1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 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 since the 1.6.x series:
17 * Changes to the distribution
19 ** Guile is now licensed with the GNU Lesser General Public License.
21 ** The manual is now licensed with the GNU Free Documentation License.
23 ** Guile now requires GNU MP (http://swox.com/gmp).
25 Guile now uses the GNU MP library for arbitrary precision arithmetic.
27 ** Guile now has separate private and public configuration headers.
29 That is, things like HAVE_STRING_H no longer leak from Guile's
32 ** Guile now provides and uses an "effective" version number.
34 Guile now provides scm_effective_version and effective-version
35 functions which return the "effective" version number. This is just
36 the normal full version string without the final micro-version number,
37 so the current effective-version is "1.7". The effective version
38 should remain unchanged during a stable series, and should be used for
39 items like the versioned share directory name
40 i.e. /usr/share/guile/1.7.
42 Providing an unchanging version number during a stable release for
43 things like the versioned share directory can be particularly
44 important for Guile "add-on" packages, since it provides a directory
45 that they can install to that won't be changed out from under them
46 with each micro release during a stable series.
48 ** Thread implementation has changed.
50 When you configure "--with-threads=null", you will get the usual
51 threading API (call-with-new-thread, make-mutex, etc), but you can't
52 actually create new threads. Also, "--with-threads=no" is now
53 equivalent to "--with-threads=null". This means that the thread API
54 is always present, although you might not be able to create new
57 When you configure "--with-threads=pthreads" or "--with-threads=yes",
58 you will get threads that are implemented with the portable POSIX
59 threads. These threads can run concurrently (unlike the previous
60 "coop" thread implementation), but need to cooperate for things like
61 the GC. See the manual for details. [XXX - write this.]
63 The default is "pthreads", unless your platform doesn't have pthreads,
64 in which case "null" threads are used.
66 ** New module (ice-9 serialize):
68 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when
69 you don't trust the thread safety of most of your program, but
70 where you have some section(s) of code which you consider can run
71 in parallel to other sections.
73 ### move rest to manual
75 They "flag" (with dynamic extent) sections of code to be of
76 "serial" or "parallel" nature and have the single effect of
77 preventing a serial section from being run in parallel with any
78 serial section (including itself).
80 Both serialize and parallelize can be nested. If so, the
81 inner-most construct is in effect.
83 NOTE 1: A serial section can run in parallel with a parallel
86 NOTE 2: If a serial section S is "interrupted" by a parallel
87 section P in the following manner: S = S1 P S2, S2 is not
88 guaranteed to be resumed by the same thread that previously
91 WARNING: Spawning new threads within a serial section have
92 undefined effects. It is OK, though, to spawn threads in unflagged
93 sections of code where neither serialize or parallelize is in
96 A typical usage is when Guile is used as scripting language in some
97 application doing heavy computations. If each thread is
98 encapsulated with a serialize form, you can then put a parallelize
99 form around the code performing the heavy computations (typically a
100 C code primitive), enabling the computations to run in parallel
101 while the scripting code runs single-threadedly.
103 ** New module (srfi srfi-26)
105 This is an implementation of SRFI-26.
107 ** New module (srfi srfi-31)
109 This is an implementation of SRFI-31 which provides a special form
110 `rec' for recursive evaluation.
112 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
113 been merged with the core, making their functionality always
116 The modules are still available, tho, and you could use them together
117 with a renaming import, for example.
119 ** Guile no longer includes its own version of libltdl.
121 The official version is good enough now.
123 ** The --enable-htmldoc option has been removed from 'configure'.
125 Support for translating the documentation into HTML is now always
126 provided. Use 'make html'.
128 * Changes to the stand-alone interpreter
130 ** New command line option `-L'.
132 This option adds a directory to the front of the load path.
134 ** New command line option `--no-debug'.
136 Specifying `--no-debug' on the command line will keep the debugging
137 evaluator turned off, even for interactive sessions.
139 ** User-init file ~/.guile is now loaded with the debugging evaluator.
141 Previously, the normal evaluator would have been used. Using the
142 debugging evaluator gives better error messages.
144 ** The '-e' option now 'read's its argument.
146 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
147 be used with '-e'. For example, you can now write a script like
150 exec guile -e '(@ (demo) main)' -s "$0" "$@"
153 (define-module (demo)
157 (format #t "Demo: ~a~%" args))
160 * Changes to Scheme functions and syntax
162 ** The default load path no longer includes "." at the end.
164 Automatically loading modules from the current directory should not
165 happen by default. If you want to allow it in a more controlled
166 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
169 ** The uniform vector and array support has been overhauled.
171 It now complies with SRFI-4 and the weird prototype based uniform
172 array creation has been deprecated. See the manual for more details.
174 Some non-compatible changes have been made:
175 - characters can no longer be stored into byte arrays.
176 - strings and bit vectors are no longer considered to be uniform vectors.
178 There is the new notion of 'generalized vectors' and corresponding
179 procedures like 'generalized-vector-ref'. Generalized vectors include
180 strings bitvectors, ordinary vectors, and uniform numeric vectors.
182 Arrays use generalized vectors their storage, so that you still have
183 arrays of characters, bits, etc.
185 ** There is now support for copy-on-write substrings, mutation-sharing
186 substrings and read-only strings.
188 Three new procedures are related to this: substring/shared,
189 substring/copy, and substring/read-only. See the manual for more
192 ** Backtraces will now highlight the value that caused the error.
194 By default, these values are enclosed in "{...}", such as in this
203 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
204 <unnamed port>:1:1: Wrong type (expecting pair): a
205 ABORT: (wrong-type-arg)
207 The prefix and suffix used for highlighting can be set via the two new
208 printer options 'highlight-prefix' and 'highlight-suffix'. For
209 example, putting this into ~/.guile will output the bad value in bold
212 (print-set! highlight-prefix "\x1b[1m")
213 (print-set! highlight-suffix "\x1b[22m")
216 ** 'gettext' support for internationalization has been added.
218 See the manual for details.
220 ** New syntax '@' and '@@':
222 You can now directly refer to variables exported from a module by
225 (@ MODULE-NAME VARIABLE-NAME)
227 For example (@ (ice-9 pretty-print) pretty-print) will directly access
228 the pretty-print variable exported from the (ice-9 pretty-print)
229 module. You don't need to 'use' that module first. You can also use
230 '@' as a target of 'set!', as in (set! (@ mod var) val).
232 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
233 but it can also access variables that have not been exported. It is
234 intended only for kluges and temporary fixes and for debugging, not
237 ** Keyword syntax has been made more disciplined.
239 Previously, the name of a keyword was read as a 'token' but printed as
240 a symbol. Now, it is read as a general Scheme datum which must be a
251 ERROR: In expression (a b c):
255 ERROR: Unbound variable: foo
260 ERROR: Wrong type (expecting symbol): 12
264 ERROR: Wrong type (expecting symbol): (a b c)
268 ** 'while' now provides 'break' and 'continue'
270 break and continue were previously bound in a while loop, but not
271 documented, and continue didn't quite work properly. The undocumented
272 parameter to break which gave a return value for the while has been
275 ** 'call-with-current-continuation' is now also available under the name
278 ** The module system now checks for duplicate bindings.
280 The module system now can check for name conflicts among imported
283 The behavior can be controlled by specifying one or more 'duplicates'
284 handlers. For example, to make Guile return an error for every name
292 The new default behavior of the module system when a name collision
293 has been detected is to
295 1. Give priority to bindings marked as a replacement.
296 2. Issue a warning (different warning if overriding core binding).
297 3. Give priority to the last encountered binding (this corresponds to
300 If you want the old behavior back without replacements or warnings you
303 (default-duplicate-binding-handler 'last)
305 to your .guile init file.
307 ### move rest to manual
309 The syntax for the :duplicates option is:
311 :duplicates HANDLER-NAME | (HANDLER1-NAME HANDLER2-NAME ...)
313 Specifying multiple handlers is useful since some handlers (such as
314 replace) can defer conflict resolution to others. Each handler is
315 tried until a binding is selected.
317 Currently available duplicates handlers are:
319 check report an error for bindings with a common name
320 warn issue a warning for bindings with a common name
321 replace replace bindings which have an imported replacement
322 warn-override-core issue a warning for imports which override core bindings
323 and accept the override
324 first select the first encountered binding (override)
325 last select the last encountered binding (override)
327 These two are provided by the (oop goops) module:
329 merge-generics merge generic functions with a common name
330 into an <extended-generic>
331 merge-accessors merge accessors with a common name
333 The default duplicates handler is:
335 (replace warn-override-core warn last)
337 A recommended handler (which is likely to correspond to future Guile
338 behavior) can be installed with:
340 (default-duplicate-binding-handler '(replace warn-override-core check))
342 ** New define-module option: :replace
344 :replace works as :export, but, in addition, marks the binding as a
347 A typical example is `format' in (ice-9 format) which is a replacement
348 for the core binding `format'.
350 ** Adding prefixes to imported bindings in the module system
352 There is now a new :use-module option :prefix. It can be used to add
353 a prefix to all imported bindings.
356 :use-module ((bar) :prefix bar:))
358 will import all bindings exported from bar, but rename them by adding
361 ** Conflicting generic functions can be automatically merged.
363 When two imported bindings conflict and they are both generic
364 functions, the two functions can now be merged automatically. This is
365 activated with the 'duplicates' handler 'merge-generics'.
367 ### move the rest to the manual
369 It is sometimes tempting to use GOOPS accessors with short names.
370 For example, it is tempting to use the name `x' for the x-coordinate
373 Assume that we work with a graphical package which needs to use two
374 independent vector packages for 2D and 3D vectors respectively. If
375 both packages export `x' we will encounter a name collision.
377 This can now be resolved automagically with the duplicates handler
378 `merge-generics' which gives the module system license to merge all
379 generic functions sharing a common name:
381 (define-module (math 2D-vectors)
382 :use-module (oop goops)
385 (define-module (math 3D-vectors)
386 :use-module (oop goops)
389 (define-module (my-module)
390 :use-module (math 2D-vectors)
391 :use-module (math 3D-vectors)
392 :duplicates merge-generics)
394 x in (my-module) will now share methods with x in both imported
397 There will, in fact, now be three distinct generic functions named
398 `x': x in (2D-vectors), x in (3D-vectors), and x in (my-module). The
399 last function will be an <extended-generic>, extending the previous
402 Let's call the imported generic functions the "ancestor functions". x
403 in (my-module) is, in turn, a "descendant function" of the imported
404 functions, extending its ancestors.
406 For any generic function G, the applicable methods are selected from
407 the union of the methods of the descendant functions, the methods of G
408 itself and the methods of the ancestor functions.
410 This, ancestor functions share methods with their descendants and vice
411 versa. This implies that x in (math 2D-vectors) can will share the
412 methods of x in (my-module) and vice versa, while x in (math 2D-vectors)
413 doesn't share the methods of x in (math 3D-vectors), thus preserving
416 Sharing is dynamic, so that adding new methods to a descendant implies
417 adding it to the ancestor.
419 If duplicates checking is desired in the above example, the following
420 form of the :duplicates option can be used instead:
422 :duplicates (merge-generics check)
424 ** New function: effective-version
426 Returns the "effective" version number. This is just the normal full
427 version string without the final micro-version number. See "Changes
428 to the distribution" above.
430 ** New feature, 'futures': future, make-future, future-ref
432 Futures are like promises, but begin execution immediately in a new
433 thread. See the "Futures" section in the reference manual.
435 ** New threading functions: parallel, letpar, par-map, and friends
437 These are convenient ways to run calculations in parallel in new
438 threads. See "Parallel forms" in the manual for details.
440 ** Fair mutexes and condition variables
442 Fair mutexes and condition variables have been added. The fairness
443 means that scheduling is arranged to give as equal time shares as
444 possible and that threads are awakened in a first-in-first-out
445 manner. This is not guaranteed with standard mutexes and condition
448 In addition, fair mutexes are recursive. Locking a fair mutex that
449 you have already locked will succeed. Every call to lock-mutex must
450 be matched with a call to unlock-mutex. Only the last call to
451 unlock-mutex will actually unlock the mutex.
453 A fair condition variable must be used together with a fair mutex,
454 just as a standard condition variable must be used together with a
457 *** New functions: make-fair-mutex, make-fair-condition-variable'
459 Make a new fair mutex and a new fair condition variable respectively.
461 ** New function 'try-mutex'.
463 This function will attempt to lock a mutex but will return immediately
464 instead if blocking and indicate failure.
466 ** Waiting on a condition variable can have a timeout.
468 The funtion 'wait-condition-variable' now takes a third, optional
469 argument that specifies the point in time where the waiting should be
472 ** New function 'broadcast-condition-variable'.
474 ** New functions 'all-threads' and 'current-thread'.
476 ** Signals and system asyncs work better with threads.
478 The function 'sigaction' now takes a fourth, optional, argument that
479 specifies the thread that the handler should run in. When the
480 argument is omitted, the handler will run in the thread that called
483 Likewise, 'system-async-mark' takes a second, optional, argument that
484 specifies the thread that the async should run in. When it is
485 omitted, the async will run in the thread that called
488 C code can use the new functions scm_sigaction_for_thread and
489 scm_system_async_mark_for_thread to pass the new thread argument.
491 ** The function 'system-async' is deprecated.
493 You can now pass any zero-argument procedure to 'system-async-mark'.
494 The function 'system-async' will just return its argument unchanged
497 ** New functions 'call-with-blocked-asyncs' and
498 'call-with-unblocked-asyncs'
500 The expression (call-with-blocked-asyncs PROC) will call PROC and will
501 block execution of system asyncs for the current thread by one level
502 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
503 procedure and will unblock the execution of system asyncs by one
504 level for the current thread.
506 Only system asyncs are affected by these functions.
508 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
510 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
511 instead. Those functions are easier to use correctly and can be
514 ** New function 'unsetenv'.
516 ** New macro 'define-syntax-public'.
518 It works like 'define-syntax' and also exports the defined macro (but
521 ** There is support for Infinity and NaNs.
523 Following PLT Scheme, Guile can now work with infinite numbers, and
526 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
527 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
528 "+nan.0"). These numbers are inexact and have no exact counterpart.
530 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
531 sign of the dividend. The infinities are integers, and they answer #t
532 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
533 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
544 ERROR: Numerical overflow
546 Two new predicates 'inf?' and 'nan?' can be used to test for the
549 ** Inexact zero can have a sign.
551 Guile can now distinguish between plus and minus inexact zero, if your
552 platform supports this, too. The two zeros are equal according to
553 '=', but not according to 'eqv?'. For example
564 ** Guile now has exact rationals.
566 Guile can now represent fractions such as 1/3 exactly. Computing with
567 them is also done exactly, of course:
572 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
575 For example: (floor 2) now returns an exact 2 where in the past it
576 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
578 ** inexact->exact no longer returns only integers.
580 Without exact rationals, the closest exact number was always an
581 integer, but now inexact->exact returns the fraction that is exactly
582 equal to a floating point number. For example:
584 (inexact->exact 1.234)
585 => 694680242521899/562949953421312
587 When you want the old behavior, use 'round' explicitely:
589 (inexact->exact (round 1.234))
592 ** New function 'rationalize'.
594 This function finds a simple fraction that is close to a given real
595 number. For example (and compare with inexact->exact above):
597 (rationalize (inexact->exact 1.234) 1/2000)
600 Note that, as required by R5RS, rationalize returns only then an exact
601 result when both its arguments are exact.
603 ** 'odd?' and 'even?' work also for inexact integers.
605 Previously, (odd? 1.0) would signal an error since only exact integers
606 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
607 returns #f and (odd? 1.5) signals an error.
609 ** Guile now has uninterned symbols.
611 The new function 'make-symbol' will return an uninterned symbol. This
612 is a symbol that is unique and is guaranteed to remain unique.
613 However, uninterned symbols can not yet be read back in.
615 Use the new function 'symbol-interned?' to check whether a symbol is
618 ** pretty-print has more options.
620 The function pretty-print from the (ice-9 pretty-print) module can now
621 also be invoked with keyword arguments that control things like
622 maximum output width. See the manual for details.
624 ** Variables have no longer a special behavior for `equal?'.
626 Previously, comparing two variables with `equal?' would recursivly
627 compare their values. This is no longer done. Variables are now only
628 `equal?' if they are `eq?'.
630 ** `(begin)' is now valid.
632 You can now use an empty `begin' form. It will yield #<unspecified>
633 when evaluated and simply be ignored in a definition context.
635 ** Deprecated: procedure->macro
637 Change your code to use 'define-macro' or r5rs macros. Also, be aware
638 that macro expansion will not be done during evaluation, but prior to
641 ** Soft ports now allow a `char-ready?' procedure
643 The vector argument to `make-soft-port' can now have a length of
644 either 5 or 6. (Previously the length had to be 5.) The optional 6th
645 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
646 that returns the number of characters that can be read immediately
647 without the soft port blocking.
649 ** New debugging feature: breakpoints.
651 Guile now has breakpoints. For details see the `Debugging Features'
652 chapter in the reference manual.
654 ** Deprecated: undefine
656 There is no replacement for undefine.
658 * The functions make-keyword-from-dash-symbol and keyword-dash-symbol
659 have been discouraged.
661 They are relics from a time where a keyword like #:foo was used
662 directly as a Tcl option "-foo" and thus keywords were internally
663 stored as a symbol with a starting dash. We now store a symbol
666 Use symbol->keyword and keyword->symbol instead.
669 * Changes to the C interface
671 ** There is the new notion of 'discouraged' features.
673 This is a milder form of deprecation.
675 Things that are discouraged should not be used in new code, but it is
676 OK to leave them in old code for now. When a discouraged feature is
677 used, no warning message is printed like there is for 'deprecated'
678 features. Also, things that are merely discouraged are nevertheless
679 implemented efficiently, while deprecated features can be very slow.
681 You can omit discouraged features from libguile by configuring it with
682 the '--disable-discouraged' option.
684 ** A new family of functions for converting between C values and
685 Scheme values has been added.
687 These functions follow a common naming scheme and are designed to be
688 easier to use, thread-safe and more future-proof than the older
693 These are predicates that return a C boolean: 1 or 0. Instead of
694 SCM_NFALSEP, you can now use scm_is_true, for example.
696 - <type> scm_to_<type> (SCM val, ...)
698 These are functions that convert a Scheme value into an appropriate
699 C value. For example, you can use scm_to_int to safely convert from
702 - SCM scm_from_<type> (<type> val, ...)
704 These functions convert from a C type to a SCM value; for example,
705 scm_from_int for ints.
707 There is a huge number of these functions, for numbers, strings,
708 symbols, vectors, etc. They are documented in the reference manual in
709 the API section together with the types that they apply to.
711 ** New functions for dealing with complex numbers in C have been added.
713 The new functions are scm_c_make_rectangular, scm_c_make_polar,
714 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
715 They work like scm_make_rectangular etc but take or return doubles
718 ** The function scm_make_complex has been discouraged.
720 Use scm_c_make_rectangular instead.
722 ** The INUM macros have been deprecated.
724 A lot of code uses these macros to do general integer conversions,
725 although the macros only work correctly with fixnums. Use the
726 following alternatives.
728 SCM_INUMP -> scm_is_integer or similar
729 SCM_NINUMP -> !scm_is_integer or similar
730 SCM_MAKINUM -> scm_from_int or similar
731 SCM_INUM -> scm_to_int or similar
733 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
734 do the validating for you.
736 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
737 have been discouraged.
739 Use the newer scm_to_<type> and scm_from_<type> functions instead for
740 new code. The functions have been discouraged since they don't fit
743 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
745 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
746 evaluates its argument twice. Use scm_is_true, etc. instead for new
749 ** The macro SCM_EQ_P has been discouraged.
751 Use scm_is_eq for new code, which fits better into the naming
754 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
757 Use the function scm_is_pair or scm_is_null instead.
759 ** The functions scm_round and scm_truncate have been deprecated and
760 are now available as scm_c_round and scm_c_truncate, respectively.
762 These functions occupy the names that scm_round_number and
763 scm_truncate_number should have.
765 ** The functions scm_c_string2str, scm_c_substring2str, and
766 scm_c_symbol2str have been deprecated.
768 Use scm_to_locale_stringbuf or similar instead, maybe together with
771 ** New functions scm_c_make_string, scm_c_string_length,
772 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
773 scm_c_substring_shared, scm_c_substring_copy.
775 These are like scm_make_string, scm_length, etc. but are slightly
776 easier to use from C.
778 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
779 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
781 They export too many assumptions about the implementation of strings
782 and symbols that are no longer true in the presence of
783 mutation-sharing substrings and when Guile switches to some form of
786 When working with strings, it is often best to use the normal string
787 functions provided by Guile, such as scm_c_string_ref,
788 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
789 manual since many more such functions are now provided than
792 When you want to convert a SCM string to a C string, use the
793 scm_to_locale_string function or similar instead. For symbols, use
794 scm_symbol_to_string and then work with that string. Because of the
795 new string representation, scm_symbol_to_string does not need to copy
796 and is thus quite efficient.
