common-list))
1715 ;; import four bindings, renaming two of them;
1716 ;; the current module sees: every some zonk-y zonk-n
1717 (use-modules ((ice-9 common-list)
1719 (remove-if . zonk-y)
1720 (remove-if-not . zonk-n))))
1722 You can also programmatically rename all selected bindings using the
1723 `:renamer' clause, which specifies a proc that takes a symbol and
1724 returns another symbol. Because it is common practice to use a prefix,
1725 we now provide the convenience procedure `symbol-prefix-proc'. For
1728 ;; import four bindings, renaming two of them specifically,
1729 ;; and all four w/ prefix "CL:";
1730 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1731 (use-modules ((ice-9 common-list)
1733 (remove-if . zonk-y)
1734 (remove-if-not . zonk-n))
1735 :renamer (symbol-prefix-proc 'CL:)))
1737 ;; import four bindings, renaming two of them specifically,
1738 ;; and all four by upcasing.
1739 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1740 (define (upcase-symbol sym)
1741 (string->symbol (string-upcase (symbol->string sym))))
1743 (use-modules ((ice-9 common-list)
1745 (remove-if . zonk-y)
1746 (remove-if-not . zonk-n))
1747 :renamer upcase-symbol))
1749 Note that programmatic renaming is done *after* individual renaming.
1750 Also, the above examples show `use-modules', but the same facilities are
1751 available for the `#:use-module' clause of `define-module'.
1753 See manual for more info.
1755 ** The semantics of guardians have changed.
1757 The changes are for the most part compatible. An important criterion
1758 was to keep the typical usage of guardians as simple as before, but to
1759 make the semantics safer and (as a result) more useful.
1761 *** All objects returned from guardians are now properly alive.
1763 It is now guaranteed that any object referenced by an object returned
1764 from a guardian is alive. It's now impossible for a guardian to
1765 return a "contained" object before its "containing" object.
1767 One incompatible (but probably not very important) change resulting
1768 from this is that it is no longer possible to guard objects that
1769 indirectly reference themselves (i.e. are parts of cycles). If you do
1770 so accidentally, you'll get a warning.
1772 *** There are now two types of guardians: greedy and sharing.
1774 If you call (make-guardian #t) or just (make-guardian), you'll get a
1775 greedy guardian, and for (make-guardian #f) a sharing guardian.
1777 Greedy guardians are the default because they are more "defensive".
1778 You can only greedily guard an object once. If you guard an object
1779 more than once, once in a greedy guardian and the rest of times in
1780 sharing guardians, then it is guaranteed that the object won't be
1781 returned from sharing guardians as long as it is greedily guarded
1784 Guardians returned by calls to `make-guardian' can now take one more
1785 optional parameter, which says whether to throw an error in case an
1786 attempt is made to greedily guard an object that is already greedily
1787 guarded. The default is true, i.e. throw an error. If the parameter
1788 is false, the guardian invocation returns #t if guarding was
1789 successful and #f if it wasn't.
1791 Also, since greedy guarding is, in effect, a side-effecting operation
1792 on objects, a new function is introduced: `destroy-guardian!'.
1793 Invoking this function on a guardian renders it unoperative and, if
1794 the guardian is greedy, clears the "greedily guarded" property of the
1795 objects that were guarded by it, thus undoing the side effect.
1797 Note that all this hair is hardly very important, since guardian
1798 objects are usually permanent.
1800 ** Continuations created by call-with-current-continuation now accept
1801 any number of arguments, as required by R5RS.
1803 ** New function `issue-deprecation-warning'
1805 This function is used to display the deprecation messages that are
1806 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1809 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1813 ;; `id' is deprecated. Use `identity' instead.
1818 ** New syntax `begin-deprecated'
1820 When deprecated features are included (as determined by the configure
1821 option --enable-deprecated), `begin-deprecated' is identical to
1822 `begin'. When deprecated features are excluded, it always evaluates
1823 to `#f', ignoring the body forms.
1825 ** New function `make-object-property'
1827 This function returns a new `procedure with setter' P that can be used
1828 to attach a property to objects. When calling P as
1832 where `obj' is any kind of object, it attaches `val' to `obj' in such
1833 a way that it can be retrieved by calling P as
1837 This function will replace procedure properties, symbol properties and
1838 source properties eventually.
1840 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1842 Instead of #&optional, #&key, etc you should now use #:optional,
1843 #:key, etc. Since #:optional is a keyword, you can write it as just
1844 :optional when (read-set! keywords 'prefix) is active.
1846 The old reader syntax `#&' is still supported, but deprecated. It
1847 will be removed in the next release.
1849 ** New define-module option: pure
1851 Tells the module system not to include any bindings from the root
1856 (define-module (totally-empty-module)
1859 ** New define-module option: export NAME1 ...
1861 Export names NAME1 ...
1863 This option is required if you want to be able to export bindings from
1864 a module which doesn't import one of `define-public' or `export'.
1868 (define-module (foo)
1870 :use-module (ice-9 r5rs)
1873 ;;; Note that we're pure R5RS below this point!
1878 ** New function: object->string OBJ
1880 Return a Scheme string obtained by printing a given object.
1882 ** New function: port? X
1884 Returns a boolean indicating whether X is a port. Equivalent to
1885 `(or (input-port? X) (output-port? X))'.
1887 ** New function: file-port?
1889 Determines whether a given object is a port that is related to a file.
1891 ** New function: port-for-each proc
1893 Apply PROC to each port in the Guile port table in turn. The return
1894 value is unspecified. More specifically, PROC is applied exactly once
1895 to every port that exists in the system at the time PORT-FOR-EACH is
1896 invoked. Changes to the port table while PORT-FOR-EACH is running
1897 have no effect as far as PORT-FOR-EACH is concerned.
1899 ** New function: dup2 oldfd newfd
1901 A simple wrapper for the `dup2' system call. Copies the file
1902 descriptor OLDFD to descriptor number NEWFD, replacing the
1903 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1904 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1905 to move away ports which are using NEWFD. The return value is
1908 ** New function: close-fdes fd
1910 A simple wrapper for the `close' system call. Close file
1911 descriptor FD, which must be an integer. Unlike close (*note
1912 close: Ports and File Descriptors.), the file descriptor will be
1913 closed even if a port is using it. The return value is
1916 ** New function: crypt password salt
1918 Encrypts `password' using the standard unix password encryption
1921 ** New function: chroot path
1923 Change the root directory of the running process to `path'.
1925 ** New functions: getlogin, cuserid
1927 Return the login name or the user name of the current effective user
1930 ** New functions: getpriority which who, setpriority which who prio
1932 Get or set the priority of the running process.
1934 ** New function: getpass prompt
1936 Read a password from the terminal, first displaying `prompt' and
1939 ** New function: flock file operation
1941 Set/remove an advisory shared or exclusive lock on `file'.
1943 ** New functions: sethostname name, gethostname
1945 Set or get the hostname of the machine the current process is running
1948 ** New function: mkstemp! tmpl
1950 mkstemp creates a new unique file in the file system and returns a
1951 new buffered port open for reading and writing to the file. TMPL
1952 is a string specifying where the file should be created: it must
1953 end with `XXXXXX' and will be changed in place to return the name
1954 of the temporary file.
1956 ** New function: open-input-string string
1958 Return an input string port which delivers the characters from
1959 `string'. This procedure, together with `open-output-string' and
1960 `get-output-string' implements SRFI-6.
1962 ** New function: open-output-string
1964 Return an output string port which collects all data written to it.
1965 The data can then be retrieved by `get-output-string'.
1967 ** New function: get-output-string
1969 Return the contents of an output string port.
1971 ** New function: identity
1973 Return the argument.
1975 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1976 are represented in Scheme as integers with normal host byte ordering.
1978 ** New function: inet-pton family address
1980 Convert a printable string network address into an integer. Note that
1981 unlike the C version of this function, the result is an integer with
1982 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1985 (inet-pton AF_INET "127.0.0.1") => 2130706433
1986 (inet-pton AF_INET6 "::1") => 1
1988 ** New function: inet-ntop family address
1990 Convert an integer network address into a printable string. Note that
1991 unlike the C version of this function, the input is an integer with
1992 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1995 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1996 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1997 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2001 Use `identity' instead.
2007 ** Deprecated: return-it
2011 ** Deprecated: string-character-length
2013 Use `string-length' instead.
2015 ** Deprecated: flags
2017 Use `logior' instead.
2019 ** Deprecated: close-all-ports-except.
2021 This was intended for closing ports in a child process after a fork,
2022 but it has the undesirable side effect of flushing buffers.
2023 port-for-each is more flexible.
2025 ** The (ice-9 popen) module now attempts to set up file descriptors in
2026 the child process from the current Scheme ports, instead of using the
2027 current values of file descriptors 0, 1, and 2 in the parent process.
2029 ** Removed function: builtin-weak-bindings
2031 There is no such concept as a weak binding any more.
2033 ** Removed constants: bignum-radix, scm-line-incrementors
2035 ** define-method: New syntax mandatory.
2037 The new method syntax is now mandatory:
2039 (define-method (NAME ARG-SPEC ...) BODY ...)
2040 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2042 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2043 REST-ARG ::= ARG-NAME
2045 If you have old code using the old syntax, import
2046 (oop goops old-define-method) before (oop goops) as in:
2048 (use-modules (oop goops old-define-method) (oop goops))
2050 ** Deprecated function: builtin-variable
2051 Removed function: builtin-bindings
2053 There is no longer a distinction between builtin or other variables.
2054 Use module system operations for all variables.
2056 ** Lazy-catch handlers are no longer allowed to return.
2058 That is, a call to `throw', `error', etc is now guaranteed to not
2061 ** Bugfixes for (ice-9 getopt-long)
2063 This module is now tested using test-suite/tests/getopt-long.test.
2064 The following bugs have been fixed:
2066 *** Parsing for options that are specified to have `optional' args now checks
2067 if the next element is an option instead of unconditionally taking it as the
2070 *** An error is now thrown for `--opt=val' when the option description
2071 does not specify `(value #t)' or `(value optional)'. This condition used to
2072 be accepted w/o error, contrary to the documentation.
2074 *** The error message for unrecognized options is now more informative.
2075 It used to be "not a record", an artifact of the implementation.
2077 *** The error message for `--opt' terminating the arg list (no value), when
2078 `(value #t)' is specified, is now more informative. It used to be "not enough
2081 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2082 The expansion used to be like so:
2084 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2086 Note that the "5d" is dropped. Now it is like so:
2088 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2090 This enables single-char options to have adjoining arguments as long as their
2091 constituent characters are not potential single-char options.
2093 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2095 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2096 property `arglist', which can be retrieved by `arity'. The result is that
2097 `arity' can give more detailed information than before:
2101 guile> (use-modules (ice-9 optargs))
2102 guile> (define* (foo #:optional a b c) a)
2104 0 or more arguments in `lambda*:G0'.
2109 3 optional arguments: `a', `b' and `c'.
2110 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2112 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2113 and `d', other keywords allowed.
2114 guile> (define* (baz a b #:optional c #:rest r) a)
2116 2 required arguments: `a' and `b', 1 optional argument: `c',
2119 * Changes to the C interface
2121 ** Types have been renamed from scm_*_t to scm_t_*.
2123 This has been done for POSIX sake. It reserves identifiers ending
2124 with "_t". What a concept.
2126 The old names are still available with status `deprecated'.
2128 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2130 ** Deprecated features have been removed.
2134 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2135 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2137 *** C Functions removed
2139 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2140 scm_fseek - replaced by scm_seek.
2141 gc-thunk - replaced by after-gc-hook.
2142 gh_int2scmb - replaced by gh_bool2scm.
2143 scm_tc_dblr - replaced by scm_tc16_real.
2144 scm_tc_dblc - replaced by scm_tc16_complex.
2145 scm_list_star - replaced by scm_cons_star.
2147 ** Deprecated: scm_makfromstr
2149 Use scm_mem2string instead.
2151 ** Deprecated: scm_make_shared_substring
2153 Explicit shared substrings will disappear from Guile.
2155 Instead, "normal" strings will be implemented using sharing
2156 internally, combined with a copy-on-write strategy.
2158 ** Deprecated: scm_read_only_string_p
2160 The concept of read-only strings will disappear in next release of
2163 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2165 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2167 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2169 Call a procedure with the indicated number of arguments. See "Fly
2170 Evaluation" in the manual.
2172 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2174 Call a procedure with the indicated number of arguments and a list of
2175 further arguments. See "Fly Evaluation" in the manual.
2177 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2179 Create a list of the given number of elements. See "List
2180 Constructors" in the manual.
2182 ** Renamed function: scm_listify has been replaced by scm_list_n.
2184 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2185 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2187 Use functions scm_list_N instead.
2189 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2191 Used by an application to read arbitrary number of bytes from a port.
2192 Same semantics as libc read, except that scm_c_read only returns less
2193 than SIZE bytes if at end-of-file.
2195 Warning: Doesn't update port line and column counts!
2197 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2199 Used by an application to write arbitrary number of bytes to an SCM
2200 port. Similar semantics as libc write. However, unlike libc
2201 write, scm_c_write writes the requested number of bytes and has no
2204 Warning: Doesn't update port line and column counts!
2206 ** New function: scm_init_guile ()
2208 In contrast to scm_boot_guile, scm_init_guile will return normally
2209 after initializing Guile. It is not available on all systems, tho.
2211 ** New functions: scm_str2symbol, scm_mem2symbol
2213 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2214 field of characters and creates a scheme symbol object from that C string.
2215 The function scm_mem2symbol takes a const char* and a number of characters and
2216 creates a symbol from the characters in that memory area.
2218 ** New functions: scm_primitive_make_property
2219 scm_primitive_property_ref
2220 scm_primitive_property_set_x
2221 scm_primitive_property_del_x
2223 These functions implement a new way to deal with object properties.
2224 See libguile/properties.c for their documentation.
2226 ** New function: scm_done_free (long size)
2228 This function is the inverse of scm_done_malloc. Use it to report the
2229 amount of smob memory you free. The previous method, which involved
2230 calling scm_done_malloc with negative argument, was somewhat
2231 unintuitive (and is still available, of course).
2233 ** New function: scm_c_memq (SCM obj, SCM list)
2235 This function provides a fast C level alternative for scm_memq for the case
2236 that the list parameter is known to be a proper list. The function is a
2237 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2238 list input parameter, since for anything else but a proper list the function's
2239 behaviour is undefined - it may even crash or loop endlessly. Further, for
2240 the case that the object is not found in the list, scm_c_memq returns #f which
2241 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2243 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2244 scm_remember_upto_here
2246 These functions replace the function scm_remember.
2248 ** Deprecated function: scm_remember
2250 Use one of the new functions scm_remember_upto_here_1,
2251 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2253 ** New function: scm_allocate_string
2255 This function replaces the function scm_makstr.
2257 ** Deprecated function: scm_makstr
2259 Use the new function scm_allocate_string instead.
2261 ** New global variable scm_gc_running_p introduced.
2263 Use this variable to find out if garbage collection is being executed. Up to
2264 now applications have used scm_gc_heap_lock to test if garbage collection was
2265 running, which also works because of the fact that up to know only the garbage
2266 collector has set this variable. But, this is an implementation detail that
2267 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2268 of this variable is (and has been) not fully safe anyway.
2270 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2272 Use these instead of SCM_LENGTH_MAX.
2274 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2275 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2276 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2278 Use these instead of SCM_LENGTH.
2280 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2281 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2282 SCM_SET_BITVECTOR_LENGTH
2284 Use these instead of SCM_SETLENGTH
2286 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2287 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2290 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2293 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2294 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2297 Use these instead of SCM_SETCHARS.
2299 ** New macro: SCM_BITVECTOR_P
2301 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2303 Use instead of SCM_COERCE_SUBSTR.
2305 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2307 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2309 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2310 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2311 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2312 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2313 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2314 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2315 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2316 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2317 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2318 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2319 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2320 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2321 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2322 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2323 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2325 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2326 Use scm_memory_error instead of SCM_NALLOC.
2327 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2328 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2329 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2330 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2331 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2332 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2333 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2334 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2335 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2336 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2337 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2338 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2339 Use a type specific setter macro instead of SCM_SETLENGTH.
2340 Use a type specific setter macro instead of SCM_SETCHARS.
2341 Use a type specific length macro instead of SCM_LENGTH_MAX.
2342 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2343 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2344 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2345 Use SCM_TYP16 instead of SCM_GCTYP16.
2346 Use SCM_CDR instead of SCM_GCCDR.
2347 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2348 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2349 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2350 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2351 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2352 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2353 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2355 ** Removed function: scm_struct_init
2357 ** Removed variable: scm_symhash_dim
2359 ** Renamed function: scm_make_cont has been replaced by
2360 scm_make_continuation, which has a different interface.
2362 ** Deprecated function: scm_call_catching_errors
2364 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2366 ** Deprecated function: scm_strhash
2368 Use scm_string_hash instead.
2370 ** Deprecated function: scm_vector_set_length_x
2372 Instead, create a fresh vector of the desired size and copy the contents.
2374 ** scm_gensym has changed prototype
2376 scm_gensym now only takes one argument.
2378 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2381 There is now only a single symbol type scm_tc7_symbol.
2382 The tag scm_tc7_lvector was not used anyway.
2384 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2386 Use scm_make_smob_type and scm_set_smob_XXX instead.
2388 ** New function scm_set_smob_apply.
2390 This can be used to set an apply function to a smob type.
2392 ** Deprecated function: scm_strprint_obj
2394 Use scm_object_to_string instead.
2396 ** Deprecated function: scm_wta
2398 Use scm_wrong_type_arg, or another appropriate error signalling function
2401 ** Explicit support for obarrays has been deprecated.
2403 Use `scm_str2symbol' and the generic hashtable functions instead.
2405 ** The concept of `vcells' has been deprecated.
2407 The data type `variable' is now used exclusively. `Vcells' have been
2408 a low-level concept so you are likely not affected by this change.
2410 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2411 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2413 Use scm_c_define or scm_c_lookup instead, as appropriate.
2415 *** New functions: scm_c_module_lookup, scm_c_lookup,
2416 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2417 scm_module_define, scm_define.
2419 These functions work with variables instead of with vcells.
2421 ** New functions for creating and defining `subr's and `gsubr's.
2423 The new functions more clearly distinguish between creating a subr (or
2424 gsubr) object and adding it to the current module.
2426 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2427 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2428 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2429 scm_c_define_gsubr_with_generic.
2431 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2432 scm_make_subr_with_generic, scm_make_gsubr,
2433 scm_make_gsubr_with_generic.
2435 Use the new ones from above instead.
2437 ** C interface to the module system has changed.
2439 While we suggest that you avoid as many explicit module system
2440 operations from C as possible for the time being, the C interface has
2441 been made more similar to the high-level Scheme module system.
2443 *** New functions: scm_c_define_module, scm_c_use_module,
2444 scm_c_export, scm_c_resolve_module.
2446 They mostly work like their Scheme namesakes. scm_c_define_module
2447 takes a function that is called a context where the new module is
2450 *** Deprecated functions: scm_the_root_module, scm_make_module,
2451 scm_ensure_user_module, scm_load_scheme_module.
2453 Use the new functions instead.
2455 ** Renamed function: scm_internal_with_fluids becomes
2458 scm_internal_with_fluids is available as a deprecated function.
2460 ** New function: scm_c_with_fluid.
2462 Just like scm_c_with_fluids, but takes one fluid and one value instead
2465 ** Deprecated typedefs: long_long, ulong_long.
2467 They are of questionable utility and they pollute the global
2470 ** Deprecated typedef: scm_sizet
2472 It is of questionable utility now that Guile requires ANSI C, and is
2475 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2476 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2477 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2479 Made more compliant with the naming policy by adding a _t at the end.
2481 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2482 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2484 With the exception of the mysterious scm_2ulong2big, they are still
2485 available under new names (scm_i_mkbig etc). These functions are not
2486 intended to be used in user code. You should avoid dealing with
2487 bignums directly, and should deal with numbers in general (which can
2490 ** Change in behavior: scm_num2long, scm_num2ulong
2492 The scm_num2[u]long functions don't any longer accept an inexact
2493 argument. This change in behavior is motivated by concordance with
2494 R5RS: It is more common that a primitive doesn't want to accept an
2495 inexact for an exact.
2497 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2498 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2499 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2502 These are conversion functions between the various ANSI C integral
2503 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2504 accept an inexact argument.
2506 ** New functions: scm_float2num, scm_double2num,
2507 scm_num2float, scm_num2double.
2509 These are conversion functions between the two ANSI C float types and
2512 ** New number validation macros:
2513 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2517 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2519 These are just nicer-named old scm_protect_object and
2520 scm_unprotect_object.
