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 ** The `(ice-9 slib)' module now works as expected
45 ** Expressions like "(set! 'x #t)" no longer yield a crash
46 ** Warnings about duplicate bindings now go to stderr
47 ** A memory leak in `make-socket-address' was fixed
48 ** Alignment issues (e.g., on SPARC) in network routines were fixed
49 ** A threading issue that showed up at least on NetBSD was fixed
50 ** Build problems on Solaris and IRIX fixed
52 * Implementation improvements
54 ** The reader is now faster, which reduces startup time
55 ** Procedures returned by `record-accessor' and `record-modifier' are faster
59 Changes in 1.8.2 (since 1.8.1):
61 * New procedures (see the manual for details)
63 ** set-program-arguments
68 ** Fractions were not `equal?' if stored in unreduced form.
69 (A subtle problem, since printing a value reduced it, making it work.)
70 ** srfi-60 `copy-bit' failed on 64-bit systems
71 ** "guile --use-srfi" option at the REPL can replace core functions
72 (Programs run with that option were ok, but in the interactive REPL
73 the core bindings got priority, preventing SRFI replacements or
75 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
76 ** `kill' on mingw throws an error for a PID other than oneself
77 ** Procedure names are attached to procedure-with-setters
78 ** Array read syntax works with negative lower bound
79 ** `array-in-bounds?' fix if an array has different lower bounds on each index
80 ** `*' returns exact 0 for "(* inexact 0)"
81 This follows what it always did for "(* 0 inexact)".
82 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
83 ** SRFI-19: `date->julian-day' did not account for timezone offset
84 ** `ttyname' no longer crashes when passed a non-tty argument
85 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
86 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
87 ** GOOPS: Fixed a bug in `method-more-specific?'
88 ** Build problems on Solaris fixed
89 ** Build problems on HP-UX IA64 fixed
90 ** Build problems on MinGW fixed
93 Changes in 1.8.1 (since 1.8.0):
95 * LFS functions are now used to access 64-bit files on 32-bit systems.
97 * New procedures (see the manual for details)
99 ** primitive-_exit - [Scheme] the-root-module
100 ** scm_primitive__exit - [C]
101 ** make-completion-function - [Scheme] (ice-9 readline)
102 ** scm_c_locale_stringn_to_number - [C]
103 ** scm_srfi1_append_reverse [C]
104 ** scm_srfi1_append_reverse_x [C]
110 * New `(ice-9 i18n)' module (see the manual for details)
114 ** Build problems have been fixed on MacOS, SunOS, and QNX.
116 ** `strftime' fix sign of %z timezone offset.
118 ** A one-dimensional array can now be 'equal?' to a vector.
120 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
122 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
124 ** `record-accessor' and `record-modifier' now have strict type checks.
126 Record accessor and modifier procedures now throw an error if the
127 record type of the record they're given is not the type expected.
128 (Previously accessors returned #f and modifiers silently did nothing).
130 ** It is now OK to use both autoload and use-modules on a given module.
132 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
134 Previously there was no checking on primatives like make-vector that
135 accept "one or two" arguments. Now there is.
137 ** The srfi-1 assoc function now calls its equality predicate properly.
139 Previously srfi-1 assoc would call the equality predicate with the key
140 last. According to the SRFI, the key should be first.
142 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
144 ** The array-set! procedure no longer segfaults when given a bit vector.
146 ** Bugs in make-shared-array have been fixed.
148 ** string<? and friends now follow char<? etc order on 8-bit chars.
150 ** The format procedure now handles inf and nan values for ~f correctly.
152 ** exact->inexact should no longer overflow when given certain large fractions.
154 ** srfi-9 accessor and modifier procedures now have strict record type checks.
156 This matches the srfi-9 specification.
158 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
160 Previously the (ice-9 ftw) procedures would ignore any file that had
161 the same inode number as a file they had already seen, even if that
162 file was on a different device.
165 Changes in 1.8.0 (changes since the 1.6.x series):
167 * Changes to the distribution
169 ** Guile is now licensed with the GNU Lesser General Public License.
171 ** The manual is now licensed with the GNU Free Documentation License.
173 ** Guile now requires GNU MP (http://swox.com/gmp).
175 Guile now uses the GNU MP library for arbitrary precision arithmetic.
177 ** Guile now has separate private and public configuration headers.
179 That is, things like HAVE_STRING_H no longer leak from Guile's
182 ** Guile now provides and uses an "effective" version number.
184 Guile now provides scm_effective_version and effective-version
185 functions which return the "effective" version number. This is just
186 the normal full version string without the final micro-version number,
187 so the current effective-version is "1.8". The effective version
188 should remain unchanged during a stable series, and should be used for
189 items like the versioned share directory name
190 i.e. /usr/share/guile/1.8.
192 Providing an unchanging version number during a stable release for
193 things like the versioned share directory can be particularly
194 important for Guile "add-on" packages, since it provides a directory
195 that they can install to that won't be changed out from under them
196 with each micro release during a stable series.
198 ** Thread implementation has changed.
200 When you configure "--with-threads=null", you will get the usual
201 threading API (call-with-new-thread, make-mutex, etc), but you can't
202 actually create new threads. Also, "--with-threads=no" is now
203 equivalent to "--with-threads=null". This means that the thread API
204 is always present, although you might not be able to create new
207 When you configure "--with-threads=pthreads" or "--with-threads=yes",
208 you will get threads that are implemented with the portable POSIX
209 threads. These threads can run concurrently (unlike the previous
210 "coop" thread implementation), but need to cooperate for things like
213 The default is "pthreads", unless your platform doesn't have pthreads,
214 in which case "null" threads are used.
216 See the manual for details, nodes "Initialization", "Multi-Threading",
217 "Blocking", and others.
219 ** There is the new notion of 'discouraged' features.
221 This is a milder form of deprecation.
223 Things that are discouraged should not be used in new code, but it is
224 OK to leave them in old code for now. When a discouraged feature is
225 used, no warning message is printed like there is for 'deprecated'
226 features. Also, things that are merely discouraged are nevertheless
227 implemented efficiently, while deprecated features can be very slow.
229 You can omit discouraged features from libguile by configuring it with
230 the '--disable-discouraged' option.
232 ** Deprecation warnings can be controlled at run-time.
234 (debug-enable 'warn-deprecated) switches them on and (debug-disable
235 'warn-deprecated) switches them off.
237 ** Support for SRFI 61, extended cond syntax for multiple values has
240 This SRFI is always available.
242 ** Support for require-extension, SRFI-55, has been added.
244 The SRFI-55 special form `require-extension' has been added. It is
245 available at startup, and provides a portable way to load Scheme
246 extensions. SRFI-55 only requires support for one type of extension,
247 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
250 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
252 The (srfi srfi-26) module is an implementation of SRFI-26 which
253 provides the `cut' and `cute' syntax. These may be used to specialize
254 parameters without currying.
256 ** New module (srfi srfi-31)
258 This is an implementation of SRFI-31 which provides a special form
259 `rec' for recursive evaluation.
261 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
262 been merged with the core, making their functionality always
265 The modules are still available, tho, and you could use them together
266 with a renaming import, for example.
268 ** Guile no longer includes its own version of libltdl.
270 The official version is good enough now.
272 ** The --enable-htmldoc option has been removed from 'configure'.
274 Support for translating the documentation into HTML is now always
275 provided. Use 'make html'.
277 ** New module (ice-9 serialize):
279 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
280 don't trust the thread safety of most of your program, but where you
281 have some section(s) of code which you consider can run in parallel to
282 other sections. See ice-9/serialize.scm for more information.
284 ** The configure option '--disable-arrays' has been removed.
286 Support for arrays and uniform numeric arrays is now always included
289 * Changes to the stand-alone interpreter
291 ** New command line option `-L'.
293 This option adds a directory to the front of the load path.
295 ** New command line option `--no-debug'.
297 Specifying `--no-debug' on the command line will keep the debugging
298 evaluator turned off, even for interactive sessions.
300 ** User-init file ~/.guile is now loaded with the debugging evaluator.
302 Previously, the normal evaluator would have been used. Using the
303 debugging evaluator gives better error messages.
305 ** The '-e' option now 'read's its argument.
307 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
308 be used with '-e'. For example, you can now write a script like
311 exec guile -e '(@ (demo) main)' -s "$0" "$@"
314 (define-module (demo)
318 (format #t "Demo: ~a~%" args))
321 * Changes to Scheme functions and syntax
323 ** Guardians have changed back to their original semantics
325 Guardians now behave like described in the paper by Dybvig et al. In
326 particular, they no longer make guarantees about the order in which
327 they return objects, and they can no longer be greedy.
329 They no longer drop cyclic data structures.
331 The C function scm_make_guardian has been changed incompatibly and no
332 longer takes the 'greedy_p' argument.
334 ** New function hashx-remove!
336 This function completes the set of 'hashx' functions.
338 ** The concept of dynamic roots has been factored into continuation
339 barriers and dynamic states.
341 Each thread has a current dynamic state that carries the values of the
342 fluids. You can create and copy dynamic states and use them as the
343 second argument for 'eval'. See "Fluids and Dynamic States" in the
346 To restrict the influence that captured continuations can have on the
347 control flow, you can errect continuation barriers. See "Continuation
348 Barriers" in the manual.
350 The function call-with-dynamic-root now essentially temporarily
351 installs a new dynamic state and errects a continuation barrier.
353 ** The default load path no longer includes "." at the end.
355 Automatically loading modules from the current directory should not
356 happen by default. If you want to allow it in a more controlled
357 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
360 ** The uniform vector and array support has been overhauled.
362 It now complies with SRFI-4 and the weird prototype based uniform
363 array creation has been deprecated. See the manual for more details.
365 Some non-compatible changes have been made:
366 - characters can no longer be stored into byte arrays.
367 - strings and bit vectors are no longer considered to be uniform numeric
369 - array-rank throws an error for non-arrays instead of returning zero.
370 - array-ref does no longer accept non-arrays when no indices are given.
372 There is the new notion of 'generalized vectors' and corresponding
373 procedures like 'generalized-vector-ref'. Generalized vectors include
374 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
376 Arrays use generalized vectors as their storage, so that you still
377 have arrays of characters, bits, etc. However, uniform-array-read!
378 and uniform-array-write can no longer read/write strings and
381 ** There is now support for copy-on-write substrings, mutation-sharing
382 substrings and read-only strings.
384 Three new procedures are related to this: substring/shared,
385 substring/copy, and substring/read-only. See the manual for more
388 ** Backtraces will now highlight the value that caused the error.
390 By default, these values are enclosed in "{...}", such as in this
399 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
400 <unnamed port>:1:1: Wrong type (expecting pair): a
401 ABORT: (wrong-type-arg)
403 The prefix and suffix used for highlighting can be set via the two new
404 printer options 'highlight-prefix' and 'highlight-suffix'. For
405 example, putting this into ~/.guile will output the bad value in bold
408 (print-set! highlight-prefix "\x1b[1m")
409 (print-set! highlight-suffix "\x1b[22m")
412 ** 'gettext' support for internationalization has been added.
414 See the manual for details.
416 ** New syntax '@' and '@@':
418 You can now directly refer to variables exported from a module by
421 (@ MODULE-NAME VARIABLE-NAME)
423 For example (@ (ice-9 pretty-print) pretty-print) will directly access
424 the pretty-print variable exported from the (ice-9 pretty-print)
425 module. You don't need to 'use' that module first. You can also use
426 '@' as a target of 'set!', as in (set! (@ mod var) val).
428 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
429 but it can also access variables that have not been exported. It is
430 intended only for kluges and temporary fixes and for debugging, not
433 ** Keyword syntax has been made more disciplined.
435 Previously, the name of a keyword was read as a 'token' but printed as
436 a symbol. Now, it is read as a general Scheme datum which must be a
447 ERROR: In expression (a b c):
451 ERROR: Unbound variable: foo
456 ERROR: Wrong type (expecting symbol): 12
460 ERROR: Wrong type (expecting symbol): (a b c)
464 ** The printing of symbols that might look like keywords can be
467 The new printer option 'quote-keywordish-symbols' controls how symbols
468 are printed that have a colon as their first or last character. The
469 default now is to only quote a symbol with #{...}# when the read
470 option 'keywords' is not '#f'. Thus:
472 guile> (define foo (string->symbol ":foo"))
473 guile> (read-set! keywords #f)
476 guile> (read-set! keywords 'prefix)
479 guile> (print-set! quote-keywordish-symbols #f)
483 ** 'while' now provides 'break' and 'continue'
485 break and continue were previously bound in a while loop, but not
486 documented, and continue didn't quite work properly. The undocumented
487 parameter to break which gave a return value for the while has been
490 ** 'call-with-current-continuation' is now also available under the name
493 ** The module system now checks for duplicate bindings.
495 The module system now can check for name conflicts among imported
498 The behavior can be controlled by specifying one or more 'duplicates'
499 handlers. For example, to make Guile return an error for every name
507 The new default behavior of the module system when a name collision
508 has been detected is to
510 1. Give priority to bindings marked as a replacement.
511 2. Issue a warning (different warning if overriding core binding).
512 3. Give priority to the last encountered binding (this corresponds to
515 If you want the old behavior back without replacements or warnings you
518 (default-duplicate-binding-handler 'last)
520 to your .guile init file.
522 ** New define-module option: :replace
524 :replace works as :export, but, in addition, marks the binding as a
527 A typical example is `format' in (ice-9 format) which is a replacement
528 for the core binding `format'.
530 ** Adding prefixes to imported bindings in the module system
532 There is now a new :use-module option :prefix. It can be used to add
533 a prefix to all imported bindings.
536 :use-module ((bar) :prefix bar:))
538 will import all bindings exported from bar, but rename them by adding
541 ** Conflicting generic functions can be automatically merged.
543 When two imported bindings conflict and they are both generic
544 functions, the two functions can now be merged automatically. This is
545 activated with the 'duplicates' handler 'merge-generics'.
547 ** New function: effective-version
549 Returns the "effective" version number. This is just the normal full
550 version string without the final micro-version number. See "Changes
551 to the distribution" above.
553 ** New threading functions: parallel, letpar, par-map, and friends
555 These are convenient ways to run calculations in parallel in new
556 threads. See "Parallel forms" in the manual for details.
558 ** New function 'try-mutex'.
560 This function will attempt to lock a mutex but will return immediately
561 instead of blocking and indicate failure.
563 ** Waiting on a condition variable can have a timeout.
565 The function 'wait-condition-variable' now takes a third, optional
566 argument that specifies the point in time where the waiting should be
569 ** New function 'broadcast-condition-variable'.
571 ** New functions 'all-threads' and 'current-thread'.
573 ** Signals and system asyncs work better with threads.
575 The function 'sigaction' now takes a fourth, optional, argument that
576 specifies the thread that the handler should run in. When the
577 argument is omitted, the handler will run in the thread that called
580 Likewise, 'system-async-mark' takes a second, optional, argument that
581 specifies the thread that the async should run in. When it is
582 omitted, the async will run in the thread that called
585 C code can use the new functions scm_sigaction_for_thread and
586 scm_system_async_mark_for_thread to pass the new thread argument.
588 When a thread blocks on a mutex, a condition variable or is waiting
589 for IO to be possible, it will still execute system asyncs. This can
590 be used to interrupt such a thread by making it execute a 'throw', for
593 ** The function 'system-async' is deprecated.
595 You can now pass any zero-argument procedure to 'system-async-mark'.
596 The function 'system-async' will just return its argument unchanged
599 ** New functions 'call-with-blocked-asyncs' and
600 'call-with-unblocked-asyncs'
602 The expression (call-with-blocked-asyncs PROC) will call PROC and will
603 block execution of system asyncs for the current thread by one level
604 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
605 procedure and will unblock the execution of system asyncs by one
606 level for the current thread.
608 Only system asyncs are affected by these functions.
610 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
612 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
613 instead. Those functions are easier to use correctly and can be
616 ** New function 'unsetenv'.
618 ** New macro 'define-syntax-public'.
620 It works like 'define-syntax' and also exports the defined macro (but
623 ** There is support for Infinity and NaNs.
625 Following PLT Scheme, Guile can now work with infinite numbers, and
628 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
629 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
630 "+nan.0"). These numbers are inexact and have no exact counterpart.
632 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
633 sign of the dividend. The infinities are integers, and they answer #t
634 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
635 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
646 ERROR: Numerical overflow
648 Two new predicates 'inf?' and 'nan?' can be used to test for the
651 ** Inexact zero can have a sign.
653 Guile can now distinguish between plus and minus inexact zero, if your
654 platform supports this, too. The two zeros are equal according to
655 '=', but not according to 'eqv?'. For example
666 ** Guile now has exact rationals.
668 Guile can now represent fractions such as 1/3 exactly. Computing with
669 them is also done exactly, of course:
674 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
677 For example: (floor 2) now returns an exact 2 where in the past it
678 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
680 ** inexact->exact no longer returns only integers.
682 Without exact rationals, the closest exact number was always an
683 integer, but now inexact->exact returns the fraction that is exactly
684 equal to a floating point number. For example:
686 (inexact->exact 1.234)
687 => 694680242521899/562949953421312
689 When you want the old behavior, use 'round' explicitly:
691 (inexact->exact (round 1.234))
694 ** New function 'rationalize'.
696 This function finds a simple fraction that is close to a given real
697 number. For example (and compare with inexact->exact above):
699 (rationalize (inexact->exact 1.234) 1/2000)
702 Note that, as required by R5RS, rationalize returns only then an exact
703 result when both its arguments are exact.
705 ** 'odd?' and 'even?' work also for inexact integers.
707 Previously, (odd? 1.0) would signal an error since only exact integers
708 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
709 returns #f and (odd? 1.5) signals an error.
711 ** Guile now has uninterned symbols.
713 The new function 'make-symbol' will return an uninterned symbol. This
714 is a symbol that is unique and is guaranteed to remain unique.
715 However, uninterned symbols can not yet be read back in.
717 Use the new function 'symbol-interned?' to check whether a symbol is
720 ** pretty-print has more options.
722 The function pretty-print from the (ice-9 pretty-print) module can now
723 also be invoked with keyword arguments that control things like
724 maximum output width. See the manual for details.
726 ** Variables have no longer a special behavior for `equal?'.
728 Previously, comparing two variables with `equal?' would recursivly
729 compare their values. This is no longer done. Variables are now only
730 `equal?' if they are `eq?'.
732 ** `(begin)' is now valid.
734 You can now use an empty `begin' form. It will yield #<unspecified>
735 when evaluated and simply be ignored in a definition context.
737 ** Deprecated: procedure->macro
739 Change your code to use 'define-macro' or r5rs macros. Also, be aware
740 that macro expansion will not be done during evaluation, but prior to
743 ** Soft ports now allow a `char-ready?' procedure
745 The vector argument to `make-soft-port' can now have a length of
746 either 5 or 6. (Previously the length had to be 5.) The optional 6th
747 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
748 that returns the number of characters that can be read immediately
749 without the soft port blocking.
751 ** Deprecated: undefine
753 There is no replacement for undefine.
755 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
756 have been discouraged.
758 They are relics from a time where a keyword like #:foo was used
759 directly as a Tcl option "-foo" and thus keywords were internally
760 stored as a symbol with a starting dash. We now store a symbol
763 Use symbol->keyword and keyword->symbol instead.
765 ** The `cheap' debug option is now obsolete
767 Evaluator trap calls are now unconditionally "cheap" - in other words,
768 they pass a debug object to the trap handler rather than a full
769 continuation. The trap handler code can capture a full continuation
770 by using `call-with-current-continuation' in the usual way, if it so
773 The `cheap' option is retained for now so as not to break existing
774 code which gets or sets it, but setting it now has no effect. It will
775 be removed in the next major Guile release.
777 ** Evaluator trap calls now support `tweaking'
779 `Tweaking' means that the trap handler code can modify the Scheme
780 expression that is about to be evaluated (in the case of an
781 enter-frame trap) or the value that is being returned (in the case of
782 an exit-frame trap). The trap handler code indicates that it wants to
783 do this by returning a pair whose car is the symbol 'instead and whose
784 cdr is the modified expression or return value.
786 * Changes to the C interface
788 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
789 take a 'delete' function argument.
791 This argument makes no sense since the delete function is used to
792 remove a pair from an alist, and this must not be configurable.
794 This is an incompatible change.
796 ** The GH interface is now subject to the deprecation mechanism
798 The GH interface has been deprecated for quite some time but now it is
799 actually removed from Guile when it is configured with
800 --disable-deprecated.
802 See the manual "Transitioning away from GH" for more information.
804 ** A new family of functions for converting between C values and
805 Scheme values has been added.
807 These functions follow a common naming scheme and are designed to be
808 easier to use, thread-safe and more future-proof than the older
813 These are predicates that return a C boolean: 1 or 0. Instead of
814 SCM_NFALSEP, you can now use scm_is_true, for example.
816 - <type> scm_to_<type> (SCM val, ...)
818 These are functions that convert a Scheme value into an appropriate
819 C value. For example, you can use scm_to_int to safely convert from
822 - SCM scm_from_<type> (<type> val, ...)
824 These functions convert from a C type to a SCM value; for example,
825 scm_from_int for ints.
827 There is a huge number of these functions, for numbers, strings,
828 symbols, vectors, etc. They are documented in the reference manual in
829 the API section together with the types that they apply to.
831 ** New functions for dealing with complex numbers in C have been added.
833 The new functions are scm_c_make_rectangular, scm_c_make_polar,
834 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
835 They work like scm_make_rectangular etc but take or return doubles
838 ** The function scm_make_complex has been discouraged.
840 Use scm_c_make_rectangular instead.
842 ** The INUM macros have been deprecated.
844 A lot of code uses these macros to do general integer conversions,
845 although the macros only work correctly with fixnums. Use the
846 following alternatives.
848 SCM_INUMP -> scm_is_integer or similar
849 SCM_NINUMP -> !scm_is_integer or similar
850 SCM_MAKINUM -> scm_from_int or similar
851 SCM_INUM -> scm_to_int or similar
853 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
854 do the validating for you.
856 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
857 have been discouraged.
859 Use the newer scm_to_<type> and scm_from_<type> functions instead for
860 new code. The functions have been discouraged since they don't fit
863 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
865 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
866 evaluates its argument twice. Use scm_is_true, etc. instead for new
869 ** The macro SCM_EQ_P has been discouraged.
871 Use scm_is_eq for new code, which fits better into the naming
874 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
877 Use the function scm_is_pair or scm_is_null instead.
879 ** The functions scm_round and scm_truncate have been deprecated and
880 are now available as scm_c_round and scm_c_truncate, respectively.
882 These functions occupy the names that scm_round_number and
883 scm_truncate_number should have.
885 ** The functions scm_c_string2str, scm_c_substring2str, and
886 scm_c_symbol2str have been deprecated.
888 Use scm_to_locale_stringbuf or similar instead, maybe together with
891 ** New functions scm_c_make_string, scm_c_string_length,
892 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
893 scm_c_substring_shared, scm_c_substring_copy.
895 These are like scm_make_string, scm_length, etc. but are slightly
896 easier to use from C.
898 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
899 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
901 They export too many assumptions about the implementation of strings
902 and symbols that are no longer true in the presence of
903 mutation-sharing substrings and when Guile switches to some form of
906 When working with strings, it is often best to use the normal string
907 functions provided by Guile, such as scm_c_string_ref,
908 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
909 manual since many more such functions are now provided than
912 When you want to convert a SCM string to a C string, use the
913 scm_to_locale_string function or similar instead. For symbols, use
914 scm_symbol_to_string and then work with that string. Because of the
915 new string representation, scm_symbol_to_string does not need to copy
916 and is thus quite efficient.
918 ** Some string, symbol and keyword functions have been discouraged.
920 They don't fit into the uniform naming scheme and are not explicit
921 about the character encoding.
923 Replace according to the following table:
925 scm_allocate_string -> scm_c_make_string
926 scm_take_str -> scm_take_locale_stringn
927 scm_take0str -> scm_take_locale_string
928 scm_mem2string -> scm_from_locale_stringn
929 scm_str2string -> scm_from_locale_string
930 scm_makfrom0str -> scm_from_locale_string
931 scm_mem2symbol -> scm_from_locale_symboln
932 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
933 scm_str2symbol -> scm_from_locale_symbol
935 SCM_SYMBOL_HASH -> scm_hashq
936 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
938 scm_c_make_keyword -> scm_from_locale_keyword
940 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
941 now also available to C code.
943 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
945 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
946 the latter returns the true name of the keyword, not the 'dash name',
947 as SCM_KEYWORDSYM used to do.
949 ** A new way to access arrays in a thread-safe and efficient way has
952 See the manual, node "Accessing Arrays From C".
954 ** The old uniform vector and bitvector implementations have been
955 unceremoniously removed.
957 This implementation exposed the details of the tagging system of
958 Guile. Use the new C API explained in the manual in node "Uniform
959 Numeric Vectors" and "Bit Vectors", respectively.
961 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
962 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
963 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
964 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
965 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
966 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
969 ** The macros dealing with vectors have been deprecated.
971 Use the new functions scm_is_vector, scm_vector_elements,
972 scm_vector_writable_elements, etc, or scm_is_simple_vector,
973 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
974 manual for more details.
976 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
977 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
979 The following macros have been removed: SCM_VECTOR_BASE,
980 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
981 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
983 ** Some C functions and macros related to arrays have been deprecated.
985 Migrate according to the following table:
987 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
988 scm_make_ra -> scm_make_array
989 scm_shap2ra -> scm_make_array
990 scm_cvref -> scm_c_generalized_vector_ref
991 scm_ra_set_contp -> do not use
992 scm_aind -> scm_array_handle_pos
993 scm_raprin1 -> scm_display or scm_write
995 SCM_ARRAYP -> scm_is_array
996 SCM_ARRAY_NDIM -> scm_c_array_rank
997 SCM_ARRAY_DIMS -> scm_array_handle_dims
998 SCM_ARRAY_CONTP -> do not use
999 SCM_ARRAY_MEM -> do not use
1000 SCM_ARRAY_V -> scm_array_handle_elements or similar
1001 SCM_ARRAY_BASE -> do not use
1003 ** SCM_CELL_WORD_LOC has been deprecated.
1005 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1006 to a SCM, as opposed to a pointer to a scm_t_bits.
1008 This was done to allow the correct use of pointers into the Scheme
1009 heap. Previously, the heap words were of type scm_t_bits and local
1010 variables and function arguments were of type SCM, making it
1011 non-standards-conformant to have a pointer that can point to both.
