1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
7 Each release reports the NEWS in the following sections:
9 * Changes to the distribution
10 * Changes to the stand-alone interpreter
11 * Changes to Scheme functions and syntax
12 * Changes to the C interface
15 Changes in 1.9.XXXXXXXX:
17 * Changes to the distribution
18 * Changes to the stand-alone interpreter
19 * Changes to Scheme functions and syntax
20 * Changes to the C interface
23 Changes in 1.8.1 (since 1.8.0):
25 * Changes to the distribution
27 ** New primitive-_exit giving the _exit() system call.
29 * Changes to Scheme functions and syntax
31 ** A one-dimensional array can now be 'equal?' to a vector.
32 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
33 ** SRFI-14 standard char sets are now recomputed upon successful `setlocale'.
35 * Changes to the C interface
37 ** New function scm_c_locale_stringn_to_number.
41 ** array-set! with bit vector.
42 ** make-shared-array fixes, including examples in the manual which failed.
43 ** string<? and friends follow char<? etc order on 8-bit chars.
44 ** n-par-for-each, n-for-each-par-map for "futures" variable.
45 ** module autoload and explicit use-modules cooperate.
46 ** ice-9 format ~f with infs and nans.
47 ** exact->inexact overflows on fractions with big num/den but small result.
48 ** srfi-1 assoc "=" procedure argument order.
49 ** Build problems on MacOS, SunOS, QNX.
52 Changes since the 1.6.x series:
54 * Changes to the distribution
56 ** Guile is now licensed with the GNU Lesser General Public License.
58 ** The manual is now licensed with the GNU Free Documentation License.
60 ** Guile now requires GNU MP (http://swox.com/gmp).
62 Guile now uses the GNU MP library for arbitrary precision arithmetic.
64 ** Guile now has separate private and public configuration headers.
66 That is, things like HAVE_STRING_H no longer leak from Guile's
69 ** Guile now provides and uses an "effective" version number.
71 Guile now provides scm_effective_version and effective-version
72 functions which return the "effective" version number. This is just
73 the normal full version string without the final micro-version number,
74 so the current effective-version is "1.8". The effective version
75 should remain unchanged during a stable series, and should be used for
76 items like the versioned share directory name
77 i.e. /usr/share/guile/1.8.
79 Providing an unchanging version number during a stable release for
80 things like the versioned share directory can be particularly
81 important for Guile "add-on" packages, since it provides a directory
82 that they can install to that won't be changed out from under them
83 with each micro release during a stable series.
85 ** Thread implementation has changed.
87 When you configure "--with-threads=null", you will get the usual
88 threading API (call-with-new-thread, make-mutex, etc), but you can't
89 actually create new threads. Also, "--with-threads=no" is now
90 equivalent to "--with-threads=null". This means that the thread API
91 is always present, although you might not be able to create new
94 When you configure "--with-threads=pthreads" or "--with-threads=yes",
95 you will get threads that are implemented with the portable POSIX
96 threads. These threads can run concurrently (unlike the previous
97 "coop" thread implementation), but need to cooperate for things like
100 The default is "pthreads", unless your platform doesn't have pthreads,
101 in which case "null" threads are used.
103 See the manual for details, nodes "Initialization", "Multi-Threading",
104 "Blocking", and others.
106 ** There is the new notion of 'discouraged' features.
108 This is a milder form of deprecation.
110 Things that are discouraged should not be used in new code, but it is
111 OK to leave them in old code for now. When a discouraged feature is
112 used, no warning message is printed like there is for 'deprecated'
113 features. Also, things that are merely discouraged are nevertheless
114 implemented efficiently, while deprecated features can be very slow.
116 You can omit discouraged features from libguile by configuring it with
117 the '--disable-discouraged' option.
119 ** Deprecation warnings can be controlled at run-time.
121 (debug-enable 'warn-deprecated) switches them on and (debug-disable
122 'warn-deprecated) switches them off.
124 ** Support for SRFI 61, extended cond syntax for multiple values has
127 This SRFI is always available.
129 ** Support for require-extension, SRFI-55, has been added.
131 The SRFI-55 special form `require-extension' has been added. It is
132 available at startup, and provides a portable way to load Scheme
133 extensions. SRFI-55 only requires support for one type of extension,
134 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
137 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
139 The (srfi srfi-26) module is an implementation of SRFI-26 which
140 provides the `cut' and `cute' syntax. These may be used to specialize
141 parameters without currying.
143 ** New module (srfi srfi-31)
145 This is an implementation of SRFI-31 which provides a special form
146 `rec' for recursive evaluation.
148 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
149 been merged with the core, making their functionality always
152 The modules are still available, tho, and you could use them together
153 with a renaming import, for example.
155 ** Guile no longer includes its own version of libltdl.
157 The official version is good enough now.
159 ** The --enable-htmldoc option has been removed from 'configure'.
161 Support for translating the documentation into HTML is now always
162 provided. Use 'make html'.
164 ** New module (ice-9 serialize):
166 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
167 don't trust the thread safety of most of your program, but where you
168 have some section(s) of code which you consider can run in parallel to
169 other sections. See ice-9/serialize.scm for more information.
171 ** The configure option '--disable-arrays' has been removed.
173 Support for arrays and uniform numeric arrays is now always included
176 * Changes to the stand-alone interpreter
178 ** New command line option `-L'.
180 This option adds a directory to the front of the load path.
182 ** New command line option `--no-debug'.
184 Specifying `--no-debug' on the command line will keep the debugging
185 evaluator turned off, even for interactive sessions.
187 ** User-init file ~/.guile is now loaded with the debugging evaluator.
189 Previously, the normal evaluator would have been used. Using the
190 debugging evaluator gives better error messages.
192 ** The '-e' option now 'read's its argument.
194 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
195 be used with '-e'. For example, you can now write a script like
198 exec guile -e '(@ (demo) main)' -s "$0" "$@"
201 (define-module (demo)
205 (format #t "Demo: ~a~%" args))
208 * Changes to Scheme functions and syntax
210 ** Guardians have changed back to their original semantics
212 Guardians now behave like described in the paper by Dybvig et al. In
213 particular, they no longer make guarantees about the order in which
214 they return objects, and they can no longer be greedy.
216 They no longer drop cyclic data structures.
218 The C function scm_make_guardian has been changed incompatibly and no
219 longer takes the 'greedy_p' argument.
221 ** New function hashx-remove!
223 This function completes the set of 'hashx' functions.
225 ** The concept of dynamic roots has been factored into continuation
226 barriers and dynamic states.
228 Each thread has a current dynamic state that carries the values of the
229 fluids. You can create and copy dynamic states and use them as the
230 second argument for 'eval'. See "Fluids and Dynamic States" in the
233 To restrict the influence that captured continuations can have on the
234 control flow, you can errect continuation barriers. See "Continuation
235 Barriers" in the manual.
237 The function call-with-dynamic-root now essentially temporarily
238 installs a new dynamic state and errects a continuation barrier.
240 ** The default load path no longer includes "." at the end.
242 Automatically loading modules from the current directory should not
243 happen by default. If you want to allow it in a more controlled
244 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
247 ** The uniform vector and array support has been overhauled.
249 It now complies with SRFI-4 and the weird prototype based uniform
250 array creation has been deprecated. See the manual for more details.
252 Some non-compatible changes have been made:
253 - characters can no longer be stored into byte arrays.
254 - strings and bit vectors are no longer considered to be uniform numeric
256 - array-rank throws an error for non-arrays instead of returning zero.
257 - array-ref does no longer accept non-arrays when no indices are given.
259 There is the new notion of 'generalized vectors' and corresponding
260 procedures like 'generalized-vector-ref'. Generalized vectors include
261 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
263 Arrays use generalized vectors as their storage, so that you still
264 have arrays of characters, bits, etc. However, uniform-array-read!
265 and uniform-array-write can no longer read/write strings and
268 ** There is now support for copy-on-write substrings, mutation-sharing
269 substrings and read-only strings.
271 Three new procedures are related to this: substring/shared,
272 substring/copy, and substring/read-only. See the manual for more
275 ** Backtraces will now highlight the value that caused the error.
277 By default, these values are enclosed in "{...}", such as in this
286 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
287 <unnamed port>:1:1: Wrong type (expecting pair): a
288 ABORT: (wrong-type-arg)
290 The prefix and suffix used for highlighting can be set via the two new
291 printer options 'highlight-prefix' and 'highlight-suffix'. For
292 example, putting this into ~/.guile will output the bad value in bold
295 (print-set! highlight-prefix "\x1b[1m")
296 (print-set! highlight-suffix "\x1b[22m")
299 ** 'gettext' support for internationalization has been added.
301 See the manual for details.
303 ** New syntax '@' and '@@':
305 You can now directly refer to variables exported from a module by
308 (@ MODULE-NAME VARIABLE-NAME)
310 For example (@ (ice-9 pretty-print) pretty-print) will directly access
311 the pretty-print variable exported from the (ice-9 pretty-print)
312 module. You don't need to 'use' that module first. You can also use
313 '@' as a target of 'set!', as in (set! (@ mod var) val).
315 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
316 but it can also access variables that have not been exported. It is
317 intended only for kluges and temporary fixes and for debugging, not
320 ** Keyword syntax has been made more disciplined.
322 Previously, the name of a keyword was read as a 'token' but printed as
323 a symbol. Now, it is read as a general Scheme datum which must be a
334 ERROR: In expression (a b c):
338 ERROR: Unbound variable: foo
343 ERROR: Wrong type (expecting symbol): 12
347 ERROR: Wrong type (expecting symbol): (a b c)
351 ** The printing of symbols that might look like keywords can be
354 The new printer option 'quote-keywordish-symbols' controls how symbols
355 are printed that have a colon as their first or last character. The
356 default now is to only quote a symbol with #{...}# when the read
357 option 'keywords' is not '#f'. Thus:
359 guile> (define foo (string->symbol ":foo"))
360 guile> (read-set! keywords #f)
363 guile> (read-set! keywords 'prefix)
366 guile> (print-set! quote-keywordish-symbols #f)
370 ** 'while' now provides 'break' and 'continue'
372 break and continue were previously bound in a while loop, but not
373 documented, and continue didn't quite work properly. The undocumented
374 parameter to break which gave a return value for the while has been
377 ** 'call-with-current-continuation' is now also available under the name
380 ** The module system now checks for duplicate bindings.
382 The module system now can check for name conflicts among imported
385 The behavior can be controlled by specifying one or more 'duplicates'
386 handlers. For example, to make Guile return an error for every name
394 The new default behavior of the module system when a name collision
395 has been detected is to
397 1. Give priority to bindings marked as a replacement.
398 2. Issue a warning (different warning if overriding core binding).
399 3. Give priority to the last encountered binding (this corresponds to
402 If you want the old behavior back without replacements or warnings you
405 (default-duplicate-binding-handler 'last)
407 to your .guile init file.
409 ** New define-module option: :replace
411 :replace works as :export, but, in addition, marks the binding as a
414 A typical example is `format' in (ice-9 format) which is a replacement
415 for the core binding `format'.
417 ** Adding prefixes to imported bindings in the module system
419 There is now a new :use-module option :prefix. It can be used to add
420 a prefix to all imported bindings.
423 :use-module ((bar) :prefix bar:))
425 will import all bindings exported from bar, but rename them by adding
428 ** Conflicting generic functions can be automatically merged.
430 When two imported bindings conflict and they are both generic
431 functions, the two functions can now be merged automatically. This is
432 activated with the 'duplicates' handler 'merge-generics'.
434 ** New function: effective-version
436 Returns the "effective" version number. This is just the normal full
437 version string without the final micro-version number. See "Changes
438 to the distribution" above.
440 ** New threading functions: parallel, letpar, par-map, and friends
442 These are convenient ways to run calculations in parallel in new
443 threads. See "Parallel forms" in the manual for details.
445 ** New function 'try-mutex'.
447 This function will attempt to lock a mutex but will return immediately
448 instead of blocking and indicate failure.
450 ** Waiting on a condition variable can have a timeout.
452 The function 'wait-condition-variable' now takes a third, optional
453 argument that specifies the point in time where the waiting should be
456 ** New function 'broadcast-condition-variable'.
458 ** New functions 'all-threads' and 'current-thread'.
460 ** Signals and system asyncs work better with threads.
462 The function 'sigaction' now takes a fourth, optional, argument that
463 specifies the thread that the handler should run in. When the
464 argument is omitted, the handler will run in the thread that called
467 Likewise, 'system-async-mark' takes a second, optional, argument that
468 specifies the thread that the async should run in. When it is
469 omitted, the async will run in the thread that called
472 C code can use the new functions scm_sigaction_for_thread and
473 scm_system_async_mark_for_thread to pass the new thread argument.
475 When a thread blocks on a mutex, a condition variable or is waiting
476 for IO to be possible, it will still execute system asyncs. This can
477 be used to interrupt such a thread by making it execute a 'throw', for
480 ** The function 'system-async' is deprecated.
482 You can now pass any zero-argument procedure to 'system-async-mark'.
483 The function 'system-async' will just return its argument unchanged
486 ** New functions 'call-with-blocked-asyncs' and
487 'call-with-unblocked-asyncs'
489 The expression (call-with-blocked-asyncs PROC) will call PROC and will
490 block execution of system asyncs for the current thread by one level
491 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
492 procedure and will unblock the execution of system asyncs by one
493 level for the current thread.
495 Only system asyncs are affected by these functions.
497 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
499 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
500 instead. Those functions are easier to use correctly and can be
503 ** New function 'unsetenv'.
505 ** New macro 'define-syntax-public'.
507 It works like 'define-syntax' and also exports the defined macro (but
510 ** There is support for Infinity and NaNs.
512 Following PLT Scheme, Guile can now work with infinite numbers, and
515 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
516 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
517 "+nan.0"). These numbers are inexact and have no exact counterpart.
519 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
520 sign of the dividend. The infinities are integers, and they answer #t
521 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
522 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
533 ERROR: Numerical overflow
535 Two new predicates 'inf?' and 'nan?' can be used to test for the
538 ** Inexact zero can have a sign.
540 Guile can now distinguish between plus and minus inexact zero, if your
541 platform supports this, too. The two zeros are equal according to
542 '=', but not according to 'eqv?'. For example
553 ** Guile now has exact rationals.
555 Guile can now represent fractions such as 1/3 exactly. Computing with
556 them is also done exactly, of course:
561 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
564 For example: (floor 2) now returns an exact 2 where in the past it
565 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
567 ** inexact->exact no longer returns only integers.
569 Without exact rationals, the closest exact number was always an
570 integer, but now inexact->exact returns the fraction that is exactly
571 equal to a floating point number. For example:
573 (inexact->exact 1.234)
574 => 694680242521899/562949953421312
576 When you want the old behavior, use 'round' explicitly:
578 (inexact->exact (round 1.234))
581 ** New function 'rationalize'.
583 This function finds a simple fraction that is close to a given real
584 number. For example (and compare with inexact->exact above):
586 (rationalize (inexact->exact 1.234) 1/2000)
589 Note that, as required by R5RS, rationalize returns only then an exact
590 result when both its arguments are exact.
592 ** 'odd?' and 'even?' work also for inexact integers.
594 Previously, (odd? 1.0) would signal an error since only exact integers
595 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
596 returns #f and (odd? 1.5) signals an error.
598 ** Guile now has uninterned symbols.
600 The new function 'make-symbol' will return an uninterned symbol. This
601 is a symbol that is unique and is guaranteed to remain unique.
602 However, uninterned symbols can not yet be read back in.
604 Use the new function 'symbol-interned?' to check whether a symbol is
607 ** pretty-print has more options.
609 The function pretty-print from the (ice-9 pretty-print) module can now
610 also be invoked with keyword arguments that control things like
611 maximum output width. See the manual for details.
613 ** Variables have no longer a special behavior for `equal?'.
615 Previously, comparing two variables with `equal?' would recursivly
616 compare their values. This is no longer done. Variables are now only
617 `equal?' if they are `eq?'.
619 ** `(begin)' is now valid.
621 You can now use an empty `begin' form. It will yield #<unspecified>
622 when evaluated and simply be ignored in a definition context.
624 ** Deprecated: procedure->macro
626 Change your code to use 'define-macro' or r5rs macros. Also, be aware
627 that macro expansion will not be done during evaluation, but prior to
630 ** Soft ports now allow a `char-ready?' procedure
632 The vector argument to `make-soft-port' can now have a length of
633 either 5 or 6. (Previously the length had to be 5.) The optional 6th
634 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
635 that returns the number of characters that can be read immediately
636 without the soft port blocking.
638 ** Deprecated: undefine
640 There is no replacement for undefine.
642 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
643 have been discouraged.
645 They are relics from a time where a keyword like #:foo was used
646 directly as a Tcl option "-foo" and thus keywords were internally
647 stored as a symbol with a starting dash. We now store a symbol
650 Use symbol->keyword and keyword->symbol instead.
652 ** The `cheap' debug option is now obsolete
654 Evaluator trap calls are now unconditionally "cheap" - in other words,
655 they pass a debug object to the trap handler rather than a full
656 continuation. The trap handler code can capture a full continuation
657 by using `call-with-current-continuation' in the usual way, if it so
660 The `cheap' option is retained for now so as not to break existing
661 code which gets or sets it, but setting it now has no effect. It will
662 be removed in the next major Guile release.
664 ** Evaluator trap calls now support `tweaking'
666 `Tweaking' means that the trap handler code can modify the Scheme
667 expression that is about to be evaluated (in the case of an
668 enter-frame trap) or the value that is being returned (in the case of
669 an exit-frame trap). The trap handler code indicates that it wants to
670 do this by returning a pair whose car is the symbol 'instead and whose
671 cdr is the modified expression or return value.
673 * Changes to the C interface
675 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
676 take a 'delete' function argument.
678 This argument makes no sense since the delete function is used to
679 remove a pair from an alist, and this must not be configurable.
681 This is an incompatible change.
683 ** The GH interface is now subject to the deprecation mechanism
685 The GH interface has been deprecated for quite some time but now it is
686 actually removed from Guile when it is configured with
687 --disable-deprecated.
689 See the manual "Transitioning away from GH" for more information.
691 ** A new family of functions for converting between C values and
692 Scheme values has been added.
694 These functions follow a common naming scheme and are designed to be
695 easier to use, thread-safe and more future-proof than the older
700 These are predicates that return a C boolean: 1 or 0. Instead of
701 SCM_NFALSEP, you can now use scm_is_true, for example.
703 - <type> scm_to_<type> (SCM val, ...)
705 These are functions that convert a Scheme value into an appropriate
706 C value. For example, you can use scm_to_int to safely convert from
709 - SCM scm_from_<type> (<type> val, ...)
711 These functions convert from a C type to a SCM value; for example,
712 scm_from_int for ints.
714 There is a huge number of these functions, for numbers, strings,
715 symbols, vectors, etc. They are documented in the reference manual in
716 the API section together with the types that they apply to.
718 ** New functions for dealing with complex numbers in C have been added.
720 The new functions are scm_c_make_rectangular, scm_c_make_polar,
721 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
722 They work like scm_make_rectangular etc but take or return doubles
725 ** The function scm_make_complex has been discouraged.
727 Use scm_c_make_rectangular instead.
729 ** The INUM macros have been deprecated.
731 A lot of code uses these macros to do general integer conversions,
732 although the macros only work correctly with fixnums. Use the
733 following alternatives.
735 SCM_INUMP -> scm_is_integer or similar
736 SCM_NINUMP -> !scm_is_integer or similar
737 SCM_MAKINUM -> scm_from_int or similar
738 SCM_INUM -> scm_to_int or similar
740 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
741 do the validating for you.
743 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
744 have been discouraged.
746 Use the newer scm_to_<type> and scm_from_<type> functions instead for
747 new code. The functions have been discouraged since they don't fit
750 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
752 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
753 evaluates its argument twice. Use scm_is_true, etc. instead for new
756 ** The macro SCM_EQ_P has been discouraged.
758 Use scm_is_eq for new code, which fits better into the naming
761 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
764 Use the function scm_is_pair or scm_is_null instead.
766 ** The functions scm_round and scm_truncate have been deprecated and
767 are now available as scm_c_round and scm_c_truncate, respectively.
769 These functions occupy the names that scm_round_number and
770 scm_truncate_number should have.
772 ** The functions scm_c_string2str, scm_c_substring2str, and
773 scm_c_symbol2str have been deprecated.
775 Use scm_to_locale_stringbuf or similar instead, maybe together with
778 ** New functions scm_c_make_string, scm_c_string_length,
779 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
780 scm_c_substring_shared, scm_c_substring_copy.
782 These are like scm_make_string, scm_length, etc. but are slightly
783 easier to use from C.
785 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
786 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
788 They export too many assumptions about the implementation of strings
789 and symbols that are no longer true in the presence of
790 mutation-sharing substrings and when Guile switches to some form of
793 When working with strings, it is often best to use the normal string
794 functions provided by Guile, such as scm_c_string_ref,
795 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
796 manual since many more such functions are now provided than
799 When you want to convert a SCM string to a C string, use the
800 scm_to_locale_string function or similar instead. For symbols, use
801 scm_symbol_to_string and then work with that string. Because of the
802 new string representation, scm_symbol_to_string does not need to copy
803 and is thus quite efficient.
805 ** Some string, symbol and keyword functions have been discouraged.
807 They don't fit into the uniform naming scheme and are not explicit
808 about the character encoding.
810 Replace according to the following table:
812 scm_allocate_string -> scm_c_make_string
813 scm_take_str -> scm_take_locale_stringn
814 scm_take0str -> scm_take_locale_string
815 scm_mem2string -> scm_from_locale_stringn
816 scm_str2string -> scm_from_locale_string
817 scm_makfrom0str -> scm_from_locale_string
818 scm_mem2symbol -> scm_from_locale_symboln
819 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
820 scm_str2symbol -> scm_from_locale_symbol
822 SCM_SYMBOL_HASH -> scm_hashq
823 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
825 scm_c_make_keyword -> scm_from_locale_keyword
827 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
828 now also available to C code.
830 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
832 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
833 the latter returns the true name of the keyword, not the 'dash name',
834 as SCM_KEYWORDSYM used to do.
836 ** A new way to access arrays in a thread-safe and efficient way has
839 See the manual, node "Accessing Arrays From C".
841 ** The old uniform vector and bitvector implementations have been
842 unceremoniously removed.
844 This implementation exposed the details of the tagging system of
845 Guile. Use the new C API explained in the manual in node "Uniform
846 Numeric Vectors" and "Bit Vectors", respectively.
848 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
849 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
850 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
851 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
852 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
853 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
856 ** The macros dealing with vectors have been deprecated.
858 Use the new functions scm_is_vector, scm_vector_elements,
859 scm_vector_writable_elements, etc, or scm_is_simple_vector,
860 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
861 manual for more details.
863 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
864 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
866 The following macros have been removed: SCM_VECTOR_BASE,
867 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
868 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
870 ** Some C functions and macros related to arrays have been deprecated.
872 Migrate according to the following table:
874 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
875 scm_make_ra -> scm_make_array
876 scm_shap2ra -> scm_make_array
877 scm_cvref -> scm_c_generalized_vector_ref
878 scm_ra_set_contp -> do not use
879 scm_aind -> scm_array_handle_pos
880 scm_raprin1 -> scm_display or scm_write
882 SCM_ARRAYP -> scm_is_array
883 SCM_ARRAY_NDIM -> scm_c_array_rank
884 SCM_ARRAY_DIMS -> scm_array_handle_dims
885 SCM_ARRAY_CONTP -> do not use
886 SCM_ARRAY_MEM -> do not use
887 SCM_ARRAY_V -> scm_array_handle_elements or similar
888 SCM_ARRAY_BASE -> do not use
890 ** SCM_CELL_WORD_LOC has been deprecated.
892 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
893 to a SCM, as opposed to a pointer to a scm_t_bits.
895 This was done to allow the correct use of pointers into the Scheme
896 heap. Previously, the heap words were of type scm_t_bits and local
897 variables and function arguments were of type SCM, making it
898 non-standards-conformant to have a pointer that can point to both.
900 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
902 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
903 second and third words of double smobs. Likewise for
904 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
906 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
907 used to get and set the 16 exra bits in the zeroth word of a smob.
909 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
910 accesing the first immediate word of a smob as a SCM value, and there
911 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
912 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
914 ** New way to deal with non-local exits and re-entries.
916 There is a new set of functions that essentially do what
917 scm_internal_dynamic_wind does, but in a way that is more convenient
918 for C code in some situations. Here is a quick example of how to
919 prevent a potential memory leak:
926 scm_dynwind_begin (0);
928 mem = scm_malloc (100);
929 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
931 /* MEM would leak if BAR throws an error.
932 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
939 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
940 SCM_DYNWIND_END as well.
944 For full documentation, see the node "Dynamic Wind" in the manual.
946 ** New function scm_dynwind_free
948 This function calls 'free' on a given pointer when a dynwind context
949 is left. Thus the call to scm_dynwind_unwind_handler above could be
950 replaced with simply scm_dynwind_free (mem).
952 ** New functions scm_c_call_with_blocked_asyncs and
953 scm_c_call_with_unblocked_asyncs
955 Like scm_call_with_blocked_asyncs etc. but for C functions.
957 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
959 In addition to scm_c_call_with_blocked_asyncs you can now also use
960 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
961 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
963 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
964 SCM_REALLOW_INTS have been deprecated.
966 They do no longer fulfill their original role of blocking signal
967 delivery. Depending on what you want to achieve, replace a pair of
968 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
969 mutex, blocks asyncs, or both. See node "Critical Sections" in the
972 ** The value 'scm_mask_ints' is no longer writable.
974 Previously, you could set scm_mask_ints directly. This is no longer
975 possible. Use scm_c_call_with_blocked_asyncs and
976 scm_c_call_with_unblocked_asyncs instead.
978 ** New way to temporarily set the current input, output or error ports
980 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
981 context' (see above). <foo> is one of "input", "output" or "error".
983 ** New way to temporarily set fluids
985 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
986 above) to temporarily set the value of a fluid.
988 ** New types scm_t_intmax and scm_t_uintmax.