2522 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2524 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2526 These functions can be used to register pointers to locations that
2529 ** Deprecated function: scm_create_hook.
2531 Its sins are: misleading name, non-modularity and lack of general
2535 Changes since Guile 1.3.4:
2537 * Changes to the distribution
2539 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2541 We've changed the way we handle generated files in the Guile source
2542 repository. As a result, the procedure for building trees obtained
2543 from the nightly FTP snapshots or via CVS has changed:
2544 - You must have appropriate versions of autoconf, automake, and
2545 libtool installed on your system. See README for info on how to
2546 obtain these programs.
2547 - Before configuring the tree, you must first run the script
2548 `autogen.sh' at the top of the source tree.
2550 The Guile repository used to contain not only source files, written by
2551 humans, but also some generated files, like configure scripts and
2552 Makefile.in files. Even though the contents of these files could be
2553 derived mechanically from other files present, we thought it would
2554 make the tree easier to build if we checked them into CVS.
2556 However, this approach means that minor differences between
2557 developer's installed tools and habits affected the whole team.
2558 So we have removed the generated files from the repository, and
2559 added the autogen.sh script, which will reconstruct them
2563 ** configure now has experimental options to remove support for certain
2566 --disable-arrays omit array and uniform array support
2567 --disable-posix omit posix interfaces
2568 --disable-networking omit networking interfaces
2569 --disable-regex omit regular expression interfaces
2571 These are likely to become separate modules some day.
2573 ** New configure option --enable-debug-freelist
2575 This enables a debugging version of SCM_NEWCELL(), and also registers
2576 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2578 Configure with the --enable-debug-freelist option to enable
2579 the gc-set-debug-check-freelist! primitive, and then use:
2581 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2582 (gc-set-debug-check-freelist! #f) # turn off checking
2584 Checking of the freelist forces a traversal of the freelist and
2585 a garbage collection before each allocation of a cell. This can
2586 slow down the interpreter dramatically, so the setter should be used to
2587 turn on this extra processing only when necessary.
2589 ** New configure option --enable-debug-malloc
2591 Include code for debugging of calls to scm_must_malloc/realloc/free.
2595 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2596 2. objects reallocated by scm_must_realloc has been allocated by
2598 3. reallocated objects are reallocated with the same what string
2600 But, most importantly, it records the number of allocated objects of
2601 each kind. This is useful when searching for memory leaks.
2603 A Guile compiled with this option provides the primitive
2604 `malloc-stats' which returns an alist with pairs of kind and the
2605 number of objects of that kind.
2607 ** All includes are now referenced relative to the root directory
2609 Since some users have had problems with mixups between Guile and
2610 system headers, we have decided to always refer to Guile headers via
2611 their parent directories. This essentially creates a "private name
2612 space" for Guile headers. This means that the compiler only is given
2613 -I options for the root build and root source directory.
2615 ** Header files kw.h and genio.h have been removed.
2617 ** The module (ice-9 getopt-gnu-style) has been removed.
2619 ** New module (ice-9 documentation)
2621 Implements the interface to documentation strings associated with
2624 ** New module (ice-9 time)
2626 Provides a macro `time', which displays execution time of a given form.
2628 ** New module (ice-9 history)
2630 Loading this module enables value history in the repl.
2632 * Changes to the stand-alone interpreter
2634 ** New command line option --debug
2636 Start Guile with debugging evaluator and backtraces enabled.
2638 This is useful when debugging your .guile init file or scripts.
2640 ** New help facility
2642 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2643 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2644 (help 'NAME) gives documentation for NAME, even if it is not an object
2645 (help ,EXPR) gives documentation for object returned by EXPR
2646 (help (my module)) gives module commentary for `(my module)'
2647 (help) gives this text
2649 `help' searches among bindings exported from loaded modules, while
2650 `apropos' searches among bindings visible from the "current" module.
2652 Examples: (help help)
2654 (help "output-string")
2656 ** `help' and `apropos' now prints full module names
2658 ** Dynamic linking now uses libltdl from the libtool package.
2660 The old system dependent code for doing dynamic linking has been
2661 replaced with calls to the libltdl functions which do all the hairy
2664 The major improvement is that you can now directly pass libtool
2665 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2666 will be able to do the best shared library job you can get, via
2669 The way dynamic libraries are found has changed and is not really
2670 portable across platforms, probably. It is therefore recommended to
2671 use absolute filenames when possible.
2673 If you pass a filename without an extension to `dynamic-link', it will
2674 try a few appropriate ones. Thus, the most platform ignorant way is
2675 to specify a name like "libfoo", without any directories and
2678 ** Guile COOP threads are now compatible with LinuxThreads
2680 Previously, COOP threading wasn't possible in applications linked with
2681 Linux POSIX threads due to their use of the stack pointer to find the
2682 thread context. This has now been fixed with a workaround which uses
2683 the pthreads to allocate the stack.
2685 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2687 ** Positions of erring expression in scripts
2689 With version 1.3.4, the location of the erring expression in Guile
2690 scipts is no longer automatically reported. (This should have been
2691 documented before the 1.3.4 release.)
2693 You can get this information by enabling recording of positions of
2694 source expressions and running the debugging evaluator. Put this at
2695 the top of your script (or in your "site" file):
2697 (read-enable 'positions)
2698 (debug-enable 'debug)
2700 ** Backtraces in scripts
2702 It is now possible to get backtraces in scripts.
2706 (debug-enable 'debug 'backtrace)
2708 at the top of the script.
2710 (The first options enables the debugging evaluator.
2711 The second enables backtraces.)
2713 ** Part of module system symbol lookup now implemented in C
2715 The eval closure of most modules is now implemented in C. Since this
2716 was one of the bottlenecks for loading speed, Guile now loads code
2717 substantially faster than before.
2719 ** Attempting to get the value of an unbound variable now produces
2720 an exception with a key of 'unbound-variable instead of 'misc-error.
2722 ** The initial default output port is now unbuffered if it's using a
2723 tty device. Previously in this situation it was line-buffered.
2725 ** New hook: after-gc-hook
2727 after-gc-hook takes over the role of gc-thunk. This hook is run at
2728 the first SCM_TICK after a GC. (Thus, the code is run at the same
2729 point during evaluation as signal handlers.)
2731 Note that this hook should be used only for diagnostic and debugging
2732 purposes. It is not certain that it will continue to be well-defined
2733 when this hook is run in the future.
2735 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2736 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2738 ** Improvements to garbage collector
2740 Guile 1.4 has a new policy for triggering heap allocation and
2741 determining the sizes of heap segments. It fixes a number of problems
2744 1. The new policy can handle two separate pools of cells
2745 (2-word/4-word) better. (The old policy would run wild, allocating
2746 more and more memory for certain programs.)
2748 2. The old code would sometimes allocate far too much heap so that the
2749 Guile process became gigantic. The new code avoids this.
2751 3. The old code would sometimes allocate too little so that few cells
2752 were freed at GC so that, in turn, too much time was spent in GC.
2754 4. The old code would often trigger heap allocation several times in a
2755 row. (The new scheme predicts how large the segments needs to be
2756 in order not to need further allocation.)
2758 All in all, the new GC policy will make larger applications more
2761 The new GC scheme also is prepared for POSIX threading. Threads can
2762 allocate private pools of cells ("clusters") with just a single
2763 function call. Allocation of single cells from such a cluster can
2764 then proceed without any need of inter-thread synchronization.
2766 ** New environment variables controlling GC parameters
2768 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2771 Allocation of 2-word cell heaps:
2773 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2776 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2777 GC in percent of total heap size
2780 Allocation of 4-word cell heaps
2781 (used for real numbers and misc other objects):
2783 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2785 (See entry "Way for application to customize GC parameters" under
2786 section "Changes to the scm_ interface" below.)
2788 ** Guile now implements reals using 4-word cells
2790 This speeds up computation with reals. (They were earlier allocated
2791 with `malloc'.) There is still some room for optimizations, however.
2793 ** Some further steps toward POSIX thread support have been taken
2795 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2796 don't have much effect any longer, and many of them will be removed in
2800 are only handled at the top of the evaluator loop, immediately after
2801 I/O, and in scm_equalp.
2803 *** The GC can allocate thread private pools of pairs.
2805 * Changes to Scheme functions and syntax
2807 ** close-input-port and close-output-port are now R5RS
2809 These procedures have been turned into primitives and have R5RS behaviour.
2811 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2813 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2814 extended by the more sophisticated version in (ice-9 format)
2816 (simple-format port message . args)
2817 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2818 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2819 the escapes are replaced with corresponding members of ARGS:
2820 ~A formats using `display' and ~S formats using `write'.
2821 If DESTINATION is #t, then use the `current-output-port',
2822 if DESTINATION is #f, then return a string containing the formatted text.
2823 Does not add a trailing newline."
2825 ** string-ref: the second argument is no longer optional.
2827 ** string, list->string: no longer accept strings in their arguments,
2828 only characters, for compatibility with R5RS.
2830 ** New procedure: port-closed? PORT
2831 Returns #t if PORT is closed or #f if it is open.
2833 ** Deprecated: list*
2835 The list* functionality is now provided by cons* (SRFI-1 compliant)
2837 ** New procedure: cons* ARG1 ARG2 ... ARGn
2839 Like `list', but the last arg provides the tail of the constructed list,
2840 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2842 Requires at least one argument. If given one argument, that argument
2843 is returned as result.
2845 This function is called `list*' in some other Schemes and in Common LISP.
2847 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2849 ** New procedure: object-documentation OBJECT
2851 Returns the documentation string associated with OBJECT. The
2852 procedure uses a caching mechanism so that subsequent lookups are
2855 Exported by (ice-9 documentation).
2857 ** module-name now returns full names of modules
2859 Previously, only the last part of the name was returned (`session' for
2860 `(ice-9 session)'). Ex: `(ice-9 session)'.
2862 * Changes to the gh_ interface
2864 ** Deprecated: gh_int2scmb
2866 Use gh_bool2scm instead.
2868 * Changes to the scm_ interface
2870 ** Guile primitives now carry docstrings!
2872 Thanks to Greg Badros!
2874 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2876 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2877 macros and must contain a docstring that is extracted into foo.doc using a new
2878 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2880 However, a major overhaul of these macros is scheduled for the next release of
2883 ** Guile primitives use a new technique for validation of arguments
2885 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2886 the readability of argument checking.
2888 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2890 ** New macros: SCM_PACK, SCM_UNPACK
2892 Compose/decompose an SCM value.
2894 The SCM type is now treated as an abstract data type and may be defined as a
2895 long, a void* or as a struct, depending on the architecture and compile time
2896 options. This makes it easier to find several types of bugs, for example when
2897 SCM values are treated as integers without conversion. Values of the SCM type
2898 should be treated as "atomic" values. These macros are used when
2899 composing/decomposing an SCM value, either because you want to access
2900 individual bits, or because you want to treat it as an integer value.
2902 E.g., in order to set bit 7 in an SCM value x, use the expression
2904 SCM_PACK (SCM_UNPACK (x) | 0x80)
2906 ** The name property of hooks is deprecated.
2907 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2909 You can emulate this feature by using object properties.
2911 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2912 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2915 These macros will be removed in a future release of Guile.
2917 ** The following types, functions and macros from numbers.h are deprecated:
2918 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2919 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2921 ** Port internals: the rw_random variable in the scm_port structure
2922 must be set to non-zero in any random access port. In recent Guile
2923 releases it was only set for bidirectional random-access ports.
2925 ** Port internals: the seek ptob procedure is now responsible for
2926 resetting the buffers if required. The change was made so that in the
2927 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2928 the fport and strport ptobs can avoid resetting the buffers,
2929 in particular to avoid discarding unread chars. An existing port
2930 type can be fixed by adding something like the following to the
2931 beginning of the ptob seek procedure:
2933 if (pt->rw_active == SCM_PORT_READ)
2934 scm_end_input (object);
2935 else if (pt->rw_active == SCM_PORT_WRITE)
2936 ptob->flush (object);
2938 although to actually avoid resetting the buffers and discard unread
2939 chars requires further hacking that depends on the characteristics
2942 ** Deprecated functions: scm_fseek, scm_tag
2944 These functions are no longer used and will be removed in a future version.
2946 ** The scm_sysmissing procedure is no longer used in libguile.
2947 Unless it turns out to be unexpectedly useful to somebody, it will be
2948 removed in a future version.
2950 ** The format of error message strings has changed
2952 The two C procedures: scm_display_error and scm_error, as well as the
2953 primitive `scm-error', now use scm_simple_format to do their work.
2954 This means that the message strings of all code must be updated to use
2955 ~A where %s was used before, and ~S where %S was used before.
2957 During the period when there still are a lot of old Guiles out there,
2958 you might want to support both old and new versions of Guile.
2960 There are basically two methods to achieve this. Both methods use
2963 AC_CHECK_FUNCS(scm_simple_format)
2965 in your configure.in.
2967 Method 1: Use the string concatenation features of ANSI C's
2972 #ifdef HAVE_SCM_SIMPLE_FORMAT
2978 Then represent each of your error messages using a preprocessor macro:
2980 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2984 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2985 (define make-message string-append)
2987 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2989 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2993 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2998 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3002 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3004 Don't use the functions coop_mutex_init and
3005 coop_condition_variable_init. They will change.
3007 Use scm_mutex_init and scm_cond_init instead.
3009 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3010 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3011 COND, as `scm_cond_wait' does, but it also bounds the duration
3012 of the wait. If COND has not been signaled before time ABSTIME,
3013 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3014 returns the error code `ETIMEDOUT'.
3016 The ABSTIME parameter specifies an absolute time, with the same
3017 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3018 to 00:00:00 GMT, January 1, 1970.
3020 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3021 `scm_cond_broadcast' restarts all the threads that are waiting
3022 on the condition variable COND. Nothing happens if no threads are
3025 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3026 `scm_key_create' allocates a new TSD key. The key is stored in
3027 the location pointed to by KEY. There is no limit on the number
3028 of keys allocated at a given time. The value initially associated
3029 with the returned key is `NULL' in all currently executing threads.
3031 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3032 function associated with the key. When a thread terminates,
3033 DESTR_FUNCTION is called on the value associated with the key in
3034 that thread. The DESTR_FUNCTION is not called if a key is deleted
3035 with `scm_key_delete' or a value is changed with
3036 `scm_setspecific'. The order in which destructor functions are
3037 called at thread termination time is unspecified.
3039 Destructors are not yet implemented.
3041 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3042 `scm_setspecific' changes the value associated with KEY in the
3043 calling thread, storing the given POINTER instead.
3045 ** New function: scm_getspecific (scm_key_t KEY)
3046 `scm_getspecific' returns the value currently associated with
3047 KEY in the calling thread.
3049 ** New function: scm_key_delete (scm_key_t KEY)
3050 `scm_key_delete' deallocates a TSD key. It does not check
3051 whether non-`NULL' values are associated with that key in the
3052 currently executing threads, nor call the destructor function
3053 associated with the key.
3055 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3057 Initialize a C level hook HOOK with associated HOOK_DATA and type
3058 TYPE. (See scm_c_hook_run ().)
3060 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3062 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3063 is true, add it last, otherwise first. The same FUNC can be added
3064 multiple times if FUNC_DATA differ and vice versa.
3066 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3068 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3069 function is only removed if both FUNC and FUNC_DATA matches.
3071 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3073 Run hook HOOK passing DATA to the hook functions.
3075 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3076 returned is undefined.
3078 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3079 returns a non-NULL value. This value is returned as the result of
3080 scm_c_hook_run. If all functions return NULL, NULL is returned.
3082 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3083 returns a NULL value, and NULL is returned. If all functions returns
3084 a non-NULL value, the last value is returned.
3086 ** New C level GC hooks
3088 Five new C level hooks has been added to the garbage collector.
3090 scm_before_gc_c_hook
3093 are run before locking and after unlocking the heap. The system is
3094 thus in a mode where evaluation can take place. (Except that
3095 scm_before_gc_c_hook must not allocate new cells.)
3097 scm_before_mark_c_hook
3098 scm_before_sweep_c_hook
3099 scm_after_sweep_c_hook
3101 are run when the heap is locked. These are intended for extension of
3102 the GC in a modular fashion. Examples are the weaks and guardians
3105 ** Way for application to customize GC parameters
3107 The application can set up other default values for the GC heap
3108 allocation parameters
3110 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3111 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3112 GUILE_MAX_SEGMENT_SIZE,
3116 scm_default_init_heap_size_1, scm_default_min_yield_1,
3117 scm_default_init_heap_size_2, scm_default_min_yield_2,
3118 scm_default_max_segment_size
3120 respectively before callong scm_boot_guile.
3122 (See entry "New environment variables ..." in section
3123 "Changes to the stand-alone interpreter" above.)
3125 ** scm_protect_object/scm_unprotect_object now nest
3127 This means that you can call scm_protect_object multiple times on an
3128 object and count on the object being protected until
3129 scm_unprotect_object has been call the same number of times.
3131 The functions also have better time complexity.
3133 Still, it is usually possible to structure the application in a way
3134 that you don't need to use these functions. For example, if you use a
3135 protected standard Guile list to keep track of live objects rather
3136 than some custom data type, objects will die a natural death when they
3137 are no longer needed.
3139 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3141 Guile does not provide the float representation for inexact real numbers any
3142 more. Now, only doubles are used to represent inexact real numbers. Further,
3143 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3144 and scm_tc16_complex, respectively.
3146 ** Removed deprecated type scm_smobfuns
3148 ** Removed deprecated function scm_newsmob
3150 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3152 There is an ongoing discussion among the developers whether to
3153 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3154 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3155 until this issue has been settled.
3157 ** Removed deprecated type tag scm_tc16_kw
3159 ** Added type tag scm_tc16_keyword
3161 (This was introduced already in release 1.3.4 but was not documented
3164 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3166 * Changes to system call interfaces:
3168 ** The "select" procedure now tests port buffers for the ability to
3169 provide input or accept output. Previously only the underlying file
3170 descriptors were checked.
3172 ** New variable PIPE_BUF: the maximum number of bytes that can be
3173 atomically written to a pipe.
3175 ** If a facility is not available on the system when Guile is
3176 compiled, the corresponding primitive procedure will not be defined.
3177 Previously it would have been defined but would throw a system-error
3178 exception if called. Exception handlers which catch this case may
3179 need minor modification: an error will be thrown with key
3180 'unbound-variable instead of 'system-error. Alternatively it's
3181 now possible to use `defined?' to check whether the facility is
3184 ** Procedures which depend on the timezone should now give the correct
3185 result on systems which cache the TZ environment variable, even if TZ
3186 is changed without calling tzset.
3188 * Changes to the networking interfaces:
3190 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3191 long integers between network and host format. For now, it's not
3192 particularly convenient to do this kind of thing, but consider:
3194 (define write-network-long
3195 (lambda (value port)
3196 (let ((v (make-uniform-vector 1 1 0)))
3197 (uniform-vector-set! v 0 (htonl value))
3198 (uniform-vector-write v port))))
3200 (define read-network-long
3202 (let ((v (make-uniform-vector 1 1 0)))
3203 (uniform-vector-read! v port)
3204 (ntohl (uniform-vector-ref v 0)))))
3206 ** If inet-aton fails, it now throws an error with key 'misc-error
3207 instead of 'system-error, since errno is not relevant.
3209 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3210 specific keys instead of 'system-error. The latter is inappropriate
3211 since errno will not have been set. The keys are:
3212 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3214 ** sethostent, setnetent, setprotoent, setservent: now take an
3215 optional argument STAYOPEN, which specifies whether the database
3216 remains open after a database entry is accessed randomly (e.g., using
3217 gethostbyname for the hosts database.) The default is #f. Previously
3221 Changes since Guile 1.3.2:
3223 * Changes to the stand-alone interpreter
3227 An initial version of the Guile debugger written by Chris Hanson has
3228 been added. The debugger is still under development but is included
3229 in the distribution anyway since it is already quite useful.
3235 after an error to enter the debugger. Type `help' inside the debugger
3236 for a description of available commands.
3238 If you prefer to have stack frames numbered and printed in
3239 anti-chronological order and prefer up in the stack to be down on the
3240 screen as is the case in gdb, you can put
3242 (debug-enable 'backwards)
3244 in your .guile startup file. (However, this means that Guile can't
3245 use indentation to indicate stack level.)
3247 The debugger is autoloaded into Guile at the first use.
3249 ** Further enhancements to backtraces
3251 There is a new debug option `width' which controls the maximum width
3252 on the screen of printed stack frames. Fancy printing parameters
3253 ("level" and "length" as in Common LISP) are adaptively adjusted for
3254 each stack frame to give maximum information while still fitting
3255 within the bounds. If the stack frame can't be made to fit by
3256 adjusting parameters, it is simply cut off at the end. This is marked
3259 ** Some modules are now only loaded when the repl is started
3261 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3262 regex) are now loaded into (guile-user) only if the repl has been
3263 started. The effect is that the startup time for scripts has been
3264 reduced to 30% of what it was previously.