1013 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1015 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1016 second and third words of double smobs. Likewise for
1017 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1019 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1020 used to get and set the 16 exra bits in the zeroth word of a smob.
1022 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1023 accesing the first immediate word of a smob as a SCM value, and there
1024 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1025 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1027 ** New way to deal with non-local exits and re-entries.
1029 There is a new set of functions that essentially do what
1030 scm_internal_dynamic_wind does, but in a way that is more convenient
1031 for C code in some situations. Here is a quick example of how to
1032 prevent a potential memory leak:
1039 scm_dynwind_begin (0);
1041 mem = scm_malloc (100);
1042 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1044 /* MEM would leak if BAR throws an error.
1045 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1052 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1053 SCM_DYNWIND_END as well.
1057 For full documentation, see the node "Dynamic Wind" in the manual.
1059 ** New function scm_dynwind_free
1061 This function calls 'free' on a given pointer when a dynwind context
1062 is left. Thus the call to scm_dynwind_unwind_handler above could be
1063 replaced with simply scm_dynwind_free (mem).
1065 ** New functions scm_c_call_with_blocked_asyncs and
1066 scm_c_call_with_unblocked_asyncs
1068 Like scm_call_with_blocked_asyncs etc. but for C functions.
1070 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1072 In addition to scm_c_call_with_blocked_asyncs you can now also use
1073 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1074 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1076 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1077 SCM_REALLOW_INTS have been deprecated.
1079 They do no longer fulfill their original role of blocking signal
1080 delivery. Depending on what you want to achieve, replace a pair of
1081 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1082 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1085 ** The value 'scm_mask_ints' is no longer writable.
1087 Previously, you could set scm_mask_ints directly. This is no longer
1088 possible. Use scm_c_call_with_blocked_asyncs and
1089 scm_c_call_with_unblocked_asyncs instead.
1091 ** New way to temporarily set the current input, output or error ports
1093 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1094 context' (see above). <foo> is one of "input", "output" or "error".
1096 ** New way to temporarily set fluids
1098 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1099 above) to temporarily set the value of a fluid.
1101 ** New types scm_t_intmax and scm_t_uintmax.
1103 On platforms that have them, these types are identical to intmax_t and
1104 uintmax_t, respectively. On other platforms, they are identical to
1105 the largest integer types that Guile knows about.
1107 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1109 You should not have used them.
1111 ** Many public #defines with generic names have been made private.
1113 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1114 private or renamed with a more suitable public name.
1116 ** The macro SCM_TYP16S has been deprecated.
1118 This macro is not intended for public use.
1120 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1122 Use scm_is_true (scm_inexact_p (...)) instead.
1124 ** The macro SCM_SLOPPY_REALP has been deprecated.
1126 Use scm_is_real instead.
1128 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1130 Use scm_is_complex instead.
1132 ** Some preprocessor defines have been deprecated.
1134 These defines indicated whether a certain feature was present in Guile
1135 or not. Going forward, assume that the features are always present.
1137 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1138 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1140 The following macros have been removed completely: MEMOIZE_LOCALS,
1141 SCM_RECKLESS, SCM_CAUTIOUS.
1143 ** The preprocessor define STACK_DIRECTION has been deprecated.
1145 There should be no need to know about the stack direction for ordinary
1148 ** New function: scm_effective_version
1150 Returns the "effective" version number. This is just the normal full
1151 version string without the final micro-version number. See "Changes
1152 to the distribution" above.
1154 ** The function scm_call_with_new_thread has a new prototype.
1156 Instead of taking a list with the thunk and handler, these two
1157 arguments are now passed directly:
1159 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1161 This is an incompatible change.
1163 ** New snarfer macro SCM_DEFINE_PUBLIC.
1165 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1166 function in the init section.
1168 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1170 ** Garbage collector rewrite.
1172 The garbage collector is cleaned up a lot, and now uses lazy
1173 sweeping. This is reflected in the output of (gc-stats); since cells
1174 are being freed when they are allocated, the cells-allocated field
1175 stays roughly constant.
1177 For malloc related triggers, the behavior is changed. It uses the same
1178 heuristic as the cell-triggered collections. It may be tuned with the
1179 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1180 for minimum yield of malloc related triggers. The default is 40.
1181 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1184 Debugging operations for the freelist have been deprecated, along with
1185 the C variables that control garbage collection. The environment
1186 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1187 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1189 For understanding the memory usage of a GUILE program, the routine
1190 gc-live-object-stats returns an alist containing the number of live
1191 objects for every type.
1194 ** The function scm_definedp has been renamed to scm_defined_p
1196 The name scm_definedp is deprecated.
1198 ** The struct scm_cell type has been renamed to scm_t_cell
1200 This is in accordance to Guile's naming scheme for types. Note that
1201 the name scm_cell is now used for a function that allocates and
1202 initializes a new cell (see below).
1204 ** New functions for memory management
1206 A new set of functions for memory management has been added since the
1207 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1208 indeed, Guile itself contained some long standing bugs that could
1209 cause aborts in long running programs.
1211 The new functions are more symmetrical and do not need cooperation
1212 from smob free routines, among other improvements.
1214 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1215 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1216 scm_gc_free, scm_gc_register_collectable_memory, and
1217 scm_gc_unregister_collectable_memory. Refer to the manual for more
1218 details and for upgrading instructions.
1220 The old functions for memory management have been deprecated. They
1221 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1222 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1224 ** Declarations of exported features are marked with SCM_API.
1226 Every declaration of a feature that belongs to the exported Guile API
1227 has been marked by adding the macro "SCM_API" to the start of the
1228 declaration. This macro can expand into different things, the most
1229 common of which is just "extern" for Unix platforms. On Win32, it can
1230 be used to control which symbols are exported from a DLL.
1232 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1233 will expand into "__declspec (dllimport) extern", which is needed for
1234 linking to the Guile DLL in Windows.
1236 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1237 SCM_SRFI4_IMPORT, for the corresponding libraries.
1239 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1241 Use the new functions scm_cell and scm_double_cell instead. The old
1242 macros had problems because with them allocation and initialization
1243 was separated and the GC could sometimes observe half initialized
1244 cells. Only careful coding by the user of SCM_NEWCELL and
1245 SCM_NEWCELL2 could make this safe and efficient.
1247 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1249 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1252 ** SRCBRKP has been deprecated.
1254 Use scm_c_source_property_breakpoint_p instead.
1256 ** Deprecated: scm_makmacro
1258 Change your code to use either scm_makmmacro or to define macros in
1259 Scheme, using 'define-macro'.
1261 ** New function scm_c_port_for_each.
1263 This function is like scm_port_for_each but takes a pointer to a C
1264 function as the callback instead of a SCM value.
1266 ** The names scm_internal_select, scm_thread_sleep, and
1267 scm_thread_usleep have been discouraged.
1269 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1271 ** The GC can no longer be blocked.
1273 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1274 The GC can now run (partially) concurrently with other code and thus
1275 blocking it is not well defined.
1277 ** Many definitions have been removed that were previously deprecated.
1279 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1280 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1281 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1282 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1283 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1284 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1285 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1286 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1287 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1288 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1289 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1290 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1291 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1292 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1293 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1294 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1295 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1296 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1297 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1298 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1299 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1300 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1301 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1302 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1303 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1304 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1305 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1306 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1307 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1308 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1309 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1311 * Changes to bundled modules
1315 Using the (ice-9 debug) module no longer automatically switches Guile
1316 to use the debugging evaluator. If you want to switch to the
1317 debugging evaluator (which is needed for backtrace information if you
1318 hit an error), please add an explicit "(debug-enable 'debug)" to your
1319 code just after the code to use (ice-9 debug).
1322 Changes since Guile 1.4:
1324 * Changes to the distribution
1326 ** A top-level TODO file is included.
1328 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1330 Guile now always uses three numbers to represent the version,
1331 i.e. "1.6.5". The first number, 1, is the major version number, the
1332 second number, 6, is the minor version number, and the third number,
1333 5, is the micro version number. Changes in major version number
1334 indicate major changes in Guile.
1336 Minor version numbers that are even denote stable releases, and odd
1337 minor version numbers denote development versions (which may be
1338 unstable). The micro version number indicates a minor sub-revision of
1339 a given MAJOR.MINOR release.
1341 In keeping with the new scheme, (minor-version) and scm_minor_version
1342 no longer return everything but the major version number. They now
1343 just return the minor version number. Two new functions
1344 (micro-version) and scm_micro_version have been added to report the
1345 micro version number.
1347 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1349 ** New preprocessor definitions are available for checking versions.
1351 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1352 SCM_MICRO_VERSION to the appropriate integer values.
1354 ** Guile now actively warns about deprecated features.
1356 The new configure option `--enable-deprecated=LEVEL' and the
1357 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1358 See INSTALL and README for more information.
1360 ** Guile is much more likely to work on 64-bit architectures.
1362 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1363 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1364 for the use of a test machine, and thanks to Stefan Jahn for ia64
1367 ** New functions: setitimer and getitimer.
1369 These implement a fairly direct interface to the libc functions of the
1372 ** The #. reader extension is now disabled by default.
1374 For safety reasons, #. evaluation is disabled by default. To
1375 re-enable it, set the fluid read-eval? to #t. For example:
1377 (fluid-set! read-eval? #t)
1379 but make sure you realize the potential security risks involved. With
1380 read-eval? enabled, reading a data file from an untrusted source can
1383 ** New SRFI modules have been added:
1385 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1388 (srfi srfi-1) is a library containing many useful pair- and list-processing
1391 (srfi srfi-2) exports and-let*.
1393 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1395 (srfi srfi-6) is a dummy module for now, since guile already provides
1396 all of the srfi-6 procedures by default: open-input-string,
1397 open-output-string, get-output-string.
1399 (srfi srfi-8) exports receive.
1401 (srfi srfi-9) exports define-record-type.
1403 (srfi srfi-10) exports define-reader-ctor and implements the reader
1406 (srfi srfi-11) exports let-values and let*-values.
1408 (srfi srfi-13) implements the SRFI String Library.
1410 (srfi srfi-14) implements the SRFI Character-Set Library.
1412 (srfi srfi-17) implements setter and getter-with-setter and redefines
1413 some accessor procedures as procedures with getters. (such as car,
1414 cdr, vector-ref etc.)
1416 (srfi srfi-19) implements the SRFI Time/Date Library.
1418 ** New scripts / "executable modules"
1420 Subdirectory "scripts" contains Scheme modules that are packaged to
1421 also be executable as scripts. At this time, these scripts are available:
1430 See README there for more info.
1432 These scripts can be invoked from the shell with the new program
1433 "guile-tools", which keeps track of installation directory for you.
1436 $ guile-tools display-commentary srfi/*.scm
1438 guile-tools is copied to the standard $bindir on "make install".
1440 ** New module (ice-9 stack-catch):
1442 stack-catch is like catch, but saves the current state of the stack in
1443 the fluid the-last-stack. This fluid can be useful when using the
1444 debugger and when re-throwing an error.
1446 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1448 This has been done to prevent problems on lesser operating systems
1449 that can't tolerate `*'s in file names. The exported macro continues
1450 to be named `and-let*', of course.
1452 On systems that support it, there is also a compatibility module named
1453 (ice-9 and-let*). It will go away in the next release.
1455 ** New modules (oop goops) etc.:
1458 (oop goops describe)
1460 (oop goops active-slot)
1461 (oop goops composite-slot)
1463 The Guile Object Oriented Programming System (GOOPS) has been
1464 integrated into Guile. For further information, consult the GOOPS
1465 manual and tutorial in the `doc' directory.
1467 ** New module (ice-9 rdelim).
1469 This exports the following procedures which were previously defined
1470 in the default environment:
1472 read-line read-line! read-delimited read-delimited! %read-delimited!
1473 %read-line write-line
1475 For backwards compatibility the definitions are still imported into the
1476 default environment in this version of Guile. However you should add:
1478 (use-modules (ice-9 rdelim))
1480 to any program which uses the definitions, since this may change in
1483 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1484 can be used for similar functionality.
1486 ** New module (ice-9 rw)
1488 This is a subset of the (scsh rw) module from guile-scsh. Currently
1489 it defines two procedures:
1491 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1493 Read characters from a port or file descriptor into a string STR.
1494 A port must have an underlying file descriptor -- a so-called
1495 fport. This procedure is scsh-compatible and can efficiently read
1498 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1500 Write characters from a string STR to a port or file descriptor.
1501 A port must have an underlying file descriptor -- a so-called
1502 fport. This procedure is mostly compatible and can efficiently
1503 write large strings.
1505 ** New module (ice-9 match)
1507 This module includes Andrew K. Wright's pattern matcher. See
1508 ice-9/match.scm for brief description or
1510 http://www.star-lab.com/wright/code.html
1512 for complete documentation.
1514 ** New module (ice-9 buffered-input)
1516 This module provides procedures to construct an input port from an
1517 underlying source of input that reads and returns its input in chunks.
1518 The underlying input source is a Scheme procedure, specified by the
1519 caller, which the port invokes whenever it needs more input.
1521 This is useful when building an input port whose back end is Readline
1522 or a UI element such as the GtkEntry widget.
1526 The reference and tutorial documentation that was previously
1527 distributed separately, as `guile-doc', is now included in the core
1528 Guile distribution. The documentation consists of the following
1531 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1534 - The Guile Reference Manual (guile.texi) contains (or is intended to
1535 contain) reference documentation on all aspects of Guile.
1537 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1538 reference documentation for using GOOPS, Guile's Object Oriented
1541 - The Revised^5 Report on the Algorithmic Language Scheme
1544 See the README file in the `doc' directory for more details.
1546 ** There are a couple of examples in the examples/ directory now.
1548 * Changes to the stand-alone interpreter
1550 ** New command line option `--use-srfi'
1552 Using this option, SRFI modules can be loaded on startup and be
1553 available right from the beginning. This makes programming portable
1554 Scheme programs easier.
1556 The option `--use-srfi' expects a comma-separated list of numbers,
1557 each representing a SRFI number to be loaded into the interpreter
1558 before starting evaluating a script file or the REPL. Additionally,
1559 the feature identifier for the loaded SRFIs is recognized by
1560 `cond-expand' when using this option.
1563 $ guile --use-srfi=8,13
1564 guile> (receive (x z) (values 1 2) (+ 1 2))
1566 guile> (string-pad "bla" 20)
1569 ** Guile now always starts up in the `(guile-user)' module.
1571 Previously, scripts executed via the `-s' option would run in the
1572 `(guile)' module and the repl would run in the `(guile-user)' module.
1573 Now every user action takes place in the `(guile-user)' module by
1576 * Changes to Scheme functions and syntax
1578 ** Character classifiers work for non-ASCII characters.
1580 The predicates `char-alphabetic?', `char-numeric?',
1581 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1582 no longer check whether their arguments are ASCII characters.
1583 Previously, a character would only be considered alphabetic when it
1584 was also ASCII, for example.
1586 ** Previously deprecated Scheme functions have been removed:
1588 tag - no replacement.
1589 fseek - replaced by seek.
1590 list* - replaced by cons*.
1592 ** It's now possible to create modules with controlled environments
1596 (use-modules (ice-9 safe))
1597 (define m (make-safe-module))
1598 ;;; m will now be a module containing only a safe subset of R5RS
1599 (eval '(+ 1 2) m) --> 3
1600 (eval 'load m) --> ERROR: Unbound variable: load
1602 ** Evaluation of "()", the empty list, is now an error.
1604 Previously, the expression "()" evaluated to the empty list. This has
1605 been changed to signal a "missing expression" error. The correct way
1606 to write the empty list as a literal constant is to use quote: "'()".
1608 ** New concept of `Guile Extensions'.
1610 A Guile Extension is just a ordinary shared library that can be linked
1611 at run-time. We found it advantageous to give this simple concept a
1612 dedicated name to distinguish the issues related to shared libraries
1613 from the issues related to the module system.
1615 *** New function: load-extension
1617 Executing (load-extension lib init) is mostly equivalent to
1619 (dynamic-call init (dynamic-link lib))
1621 except when scm_register_extension has been called previously.
1622 Whenever appropriate, you should use `load-extension' instead of
1623 dynamic-link and dynamic-call.
1625 *** New C function: scm_c_register_extension
1627 This function registers a initialization function for use by
1628 `load-extension'. Use it when you don't want specific extensions to
1629 be loaded as shared libraries (for example on platforms that don't
1630 support dynamic linking).
1632 ** Auto-loading of compiled-code modules is deprecated.
1634 Guile used to be able to automatically find and link a shared
1635 library to satisfy requests for a module. For example, the module
1636 `(foo bar)' could be implemented by placing a shared library named
1637 "foo/libbar.so" (or with a different extension) in a directory on the
1640 This has been found to be too tricky, and is no longer supported. The
1641 shared libraries are now called "extensions". You should now write a
1642 small Scheme file that calls `load-extension' to load the shared
1643 library and initialize it explicitly.
1645 The shared libraries themselves should be installed in the usual
1646 places for shared libraries, with names like "libguile-foo-bar".
1648 For example, place this into a file "foo/bar.scm"
1650 (define-module (foo bar))
1652 (load-extension "libguile-foo-bar" "foobar_init")
1654 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1656 `eval' is now R5RS, that is it takes two arguments.
1657 The second argument is an environment specifier, i.e. either
1659 (scheme-report-environment 5)
1660 (null-environment 5)
1661 (interaction-environment)
1667 ** The module system has been made more disciplined.
1669 The function `eval' will save and restore the current module around
1670 the evaluation of the specified expression. While this expression is
1671 evaluated, `(current-module)' will now return the right module, which
1672 is the module specified as the second argument to `eval'.
1674 A consequence of this change is that `eval' is not particularly
1675 useful when you want allow the evaluated code to change what module is
1676 designated as the current module and have this change persist from one
1677 call to `eval' to the next. The read-eval-print-loop is an example
1678 where `eval' is now inadequate. To compensate, there is a new
1679 function `primitive-eval' that does not take a module specifier and
1680 that does not save/restore the current module. You should use this
1681 function together with `set-current-module', `current-module', etc
1682 when you want to have more control over the state that is carried from
1683 one eval to the next.
1685 Additionally, it has been made sure that forms that are evaluated at
1686 the top level are always evaluated with respect to the current module.
1687 Previously, subforms of top-level forms such as `begin', `case',
1688 etc. did not respect changes to the current module although these
1689 subforms are at the top-level as well.
1691 To prevent strange behavior, the forms `define-module',
1692 `use-modules', `use-syntax', and `export' have been restricted to only
1693 work on the top level. The forms `define-public' and
1694 `defmacro-public' only export the new binding on the top level. They
1695 behave just like `define' and `defmacro', respectively, when they are
1696 used in a lexical environment.
1698 Also, `export' will no longer silently re-export bindings imported
1699 from a used module. It will emit a `deprecation' warning and will
1700 cease to perform any re-export in the next version. If you actually
1701 want to re-export bindings, use the new `re-export' in place of
1702 `export'. The new `re-export' will not make copies of variables when
1703 rexporting them, as `export' did wrongly.
1705 ** Module system now allows selection and renaming of imported bindings
1707 Previously, when using `use-modules' or the `#:use-module' clause in
1708 the `define-module' form, all the bindings (association of symbols to
1709 values) for imported modules were added to the "current module" on an
1710 as-is basis. This has been changed to allow finer control through two
1711 new facilities: selection and renaming.
1713 You can now select which of the imported module's bindings are to be
1714 visible in the current module by using the `:select' clause. This
1715 clause also can be used to rename individual bindings. For example:
1717 ;; import all bindings no questions asked
1718 (use-modules (ice-9 common-list))
1720 ;; import four bindings, renaming two of them;
1721 ;; the current module sees: every some zonk-y zonk-n
1722 (use-modules ((ice-9 common-list)
1724 (remove-if . zonk-y)
1725 (remove-if-not . zonk-n))))
1727 You can also programmatically rename all selected bindings using the
1728 `:renamer' clause, which specifies a proc that takes a symbol and
1729 returns another symbol. Because it is common practice to use a prefix,
1730 we now provide the convenience procedure `symbol-prefix-proc'. For
1733 ;; import four bindings, renaming two of them specifically,
1734 ;; and all four w/ prefix "CL:";
1735 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1736 (use-modules ((ice-9 common-list)
1738 (remove-if . zonk-y)
1739 (remove-if-not . zonk-n))
1740 :renamer (symbol-prefix-proc 'CL:)))
1742 ;; import four bindings, renaming two of them specifically,
1743 ;; and all four by upcasing.
1744 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1745 (define (upcase-symbol sym)
1746 (string->symbol (string-upcase (symbol->string sym))))
1748 (use-modules ((ice-9 common-list)
1750 (remove-if . zonk-y)
1751 (remove-if-not . zonk-n))
1752 :renamer upcase-symbol))
1754 Note that programmatic renaming is done *after* individual renaming.
1755 Also, the above examples show `use-modules', but the same facilities are
1756 available for the `#:use-module' clause of `define-module'.
1758 See manual for more info.
1760 ** The semantics of guardians have changed.
1762 The changes are for the most part compatible. An important criterion
1763 was to keep the typical usage of guardians as simple as before, but to
1764 make the semantics safer and (as a result) more useful.
1766 *** All objects returned from guardians are now properly alive.
1768 It is now guaranteed that any object referenced by an object returned
1769 from a guardian is alive. It's now impossible for a guardian to
1770 return a "contained" object before its "containing" object.
1772 One incompatible (but probably not very important) change resulting
1773 from this is that it is no longer possible to guard objects that
1774 indirectly reference themselves (i.e. are parts of cycles). If you do
1775 so accidentally, you'll get a warning.
1777 *** There are now two types of guardians: greedy and sharing.
1779 If you call (make-guardian #t) or just (make-guardian), you'll get a
1780 greedy guardian, and for (make-guardian #f) a sharing guardian.
1782 Greedy guardians are the default because they are more "defensive".
1783 You can only greedily guard an object once. If you guard an object
1784 more than once, once in a greedy guardian and the rest of times in
1785 sharing guardians, then it is guaranteed that the object won't be
1786 returned from sharing guardians as long as it is greedily guarded
1789 Guardians returned by calls to `make-guardian' can now take one more
1790 optional parameter, which says whether to throw an error in case an
1791 attempt is made to greedily guard an object that is already greedily
1792 guarded. The default is true, i.e. throw an error. If the parameter
1793 is false, the guardian invocation returns #t if guarding was
1794 successful and #f if it wasn't.
1796 Also, since greedy guarding is, in effect, a side-effecting operation
1797 on objects, a new function is introduced: `destroy-guardian!'.
1798 Invoking this function on a guardian renders it unoperative and, if
1799 the guardian is greedy, clears the "greedily guarded" property of the
1800 objects that were guarded by it, thus undoing the side effect.
1802 Note that all this hair is hardly very important, since guardian
1803 objects are usually permanent.
1805 ** Continuations created by call-with-current-continuation now accept
1806 any number of arguments, as required by R5RS.
1808 ** New function `issue-deprecation-warning'
1810 This function is used to display the deprecation messages that are
1811 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1814 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1818 ;; `id' is deprecated. Use `identity' instead.
1823 ** New syntax `begin-deprecated'
1825 When deprecated features are included (as determined by the configure
1826 option --enable-deprecated), `begin-deprecated' is identical to
1827 `begin'. When deprecated features are excluded, it always evaluates
1828 to `#f', ignoring the body forms.
1830 ** New function `make-object-property'
1832 This function returns a new `procedure with setter' P that can be used
1833 to attach a property to objects. When calling P as
1837 where `obj' is any kind of object, it attaches `val' to `obj' in such
1838 a way that it can be retrieved by calling P as
1842 This function will replace procedure properties, symbol properties and
1843 source properties eventually.
1845 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1847 Instead of #&optional, #&key, etc you should now use #:optional,
1848 #:key, etc. Since #:optional is a keyword, you can write it as just
1849 :optional when (read-set! keywords 'prefix) is active.
1851 The old reader syntax `#&' is still supported, but deprecated. It
1852 will be removed in the next release.
1854 ** New define-module option: pure
1856 Tells the module system not to include any bindings from the root
1861 (define-module (totally-empty-module)
1864 ** New define-module option: export NAME1 ...
1866 Export names NAME1 ...
1868 This option is required if you want to be able to export bindings from
1869 a module which doesn't import one of `define-public' or `export'.
1873 (define-module (foo)
1875 :use-module (ice-9 r5rs)
1878 ;;; Note that we're pure R5RS below this point!
1883 ** New function: object->string OBJ
1885 Return a Scheme string obtained by printing a given object.
1887 ** New function: port? X
1889 Returns a boolean indicating whether X is a port. Equivalent to
1890 `(or (input-port? X) (output-port? X))'.
1892 ** New function: file-port?
1894 Determines whether a given object is a port that is related to a file.
1896 ** New function: port-for-each proc
1898 Apply PROC to each port in the Guile port table in turn. The return
1899 value is unspecified. More specifically, PROC is applied exactly once
1900 to every port that exists in the system at the time PORT-FOR-EACH is
1901 invoked. Changes to the port table while PORT-FOR-EACH is running
1902 have no effect as far as PORT-FOR-EACH is concerned.
1904 ** New function: dup2 oldfd newfd
1906 A simple wrapper for the `dup2' system call. Copies the file
1907 descriptor OLDFD to descriptor number NEWFD, replacing the
1908 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1909 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1910 to move away ports which are using NEWFD. The return value is
1913 ** New function: close-fdes fd
1915 A simple wrapper for the `close' system call. Close file
1916 descriptor FD, which must be an integer. Unlike close (*note
1917 close: Ports and File Descriptors.), the file descriptor will be
1918 closed even if a port is using it. The return value is
1921 ** New function: crypt password salt
1923 Encrypts `password' using the standard unix password encryption
1926 ** New function: chroot path
1928 Change the root directory of the running process to `path'.
1930 ** New functions: getlogin, cuserid
1932 Return the login name or the user name of the current effective user
1935 ** New functions: getpriority which who, setpriority which who prio
1937 Get or set the priority of the running process.
1939 ** New function: getpass prompt
1941 Read a password from the terminal, first displaying `prompt' and
1944 ** New function: flock file operation
1946 Set/remove an advisory shared or exclusive lock on `file'.
1948 ** New functions: sethostname name, gethostname
1950 Set or get the hostname of the machine the current process is running
1953 ** New function: mkstemp! tmpl
1955 mkstemp creates a new unique file in the file system and returns a
1956 new buffered port open for reading and writing to the file. TMPL
1957 is a string specifying where the file should be created: it must
1958 end with `XXXXXX' and will be changed in place to return the name
1959 of the temporary file.