798 ** Some string, symbol and keyword functions have been discouraged.
800 They don't fit into the uniform naming scheme and are not explicit
801 about the character encoding.
803 Replace according to the following table:
805 scm_allocate_string -> scm_c_make_string
806 scm_take_str -> scm_take_locale_stringn
807 scm_take0str -> scm_take_locale_string
808 scm_mem2string -> scm_from_locale_stringn
809 scm_str2string -> scm_from_locale_string
810 scm_makfrom0str -> scm_from_locale_string
811 scm_mem2symbol -> scm_from_locale_symboln
812 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
813 scm_str2symbol -> scm_from_locale_symbol
815 SCM_SYMBOL_HASH -> scm_hashq
816 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
818 scm_c_make_keyword -> scm_from_locale_keyword
820 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
821 now also available to C code.
823 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
825 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
826 the latter returns the true name of the keyword, not the 'dash name',
827 as SCM_KEYWORDSYM used to do.
829 ** The old uniform vector implementation has been unceremoniously
832 This implementation exposed the detailes of the tagging system of
833 Guile. Use the new C API explained in the manual in node "Uniform
836 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
837 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
838 SCM_SET_UVECTOR_LENGTH.
840 ** SCM_CELL_WORD_LOC has been deprecated.
842 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
843 to a SCM, as opposed to a pointer to a scm_t_bits.
845 This was done to allow the correct use of pointers into the Scheme
846 heap. Previously, the heap words were of type scm_t_bits and local
847 variables and function arguments were of type SCM, making it
848 non-standards-conformant to have a pointer that can point to both.
850 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
852 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
853 second and third words of double smobs. Likewise for
854 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
856 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
857 used to get and set the 16 exra bits in the zeroth word of a smob.
859 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
860 accesing the first immediate word of a smob as a SCM value, and there
861 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
862 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
864 ** New way to deal with non-local exits and re-entries.
866 There is a new set of functions that essentially do what
867 scm_internal_dynamic_wind does, but in a way that is more convenient
868 for C code in some situations. Here is a quick example of how to
869 prevent a potential memory leak:
878 mem = scm_malloc (100);
879 scm_frame_unwind_handler (free, mem, SCM_F_WIND_EXPLICITELY);
881 /* MEM would leak if BAR throws an error.
882 SCM_FRAME_UNWIND_HANDLER frees it nevertheless.
889 /* Because of SCM_F_WIND_EXPLICITELY, MEM will be freed by
890 SCM_FRAME_END as well.
894 For full documentation, see the node "Frames" in the manual.
896 ** New function scm_frame_free
898 This function calls 'free' on a given pointer when a frame is left.
899 Thus the call to scm_frame_unwind_handler above could be replaced with
900 simply scm_frame_free (mem).
902 ** New way to block and unblock asyncs
904 In addition to scm_c_call_with_blocked_asyncs you can now also use
905 scm_frame_block_asyncs in a 'frame' (see above). Likewise for
906 scm_c_call_with_unblocked_asyncs and scm_frame_unblock_asyncs.
908 ** New way to temporarily set the current input, output or error ports
910 C code can now use scm_frame_current_<foo>_port in a 'frame' (see
911 above). <foo> is one of "input", "output" or "error".
913 ** New way to temporarily set fluids
915 C code can now use scm_frame_fluid in a 'frame' (see
916 above) to temporarily set the value of a fluid.
918 ** New types scm_t_intmax and scm_t_uintmax.
920 On platforms that have them, these types are identical to intmax_t and
921 uintmax_t, respectively. On other platforms, they are identical to
922 the largest integer types that Guile knows about.
924 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
926 You should not have used them.
928 ** Many public #defines with generic names have been made private.
930 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
931 private or renamed with a more suitable public name.
933 ** The macro SCM_TYP16S has been deprecated.
935 This macro is not intended for public use.
937 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
939 Use scm_is_true (scm_inexact_p (...)) instead.
941 ** The macro SCM_SLOPPY_REALP has been deprecated.
943 Use scm_is_real instead.
945 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
947 Use scm_is_complex instead.
949 ** Some preprocessor defines have been deprecated.
951 These defines indicated whether a certain feature was present in Guile
952 or not. Going forward, assume that the features are always present.
954 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
955 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
957 The following macros have been removed completely: MEMOIZE_LOCALS,
958 SCM_RECKLESS, SCM_CAUTIOUS.
960 ** The preprocessor define STACK_DIRECTION has been deprecated.
962 There should be no need to know about the stack direction for ordinary
965 ** New function: scm_effective_version
967 Returns the "effective" version number. This is just the normal full
968 version string without the final micro-version number. See "Changes
969 to the distribution" above.
971 ** The function scm_call_with_new_thread has a new prototype.
973 Instead of taking a list with the thunk and handler, these two
974 arguments are now passed directly:
976 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
978 This is an incompatible change.
980 ** The value 'scm_mask_ints' is no longer writable.
982 Previously, you could set scm_mask_ints directly. This is no longer
983 possible. Use scm_c_call_with_blocked_asyncs and
984 scm_c_call_with_unblocked_asyncs instead.
986 ** New functions scm_c_call_with_blocked_asyncs and
987 scm_c_call_with_unblocked_asyncs
989 Like scm_call_with_blocked_asyncs etc. but for C functions.
991 ** New snarfer macro SCM_DEFINE_PUBLIC.
993 This is like SCM_DEFINE, but also calls scm_c_export for the defined
994 function in the init section.
996 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
998 ** New macros SCM_VECTOR_REF and SCM_VECTOR_SET.
1000 Use these in preference to SCM_VELTS.
1002 ** The SCM_VELTS macros now returns a read-only vector. For writing,
1003 use the new macros SCM_WRITABLE_VELTS or SCM_VECTOR_SET. The use of
1004 SCM_WRITABLE_VELTS is discouraged, though.
1006 ** Garbage collector rewrite.
1008 The garbage collector is cleaned up a lot, and now uses lazy
1009 sweeping. This is reflected in the output of (gc-stats); since cells
1010 are being freed when they are allocated, the cells-allocated field
1011 stays roughly constant.
1013 For malloc related triggers, the behavior is changed. It uses the same
1014 heuristic as the cell-triggered collections. It may be tuned with the
1015 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1016 for minimum yield of malloc related triggers. The default is 40.
1017 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1020 Debugging operations for the freelist have been deprecated, along with
1021 the C variables that control garbage collection. The environment
1022 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1023 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1025 ** The function scm_definedp has been renamed to scm_defined_p
1027 The name scm_definedp is deprecated.
1029 ** The struct scm_cell type has been renamed to scm_t_cell
1031 This is in accordance to Guile's naming scheme for types. Note that
1032 the name scm_cell is now used for a function that allocates and
1033 initializes a new cell (see below).
1035 ** New functions for memory management
1037 A new set of functions for memory management has been added since the
1038 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1039 indeed, Guile itself contained some long standing bugs that could
1040 cause aborts in long running programs.
1042 The new functions are more symmetrical and do not need cooperation
1043 from smob free routines, among other improvements.
1045 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1046 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1047 scm_gc_free, scm_gc_register_collectable_memory, and
1048 scm_gc_unregister_collectable_memory. Refer to the manual for more
1049 details and for upgrading instructions.
1051 The old functions for memory management have been deprecated. They
1052 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1053 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1055 ** Declarations of exported features are marked with SCM_API.
1057 Every declaration of a feature that belongs to the exported Guile API
1058 has been marked by adding the macro "SCM_API" to the start of the
1059 declaration. This macro can expand into different things, the most
1060 common of which is just "extern" for Unix platforms. On Win32, it can
1061 be used to control which symbols are exported from a DLL.
1063 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1064 will expand into "__declspec (dllimport) extern", which is needed for
1065 linking to the Guile DLL in Windows.
1067 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1068 SCM_SRFI4_IMPORT, for the corresponding libraries.
1070 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1072 Use the new functions scm_cell and scm_double_cell instead. The old
1073 macros had problems because with them allocation and initialization
1074 was separated and the GC could sometimes observe half initialized
1075 cells. Only careful coding by the user of SCM_NEWCELL and
1076 SCM_NEWCELL2 could make this safe and efficient.
1078 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1080 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1083 ** SRCBRKP has been deprecated.
1085 Use scm_c_source_property_breakpoint_p instead.
1087 ** Deprecated: scm_makmacro
1089 Change your code to use either scm_makmmacro or to define macros in
1090 Scheme, using 'define-macro'.
1092 ** New function scm_c_port_for_each.
1094 This function is like scm_port_for_each but takes a pointer to a C
1095 function as the callback instead of a SCM value.
1097 ** Many definitions have been removed that were previously deprecated.
1099 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1100 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1101 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1102 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1103 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1104 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1105 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1106 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1107 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1108 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1109 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1110 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1111 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1112 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1113 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1114 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1115 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1116 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1117 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1118 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1119 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1120 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1121 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1122 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1123 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1124 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1125 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1126 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1127 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1128 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1129 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1132 Changes since Guile 1.4:
1134 * Changes to the distribution
1136 ** A top-level TODO file is included.
1138 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1140 Guile now always uses three numbers to represent the version,
1141 i.e. "1.6.5". The first number, 1, is the major version number, the
1142 second number, 6, is the minor version number, and the third number,
1143 5, is the micro version number. Changes in major version number
1144 indicate major changes in Guile.
1146 Minor version numbers that are even denote stable releases, and odd
1147 minor version numbers denote development versions (which may be
1148 unstable). The micro version number indicates a minor sub-revision of
1149 a given MAJOR.MINOR release.
1151 In keeping with the new scheme, (minor-version) and scm_minor_version
1152 no longer return everything but the major version number. They now
1153 just return the minor version number. Two new functions
1154 (micro-version) and scm_micro_version have been added to report the
1155 micro version number.
1157 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1159 ** New preprocessor definitions are available for checking versions.
1161 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1162 SCM_MICRO_VERSION to the appropriate integer values.
1164 ** Guile now actively warns about deprecated features.
1166 The new configure option `--enable-deprecated=LEVEL' and the
1167 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1168 See INSTALL and README for more information.
1170 ** Guile is much more likely to work on 64-bit architectures.
1172 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1173 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1174 for the use of a test machine, and thanks to Stefan Jahn for ia64
1177 ** New functions: setitimer and getitimer.
1179 These implement a fairly direct interface to the libc functions of the
1182 ** The #. reader extension is now disabled by default.
1184 For safety reasons, #. evaluation is disabled by default. To
1185 re-enable it, set the fluid read-eval? to #t. For example:
1187 (fluid-set! read-eval? #t)
1189 but make sure you realize the potential security risks involved. With
1190 read-eval? enabled, reading a data file from an untrusted source can
1193 ** New SRFI modules have been added:
1195 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1198 (srfi srfi-1) is a library containing many useful pair- and list-processing
1201 (srfi srfi-2) exports and-let*.
1203 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1205 (srfi srfi-6) is a dummy module for now, since guile already provides
1206 all of the srfi-6 procedures by default: open-input-string,
1207 open-output-string, get-output-string.
1209 (srfi srfi-8) exports receive.
1211 (srfi srfi-9) exports define-record-type.
1213 (srfi srfi-10) exports define-reader-ctor and implements the reader
1216 (srfi srfi-11) exports let-values and let*-values.
1218 (srfi srfi-13) implements the SRFI String Library.
1220 (srfi srfi-14) implements the SRFI Character-Set Library.
1222 (srfi srfi-17) implements setter and getter-with-setter and redefines
1223 some accessor procedures as procedures with getters. (such as car,
1224 cdr, vector-ref etc.)
1226 (srfi srfi-19) implements the SRFI Time/Date Library.
1228 ** New scripts / "executable modules"
1230 Subdirectory "scripts" contains Scheme modules that are packaged to
1231 also be executable as scripts. At this time, these scripts are available:
1240 See README there for more info.
1242 These scripts can be invoked from the shell with the new program
1243 "guile-tools", which keeps track of installation directory for you.
1246 $ guile-tools display-commentary srfi/*.scm
1248 guile-tools is copied to the standard $bindir on "make install".
1250 ** New module (ice-9 stack-catch):
1252 stack-catch is like catch, but saves the current state of the stack in
1253 the fluid the-last-stack. This fluid can be useful when using the
1254 debugger and when re-throwing an error.
1256 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1258 This has been done to prevent problems on lesser operating systems
1259 that can't tolerate `*'s in file names. The exported macro continues
1260 to be named `and-let*', of course.
1262 On systems that support it, there is also a compatibility module named
1263 (ice-9 and-let*). It will go away in the next release.
1265 ** New modules (oop goops) etc.:
1268 (oop goops describe)
1270 (oop goops active-slot)
1271 (oop goops composite-slot)
1273 The Guile Object Oriented Programming System (GOOPS) has been
1274 integrated into Guile. For further information, consult the GOOPS
1275 manual and tutorial in the `doc' directory.
1277 ** New module (ice-9 rdelim).
1279 This exports the following procedures which were previously defined
1280 in the default environment:
1282 read-line read-line! read-delimited read-delimited! %read-delimited!
1283 %read-line write-line
1285 For backwards compatibility the definitions are still imported into the
1286 default environment in this version of Guile. However you should add:
1288 (use-modules (ice-9 rdelim))
1290 to any program which uses the definitions, since this may change in
1293 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1294 can be used for similar functionality.
1296 ** New module (ice-9 rw)
1298 This is a subset of the (scsh rw) module from guile-scsh. Currently
1299 it defines two procedures:
1301 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1303 Read characters from a port or file descriptor into a string STR.
1304 A port must have an underlying file descriptor -- a so-called
1305 fport. This procedure is scsh-compatible and can efficiently read
1308 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1310 Write characters from a string STR to a port or file descriptor.
1311 A port must have an underlying file descriptor -- a so-called
1312 fport. This procedure is mostly compatible and can efficiently
1313 write large strings.
1315 ** New module (ice-9 match)
1317 This module includes Andrew K. Wright's pattern matcher. See
1318 ice-9/match.scm for brief description or
1320 http://www.star-lab.com/wright/code.html
1322 for complete documentation.
1324 ** New module (ice-9 buffered-input)
1326 This module provides procedures to construct an input port from an
1327 underlying source of input that reads and returns its input in chunks.
1328 The underlying input source is a Scheme procedure, specified by the
1329 caller, which the port invokes whenever it needs more input.
1331 This is useful when building an input port whose back end is Readline
1332 or a UI element such as the GtkEntry widget.
1336 The reference and tutorial documentation that was previously
1337 distributed separately, as `guile-doc', is now included in the core
1338 Guile distribution. The documentation consists of the following
1341 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1344 - The Guile Reference Manual (guile.texi) contains (or is intended to
1345 contain) reference documentation on all aspects of Guile.
1347 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1348 reference documentation for using GOOPS, Guile's Object Oriented
1351 - The Revised^5 Report on the Algorithmic Language Scheme
1354 See the README file in the `doc' directory for more details.
1356 ** There are a couple of examples in the examples/ directory now.
1358 * Changes to the stand-alone interpreter
1360 ** New command line option `--use-srfi'
1362 Using this option, SRFI modules can be loaded on startup and be
1363 available right from the beginning. This makes programming portable
1364 Scheme programs easier.
1366 The option `--use-srfi' expects a comma-separated list of numbers,
1367 each representing a SRFI number to be loaded into the interpreter
1368 before starting evaluating a script file or the REPL. Additionally,
1369 the feature identifier for the loaded SRFIs is recognized by
1370 `cond-expand' when using this option.
1373 $ guile --use-srfi=8,13
1374 guile> (receive (x z) (values 1 2) (+ 1 2))
1376 guile> (string-pad "bla" 20)
1379 ** Guile now always starts up in the `(guile-user)' module.
1381 Previously, scripts executed via the `-s' option would run in the
1382 `(guile)' module and the repl would run in the `(guile-user)' module.
1383 Now every user action takes place in the `(guile-user)' module by
1386 * Changes to Scheme functions and syntax
1388 ** Character classifiers work for non-ASCII characters.
1390 The predicates `char-alphabetic?', `char-numeric?',
1391 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1392 no longer check whether their arguments are ASCII characters.
1393 Previously, a character would only be considered alphabetic when it
1394 was also ASCII, for example.
1396 ** Previously deprecated Scheme functions have been removed:
1398 tag - no replacement.
1399 fseek - replaced by seek.
1400 list* - replaced by cons*.
1402 ** It's now possible to create modules with controlled environments
1406 (use-modules (ice-9 safe))
1407 (define m (make-safe-module))
1408 ;;; m will now be a module containing only a safe subset of R5RS
1409 (eval '(+ 1 2) m) --> 3
1410 (eval 'load m) --> ERROR: Unbound variable: load
1412 ** Evaluation of "()", the empty list, is now an error.
1414 Previously, the expression "()" evaluated to the empty list. This has
1415 been changed to signal a "missing expression" error. The correct way
1416 to write the empty list as a literal constant is to use quote: "'()".
1418 ** New concept of `Guile Extensions'.
1420 A Guile Extension is just a ordinary shared library that can be linked
1421 at run-time. We found it advantageous to give this simple concept a
1422 dedicated name to distinguish the issues related to shared libraries
1423 from the issues related to the module system.
1425 *** New function: load-extension
1427 Executing (load-extension lib init) is mostly equivalent to
1429 (dynamic-call init (dynamic-link lib))
1431 except when scm_register_extension has been called previously.
1432 Whenever appropriate, you should use `load-extension' instead of
1433 dynamic-link and dynamic-call.
1435 *** New C function: scm_c_register_extension
1437 This function registers a initialization function for use by
1438 `load-extension'. Use it when you don't want specific extensions to
1439 be loaded as shared libraries (for example on platforms that don't
1440 support dynamic linking).
1442 ** Auto-loading of compiled-code modules is deprecated.
1444 Guile used to be able to automatically find and link a shared
1445 library to satisfy requests for a module. For example, the module
1446 `(foo bar)' could be implemented by placing a shared library named
1447 "foo/libbar.so" (or with a different extension) in a directory on the
1450 This has been found to be too tricky, and is no longer supported. The
1451 shared libraries are now called "extensions". You should now write a
1452 small Scheme file that calls `load-extension' to load the shared
1453 library and initialize it explicitely.
1455 The shared libraries themselves should be installed in the usual
1456 places for shared libraries, with names like "libguile-foo-bar".
1458 For example, place this into a file "foo/bar.scm"
1460 (define-module (foo bar))
1462 (load-extension "libguile-foo-bar" "foobar_init")
1464 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1466 `eval' is now R5RS, that is it takes two arguments.
1467 The second argument is an environment specifier, i.e. either
1469 (scheme-report-environment 5)
1470 (null-environment 5)
1471 (interaction-environment)
1477 ** The module system has been made more disciplined.
1479 The function `eval' will save and restore the current module around
1480 the evaluation of the specified expression. While this expression is
1481 evaluated, `(current-module)' will now return the right module, which
1482 is the module specified as the second argument to `eval'.
1484 A consequence of this change is that `eval' is not particularly
1485 useful when you want allow the evaluated code to change what module is
1486 designated as the current module and have this change persist from one
1487 call to `eval' to the next. The read-eval-print-loop is an example
1488 where `eval' is now inadequate. To compensate, there is a new
1489 function `primitive-eval' that does not take a module specifier and
1490 that does not save/restore the current module. You should use this
1491 function together with `set-current-module', `current-module', etc
1492 when you want to have more control over the state that is carried from
1493 one eval to the next.
1495 Additionally, it has been made sure that forms that are evaluated at
1496 the top level are always evaluated with respect to the current module.
1497 Previously, subforms of top-level forms such as `begin', `case',
1498 etc. did not respect changes to the current module although these
1499 subforms are at the top-level as well.
1501 To prevent strange behavior, the forms `define-module',
1502 `use-modules', `use-syntax', and `export' have been restricted to only
1503 work on the top level. The forms `define-public' and
1504 `defmacro-public' only export the new binding on the top level. They
1505 behave just like `define' and `defmacro', respectively, when they are
1506 used in a lexical environment.
1508 Also, `export' will no longer silently re-export bindings imported
1509 from a used module. It will emit a `deprecation' warning and will
1510 cease to perform any re-export in the next version. If you actually
1511 want to re-export bindings, use the new `re-export' in place of
1512 `export'. The new `re-export' will not make copies of variables when
1513 rexporting them, as `export' did wrongly.
1515 ** Module system now allows selection and renaming of imported bindings
1517 Previously, when using `use-modules' or the `#:use-module' clause in
1518 the `define-module' form, all the bindings (association of symbols to
1519 values) for imported modules were added to the "current module" on an
1520 as-is basis. This has been changed to allow finer control through two
1521 new facilities: selection and renaming.