3266 Correctly written scripts load the modules they require at the top of
3267 the file and should not be affected by this change.
3269 ** Hooks are now represented as smobs
3271 * Changes to Scheme functions and syntax
3273 ** Readline support has changed again.
3275 The old (readline-activator) module is gone. Use (ice-9 readline)
3276 instead, which now contains all readline functionality. So the code
3277 to activate readline is now
3279 (use-modules (ice-9 readline))
3282 This should work at any time, including from the guile prompt.
3284 To avoid confusion about the terms of Guile's license, please only
3285 enable readline for your personal use; please don't make it the
3286 default for others. Here is why we make this rather odd-sounding
3289 Guile is normally licensed under a weakened form of the GNU General
3290 Public License, which allows you to link code with Guile without
3291 placing that code under the GPL. This exception is important to some
3294 However, since readline is distributed under the GNU General Public
3295 License, when you link Guile with readline, either statically or
3296 dynamically, you effectively change Guile's license to the strict GPL.
3297 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3298 which are normally permitted become forbidden. This is a rather
3299 non-obvious consequence of the licensing terms.
3301 So, to make sure things remain clear, please let people choose for
3302 themselves whether to link GPL'd libraries like readline with Guile.
3304 ** regexp-substitute/global has changed slightly, but incompatibly.
3306 If you include a function in the item list, the string of the match
3307 object it receives is the same string passed to
3308 regexp-substitute/global, not some suffix of that string.
3309 Correspondingly, the match's positions are relative to the entire
3310 string, not the suffix.
3312 If the regexp can match the empty string, the way matches are chosen
3313 from the string has changed. regexp-substitute/global recognizes the
3314 same set of matches that list-matches does; see below.
3316 ** New function: list-matches REGEXP STRING [FLAGS]
3318 Return a list of match objects, one for every non-overlapping, maximal
3319 match of REGEXP in STRING. The matches appear in left-to-right order.
3320 list-matches only reports matches of the empty string if there are no
3321 other matches which begin on, end at, or include the empty match's
3324 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3326 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3328 For each match of REGEXP in STRING, apply PROC to the match object,
3329 and the last value PROC returned, or INIT for the first call. Return
3330 the last value returned by PROC. We apply PROC to the matches as they
3331 appear from left to right.
3333 This function recognizes matches according to the same criteria as
3336 Thus, you could define list-matches like this:
3338 (define (list-matches regexp string . flags)
3339 (reverse! (apply fold-matches regexp string '() cons flags)))
3341 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3345 *** New function: hook? OBJ
3347 Return #t if OBJ is a hook, otherwise #f.
3349 *** New function: make-hook-with-name NAME [ARITY]
3351 Return a hook with name NAME and arity ARITY. The default value for
3352 ARITY is 0. The only effect of NAME is that it will appear when the
3353 hook object is printed to ease debugging.
3355 *** New function: hook-empty? HOOK
3357 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3359 *** New function: hook->list HOOK
3361 Return a list of the procedures that are called when run-hook is
3364 ** `map' signals an error if its argument lists are not all the same length.
3366 This is the behavior required by R5RS, so this change is really a bug
3367 fix. But it seems to affect a lot of people's code, so we're
3368 mentioning it here anyway.
3370 ** Print-state handling has been made more transparent
3372 Under certain circumstances, ports are represented as a port with an
3373 associated print state. Earlier, this pair was represented as a pair
3374 (see "Some magic has been added to the printer" below). It is now
3375 indistinguishable (almost; see `get-print-state') from a port on the
3378 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3380 Return a new port with the associated print state PRINT-STATE.
3382 *** New function: get-print-state OUTPUT-PORT
3384 Return the print state associated with this port if it exists,
3385 otherwise return #f.
3387 *** New function: directory-stream? OBJECT
3389 Returns true iff OBJECT is a directory stream --- the sort of object
3390 returned by `opendir'.
3392 ** New function: using-readline?
3394 Return #t if readline is in use in the current repl.
3396 ** structs will be removed in 1.4
3398 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3399 and use GOOPS objects as the fundamental record type.
3401 * Changes to the scm_ interface
3403 ** structs will be removed in 1.4
3405 The entire current struct interface (struct.c, struct.h) will be
3406 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3407 GOOPS objects as the fundamental record type.
3409 ** The internal representation of subr's has changed
3411 Instead of giving a hint to the subr name, the CAR field of the subr
3412 now contains an index to a subr entry in scm_subr_table.
3414 *** New variable: scm_subr_table
3416 An array of subr entries. A subr entry contains the name, properties
3417 and documentation associated with the subr. The properties and
3418 documentation slots are not yet used.
3420 ** A new scheme for "forwarding" calls to a builtin to a generic function
3422 It is now possible to extend the functionality of some Guile
3423 primitives by letting them defer a call to a GOOPS generic function on
3424 argument mismatch. This means that there is no loss of efficiency in
3429 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3430 (define-method + ((x <string>) (y <string>))
3431 (string-append x y))
3433 + will still be as efficient as usual in numerical calculations, but
3434 can also be used for concatenating strings.
3436 Who will be the first one to extend Guile's numerical tower to
3437 rationals? :) [OK, there a few other things to fix before this can
3438 be made in a clean way.]
3440 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3442 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3444 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3446 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3447 a variable GENERIC which can be used by the dispatch macros below.
3449 [This is experimental code which may change soon.]
3451 *** New macros for forwarding control to a generic on arg type error
3453 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3455 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3457 These correspond to the scm_wta function call, and have the same
3458 behaviour until the user has called the GOOPS primitive
3459 `enable-primitive-generic!'. After that, these macros will apply the
3460 generic function GENERIC to the argument(s) instead of calling
3463 [This is experimental code which may change soon.]
3465 *** New macros for argument testing with generic dispatch
3467 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3469 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3471 These correspond to the SCM_ASSERT macro, but will defer control to
3472 GENERIC on error after `enable-primitive-generic!' has been called.
3474 [This is experimental code which may change soon.]
3476 ** New function: SCM scm_eval_body (SCM body, SCM env)
3478 Evaluates the body of a special form.
3480 ** The internal representation of struct's has changed
3482 Previously, four slots were allocated for the procedure(s) of entities
3483 and operators. The motivation for this representation had to do with
3484 the structure of the evaluator, the wish to support tail-recursive
3485 generic functions, and efficiency. Since the generic function
3486 dispatch mechanism has changed, there is no longer a need for such an
3487 expensive representation, and the representation has been simplified.
3489 This should not make any difference for most users.
3491 ** GOOPS support has been cleaned up.
3493 Some code has been moved from eval.c to objects.c and code in both of
3494 these compilation units has been cleaned up and better structured.
3496 *** New functions for applying generic functions
3498 New function: SCM scm_apply_generic (GENERIC, ARGS)
3499 New function: SCM scm_call_generic_0 (GENERIC)
3500 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3501 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3502 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3504 ** Deprecated function: scm_make_named_hook
3506 It is now replaced by:
3508 ** New function: SCM scm_create_hook (const char *name, int arity)
3510 Creates a hook in the same way as make-hook above but also
3511 binds a variable named NAME to it.
3513 This is the typical way of creating a hook from C code.
3515 Currently, the variable is created in the "current" module.
3516 This might change when we get the new module system.
3518 [The behaviour is identical to scm_make_named_hook.]
3522 Changes since Guile 1.3:
3524 * Changes to mailing lists
3526 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3528 See the README file to find current addresses for all the Guile
3531 * Changes to the distribution
3533 ** Readline support is no longer included with Guile by default.
3535 Based on the different license terms of Guile and Readline, we
3536 concluded that Guile should not *by default* cause the linking of
3537 Readline into an application program. Readline support is now offered
3538 as a separate module, which is linked into an application only when
3539 you explicitly specify it.
3541 Although Guile is GNU software, its distribution terms add a special
3542 exception to the usual GNU General Public License (GPL). Guile's
3543 license includes a clause that allows you to link Guile with non-free
3544 programs. We add this exception so as not to put Guile at a
3545 disadvantage vis-a-vis other extensibility packages that support other
3548 In contrast, the GNU Readline library is distributed under the GNU
3549 General Public License pure and simple. This means that you may not
3550 link Readline, even dynamically, into an application unless it is
3551 distributed under a free software license that is compatible the GPL.
3553 Because of this difference in distribution terms, an application that
3554 can use Guile may not be able to use Readline. Now users will be
3555 explicitly offered two independent decisions about the use of these
3558 You can activate the readline support by issuing
3560 (use-modules (readline-activator))
3563 from your ".guile" file, for example.
3565 * Changes to the stand-alone interpreter
3567 ** All builtins now print as primitives.
3568 Previously builtin procedures not belonging to the fundamental subr
3569 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3570 Now, they print as #<primitive-procedure NAME>.
3572 ** Backtraces slightly more intelligible.
3573 gsubr-apply and macro transformer application frames no longer appear
3576 * Changes to Scheme functions and syntax
3578 ** Guile now correctly handles internal defines by rewriting them into
3579 their equivalent letrec. Previously, internal defines would
3580 incrementally add to the innermost environment, without checking
3581 whether the restrictions specified in RnRS were met. This lead to the
3582 correct behaviour when these restriction actually were met, but didn't
3583 catch all illegal uses. Such an illegal use could lead to crashes of
3584 the Guile interpreter or or other unwanted results. An example of
3585 incorrect internal defines that made Guile behave erratically:
3597 The problem with this example is that the definition of `c' uses the
3598 value of `b' directly. This confuses the meoization machine of Guile
3599 so that the second call of `b' (this time in a larger environment that
3600 also contains bindings for `c' and `d') refers to the binding of `c'
3601 instead of `a'. You could also make Guile crash with a variation on
3606 (define (b flag) (if flag a 1))
3607 (define c (1+ (b flag)))
3615 From now on, Guile will issue an `Unbound variable: b' error message
3620 A hook contains a list of functions which should be called on
3621 particular occasions in an existing program. Hooks are used for
3624 A window manager might have a hook before-window-map-hook. The window
3625 manager uses the function run-hooks to call all functions stored in
3626 before-window-map-hook each time a window is mapped. The user can
3627 store functions in the hook using add-hook!.
3629 In Guile, hooks are first class objects.
3631 *** New function: make-hook [N_ARGS]
3633 Return a hook for hook functions which can take N_ARGS arguments.
3634 The default value for N_ARGS is 0.
3636 (See also scm_make_named_hook below.)
3638 *** New function: add-hook! HOOK PROC [APPEND_P]
3640 Put PROC at the beginning of the list of functions stored in HOOK.
3641 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3643 PROC must be able to take the number of arguments specified when the
3646 If PROC already exists in HOOK, then remove it first.
3648 *** New function: remove-hook! HOOK PROC
3650 Remove PROC from the list of functions in HOOK.
3652 *** New function: reset-hook! HOOK
3654 Clear the list of hook functions stored in HOOK.
3656 *** New function: run-hook HOOK ARG1 ...
3658 Run all hook functions stored in HOOK with arguments ARG1 ... .
3659 The number of arguments supplied must correspond to the number given
3660 when the hook was created.
3662 ** The function `dynamic-link' now takes optional keyword arguments.
3663 The only keyword argument that is currently defined is `:global
3664 BOOL'. With it, you can control whether the shared library will be
3665 linked in global mode or not. In global mode, the symbols from the
3666 linked library can be used to resolve references from other
3667 dynamically linked libraries. In non-global mode, the linked
3668 library is essentially invisible and can only be accessed via
3669 `dynamic-func', etc. The default is now to link in global mode.
3670 Previously, the default has been non-global mode.
3672 The `#:global' keyword is only effective on platforms that support
3673 the dlopen family of functions.
3675 ** New function `provided?'
3677 - Function: provided? FEATURE
3678 Return true iff FEATURE is supported by this installation of
3679 Guile. FEATURE must be a symbol naming a feature; the global
3680 variable `*features*' is a list of available features.
3682 ** Changes to the module (ice-9 expect):
3684 *** The expect-strings macro now matches `$' in a regular expression
3685 only at a line-break or end-of-file by default. Previously it would
3686 match the end of the string accumulated so far. The old behaviour
3687 can be obtained by setting the variable `expect-strings-exec-flags'
3690 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3691 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3692 in a regular expression will still match before a line-break or
3693 end-of-file. The default is `regexp/noteol'.
3695 *** The expect-strings macro now uses a variable
3696 `expect-strings-compile-flags' for the flags to be supplied to
3697 `make-regexp'. The default is `regexp/newline', which was previously
3700 *** The expect macro now supplies two arguments to a match procedure:
3701 the current accumulated string and a flag to indicate whether
3702 end-of-file has been reached. Previously only the string was supplied.
3703 If end-of-file is reached, the match procedure will be called an
3704 additional time with the same accumulated string as the previous call
3705 but with the flag set.
3707 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3709 This code, and the documentation for it that appears here, was
3710 borrowed from SLIB, with minor adaptations for Guile.
3712 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3713 An almost complete implementation of Common LISP format description
3714 according to the CL reference book `Common LISP' from Guy L.
3715 Steele, Digital Press. Backward compatible to most of the
3716 available Scheme format implementations.
3718 Returns `#t', `#f' or a string; has side effect of printing
3719 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3720 to the current output port and `#t' is returned. If DESTINATION
3721 is `#f', a formatted string is returned as the result of the call.
3722 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3723 format string; FORMAT-STRING is then the first argument and the
3724 output is returned as a string. If DESTINATION is a number, the
3725 output is to the current error port if available by the
3726 implementation. Otherwise DESTINATION must be an output port and
3729 FORMAT-STRING must be a string. In case of a formatting error
3730 format returns `#f' and prints a message on the current output or
3731 error port. Characters are output as if the string were output by
3732 the `display' function with the exception of those prefixed by a
3733 tilde (~). For a detailed description of the FORMAT-STRING syntax
3734 please consult a Common LISP format reference manual. For a test
3735 suite to verify this format implementation load `formatst.scm'.
3736 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3738 Note: `format' is not reentrant, i.e. only one `format'-call may
3739 be executed at a time.
3742 *** Format Specification (Format version 3.0)
3744 Please consult a Common LISP format reference manual for a detailed
3745 description of the format string syntax. For a demonstration of the
3746 implemented directives see `formatst.scm'.
3748 This implementation supports directive parameters and modifiers (`:'
3749 and `@' characters). Multiple parameters must be separated by a comma
3750 (`,'). Parameters can be numerical parameters (positive or negative),
3751 character parameters (prefixed by a quote character (`''), variable
3752 parameters (`v'), number of rest arguments parameter (`#'), empty and
3753 default parameters. Directive characters are case independent. The
3754 general form of a directive is:
3756 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3758 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3760 *** Implemented CL Format Control Directives
3762 Documentation syntax: Uppercase characters represent the
3763 corresponding control directive characters. Lowercase characters
3764 represent control directive parameter descriptions.
3767 Any (print as `display' does).
3771 `~MINCOL,COLINC,MINPAD,PADCHARA'
3775 S-expression (print as `write' does).
3779 `~MINCOL,COLINC,MINPAD,PADCHARS'
3785 print number sign always.
3788 print comma separated.
3790 `~MINCOL,PADCHAR,COMMACHARD'
3796 print number sign always.
3799 print comma separated.
3801 `~MINCOL,PADCHAR,COMMACHARX'
3807 print number sign always.
3810 print comma separated.
3812 `~MINCOL,PADCHAR,COMMACHARO'
3818 print number sign always.
3821 print comma separated.
3823 `~MINCOL,PADCHAR,COMMACHARB'
3828 `~N,MINCOL,PADCHAR,COMMACHARR'
3832 print a number as a Roman numeral.
3835 print a number as an "old fashioned" Roman numeral.
3838 print a number as an ordinal English number.
3841 print a number as a cardinal English number.
3846 prints `y' and `ies'.
3849 as `~P but jumps 1 argument backward.'
3852 as `~@P but jumps 1 argument backward.'
3857 prints a character as the reader can understand it (i.e. `#\'
3861 prints a character as emacs does (eg. `^C' for ASCII 03).
3864 Fixed-format floating-point (prints a flonum like MMM.NNN).
3865 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3867 If the number is positive a plus sign is printed.
3870 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3871 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3873 If the number is positive a plus sign is printed.
3876 General floating-point (prints a flonum either fixed or
3878 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3880 If the number is positive a plus sign is printed.
3883 Dollars floating-point (prints a flonum in fixed with signs
3885 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3887 If the number is positive a plus sign is printed.
3890 A sign is always printed and appears before the padding.
3893 The sign appears before the padding.
3901 print newline if not at the beginning of the output line.
3903 prints `~&' and then N-1 newlines.
3908 print N page separators.
3918 newline is ignored, white space left.
3921 newline is left, white space ignored.
3926 relative tabulation.
3932 Indirection (expects indirect arguments as a list).
3934 extracts indirect arguments from format arguments.
3937 Case conversion (converts by `string-downcase').
3939 converts by `string-capitalize'.
3942 converts by `string-capitalize-first'.
3945 converts by `string-upcase'.
3948 Argument Jumping (jumps 1 argument forward).
3950 jumps N arguments forward.
3953 jumps 1 argument backward.
3956 jumps N arguments backward.
3959 jumps to the 0th argument.
3962 jumps to the Nth argument (beginning from 0)
3964 `~[STR0~;STR1~;...~;STRN~]'
3965 Conditional Expression (numerical clause conditional).
3967 take argument from N.
3970 true test conditional.
3973 if-else-then conditional.
3979 default clause follows.
3982 Iteration (args come from the next argument (a list)).
3984 at most N iterations.
3987 args from next arg (a list of lists).
3990 args from the rest of arguments.
3993 args from the rest args (lists).
4004 aborts if N <= M <= K
4006 *** Not Implemented CL Format Control Directives
4009 print `#f' as an empty list (see below).
4012 print `#f' as an empty list (see below).
4018 (sorry I don't understand its semantics completely)
4020 *** Extended, Replaced and Additional Control Directives
4022 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4023 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4024 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4025 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4026 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4027 COMMAWIDTH is the number of characters between two comma
4031 print a R4RS complex number as `~F~@Fi' with passed parameters for
4035 Pretty print formatting of an argument for scheme code lists.
4041 Flushes the output if format DESTINATION is a port.
4044 Print a `#\space' character
4046 print N `#\space' characters.
4049 Print a `#\tab' character
4051 print N `#\tab' characters.
4054 Takes N as an integer representation for a character. No arguments
4055 are consumed. N is converted to a character by `integer->char'. N
4056 must be a positive decimal number.
4059 Print out readproof. Prints out internal objects represented as
4060 `#<...>' as strings `"#<...>"' so that the format output can always
4061 be processed by `read'.
4064 Print out readproof. Prints out internal objects represented as
4065 `#<...>' as strings `"#<...>"' so that the format output can always
4066 be processed by `read'.
4069 Prints information and a copyright notice on the format
4072 prints format version.
4075 may also print number strings, i.e. passing a number as a string
4076 and format it accordingly.
4078 *** Configuration Variables
4080 The format module exports some configuration variables to suit the
4081 systems and users needs. There should be no modification necessary for
4082 the configuration that comes with Guile. Format detects automatically
4083 if the running scheme system implements floating point numbers and
4086 format:symbol-case-conv
4087 Symbols are converted by `symbol->string' so the case type of the
4088 printed symbols is implementation dependent.
4089 `format:symbol-case-conv' is a one arg closure which is either
4090 `#f' (no conversion), `string-upcase', `string-downcase' or
4091 `string-capitalize'. (default `#f')
4093 format:iobj-case-conv
4094 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4095 implementation internal objects. (default `#f')
4098 The character prefixing the exponent value in `~E' printing.
4101 *** Compatibility With Other Format Implementations
4107 Downward compatible except for padding support and `~A', `~S',
4108 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4109 `printf' padding support which is completely replaced by the CL
4110 `format' padding style.
4113 Downward compatible except for `~', which is not documented
4114 (ignores all characters inside the format string up to a newline
4115 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4116 numerical and variable parameters and `:/@' modifiers in the CL
4120 Downward compatible except for `~A' and `~S' which print in
4121 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4122 directive parameters or modifiers)).
4125 Downward compatible except for an optional destination parameter:
4126 S2C accepts a format call without a destination which returns a
4127 formatted string. This is equivalent to a #f destination in S2C.
4128 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4129 parameters or modifiers)).
4132 ** Changes to string-handling functions.
4134 These functions were added to support the (ice-9 format) module, above.
4136 *** New function: string-upcase STRING
4137 *** New function: string-downcase STRING
4139 These are non-destructive versions of the existing string-upcase! and
4140 string-downcase! functions.