990 On platforms that have them, these types are identical to intmax_t and
991 uintmax_t, respectively. On other platforms, they are identical to
992 the largest integer types that Guile knows about.
994 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
996 You should not have used them.
998 ** Many public #defines with generic names have been made private.
1000 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1001 private or renamed with a more suitable public name.
1003 ** The macro SCM_TYP16S has been deprecated.
1005 This macro is not intended for public use.
1007 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1009 Use scm_is_true (scm_inexact_p (...)) instead.
1011 ** The macro SCM_SLOPPY_REALP has been deprecated.
1013 Use scm_is_real instead.
1015 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1017 Use scm_is_complex instead.
1019 ** Some preprocessor defines have been deprecated.
1021 These defines indicated whether a certain feature was present in Guile
1022 or not. Going forward, assume that the features are always present.
1024 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1025 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1027 The following macros have been removed completely: MEMOIZE_LOCALS,
1028 SCM_RECKLESS, SCM_CAUTIOUS.
1030 ** The preprocessor define STACK_DIRECTION has been deprecated.
1032 There should be no need to know about the stack direction for ordinary
1035 ** New function: scm_effective_version
1037 Returns the "effective" version number. This is just the normal full
1038 version string without the final micro-version number. See "Changes
1039 to the distribution" above.
1041 ** The function scm_call_with_new_thread has a new prototype.
1043 Instead of taking a list with the thunk and handler, these two
1044 arguments are now passed directly:
1046 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1048 This is an incompatible change.
1050 ** New snarfer macro SCM_DEFINE_PUBLIC.
1052 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1053 function in the init section.
1055 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1057 ** Garbage collector rewrite.
1059 The garbage collector is cleaned up a lot, and now uses lazy
1060 sweeping. This is reflected in the output of (gc-stats); since cells
1061 are being freed when they are allocated, the cells-allocated field
1062 stays roughly constant.
1064 For malloc related triggers, the behavior is changed. It uses the same
1065 heuristic as the cell-triggered collections. It may be tuned with the
1066 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1067 for minimum yield of malloc related triggers. The default is 40.
1068 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1071 Debugging operations for the freelist have been deprecated, along with
1072 the C variables that control garbage collection. The environment
1073 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1074 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1076 For understanding the memory usage of a GUILE program, the routine
1077 gc-live-object-stats returns an alist containing the number of live
1078 objects for every type.
1081 ** The function scm_definedp has been renamed to scm_defined_p
1083 The name scm_definedp is deprecated.
1085 ** The struct scm_cell type has been renamed to scm_t_cell
1087 This is in accordance to Guile's naming scheme for types. Note that
1088 the name scm_cell is now used for a function that allocates and
1089 initializes a new cell (see below).
1091 ** New functions for memory management
1093 A new set of functions for memory management has been added since the
1094 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1095 indeed, Guile itself contained some long standing bugs that could
1096 cause aborts in long running programs.
1098 The new functions are more symmetrical and do not need cooperation
1099 from smob free routines, among other improvements.
1101 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1102 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1103 scm_gc_free, scm_gc_register_collectable_memory, and
1104 scm_gc_unregister_collectable_memory. Refer to the manual for more
1105 details and for upgrading instructions.
1107 The old functions for memory management have been deprecated. They
1108 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1109 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1111 ** Declarations of exported features are marked with SCM_API.
1113 Every declaration of a feature that belongs to the exported Guile API
1114 has been marked by adding the macro "SCM_API" to the start of the
1115 declaration. This macro can expand into different things, the most
1116 common of which is just "extern" for Unix platforms. On Win32, it can
1117 be used to control which symbols are exported from a DLL.
1119 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1120 will expand into "__declspec (dllimport) extern", which is needed for
1121 linking to the Guile DLL in Windows.
1123 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1124 SCM_SRFI4_IMPORT, for the corresponding libraries.
1126 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1128 Use the new functions scm_cell and scm_double_cell instead. The old
1129 macros had problems because with them allocation and initialization
1130 was separated and the GC could sometimes observe half initialized
1131 cells. Only careful coding by the user of SCM_NEWCELL and
1132 SCM_NEWCELL2 could make this safe and efficient.
1134 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1136 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1139 ** SRCBRKP has been deprecated.
1141 Use scm_c_source_property_breakpoint_p instead.
1143 ** Deprecated: scm_makmacro
1145 Change your code to use either scm_makmmacro or to define macros in
1146 Scheme, using 'define-macro'.
1148 ** New function scm_c_port_for_each.
1150 This function is like scm_port_for_each but takes a pointer to a C
1151 function as the callback instead of a SCM value.
1153 ** The names scm_internal_select, scm_thread_sleep, and
1154 scm_thread_usleep have been discouraged.
1156 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1158 ** The GC can no longer be blocked.
1160 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1161 The GC can now run (partially) concurrently with other code and thus
1162 blocking it is not well defined.
1164 ** Many definitions have been removed that were previously deprecated.
1166 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1167 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1168 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1169 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1170 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1171 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1172 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1173 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1174 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1175 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1176 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1177 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1178 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1179 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1180 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1181 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1182 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1183 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1184 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1185 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1186 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1187 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1188 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1189 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1190 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1191 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1192 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1193 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1194 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1195 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1196 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1198 * Changes to bundled modules
1202 Using the (ice-9 debug) module no longer automatically switches Guile
1203 to use the debugging evaluator. If you want to switch to the
1204 debugging evaluator (which is needed for backtrace information if you
1205 hit an error), please add an explicit "(debug-enable 'debug)" to your
1206 code just after the code to use (ice-9 debug).
1209 Changes since Guile 1.4:
1211 * Changes to the distribution
1213 ** A top-level TODO file is included.
1215 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1217 Guile now always uses three numbers to represent the version,
1218 i.e. "1.6.5". The first number, 1, is the major version number, the
1219 second number, 6, is the minor version number, and the third number,
1220 5, is the micro version number. Changes in major version number
1221 indicate major changes in Guile.
1223 Minor version numbers that are even denote stable releases, and odd
1224 minor version numbers denote development versions (which may be
1225 unstable). The micro version number indicates a minor sub-revision of
1226 a given MAJOR.MINOR release.
1228 In keeping with the new scheme, (minor-version) and scm_minor_version
1229 no longer return everything but the major version number. They now
1230 just return the minor version number. Two new functions
1231 (micro-version) and scm_micro_version have been added to report the
1232 micro version number.
1234 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1236 ** New preprocessor definitions are available for checking versions.
1238 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1239 SCM_MICRO_VERSION to the appropriate integer values.
1241 ** Guile now actively warns about deprecated features.
1243 The new configure option `--enable-deprecated=LEVEL' and the
1244 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1245 See INSTALL and README for more information.
1247 ** Guile is much more likely to work on 64-bit architectures.
1249 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1250 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1251 for the use of a test machine, and thanks to Stefan Jahn for ia64
1254 ** New functions: setitimer and getitimer.
1256 These implement a fairly direct interface to the libc functions of the
1259 ** The #. reader extension is now disabled by default.
1261 For safety reasons, #. evaluation is disabled by default. To
1262 re-enable it, set the fluid read-eval? to #t. For example:
1264 (fluid-set! read-eval? #t)
1266 but make sure you realize the potential security risks involved. With
1267 read-eval? enabled, reading a data file from an untrusted source can
1270 ** New SRFI modules have been added:
1272 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1275 (srfi srfi-1) is a library containing many useful pair- and list-processing
1278 (srfi srfi-2) exports and-let*.
1280 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1282 (srfi srfi-6) is a dummy module for now, since guile already provides
1283 all of the srfi-6 procedures by default: open-input-string,
1284 open-output-string, get-output-string.
1286 (srfi srfi-8) exports receive.
1288 (srfi srfi-9) exports define-record-type.
1290 (srfi srfi-10) exports define-reader-ctor and implements the reader
1293 (srfi srfi-11) exports let-values and let*-values.
1295 (srfi srfi-13) implements the SRFI String Library.
1297 (srfi srfi-14) implements the SRFI Character-Set Library.
1299 (srfi srfi-17) implements setter and getter-with-setter and redefines
1300 some accessor procedures as procedures with getters. (such as car,
1301 cdr, vector-ref etc.)
1303 (srfi srfi-19) implements the SRFI Time/Date Library.
1305 ** New scripts / "executable modules"
1307 Subdirectory "scripts" contains Scheme modules that are packaged to
1308 also be executable as scripts. At this time, these scripts are available:
1317 See README there for more info.
1319 These scripts can be invoked from the shell with the new program
1320 "guile-tools", which keeps track of installation directory for you.
1323 $ guile-tools display-commentary srfi/*.scm
1325 guile-tools is copied to the standard $bindir on "make install".
1327 ** New module (ice-9 stack-catch):
1329 stack-catch is like catch, but saves the current state of the stack in
1330 the fluid the-last-stack. This fluid can be useful when using the
1331 debugger and when re-throwing an error.
1333 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1335 This has been done to prevent problems on lesser operating systems
1336 that can't tolerate `*'s in file names. The exported macro continues
1337 to be named `and-let*', of course.
1339 On systems that support it, there is also a compatibility module named
1340 (ice-9 and-let*). It will go away in the next release.
1342 ** New modules (oop goops) etc.:
1345 (oop goops describe)
1347 (oop goops active-slot)
1348 (oop goops composite-slot)
1350 The Guile Object Oriented Programming System (GOOPS) has been
1351 integrated into Guile. For further information, consult the GOOPS
1352 manual and tutorial in the `doc' directory.
1354 ** New module (ice-9 rdelim).
1356 This exports the following procedures which were previously defined
1357 in the default environment:
1359 read-line read-line! read-delimited read-delimited! %read-delimited!
1360 %read-line write-line
1362 For backwards compatibility the definitions are still imported into the
1363 default environment in this version of Guile. However you should add:
1365 (use-modules (ice-9 rdelim))
1367 to any program which uses the definitions, since this may change in
1370 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1371 can be used for similar functionality.
1373 ** New module (ice-9 rw)
1375 This is a subset of the (scsh rw) module from guile-scsh. Currently
1376 it defines two procedures:
1378 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1380 Read characters from a port or file descriptor into a string STR.
1381 A port must have an underlying file descriptor -- a so-called
1382 fport. This procedure is scsh-compatible and can efficiently read
1385 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1387 Write characters from a string STR to a port or file descriptor.
1388 A port must have an underlying file descriptor -- a so-called
1389 fport. This procedure is mostly compatible and can efficiently
1390 write large strings.
1392 ** New module (ice-9 match)
1394 This module includes Andrew K. Wright's pattern matcher. See
1395 ice-9/match.scm for brief description or
1397 http://www.star-lab.com/wright/code.html
1399 for complete documentation.
1401 ** New module (ice-9 buffered-input)
1403 This module provides procedures to construct an input port from an
1404 underlying source of input that reads and returns its input in chunks.
1405 The underlying input source is a Scheme procedure, specified by the
1406 caller, which the port invokes whenever it needs more input.
1408 This is useful when building an input port whose back end is Readline
1409 or a UI element such as the GtkEntry widget.
1413 The reference and tutorial documentation that was previously
1414 distributed separately, as `guile-doc', is now included in the core
1415 Guile distribution. The documentation consists of the following
1418 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1421 - The Guile Reference Manual (guile.texi) contains (or is intended to
1422 contain) reference documentation on all aspects of Guile.
1424 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1425 reference documentation for using GOOPS, Guile's Object Oriented
1428 - The Revised^5 Report on the Algorithmic Language Scheme
1431 See the README file in the `doc' directory for more details.
1433 ** There are a couple of examples in the examples/ directory now.
1435 * Changes to the stand-alone interpreter
1437 ** New command line option `--use-srfi'
1439 Using this option, SRFI modules can be loaded on startup and be
1440 available right from the beginning. This makes programming portable
1441 Scheme programs easier.
1443 The option `--use-srfi' expects a comma-separated list of numbers,
1444 each representing a SRFI number to be loaded into the interpreter
1445 before starting evaluating a script file or the REPL. Additionally,
1446 the feature identifier for the loaded SRFIs is recognized by
1447 `cond-expand' when using this option.
1450 $ guile --use-srfi=8,13
1451 guile> (receive (x z) (values 1 2) (+ 1 2))
1453 guile> (string-pad "bla" 20)
1456 ** Guile now always starts up in the `(guile-user)' module.
1458 Previously, scripts executed via the `-s' option would run in the
1459 `(guile)' module and the repl would run in the `(guile-user)' module.
1460 Now every user action takes place in the `(guile-user)' module by
1463 * Changes to Scheme functions and syntax
1465 ** Character classifiers work for non-ASCII characters.
1467 The predicates `char-alphabetic?', `char-numeric?',
1468 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1469 no longer check whether their arguments are ASCII characters.
1470 Previously, a character would only be considered alphabetic when it
1471 was also ASCII, for example.
1473 ** Previously deprecated Scheme functions have been removed:
1475 tag - no replacement.
1476 fseek - replaced by seek.
1477 list* - replaced by cons*.
1479 ** It's now possible to create modules with controlled environments
1483 (use-modules (ice-9 safe))
1484 (define m (make-safe-module))
1485 ;;; m will now be a module containing only a safe subset of R5RS
1486 (eval '(+ 1 2) m) --> 3
1487 (eval 'load m) --> ERROR: Unbound variable: load
1489 ** Evaluation of "()", the empty list, is now an error.
1491 Previously, the expression "()" evaluated to the empty list. This has
1492 been changed to signal a "missing expression" error. The correct way
1493 to write the empty list as a literal constant is to use quote: "'()".
1495 ** New concept of `Guile Extensions'.
1497 A Guile Extension is just a ordinary shared library that can be linked
1498 at run-time. We found it advantageous to give this simple concept a
1499 dedicated name to distinguish the issues related to shared libraries
1500 from the issues related to the module system.
1502 *** New function: load-extension
1504 Executing (load-extension lib init) is mostly equivalent to
1506 (dynamic-call init (dynamic-link lib))
1508 except when scm_register_extension has been called previously.
1509 Whenever appropriate, you should use `load-extension' instead of
1510 dynamic-link and dynamic-call.
1512 *** New C function: scm_c_register_extension
1514 This function registers a initialization function for use by
1515 `load-extension'. Use it when you don't want specific extensions to
1516 be loaded as shared libraries (for example on platforms that don't
1517 support dynamic linking).
1519 ** Auto-loading of compiled-code modules is deprecated.
1521 Guile used to be able to automatically find and link a shared
1522 library to satisfy requests for a module. For example, the module
1523 `(foo bar)' could be implemented by placing a shared library named
1524 "foo/libbar.so" (or with a different extension) in a directory on the
1527 This has been found to be too tricky, and is no longer supported. The
1528 shared libraries are now called "extensions". You should now write a
1529 small Scheme file that calls `load-extension' to load the shared
1530 library and initialize it explicitly.
1532 The shared libraries themselves should be installed in the usual
1533 places for shared libraries, with names like "libguile-foo-bar".
1535 For example, place this into a file "foo/bar.scm"
1537 (define-module (foo bar))
1539 (load-extension "libguile-foo-bar" "foobar_init")
1541 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1543 `eval' is now R5RS, that is it takes two arguments.
1544 The second argument is an environment specifier, i.e. either
1546 (scheme-report-environment 5)
1547 (null-environment 5)
1548 (interaction-environment)
1554 ** The module system has been made more disciplined.
1556 The function `eval' will save and restore the current module around
1557 the evaluation of the specified expression. While this expression is
1558 evaluated, `(current-module)' will now return the right module, which
1559 is the module specified as the second argument to `eval'.
1561 A consequence of this change is that `eval' is not particularly
1562 useful when you want allow the evaluated code to change what module is
1563 designated as the current module and have this change persist from one
1564 call to `eval' to the next. The read-eval-print-loop is an example
1565 where `eval' is now inadequate. To compensate, there is a new
1566 function `primitive-eval' that does not take a module specifier and
1567 that does not save/restore the current module. You should use this
1568 function together with `set-current-module', `current-module', etc
1569 when you want to have more control over the state that is carried from
1570 one eval to the next.
1572 Additionally, it has been made sure that forms that are evaluated at
1573 the top level are always evaluated with respect to the current module.
1574 Previously, subforms of top-level forms such as `begin', `case',
1575 etc. did not respect changes to the current module although these
1576 subforms are at the top-level as well.
1578 To prevent strange behavior, the forms `define-module',
1579 `use-modules', `use-syntax', and `export' have been restricted to only
1580 work on the top level. The forms `define-public' and
1581 `defmacro-public' only export the new binding on the top level. They
1582 behave just like `define' and `defmacro', respectively, when they are
1583 used in a lexical environment.
1585 Also, `export' will no longer silently re-export bindings imported
1586 from a used module. It will emit a `deprecation' warning and will
1587 cease to perform any re-export in the next version. If you actually
1588 want to re-export bindings, use the new `re-export' in place of
1589 `export'. The new `re-export' will not make copies of variables when
1590 rexporting them, as `export' did wrongly.
1592 ** Module system now allows selection and renaming of imported bindings
1594 Previously, when using `use-modules' or the `#:use-module' clause in
1595 the `define-module' form, all the bindings (association of symbols to
1596 values) for imported modules were added to the "current module" on an
1597 as-is basis. This has been changed to allow finer control through two
1598 new facilities: selection and renaming.
1600 You can now select which of the imported module's bindings are to be
1601 visible in the current module by using the `:select' clause. This
1602 clause also can be used to rename individual bindings. For example:
1604 ;; import all bindings no questions asked
1605 (use-modules (ice-9 common-list))
1607 ;; import four bindings, renaming two of them;
1608 ;; the current module sees: every some zonk-y zonk-n
1609 (use-modules ((ice-9 common-list)
1611 (remove-if . zonk-y)
1612 (remove-if-not . zonk-n))))
1614 You can also programmatically rename all selected bindings using the
1615 `:renamer' clause, which specifies a proc that takes a symbol and
1616 returns another symbol. Because it is common practice to use a prefix,
1617 we now provide the convenience procedure `symbol-prefix-proc'. For
1620 ;; import four bindings, renaming two of them specifically,
1621 ;; and all four w/ prefix "CL:";
1622 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1623 (use-modules ((ice-9 common-list)
1625 (remove-if . zonk-y)
1626 (remove-if-not . zonk-n))
1627 :renamer (symbol-prefix-proc 'CL:)))
1629 ;; import four bindings, renaming two of them specifically,
1630 ;; and all four by upcasing.
1631 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1632 (define (upcase-symbol sym)
1633 (string->symbol (string-upcase (symbol->string sym))))
1635 (use-modules ((ice-9 common-list)
1637 (remove-if . zonk-y)
1638 (remove-if-not . zonk-n))
1639 :renamer upcase-symbol))
1641 Note that programmatic renaming is done *after* individual renaming.
1642 Also, the above examples show `use-modules', but the same facilities are
1643 available for the `#:use-module' clause of `define-module'.
1645 See manual for more info.
1647 ** The semantics of guardians have changed.
1649 The changes are for the most part compatible. An important criterion
1650 was to keep the typical usage of guardians as simple as before, but to
1651 make the semantics safer and (as a result) more useful.
1653 *** All objects returned from guardians are now properly alive.
1655 It is now guaranteed that any object referenced by an object returned
1656 from a guardian is alive. It's now impossible for a guardian to
1657 return a "contained" object before its "containing" object.
1659 One incompatible (but probably not very important) change resulting
1660 from this is that it is no longer possible to guard objects that
1661 indirectly reference themselves (i.e. are parts of cycles). If you do
1662 so accidentally, you'll get a warning.
1664 *** There are now two types of guardians: greedy and sharing.
1666 If you call (make-guardian #t) or just (make-guardian), you'll get a
1667 greedy guardian, and for (make-guardian #f) a sharing guardian.
1669 Greedy guardians are the default because they are more "defensive".
1670 You can only greedily guard an object once. If you guard an object
1671 more than once, once in a greedy guardian and the rest of times in
1672 sharing guardians, then it is guaranteed that the object won't be
1673 returned from sharing guardians as long as it is greedily guarded
1676 Guardians returned by calls to `make-guardian' can now take one more
1677 optional parameter, which says whether to throw an error in case an
1678 attempt is made to greedily guard an object that is already greedily
1679 guarded. The default is true, i.e. throw an error. If the parameter
1680 is false, the guardian invocation returns #t if guarding was
1681 successful and #f if it wasn't.
1683 Also, since greedy guarding is, in effect, a side-effecting operation
1684 on objects, a new function is introduced: `destroy-guardian!'.
1685 Invoking this function on a guardian renders it unoperative and, if
1686 the guardian is greedy, clears the "greedily guarded" property of the
1687 objects that were guarded by it, thus undoing the side effect.
1689 Note that all this hair is hardly very important, since guardian
1690 objects are usually permanent.
1692 ** Continuations created by call-with-current-continuation now accept
1693 any number of arguments, as required by R5RS.
1695 ** New function `issue-deprecation-warning'
1697 This function is used to display the deprecation messages that are
1698 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1701 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1705 ;; `id' is deprecated. Use `identity' instead.
1710 ** New syntax `begin-deprecated'
1712 When deprecated features are included (as determined by the configure
1713 option --enable-deprecated), `begin-deprecated' is identical to
1714 `begin'. When deprecated features are excluded, it always evaluates
1715 to `#f', ignoring the body forms.
1717 ** New function `make-object-property'
1719 This function returns a new `procedure with setter' P that can be used
1720 to attach a property to objects. When calling P as
1724 where `obj' is any kind of object, it attaches `val' to `obj' in such
1725 a way that it can be retrieved by calling P as
1729 This function will replace procedure properties, symbol properties and
1730 source properties eventually.
1732 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1734 Instead of #&optional, #&key, etc you should now use #:optional,
1735 #:key, etc. Since #:optional is a keyword, you can write it as just
1736 :optional when (read-set! keywords 'prefix) is active.
1738 The old reader syntax `#&' is still supported, but deprecated. It
1739 will be removed in the next release.
1741 ** New define-module option: pure
1743 Tells the module system not to include any bindings from the root
1748 (define-module (totally-empty-module)
1751 ** New define-module option: export NAME1 ...
1753 Export names NAME1 ...
1755 This option is required if you want to be able to export bindings from
1756 a module which doesn't import one of `define-public' or `export'.
1760 (define-module (foo)
1762 :use-module (ice-9 r5rs)
1765 ;;; Note that we're pure R5RS below this point!
1770 ** New function: object->string OBJ
1772 Return a Scheme string obtained by printing a given object.
1774 ** New function: port? X
1776 Returns a boolean indicating whether X is a port. Equivalent to
1777 `(or (input-port? X) (output-port? X))'.
1779 ** New function: file-port?
1781 Determines whether a given object is a port that is related to a file.
1783 ** New function: port-for-each proc
1785 Apply PROC to each port in the Guile port table in turn. The return
1786 value is unspecified. More specifically, PROC is applied exactly once
1787 to every port that exists in the system at the time PORT-FOR-EACH is
1788 invoked. Changes to the port table while PORT-FOR-EACH is running
1789 have no effect as far as PORT-FOR-EACH is concerned.
1791 ** New function: dup2 oldfd newfd
1793 A simple wrapper for the `dup2' system call. Copies the file
1794 descriptor OLDFD to descriptor number NEWFD, replacing the
1795 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1796 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1797 to move away ports which are using NEWFD. The return value is
1800 ** New function: close-fdes fd
1802 A simple wrapper for the `close' system call. Close file
1803 descriptor FD, which must be an integer. Unlike close (*note
1804 close: Ports and File Descriptors.), the file descriptor will be
1805 closed even if a port is using it. The return value is
1808 ** New function: crypt password salt
1810 Encrypts `password' using the standard unix password encryption
1813 ** New function: chroot path
1815 Change the root directory of the running process to `path'.
1817 ** New functions: getlogin, cuserid
1819 Return the login name or the user name of the current effective user
1822 ** New functions: getpriority which who, setpriority which who prio
1824 Get or set the priority of the running process.
1826 ** New function: getpass prompt
1828 Read a password from the terminal, first displaying `prompt' and
1831 ** New function: flock file operation
1833 Set/remove an advisory shared or exclusive lock on `file'.
1835 ** New functions: sethostname name, gethostname
1837 Set or get the hostname of the machine the current process is running
1840 ** New function: mkstemp! tmpl
1842 mkstemp creates a new unique file in the file system and returns a
1843 new buffered port open for reading and writing to the file. TMPL
1844 is a string specifying where the file should be created: it must
1845 end with `XXXXXX' and will be changed in place to return the name
1846 of the temporary file.
1848 ** New function: open-input-string string
1850 Return an input string port which delivers the characters from
1851 `string'. This procedure, together with `open-output-string' and
1852 `get-output-string' implements SRFI-6.
1854 ** New function: open-output-string
1856 Return an output string port which collects all data written to it.
1857 The data can then be retrieved by `get-output-string'.
1859 ** New function: get-output-string
1861 Return the contents of an output string port.
1863 ** New function: identity
1865 Return the argument.
1867 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1868 are represented in Scheme as integers with normal host byte ordering.
1870 ** New function: inet-pton family address
1872 Convert a printable string network address into an integer. Note that
1873 unlike the C version of this function, the result is an integer with
1874 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1877 (inet-pton AF_INET "127.0.0.1") => 2130706433
1878 (inet-pton AF_INET6 "::1") => 1
1880 ** New function: inet-ntop family address
1882 Convert an integer network address into a printable string. Note that
1883 unlike the C version of this function, the input is an integer with
1884 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1887 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1888 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1889 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1893 Use `identity' instead.
1899 ** Deprecated: return-it
1903 ** Deprecated: string-character-length
1905 Use `string-length' instead.
1907 ** Deprecated: flags
1909 Use `logior' instead.
1911 ** Deprecated: close-all-ports-except.
1913 This was intended for closing ports in a child process after a fork,
1914 but it has the undesirable side effect of flushing buffers.
1915 port-for-each is more flexible.
1917 ** The (ice-9 popen) module now attempts to set up file descriptors in
1918 the child process from the current Scheme ports, instead of using the
1919 current values of file descriptors 0, 1, and 2 in the parent process.