1961 ** New function: open-input-string string
1963 Return an input string port which delivers the characters from
1964 `string'. This procedure, together with `open-output-string' and
1965 `get-output-string' implements SRFI-6.
1967 ** New function: open-output-string
1969 Return an output string port which collects all data written to it.
1970 The data can then be retrieved by `get-output-string'.
1972 ** New function: get-output-string
1974 Return the contents of an output string port.
1976 ** New function: identity
1978 Return the argument.
1980 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1981 are represented in Scheme as integers with normal host byte ordering.
1983 ** New function: inet-pton family address
1985 Convert a printable string network address into an integer. Note that
1986 unlike the C version of this function, the result is an integer with
1987 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1990 (inet-pton AF_INET "127.0.0.1") => 2130706433
1991 (inet-pton AF_INET6 "::1") => 1
1993 ** New function: inet-ntop family address
1995 Convert an integer network address into a printable string. Note that
1996 unlike the C version of this function, the input is an integer with
1997 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2000 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2001 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2002 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2006 Use `identity' instead.
2012 ** Deprecated: return-it
2016 ** Deprecated: string-character-length
2018 Use `string-length' instead.
2020 ** Deprecated: flags
2022 Use `logior' instead.
2024 ** Deprecated: close-all-ports-except.
2026 This was intended for closing ports in a child process after a fork,
2027 but it has the undesirable side effect of flushing buffers.
2028 port-for-each is more flexible.
2030 ** The (ice-9 popen) module now attempts to set up file descriptors in
2031 the child process from the current Scheme ports, instead of using the
2032 current values of file descriptors 0, 1, and 2 in the parent process.
2034 ** Removed function: builtin-weak-bindings
2036 There is no such concept as a weak binding any more.
2038 ** Removed constants: bignum-radix, scm-line-incrementors
2040 ** define-method: New syntax mandatory.
2042 The new method syntax is now mandatory:
2044 (define-method (NAME ARG-SPEC ...) BODY ...)
2045 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2047 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2048 REST-ARG ::= ARG-NAME
2050 If you have old code using the old syntax, import
2051 (oop goops old-define-method) before (oop goops) as in:
2053 (use-modules (oop goops old-define-method) (oop goops))
2055 ** Deprecated function: builtin-variable
2056 Removed function: builtin-bindings
2058 There is no longer a distinction between builtin or other variables.
2059 Use module system operations for all variables.
2061 ** Lazy-catch handlers are no longer allowed to return.
2063 That is, a call to `throw', `error', etc is now guaranteed to not
2066 ** Bugfixes for (ice-9 getopt-long)
2068 This module is now tested using test-suite/tests/getopt-long.test.
2069 The following bugs have been fixed:
2071 *** Parsing for options that are specified to have `optional' args now checks
2072 if the next element is an option instead of unconditionally taking it as the
2075 *** An error is now thrown for `--opt=val' when the option description
2076 does not specify `(value #t)' or `(value optional)'. This condition used to
2077 be accepted w/o error, contrary to the documentation.
2079 *** The error message for unrecognized options is now more informative.
2080 It used to be "not a record", an artifact of the implementation.
2082 *** The error message for `--opt' terminating the arg list (no value), when
2083 `(value #t)' is specified, is now more informative. It used to be "not enough
2086 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2087 The expansion used to be like so:
2089 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2091 Note that the "5d" is dropped. Now it is like so:
2093 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2095 This enables single-char options to have adjoining arguments as long as their
2096 constituent characters are not potential single-char options.
2098 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2100 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2101 property `arglist', which can be retrieved by `arity'. The result is that
2102 `arity' can give more detailed information than before:
2106 guile> (use-modules (ice-9 optargs))
2107 guile> (define* (foo #:optional a b c) a)
2109 0 or more arguments in `lambda*:G0'.
2114 3 optional arguments: `a', `b' and `c'.
2115 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2117 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2118 and `d', other keywords allowed.
2119 guile> (define* (baz a b #:optional c #:rest r) a)
2121 2 required arguments: `a' and `b', 1 optional argument: `c',
2124 * Changes to the C interface
2126 ** Types have been renamed from scm_*_t to scm_t_*.
2128 This has been done for POSIX sake. It reserves identifiers ending
2129 with "_t". What a concept.
2131 The old names are still available with status `deprecated'.
2133 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2135 ** Deprecated features have been removed.
2139 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2140 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2142 *** C Functions removed
2144 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2145 scm_fseek - replaced by scm_seek.
2146 gc-thunk - replaced by after-gc-hook.
2147 gh_int2scmb - replaced by gh_bool2scm.
2148 scm_tc_dblr - replaced by scm_tc16_real.
2149 scm_tc_dblc - replaced by scm_tc16_complex.
2150 scm_list_star - replaced by scm_cons_star.
2152 ** Deprecated: scm_makfromstr
2154 Use scm_mem2string instead.
2156 ** Deprecated: scm_make_shared_substring
2158 Explicit shared substrings will disappear from Guile.
2160 Instead, "normal" strings will be implemented using sharing
2161 internally, combined with a copy-on-write strategy.
2163 ** Deprecated: scm_read_only_string_p
2165 The concept of read-only strings will disappear in next release of
2168 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2170 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2172 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2174 Call a procedure with the indicated number of arguments. See "Fly
2175 Evaluation" in the manual.
2177 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2179 Call a procedure with the indicated number of arguments and a list of
2180 further arguments. See "Fly Evaluation" in the manual.
2182 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2184 Create a list of the given number of elements. See "List
2185 Constructors" in the manual.
2187 ** Renamed function: scm_listify has been replaced by scm_list_n.
2189 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2190 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2192 Use functions scm_list_N instead.
2194 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2196 Used by an application to read arbitrary number of bytes from a port.
2197 Same semantics as libc read, except that scm_c_read only returns less
2198 than SIZE bytes if at end-of-file.
2200 Warning: Doesn't update port line and column counts!
2202 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2204 Used by an application to write arbitrary number of bytes to an SCM
2205 port. Similar semantics as libc write. However, unlike libc
2206 write, scm_c_write writes the requested number of bytes and has no
2209 Warning: Doesn't update port line and column counts!
2211 ** New function: scm_init_guile ()
2213 In contrast to scm_boot_guile, scm_init_guile will return normally
2214 after initializing Guile. It is not available on all systems, tho.
2216 ** New functions: scm_str2symbol, scm_mem2symbol
2218 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2219 field of characters and creates a scheme symbol object from that C string.
2220 The function scm_mem2symbol takes a const char* and a number of characters and
2221 creates a symbol from the characters in that memory area.
2223 ** New functions: scm_primitive_make_property
2224 scm_primitive_property_ref
2225 scm_primitive_property_set_x
2226 scm_primitive_property_del_x
2228 These functions implement a new way to deal with object properties.
2229 See libguile/properties.c for their documentation.
2231 ** New function: scm_done_free (long size)
2233 This function is the inverse of scm_done_malloc. Use it to report the
2234 amount of smob memory you free. The previous method, which involved
2235 calling scm_done_malloc with negative argument, was somewhat
2236 unintuitive (and is still available, of course).
2238 ** New function: scm_c_memq (SCM obj, SCM list)
2240 This function provides a fast C level alternative for scm_memq for the case
2241 that the list parameter is known to be a proper list. The function is a
2242 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2243 list input parameter, since for anything else but a proper list the function's
2244 behaviour is undefined - it may even crash or loop endlessly. Further, for
2245 the case that the object is not found in the list, scm_c_memq returns #f which
2246 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2248 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2249 scm_remember_upto_here
2251 These functions replace the function scm_remember.
2253 ** Deprecated function: scm_remember
2255 Use one of the new functions scm_remember_upto_here_1,
2256 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2258 ** New function: scm_allocate_string
2260 This function replaces the function scm_makstr.
2262 ** Deprecated function: scm_makstr
2264 Use the new function scm_allocate_string instead.
2266 ** New global variable scm_gc_running_p introduced.
2268 Use this variable to find out if garbage collection is being executed. Up to
2269 now applications have used scm_gc_heap_lock to test if garbage collection was
2270 running, which also works because of the fact that up to know only the garbage
2271 collector has set this variable. But, this is an implementation detail that
2272 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2273 of this variable is (and has been) not fully safe anyway.
2275 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2277 Use these instead of SCM_LENGTH_MAX.
2279 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2280 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2281 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2283 Use these instead of SCM_LENGTH.
2285 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2286 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2287 SCM_SET_BITVECTOR_LENGTH
2289 Use these instead of SCM_SETLENGTH
2291 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2292 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2295 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2298 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2299 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2302 Use these instead of SCM_SETCHARS.
2304 ** New macro: SCM_BITVECTOR_P
2306 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2308 Use instead of SCM_COERCE_SUBSTR.
2310 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2312 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2314 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2315 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2316 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2317 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2318 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2319 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2320 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2321 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2322 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2323 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2324 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2325 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2326 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2327 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2328 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2330 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2331 Use scm_memory_error instead of SCM_NALLOC.
2332 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2333 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2334 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2335 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2336 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2337 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2338 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2339 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2340 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2341 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2342 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2343 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2344 Use a type specific setter macro instead of SCM_SETLENGTH.
2345 Use a type specific setter macro instead of SCM_SETCHARS.
2346 Use a type specific length macro instead of SCM_LENGTH_MAX.
2347 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2348 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2349 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2350 Use SCM_TYP16 instead of SCM_GCTYP16.
2351 Use SCM_CDR instead of SCM_GCCDR.
2352 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2353 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2354 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2355 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2356 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2357 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2358 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2360 ** Removed function: scm_struct_init
2362 ** Removed variable: scm_symhash_dim
2364 ** Renamed function: scm_make_cont has been replaced by
2365 scm_make_continuation, which has a different interface.
2367 ** Deprecated function: scm_call_catching_errors
2369 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2371 ** Deprecated function: scm_strhash
2373 Use scm_string_hash instead.
2375 ** Deprecated function: scm_vector_set_length_x
2377 Instead, create a fresh vector of the desired size and copy the contents.
2379 ** scm_gensym has changed prototype
2381 scm_gensym now only takes one argument.
2383 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2386 There is now only a single symbol type scm_tc7_symbol.
2387 The tag scm_tc7_lvector was not used anyway.
2389 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2391 Use scm_make_smob_type and scm_set_smob_XXX instead.
2393 ** New function scm_set_smob_apply.
2395 This can be used to set an apply function to a smob type.
2397 ** Deprecated function: scm_strprint_obj
2399 Use scm_object_to_string instead.
2401 ** Deprecated function: scm_wta
2403 Use scm_wrong_type_arg, or another appropriate error signalling function
2406 ** Explicit support for obarrays has been deprecated.
2408 Use `scm_str2symbol' and the generic hashtable functions instead.
2410 ** The concept of `vcells' has been deprecated.
2412 The data type `variable' is now used exclusively. `Vcells' have been
2413 a low-level concept so you are likely not affected by this change.
2415 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2416 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2418 Use scm_c_define or scm_c_lookup instead, as appropriate.
2420 *** New functions: scm_c_module_lookup, scm_c_lookup,
2421 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2422 scm_module_define, scm_define.
2424 These functions work with variables instead of with vcells.
2426 ** New functions for creating and defining `subr's and `gsubr's.
2428 The new functions more clearly distinguish between creating a subr (or
2429 gsubr) object and adding it to the current module.
2431 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2432 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2433 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2434 scm_c_define_gsubr_with_generic.
2436 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2437 scm_make_subr_with_generic, scm_make_gsubr,
2438 scm_make_gsubr_with_generic.
2440 Use the new ones from above instead.
2442 ** C interface to the module system has changed.
2444 While we suggest that you avoid as many explicit module system
2445 operations from C as possible for the time being, the C interface has
2446 been made more similar to the high-level Scheme module system.
2448 *** New functions: scm_c_define_module, scm_c_use_module,
2449 scm_c_export, scm_c_resolve_module.
2451 They mostly work like their Scheme namesakes. scm_c_define_module
2452 takes a function that is called a context where the new module is
2455 *** Deprecated functions: scm_the_root_module, scm_make_module,
2456 scm_ensure_user_module, scm_load_scheme_module.
2458 Use the new functions instead.
2460 ** Renamed function: scm_internal_with_fluids becomes
2463 scm_internal_with_fluids is available as a deprecated function.
2465 ** New function: scm_c_with_fluid.
2467 Just like scm_c_with_fluids, but takes one fluid and one value instead
2470 ** Deprecated typedefs: long_long, ulong_long.
2472 They are of questionable utility and they pollute the global
2475 ** Deprecated typedef: scm_sizet
2477 It is of questionable utility now that Guile requires ANSI C, and is
2480 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2481 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2482 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2484 Made more compliant with the naming policy by adding a _t at the end.
2486 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2487 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2489 With the exception of the mysterious scm_2ulong2big, they are still
2490 available under new names (scm_i_mkbig etc). These functions are not
2491 intended to be used in user code. You should avoid dealing with
2492 bignums directly, and should deal with numbers in general (which can
2495 ** Change in behavior: scm_num2long, scm_num2ulong
2497 The scm_num2[u]long functions don't any longer accept an inexact
2498 argument. This change in behavior is motivated by concordance with
2499 R5RS: It is more common that a primitive doesn't want to accept an
2500 inexact for an exact.
2502 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2503 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2504 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2507 These are conversion functions between the various ANSI C integral
2508 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2509 accept an inexact argument.
2511 ** New functions: scm_float2num, scm_double2num,
2512 scm_num2float, scm_num2double.
2514 These are conversion functions between the two ANSI C float types and
2517 ** New number validation macros:
2518 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2522 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2524 These are just nicer-named old scm_protect_object and
2525 scm_unprotect_object.
2527 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2529 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2531 These functions can be used to register pointers to locations that
2534 ** Deprecated function: scm_create_hook.
2536 Its sins are: misleading name, non-modularity and lack of general
2540 Changes since Guile 1.3.4:
2542 * Changes to the distribution
2544 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2546 We've changed the way we handle generated files in the Guile source
2547 repository. As a result, the procedure for building trees obtained
2548 from the nightly FTP snapshots or via CVS has changed:
2549 - You must have appropriate versions of autoconf, automake, and
2550 libtool installed on your system. See README for info on how to
2551 obtain these programs.
2552 - Before configuring the tree, you must first run the script
2553 `autogen.sh' at the top of the source tree.
2555 The Guile repository used to contain not only source files, written by
2556 humans, but also some generated files, like configure scripts and
2557 Makefile.in files. Even though the contents of these files could be
2558 derived mechanically from other files present, we thought it would
2559 make the tree easier to build if we checked them into CVS.
2561 However, this approach means that minor differences between
2562 developer's installed tools and habits affected the whole team.
2563 So we have removed the generated files from the repository, and
2564 added the autogen.sh script, which will reconstruct them
2568 ** configure now has experimental options to remove support for certain
2571 --disable-arrays omit array and uniform array support
2572 --disable-posix omit posix interfaces
2573 --disable-networking omit networking interfaces
2574 --disable-regex omit regular expression interfaces
2576 These are likely to become separate modules some day.
2578 ** New configure option --enable-debug-freelist
2580 This enables a debugging version of SCM_NEWCELL(), and also registers
2581 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2583 Configure with the --enable-debug-freelist option to enable
2584 the gc-set-debug-check-freelist! primitive, and then use:
2586 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2587 (gc-set-debug-check-freelist! #f) # turn off checking
2589 Checking of the freelist forces a traversal of the freelist and
2590 a garbage collection before each allocation of a cell. This can
2591 slow down the interpreter dramatically, so the setter should be used to
2592 turn on this extra processing only when necessary.
2594 ** New configure option --enable-debug-malloc
2596 Include code for debugging of calls to scm_must_malloc/realloc/free.
2600 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2601 2. objects reallocated by scm_must_realloc has been allocated by
2603 3. reallocated objects are reallocated with the same what string
2605 But, most importantly, it records the number of allocated objects of
2606 each kind. This is useful when searching for memory leaks.
2608 A Guile compiled with this option provides the primitive
2609 `malloc-stats' which returns an alist with pairs of kind and the
2610 number of objects of that kind.
2612 ** All includes are now referenced relative to the root directory
2614 Since some users have had problems with mixups between Guile and
2615 system headers, we have decided to always refer to Guile headers via
2616 their parent directories. This essentially creates a "private name
2617 space" for Guile headers. This means that the compiler only is given
2618 -I options for the root build and root source directory.
2620 ** Header files kw.h and genio.h have been removed.
2622 ** The module (ice-9 getopt-gnu-style) has been removed.
2624 ** New module (ice-9 documentation)
2626 Implements the interface to documentation strings associated with
2629 ** New module (ice-9 time)
2631 Provides a macro `time', which displays execution time of a given form.
2633 ** New module (ice-9 history)
2635 Loading this module enables value history in the repl.
2637 * Changes to the stand-alone interpreter
2639 ** New command line option --debug
2641 Start Guile with debugging evaluator and backtraces enabled.
2643 This is useful when debugging your .guile init file or scripts.
2645 ** New help facility
2647 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2648 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2649 (help 'NAME) gives documentation for NAME, even if it is not an object
2650 (help ,EXPR) gives documentation for object returned by EXPR
2651 (help (my module)) gives module commentary for `(my module)'
2652 (help) gives this text
2654 `help' searches among bindings exported from loaded modules, while
2655 `apropos' searches among bindings visible from the "current" module.
2657 Examples: (help help)
2659 (help "output-string")
2661 ** `help' and `apropos' now prints full module names
2663 ** Dynamic linking now uses libltdl from the libtool package.
2665 The old system dependent code for doing dynamic linking has been
2666 replaced with calls to the libltdl functions which do all the hairy
2669 The major improvement is that you can now directly pass libtool
2670 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2671 will be able to do the best shared library job you can get, via
2674 The way dynamic libraries are found has changed and is not really
2675 portable across platforms, probably. It is therefore recommended to
2676 use absolute filenames when possible.
2678 If you pass a filename without an extension to `dynamic-link', it will
2679 try a few appropriate ones. Thus, the most platform ignorant way is
2680 to specify a name like "libfoo", without any directories and
2683 ** Guile COOP threads are now compatible with LinuxThreads
2685 Previously, COOP threading wasn't possible in applications linked with
2686 Linux POSIX threads due to their use of the stack pointer to find the
2687 thread context. This has now been fixed with a workaround which uses
2688 the pthreads to allocate the stack.
2690 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2692 ** Positions of erring expression in scripts
2694 With version 1.3.4, the location of the erring expression in Guile
2695 scipts is no longer automatically reported. (This should have been
2696 documented before the 1.3.4 release.)
2698 You can get this information by enabling recording of positions of
2699 source expressions and running the debugging evaluator. Put this at
2700 the top of your script (or in your "site" file):
2702 (read-enable 'positions)
2703 (debug-enable 'debug)
2705 ** Backtraces in scripts
2707 It is now possible to get backtraces in scripts.
2711 (debug-enable 'debug 'backtrace)
2713 at the top of the script.
2715 (The first options enables the debugging evaluator.
2716 The second enables backtraces.)
2718 ** Part of module system symbol lookup now implemented in C
2720 The eval closure of most modules is now implemented in C. Since this
2721 was one of the bottlenecks for loading speed, Guile now loads code
2722 substantially faster than before.
2724 ** Attempting to get the value of an unbound variable now produces
2725 an exception with a key of 'unbound-variable instead of 'misc-error.
2727 ** The initial default output port is now unbuffered if it's using a
2728 tty device. Previously in this situation it was line-buffered.
2730 ** New hook: after-gc-hook
2732 after-gc-hook takes over the role of gc-thunk. This hook is run at
2733 the first SCM_TICK after a GC. (Thus, the code is run at the same
2734 point during evaluation as signal handlers.)
2736 Note that this hook should be used only for diagnostic and debugging
2737 purposes. It is not certain that it will continue to be well-defined
2738 when this hook is run in the future.
2740 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2741 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2743 ** Improvements to garbage collector
2745 Guile 1.4 has a new policy for triggering heap allocation and
2746 determining the sizes of heap segments. It fixes a number of problems
2749 1. The new policy can handle two separate pools of cells
2750 (2-word/4-word) better. (The old policy would run wild, allocating
2751 more and more memory for certain programs.)
2753 2. The old code would sometimes allocate far too much heap so that the
2754 Guile process became gigantic. The new code avoids this.
2756 3. The old code would sometimes allocate too little so that few cells
2757 were freed at GC so that, in turn, too much time was spent in GC.
2759 4. The old code would often trigger heap allocation several times in a
2760 row. (The new scheme predicts how large the segments needs to be
2761 in order not to need further allocation.)
2763 All in all, the new GC policy will make larger applications more
2766 The new GC scheme also is prepared for POSIX threading. Threads can
2767 allocate private pools of cells ("clusters") with just a single
2768 function call. Allocation of single cells from such a cluster can
2769 then proceed without any need of inter-thread synchronization.
2771 ** New environment variables controlling GC parameters
2773 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2776 Allocation of 2-word cell heaps:
2778 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2781 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2782 GC in percent of total heap size
2785 Allocation of 4-word cell heaps
2786 (used for real numbers and misc other objects):
2788 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2790 (See entry "Way for application to customize GC parameters" under
2791 section "Changes to the scm_ interface" below.)
2793 ** Guile now implements reals using 4-word cells
2795 This speeds up computation with reals. (They were earlier allocated
2796 with `malloc'.) There is still some room for optimizations, however.
2798 ** Some further steps toward POSIX thread support have been taken
2800 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2801 don't have much effect any longer, and many of them will be removed in
2805 are only handled at the top of the evaluator loop, immediately after
2806 I/O, and in scm_equalp.
2808 *** The GC can allocate thread private pools of pairs.
2810 * Changes to Scheme functions and syntax
2812 ** close-input-port and close-output-port are now R5RS
2814 These procedures have been turned into primitives and have R5RS behaviour.
2816 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2818 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2819 extended by the more sophisticated version in (ice-9 format)
2821 (simple-format port message . args)
2822 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2823 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2824 the escapes are replaced with corresponding members of ARGS:
2825 ~A formats using `display' and ~S formats using `write'.
2826 If DESTINATION is #t, then use the `current-output-port',
2827 if DESTINATION is #f, then return a string containing the formatted text.
2828 Does not add a trailing newline."
2830 ** string-ref: the second argument is no longer optional.
2832 ** string, list->string: no longer accept strings in their arguments,
2833 only characters, for compatibility with R5RS.
2835 ** New procedure: port-closed? PORT
2836 Returns #t if PORT is closed or #f if it is open.
2838 ** Deprecated: list*
2840 The list* functionality is now provided by cons* (SRFI-1 compliant)
2842 ** New procedure: cons* ARG1 ARG2 ... ARGn
2844 Like `list', but the last arg provides the tail of the constructed list,
2845 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2847 Requires at least one argument. If given one argument, that argument
2848 is returned as result.
2850 This function is called `list*' in some other Schemes and in Common LISP.
2852 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2854 ** New procedure: object-documentation OBJECT
2856 Returns the documentation string associated with OBJECT. The
2857 procedure uses a caching mechanism so that subsequent lookups are
2860 Exported by (ice-9 documentation).
2862 ** module-name now returns full names of modules
2864 Previously, only the last part of the name was returned (`session' for
2865 `(ice-9 session)'). Ex: `(ice-9 session)'.
2867 * Changes to the gh_ interface
2869 ** Deprecated: gh_int2scmb
2871 Use gh_bool2scm instead.
2873 * Changes to the scm_ interface
2875 ** Guile primitives now carry docstrings!
2877 Thanks to Greg Badros!
2879 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2881 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2882 macros and must contain a docstring that is extracted into foo.doc using a new
2883 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2885 However, a major overhaul of these macros is scheduled for the next release of
2888 ** Guile primitives use a new technique for validation of arguments
2890 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2891 the readability of argument checking.
2893 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2895 ** New macros: SCM_PACK, SCM_UNPACK
2897 Compose/decompose an SCM value.
2899 The SCM type is now treated as an abstract data type and may be defined as a
2900 long, a void* or as a struct, depending on the architecture and compile time
2901 options. This makes it easier to find several types of bugs, for example when
2902 SCM values are treated as integers without conversion. Values of the SCM type
2903 should be treated as "atomic" values. These macros are used when
2904 composing/decomposing an SCM value, either because you want to access
2905 individual bits, or because you want to treat it as an integer value.
2907 E.g., in order to set bit 7 in an SCM value x, use the expression
2909 SCM_PACK (SCM_UNPACK (x) | 0x80)
2911 ** The name property of hooks is deprecated.
2912 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2914 You can emulate this feature by using object properties.
2916 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2917 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2920 These macros will be removed in a future release of Guile.
2922 ** The following types, functions and macros from numbers.h are deprecated:
2923 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2924 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2926 ** Port internals: the rw_random variable in the scm_port structure
2927 must be set to non-zero in any random access port. In recent Guile
2928 releases it was only set for bidirectional random-access ports.
2930 ** Port internals: the seek ptob procedure is now responsible for
2931 resetting the buffers if required. The change was made so that in the
2932 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2933 the fport and strport ptobs can avoid resetting the buffers,
2934 in particular to avoid discarding unread chars. An existing port
2935 type can be fixed by adding something like the following to the
2936 beginning of the ptob seek procedure:
2938 if (pt->rw_active == SCM_PORT_READ)
2939 scm_end_input (object);
2940 else if (pt->rw_active == SCM_PORT_WRITE)
2941 ptob->flush (object);
2943 although to actually avoid resetting the buffers and discard unread
2944 chars requires further hacking that depends on the characteristics
2947 ** Deprecated functions: scm_fseek, scm_tag
2949 These functions are no longer used and will be removed in a future version.
2951 ** The scm_sysmissing procedure is no longer used in libguile.
2952 Unless it turns out to be unexpectedly useful to somebody, it will be
2953 removed in a future version.
2955 ** The format of error message strings has changed
2957 The two C procedures: scm_display_error and scm_error, as well as the
2958 primitive `scm-error', now use scm_simple_format to do their work.
2959 This means that the message strings of all code must be updated to use
2960 ~A where %s was used before, and ~S where %S was used before.