4142 *** New function: string-capitalize! STRING
4143 *** New function: string-capitalize STRING
4145 These functions convert the first letter of each word in the string to
4148 (string-capitalize "howdy there")
4151 As with the other functions, string-capitalize! modifies the string in
4152 place, while string-capitalize returns a modified copy of its argument.
4154 *** New function: string-ci->symbol STRING
4156 Return a symbol whose name is STRING, but having the same case as if
4157 the symbol had be read by `read'.
4159 Guile can be configured to be sensitive or insensitive to case
4160 differences in Scheme identifiers. If Guile is case-insensitive, all
4161 symbols are converted to lower case on input. The `string-ci->symbol'
4162 function returns a symbol whose name in STRING, transformed as Guile
4163 would if STRING were input.
4165 *** New function: substring-move! STRING1 START END STRING2 START
4167 Copy the substring of STRING1 from START (inclusive) to END
4168 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4169 string, and the source and destination areas may overlap; in all
4170 cases, the function behaves as if all the characters were copied
4173 *** Extended functions: substring-move-left! substring-move-right!
4175 These functions now correctly copy arbitrarily overlapping substrings;
4176 they are both synonyms for substring-move!.
4179 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4181 getopt-long is a function for parsing command-line arguments in a
4182 manner consistent with other GNU programs.
4184 (getopt-long ARGS GRAMMAR)
4185 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4187 ARGS should be a list of strings. Its first element should be the
4188 name of the program; subsequent elements should be the arguments
4189 that were passed to the program on the command line. The
4190 `program-arguments' procedure returns a list of this form.
4192 GRAMMAR is a list of the form:
4193 ((OPTION (PROPERTY VALUE) ...) ...)
4195 Each OPTION should be a symbol. `getopt-long' will accept a
4196 command-line option named `--OPTION'.
4197 Each option can have the following (PROPERTY VALUE) pairs:
4199 (single-char CHAR) --- Accept `-CHAR' as a single-character
4200 equivalent to `--OPTION'. This is how to specify traditional
4202 (required? BOOL) --- If BOOL is true, the option is required.
4203 getopt-long will raise an error if it is not found in ARGS.
4204 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4205 it is #f, it does not; and if it is the symbol
4206 `optional', the option may appear in ARGS with or
4208 (predicate FUNC) --- If the option accepts a value (i.e. you
4209 specified `(value #t)' for this option), then getopt
4210 will apply FUNC to the value, and throw an exception
4211 if it returns #f. FUNC should be a procedure which
4212 accepts a string and returns a boolean value; you may
4213 need to use quasiquotes to get it into GRAMMAR.
4215 The (PROPERTY VALUE) pairs may occur in any order, but each
4216 property may occur only once. By default, options do not have
4217 single-character equivalents, are not required, and do not take
4220 In ARGS, single-character options may be combined, in the usual
4221 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4222 accepts values, then it must be the last option in the
4223 combination; the value is the next argument. So, for example, using
4224 the following grammar:
4225 ((apples (single-char #\a))
4226 (blimps (single-char #\b) (value #t))
4227 (catalexis (single-char #\c) (value #t)))
4228 the following argument lists would be acceptable:
4229 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4230 for "blimps" and "catalexis")
4231 ("-ab" "bang" "-c" "couth") (same)
4232 ("-ac" "couth" "-b" "bang") (same)
4233 ("-abc" "couth" "bang") (an error, since `-b' is not the
4234 last option in its combination)
4236 If an option's value is optional, then `getopt-long' decides
4237 whether it has a value by looking at what follows it in ARGS. If
4238 the next element is a string, and it does not appear to be an
4239 option itself, then that string is the option's value.
4241 The value of a long option can appear as the next element in ARGS,
4242 or it can follow the option name, separated by an `=' character.
4243 Thus, using the same grammar as above, the following argument lists
4245 ("--apples" "Braeburn" "--blimps" "Goodyear")
4246 ("--apples=Braeburn" "--blimps" "Goodyear")
4247 ("--blimps" "Goodyear" "--apples=Braeburn")
4249 If the option "--" appears in ARGS, argument parsing stops there;
4250 subsequent arguments are returned as ordinary arguments, even if
4251 they resemble options. So, in the argument list:
4252 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4253 `getopt-long' will recognize the `apples' option as having the
4254 value "Granny Smith", but it will not recognize the `blimp'
4255 option; it will return the strings "--blimp" and "Goodyear" as
4256 ordinary argument strings.
4258 The `getopt-long' function returns the parsed argument list as an
4259 assocation list, mapping option names --- the symbols from GRAMMAR
4260 --- onto their values, or #t if the option does not accept a value.
4261 Unused options do not appear in the alist.
4263 All arguments that are not the value of any option are returned
4264 as a list, associated with the empty list.
4266 `getopt-long' throws an exception if:
4267 - it finds an unrecognized option in ARGS
4268 - a required option is omitted
4269 - an option that requires an argument doesn't get one
4270 - an option that doesn't accept an argument does get one (this can
4271 only happen using the long option `--opt=value' syntax)
4272 - an option predicate fails
4277 `((lockfile-dir (required? #t)
4280 (predicate ,file-is-directory?))
4281 (verbose (required? #f)
4284 (x-includes (single-char #\x))
4285 (rnet-server (single-char #\y)
4286 (predicate ,string?))))
4288 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4289 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4291 => ((() "foo1" "-fred" "foo2" "foo3")
4292 (rnet-server . "lamprod")
4293 (x-includes . "/usr/include")
4294 (lockfile-dir . "/tmp")
4297 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4299 It will be removed in a few releases.
4301 ** New syntax: lambda*
4302 ** New syntax: define*
4303 ** New syntax: define*-public
4304 ** New syntax: defmacro*
4305 ** New syntax: defmacro*-public
4306 Guile now supports optional arguments.
4308 `lambda*', `define*', `define*-public', `defmacro*' and
4309 `defmacro*-public' are identical to the non-* versions except that
4310 they use an extended type of parameter list that has the following BNF
4311 syntax (parentheses are literal, square brackets indicate grouping,
4312 and `*', `+' and `?' have the usual meaning):
4314 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4315 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4316 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4318 ext-var-decl ::= identifier | ( identifier expression )
4320 The semantics are best illustrated with the following documentation
4321 and examples for `lambda*':
4324 lambda extended for optional and keyword arguments
4326 lambda* creates a procedure that takes optional arguments. These
4327 are specified by putting them inside brackets at the end of the
4328 paramater list, but before any dotted rest argument. For example,
4329 (lambda* (a b #&optional c d . e) '())
4330 creates a procedure with fixed arguments a and b, optional arguments c
4331 and d, and rest argument e. If the optional arguments are omitted
4332 in a call, the variables for them are unbound in the procedure. This
4333 can be checked with the bound? macro.
4335 lambda* can also take keyword arguments. For example, a procedure
4337 (lambda* (#&key xyzzy larch) '())
4338 can be called with any of the argument lists (#:xyzzy 11)
4339 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4340 are given as keywords are bound to values.
4342 Optional and keyword arguments can also be given default values
4343 which they take on when they are not present in a call, by giving a
4344 two-item list in place of an optional argument, for example in:
4345 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4346 foo is a fixed argument, bar is an optional argument with default
4347 value 42, and baz is a keyword argument with default value 73.
4348 Default value expressions are not evaluated unless they are needed
4349 and until the procedure is called.
4351 lambda* now supports two more special parameter list keywords.
4353 lambda*-defined procedures now throw an error by default if a
4354 keyword other than one of those specified is found in the actual
4355 passed arguments. However, specifying #&allow-other-keys
4356 immediately after the kyword argument declarations restores the
4357 previous behavior of ignoring unknown keywords. lambda* also now
4358 guarantees that if the same keyword is passed more than once, the
4359 last one passed is the one that takes effect. For example,
4360 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4361 #:heads 37 #:tails 42 #:heads 99)
4362 would result in (99 47) being displayed.
4364 #&rest is also now provided as a synonym for the dotted syntax rest
4365 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4366 all respects to lambda*. This is provided for more similarity to DSSSL,
4367 MIT-Scheme and Kawa among others, as well as for refugees from other
4370 Further documentation may be found in the optargs.scm file itself.
4372 The optional argument module also exports the macros `let-optional',
4373 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4374 are not documented here because they may be removed in the future, but
4375 full documentation is still available in optargs.scm.
4377 ** New syntax: and-let*
4378 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4380 Syntax: (land* (<clause> ...) <body> ...)
4381 Each <clause> should have one of the following forms:
4382 (<variable> <expression>)
4385 Each <variable> or <bound-variable> should be an identifier. Each
4386 <expression> should be a valid expression. The <body> should be a
4387 possibly empty sequence of expressions, like the <body> of a
4390 Semantics: A LAND* expression is evaluated by evaluating the
4391 <expression> or <bound-variable> of each of the <clause>s from
4392 left to right. The value of the first <expression> or
4393 <bound-variable> that evaluates to a false value is returned; the
4394 remaining <expression>s and <bound-variable>s are not evaluated.
4395 The <body> forms are evaluated iff all the <expression>s and
4396 <bound-variable>s evaluate to true values.
4398 The <expression>s and the <body> are evaluated in an environment
4399 binding each <variable> of the preceding (<variable> <expression>)
4400 clauses to the value of the <expression>. Later bindings
4401 shadow earlier bindings.
4403 Guile's and-let* macro was contributed by Michael Livshin.
4405 ** New sorting functions
4407 *** New function: sorted? SEQUENCE LESS?
4408 Returns `#t' when the sequence argument is in non-decreasing order
4409 according to LESS? (that is, there is no adjacent pair `... x y
4410 ...' for which `(less? y x)').
4412 Returns `#f' when the sequence contains at least one out-of-order
4413 pair. It is an error if the sequence is neither a list nor a
4416 *** New function: merge LIST1 LIST2 LESS?
4417 LIST1 and LIST2 are sorted lists.
4418 Returns the sorted list of all elements in LIST1 and LIST2.
4420 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4421 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4422 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4423 (Here "<" should read "comes before".)
4425 *** New procedure: merge! LIST1 LIST2 LESS?
4426 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4427 the result. If the code is compiled, and LESS? constructs no new
4428 pairs, no pairs at all will be allocated. The first pair of the
4429 result will be either the first pair of LIST1 or the first pair of
4432 *** New function: sort SEQUENCE LESS?
4433 Accepts either a list or a vector, and returns a new sequence
4434 which is sorted. The new sequence is the same type as the input.
4435 Always `(sorted? (sort sequence less?) less?)'. The original
4436 sequence is not altered in any way. The new sequence shares its
4437 elements with the old one; no elements are copied.
4439 *** New procedure: sort! SEQUENCE LESS
4440 Returns its sorted result in the original boxes. No new storage is
4441 allocated at all. Proper usage: (set! slist (sort! slist <))
4443 *** New function: stable-sort SEQUENCE LESS?
4444 Similar to `sort' but stable. That is, if "equal" elements are
4445 ordered a < b in the original sequence, they will have the same order
4448 *** New function: stable-sort! SEQUENCE LESS?
4449 Similar to `sort!' but stable.
4450 Uses temporary storage when sorting vectors.
4452 *** New functions: sort-list, sort-list!
4453 Added for compatibility with scsh.
4455 ** New built-in random number support
4457 *** New function: random N [STATE]
4458 Accepts a positive integer or real N and returns a number of the
4459 same type between zero (inclusive) and N (exclusive). The values
4460 returned have a uniform distribution.
4462 The optional argument STATE must be of the type produced by
4463 `copy-random-state' or `seed->random-state'. It defaults to the value
4464 of the variable `*random-state*'. This object is used to maintain the
4465 state of the pseudo-random-number generator and is altered as a side
4466 effect of the `random' operation.
4468 *** New variable: *random-state*
4469 Holds a data structure that encodes the internal state of the
4470 random-number generator that `random' uses by default. The nature
4471 of this data structure is implementation-dependent. It may be
4472 printed out and successfully read back in, but may or may not
4473 function correctly as a random-number state object in another
4476 *** New function: copy-random-state [STATE]
4477 Returns a new object of type suitable for use as the value of the
4478 variable `*random-state*' and as a second argument to `random'.
4479 If argument STATE is given, a copy of it is returned. Otherwise a
4480 copy of `*random-state*' is returned.
4482 *** New function: seed->random-state SEED
4483 Returns a new object of type suitable for use as the value of the
4484 variable `*random-state*' and as a second argument to `random'.
4485 SEED is a string or a number. A new state is generated and
4486 initialized using SEED.
4488 *** New function: random:uniform [STATE]
4489 Returns an uniformly distributed inexact real random number in the
4490 range between 0 and 1.
4492 *** New procedure: random:solid-sphere! VECT [STATE]
4493 Fills VECT with inexact real random numbers the sum of whose
4494 squares is less than 1.0. Thinking of VECT as coordinates in
4495 space of dimension N = `(vector-length VECT)', the coordinates are
4496 uniformly distributed within the unit N-shere. The sum of the
4497 squares of the numbers is returned. VECT can be either a vector
4498 or a uniform vector of doubles.
4500 *** New procedure: random:hollow-sphere! VECT [STATE]
4501 Fills VECT with inexact real random numbers the sum of whose squares
4502 is equal to 1.0. Thinking of VECT as coordinates in space of
4503 dimension n = `(vector-length VECT)', the coordinates are uniformly
4504 distributed over the surface of the unit n-shere. VECT can be either
4505 a vector or a uniform vector of doubles.
4507 *** New function: random:normal [STATE]
4508 Returns an inexact real in a normal distribution with mean 0 and
4509 standard deviation 1. For a normal distribution with mean M and
4510 standard deviation D use `(+ M (* D (random:normal)))'.
4512 *** New procedure: random:normal-vector! VECT [STATE]
4513 Fills VECT with inexact real random numbers which are independent and
4514 standard normally distributed (i.e., with mean 0 and variance 1).
4515 VECT can be either a vector or a uniform vector of doubles.
4517 *** New function: random:exp STATE
4518 Returns an inexact real in an exponential distribution with mean 1.
4519 For an exponential distribution with mean U use (* U (random:exp)).
4521 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4523 These functions now operate on numbers in the range of a C unsigned
4526 These functions used to operate on numbers in the range of a C signed
4527 long; however, this seems inappropriate, because Guile integers don't
4530 ** New function: make-guardian
4531 This is an implementation of guardians as described in
4532 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4533 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4534 Programming Language Design and Implementation, June 1993
4535 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4537 ** New functions: delq1!, delv1!, delete1!
4538 These procedures behave similar to delq! and friends but delete only
4539 one object if at all.
4541 ** New function: unread-string STRING PORT
4542 Unread STRING to PORT, that is, push it back onto the port so that
4543 next read operation will work on the pushed back characters.
4545 ** unread-char can now be called multiple times
4546 If unread-char is called multiple times, the unread characters will be
4547 read again in last-in first-out order.
4549 ** the procedures uniform-array-read! and uniform-array-write! now
4550 work on any kind of port, not just ports which are open on a file.
4552 ** Now 'l' in a port mode requests line buffering.
4554 ** The procedure truncate-file now works on string ports as well
4555 as file ports. If the size argument is omitted, the current
4556 file position is used.
4558 ** new procedure: seek PORT/FDES OFFSET WHENCE
4559 The arguments are the same as for the old fseek procedure, but it
4560 works on string ports as well as random-access file ports.
4562 ** the fseek procedure now works on string ports, since it has been
4563 redefined using seek.
4565 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4566 size is not supplied.
4568 ** the newline procedure no longer flushes the port if it's not
4569 line-buffered: previously it did if it was the current output port.
4571 ** open-pipe and close-pipe are no longer primitive procedures, but
4572 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4574 ** the freopen procedure has been removed.
4576 ** new procedure: drain-input PORT
4577 Drains PORT's read buffers (including any pushed-back characters)
4578 and returns the contents as a single string.
4580 ** New function: map-in-order PROC LIST1 LIST2 ...
4581 Version of `map' which guarantees that the procedure is applied to the
4582 lists in serial order.
4584 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4585 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4586 now obsolete and will go away in release 1.5.
4588 ** New syntax: collect BODY1 ...
4589 Version of `begin' which returns a list of the results of the body
4590 forms instead of the result of the last body form. In contrast to
4591 `begin', `collect' allows an empty body.
4593 ** New functions: read-history FILENAME, write-history FILENAME
4594 Read/write command line history from/to file. Returns #t on success
4595 and #f if an error occured.
4597 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4599 These procedures return a list of definitions available in the specified
4600 argument, a relative module reference. In the case of no argument,
4601 `(current-module)' is now consulted for definitions to return, instead
4602 of simply returning #f, the former behavior.
4604 ** The #/ syntax for lists is no longer supported.
4606 Earlier versions of Scheme accepted this syntax, but printed a
4609 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4611 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4614 * Changes to the gh_ interface
4618 Now takes a second argument which is the result array. If this
4619 pointer is NULL, a new array is malloced (the old behaviour).
4621 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4622 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4626 * Changes to the scm_ interface
4628 ** Function: scm_make_named_hook (char* name, int n_args)
4630 Creates a hook in the same way as make-hook above but also
4631 binds a variable named NAME to it.
4633 This is the typical way of creating a hook from C code.
4635 Currently, the variable is created in the "current" module. This
4636 might change when we get the new module system.
4638 ** The smob interface
4640 The interface for creating smobs has changed. For documentation, see
4641 data-rep.info (made from guile-core/doc/data-rep.texi).
4643 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4645 >>> This function will be removed in 1.3.4. <<<
4649 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4650 This function adds a new smob type, named NAME, with instance size
4651 SIZE to the system. The return value is a tag that is used in
4652 creating instances of the type. If SIZE is 0, then no memory will
4653 be allocated when instances of the smob are created, and nothing
4654 will be freed by the default free function.
4656 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4657 This function sets the smob marking procedure for the smob type
4658 specified by the tag TC. TC is the tag returned by
4659 `scm_make_smob_type'.
4661 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4662 This function sets the smob freeing procedure for the smob type
4663 specified by the tag TC. TC is the tag returned by
4664 `scm_make_smob_type'.
4666 *** Function: void scm_set_smob_print (tc, print)
4668 - Function: void scm_set_smob_print (long tc,
4669 scm_sizet (*print) (SCM,
4673 This function sets the smob printing procedure for the smob type
4674 specified by the tag TC. TC is the tag returned by
4675 `scm_make_smob_type'.
4677 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4678 This function sets the smob equality-testing predicate for the
4679 smob type specified by the tag TC. TC is the tag returned by
4680 `scm_make_smob_type'.
4682 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4683 Make VALUE contain a smob instance of the type with type code TC and
4684 smob data DATA. VALUE must be previously declared as C type `SCM'.
4686 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4687 This macro expands to a block of code that creates a smob instance
4688 of the type with type code TC and smob data DATA, and returns that
4689 `SCM' value. It should be the last piece of code in a block.
4691 ** The interfaces for using I/O ports and implementing port types
4692 (ptobs) have changed significantly. The new interface is based on
4693 shared access to buffers and a new set of ptob procedures.
4695 *** scm_newptob has been removed
4699 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4701 - Function: SCM scm_make_port_type (char *type_name,
4702 int (*fill_buffer) (SCM port),
4703 void (*write_flush) (SCM port));
4705 Similarly to the new smob interface, there is a set of function
4706 setters by which the user can customize the behaviour of his port
4707 type. See ports.h (scm_set_port_XXX).
4709 ** scm_strport_to_string: New function: creates a new string from
4710 a string port's buffer.
4712 ** Plug in interface for random number generators
4713 The variable `scm_the_rng' in random.c contains a value and three
4714 function pointers which together define the current random number
4715 generator being used by the Scheme level interface and the random
4716 number library functions.
4718 The user is free to replace the default generator with the generator
4721 *** Variable: size_t scm_the_rng.rstate_size
4722 The size of the random state type used by the current RNG
4725 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4726 Given the random STATE, return 32 random bits.
4728 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4729 Seed random state STATE using string S of length N.
4731 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4732 Given random state STATE, return a malloced copy.
4735 The default RNG is the MWC (Multiply With Carry) random number
4736 generator described by George Marsaglia at the Department of
4737 Statistics and Supercomputer Computations Research Institute, The
4738 Florida State University (http://stat.fsu.edu/~geo).
4740 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4741 passes all tests in the DIEHARD test suite
4742 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4743 costs one multiply and one add on platforms which either supports long
4744 longs (gcc does this on most systems) or have 64 bit longs. The cost
4745 is four multiply on other systems but this can be optimized by writing
4746 scm_i_uniform32 in assembler.
4748 These functions are provided through the scm_the_rng interface for use
4749 by libguile and the application.