1523 You can now select which of the imported module's bindings are to be
1524 visible in the current module by using the `:select' clause. This
1525 clause also can be used to rename individual bindings. For example:
1527 ;; import all bindings no questions asked
1528 (use-modules (ice-9 common-list))
1530 ;; import four bindings, renaming two of them;
1531 ;; the current module sees: every some zonk-y zonk-n
1532 (use-modules ((ice-9 common-list)
1534 (remove-if . zonk-y)
1535 (remove-if-not . zonk-n))))
1537 You can also programmatically rename all selected bindings using the
1538 `:renamer' clause, which specifies a proc that takes a symbol and
1539 returns another symbol. Because it is common practice to use a prefix,
1540 we now provide the convenience procedure `symbol-prefix-proc'. For
1543 ;; import four bindings, renaming two of them specifically,
1544 ;; and all four w/ prefix "CL:";
1545 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1546 (use-modules ((ice-9 common-list)
1548 (remove-if . zonk-y)
1549 (remove-if-not . zonk-n))
1550 :renamer (symbol-prefix-proc 'CL:)))
1552 ;; import four bindings, renaming two of them specifically,
1553 ;; and all four by upcasing.
1554 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1555 (define (upcase-symbol sym)
1556 (string->symbol (string-upcase (symbol->string sym))))
1558 (use-modules ((ice-9 common-list)
1560 (remove-if . zonk-y)
1561 (remove-if-not . zonk-n))
1562 :renamer upcase-symbol))
1564 Note that programmatic renaming is done *after* individual renaming.
1565 Also, the above examples show `use-modules', but the same facilities are
1566 available for the `#:use-module' clause of `define-module'.
1568 See manual for more info.
1570 ** The semantics of guardians have changed.
1572 The changes are for the most part compatible. An important criterion
1573 was to keep the typical usage of guardians as simple as before, but to
1574 make the semantics safer and (as a result) more useful.
1576 *** All objects returned from guardians are now properly alive.
1578 It is now guaranteed that any object referenced by an object returned
1579 from a guardian is alive. It's now impossible for a guardian to
1580 return a "contained" object before its "containing" object.
1582 One incompatible (but probably not very important) change resulting
1583 from this is that it is no longer possible to guard objects that
1584 indirectly reference themselves (i.e. are parts of cycles). If you do
1585 so accidentally, you'll get a warning.
1587 *** There are now two types of guardians: greedy and sharing.
1589 If you call (make-guardian #t) or just (make-guardian), you'll get a
1590 greedy guardian, and for (make-guardian #f) a sharing guardian.
1592 Greedy guardians are the default because they are more "defensive".
1593 You can only greedily guard an object once. If you guard an object
1594 more than once, once in a greedy guardian and the rest of times in
1595 sharing guardians, then it is guaranteed that the object won't be
1596 returned from sharing guardians as long as it is greedily guarded
1599 Guardians returned by calls to `make-guardian' can now take one more
1600 optional parameter, which says whether to throw an error in case an
1601 attempt is made to greedily guard an object that is already greedily
1602 guarded. The default is true, i.e. throw an error. If the parameter
1603 is false, the guardian invocation returns #t if guarding was
1604 successful and #f if it wasn't.
1606 Also, since greedy guarding is, in effect, a side-effecting operation
1607 on objects, a new function is introduced: `destroy-guardian!'.
1608 Invoking this function on a guardian renders it unoperative and, if
1609 the guardian is greedy, clears the "greedily guarded" property of the
1610 objects that were guarded by it, thus undoing the side effect.
1612 Note that all this hair is hardly very important, since guardian
1613 objects are usually permanent.
1615 ** Continuations created by call-with-current-continuation now accept
1616 any number of arguments, as required by R5RS.
1618 ** New function `issue-deprecation-warning'
1620 This function is used to display the deprecation messages that are
1621 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1624 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1628 ;; `id' is deprecated. Use `identity' instead.
1633 ** New syntax `begin-deprecated'
1635 When deprecated features are included (as determined by the configure
1636 option --enable-deprecated), `begin-deprecated' is identical to
1637 `begin'. When deprecated features are excluded, it always evaluates
1638 to `#f', ignoring the body forms.
1640 ** New function `make-object-property'
1642 This function returns a new `procedure with setter' P that can be used
1643 to attach a property to objects. When calling P as
1647 where `obj' is any kind of object, it attaches `val' to `obj' in such
1648 a way that it can be retrieved by calling P as
1652 This function will replace procedure properties, symbol properties and
1653 source properties eventually.
1655 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1657 Instead of #&optional, #&key, etc you should now use #:optional,
1658 #:key, etc. Since #:optional is a keyword, you can write it as just
1659 :optional when (read-set! keywords 'prefix) is active.
1661 The old reader syntax `#&' is still supported, but deprecated. It
1662 will be removed in the next release.
1664 ** New define-module option: pure
1666 Tells the module system not to include any bindings from the root
1671 (define-module (totally-empty-module)
1674 ** New define-module option: export NAME1 ...
1676 Export names NAME1 ...
1678 This option is required if you want to be able to export bindings from
1679 a module which doesn't import one of `define-public' or `export'.
1683 (define-module (foo)
1685 :use-module (ice-9 r5rs)
1688 ;;; Note that we're pure R5RS below this point!
1693 ** New function: object->string OBJ
1695 Return a Scheme string obtained by printing a given object.
1697 ** New function: port? X
1699 Returns a boolean indicating whether X is a port. Equivalent to
1700 `(or (input-port? X) (output-port? X))'.
1702 ** New function: file-port?
1704 Determines whether a given object is a port that is related to a file.
1706 ** New function: port-for-each proc
1708 Apply PROC to each port in the Guile port table in turn. The return
1709 value is unspecified. More specifically, PROC is applied exactly once
1710 to every port that exists in the system at the time PORT-FOR-EACH is
1711 invoked. Changes to the port table while PORT-FOR-EACH is running
1712 have no effect as far as PORT-FOR-EACH is concerned.
1714 ** New function: dup2 oldfd newfd
1716 A simple wrapper for the `dup2' system call. Copies the file
1717 descriptor OLDFD to descriptor number NEWFD, replacing the
1718 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1719 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1720 to move away ports which are using NEWFD. The return value is
1723 ** New function: close-fdes fd
1725 A simple wrapper for the `close' system call. Close file
1726 descriptor FD, which must be an integer. Unlike close (*note
1727 close: Ports and File Descriptors.), the file descriptor will be
1728 closed even if a port is using it. The return value is
1731 ** New function: crypt password salt
1733 Encrypts `password' using the standard unix password encryption
1736 ** New function: chroot path
1738 Change the root directory of the running process to `path'.
1740 ** New functions: getlogin, cuserid
1742 Return the login name or the user name of the current effective user
1745 ** New functions: getpriority which who, setpriority which who prio
1747 Get or set the priority of the running process.
1749 ** New function: getpass prompt
1751 Read a password from the terminal, first displaying `prompt' and
1754 ** New function: flock file operation
1756 Set/remove an advisory shared or exclusive lock on `file'.
1758 ** New functions: sethostname name, gethostname
1760 Set or get the hostname of the machine the current process is running
1763 ** New function: mkstemp! tmpl
1765 mkstemp creates a new unique file in the file system and returns a
1766 new buffered port open for reading and writing to the file. TMPL
1767 is a string specifying where the file should be created: it must
1768 end with `XXXXXX' and will be changed in place to return the name
1769 of the temporary file.
1771 ** New function: open-input-string string
1773 Return an input string port which delivers the characters from
1774 `string'. This procedure, together with `open-output-string' and
1775 `get-output-string' implements SRFI-6.
1777 ** New function: open-output-string
1779 Return an output string port which collects all data written to it.
1780 The data can then be retrieved by `get-output-string'.
1782 ** New function: get-output-string
1784 Return the contents of an output string port.
1786 ** New function: identity
1788 Return the argument.
1790 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1791 are represented in Scheme as integers with normal host byte ordering.
1793 ** New function: inet-pton family address
1795 Convert a printable string network address into an integer. Note that
1796 unlike the C version of this function, the result is an integer with
1797 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1800 (inet-pton AF_INET "127.0.0.1") => 2130706433
1801 (inet-pton AF_INET6 "::1") => 1
1803 ** New function: inet-ntop family address
1805 Convert an integer network address into a printable string. Note that
1806 unlike the C version of this function, the input is an integer with
1807 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1810 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1811 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1812 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1816 Use `identity' instead.
1822 ** Deprecated: return-it
1826 ** Deprecated: string-character-length
1828 Use `string-length' instead.
1830 ** Deprecated: flags
1832 Use `logior' instead.
1834 ** Deprecated: close-all-ports-except.
1836 This was intended for closing ports in a child process after a fork,
1837 but it has the undesirable side effect of flushing buffers.
1838 port-for-each is more flexible.
1840 ** The (ice-9 popen) module now attempts to set up file descriptors in
1841 the child process from the current Scheme ports, instead of using the
1842 current values of file descriptors 0, 1, and 2 in the parent process.
1844 ** Removed function: builtin-weak-bindings
1846 There is no such concept as a weak binding any more.
1848 ** Removed constants: bignum-radix, scm-line-incrementors
1850 ** define-method: New syntax mandatory.
1852 The new method syntax is now mandatory:
1854 (define-method (NAME ARG-SPEC ...) BODY ...)
1855 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1857 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1858 REST-ARG ::= ARG-NAME
1860 If you have old code using the old syntax, import
1861 (oop goops old-define-method) before (oop goops) as in:
1863 (use-modules (oop goops old-define-method) (oop goops))
1865 ** Deprecated function: builtin-variable
1866 Removed function: builtin-bindings
1868 There is no longer a distinction between builtin or other variables.
1869 Use module system operations for all variables.
1871 ** Lazy-catch handlers are no longer allowed to return.
1873 That is, a call to `throw', `error', etc is now guaranteed to not
1876 ** Bugfixes for (ice-9 getopt-long)
1878 This module is now tested using test-suite/tests/getopt-long.test.
1879 The following bugs have been fixed:
1881 *** Parsing for options that are specified to have `optional' args now checks
1882 if the next element is an option instead of unconditionally taking it as the
1885 *** An error is now thrown for `--opt=val' when the option description
1886 does not specify `(value #t)' or `(value optional)'. This condition used to
1887 be accepted w/o error, contrary to the documentation.
1889 *** The error message for unrecognized options is now more informative.
1890 It used to be "not a record", an artifact of the implementation.
1892 *** The error message for `--opt' terminating the arg list (no value), when
1893 `(value #t)' is specified, is now more informative. It used to be "not enough
1896 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1897 The expansion used to be like so:
1899 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1901 Note that the "5d" is dropped. Now it is like so:
1903 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1905 This enables single-char options to have adjoining arguments as long as their
1906 constituent characters are not potential single-char options.
1908 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1910 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1911 property `arglist', which can be retrieved by `arity'. The result is that
1912 `arity' can give more detailed information than before:
1916 guile> (use-modules (ice-9 optargs))
1917 guile> (define* (foo #:optional a b c) a)
1919 0 or more arguments in `lambda*:G0'.
1924 3 optional arguments: `a', `b' and `c'.
1925 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
1927 2 required arguments: `a' and `b', 2 keyword arguments: `c'
1928 and `d', other keywords allowed.
1929 guile> (define* (baz a b #:optional c #:rest r) a)
1931 2 required arguments: `a' and `b', 1 optional argument: `c',
1934 * Changes to the C interface
1936 ** Types have been renamed from scm_*_t to scm_t_*.
1938 This has been done for POSIX sake. It reserves identifiers ending
1939 with "_t". What a concept.
1941 The old names are still available with status `deprecated'.
1943 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
1945 ** Deprecated features have been removed.
1949 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
1950 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
1952 *** C Functions removed
1954 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
1955 scm_fseek - replaced by scm_seek.
1956 gc-thunk - replaced by after-gc-hook.
1957 gh_int2scmb - replaced by gh_bool2scm.
1958 scm_tc_dblr - replaced by scm_tc16_real.
1959 scm_tc_dblc - replaced by scm_tc16_complex.
1960 scm_list_star - replaced by scm_cons_star.
1962 ** Deprecated: scm_makfromstr
1964 Use scm_mem2string instead.
1966 ** Deprecated: scm_make_shared_substring
1968 Explicit shared substrings will disappear from Guile.
1970 Instead, "normal" strings will be implemented using sharing
1971 internally, combined with a copy-on-write strategy.
1973 ** Deprecated: scm_read_only_string_p
1975 The concept of read-only strings will disappear in next release of
1978 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
1980 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
1982 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
1984 Call a procedure with the indicated number of arguments. See "Fly
1985 Evaluation" in the manual.
1987 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
1989 Call a procedure with the indicated number of arguments and a list of
1990 further arguments. See "Fly Evaluation" in the manual.
1992 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
1994 Create a list of the given number of elements. See "List
1995 Constructors" in the manual.
1997 ** Renamed function: scm_listify has been replaced by scm_list_n.
1999 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2000 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2002 Use functions scm_list_N instead.
2004 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2006 Used by an application to read arbitrary number of bytes from a port.
2007 Same semantics as libc read, except that scm_c_read only returns less
2008 than SIZE bytes if at end-of-file.
2010 Warning: Doesn't update port line and column counts!
2012 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2014 Used by an application to write arbitrary number of bytes to an SCM
2015 port. Similar semantics as libc write. However, unlike libc
2016 write, scm_c_write writes the requested number of bytes and has no
2019 Warning: Doesn't update port line and column counts!
2021 ** New function: scm_init_guile ()
2023 In contrast to scm_boot_guile, scm_init_guile will return normally
2024 after initializing Guile. It is not available on all systems, tho.
2026 ** New functions: scm_str2symbol, scm_mem2symbol
2028 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2029 field of characters and creates a scheme symbol object from that C string.
2030 The function scm_mem2symbol takes a const char* and a number of characters and
2031 creates a symbol from the characters in that memory area.
2033 ** New functions: scm_primitive_make_property
2034 scm_primitive_property_ref
2035 scm_primitive_property_set_x
2036 scm_primitive_property_del_x
2038 These functions implement a new way to deal with object properties.
2039 See libguile/properties.c for their documentation.
2041 ** New function: scm_done_free (long size)
2043 This function is the inverse of scm_done_malloc. Use it to report the
2044 amount of smob memory you free. The previous method, which involved
2045 calling scm_done_malloc with negative argument, was somewhat
2046 unintuitive (and is still available, of course).
2048 ** New function: scm_c_memq (SCM obj, SCM list)
2050 This function provides a fast C level alternative for scm_memq for the case
2051 that the list parameter is known to be a proper list. The function is a
2052 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2053 list input parameter, since for anything else but a proper list the function's
2054 behaviour is undefined - it may even crash or loop endlessly. Further, for
2055 the case that the object is not found in the list, scm_c_memq returns #f which
2056 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2058 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2059 scm_remember_upto_here
2061 These functions replace the function scm_remember.
2063 ** Deprecated function: scm_remember
2065 Use one of the new functions scm_remember_upto_here_1,
2066 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2068 ** New function: scm_allocate_string
2070 This function replaces the function scm_makstr.
2072 ** Deprecated function: scm_makstr
2074 Use the new function scm_allocate_string instead.
2076 ** New global variable scm_gc_running_p introduced.
2078 Use this variable to find out if garbage collection is being executed. Up to
2079 now applications have used scm_gc_heap_lock to test if garbage collection was
2080 running, which also works because of the fact that up to know only the garbage
2081 collector has set this variable. But, this is an implementation detail that
2082 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2083 of this variable is (and has been) not fully safe anyway.
2085 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2087 Use these instead of SCM_LENGTH_MAX.
2089 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2090 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2091 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2093 Use these instead of SCM_LENGTH.
2095 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2096 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2097 SCM_SET_BITVECTOR_LENGTH
2099 Use these instead of SCM_SETLENGTH
2101 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2102 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2105 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2108 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2109 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2112 Use these instead of SCM_SETCHARS.
2114 ** New macro: SCM_BITVECTOR_P
2116 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2118 Use instead of SCM_COERCE_SUBSTR.
2120 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2122 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2124 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2125 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2126 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2127 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2128 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2129 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2130 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2131 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2132 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2133 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2134 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2135 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2136 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2137 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2138 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2140 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2141 Use scm_memory_error instead of SCM_NALLOC.
2142 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2143 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2144 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2145 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2146 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2147 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2148 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2149 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2150 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2151 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2152 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2153 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2154 Use a type specific setter macro instead of SCM_SETLENGTH.
2155 Use a type specific setter macro instead of SCM_SETCHARS.
2156 Use a type specific length macro instead of SCM_LENGTH_MAX.
2157 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2158 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2159 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2160 Use SCM_TYP16 instead of SCM_GCTYP16.
2161 Use SCM_CDR instead of SCM_GCCDR.
2162 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2163 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2164 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2165 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2166 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2167 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2168 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2170 ** Removed function: scm_struct_init
2172 ** Removed variable: scm_symhash_dim
2174 ** Renamed function: scm_make_cont has been replaced by
2175 scm_make_continuation, which has a different interface.
2177 ** Deprecated function: scm_call_catching_errors
2179 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2181 ** Deprecated function: scm_strhash
2183 Use scm_string_hash instead.
2185 ** Deprecated function: scm_vector_set_length_x
2187 Instead, create a fresh vector of the desired size and copy the contents.
2189 ** scm_gensym has changed prototype
2191 scm_gensym now only takes one argument.
2193 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2196 There is now only a single symbol type scm_tc7_symbol.
2197 The tag scm_tc7_lvector was not used anyway.
2199 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2201 Use scm_make_smob_type and scm_set_smob_XXX instead.
2203 ** New function scm_set_smob_apply.
2205 This can be used to set an apply function to a smob type.
2207 ** Deprecated function: scm_strprint_obj
2209 Use scm_object_to_string instead.
2211 ** Deprecated function: scm_wta
2213 Use scm_wrong_type_arg, or another appropriate error signalling function
2216 ** Explicit support for obarrays has been deprecated.
2218 Use `scm_str2symbol' and the generic hashtable functions instead.
2220 ** The concept of `vcells' has been deprecated.
2222 The data type `variable' is now used exclusively. `Vcells' have been
2223 a low-level concept so you are likely not affected by this change.
2225 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2226 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2228 Use scm_c_define or scm_c_lookup instead, as appropriate.
2230 *** New functions: scm_c_module_lookup, scm_c_lookup,
2231 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2232 scm_module_define, scm_define.
2234 These functions work with variables instead of with vcells.
2236 ** New functions for creating and defining `subr's and `gsubr's.
2238 The new functions more clearly distinguish between creating a subr (or
2239 gsubr) object and adding it to the current module.
2241 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2242 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2243 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2244 scm_c_define_gsubr_with_generic.
2246 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2247 scm_make_subr_with_generic, scm_make_gsubr,
2248 scm_make_gsubr_with_generic.
2250 Use the new ones from above instead.
2252 ** C interface to the module system has changed.
2254 While we suggest that you avoid as many explicit module system
2255 operations from C as possible for the time being, the C interface has
2256 been made more similar to the high-level Scheme module system.
2258 *** New functions: scm_c_define_module, scm_c_use_module,
2259 scm_c_export, scm_c_resolve_module.
2261 They mostly work like their Scheme namesakes. scm_c_define_module
2262 takes a function that is called a context where the new module is
2265 *** Deprecated functions: scm_the_root_module, scm_make_module,
2266 scm_ensure_user_module, scm_load_scheme_module.
2268 Use the new functions instead.
2270 ** Renamed function: scm_internal_with_fluids becomes
2273 scm_internal_with_fluids is available as a deprecated function.
2275 ** New function: scm_c_with_fluid.
2277 Just like scm_c_with_fluids, but takes one fluid and one value instead
2280 ** Deprecated typedefs: long_long, ulong_long.
2282 They are of questionable utility and they pollute the global
2285 ** Deprecated typedef: scm_sizet
2287 It is of questionable utility now that Guile requires ANSI C, and is
2290 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2291 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2292 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2294 Made more compliant with the naming policy by adding a _t at the end.
2296 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2297 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2299 With the exception of the mysterious scm_2ulong2big, they are still
2300 available under new names (scm_i_mkbig etc). These functions are not
2301 intended to be used in user code. You should avoid dealing with
2302 bignums directly, and should deal with numbers in general (which can
2305 ** Change in behavior: scm_num2long, scm_num2ulong
2307 The scm_num2[u]long functions don't any longer accept an inexact
2308 argument. This change in behavior is motivated by concordance with
2309 R5RS: It is more common that a primitive doesn't want to accept an
2310 inexact for an exact.
2312 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2313 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2314 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2317 These are conversion functions between the various ANSI C integral
2318 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2319 accept an inexact argument.
2321 ** New functions: scm_float2num, scm_double2num,
2322 scm_num2float, scm_num2double.
2324 These are conversion functions between the two ANSI C float types and
2327 ** New number validation macros:
2328 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2332 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2334 These are just nicer-named old scm_protect_object and
2335 scm_unprotect_object.
2337 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2339 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2341 These functions can be used to register pointers to locations that
2344 ** Deprecated function: scm_create_hook.