2962 During the period when there still are a lot of old Guiles out there,
2963 you might want to support both old and new versions of Guile.
2965 There are basically two methods to achieve this. Both methods use
2968 AC_CHECK_FUNCS(scm_simple_format)
2970 in your configure.in.
2972 Method 1: Use the string concatenation features of ANSI C's
2977 #ifdef HAVE_SCM_SIMPLE_FORMAT
2983 Then represent each of your error messages using a preprocessor macro:
2985 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2989 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2990 (define make-message string-append)
2992 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2994 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2998 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3003 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3007 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3009 Don't use the functions coop_mutex_init and
3010 coop_condition_variable_init. They will change.
3012 Use scm_mutex_init and scm_cond_init instead.
3014 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3015 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3016 COND, as `scm_cond_wait' does, but it also bounds the duration
3017 of the wait. If COND has not been signaled before time ABSTIME,
3018 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3019 returns the error code `ETIMEDOUT'.
3021 The ABSTIME parameter specifies an absolute time, with the same
3022 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3023 to 00:00:00 GMT, January 1, 1970.
3025 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3026 `scm_cond_broadcast' restarts all the threads that are waiting
3027 on the condition variable COND. Nothing happens if no threads are
3030 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3031 `scm_key_create' allocates a new TSD key. The key is stored in
3032 the location pointed to by KEY. There is no limit on the number
3033 of keys allocated at a given time. The value initially associated
3034 with the returned key is `NULL' in all currently executing threads.
3036 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3037 function associated with the key. When a thread terminates,
3038 DESTR_FUNCTION is called on the value associated with the key in
3039 that thread. The DESTR_FUNCTION is not called if a key is deleted
3040 with `scm_key_delete' or a value is changed with
3041 `scm_setspecific'. The order in which destructor functions are
3042 called at thread termination time is unspecified.
3044 Destructors are not yet implemented.
3046 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3047 `scm_setspecific' changes the value associated with KEY in the
3048 calling thread, storing the given POINTER instead.
3050 ** New function: scm_getspecific (scm_key_t KEY)
3051 `scm_getspecific' returns the value currently associated with
3052 KEY in the calling thread.
3054 ** New function: scm_key_delete (scm_key_t KEY)
3055 `scm_key_delete' deallocates a TSD key. It does not check
3056 whether non-`NULL' values are associated with that key in the
3057 currently executing threads, nor call the destructor function
3058 associated with the key.
3060 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3062 Initialize a C level hook HOOK with associated HOOK_DATA and type
3063 TYPE. (See scm_c_hook_run ().)
3065 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3067 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3068 is true, add it last, otherwise first. The same FUNC can be added
3069 multiple times if FUNC_DATA differ and vice versa.
3071 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3073 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3074 function is only removed if both FUNC and FUNC_DATA matches.
3076 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3078 Run hook HOOK passing DATA to the hook functions.
3080 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3081 returned is undefined.
3083 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3084 returns a non-NULL value. This value is returned as the result of
3085 scm_c_hook_run. If all functions return NULL, NULL is returned.
3087 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3088 returns a NULL value, and NULL is returned. If all functions returns
3089 a non-NULL value, the last value is returned.
3091 ** New C level GC hooks
3093 Five new C level hooks has been added to the garbage collector.
3095 scm_before_gc_c_hook
3098 are run before locking and after unlocking the heap. The system is
3099 thus in a mode where evaluation can take place. (Except that
3100 scm_before_gc_c_hook must not allocate new cells.)
3102 scm_before_mark_c_hook
3103 scm_before_sweep_c_hook
3104 scm_after_sweep_c_hook
3106 are run when the heap is locked. These are intended for extension of
3107 the GC in a modular fashion. Examples are the weaks and guardians
3110 ** Way for application to customize GC parameters
3112 The application can set up other default values for the GC heap
3113 allocation parameters
3115 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3116 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3117 GUILE_MAX_SEGMENT_SIZE,
3121 scm_default_init_heap_size_1, scm_default_min_yield_1,
3122 scm_default_init_heap_size_2, scm_default_min_yield_2,
3123 scm_default_max_segment_size
3125 respectively before callong scm_boot_guile.
3127 (See entry "New environment variables ..." in section
3128 "Changes to the stand-alone interpreter" above.)
3130 ** scm_protect_object/scm_unprotect_object now nest
3132 This means that you can call scm_protect_object multiple times on an
3133 object and count on the object being protected until
3134 scm_unprotect_object has been call the same number of times.
3136 The functions also have better time complexity.
3138 Still, it is usually possible to structure the application in a way
3139 that you don't need to use these functions. For example, if you use a
3140 protected standard Guile list to keep track of live objects rather
3141 than some custom data type, objects will die a natural death when they
3142 are no longer needed.
3144 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3146 Guile does not provide the float representation for inexact real numbers any
3147 more. Now, only doubles are used to represent inexact real numbers. Further,
3148 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3149 and scm_tc16_complex, respectively.
3151 ** Removed deprecated type scm_smobfuns
3153 ** Removed deprecated function scm_newsmob
3155 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3157 There is an ongoing discussion among the developers whether to
3158 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3159 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3160 until this issue has been settled.
3162 ** Removed deprecated type tag scm_tc16_kw
3164 ** Added type tag scm_tc16_keyword
3166 (This was introduced already in release 1.3.4 but was not documented
3169 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3171 * Changes to system call interfaces:
3173 ** The "select" procedure now tests port buffers for the ability to
3174 provide input or accept output. Previously only the underlying file
3175 descriptors were checked.
3177 ** New variable PIPE_BUF: the maximum number of bytes that can be
3178 atomically written to a pipe.
3180 ** If a facility is not available on the system when Guile is
3181 compiled, the corresponding primitive procedure will not be defined.
3182 Previously it would have been defined but would throw a system-error
3183 exception if called. Exception handlers which catch this case may
3184 need minor modification: an error will be thrown with key
3185 'unbound-variable instead of 'system-error. Alternatively it's
3186 now possible to use `defined?' to check whether the facility is
3189 ** Procedures which depend on the timezone should now give the correct
3190 result on systems which cache the TZ environment variable, even if TZ
3191 is changed without calling tzset.
3193 * Changes to the networking interfaces:
3195 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3196 long integers between network and host format. For now, it's not
3197 particularly convenient to do this kind of thing, but consider:
3199 (define write-network-long
3200 (lambda (value port)
3201 (let ((v (make-uniform-vector 1 1 0)))
3202 (uniform-vector-set! v 0 (htonl value))
3203 (uniform-vector-write v port))))
3205 (define read-network-long
3207 (let ((v (make-uniform-vector 1 1 0)))
3208 (uniform-vector-read! v port)
3209 (ntohl (uniform-vector-ref v 0)))))
3211 ** If inet-aton fails, it now throws an error with key 'misc-error
3212 instead of 'system-error, since errno is not relevant.
3214 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3215 specific keys instead of 'system-error. The latter is inappropriate
3216 since errno will not have been set. The keys are:
3217 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3219 ** sethostent, setnetent, setprotoent, setservent: now take an
3220 optional argument STAYOPEN, which specifies whether the database
3221 remains open after a database entry is accessed randomly (e.g., using
3222 gethostbyname for the hosts database.) The default is #f. Previously
3226 Changes since Guile 1.3.2:
3228 * Changes to the stand-alone interpreter
3232 An initial version of the Guile debugger written by Chris Hanson has
3233 been added. The debugger is still under development but is included
3234 in the distribution anyway since it is already quite useful.
3240 after an error to enter the debugger. Type `help' inside the debugger
3241 for a description of available commands.
3243 If you prefer to have stack frames numbered and printed in
3244 anti-chronological order and prefer up in the stack to be down on the
3245 screen as is the case in gdb, you can put
3247 (debug-enable 'backwards)
3249 in your .guile startup file. (However, this means that Guile can't
3250 use indentation to indicate stack level.)
3252 The debugger is autoloaded into Guile at the first use.
3254 ** Further enhancements to backtraces
3256 There is a new debug option `width' which controls the maximum width
3257 on the screen of printed stack frames. Fancy printing parameters
3258 ("level" and "length" as in Common LISP) are adaptively adjusted for
3259 each stack frame to give maximum information while still fitting
3260 within the bounds. If the stack frame can't be made to fit by
3261 adjusting parameters, it is simply cut off at the end. This is marked
3264 ** Some modules are now only loaded when the repl is started
3266 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3267 regex) are now loaded into (guile-user) only if the repl has been
3268 started. The effect is that the startup time for scripts has been
3269 reduced to 30% of what it was previously.
3271 Correctly written scripts load the modules they require at the top of
3272 the file and should not be affected by this change.
3274 ** Hooks are now represented as smobs
3276 * Changes to Scheme functions and syntax
3278 ** Readline support has changed again.
3280 The old (readline-activator) module is gone. Use (ice-9 readline)
3281 instead, which now contains all readline functionality. So the code
3282 to activate readline is now
3284 (use-modules (ice-9 readline))
3287 This should work at any time, including from the guile prompt.
3289 To avoid confusion about the terms of Guile's license, please only
3290 enable readline for your personal use; please don't make it the
3291 default for others. Here is why we make this rather odd-sounding
3294 Guile is normally licensed under a weakened form of the GNU General
3295 Public License, which allows you to link code with Guile without
3296 placing that code under the GPL. This exception is important to some
3299 However, since readline is distributed under the GNU General Public
3300 License, when you link Guile with readline, either statically or
3301 dynamically, you effectively change Guile's license to the strict GPL.
3302 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3303 which are normally permitted become forbidden. This is a rather
3304 non-obvious consequence of the licensing terms.
3306 So, to make sure things remain clear, please let people choose for
3307 themselves whether to link GPL'd libraries like readline with Guile.
3309 ** regexp-substitute/global has changed slightly, but incompatibly.
3311 If you include a function in the item list, the string of the match
3312 object it receives is the same string passed to
3313 regexp-substitute/global, not some suffix of that string.
3314 Correspondingly, the match's positions are relative to the entire
3315 string, not the suffix.
3317 If the regexp can match the empty string, the way matches are chosen
3318 from the string has changed. regexp-substitute/global recognizes the
3319 same set of matches that list-matches does; see below.
3321 ** New function: list-matches REGEXP STRING [FLAGS]
3323 Return a list of match objects, one for every non-overlapping, maximal
3324 match of REGEXP in STRING. The matches appear in left-to-right order.
3325 list-matches only reports matches of the empty string if there are no
3326 other matches which begin on, end at, or include the empty match's
3329 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3331 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3333 For each match of REGEXP in STRING, apply PROC to the match object,
3334 and the last value PROC returned, or INIT for the first call. Return
3335 the last value returned by PROC. We apply PROC to the matches as they
3336 appear from left to right.
3338 This function recognizes matches according to the same criteria as
3341 Thus, you could define list-matches like this:
3343 (define (list-matches regexp string . flags)
3344 (reverse! (apply fold-matches regexp string '() cons flags)))
3346 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3350 *** New function: hook? OBJ
3352 Return #t if OBJ is a hook, otherwise #f.
3354 *** New function: make-hook-with-name NAME [ARITY]
3356 Return a hook with name NAME and arity ARITY. The default value for
3357 ARITY is 0. The only effect of NAME is that it will appear when the
3358 hook object is printed to ease debugging.
3360 *** New function: hook-empty? HOOK
3362 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3364 *** New function: hook->list HOOK
3366 Return a list of the procedures that are called when run-hook is
3369 ** `map' signals an error if its argument lists are not all the same length.
3371 This is the behavior required by R5RS, so this change is really a bug
3372 fix. But it seems to affect a lot of people's code, so we're
3373 mentioning it here anyway.
3375 ** Print-state handling has been made more transparent
3377 Under certain circumstances, ports are represented as a port with an
3378 associated print state. Earlier, this pair was represented as a pair
3379 (see "Some magic has been added to the printer" below). It is now
3380 indistinguishable (almost; see `get-print-state') from a port on the
3383 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3385 Return a new port with the associated print state PRINT-STATE.
3387 *** New function: get-print-state OUTPUT-PORT
3389 Return the print state associated with this port if it exists,
3390 otherwise return #f.
3392 *** New function: directory-stream? OBJECT
3394 Returns true iff OBJECT is a directory stream --- the sort of object
3395 returned by `opendir'.
3397 ** New function: using-readline?
3399 Return #t if readline is in use in the current repl.
3401 ** structs will be removed in 1.4
3403 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3404 and use GOOPS objects as the fundamental record type.
3406 * Changes to the scm_ interface
3408 ** structs will be removed in 1.4
3410 The entire current struct interface (struct.c, struct.h) will be
3411 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3412 GOOPS objects as the fundamental record type.
3414 ** The internal representation of subr's has changed
3416 Instead of giving a hint to the subr name, the CAR field of the subr
3417 now contains an index to a subr entry in scm_subr_table.
3419 *** New variable: scm_subr_table
3421 An array of subr entries. A subr entry contains the name, properties
3422 and documentation associated with the subr. The properties and
3423 documentation slots are not yet used.
3425 ** A new scheme for "forwarding" calls to a builtin to a generic function
3427 It is now possible to extend the functionality of some Guile
3428 primitives by letting them defer a call to a GOOPS generic function on
3429 argument mismatch. This means that there is no loss of efficiency in
3434 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3435 (define-method + ((x <string>) (y <string>))
3436 (string-append x y))
3438 + will still be as efficient as usual in numerical calculations, but
3439 can also be used for concatenating strings.
3441 Who will be the first one to extend Guile's numerical tower to
3442 rationals? :) [OK, there a few other things to fix before this can
3443 be made in a clean way.]
3445 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3447 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3449 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3451 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3452 a variable GENERIC which can be used by the dispatch macros below.
3454 [This is experimental code which may change soon.]
3456 *** New macros for forwarding control to a generic on arg type error
3458 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3460 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3462 These correspond to the scm_wta function call, and have the same
3463 behaviour until the user has called the GOOPS primitive
3464 `enable-primitive-generic!'. After that, these macros will apply the
3465 generic function GENERIC to the argument(s) instead of calling
3468 [This is experimental code which may change soon.]
3470 *** New macros for argument testing with generic dispatch
3472 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3474 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3476 These correspond to the SCM_ASSERT macro, but will defer control to
3477 GENERIC on error after `enable-primitive-generic!' has been called.
3479 [This is experimental code which may change soon.]
3481 ** New function: SCM scm_eval_body (SCM body, SCM env)
3483 Evaluates the body of a special form.
3485 ** The internal representation of struct's has changed
3487 Previously, four slots were allocated for the procedure(s) of entities
3488 and operators. The motivation for this representation had to do with
3489 the structure of the evaluator, the wish to support tail-recursive
3490 generic functions, and efficiency. Since the generic function
3491 dispatch mechanism has changed, there is no longer a need for such an
3492 expensive representation, and the representation has been simplified.
3494 This should not make any difference for most users.
3496 ** GOOPS support has been cleaned up.
3498 Some code has been moved from eval.c to objects.c and code in both of
3499 these compilation units has been cleaned up and better structured.
3501 *** New functions for applying generic functions
3503 New function: SCM scm_apply_generic (GENERIC, ARGS)
3504 New function: SCM scm_call_generic_0 (GENERIC)
3505 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3506 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3507 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3509 ** Deprecated function: scm_make_named_hook
3511 It is now replaced by:
3513 ** New function: SCM scm_create_hook (const char *name, int arity)
3515 Creates a hook in the same way as make-hook above but also
3516 binds a variable named NAME to it.
3518 This is the typical way of creating a hook from C code.
3520 Currently, the variable is created in the "current" module.
3521 This might change when we get the new module system.
3523 [The behaviour is identical to scm_make_named_hook.]
3527 Changes since Guile 1.3:
3529 * Changes to mailing lists
3531 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3533 See the README file to find current addresses for all the Guile
3536 * Changes to the distribution
3538 ** Readline support is no longer included with Guile by default.
3540 Based on the different license terms of Guile and Readline, we
3541 concluded that Guile should not *by default* cause the linking of
3542 Readline into an application program. Readline support is now offered
3543 as a separate module, which is linked into an application only when
3544 you explicitly specify it.
3546 Although Guile is GNU software, its distribution terms add a special
3547 exception to the usual GNU General Public License (GPL). Guile's
3548 license includes a clause that allows you to link Guile with non-free
3549 programs. We add this exception so as not to put Guile at a
3550 disadvantage vis-a-vis other extensibility packages that support other
3553 In contrast, the GNU Readline library is distributed under the GNU
3554 General Public License pure and simple. This means that you may not
3555 link Readline, even dynamically, into an application unless it is
3556 distributed under a free software license that is compatible the GPL.
3558 Because of this difference in distribution terms, an application that
3559 can use Guile may not be able to use Readline. Now users will be
3560 explicitly offered two independent decisions about the use of these
3563 You can activate the readline support by issuing
3565 (use-modules (readline-activator))
3568 from your ".guile" file, for example.
3570 * Changes to the stand-alone interpreter
3572 ** All builtins now print as primitives.
3573 Previously builtin procedures not belonging to the fundamental subr
3574 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3575 Now, they print as #<primitive-procedure NAME>.
3577 ** Backtraces slightly more intelligible.
3578 gsubr-apply and macro transformer application frames no longer appear
3581 * Changes to Scheme functions and syntax
3583 ** Guile now correctly handles internal defines by rewriting them into
3584 their equivalent letrec. Previously, internal defines would
3585 incrementally add to the innermost environment, without checking
3586 whether the restrictions specified in RnRS were met. This lead to the
3587 correct behaviour when these restriction actually were met, but didn't
3588 catch all illegal uses. Such an illegal use could lead to crashes of
3589 the Guile interpreter or or other unwanted results. An example of
3590 incorrect internal defines that made Guile behave erratically:
3602 The problem with this example is that the definition of `c' uses the
3603 value of `b' directly. This confuses the meoization machine of Guile
3604 so that the second call of `b' (this time in a larger environment that
3605 also contains bindings for `c' and `d') refers to the binding of `c'
3606 instead of `a'. You could also make Guile crash with a variation on
3611 (define (b flag) (if flag a 1))
3612 (define c (1+ (b flag)))
3620 From now on, Guile will issue an `Unbound variable: b' error message
3625 A hook contains a list of functions which should be called on
3626 particular occasions in an existing program. Hooks are used for
3629 A window manager might have a hook before-window-map-hook. The window
3630 manager uses the function run-hooks to call all functions stored in
3631 before-window-map-hook each time a window is mapped. The user can
3632 store functions in the hook using add-hook!.
3634 In Guile, hooks are first class objects.
3636 *** New function: make-hook [N_ARGS]
3638 Return a hook for hook functions which can take N_ARGS arguments.
3639 The default value for N_ARGS is 0.
3641 (See also scm_make_named_hook below.)
3643 *** New function: add-hook! HOOK PROC [APPEND_P]
3645 Put PROC at the beginning of the list of functions stored in HOOK.
3646 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3648 PROC must be able to take the number of arguments specified when the
3651 If PROC already exists in HOOK, then remove it first.
3653 *** New function: remove-hook! HOOK PROC
3655 Remove PROC from the list of functions in HOOK.
3657 *** New function: reset-hook! HOOK
3659 Clear the list of hook functions stored in HOOK.
3661 *** New function: run-hook HOOK ARG1 ...
3663 Run all hook functions stored in HOOK with arguments ARG1 ... .
3664 The number of arguments supplied must correspond to the number given
3665 when the hook was created.
3667 ** The function `dynamic-link' now takes optional keyword arguments.
3668 The only keyword argument that is currently defined is `:global
3669 BOOL'. With it, you can control whether the shared library will be
3670 linked in global mode or not. In global mode, the symbols from the
3671 linked library can be used to resolve references from other
3672 dynamically linked libraries. In non-global mode, the linked
3673 library is essentially invisible and can only be accessed via
3674 `dynamic-func', etc. The default is now to link in global mode.
3675 Previously, the default has been non-global mode.
3677 The `#:global' keyword is only effective on platforms that support
3678 the dlopen family of functions.
3680 ** New function `provided?'
3682 - Function: provided? FEATURE
3683 Return true iff FEATURE is supported by this installation of
3684 Guile. FEATURE must be a symbol naming a feature; the global
3685 variable `*features*' is a list of available features.
3687 ** Changes to the module (ice-9 expect):
3689 *** The expect-strings macro now matches `$' in a regular expression
3690 only at a line-break or end-of-file by default. Previously it would
3691 match the end of the string accumulated so far. The old behaviour
3692 can be obtained by setting the variable `expect-strings-exec-flags'
3695 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3696 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3697 in a regular expression will still match before a line-break or
3698 end-of-file. The default is `regexp/noteol'.
3700 *** The expect-strings macro now uses a variable
3701 `expect-strings-compile-flags' for the flags to be supplied to
3702 `make-regexp'. The default is `regexp/newline', which was previously
3705 *** The expect macro now supplies two arguments to a match procedure:
3706 the current accumulated string and a flag to indicate whether
3707 end-of-file has been reached. Previously only the string was supplied.
3708 If end-of-file is reached, the match procedure will be called an
3709 additional time with the same accumulated string as the previous call
3710 but with the flag set.
3712 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3714 This code, and the documentation for it that appears here, was
3715 borrowed from SLIB, with minor adaptations for Guile.
3717 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3718 An almost complete implementation of Common LISP format description
3719 according to the CL reference book `Common LISP' from Guy L.
3720 Steele, Digital Press. Backward compatible to most of the
3721 available Scheme format implementations.
3723 Returns `#t', `#f' or a string; has side effect of printing
3724 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3725 to the current output port and `#t' is returned. If DESTINATION
3726 is `#f', a formatted string is returned as the result of the call.
3727 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3728 format string; FORMAT-STRING is then the first argument and the
3729 output is returned as a string. If DESTINATION is a number, the
3730 output is to the current error port if available by the
3731 implementation. Otherwise DESTINATION must be an output port and
3734 FORMAT-STRING must be a string. In case of a formatting error
3735 format returns `#f' and prints a message on the current output or
3736 error port. Characters are output as if the string were output by
3737 the `display' function with the exception of those prefixed by a
3738 tilde (~). For a detailed description of the FORMAT-STRING syntax
3739 please consult a Common LISP format reference manual. For a test
3740 suite to verify this format implementation load `formatst.scm'.
3741 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3743 Note: `format' is not reentrant, i.e. only one `format'-call may
3744 be executed at a time.
3747 *** Format Specification (Format version 3.0)
3749 Please consult a Common LISP format reference manual for a detailed
3750 description of the format string syntax. For a demonstration of the
3751 implemented directives see `formatst.scm'.
3753 This implementation supports directive parameters and modifiers (`:'
3754 and `@' characters). Multiple parameters must be separated by a comma
3755 (`,'). Parameters can be numerical parameters (positive or negative),
3756 character parameters (prefixed by a quote character (`''), variable
3757 parameters (`v'), number of rest arguments parameter (`#'), empty and
3758 default parameters. Directive characters are case independent. The
3759 general form of a directive is:
3761 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3763 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3765 *** Implemented CL Format Control Directives
3767 Documentation syntax: Uppercase characters represent the
3768 corresponding control directive characters. Lowercase characters
3769 represent control directive parameter descriptions.
3772 Any (print as `display' does).
3776 `~MINCOL,COLINC,MINPAD,PADCHARA'
3780 S-expression (print as `write' does).
3784 `~MINCOL,COLINC,MINPAD,PADCHARS'
3790 print number sign always.
3793 print comma separated.
3795 `~MINCOL,PADCHAR,COMMACHARD'
3801 print number sign always.
3804 print comma separated.
3806 `~MINCOL,PADCHAR,COMMACHARX'
3812 print number sign always.
3815 print comma separated.
3817 `~MINCOL,PADCHAR,COMMACHARO'
3823 print number sign always.
3826 print comma separated.
3828 `~MINCOL,PADCHAR,COMMACHARB'
3833 `~N,MINCOL,PADCHAR,COMMACHARR'
3837 print a number as a Roman numeral.
3840 print a number as an "old fashioned" Roman numeral.
3843 print a number as an ordinal English number.
3846 print a number as a cardinal English number.
3851 prints `y' and `ies'.
3854 as `~P but jumps 1 argument backward.'
3857 as `~@P but jumps 1 argument backward.'
3862 prints a character as the reader can understand it (i.e. `#\'
3866 prints a character as emacs does (eg. `^C' for ASCII 03).
3869 Fixed-format floating-point (prints a flonum like MMM.NNN).
3870 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3872 If the number is positive a plus sign is printed.
3875 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3876 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3878 If the number is positive a plus sign is printed.
3881 General floating-point (prints a flonum either fixed or
3883 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3885 If the number is positive a plus sign is printed.
3888 Dollars floating-point (prints a flonum in fixed with signs
3890 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3892 If the number is positive a plus sign is printed.
3895 A sign is always printed and appears before the padding.
3898 The sign appears before the padding.
3906 print newline if not at the beginning of the output line.
3908 prints `~&' and then N-1 newlines.
3913 print N page separators.
3923 newline is ignored, white space left.
3926 newline is left, white space ignored.
3931 relative tabulation.
3937 Indirection (expects indirect arguments as a list).
3939 extracts indirect arguments from format arguments.
3942 Case conversion (converts by `string-downcase').
3944 converts by `string-capitalize'.
3947 converts by `string-capitalize-first'.
3950 converts by `string-upcase'.
3953 Argument Jumping (jumps 1 argument forward).
3955 jumps N arguments forward.
3958 jumps 1 argument backward.
3961 jumps N arguments backward.
3964 jumps to the 0th argument.
3967 jumps to the Nth argument (beginning from 0)
3969 `~[STR0~;STR1~;...~;STRN~]'
3970 Conditional Expression (numerical clause conditional).
3972 take argument from N.
3975 true test conditional.
3978 if-else-then conditional.
3984 default clause follows.
3987 Iteration (args come from the next argument (a list)).
3989 at most N iterations.
3992 args from next arg (a list of lists).
3995 args from the rest of arguments.
3998 args from the rest args (lists).
4009 aborts if N <= M <= K
4011 *** Not Implemented CL Format Control Directives
4014 print `#f' as an empty list (see below).
4017 print `#f' as an empty list (see below).
4023 (sorry I don't understand its semantics completely)
4025 *** Extended, Replaced and Additional Control Directives
4027 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4028 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4029 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4030 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4031 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4032 COMMAWIDTH is the number of characters between two comma
4036 print a R4RS complex number as `~F~@Fi' with passed parameters for
4040 Pretty print formatting of an argument for scheme code lists.
4046 Flushes the output if format DESTINATION is a port.
4049 Print a `#\space' character
4051 print N `#\space' characters.