4751 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4752 Given the random STATE, return 32 random bits.
4753 Don't use this function directly. Instead go through the plugin
4754 interface (see "Plug in interface" above).
4756 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4757 Initialize STATE using SEED of length N.
4759 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4760 Return a malloc:ed copy of STATE. This function can easily be re-used
4761 in the interfaces to other RNGs.
4763 ** Random number library functions
4764 These functions use the current RNG through the scm_the_rng interface.
4765 It might be a good idea to use these functions from your C code so
4766 that only one random generator is used by all code in your program.
4768 The default random state is stored in:
4770 *** Variable: SCM scm_var_random_state
4771 Contains the vcell of the Scheme variable "*random-state*" which is
4772 used as default state by all random number functions in the Scheme
4777 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4779 *** Function: scm_rstate *scm_c_default_rstate (void)
4780 This is a convenience function which returns the value of
4781 scm_var_random_state. An error message is generated if this value
4782 isn't a random state.
4784 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4785 Make a new random state from the string SEED of length LENGTH.
4787 It is generally not a good idea to use multiple random states in a
4788 program. While subsequent random numbers generated from one random
4789 state are guaranteed to be reasonably independent, there is no such
4790 guarantee for numbers generated from different random states.
4792 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4793 Return 32 random bits.
4795 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4796 Return a sample from the uniform(0,1) distribution.
4798 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4799 Return a sample from the normal(0,1) distribution.
4801 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4802 Return a sample from the exp(1) distribution.
4804 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4805 Return a sample from the discrete uniform(0,M) distribution.
4807 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4808 Return a sample from the discrete uniform(0,M) distribution.
4809 M must be a bignum object. The returned value may be an INUM.
4813 Changes in Guile 1.3 (released Monday, October 19, 1998):
4815 * Changes to the distribution
4817 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4818 To avoid conflicts, programs should name environment variables after
4819 themselves, except when there's a common practice establishing some
4822 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4823 giving the former precedence, and printing a warning message if the
4824 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4826 ** The header files related to multi-byte characters have been removed.
4827 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4828 which referred to these explicitly will probably need to be rewritten,
4829 since the support for the variant string types has been removed; see
4832 ** The header files append.h and sequences.h have been removed. These
4833 files implemented non-R4RS operations which would encourage
4834 non-portable programming style and less easy-to-read code.
4836 * Changes to the stand-alone interpreter
4838 ** New procedures have been added to implement a "batch mode":
4840 *** Function: batch-mode?
4842 Returns a boolean indicating whether the interpreter is in batch
4845 *** Function: set-batch-mode?! ARG
4847 If ARG is true, switches the interpreter to batch mode. The `#f'
4848 case has not been implemented.
4850 ** Guile now provides full command-line editing, when run interactively.
4851 To use this feature, you must have the readline library installed.
4852 The Guile build process will notice it, and automatically include
4855 The readline library is available via anonymous FTP from any GNU
4856 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4858 ** the-last-stack is now a fluid.
4860 * Changes to the procedure for linking libguile with your programs
4862 ** You can now use the `guile-config' utility to build programs that use Guile.
4864 Guile now includes a command-line utility called `guile-config', which
4865 can provide information about how to compile and link programs that
4868 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4869 You should include this command's output on the command line you use
4870 to compile C or C++ code that #includes the Guile header files. It's
4871 usually just a `-I' flag to help the compiler find the Guile headers.
4874 *** `guile-config link' prints any linker flags necessary to link with Guile.
4876 This command writes to its standard output a list of flags which you
4877 must pass to the linker to link your code against the Guile library.
4878 The flags include '-lguile' itself, any other libraries the Guile
4879 library depends upon, and any `-L' flags needed to help the linker
4880 find those libraries.
4882 For example, here is a Makefile rule that builds a program named 'foo'
4883 from the object files ${FOO_OBJECTS}, and links them against Guile:
4886 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4888 Previous Guile releases recommended that you use autoconf to detect
4889 which of a predefined set of libraries were present on your system.
4890 It is more robust to use `guile-config', since it records exactly which
4891 libraries the installed Guile library requires.
4893 This was originally called `build-guile', but was renamed to
4894 `guile-config' before Guile 1.3 was released, to be consistent with
4895 the analogous script for the GTK+ GUI toolkit, which is called
4899 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4901 If you are using the GNU autoconf package to configure your program,
4902 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4903 (described above) and gather the necessary values for use in your
4906 The GUILE_FLAGS macro expands to configure script code which runs the
4907 `guile-config' script, to find out where Guile's header files and
4908 libraries are installed. It sets two variables, marked for
4909 substitution, as by AC_SUBST.
4911 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4912 code that uses Guile header files. This is almost always just a
4915 GUILE_LDFLAGS --- flags to pass to the linker to link a
4916 program against Guile. This includes `-lguile' for the Guile
4917 library itself, any libraries that Guile itself requires (like
4918 -lqthreads), and so on. It may also include a -L flag to tell the
4919 compiler where to find the libraries.
4921 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4922 directory of the Guile distribution. You can copy it into your
4923 package's aclocal.m4 file, and then use it in your configure.in file.
4925 If you are using the `aclocal' program, distributed with GNU automake,
4926 to maintain your aclocal.m4 file, the Guile installation process
4927 installs guile.m4 where aclocal will find it. All you need to do is
4928 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4929 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4933 * Changes to Scheme functions and syntax
4935 ** Multi-byte strings have been removed, as have multi-byte and wide
4936 ports. We felt that these were the wrong approach to
4937 internationalization support.
4939 ** New function: readline [PROMPT]
4940 Read a line from the terminal, and allow the user to edit it,
4941 prompting with PROMPT. READLINE provides a large set of Emacs-like
4942 editing commands, lets the user recall previously typed lines, and
4943 works on almost every kind of terminal, including dumb terminals.
4945 READLINE assumes that the cursor is at the beginning of the line when
4946 it is invoked. Thus, you can't print a prompt yourself, and then call
4947 READLINE; you need to package up your prompt as a string, pass it to
4948 the function, and let READLINE print the prompt itself. This is
4949 because READLINE needs to know the prompt's screen width.
4951 For Guile to provide this function, you must have the readline
4952 library, version 2.1 or later, installed on your system. Readline is
4953 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4954 any GNU mirror site.
4956 See also ADD-HISTORY function.
4958 ** New function: add-history STRING
4959 Add STRING as the most recent line in the history used by the READLINE
4960 command. READLINE does not add lines to the history itself; you must
4961 call ADD-HISTORY to make previous input available to the user.
4963 ** The behavior of the read-line function has changed.
4965 This function now uses standard C library functions to read the line,
4966 for speed. This means that it doesn not respect the value of
4967 scm-line-incrementors; it assumes that lines are delimited with
4970 (Note that this is read-line, the function that reads a line of text
4971 from a port, not readline, the function that reads a line from a
4972 terminal, providing full editing capabilities.)
4974 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4976 This module provides some simple argument parsing. It exports one
4979 Function: getopt-gnu-style ARG-LS
4980 Parse a list of program arguments into an alist of option
4983 Each item in the list of program arguments is examined to see if
4984 it meets the syntax of a GNU long-named option. An argument like
4985 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4986 returned alist, where MUMBLE is a keyword object with the same
4987 name as the argument. An argument like `--MUMBLE=FROB' produces
4988 an element of the form (MUMBLE . FROB), where FROB is a string.
4990 As a special case, the returned alist also contains a pair whose
4991 car is the symbol `rest'. The cdr of this pair is a list
4992 containing all the items in the argument list that are not options
4993 of the form mentioned above.
4995 The argument `--' is treated specially: all items in the argument
4996 list appearing after such an argument are not examined, and are
4997 returned in the special `rest' list.
4999 This function does not parse normal single-character switches.
5000 You will need to parse them out of the `rest' list yourself.
5002 ** The read syntax for byte vectors and short vectors has changed.
5004 Instead of #bytes(...), write #y(...).
5006 Instead of #short(...), write #h(...).
5008 This may seem nutty, but, like the other uniform vectors, byte vectors
5009 and short vectors want to have the same print and read syntax (and,
5010 more basic, want to have read syntax!). Changing the read syntax to
5011 use multiple characters after the hash sign breaks with the
5012 conventions used in R5RS and the conventions used for the other
5013 uniform vectors. It also introduces complexity in the current reader,
5014 both on the C and Scheme levels. (The Right solution is probably to
5015 change the syntax and prototypes for uniform vectors entirely.)
5018 ** The new module (ice-9 session) provides useful interactive functions.
5020 *** New procedure: (apropos REGEXP OPTION ...)
5022 Display a list of top-level variables whose names match REGEXP, and
5023 the modules they are imported from. Each OPTION should be one of the
5026 value --- Show the value of each matching variable.
5027 shadow --- Show bindings shadowed by subsequently imported modules.
5028 full --- Same as both `shadow' and `value'.
5032 guile> (apropos "trace" 'full)
5033 debug: trace #<procedure trace args>
5034 debug: untrace #<procedure untrace args>
5035 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5036 the-scm-module: before-backtrace-hook ()
5037 the-scm-module: backtrace #<primitive-procedure backtrace>
5038 the-scm-module: after-backtrace-hook ()
5039 the-scm-module: has-shown-backtrace-hint? #f
5042 ** There are new functions and syntax for working with macros.
5044 Guile implements macros as a special object type. Any variable whose
5045 top-level binding is a macro object acts as a macro. The macro object
5046 specifies how the expression should be transformed before evaluation.
5048 *** Macro objects now print in a reasonable way, resembling procedures.
5050 *** New function: (macro? OBJ)
5051 True iff OBJ is a macro object.
5053 *** New function: (primitive-macro? OBJ)
5054 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5055 macro transformers, implemented in eval.c rather than Scheme code.
5057 Why do we have this function?
5058 - For symmetry with procedure? and primitive-procedure?,
5059 - to allow custom print procedures to tell whether a macro is
5060 primitive, and display it differently, and
5061 - to allow compilers and user-written evaluators to distinguish
5062 builtin special forms from user-defined ones, which could be
5065 *** New function: (macro-type OBJ)
5066 Return a value indicating what kind of macro OBJ is. Possible return
5069 The symbol `syntax' --- a macro created by procedure->syntax.
5070 The symbol `macro' --- a macro created by procedure->macro.
5071 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5072 The boolean #f --- if OBJ is not a macro object.
5074 *** New function: (macro-name MACRO)
5075 Return the name of the macro object MACRO's procedure, as returned by
5078 *** New function: (macro-transformer MACRO)
5079 Return the transformer procedure for MACRO.
5081 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5083 Specify a new macro expander to use in the current module. Each
5084 MODULE is a module name, with the same meaning as in the `use-modules'
5085 form; each named module's exported bindings are added to the current
5086 top-level environment. TRANSFORMER is an expression evaluated in the
5087 resulting environment which must yield a procedure to use as the
5088 module's eval transformer: every expression evaluated in this module
5089 is passed to this function, and the result passed to the Guile
5092 *** macro-eval! is removed. Use local-eval instead.
5094 ** Some magic has been added to the printer to better handle user
5095 written printing routines (like record printers, closure printers).
5097 The problem is that these user written routines must have access to
5098 the current `print-state' to be able to handle fancy things like
5099 detection of circular references. These print-states have to be
5100 passed to the builtin printing routines (display, write, etc) to
5101 properly continue the print chain.
5103 We didn't want to change all existing print code so that it
5104 explicitly passes thru a print state in addition to a port. Instead,
5105 we extented the possible values that the builtin printing routines
5106 accept as a `port'. In addition to a normal port, they now also take
5107 a pair of a normal port and a print-state. Printing will go to the
5108 port and the print-state will be used to control the detection of
5109 circular references, etc. If the builtin function does not care for a
5110 print-state, it is simply ignored.
5112 User written callbacks are now called with such a pair as their
5113 `port', but because every function now accepts this pair as a PORT
5114 argument, you don't have to worry about that. In fact, it is probably
5115 safest to not check for these pairs.
5117 However, it is sometimes necessary to continue a print chain on a
5118 different port, for example to get a intermediate string
5119 representation of the printed value, mangle that string somehow, and
5120 then to finally print the mangled string. Use the new function
5122 inherit-print-state OLD-PORT NEW-PORT
5124 for this. It constructs a new `port' that prints to NEW-PORT but
5125 inherits the print-state of OLD-PORT.
5127 ** struct-vtable-offset renamed to vtable-offset-user
5129 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5131 ** There is now a third optional argument to make-vtable-vtable
5132 (and fourth to make-struct) when constructing new types (vtables).
5133 This argument initializes field vtable-index-printer of the vtable.
5135 ** The detection of circular references has been extended to structs.
5136 That is, a structure that -- in the process of being printed -- prints
5137 itself does not lead to infinite recursion.
5139 ** There is now some basic support for fluids. Please read
5140 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5141 the following functions and macros:
5143 Function: make-fluid
5145 Create a new fluid object. Fluids are not special variables or
5146 some other extension to the semantics of Scheme, but rather
5147 ordinary Scheme objects. You can store them into variables (that
5148 are still lexically scoped, of course) or into any other place you
5149 like. Every fluid has a initial value of `#f'.
5151 Function: fluid? OBJ
5153 Test whether OBJ is a fluid.
5155 Function: fluid-ref FLUID
5156 Function: fluid-set! FLUID VAL
5158 Access/modify the fluid FLUID. Modifications are only visible
5159 within the current dynamic root (that includes threads).
5161 Function: with-fluids* FLUIDS VALUES THUNK
5163 FLUIDS is a list of fluids and VALUES a corresponding list of
5164 values for these fluids. Before THUNK gets called the values are
5165 installed in the fluids and the old values of the fluids are
5166 saved in the VALUES list. When the flow of control leaves THUNK
5167 or reenters it, the values get swapped again. You might think of
5168 this as a `safe-fluid-excursion'. Note that the VALUES list is
5169 modified by `with-fluids*'.
5171 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5173 The same as `with-fluids*' but with a different syntax. It looks
5174 just like `let', but both FLUID and VALUE are evaluated. Remember,
5175 fluids are not special variables but ordinary objects. FLUID
5176 should evaluate to a fluid.
5178 ** Changes to system call interfaces:
5180 *** close-port, close-input-port and close-output-port now return a
5181 boolean instead of an `unspecified' object. #t means that the port
5182 was successfully closed, while #f means it was already closed. It is
5183 also now possible for these procedures to raise an exception if an
5184 error occurs (some errors from write can be delayed until close.)
5186 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5189 *** the third argument to fcntl is now optional.
5191 *** the first argument to chown can now be a file descriptor or a port.
5193 *** the argument to stat can now be a port.
5195 *** The following new procedures have been added (most use scsh
5198 *** procedure: close PORT/FD
5199 Similar to close-port (*note close-port: Closing Ports.), but also
5200 works on file descriptors. A side effect of closing a file
5201 descriptor is that any ports using that file descriptor are moved
5202 to a different file descriptor and have their revealed counts set
5205 *** procedure: port->fdes PORT
5206 Returns the integer file descriptor underlying PORT. As a side
5207 effect the revealed count of PORT is incremented.
5209 *** procedure: fdes->ports FDES
5210 Returns a list of existing ports which have FDES as an underlying
5211 file descriptor, without changing their revealed counts.
5213 *** procedure: fdes->inport FDES
5214 Returns an existing input port which has FDES as its underlying
5215 file descriptor, if one exists, and increments its revealed count.
5216 Otherwise, returns a new input port with a revealed count of 1.
5218 *** procedure: fdes->outport FDES
5219 Returns an existing output port which has FDES as its underlying
5220 file descriptor, if one exists, and increments its revealed count.
5221 Otherwise, returns a new output port with a revealed count of 1.
5223 The next group of procedures perform a `dup2' system call, if NEWFD
5224 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5225 duplicated can be supplied as an integer or contained in a port. The
5226 type of value returned varies depending on which procedure is used.
5228 All procedures also have the side effect when performing `dup2' that
5229 any ports using NEWFD are moved to a different file descriptor and have
5230 their revealed counts set to zero.
5232 *** procedure: dup->fdes PORT/FD [NEWFD]
5233 Returns an integer file descriptor.
5235 *** procedure: dup->inport PORT/FD [NEWFD]
5236 Returns a new input port using the new file descriptor.
5238 *** procedure: dup->outport PORT/FD [NEWFD]
5239 Returns a new output port using the new file descriptor.
5241 *** procedure: dup PORT/FD [NEWFD]
5242 Returns a new port if PORT/FD is a port, with the same mode as the
5243 supplied port, otherwise returns an integer file descriptor.
5245 *** procedure: dup->port PORT/FD MODE [NEWFD]
5246 Returns a new port using the new file descriptor. MODE supplies a
5247 mode string for the port (*note open-file: File Ports.).
5249 *** procedure: setenv NAME VALUE
5250 Modifies the environment of the current process, which is also the
5251 default environment inherited by child processes.
5253 If VALUE is `#f', then NAME is removed from the environment.
5254 Otherwise, the string NAME=VALUE is added to the environment,
5255 replacing any existing string with name matching NAME.
5257 The return value is unspecified.
5259 *** procedure: truncate-file OBJ SIZE
5260 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5261 can be a string containing a file name or an integer file
5262 descriptor or port open for output on the file. The underlying
5263 system calls are `truncate' and `ftruncate'.
5265 The return value is unspecified.
5267 *** procedure: setvbuf PORT MODE [SIZE]
5268 Set the buffering mode for PORT. MODE can be:
5276 block buffered, using a newly allocated buffer of SIZE bytes.
5277 However if SIZE is zero or unspecified, the port will be made
5280 This procedure should not be used after I/O has been performed with
5283 Ports are usually block buffered by default, with a default buffer
5284 size. Procedures e.g., *Note open-file: File Ports, which accept a
5285 mode string allow `0' to be added to request an unbuffered port.
5287 *** procedure: fsync PORT/FD
5288 Copies any unwritten data for the specified output file descriptor
5289 to disk. If PORT/FD is a port, its buffer is flushed before the
5290 underlying file descriptor is fsync'd. The return value is
5293 *** procedure: open-fdes PATH FLAGS [MODES]
5294 Similar to `open' but returns a file descriptor instead of a port.
5296 *** procedure: execle PATH ENV [ARG] ...
5297 Similar to `execl', but the environment of the new process is
5298 specified by ENV, which must be a list of strings as returned by
5299 the `environ' procedure.
5301 This procedure is currently implemented using the `execve' system
5302 call, but we call it `execle' because of its Scheme calling
5305 *** procedure: strerror ERRNO
5306 Returns the Unix error message corresponding to ERRNO, an integer.
5308 *** procedure: primitive-exit [STATUS]
5309 Terminate the current process without unwinding the Scheme stack.
5310 This is would typically be useful after a fork. The exit status
5311 is STATUS if supplied, otherwise zero.
5313 *** procedure: times
5314 Returns an object with information about real and processor time.
5315 The following procedures accept such an object as an argument and
5316 return a selected component:
5319 The current real time, expressed as time units relative to an
5323 The CPU time units used by the calling process.
5326 The CPU time units used by the system on behalf of the
5330 The CPU time units used by terminated child processes of the
5331 calling process, whose status has been collected (e.g., using
5335 Similarly, the CPU times units used by the system on behalf of
5336 terminated child processes.
5338 ** Removed: list-length
5339 ** Removed: list-append, list-append!
5340 ** Removed: list-reverse, list-reverse!
5342 ** array-map renamed to array-map!
5344 ** serial-array-map renamed to serial-array-map!
5346 ** catch doesn't take #f as first argument any longer
5348 Previously, it was possible to pass #f instead of a key to `catch'.
5349 That would cause `catch' to pass a jump buffer object to the procedure
5350 passed as second argument. The procedure could then use this jump
5351 buffer objekt as an argument to throw.
5353 This mechanism has been removed since its utility doesn't motivate the
5354 extra complexity it introduces.
5356 ** The `#/' notation for lists now provokes a warning message from Guile.
5357 This syntax will be removed from Guile in the near future.
5359 To disable the warning message, set the GUILE_HUSH environment
5360 variable to any non-empty value.
5362 ** The newline character now prints as `#\newline', following the
5363 normal Scheme notation, not `#\nl'.
5365 * Changes to the gh_ interface
5367 ** The gh_enter function now takes care of loading the Guile startup files.
5368 gh_enter works by calling scm_boot_guile; see the remarks below.
5370 ** Function: void gh_write (SCM x)
5372 Write the printed representation of the scheme object x to the current
5373 output port. Corresponds to the scheme level `write'.
5375 ** gh_list_length renamed to gh_length.
5377 ** vector handling routines
5379 Several major changes. In particular, gh_vector() now resembles
5380 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5381 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5382 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5383 vector-related gh_ functions have been implemented.