1921 ** Removed function: builtin-weak-bindings
1923 There is no such concept as a weak binding any more.
1925 ** Removed constants: bignum-radix, scm-line-incrementors
1927 ** define-method: New syntax mandatory.
1929 The new method syntax is now mandatory:
1931 (define-method (NAME ARG-SPEC ...) BODY ...)
1932 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1934 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1935 REST-ARG ::= ARG-NAME
1937 If you have old code using the old syntax, import
1938 (oop goops old-define-method) before (oop goops) as in:
1940 (use-modules (oop goops old-define-method) (oop goops))
1942 ** Deprecated function: builtin-variable
1943 Removed function: builtin-bindings
1945 There is no longer a distinction between builtin or other variables.
1946 Use module system operations for all variables.
1948 ** Lazy-catch handlers are no longer allowed to return.
1950 That is, a call to `throw', `error', etc is now guaranteed to not
1953 ** Bugfixes for (ice-9 getopt-long)
1955 This module is now tested using test-suite/tests/getopt-long.test.
1956 The following bugs have been fixed:
1958 *** Parsing for options that are specified to have `optional' args now checks
1959 if the next element is an option instead of unconditionally taking it as the
1962 *** An error is now thrown for `--opt=val' when the option description
1963 does not specify `(value #t)' or `(value optional)'. This condition used to
1964 be accepted w/o error, contrary to the documentation.
1966 *** The error message for unrecognized options is now more informative.
1967 It used to be "not a record", an artifact of the implementation.
1969 *** The error message for `--opt' terminating the arg list (no value), when
1970 `(value #t)' is specified, is now more informative. It used to be "not enough
1973 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1974 The expansion used to be like so:
1976 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1978 Note that the "5d" is dropped. Now it is like so:
1980 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1982 This enables single-char options to have adjoining arguments as long as their
1983 constituent characters are not potential single-char options.
1985 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1987 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1988 property `arglist', which can be retrieved by `arity'. The result is that
1989 `arity' can give more detailed information than before:
1993 guile> (use-modules (ice-9 optargs))
1994 guile> (define* (foo #:optional a b c) a)
1996 0 or more arguments in `lambda*:G0'.
2001 3 optional arguments: `a', `b' and `c'.
2002 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2004 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2005 and `d', other keywords allowed.
2006 guile> (define* (baz a b #:optional c #:rest r) a)
2008 2 required arguments: `a' and `b', 1 optional argument: `c',
2011 * Changes to the C interface
2013 ** Types have been renamed from scm_*_t to scm_t_*.
2015 This has been done for POSIX sake. It reserves identifiers ending
2016 with "_t". What a concept.
2018 The old names are still available with status `deprecated'.
2020 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2022 ** Deprecated features have been removed.
2026 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2027 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2029 *** C Functions removed
2031 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2032 scm_fseek - replaced by scm_seek.
2033 gc-thunk - replaced by after-gc-hook.
2034 gh_int2scmb - replaced by gh_bool2scm.
2035 scm_tc_dblr - replaced by scm_tc16_real.
2036 scm_tc_dblc - replaced by scm_tc16_complex.
2037 scm_list_star - replaced by scm_cons_star.
2039 ** Deprecated: scm_makfromstr
2041 Use scm_mem2string instead.
2043 ** Deprecated: scm_make_shared_substring
2045 Explicit shared substrings will disappear from Guile.
2047 Instead, "normal" strings will be implemented using sharing
2048 internally, combined with a copy-on-write strategy.
2050 ** Deprecated: scm_read_only_string_p
2052 The concept of read-only strings will disappear in next release of
2055 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2057 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2059 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2061 Call a procedure with the indicated number of arguments. See "Fly
2062 Evaluation" in the manual.
2064 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2066 Call a procedure with the indicated number of arguments and a list of
2067 further arguments. See "Fly Evaluation" in the manual.
2069 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2071 Create a list of the given number of elements. See "List
2072 Constructors" in the manual.
2074 ** Renamed function: scm_listify has been replaced by scm_list_n.
2076 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2077 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2079 Use functions scm_list_N instead.
2081 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2083 Used by an application to read arbitrary number of bytes from a port.
2084 Same semantics as libc read, except that scm_c_read only returns less
2085 than SIZE bytes if at end-of-file.
2087 Warning: Doesn't update port line and column counts!
2089 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2091 Used by an application to write arbitrary number of bytes to an SCM
2092 port. Similar semantics as libc write. However, unlike libc
2093 write, scm_c_write writes the requested number of bytes and has no
2096 Warning: Doesn't update port line and column counts!
2098 ** New function: scm_init_guile ()
2100 In contrast to scm_boot_guile, scm_init_guile will return normally
2101 after initializing Guile. It is not available on all systems, tho.
2103 ** New functions: scm_str2symbol, scm_mem2symbol
2105 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2106 field of characters and creates a scheme symbol object from that C string.
2107 The function scm_mem2symbol takes a const char* and a number of characters and
2108 creates a symbol from the characters in that memory area.
2110 ** New functions: scm_primitive_make_property
2111 scm_primitive_property_ref
2112 scm_primitive_property_set_x
2113 scm_primitive_property_del_x
2115 These functions implement a new way to deal with object properties.
2116 See libguile/properties.c for their documentation.
2118 ** New function: scm_done_free (long size)
2120 This function is the inverse of scm_done_malloc. Use it to report the
2121 amount of smob memory you free. The previous method, which involved
2122 calling scm_done_malloc with negative argument, was somewhat
2123 unintuitive (and is still available, of course).
2125 ** New function: scm_c_memq (SCM obj, SCM list)
2127 This function provides a fast C level alternative for scm_memq for the case
2128 that the list parameter is known to be a proper list. The function is a
2129 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2130 list input parameter, since for anything else but a proper list the function's
2131 behaviour is undefined - it may even crash or loop endlessly. Further, for
2132 the case that the object is not found in the list, scm_c_memq returns #f which
2133 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2135 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2136 scm_remember_upto_here
2138 These functions replace the function scm_remember.
2140 ** Deprecated function: scm_remember
2142 Use one of the new functions scm_remember_upto_here_1,
2143 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2145 ** New function: scm_allocate_string
2147 This function replaces the function scm_makstr.
2149 ** Deprecated function: scm_makstr
2151 Use the new function scm_allocate_string instead.
2153 ** New global variable scm_gc_running_p introduced.
2155 Use this variable to find out if garbage collection is being executed. Up to
2156 now applications have used scm_gc_heap_lock to test if garbage collection was
2157 running, which also works because of the fact that up to know only the garbage
2158 collector has set this variable. But, this is an implementation detail that
2159 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2160 of this variable is (and has been) not fully safe anyway.
2162 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2164 Use these instead of SCM_LENGTH_MAX.
2166 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2167 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2168 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2170 Use these instead of SCM_LENGTH.
2172 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2173 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2174 SCM_SET_BITVECTOR_LENGTH
2176 Use these instead of SCM_SETLENGTH
2178 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2179 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2182 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2185 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2186 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2189 Use these instead of SCM_SETCHARS.
2191 ** New macro: SCM_BITVECTOR_P
2193 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2195 Use instead of SCM_COERCE_SUBSTR.
2197 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2199 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2201 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2202 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2203 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2204 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2205 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2206 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2207 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2208 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2209 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2210 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2211 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2212 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2213 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2214 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2215 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2217 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2218 Use scm_memory_error instead of SCM_NALLOC.
2219 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2220 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2221 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2222 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2223 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2224 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2225 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2226 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2227 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2228 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2229 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2230 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2231 Use a type specific setter macro instead of SCM_SETLENGTH.
2232 Use a type specific setter macro instead of SCM_SETCHARS.
2233 Use a type specific length macro instead of SCM_LENGTH_MAX.
2234 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2235 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2236 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2237 Use SCM_TYP16 instead of SCM_GCTYP16.
2238 Use SCM_CDR instead of SCM_GCCDR.
2239 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2240 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2241 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2242 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2243 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2244 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2245 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2247 ** Removed function: scm_struct_init
2249 ** Removed variable: scm_symhash_dim
2251 ** Renamed function: scm_make_cont has been replaced by
2252 scm_make_continuation, which has a different interface.
2254 ** Deprecated function: scm_call_catching_errors
2256 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2258 ** Deprecated function: scm_strhash
2260 Use scm_string_hash instead.
2262 ** Deprecated function: scm_vector_set_length_x
2264 Instead, create a fresh vector of the desired size and copy the contents.
2266 ** scm_gensym has changed prototype
2268 scm_gensym now only takes one argument.
2270 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2273 There is now only a single symbol type scm_tc7_symbol.
2274 The tag scm_tc7_lvector was not used anyway.
2276 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2278 Use scm_make_smob_type and scm_set_smob_XXX instead.
2280 ** New function scm_set_smob_apply.
2282 This can be used to set an apply function to a smob type.
2284 ** Deprecated function: scm_strprint_obj
2286 Use scm_object_to_string instead.
2288 ** Deprecated function: scm_wta
2290 Use scm_wrong_type_arg, or another appropriate error signalling function
2293 ** Explicit support for obarrays has been deprecated.
2295 Use `scm_str2symbol' and the generic hashtable functions instead.
2297 ** The concept of `vcells' has been deprecated.
2299 The data type `variable' is now used exclusively. `Vcells' have been
2300 a low-level concept so you are likely not affected by this change.
2302 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2303 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2305 Use scm_c_define or scm_c_lookup instead, as appropriate.
2307 *** New functions: scm_c_module_lookup, scm_c_lookup,
2308 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2309 scm_module_define, scm_define.
2311 These functions work with variables instead of with vcells.
2313 ** New functions for creating and defining `subr's and `gsubr's.
2315 The new functions more clearly distinguish between creating a subr (or
2316 gsubr) object and adding it to the current module.
2318 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2319 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2320 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2321 scm_c_define_gsubr_with_generic.
2323 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2324 scm_make_subr_with_generic, scm_make_gsubr,
2325 scm_make_gsubr_with_generic.
2327 Use the new ones from above instead.
2329 ** C interface to the module system has changed.
2331 While we suggest that you avoid as many explicit module system
2332 operations from C as possible for the time being, the C interface has
2333 been made more similar to the high-level Scheme module system.
2335 *** New functions: scm_c_define_module, scm_c_use_module,
2336 scm_c_export, scm_c_resolve_module.
2338 They mostly work like their Scheme namesakes. scm_c_define_module
2339 takes a function that is called a context where the new module is
2342 *** Deprecated functions: scm_the_root_module, scm_make_module,
2343 scm_ensure_user_module, scm_load_scheme_module.
2345 Use the new functions instead.
2347 ** Renamed function: scm_internal_with_fluids becomes
2350 scm_internal_with_fluids is available as a deprecated function.
2352 ** New function: scm_c_with_fluid.
2354 Just like scm_c_with_fluids, but takes one fluid and one value instead
2357 ** Deprecated typedefs: long_long, ulong_long.
2359 They are of questionable utility and they pollute the global
2362 ** Deprecated typedef: scm_sizet
2364 It is of questionable utility now that Guile requires ANSI C, and is
2367 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2368 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2369 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2371 Made more compliant with the naming policy by adding a _t at the end.
2373 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2374 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2376 With the exception of the mysterious scm_2ulong2big, they are still
2377 available under new names (scm_i_mkbig etc). These functions are not
2378 intended to be used in user code. You should avoid dealing with
2379 bignums directly, and should deal with numbers in general (which can
2382 ** Change in behavior: scm_num2long, scm_num2ulong
2384 The scm_num2[u]long functions don't any longer accept an inexact
2385 argument. This change in behavior is motivated by concordance with
2386 R5RS: It is more common that a primitive doesn't want to accept an
2387 inexact for an exact.
2389 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2390 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2391 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2394 These are conversion functions between the various ANSI C integral
2395 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2396 accept an inexact argument.
2398 ** New functions: scm_float2num, scm_double2num,
2399 scm_num2float, scm_num2double.
2401 These are conversion functions between the two ANSI C float types and
2404 ** New number validation macros:
2405 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2409 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2411 These are just nicer-named old scm_protect_object and
2412 scm_unprotect_object.
2414 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2416 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2418 These functions can be used to register pointers to locations that
2421 ** Deprecated function: scm_create_hook.
2423 Its sins are: misleading name, non-modularity and lack of general
2427 Changes since Guile 1.3.4:
2429 * Changes to the distribution
2431 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2433 We've changed the way we handle generated files in the Guile source
2434 repository. As a result, the procedure for building trees obtained
2435 from the nightly FTP snapshots or via CVS has changed:
2436 - You must have appropriate versions of autoconf, automake, and
2437 libtool installed on your system. See README for info on how to
2438 obtain these programs.
2439 - Before configuring the tree, you must first run the script
2440 `autogen.sh' at the top of the source tree.
2442 The Guile repository used to contain not only source files, written by
2443 humans, but also some generated files, like configure scripts and
2444 Makefile.in files. Even though the contents of these files could be
2445 derived mechanically from other files present, we thought it would
2446 make the tree easier to build if we checked them into CVS.
2448 However, this approach means that minor differences between
2449 developer's installed tools and habits affected the whole team.
2450 So we have removed the generated files from the repository, and
2451 added the autogen.sh script, which will reconstruct them
2455 ** configure now has experimental options to remove support for certain
2458 --disable-arrays omit array and uniform array support
2459 --disable-posix omit posix interfaces
2460 --disable-networking omit networking interfaces
2461 --disable-regex omit regular expression interfaces
2463 These are likely to become separate modules some day.
2465 ** New configure option --enable-debug-freelist
2467 This enables a debugging version of SCM_NEWCELL(), and also registers
2468 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2470 Configure with the --enable-debug-freelist option to enable
2471 the gc-set-debug-check-freelist! primitive, and then use:
2473 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2474 (gc-set-debug-check-freelist! #f) # turn off checking
2476 Checking of the freelist forces a traversal of the freelist and
2477 a garbage collection before each allocation of a cell. This can
2478 slow down the interpreter dramatically, so the setter should be used to
2479 turn on this extra processing only when necessary.
2481 ** New configure option --enable-debug-malloc
2483 Include code for debugging of calls to scm_must_malloc/realloc/free.
2487 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2488 2. objects reallocated by scm_must_realloc has been allocated by
2490 3. reallocated objects are reallocated with the same what string
2492 But, most importantly, it records the number of allocated objects of
2493 each kind. This is useful when searching for memory leaks.
2495 A Guile compiled with this option provides the primitive
2496 `malloc-stats' which returns an alist with pairs of kind and the
2497 number of objects of that kind.
2499 ** All includes are now referenced relative to the root directory
2501 Since some users have had problems with mixups between Guile and
2502 system headers, we have decided to always refer to Guile headers via
2503 their parent directories. This essentially creates a "private name
2504 space" for Guile headers. This means that the compiler only is given
2505 -I options for the root build and root source directory.
2507 ** Header files kw.h and genio.h have been removed.
2509 ** The module (ice-9 getopt-gnu-style) has been removed.
2511 ** New module (ice-9 documentation)
2513 Implements the interface to documentation strings associated with
2516 ** New module (ice-9 time)
2518 Provides a macro `time', which displays execution time of a given form.
2520 ** New module (ice-9 history)
2522 Loading this module enables value history in the repl.
2524 * Changes to the stand-alone interpreter
2526 ** New command line option --debug
2528 Start Guile with debugging evaluator and backtraces enabled.
2530 This is useful when debugging your .guile init file or scripts.
2532 ** New help facility
2534 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2535 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2536 (help 'NAME) gives documentation for NAME, even if it is not an object
2537 (help ,EXPR) gives documentation for object returned by EXPR
2538 (help (my module)) gives module commentary for `(my module)'
2539 (help) gives this text
2541 `help' searches among bindings exported from loaded modules, while
2542 `apropos' searches among bindings visible from the "current" module.
2544 Examples: (help help)
2546 (help "output-string")
2548 ** `help' and `apropos' now prints full module names
2550 ** Dynamic linking now uses libltdl from the libtool package.
2552 The old system dependent code for doing dynamic linking has been
2553 replaced with calls to the libltdl functions which do all the hairy
2556 The major improvement is that you can now directly pass libtool
2557 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2558 will be able to do the best shared library job you can get, via
2561 The way dynamic libraries are found has changed and is not really
2562 portable across platforms, probably. It is therefore recommended to
2563 use absolute filenames when possible.
2565 If you pass a filename without an extension to `dynamic-link', it will
2566 try a few appropriate ones. Thus, the most platform ignorant way is
2567 to specify a name like "libfoo", without any directories and
2570 ** Guile COOP threads are now compatible with LinuxThreads
2572 Previously, COOP threading wasn't possible in applications linked with
2573 Linux POSIX threads due to their use of the stack pointer to find the
2574 thread context. This has now been fixed with a workaround which uses
2575 the pthreads to allocate the stack.
2577 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2579 ** Positions of erring expression in scripts
2581 With version 1.3.4, the location of the erring expression in Guile
2582 scipts is no longer automatically reported. (This should have been
2583 documented before the 1.3.4 release.)
2585 You can get this information by enabling recording of positions of
2586 source expressions and running the debugging evaluator. Put this at
2587 the top of your script (or in your "site" file):
2589 (read-enable 'positions)
2590 (debug-enable 'debug)
2592 ** Backtraces in scripts
2594 It is now possible to get backtraces in scripts.
2598 (debug-enable 'debug 'backtrace)
2600 at the top of the script.
2602 (The first options enables the debugging evaluator.
2603 The second enables backtraces.)
2605 ** Part of module system symbol lookup now implemented in C
2607 The eval closure of most modules is now implemented in C. Since this
2608 was one of the bottlenecks for loading speed, Guile now loads code
2609 substantially faster than before.
2611 ** Attempting to get the value of an unbound variable now produces
2612 an exception with a key of 'unbound-variable instead of 'misc-error.
2614 ** The initial default output port is now unbuffered if it's using a
2615 tty device. Previously in this situation it was line-buffered.
2617 ** New hook: after-gc-hook
2619 after-gc-hook takes over the role of gc-thunk. This hook is run at
2620 the first SCM_TICK after a GC. (Thus, the code is run at the same
2621 point during evaluation as signal handlers.)
2623 Note that this hook should be used only for diagnostic and debugging
2624 purposes. It is not certain that it will continue to be well-defined
2625 when this hook is run in the future.
2627 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2628 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2630 ** Improvements to garbage collector
2632 Guile 1.4 has a new policy for triggering heap allocation and
2633 determining the sizes of heap segments. It fixes a number of problems
2636 1. The new policy can handle two separate pools of cells
2637 (2-word/4-word) better. (The old policy would run wild, allocating
2638 more and more memory for certain programs.)
2640 2. The old code would sometimes allocate far too much heap so that the
2641 Guile process became gigantic. The new code avoids this.
2643 3. The old code would sometimes allocate too little so that few cells
2644 were freed at GC so that, in turn, too much time was spent in GC.
2646 4. The old code would often trigger heap allocation several times in a
2647 row. (The new scheme predicts how large the segments needs to be
2648 in order not to need further allocation.)
2650 All in all, the new GC policy will make larger applications more
2653 The new GC scheme also is prepared for POSIX threading. Threads can
2654 allocate private pools of cells ("clusters") with just a single
2655 function call. Allocation of single cells from such a cluster can
2656 then proceed without any need of inter-thread synchronization.
2658 ** New environment variables controlling GC parameters
2660 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2663 Allocation of 2-word cell heaps:
2665 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2668 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2669 GC in percent of total heap size
2672 Allocation of 4-word cell heaps
2673 (used for real numbers and misc other objects):
2675 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2677 (See entry "Way for application to customize GC parameters" under
2678 section "Changes to the scm_ interface" below.)
2680 ** Guile now implements reals using 4-word cells
2682 This speeds up computation with reals. (They were earlier allocated
2683 with `malloc'.) There is still some room for optimizations, however.
2685 ** Some further steps toward POSIX thread support have been taken
2687 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2688 don't have much effect any longer, and many of them will be removed in
2692 are only handled at the top of the evaluator loop, immediately after
2693 I/O, and in scm_equalp.
2695 *** The GC can allocate thread private pools of pairs.
2697 * Changes to Scheme functions and syntax
2699 ** close-input-port and close-output-port are now R5RS
2701 These procedures have been turned into primitives and have R5RS behaviour.
2703 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2705 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2706 extended by the more sophisticated version in (ice-9 format)
2708 (simple-format port message . args)
2709 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2710 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2711 the escapes are replaced with corresponding members of ARGS:
2712 ~A formats using `display' and ~S formats using `write'.
2713 If DESTINATION is #t, then use the `current-output-port',
2714 if DESTINATION is #f, then return a string containing the formatted text.
2715 Does not add a trailing newline."
2717 ** string-ref: the second argument is no longer optional.
2719 ** string, list->string: no longer accept strings in their arguments,
2720 only characters, for compatibility with R5RS.
2722 ** New procedure: port-closed? PORT
2723 Returns #t if PORT is closed or #f if it is open.
2725 ** Deprecated: list*
2727 The list* functionality is now provided by cons* (SRFI-1 compliant)
2729 ** New procedure: cons* ARG1 ARG2 ... ARGn
2731 Like `list', but the last arg provides the tail of the constructed list,
2732 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2734 Requires at least one argument. If given one argument, that argument
2735 is returned as result.
2737 This function is called `list*' in some other Schemes and in Common LISP.
2739 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2741 ** New procedure: object-documentation OBJECT
2743 Returns the documentation string associated with OBJECT. The
2744 procedure uses a caching mechanism so that subsequent lookups are
2747 Exported by (ice-9 documentation).
2749 ** module-name now returns full names of modules
2751 Previously, only the last part of the name was returned (`session' for
2752 `(ice-9 session)'). Ex: `(ice-9 session)'.
2754 * Changes to the gh_ interface
2756 ** Deprecated: gh_int2scmb
2758 Use gh_bool2scm instead.
2760 * Changes to the scm_ interface
2762 ** Guile primitives now carry docstrings!
2764 Thanks to Greg Badros!
2766 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2768 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2769 macros and must contain a docstring that is extracted into foo.doc using a new
2770 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2772 However, a major overhaul of these macros is scheduled for the next release of
2775 ** Guile primitives use a new technique for validation of arguments
2777 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2778 the readability of argument checking.
2780 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2782 ** New macros: SCM_PACK, SCM_UNPACK
2784 Compose/decompose an SCM value.
2786 The SCM type is now treated as an abstract data type and may be defined as a
2787 long, a void* or as a struct, depending on the architecture and compile time
2788 options. This makes it easier to find several types of bugs, for example when
2789 SCM values are treated as integers without conversion. Values of the SCM type
2790 should be treated as "atomic" values. These macros are used when
2791 composing/decomposing an SCM value, either because you want to access
2792 individual bits, or because you want to treat it as an integer value.
2794 E.g., in order to set bit 7 in an SCM value x, use the expression
2796 SCM_PACK (SCM_UNPACK (x) | 0x80)
2798 ** The name property of hooks is deprecated.
2799 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2801 You can emulate this feature by using object properties.
2803 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2804 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2807 These macros will be removed in a future release of Guile.
2809 ** The following types, functions and macros from numbers.h are deprecated:
2810 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2811 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2813 ** Port internals: the rw_random variable in the scm_port structure
2814 must be set to non-zero in any random access port. In recent Guile
2815 releases it was only set for bidirectional random-access ports.
2817 ** Port internals: the seek ptob procedure is now responsible for
2818 resetting the buffers if required. The change was made so that in the
2819 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2820 the fport and strport ptobs can avoid resetting the buffers,
2821 in particular to avoid discarding unread chars. An existing port
2822 type can be fixed by adding something like the following to the
2823 beginning of the ptob seek procedure:
2825 if (pt->rw_active == SCM_PORT_READ)
2826 scm_end_input (object);
2827 else if (pt->rw_active == SCM_PORT_WRITE)
2828 ptob->flush (object);
2830 although to actually avoid resetting the buffers and discard unread
2831 chars requires further hacking that depends on the characteristics
2834 ** Deprecated functions: scm_fseek, scm_tag
2836 These functions are no longer used and will be removed in a future version.
2838 ** The scm_sysmissing procedure is no longer used in libguile.
2839 Unless it turns out to be unexpectedly useful to somebody, it will be
2840 removed in a future version.
2842 ** The format of error message strings has changed
2844 The two C procedures: scm_display_error and scm_error, as well as the
2845 primitive `scm-error', now use scm_simple_format to do their work.
2846 This means that the message strings of all code must be updated to use
2847 ~A where %s was used before, and ~S where %S was used before.
2849 During the period when there still are a lot of old Guiles out there,
2850 you might want to support both old and new versions of Guile.
2852 There are basically two methods to achieve this. Both methods use
2855 AC_CHECK_FUNCS(scm_simple_format)
2857 in your configure.in.
2859 Method 1: Use the string concatenation features of ANSI C's
2864 #ifdef HAVE_SCM_SIMPLE_FORMAT
2870 Then represent each of your error messages using a preprocessor macro:
2872 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2876 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2877 (define make-message string-append)
2879 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2881 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2885 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2890 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2894 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2896 Don't use the functions coop_mutex_init and
2897 coop_condition_variable_init. They will change.
2899 Use scm_mutex_init and scm_cond_init instead.
2901 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2902 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2903 COND, as `scm_cond_wait' does, but it also bounds the duration
2904 of the wait. If COND has not been signaled before time ABSTIME,
2905 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2906 returns the error code `ETIMEDOUT'.
2908 The ABSTIME parameter specifies an absolute time, with the same
2909 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2910 to 00:00:00 GMT, January 1, 1970.
2912 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2913 `scm_cond_broadcast' restarts all the threads that are waiting
2914 on the condition variable COND. Nothing happens if no threads are
2917 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2918 `scm_key_create' allocates a new TSD key. The key is stored in
2919 the location pointed to by KEY. There is no limit on the number
2920 of keys allocated at a given time. The value initially associated
2921 with the returned key is `NULL' in all currently executing threads.
2923 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2924 function associated with the key. When a thread terminates,
2925 DESTR_FUNCTION is called on the value associated with the key in
2926 that thread. The DESTR_FUNCTION is not called if a key is deleted
2927 with `scm_key_delete' or a value is changed with
2928 `scm_setspecific'. The order in which destructor functions are
2929 called at thread termination time is unspecified.