4054 Print a `#\tab' character
4056 print N `#\tab' characters.
4059 Takes N as an integer representation for a character. No arguments
4060 are consumed. N is converted to a character by `integer->char'. N
4061 must be a positive decimal number.
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 Print out readproof. Prints out internal objects represented as
4070 `#<...>' as strings `"#<...>"' so that the format output can always
4071 be processed by `read'.
4074 Prints information and a copyright notice on the format
4077 prints format version.
4080 may also print number strings, i.e. passing a number as a string
4081 and format it accordingly.
4083 *** Configuration Variables
4085 The format module exports some configuration variables to suit the
4086 systems and users needs. There should be no modification necessary for
4087 the configuration that comes with Guile. Format detects automatically
4088 if the running scheme system implements floating point numbers and
4091 format:symbol-case-conv
4092 Symbols are converted by `symbol->string' so the case type of the
4093 printed symbols is implementation dependent.
4094 `format:symbol-case-conv' is a one arg closure which is either
4095 `#f' (no conversion), `string-upcase', `string-downcase' or
4096 `string-capitalize'. (default `#f')
4098 format:iobj-case-conv
4099 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4100 implementation internal objects. (default `#f')
4103 The character prefixing the exponent value in `~E' printing.
4106 *** Compatibility With Other Format Implementations
4112 Downward compatible except for padding support and `~A', `~S',
4113 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4114 `printf' padding support which is completely replaced by the CL
4115 `format' padding style.
4118 Downward compatible except for `~', which is not documented
4119 (ignores all characters inside the format string up to a newline
4120 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4121 numerical and variable parameters and `:/@' modifiers in the CL
4125 Downward compatible except for `~A' and `~S' which print in
4126 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4127 directive parameters or modifiers)).
4130 Downward compatible except for an optional destination parameter:
4131 S2C accepts a format call without a destination which returns a
4132 formatted string. This is equivalent to a #f destination in S2C.
4133 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4134 parameters or modifiers)).
4137 ** Changes to string-handling functions.
4139 These functions were added to support the (ice-9 format) module, above.
4141 *** New function: string-upcase STRING
4142 *** New function: string-downcase STRING
4144 These are non-destructive versions of the existing string-upcase! and
4145 string-downcase! functions.
4147 *** New function: string-capitalize! STRING
4148 *** New function: string-capitalize STRING
4150 These functions convert the first letter of each word in the string to
4153 (string-capitalize "howdy there")
4156 As with the other functions, string-capitalize! modifies the string in
4157 place, while string-capitalize returns a modified copy of its argument.
4159 *** New function: string-ci->symbol STRING
4161 Return a symbol whose name is STRING, but having the same case as if
4162 the symbol had be read by `read'.
4164 Guile can be configured to be sensitive or insensitive to case
4165 differences in Scheme identifiers. If Guile is case-insensitive, all
4166 symbols are converted to lower case on input. The `string-ci->symbol'
4167 function returns a symbol whose name in STRING, transformed as Guile
4168 would if STRING were input.
4170 *** New function: substring-move! STRING1 START END STRING2 START
4172 Copy the substring of STRING1 from START (inclusive) to END
4173 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4174 string, and the source and destination areas may overlap; in all
4175 cases, the function behaves as if all the characters were copied
4178 *** Extended functions: substring-move-left! substring-move-right!
4180 These functions now correctly copy arbitrarily overlapping substrings;
4181 they are both synonyms for substring-move!.
4184 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4186 getopt-long is a function for parsing command-line arguments in a
4187 manner consistent with other GNU programs.
4189 (getopt-long ARGS GRAMMAR)
4190 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4192 ARGS should be a list of strings. Its first element should be the
4193 name of the program; subsequent elements should be the arguments
4194 that were passed to the program on the command line. The
4195 `program-arguments' procedure returns a list of this form.
4197 GRAMMAR is a list of the form:
4198 ((OPTION (PROPERTY VALUE) ...) ...)
4200 Each OPTION should be a symbol. `getopt-long' will accept a
4201 command-line option named `--OPTION'.
4202 Each option can have the following (PROPERTY VALUE) pairs:
4204 (single-char CHAR) --- Accept `-CHAR' as a single-character
4205 equivalent to `--OPTION'. This is how to specify traditional
4207 (required? BOOL) --- If BOOL is true, the option is required.
4208 getopt-long will raise an error if it is not found in ARGS.
4209 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4210 it is #f, it does not; and if it is the symbol
4211 `optional', the option may appear in ARGS with or
4213 (predicate FUNC) --- If the option accepts a value (i.e. you
4214 specified `(value #t)' for this option), then getopt
4215 will apply FUNC to the value, and throw an exception
4216 if it returns #f. FUNC should be a procedure which
4217 accepts a string and returns a boolean value; you may
4218 need to use quasiquotes to get it into GRAMMAR.
4220 The (PROPERTY VALUE) pairs may occur in any order, but each
4221 property may occur only once. By default, options do not have
4222 single-character equivalents, are not required, and do not take
4225 In ARGS, single-character options may be combined, in the usual
4226 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4227 accepts values, then it must be the last option in the
4228 combination; the value is the next argument. So, for example, using
4229 the following grammar:
4230 ((apples (single-char #\a))
4231 (blimps (single-char #\b) (value #t))
4232 (catalexis (single-char #\c) (value #t)))
4233 the following argument lists would be acceptable:
4234 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4235 for "blimps" and "catalexis")
4236 ("-ab" "bang" "-c" "couth") (same)
4237 ("-ac" "couth" "-b" "bang") (same)
4238 ("-abc" "couth" "bang") (an error, since `-b' is not the
4239 last option in its combination)
4241 If an option's value is optional, then `getopt-long' decides
4242 whether it has a value by looking at what follows it in ARGS. If
4243 the next element is a string, and it does not appear to be an
4244 option itself, then that string is the option's value.
4246 The value of a long option can appear as the next element in ARGS,
4247 or it can follow the option name, separated by an `=' character.
4248 Thus, using the same grammar as above, the following argument lists
4250 ("--apples" "Braeburn" "--blimps" "Goodyear")
4251 ("--apples=Braeburn" "--blimps" "Goodyear")
4252 ("--blimps" "Goodyear" "--apples=Braeburn")
4254 If the option "--" appears in ARGS, argument parsing stops there;
4255 subsequent arguments are returned as ordinary arguments, even if
4256 they resemble options. So, in the argument list:
4257 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4258 `getopt-long' will recognize the `apples' option as having the
4259 value "Granny Smith", but it will not recognize the `blimp'
4260 option; it will return the strings "--blimp" and "Goodyear" as
4261 ordinary argument strings.
4263 The `getopt-long' function returns the parsed argument list as an
4264 assocation list, mapping option names --- the symbols from GRAMMAR
4265 --- onto their values, or #t if the option does not accept a value.
4266 Unused options do not appear in the alist.
4268 All arguments that are not the value of any option are returned
4269 as a list, associated with the empty list.
4271 `getopt-long' throws an exception if:
4272 - it finds an unrecognized option in ARGS
4273 - a required option is omitted
4274 - an option that requires an argument doesn't get one
4275 - an option that doesn't accept an argument does get one (this can
4276 only happen using the long option `--opt=value' syntax)
4277 - an option predicate fails
4282 `((lockfile-dir (required? #t)
4285 (predicate ,file-is-directory?))
4286 (verbose (required? #f)
4289 (x-includes (single-char #\x))
4290 (rnet-server (single-char #\y)
4291 (predicate ,string?))))
4293 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4294 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4296 => ((() "foo1" "-fred" "foo2" "foo3")
4297 (rnet-server . "lamprod")
4298 (x-includes . "/usr/include")
4299 (lockfile-dir . "/tmp")
4302 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4304 It will be removed in a few releases.
4306 ** New syntax: lambda*
4307 ** New syntax: define*
4308 ** New syntax: define*-public
4309 ** New syntax: defmacro*
4310 ** New syntax: defmacro*-public
4311 Guile now supports optional arguments.
4313 `lambda*', `define*', `define*-public', `defmacro*' and
4314 `defmacro*-public' are identical to the non-* versions except that
4315 they use an extended type of parameter list that has the following BNF
4316 syntax (parentheses are literal, square brackets indicate grouping,
4317 and `*', `+' and `?' have the usual meaning):
4319 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4320 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4321 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4323 ext-var-decl ::= identifier | ( identifier expression )
4325 The semantics are best illustrated with the following documentation
4326 and examples for `lambda*':
4329 lambda extended for optional and keyword arguments
4331 lambda* creates a procedure that takes optional arguments. These
4332 are specified by putting them inside brackets at the end of the
4333 paramater list, but before any dotted rest argument. For example,
4334 (lambda* (a b #&optional c d . e) '())
4335 creates a procedure with fixed arguments a and b, optional arguments c
4336 and d, and rest argument e. If the optional arguments are omitted
4337 in a call, the variables for them are unbound in the procedure. This
4338 can be checked with the bound? macro.
4340 lambda* can also take keyword arguments. For example, a procedure
4342 (lambda* (#&key xyzzy larch) '())
4343 can be called with any of the argument lists (#:xyzzy 11)
4344 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4345 are given as keywords are bound to values.
4347 Optional and keyword arguments can also be given default values
4348 which they take on when they are not present in a call, by giving a
4349 two-item list in place of an optional argument, for example in:
4350 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4351 foo is a fixed argument, bar is an optional argument with default
4352 value 42, and baz is a keyword argument with default value 73.
4353 Default value expressions are not evaluated unless they are needed
4354 and until the procedure is called.
4356 lambda* now supports two more special parameter list keywords.
4358 lambda*-defined procedures now throw an error by default if a
4359 keyword other than one of those specified is found in the actual
4360 passed arguments. However, specifying #&allow-other-keys
4361 immediately after the kyword argument declarations restores the
4362 previous behavior of ignoring unknown keywords. lambda* also now
4363 guarantees that if the same keyword is passed more than once, the
4364 last one passed is the one that takes effect. For example,
4365 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4366 #:heads 37 #:tails 42 #:heads 99)
4367 would result in (99 47) being displayed.
4369 #&rest is also now provided as a synonym for the dotted syntax rest
4370 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4371 all respects to lambda*. This is provided for more similarity to DSSSL,
4372 MIT-Scheme and Kawa among others, as well as for refugees from other
4375 Further documentation may be found in the optargs.scm file itself.
4377 The optional argument module also exports the macros `let-optional',
4378 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4379 are not documented here because they may be removed in the future, but
4380 full documentation is still available in optargs.scm.
4382 ** New syntax: and-let*
4383 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4385 Syntax: (land* (<clause> ...) <body> ...)
4386 Each <clause> should have one of the following forms:
4387 (<variable> <expression>)
4390 Each <variable> or <bound-variable> should be an identifier. Each
4391 <expression> should be a valid expression. The <body> should be a
4392 possibly empty sequence of expressions, like the <body> of a
4395 Semantics: A LAND* expression is evaluated by evaluating the
4396 <expression> or <bound-variable> of each of the <clause>s from
4397 left to right. The value of the first <expression> or
4398 <bound-variable> that evaluates to a false value is returned; the
4399 remaining <expression>s and <bound-variable>s are not evaluated.
4400 The <body> forms are evaluated iff all the <expression>s and
4401 <bound-variable>s evaluate to true values.
4403 The <expression>s and the <body> are evaluated in an environment
4404 binding each <variable> of the preceding (<variable> <expression>)
4405 clauses to the value of the <expression>. Later bindings
4406 shadow earlier bindings.
4408 Guile's and-let* macro was contributed by Michael Livshin.
4410 ** New sorting functions
4412 *** New function: sorted? SEQUENCE LESS?
4413 Returns `#t' when the sequence argument is in non-decreasing order
4414 according to LESS? (that is, there is no adjacent pair `... x y
4415 ...' for which `(less? y x)').
4417 Returns `#f' when the sequence contains at least one out-of-order
4418 pair. It is an error if the sequence is neither a list nor a
4421 *** New function: merge LIST1 LIST2 LESS?
4422 LIST1 and LIST2 are sorted lists.
4423 Returns the sorted list of all elements in LIST1 and LIST2.
4425 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4426 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4427 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4428 (Here "<" should read "comes before".)
4430 *** New procedure: merge! LIST1 LIST2 LESS?
4431 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4432 the result. If the code is compiled, and LESS? constructs no new
4433 pairs, no pairs at all will be allocated. The first pair of the
4434 result will be either the first pair of LIST1 or the first pair of
4437 *** New function: sort SEQUENCE LESS?
4438 Accepts either a list or a vector, and returns a new sequence
4439 which is sorted. The new sequence is the same type as the input.
4440 Always `(sorted? (sort sequence less?) less?)'. The original
4441 sequence is not altered in any way. The new sequence shares its
4442 elements with the old one; no elements are copied.
4444 *** New procedure: sort! SEQUENCE LESS
4445 Returns its sorted result in the original boxes. No new storage is
4446 allocated at all. Proper usage: (set! slist (sort! slist <))
4448 *** New function: stable-sort SEQUENCE LESS?
4449 Similar to `sort' but stable. That is, if "equal" elements are
4450 ordered a < b in the original sequence, they will have the same order
4453 *** New function: stable-sort! SEQUENCE LESS?
4454 Similar to `sort!' but stable.
4455 Uses temporary storage when sorting vectors.
4457 *** New functions: sort-list, sort-list!
4458 Added for compatibility with scsh.
4460 ** New built-in random number support
4462 *** New function: random N [STATE]
4463 Accepts a positive integer or real N and returns a number of the
4464 same type between zero (inclusive) and N (exclusive). The values
4465 returned have a uniform distribution.
4467 The optional argument STATE must be of the type produced by
4468 `copy-random-state' or `seed->random-state'. It defaults to the value
4469 of the variable `*random-state*'. This object is used to maintain the
4470 state of the pseudo-random-number generator and is altered as a side
4471 effect of the `random' operation.
4473 *** New variable: *random-state*
4474 Holds a data structure that encodes the internal state of the
4475 random-number generator that `random' uses by default. The nature
4476 of this data structure is implementation-dependent. It may be
4477 printed out and successfully read back in, but may or may not
4478 function correctly as a random-number state object in another
4481 *** New function: copy-random-state [STATE]
4482 Returns a new object of type suitable for use as the value of the
4483 variable `*random-state*' and as a second argument to `random'.
4484 If argument STATE is given, a copy of it is returned. Otherwise a
4485 copy of `*random-state*' is returned.
4487 *** New function: seed->random-state SEED
4488 Returns a new object of type suitable for use as the value of the
4489 variable `*random-state*' and as a second argument to `random'.
4490 SEED is a string or a number. A new state is generated and
4491 initialized using SEED.
4493 *** New function: random:uniform [STATE]
4494 Returns an uniformly distributed inexact real random number in the
4495 range between 0 and 1.
4497 *** New procedure: random:solid-sphere! VECT [STATE]
4498 Fills VECT with inexact real random numbers the sum of whose
4499 squares is less than 1.0. Thinking of VECT as coordinates in
4500 space of dimension N = `(vector-length VECT)', the coordinates are
4501 uniformly distributed within the unit N-shere. The sum of the
4502 squares of the numbers is returned. VECT can be either a vector
4503 or a uniform vector of doubles.
4505 *** New procedure: random:hollow-sphere! VECT [STATE]
4506 Fills VECT with inexact real random numbers the sum of whose squares
4507 is equal to 1.0. Thinking of VECT as coordinates in space of
4508 dimension n = `(vector-length VECT)', the coordinates are uniformly
4509 distributed over the surface of the unit n-shere. VECT can be either
4510 a vector or a uniform vector of doubles.
4512 *** New function: random:normal [STATE]
4513 Returns an inexact real in a normal distribution with mean 0 and
4514 standard deviation 1. For a normal distribution with mean M and
4515 standard deviation D use `(+ M (* D (random:normal)))'.
4517 *** New procedure: random:normal-vector! VECT [STATE]
4518 Fills VECT with inexact real random numbers which are independent and
4519 standard normally distributed (i.e., with mean 0 and variance 1).
4520 VECT can be either a vector or a uniform vector of doubles.
4522 *** New function: random:exp STATE
4523 Returns an inexact real in an exponential distribution with mean 1.
4524 For an exponential distribution with mean U use (* U (random:exp)).
4526 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4528 These functions now operate on numbers in the range of a C unsigned
4531 These functions used to operate on numbers in the range of a C signed
4532 long; however, this seems inappropriate, because Guile integers don't
4535 ** New function: make-guardian
4536 This is an implementation of guardians as described in
4537 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4538 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4539 Programming Language Design and Implementation, June 1993
4540 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4542 ** New functions: delq1!, delv1!, delete1!
4543 These procedures behave similar to delq! and friends but delete only
4544 one object if at all.
4546 ** New function: unread-string STRING PORT
4547 Unread STRING to PORT, that is, push it back onto the port so that
4548 next read operation will work on the pushed back characters.
4550 ** unread-char can now be called multiple times
4551 If unread-char is called multiple times, the unread characters will be
4552 read again in last-in first-out order.
4554 ** the procedures uniform-array-read! and uniform-array-write! now
4555 work on any kind of port, not just ports which are open on a file.
4557 ** Now 'l' in a port mode requests line buffering.
4559 ** The procedure truncate-file now works on string ports as well
4560 as file ports. If the size argument is omitted, the current
4561 file position is used.
4563 ** new procedure: seek PORT/FDES OFFSET WHENCE
4564 The arguments are the same as for the old fseek procedure, but it
4565 works on string ports as well as random-access file ports.
4567 ** the fseek procedure now works on string ports, since it has been
4568 redefined using seek.
4570 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4571 size is not supplied.
4573 ** the newline procedure no longer flushes the port if it's not
4574 line-buffered: previously it did if it was the current output port.
4576 ** open-pipe and close-pipe are no longer primitive procedures, but
4577 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4579 ** the freopen procedure has been removed.
4581 ** new procedure: drain-input PORT
4582 Drains PORT's read buffers (including any pushed-back characters)
4583 and returns the contents as a single string.
4585 ** New function: map-in-order PROC LIST1 LIST2 ...
4586 Version of `map' which guarantees that the procedure is applied to the
4587 lists in serial order.
4589 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4590 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4591 now obsolete and will go away in release 1.5.
4593 ** New syntax: collect BODY1 ...
4594 Version of `begin' which returns a list of the results of the body
4595 forms instead of the result of the last body form. In contrast to
4596 `begin', `collect' allows an empty body.
4598 ** New functions: read-history FILENAME, write-history FILENAME
4599 Read/write command line history from/to file. Returns #t on success
4600 and #f if an error occured.
4602 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4604 These procedures return a list of definitions available in the specified
4605 argument, a relative module reference. In the case of no argument,
4606 `(current-module)' is now consulted for definitions to return, instead
4607 of simply returning #f, the former behavior.
4609 ** The #/ syntax for lists is no longer supported.
4611 Earlier versions of Scheme accepted this syntax, but printed a
4614 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4616 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4619 * Changes to the gh_ interface
4623 Now takes a second argument which is the result array. If this
4624 pointer is NULL, a new array is malloced (the old behaviour).
4626 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4627 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4631 * Changes to the scm_ interface
4633 ** Function: scm_make_named_hook (char* name, int n_args)
4635 Creates a hook in the same way as make-hook above but also
4636 binds a variable named NAME to it.
4638 This is the typical way of creating a hook from C code.
4640 Currently, the variable is created in the "current" module. This
4641 might change when we get the new module system.
4643 ** The smob interface
4645 The interface for creating smobs has changed. For documentation, see
4646 data-rep.info (made from guile-core/doc/data-rep.texi).
4648 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4650 >>> This function will be removed in 1.3.4. <<<
4654 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4655 This function adds a new smob type, named NAME, with instance size
4656 SIZE to the system. The return value is a tag that is used in
4657 creating instances of the type. If SIZE is 0, then no memory will
4658 be allocated when instances of the smob are created, and nothing
4659 will be freed by the default free function.
4661 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4662 This function sets the smob marking 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_free (long tc, SCM (*mark) (SCM))
4667 This function sets the smob freeing procedure for the smob type
4668 specified by the tag TC. TC is the tag returned by
4669 `scm_make_smob_type'.
4671 *** Function: void scm_set_smob_print (tc, print)
4673 - Function: void scm_set_smob_print (long tc,
4674 scm_sizet (*print) (SCM,
4678 This function sets the smob printing procedure for the smob type
4679 specified by the tag TC. TC is the tag returned by
4680 `scm_make_smob_type'.
4682 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4683 This function sets the smob equality-testing predicate for the
4684 smob type specified by the tag TC. TC is the tag returned by
4685 `scm_make_smob_type'.
4687 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4688 Make VALUE contain a smob instance of the type with type code TC and
4689 smob data DATA. VALUE must be previously declared as C type `SCM'.
4691 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4692 This macro expands to a block of code that creates a smob instance
4693 of the type with type code TC and smob data DATA, and returns that
4694 `SCM' value. It should be the last piece of code in a block.
4696 ** The interfaces for using I/O ports and implementing port types
4697 (ptobs) have changed significantly. The new interface is based on
4698 shared access to buffers and a new set of ptob procedures.
4700 *** scm_newptob has been removed
4704 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4706 - Function: SCM scm_make_port_type (char *type_name,
4707 int (*fill_buffer) (SCM port),
4708 void (*write_flush) (SCM port));
4710 Similarly to the new smob interface, there is a set of function
4711 setters by which the user can customize the behaviour of his port
4712 type. See ports.h (scm_set_port_XXX).
4714 ** scm_strport_to_string: New function: creates a new string from
4715 a string port's buffer.
4717 ** Plug in interface for random number generators
4718 The variable `scm_the_rng' in random.c contains a value and three
4719 function pointers which together define the current random number
4720 generator being used by the Scheme level interface and the random
4721 number library functions.
4723 The user is free to replace the default generator with the generator
4726 *** Variable: size_t scm_the_rng.rstate_size
4727 The size of the random state type used by the current RNG
4730 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4731 Given the random STATE, return 32 random bits.
4733 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4734 Seed random state STATE using string S of length N.
4736 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4737 Given random state STATE, return a malloced copy.
4740 The default RNG is the MWC (Multiply With Carry) random number
4741 generator described by George Marsaglia at the Department of
4742 Statistics and Supercomputer Computations Research Institute, The
4743 Florida State University (http://stat.fsu.edu/~geo).
4745 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4746 passes all tests in the DIEHARD test suite
4747 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4748 costs one multiply and one add on platforms which either supports long
4749 longs (gcc does this on most systems) or have 64 bit longs. The cost
4750 is four multiply on other systems but this can be optimized by writing
4751 scm_i_uniform32 in assembler.
4753 These functions are provided through the scm_the_rng interface for use
4754 by libguile and the application.
4756 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4757 Given the random STATE, return 32 random bits.
4758 Don't use this function directly. Instead go through the plugin
4759 interface (see "Plug in interface" above).
4761 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4762 Initialize STATE using SEED of length N.
4764 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4765 Return a malloc:ed copy of STATE. This function can easily be re-used
4766 in the interfaces to other RNGs.
4768 ** Random number library functions
4769 These functions use the current RNG through the scm_the_rng interface.
4770 It might be a good idea to use these functions from your C code so
4771 that only one random generator is used by all code in your program.
4773 The default random state is stored in:
4775 *** Variable: SCM scm_var_random_state
4776 Contains the vcell of the Scheme variable "*random-state*" which is
4777 used as default state by all random number functions in the Scheme
4782 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4784 *** Function: scm_rstate *scm_c_default_rstate (void)
4785 This is a convenience function which returns the value of
4786 scm_var_random_state. An error message is generated if this value
4787 isn't a random state.
4789 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4790 Make a new random state from the string SEED of length LENGTH.
4792 It is generally not a good idea to use multiple random states in a
4793 program. While subsequent random numbers generated from one random
4794 state are guaranteed to be reasonably independent, there is no such
4795 guarantee for numbers generated from different random states.
4797 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4798 Return 32 random bits.
4800 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4801 Return a sample from the uniform(0,1) distribution.
4803 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4804 Return a sample from the normal(0,1) distribution.
4806 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4807 Return a sample from the exp(1) distribution.
4809 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4810 Return a sample from the discrete uniform(0,M) distribution.
4812 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4813 Return a sample from the discrete uniform(0,M) distribution.
4814 M must be a bignum object. The returned value may be an INUM.
4818 Changes in Guile 1.3 (released Monday, October 19, 1998):
4820 * Changes to the distribution
4822 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4823 To avoid conflicts, programs should name environment variables after
4824 themselves, except when there's a common practice establishing some
4827 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4828 giving the former precedence, and printing a warning message if the
4829 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4831 ** The header files related to multi-byte characters have been removed.
4832 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4833 which referred to these explicitly will probably need to be rewritten,
4834 since the support for the variant string types has been removed; see
4837 ** The header files append.h and sequences.h have been removed. These
4838 files implemented non-R4RS operations which would encourage
4839 non-portable programming style and less easy-to-read code.
4841 * Changes to the stand-alone interpreter
4843 ** New procedures have been added to implement a "batch mode":
4845 *** Function: batch-mode?
4847 Returns a boolean indicating whether the interpreter is in batch
4850 *** Function: set-batch-mode?! ARG
4852 If ARG is true, switches the interpreter to batch mode. The `#f'
4853 case has not been implemented.
4855 ** Guile now provides full command-line editing, when run interactively.
4856 To use this feature, you must have the readline library installed.
4857 The Guile build process will notice it, and automatically include
4860 The readline library is available via anonymous FTP from any GNU
4861 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4863 ** the-last-stack is now a fluid.
4865 * Changes to the procedure for linking libguile with your programs
4867 ** You can now use the `guile-config' utility to build programs that use Guile.
4869 Guile now includes a command-line utility called `guile-config', which
4870 can provide information about how to compile and link programs that
4873 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4874 You should include this command's output on the command line you use
4875 to compile C or C++ code that #includes the Guile header files. It's
4876 usually just a `-I' flag to help the compiler find the Guile headers.
4879 *** `guile-config link' prints any linker flags necessary to link with Guile.
4881 This command writes to its standard output a list of flags which you
4882 must pass to the linker to link your code against the Guile library.