5385 ** pair and list routines
5387 Implemented several of the R4RS pair and list functions that were
5390 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5392 New function. Converts double arrays back and forth between Scheme
5395 * Changes to the scm_ interface
5397 ** The function scm_boot_guile now takes care of loading the startup files.
5399 Guile's primary initialization function, scm_boot_guile, now takes
5400 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5401 Guile, define the module system, and put together some standard
5402 bindings. It also loads `init.scm', which is intended to hold
5403 site-specific initialization code.
5405 Since Guile cannot operate properly until boot-9.scm is loaded, there
5406 is no reason to separate loading boot-9.scm from Guile's other
5407 initialization processes.
5409 This job used to be done by scm_compile_shell_switches, which didn't
5410 make much sense; in particular, it meant that people using Guile for
5411 non-shell-like applications had to jump through hoops to get Guile
5412 initialized properly.
5414 ** The function scm_compile_shell_switches no longer loads the startup files.
5415 Now, Guile always loads the startup files, whenever it is initialized;
5416 see the notes above for scm_boot_guile and scm_load_startup_files.
5418 ** Function: scm_load_startup_files
5419 This new function takes care of loading Guile's initialization file
5420 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5421 this is always called by the Guile initialization process, it's
5422 probably not too useful to call this yourself, but it's there anyway.
5424 ** The semantics of smob marking have changed slightly.
5426 The smob marking function (the `mark' member of the scm_smobfuns
5427 structure) is no longer responsible for setting the mark bit on the
5428 smob. The generic smob handling code in the garbage collector will
5429 set this bit. The mark function need only ensure that any other
5430 objects the smob refers to get marked.
5432 Note that this change means that the smob's GC8MARK bit is typically
5433 already set upon entry to the mark function. Thus, marking functions
5434 which look like this:
5437 if (SCM_GC8MARKP (ptr))
5439 SCM_SETGC8MARK (ptr);
5440 ... mark objects to which the smob refers ...
5443 are now incorrect, since they will return early, and fail to mark any
5444 other objects the smob refers to. Some code in the Guile library used
5447 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5449 If you have implemented your own I/O port type, by writing the
5450 functions required by the scm_ptobfuns and then calling scm_newptob,
5451 you will need to change your functions slightly.
5453 The functions in a scm_ptobfuns structure now expect the port itself
5454 as their argument; they used to expect the `stream' member of the
5455 port's scm_port_table structure. This allows functions in an
5456 scm_ptobfuns structure to easily access the port's cell (and any flags
5457 it its CAR), and the port's scm_port_table structure.
5459 Guile now passes the I/O port itself as the `port' argument in the
5460 following scm_ptobfuns functions:
5462 int (*free) (SCM port);
5463 int (*fputc) (int, SCM port);
5464 int (*fputs) (char *, SCM port);
5465 scm_sizet (*fwrite) SCM_P ((char *ptr,
5469 int (*fflush) (SCM port);
5470 int (*fgetc) (SCM port);
5471 int (*fclose) (SCM port);
5473 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5476 If you have existing code which defines its own port types, it is easy
5477 to convert your code to the new interface; simply apply SCM_STREAM to
5478 the port argument to yield the value you code used to expect.
5480 Note that since both the port and the stream have the same type in the
5481 C code --- they are both SCM values --- the C compiler will not remind
5482 you if you forget to update your scm_ptobfuns functions.
5485 ** Function: int scm_internal_select (int fds,
5489 struct timeval *timeout);
5491 This is a replacement for the `select' function provided by the OS.
5492 It enables I/O blocking and sleeping to happen for one cooperative
5493 thread without blocking other threads. It also avoids busy-loops in
5494 these situations. It is intended that all I/O blocking and sleeping
5495 will finally go through this function. Currently, this function is
5496 only available on systems providing `gettimeofday' and `select'.
5498 ** Function: SCM scm_internal_stack_catch (SCM tag,
5499 scm_catch_body_t body,
5501 scm_catch_handler_t handler,
5504 A new sibling to the other two C level `catch' functions
5505 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5506 the stack to be saved automatically into the variable `the-last-stack'
5507 (scm_the_last_stack_var) on error. This is necessary if you want to
5508 use advanced error reporting, such as calling scm_display_error and
5509 scm_display_backtrace. (They both take a stack object as argument.)
5511 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5513 scm_catch_handler_t handler,
5516 Spawns a new thread. It does a job similar to
5517 scm_call_with_new_thread but takes arguments more suitable when
5518 spawning threads from application C code.
5520 ** The hook scm_error_callback has been removed. It was originally
5521 intended as a way for the user to install his own error handler. But
5522 that method works badly since it intervenes between throw and catch,
5523 thereby changing the semantics of expressions like (catch #t ...).
5524 The correct way to do it is to use one of the C level catch functions
5525 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5527 ** Removed functions:
5529 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5530 scm_list_reverse, scm_list_reverse_x
5532 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5534 These can be used for pretty list creation from C. The idea is taken
5535 from Erick Gallesio's STk.
5537 ** scm_array_map renamed to scm_array_map_x
5539 ** mbstrings are now removed
5541 This means that the type codes scm_tc7_mb_string and
5542 scm_tc7_mb_substring has been removed.
5544 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5546 Since we no longer support multi-byte strings, these I/O functions
5547 have been simplified, and renamed. Here are their old names, and
5548 their new names and arguments:
5550 scm_gen_putc -> void scm_putc (int c, SCM port);
5551 scm_gen_puts -> void scm_puts (char *s, SCM port);
5552 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5553 scm_gen_getc -> void scm_getc (SCM port);
5556 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5558 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5560 SCM_TYP7S now masks away the bit which distinguishes substrings from
5563 ** scm_catch_body_t: Backward incompatible change!
5565 Body functions to scm_internal_catch and friends do not any longer
5566 take a second argument. This is because it is no longer possible to
5567 pass a #f arg to catch.
5569 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5571 The function scm_protect_object protects its argument from being freed
5572 by the garbage collector. scm_unprotect_object removes that
5575 These functions now nest properly. That is, for every object O, there
5576 is a counter which scm_protect_object(O) increments and
5577 scm_unprotect_object(O) decrements, if the counter is greater than
5578 zero. Every object's counter is zero when it is first created. If an
5579 object's counter is greater than zero, the garbage collector will not
5580 reclaim its storage.
5582 This allows you to use scm_protect_object in your code without
5583 worrying that some other function you call will call
5584 scm_unprotect_object, and allow it to be freed. Assuming that the
5585 functions you call are well-behaved, and unprotect only those objects
5586 they protect, you can follow the same rule and have confidence that
5587 objects will be freed only at appropriate times.
5590 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5592 * Changes to the distribution
5594 ** Nightly snapshots are now available from ftp.red-bean.com.
5595 The old server, ftp.cyclic.com, has been relinquished to its rightful
5598 Nightly snapshots of the Guile development sources are now available via
5599 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5601 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5602 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5604 ** To run Guile without installing it, the procedure has changed a bit.
5606 If you used a separate build directory to compile Guile, you'll need
5607 to include the build directory in SCHEME_LOAD_PATH, as well as the
5608 source directory. See the `INSTALL' file for examples.
5610 * Changes to the procedure for linking libguile with your programs
5612 ** The standard Guile load path for Scheme code now includes
5613 $(datadir)/guile (usually /usr/local/share/guile). This means that
5614 you can install your own Scheme files there, and Guile will find them.
5615 (Previous versions of Guile only checked a directory whose name
5616 contained the Guile version number, so you had to re-install or move
5617 your Scheme sources each time you installed a fresh version of Guile.)
5619 The load path also includes $(datadir)/guile/site; we recommend
5620 putting individual Scheme files there. If you want to install a
5621 package with multiple source files, create a directory for them under
5624 ** Guile 1.2 will now use the Rx regular expression library, if it is
5625 installed on your system. When you are linking libguile into your own
5626 programs, this means you will have to link against -lguile, -lqt (if
5627 you configured Guile with thread support), and -lrx.
5629 If you are using autoconf to generate configuration scripts for your
5630 application, the following lines should suffice to add the appropriate
5631 libraries to your link command:
5633 ### Find Rx, quickthreads and libguile.
5634 AC_CHECK_LIB(rx, main)
5635 AC_CHECK_LIB(qt, main)
5636 AC_CHECK_LIB(guile, scm_shell)
5638 The Guile 1.2 distribution does not contain sources for the Rx
5639 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5640 retrieve it from a GNU FTP site and install it separately.
5642 * Changes to Scheme functions and syntax
5644 ** The dynamic linking features of Guile are now enabled by default.
5645 You can disable them by giving the `--disable-dynamic-linking' option
5648 (dynamic-link FILENAME)
5650 Find the object file denoted by FILENAME (a string) and link it
5651 into the running Guile application. When everything works out,
5652 return a Scheme object suitable for representing the linked object
5653 file. Otherwise an error is thrown. How object files are
5654 searched is system dependent.
5656 (dynamic-object? VAL)
5658 Determine whether VAL represents a dynamically linked object file.
5660 (dynamic-unlink DYNOBJ)
5662 Unlink the indicated object file from the application. DYNOBJ
5663 should be one of the values returned by `dynamic-link'.
5665 (dynamic-func FUNCTION DYNOBJ)
5667 Search the C function indicated by FUNCTION (a string or symbol)
5668 in DYNOBJ and return some Scheme object that can later be used
5669 with `dynamic-call' to actually call this function. Right now,
5670 these Scheme objects are formed by casting the address of the
5671 function to `long' and converting this number to its Scheme
5674 (dynamic-call FUNCTION DYNOBJ)
5676 Call the C function indicated by FUNCTION and DYNOBJ. The
5677 function is passed no arguments and its return value is ignored.
5678 When FUNCTION is something returned by `dynamic-func', call that
5679 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5680 etc.), look it up in DYNOBJ; this is equivalent to
5682 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5684 Interrupts are deferred while the C function is executing (with
5685 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5687 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5689 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5690 some arguments and return its return value. The C function is
5691 expected to take two arguments and return an `int', just like
5694 int c_func (int argc, char **argv);
5696 ARGS must be a list of strings and is converted into an array of
5697 `char *'. The array is passed in ARGV and its size in ARGC. The
5698 return value is converted to a Scheme number and returned from the
5699 call to `dynamic-args-call'.
5701 When dynamic linking is disabled or not supported on your system,
5702 the above functions throw errors, but they are still available.
5704 Here is a small example that works on GNU/Linux:
5706 (define libc-obj (dynamic-link "libc.so"))
5707 (dynamic-args-call 'rand libc-obj '())
5709 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5711 ** The #/ syntax for module names is depreciated, and will be removed
5712 in a future version of Guile. Instead of
5720 The latter syntax is more consistent with existing Lisp practice.
5722 ** Guile now does fancier printing of structures. Structures are the
5723 underlying implementation for records, which in turn are used to
5724 implement modules, so all of these object now print differently and in
5725 a more informative way.
5727 The Scheme printer will examine the builtin variable *struct-printer*
5728 whenever it needs to print a structure object. When this variable is
5729 not `#f' it is deemed to be a procedure and will be applied to the
5730 structure object and the output port. When *struct-printer* is `#f'
5731 or the procedure return `#f' the structure object will be printed in
5732 the boring #<struct 80458270> form.
5734 This hook is used by some routines in ice-9/boot-9.scm to implement
5735 type specific printing routines. Please read the comments there about
5738 One of the more specific uses of structs are records. The printing
5739 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5740 called. It should behave like a *struct-printer* procedure (described
5743 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5744 token of the form #:NAME, where NAME has the same syntax as a Scheme
5745 symbol, is the external representation of the keyword named NAME.
5746 Keyword objects print using this syntax as well, so values containing
5747 keyword objects can be read back into Guile. When used in an
5748 expression, keywords are self-quoting objects.
5750 Guile suports this read syntax, and uses this print syntax, regardless
5751 of the current setting of the `keyword' read option. The `keyword'
5752 read option only controls whether Guile recognizes the `:NAME' syntax,
5753 which is incompatible with R4RS. (R4RS says such token represent
5756 ** Guile has regular expression support again. Guile 1.0 included
5757 functions for matching regular expressions, based on the Rx library.
5758 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5759 distribution, and thus Guile had no regular expression support. Guile
5760 1.2 again supports the most commonly used functions, and supports all
5761 of SCSH's regular expression functions.
5763 If your system does not include a POSIX regular expression library,
5764 and you have not linked Guile with a third-party regexp library such as
5765 Rx, these functions will not be available. You can tell whether your
5766 Guile installation includes regular expression support by checking
5767 whether the `*features*' list includes the `regex' symbol.
5769 *** regexp functions
5771 By default, Guile supports POSIX extended regular expressions. That
5772 means that the characters `(', `)', `+' and `?' are special, and must
5773 be escaped if you wish to match the literal characters.
5775 This regular expression interface was modeled after that implemented
5776 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5777 with SCSH regular expressions.
5779 **** Function: string-match PATTERN STR [START]
5780 Compile the string PATTERN into a regular expression and compare
5781 it with STR. The optional numeric argument START specifies the
5782 position of STR at which to begin matching.
5784 `string-match' returns a "match structure" which describes what,
5785 if anything, was matched by the regular expression. *Note Match
5786 Structures::. If STR does not match PATTERN at all,
5787 `string-match' returns `#f'.
5789 Each time `string-match' is called, it must compile its PATTERN
5790 argument into a regular expression structure. This operation is
5791 expensive, which makes `string-match' inefficient if the same regular
5792 expression is used several times (for example, in a loop). For better
5793 performance, you can compile a regular expression in advance and then
5794 match strings against the compiled regexp.
5796 **** Function: make-regexp STR [FLAGS]
5797 Compile the regular expression described by STR, and return the
5798 compiled regexp structure. If STR does not describe a legal
5799 regular expression, `make-regexp' throws a
5800 `regular-expression-syntax' error.
5802 FLAGS may be the bitwise-or of one or more of the following:
5804 **** Constant: regexp/extended
5805 Use POSIX Extended Regular Expression syntax when interpreting
5806 STR. If not set, POSIX Basic Regular Expression syntax is used.
5807 If the FLAGS argument is omitted, we assume regexp/extended.
5809 **** Constant: regexp/icase
5810 Do not differentiate case. Subsequent searches using the
5811 returned regular expression will be case insensitive.
5813 **** Constant: regexp/newline
5814 Match-any-character operators don't match a newline.
5816 A non-matching list ([^...]) not containing a newline matches a
5819 Match-beginning-of-line operator (^) matches the empty string
5820 immediately after a newline, regardless of whether the FLAGS
5821 passed to regexp-exec contain regexp/notbol.
5823 Match-end-of-line operator ($) matches the empty string
5824 immediately before a newline, regardless of whether the FLAGS
5825 passed to regexp-exec contain regexp/noteol.
5827 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5828 Match the compiled regular expression REGEXP against `str'. If
5829 the optional integer START argument is provided, begin matching
5830 from that position in the string. Return a match structure
5831 describing the results of the match, or `#f' if no match could be
5834 FLAGS may be the bitwise-or of one or more of the following:
5836 **** Constant: regexp/notbol
5837 The match-beginning-of-line operator always fails to match (but
5838 see the compilation flag regexp/newline above) This flag may be
5839 used when different portions of a string are passed to
5840 regexp-exec and the beginning of the string should not be
5841 interpreted as the beginning of the line.
5843 **** Constant: regexp/noteol
5844 The match-end-of-line operator always fails to match (but see the
5845 compilation flag regexp/newline above)
5847 **** Function: regexp? OBJ
5848 Return `#t' if OBJ is a compiled regular expression, or `#f'
5851 Regular expressions are commonly used to find patterns in one string
5852 and replace them with the contents of another string.
5854 **** Function: regexp-substitute PORT MATCH [ITEM...]
5855 Write to the output port PORT selected contents of the match
5856 structure MATCH. Each ITEM specifies what should be written, and
5857 may be one of the following arguments:
5859 * A string. String arguments are written out verbatim.
5861 * An integer. The submatch with that number is written.
5863 * The symbol `pre'. The portion of the matched string preceding
5864 the regexp match is written.
5866 * The symbol `post'. The portion of the matched string
5867 following the regexp match is written.
5869 PORT may be `#f', in which case nothing is written; instead,
5870 `regexp-substitute' constructs a string from the specified ITEMs
5873 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5874 Similar to `regexp-substitute', but can be used to perform global
5875 substitutions on STR. Instead of taking a match structure as an
5876 argument, `regexp-substitute/global' takes two string arguments: a
5877 REGEXP string describing a regular expression, and a TARGET string
5878 which should be matched against this regular expression.
5880 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5883 * A function may be supplied. When this function is called, it
5884 will be passed one argument: a match structure for a given
5885 regular expression match. It should return a string to be
5886 written out to PORT.
5888 * The `post' symbol causes `regexp-substitute/global' to recurse
5889 on the unmatched portion of STR. This *must* be supplied in
5890 order to perform global search-and-replace on STR; if it is
5891 not present among the ITEMs, then `regexp-substitute/global'
5892 will return after processing a single match.
5894 *** Match Structures
5896 A "match structure" is the object returned by `string-match' and
5897 `regexp-exec'. It describes which portion of a string, if any, matched
5898 the given regular expression. Match structures include: a reference to
5899 the string that was checked for matches; the starting and ending
5900 positions of the regexp match; and, if the regexp included any
5901 parenthesized subexpressions, the starting and ending positions of each
5904 In each of the regexp match functions described below, the `match'
5905 argument must be a match structure returned by a previous call to
5906 `string-match' or `regexp-exec'. Most of these functions return some
5907 information about the original target string that was matched against a
5908 regular expression; we will call that string TARGET for easy reference.
5910 **** Function: regexp-match? OBJ
5911 Return `#t' if OBJ is a match structure returned by a previous
5912 call to `regexp-exec', or `#f' otherwise.
5914 **** Function: match:substring MATCH [N]
5915 Return the portion of TARGET matched by subexpression number N.
5916 Submatch 0 (the default) represents the entire regexp match. If
5917 the regular expression as a whole matched, but the subexpression
5918 number N did not match, return `#f'.
5920 **** Function: match:start MATCH [N]
5921 Return the starting position of submatch number N.
5923 **** Function: match:end MATCH [N]
5924 Return the ending position of submatch number N.
5926 **** Function: match:prefix MATCH
5927 Return the unmatched portion of TARGET preceding the regexp match.
5929 **** Function: match:suffix MATCH
5930 Return the unmatched portion of TARGET following the regexp match.
5932 **** Function: match:count MATCH
5933 Return the number of parenthesized subexpressions from MATCH.
5934 Note that the entire regular expression match itself counts as a
5935 subexpression, and failed submatches are included in the count.
5937 **** Function: match:string MATCH
5938 Return the original TARGET string.
5940 *** Backslash Escapes
5942 Sometimes you will want a regexp to match characters like `*' or `$'
5943 exactly. For example, to check whether a particular string represents
5944 a menu entry from an Info node, it would be useful to match it against
5945 a regexp like `^* [^:]*::'. However, this won't work; because the
5946 asterisk is a metacharacter, it won't match the `*' at the beginning of
5947 the string. In this case, we want to make the first asterisk un-magic.
5949 You can do this by preceding the metacharacter with a backslash
5950 character `\'. (This is also called "quoting" the metacharacter, and
5951 is known as a "backslash escape".) When Guile sees a backslash in a
5952 regular expression, it considers the following glyph to be an ordinary
5953 character, no matter what special meaning it would ordinarily have.
5954 Therefore, we can make the above example work by changing the regexp to
5955 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5956 to match only a single asterisk in the target string.
5958 Since the backslash is itself a metacharacter, you may force a
5959 regexp to match a backslash in the target string by preceding the
5960 backslash with itself. For example, to find variable references in a
5961 TeX program, you might want to find occurrences of the string `\let\'
5962 followed by any number of alphabetic characters. The regular expression
5963 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5964 each match a single backslash in the target string.
5966 **** Function: regexp-quote STR
5967 Quote each special character found in STR with a backslash, and
5968 return the resulting string.
5970 *Very important:* Using backslash escapes in Guile source code (as
5971 in Emacs Lisp or C) can be tricky, because the backslash character has
5972 special meaning for the Guile reader. For example, if Guile encounters
5973 the character sequence `\n' in the middle of a string while processing
5974 Scheme code, it replaces those characters with a newline character.
5975 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5976 Several of these "escape sequences" are processed by the Guile reader
5977 before your code is executed. Unrecognized escape sequences are
5978 ignored: if the characters `\*' appear in a string, they will be
5979 translated to the single character `*'.