2931 Destructors are not yet implemented.
2933 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2934 `scm_setspecific' changes the value associated with KEY in the
2935 calling thread, storing the given POINTER instead.
2937 ** New function: scm_getspecific (scm_key_t KEY)
2938 `scm_getspecific' returns the value currently associated with
2939 KEY in the calling thread.
2941 ** New function: scm_key_delete (scm_key_t KEY)
2942 `scm_key_delete' deallocates a TSD key. It does not check
2943 whether non-`NULL' values are associated with that key in the
2944 currently executing threads, nor call the destructor function
2945 associated with the key.
2947 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2949 Initialize a C level hook HOOK with associated HOOK_DATA and type
2950 TYPE. (See scm_c_hook_run ().)
2952 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2954 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2955 is true, add it last, otherwise first. The same FUNC can be added
2956 multiple times if FUNC_DATA differ and vice versa.
2958 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2960 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2961 function is only removed if both FUNC and FUNC_DATA matches.
2963 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2965 Run hook HOOK passing DATA to the hook functions.
2967 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2968 returned is undefined.
2970 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2971 returns a non-NULL value. This value is returned as the result of
2972 scm_c_hook_run. If all functions return NULL, NULL is returned.
2974 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2975 returns a NULL value, and NULL is returned. If all functions returns
2976 a non-NULL value, the last value is returned.
2978 ** New C level GC hooks
2980 Five new C level hooks has been added to the garbage collector.
2982 scm_before_gc_c_hook
2985 are run before locking and after unlocking the heap. The system is
2986 thus in a mode where evaluation can take place. (Except that
2987 scm_before_gc_c_hook must not allocate new cells.)
2989 scm_before_mark_c_hook
2990 scm_before_sweep_c_hook
2991 scm_after_sweep_c_hook
2993 are run when the heap is locked. These are intended for extension of
2994 the GC in a modular fashion. Examples are the weaks and guardians
2997 ** Way for application to customize GC parameters
2999 The application can set up other default values for the GC heap
3000 allocation parameters
3002 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3003 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3004 GUILE_MAX_SEGMENT_SIZE,
3008 scm_default_init_heap_size_1, scm_default_min_yield_1,
3009 scm_default_init_heap_size_2, scm_default_min_yield_2,
3010 scm_default_max_segment_size
3012 respectively before callong scm_boot_guile.
3014 (See entry "New environment variables ..." in section
3015 "Changes to the stand-alone interpreter" above.)
3017 ** scm_protect_object/scm_unprotect_object now nest
3019 This means that you can call scm_protect_object multiple times on an
3020 object and count on the object being protected until
3021 scm_unprotect_object has been call the same number of times.
3023 The functions also have better time complexity.
3025 Still, it is usually possible to structure the application in a way
3026 that you don't need to use these functions. For example, if you use a
3027 protected standard Guile list to keep track of live objects rather
3028 than some custom data type, objects will die a natural death when they
3029 are no longer needed.
3031 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3033 Guile does not provide the float representation for inexact real numbers any
3034 more. Now, only doubles are used to represent inexact real numbers. Further,
3035 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3036 and scm_tc16_complex, respectively.
3038 ** Removed deprecated type scm_smobfuns
3040 ** Removed deprecated function scm_newsmob
3042 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3044 There is an ongoing discussion among the developers whether to
3045 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3046 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3047 until this issue has been settled.
3049 ** Removed deprecated type tag scm_tc16_kw
3051 ** Added type tag scm_tc16_keyword
3053 (This was introduced already in release 1.3.4 but was not documented
3056 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3058 * Changes to system call interfaces:
3060 ** The "select" procedure now tests port buffers for the ability to
3061 provide input or accept output. Previously only the underlying file
3062 descriptors were checked.
3064 ** New variable PIPE_BUF: the maximum number of bytes that can be
3065 atomically written to a pipe.
3067 ** If a facility is not available on the system when Guile is
3068 compiled, the corresponding primitive procedure will not be defined.
3069 Previously it would have been defined but would throw a system-error
3070 exception if called. Exception handlers which catch this case may
3071 need minor modification: an error will be thrown with key
3072 'unbound-variable instead of 'system-error. Alternatively it's
3073 now possible to use `defined?' to check whether the facility is
3076 ** Procedures which depend on the timezone should now give the correct
3077 result on systems which cache the TZ environment variable, even if TZ
3078 is changed without calling tzset.
3080 * Changes to the networking interfaces:
3082 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3083 long integers between network and host format. For now, it's not
3084 particularly convenient to do this kind of thing, but consider:
3086 (define write-network-long
3087 (lambda (value port)
3088 (let ((v (make-uniform-vector 1 1 0)))
3089 (uniform-vector-set! v 0 (htonl value))
3090 (uniform-vector-write v port))))
3092 (define read-network-long
3094 (let ((v (make-uniform-vector 1 1 0)))
3095 (uniform-vector-read! v port)
3096 (ntohl (uniform-vector-ref v 0)))))
3098 ** If inet-aton fails, it now throws an error with key 'misc-error
3099 instead of 'system-error, since errno is not relevant.
3101 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3102 specific keys instead of 'system-error. The latter is inappropriate
3103 since errno will not have been set. The keys are:
3104 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3106 ** sethostent, setnetent, setprotoent, setservent: now take an
3107 optional argument STAYOPEN, which specifies whether the database
3108 remains open after a database entry is accessed randomly (e.g., using
3109 gethostbyname for the hosts database.) The default is #f. Previously
3113 Changes since Guile 1.3.2:
3115 * Changes to the stand-alone interpreter
3119 An initial version of the Guile debugger written by Chris Hanson has
3120 been added. The debugger is still under development but is included
3121 in the distribution anyway since it is already quite useful.
3127 after an error to enter the debugger. Type `help' inside the debugger
3128 for a description of available commands.
3130 If you prefer to have stack frames numbered and printed in
3131 anti-chronological order and prefer up in the stack to be down on the
3132 screen as is the case in gdb, you can put
3134 (debug-enable 'backwards)
3136 in your .guile startup file. (However, this means that Guile can't
3137 use indentation to indicate stack level.)
3139 The debugger is autoloaded into Guile at the first use.
3141 ** Further enhancements to backtraces
3143 There is a new debug option `width' which controls the maximum width
3144 on the screen of printed stack frames. Fancy printing parameters
3145 ("level" and "length" as in Common LISP) are adaptively adjusted for
3146 each stack frame to give maximum information while still fitting
3147 within the bounds. If the stack frame can't be made to fit by
3148 adjusting parameters, it is simply cut off at the end. This is marked
3151 ** Some modules are now only loaded when the repl is started
3153 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3154 regex) are now loaded into (guile-user) only if the repl has been
3155 started. The effect is that the startup time for scripts has been
3156 reduced to 30% of what it was previously.
3158 Correctly written scripts load the modules they require at the top of
3159 the file and should not be affected by this change.
3161 ** Hooks are now represented as smobs
3163 * Changes to Scheme functions and syntax
3165 ** Readline support has changed again.
3167 The old (readline-activator) module is gone. Use (ice-9 readline)
3168 instead, which now contains all readline functionality. So the code
3169 to activate readline is now
3171 (use-modules (ice-9 readline))
3174 This should work at any time, including from the guile prompt.
3176 To avoid confusion about the terms of Guile's license, please only
3177 enable readline for your personal use; please don't make it the
3178 default for others. Here is why we make this rather odd-sounding
3181 Guile is normally licensed under a weakened form of the GNU General
3182 Public License, which allows you to link code with Guile without
3183 placing that code under the GPL. This exception is important to some
3186 However, since readline is distributed under the GNU General Public
3187 License, when you link Guile with readline, either statically or
3188 dynamically, you effectively change Guile's license to the strict GPL.
3189 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3190 which are normally permitted become forbidden. This is a rather
3191 non-obvious consequence of the licensing terms.
3193 So, to make sure things remain clear, please let people choose for
3194 themselves whether to link GPL'd libraries like readline with Guile.
3196 ** regexp-substitute/global has changed slightly, but incompatibly.
3198 If you include a function in the item list, the string of the match
3199 object it receives is the same string passed to
3200 regexp-substitute/global, not some suffix of that string.
3201 Correspondingly, the match's positions are relative to the entire
3202 string, not the suffix.
3204 If the regexp can match the empty string, the way matches are chosen
3205 from the string has changed. regexp-substitute/global recognizes the
3206 same set of matches that list-matches does; see below.
3208 ** New function: list-matches REGEXP STRING [FLAGS]
3210 Return a list of match objects, one for every non-overlapping, maximal
3211 match of REGEXP in STRING. The matches appear in left-to-right order.
3212 list-matches only reports matches of the empty string if there are no
3213 other matches which begin on, end at, or include the empty match's
3216 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3218 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3220 For each match of REGEXP in STRING, apply PROC to the match object,
3221 and the last value PROC returned, or INIT for the first call. Return
3222 the last value returned by PROC. We apply PROC to the matches as they
3223 appear from left to right.
3225 This function recognizes matches according to the same criteria as
3228 Thus, you could define list-matches like this:
3230 (define (list-matches regexp string . flags)
3231 (reverse! (apply fold-matches regexp string '() cons flags)))
3233 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3237 *** New function: hook? OBJ
3239 Return #t if OBJ is a hook, otherwise #f.
3241 *** New function: make-hook-with-name NAME [ARITY]
3243 Return a hook with name NAME and arity ARITY. The default value for
3244 ARITY is 0. The only effect of NAME is that it will appear when the
3245 hook object is printed to ease debugging.
3247 *** New function: hook-empty? HOOK
3249 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3251 *** New function: hook->list HOOK
3253 Return a list of the procedures that are called when run-hook is
3256 ** `map' signals an error if its argument lists are not all the same length.
3258 This is the behavior required by R5RS, so this change is really a bug
3259 fix. But it seems to affect a lot of people's code, so we're
3260 mentioning it here anyway.
3262 ** Print-state handling has been made more transparent
3264 Under certain circumstances, ports are represented as a port with an
3265 associated print state. Earlier, this pair was represented as a pair
3266 (see "Some magic has been added to the printer" below). It is now
3267 indistinguishable (almost; see `get-print-state') from a port on the
3270 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3272 Return a new port with the associated print state PRINT-STATE.
3274 *** New function: get-print-state OUTPUT-PORT
3276 Return the print state associated with this port if it exists,
3277 otherwise return #f.
3279 *** New function: directory-stream? OBJECT
3281 Returns true iff OBJECT is a directory stream --- the sort of object
3282 returned by `opendir'.
3284 ** New function: using-readline?
3286 Return #t if readline is in use in the current repl.
3288 ** structs will be removed in 1.4
3290 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3291 and use GOOPS objects as the fundamental record type.
3293 * Changes to the scm_ interface
3295 ** structs will be removed in 1.4
3297 The entire current struct interface (struct.c, struct.h) will be
3298 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3299 GOOPS objects as the fundamental record type.
3301 ** The internal representation of subr's has changed
3303 Instead of giving a hint to the subr name, the CAR field of the subr
3304 now contains an index to a subr entry in scm_subr_table.
3306 *** New variable: scm_subr_table
3308 An array of subr entries. A subr entry contains the name, properties
3309 and documentation associated with the subr. The properties and
3310 documentation slots are not yet used.
3312 ** A new scheme for "forwarding" calls to a builtin to a generic function
3314 It is now possible to extend the functionality of some Guile
3315 primitives by letting them defer a call to a GOOPS generic function on
3316 argument mismatch. This means that there is no loss of efficiency in
3321 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3322 (define-method + ((x <string>) (y <string>))
3323 (string-append x y))
3325 + will still be as efficient as usual in numerical calculations, but
3326 can also be used for concatenating strings.
3328 Who will be the first one to extend Guile's numerical tower to
3329 rationals? :) [OK, there a few other things to fix before this can
3330 be made in a clean way.]
3332 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3334 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3336 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3338 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3339 a variable GENERIC which can be used by the dispatch macros below.
3341 [This is experimental code which may change soon.]
3343 *** New macros for forwarding control to a generic on arg type error
3345 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3347 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3349 These correspond to the scm_wta function call, and have the same
3350 behaviour until the user has called the GOOPS primitive
3351 `enable-primitive-generic!'. After that, these macros will apply the
3352 generic function GENERIC to the argument(s) instead of calling
3355 [This is experimental code which may change soon.]
3357 *** New macros for argument testing with generic dispatch
3359 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3361 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3363 These correspond to the SCM_ASSERT macro, but will defer control to
3364 GENERIC on error after `enable-primitive-generic!' has been called.
3366 [This is experimental code which may change soon.]
3368 ** New function: SCM scm_eval_body (SCM body, SCM env)
3370 Evaluates the body of a special form.
3372 ** The internal representation of struct's has changed
3374 Previously, four slots were allocated for the procedure(s) of entities
3375 and operators. The motivation for this representation had to do with
3376 the structure of the evaluator, the wish to support tail-recursive
3377 generic functions, and efficiency. Since the generic function
3378 dispatch mechanism has changed, there is no longer a need for such an
3379 expensive representation, and the representation has been simplified.
3381 This should not make any difference for most users.
3383 ** GOOPS support has been cleaned up.
3385 Some code has been moved from eval.c to objects.c and code in both of
3386 these compilation units has been cleaned up and better structured.
3388 *** New functions for applying generic functions
3390 New function: SCM scm_apply_generic (GENERIC, ARGS)
3391 New function: SCM scm_call_generic_0 (GENERIC)
3392 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3393 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3394 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3396 ** Deprecated function: scm_make_named_hook
3398 It is now replaced by:
3400 ** New function: SCM scm_create_hook (const char *name, int arity)
3402 Creates a hook in the same way as make-hook above but also
3403 binds a variable named NAME to it.
3405 This is the typical way of creating a hook from C code.
3407 Currently, the variable is created in the "current" module.
3408 This might change when we get the new module system.
3410 [The behaviour is identical to scm_make_named_hook.]
3414 Changes since Guile 1.3:
3416 * Changes to mailing lists
3418 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3420 See the README file to find current addresses for all the Guile
3423 * Changes to the distribution
3425 ** Readline support is no longer included with Guile by default.
3427 Based on the different license terms of Guile and Readline, we
3428 concluded that Guile should not *by default* cause the linking of
3429 Readline into an application program. Readline support is now offered
3430 as a separate module, which is linked into an application only when
3431 you explicitly specify it.
3433 Although Guile is GNU software, its distribution terms add a special
3434 exception to the usual GNU General Public License (GPL). Guile's
3435 license includes a clause that allows you to link Guile with non-free
3436 programs. We add this exception so as not to put Guile at a
3437 disadvantage vis-a-vis other extensibility packages that support other
3440 In contrast, the GNU Readline library is distributed under the GNU
3441 General Public License pure and simple. This means that you may not
3442 link Readline, even dynamically, into an application unless it is
3443 distributed under a free software license that is compatible the GPL.
3445 Because of this difference in distribution terms, an application that
3446 can use Guile may not be able to use Readline. Now users will be
3447 explicitly offered two independent decisions about the use of these
3450 You can activate the readline support by issuing
3452 (use-modules (readline-activator))
3455 from your ".guile" file, for example.
3457 * Changes to the stand-alone interpreter
3459 ** All builtins now print as primitives.
3460 Previously builtin procedures not belonging to the fundamental subr
3461 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3462 Now, they print as #<primitive-procedure NAME>.
3464 ** Backtraces slightly more intelligible.
3465 gsubr-apply and macro transformer application frames no longer appear
3468 * Changes to Scheme functions and syntax
3470 ** Guile now correctly handles internal defines by rewriting them into
3471 their equivalent letrec. Previously, internal defines would
3472 incrementally add to the innermost environment, without checking
3473 whether the restrictions specified in RnRS were met. This lead to the
3474 correct behaviour when these restriction actually were met, but didn't
3475 catch all illegal uses. Such an illegal use could lead to crashes of
3476 the Guile interpreter or or other unwanted results. An example of
3477 incorrect internal defines that made Guile behave erratically:
3489 The problem with this example is that the definition of `c' uses the
3490 value of `b' directly. This confuses the meoization machine of Guile
3491 so that the second call of `b' (this time in a larger environment that
3492 also contains bindings for `c' and `d') refers to the binding of `c'
3493 instead of `a'. You could also make Guile crash with a variation on
3498 (define (b flag) (if flag a 1))
3499 (define c (1+ (b flag)))
3507 From now on, Guile will issue an `Unbound variable: b' error message
3512 A hook contains a list of functions which should be called on
3513 particular occasions in an existing program. Hooks are used for
3516 A window manager might have a hook before-window-map-hook. The window
3517 manager uses the function run-hooks to call all functions stored in
3518 before-window-map-hook each time a window is mapped. The user can
3519 store functions in the hook using add-hook!.
3521 In Guile, hooks are first class objects.
3523 *** New function: make-hook [N_ARGS]
3525 Return a hook for hook functions which can take N_ARGS arguments.
3526 The default value for N_ARGS is 0.
3528 (See also scm_make_named_hook below.)
3530 *** New function: add-hook! HOOK PROC [APPEND_P]
3532 Put PROC at the beginning of the list of functions stored in HOOK.
3533 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3535 PROC must be able to take the number of arguments specified when the
3538 If PROC already exists in HOOK, then remove it first.
3540 *** New function: remove-hook! HOOK PROC
3542 Remove PROC from the list of functions in HOOK.
3544 *** New function: reset-hook! HOOK
3546 Clear the list of hook functions stored in HOOK.
3548 *** New function: run-hook HOOK ARG1 ...
3550 Run all hook functions stored in HOOK with arguments ARG1 ... .
3551 The number of arguments supplied must correspond to the number given
3552 when the hook was created.
3554 ** The function `dynamic-link' now takes optional keyword arguments.
3555 The only keyword argument that is currently defined is `:global
3556 BOOL'. With it, you can control whether the shared library will be
3557 linked in global mode or not. In global mode, the symbols from the
3558 linked library can be used to resolve references from other
3559 dynamically linked libraries. In non-global mode, the linked
3560 library is essentially invisible and can only be accessed via
3561 `dynamic-func', etc. The default is now to link in global mode.
3562 Previously, the default has been non-global mode.
3564 The `#:global' keyword is only effective on platforms that support
3565 the dlopen family of functions.
3567 ** New function `provided?'
3569 - Function: provided? FEATURE
3570 Return true iff FEATURE is supported by this installation of
3571 Guile. FEATURE must be a symbol naming a feature; the global
3572 variable `*features*' is a list of available features.
3574 ** Changes to the module (ice-9 expect):
3576 *** The expect-strings macro now matches `$' in a regular expression
3577 only at a line-break or end-of-file by default. Previously it would
3578 match the end of the string accumulated so far. The old behaviour
3579 can be obtained by setting the variable `expect-strings-exec-flags'
3582 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3583 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3584 in a regular expression will still match before a line-break or
3585 end-of-file. The default is `regexp/noteol'.
3587 *** The expect-strings macro now uses a variable
3588 `expect-strings-compile-flags' for the flags to be supplied to
3589 `make-regexp'. The default is `regexp/newline', which was previously
3592 *** The expect macro now supplies two arguments to a match procedure:
3593 the current accumulated string and a flag to indicate whether
3594 end-of-file has been reached. Previously only the string was supplied.
3595 If end-of-file is reached, the match procedure will be called an
3596 additional time with the same accumulated string as the previous call
3597 but with the flag set.
3599 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3601 This code, and the documentation for it that appears here, was
3602 borrowed from SLIB, with minor adaptations for Guile.
3604 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3605 An almost complete implementation of Common LISP format description
3606 according to the CL reference book `Common LISP' from Guy L.
3607 Steele, Digital Press. Backward compatible to most of the
3608 available Scheme format implementations.
3610 Returns `#t', `#f' or a string; has side effect of printing
3611 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3612 to the current output port and `#t' is returned. If DESTINATION
3613 is `#f', a formatted string is returned as the result of the call.
3614 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3615 format string; FORMAT-STRING is then the first argument and the
3616 output is returned as a string. If DESTINATION is a number, the
3617 output is to the current error port if available by the
3618 implementation. Otherwise DESTINATION must be an output port and
3621 FORMAT-STRING must be a string. In case of a formatting error
3622 format returns `#f' and prints a message on the current output or
3623 error port. Characters are output as if the string were output by
3624 the `display' function with the exception of those prefixed by a
3625 tilde (~). For a detailed description of the FORMAT-STRING syntax
3626 please consult a Common LISP format reference manual. For a test
3627 suite to verify this format implementation load `formatst.scm'.
3628 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3630 Note: `format' is not reentrant, i.e. only one `format'-call may
3631 be executed at a time.
3634 *** Format Specification (Format version 3.0)
3636 Please consult a Common LISP format reference manual for a detailed
3637 description of the format string syntax. For a demonstration of the
3638 implemented directives see `formatst.scm'.
3640 This implementation supports directive parameters and modifiers (`:'
3641 and `@' characters). Multiple parameters must be separated by a comma
3642 (`,'). Parameters can be numerical parameters (positive or negative),
3643 character parameters (prefixed by a quote character (`''), variable
3644 parameters (`v'), number of rest arguments parameter (`#'), empty and
3645 default parameters. Directive characters are case independent. The
3646 general form of a directive is:
3648 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3650 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3652 *** Implemented CL Format Control Directives
3654 Documentation syntax: Uppercase characters represent the
3655 corresponding control directive characters. Lowercase characters
3656 represent control directive parameter descriptions.
3659 Any (print as `display' does).
3663 `~MINCOL,COLINC,MINPAD,PADCHARA'
3667 S-expression (print as `write' does).
3671 `~MINCOL,COLINC,MINPAD,PADCHARS'
3677 print number sign always.
3680 print comma separated.
3682 `~MINCOL,PADCHAR,COMMACHARD'
3688 print number sign always.
3691 print comma separated.
3693 `~MINCOL,PADCHAR,COMMACHARX'
3699 print number sign always.
3702 print comma separated.
3704 `~MINCOL,PADCHAR,COMMACHARO'
3710 print number sign always.
3713 print comma separated.
3715 `~MINCOL,PADCHAR,COMMACHARB'
3720 `~N,MINCOL,PADCHAR,COMMACHARR'
3724 print a number as a Roman numeral.
3727 print a number as an "old fashioned" Roman numeral.
3730 print a number as an ordinal English number.
3733 print a number as a cardinal English number.
3738 prints `y' and `ies'.
3741 as `~P but jumps 1 argument backward.'
3744 as `~@P but jumps 1 argument backward.'
3749 prints a character as the reader can understand it (i.e. `#\'
3753 prints a character as emacs does (eg. `^C' for ASCII 03).
3756 Fixed-format floating-point (prints a flonum like MMM.NNN).
3757 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3759 If the number is positive a plus sign is printed.
3762 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3763 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3765 If the number is positive a plus sign is printed.
3768 General floating-point (prints a flonum either fixed or
3770 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3772 If the number is positive a plus sign is printed.
3775 Dollars floating-point (prints a flonum in fixed with signs
3777 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3779 If the number is positive a plus sign is printed.
3782 A sign is always printed and appears before the padding.
3785 The sign appears before the padding.
3793 print newline if not at the beginning of the output line.
3795 prints `~&' and then N-1 newlines.
3800 print N page separators.
3810 newline is ignored, white space left.
3813 newline is left, white space ignored.
3818 relative tabulation.
3824 Indirection (expects indirect arguments as a list).
3826 extracts indirect arguments from format arguments.
3829 Case conversion (converts by `string-downcase').
3831 converts by `string-capitalize'.
3834 converts by `string-capitalize-first'.
3837 converts by `string-upcase'.
3840 Argument Jumping (jumps 1 argument forward).
3842 jumps N arguments forward.
3845 jumps 1 argument backward.
3848 jumps N arguments backward.
3851 jumps to the 0th argument.
3854 jumps to the Nth argument (beginning from 0)
3856 `~[STR0~;STR1~;...~;STRN~]'
3857 Conditional Expression (numerical clause conditional).
3859 take argument from N.
3862 true test conditional.
3865 if-else-then conditional.
3871 default clause follows.
3874 Iteration (args come from the next argument (a list)).
3876 at most N iterations.
3879 args from next arg (a list of lists).
3882 args from the rest of arguments.
3885 args from the rest args (lists).
3896 aborts if N <= M <= K
3898 *** Not Implemented CL Format Control Directives
3901 print `#f' as an empty list (see below).
3904 print `#f' as an empty list (see below).
3910 (sorry I don't understand its semantics completely)
3912 *** Extended, Replaced and Additional Control Directives
3914 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3915 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3916 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3917 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3918 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3919 COMMAWIDTH is the number of characters between two comma
3923 print a R4RS complex number as `~F~@Fi' with passed parameters for
3927 Pretty print formatting of an argument for scheme code lists.
3933 Flushes the output if format DESTINATION is a port.
3936 Print a `#\space' character
3938 print N `#\space' characters.
3941 Print a `#\tab' character
3943 print N `#\tab' characters.
3946 Takes N as an integer representation for a character. No arguments
3947 are consumed. N is converted to a character by `integer->char'. N
3948 must be a positive decimal number.
3951 Print out readproof. Prints out internal objects represented as
3952 `#<...>' as strings `"#<...>"' so that the format output can always
3953 be processed by `read'.
3956 Print out readproof. Prints out internal objects represented as
3957 `#<...>' as strings `"#<...>"' so that the format output can always
3958 be processed by `read'.
3961 Prints information and a copyright notice on the format
3964 prints format version.
3967 may also print number strings, i.e. passing a number as a string
3968 and format it accordingly.
3970 *** Configuration Variables
3972 The format module exports some configuration variables to suit the
3973 systems and users needs. There should be no modification necessary for
3974 the configuration that comes with Guile. Format detects automatically
3975 if the running scheme system implements floating point numbers and
3978 format:symbol-case-conv
3979 Symbols are converted by `symbol->string' so the case type of the
3980 printed symbols is implementation dependent.