4883 The flags include '-lguile' itself, any other libraries the Guile
4884 library depends upon, and any `-L' flags needed to help the linker
4885 find those libraries.
4887 For example, here is a Makefile rule that builds a program named 'foo'
4888 from the object files ${FOO_OBJECTS}, and links them against Guile:
4891 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4893 Previous Guile releases recommended that you use autoconf to detect
4894 which of a predefined set of libraries were present on your system.
4895 It is more robust to use `guile-config', since it records exactly which
4896 libraries the installed Guile library requires.
4898 This was originally called `build-guile', but was renamed to
4899 `guile-config' before Guile 1.3 was released, to be consistent with
4900 the analogous script for the GTK+ GUI toolkit, which is called
4904 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4906 If you are using the GNU autoconf package to configure your program,
4907 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4908 (described above) and gather the necessary values for use in your
4911 The GUILE_FLAGS macro expands to configure script code which runs the
4912 `guile-config' script, to find out where Guile's header files and
4913 libraries are installed. It sets two variables, marked for
4914 substitution, as by AC_SUBST.
4916 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4917 code that uses Guile header files. This is almost always just a
4920 GUILE_LDFLAGS --- flags to pass to the linker to link a
4921 program against Guile. This includes `-lguile' for the Guile
4922 library itself, any libraries that Guile itself requires (like
4923 -lqthreads), and so on. It may also include a -L flag to tell the
4924 compiler where to find the libraries.
4926 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4927 directory of the Guile distribution. You can copy it into your
4928 package's aclocal.m4 file, and then use it in your configure.in file.
4930 If you are using the `aclocal' program, distributed with GNU automake,
4931 to maintain your aclocal.m4 file, the Guile installation process
4932 installs guile.m4 where aclocal will find it. All you need to do is
4933 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4934 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4938 * Changes to Scheme functions and syntax
4940 ** Multi-byte strings have been removed, as have multi-byte and wide
4941 ports. We felt that these were the wrong approach to
4942 internationalization support.
4944 ** New function: readline [PROMPT]
4945 Read a line from the terminal, and allow the user to edit it,
4946 prompting with PROMPT. READLINE provides a large set of Emacs-like
4947 editing commands, lets the user recall previously typed lines, and
4948 works on almost every kind of terminal, including dumb terminals.
4950 READLINE assumes that the cursor is at the beginning of the line when
4951 it is invoked. Thus, you can't print a prompt yourself, and then call
4952 READLINE; you need to package up your prompt as a string, pass it to
4953 the function, and let READLINE print the prompt itself. This is
4954 because READLINE needs to know the prompt's screen width.
4956 For Guile to provide this function, you must have the readline
4957 library, version 2.1 or later, installed on your system. Readline is
4958 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4959 any GNU mirror site.
4961 See also ADD-HISTORY function.
4963 ** New function: add-history STRING
4964 Add STRING as the most recent line in the history used by the READLINE
4965 command. READLINE does not add lines to the history itself; you must
4966 call ADD-HISTORY to make previous input available to the user.
4968 ** The behavior of the read-line function has changed.
4970 This function now uses standard C library functions to read the line,
4971 for speed. This means that it doesn not respect the value of
4972 scm-line-incrementors; it assumes that lines are delimited with
4975 (Note that this is read-line, the function that reads a line of text
4976 from a port, not readline, the function that reads a line from a
4977 terminal, providing full editing capabilities.)
4979 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4981 This module provides some simple argument parsing. It exports one
4984 Function: getopt-gnu-style ARG-LS
4985 Parse a list of program arguments into an alist of option
4988 Each item in the list of program arguments is examined to see if
4989 it meets the syntax of a GNU long-named option. An argument like
4990 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4991 returned alist, where MUMBLE is a keyword object with the same
4992 name as the argument. An argument like `--MUMBLE=FROB' produces
4993 an element of the form (MUMBLE . FROB), where FROB is a string.
4995 As a special case, the returned alist also contains a pair whose
4996 car is the symbol `rest'. The cdr of this pair is a list
4997 containing all the items in the argument list that are not options
4998 of the form mentioned above.
5000 The argument `--' is treated specially: all items in the argument
5001 list appearing after such an argument are not examined, and are
5002 returned in the special `rest' list.
5004 This function does not parse normal single-character switches.
5005 You will need to parse them out of the `rest' list yourself.
5007 ** The read syntax for byte vectors and short vectors has changed.
5009 Instead of #bytes(...), write #y(...).
5011 Instead of #short(...), write #h(...).
5013 This may seem nutty, but, like the other uniform vectors, byte vectors
5014 and short vectors want to have the same print and read syntax (and,
5015 more basic, want to have read syntax!). Changing the read syntax to
5016 use multiple characters after the hash sign breaks with the
5017 conventions used in R5RS and the conventions used for the other
5018 uniform vectors. It also introduces complexity in the current reader,
5019 both on the C and Scheme levels. (The Right solution is probably to
5020 change the syntax and prototypes for uniform vectors entirely.)
5023 ** The new module (ice-9 session) provides useful interactive functions.
5025 *** New procedure: (apropos REGEXP OPTION ...)
5027 Display a list of top-level variables whose names match REGEXP, and
5028 the modules they are imported from. Each OPTION should be one of the
5031 value --- Show the value of each matching variable.
5032 shadow --- Show bindings shadowed by subsequently imported modules.
5033 full --- Same as both `shadow' and `value'.
5037 guile> (apropos "trace" 'full)
5038 debug: trace #<procedure trace args>
5039 debug: untrace #<procedure untrace args>
5040 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5041 the-scm-module: before-backtrace-hook ()
5042 the-scm-module: backtrace #<primitive-procedure backtrace>
5043 the-scm-module: after-backtrace-hook ()
5044 the-scm-module: has-shown-backtrace-hint? #f
5047 ** There are new functions and syntax for working with macros.
5049 Guile implements macros as a special object type. Any variable whose
5050 top-level binding is a macro object acts as a macro. The macro object
5051 specifies how the expression should be transformed before evaluation.
5053 *** Macro objects now print in a reasonable way, resembling procedures.
5055 *** New function: (macro? OBJ)
5056 True iff OBJ is a macro object.
5058 *** New function: (primitive-macro? OBJ)
5059 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5060 macro transformers, implemented in eval.c rather than Scheme code.
5062 Why do we have this function?
5063 - For symmetry with procedure? and primitive-procedure?,
5064 - to allow custom print procedures to tell whether a macro is
5065 primitive, and display it differently, and
5066 - to allow compilers and user-written evaluators to distinguish
5067 builtin special forms from user-defined ones, which could be
5070 *** New function: (macro-type OBJ)
5071 Return a value indicating what kind of macro OBJ is. Possible return
5074 The symbol `syntax' --- a macro created by procedure->syntax.
5075 The symbol `macro' --- a macro created by procedure->macro.
5076 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5077 The boolean #f --- if OBJ is not a macro object.
5079 *** New function: (macro-name MACRO)
5080 Return the name of the macro object MACRO's procedure, as returned by
5083 *** New function: (macro-transformer MACRO)
5084 Return the transformer procedure for MACRO.
5086 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5088 Specify a new macro expander to use in the current module. Each
5089 MODULE is a module name, with the same meaning as in the `use-modules'
5090 form; each named module's exported bindings are added to the current
5091 top-level environment. TRANSFORMER is an expression evaluated in the
5092 resulting environment which must yield a procedure to use as the
5093 module's eval transformer: every expression evaluated in this module
5094 is passed to this function, and the result passed to the Guile
5097 *** macro-eval! is removed. Use local-eval instead.
5099 ** Some magic has been added to the printer to better handle user
5100 written printing routines (like record printers, closure printers).
5102 The problem is that these user written routines must have access to
5103 the current `print-state' to be able to handle fancy things like
5104 detection of circular references. These print-states have to be
5105 passed to the builtin printing routines (display, write, etc) to
5106 properly continue the print chain.
5108 We didn't want to change all existing print code so that it
5109 explicitly passes thru a print state in addition to a port. Instead,
5110 we extented the possible values that the builtin printing routines
5111 accept as a `port'. In addition to a normal port, they now also take
5112 a pair of a normal port and a print-state. Printing will go to the
5113 port and the print-state will be used to control the detection of
5114 circular references, etc. If the builtin function does not care for a
5115 print-state, it is simply ignored.
5117 User written callbacks are now called with such a pair as their
5118 `port', but because every function now accepts this pair as a PORT
5119 argument, you don't have to worry about that. In fact, it is probably
5120 safest to not check for these pairs.
5122 However, it is sometimes necessary to continue a print chain on a
5123 different port, for example to get a intermediate string
5124 representation of the printed value, mangle that string somehow, and
5125 then to finally print the mangled string. Use the new function
5127 inherit-print-state OLD-PORT NEW-PORT
5129 for this. It constructs a new `port' that prints to NEW-PORT but
5130 inherits the print-state of OLD-PORT.
5132 ** struct-vtable-offset renamed to vtable-offset-user
5134 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5136 ** There is now a third optional argument to make-vtable-vtable
5137 (and fourth to make-struct) when constructing new types (vtables).
5138 This argument initializes field vtable-index-printer of the vtable.
5140 ** The detection of circular references has been extended to structs.
5141 That is, a structure that -- in the process of being printed -- prints
5142 itself does not lead to infinite recursion.
5144 ** There is now some basic support for fluids. Please read
5145 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5146 the following functions and macros:
5148 Function: make-fluid
5150 Create a new fluid object. Fluids are not special variables or
5151 some other extension to the semantics of Scheme, but rather
5152 ordinary Scheme objects. You can store them into variables (that
5153 are still lexically scoped, of course) or into any other place you
5154 like. Every fluid has a initial value of `#f'.
5156 Function: fluid? OBJ
5158 Test whether OBJ is a fluid.
5160 Function: fluid-ref FLUID
5161 Function: fluid-set! FLUID VAL
5163 Access/modify the fluid FLUID. Modifications are only visible
5164 within the current dynamic root (that includes threads).
5166 Function: with-fluids* FLUIDS VALUES THUNK
5168 FLUIDS is a list of fluids and VALUES a corresponding list of
5169 values for these fluids. Before THUNK gets called the values are
5170 installed in the fluids and the old values of the fluids are
5171 saved in the VALUES list. When the flow of control leaves THUNK
5172 or reenters it, the values get swapped again. You might think of
5173 this as a `safe-fluid-excursion'. Note that the VALUES list is
5174 modified by `with-fluids*'.
5176 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5178 The same as `with-fluids*' but with a different syntax. It looks
5179 just like `let', but both FLUID and VALUE are evaluated. Remember,
5180 fluids are not special variables but ordinary objects. FLUID
5181 should evaluate to a fluid.
5183 ** Changes to system call interfaces:
5185 *** close-port, close-input-port and close-output-port now return a
5186 boolean instead of an `unspecified' object. #t means that the port
5187 was successfully closed, while #f means it was already closed. It is
5188 also now possible for these procedures to raise an exception if an
5189 error occurs (some errors from write can be delayed until close.)
5191 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5194 *** the third argument to fcntl is now optional.
5196 *** the first argument to chown can now be a file descriptor or a port.
5198 *** the argument to stat can now be a port.
5200 *** The following new procedures have been added (most use scsh
5203 *** procedure: close PORT/FD
5204 Similar to close-port (*note close-port: Closing Ports.), but also
5205 works on file descriptors. A side effect of closing a file
5206 descriptor is that any ports using that file descriptor are moved
5207 to a different file descriptor and have their revealed counts set
5210 *** procedure: port->fdes PORT
5211 Returns the integer file descriptor underlying PORT. As a side
5212 effect the revealed count of PORT is incremented.
5214 *** procedure: fdes->ports FDES
5215 Returns a list of existing ports which have FDES as an underlying
5216 file descriptor, without changing their revealed counts.
5218 *** procedure: fdes->inport FDES
5219 Returns an existing input port which has FDES as its underlying
5220 file descriptor, if one exists, and increments its revealed count.
5221 Otherwise, returns a new input port with a revealed count of 1.
5223 *** procedure: fdes->outport FDES
5224 Returns an existing output port which has FDES as its underlying
5225 file descriptor, if one exists, and increments its revealed count.
5226 Otherwise, returns a new output port with a revealed count of 1.
5228 The next group of procedures perform a `dup2' system call, if NEWFD
5229 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5230 duplicated can be supplied as an integer or contained in a port. The
5231 type of value returned varies depending on which procedure is used.
5233 All procedures also have the side effect when performing `dup2' that
5234 any ports using NEWFD are moved to a different file descriptor and have
5235 their revealed counts set to zero.
5237 *** procedure: dup->fdes PORT/FD [NEWFD]
5238 Returns an integer file descriptor.
5240 *** procedure: dup->inport PORT/FD [NEWFD]
5241 Returns a new input port using the new file descriptor.
5243 *** procedure: dup->outport PORT/FD [NEWFD]
5244 Returns a new output port using the new file descriptor.
5246 *** procedure: dup PORT/FD [NEWFD]
5247 Returns a new port if PORT/FD is a port, with the same mode as the
5248 supplied port, otherwise returns an integer file descriptor.
5250 *** procedure: dup->port PORT/FD MODE [NEWFD]
5251 Returns a new port using the new file descriptor. MODE supplies a
5252 mode string for the port (*note open-file: File Ports.).
5254 *** procedure: setenv NAME VALUE
5255 Modifies the environment of the current process, which is also the
5256 default environment inherited by child processes.
5258 If VALUE is `#f', then NAME is removed from the environment.
5259 Otherwise, the string NAME=VALUE is added to the environment,
5260 replacing any existing string with name matching NAME.
5262 The return value is unspecified.
5264 *** procedure: truncate-file OBJ SIZE
5265 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5266 can be a string containing a file name or an integer file
5267 descriptor or port open for output on the file. The underlying
5268 system calls are `truncate' and `ftruncate'.
5270 The return value is unspecified.
5272 *** procedure: setvbuf PORT MODE [SIZE]
5273 Set the buffering mode for PORT. MODE can be:
5281 block buffered, using a newly allocated buffer of SIZE bytes.
5282 However if SIZE is zero or unspecified, the port will be made
5285 This procedure should not be used after I/O has been performed with
5288 Ports are usually block buffered by default, with a default buffer
5289 size. Procedures e.g., *Note open-file: File Ports, which accept a
5290 mode string allow `0' to be added to request an unbuffered port.
5292 *** procedure: fsync PORT/FD
5293 Copies any unwritten data for the specified output file descriptor
5294 to disk. If PORT/FD is a port, its buffer is flushed before the
5295 underlying file descriptor is fsync'd. The return value is
5298 *** procedure: open-fdes PATH FLAGS [MODES]
5299 Similar to `open' but returns a file descriptor instead of a port.
5301 *** procedure: execle PATH ENV [ARG] ...
5302 Similar to `execl', but the environment of the new process is
5303 specified by ENV, which must be a list of strings as returned by
5304 the `environ' procedure.
5306 This procedure is currently implemented using the `execve' system
5307 call, but we call it `execle' because of its Scheme calling
5310 *** procedure: strerror ERRNO
5311 Returns the Unix error message corresponding to ERRNO, an integer.
5313 *** procedure: primitive-exit [STATUS]
5314 Terminate the current process without unwinding the Scheme stack.
5315 This is would typically be useful after a fork. The exit status
5316 is STATUS if supplied, otherwise zero.
5318 *** procedure: times
5319 Returns an object with information about real and processor time.
5320 The following procedures accept such an object as an argument and
5321 return a selected component:
5324 The current real time, expressed as time units relative to an
5328 The CPU time units used by the calling process.
5331 The CPU time units used by the system on behalf of the
5335 The CPU time units used by terminated child processes of the
5336 calling process, whose status has been collected (e.g., using
5340 Similarly, the CPU times units used by the system on behalf of
5341 terminated child processes.
5343 ** Removed: list-length
5344 ** Removed: list-append, list-append!
5345 ** Removed: list-reverse, list-reverse!
5347 ** array-map renamed to array-map!
5349 ** serial-array-map renamed to serial-array-map!
5351 ** catch doesn't take #f as first argument any longer
5353 Previously, it was possible to pass #f instead of a key to `catch'.
5354 That would cause `catch' to pass a jump buffer object to the procedure
5355 passed as second argument. The procedure could then use this jump
5356 buffer objekt as an argument to throw.
5358 This mechanism has been removed since its utility doesn't motivate the
5359 extra complexity it introduces.
5361 ** The `#/' notation for lists now provokes a warning message from Guile.
5362 This syntax will be removed from Guile in the near future.
5364 To disable the warning message, set the GUILE_HUSH environment
5365 variable to any non-empty value.
5367 ** The newline character now prints as `#\newline', following the
5368 normal Scheme notation, not `#\nl'.
5370 * Changes to the gh_ interface
5372 ** The gh_enter function now takes care of loading the Guile startup files.
5373 gh_enter works by calling scm_boot_guile; see the remarks below.
5375 ** Function: void gh_write (SCM x)
5377 Write the printed representation of the scheme object x to the current
5378 output port. Corresponds to the scheme level `write'.
5380 ** gh_list_length renamed to gh_length.
5382 ** vector handling routines
5384 Several major changes. In particular, gh_vector() now resembles
5385 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5386 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5387 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5388 vector-related gh_ functions have been implemented.
5390 ** pair and list routines
5392 Implemented several of the R4RS pair and list functions that were
5395 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5397 New function. Converts double arrays back and forth between Scheme
5400 * Changes to the scm_ interface
5402 ** The function scm_boot_guile now takes care of loading the startup files.
5404 Guile's primary initialization function, scm_boot_guile, now takes
5405 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5406 Guile, define the module system, and put together some standard
5407 bindings. It also loads `init.scm', which is intended to hold
5408 site-specific initialization code.
5410 Since Guile cannot operate properly until boot-9.scm is loaded, there
5411 is no reason to separate loading boot-9.scm from Guile's other
5412 initialization processes.
5414 This job used to be done by scm_compile_shell_switches, which didn't
5415 make much sense; in particular, it meant that people using Guile for
5416 non-shell-like applications had to jump through hoops to get Guile
5417 initialized properly.
5419 ** The function scm_compile_shell_switches no longer loads the startup files.
5420 Now, Guile always loads the startup files, whenever it is initialized;
5421 see the notes above for scm_boot_guile and scm_load_startup_files.
5423 ** Function: scm_load_startup_files
5424 This new function takes care of loading Guile's initialization file
5425 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5426 this is always called by the Guile initialization process, it's
5427 probably not too useful to call this yourself, but it's there anyway.
5429 ** The semantics of smob marking have changed slightly.
5431 The smob marking function (the `mark' member of the scm_smobfuns
5432 structure) is no longer responsible for setting the mark bit on the
5433 smob. The generic smob handling code in the garbage collector will
5434 set this bit. The mark function need only ensure that any other
5435 objects the smob refers to get marked.
5437 Note that this change means that the smob's GC8MARK bit is typically
5438 already set upon entry to the mark function. Thus, marking functions
5439 which look like this:
5442 if (SCM_GC8MARKP (ptr))
5444 SCM_SETGC8MARK (ptr);
5445 ... mark objects to which the smob refers ...
5448 are now incorrect, since they will return early, and fail to mark any
5449 other objects the smob refers to. Some code in the Guile library used
5452 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5454 If you have implemented your own I/O port type, by writing the
5455 functions required by the scm_ptobfuns and then calling scm_newptob,
5456 you will need to change your functions slightly.
5458 The functions in a scm_ptobfuns structure now expect the port itself
5459 as their argument; they used to expect the `stream' member of the
5460 port's scm_port_table structure. This allows functions in an
5461 scm_ptobfuns structure to easily access the port's cell (and any flags
5462 it its CAR), and the port's scm_port_table structure.
5464 Guile now passes the I/O port itself as the `port' argument in the
5465 following scm_ptobfuns functions:
5467 int (*free) (SCM port);
5468 int (*fputc) (int, SCM port);
5469 int (*fputs) (char *, SCM port);
5470 scm_sizet (*fwrite) SCM_P ((char *ptr,
5474 int (*fflush) (SCM port);
5475 int (*fgetc) (SCM port);
5476 int (*fclose) (SCM port);
5478 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5481 If you have existing code which defines its own port types, it is easy
5482 to convert your code to the new interface; simply apply SCM_STREAM to
5483 the port argument to yield the value you code used to expect.
5485 Note that since both the port and the stream have the same type in the
5486 C code --- they are both SCM values --- the C compiler will not remind
5487 you if you forget to update your scm_ptobfuns functions.
5490 ** Function: int scm_internal_select (int fds,
5494 struct timeval *timeout);
5496 This is a replacement for the `select' function provided by the OS.
5497 It enables I/O blocking and sleeping to happen for one cooperative
5498 thread without blocking other threads. It also avoids busy-loops in
5499 these situations. It is intended that all I/O blocking and sleeping
5500 will finally go through this function. Currently, this function is
5501 only available on systems providing `gettimeofday' and `select'.
5503 ** Function: SCM scm_internal_stack_catch (SCM tag,
5504 scm_catch_body_t body,
5506 scm_catch_handler_t handler,
5509 A new sibling to the other two C level `catch' functions
5510 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5511 the stack to be saved automatically into the variable `the-last-stack'
5512 (scm_the_last_stack_var) on error. This is necessary if you want to
5513 use advanced error reporting, such as calling scm_display_error and
5514 scm_display_backtrace. (They both take a stack object as argument.)
5516 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5518 scm_catch_handler_t handler,
5521 Spawns a new thread. It does a job similar to
5522 scm_call_with_new_thread but takes arguments more suitable when
5523 spawning threads from application C code.
5525 ** The hook scm_error_callback has been removed. It was originally
5526 intended as a way for the user to install his own error handler. But
5527 that method works badly since it intervenes between throw and catch,
5528 thereby changing the semantics of expressions like (catch #t ...).
5529 The correct way to do it is to use one of the C level catch functions
5530 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5532 ** Removed functions:
5534 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5535 scm_list_reverse, scm_list_reverse_x
5537 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5539 These can be used for pretty list creation from C. The idea is taken
5540 from Erick Gallesio's STk.
5542 ** scm_array_map renamed to scm_array_map_x
5544 ** mbstrings are now removed
5546 This means that the type codes scm_tc7_mb_string and
5547 scm_tc7_mb_substring has been removed.
5549 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5551 Since we no longer support multi-byte strings, these I/O functions
5552 have been simplified, and renamed. Here are their old names, and
5553 their new names and arguments:
5555 scm_gen_putc -> void scm_putc (int c, SCM port);
5556 scm_gen_puts -> void scm_puts (char *s, SCM port);
5557 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5558 scm_gen_getc -> void scm_getc (SCM port);
5561 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5563 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5565 SCM_TYP7S now masks away the bit which distinguishes substrings from
5568 ** scm_catch_body_t: Backward incompatible change!
5570 Body functions to scm_internal_catch and friends do not any longer
5571 take a second argument. This is because it is no longer possible to
5572 pass a #f arg to catch.
5574 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5576 The function scm_protect_object protects its argument from being freed
5577 by the garbage collector. scm_unprotect_object removes that
5580 These functions now nest properly. That is, for every object O, there
5581 is a counter which scm_protect_object(O) increments and
5582 scm_unprotect_object(O) decrements, if the counter is greater than
5583 zero. Every object's counter is zero when it is first created. If an
5584 object's counter is greater than zero, the garbage collector will not
5585 reclaim its storage.
5587 This allows you to use scm_protect_object in your code without
5588 worrying that some other function you call will call
5589 scm_unprotect_object, and allow it to be freed. Assuming that the
5590 functions you call are well-behaved, and unprotect only those objects
5591 they protect, you can follow the same rule and have confidence that
5592 objects will be freed only at appropriate times.
5595 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5597 * Changes to the distribution
5599 ** Nightly snapshots are now available from ftp.red-bean.com.
5600 The old server, ftp.cyclic.com, has been relinquished to its rightful
5603 Nightly snapshots of the Guile development sources are now available via
5604 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5606 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5607 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5609 ** To run Guile without installing it, the procedure has changed a bit.
5611 If you used a separate build directory to compile Guile, you'll need
5612 to include the build directory in SCHEME_LOAD_PATH, as well as the
5613 source directory. See the `INSTALL' file for examples.
5615 * Changes to the procedure for linking libguile with your programs
5617 ** The standard Guile load path for Scheme code now includes
5618 $(datadir)/guile (usually /usr/local/share/guile). This means that
5619 you can install your own Scheme files there, and Guile will find them.
5620 (Previous versions of Guile only checked a directory whose name
5621 contained the Guile version number, so you had to re-install or move
5622 your Scheme sources each time you installed a fresh version of Guile.)
5624 The load path also includes $(datadir)/guile/site; we recommend
5625 putting individual Scheme files there. If you want to install a
5626 package with multiple source files, create a directory for them under
5629 ** Guile 1.2 will now use the Rx regular expression library, if it is
5630 installed on your system. When you are linking libguile into your own
5631 programs, this means you will have to link against -lguile, -lqt (if
5632 you configured Guile with thread support), and -lrx.
5634 If you are using autoconf to generate configuration scripts for your
5635 application, the following lines should suffice to add the appropriate
5636 libraries to your link command:
5638 ### Find Rx, quickthreads and libguile.
5639 AC_CHECK_LIB(rx, main)
5640 AC_CHECK_LIB(qt, main)
5641 AC_CHECK_LIB(guile, scm_shell)
5643 The Guile 1.2 distribution does not contain sources for the Rx
5644 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5645 retrieve it from a GNU FTP site and install it separately.
5647 * Changes to Scheme functions and syntax
5649 ** The dynamic linking features of Guile are now enabled by default.
5650 You can disable them by giving the `--disable-dynamic-linking' option
5653 (dynamic-link FILENAME)
5655 Find the object file denoted by FILENAME (a string) and link it
5656 into the running Guile application. When everything works out,
5657 return a Scheme object suitable for representing the linked object
5658 file. Otherwise an error is thrown. How object files are
5659 searched is system dependent.
5661 (dynamic-object? VAL)
5663 Determine whether VAL represents a dynamically linked object file.
5665 (dynamic-unlink DYNOBJ)
5667 Unlink the indicated object file from the application. DYNOBJ
5668 should be one of the values returned by `dynamic-link'.