5981 This translation is obviously undesirable for regular expressions,
5982 since we want to be able to include backslashes in a string in order to
5983 escape regexp metacharacters. Therefore, to make sure that a backslash
5984 is preserved in a string in your Guile program, you must use *two*
5985 consecutive backslashes:
5987 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5989 The string in this example is preprocessed by the Guile reader before
5990 any code is executed. The resulting argument to `make-regexp' is the
5991 string `^\* [^:]*', which is what we really want.
5993 This also means that in order to write a regular expression that
5994 matches a single backslash character, the regular expression string in
5995 the source code must include *four* backslashes. Each consecutive pair
5996 of backslashes gets translated by the Guile reader to a single
5997 backslash, and the resulting double-backslash is interpreted by the
5998 regexp engine as matching a single backslash character. Hence:
6000 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6002 The reason for the unwieldiness of this syntax is historical. Both
6003 regular expression pattern matchers and Unix string processing systems
6004 have traditionally used backslashes with the special meanings described
6005 above. The POSIX regular expression specification and ANSI C standard
6006 both require these semantics. Attempting to abandon either convention
6007 would cause other kinds of compatibility problems, possibly more severe
6008 ones. Therefore, without extending the Scheme reader to support
6009 strings with different quoting conventions (an ungainly and confusing
6010 extension when implemented in other languages), we must adhere to this
6011 cumbersome escape syntax.
6013 * Changes to the gh_ interface
6015 * Changes to the scm_ interface
6017 * Changes to system call interfaces:
6019 ** The value returned by `raise' is now unspecified. It throws an exception
6022 *** A new procedure `sigaction' can be used to install signal handlers
6024 (sigaction signum [action] [flags])
6026 signum is the signal number, which can be specified using the value
6029 If action is omitted, sigaction returns a pair: the CAR is the current
6030 signal hander, which will be either an integer with the value SIG_DFL
6031 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6032 handles the signal, or #f if a non-Scheme procedure handles the
6033 signal. The CDR contains the current sigaction flags for the handler.
6035 If action is provided, it is installed as the new handler for signum.
6036 action can be a Scheme procedure taking one argument, or the value of
6037 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6038 whatever signal handler was installed before sigaction was first used.
6039 Flags can optionally be specified for the new handler (SA_RESTART is
6040 always used if the system provides it, so need not be specified.) The
6041 return value is a pair with information about the old handler as
6044 This interface does not provide access to the "signal blocking"
6045 facility. Maybe this is not needed, since the thread support may
6046 provide solutions to the problem of consistent access to data
6049 *** A new procedure `flush-all-ports' is equivalent to running
6050 `force-output' on every port open for output.
6052 ** Guile now provides information on how it was built, via the new
6053 global variable, %guile-build-info. This variable records the values
6054 of the standard GNU makefile directory variables as an assocation
6055 list, mapping variable names (symbols) onto directory paths (strings).
6056 For example, to find out where the Guile link libraries were
6057 installed, you can say:
6059 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6062 * Changes to the scm_ interface
6064 ** The new function scm_handle_by_message_noexit is just like the
6065 existing scm_handle_by_message function, except that it doesn't call
6066 exit to terminate the process. Instead, it prints a message and just
6067 returns #f. This might be a more appropriate catch-all handler for
6068 new dynamic roots and threads.
6071 Changes in Guile 1.1 (released Friday, May 16 1997):
6073 * Changes to the distribution.
6075 The Guile 1.0 distribution has been split up into several smaller
6077 guile-core --- the Guile interpreter itself.
6078 guile-tcltk --- the interface between the Guile interpreter and
6079 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6080 is a toolkit for building graphical user interfaces.
6081 guile-rgx-ctax --- the interface between Guile and the Rx regular
6082 expression matcher, and the translator for the Ctax
6083 programming language. These are packaged together because the
6084 Ctax translator uses Rx to parse Ctax source code.
6086 This NEWS file describes the changes made to guile-core since the 1.0
6089 We no longer distribute the documentation, since it was either out of
6090 date, or incomplete. As soon as we have current documentation, we
6095 * Changes to the stand-alone interpreter
6097 ** guile now accepts command-line arguments compatible with SCSH, Olin
6098 Shivers' Scheme Shell.
6100 In general, arguments are evaluated from left to right, but there are
6101 exceptions. The following switches stop argument processing, and
6102 stash all remaining command-line arguments as the value returned by
6103 the (command-line) function.
6104 -s SCRIPT load Scheme source code from FILE, and exit
6105 -c EXPR evalute Scheme expression EXPR, and exit
6106 -- stop scanning arguments; run interactively
6108 The switches below are processed as they are encountered.
6109 -l FILE load Scheme source code from FILE
6110 -e FUNCTION after reading script, apply FUNCTION to
6111 command line arguments
6112 -ds do -s script at this point
6113 --emacs enable Emacs protocol (experimental)
6114 -h, --help display this help and exit
6115 -v, --version display version information and exit
6116 \ read arguments from following script lines
6118 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6119 which re-implements the traditional "echo" command:
6121 #!/usr/local/bin/guile -s
6124 (map (lambda (arg) (display arg) (display " "))
6128 (main (command-line))
6130 Suppose we invoke this script as follows:
6132 ekko a speckled gecko
6134 Through the magic of Unix script processing (triggered by the `#!'
6135 token at the top of the file), /usr/local/bin/guile receives the
6136 following list of command-line arguments:
6138 ("-s" "./ekko" "a" "speckled" "gecko")
6140 Unix inserts the name of the script after the argument specified on
6141 the first line of the file (in this case, "-s"), and then follows that
6142 with the arguments given to the script. Guile loads the script, which
6143 defines the `main' function, and then applies it to the list of
6144 remaining command-line arguments, ("a" "speckled" "gecko").
6146 In Unix, the first line of a script file must take the following form:
6148 #!INTERPRETER ARGUMENT
6150 where INTERPRETER is the absolute filename of the interpreter
6151 executable, and ARGUMENT is a single command-line argument to pass to
6154 You may only pass one argument to the interpreter, and its length is
6155 limited. These restrictions can be annoying to work around, so Guile
6156 provides a general mechanism (borrowed from, and compatible with,
6157 SCSH) for circumventing them.
6159 If the ARGUMENT in a Guile script is a single backslash character,
6160 `\', Guile will open the script file, parse arguments from its second
6161 and subsequent lines, and replace the `\' with them. So, for example,
6162 here is another implementation of the `ekko' script:
6164 #!/usr/local/bin/guile \
6168 (for-each (lambda (arg) (display arg) (display " "))
6172 If the user invokes this script as follows:
6174 ekko a speckled gecko
6176 Unix expands this into
6178 /usr/local/bin/guile \ ekko a speckled gecko
6180 When Guile sees the `\' argument, it replaces it with the arguments
6181 read from the second line of the script, producing:
6183 /usr/local/bin/guile -e main -s ekko a speckled gecko
6185 This tells Guile to load the `ekko' script, and apply the function
6186 `main' to the argument list ("a" "speckled" "gecko").
6188 Here is how Guile parses the command-line arguments:
6189 - Each space character terminates an argument. This means that two
6190 spaces in a row introduce an empty-string argument.
6191 - The tab character is not permitted (unless you quote it with the
6192 backslash character, as described below), to avoid confusion.
6193 - The newline character terminates the sequence of arguments, and will
6194 also terminate a final non-empty argument. (However, a newline
6195 following a space will not introduce a final empty-string argument;
6196 it only terminates the argument list.)
6197 - The backslash character is the escape character. It escapes
6198 backslash, space, tab, and newline. The ANSI C escape sequences
6199 like \n and \t are also supported. These produce argument
6200 constituents; the two-character combination \n doesn't act like a
6201 terminating newline. The escape sequence \NNN for exactly three
6202 octal digits reads as the character whose ASCII code is NNN. As
6203 above, characters produced this way are argument constituents.
6204 Backslash followed by other characters is not allowed.
6206 * Changes to the procedure for linking libguile with your programs
6208 ** Guile now builds and installs a shared guile library, if your
6209 system support shared libraries. (It still builds a static library on
6210 all systems.) Guile automatically detects whether your system
6211 supports shared libraries. To prevent Guile from buildisg shared
6212 libraries, pass the `--disable-shared' flag to the configure script.
6214 Guile takes longer to compile when it builds shared libraries, because
6215 it must compile every file twice --- once to produce position-
6216 independent object code, and once to produce normal object code.
6218 ** The libthreads library has been merged into libguile.
6220 To link a program against Guile, you now need only link against
6221 -lguile and -lqt; -lthreads is no longer needed. If you are using
6222 autoconf to generate configuration scripts for your application, the
6223 following lines should suffice to add the appropriate libraries to
6226 ### Find quickthreads and libguile.
6227 AC_CHECK_LIB(qt, main)
6228 AC_CHECK_LIB(guile, scm_shell)
6230 * Changes to Scheme functions
6232 ** Guile Scheme's special syntax for keyword objects is now optional,
6233 and disabled by default.
6235 The syntax variation from R4RS made it difficult to port some
6236 interesting packages to Guile. The routines which accepted keyword
6237 arguments (mostly in the module system) have been modified to also
6238 accept symbols whose names begin with `:'.
6240 To change the keyword syntax, you must first import the (ice-9 debug)
6242 (use-modules (ice-9 debug))
6244 Then you can enable the keyword syntax as follows:
6245 (read-set! keywords 'prefix)
6247 To disable keyword syntax, do this:
6248 (read-set! keywords #f)
6250 ** Many more primitive functions accept shared substrings as
6251 arguments. In the past, these functions required normal, mutable
6252 strings as arguments, although they never made use of this
6255 ** The uniform array functions now operate on byte vectors. These
6256 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6257 `serial-array-map', `array-map', `array-for-each', and
6260 ** The new functions `trace' and `untrace' implement simple debugging
6261 support for Scheme functions.
6263 The `trace' function accepts any number of procedures as arguments,
6264 and tells the Guile interpreter to display each procedure's name and
6265 arguments each time the procedure is invoked. When invoked with no
6266 arguments, `trace' returns the list of procedures currently being
6269 The `untrace' function accepts any number of procedures as arguments,
6270 and tells the Guile interpreter not to trace them any more. When
6271 invoked with no arguments, `untrace' untraces all curretly traced
6274 The tracing in Guile has an advantage over most other systems: we
6275 don't create new procedure objects, but mark the procedure objects
6276 themselves. This means that anonymous and internal procedures can be
6279 ** The function `assert-repl-prompt' has been renamed to
6280 `set-repl-prompt!'. It takes one argument, PROMPT.
6281 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6282 - If PROMPT is a string, we use it as a prompt.
6283 - If PROMPT is a procedure accepting no arguments, we call it, and
6284 display the result as a prompt.
6285 - Otherwise, we display "> ".
6287 ** The new function `eval-string' reads Scheme expressions from a
6288 string and evaluates them, returning the value of the last expression
6289 in the string. If the string contains no expressions, it returns an
6292 ** The new function `thunk?' returns true iff its argument is a
6293 procedure of zero arguments.
6295 ** `defined?' is now a builtin function, instead of syntax. This
6296 means that its argument should be quoted. It returns #t iff its
6297 argument is bound in the current module.
6299 ** The new syntax `use-modules' allows you to add new modules to your
6300 environment without re-typing a complete `define-module' form. It
6301 accepts any number of module names as arguments, and imports their
6302 public bindings into the current module.
6304 ** The new function (module-defined? NAME MODULE) returns true iff
6305 NAME, a symbol, is defined in MODULE, a module object.
6307 ** The new function `builtin-bindings' creates and returns a hash
6308 table containing copies of all the root module's bindings.
6310 ** The new function `builtin-weak-bindings' does the same as
6311 `builtin-bindings', but creates a doubly-weak hash table.
6313 ** The `equal?' function now considers variable objects to be
6314 equivalent if they have the same name and the same value.
6316 ** The new function `command-line' returns the command-line arguments
6317 given to Guile, as a list of strings.
6319 When using guile as a script interpreter, `command-line' returns the
6320 script's arguments; those processed by the interpreter (like `-s' or
6321 `-c') are omitted. (In other words, you get the normal, expected
6322 behavior.) Any application that uses scm_shell to process its
6323 command-line arguments gets this behavior as well.
6325 ** The new function `load-user-init' looks for a file called `.guile'
6326 in the user's home directory, and loads it if it exists. This is
6327 mostly for use by the code generated by scm_compile_shell_switches,
6328 but we thought it might also be useful in other circumstances.
6330 ** The new function `log10' returns the base-10 logarithm of its
6333 ** Changes to I/O functions
6335 *** The functions `read', `primitive-load', `read-and-eval!', and
6336 `primitive-load-path' no longer take optional arguments controlling
6337 case insensitivity and a `#' parser.
6339 Case sensitivity is now controlled by a read option called
6340 `case-insensitive'. The user can add new `#' syntaxes with the
6341 `read-hash-extend' function (see below).
6343 *** The new function `read-hash-extend' allows the user to change the
6344 syntax of Guile Scheme in a somewhat controlled way.
6346 (read-hash-extend CHAR PROC)
6347 When parsing S-expressions, if we read a `#' character followed by
6348 the character CHAR, use PROC to parse an object from the stream.
6349 If PROC is #f, remove any parsing procedure registered for CHAR.
6351 The reader applies PROC to two arguments: CHAR and an input port.
6353 *** The new functions read-delimited and read-delimited! provide a
6354 general mechanism for doing delimited input on streams.
6356 (read-delimited DELIMS [PORT HANDLE-DELIM])
6357 Read until we encounter one of the characters in DELIMS (a string),
6358 or end-of-file. PORT is the input port to read from; it defaults to
6359 the current input port. The HANDLE-DELIM parameter determines how
6360 the terminating character is handled; it should be one of the
6363 'trim omit delimiter from result
6364 'peek leave delimiter character in input stream
6365 'concat append delimiter character to returned value
6366 'split return a pair: (RESULT . TERMINATOR)
6368 HANDLE-DELIM defaults to 'peek.
6370 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6371 A side-effecting variant of `read-delimited'.
6373 The data is written into the string BUF at the indices in the
6374 half-open interval [START, END); the default interval is the whole
6375 string: START = 0 and END = (string-length BUF). The values of
6376 START and END must specify a well-defined interval in BUF, i.e.
6377 0 <= START <= END <= (string-length BUF).
6379 It returns NBYTES, the number of bytes read. If the buffer filled
6380 up without a delimiter character being found, it returns #f. If the
6381 port is at EOF when the read starts, it returns the EOF object.
6383 If an integer is returned (i.e., the read is successfully terminated
6384 by reading a delimiter character), then the HANDLE-DELIM parameter
6385 determines how to handle the terminating character. It is described
6386 above, and defaults to 'peek.
6388 (The descriptions of these functions were borrowed from the SCSH
6389 manual, by Olin Shivers and Brian Carlstrom.)
6391 *** The `%read-delimited!' function is the primitive used to implement
6392 `read-delimited' and `read-delimited!'.
6394 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6396 This returns a pair of values: (TERMINATOR . NUM-READ).
6397 - TERMINATOR describes why the read was terminated. If it is a
6398 character or the eof object, then that is the value that terminated
6399 the read. If it is #f, the function filled the buffer without finding
6400 a delimiting character.
6401 - NUM-READ is the number of characters read into BUF.
6403 If the read is successfully terminated by reading a delimiter
6404 character, then the gobble? parameter determines what to do with the
6405 terminating character. If true, the character is removed from the
6406 input stream; if false, the character is left in the input stream
6407 where a subsequent read operation will retrieve it. In either case,
6408 the character is also the first value returned by the procedure call.
6410 (The descriptions of this function was borrowed from the SCSH manual,
6411 by Olin Shivers and Brian Carlstrom.)
6413 *** The `read-line' and `read-line!' functions have changed; they now
6414 trim the terminator by default; previously they appended it to the
6415 returned string. For the old behavior, use (read-line PORT 'concat).
6417 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6418 take new optional START and END arguments, specifying the region of
6419 the array to read and write.
6421 *** The `ungetc-char-ready?' function has been removed. We feel it's
6422 inappropriate for an interface to expose implementation details this
6425 ** Changes to the Unix library and system call interface
6427 *** The new fcntl function provides access to the Unix `fcntl' system
6430 (fcntl PORT COMMAND VALUE)
6431 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6432 Values for COMMAND are:
6434 F_DUPFD duplicate a file descriptor
6435 F_GETFD read the descriptor's close-on-exec flag
6436 F_SETFD set the descriptor's close-on-exec flag to VALUE
6437 F_GETFL read the descriptor's flags, as set on open
6438 F_SETFL set the descriptor's flags, as set on open to VALUE
6439 F_GETOWN return the process ID of a socket's owner, for SIGIO
6440 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6441 FD_CLOEXEC not sure what this is
6443 For details, see the documentation for the fcntl system call.
6445 *** The arguments to `select' have changed, for compatibility with
6446 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6447 expected behavior. The MILLISECONDS parameter has been changed to
6448 MICROSECONDS, to more closely resemble the underlying system call.
6449 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6450 corresponding return set will be the same.
6452 *** The arguments to the `mknod' system call have changed. They are
6455 (mknod PATH TYPE PERMS DEV)
6456 Create a new file (`node') in the file system. PATH is the name of
6457 the file to create. TYPE is the kind of file to create; it should
6458 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6459 permission bits to give the newly created file. If TYPE is
6460 'block-special or 'char-special, DEV specifies which device the
6461 special file refers to; its interpretation depends on the kind of
6462 special file being created.
6464 *** The `fork' function has been renamed to `primitive-fork', to avoid
6465 clashing with various SCSH forks.
6467 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6468 and `recvfrom!'. They no longer accept a size for a second argument;
6469 you must pass a string to hold the received value. They no longer
6470 return the buffer. Instead, `recv' returns the length of the message
6471 received, and `recvfrom' returns a pair containing the packet's length
6472 and originating address.
6474 *** The file descriptor datatype has been removed, as have the
6475 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6476 We plan to replace these functions with a SCSH-compatible interface.
6478 *** The `create' function has been removed; it's just a special case
6481 *** There are new functions to break down process termination status
6482 values. In the descriptions below, STATUS is a value returned by
6485 (status:exit-val STATUS)
6486 If the child process exited normally, this function returns the exit
6487 code for the child process (i.e., the value passed to exit, or
6488 returned from main). If the child process did not exit normally,
6489 this function returns #f.
6491 (status:stop-sig STATUS)
6492 If the child process was suspended by a signal, this function
6493 returns the signal that suspended the child. Otherwise, it returns
6496 (status:term-sig STATUS)
6497 If the child process terminated abnormally, this function returns
6498 the signal that terminated the child. Otherwise, this function
6501 POSIX promises that exactly one of these functions will return true on
6502 a valid STATUS value.
6504 These functions are compatible with SCSH.
6506 *** There are new accessors and setters for the broken-out time vectors
6507 returned by `localtime', `gmtime', and that ilk. They are:
6509 Component Accessor Setter
6510 ========================= ============ ============
6511 seconds tm:sec set-tm:sec
6512 minutes tm:min set-tm:min
6513 hours tm:hour set-tm:hour
6514 day of the month tm:mday set-tm:mday
6515 month tm:mon set-tm:mon
6516 year tm:year set-tm:year
6517 day of the week tm:wday set-tm:wday
6518 day in the year tm:yday set-tm:yday
6519 daylight saving time tm:isdst set-tm:isdst
6520 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6521 name of time zone tm:zone set-tm:zone
6523 *** There are new accessors for the vectors returned by `uname',
6524 describing the host system:
6527 ============================================== ================
6528 name of the operating system implementation utsname:sysname
6529 network name of this machine utsname:nodename
6530 release level of the operating system utsname:release
6531 version level of the operating system utsname:version
6532 machine hardware platform utsname:machine
6534 *** There are new accessors for the vectors returned by `getpw',
6535 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6536 system's user database:
6539 ====================== =================
6540 user name passwd:name
6541 user password passwd:passwd
6544 real name passwd:gecos
6545 home directory passwd:dir
6546 shell program passwd:shell
6548 *** There are new accessors for the vectors returned by `getgr',
6549 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6550 system's group database:
6553 ======================= ============
6554 group name group:name
6555 group password group:passwd
6557 group members group:mem
6559 *** There are new accessors for the vectors returned by `gethost',
6560 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6564 ========================= ===============
6565 official name of host hostent:name
6566 alias list hostent:aliases
6567 host address type hostent:addrtype
6568 length of address hostent:length
6569 list of addresses hostent:addr-list
6571 *** There are new accessors for the vectors returned by `getnet',
6572 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6576 ========================= ===============
6577 official name of net netent:name
6578 alias list netent:aliases
6579 net number type netent:addrtype
6580 net number netent:net
6582 *** There are new accessors for the vectors returned by `getproto',
6583 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6587 ========================= ===============
6588 official protocol name protoent:name
6589 alias list protoent:aliases
6590 protocol number protoent:proto
6592 *** There are new accessors for the vectors returned by `getserv',
6593 `getservbyname', `getservbyport', and `getservent', describing
6597 ========================= ===============
6598 official service name servent:name
6599 alias list servent:aliases
6600 port number servent:port
6601 protocol to use servent:proto
6603 *** There are new accessors for the sockaddr structures returned by
6604 `accept', `getsockname', `getpeername', `recvfrom!':
6607 ======================================== ===============
6608 address format (`family') sockaddr:fam
6609 path, for file domain addresses sockaddr:path
6610 address, for internet domain addresses sockaddr:addr
6611 TCP or UDP port, for internet sockaddr:port
6613 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6614 `getprotoent', and `getservent' functions now return #f at the end of
6615 the user database. (They used to throw an exception.)