3981 `format:symbol-case-conv' is a one arg closure which is either
3982 `#f' (no conversion), `string-upcase', `string-downcase' or
3983 `string-capitalize'. (default `#f')
3985 format:iobj-case-conv
3986 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3987 implementation internal objects. (default `#f')
3990 The character prefixing the exponent value in `~E' printing.
3993 *** Compatibility With Other Format Implementations
3999 Downward compatible except for padding support and `~A', `~S',
4000 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4001 `printf' padding support which is completely replaced by the CL
4002 `format' padding style.
4005 Downward compatible except for `~', which is not documented
4006 (ignores all characters inside the format string up to a newline
4007 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4008 numerical and variable parameters and `:/@' modifiers in the CL
4012 Downward compatible except for `~A' and `~S' which print in
4013 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4014 directive parameters or modifiers)).
4017 Downward compatible except for an optional destination parameter:
4018 S2C accepts a format call without a destination which returns a
4019 formatted string. This is equivalent to a #f destination in S2C.
4020 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4021 parameters or modifiers)).
4024 ** Changes to string-handling functions.
4026 These functions were added to support the (ice-9 format) module, above.
4028 *** New function: string-upcase STRING
4029 *** New function: string-downcase STRING
4031 These are non-destructive versions of the existing string-upcase! and
4032 string-downcase! functions.
4034 *** New function: string-capitalize! STRING
4035 *** New function: string-capitalize STRING
4037 These functions convert the first letter of each word in the string to
4040 (string-capitalize "howdy there")
4043 As with the other functions, string-capitalize! modifies the string in
4044 place, while string-capitalize returns a modified copy of its argument.
4046 *** New function: string-ci->symbol STRING
4048 Return a symbol whose name is STRING, but having the same case as if
4049 the symbol had be read by `read'.
4051 Guile can be configured to be sensitive or insensitive to case
4052 differences in Scheme identifiers. If Guile is case-insensitive, all
4053 symbols are converted to lower case on input. The `string-ci->symbol'
4054 function returns a symbol whose name in STRING, transformed as Guile
4055 would if STRING were input.
4057 *** New function: substring-move! STRING1 START END STRING2 START
4059 Copy the substring of STRING1 from START (inclusive) to END
4060 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4061 string, and the source and destination areas may overlap; in all
4062 cases, the function behaves as if all the characters were copied
4065 *** Extended functions: substring-move-left! substring-move-right!
4067 These functions now correctly copy arbitrarily overlapping substrings;
4068 they are both synonyms for substring-move!.
4071 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4073 getopt-long is a function for parsing command-line arguments in a
4074 manner consistent with other GNU programs.
4076 (getopt-long ARGS GRAMMAR)
4077 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4079 ARGS should be a list of strings. Its first element should be the
4080 name of the program; subsequent elements should be the arguments
4081 that were passed to the program on the command line. The
4082 `program-arguments' procedure returns a list of this form.
4084 GRAMMAR is a list of the form:
4085 ((OPTION (PROPERTY VALUE) ...) ...)
4087 Each OPTION should be a symbol. `getopt-long' will accept a
4088 command-line option named `--OPTION'.
4089 Each option can have the following (PROPERTY VALUE) pairs:
4091 (single-char CHAR) --- Accept `-CHAR' as a single-character
4092 equivalent to `--OPTION'. This is how to specify traditional
4094 (required? BOOL) --- If BOOL is true, the option is required.
4095 getopt-long will raise an error if it is not found in ARGS.
4096 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4097 it is #f, it does not; and if it is the symbol
4098 `optional', the option may appear in ARGS with or
4100 (predicate FUNC) --- If the option accepts a value (i.e. you
4101 specified `(value #t)' for this option), then getopt
4102 will apply FUNC to the value, and throw an exception
4103 if it returns #f. FUNC should be a procedure which
4104 accepts a string and returns a boolean value; you may
4105 need to use quasiquotes to get it into GRAMMAR.
4107 The (PROPERTY VALUE) pairs may occur in any order, but each
4108 property may occur only once. By default, options do not have
4109 single-character equivalents, are not required, and do not take
4112 In ARGS, single-character options may be combined, in the usual
4113 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4114 accepts values, then it must be the last option in the
4115 combination; the value is the next argument. So, for example, using
4116 the following grammar:
4117 ((apples (single-char #\a))
4118 (blimps (single-char #\b) (value #t))
4119 (catalexis (single-char #\c) (value #t)))
4120 the following argument lists would be acceptable:
4121 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4122 for "blimps" and "catalexis")
4123 ("-ab" "bang" "-c" "couth") (same)
4124 ("-ac" "couth" "-b" "bang") (same)
4125 ("-abc" "couth" "bang") (an error, since `-b' is not the
4126 last option in its combination)
4128 If an option's value is optional, then `getopt-long' decides
4129 whether it has a value by looking at what follows it in ARGS. If
4130 the next element is a string, and it does not appear to be an
4131 option itself, then that string is the option's value.
4133 The value of a long option can appear as the next element in ARGS,
4134 or it can follow the option name, separated by an `=' character.
4135 Thus, using the same grammar as above, the following argument lists
4137 ("--apples" "Braeburn" "--blimps" "Goodyear")
4138 ("--apples=Braeburn" "--blimps" "Goodyear")
4139 ("--blimps" "Goodyear" "--apples=Braeburn")
4141 If the option "--" appears in ARGS, argument parsing stops there;
4142 subsequent arguments are returned as ordinary arguments, even if
4143 they resemble options. So, in the argument list:
4144 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4145 `getopt-long' will recognize the `apples' option as having the
4146 value "Granny Smith", but it will not recognize the `blimp'
4147 option; it will return the strings "--blimp" and "Goodyear" as
4148 ordinary argument strings.
4150 The `getopt-long' function returns the parsed argument list as an
4151 assocation list, mapping option names --- the symbols from GRAMMAR
4152 --- onto their values, or #t if the option does not accept a value.
4153 Unused options do not appear in the alist.
4155 All arguments that are not the value of any option are returned
4156 as a list, associated with the empty list.
4158 `getopt-long' throws an exception if:
4159 - it finds an unrecognized option in ARGS
4160 - a required option is omitted
4161 - an option that requires an argument doesn't get one
4162 - an option that doesn't accept an argument does get one (this can
4163 only happen using the long option `--opt=value' syntax)
4164 - an option predicate fails
4169 `((lockfile-dir (required? #t)
4172 (predicate ,file-is-directory?))
4173 (verbose (required? #f)
4176 (x-includes (single-char #\x))
4177 (rnet-server (single-char #\y)
4178 (predicate ,string?))))
4180 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4181 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4183 => ((() "foo1" "-fred" "foo2" "foo3")
4184 (rnet-server . "lamprod")
4185 (x-includes . "/usr/include")
4186 (lockfile-dir . "/tmp")
4189 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4191 It will be removed in a few releases.
4193 ** New syntax: lambda*
4194 ** New syntax: define*
4195 ** New syntax: define*-public
4196 ** New syntax: defmacro*
4197 ** New syntax: defmacro*-public
4198 Guile now supports optional arguments.
4200 `lambda*', `define*', `define*-public', `defmacro*' and
4201 `defmacro*-public' are identical to the non-* versions except that
4202 they use an extended type of parameter list that has the following BNF
4203 syntax (parentheses are literal, square brackets indicate grouping,
4204 and `*', `+' and `?' have the usual meaning):
4206 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4207 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4208 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4210 ext-var-decl ::= identifier | ( identifier expression )
4212 The semantics are best illustrated with the following documentation
4213 and examples for `lambda*':
4216 lambda extended for optional and keyword arguments
4218 lambda* creates a procedure that takes optional arguments. These
4219 are specified by putting them inside brackets at the end of the
4220 paramater list, but before any dotted rest argument. For example,
4221 (lambda* (a b #&optional c d . e) '())
4222 creates a procedure with fixed arguments a and b, optional arguments c
4223 and d, and rest argument e. If the optional arguments are omitted
4224 in a call, the variables for them are unbound in the procedure. This
4225 can be checked with the bound? macro.
4227 lambda* can also take keyword arguments. For example, a procedure
4229 (lambda* (#&key xyzzy larch) '())
4230 can be called with any of the argument lists (#:xyzzy 11)
4231 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4232 are given as keywords are bound to values.
4234 Optional and keyword arguments can also be given default values
4235 which they take on when they are not present in a call, by giving a
4236 two-item list in place of an optional argument, for example in:
4237 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4238 foo is a fixed argument, bar is an optional argument with default
4239 value 42, and baz is a keyword argument with default value 73.
4240 Default value expressions are not evaluated unless they are needed
4241 and until the procedure is called.
4243 lambda* now supports two more special parameter list keywords.
4245 lambda*-defined procedures now throw an error by default if a
4246 keyword other than one of those specified is found in the actual
4247 passed arguments. However, specifying #&allow-other-keys
4248 immediately after the kyword argument declarations restores the
4249 previous behavior of ignoring unknown keywords. lambda* also now
4250 guarantees that if the same keyword is passed more than once, the
4251 last one passed is the one that takes effect. For example,
4252 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4253 #:heads 37 #:tails 42 #:heads 99)
4254 would result in (99 47) being displayed.
4256 #&rest is also now provided as a synonym for the dotted syntax rest
4257 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4258 all respects to lambda*. This is provided for more similarity to DSSSL,
4259 MIT-Scheme and Kawa among others, as well as for refugees from other
4262 Further documentation may be found in the optargs.scm file itself.
4264 The optional argument module also exports the macros `let-optional',
4265 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4266 are not documented here because they may be removed in the future, but
4267 full documentation is still available in optargs.scm.
4269 ** New syntax: and-let*
4270 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4272 Syntax: (land* (<clause> ...) <body> ...)
4273 Each <clause> should have one of the following forms:
4274 (<variable> <expression>)
4277 Each <variable> or <bound-variable> should be an identifier. Each
4278 <expression> should be a valid expression. The <body> should be a
4279 possibly empty sequence of expressions, like the <body> of a
4282 Semantics: A LAND* expression is evaluated by evaluating the
4283 <expression> or <bound-variable> of each of the <clause>s from
4284 left to right. The value of the first <expression> or
4285 <bound-variable> that evaluates to a false value is returned; the
4286 remaining <expression>s and <bound-variable>s are not evaluated.
4287 The <body> forms are evaluated iff all the <expression>s and
4288 <bound-variable>s evaluate to true values.
4290 The <expression>s and the <body> are evaluated in an environment
4291 binding each <variable> of the preceding (<variable> <expression>)
4292 clauses to the value of the <expression>. Later bindings
4293 shadow earlier bindings.
4295 Guile's and-let* macro was contributed by Michael Livshin.
4297 ** New sorting functions
4299 *** New function: sorted? SEQUENCE LESS?
4300 Returns `#t' when the sequence argument is in non-decreasing order
4301 according to LESS? (that is, there is no adjacent pair `... x y
4302 ...' for which `(less? y x)').
4304 Returns `#f' when the sequence contains at least one out-of-order
4305 pair. It is an error if the sequence is neither a list nor a
4308 *** New function: merge LIST1 LIST2 LESS?
4309 LIST1 and LIST2 are sorted lists.
4310 Returns the sorted list of all elements in LIST1 and LIST2.
4312 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4313 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4314 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4315 (Here "<" should read "comes before".)
4317 *** New procedure: merge! LIST1 LIST2 LESS?
4318 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4319 the result. If the code is compiled, and LESS? constructs no new
4320 pairs, no pairs at all will be allocated. The first pair of the
4321 result will be either the first pair of LIST1 or the first pair of
4324 *** New function: sort SEQUENCE LESS?
4325 Accepts either a list or a vector, and returns a new sequence
4326 which is sorted. The new sequence is the same type as the input.
4327 Always `(sorted? (sort sequence less?) less?)'. The original
4328 sequence is not altered in any way. The new sequence shares its
4329 elements with the old one; no elements are copied.
4331 *** New procedure: sort! SEQUENCE LESS
4332 Returns its sorted result in the original boxes. No new storage is
4333 allocated at all. Proper usage: (set! slist (sort! slist <))
4335 *** New function: stable-sort SEQUENCE LESS?
4336 Similar to `sort' but stable. That is, if "equal" elements are
4337 ordered a < b in the original sequence, they will have the same order
4340 *** New function: stable-sort! SEQUENCE LESS?
4341 Similar to `sort!' but stable.
4342 Uses temporary storage when sorting vectors.
4344 *** New functions: sort-list, sort-list!
4345 Added for compatibility with scsh.
4347 ** New built-in random number support
4349 *** New function: random N [STATE]
4350 Accepts a positive integer or real N and returns a number of the
4351 same type between zero (inclusive) and N (exclusive). The values
4352 returned have a uniform distribution.
4354 The optional argument STATE must be of the type produced by
4355 `copy-random-state' or `seed->random-state'. It defaults to the value
4356 of the variable `*random-state*'. This object is used to maintain the
4357 state of the pseudo-random-number generator and is altered as a side
4358 effect of the `random' operation.
4360 *** New variable: *random-state*
4361 Holds a data structure that encodes the internal state of the
4362 random-number generator that `random' uses by default. The nature
4363 of this data structure is implementation-dependent. It may be
4364 printed out and successfully read back in, but may or may not
4365 function correctly as a random-number state object in another
4368 *** New function: copy-random-state [STATE]
4369 Returns a new object of type suitable for use as the value of the
4370 variable `*random-state*' and as a second argument to `random'.
4371 If argument STATE is given, a copy of it is returned. Otherwise a
4372 copy of `*random-state*' is returned.
4374 *** New function: seed->random-state SEED
4375 Returns a new object of type suitable for use as the value of the
4376 variable `*random-state*' and as a second argument to `random'.
4377 SEED is a string or a number. A new state is generated and
4378 initialized using SEED.
4380 *** New function: random:uniform [STATE]
4381 Returns an uniformly distributed inexact real random number in the
4382 range between 0 and 1.
4384 *** New procedure: random:solid-sphere! VECT [STATE]
4385 Fills VECT with inexact real random numbers the sum of whose
4386 squares is less than 1.0. Thinking of VECT as coordinates in
4387 space of dimension N = `(vector-length VECT)', the coordinates are
4388 uniformly distributed within the unit N-shere. The sum of the
4389 squares of the numbers is returned. VECT can be either a vector
4390 or a uniform vector of doubles.
4392 *** New procedure: random:hollow-sphere! VECT [STATE]
4393 Fills VECT with inexact real random numbers the sum of whose squares
4394 is equal to 1.0. Thinking of VECT as coordinates in space of
4395 dimension n = `(vector-length VECT)', the coordinates are uniformly
4396 distributed over the surface of the unit n-shere. VECT can be either
4397 a vector or a uniform vector of doubles.
4399 *** New function: random:normal [STATE]
4400 Returns an inexact real in a normal distribution with mean 0 and
4401 standard deviation 1. For a normal distribution with mean M and
4402 standard deviation D use `(+ M (* D (random:normal)))'.
4404 *** New procedure: random:normal-vector! VECT [STATE]
4405 Fills VECT with inexact real random numbers which are independent and
4406 standard normally distributed (i.e., with mean 0 and variance 1).
4407 VECT can be either a vector or a uniform vector of doubles.
4409 *** New function: random:exp STATE
4410 Returns an inexact real in an exponential distribution with mean 1.
4411 For an exponential distribution with mean U use (* U (random:exp)).
4413 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4415 These functions now operate on numbers in the range of a C unsigned
4418 These functions used to operate on numbers in the range of a C signed
4419 long; however, this seems inappropriate, because Guile integers don't
4422 ** New function: make-guardian
4423 This is an implementation of guardians as described in
4424 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4425 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4426 Programming Language Design and Implementation, June 1993
4427 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4429 ** New functions: delq1!, delv1!, delete1!
4430 These procedures behave similar to delq! and friends but delete only
4431 one object if at all.
4433 ** New function: unread-string STRING PORT
4434 Unread STRING to PORT, that is, push it back onto the port so that
4435 next read operation will work on the pushed back characters.
4437 ** unread-char can now be called multiple times
4438 If unread-char is called multiple times, the unread characters will be
4439 read again in last-in first-out order.
4441 ** the procedures uniform-array-read! and uniform-array-write! now
4442 work on any kind of port, not just ports which are open on a file.
4444 ** Now 'l' in a port mode requests line buffering.
4446 ** The procedure truncate-file now works on string ports as well
4447 as file ports. If the size argument is omitted, the current
4448 file position is used.
4450 ** new procedure: seek PORT/FDES OFFSET WHENCE
4451 The arguments are the same as for the old fseek procedure, but it
4452 works on string ports as well as random-access file ports.
4454 ** the fseek procedure now works on string ports, since it has been
4455 redefined using seek.
4457 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4458 size is not supplied.
4460 ** the newline procedure no longer flushes the port if it's not
4461 line-buffered: previously it did if it was the current output port.
4463 ** open-pipe and close-pipe are no longer primitive procedures, but
4464 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4466 ** the freopen procedure has been removed.
4468 ** new procedure: drain-input PORT
4469 Drains PORT's read buffers (including any pushed-back characters)
4470 and returns the contents as a single string.
4472 ** New function: map-in-order PROC LIST1 LIST2 ...
4473 Version of `map' which guarantees that the procedure is applied to the
4474 lists in serial order.
4476 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4477 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4478 now obsolete and will go away in release 1.5.
4480 ** New syntax: collect BODY1 ...
4481 Version of `begin' which returns a list of the results of the body
4482 forms instead of the result of the last body form. In contrast to
4483 `begin', `collect' allows an empty body.
4485 ** New functions: read-history FILENAME, write-history FILENAME
4486 Read/write command line history from/to file. Returns #t on success
4487 and #f if an error occured.
4489 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4491 These procedures return a list of definitions available in the specified
4492 argument, a relative module reference. In the case of no argument,
4493 `(current-module)' is now consulted for definitions to return, instead
4494 of simply returning #f, the former behavior.
4496 ** The #/ syntax for lists is no longer supported.
4498 Earlier versions of Scheme accepted this syntax, but printed a
4501 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4503 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4506 * Changes to the gh_ interface
4510 Now takes a second argument which is the result array. If this
4511 pointer is NULL, a new array is malloced (the old behaviour).
4513 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4514 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4518 * Changes to the scm_ interface
4520 ** Function: scm_make_named_hook (char* name, int n_args)
4522 Creates a hook in the same way as make-hook above but also
4523 binds a variable named NAME to it.
4525 This is the typical way of creating a hook from C code.
4527 Currently, the variable is created in the "current" module. This
4528 might change when we get the new module system.
4530 ** The smob interface
4532 The interface for creating smobs has changed. For documentation, see
4533 data-rep.info (made from guile-core/doc/data-rep.texi).
4535 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4537 >>> This function will be removed in 1.3.4. <<<
4541 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4542 This function adds a new smob type, named NAME, with instance size
4543 SIZE to the system. The return value is a tag that is used in
4544 creating instances of the type. If SIZE is 0, then no memory will
4545 be allocated when instances of the smob are created, and nothing
4546 will be freed by the default free function.
4548 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4549 This function sets the smob marking procedure for the smob type
4550 specified by the tag TC. TC is the tag returned by
4551 `scm_make_smob_type'.
4553 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4554 This function sets the smob freeing procedure for the smob type
4555 specified by the tag TC. TC is the tag returned by
4556 `scm_make_smob_type'.
4558 *** Function: void scm_set_smob_print (tc, print)
4560 - Function: void scm_set_smob_print (long tc,
4561 scm_sizet (*print) (SCM,
4565 This function sets the smob printing procedure for the smob type
4566 specified by the tag TC. TC is the tag returned by
4567 `scm_make_smob_type'.
4569 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4570 This function sets the smob equality-testing predicate for the
4571 smob type specified by the tag TC. TC is the tag returned by
4572 `scm_make_smob_type'.
4574 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4575 Make VALUE contain a smob instance of the type with type code TC and
4576 smob data DATA. VALUE must be previously declared as C type `SCM'.
4578 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4579 This macro expands to a block of code that creates a smob instance
4580 of the type with type code TC and smob data DATA, and returns that
4581 `SCM' value. It should be the last piece of code in a block.
4583 ** The interfaces for using I/O ports and implementing port types
4584 (ptobs) have changed significantly. The new interface is based on
4585 shared access to buffers and a new set of ptob procedures.
4587 *** scm_newptob has been removed
4591 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4593 - Function: SCM scm_make_port_type (char *type_name,
4594 int (*fill_buffer) (SCM port),
4595 void (*write_flush) (SCM port));
4597 Similarly to the new smob interface, there is a set of function
4598 setters by which the user can customize the behaviour of his port
4599 type. See ports.h (scm_set_port_XXX).
4601 ** scm_strport_to_string: New function: creates a new string from
4602 a string port's buffer.
4604 ** Plug in interface for random number generators
4605 The variable `scm_the_rng' in random.c contains a value and three
4606 function pointers which together define the current random number
4607 generator being used by the Scheme level interface and the random
4608 number library functions.
4610 The user is free to replace the default generator with the generator
4613 *** Variable: size_t scm_the_rng.rstate_size
4614 The size of the random state type used by the current RNG
4617 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4618 Given the random STATE, return 32 random bits.
4620 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4621 Seed random state STATE using string S of length N.
4623 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4624 Given random state STATE, return a malloced copy.
4627 The default RNG is the MWC (Multiply With Carry) random number
4628 generator described by George Marsaglia at the Department of
4629 Statistics and Supercomputer Computations Research Institute, The
4630 Florida State University (http://stat.fsu.edu/~geo).
4632 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4633 passes all tests in the DIEHARD test suite
4634 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4635 costs one multiply and one add on platforms which either supports long
4636 longs (gcc does this on most systems) or have 64 bit longs. The cost
4637 is four multiply on other systems but this can be optimized by writing
4638 scm_i_uniform32 in assembler.
4640 These functions are provided through the scm_the_rng interface for use
4641 by libguile and the application.
4643 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4644 Given the random STATE, return 32 random bits.
4645 Don't use this function directly. Instead go through the plugin
4646 interface (see "Plug in interface" above).
4648 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4649 Initialize STATE using SEED of length N.
4651 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4652 Return a malloc:ed copy of STATE. This function can easily be re-used
4653 in the interfaces to other RNGs.
4655 ** Random number library functions
4656 These functions use the current RNG through the scm_the_rng interface.
4657 It might be a good idea to use these functions from your C code so
4658 that only one random generator is used by all code in your program.
4660 The default random state is stored in:
4662 *** Variable: SCM scm_var_random_state
4663 Contains the vcell of the Scheme variable "*random-state*" which is
4664 used as default state by all random number functions in the Scheme
4669 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4671 *** Function: scm_rstate *scm_c_default_rstate (void)
4672 This is a convenience function which returns the value of
4673 scm_var_random_state. An error message is generated if this value
4674 isn't a random state.
4676 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4677 Make a new random state from the string SEED of length LENGTH.
4679 It is generally not a good idea to use multiple random states in a
4680 program. While subsequent random numbers generated from one random
4681 state are guaranteed to be reasonably independent, there is no such
4682 guarantee for numbers generated from different random states.
4684 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4685 Return 32 random bits.
4687 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4688 Return a sample from the uniform(0,1) distribution.
4690 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4691 Return a sample from the normal(0,1) distribution.
4693 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4694 Return a sample from the exp(1) distribution.
4696 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4697 Return a sample from the discrete uniform(0,M) distribution.
4699 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4700 Return a sample from the discrete uniform(0,M) distribution.
4701 M must be a bignum object. The returned value may be an INUM.
4705 Changes in Guile 1.3 (released Monday, October 19, 1998):
4707 * Changes to the distribution
4709 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4710 To avoid conflicts, programs should name environment variables after
4711 themselves, except when there's a common practice establishing some
4714 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4715 giving the former precedence, and printing a warning message if the
4716 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4718 ** The header files related to multi-byte characters have been removed.
4719 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4720 which referred to these explicitly will probably need to be rewritten,
4721 since the support for the variant string types has been removed; see
4724 ** The header files append.h and sequences.h have been removed. These
4725 files implemented non-R4RS operations which would encourage
4726 non-portable programming style and less easy-to-read code.
4728 * Changes to the stand-alone interpreter
4730 ** New procedures have been added to implement a "batch mode":
4732 *** Function: batch-mode?
4734 Returns a boolean indicating whether the interpreter is in batch
4737 *** Function: set-batch-mode?! ARG
4739 If ARG is true, switches the interpreter to batch mode. The `#f'
4740 case has not been implemented.
4742 ** Guile now provides full command-line editing, when run interactively.
4743 To use this feature, you must have the readline library installed.
4744 The Guile build process will notice it, and automatically include
4747 The readline library is available via anonymous FTP from any GNU
4748 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4750 ** the-last-stack is now a fluid.
4752 * Changes to the procedure for linking libguile with your programs
4754 ** You can now use the `guile-config' utility to build programs that use Guile.
4756 Guile now includes a command-line utility called `guile-config', which
4757 can provide information about how to compile and link programs that
4760 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4761 You should include this command's output on the command line you use
4762 to compile C or C++ code that #includes the Guile header files. It's
4763 usually just a `-I' flag to help the compiler find the Guile headers.
4766 *** `guile-config link' prints any linker flags necessary to link with Guile.
4768 This command writes to its standard output a list of flags which you
4769 must pass to the linker to link your code against the Guile library.
4770 The flags include '-lguile' itself, any other libraries the Guile
4771 library depends upon, and any `-L' flags needed to help the linker
4772 find those libraries.
4774 For example, here is a Makefile rule that builds a program named 'foo'
4775 from the object files ${FOO_OBJECTS}, and links them against Guile:
4778 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4780 Previous Guile releases recommended that you use autoconf to detect
4781 which of a predefined set of libraries were present on your system.
4782 It is more robust to use `guile-config', since it records exactly which
4783 libraries the installed Guile library requires.
4785 This was originally called `build-guile', but was renamed to
4786 `guile-config' before Guile 1.3 was released, to be consistent with
4787 the analogous script for the GTK+ GUI toolkit, which is called
4791 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4793 If you are using the GNU autoconf package to configure your program,
4794 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4795 (described above) and gather the necessary values for use in your
4798 The GUILE_FLAGS macro expands to configure script code which runs the
4799 `guile-config' script, to find out where Guile's header files and
4800 libraries are installed. It sets two variables, marked for
4801 substitution, as by AC_SUBST.