5670 (dynamic-func FUNCTION DYNOBJ)
5672 Search the C function indicated by FUNCTION (a string or symbol)
5673 in DYNOBJ and return some Scheme object that can later be used
5674 with `dynamic-call' to actually call this function. Right now,
5675 these Scheme objects are formed by casting the address of the
5676 function to `long' and converting this number to its Scheme
5679 (dynamic-call FUNCTION DYNOBJ)
5681 Call the C function indicated by FUNCTION and DYNOBJ. The
5682 function is passed no arguments and its return value is ignored.
5683 When FUNCTION is something returned by `dynamic-func', call that
5684 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5685 etc.), look it up in DYNOBJ; this is equivalent to
5687 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5689 Interrupts are deferred while the C function is executing (with
5690 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5692 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5694 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5695 some arguments and return its return value. The C function is
5696 expected to take two arguments and return an `int', just like
5699 int c_func (int argc, char **argv);
5701 ARGS must be a list of strings and is converted into an array of
5702 `char *'. The array is passed in ARGV and its size in ARGC. The
5703 return value is converted to a Scheme number and returned from the
5704 call to `dynamic-args-call'.
5706 When dynamic linking is disabled or not supported on your system,
5707 the above functions throw errors, but they are still available.
5709 Here is a small example that works on GNU/Linux:
5711 (define libc-obj (dynamic-link "libc.so"))
5712 (dynamic-args-call 'rand libc-obj '())
5714 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5716 ** The #/ syntax for module names is depreciated, and will be removed
5717 in a future version of Guile. Instead of
5725 The latter syntax is more consistent with existing Lisp practice.
5727 ** Guile now does fancier printing of structures. Structures are the
5728 underlying implementation for records, which in turn are used to
5729 implement modules, so all of these object now print differently and in
5730 a more informative way.
5732 The Scheme printer will examine the builtin variable *struct-printer*
5733 whenever it needs to print a structure object. When this variable is
5734 not `#f' it is deemed to be a procedure and will be applied to the
5735 structure object and the output port. When *struct-printer* is `#f'
5736 or the procedure return `#f' the structure object will be printed in
5737 the boring #<struct 80458270> form.
5739 This hook is used by some routines in ice-9/boot-9.scm to implement
5740 type specific printing routines. Please read the comments there about
5743 One of the more specific uses of structs are records. The printing
5744 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5745 called. It should behave like a *struct-printer* procedure (described
5748 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5749 token of the form #:NAME, where NAME has the same syntax as a Scheme
5750 symbol, is the external representation of the keyword named NAME.
5751 Keyword objects print using this syntax as well, so values containing
5752 keyword objects can be read back into Guile. When used in an
5753 expression, keywords are self-quoting objects.
5755 Guile suports this read syntax, and uses this print syntax, regardless
5756 of the current setting of the `keyword' read option. The `keyword'
5757 read option only controls whether Guile recognizes the `:NAME' syntax,
5758 which is incompatible with R4RS. (R4RS says such token represent
5761 ** Guile has regular expression support again. Guile 1.0 included
5762 functions for matching regular expressions, based on the Rx library.
5763 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5764 distribution, and thus Guile had no regular expression support. Guile
5765 1.2 again supports the most commonly used functions, and supports all
5766 of SCSH's regular expression functions.
5768 If your system does not include a POSIX regular expression library,
5769 and you have not linked Guile with a third-party regexp library such as
5770 Rx, these functions will not be available. You can tell whether your
5771 Guile installation includes regular expression support by checking
5772 whether the `*features*' list includes the `regex' symbol.
5774 *** regexp functions
5776 By default, Guile supports POSIX extended regular expressions. That
5777 means that the characters `(', `)', `+' and `?' are special, and must
5778 be escaped if you wish to match the literal characters.
5780 This regular expression interface was modeled after that implemented
5781 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5782 with SCSH regular expressions.
5784 **** Function: string-match PATTERN STR [START]
5785 Compile the string PATTERN into a regular expression and compare
5786 it with STR. The optional numeric argument START specifies the
5787 position of STR at which to begin matching.
5789 `string-match' returns a "match structure" which describes what,
5790 if anything, was matched by the regular expression. *Note Match
5791 Structures::. If STR does not match PATTERN at all,
5792 `string-match' returns `#f'.
5794 Each time `string-match' is called, it must compile its PATTERN
5795 argument into a regular expression structure. This operation is
5796 expensive, which makes `string-match' inefficient if the same regular
5797 expression is used several times (for example, in a loop). For better
5798 performance, you can compile a regular expression in advance and then
5799 match strings against the compiled regexp.
5801 **** Function: make-regexp STR [FLAGS]
5802 Compile the regular expression described by STR, and return the
5803 compiled regexp structure. If STR does not describe a legal
5804 regular expression, `make-regexp' throws a
5805 `regular-expression-syntax' error.
5807 FLAGS may be the bitwise-or of one or more of the following:
5809 **** Constant: regexp/extended
5810 Use POSIX Extended Regular Expression syntax when interpreting
5811 STR. If not set, POSIX Basic Regular Expression syntax is used.
5812 If the FLAGS argument is omitted, we assume regexp/extended.
5814 **** Constant: regexp/icase
5815 Do not differentiate case. Subsequent searches using the
5816 returned regular expression will be case insensitive.
5818 **** Constant: regexp/newline
5819 Match-any-character operators don't match a newline.
5821 A non-matching list ([^...]) not containing a newline matches a
5824 Match-beginning-of-line operator (^) matches the empty string
5825 immediately after a newline, regardless of whether the FLAGS
5826 passed to regexp-exec contain regexp/notbol.
5828 Match-end-of-line operator ($) matches the empty string
5829 immediately before a newline, regardless of whether the FLAGS
5830 passed to regexp-exec contain regexp/noteol.
5832 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5833 Match the compiled regular expression REGEXP against `str'. If
5834 the optional integer START argument is provided, begin matching
5835 from that position in the string. Return a match structure
5836 describing the results of the match, or `#f' if no match could be
5839 FLAGS may be the bitwise-or of one or more of the following:
5841 **** Constant: regexp/notbol
5842 The match-beginning-of-line operator always fails to match (but
5843 see the compilation flag regexp/newline above) This flag may be
5844 used when different portions of a string are passed to
5845 regexp-exec and the beginning of the string should not be
5846 interpreted as the beginning of the line.
5848 **** Constant: regexp/noteol
5849 The match-end-of-line operator always fails to match (but see the
5850 compilation flag regexp/newline above)
5852 **** Function: regexp? OBJ
5853 Return `#t' if OBJ is a compiled regular expression, or `#f'
5856 Regular expressions are commonly used to find patterns in one string
5857 and replace them with the contents of another string.
5859 **** Function: regexp-substitute PORT MATCH [ITEM...]
5860 Write to the output port PORT selected contents of the match
5861 structure MATCH. Each ITEM specifies what should be written, and
5862 may be one of the following arguments:
5864 * A string. String arguments are written out verbatim.
5866 * An integer. The submatch with that number is written.
5868 * The symbol `pre'. The portion of the matched string preceding
5869 the regexp match is written.
5871 * The symbol `post'. The portion of the matched string
5872 following the regexp match is written.
5874 PORT may be `#f', in which case nothing is written; instead,
5875 `regexp-substitute' constructs a string from the specified ITEMs
5878 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5879 Similar to `regexp-substitute', but can be used to perform global
5880 substitutions on STR. Instead of taking a match structure as an
5881 argument, `regexp-substitute/global' takes two string arguments: a
5882 REGEXP string describing a regular expression, and a TARGET string
5883 which should be matched against this regular expression.
5885 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5888 * A function may be supplied. When this function is called, it
5889 will be passed one argument: a match structure for a given
5890 regular expression match. It should return a string to be
5891 written out to PORT.
5893 * The `post' symbol causes `regexp-substitute/global' to recurse
5894 on the unmatched portion of STR. This *must* be supplied in
5895 order to perform global search-and-replace on STR; if it is
5896 not present among the ITEMs, then `regexp-substitute/global'
5897 will return after processing a single match.
5899 *** Match Structures
5901 A "match structure" is the object returned by `string-match' and
5902 `regexp-exec'. It describes which portion of a string, if any, matched
5903 the given regular expression. Match structures include: a reference to
5904 the string that was checked for matches; the starting and ending
5905 positions of the regexp match; and, if the regexp included any
5906 parenthesized subexpressions, the starting and ending positions of each
5909 In each of the regexp match functions described below, the `match'
5910 argument must be a match structure returned by a previous call to
5911 `string-match' or `regexp-exec'. Most of these functions return some
5912 information about the original target string that was matched against a
5913 regular expression; we will call that string TARGET for easy reference.
5915 **** Function: regexp-match? OBJ
5916 Return `#t' if OBJ is a match structure returned by a previous
5917 call to `regexp-exec', or `#f' otherwise.
5919 **** Function: match:substring MATCH [N]
5920 Return the portion of TARGET matched by subexpression number N.
5921 Submatch 0 (the default) represents the entire regexp match. If
5922 the regular expression as a whole matched, but the subexpression
5923 number N did not match, return `#f'.
5925 **** Function: match:start MATCH [N]
5926 Return the starting position of submatch number N.
5928 **** Function: match:end MATCH [N]
5929 Return the ending position of submatch number N.
5931 **** Function: match:prefix MATCH
5932 Return the unmatched portion of TARGET preceding the regexp match.
5934 **** Function: match:suffix MATCH
5935 Return the unmatched portion of TARGET following the regexp match.
5937 **** Function: match:count MATCH
5938 Return the number of parenthesized subexpressions from MATCH.
5939 Note that the entire regular expression match itself counts as a
5940 subexpression, and failed submatches are included in the count.
5942 **** Function: match:string MATCH
5943 Return the original TARGET string.
5945 *** Backslash Escapes
5947 Sometimes you will want a regexp to match characters like `*' or `$'
5948 exactly. For example, to check whether a particular string represents
5949 a menu entry from an Info node, it would be useful to match it against
5950 a regexp like `^* [^:]*::'. However, this won't work; because the
5951 asterisk is a metacharacter, it won't match the `*' at the beginning of
5952 the string. In this case, we want to make the first asterisk un-magic.
5954 You can do this by preceding the metacharacter with a backslash
5955 character `\'. (This is also called "quoting" the metacharacter, and
5956 is known as a "backslash escape".) When Guile sees a backslash in a
5957 regular expression, it considers the following glyph to be an ordinary
5958 character, no matter what special meaning it would ordinarily have.
5959 Therefore, we can make the above example work by changing the regexp to
5960 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5961 to match only a single asterisk in the target string.
5963 Since the backslash is itself a metacharacter, you may force a
5964 regexp to match a backslash in the target string by preceding the
5965 backslash with itself. For example, to find variable references in a
5966 TeX program, you might want to find occurrences of the string `\let\'
5967 followed by any number of alphabetic characters. The regular expression
5968 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5969 each match a single backslash in the target string.
5971 **** Function: regexp-quote STR
5972 Quote each special character found in STR with a backslash, and
5973 return the resulting string.
5975 *Very important:* Using backslash escapes in Guile source code (as
5976 in Emacs Lisp or C) can be tricky, because the backslash character has
5977 special meaning for the Guile reader. For example, if Guile encounters
5978 the character sequence `\n' in the middle of a string while processing
5979 Scheme code, it replaces those characters with a newline character.
5980 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5981 Several of these "escape sequences" are processed by the Guile reader
5982 before your code is executed. Unrecognized escape sequences are
5983 ignored: if the characters `\*' appear in a string, they will be
5984 translated to the single character `*'.
5986 This translation is obviously undesirable for regular expressions,
5987 since we want to be able to include backslashes in a string in order to
5988 escape regexp metacharacters. Therefore, to make sure that a backslash
5989 is preserved in a string in your Guile program, you must use *two*
5990 consecutive backslashes:
5992 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5994 The string in this example is preprocessed by the Guile reader before
5995 any code is executed. The resulting argument to `make-regexp' is the
5996 string `^\* [^:]*', which is what we really want.
5998 This also means that in order to write a regular expression that
5999 matches a single backslash character, the regular expression string in
6000 the source code must include *four* backslashes. Each consecutive pair
6001 of backslashes gets translated by the Guile reader to a single
6002 backslash, and the resulting double-backslash is interpreted by the
6003 regexp engine as matching a single backslash character. Hence:
6005 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6007 The reason for the unwieldiness of this syntax is historical. Both
6008 regular expression pattern matchers and Unix string processing systems
6009 have traditionally used backslashes with the special meanings described
6010 above. The POSIX regular expression specification and ANSI C standard
6011 both require these semantics. Attempting to abandon either convention
6012 would cause other kinds of compatibility problems, possibly more severe
6013 ones. Therefore, without extending the Scheme reader to support
6014 strings with different quoting conventions (an ungainly and confusing
6015 extension when implemented in other languages), we must adhere to this
6016 cumbersome escape syntax.
6018 * Changes to the gh_ interface
6020 * Changes to the scm_ interface
6022 * Changes to system call interfaces:
6024 ** The value returned by `raise' is now unspecified. It throws an exception
6027 *** A new procedure `sigaction' can be used to install signal handlers
6029 (sigaction signum [action] [flags])
6031 signum is the signal number, which can be specified using the value
6034 If action is omitted, sigaction returns a pair: the CAR is the current
6035 signal hander, which will be either an integer with the value SIG_DFL
6036 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6037 handles the signal, or #f if a non-Scheme procedure handles the
6038 signal. The CDR contains the current sigaction flags for the handler.
6040 If action is provided, it is installed as the new handler for signum.
6041 action can be a Scheme procedure taking one argument, or the value of
6042 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6043 whatever signal handler was installed before sigaction was first used.
6044 Flags can optionally be specified for the new handler (SA_RESTART is
6045 always used if the system provides it, so need not be specified.) The
6046 return value is a pair with information about the old handler as
6049 This interface does not provide access to the "signal blocking"
6050 facility. Maybe this is not needed, since the thread support may
6051 provide solutions to the problem of consistent access to data
6054 *** A new procedure `flush-all-ports' is equivalent to running
6055 `force-output' on every port open for output.
6057 ** Guile now provides information on how it was built, via the new
6058 global variable, %guile-build-info. This variable records the values
6059 of the standard GNU makefile directory variables as an assocation
6060 list, mapping variable names (symbols) onto directory paths (strings).
6061 For example, to find out where the Guile link libraries were
6062 installed, you can say:
6064 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6067 * Changes to the scm_ interface
6069 ** The new function scm_handle_by_message_noexit is just like the
6070 existing scm_handle_by_message function, except that it doesn't call
6071 exit to terminate the process. Instead, it prints a message and just
6072 returns #f. This might be a more appropriate catch-all handler for
6073 new dynamic roots and threads.
6076 Changes in Guile 1.1 (released Friday, May 16 1997):
6078 * Changes to the distribution.
6080 The Guile 1.0 distribution has been split up into several smaller
6082 guile-core --- the Guile interpreter itself.
6083 guile-tcltk --- the interface between the Guile interpreter and
6084 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6085 is a toolkit for building graphical user interfaces.
6086 guile-rgx-ctax --- the interface between Guile and the Rx regular
6087 expression matcher, and the translator for the Ctax
6088 programming language. These are packaged together because the
6089 Ctax translator uses Rx to parse Ctax source code.
6091 This NEWS file describes the changes made to guile-core since the 1.0
6094 We no longer distribute the documentation, since it was either out of
6095 date, or incomplete. As soon as we have current documentation, we
6100 * Changes to the stand-alone interpreter
6102 ** guile now accepts command-line arguments compatible with SCSH, Olin
6103 Shivers' Scheme Shell.
6105 In general, arguments are evaluated from left to right, but there are
6106 exceptions. The following switches stop argument processing, and
6107 stash all remaining command-line arguments as the value returned by
6108 the (command-line) function.
6109 -s SCRIPT load Scheme source code from FILE, and exit
6110 -c EXPR evalute Scheme expression EXPR, and exit
6111 -- stop scanning arguments; run interactively
6113 The switches below are processed as they are encountered.
6114 -l FILE load Scheme source code from FILE
6115 -e FUNCTION after reading script, apply FUNCTION to
6116 command line arguments
6117 -ds do -s script at this point
6118 --emacs enable Emacs protocol (experimental)
6119 -h, --help display this help and exit
6120 -v, --version display version information and exit
6121 \ read arguments from following script lines
6123 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6124 which re-implements the traditional "echo" command:
6126 #!/usr/local/bin/guile -s
6129 (map (lambda (arg) (display arg) (display " "))
6133 (main (command-line))
6135 Suppose we invoke this script as follows:
6137 ekko a speckled gecko
6139 Through the magic of Unix script processing (triggered by the `#!'
6140 token at the top of the file), /usr/local/bin/guile receives the
6141 following list of command-line arguments:
6143 ("-s" "./ekko" "a" "speckled" "gecko")
6145 Unix inserts the name of the script after the argument specified on
6146 the first line of the file (in this case, "-s"), and then follows that
6147 with the arguments given to the script. Guile loads the script, which
6148 defines the `main' function, and then applies it to the list of
6149 remaining command-line arguments, ("a" "speckled" "gecko").
6151 In Unix, the first line of a script file must take the following form:
6153 #!INTERPRETER ARGUMENT
6155 where INTERPRETER is the absolute filename of the interpreter
6156 executable, and ARGUMENT is a single command-line argument to pass to
6159 You may only pass one argument to the interpreter, and its length is
6160 limited. These restrictions can be annoying to work around, so Guile
6161 provides a general mechanism (borrowed from, and compatible with,
6162 SCSH) for circumventing them.
6164 If the ARGUMENT in a Guile script is a single backslash character,
6165 `\', Guile will open the script file, parse arguments from its second
6166 and subsequent lines, and replace the `\' with them. So, for example,
6167 here is another implementation of the `ekko' script:
6169 #!/usr/local/bin/guile \
6173 (for-each (lambda (arg) (display arg) (display " "))
6177 If the user invokes this script as follows:
6179 ekko a speckled gecko
6181 Unix expands this into
6183 /usr/local/bin/guile \ ekko a speckled gecko
6185 When Guile sees the `\' argument, it replaces it with the arguments
6186 read from the second line of the script, producing:
6188 /usr/local/bin/guile -e main -s ekko a speckled gecko
6190 This tells Guile to load the `ekko' script, and apply the function
6191 `main' to the argument list ("a" "speckled" "gecko").
6193 Here is how Guile parses the command-line arguments:
6194 - Each space character terminates an argument. This means that two
6195 spaces in a row introduce an empty-string argument.
6196 - The tab character is not permitted (unless you quote it with the
6197 backslash character, as described below), to avoid confusion.
6198 - The newline character terminates the sequence of arguments, and will
6199 also terminate a final non-empty argument. (However, a newline
6200 following a space will not introduce a final empty-string argument;
6201 it only terminates the argument list.)
6202 - The backslash character is the escape character. It escapes
6203 backslash, space, tab, and newline. The ANSI C escape sequences
6204 like \n and \t are also supported. These produce argument
6205 constituents; the two-character combination \n doesn't act like a
6206 terminating newline. The escape sequence \NNN for exactly three
6207 octal digits reads as the character whose ASCII code is NNN. As
6208 above, characters produced this way are argument constituents.
6209 Backslash followed by other characters is not allowed.
6211 * Changes to the procedure for linking libguile with your programs
6213 ** Guile now builds and installs a shared guile library, if your
6214 system support shared libraries. (It still builds a static library on
6215 all systems.) Guile automatically detects whether your system
6216 supports shared libraries. To prevent Guile from buildisg shared
6217 libraries, pass the `--disable-shared' flag to the configure script.
6219 Guile takes longer to compile when it builds shared libraries, because
6220 it must compile every file twice --- once to produce position-
6221 independent object code, and once to produce normal object code.
6223 ** The libthreads library has been merged into libguile.
6225 To link a program against Guile, you now need only link against
6226 -lguile and -lqt; -lthreads is no longer needed. If you are using
6227 autoconf to generate configuration scripts for your application, the
6228 following lines should suffice to add the appropriate libraries to
6231 ### Find quickthreads and libguile.
6232 AC_CHECK_LIB(qt, main)
6233 AC_CHECK_LIB(guile, scm_shell)
6235 * Changes to Scheme functions
6237 ** Guile Scheme's special syntax for keyword objects is now optional,
6238 and disabled by default.
6240 The syntax variation from R4RS made it difficult to port some
6241 interesting packages to Guile. The routines which accepted keyword
6242 arguments (mostly in the module system) have been modified to also
6243 accept symbols whose names begin with `:'.
6245 To change the keyword syntax, you must first import the (ice-9 debug)
6247 (use-modules (ice-9 debug))
6249 Then you can enable the keyword syntax as follows:
6250 (read-set! keywords 'prefix)
6252 To disable keyword syntax, do this:
6253 (read-set! keywords #f)
6255 ** Many more primitive functions accept shared substrings as
6256 arguments. In the past, these functions required normal, mutable
6257 strings as arguments, although they never made use of this
6260 ** The uniform array functions now operate on byte vectors. These
6261 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6262 `serial-array-map', `array-map', `array-for-each', and
6265 ** The new functions `trace' and `untrace' implement simple debugging
6266 support for Scheme functions.
6268 The `trace' function accepts any number of procedures as arguments,
6269 and tells the Guile interpreter to display each procedure's name and
6270 arguments each time the procedure is invoked. When invoked with no
6271 arguments, `trace' returns the list of procedures currently being
6274 The `untrace' function accepts any number of procedures as arguments,
6275 and tells the Guile interpreter not to trace them any more. When
6276 invoked with no arguments, `untrace' untraces all curretly traced
6279 The tracing in Guile has an advantage over most other systems: we
6280 don't create new procedure objects, but mark the procedure objects
6281 themselves. This means that anonymous and internal procedures can be
6284 ** The function `assert-repl-prompt' has been renamed to
6285 `set-repl-prompt!'. It takes one argument, PROMPT.
6286 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6287 - If PROMPT is a string, we use it as a prompt.
6288 - If PROMPT is a procedure accepting no arguments, we call it, and
6289 display the result as a prompt.
6290 - Otherwise, we display "> ".
6292 ** The new function `eval-string' reads Scheme expressions from a
6293 string and evaluates them, returning the value of the last expression
6294 in the string. If the string contains no expressions, it returns an
6297 ** The new function `thunk?' returns true iff its argument is a
6298 procedure of zero arguments.
6300 ** `defined?' is now a builtin function, instead of syntax. This
6301 means that its argument should be quoted. It returns #t iff its
6302 argument is bound in the current module.
6304 ** The new syntax `use-modules' allows you to add new modules to your
6305 environment without re-typing a complete `define-module' form. It
6306 accepts any number of module names as arguments, and imports their
6307 public bindings into the current module.
6309 ** The new function (module-defined? NAME MODULE) returns true iff
6310 NAME, a symbol, is defined in MODULE, a module object.
6312 ** The new function `builtin-bindings' creates and returns a hash
6313 table containing copies of all the root module's bindings.
6315 ** The new function `builtin-weak-bindings' does the same as
6316 `builtin-bindings', but creates a doubly-weak hash table.
6318 ** The `equal?' function now considers variable objects to be
6319 equivalent if they have the same name and the same value.
6321 ** The new function `command-line' returns the command-line arguments
6322 given to Guile, as a list of strings.
6324 When using guile as a script interpreter, `command-line' returns the
6325 script's arguments; those processed by the interpreter (like `-s' or
6326 `-c') are omitted. (In other words, you get the normal, expected
6327 behavior.) Any application that uses scm_shell to process its
6328 command-line arguments gets this behavior as well.
6330 ** The new function `load-user-init' looks for a file called `.guile'
6331 in the user's home directory, and loads it if it exists. This is
6332 mostly for use by the code generated by scm_compile_shell_switches,
6333 but we thought it might also be useful in other circumstances.
6335 ** The new function `log10' returns the base-10 logarithm of its
6338 ** Changes to I/O functions
6340 *** The functions `read', `primitive-load', `read-and-eval!', and
6341 `primitive-load-path' no longer take optional arguments controlling
6342 case insensitivity and a `#' parser.
6344 Case sensitivity is now controlled by a read option called
6345 `case-insensitive'. The user can add new `#' syntaxes with the
6346 `read-hash-extend' function (see below).
6348 *** The new function `read-hash-extend' allows the user to change the
6349 syntax of Guile Scheme in a somewhat controlled way.
6351 (read-hash-extend CHAR PROC)
6352 When parsing S-expressions, if we read a `#' character followed by
6353 the character CHAR, use PROC to parse an object from the stream.
6354 If PROC is #f, remove any parsing procedure registered for CHAR.
6356 The reader applies PROC to two arguments: CHAR and an input port.
6358 *** The new functions read-delimited and read-delimited! provide a
6359 general mechanism for doing delimited input on streams.
6361 (read-delimited DELIMS [PORT HANDLE-DELIM])
6362 Read until we encounter one of the characters in DELIMS (a string),
6363 or end-of-file. PORT is the input port to read from; it defaults to
6364 the current input port. The HANDLE-DELIM parameter determines how
6365 the terminating character is handled; it should be one of the
6368 'trim omit delimiter from result
6369 'peek leave delimiter character in input stream
6370 'concat append delimiter character to returned value
6371 'split return a pair: (RESULT . TERMINATOR)
6373 HANDLE-DELIM defaults to 'peek.
6375 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6376 A side-effecting variant of `read-delimited'.
6378 The data is written into the string BUF at the indices in the
6379 half-open interval [START, END); the default interval is the whole
6380 string: START = 0 and END = (string-length BUF). The values of
6381 START and END must specify a well-defined interval in BUF, i.e.
6382 0 <= START <= END <= (string-length BUF).
6384 It returns NBYTES, the number of bytes read. If the buffer filled
6385 up without a delimiter character being found, it returns #f. If the
6386 port is at EOF when the read starts, it returns the EOF object.
6388 If an integer is returned (i.e., the read is successfully terminated
6389 by reading a delimiter character), then the HANDLE-DELIM parameter
6390 determines how to handle the terminating character. It is described
6391 above, and defaults to 'peek.
6393 (The descriptions of these functions were borrowed from the SCSH
6394 manual, by Olin Shivers and Brian Carlstrom.)
6396 *** The `%read-delimited!' function is the primitive used to implement
6397 `read-delimited' and `read-delimited!'.
6399 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6401 This returns a pair of values: (TERMINATOR . NUM-READ).
6402 - TERMINATOR describes why the read was terminated. If it is a
6403 character or the eof object, then that is the value that terminated
6404 the read. If it is #f, the function filled the buffer without finding
6405 a delimiting character.