6617 Note that calling MUMBLEent function is equivalent to calling the
6618 corresponding MUMBLE function with no arguments.
6620 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6621 `setprotoent', and `setservent' routines now take no arguments.
6623 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6624 provide more useful information when they throw an exception.
6626 *** The `lnaof' function has been renamed to `inet-lnaof'.
6628 *** Guile now claims to have the `current-time' feature.
6630 *** The `mktime' function now takes an optional second argument ZONE,
6631 giving the time zone to use for the conversion. ZONE should be a
6632 string, in the same format as expected for the "TZ" environment variable.
6634 *** The `strptime' function now returns a pair (TIME . COUNT), where
6635 TIME is the parsed time as a vector, and COUNT is the number of
6636 characters from the string left unparsed. This function used to
6637 return the remaining characters as a string.
6639 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6640 The return value is now (SECONDS . MICROSECONDS); the fractional
6641 component is no longer expressed in "ticks".
6643 *** The `ticks/sec' constant has been removed, in light of the above change.
6645 * Changes to the gh_ interface
6647 ** gh_eval_str() now returns an SCM object which is the result of the
6650 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6653 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6654 and returns the array
6656 ** gh_scm2str0() is gone: there is no need to distinguish
6657 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6658 the user to interpret the data both ways.
6660 * Changes to the scm_ interface
6662 ** The new function scm_symbol_value0 provides an easy way to get a
6663 symbol's value from C code:
6665 SCM scm_symbol_value0 (char *NAME)
6666 Return the value of the symbol named by the null-terminated string
6667 NAME in the current module. If the symbol named NAME is unbound in
6668 the current module, return SCM_UNDEFINED.
6670 ** The new function scm_sysintern0 creates new top-level variables,
6671 without assigning them a value.
6673 SCM scm_sysintern0 (char *NAME)
6674 Create a new Scheme top-level variable named NAME. NAME is a
6675 null-terminated string. Return the variable's value cell.
6677 ** The function scm_internal_catch is the guts of catch. It handles
6678 all the mechanics of setting up a catch target, invoking the catch
6679 body, and perhaps invoking the handler if the body does a throw.
6681 The function is designed to be usable from C code, but is general
6682 enough to implement all the semantics Guile Scheme expects from throw.
6684 TAG is the catch tag. Typically, this is a symbol, but this function
6685 doesn't actually care about that.
6687 BODY is a pointer to a C function which runs the body of the catch;
6688 this is the code you can throw from. We call it like this:
6689 BODY (BODY_DATA, JMPBUF)
6691 BODY_DATA is just the BODY_DATA argument we received; we pass it
6692 through to BODY as its first argument. The caller can make
6693 BODY_DATA point to anything useful that BODY might need.
6694 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6695 which we have just created and initialized.
6697 HANDLER is a pointer to a C function to deal with a throw to TAG,
6698 should one occur. We call it like this:
6699 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6701 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6702 same idea as BODY_DATA above.
6703 THROWN_TAG is the tag that the user threw to; usually this is
6704 TAG, but it could be something else if TAG was #t (i.e., a
6705 catch-all), or the user threw to a jmpbuf.
6706 THROW_ARGS is the list of arguments the user passed to the THROW
6709 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6710 is just a pointer we pass through to HANDLER. We don't actually
6711 use either of those pointers otherwise ourselves. The idea is
6712 that, if our caller wants to communicate something to BODY or
6713 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6714 HANDLER can then use. Think of it as a way to make BODY and
6715 HANDLER closures, not just functions; MUMBLE_DATA points to the
6718 Of course, it's up to the caller to make sure that any data a
6719 MUMBLE_DATA needs is protected from GC. A common way to do this is
6720 to make MUMBLE_DATA a pointer to data stored in an automatic
6721 structure variable; since the collector must scan the stack for
6722 references anyway, this assures that any references in MUMBLE_DATA
6725 ** The new function scm_internal_lazy_catch is exactly like
6726 scm_internal_catch, except:
6728 - It does not unwind the stack (this is the major difference).
6729 - If handler returns, its value is returned from the throw.
6730 - BODY always receives #f as its JMPBUF argument (since there's no
6731 jmpbuf associated with a lazy catch, because we don't unwind the
6734 ** scm_body_thunk is a new body function you can pass to
6735 scm_internal_catch if you want the body to be like Scheme's `catch'
6736 --- a thunk, or a function of one argument if the tag is #f.
6738 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6739 contains the Scheme procedure to invoke as the body, and the tag
6740 we're catching. If the tag is #f, then we pass JMPBUF (created by
6741 scm_internal_catch) to the body procedure; otherwise, the body gets
6744 ** scm_handle_by_proc is a new handler function you can pass to
6745 scm_internal_catch if you want the handler to act like Scheme's catch
6746 --- call a procedure with the tag and the throw arguments.
6748 If the user does a throw to this catch, this function runs a handler
6749 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6750 variable holding the Scheme procedure object to invoke. It ought to
6751 be a pointer to an automatic variable (i.e., one living on the stack),
6752 or the procedure object should be otherwise protected from GC.
6754 ** scm_handle_by_message is a new handler function to use with
6755 `scm_internal_catch' if you want Guile to print a message and die.
6756 It's useful for dealing with throws to uncaught keys at the top level.
6758 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6759 message header to print; if zero, we use "guile" instead. That
6760 text is followed by a colon, then the message described by ARGS.
6762 ** The return type of scm_boot_guile is now void; the function does
6763 not return a value, and indeed, never returns at all.
6765 ** The new function scm_shell makes it easy for user applications to
6766 process command-line arguments in a way that is compatible with the
6767 stand-alone guile interpreter (which is in turn compatible with SCSH,
6770 To use the scm_shell function, first initialize any guile modules
6771 linked into your application, and then call scm_shell with the values
6772 of ARGC and ARGV your `main' function received. scm_shell will add
6773 any SCSH-style meta-arguments from the top of the script file to the
6774 argument vector, and then process the command-line arguments. This
6775 generally means loading a script file or starting up an interactive
6776 command interpreter. For details, see "Changes to the stand-alone
6779 ** The new functions scm_get_meta_args and scm_count_argv help you
6780 implement the SCSH-style meta-argument, `\'.
6782 char **scm_get_meta_args (int ARGC, char **ARGV)
6783 If the second element of ARGV is a string consisting of a single
6784 backslash character (i.e. "\\" in Scheme notation), open the file
6785 named by the following argument, parse arguments from it, and return
6786 the spliced command line. The returned array is terminated by a
6789 For details of argument parsing, see above, under "guile now accepts
6790 command-line arguments compatible with SCSH..."
6792 int scm_count_argv (char **ARGV)
6793 Count the arguments in ARGV, assuming it is terminated by a null
6796 For an example of how these functions might be used, see the source
6797 code for the function scm_shell in libguile/script.c.
6799 You will usually want to use scm_shell instead of calling this
6802 ** The new function scm_compile_shell_switches turns an array of
6803 command-line arguments into Scheme code to carry out the actions they
6804 describe. Given ARGC and ARGV, it returns a Scheme expression to
6805 evaluate, and calls scm_set_program_arguments to make any remaining
6806 command-line arguments available to the Scheme code. For example,
6807 given the following arguments:
6809 -e main -s ekko a speckled gecko
6811 scm_set_program_arguments will return the following expression:
6813 (begin (load "ekko") (main (command-line)) (quit))
6815 You will usually want to use scm_shell instead of calling this
6818 ** The function scm_shell_usage prints a usage message appropriate for
6819 an interpreter that uses scm_compile_shell_switches to handle its
6820 command-line arguments.
6822 void scm_shell_usage (int FATAL, char *MESSAGE)
6823 Print a usage message to the standard error output. If MESSAGE is
6824 non-zero, write it before the usage message, followed by a newline.
6825 If FATAL is non-zero, exit the process, using FATAL as the
6826 termination status. (If you want to be compatible with Guile,
6827 always use 1 as the exit status when terminating due to command-line
6830 You will usually want to use scm_shell instead of calling this
6833 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6834 expressions. It used to return SCM_EOL. Earth-shattering.
6836 ** The macros for declaring scheme objects in C code have been
6837 rearranged slightly. They are now:
6839 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6840 Declare a static SCM variable named C_NAME, and initialize it to
6841 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6842 be a C identifier, and SCHEME_NAME should be a C string.
6844 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6845 Just like SCM_SYMBOL, but make C_NAME globally visible.
6847 SCM_VCELL (C_NAME, SCHEME_NAME)
6848 Create a global variable at the Scheme level named SCHEME_NAME.
6849 Declare a static SCM variable named C_NAME, and initialize it to
6850 point to the Scheme variable's value cell.
6852 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6853 Just like SCM_VCELL, but make C_NAME globally visible.
6855 The `guile-snarf' script writes initialization code for these macros
6856 to its standard output, given C source code as input.
6858 The SCM_GLOBAL macro is gone.
6860 ** The scm_read_line and scm_read_line_x functions have been replaced
6861 by Scheme code based on the %read-delimited! procedure (known to C
6862 code as scm_read_delimited_x). See its description above for more
6865 ** The function scm_sys_open has been renamed to scm_open. It now
6866 returns a port instead of an FD object.
6868 * The dynamic linking support has changed. For more information, see
6869 libguile/DYNAMIC-LINKING.
6874 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6877 * Changes to the 'guile' program:
6879 ** Guile now loads some new files when it starts up. Guile first
6880 searches the load path for init.scm, and loads it if found. Then, if
6881 Guile is not being used to execute a script, and the user's home
6882 directory contains a file named `.guile', Guile loads that.
6884 ** You can now use Guile as a shell script interpreter.
6886 To paraphrase the SCSH manual:
6888 When Unix tries to execute an executable file whose first two
6889 characters are the `#!', it treats the file not as machine code to
6890 be directly executed by the native processor, but as source code
6891 to be executed by some interpreter. The interpreter to use is
6892 specified immediately after the #! sequence on the first line of
6893 the source file. The kernel reads in the name of the interpreter,
6894 and executes that instead. It passes the interpreter the source
6895 filename as its first argument, with the original arguments
6896 following. Consult the Unix man page for the `exec' system call
6897 for more information.
6899 Now you can use Guile as an interpreter, using a mechanism which is a
6900 compatible subset of that provided by SCSH.
6902 Guile now recognizes a '-s' command line switch, whose argument is the
6903 name of a file of Scheme code to load. It also treats the two
6904 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6905 to make a file of Scheme code directly executable by Unix, insert the
6906 following two lines at the top of the file:
6908 #!/usr/local/bin/guile -s
6911 Guile treats the argument of the `-s' command-line switch as the name
6912 of a file of Scheme code to load, and treats the sequence `#!' as the
6913 start of a block comment, terminated by `!#'.
6915 For example, here's a version of 'echo' written in Scheme:
6917 #!/usr/local/bin/guile -s
6919 (let loop ((args (cdr (program-arguments))))
6922 (display (car args))
6923 (if (pair? (cdr args))
6925 (loop (cdr args)))))
6928 Why does `#!' start a block comment terminated by `!#', instead of the
6929 end of the line? That is the notation SCSH uses, and although we
6930 don't yet support the other SCSH features that motivate that choice,
6931 we would like to be backward-compatible with any existing Guile
6932 scripts once we do. Furthermore, if the path to Guile on your system
6933 is too long for your kernel, you can start the script with this
6937 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6940 Note that some very old Unix systems don't support the `#!' syntax.
6943 ** You can now run Guile without installing it.
6945 Previous versions of the interactive Guile interpreter (`guile')
6946 couldn't start up unless Guile's Scheme library had been installed;
6947 they used the value of the environment variable `SCHEME_LOAD_PATH'
6948 later on in the startup process, but not to find the startup code
6949 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6952 To run Guile without installing it, build it in the normal way, and
6953 then set the environment variable `SCHEME_LOAD_PATH' to a
6954 colon-separated list of directories, including the top-level directory
6955 of the Guile sources. For example, if you unpacked Guile so that the
6956 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6959 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6962 ** Guile's read-eval-print loop no longer prints #<unspecified>
6963 results. If the user wants to see this, she can evaluate the
6964 expression (assert-repl-print-unspecified #t), perhaps in her startup
6967 ** Guile no longer shows backtraces by default when an error occurs;
6968 however, it does display a message saying how to get one, and how to
6969 request that they be displayed by default. After an error, evaluate
6971 to see a backtrace, and
6972 (debug-enable 'backtrace)
6973 to see them by default.
6977 * Changes to Guile Scheme:
6979 ** Guile now distinguishes between #f and the empty list.
6981 This is for compatibility with the IEEE standard, the (possibly)
6982 upcoming Revised^5 Report on Scheme, and many extant Scheme
6985 Guile used to have #f and '() denote the same object, to make Scheme's
6986 type system more compatible with Emacs Lisp's. However, the change
6987 caused too much trouble for Scheme programmers, and we found another
6988 way to reconcile Emacs Lisp with Scheme that didn't require this.
6991 ** Guile's delq, delv, delete functions, and their destructive
6992 counterparts, delq!, delv!, and delete!, now remove all matching
6993 elements from the list, not just the first. This matches the behavior
6994 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6995 functions which inspired them.
6997 I recognize that this change may break code in subtle ways, but it
6998 seems best to make the change before the FSF's first Guile release,
7002 ** The compiled-library-path function has been deleted from libguile.
7004 ** The facilities for loading Scheme source files have changed.
7006 *** The variable %load-path now tells Guile which directories to search
7007 for Scheme code. Its value is a list of strings, each of which names
7010 *** The variable %load-extensions now tells Guile which extensions to
7011 try appending to a filename when searching the load path. Its value
7012 is a list of strings. Its default value is ("" ".scm").
7014 *** (%search-load-path FILENAME) searches the directories listed in the
7015 value of the %load-path variable for a Scheme file named FILENAME,
7016 with all the extensions listed in %load-extensions. If it finds a
7017 match, then it returns its full filename. If FILENAME is absolute, it
7018 returns it unchanged. Otherwise, it returns #f.
7020 %search-load-path will not return matches that refer to directories.
7022 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7023 uses %seach-load-path to find a file named FILENAME, and loads it if
7024 it finds it. If it can't read FILENAME for any reason, it throws an
7027 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7030 *** load uses the same searching semantics as primitive-load.
7032 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7033 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7034 path, and load-module-with-path have been deleted. The functions
7035 above should serve their purposes.
7037 *** If the value of the variable %load-hook is a procedure,
7038 `primitive-load' applies its value to the name of the file being
7039 loaded (without the load path directory name prepended). If its value
7040 is #f, it is ignored. Otherwise, an error occurs.
7042 This is mostly useful for printing load notification messages.
7045 ** The function `eval!' is no longer accessible from the scheme level.
7046 We can't allow operations which introduce glocs into the scheme level,
7047 because Guile's type system can't handle these as data. Use `eval' or
7048 `read-and-eval!' (see below) as replacement.
7050 ** The new function read-and-eval! reads an expression from PORT,
7051 evaluates it, and returns the result. This is more efficient than
7052 simply calling `read' and `eval', since it is not necessary to make a
7053 copy of the expression for the evaluator to munge.
7055 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7056 for the `read' function.
7059 ** The function `int?' has been removed; its definition was identical
7060 to that of `integer?'.
7062 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7063 use the R4RS names for these functions.
7065 ** The function object-properties no longer returns the hash handle;
7066 it simply returns the object's property list.
7068 ** Many functions have been changed to throw errors, instead of
7069 returning #f on failure. The point of providing exception handling in
7070 the language is to simplify the logic of user code, but this is less
7071 useful if Guile's primitives don't throw exceptions.
7073 ** The function `fileno' has been renamed from `%fileno'.
7075 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7078 * Changes to Guile's C interface:
7080 ** The library's initialization procedure has been simplified.
7081 scm_boot_guile now has the prototype:
7083 void scm_boot_guile (int ARGC,
7085 void (*main_func) (),
7088 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7089 MAIN_FUNC should do all the work of the program (initializing other
7090 packages, reading user input, etc.) before returning. When MAIN_FUNC
7091 returns, call exit (0); this function never returns. If you want some
7092 other exit value, MAIN_FUNC may call exit itself.
7094 scm_boot_guile arranges for program-arguments to return the strings
7095 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7096 scm_set_program_arguments with the final list, so Scheme code will
7097 know which arguments have been processed.
7099 scm_boot_guile establishes a catch-all catch handler which prints an
7100 error message and exits the process. This means that Guile exits in a
7101 coherent way when system errors occur and the user isn't prepared to
7102 handle it. If the user doesn't like this behavior, they can establish
7103 their own universal catcher in MAIN_FUNC to shadow this one.
7105 Why must the caller do all the real work from MAIN_FUNC? The garbage
7106 collector assumes that all local variables of type SCM will be above
7107 scm_boot_guile's stack frame on the stack. If you try to manipulate
7108 SCM values after this function returns, it's the luck of the draw
7109 whether the GC will be able to find the objects you allocate. So,
7110 scm_boot_guile function exits, rather than returning, to discourage
7111 people from making that mistake.
7113 The IN, OUT, and ERR arguments were removed; there are other
7114 convenient ways to override these when desired.
7116 The RESULT argument was deleted; this function should never return.
7118 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7122 ** Guile's header files should no longer conflict with your system's
7125 In order to compile code which #included <libguile.h>, previous
7126 versions of Guile required you to add a directory containing all the
7127 Guile header files to your #include path. This was a problem, since
7128 Guile's header files have names which conflict with many systems'
7131 Now only <libguile.h> need appear in your #include path; you must
7132 refer to all Guile's other header files as <libguile/mumble.h>.
7133 Guile's installation procedure puts libguile.h in $(includedir), and
7134 the rest in $(includedir)/libguile.
7137 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7138 have been added to the Guile library.
7140 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7141 OBJ will not be freed, even if all other references are dropped,
7142 until someone does scm_unprotect_object (OBJ). Both functions
7145 Note that calls to scm_protect_object do not nest. You can call
7146 scm_protect_object any number of times on a given object, and the
7147 next call to scm_unprotect_object will unprotect it completely.
7149 Basically, scm_protect_object and scm_unprotect_object just
7150 maintain a list of references to things. Since the GC knows about
7151 this list, all objects it mentions stay alive. scm_protect_object
7152 adds its argument to the list; scm_unprotect_object remove its
7153 argument from the list.
7156 ** scm_eval_0str now returns the value of the last expression
7159 ** The new function scm_read_0str reads an s-expression from a
7160 null-terminated string, and returns it.
7162 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7163 to a Scheme port object.
7165 ** The new function `scm_set_program_arguments' allows C code to set
7166 the value returned by the Scheme `program-arguments' function.
7171 * Guile no longer includes sophisticated Tcl/Tk support.
7173 The old Tcl/Tk support was unsatisfying to us, because it required the
7174 user to link against the Tcl library, as well as Tk and Guile. The
7175 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7176 referring to widgets by names, rather than exporting widgets to Scheme
7177 code as a special datatype.
7179 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7180 maintainers described some very interesting changes in progress to the
7181 Tcl/Tk internals, which would facilitate clean interfaces between lone
7182 Tk and other interpreters --- even for garbage-collected languages
7183 like Scheme. They expected the new Tk to be publicly available in the
7186 Since it seems that Guile might soon have a new, cleaner interface to
7187 lone Tk, and that the old Guile/Tk glue code would probably need to be
7188 completely rewritten, we (Jim Blandy and Richard Stallman) have
7189 decided not to support the old code. We'll spend the time instead on
7190 a good interface to the newer Tk, as soon as it is available.
7192 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7195 Copyright information:
7197 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7199 Permission is granted to anyone to make or distribute verbatim copies
7200 of this document as received, in any medium, provided that the
7201 copyright notice and this permission notice are preserved,
7202 thus giving the recipient permission to redistribute in turn.
7204 Permission is granted to distribute modified versions
7205 of this document, or of portions of it,
7206 under the above conditions, provided also that they
7207 carry prominent notices stating who last changed them.
7212 paragraph-separate: "[
\f]*$"