2346 Its sins are: misleading name, non-modularity and lack of general
2350 Changes since Guile 1.3.4:
2352 * Changes to the distribution
2354 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2356 We've changed the way we handle generated files in the Guile source
2357 repository. As a result, the procedure for building trees obtained
2358 from the nightly FTP snapshots or via CVS has changed:
2359 - You must have appropriate versions of autoconf, automake, and
2360 libtool installed on your system. See README for info on how to
2361 obtain these programs.
2362 - Before configuring the tree, you must first run the script
2363 `autogen.sh' at the top of the source tree.
2365 The Guile repository used to contain not only source files, written by
2366 humans, but also some generated files, like configure scripts and
2367 Makefile.in files. Even though the contents of these files could be
2368 derived mechanically from other files present, we thought it would
2369 make the tree easier to build if we checked them into CVS.
2371 However, this approach means that minor differences between
2372 developer's installed tools and habits affected the whole team.
2373 So we have removed the generated files from the repository, and
2374 added the autogen.sh script, which will reconstruct them
2378 ** configure now has experimental options to remove support for certain
2381 --disable-arrays omit array and uniform array support
2382 --disable-posix omit posix interfaces
2383 --disable-networking omit networking interfaces
2384 --disable-regex omit regular expression interfaces
2386 These are likely to become separate modules some day.
2388 ** New configure option --enable-debug-freelist
2390 This enables a debugging version of SCM_NEWCELL(), and also registers
2391 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2393 Configure with the --enable-debug-freelist option to enable
2394 the gc-set-debug-check-freelist! primitive, and then use:
2396 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2397 (gc-set-debug-check-freelist! #f) # turn off checking
2399 Checking of the freelist forces a traversal of the freelist and
2400 a garbage collection before each allocation of a cell. This can
2401 slow down the interpreter dramatically, so the setter should be used to
2402 turn on this extra processing only when necessary.
2404 ** New configure option --enable-debug-malloc
2406 Include code for debugging of calls to scm_must_malloc/realloc/free.
2410 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2411 2. objects reallocated by scm_must_realloc has been allocated by
2413 3. reallocated objects are reallocated with the same what string
2415 But, most importantly, it records the number of allocated objects of
2416 each kind. This is useful when searching for memory leaks.
2418 A Guile compiled with this option provides the primitive
2419 `malloc-stats' which returns an alist with pairs of kind and the
2420 number of objects of that kind.
2422 ** All includes are now referenced relative to the root directory
2424 Since some users have had problems with mixups between Guile and
2425 system headers, we have decided to always refer to Guile headers via
2426 their parent directories. This essentially creates a "private name
2427 space" for Guile headers. This means that the compiler only is given
2428 -I options for the root build and root source directory.
2430 ** Header files kw.h and genio.h have been removed.
2432 ** The module (ice-9 getopt-gnu-style) has been removed.
2434 ** New module (ice-9 documentation)
2436 Implements the interface to documentation strings associated with
2439 ** New module (ice-9 time)
2441 Provides a macro `time', which displays execution time of a given form.
2443 ** New module (ice-9 history)
2445 Loading this module enables value history in the repl.
2447 * Changes to the stand-alone interpreter
2449 ** New command line option --debug
2451 Start Guile with debugging evaluator and backtraces enabled.
2453 This is useful when debugging your .guile init file or scripts.
2455 ** New help facility
2457 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2458 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2459 (help 'NAME) gives documentation for NAME, even if it is not an object
2460 (help ,EXPR) gives documentation for object returned by EXPR
2461 (help (my module)) gives module commentary for `(my module)'
2462 (help) gives this text
2464 `help' searches among bindings exported from loaded modules, while
2465 `apropos' searches among bindings visible from the "current" module.
2467 Examples: (help help)
2469 (help "output-string")
2471 ** `help' and `apropos' now prints full module names
2473 ** Dynamic linking now uses libltdl from the libtool package.
2475 The old system dependent code for doing dynamic linking has been
2476 replaced with calls to the libltdl functions which do all the hairy
2479 The major improvement is that you can now directly pass libtool
2480 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2481 will be able to do the best shared library job you can get, via
2484 The way dynamic libraries are found has changed and is not really
2485 portable across platforms, probably. It is therefore recommended to
2486 use absolute filenames when possible.
2488 If you pass a filename without an extension to `dynamic-link', it will
2489 try a few appropriate ones. Thus, the most platform ignorant way is
2490 to specify a name like "libfoo", without any directories and
2493 ** Guile COOP threads are now compatible with LinuxThreads
2495 Previously, COOP threading wasn't possible in applications linked with
2496 Linux POSIX threads due to their use of the stack pointer to find the
2497 thread context. This has now been fixed with a workaround which uses
2498 the pthreads to allocate the stack.
2500 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2502 ** Positions of erring expression in scripts
2504 With version 1.3.4, the location of the erring expression in Guile
2505 scipts is no longer automatically reported. (This should have been
2506 documented before the 1.3.4 release.)
2508 You can get this information by enabling recording of positions of
2509 source expressions and running the debugging evaluator. Put this at
2510 the top of your script (or in your "site" file):
2512 (read-enable 'positions)
2513 (debug-enable 'debug)
2515 ** Backtraces in scripts
2517 It is now possible to get backtraces in scripts.
2521 (debug-enable 'debug 'backtrace)
2523 at the top of the script.
2525 (The first options enables the debugging evaluator.
2526 The second enables backtraces.)
2528 ** Part of module system symbol lookup now implemented in C
2530 The eval closure of most modules is now implemented in C. Since this
2531 was one of the bottlenecks for loading speed, Guile now loads code
2532 substantially faster than before.
2534 ** Attempting to get the value of an unbound variable now produces
2535 an exception with a key of 'unbound-variable instead of 'misc-error.
2537 ** The initial default output port is now unbuffered if it's using a
2538 tty device. Previously in this situation it was line-buffered.
2540 ** New hook: after-gc-hook
2542 after-gc-hook takes over the role of gc-thunk. This hook is run at
2543 the first SCM_TICK after a GC. (Thus, the code is run at the same
2544 point during evaluation as signal handlers.)
2546 Note that this hook should be used only for diagnostic and debugging
2547 purposes. It is not certain that it will continue to be well-defined
2548 when this hook is run in the future.
2550 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2551 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2553 ** Improvements to garbage collector
2555 Guile 1.4 has a new policy for triggering heap allocation and
2556 determining the sizes of heap segments. It fixes a number of problems
2559 1. The new policy can handle two separate pools of cells
2560 (2-word/4-word) better. (The old policy would run wild, allocating
2561 more and more memory for certain programs.)
2563 2. The old code would sometimes allocate far too much heap so that the
2564 Guile process became gigantic. The new code avoids this.
2566 3. The old code would sometimes allocate too little so that few cells
2567 were freed at GC so that, in turn, too much time was spent in GC.
2569 4. The old code would often trigger heap allocation several times in a
2570 row. (The new scheme predicts how large the segments needs to be
2571 in order not to need further allocation.)
2573 All in all, the new GC policy will make larger applications more
2576 The new GC scheme also is prepared for POSIX threading. Threads can
2577 allocate private pools of cells ("clusters") with just a single
2578 function call. Allocation of single cells from such a cluster can
2579 then proceed without any need of inter-thread synchronization.
2581 ** New environment variables controlling GC parameters
2583 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2586 Allocation of 2-word cell heaps:
2588 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2591 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2592 GC in percent of total heap size
2595 Allocation of 4-word cell heaps
2596 (used for real numbers and misc other objects):
2598 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2600 (See entry "Way for application to customize GC parameters" under
2601 section "Changes to the scm_ interface" below.)
2603 ** Guile now implements reals using 4-word cells
2605 This speeds up computation with reals. (They were earlier allocated
2606 with `malloc'.) There is still some room for optimizations, however.
2608 ** Some further steps toward POSIX thread support have been taken
2610 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2611 don't have much effect any longer, and many of them will be removed in
2615 are only handled at the top of the evaluator loop, immediately after
2616 I/O, and in scm_equalp.
2618 *** The GC can allocate thread private pools of pairs.
2620 * Changes to Scheme functions and syntax
2622 ** close-input-port and close-output-port are now R5RS
2624 These procedures have been turned into primitives and have R5RS behaviour.
2626 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2628 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2629 extended by the more sophisticated version in (ice-9 format)
2631 (simple-format port message . args)
2632 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2633 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2634 the escapes are replaced with corresponding members of ARGS:
2635 ~A formats using `display' and ~S formats using `write'.
2636 If DESTINATION is #t, then use the `current-output-port',
2637 if DESTINATION is #f, then return a string containing the formatted text.
2638 Does not add a trailing newline."
2640 ** string-ref: the second argument is no longer optional.
2642 ** string, list->string: no longer accept strings in their arguments,
2643 only characters, for compatibility with R5RS.
2645 ** New procedure: port-closed? PORT
2646 Returns #t if PORT is closed or #f if it is open.
2648 ** Deprecated: list*
2650 The list* functionality is now provided by cons* (SRFI-1 compliant)
2652 ** New procedure: cons* ARG1 ARG2 ... ARGn
2654 Like `list', but the last arg provides the tail of the constructed list,
2655 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2657 Requires at least one argument. If given one argument, that argument
2658 is returned as result.
2660 This function is called `list*' in some other Schemes and in Common LISP.
2662 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2664 ** New procedure: object-documentation OBJECT
2666 Returns the documentation string associated with OBJECT. The
2667 procedure uses a caching mechanism so that subsequent lookups are
2670 Exported by (ice-9 documentation).
2672 ** module-name now returns full names of modules
2674 Previously, only the last part of the name was returned (`session' for
2675 `(ice-9 session)'). Ex: `(ice-9 session)'.
2677 * Changes to the gh_ interface
2679 ** Deprecated: gh_int2scmb
2681 Use gh_bool2scm instead.
2683 * Changes to the scm_ interface
2685 ** Guile primitives now carry docstrings!
2687 Thanks to Greg Badros!
2689 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2691 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2692 macros and must contain a docstring that is extracted into foo.doc using a new
2693 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2695 However, a major overhaul of these macros is scheduled for the next release of
2698 ** Guile primitives use a new technique for validation of arguments
2700 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2701 the readability of argument checking.
2703 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2705 ** New macros: SCM_PACK, SCM_UNPACK
2707 Compose/decompose an SCM value.
2709 The SCM type is now treated as an abstract data type and may be defined as a
2710 long, a void* or as a struct, depending on the architecture and compile time
2711 options. This makes it easier to find several types of bugs, for example when
2712 SCM values are treated as integers without conversion. Values of the SCM type
2713 should be treated as "atomic" values. These macros are used when
2714 composing/decomposing an SCM value, either because you want to access
2715 individual bits, or because you want to treat it as an integer value.
2717 E.g., in order to set bit 7 in an SCM value x, use the expression
2719 SCM_PACK (SCM_UNPACK (x) | 0x80)
2721 ** The name property of hooks is deprecated.
2722 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2724 You can emulate this feature by using object properties.
2726 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2727 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2730 These macros will be removed in a future release of Guile.
2732 ** The following types, functions and macros from numbers.h are deprecated:
2733 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2734 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2736 ** Port internals: the rw_random variable in the scm_port structure
2737 must be set to non-zero in any random access port. In recent Guile
2738 releases it was only set for bidirectional random-access ports.
2740 ** Port internals: the seek ptob procedure is now responsible for
2741 resetting the buffers if required. The change was made so that in the
2742 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2743 the fport and strport ptobs can avoid resetting the buffers,
2744 in particular to avoid discarding unread chars. An existing port
2745 type can be fixed by adding something like the following to the
2746 beginning of the ptob seek procedure:
2748 if (pt->rw_active == SCM_PORT_READ)
2749 scm_end_input (object);
2750 else if (pt->rw_active == SCM_PORT_WRITE)
2751 ptob->flush (object);
2753 although to actually avoid resetting the buffers and discard unread
2754 chars requires further hacking that depends on the characteristics
2757 ** Deprecated functions: scm_fseek, scm_tag
2759 These functions are no longer used and will be removed in a future version.
2761 ** The scm_sysmissing procedure is no longer used in libguile.
2762 Unless it turns out to be unexpectedly useful to somebody, it will be
2763 removed in a future version.
2765 ** The format of error message strings has changed
2767 The two C procedures: scm_display_error and scm_error, as well as the
2768 primitive `scm-error', now use scm_simple_format to do their work.
2769 This means that the message strings of all code must be updated to use
2770 ~A where %s was used before, and ~S where %S was used before.
2772 During the period when there still are a lot of old Guiles out there,
2773 you might want to support both old and new versions of Guile.
2775 There are basically two methods to achieve this. Both methods use
2778 AC_CHECK_FUNCS(scm_simple_format)
2780 in your configure.in.
2782 Method 1: Use the string concatenation features of ANSI C's
2787 #ifdef HAVE_SCM_SIMPLE_FORMAT
2793 Then represent each of your error messages using a preprocessor macro:
2795 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2799 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2800 (define make-message string-append)
2802 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2804 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2808 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2813 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2817 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2819 Don't use the functions coop_mutex_init and
2820 coop_condition_variable_init. They will change.
2822 Use scm_mutex_init and scm_cond_init instead.
2824 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2825 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2826 COND, as `scm_cond_wait' does, but it also bounds the duration
2827 of the wait. If COND has not been signaled before time ABSTIME,
2828 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2829 returns the error code `ETIMEDOUT'.
2831 The ABSTIME parameter specifies an absolute time, with the same
2832 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2833 to 00:00:00 GMT, January 1, 1970.
2835 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2836 `scm_cond_broadcast' restarts all the threads that are waiting
2837 on the condition variable COND. Nothing happens if no threads are
2840 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2841 `scm_key_create' allocates a new TSD key. The key is stored in
2842 the location pointed to by KEY. There is no limit on the number
2843 of keys allocated at a given time. The value initially associated
2844 with the returned key is `NULL' in all currently executing threads.
2846 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2847 function associated with the key. When a thread terminates,
2848 DESTR_FUNCTION is called on the value associated with the key in
2849 that thread. The DESTR_FUNCTION is not called if a key is deleted
2850 with `scm_key_delete' or a value is changed with
2851 `scm_setspecific'. The order in which destructor functions are
2852 called at thread termination time is unspecified.
2854 Destructors are not yet implemented.
2856 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2857 `scm_setspecific' changes the value associated with KEY in the
2858 calling thread, storing the given POINTER instead.
2860 ** New function: scm_getspecific (scm_key_t KEY)
2861 `scm_getspecific' returns the value currently associated with
2862 KEY in the calling thread.
2864 ** New function: scm_key_delete (scm_key_t KEY)
2865 `scm_key_delete' deallocates a TSD key. It does not check
2866 whether non-`NULL' values are associated with that key in the
2867 currently executing threads, nor call the destructor function
2868 associated with the key.
2870 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2872 Initialize a C level hook HOOK with associated HOOK_DATA and type
2873 TYPE. (See scm_c_hook_run ().)
2875 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2877 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2878 is true, add it last, otherwise first. The same FUNC can be added
2879 multiple times if FUNC_DATA differ and vice versa.
2881 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2883 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2884 function is only removed if both FUNC and FUNC_DATA matches.
2886 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2888 Run hook HOOK passing DATA to the hook functions.
2890 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2891 returned is undefined.
2893 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2894 returns a non-NULL value. This value is returned as the result of
2895 scm_c_hook_run. If all functions return NULL, NULL is returned.
2897 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2898 returns a NULL value, and NULL is returned. If all functions returns
2899 a non-NULL value, the last value is returned.
2901 ** New C level GC hooks
2903 Five new C level hooks has been added to the garbage collector.
2905 scm_before_gc_c_hook
2908 are run before locking and after unlocking the heap. The system is
2909 thus in a mode where evaluation can take place. (Except that
2910 scm_before_gc_c_hook must not allocate new cells.)
2912 scm_before_mark_c_hook
2913 scm_before_sweep_c_hook
2914 scm_after_sweep_c_hook
2916 are run when the heap is locked. These are intended for extension of
2917 the GC in a modular fashion. Examples are the weaks and guardians
2920 ** Way for application to customize GC parameters
2922 The application can set up other default values for the GC heap
2923 allocation parameters
2925 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
2926 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
2927 GUILE_MAX_SEGMENT_SIZE,
2931 scm_default_init_heap_size_1, scm_default_min_yield_1,
2932 scm_default_init_heap_size_2, scm_default_min_yield_2,
2933 scm_default_max_segment_size
2935 respectively before callong scm_boot_guile.
2937 (See entry "New environment variables ..." in section
2938 "Changes to the stand-alone interpreter" above.)
2940 ** scm_protect_object/scm_unprotect_object now nest
2942 This means that you can call scm_protect_object multiple times on an
2943 object and count on the object being protected until
2944 scm_unprotect_object has been call the same number of times.
2946 The functions also have better time complexity.
2948 Still, it is usually possible to structure the application in a way
2949 that you don't need to use these functions. For example, if you use a
2950 protected standard Guile list to keep track of live objects rather
2951 than some custom data type, objects will die a natural death when they
2952 are no longer needed.
2954 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
2956 Guile does not provide the float representation for inexact real numbers any
2957 more. Now, only doubles are used to represent inexact real numbers. Further,
2958 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
2959 and scm_tc16_complex, respectively.
2961 ** Removed deprecated type scm_smobfuns
2963 ** Removed deprecated function scm_newsmob
2965 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
2967 There is an ongoing discussion among the developers whether to
2968 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
2969 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
2970 until this issue has been settled.
2972 ** Removed deprecated type tag scm_tc16_kw
2974 ** Added type tag scm_tc16_keyword
2976 (This was introduced already in release 1.3.4 but was not documented
2979 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
2981 * Changes to system call interfaces:
2983 ** The "select" procedure now tests port buffers for the ability to
2984 provide input or accept output. Previously only the underlying file
2985 descriptors were checked.
2987 ** New variable PIPE_BUF: the maximum number of bytes that can be
2988 atomically written to a pipe.
2990 ** If a facility is not available on the system when Guile is
2991 compiled, the corresponding primitive procedure will not be defined.
2992 Previously it would have been defined but would throw a system-error
2993 exception if called. Exception handlers which catch this case may
2994 need minor modification: an error will be thrown with key
2995 'unbound-variable instead of 'system-error. Alternatively it's
2996 now possible to use `defined?' to check whether the facility is
2999 ** Procedures which depend on the timezone should now give the correct
3000 result on systems which cache the TZ environment variable, even if TZ
3001 is changed without calling tzset.
3003 * Changes to the networking interfaces:
3005 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3006 long integers between network and host format. For now, it's not
3007 particularly convenient to do this kind of thing, but consider:
3009 (define write-network-long
3010 (lambda (value port)
3011 (let ((v (make-uniform-vector 1 1 0)))
3012 (uniform-vector-set! v 0 (htonl value))
3013 (uniform-vector-write v port))))
3015 (define read-network-long
3017 (let ((v (make-uniform-vector 1 1 0)))
3018 (uniform-vector-read! v port)
3019 (ntohl (uniform-vector-ref v 0)))))
3021 ** If inet-aton fails, it now throws an error with key 'misc-error
3022 instead of 'system-error, since errno is not relevant.
3024 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3025 specific keys instead of 'system-error. The latter is inappropriate
3026 since errno will not have been set. The keys are:
3027 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3029 ** sethostent, setnetent, setprotoent, setservent: now take an
3030 optional argument STAYOPEN, which specifies whether the database
3031 remains open after a database entry is accessed randomly (e.g., using
3032 gethostbyname for the hosts database.) The default is #f. Previously
3036 Changes since Guile 1.3.2:
3038 * Changes to the stand-alone interpreter
3042 An initial version of the Guile debugger written by Chris Hanson has
3043 been added. The debugger is still under development but is included
3044 in the distribution anyway since it is already quite useful.
3050 after an error to enter the debugger. Type `help' inside the debugger
3051 for a description of available commands.
3053 If you prefer to have stack frames numbered and printed in
3054 anti-chronological order and prefer up in the stack to be down on the
3055 screen as is the case in gdb, you can put
3057 (debug-enable 'backwards)
3059 in your .guile startup file. (However, this means that Guile can't
3060 use indentation to indicate stack level.)
3062 The debugger is autoloaded into Guile at the first use.
3064 ** Further enhancements to backtraces
3066 There is a new debug option `width' which controls the maximum width
3067 on the screen of printed stack frames. Fancy printing parameters
3068 ("level" and "length" as in Common LISP) are adaptively adjusted for
3069 each stack frame to give maximum information while still fitting
3070 within the bounds. If the stack frame can't be made to fit by
3071 adjusting parameters, it is simply cut off at the end. This is marked
3074 ** Some modules are now only loaded when the repl is started
3076 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3077 regex) are now loaded into (guile-user) only if the repl has been
3078 started. The effect is that the startup time for scripts has been
3079 reduced to 30% of what it was previously.
3081 Correctly written scripts load the modules they require at the top of
3082 the file and should not be affected by this change.