4803 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4804 code that uses Guile header files. This is almost always just a
4807 GUILE_LDFLAGS --- flags to pass to the linker to link a
4808 program against Guile. This includes `-lguile' for the Guile
4809 library itself, any libraries that Guile itself requires (like
4810 -lqthreads), and so on. It may also include a -L flag to tell the
4811 compiler where to find the libraries.
4813 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4814 directory of the Guile distribution. You can copy it into your
4815 package's aclocal.m4 file, and then use it in your configure.in file.
4817 If you are using the `aclocal' program, distributed with GNU automake,
4818 to maintain your aclocal.m4 file, the Guile installation process
4819 installs guile.m4 where aclocal will find it. All you need to do is
4820 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4821 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4825 * Changes to Scheme functions and syntax
4827 ** Multi-byte strings have been removed, as have multi-byte and wide
4828 ports. We felt that these were the wrong approach to
4829 internationalization support.
4831 ** New function: readline [PROMPT]
4832 Read a line from the terminal, and allow the user to edit it,
4833 prompting with PROMPT. READLINE provides a large set of Emacs-like
4834 editing commands, lets the user recall previously typed lines, and
4835 works on almost every kind of terminal, including dumb terminals.
4837 READLINE assumes that the cursor is at the beginning of the line when
4838 it is invoked. Thus, you can't print a prompt yourself, and then call
4839 READLINE; you need to package up your prompt as a string, pass it to
4840 the function, and let READLINE print the prompt itself. This is
4841 because READLINE needs to know the prompt's screen width.
4843 For Guile to provide this function, you must have the readline
4844 library, version 2.1 or later, installed on your system. Readline is
4845 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4846 any GNU mirror site.
4848 See also ADD-HISTORY function.
4850 ** New function: add-history STRING
4851 Add STRING as the most recent line in the history used by the READLINE
4852 command. READLINE does not add lines to the history itself; you must
4853 call ADD-HISTORY to make previous input available to the user.
4855 ** The behavior of the read-line function has changed.
4857 This function now uses standard C library functions to read the line,
4858 for speed. This means that it doesn not respect the value of
4859 scm-line-incrementors; it assumes that lines are delimited with
4862 (Note that this is read-line, the function that reads a line of text
4863 from a port, not readline, the function that reads a line from a
4864 terminal, providing full editing capabilities.)
4866 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4868 This module provides some simple argument parsing. It exports one
4871 Function: getopt-gnu-style ARG-LS
4872 Parse a list of program arguments into an alist of option
4875 Each item in the list of program arguments is examined to see if
4876 it meets the syntax of a GNU long-named option. An argument like
4877 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4878 returned alist, where MUMBLE is a keyword object with the same
4879 name as the argument. An argument like `--MUMBLE=FROB' produces
4880 an element of the form (MUMBLE . FROB), where FROB is a string.
4882 As a special case, the returned alist also contains a pair whose
4883 car is the symbol `rest'. The cdr of this pair is a list
4884 containing all the items in the argument list that are not options
4885 of the form mentioned above.
4887 The argument `--' is treated specially: all items in the argument
4888 list appearing after such an argument are not examined, and are
4889 returned in the special `rest' list.
4891 This function does not parse normal single-character switches.
4892 You will need to parse them out of the `rest' list yourself.
4894 ** The read syntax for byte vectors and short vectors has changed.
4896 Instead of #bytes(...), write #y(...).
4898 Instead of #short(...), write #h(...).
4900 This may seem nutty, but, like the other uniform vectors, byte vectors
4901 and short vectors want to have the same print and read syntax (and,
4902 more basic, want to have read syntax!). Changing the read syntax to
4903 use multiple characters after the hash sign breaks with the
4904 conventions used in R5RS and the conventions used for the other
4905 uniform vectors. It also introduces complexity in the current reader,
4906 both on the C and Scheme levels. (The Right solution is probably to
4907 change the syntax and prototypes for uniform vectors entirely.)
4910 ** The new module (ice-9 session) provides useful interactive functions.
4912 *** New procedure: (apropos REGEXP OPTION ...)
4914 Display a list of top-level variables whose names match REGEXP, and
4915 the modules they are imported from. Each OPTION should be one of the
4918 value --- Show the value of each matching variable.
4919 shadow --- Show bindings shadowed by subsequently imported modules.
4920 full --- Same as both `shadow' and `value'.
4924 guile> (apropos "trace" 'full)
4925 debug: trace #<procedure trace args>
4926 debug: untrace #<procedure untrace args>
4927 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4928 the-scm-module: before-backtrace-hook ()
4929 the-scm-module: backtrace #<primitive-procedure backtrace>
4930 the-scm-module: after-backtrace-hook ()
4931 the-scm-module: has-shown-backtrace-hint? #f
4934 ** There are new functions and syntax for working with macros.
4936 Guile implements macros as a special object type. Any variable whose
4937 top-level binding is a macro object acts as a macro. The macro object
4938 specifies how the expression should be transformed before evaluation.
4940 *** Macro objects now print in a reasonable way, resembling procedures.
4942 *** New function: (macro? OBJ)
4943 True iff OBJ is a macro object.
4945 *** New function: (primitive-macro? OBJ)
4946 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4947 macro transformers, implemented in eval.c rather than Scheme code.
4949 Why do we have this function?
4950 - For symmetry with procedure? and primitive-procedure?,
4951 - to allow custom print procedures to tell whether a macro is
4952 primitive, and display it differently, and
4953 - to allow compilers and user-written evaluators to distinguish
4954 builtin special forms from user-defined ones, which could be
4957 *** New function: (macro-type OBJ)
4958 Return a value indicating what kind of macro OBJ is. Possible return
4961 The symbol `syntax' --- a macro created by procedure->syntax.
4962 The symbol `macro' --- a macro created by procedure->macro.
4963 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4964 The boolean #f --- if OBJ is not a macro object.
4966 *** New function: (macro-name MACRO)
4967 Return the name of the macro object MACRO's procedure, as returned by
4970 *** New function: (macro-transformer MACRO)
4971 Return the transformer procedure for MACRO.
4973 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4975 Specify a new macro expander to use in the current module. Each
4976 MODULE is a module name, with the same meaning as in the `use-modules'
4977 form; each named module's exported bindings are added to the current
4978 top-level environment. TRANSFORMER is an expression evaluated in the
4979 resulting environment which must yield a procedure to use as the
4980 module's eval transformer: every expression evaluated in this module
4981 is passed to this function, and the result passed to the Guile
4984 *** macro-eval! is removed. Use local-eval instead.
4986 ** Some magic has been added to the printer to better handle user
4987 written printing routines (like record printers, closure printers).
4989 The problem is that these user written routines must have access to
4990 the current `print-state' to be able to handle fancy things like
4991 detection of circular references. These print-states have to be
4992 passed to the builtin printing routines (display, write, etc) to
4993 properly continue the print chain.
4995 We didn't want to change all existing print code so that it
4996 explicitly passes thru a print state in addition to a port. Instead,
4997 we extented the possible values that the builtin printing routines
4998 accept as a `port'. In addition to a normal port, they now also take
4999 a pair of a normal port and a print-state. Printing will go to the
5000 port and the print-state will be used to control the detection of
5001 circular references, etc. If the builtin function does not care for a
5002 print-state, it is simply ignored.
5004 User written callbacks are now called with such a pair as their
5005 `port', but because every function now accepts this pair as a PORT
5006 argument, you don't have to worry about that. In fact, it is probably
5007 safest to not check for these pairs.
5009 However, it is sometimes necessary to continue a print chain on a
5010 different port, for example to get a intermediate string
5011 representation of the printed value, mangle that string somehow, and
5012 then to finally print the mangled string. Use the new function
5014 inherit-print-state OLD-PORT NEW-PORT
5016 for this. It constructs a new `port' that prints to NEW-PORT but
5017 inherits the print-state of OLD-PORT.
5019 ** struct-vtable-offset renamed to vtable-offset-user
5021 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5023 ** There is now a third optional argument to make-vtable-vtable
5024 (and fourth to make-struct) when constructing new types (vtables).
5025 This argument initializes field vtable-index-printer of the vtable.
5027 ** The detection of circular references has been extended to structs.
5028 That is, a structure that -- in the process of being printed -- prints
5029 itself does not lead to infinite recursion.
5031 ** There is now some basic support for fluids. Please read
5032 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5033 the following functions and macros:
5035 Function: make-fluid
5037 Create a new fluid object. Fluids are not special variables or
5038 some other extension to the semantics of Scheme, but rather
5039 ordinary Scheme objects. You can store them into variables (that
5040 are still lexically scoped, of course) or into any other place you
5041 like. Every fluid has a initial value of `#f'.
5043 Function: fluid? OBJ
5045 Test whether OBJ is a fluid.
5047 Function: fluid-ref FLUID
5048 Function: fluid-set! FLUID VAL
5050 Access/modify the fluid FLUID. Modifications are only visible
5051 within the current dynamic root (that includes threads).
5053 Function: with-fluids* FLUIDS VALUES THUNK
5055 FLUIDS is a list of fluids and VALUES a corresponding list of
5056 values for these fluids. Before THUNK gets called the values are
5057 installed in the fluids and the old values of the fluids are
5058 saved in the VALUES list. When the flow of control leaves THUNK
5059 or reenters it, the values get swapped again. You might think of
5060 this as a `safe-fluid-excursion'. Note that the VALUES list is
5061 modified by `with-fluids*'.
5063 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5065 The same as `with-fluids*' but with a different syntax. It looks
5066 just like `let', but both FLUID and VALUE are evaluated. Remember,
5067 fluids are not special variables but ordinary objects. FLUID
5068 should evaluate to a fluid.
5070 ** Changes to system call interfaces:
5072 *** close-port, close-input-port and close-output-port now return a
5073 boolean instead of an `unspecified' object. #t means that the port
5074 was successfully closed, while #f means it was already closed. It is
5075 also now possible for these procedures to raise an exception if an
5076 error occurs (some errors from write can be delayed until close.)
5078 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5081 *** the third argument to fcntl is now optional.
5083 *** the first argument to chown can now be a file descriptor or a port.
5085 *** the argument to stat can now be a port.
5087 *** The following new procedures have been added (most use scsh
5090 *** procedure: close PORT/FD
5091 Similar to close-port (*note close-port: Closing Ports.), but also
5092 works on file descriptors. A side effect of closing a file
5093 descriptor is that any ports using that file descriptor are moved
5094 to a different file descriptor and have their revealed counts set
5097 *** procedure: port->fdes PORT
5098 Returns the integer file descriptor underlying PORT. As a side
5099 effect the revealed count of PORT is incremented.
5101 *** procedure: fdes->ports FDES
5102 Returns a list of existing ports which have FDES as an underlying
5103 file descriptor, without changing their revealed counts.
5105 *** procedure: fdes->inport FDES
5106 Returns an existing input port which has FDES as its underlying
5107 file descriptor, if one exists, and increments its revealed count.
5108 Otherwise, returns a new input port with a revealed count of 1.
5110 *** procedure: fdes->outport FDES
5111 Returns an existing output port which has FDES as its underlying
5112 file descriptor, if one exists, and increments its revealed count.
5113 Otherwise, returns a new output port with a revealed count of 1.
5115 The next group of procedures perform a `dup2' system call, if NEWFD
5116 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5117 duplicated can be supplied as an integer or contained in a port. The
5118 type of value returned varies depending on which procedure is used.
5120 All procedures also have the side effect when performing `dup2' that
5121 any ports using NEWFD are moved to a different file descriptor and have
5122 their revealed counts set to zero.
5124 *** procedure: dup->fdes PORT/FD [NEWFD]
5125 Returns an integer file descriptor.
5127 *** procedure: dup->inport PORT/FD [NEWFD]
5128 Returns a new input port using the new file descriptor.
5130 *** procedure: dup->outport PORT/FD [NEWFD]
5131 Returns a new output port using the new file descriptor.
5133 *** procedure: dup PORT/FD [NEWFD]
5134 Returns a new port if PORT/FD is a port, with the same mode as the
5135 supplied port, otherwise returns an integer file descriptor.
5137 *** procedure: dup->port PORT/FD MODE [NEWFD]
5138 Returns a new port using the new file descriptor. MODE supplies a
5139 mode string for the port (*note open-file: File Ports.).
5141 *** procedure: setenv NAME VALUE
5142 Modifies the environment of the current process, which is also the
5143 default environment inherited by child processes.
5145 If VALUE is `#f', then NAME is removed from the environment.
5146 Otherwise, the string NAME=VALUE is added to the environment,
5147 replacing any existing string with name matching NAME.
5149 The return value is unspecified.
5151 *** procedure: truncate-file OBJ SIZE
5152 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5153 can be a string containing a file name or an integer file
5154 descriptor or port open for output on the file. The underlying
5155 system calls are `truncate' and `ftruncate'.
5157 The return value is unspecified.
5159 *** procedure: setvbuf PORT MODE [SIZE]
5160 Set the buffering mode for PORT. MODE can be:
5168 block buffered, using a newly allocated buffer of SIZE bytes.
5169 However if SIZE is zero or unspecified, the port will be made
5172 This procedure should not be used after I/O has been performed with
5175 Ports are usually block buffered by default, with a default buffer
5176 size. Procedures e.g., *Note open-file: File Ports, which accept a
5177 mode string allow `0' to be added to request an unbuffered port.
5179 *** procedure: fsync PORT/FD
5180 Copies any unwritten data for the specified output file descriptor
5181 to disk. If PORT/FD is a port, its buffer is flushed before the
5182 underlying file descriptor is fsync'd. The return value is
5185 *** procedure: open-fdes PATH FLAGS [MODES]
5186 Similar to `open' but returns a file descriptor instead of a port.
5188 *** procedure: execle PATH ENV [ARG] ...
5189 Similar to `execl', but the environment of the new process is
5190 specified by ENV, which must be a list of strings as returned by
5191 the `environ' procedure.
5193 This procedure is currently implemented using the `execve' system
5194 call, but we call it `execle' because of its Scheme calling
5197 *** procedure: strerror ERRNO
5198 Returns the Unix error message corresponding to ERRNO, an integer.
5200 *** procedure: primitive-exit [STATUS]
5201 Terminate the current process without unwinding the Scheme stack.
5202 This is would typically be useful after a fork. The exit status
5203 is STATUS if supplied, otherwise zero.
5205 *** procedure: times
5206 Returns an object with information about real and processor time.
5207 The following procedures accept such an object as an argument and
5208 return a selected component:
5211 The current real time, expressed as time units relative to an
5215 The CPU time units used by the calling process.
5218 The CPU time units used by the system on behalf of the
5222 The CPU time units used by terminated child processes of the
5223 calling process, whose status has been collected (e.g., using
5227 Similarly, the CPU times units used by the system on behalf of
5228 terminated child processes.
5230 ** Removed: list-length
5231 ** Removed: list-append, list-append!
5232 ** Removed: list-reverse, list-reverse!
5234 ** array-map renamed to array-map!
5236 ** serial-array-map renamed to serial-array-map!
5238 ** catch doesn't take #f as first argument any longer
5240 Previously, it was possible to pass #f instead of a key to `catch'.
5241 That would cause `catch' to pass a jump buffer object to the procedure
5242 passed as second argument. The procedure could then use this jump
5243 buffer objekt as an argument to throw.
5245 This mechanism has been removed since its utility doesn't motivate the
5246 extra complexity it introduces.
5248 ** The `#/' notation for lists now provokes a warning message from Guile.
5249 This syntax will be removed from Guile in the near future.
5251 To disable the warning message, set the GUILE_HUSH environment
5252 variable to any non-empty value.
5254 ** The newline character now prints as `#\newline', following the
5255 normal Scheme notation, not `#\nl'.
5257 * Changes to the gh_ interface
5259 ** The gh_enter function now takes care of loading the Guile startup files.
5260 gh_enter works by calling scm_boot_guile; see the remarks below.
5262 ** Function: void gh_write (SCM x)
5264 Write the printed representation of the scheme object x to the current
5265 output port. Corresponds to the scheme level `write'.
5267 ** gh_list_length renamed to gh_length.
5269 ** vector handling routines
5271 Several major changes. In particular, gh_vector() now resembles
5272 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5273 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5274 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5275 vector-related gh_ functions have been implemented.
5277 ** pair and list routines
5279 Implemented several of the R4RS pair and list functions that were
5282 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5284 New function. Converts double arrays back and forth between Scheme
5287 * Changes to the scm_ interface
5289 ** The function scm_boot_guile now takes care of loading the startup files.
5291 Guile's primary initialization function, scm_boot_guile, now takes
5292 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5293 Guile, define the module system, and put together some standard
5294 bindings. It also loads `init.scm', which is intended to hold
5295 site-specific initialization code.
5297 Since Guile cannot operate properly until boot-9.scm is loaded, there
5298 is no reason to separate loading boot-9.scm from Guile's other
5299 initialization processes.
5301 This job used to be done by scm_compile_shell_switches, which didn't
5302 make much sense; in particular, it meant that people using Guile for
5303 non-shell-like applications had to jump through hoops to get Guile
5304 initialized properly.
5306 ** The function scm_compile_shell_switches no longer loads the startup files.
5307 Now, Guile always loads the startup files, whenever it is initialized;
5308 see the notes above for scm_boot_guile and scm_load_startup_files.
5310 ** Function: scm_load_startup_files
5311 This new function takes care of loading Guile's initialization file
5312 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5313 this is always called by the Guile initialization process, it's
5314 probably not too useful to call this yourself, but it's there anyway.
5316 ** The semantics of smob marking have changed slightly.
5318 The smob marking function (the `mark' member of the scm_smobfuns
5319 structure) is no longer responsible for setting the mark bit on the
5320 smob. The generic smob handling code in the garbage collector will
5321 set this bit. The mark function need only ensure that any other
5322 objects the smob refers to get marked.
5324 Note that this change means that the smob's GC8MARK bit is typically
5325 already set upon entry to the mark function. Thus, marking functions
5326 which look like this:
5329 if (SCM_GC8MARKP (ptr))
5331 SCM_SETGC8MARK (ptr);
5332 ... mark objects to which the smob refers ...
5335 are now incorrect, since they will return early, and fail to mark any
5336 other objects the smob refers to. Some code in the Guile library used
5339 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5341 If you have implemented your own I/O port type, by writing the
5342 functions required by the scm_ptobfuns and then calling scm_newptob,
5343 you will need to change your functions slightly.
5345 The functions in a scm_ptobfuns structure now expect the port itself
5346 as their argument; they used to expect the `stream' member of the
5347 port's scm_port_table structure. This allows functions in an
5348 scm_ptobfuns structure to easily access the port's cell (and any flags
5349 it its CAR), and the port's scm_port_table structure.
5351 Guile now passes the I/O port itself as the `port' argument in the
5352 following scm_ptobfuns functions:
5354 int (*free) (SCM port);
5355 int (*fputc) (int, SCM port);
5356 int (*fputs) (char *, SCM port);
5357 scm_sizet (*fwrite) SCM_P ((char *ptr,
5361 int (*fflush) (SCM port);
5362 int (*fgetc) (SCM port);
5363 int (*fclose) (SCM port);
5365 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5368 If you have existing code which defines its own port types, it is easy
5369 to convert your code to the new interface; simply apply SCM_STREAM to
5370 the port argument to yield the value you code used to expect.
5372 Note that since both the port and the stream have the same type in the
5373 C code --- they are both SCM values --- the C compiler will not remind
5374 you if you forget to update your scm_ptobfuns functions.
5377 ** Function: int scm_internal_select (int fds,
5381 struct timeval *timeout);
5383 This is a replacement for the `select' function provided by the OS.
5384 It enables I/O blocking and sleeping to happen for one cooperative
5385 thread without blocking other threads. It also avoids busy-loops in
5386 these situations. It is intended that all I/O blocking and sleeping
5387 will finally go through this function. Currently, this function is
5388 only available on systems providing `gettimeofday' and `select'.
5390 ** Function: SCM scm_internal_stack_catch (SCM tag,
5391 scm_catch_body_t body,
5393 scm_catch_handler_t handler,
5396 A new sibling to the other two C level `catch' functions
5397 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5398 the stack to be saved automatically into the variable `the-last-stack'
5399 (scm_the_last_stack_var) on error. This is necessary if you want to
5400 use advanced error reporting, such as calling scm_display_error and
5401 scm_display_backtrace. (They both take a stack object as argument.)
5403 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5405 scm_catch_handler_t handler,
5408 Spawns a new thread. It does a job similar to
5409 scm_call_with_new_thread but takes arguments more suitable when
5410 spawning threads from application C code.
5412 ** The hook scm_error_callback has been removed. It was originally
5413 intended as a way for the user to install his own error handler. But
5414 that method works badly since it intervenes between throw and catch,
5415 thereby changing the semantics of expressions like (catch #t ...).
5416 The correct way to do it is to use one of the C level catch functions
5417 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5419 ** Removed functions:
5421 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5422 scm_list_reverse, scm_list_reverse_x
5424 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5426 These can be used for pretty list creation from C. The idea is taken
5427 from Erick Gallesio's STk.
5429 ** scm_array_map renamed to scm_array_map_x
5431 ** mbstrings are now removed
5433 This means that the type codes scm_tc7_mb_string and
5434 scm_tc7_mb_substring has been removed.
5436 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5438 Since we no longer support multi-byte strings, these I/O functions
5439 have been simplified, and renamed. Here are their old names, and
5440 their new names and arguments:
5442 scm_gen_putc -> void scm_putc (int c, SCM port);
5443 scm_gen_puts -> void scm_puts (char *s, SCM port);
5444 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5445 scm_gen_getc -> void scm_getc (SCM port);
5448 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5450 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5452 SCM_TYP7S now masks away the bit which distinguishes substrings from
5455 ** scm_catch_body_t: Backward incompatible change!
5457 Body functions to scm_internal_catch and friends do not any longer
5458 take a second argument. This is because it is no longer possible to
5459 pass a #f arg to catch.
5461 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5463 The function scm_protect_object protects its argument from being freed
5464 by the garbage collector. scm_unprotect_object removes that
5467 These functions now nest properly. That is, for every object O, there
5468 is a counter which scm_protect_object(O) increments and
5469 scm_unprotect_object(O) decrements, if the counter is greater than
5470 zero. Every object's counter is zero when it is first created. If an
5471 object's counter is greater than zero, the garbage collector will not
5472 reclaim its storage.
5474 This allows you to use scm_protect_object in your code without
5475 worrying that some other function you call will call
5476 scm_unprotect_object, and allow it to be freed. Assuming that the
5477 functions you call are well-behaved, and unprotect only those objects
5478 they protect, you can follow the same rule and have confidence that
5479 objects will be freed only at appropriate times.
5482 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5484 * Changes to the distribution
5486 ** Nightly snapshots are now available from ftp.red-bean.com.
5487 The old server, ftp.cyclic.com, has been relinquished to its rightful
5490 Nightly snapshots of the Guile development sources are now available via
5491 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5493 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5494 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5496 ** To run Guile without installing it, the procedure has changed a bit.
5498 If you used a separate build directory to compile Guile, you'll need
5499 to include the build directory in SCHEME_LOAD_PATH, as well as the
5500 source directory. See the `INSTALL' file for examples.
5502 * Changes to the procedure for linking libguile with your programs
5504 ** The standard Guile load path for Scheme code now includes
5505 $(datadir)/guile (usually /usr/local/share/guile). This means that
5506 you can install your own Scheme files there, and Guile will find them.
5507 (Previous versions of Guile only checked a directory whose name
5508 contained the Guile version number, so you had to re-install or move
5509 your Scheme sources each time you installed a fresh version of Guile.)
5511 The load path also includes $(datadir)/guile/site; we recommend
5512 putting individual Scheme files there. If you want to install a
5513 package with multiple source files, create a directory for them under
5516 ** Guile 1.2 will now use the Rx regular expression library, if it is
5517 installed on your system. When you are linking libguile into your own
5518 programs, this means you will have to link against -lguile, -lqt (if
5519 you configured Guile with thread support), and -lrx.
5521 If you are using autoconf to generate configuration scripts for your
5522 application, the following lines should suffice to add the appropriate
5523 libraries to your link command:
5525 ### Find Rx, quickthreads and libguile.
5526 AC_CHECK_LIB(rx, main)
5527 AC_CHECK_LIB(qt, main)
5528 AC_CHECK_LIB(guile, scm_shell)
5530 The Guile 1.2 distribution does not contain sources for the Rx
5531 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5532 retrieve it from a GNU FTP site and install it separately.
5534 * Changes to Scheme functions and syntax
5536 ** The dynamic linking features of Guile are now enabled by default.
5537 You can disable them by giving the `--disable-dynamic-linking' option
5540 (dynamic-link FILENAME)
5542 Find the object file denoted by FILENAME (a string) and link it
5543 into the running Guile application. When everything works out,
5544 return a Scheme object suitable for representing the linked object
5545 file. Otherwise an error is thrown. How object files are
5546 searched is system dependent.
5548 (dynamic-object? VAL)
5550 Determine whether VAL represents a dynamically linked object file.
5552 (dynamic-unlink DYNOBJ)
5554 Unlink the indicated object file from the application. DYNOBJ
5555 should be one of the values returned by `dynamic-link'.
5557 (dynamic-func FUNCTION DYNOBJ)
5559 Search the C function indicated by FUNCTION (a string or symbol)
5560 in DYNOBJ and return some Scheme object that can later be used
5561 with `dynamic-call' to actually call this function. Right now,
5562 these Scheme objects are formed by casting the address of the
5563 function to `long' and converting this number to its Scheme
5566 (dynamic-call FUNCTION DYNOBJ)
5568 Call the C function indicated by FUNCTION and DYNOBJ. The
5569 function is passed no arguments and its return value is ignored.