6406 - NUM-READ is the number of characters read into BUF.
6408 If the read is successfully terminated by reading a delimiter
6409 character, then the gobble? parameter determines what to do with the
6410 terminating character. If true, the character is removed from the
6411 input stream; if false, the character is left in the input stream
6412 where a subsequent read operation will retrieve it. In either case,
6413 the character is also the first value returned by the procedure call.
6415 (The descriptions of this function was borrowed from the SCSH manual,
6416 by Olin Shivers and Brian Carlstrom.)
6418 *** The `read-line' and `read-line!' functions have changed; they now
6419 trim the terminator by default; previously they appended it to the
6420 returned string. For the old behavior, use (read-line PORT 'concat).
6422 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6423 take new optional START and END arguments, specifying the region of
6424 the array to read and write.
6426 *** The `ungetc-char-ready?' function has been removed. We feel it's
6427 inappropriate for an interface to expose implementation details this
6430 ** Changes to the Unix library and system call interface
6432 *** The new fcntl function provides access to the Unix `fcntl' system
6435 (fcntl PORT COMMAND VALUE)
6436 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6437 Values for COMMAND are:
6439 F_DUPFD duplicate a file descriptor
6440 F_GETFD read the descriptor's close-on-exec flag
6441 F_SETFD set the descriptor's close-on-exec flag to VALUE
6442 F_GETFL read the descriptor's flags, as set on open
6443 F_SETFL set the descriptor's flags, as set on open to VALUE
6444 F_GETOWN return the process ID of a socket's owner, for SIGIO
6445 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6446 FD_CLOEXEC not sure what this is
6448 For details, see the documentation for the fcntl system call.
6450 *** The arguments to `select' have changed, for compatibility with
6451 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6452 expected behavior. The MILLISECONDS parameter has been changed to
6453 MICROSECONDS, to more closely resemble the underlying system call.
6454 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6455 corresponding return set will be the same.
6457 *** The arguments to the `mknod' system call have changed. They are
6460 (mknod PATH TYPE PERMS DEV)
6461 Create a new file (`node') in the file system. PATH is the name of
6462 the file to create. TYPE is the kind of file to create; it should
6463 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6464 permission bits to give the newly created file. If TYPE is
6465 'block-special or 'char-special, DEV specifies which device the
6466 special file refers to; its interpretation depends on the kind of
6467 special file being created.
6469 *** The `fork' function has been renamed to `primitive-fork', to avoid
6470 clashing with various SCSH forks.
6472 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6473 and `recvfrom!'. They no longer accept a size for a second argument;
6474 you must pass a string to hold the received value. They no longer
6475 return the buffer. Instead, `recv' returns the length of the message
6476 received, and `recvfrom' returns a pair containing the packet's length
6477 and originating address.
6479 *** The file descriptor datatype has been removed, as have the
6480 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6481 We plan to replace these functions with a SCSH-compatible interface.
6483 *** The `create' function has been removed; it's just a special case
6486 *** There are new functions to break down process termination status
6487 values. In the descriptions below, STATUS is a value returned by
6490 (status:exit-val STATUS)
6491 If the child process exited normally, this function returns the exit
6492 code for the child process (i.e., the value passed to exit, or
6493 returned from main). If the child process did not exit normally,
6494 this function returns #f.
6496 (status:stop-sig STATUS)
6497 If the child process was suspended by a signal, this function
6498 returns the signal that suspended the child. Otherwise, it returns
6501 (status:term-sig STATUS)
6502 If the child process terminated abnormally, this function returns
6503 the signal that terminated the child. Otherwise, this function
6506 POSIX promises that exactly one of these functions will return true on
6507 a valid STATUS value.
6509 These functions are compatible with SCSH.
6511 *** There are new accessors and setters for the broken-out time vectors
6512 returned by `localtime', `gmtime', and that ilk. They are:
6514 Component Accessor Setter
6515 ========================= ============ ============
6516 seconds tm:sec set-tm:sec
6517 minutes tm:min set-tm:min
6518 hours tm:hour set-tm:hour
6519 day of the month tm:mday set-tm:mday
6520 month tm:mon set-tm:mon
6521 year tm:year set-tm:year
6522 day of the week tm:wday set-tm:wday
6523 day in the year tm:yday set-tm:yday
6524 daylight saving time tm:isdst set-tm:isdst
6525 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6526 name of time zone tm:zone set-tm:zone
6528 *** There are new accessors for the vectors returned by `uname',
6529 describing the host system:
6532 ============================================== ================
6533 name of the operating system implementation utsname:sysname
6534 network name of this machine utsname:nodename
6535 release level of the operating system utsname:release
6536 version level of the operating system utsname:version
6537 machine hardware platform utsname:machine
6539 *** There are new accessors for the vectors returned by `getpw',
6540 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6541 system's user database:
6544 ====================== =================
6545 user name passwd:name
6546 user password passwd:passwd
6549 real name passwd:gecos
6550 home directory passwd:dir
6551 shell program passwd:shell
6553 *** There are new accessors for the vectors returned by `getgr',
6554 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6555 system's group database:
6558 ======================= ============
6559 group name group:name
6560 group password group:passwd
6562 group members group:mem
6564 *** There are new accessors for the vectors returned by `gethost',
6565 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6569 ========================= ===============
6570 official name of host hostent:name
6571 alias list hostent:aliases
6572 host address type hostent:addrtype
6573 length of address hostent:length
6574 list of addresses hostent:addr-list
6576 *** There are new accessors for the vectors returned by `getnet',
6577 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6581 ========================= ===============
6582 official name of net netent:name
6583 alias list netent:aliases
6584 net number type netent:addrtype
6585 net number netent:net
6587 *** There are new accessors for the vectors returned by `getproto',
6588 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6592 ========================= ===============
6593 official protocol name protoent:name
6594 alias list protoent:aliases
6595 protocol number protoent:proto
6597 *** There are new accessors for the vectors returned by `getserv',
6598 `getservbyname', `getservbyport', and `getservent', describing
6602 ========================= ===============
6603 official service name servent:name
6604 alias list servent:aliases
6605 port number servent:port
6606 protocol to use servent:proto
6608 *** There are new accessors for the sockaddr structures returned by
6609 `accept', `getsockname', `getpeername', `recvfrom!':
6612 ======================================== ===============
6613 address format (`family') sockaddr:fam
6614 path, for file domain addresses sockaddr:path
6615 address, for internet domain addresses sockaddr:addr
6616 TCP or UDP port, for internet sockaddr:port
6618 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6619 `getprotoent', and `getservent' functions now return #f at the end of
6620 the user database. (They used to throw an exception.)
6622 Note that calling MUMBLEent function is equivalent to calling the
6623 corresponding MUMBLE function with no arguments.
6625 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6626 `setprotoent', and `setservent' routines now take no arguments.
6628 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6629 provide more useful information when they throw an exception.
6631 *** The `lnaof' function has been renamed to `inet-lnaof'.
6633 *** Guile now claims to have the `current-time' feature.
6635 *** The `mktime' function now takes an optional second argument ZONE,
6636 giving the time zone to use for the conversion. ZONE should be a
6637 string, in the same format as expected for the "TZ" environment variable.
6639 *** The `strptime' function now returns a pair (TIME . COUNT), where
6640 TIME is the parsed time as a vector, and COUNT is the number of
6641 characters from the string left unparsed. This function used to
6642 return the remaining characters as a string.
6644 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6645 The return value is now (SECONDS . MICROSECONDS); the fractional
6646 component is no longer expressed in "ticks".
6648 *** The `ticks/sec' constant has been removed, in light of the above change.
6650 * Changes to the gh_ interface
6652 ** gh_eval_str() now returns an SCM object which is the result of the
6655 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6658 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6659 and returns the array
6661 ** gh_scm2str0() is gone: there is no need to distinguish
6662 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6663 the user to interpret the data both ways.
6665 * Changes to the scm_ interface
6667 ** The new function scm_symbol_value0 provides an easy way to get a
6668 symbol's value from C code:
6670 SCM scm_symbol_value0 (char *NAME)
6671 Return the value of the symbol named by the null-terminated string
6672 NAME in the current module. If the symbol named NAME is unbound in
6673 the current module, return SCM_UNDEFINED.
6675 ** The new function scm_sysintern0 creates new top-level variables,
6676 without assigning them a value.
6678 SCM scm_sysintern0 (char *NAME)
6679 Create a new Scheme top-level variable named NAME. NAME is a
6680 null-terminated string. Return the variable's value cell.
6682 ** The function scm_internal_catch is the guts of catch. It handles
6683 all the mechanics of setting up a catch target, invoking the catch
6684 body, and perhaps invoking the handler if the body does a throw.
6686 The function is designed to be usable from C code, but is general
6687 enough to implement all the semantics Guile Scheme expects from throw.
6689 TAG is the catch tag. Typically, this is a symbol, but this function
6690 doesn't actually care about that.
6692 BODY is a pointer to a C function which runs the body of the catch;
6693 this is the code you can throw from. We call it like this:
6694 BODY (BODY_DATA, JMPBUF)
6696 BODY_DATA is just the BODY_DATA argument we received; we pass it
6697 through to BODY as its first argument. The caller can make
6698 BODY_DATA point to anything useful that BODY might need.
6699 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6700 which we have just created and initialized.
6702 HANDLER is a pointer to a C function to deal with a throw to TAG,
6703 should one occur. We call it like this:
6704 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6706 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6707 same idea as BODY_DATA above.
6708 THROWN_TAG is the tag that the user threw to; usually this is
6709 TAG, but it could be something else if TAG was #t (i.e., a
6710 catch-all), or the user threw to a jmpbuf.
6711 THROW_ARGS is the list of arguments the user passed to the THROW
6714 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6715 is just a pointer we pass through to HANDLER. We don't actually
6716 use either of those pointers otherwise ourselves. The idea is
6717 that, if our caller wants to communicate something to BODY or
6718 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6719 HANDLER can then use. Think of it as a way to make BODY and
6720 HANDLER closures, not just functions; MUMBLE_DATA points to the
6723 Of course, it's up to the caller to make sure that any data a
6724 MUMBLE_DATA needs is protected from GC. A common way to do this is
6725 to make MUMBLE_DATA a pointer to data stored in an automatic
6726 structure variable; since the collector must scan the stack for
6727 references anyway, this assures that any references in MUMBLE_DATA
6730 ** The new function scm_internal_lazy_catch is exactly like
6731 scm_internal_catch, except:
6733 - It does not unwind the stack (this is the major difference).
6734 - If handler returns, its value is returned from the throw.
6735 - BODY always receives #f as its JMPBUF argument (since there's no
6736 jmpbuf associated with a lazy catch, because we don't unwind the
6739 ** scm_body_thunk is a new body function you can pass to
6740 scm_internal_catch if you want the body to be like Scheme's `catch'
6741 --- a thunk, or a function of one argument if the tag is #f.
6743 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6744 contains the Scheme procedure to invoke as the body, and the tag
6745 we're catching. If the tag is #f, then we pass JMPBUF (created by
6746 scm_internal_catch) to the body procedure; otherwise, the body gets
6749 ** scm_handle_by_proc is a new handler function you can pass to
6750 scm_internal_catch if you want the handler to act like Scheme's catch
6751 --- call a procedure with the tag and the throw arguments.
6753 If the user does a throw to this catch, this function runs a handler
6754 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6755 variable holding the Scheme procedure object to invoke. It ought to
6756 be a pointer to an automatic variable (i.e., one living on the stack),
6757 or the procedure object should be otherwise protected from GC.
6759 ** scm_handle_by_message is a new handler function to use with
6760 `scm_internal_catch' if you want Guile to print a message and die.
6761 It's useful for dealing with throws to uncaught keys at the top level.
6763 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6764 message header to print; if zero, we use "guile" instead. That
6765 text is followed by a colon, then the message described by ARGS.
6767 ** The return type of scm_boot_guile is now void; the function does
6768 not return a value, and indeed, never returns at all.
6770 ** The new function scm_shell makes it easy for user applications to
6771 process command-line arguments in a way that is compatible with the
6772 stand-alone guile interpreter (which is in turn compatible with SCSH,
6775 To use the scm_shell function, first initialize any guile modules
6776 linked into your application, and then call scm_shell with the values
6777 of ARGC and ARGV your `main' function received. scm_shell will add
6778 any SCSH-style meta-arguments from the top of the script file to the
6779 argument vector, and then process the command-line arguments. This
6780 generally means loading a script file or starting up an interactive
6781 command interpreter. For details, see "Changes to the stand-alone
6784 ** The new functions scm_get_meta_args and scm_count_argv help you
6785 implement the SCSH-style meta-argument, `\'.
6787 char **scm_get_meta_args (int ARGC, char **ARGV)
6788 If the second element of ARGV is a string consisting of a single
6789 backslash character (i.e. "\\" in Scheme notation), open the file
6790 named by the following argument, parse arguments from it, and return
6791 the spliced command line. The returned array is terminated by a
6794 For details of argument parsing, see above, under "guile now accepts
6795 command-line arguments compatible with SCSH..."
6797 int scm_count_argv (char **ARGV)
6798 Count the arguments in ARGV, assuming it is terminated by a null
6801 For an example of how these functions might be used, see the source
6802 code for the function scm_shell in libguile/script.c.
6804 You will usually want to use scm_shell instead of calling this
6807 ** The new function scm_compile_shell_switches turns an array of
6808 command-line arguments into Scheme code to carry out the actions they
6809 describe. Given ARGC and ARGV, it returns a Scheme expression to
6810 evaluate, and calls scm_set_program_arguments to make any remaining
6811 command-line arguments available to the Scheme code. For example,
6812 given the following arguments:
6814 -e main -s ekko a speckled gecko
6816 scm_set_program_arguments will return the following expression:
6818 (begin (load "ekko") (main (command-line)) (quit))
6820 You will usually want to use scm_shell instead of calling this
6823 ** The function scm_shell_usage prints a usage message appropriate for
6824 an interpreter that uses scm_compile_shell_switches to handle its
6825 command-line arguments.
6827 void scm_shell_usage (int FATAL, char *MESSAGE)
6828 Print a usage message to the standard error output. If MESSAGE is
6829 non-zero, write it before the usage message, followed by a newline.
6830 If FATAL is non-zero, exit the process, using FATAL as the
6831 termination status. (If you want to be compatible with Guile,
6832 always use 1 as the exit status when terminating due to command-line
6835 You will usually want to use scm_shell instead of calling this
6838 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6839 expressions. It used to return SCM_EOL. Earth-shattering.
6841 ** The macros for declaring scheme objects in C code have been
6842 rearranged slightly. They are now:
6844 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6845 Declare a static SCM variable named C_NAME, and initialize it to
6846 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6847 be a C identifier, and SCHEME_NAME should be a C string.
6849 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6850 Just like SCM_SYMBOL, but make C_NAME globally visible.
6852 SCM_VCELL (C_NAME, SCHEME_NAME)
6853 Create a global variable at the Scheme level named SCHEME_NAME.
6854 Declare a static SCM variable named C_NAME, and initialize it to
6855 point to the Scheme variable's value cell.
6857 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6858 Just like SCM_VCELL, but make C_NAME globally visible.
6860 The `guile-snarf' script writes initialization code for these macros
6861 to its standard output, given C source code as input.
6863 The SCM_GLOBAL macro is gone.
6865 ** The scm_read_line and scm_read_line_x functions have been replaced
6866 by Scheme code based on the %read-delimited! procedure (known to C
6867 code as scm_read_delimited_x). See its description above for more
6870 ** The function scm_sys_open has been renamed to scm_open. It now
6871 returns a port instead of an FD object.
6873 * The dynamic linking support has changed. For more information, see
6874 libguile/DYNAMIC-LINKING.
6879 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6882 * Changes to the 'guile' program:
6884 ** Guile now loads some new files when it starts up. Guile first
6885 searches the load path for init.scm, and loads it if found. Then, if
6886 Guile is not being used to execute a script, and the user's home
6887 directory contains a file named `.guile', Guile loads that.
6889 ** You can now use Guile as a shell script interpreter.
6891 To paraphrase the SCSH manual:
6893 When Unix tries to execute an executable file whose first two
6894 characters are the `#!', it treats the file not as machine code to
6895 be directly executed by the native processor, but as source code
6896 to be executed by some interpreter. The interpreter to use is
6897 specified immediately after the #! sequence on the first line of
6898 the source file. The kernel reads in the name of the interpreter,
6899 and executes that instead. It passes the interpreter the source
6900 filename as its first argument, with the original arguments
6901 following. Consult the Unix man page for the `exec' system call
6902 for more information.
6904 Now you can use Guile as an interpreter, using a mechanism which is a
6905 compatible subset of that provided by SCSH.
6907 Guile now recognizes a '-s' command line switch, whose argument is the
6908 name of a file of Scheme code to load. It also treats the two
6909 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6910 to make a file of Scheme code directly executable by Unix, insert the
6911 following two lines at the top of the file:
6913 #!/usr/local/bin/guile -s
6916 Guile treats the argument of the `-s' command-line switch as the name
6917 of a file of Scheme code to load, and treats the sequence `#!' as the
6918 start of a block comment, terminated by `!#'.
6920 For example, here's a version of 'echo' written in Scheme:
6922 #!/usr/local/bin/guile -s
6924 (let loop ((args (cdr (program-arguments))))
6927 (display (car args))
6928 (if (pair? (cdr args))
6930 (loop (cdr args)))))
6933 Why does `#!' start a block comment terminated by `!#', instead of the
6934 end of the line? That is the notation SCSH uses, and although we
6935 don't yet support the other SCSH features that motivate that choice,
6936 we would like to be backward-compatible with any existing Guile
6937 scripts once we do. Furthermore, if the path to Guile on your system
6938 is too long for your kernel, you can start the script with this
6942 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6945 Note that some very old Unix systems don't support the `#!' syntax.
6948 ** You can now run Guile without installing it.
6950 Previous versions of the interactive Guile interpreter (`guile')
6951 couldn't start up unless Guile's Scheme library had been installed;
6952 they used the value of the environment variable `SCHEME_LOAD_PATH'
6953 later on in the startup process, but not to find the startup code
6954 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6957 To run Guile without installing it, build it in the normal way, and
6958 then set the environment variable `SCHEME_LOAD_PATH' to a
6959 colon-separated list of directories, including the top-level directory
6960 of the Guile sources. For example, if you unpacked Guile so that the
6961 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6964 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6967 ** Guile's read-eval-print loop no longer prints #<unspecified>
6968 results. If the user wants to see this, she can evaluate the
6969 expression (assert-repl-print-unspecified #t), perhaps in her startup
6972 ** Guile no longer shows backtraces by default when an error occurs;
6973 however, it does display a message saying how to get one, and how to
6974 request that they be displayed by default. After an error, evaluate
6976 to see a backtrace, and
6977 (debug-enable 'backtrace)
6978 to see them by default.
6982 * Changes to Guile Scheme:
6984 ** Guile now distinguishes between #f and the empty list.
6986 This is for compatibility with the IEEE standard, the (possibly)
6987 upcoming Revised^5 Report on Scheme, and many extant Scheme
6990 Guile used to have #f and '() denote the same object, to make Scheme's
6991 type system more compatible with Emacs Lisp's. However, the change
6992 caused too much trouble for Scheme programmers, and we found another
6993 way to reconcile Emacs Lisp with Scheme that didn't require this.
6996 ** Guile's delq, delv, delete functions, and their destructive
6997 counterparts, delq!, delv!, and delete!, now remove all matching
6998 elements from the list, not just the first. This matches the behavior
6999 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7000 functions which inspired them.
7002 I recognize that this change may break code in subtle ways, but it
7003 seems best to make the change before the FSF's first Guile release,
7007 ** The compiled-library-path function has been deleted from libguile.
7009 ** The facilities for loading Scheme source files have changed.
7011 *** The variable %load-path now tells Guile which directories to search
7012 for Scheme code. Its value is a list of strings, each of which names
7015 *** The variable %load-extensions now tells Guile which extensions to
7016 try appending to a filename when searching the load path. Its value
7017 is a list of strings. Its default value is ("" ".scm").
7019 *** (%search-load-path FILENAME) searches the directories listed in the
7020 value of the %load-path variable for a Scheme file named FILENAME,
7021 with all the extensions listed in %load-extensions. If it finds a
7022 match, then it returns its full filename. If FILENAME is absolute, it
7023 returns it unchanged. Otherwise, it returns #f.
7025 %search-load-path will not return matches that refer to directories.
7027 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7028 uses %seach-load-path to find a file named FILENAME, and loads it if
7029 it finds it. If it can't read FILENAME for any reason, it throws an
7032 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7035 *** load uses the same searching semantics as primitive-load.
7037 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7038 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7039 path, and load-module-with-path have been deleted. The functions
7040 above should serve their purposes.
7042 *** If the value of the variable %load-hook is a procedure,
7043 `primitive-load' applies its value to the name of the file being
7044 loaded (without the load path directory name prepended). If its value
7045 is #f, it is ignored. Otherwise, an error occurs.
7047 This is mostly useful for printing load notification messages.
7050 ** The function `eval!' is no longer accessible from the scheme level.
7051 We can't allow operations which introduce glocs into the scheme level,
7052 because Guile's type system can't handle these as data. Use `eval' or
7053 `read-and-eval!' (see below) as replacement.
7055 ** The new function read-and-eval! reads an expression from PORT,
7056 evaluates it, and returns the result. This is more efficient than
7057 simply calling `read' and `eval', since it is not necessary to make a
7058 copy of the expression for the evaluator to munge.
7060 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7061 for the `read' function.
7064 ** The function `int?' has been removed; its definition was identical
7065 to that of `integer?'.
7067 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7068 use the R4RS names for these functions.
7070 ** The function object-properties no longer returns the hash handle;
7071 it simply returns the object's property list.
7073 ** Many functions have been changed to throw errors, instead of
7074 returning #f on failure. The point of providing exception handling in
7075 the language is to simplify the logic of user code, but this is less
7076 useful if Guile's primitives don't throw exceptions.
7078 ** The function `fileno' has been renamed from `%fileno'.
7080 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7083 * Changes to Guile's C interface:
7085 ** The library's initialization procedure has been simplified.
7086 scm_boot_guile now has the prototype:
7088 void scm_boot_guile (int ARGC,
7090 void (*main_func) (),
7093 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7094 MAIN_FUNC should do all the work of the program (initializing other
7095 packages, reading user input, etc.) before returning. When MAIN_FUNC
7096 returns, call exit (0); this function never returns. If you want some
7097 other exit value, MAIN_FUNC may call exit itself.
7099 scm_boot_guile arranges for program-arguments to return the strings
7100 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7101 scm_set_program_arguments with the final list, so Scheme code will
7102 know which arguments have been processed.
7104 scm_boot_guile establishes a catch-all catch handler which prints an
7105 error message and exits the process. This means that Guile exits in a
7106 coherent way when system errors occur and the user isn't prepared to
7107 handle it. If the user doesn't like this behavior, they can establish
7108 their own universal catcher in MAIN_FUNC to shadow this one.
7110 Why must the caller do all the real work from MAIN_FUNC? The garbage
7111 collector assumes that all local variables of type SCM will be above
7112 scm_boot_guile's stack frame on the stack. If you try to manipulate
7113 SCM values after this function returns, it's the luck of the draw
7114 whether the GC will be able to find the objects you allocate. So,
7115 scm_boot_guile function exits, rather than returning, to discourage
7116 people from making that mistake.
7118 The IN, OUT, and ERR arguments were removed; there are other
7119 convenient ways to override these when desired.
7121 The RESULT argument was deleted; this function should never return.
7123 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7127 ** Guile's header files should no longer conflict with your system's
7130 In order to compile code which #included <libguile.h>, previous
7131 versions of Guile required you to add a directory containing all the
7132 Guile header files to your #include path. This was a problem, since
7133 Guile's header files have names which conflict with many systems'
7136 Now only <libguile.h> need appear in your #include path; you must
7137 refer to all Guile's other header files as <libguile/mumble.h>.
7138 Guile's installation procedure puts libguile.h in $(includedir), and
7139 the rest in $(includedir)/libguile.
7142 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7143 have been added to the Guile library.
7145 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7146 OBJ will not be freed, even if all other references are dropped,
7147 until someone does scm_unprotect_object (OBJ). Both functions
7150 Note that calls to scm_protect_object do not nest. You can call
7151 scm_protect_object any number of times on a given object, and the
7152 next call to scm_unprotect_object will unprotect it completely.
7154 Basically, scm_protect_object and scm_unprotect_object just
7155 maintain a list of references to things. Since the GC knows about
7156 this list, all objects it mentions stay alive. scm_protect_object
7157 adds its argument to the list; scm_unprotect_object remove its
7158 argument from the list.
7161 ** scm_eval_0str now returns the value of the last expression
7164 ** The new function scm_read_0str reads an s-expression from a
7165 null-terminated string, and returns it.
7167 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7168 to a Scheme port object.
7170 ** The new function `scm_set_program_arguments' allows C code to set
7171 the value returned by the Scheme `program-arguments' function.
7176 * Guile no longer includes sophisticated Tcl/Tk support.
7178 The old Tcl/Tk support was unsatisfying to us, because it required the
7179 user to link against the Tcl library, as well as Tk and Guile. The
7180 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7181 referring to widgets by names, rather than exporting widgets to Scheme
7182 code as a special datatype.
7184 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7185 maintainers described some very interesting changes in progress to the
7186 Tcl/Tk internals, which would facilitate clean interfaces between lone
7187 Tk and other interpreters --- even for garbage-collected languages
7188 like Scheme. They expected the new Tk to be publicly available in the
7191 Since it seems that Guile might soon have a new, cleaner interface to
7192 lone Tk, and that the old Guile/Tk glue code would probably need to be
7193 completely rewritten, we (Jim Blandy and Richard Stallman) have
7194 decided not to support the old code. We'll spend the time instead on
7195 a good interface to the newer Tk, as soon as it is available.
7197 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7200 Copyright information:
7202 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7204 Permission is granted to anyone to make or distribute verbatim copies
7205 of this document as received, in any medium, provided that the
7206 copyright notice and this permission notice are preserved,
7207 thus giving the recipient permission to redistribute in turn.
7209 Permission is granted to distribute modified versions
7210 of this document, or of portions of it,
7211 under the above conditions, provided also that they
7212 carry prominent notices stating who last changed them.
7217 paragraph-separate: "[
\f]*$"