3084 ** Hooks are now represented as smobs
3086 * Changes to Scheme functions and syntax
3088 ** Readline support has changed again.
3090 The old (readline-activator) module is gone. Use (ice-9 readline)
3091 instead, which now contains all readline functionality. So the code
3092 to activate readline is now
3094 (use-modules (ice-9 readline))
3097 This should work at any time, including from the guile prompt.
3099 To avoid confusion about the terms of Guile's license, please only
3100 enable readline for your personal use; please don't make it the
3101 default for others. Here is why we make this rather odd-sounding
3104 Guile is normally licensed under a weakened form of the GNU General
3105 Public License, which allows you to link code with Guile without
3106 placing that code under the GPL. This exception is important to some
3109 However, since readline is distributed under the GNU General Public
3110 License, when you link Guile with readline, either statically or
3111 dynamically, you effectively change Guile's license to the strict GPL.
3112 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3113 which are normally permitted become forbidden. This is a rather
3114 non-obvious consequence of the licensing terms.
3116 So, to make sure things remain clear, please let people choose for
3117 themselves whether to link GPL'd libraries like readline with Guile.
3119 ** regexp-substitute/global has changed slightly, but incompatibly.
3121 If you include a function in the item list, the string of the match
3122 object it receives is the same string passed to
3123 regexp-substitute/global, not some suffix of that string.
3124 Correspondingly, the match's positions are relative to the entire
3125 string, not the suffix.
3127 If the regexp can match the empty string, the way matches are chosen
3128 from the string has changed. regexp-substitute/global recognizes the
3129 same set of matches that list-matches does; see below.
3131 ** New function: list-matches REGEXP STRING [FLAGS]
3133 Return a list of match objects, one for every non-overlapping, maximal
3134 match of REGEXP in STRING. The matches appear in left-to-right order.
3135 list-matches only reports matches of the empty string if there are no
3136 other matches which begin on, end at, or include the empty match's
3139 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3141 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3143 For each match of REGEXP in STRING, apply PROC to the match object,
3144 and the last value PROC returned, or INIT for the first call. Return
3145 the last value returned by PROC. We apply PROC to the matches as they
3146 appear from left to right.
3148 This function recognizes matches according to the same criteria as
3151 Thus, you could define list-matches like this:
3153 (define (list-matches regexp string . flags)
3154 (reverse! (apply fold-matches regexp string '() cons flags)))
3156 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3160 *** New function: hook? OBJ
3162 Return #t if OBJ is a hook, otherwise #f.
3164 *** New function: make-hook-with-name NAME [ARITY]
3166 Return a hook with name NAME and arity ARITY. The default value for
3167 ARITY is 0. The only effect of NAME is that it will appear when the
3168 hook object is printed to ease debugging.
3170 *** New function: hook-empty? HOOK
3172 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3174 *** New function: hook->list HOOK
3176 Return a list of the procedures that are called when run-hook is
3179 ** `map' signals an error if its argument lists are not all the same length.
3181 This is the behavior required by R5RS, so this change is really a bug
3182 fix. But it seems to affect a lot of people's code, so we're
3183 mentioning it here anyway.
3185 ** Print-state handling has been made more transparent
3187 Under certain circumstances, ports are represented as a port with an
3188 associated print state. Earlier, this pair was represented as a pair
3189 (see "Some magic has been added to the printer" below). It is now
3190 indistinguishable (almost; see `get-print-state') from a port on the
3193 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3195 Return a new port with the associated print state PRINT-STATE.
3197 *** New function: get-print-state OUTPUT-PORT
3199 Return the print state associated with this port if it exists,
3200 otherwise return #f.
3202 *** New function: directory-stream? OBJECT
3204 Returns true iff OBJECT is a directory stream --- the sort of object
3205 returned by `opendir'.
3207 ** New function: using-readline?
3209 Return #t if readline is in use in the current repl.
3211 ** structs will be removed in 1.4
3213 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3214 and use GOOPS objects as the fundamental record type.
3216 * Changes to the scm_ interface
3218 ** structs will be removed in 1.4
3220 The entire current struct interface (struct.c, struct.h) will be
3221 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3222 GOOPS objects as the fundamental record type.
3224 ** The internal representation of subr's has changed
3226 Instead of giving a hint to the subr name, the CAR field of the subr
3227 now contains an index to a subr entry in scm_subr_table.
3229 *** New variable: scm_subr_table
3231 An array of subr entries. A subr entry contains the name, properties
3232 and documentation associated with the subr. The properties and
3233 documentation slots are not yet used.
3235 ** A new scheme for "forwarding" calls to a builtin to a generic function
3237 It is now possible to extend the functionality of some Guile
3238 primitives by letting them defer a call to a GOOPS generic function on
3239 argument mismatch. This means that there is no loss of efficiency in
3244 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3245 (define-method + ((x <string>) (y <string>))
3246 (string-append x y))
3248 + will still be as efficient as usual in numerical calculations, but
3249 can also be used for concatenating strings.
3251 Who will be the first one to extend Guile's numerical tower to
3252 rationals? :) [OK, there a few other things to fix before this can
3253 be made in a clean way.]
3255 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3257 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3259 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3261 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3262 a variable GENERIC which can be used by the dispatch macros below.
3264 [This is experimental code which may change soon.]
3266 *** New macros for forwarding control to a generic on arg type error
3268 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3270 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3272 These correspond to the scm_wta function call, and have the same
3273 behaviour until the user has called the GOOPS primitive
3274 `enable-primitive-generic!'. After that, these macros will apply the
3275 generic function GENERIC to the argument(s) instead of calling
3278 [This is experimental code which may change soon.]
3280 *** New macros for argument testing with generic dispatch
3282 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3284 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3286 These correspond to the SCM_ASSERT macro, but will defer control to
3287 GENERIC on error after `enable-primitive-generic!' has been called.
3289 [This is experimental code which may change soon.]
3291 ** New function: SCM scm_eval_body (SCM body, SCM env)
3293 Evaluates the body of a special form.
3295 ** The internal representation of struct's has changed
3297 Previously, four slots were allocated for the procedure(s) of entities
3298 and operators. The motivation for this representation had to do with
3299 the structure of the evaluator, the wish to support tail-recursive
3300 generic functions, and efficiency. Since the generic function
3301 dispatch mechanism has changed, there is no longer a need for such an
3302 expensive representation, and the representation has been simplified.
3304 This should not make any difference for most users.
3306 ** GOOPS support has been cleaned up.
3308 Some code has been moved from eval.c to objects.c and code in both of
3309 these compilation units has been cleaned up and better structured.
3311 *** New functions for applying generic functions
3313 New function: SCM scm_apply_generic (GENERIC, ARGS)
3314 New function: SCM scm_call_generic_0 (GENERIC)
3315 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3316 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3317 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3319 ** Deprecated function: scm_make_named_hook
3321 It is now replaced by:
3323 ** New function: SCM scm_create_hook (const char *name, int arity)
3325 Creates a hook in the same way as make-hook above but also
3326 binds a variable named NAME to it.
3328 This is the typical way of creating a hook from C code.
3330 Currently, the variable is created in the "current" module.
3331 This might change when we get the new module system.
3333 [The behaviour is identical to scm_make_named_hook.]
3337 Changes since Guile 1.3:
3339 * Changes to mailing lists
3341 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3343 See the README file to find current addresses for all the Guile
3346 * Changes to the distribution
3348 ** Readline support is no longer included with Guile by default.
3350 Based on the different license terms of Guile and Readline, we
3351 concluded that Guile should not *by default* cause the linking of
3352 Readline into an application program. Readline support is now offered
3353 as a separate module, which is linked into an application only when
3354 you explicitly specify it.
3356 Although Guile is GNU software, its distribution terms add a special
3357 exception to the usual GNU General Public License (GPL). Guile's
3358 license includes a clause that allows you to link Guile with non-free
3359 programs. We add this exception so as not to put Guile at a
3360 disadvantage vis-a-vis other extensibility packages that support other
3363 In contrast, the GNU Readline library is distributed under the GNU
3364 General Public License pure and simple. This means that you may not
3365 link Readline, even dynamically, into an application unless it is
3366 distributed under a free software license that is compatible the GPL.
3368 Because of this difference in distribution terms, an application that
3369 can use Guile may not be able to use Readline. Now users will be
3370 explicitly offered two independent decisions about the use of these
3373 You can activate the readline support by issuing
3375 (use-modules (readline-activator))
3378 from your ".guile" file, for example.
3380 * Changes to the stand-alone interpreter
3382 ** All builtins now print as primitives.
3383 Previously builtin procedures not belonging to the fundamental subr
3384 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3385 Now, they print as #<primitive-procedure NAME>.
3387 ** Backtraces slightly more intelligible.
3388 gsubr-apply and macro transformer application frames no longer appear
3391 * Changes to Scheme functions and syntax
3393 ** Guile now correctly handles internal defines by rewriting them into
3394 their equivalent letrec. Previously, internal defines would
3395 incrementally add to the innermost environment, without checking
3396 whether the restrictions specified in RnRS were met. This lead to the
3397 correct behaviour when these restriction actually were met, but didn't
3398 catch all illegal uses. Such an illegal use could lead to crashes of
3399 the Guile interpreter or or other unwanted results. An example of
3400 incorrect internal defines that made Guile behave erratically:
3412 The problem with this example is that the definition of `c' uses the
3413 value of `b' directly. This confuses the meoization machine of Guile
3414 so that the second call of `b' (this time in a larger environment that
3415 also contains bindings for `c' and `d') refers to the binding of `c'
3416 instead of `a'. You could also make Guile crash with a variation on
3421 (define (b flag) (if flag a 1))
3422 (define c (1+ (b flag)))
3430 From now on, Guile will issue an `Unbound variable: b' error message
3435 A hook contains a list of functions which should be called on
3436 particular occasions in an existing program. Hooks are used for
3439 A window manager might have a hook before-window-map-hook. The window
3440 manager uses the function run-hooks to call all functions stored in
3441 before-window-map-hook each time a window is mapped. The user can
3442 store functions in the hook using add-hook!.
3444 In Guile, hooks are first class objects.
3446 *** New function: make-hook [N_ARGS]
3448 Return a hook for hook functions which can take N_ARGS arguments.
3449 The default value for N_ARGS is 0.
3451 (See also scm_make_named_hook below.)
3453 *** New function: add-hook! HOOK PROC [APPEND_P]
3455 Put PROC at the beginning of the list of functions stored in HOOK.
3456 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3458 PROC must be able to take the number of arguments specified when the
3461 If PROC already exists in HOOK, then remove it first.
3463 *** New function: remove-hook! HOOK PROC
3465 Remove PROC from the list of functions in HOOK.
3467 *** New function: reset-hook! HOOK
3469 Clear the list of hook functions stored in HOOK.
3471 *** New function: run-hook HOOK ARG1 ...
3473 Run all hook functions stored in HOOK with arguments ARG1 ... .
3474 The number of arguments supplied must correspond to the number given
3475 when the hook was created.
3477 ** The function `dynamic-link' now takes optional keyword arguments.
3478 The only keyword argument that is currently defined is `:global
3479 BOOL'. With it, you can control whether the shared library will be
3480 linked in global mode or not. In global mode, the symbols from the
3481 linked library can be used to resolve references from other
3482 dynamically linked libraries. In non-global mode, the linked
3483 library is essentially invisible and can only be accessed via
3484 `dynamic-func', etc. The default is now to link in global mode.
3485 Previously, the default has been non-global mode.
3487 The `#:global' keyword is only effective on platforms that support
3488 the dlopen family of functions.
3490 ** New function `provided?'
3492 - Function: provided? FEATURE
3493 Return true iff FEATURE is supported by this installation of
3494 Guile. FEATURE must be a symbol naming a feature; the global
3495 variable `*features*' is a list of available features.
3497 ** Changes to the module (ice-9 expect):
3499 *** The expect-strings macro now matches `$' in a regular expression
3500 only at a line-break or end-of-file by default. Previously it would
3501 match the end of the string accumulated so far. The old behaviour
3502 can be obtained by setting the variable `expect-strings-exec-flags'
3505 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3506 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3507 in a regular expression will still match before a line-break or
3508 end-of-file. The default is `regexp/noteol'.
3510 *** The expect-strings macro now uses a variable
3511 `expect-strings-compile-flags' for the flags to be supplied to
3512 `make-regexp'. The default is `regexp/newline', which was previously
3515 *** The expect macro now supplies two arguments to a match procedure:
3516 the current accumulated string and a flag to indicate whether
3517 end-of-file has been reached. Previously only the string was supplied.
3518 If end-of-file is reached, the match procedure will be called an
3519 additional time with the same accumulated string as the previous call
3520 but with the flag set.
3522 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3524 This code, and the documentation for it that appears here, was
3525 borrowed from SLIB, with minor adaptations for Guile.
3527 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3528 An almost complete implementation of Common LISP format description
3529 according to the CL reference book `Common LISP' from Guy L.
3530 Steele, Digital Press. Backward compatible to most of the
3531 available Scheme format implementations.
3533 Returns `#t', `#f' or a string; has side effect of printing
3534 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3535 to the current output port and `#t' is returned. If DESTINATION
3536 is `#f', a formatted string is returned as the result of the call.
3537 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3538 format string; FORMAT-STRING is then the first argument and the
3539 output is returned as a string. If DESTINATION is a number, the
3540 output is to the current error port if available by the
3541 implementation. Otherwise DESTINATION must be an output port and
3544 FORMAT-STRING must be a string. In case of a formatting error
3545 format returns `#f' and prints a message on the current output or
3546 error port. Characters are output as if the string were output by
3547 the `display' function with the exception of those prefixed by a
3548 tilde (~). For a detailed description of the FORMAT-STRING syntax
3549 please consult a Common LISP format reference manual. For a test
3550 suite to verify this format implementation load `formatst.scm'.
3551 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3553 Note: `format' is not reentrant, i.e. only one `format'-call may
3554 be executed at a time.
3557 *** Format Specification (Format version 3.0)
3559 Please consult a Common LISP format reference manual for a detailed
3560 description of the format string syntax. For a demonstration of the
3561 implemented directives see `formatst.scm'.
3563 This implementation supports directive parameters and modifiers (`:'
3564 and `@' characters). Multiple parameters must be separated by a comma
3565 (`,'). Parameters can be numerical parameters (positive or negative),
3566 character parameters (prefixed by a quote character (`''), variable
3567 parameters (`v'), number of rest arguments parameter (`#'), empty and
3568 default parameters. Directive characters are case independent. The
3569 general form of a directive is:
3571 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3573 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3575 *** Implemented CL Format Control Directives
3577 Documentation syntax: Uppercase characters represent the
3578 corresponding control directive characters. Lowercase characters
3579 represent control directive parameter descriptions.
3582 Any (print as `display' does).
3586 `~MINCOL,COLINC,MINPAD,PADCHARA'
3590 S-expression (print as `write' does).
3594 `~MINCOL,COLINC,MINPAD,PADCHARS'
3600 print number sign always.
3603 print comma separated.
3605 `~MINCOL,PADCHAR,COMMACHARD'
3611 print number sign always.
3614 print comma separated.
3616 `~MINCOL,PADCHAR,COMMACHARX'
3622 print number sign always.
3625 print comma separated.
3627 `~MINCOL,PADCHAR,COMMACHARO'
3633 print number sign always.
3636 print comma separated.
3638 `~MINCOL,PADCHAR,COMMACHARB'
3643 `~N,MINCOL,PADCHAR,COMMACHARR'
3647 print a number as a Roman numeral.
3650 print a number as an "old fashioned" Roman numeral.
3653 print a number as an ordinal English number.
3656 print a number as a cardinal English number.
3661 prints `y' and `ies'.
3664 as `~P but jumps 1 argument backward.'
3667 as `~@P but jumps 1 argument backward.'
3672 prints a character as the reader can understand it (i.e. `#\'
3676 prints a character as emacs does (eg. `^C' for ASCII 03).
3679 Fixed-format floating-point (prints a flonum like MMM.NNN).
3680 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3682 If the number is positive a plus sign is printed.
3685 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3686 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3688 If the number is positive a plus sign is printed.
3691 General floating-point (prints a flonum either fixed or
3693 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3695 If the number is positive a plus sign is printed.
3698 Dollars floating-point (prints a flonum in fixed with signs
3700 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3702 If the number is positive a plus sign is printed.
3705 A sign is always printed and appears before the padding.
3708 The sign appears before the padding.
3716 print newline if not at the beginning of the output line.
3718 prints `~&' and then N-1 newlines.
3723 print N page separators.
3733 newline is ignored, white space left.
3736 newline is left, white space ignored.
3741 relative tabulation.
3747 Indirection (expects indirect arguments as a list).
3749 extracts indirect arguments from format arguments.
3752 Case conversion (converts by `string-downcase').
3754 converts by `string-capitalize'.
3757 converts by `string-capitalize-first'.
3760 converts by `string-upcase'.
3763 Argument Jumping (jumps 1 argument forward).
3765 jumps N arguments forward.
3768 jumps 1 argument backward.
3771 jumps N arguments backward.
3774 jumps to the 0th argument.
3777 jumps to the Nth argument (beginning from 0)
3779 `~[STR0~;STR1~;...~;STRN~]'
3780 Conditional Expression (numerical clause conditional).
3782 take argument from N.
3785 true test conditional.
3788 if-else-then conditional.
3794 default clause follows.
3797 Iteration (args come from the next argument (a list)).
3799 at most N iterations.
3802 args from next arg (a list of lists).
3805 args from the rest of arguments.
3808 args from the rest args (lists).
3819 aborts if N <= M <= K
3821 *** Not Implemented CL Format Control Directives
3824 print `#f' as an empty list (see below).
3827 print `#f' as an empty list (see below).
3833 (sorry I don't understand its semantics completely)
3835 *** Extended, Replaced and Additional Control Directives
3837 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3838 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3839 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3840 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3841 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3842 COMMAWIDTH is the number of characters between two comma
3846 print a R4RS complex number as `~F~@Fi' with passed parameters for
3850 Pretty print formatting of an argument for scheme code lists.
3856 Flushes the output if format DESTINATION is a port.
3859 Print a `#\space' character
3861 print N `#\space' characters.
3864 Print a `#\tab' character
3866 print N `#\tab' characters.
3869 Takes N as an integer representation for a character. No arguments
3870 are consumed. N is converted to a character by `integer->char'. N
3871 must be a positive decimal number.
3874 Print out readproof. Prints out internal objects represented as
3875 `#<...>' as strings `"#<...>"' so that the format output can always
3876 be processed by `read'.
3879 Print out readproof. Prints out internal objects represented as
3880 `#<...>' as strings `"#<...>"' so that the format output can always
3881 be processed by `read'.
3884 Prints information and a copyright notice on the format
3887 prints format version.
3890 may also print number strings, i.e. passing a number as a string
3891 and format it accordingly.
3893 *** Configuration Variables
3895 The format module exports some configuration variables to suit the
3896 systems and users needs. There should be no modification necessary for
3897 the configuration that comes with Guile. Format detects automatically
3898 if the running scheme system implements floating point numbers and
3901 format:symbol-case-conv
3902 Symbols are converted by `symbol->string' so the case type of the
3903 printed symbols is implementation dependent.
3904 `format:symbol-case-conv' is a one arg closure which is either
3905 `#f' (no conversion), `string-upcase', `string-downcase' or
3906 `string-capitalize'. (default `#f')
3908 format:iobj-case-conv
3909 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3910 implementation internal objects. (default `#f')
3913 The character prefixing the exponent value in `~E' printing.
3916 *** Compatibility With Other Format Implementations
3922 Downward compatible except for padding support and `~A', `~S',
3923 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
3924 `printf' padding support which is completely replaced by the CL
3925 `format' padding style.
3928 Downward compatible except for `~', which is not documented
3929 (ignores all characters inside the format string up to a newline
3930 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
3931 numerical and variable parameters and `:/@' modifiers in the CL
3935 Downward compatible except for `~A' and `~S' which print in
3936 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
3937 directive parameters or modifiers)).
3940 Downward compatible except for an optional destination parameter:
3941 S2C accepts a format call without a destination which returns a
3942 formatted string. This is equivalent to a #f destination in S2C.
3943 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
3944 parameters or modifiers)).
3947 ** Changes to string-handling functions.
3949 These functions were added to support the (ice-9 format) module, above.
3951 *** New function: string-upcase STRING
3952 *** New function: string-downcase STRING
3954 These are non-destructive versions of the existing string-upcase! and
3955 string-downcase! functions.
3957 *** New function: string-capitalize! STRING
3958 *** New function: string-capitalize STRING
3960 These functions convert the first letter of each word in the string to
3963 (string-capitalize "howdy there")
3966 As with the other functions, string-capitalize! modifies the string in
3967 place, while string-capitalize returns a modified copy of its argument.
3969 *** New function: string-ci->symbol STRING
3971 Return a symbol whose name is STRING, but having the same case as if
3972 the symbol had be read by `read'.