5570 When FUNCTION is something returned by `dynamic-func', call that
5571 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5572 etc.), look it up in DYNOBJ; this is equivalent to
5574 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5576 Interrupts are deferred while the C function is executing (with
5577 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5579 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5581 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5582 some arguments and return its return value. The C function is
5583 expected to take two arguments and return an `int', just like
5586 int c_func (int argc, char **argv);
5588 ARGS must be a list of strings and is converted into an array of
5589 `char *'. The array is passed in ARGV and its size in ARGC. The
5590 return value is converted to a Scheme number and returned from the
5591 call to `dynamic-args-call'.
5593 When dynamic linking is disabled or not supported on your system,
5594 the above functions throw errors, but they are still available.
5596 Here is a small example that works on GNU/Linux:
5598 (define libc-obj (dynamic-link "libc.so"))
5599 (dynamic-args-call 'rand libc-obj '())
5601 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5603 ** The #/ syntax for module names is depreciated, and will be removed
5604 in a future version of Guile. Instead of
5612 The latter syntax is more consistent with existing Lisp practice.
5614 ** Guile now does fancier printing of structures. Structures are the
5615 underlying implementation for records, which in turn are used to
5616 implement modules, so all of these object now print differently and in
5617 a more informative way.
5619 The Scheme printer will examine the builtin variable *struct-printer*
5620 whenever it needs to print a structure object. When this variable is
5621 not `#f' it is deemed to be a procedure and will be applied to the
5622 structure object and the output port. When *struct-printer* is `#f'
5623 or the procedure return `#f' the structure object will be printed in
5624 the boring #<struct 80458270> form.
5626 This hook is used by some routines in ice-9/boot-9.scm to implement
5627 type specific printing routines. Please read the comments there about
5630 One of the more specific uses of structs are records. The printing
5631 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5632 called. It should behave like a *struct-printer* procedure (described
5635 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5636 token of the form #:NAME, where NAME has the same syntax as a Scheme
5637 symbol, is the external representation of the keyword named NAME.
5638 Keyword objects print using this syntax as well, so values containing
5639 keyword objects can be read back into Guile. When used in an
5640 expression, keywords are self-quoting objects.
5642 Guile suports this read syntax, and uses this print syntax, regardless
5643 of the current setting of the `keyword' read option. The `keyword'
5644 read option only controls whether Guile recognizes the `:NAME' syntax,
5645 which is incompatible with R4RS. (R4RS says such token represent
5648 ** Guile has regular expression support again. Guile 1.0 included
5649 functions for matching regular expressions, based on the Rx library.
5650 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5651 distribution, and thus Guile had no regular expression support. Guile
5652 1.2 again supports the most commonly used functions, and supports all
5653 of SCSH's regular expression functions.
5655 If your system does not include a POSIX regular expression library,
5656 and you have not linked Guile with a third-party regexp library such as
5657 Rx, these functions will not be available. You can tell whether your
5658 Guile installation includes regular expression support by checking
5659 whether the `*features*' list includes the `regex' symbol.
5661 *** regexp functions
5663 By default, Guile supports POSIX extended regular expressions. That
5664 means that the characters `(', `)', `+' and `?' are special, and must
5665 be escaped if you wish to match the literal characters.
5667 This regular expression interface was modeled after that implemented
5668 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5669 with SCSH regular expressions.
5671 **** Function: string-match PATTERN STR [START]
5672 Compile the string PATTERN into a regular expression and compare
5673 it with STR. The optional numeric argument START specifies the
5674 position of STR at which to begin matching.
5676 `string-match' returns a "match structure" which describes what,
5677 if anything, was matched by the regular expression. *Note Match
5678 Structures::. If STR does not match PATTERN at all,
5679 `string-match' returns `#f'.
5681 Each time `string-match' is called, it must compile its PATTERN
5682 argument into a regular expression structure. This operation is
5683 expensive, which makes `string-match' inefficient if the same regular
5684 expression is used several times (for example, in a loop). For better
5685 performance, you can compile a regular expression in advance and then
5686 match strings against the compiled regexp.
5688 **** Function: make-regexp STR [FLAGS]
5689 Compile the regular expression described by STR, and return the
5690 compiled regexp structure. If STR does not describe a legal
5691 regular expression, `make-regexp' throws a
5692 `regular-expression-syntax' error.
5694 FLAGS may be the bitwise-or of one or more of the following:
5696 **** Constant: regexp/extended
5697 Use POSIX Extended Regular Expression syntax when interpreting
5698 STR. If not set, POSIX Basic Regular Expression syntax is used.
5699 If the FLAGS argument is omitted, we assume regexp/extended.
5701 **** Constant: regexp/icase
5702 Do not differentiate case. Subsequent searches using the
5703 returned regular expression will be case insensitive.
5705 **** Constant: regexp/newline
5706 Match-any-character operators don't match a newline.
5708 A non-matching list ([^...]) not containing a newline matches a
5711 Match-beginning-of-line operator (^) matches the empty string
5712 immediately after a newline, regardless of whether the FLAGS
5713 passed to regexp-exec contain regexp/notbol.
5715 Match-end-of-line operator ($) matches the empty string
5716 immediately before a newline, regardless of whether the FLAGS
5717 passed to regexp-exec contain regexp/noteol.
5719 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5720 Match the compiled regular expression REGEXP against `str'. If
5721 the optional integer START argument is provided, begin matching
5722 from that position in the string. Return a match structure
5723 describing the results of the match, or `#f' if no match could be
5726 FLAGS may be the bitwise-or of one or more of the following:
5728 **** Constant: regexp/notbol
5729 The match-beginning-of-line operator always fails to match (but
5730 see the compilation flag regexp/newline above) This flag may be
5731 used when different portions of a string are passed to
5732 regexp-exec and the beginning of the string should not be
5733 interpreted as the beginning of the line.
5735 **** Constant: regexp/noteol
5736 The match-end-of-line operator always fails to match (but see the
5737 compilation flag regexp/newline above)
5739 **** Function: regexp? OBJ
5740 Return `#t' if OBJ is a compiled regular expression, or `#f'
5743 Regular expressions are commonly used to find patterns in one string
5744 and replace them with the contents of another string.
5746 **** Function: regexp-substitute PORT MATCH [ITEM...]
5747 Write to the output port PORT selected contents of the match
5748 structure MATCH. Each ITEM specifies what should be written, and
5749 may be one of the following arguments:
5751 * A string. String arguments are written out verbatim.
5753 * An integer. The submatch with that number is written.
5755 * The symbol `pre'. The portion of the matched string preceding
5756 the regexp match is written.
5758 * The symbol `post'. The portion of the matched string
5759 following the regexp match is written.
5761 PORT may be `#f', in which case nothing is written; instead,
5762 `regexp-substitute' constructs a string from the specified ITEMs
5765 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5766 Similar to `regexp-substitute', but can be used to perform global
5767 substitutions on STR. Instead of taking a match structure as an
5768 argument, `regexp-substitute/global' takes two string arguments: a
5769 REGEXP string describing a regular expression, and a TARGET string
5770 which should be matched against this regular expression.
5772 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5775 * A function may be supplied. When this function is called, it
5776 will be passed one argument: a match structure for a given
5777 regular expression match. It should return a string to be
5778 written out to PORT.
5780 * The `post' symbol causes `regexp-substitute/global' to recurse
5781 on the unmatched portion of STR. This *must* be supplied in
5782 order to perform global search-and-replace on STR; if it is
5783 not present among the ITEMs, then `regexp-substitute/global'
5784 will return after processing a single match.
5786 *** Match Structures
5788 A "match structure" is the object returned by `string-match' and
5789 `regexp-exec'. It describes which portion of a string, if any, matched
5790 the given regular expression. Match structures include: a reference to
5791 the string that was checked for matches; the starting and ending
5792 positions of the regexp match; and, if the regexp included any
5793 parenthesized subexpressions, the starting and ending positions of each
5796 In each of the regexp match functions described below, the `match'
5797 argument must be a match structure returned by a previous call to
5798 `string-match' or `regexp-exec'. Most of these functions return some
5799 information about the original target string that was matched against a
5800 regular expression; we will call that string TARGET for easy reference.
5802 **** Function: regexp-match? OBJ
5803 Return `#t' if OBJ is a match structure returned by a previous
5804 call to `regexp-exec', or `#f' otherwise.
5806 **** Function: match:substring MATCH [N]
5807 Return the portion of TARGET matched by subexpression number N.
5808 Submatch 0 (the default) represents the entire regexp match. If
5809 the regular expression as a whole matched, but the subexpression
5810 number N did not match, return `#f'.
5812 **** Function: match:start MATCH [N]
5813 Return the starting position of submatch number N.
5815 **** Function: match:end MATCH [N]
5816 Return the ending position of submatch number N.
5818 **** Function: match:prefix MATCH
5819 Return the unmatched portion of TARGET preceding the regexp match.
5821 **** Function: match:suffix MATCH
5822 Return the unmatched portion of TARGET following the regexp match.
5824 **** Function: match:count MATCH
5825 Return the number of parenthesized subexpressions from MATCH.
5826 Note that the entire regular expression match itself counts as a
5827 subexpression, and failed submatches are included in the count.
5829 **** Function: match:string MATCH
5830 Return the original TARGET string.
5832 *** Backslash Escapes
5834 Sometimes you will want a regexp to match characters like `*' or `$'
5835 exactly. For example, to check whether a particular string represents
5836 a menu entry from an Info node, it would be useful to match it against
5837 a regexp like `^* [^:]*::'. However, this won't work; because the
5838 asterisk is a metacharacter, it won't match the `*' at the beginning of
5839 the string. In this case, we want to make the first asterisk un-magic.
5841 You can do this by preceding the metacharacter with a backslash
5842 character `\'. (This is also called "quoting" the metacharacter, and
5843 is known as a "backslash escape".) When Guile sees a backslash in a
5844 regular expression, it considers the following glyph to be an ordinary
5845 character, no matter what special meaning it would ordinarily have.
5846 Therefore, we can make the above example work by changing the regexp to
5847 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5848 to match only a single asterisk in the target string.
5850 Since the backslash is itself a metacharacter, you may force a
5851 regexp to match a backslash in the target string by preceding the
5852 backslash with itself. For example, to find variable references in a
5853 TeX program, you might want to find occurrences of the string `\let\'
5854 followed by any number of alphabetic characters. The regular expression
5855 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5856 each match a single backslash in the target string.
5858 **** Function: regexp-quote STR
5859 Quote each special character found in STR with a backslash, and
5860 return the resulting string.
5862 *Very important:* Using backslash escapes in Guile source code (as
5863 in Emacs Lisp or C) can be tricky, because the backslash character has
5864 special meaning for the Guile reader. For example, if Guile encounters
5865 the character sequence `\n' in the middle of a string while processing
5866 Scheme code, it replaces those characters with a newline character.
5867 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5868 Several of these "escape sequences" are processed by the Guile reader
5869 before your code is executed. Unrecognized escape sequences are
5870 ignored: if the characters `\*' appear in a string, they will be
5871 translated to the single character `*'.
5873 This translation is obviously undesirable for regular expressions,
5874 since we want to be able to include backslashes in a string in order to
5875 escape regexp metacharacters. Therefore, to make sure that a backslash
5876 is preserved in a string in your Guile program, you must use *two*
5877 consecutive backslashes:
5879 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5881 The string in this example is preprocessed by the Guile reader before
5882 any code is executed. The resulting argument to `make-regexp' is the
5883 string `^\* [^:]*', which is what we really want.
5885 This also means that in order to write a regular expression that
5886 matches a single backslash character, the regular expression string in
5887 the source code must include *four* backslashes. Each consecutive pair
5888 of backslashes gets translated by the Guile reader to a single
5889 backslash, and the resulting double-backslash is interpreted by the
5890 regexp engine as matching a single backslash character. Hence:
5892 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5894 The reason for the unwieldiness of this syntax is historical. Both
5895 regular expression pattern matchers and Unix string processing systems
5896 have traditionally used backslashes with the special meanings described
5897 above. The POSIX regular expression specification and ANSI C standard
5898 both require these semantics. Attempting to abandon either convention
5899 would cause other kinds of compatibility problems, possibly more severe
5900 ones. Therefore, without extending the Scheme reader to support
5901 strings with different quoting conventions (an ungainly and confusing
5902 extension when implemented in other languages), we must adhere to this
5903 cumbersome escape syntax.
5905 * Changes to the gh_ interface
5907 * Changes to the scm_ interface
5909 * Changes to system call interfaces:
5911 ** The value returned by `raise' is now unspecified. It throws an exception
5914 *** A new procedure `sigaction' can be used to install signal handlers
5916 (sigaction signum [action] [flags])
5918 signum is the signal number, which can be specified using the value
5921 If action is omitted, sigaction returns a pair: the CAR is the current
5922 signal hander, which will be either an integer with the value SIG_DFL
5923 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5924 handles the signal, or #f if a non-Scheme procedure handles the
5925 signal. The CDR contains the current sigaction flags for the handler.
5927 If action is provided, it is installed as the new handler for signum.
5928 action can be a Scheme procedure taking one argument, or the value of
5929 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5930 whatever signal handler was installed before sigaction was first used.
5931 Flags can optionally be specified for the new handler (SA_RESTART is
5932 always used if the system provides it, so need not be specified.) The
5933 return value is a pair with information about the old handler as
5936 This interface does not provide access to the "signal blocking"
5937 facility. Maybe this is not needed, since the thread support may
5938 provide solutions to the problem of consistent access to data
5941 *** A new procedure `flush-all-ports' is equivalent to running
5942 `force-output' on every port open for output.
5944 ** Guile now provides information on how it was built, via the new
5945 global variable, %guile-build-info. This variable records the values
5946 of the standard GNU makefile directory variables as an assocation
5947 list, mapping variable names (symbols) onto directory paths (strings).
5948 For example, to find out where the Guile link libraries were
5949 installed, you can say:
5951 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5954 * Changes to the scm_ interface
5956 ** The new function scm_handle_by_message_noexit is just like the
5957 existing scm_handle_by_message function, except that it doesn't call
5958 exit to terminate the process. Instead, it prints a message and just
5959 returns #f. This might be a more appropriate catch-all handler for
5960 new dynamic roots and threads.
5963 Changes in Guile 1.1 (released Friday, May 16 1997):
5965 * Changes to the distribution.
5967 The Guile 1.0 distribution has been split up into several smaller
5969 guile-core --- the Guile interpreter itself.
5970 guile-tcltk --- the interface between the Guile interpreter and
5971 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5972 is a toolkit for building graphical user interfaces.
5973 guile-rgx-ctax --- the interface between Guile and the Rx regular
5974 expression matcher, and the translator for the Ctax
5975 programming language. These are packaged together because the
5976 Ctax translator uses Rx to parse Ctax source code.
5978 This NEWS file describes the changes made to guile-core since the 1.0
5981 We no longer distribute the documentation, since it was either out of
5982 date, or incomplete. As soon as we have current documentation, we
5987 * Changes to the stand-alone interpreter
5989 ** guile now accepts command-line arguments compatible with SCSH, Olin
5990 Shivers' Scheme Shell.
5992 In general, arguments are evaluated from left to right, but there are
5993 exceptions. The following switches stop argument processing, and
5994 stash all remaining command-line arguments as the value returned by
5995 the (command-line) function.
5996 -s SCRIPT load Scheme source code from FILE, and exit
5997 -c EXPR evalute Scheme expression EXPR, and exit
5998 -- stop scanning arguments; run interactively
6000 The switches below are processed as they are encountered.
6001 -l FILE load Scheme source code from FILE
6002 -e FUNCTION after reading script, apply FUNCTION to
6003 command line arguments
6004 -ds do -s script at this point
6005 --emacs enable Emacs protocol (experimental)
6006 -h, --help display this help and exit
6007 -v, --version display version information and exit
6008 \ read arguments from following script lines
6010 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6011 which re-implements the traditional "echo" command:
6013 #!/usr/local/bin/guile -s
6016 (map (lambda (arg) (display arg) (display " "))
6020 (main (command-line))
6022 Suppose we invoke this script as follows:
6024 ekko a speckled gecko
6026 Through the magic of Unix script processing (triggered by the `#!'
6027 token at the top of the file), /usr/local/bin/guile receives the
6028 following list of command-line arguments:
6030 ("-s" "./ekko" "a" "speckled" "gecko")
6032 Unix inserts the name of the script after the argument specified on
6033 the first line of the file (in this case, "-s"), and then follows that
6034 with the arguments given to the script. Guile loads the script, which
6035 defines the `main' function, and then applies it to the list of
6036 remaining command-line arguments, ("a" "speckled" "gecko").
6038 In Unix, the first line of a script file must take the following form:
6040 #!INTERPRETER ARGUMENT
6042 where INTERPRETER is the absolute filename of the interpreter
6043 executable, and ARGUMENT is a single command-line argument to pass to
6046 You may only pass one argument to the interpreter, and its length is
6047 limited. These restrictions can be annoying to work around, so Guile
6048 provides a general mechanism (borrowed from, and compatible with,
6049 SCSH) for circumventing them.
6051 If the ARGUMENT in a Guile script is a single backslash character,
6052 `\', Guile will open the script file, parse arguments from its second
6053 and subsequent lines, and replace the `\' with them. So, for example,
6054 here is another implementation of the `ekko' script:
6056 #!/usr/local/bin/guile \
6060 (for-each (lambda (arg) (display arg) (display " "))
6064 If the user invokes this script as follows:
6066 ekko a speckled gecko
6068 Unix expands this into
6070 /usr/local/bin/guile \ ekko a speckled gecko
6072 When Guile sees the `\' argument, it replaces it with the arguments
6073 read from the second line of the script, producing:
6075 /usr/local/bin/guile -e main -s ekko a speckled gecko
6077 This tells Guile to load the `ekko' script, and apply the function
6078 `main' to the argument list ("a" "speckled" "gecko").
6080 Here is how Guile parses the command-line arguments:
6081 - Each space character terminates an argument. This means that two
6082 spaces in a row introduce an empty-string argument.
6083 - The tab character is not permitted (unless you quote it with the
6084 backslash character, as described below), to avoid confusion.
6085 - The newline character terminates the sequence of arguments, and will
6086 also terminate a final non-empty argument. (However, a newline
6087 following a space will not introduce a final empty-string argument;
6088 it only terminates the argument list.)
6089 - The backslash character is the escape character. It escapes
6090 backslash, space, tab, and newline. The ANSI C escape sequences
6091 like \n and \t are also supported. These produce argument
6092 constituents; the two-character combination \n doesn't act like a
6093 terminating newline. The escape sequence \NNN for exactly three
6094 octal digits reads as the character whose ASCII code is NNN. As
6095 above, characters produced this way are argument constituents.
6096 Backslash followed by other characters is not allowed.
6098 * Changes to the procedure for linking libguile with your programs
6100 ** Guile now builds and installs a shared guile library, if your
6101 system support shared libraries. (It still builds a static library on
6102 all systems.) Guile automatically detects whether your system
6103 supports shared libraries. To prevent Guile from buildisg shared
6104 libraries, pass the `--disable-shared' flag to the configure script.
6106 Guile takes longer to compile when it builds shared libraries, because
6107 it must compile every file twice --- once to produce position-
6108 independent object code, and once to produce normal object code.
6110 ** The libthreads library has been merged into libguile.
6112 To link a program against Guile, you now need only link against
6113 -lguile and -lqt; -lthreads is no longer needed. If you are using
6114 autoconf to generate configuration scripts for your application, the
6115 following lines should suffice to add the appropriate libraries to
6118 ### Find quickthreads and libguile.
6119 AC_CHECK_LIB(qt, main)
6120 AC_CHECK_LIB(guile, scm_shell)
6122 * Changes to Scheme functions
6124 ** Guile Scheme's special syntax for keyword objects is now optional,
6125 and disabled by default.
6127 The syntax variation from R4RS made it difficult to port some
6128 interesting packages to Guile. The routines which accepted keyword
6129 arguments (mostly in the module system) have been modified to also
6130 accept symbols whose names begin with `:'.
6132 To change the keyword syntax, you must first import the (ice-9 debug)
6134 (use-modules (ice-9 debug))
6136 Then you can enable the keyword syntax as follows:
6137 (read-set! keywords 'prefix)
6139 To disable keyword syntax, do this:
6140 (read-set! keywords #f)
6142 ** Many more primitive functions accept shared substrings as
6143 arguments. In the past, these functions required normal, mutable
6144 strings as arguments, although they never made use of this
6147 ** The uniform array functions now operate on byte vectors. These
6148 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6149 `serial-array-map', `array-map', `array-for-each', and
6152 ** The new functions `trace' and `untrace' implement simple debugging
6153 support for Scheme functions.
6155 The `trace' function accepts any number of procedures as arguments,
6156 and tells the Guile interpreter to display each procedure's name and
6157 arguments each time the procedure is invoked. When invoked with no
6158 arguments, `trace' returns the list of procedures currently being
6161 The `untrace' function accepts any number of procedures as arguments,
6162 and tells the Guile interpreter not to trace them any more. When
6163 invoked with no arguments, `untrace' untraces all curretly traced
6166 The tracing in Guile has an advantage over most other systems: we
6167 don't create new procedure objects, but mark the procedure objects
6168 themselves. This means that anonymous and internal procedures can be
6171 ** The function `assert-repl-prompt' has been renamed to
6172 `set-repl-prompt!'. It takes one argument, PROMPT.
6173 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6174 - If PROMPT is a string, we use it as a prompt.
6175 - If PROMPT is a procedure accepting no arguments, we call it, and
6176 display the result as a prompt.
6177 - Otherwise, we display "> ".
6179 ** The new function `eval-string' reads Scheme expressions from a
6180 string and evaluates them, returning the value of the last expression
6181 in the string. If the string contains no expressions, it returns an
6184 ** The new function `thunk?' returns true iff its argument is a
6185 procedure of zero arguments.
6187 ** `defined?' is now a builtin function, instead of syntax. This
6188 means that its argument should be quoted. It returns #t iff its
6189 argument is bound in the current module.
6191 ** The new syntax `use-modules' allows you to add new modules to your
6192 environment without re-typing a complete `define-module' form. It
6193 accepts any number of module names as arguments, and imports their
6194 public bindings into the current module.
6196 ** The new function (module-defined? NAME MODULE) returns true iff
6197 NAME, a symbol, is defined in MODULE, a module object.
6199 ** The new function `builtin-bindings' creates and returns a hash
6200 table containing copies of all the root module's bindings.
6202 ** The new function `builtin-weak-bindings' does the same as
6203 `builtin-bindings', but creates a doubly-weak hash table.
6205 ** The `equal?' function now considers variable objects to be
6206 equivalent if they have the same name and the same value.
6208 ** The new function `command-line' returns the command-line arguments
6209 given to Guile, as a list of strings.
6211 When using guile as a script interpreter, `command-line' returns the
6212 script's arguments; those processed by the interpreter (like `-s' or
6213 `-c') are omitted. (In other words, you get the normal, expected
6214 behavior.) Any application that uses scm_shell to process its
6215 command-line arguments gets this behavior as well.
6217 ** The new function `load-user-init' looks for a file called `.guile'
6218 in the user's home directory, and loads it if it exists. This is
6219 mostly for use by the code generated by scm_compile_shell_switches,
6220 but we thought it might also be useful in other circumstances.
6222 ** The new function `log10' returns the base-10 logarithm of its
6225 ** Changes to I/O functions
6227 *** The functions `read', `primitive-load', `read-and-eval!', and
6228 `primitive-load-path' no longer take optional arguments controlling
6229 case insensitivity and a `#' parser.
6231 Case sensitivity is now controlled by a read option called
6232 `case-insensitive'. The user can add new `#' syntaxes with the
6233 `read-hash-extend' function (see below).
6235 *** The new function `read-hash-extend' allows the user to change the
6236 syntax of Guile Scheme in a somewhat controlled way.
6238 (read-hash-extend CHAR PROC)
6239 When parsing S-expressions, if we read a `#' character followed by
6240 the character CHAR, use PROC to parse an object from the stream.
6241 If PROC is #f, remove any parsing procedure registered for CHAR.
6243 The reader applies PROC to two arguments: CHAR and an input port.
6245 *** The new functions read-delimited and read-delimited! provide a
6246 general mechanism for doing delimited input on streams.
6248 (read-delimited DELIMS [PORT HANDLE-DELIM])
6249 Read until we encounter one of the characters in DELIMS (a string),
6250 or end-of-file. PORT is the input port to read from; it defaults to
6251 the current input port. The HANDLE-DELIM parameter determines how
6252 the terminating character is handled; it should be one of the
6255 'trim omit delimiter from result
6256 'peek leave delimiter character in input stream
6257 'concat append delimiter character to returned value
6258 'split return a pair: (RESULT . TERMINATOR)
6260 HANDLE-DELIM defaults to 'peek.
6262 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6263 A side-effecting variant of `read-delimited'.
6265 The data is written into the string BUF at the indices in the
6266 half-open interval [START, END); the default interval is the whole
6267 string: START = 0 and END = (string-length BUF). The values of
6268 START and END must specify a well-defined interval in BUF, i.e.
6269 0 <= START <= END <= (string-length BUF).
6271 It returns NBYTES, the number of bytes read. If the buffer filled
6272 up without a delimiter character being found, it returns #f. If the
6273 port is at EOF when the read starts, it returns the EOF object.
6275 If an integer is returned (i.e., the read is successfully terminated
6276 by reading a delimiter character), then the HANDLE-DELIM parameter
6277 determines how to handle the terminating character. It is described
6278 above, and defaults to 'peek.
6280 (The descriptions of these functions were borrowed from the SCSH
6281 manual, by Olin Shivers and Brian Carlstrom.)
6283 *** The `%read-delimited!' function is the primitive used to implement
6284 `read-delimited' and `read-delimited!'.
6286 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6288 This returns a pair of values: (TERMINATOR . NUM-READ).
6289 - TERMINATOR describes why the read was terminated. If it is a
6290 character or the eof object, then that is the value that terminated
6291 the read. If it is #f, the function filled the buffer without finding
6292 a delimiting character.
6293 - NUM-READ is the number of characters read into BUF.
6295 If the read is successfully terminated by reading a delimiter
6296 character, then the gobble? parameter determines what to do with the
6297 terminating character. If true, the character is removed from the
6298 input stream; if false, the character is left in the input stream
6299 where a subsequent read operation will retrieve it. In either case,
6300 the character is also the first value returned by the procedure call.