3974 Guile can be configured to be sensitive or insensitive to case
3975 differences in Scheme identifiers. If Guile is case-insensitive, all
3976 symbols are converted to lower case on input. The `string-ci->symbol'
3977 function returns a symbol whose name in STRING, transformed as Guile
3978 would if STRING were input.
3980 *** New function: substring-move! STRING1 START END STRING2 START
3982 Copy the substring of STRING1 from START (inclusive) to END
3983 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
3984 string, and the source and destination areas may overlap; in all
3985 cases, the function behaves as if all the characters were copied
3988 *** Extended functions: substring-move-left! substring-move-right!
3990 These functions now correctly copy arbitrarily overlapping substrings;
3991 they are both synonyms for substring-move!.
3994 ** New module (ice-9 getopt-long), with the function `getopt-long'.
3996 getopt-long is a function for parsing command-line arguments in a
3997 manner consistent with other GNU programs.
3999 (getopt-long ARGS GRAMMAR)
4000 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4002 ARGS should be a list of strings. Its first element should be the
4003 name of the program; subsequent elements should be the arguments
4004 that were passed to the program on the command line. The
4005 `program-arguments' procedure returns a list of this form.
4007 GRAMMAR is a list of the form:
4008 ((OPTION (PROPERTY VALUE) ...) ...)
4010 Each OPTION should be a symbol. `getopt-long' will accept a
4011 command-line option named `--OPTION'.
4012 Each option can have the following (PROPERTY VALUE) pairs:
4014 (single-char CHAR) --- Accept `-CHAR' as a single-character
4015 equivalent to `--OPTION'. This is how to specify traditional
4017 (required? BOOL) --- If BOOL is true, the option is required.
4018 getopt-long will raise an error if it is not found in ARGS.
4019 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4020 it is #f, it does not; and if it is the symbol
4021 `optional', the option may appear in ARGS with or
4023 (predicate FUNC) --- If the option accepts a value (i.e. you
4024 specified `(value #t)' for this option), then getopt
4025 will apply FUNC to the value, and throw an exception
4026 if it returns #f. FUNC should be a procedure which
4027 accepts a string and returns a boolean value; you may
4028 need to use quasiquotes to get it into GRAMMAR.
4030 The (PROPERTY VALUE) pairs may occur in any order, but each
4031 property may occur only once. By default, options do not have
4032 single-character equivalents, are not required, and do not take
4035 In ARGS, single-character options may be combined, in the usual
4036 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4037 accepts values, then it must be the last option in the
4038 combination; the value is the next argument. So, for example, using
4039 the following grammar:
4040 ((apples (single-char #\a))
4041 (blimps (single-char #\b) (value #t))
4042 (catalexis (single-char #\c) (value #t)))
4043 the following argument lists would be acceptable:
4044 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4045 for "blimps" and "catalexis")
4046 ("-ab" "bang" "-c" "couth") (same)
4047 ("-ac" "couth" "-b" "bang") (same)
4048 ("-abc" "couth" "bang") (an error, since `-b' is not the
4049 last option in its combination)
4051 If an option's value is optional, then `getopt-long' decides
4052 whether it has a value by looking at what follows it in ARGS. If
4053 the next element is a string, and it does not appear to be an
4054 option itself, then that string is the option's value.
4056 The value of a long option can appear as the next element in ARGS,
4057 or it can follow the option name, separated by an `=' character.
4058 Thus, using the same grammar as above, the following argument lists
4060 ("--apples" "Braeburn" "--blimps" "Goodyear")
4061 ("--apples=Braeburn" "--blimps" "Goodyear")
4062 ("--blimps" "Goodyear" "--apples=Braeburn")
4064 If the option "--" appears in ARGS, argument parsing stops there;
4065 subsequent arguments are returned as ordinary arguments, even if
4066 they resemble options. So, in the argument list:
4067 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4068 `getopt-long' will recognize the `apples' option as having the
4069 value "Granny Smith", but it will not recognize the `blimp'
4070 option; it will return the strings "--blimp" and "Goodyear" as
4071 ordinary argument strings.
4073 The `getopt-long' function returns the parsed argument list as an
4074 assocation list, mapping option names --- the symbols from GRAMMAR
4075 --- onto their values, or #t if the option does not accept a value.
4076 Unused options do not appear in the alist.
4078 All arguments that are not the value of any option are returned
4079 as a list, associated with the empty list.
4081 `getopt-long' throws an exception if:
4082 - it finds an unrecognized option in ARGS
4083 - a required option is omitted
4084 - an option that requires an argument doesn't get one
4085 - an option that doesn't accept an argument does get one (this can
4086 only happen using the long option `--opt=value' syntax)
4087 - an option predicate fails
4092 `((lockfile-dir (required? #t)
4095 (predicate ,file-is-directory?))
4096 (verbose (required? #f)
4099 (x-includes (single-char #\x))
4100 (rnet-server (single-char #\y)
4101 (predicate ,string?))))
4103 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4104 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4106 => ((() "foo1" "-fred" "foo2" "foo3")
4107 (rnet-server . "lamprod")
4108 (x-includes . "/usr/include")
4109 (lockfile-dir . "/tmp")
4112 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4114 It will be removed in a few releases.
4116 ** New syntax: lambda*
4117 ** New syntax: define*
4118 ** New syntax: define*-public
4119 ** New syntax: defmacro*
4120 ** New syntax: defmacro*-public
4121 Guile now supports optional arguments.
4123 `lambda*', `define*', `define*-public', `defmacro*' and
4124 `defmacro*-public' are identical to the non-* versions except that
4125 they use an extended type of parameter list that has the following BNF
4126 syntax (parentheses are literal, square brackets indicate grouping,
4127 and `*', `+' and `?' have the usual meaning):
4129 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4130 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4131 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4133 ext-var-decl ::= identifier | ( identifier expression )
4135 The semantics are best illustrated with the following documentation
4136 and examples for `lambda*':
4139 lambda extended for optional and keyword arguments
4141 lambda* creates a procedure that takes optional arguments. These
4142 are specified by putting them inside brackets at the end of the
4143 paramater list, but before any dotted rest argument. For example,
4144 (lambda* (a b #&optional c d . e) '())
4145 creates a procedure with fixed arguments a and b, optional arguments c
4146 and d, and rest argument e. If the optional arguments are omitted
4147 in a call, the variables for them are unbound in the procedure. This
4148 can be checked with the bound? macro.
4150 lambda* can also take keyword arguments. For example, a procedure
4152 (lambda* (#&key xyzzy larch) '())
4153 can be called with any of the argument lists (#:xyzzy 11)
4154 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4155 are given as keywords are bound to values.
4157 Optional and keyword arguments can also be given default values
4158 which they take on when they are not present in a call, by giving a
4159 two-item list in place of an optional argument, for example in:
4160 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4161 foo is a fixed argument, bar is an optional argument with default
4162 value 42, and baz is a keyword argument with default value 73.
4163 Default value expressions are not evaluated unless they are needed
4164 and until the procedure is called.
4166 lambda* now supports two more special parameter list keywords.
4168 lambda*-defined procedures now throw an error by default if a
4169 keyword other than one of those specified is found in the actual
4170 passed arguments. However, specifying #&allow-other-keys
4171 immediately after the kyword argument declarations restores the
4172 previous behavior of ignoring unknown keywords. lambda* also now
4173 guarantees that if the same keyword is passed more than once, the
4174 last one passed is the one that takes effect. For example,
4175 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4176 #:heads 37 #:tails 42 #:heads 99)
4177 would result in (99 47) being displayed.
4179 #&rest is also now provided as a synonym for the dotted syntax rest
4180 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4181 all respects to lambda*. This is provided for more similarity to DSSSL,
4182 MIT-Scheme and Kawa among others, as well as for refugees from other
4185 Further documentation may be found in the optargs.scm file itself.
4187 The optional argument module also exports the macros `let-optional',
4188 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4189 are not documented here because they may be removed in the future, but
4190 full documentation is still available in optargs.scm.
4192 ** New syntax: and-let*
4193 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4195 Syntax: (land* (<clause> ...) <body> ...)
4196 Each <clause> should have one of the following forms:
4197 (<variable> <expression>)
4200 Each <variable> or <bound-variable> should be an identifier. Each
4201 <expression> should be a valid expression. The <body> should be a
4202 possibly empty sequence of expressions, like the <body> of a
4205 Semantics: A LAND* expression is evaluated by evaluating the
4206 <expression> or <bound-variable> of each of the <clause>s from
4207 left to right. The value of the first <expression> or
4208 <bound-variable> that evaluates to a false value is returned; the
4209 remaining <expression>s and <bound-variable>s are not evaluated.
4210 The <body> forms are evaluated iff all the <expression>s and
4211 <bound-variable>s evaluate to true values.
4213 The <expression>s and the <body> are evaluated in an environment
4214 binding each <variable> of the preceding (<variable> <expression>)
4215 clauses to the value of the <expression>. Later bindings
4216 shadow earlier bindings.
4218 Guile's and-let* macro was contributed by Michael Livshin.
4220 ** New sorting functions
4222 *** New function: sorted? SEQUENCE LESS?
4223 Returns `#t' when the sequence argument is in non-decreasing order
4224 according to LESS? (that is, there is no adjacent pair `... x y
4225 ...' for which `(less? y x)').
4227 Returns `#f' when the sequence contains at least one out-of-order
4228 pair. It is an error if the sequence is neither a list nor a
4231 *** New function: merge LIST1 LIST2 LESS?
4232 LIST1 and LIST2 are sorted lists.
4233 Returns the sorted list of all elements in LIST1 and LIST2.
4235 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4236 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4237 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4238 (Here "<" should read "comes before".)
4240 *** New procedure: merge! LIST1 LIST2 LESS?
4241 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4242 the result. If the code is compiled, and LESS? constructs no new
4243 pairs, no pairs at all will be allocated. The first pair of the
4244 result will be either the first pair of LIST1 or the first pair of
4247 *** New function: sort SEQUENCE LESS?
4248 Accepts either a list or a vector, and returns a new sequence
4249 which is sorted. The new sequence is the same type as the input.
4250 Always `(sorted? (sort sequence less?) less?)'. The original
4251 sequence is not altered in any way. The new sequence shares its
4252 elements with the old one; no elements are copied.
4254 *** New procedure: sort! SEQUENCE LESS
4255 Returns its sorted result in the original boxes. No new storage is
4256 allocated at all. Proper usage: (set! slist (sort! slist <))
4258 *** New function: stable-sort SEQUENCE LESS?
4259 Similar to `sort' but stable. That is, if "equal" elements are
4260 ordered a < b in the original sequence, they will have the same order
4263 *** New function: stable-sort! SEQUENCE LESS?
4264 Similar to `sort!' but stable.
4265 Uses temporary storage when sorting vectors.
4267 *** New functions: sort-list, sort-list!
4268 Added for compatibility with scsh.
4270 ** New built-in random number support
4272 *** New function: random N [STATE]
4273 Accepts a positive integer or real N and returns a number of the
4274 same type between zero (inclusive) and N (exclusive). The values
4275 returned have a uniform distribution.
4277 The optional argument STATE must be of the type produced by
4278 `copy-random-state' or `seed->random-state'. It defaults to the value
4279 of the variable `*random-state*'. This object is used to maintain the
4280 state of the pseudo-random-number generator and is altered as a side
4281 effect of the `random' operation.
4283 *** New variable: *random-state*
4284 Holds a data structure that encodes the internal state of the
4285 random-number generator that `random' uses by default. The nature
4286 of this data structure is implementation-dependent. It may be
4287 printed out and successfully read back in, but may or may not
4288 function correctly as a random-number state object in another
4291 *** New function: copy-random-state [STATE]
4292 Returns a new object of type suitable for use as the value of the
4293 variable `*random-state*' and as a second argument to `random'.
4294 If argument STATE is given, a copy of it is returned. Otherwise a
4295 copy of `*random-state*' is returned.
4297 *** New function: seed->random-state SEED
4298 Returns a new object of type suitable for use as the value of the
4299 variable `*random-state*' and as a second argument to `random'.
4300 SEED is a string or a number. A new state is generated and
4301 initialized using SEED.
4303 *** New function: random:uniform [STATE]
4304 Returns an uniformly distributed inexact real random number in the
4305 range between 0 and 1.
4307 *** New procedure: random:solid-sphere! VECT [STATE]
4308 Fills VECT with inexact real random numbers the sum of whose
4309 squares is less than 1.0. Thinking of VECT as coordinates in
4310 space of dimension N = `(vector-length VECT)', the coordinates are
4311 uniformly distributed within the unit N-shere. The sum of the
4312 squares of the numbers is returned. VECT can be either a vector
4313 or a uniform vector of doubles.
4315 *** New procedure: random:hollow-sphere! VECT [STATE]
4316 Fills VECT with inexact real random numbers the sum of whose squares
4317 is equal to 1.0. Thinking of VECT as coordinates in space of
4318 dimension n = `(vector-length VECT)', the coordinates are uniformly
4319 distributed over the surface of the unit n-shere. VECT can be either
4320 a vector or a uniform vector of doubles.
4322 *** New function: random:normal [STATE]
4323 Returns an inexact real in a normal distribution with mean 0 and
4324 standard deviation 1. For a normal distribution with mean M and
4325 standard deviation D use `(+ M (* D (random:normal)))'.
4327 *** New procedure: random:normal-vector! VECT [STATE]
4328 Fills VECT with inexact real random numbers which are independent and
4329 standard normally distributed (i.e., with mean 0 and variance 1).
4330 VECT can be either a vector or a uniform vector of doubles.
4332 *** New function: random:exp STATE
4333 Returns an inexact real in an exponential distribution with mean 1.
4334 For an exponential distribution with mean U use (* U (random:exp)).
4336 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4338 These functions now operate on numbers in the range of a C unsigned
4341 These functions used to operate on numbers in the range of a C signed
4342 long; however, this seems inappropriate, because Guile integers don't
4345 ** New function: make-guardian
4346 This is an implementation of guardians as described in
4347 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4348 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4349 Programming Language Design and Implementation, June 1993
4350 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4352 ** New functions: delq1!, delv1!, delete1!
4353 These procedures behave similar to delq! and friends but delete only
4354 one object if at all.
4356 ** New function: unread-string STRING PORT
4357 Unread STRING to PORT, that is, push it back onto the port so that
4358 next read operation will work on the pushed back characters.
4360 ** unread-char can now be called multiple times
4361 If unread-char is called multiple times, the unread characters will be
4362 read again in last-in first-out order.
4364 ** the procedures uniform-array-read! and uniform-array-write! now
4365 work on any kind of port, not just ports which are open on a file.
4367 ** Now 'l' in a port mode requests line buffering.
4369 ** The procedure truncate-file now works on string ports as well
4370 as file ports. If the size argument is omitted, the current
4371 file position is used.
4373 ** new procedure: seek PORT/FDES OFFSET WHENCE
4374 The arguments are the same as for the old fseek procedure, but it
4375 works on string ports as well as random-access file ports.
4377 ** the fseek procedure now works on string ports, since it has been
4378 redefined using seek.
4380 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4381 size is not supplied.
4383 ** the newline procedure no longer flushes the port if it's not
4384 line-buffered: previously it did if it was the current output port.
4386 ** open-pipe and close-pipe are no longer primitive procedures, but
4387 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4389 ** the freopen procedure has been removed.
4391 ** new procedure: drain-input PORT
4392 Drains PORT's read buffers (including any pushed-back characters)
4393 and returns the contents as a single string.
4395 ** New function: map-in-order PROC LIST1 LIST2 ...
4396 Version of `map' which guarantees that the procedure is applied to the
4397 lists in serial order.
4399 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4400 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4401 now obsolete and will go away in release 1.5.
4403 ** New syntax: collect BODY1 ...
4404 Version of `begin' which returns a list of the results of the body
4405 forms instead of the result of the last body form. In contrast to
4406 `begin', `collect' allows an empty body.
4408 ** New functions: read-history FILENAME, write-history FILENAME
4409 Read/write command line history from/to file. Returns #t on success
4410 and #f if an error occured.
4412 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4414 These procedures return a list of definitions available in the specified
4415 argument, a relative module reference. In the case of no argument,
4416 `(current-module)' is now consulted for definitions to return, instead
4417 of simply returning #f, the former behavior.
4419 ** The #/ syntax for lists is no longer supported.
4421 Earlier versions of Scheme accepted this syntax, but printed a
4424 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4426 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4429 * Changes to the gh_ interface
4433 Now takes a second argument which is the result array. If this
4434 pointer is NULL, a new array is malloced (the old behaviour).
4436 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4437 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4441 * Changes to the scm_ interface
4443 ** Function: scm_make_named_hook (char* name, int n_args)
4445 Creates a hook in the same way as make-hook above but also
4446 binds a variable named NAME to it.
4448 This is the typical way of creating a hook from C code.
4450 Currently, the variable is created in the "current" module. This
4451 might change when we get the new module system.
4453 ** The smob interface
4455 The interface for creating smobs has changed. For documentation, see
4456 data-rep.info (made from guile-core/doc/data-rep.texi).
4458 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4460 >>> This function will be removed in 1.3.4. <<<
4464 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4465 This function adds a new smob type, named NAME, with instance size
4466 SIZE to the system. The return value is a tag that is used in
4467 creating instances of the type. If SIZE is 0, then no memory will
4468 be allocated when instances of the smob are created, and nothing
4469 will be freed by the default free function.
4471 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4472 This function sets the smob marking procedure for the smob type
4473 specified by the tag TC. TC is the tag returned by
4474 `scm_make_smob_type'.
4476 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4477 This function sets the smob freeing procedure for the smob type
4478 specified by the tag TC. TC is the tag returned by
4479 `scm_make_smob_type'.
4481 *** Function: void scm_set_smob_print (tc, print)
4483 - Function: void scm_set_smob_print (long tc,
4484 scm_sizet (*print) (SCM,
4488 This function sets the smob printing procedure for the smob type
4489 specified by the tag TC. TC is the tag returned by
4490 `scm_make_smob_type'.
4492 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4493 This function sets the smob equality-testing predicate for the
4494 smob type specified by the tag TC. TC is the tag returned by
4495 `scm_make_smob_type'.
4497 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4498 Make VALUE contain a smob instance of the type with type code TC and
4499 smob data DATA. VALUE must be previously declared as C type `SCM'.
4501 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4502 This macro expands to a block of code that creates a smob instance
4503 of the type with type code TC and smob data DATA, and returns that
4504 `SCM' value. It should be the last piece of code in a block.
4506 ** The interfaces for using I/O ports and implementing port types
4507 (ptobs) have changed significantly. The new interface is based on
4508 shared access to buffers and a new set of ptob procedures.
4510 *** scm_newptob has been removed
4514 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4516 - Function: SCM scm_make_port_type (char *type_name,
4517 int (*fill_buffer) (SCM port),
4518 void (*write_flush) (SCM port));
4520 Similarly to the new smob interface, there is a set of function
4521 setters by which the user can customize the behaviour of his port
4522 type. See ports.h (scm_set_port_XXX).
4524 ** scm_strport_to_string: New function: creates a new string from
4525 a string port's buffer.
4527 ** Plug in interface for random number generators
4528 The variable `scm_the_rng' in random.c contains a value and three
4529 function pointers which together define the current random number
4530 generator being used by the Scheme level interface and the random
4531 number library functions.
4533 The user is free to replace the default generator with the generator
4536 *** Variable: size_t scm_the_rng.rstate_size
4537 The size of the random state type used by the current RNG
4540 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4541 Given the random STATE, return 32 random bits.
4543 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4544 Seed random state STATE using string S of length N.
4546 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4547 Given random state STATE, return a malloced copy.
4550 The default RNG is the MWC (Multiply With Carry) random number
4551 generator described by George Marsaglia at the Department of
4552 Statistics and Supercomputer Computations Research Institute, The
4553 Florida State University (http://stat.fsu.edu/~geo).
4555 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4556 passes all tests in the DIEHARD test suite
4557 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4558 costs one multiply and one add on platforms which either supports long
4559 longs (gcc does this on most systems) or have 64 bit longs. The cost
4560 is four multiply on other systems but this can be optimized by writing
4561 scm_i_uniform32 in assembler.
4563 These functions are provided through the scm_the_rng interface for use
4564 by libguile and the application.
4566 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4567 Given the random STATE, return 32 random bits.
4568 Don't use this function directly. Instead go through the plugin
4569 interface (see "Plug in interface" above).
4571 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4572 Initialize STATE using SEED of length N.
4574 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4575 Return a malloc:ed copy of STATE. This function can easily be re-used
4576 in the interfaces to other RNGs.
4578 ** Random number library functions
4579 These functions use the current RNG through the scm_the_rng interface.
4580 It might be a good idea to use these functions from your C code so
4581 that only one random generator is used by all code in your program.
4583 The default random state is stored in:
4585 *** Variable: SCM scm_var_random_state
4586 Contains the vcell of the Scheme variable "*random-state*" which is
4587 used as default state by all random number functions in the Scheme
4592 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4594 *** Function: scm_rstate *scm_c_default_rstate (void)
4595 This is a convenience function which returns the value of
4596 scm_var_random_state. An error message is generated if this value
4597 isn't a random state.