6302 (The descriptions of this function was borrowed from the SCSH manual,
6303 by Olin Shivers and Brian Carlstrom.)
6305 *** The `read-line' and `read-line!' functions have changed; they now
6306 trim the terminator by default; previously they appended it to the
6307 returned string. For the old behavior, use (read-line PORT 'concat).
6309 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6310 take new optional START and END arguments, specifying the region of
6311 the array to read and write.
6313 *** The `ungetc-char-ready?' function has been removed. We feel it's
6314 inappropriate for an interface to expose implementation details this
6317 ** Changes to the Unix library and system call interface
6319 *** The new fcntl function provides access to the Unix `fcntl' system
6322 (fcntl PORT COMMAND VALUE)
6323 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6324 Values for COMMAND are:
6326 F_DUPFD duplicate a file descriptor
6327 F_GETFD read the descriptor's close-on-exec flag
6328 F_SETFD set the descriptor's close-on-exec flag to VALUE
6329 F_GETFL read the descriptor's flags, as set on open
6330 F_SETFL set the descriptor's flags, as set on open to VALUE
6331 F_GETOWN return the process ID of a socket's owner, for SIGIO
6332 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6333 FD_CLOEXEC not sure what this is
6335 For details, see the documentation for the fcntl system call.
6337 *** The arguments to `select' have changed, for compatibility with
6338 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6339 expected behavior. The MILLISECONDS parameter has been changed to
6340 MICROSECONDS, to more closely resemble the underlying system call.
6341 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6342 corresponding return set will be the same.
6344 *** The arguments to the `mknod' system call have changed. They are
6347 (mknod PATH TYPE PERMS DEV)
6348 Create a new file (`node') in the file system. PATH is the name of
6349 the file to create. TYPE is the kind of file to create; it should
6350 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6351 permission bits to give the newly created file. If TYPE is
6352 'block-special or 'char-special, DEV specifies which device the
6353 special file refers to; its interpretation depends on the kind of
6354 special file being created.
6356 *** The `fork' function has been renamed to `primitive-fork', to avoid
6357 clashing with various SCSH forks.
6359 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6360 and `recvfrom!'. They no longer accept a size for a second argument;
6361 you must pass a string to hold the received value. They no longer
6362 return the buffer. Instead, `recv' returns the length of the message
6363 received, and `recvfrom' returns a pair containing the packet's length
6364 and originating address.
6366 *** The file descriptor datatype has been removed, as have the
6367 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6368 We plan to replace these functions with a SCSH-compatible interface.
6370 *** The `create' function has been removed; it's just a special case
6373 *** There are new functions to break down process termination status
6374 values. In the descriptions below, STATUS is a value returned by
6377 (status:exit-val STATUS)
6378 If the child process exited normally, this function returns the exit
6379 code for the child process (i.e., the value passed to exit, or
6380 returned from main). If the child process did not exit normally,
6381 this function returns #f.
6383 (status:stop-sig STATUS)
6384 If the child process was suspended by a signal, this function
6385 returns the signal that suspended the child. Otherwise, it returns
6388 (status:term-sig STATUS)
6389 If the child process terminated abnormally, this function returns
6390 the signal that terminated the child. Otherwise, this function
6393 POSIX promises that exactly one of these functions will return true on
6394 a valid STATUS value.
6396 These functions are compatible with SCSH.
6398 *** There are new accessors and setters for the broken-out time vectors
6399 returned by `localtime', `gmtime', and that ilk. They are:
6401 Component Accessor Setter
6402 ========================= ============ ============
6403 seconds tm:sec set-tm:sec
6404 minutes tm:min set-tm:min
6405 hours tm:hour set-tm:hour
6406 day of the month tm:mday set-tm:mday
6407 month tm:mon set-tm:mon
6408 year tm:year set-tm:year
6409 day of the week tm:wday set-tm:wday
6410 day in the year tm:yday set-tm:yday
6411 daylight saving time tm:isdst set-tm:isdst
6412 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6413 name of time zone tm:zone set-tm:zone
6415 *** There are new accessors for the vectors returned by `uname',
6416 describing the host system:
6419 ============================================== ================
6420 name of the operating system implementation utsname:sysname
6421 network name of this machine utsname:nodename
6422 release level of the operating system utsname:release
6423 version level of the operating system utsname:version
6424 machine hardware platform utsname:machine
6426 *** There are new accessors for the vectors returned by `getpw',
6427 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6428 system's user database:
6431 ====================== =================
6432 user name passwd:name
6433 user password passwd:passwd
6436 real name passwd:gecos
6437 home directory passwd:dir
6438 shell program passwd:shell
6440 *** There are new accessors for the vectors returned by `getgr',
6441 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6442 system's group database:
6445 ======================= ============
6446 group name group:name
6447 group password group:passwd
6449 group members group:mem
6451 *** There are new accessors for the vectors returned by `gethost',
6452 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6456 ========================= ===============
6457 official name of host hostent:name
6458 alias list hostent:aliases
6459 host address type hostent:addrtype
6460 length of address hostent:length
6461 list of addresses hostent:addr-list
6463 *** There are new accessors for the vectors returned by `getnet',
6464 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6468 ========================= ===============
6469 official name of net netent:name
6470 alias list netent:aliases
6471 net number type netent:addrtype
6472 net number netent:net
6474 *** There are new accessors for the vectors returned by `getproto',
6475 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6479 ========================= ===============
6480 official protocol name protoent:name
6481 alias list protoent:aliases
6482 protocol number protoent:proto
6484 *** There are new accessors for the vectors returned by `getserv',
6485 `getservbyname', `getservbyport', and `getservent', describing
6489 ========================= ===============
6490 official service name servent:name
6491 alias list servent:aliases
6492 port number servent:port
6493 protocol to use servent:proto
6495 *** There are new accessors for the sockaddr structures returned by
6496 `accept', `getsockname', `getpeername', `recvfrom!':
6499 ======================================== ===============
6500 address format (`family') sockaddr:fam
6501 path, for file domain addresses sockaddr:path
6502 address, for internet domain addresses sockaddr:addr
6503 TCP or UDP port, for internet sockaddr:port
6505 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6506 `getprotoent', and `getservent' functions now return #f at the end of
6507 the user database. (They used to throw an exception.)
6509 Note that calling MUMBLEent function is equivalent to calling the
6510 corresponding MUMBLE function with no arguments.
6512 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6513 `setprotoent', and `setservent' routines now take no arguments.
6515 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6516 provide more useful information when they throw an exception.
6518 *** The `lnaof' function has been renamed to `inet-lnaof'.
6520 *** Guile now claims to have the `current-time' feature.
6522 *** The `mktime' function now takes an optional second argument ZONE,
6523 giving the time zone to use for the conversion. ZONE should be a
6524 string, in the same format as expected for the "TZ" environment variable.
6526 *** The `strptime' function now returns a pair (TIME . COUNT), where
6527 TIME is the parsed time as a vector, and COUNT is the number of
6528 characters from the string left unparsed. This function used to
6529 return the remaining characters as a string.
6531 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6532 The return value is now (SECONDS . MICROSECONDS); the fractional
6533 component is no longer expressed in "ticks".
6535 *** The `ticks/sec' constant has been removed, in light of the above change.
6537 * Changes to the gh_ interface
6539 ** gh_eval_str() now returns an SCM object which is the result of the
6542 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6545 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6546 and returns the array
6548 ** gh_scm2str0() is gone: there is no need to distinguish
6549 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6550 the user to interpret the data both ways.
6552 * Changes to the scm_ interface
6554 ** The new function scm_symbol_value0 provides an easy way to get a
6555 symbol's value from C code:
6557 SCM scm_symbol_value0 (char *NAME)
6558 Return the value of the symbol named by the null-terminated string
6559 NAME in the current module. If the symbol named NAME is unbound in
6560 the current module, return SCM_UNDEFINED.
6562 ** The new function scm_sysintern0 creates new top-level variables,
6563 without assigning them a value.
6565 SCM scm_sysintern0 (char *NAME)
6566 Create a new Scheme top-level variable named NAME. NAME is a
6567 null-terminated string. Return the variable's value cell.
6569 ** The function scm_internal_catch is the guts of catch. It handles
6570 all the mechanics of setting up a catch target, invoking the catch
6571 body, and perhaps invoking the handler if the body does a throw.
6573 The function is designed to be usable from C code, but is general
6574 enough to implement all the semantics Guile Scheme expects from throw.
6576 TAG is the catch tag. Typically, this is a symbol, but this function
6577 doesn't actually care about that.
6579 BODY is a pointer to a C function which runs the body of the catch;
6580 this is the code you can throw from. We call it like this:
6581 BODY (BODY_DATA, JMPBUF)
6583 BODY_DATA is just the BODY_DATA argument we received; we pass it
6584 through to BODY as its first argument. The caller can make
6585 BODY_DATA point to anything useful that BODY might need.
6586 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6587 which we have just created and initialized.
6589 HANDLER is a pointer to a C function to deal with a throw to TAG,
6590 should one occur. We call it like this:
6591 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6593 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6594 same idea as BODY_DATA above.
6595 THROWN_TAG is the tag that the user threw to; usually this is
6596 TAG, but it could be something else if TAG was #t (i.e., a
6597 catch-all), or the user threw to a jmpbuf.
6598 THROW_ARGS is the list of arguments the user passed to the THROW
6601 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6602 is just a pointer we pass through to HANDLER. We don't actually
6603 use either of those pointers otherwise ourselves. The idea is
6604 that, if our caller wants to communicate something to BODY or
6605 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6606 HANDLER can then use. Think of it as a way to make BODY and
6607 HANDLER closures, not just functions; MUMBLE_DATA points to the
6610 Of course, it's up to the caller to make sure that any data a
6611 MUMBLE_DATA needs is protected from GC. A common way to do this is
6612 to make MUMBLE_DATA a pointer to data stored in an automatic
6613 structure variable; since the collector must scan the stack for
6614 references anyway, this assures that any references in MUMBLE_DATA
6617 ** The new function scm_internal_lazy_catch is exactly like
6618 scm_internal_catch, except:
6620 - It does not unwind the stack (this is the major difference).
6621 - If handler returns, its value is returned from the throw.
6622 - BODY always receives #f as its JMPBUF argument (since there's no
6623 jmpbuf associated with a lazy catch, because we don't unwind the
6626 ** scm_body_thunk is a new body function you can pass to
6627 scm_internal_catch if you want the body to be like Scheme's `catch'
6628 --- a thunk, or a function of one argument if the tag is #f.
6630 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6631 contains the Scheme procedure to invoke as the body, and the tag
6632 we're catching. If the tag is #f, then we pass JMPBUF (created by
6633 scm_internal_catch) to the body procedure; otherwise, the body gets
6636 ** scm_handle_by_proc is a new handler function you can pass to
6637 scm_internal_catch if you want the handler to act like Scheme's catch
6638 --- call a procedure with the tag and the throw arguments.
6640 If the user does a throw to this catch, this function runs a handler
6641 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6642 variable holding the Scheme procedure object to invoke. It ought to
6643 be a pointer to an automatic variable (i.e., one living on the stack),
6644 or the procedure object should be otherwise protected from GC.
6646 ** scm_handle_by_message is a new handler function to use with
6647 `scm_internal_catch' if you want Guile to print a message and die.
6648 It's useful for dealing with throws to uncaught keys at the top level.
6650 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6651 message header to print; if zero, we use "guile" instead. That
6652 text is followed by a colon, then the message described by ARGS.
6654 ** The return type of scm_boot_guile is now void; the function does
6655 not return a value, and indeed, never returns at all.
6657 ** The new function scm_shell makes it easy for user applications to
6658 process command-line arguments in a way that is compatible with the
6659 stand-alone guile interpreter (which is in turn compatible with SCSH,
6662 To use the scm_shell function, first initialize any guile modules
6663 linked into your application, and then call scm_shell with the values
6664 of ARGC and ARGV your `main' function received. scm_shell will add
6665 any SCSH-style meta-arguments from the top of the script file to the
6666 argument vector, and then process the command-line arguments. This
6667 generally means loading a script file or starting up an interactive
6668 command interpreter. For details, see "Changes to the stand-alone
6671 ** The new functions scm_get_meta_args and scm_count_argv help you
6672 implement the SCSH-style meta-argument, `\'.
6674 char **scm_get_meta_args (int ARGC, char **ARGV)
6675 If the second element of ARGV is a string consisting of a single
6676 backslash character (i.e. "\\" in Scheme notation), open the file
6677 named by the following argument, parse arguments from it, and return
6678 the spliced command line. The returned array is terminated by a
6681 For details of argument parsing, see above, under "guile now accepts
6682 command-line arguments compatible with SCSH..."
6684 int scm_count_argv (char **ARGV)
6685 Count the arguments in ARGV, assuming it is terminated by a null
6688 For an example of how these functions might be used, see the source
6689 code for the function scm_shell in libguile/script.c.
6691 You will usually want to use scm_shell instead of calling this
6694 ** The new function scm_compile_shell_switches turns an array of
6695 command-line arguments into Scheme code to carry out the actions they
6696 describe. Given ARGC and ARGV, it returns a Scheme expression to
6697 evaluate, and calls scm_set_program_arguments to make any remaining
6698 command-line arguments available to the Scheme code. For example,
6699 given the following arguments:
6701 -e main -s ekko a speckled gecko
6703 scm_set_program_arguments will return the following expression:
6705 (begin (load "ekko") (main (command-line)) (quit))
6707 You will usually want to use scm_shell instead of calling this
6710 ** The function scm_shell_usage prints a usage message appropriate for
6711 an interpreter that uses scm_compile_shell_switches to handle its
6712 command-line arguments.
6714 void scm_shell_usage (int FATAL, char *MESSAGE)
6715 Print a usage message to the standard error output. If MESSAGE is
6716 non-zero, write it before the usage message, followed by a newline.
6717 If FATAL is non-zero, exit the process, using FATAL as the
6718 termination status. (If you want to be compatible with Guile,
6719 always use 1 as the exit status when terminating due to command-line
6722 You will usually want to use scm_shell instead of calling this
6725 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6726 expressions. It used to return SCM_EOL. Earth-shattering.
6728 ** The macros for declaring scheme objects in C code have been
6729 rearranged slightly. They are now:
6731 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6732 Declare a static SCM variable named C_NAME, and initialize it to
6733 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6734 be a C identifier, and SCHEME_NAME should be a C string.
6736 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6737 Just like SCM_SYMBOL, but make C_NAME globally visible.
6739 SCM_VCELL (C_NAME, SCHEME_NAME)
6740 Create a global variable at the Scheme level named SCHEME_NAME.
6741 Declare a static SCM variable named C_NAME, and initialize it to
6742 point to the Scheme variable's value cell.
6744 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6745 Just like SCM_VCELL, but make C_NAME globally visible.
6747 The `guile-snarf' script writes initialization code for these macros
6748 to its standard output, given C source code as input.
6750 The SCM_GLOBAL macro is gone.
6752 ** The scm_read_line and scm_read_line_x functions have been replaced
6753 by Scheme code based on the %read-delimited! procedure (known to C
6754 code as scm_read_delimited_x). See its description above for more
6757 ** The function scm_sys_open has been renamed to scm_open. It now
6758 returns a port instead of an FD object.
6760 * The dynamic linking support has changed. For more information, see
6761 libguile/DYNAMIC-LINKING.
6766 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6769 * Changes to the 'guile' program:
6771 ** Guile now loads some new files when it starts up. Guile first
6772 searches the load path for init.scm, and loads it if found. Then, if
6773 Guile is not being used to execute a script, and the user's home
6774 directory contains a file named `.guile', Guile loads that.
6776 ** You can now use Guile as a shell script interpreter.
6778 To paraphrase the SCSH manual:
6780 When Unix tries to execute an executable file whose first two
6781 characters are the `#!', it treats the file not as machine code to
6782 be directly executed by the native processor, but as source code
6783 to be executed by some interpreter. The interpreter to use is
6784 specified immediately after the #! sequence on the first line of
6785 the source file. The kernel reads in the name of the interpreter,
6786 and executes that instead. It passes the interpreter the source
6787 filename as its first argument, with the original arguments
6788 following. Consult the Unix man page for the `exec' system call
6789 for more information.
6791 Now you can use Guile as an interpreter, using a mechanism which is a
6792 compatible subset of that provided by SCSH.
6794 Guile now recognizes a '-s' command line switch, whose argument is the
6795 name of a file of Scheme code to load. It also treats the two
6796 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6797 to make a file of Scheme code directly executable by Unix, insert the
6798 following two lines at the top of the file:
6800 #!/usr/local/bin/guile -s
6803 Guile treats the argument of the `-s' command-line switch as the name
6804 of a file of Scheme code to load, and treats the sequence `#!' as the
6805 start of a block comment, terminated by `!#'.
6807 For example, here's a version of 'echo' written in Scheme:
6809 #!/usr/local/bin/guile -s
6811 (let loop ((args (cdr (program-arguments))))
6814 (display (car args))
6815 (if (pair? (cdr args))
6817 (loop (cdr args)))))
6820 Why does `#!' start a block comment terminated by `!#', instead of the
6821 end of the line? That is the notation SCSH uses, and although we
6822 don't yet support the other SCSH features that motivate that choice,
6823 we would like to be backward-compatible with any existing Guile
6824 scripts once we do. Furthermore, if the path to Guile on your system
6825 is too long for your kernel, you can start the script with this
6829 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6832 Note that some very old Unix systems don't support the `#!' syntax.
6835 ** You can now run Guile without installing it.
6837 Previous versions of the interactive Guile interpreter (`guile')
6838 couldn't start up unless Guile's Scheme library had been installed;
6839 they used the value of the environment variable `SCHEME_LOAD_PATH'
6840 later on in the startup process, but not to find the startup code
6841 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6844 To run Guile without installing it, build it in the normal way, and
6845 then set the environment variable `SCHEME_LOAD_PATH' to a
6846 colon-separated list of directories, including the top-level directory
6847 of the Guile sources. For example, if you unpacked Guile so that the
6848 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6851 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6854 ** Guile's read-eval-print loop no longer prints #<unspecified>
6855 results. If the user wants to see this, she can evaluate the
6856 expression (assert-repl-print-unspecified #t), perhaps in her startup
6859 ** Guile no longer shows backtraces by default when an error occurs;
6860 however, it does display a message saying how to get one, and how to
6861 request that they be displayed by default. After an error, evaluate
6863 to see a backtrace, and
6864 (debug-enable 'backtrace)
6865 to see them by default.
6869 * Changes to Guile Scheme:
6871 ** Guile now distinguishes between #f and the empty list.
6873 This is for compatibility with the IEEE standard, the (possibly)
6874 upcoming Revised^5 Report on Scheme, and many extant Scheme
6877 Guile used to have #f and '() denote the same object, to make Scheme's
6878 type system more compatible with Emacs Lisp's. However, the change
6879 caused too much trouble for Scheme programmers, and we found another
6880 way to reconcile Emacs Lisp with Scheme that didn't require this.
6883 ** Guile's delq, delv, delete functions, and their destructive
6884 counterparts, delq!, delv!, and delete!, now remove all matching
6885 elements from the list, not just the first. This matches the behavior
6886 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6887 functions which inspired them.
6889 I recognize that this change may break code in subtle ways, but it
6890 seems best to make the change before the FSF's first Guile release,
6894 ** The compiled-library-path function has been deleted from libguile.
6896 ** The facilities for loading Scheme source files have changed.
6898 *** The variable %load-path now tells Guile which directories to search
6899 for Scheme code. Its value is a list of strings, each of which names
6902 *** The variable %load-extensions now tells Guile which extensions to
6903 try appending to a filename when searching the load path. Its value
6904 is a list of strings. Its default value is ("" ".scm").
6906 *** (%search-load-path FILENAME) searches the directories listed in the
6907 value of the %load-path variable for a Scheme file named FILENAME,
6908 with all the extensions listed in %load-extensions. If it finds a
6909 match, then it returns its full filename. If FILENAME is absolute, it
6910 returns it unchanged. Otherwise, it returns #f.
6912 %search-load-path will not return matches that refer to directories.
6914 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6915 uses %seach-load-path to find a file named FILENAME, and loads it if
6916 it finds it. If it can't read FILENAME for any reason, it throws an
6919 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6922 *** load uses the same searching semantics as primitive-load.
6924 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6925 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6926 path, and load-module-with-path have been deleted. The functions
6927 above should serve their purposes.
6929 *** If the value of the variable %load-hook is a procedure,
6930 `primitive-load' applies its value to the name of the file being
6931 loaded (without the load path directory name prepended). If its value
6932 is #f, it is ignored. Otherwise, an error occurs.
6934 This is mostly useful for printing load notification messages.
6937 ** The function `eval!' is no longer accessible from the scheme level.
6938 We can't allow operations which introduce glocs into the scheme level,
6939 because Guile's type system can't handle these as data. Use `eval' or
6940 `read-and-eval!' (see below) as replacement.
6942 ** The new function read-and-eval! reads an expression from PORT,
6943 evaluates it, and returns the result. This is more efficient than
6944 simply calling `read' and `eval', since it is not necessary to make a
6945 copy of the expression for the evaluator to munge.
6947 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6948 for the `read' function.
6951 ** The function `int?' has been removed; its definition was identical
6952 to that of `integer?'.
6954 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6955 use the R4RS names for these functions.
6957 ** The function object-properties no longer returns the hash handle;
6958 it simply returns the object's property list.
6960 ** Many functions have been changed to throw errors, instead of
6961 returning #f on failure. The point of providing exception handling in
6962 the language is to simplify the logic of user code, but this is less
6963 useful if Guile's primitives don't throw exceptions.
6965 ** The function `fileno' has been renamed from `%fileno'.
6967 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6970 * Changes to Guile's C interface:
6972 ** The library's initialization procedure has been simplified.
6973 scm_boot_guile now has the prototype:
6975 void scm_boot_guile (int ARGC,
6977 void (*main_func) (),
6980 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6981 MAIN_FUNC should do all the work of the program (initializing other
6982 packages, reading user input, etc.) before returning. When MAIN_FUNC
6983 returns, call exit (0); this function never returns. If you want some
6984 other exit value, MAIN_FUNC may call exit itself.
6986 scm_boot_guile arranges for program-arguments to return the strings
6987 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6988 scm_set_program_arguments with the final list, so Scheme code will
6989 know which arguments have been processed.
6991 scm_boot_guile establishes a catch-all catch handler which prints an
6992 error message and exits the process. This means that Guile exits in a
6993 coherent way when system errors occur and the user isn't prepared to
6994 handle it. If the user doesn't like this behavior, they can establish
6995 their own universal catcher in MAIN_FUNC to shadow this one.
6997 Why must the caller do all the real work from MAIN_FUNC? The garbage
6998 collector assumes that all local variables of type SCM will be above
6999 scm_boot_guile's stack frame on the stack. If you try to manipulate
7000 SCM values after this function returns, it's the luck of the draw
7001 whether the GC will be able to find the objects you allocate. So,
7002 scm_boot_guile function exits, rather than returning, to discourage
7003 people from making that mistake.
7005 The IN, OUT, and ERR arguments were removed; there are other
7006 convenient ways to override these when desired.
7008 The RESULT argument was deleted; this function should never return.
7010 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7014 ** Guile's header files should no longer conflict with your system's
7017 In order to compile code which #included <libguile.h>, previous
7018 versions of Guile required you to add a directory containing all the
7019 Guile header files to your #include path. This was a problem, since
7020 Guile's header files have names which conflict with many systems'
7023 Now only <libguile.h> need appear in your #include path; you must
7024 refer to all Guile's other header files as <libguile/mumble.h>.
7025 Guile's installation procedure puts libguile.h in $(includedir), and
7026 the rest in $(includedir)/libguile.
7029 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7030 have been added to the Guile library.
7032 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7033 OBJ will not be freed, even if all other references are dropped,
7034 until someone does scm_unprotect_object (OBJ). Both functions
7037 Note that calls to scm_protect_object do not nest. You can call
7038 scm_protect_object any number of times on a given object, and the
7039 next call to scm_unprotect_object will unprotect it completely.
7041 Basically, scm_protect_object and scm_unprotect_object just
7042 maintain a list of references to things. Since the GC knows about
7043 this list, all objects it mentions stay alive. scm_protect_object
7044 adds its argument to the list; scm_unprotect_object remove its
7045 argument from the list.
7048 ** scm_eval_0str now returns the value of the last expression
7051 ** The new function scm_read_0str reads an s-expression from a
7052 null-terminated string, and returns it.
7054 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7055 to a Scheme port object.
7057 ** The new function `scm_set_program_arguments' allows C code to set
7058 the value returned by the Scheme `program-arguments' function.
7063 * Guile no longer includes sophisticated Tcl/Tk support.
7065 The old Tcl/Tk support was unsatisfying to us, because it required the
7066 user to link against the Tcl library, as well as Tk and Guile. The
7067 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7068 referring to widgets by names, rather than exporting widgets to Scheme
7069 code as a special datatype.
7071 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7072 maintainers described some very interesting changes in progress to the
7073 Tcl/Tk internals, which would facilitate clean interfaces between lone
7074 Tk and other interpreters --- even for garbage-collected languages
7075 like Scheme. They expected the new Tk to be publicly available in the
7078 Since it seems that Guile might soon have a new, cleaner interface to
7079 lone Tk, and that the old Guile/Tk glue code would probably need to be
7080 completely rewritten, we (Jim Blandy and Richard Stallman) have
7081 decided not to support the old code. We'll spend the time instead on
7082 a good interface to the newer Tk, as soon as it is available.
7084 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7087 Copyright information:
7089 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7091 Permission is granted to anyone to make or distribute verbatim copies
7092 of this document as received, in any medium, provided that the
7093 copyright notice and this permission notice are preserved,
7094 thus giving the recipient permission to redistribute in turn.
7096 Permission is granted to distribute modified versions
7097 of this document, or of portions of it,
7098 under the above conditions, provided also that they
7099 carry prominent notices stating who last changed them.
7104 paragraph-separate: "[
\f]*$"