4599 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4600 Make a new random state from the string SEED of length LENGTH.
4602 It is generally not a good idea to use multiple random states in a
4603 program. While subsequent random numbers generated from one random
4604 state are guaranteed to be reasonably independent, there is no such
4605 guarantee for numbers generated from different random states.
4607 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4608 Return 32 random bits.
4610 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4611 Return a sample from the uniform(0,1) distribution.
4613 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4614 Return a sample from the normal(0,1) distribution.
4616 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4617 Return a sample from the exp(1) distribution.
4619 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4620 Return a sample from the discrete uniform(0,M) distribution.
4622 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4623 Return a sample from the discrete uniform(0,M) distribution.
4624 M must be a bignum object. The returned value may be an INUM.
4628 Changes in Guile 1.3 (released Monday, October 19, 1998):
4630 * Changes to the distribution
4632 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4633 To avoid conflicts, programs should name environment variables after
4634 themselves, except when there's a common practice establishing some
4637 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4638 giving the former precedence, and printing a warning message if the
4639 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4641 ** The header files related to multi-byte characters have been removed.
4642 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4643 which referred to these explicitly will probably need to be rewritten,
4644 since the support for the variant string types has been removed; see
4647 ** The header files append.h and sequences.h have been removed. These
4648 files implemented non-R4RS operations which would encourage
4649 non-portable programming style and less easy-to-read code.
4651 * Changes to the stand-alone interpreter
4653 ** New procedures have been added to implement a "batch mode":
4655 *** Function: batch-mode?
4657 Returns a boolean indicating whether the interpreter is in batch
4660 *** Function: set-batch-mode?! ARG
4662 If ARG is true, switches the interpreter to batch mode. The `#f'
4663 case has not been implemented.
4665 ** Guile now provides full command-line editing, when run interactively.
4666 To use this feature, you must have the readline library installed.
4667 The Guile build process will notice it, and automatically include
4670 The readline library is available via anonymous FTP from any GNU
4671 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4673 ** the-last-stack is now a fluid.
4675 * Changes to the procedure for linking libguile with your programs
4677 ** You can now use the `guile-config' utility to build programs that use Guile.
4679 Guile now includes a command-line utility called `guile-config', which
4680 can provide information about how to compile and link programs that
4683 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4684 You should include this command's output on the command line you use
4685 to compile C or C++ code that #includes the Guile header files. It's
4686 usually just a `-I' flag to help the compiler find the Guile headers.
4689 *** `guile-config link' prints any linker flags necessary to link with Guile.
4691 This command writes to its standard output a list of flags which you
4692 must pass to the linker to link your code against the Guile library.
4693 The flags include '-lguile' itself, any other libraries the Guile
4694 library depends upon, and any `-L' flags needed to help the linker
4695 find those libraries.
4697 For example, here is a Makefile rule that builds a program named 'foo'
4698 from the object files ${FOO_OBJECTS}, and links them against Guile:
4701 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4703 Previous Guile releases recommended that you use autoconf to detect
4704 which of a predefined set of libraries were present on your system.
4705 It is more robust to use `guile-config', since it records exactly which
4706 libraries the installed Guile library requires.
4708 This was originally called `build-guile', but was renamed to
4709 `guile-config' before Guile 1.3 was released, to be consistent with
4710 the analogous script for the GTK+ GUI toolkit, which is called
4714 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4716 If you are using the GNU autoconf package to configure your program,
4717 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4718 (described above) and gather the necessary values for use in your
4721 The GUILE_FLAGS macro expands to configure script code which runs the
4722 `guile-config' script, to find out where Guile's header files and
4723 libraries are installed. It sets two variables, marked for
4724 substitution, as by AC_SUBST.
4726 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4727 code that uses Guile header files. This is almost always just a
4730 GUILE_LDFLAGS --- flags to pass to the linker to link a
4731 program against Guile. This includes `-lguile' for the Guile
4732 library itself, any libraries that Guile itself requires (like
4733 -lqthreads), and so on. It may also include a -L flag to tell the
4734 compiler where to find the libraries.
4736 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4737 directory of the Guile distribution. You can copy it into your
4738 package's aclocal.m4 file, and then use it in your configure.in file.
4740 If you are using the `aclocal' program, distributed with GNU automake,
4741 to maintain your aclocal.m4 file, the Guile installation process
4742 installs guile.m4 where aclocal will find it. All you need to do is
4743 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4744 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4748 * Changes to Scheme functions and syntax
4750 ** Multi-byte strings have been removed, as have multi-byte and wide
4751 ports. We felt that these were the wrong approach to
4752 internationalization support.
4754 ** New function: readline [PROMPT]
4755 Read a line from the terminal, and allow the user to edit it,
4756 prompting with PROMPT. READLINE provides a large set of Emacs-like
4757 editing commands, lets the user recall previously typed lines, and
4758 works on almost every kind of terminal, including dumb terminals.
4760 READLINE assumes that the cursor is at the beginning of the line when
4761 it is invoked. Thus, you can't print a prompt yourself, and then call
4762 READLINE; you need to package up your prompt as a string, pass it to
4763 the function, and let READLINE print the prompt itself. This is
4764 because READLINE needs to know the prompt's screen width.
4766 For Guile to provide this function, you must have the readline
4767 library, version 2.1 or later, installed on your system. Readline is
4768 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4769 any GNU mirror site.
4771 See also ADD-HISTORY function.
4773 ** New function: add-history STRING
4774 Add STRING as the most recent line in the history used by the READLINE
4775 command. READLINE does not add lines to the history itself; you must
4776 call ADD-HISTORY to make previous input available to the user.
4778 ** The behavior of the read-line function has changed.
4780 This function now uses standard C library functions to read the line,
4781 for speed. This means that it doesn not respect the value of
4782 scm-line-incrementors; it assumes that lines are delimited with
4785 (Note that this is read-line, the function that reads a line of text
4786 from a port, not readline, the function that reads a line from a
4787 terminal, providing full editing capabilities.)
4789 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4791 This module provides some simple argument parsing. It exports one
4794 Function: getopt-gnu-style ARG-LS
4795 Parse a list of program arguments into an alist of option
4798 Each item in the list of program arguments is examined to see if
4799 it meets the syntax of a GNU long-named option. An argument like
4800 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4801 returned alist, where MUMBLE is a keyword object with the same
4802 name as the argument. An argument like `--MUMBLE=FROB' produces
4803 an element of the form (MUMBLE . FROB), where FROB is a string.
4805 As a special case, the returned alist also contains a pair whose
4806 car is the symbol `rest'. The cdr of this pair is a list
4807 containing all the items in the argument list that are not options
4808 of the form mentioned above.
4810 The argument `--' is treated specially: all items in the argument
4811 list appearing after such an argument are not examined, and are
4812 returned in the special `rest' list.
4814 This function does not parse normal single-character switches.
4815 You will need to parse them out of the `rest' list yourself.
4817 ** The read syntax for byte vectors and short vectors has changed.
4819 Instead of #bytes(...), write #y(...).
4821 Instead of #short(...), write #h(...).
4823 This may seem nutty, but, like the other uniform vectors, byte vectors
4824 and short vectors want to have the same print and read syntax (and,
4825 more basic, want to have read syntax!). Changing the read syntax to
4826 use multiple characters after the hash sign breaks with the
4827 conventions used in R5RS and the conventions used for the other
4828 uniform vectors. It also introduces complexity in the current reader,
4829 both on the C and Scheme levels. (The Right solution is probably to
4830 change the syntax and prototypes for uniform vectors entirely.)
4833 ** The new module (ice-9 session) provides useful interactive functions.
4835 *** New procedure: (apropos REGEXP OPTION ...)
4837 Display a list of top-level variables whose names match REGEXP, and
4838 the modules they are imported from. Each OPTION should be one of the
4841 value --- Show the value of each matching variable.
4842 shadow --- Show bindings shadowed by subsequently imported modules.
4843 full --- Same as both `shadow' and `value'.
4847 guile> (apropos "trace" 'full)
4848 debug: trace #<procedure trace args>
4849 debug: untrace #<procedure untrace args>
4850 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4851 the-scm-module: before-backtrace-hook ()
4852 the-scm-module: backtrace #<primitive-procedure backtrace>
4853 the-scm-module: after-backtrace-hook ()
4854 the-scm-module: has-shown-backtrace-hint? #f
4857 ** There are new functions and syntax for working with macros.
4859 Guile implements macros as a special object type. Any variable whose
4860 top-level binding is a macro object acts as a macro. The macro object
4861 specifies how the expression should be transformed before evaluation.
4863 *** Macro objects now print in a reasonable way, resembling procedures.
4865 *** New function: (macro? OBJ)
4866 True iff OBJ is a macro object.
4868 *** New function: (primitive-macro? OBJ)
4869 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4870 macro transformers, implemented in eval.c rather than Scheme code.
4872 Why do we have this function?
4873 - For symmetry with procedure? and primitive-procedure?,
4874 - to allow custom print procedures to tell whether a macro is
4875 primitive, and display it differently, and
4876 - to allow compilers and user-written evaluators to distinguish
4877 builtin special forms from user-defined ones, which could be
4880 *** New function: (macro-type OBJ)
4881 Return a value indicating what kind of macro OBJ is. Possible return
4884 The symbol `syntax' --- a macro created by procedure->syntax.
4885 The symbol `macro' --- a macro created by procedure->macro.
4886 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4887 The boolean #f --- if OBJ is not a macro object.
4889 *** New function: (macro-name MACRO)
4890 Return the name of the macro object MACRO's procedure, as returned by
4893 *** New function: (macro-transformer MACRO)
4894 Return the transformer procedure for MACRO.
4896 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4898 Specify a new macro expander to use in the current module. Each
4899 MODULE is a module name, with the same meaning as in the `use-modules'
4900 form; each named module's exported bindings are added to the current
4901 top-level environment. TRANSFORMER is an expression evaluated in the
4902 resulting environment which must yield a procedure to use as the
4903 module's eval transformer: every expression evaluated in this module
4904 is passed to this function, and the result passed to the Guile
4907 *** macro-eval! is removed. Use local-eval instead.
4909 ** Some magic has been added to the printer to better handle user
4910 written printing routines (like record printers, closure printers).
4912 The problem is that these user written routines must have access to
4913 the current `print-state' to be able to handle fancy things like
4914 detection of circular references. These print-states have to be
4915 passed to the builtin printing routines (display, write, etc) to
4916 properly continue the print chain.
4918 We didn't want to change all existing print code so that it
4919 explicitly passes thru a print state in addition to a port. Instead,
4920 we extented the possible values that the builtin printing routines
4921 accept as a `port'. In addition to a normal port, they now also take
4922 a pair of a normal port and a print-state. Printing will go to the
4923 port and the print-state will be used to control the detection of
4924 circular references, etc. If the builtin function does not care for a
4925 print-state, it is simply ignored.
4927 User written callbacks are now called with such a pair as their
4928 `port', but because every function now accepts this pair as a PORT
4929 argument, you don't have to worry about that. In fact, it is probably
4930 safest to not check for these pairs.
4932 However, it is sometimes necessary to continue a print chain on a
4933 different port, for example to get a intermediate string
4934 representation of the printed value, mangle that string somehow, and
4935 then to finally print the mangled string. Use the new function
4937 inherit-print-state OLD-PORT NEW-PORT
4939 for this. It constructs a new `port' that prints to NEW-PORT but
4940 inherits the print-state of OLD-PORT.
4942 ** struct-vtable-offset renamed to vtable-offset-user
4944 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
4946 ** There is now a third optional argument to make-vtable-vtable
4947 (and fourth to make-struct) when constructing new types (vtables).
4948 This argument initializes field vtable-index-printer of the vtable.
4950 ** The detection of circular references has been extended to structs.
4951 That is, a structure that -- in the process of being printed -- prints
4952 itself does not lead to infinite recursion.
4954 ** There is now some basic support for fluids. Please read
4955 "libguile/fluid.h" to find out more. It is accessible from Scheme with
4956 the following functions and macros:
4958 Function: make-fluid
4960 Create a new fluid object. Fluids are not special variables or
4961 some other extension to the semantics of Scheme, but rather
4962 ordinary Scheme objects. You can store them into variables (that
4963 are still lexically scoped, of course) or into any other place you
4964 like. Every fluid has a initial value of `#f'.
4966 Function: fluid? OBJ
4968 Test whether OBJ is a fluid.
4970 Function: fluid-ref FLUID
4971 Function: fluid-set! FLUID VAL
4973 Access/modify the fluid FLUID. Modifications are only visible
4974 within the current dynamic root (that includes threads).
4976 Function: with-fluids* FLUIDS VALUES THUNK
4978 FLUIDS is a list of fluids and VALUES a corresponding list of
4979 values for these fluids. Before THUNK gets called the values are
4980 installed in the fluids and the old values of the fluids are
4981 saved in the VALUES list. When the flow of control leaves THUNK
4982 or reenters it, the values get swapped again. You might think of
4983 this as a `safe-fluid-excursion'. Note that the VALUES list is
4984 modified by `with-fluids*'.
4986 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
4988 The same as `with-fluids*' but with a different syntax. It looks
4989 just like `let', but both FLUID and VALUE are evaluated. Remember,
4990 fluids are not special variables but ordinary objects. FLUID
4991 should evaluate to a fluid.
4993 ** Changes to system call interfaces:
4995 *** close-port, close-input-port and close-output-port now return a
4996 boolean instead of an `unspecified' object. #t means that the port
4997 was successfully closed, while #f means it was already closed. It is
4998 also now possible for these procedures to raise an exception if an
4999 error occurs (some errors from write can be delayed until close.)
5001 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5004 *** the third argument to fcntl is now optional.
5006 *** the first argument to chown can now be a file descriptor or a port.
5008 *** the argument to stat can now be a port.
5010 *** The following new procedures have been added (most use scsh
5013 *** procedure: close PORT/FD
5014 Similar to close-port (*note close-port: Closing Ports.), but also
5015 works on file descriptors. A side effect of closing a file
5016 descriptor is that any ports using that file descriptor are moved
5017 to a different file descriptor and have their revealed counts set
5020 *** procedure: port->fdes PORT
5021 Returns the integer file descriptor underlying PORT. As a side
5022 effect the revealed count of PORT is incremented.
5024 *** procedure: fdes->ports FDES
5025 Returns a list of existing ports which have FDES as an underlying
5026 file descriptor, without changing their revealed counts.
5028 *** procedure: fdes->inport FDES
5029 Returns an existing input port which has FDES as its underlying
5030 file descriptor, if one exists, and increments its revealed count.
5031 Otherwise, returns a new input port with a revealed count of 1.
5033 *** procedure: fdes->outport FDES
5034 Returns an existing output port which has FDES as its underlying
5035 file descriptor, if one exists, and increments its revealed count.
5036 Otherwise, returns a new output port with a revealed count of 1.
5038 The next group of procedures perform a `dup2' system call, if NEWFD
5039 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5040 duplicated can be supplied as an integer or contained in a port. The
5041 type of value returned varies depending on which procedure is used.
5043 All procedures also have the side effect when performing `dup2' that
5044 any ports using NEWFD are moved to a different file descriptor and have
5045 their revealed counts set to zero.
5047 *** procedure: dup->fdes PORT/FD [NEWFD]
5048 Returns an integer file descriptor.
5050 *** procedure: dup->inport PORT/FD [NEWFD]
5051 Returns a new input port using the new file descriptor.
5053 *** procedure: dup->outport PORT/FD [NEWFD]
5054 Returns a new output port using the new file descriptor.
5056 *** procedure: dup PORT/FD [NEWFD]
5057 Returns a new port if PORT/FD is a port, with the same mode as the
5058 supplied port, otherwise returns an integer file descriptor.
5060 *** procedure: dup->port PORT/FD MODE [NEWFD]
5061 Returns a new port using the new file descriptor. MODE supplies a
5062 mode string for the port (*note open-file: File Ports.).
5064 *** procedure: setenv NAME VALUE
5065 Modifies the environment of the current process, which is also the
5066 default environment inherited by child processes.
5068 If VALUE is `#f', then NAME is removed from the environment.
5069 Otherwise, the string NAME=VALUE is added to the environment,
5070 replacing any existing string with name matching NAME.
5072 The return value is unspecified.
5074 *** procedure: truncate-file OBJ SIZE
5075 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5076 can be a string containing a file name or an integer file
5077 descriptor or port open for output on the file. The underlying
5078 system calls are `truncate' and `ftruncate'.
5080 The return value is unspecified.
5082 *** procedure: setvbuf PORT MODE [SIZE]
5083 Set the buffering mode for PORT. MODE can be:
5091 block buffered, using a newly allocated buffer of SIZE bytes.
5092 However if SIZE is zero or unspecified, the port will be made
5095 This procedure should not be used after I/O has been performed with
5098 Ports are usually block buffered by default, with a default buffer
5099 size. Procedures e.g., *Note open-file: File Ports, which accept a
5100 mode string allow `0' to be added to request an unbuffered port.
5102 *** procedure: fsync PORT/FD
5103 Copies any unwritten data for the specified output file descriptor
5104 to disk. If PORT/FD is a port, its buffer is flushed before the
5105 underlying file descriptor is fsync'd. The return value is
5108 *** procedure: open-fdes PATH FLAGS [MODES]
5109 Similar to `open' but returns a file descriptor instead of a port.
5111 *** procedure: execle PATH ENV [ARG] ...
5112 Similar to `execl', but the environment of the new process is
5113 specified by ENV, which must be a list of strings as returned by
5114 the `environ' procedure.
5116 This procedure is currently implemented using the `execve' system
5117 call, but we call it `execle' because of its Scheme calling
5120 *** procedure: strerror ERRNO
5121 Returns the Unix error message corresponding to ERRNO, an integer.
5123 *** procedure: primitive-exit [STATUS]
5124 Terminate the current process without unwinding the Scheme stack.
5125 This is would typically be useful after a fork. The exit status
5126 is STATUS if supplied, otherwise zero.
5128 *** procedure: times
5129 Returns an object with information about real and processor time.
5130 The following procedures accept such an object as an argument and
5131 return a selected component:
5134 The current real time, expressed as time units relative to an
5138 The CPU time units used by the calling process.
5141 The CPU time units used by the system on behalf of the
5145 The CPU time units used by terminated child processes of the
5146 calling process, whose status has been collected (e.g., using
5150 Similarly, the CPU times units used by the system on behalf of
5151 terminated child processes.
5153 ** Removed: list-length
5154 ** Removed: list-append, list-append!
5155 ** Removed: list-reverse, list-reverse!
5157 ** array-map renamed to array-map!
5159 ** serial-array-map renamed to serial-array-map!
5161 ** catch doesn't take #f as first argument any longer
5163 Previously, it was possible to pass #f instead of a key to `catch'.
5164 That would cause `catch' to pass a jump buffer object to the procedure
5165 passed as second argument. The procedure could then use this jump
5166 buffer objekt as an argument to throw.
5168 This mechanism has been removed since its utility doesn't motivate the
5169 extra complexity it introduces.
5171 ** The `#/' notation for lists now provokes a warning message from Guile.
5172 This syntax will be removed from Guile in the near future.
5174 To disable the warning message, set the GUILE_HUSH environment
5175 variable to any non-empty value.
5177 ** The newline character now prints as `#\newline', following the
5178 normal Scheme notation, not `#\nl'.
5180 * Changes to the gh_ interface
5182 ** The gh_enter function now takes care of loading the Guile startup files.
5183 gh_enter works by calling scm_boot_guile; see the remarks below.
5185 ** Function: void gh_write (SCM x)
5187 Write the printed representation of the scheme object x to the current
5188 output port. Corresponds to the scheme level `write'.
5190 ** gh_list_length renamed to gh_length.
5192 ** vector handling routines
5194 Several major changes. In particular, gh_vector() now resembles
5195 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5196 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5197 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5198 vector-related gh_ functions have been implemented.
5200 ** pair and list routines
5202 Implemented several of the R4RS pair and list functions that were
5205 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5207 New function. Converts double arrays back and forth between Scheme
5210 * Changes to the scm_ interface
5212 ** The function scm_boot_guile now takes care of loading the startup files.
5214 Guile's primary initialization function, scm_boot_guile, now takes
5215 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5216 Guile, define the module system, and put together some standard
5217 bindings. It also loads `init.scm', which is intended to hold
5218 site-specific initialization code.
5220 Since Guile cannot operate properly until boot-9.scm is loaded, there
5221 is no reason to separate loading boot-9.scm from Guile's other
5222 initialization processes.
5224 This job used to be done by scm_compile_shell_switches, which didn't
5225 make much sense; in particular, it meant that people using Guile for
5226 non-shell-like applications had to jump through hoops to get Guile
5227 initialized properly.
5229 ** The function scm_compile_shell_switches no longer loads the startup files.
5230 Now, Guile always loads the startup files, whenever it is initialized;
5231 see the notes above for scm_boot_guile and scm_load_startup_files.
5233 ** Function: scm_load_startup_files
5234 This new function takes care of loading Guile's initialization file
5235 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5236 this is always called by the Guile initialization process, it's
5237 probably not too useful to call this yourself, but it's there anyway.
5239 ** The semantics of smob marking have changed slightly.
5241 The smob marking function (the `mark' member of the scm_smobfuns
5242 structure) is no longer responsible for setting the mark bit on the
5243 smob. The generic smob handling code in the garbage collector will
5244 set this bit. The mark function need only ensure that any other
5245 objects the smob refers to get marked.
5247 Note that this change means that the smob's GC8MARK bit is typically
5248 already set upon entry to the mark function. Thus, marking functions
5249 which look like this:
5252 if (SCM_GC8MARKP (ptr))
5254 SCM_SETGC8MARK (ptr);
5255 ... mark objects to which the smob refers ...
5258 are now incorrect, since they will return early, and fail to mark any
5259 other objects the smob refers to. Some code in the Guile library used
5262 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5264 If you have implemented your own I/O port type, by writing the
5265 functions required by the scm_ptobfuns and then calling scm_newptob,
5266 you will need to change your functions slightly.
5268 The functions in a scm_ptobfuns structure now expect the port itself
5269 as their argument; they used to expect the `stream' member of the
5270 port's scm_port_table structure. This allows functions in an
5271 scm_ptobfuns structure to easily access the port's cell (and any flags
5272 it its CAR), and the port's scm_port_table structure.
5274 Guile now passes the I/O port itself as the `port' argument in the
5275 following scm_ptobfuns functions:
5277 int (*free) (SCM port);
5278 int (*fputc) (int, SCM port);
5279 int (*fputs) (char *, SCM port);
5280 scm_sizet (*fwrite) SCM_P ((char *ptr,
5284 int (*fflush) (SCM port);
5285 int (*fgetc) (SCM port);
5286 int (*fclose) (SCM port);
5288 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5291 If you have existing code which defines its own port types, it is easy
5292 to convert your code to the new interface; simply apply SCM_STREAM to
5293 the port argument to yield the value you code used to expect.
5295 Note that since both the port and the stream have the same type in the
5296 C code --- they are both SCM values --- the C compiler will not remind
5297 you if you forget to update your scm_ptobfuns functions.
5300 ** Function: int scm_internal_select (int fds,
5304 struct timeval *timeout);
5306 This is a replacement for the `select' function provided by the OS.
5307 It enables I/O blocking and sleeping to happen for one cooperative
5308 thread without blocking other threads. It also avoids busy-loops in
5309 these situations. It is intended that all I/O blocking and sleeping
5310 will finally go through this function. Currently, this function is
5311 only available on systems providing `gettimeofday' and `select'.
5313 ** Function: SCM scm_internal_stack_catch (SCM tag,
5314 scm_catch_body_t body,
5316 scm_catch_handler_t handler,
5319 A new sibling to the other two C level `catch' functions
5320 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5321 the stack to be saved automatically into the variable `the-last-stack'
5322 (scm_the_last_stack_var) on error. This is necessary if you want to
5323 use advanced error reporting, such as calling scm_display_error and
5324 scm_display_backtrace. (They both take a stack object as argument.)
5326 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5328 scm_catch_handler_t handler,
5331 Spawns a new thread. It does a job similar to
5332 scm_call_with_new_thread but takes arguments more suitable when
5333 spawning threads from application C code.
5335 ** The hook scm_error_callback has been removed. It was originally
5336 intended as a way for the user to install his own error handler. But
5337 that method works badly since it intervenes between throw and catch,
5338 thereby changing the semantics of expressions like (catch #t ...).
5339 The correct way to do it is to use one of the C level catch functions
5340 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5342 ** Removed functions:
5344 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5345 scm_list_reverse, scm_list_reverse_x
5347 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5349 These can be used for pretty list creation from C. The idea is taken
5350 from Erick Gallesio's STk.
5352 ** scm_array_map renamed to scm_array_map_x
5354 ** mbstrings are now removed
5356 This means that the type codes scm_tc7_mb_string and
5357 scm_tc7_mb_substring has been removed.
5359 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5361 Since we no longer support multi-byte strings, these I/O functions
5362 have been simplified, and renamed. Here are their old names, and
5363 their new names and arguments:
5365 scm_gen_putc -> void scm_putc (int c, SCM port);
5366 scm_gen_puts -> void scm_puts (char *s, SCM port);
5367 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5368 scm_gen_getc -> void scm_getc (SCM port);
5371 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5373 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5375 SCM_TYP7S now masks away the bit which distinguishes substrings from
5378 ** scm_catch_body_t: Backward incompatible change!
5380 Body functions to scm_internal_catch and friends do not any longer
5381 take a second argument. This is because it is no longer possible to
5382 pass a #f arg to catch.
5384 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5386 The function scm_protect_object protects its argument from being freed
5387 by the garbage collector. scm_unprotect_object removes that
5390 These functions now nest properly. That is, for every object O, there
5391 is a counter which scm_protect_object(O) increments and
5392 scm_unprotect_object(O) decrements, if the counter is greater than
5393 zero. Every object's counter is zero when it is first created. If an
5394 object's counter is greater than zero, the garbage collector will not
5395 reclaim its storage.
5397 This allows you to use scm_protect_object in your code without
5398 worrying that some other function you call will call
5399 scm_unprotect_object, and allow it to be freed. Assuming that the
5400 functions you call are well-behaved, and unprotect only those objects
5401 they protect, you can follow the same rule and have confidence that
5402 objects will be freed only at appropriate times.
5405 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5407 * Changes to the distribution
5409 ** Nightly snapshots are now available from ftp.red-bean.com.
5410 The old server, ftp.cyclic.com, has been relinquished to its rightful
5413 Nightly snapshots of the Guile development sources are now available via
5414 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5416 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5417 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5419 ** To run Guile without installing it, the procedure has changed a bit.
5421 If you used a separate build directory to compile Guile, you'll need
5422 to include the build directory in SCHEME_LOAD_PATH, as well as the
5423 source directory. See the `INSTALL' file for examples.
5425 * Changes to the procedure for linking libguile with your programs
5427 ** The standard Guile load path for Scheme code now includes
5428 $(datadir)/guile (usually /usr/local/share/guile). This means that
5429 you can install your own Scheme files there, and Guile will find them.
5430 (Previous versions of Guile only checked a directory whose name
5431 contained the Guile version number, so you had to re-install or move
5432 your Scheme sources each time you installed a fresh version of Guile.)
5434 The load path also includes $(datadir)/guile/site; we recommend
5435 putting individual Scheme files there. If you want to install a
5436 package with multiple source files, create a directory for them under
5439 ** Guile 1.2 will now use the Rx regular expression library, if it is
5440 installed on your system. When you are linking libguile into your own
5441 programs, this means you will have to link against -lguile, -lqt (if
5442 you configured Guile with thread support), and -lrx.
5444 If you are using autoconf to generate configuration scripts for your
5445 application, the following lines should suffice to add the appropriate
5446 libraries to your link command:
5448 ### Find Rx, quickthreads and libguile.
5449 AC_CHECK_LIB(rx, main)
5450 AC_CHECK_LIB(qt, main)
5451 AC_CHECK_LIB(guile, scm_shell)
5453 The Guile 1.2 distribution does not contain sources for the Rx
5454 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5455 retrieve it from a GNU FTP site and install it separately.
5457 * Changes to Scheme functions and syntax
5459 ** The dynamic linking features of Guile are now enabled by default.
5460 You can disable them by giving the `--disable-dynamic-linking' option
5463 (dynamic-link FILENAME)
5465 Find the object file denoted by FILENAME (a string) and link it
5466 into the running Guile application. When everything works out,
5467 return a Scheme object suitable for representing the linked object
5468 file. Otherwise an error is thrown. How object files are
5469 searched is system dependent.
5471 (dynamic-object? VAL)
5473 Determine whether VAL represents a dynamically linked object file.
5475 (dynamic-unlink DYNOBJ)
5477 Unlink the indicated object file from the application. DYNOBJ
5478 should be one of the values returned by `dynamic-link'.
5480 (dynamic-func FUNCTION DYNOBJ)
5482 Search the C function indicated by FUNCTION (a string or symbol)
5483 in DYNOBJ and return some Scheme object that can later be used
5484 with `dynamic-call' to actually call this function. Right now,
5485 these Scheme objects are formed by casting the address of the
5486 function to `long' and converting this number to its Scheme
5489 (dynamic-call FUNCTION DYNOBJ)
5491 Call the C function indicated by FUNCTION and DYNOBJ. The
5492 function is passed no arguments and its return value is ignored.
5493 When FUNCTION is something returned by `dynamic-func', call that
5494 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5495 etc.), look it up in DYNOBJ; this is equivalent to
5497 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5499 Interrupts are deferred while the C function is executing (with
5500 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5502 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5504 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5505 some arguments and return its return value. The C function is
5506 expected to take two arguments and return an `int', just like
5509 int c_func (int argc, char **argv);
5511 ARGS must be a list of strings and is converted into an array of
5512 `char *'. The array is passed in ARGV and its size in ARGC. The
5513 return value is converted to a Scheme number and returned from the
5514 call to `dynamic-args-call'.
5516 When dynamic linking is disabled or not supported on your system,
5517 the above functions throw errors, but they are still available.
5519 Here is a small example that works on GNU/Linux:
5521 (define libc-obj (dynamic-link "libc.so"))
5522 (dynamic-args-call 'rand libc-obj '())
5524 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5526 ** The #/ syntax for module names is depreciated, and will be removed
5527 in a future version of Guile. Instead of
5535 The latter syntax is more consistent with existing Lisp practice.
5537 ** Guile now does fancier printing of structures. Structures are the
5538 underlying implementation for records, which in turn are used to
5539 implement modules, so all of these object now print differently and in
5540 a more informative way.
5542 The Scheme printer will examine the builtin variable *struct-printer*
5543 whenever it needs to print a structure object. When this variable is
5544 not `#f' it is deemed to be a procedure and will be applied to the
5545 structure object and the output port. When *struct-printer* is `#f'
5546 or the procedure return `#f' the structure object will be printed in
5547 the boring #<struct 80458270> form.
5549 This hook is used by some routines in ice-9/boot-9.scm to implement
5550 type specific printing routines. Please read the comments there about
5553 One of the more specific uses of structs are records. The printing
5554 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5555 called. It should behave like a *struct-printer* procedure (described
5558 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5559 token of the form #:NAME, where NAME has the same syntax as a Scheme
5560 symbol, is the external representation of the keyword named NAME.
5561 Keyword objects print using this syntax as well, so values containing
5562 keyword objects can be read back into Guile. When used in an
5563 expression, keywords are self-quoting objects.
5565 Guile suports this read syntax, and uses this print syntax, regardless
5566 of the current setting of the `keyword' read option. The `keyword'
5567 read option only controls whether Guile recognizes the `:NAME' syntax,
5568 which is incompatible with R4RS. (R4RS says such token represent
5571 ** Guile has regular expression support again. Guile 1.0 included
5572 functions for matching regular expressions, based on the Rx library.
5573 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5574 distribution, and thus Guile had no regular expression support. Guile
5575 1.2 again supports the most commonly used functions, and supports all
5576 of SCSH's regular expression functions.
5578 If your system does not include a POSIX regular expression library,
5579 and you have not linked Guile with a third-party regexp library such as
5580 Rx, these functions will not be available. You can tell whether your
5581 Guile installation includes regular expression support by checking
5582 whether the `*features*' list includes the `regex' symbol.
5584 *** regexp functions
5586 By default, Guile supports POSIX extended regular expressions. That
5587 means that the characters `(', `)', `+' and `?' are special, and must
5588 be escaped if you wish to match the literal characters.
5590 This regular expression interface was modeled after that implemented
5591 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5592 with SCSH regular expressions.
5594 **** Function: string-match PATTERN STR [START]
5595 Compile the string PATTERN into a regular expression and compare
5596 it with STR. The optional numeric argument START specifies the
5597 position of STR at which to begin matching.
5599 `string-match' returns a "match structure" which describes what,
5600 if anything, was matched by the regular expression. *Note Match
5601 Structures::. If STR does not match PATTERN at all,
5602 `string-match' returns `#f'.
5604 Each time `string-match' is called, it must compile its PATTERN
5605 argument into a regular expression structure. This operation is
5606 expensive, which makes `string-match' inefficient if the same regular
5607 expression is used several times (for example, in a loop). For better
5608 performance, you can compile a regular expression in advance and then
5609 match strings against the compiled regexp.
5611 **** Function: make-regexp STR [FLAGS]
5612 Compile the regular expression described by STR, and return the
5613 compiled regexp structure. If STR does not describe a legal
5614 regular expression, `make-regexp' throws a
5615 `regular-expression-syntax' error.
5617 FLAGS may be the bitwise-or of one or more of the following:
5619 **** Constant: regexp/extended
5620 Use POSIX Extended Regular Expression syntax when interpreting
5621 STR. If not set, POSIX Basic Regular Expression syntax is used.
5622 If the FLAGS argument is omitted, we assume regexp/extended.
5624 **** Constant: regexp/icase
5625 Do not differentiate case. Subsequent searches using the
5626 returned regular expression will be case insensitive.
5628 **** Constant: regexp/newline
5629 Match-any-character operators don't match a newline.
5631 A non-matching list ([^...]) not containing a newline matches a
5634 Match-beginning-of-line operator (^) matches the empty string
5635 immediately after a newline, regardless of whether the FLAGS
5636 passed to regexp-exec contain regexp/notbol.
5638 Match-end-of-line operator ($) matches the empty string
5639 immediately before a newline, regardless of whether the FLAGS
5640 passed to regexp-exec contain regexp/noteol.
5642 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5643 Match the compiled regular expression REGEXP against `str'. If
5644 the optional integer START argument is provided, begin matching
5645 from that position in the string. Return a match structure
5646 describing the results of the match, or `#f' if no match could be
5649 FLAGS may be the bitwise-or of one or more of the following:
5651 **** Constant: regexp/notbol
5652 The match-beginning-of-line operator always fails to match (but
5653 see the compilation flag regexp/newline above) This flag may be
5654 used when different portions of a string are passed to
5655 regexp-exec and the beginning of the string should not be
5656 interpreted as the beginning of the line.
5658 **** Constant: regexp/noteol
5659 The match-end-of-line operator always fails to match (but see the
5660 compilation flag regexp/newline above)
5662 **** Function: regexp? OBJ
5663 Return `#t' if OBJ is a compiled regular expression, or `#f'
5666 Regular expressions are commonly used to find patterns in one string
5667 and replace them with the contents of another string.
5669 **** Function: regexp-substitute PORT MATCH [ITEM...]
5670 Write to the output port PORT selected contents of the match
5671 structure MATCH. Each ITEM specifies what should be written, and
5672 may be one of the following arguments:
5674 * A string. String arguments are written out verbatim.
5676 * An integer. The submatch with that number is written.
5678 * The symbol `pre'. The portion of the matched string preceding
5679 the regexp match is written.
5681 * The symbol `post'. The portion of the matched string
5682 following the regexp match is written.
5684 PORT may be `#f', in which case nothing is written; instead,
5685 `regexp-substitute' constructs a string from the specified ITEMs
5688 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5689 Similar to `regexp-substitute', but can be used to perform global
5690 substitutions on STR. Instead of taking a match structure as an
5691 argument, `regexp-substitute/global' takes two string arguments: a
5692 REGEXP string describing a regular expression, and a TARGET string
5693 which should be matched against this regular expression.
5695 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5698 * A function may be supplied. When this function is called, it
5699 will be passed one argument: a match structure for a given
5700 regular expression match. It should return a string to be
5701 written out to PORT.
5703 * The `post' symbol causes `regexp-substitute/global' to recurse
5704 on the unmatched portion of STR. This *must* be supplied in
5705 order to perform global search-and-replace on STR; if it is
5706 not present among the ITEMs, then `regexp-substitute/global'
5707 will return after processing a single match.
5709 *** Match Structures
5711 A "match structure" is the object returned by `string-match' and
5712 `regexp-exec'. It describes which portion of a string, if any, matched
5713 the given regular expression. Match structures include: a reference to
5714 the string that was checked for matches; the starting and ending
5715 positions of the regexp match; and, if the regexp included any
5716 parenthesized subexpressions, the starting and ending positions of each
5719 In each of the regexp match functions described below, the `match'
5720 argument must be a match structure returned by a previous call to
5721 `string-match' or `regexp-exec'. Most of these functions return some
5722 information about the original target string that was matched against a
5723 regular expression; we will call that string TARGET for easy reference.
5725 **** Function: regexp-match? OBJ
5726 Return `#t' if OBJ is a match structure returned by a previous
5727 call to `regexp-exec', or `#f' otherwise.
5729 **** Function: match:substring MATCH [N]
5730 Return the portion of TARGET matched by subexpression number N.
5731 Submatch 0 (the default) represents the entire regexp match. If
5732 the regular expression as a whole matched, but the subexpression
5733 number N did not match, return `#f'.
5735 **** Function: match:start MATCH [N]
5736 Return the starting position of submatch number N.
5738 **** Function: match:end MATCH [N]
5739 Return the ending position of submatch number N.
5741 **** Function: match:prefix MATCH
5742 Return the unmatched portion of TARGET preceding the regexp match.
5744 **** Function: match:suffix MATCH
5745 Return the unmatched portion of TARGET following the regexp match.
5747 **** Function: match:count MATCH
5748 Return the number of parenthesized subexpressions from MATCH.
5749 Note that the entire regular expression match itself counts as a
5750 subexpression, and failed submatches are included in the count.
5752 **** Function: match:string MATCH
5753 Return the original TARGET string.
5755 *** Backslash Escapes
5757 Sometimes you will want a regexp to match characters like `*' or `$'
5758 exactly. For example, to check whether a particular string represents
5759 a menu entry from an Info node, it would be useful to match it against
5760 a regexp like `^* [^:]*::'. However, this won't work; because the
5761 asterisk is a metacharacter, it won't match the `*' at the beginning of
5762 the string. In this case, we want to make the first asterisk un-magic.
5764 You can do this by preceding the metacharacter with a backslash
5765 character `\'. (This is also called "quoting" the metacharacter, and
5766 is known as a "backslash escape".) When Guile sees a backslash in a
5767 regular expression, it considers the following glyph to be an ordinary
5768 character, no matter what special meaning it would ordinarily have.
5769 Therefore, we can make the above example work by changing the regexp to
5770 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5771 to match only a single asterisk in the target string.
5773 Since the backslash is itself a metacharacter, you may force a
5774 regexp to match a backslash in the target string by preceding the
5775 backslash with itself. For example, to find variable references in a
5776 TeX program, you might want to find occurrences of the string `\let\'
5777 followed by any number of alphabetic characters. The regular expression
5778 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5779 each match a single backslash in the target string.
5781 **** Function: regexp-quote STR
5782 Quote each special character found in STR with a backslash, and
5783 return the resulting string.
5785 *Very important:* Using backslash escapes in Guile source code (as
5786 in Emacs Lisp or C) can be tricky, because the backslash character has
5787 special meaning for the Guile reader. For example, if Guile encounters
5788 the character sequence `\n' in the middle of a string while processing
5789 Scheme code, it replaces those characters with a newline character.
5790 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5791 Several of these "escape sequences" are processed by the Guile reader
5792 before your code is executed. Unrecognized escape sequences are
5793 ignored: if the characters `\*' appear in a string, they will be
5794 translated to the single character `*'.
5796 This translation is obviously undesirable for regular expressions,
5797 since we want to be able to include backslashes in a string in order to
5798 escape regexp metacharacters. Therefore, to make sure that a backslash
5799 is preserved in a string in your Guile program, you must use *two*
5800 consecutive backslashes:
5802 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5804 The string in this example is preprocessed by the Guile reader before
5805 any code is executed. The resulting argument to `make-regexp' is the
5806 string `^\* [^:]*', which is what we really want.
5808 This also means that in order to write a regular expression that
5809 matches a single backslash character, the regular expression string in
5810 the source code must include *four* backslashes. Each consecutive pair
5811 of backslashes gets translated by the Guile reader to a single
5812 backslash, and the resulting double-backslash is interpreted by the
5813 regexp engine as matching a single backslash character. Hence:
5815 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5817 The reason for the unwieldiness of this syntax is historical. Both
5818 regular expression pattern matchers and Unix string processing systems
5819 have traditionally used backslashes with the special meanings described
5820 above. The POSIX regular expression specification and ANSI C standard
5821 both require these semantics. Attempting to abandon either convention
5822 would cause other kinds of compatibility problems, possibly more severe
5823 ones. Therefore, without extending the Scheme reader to support
5824 strings with different quoting conventions (an ungainly and confusing
5825 extension when implemented in other languages), we must adhere to this
5826 cumbersome escape syntax.
5828 * Changes to the gh_ interface
5830 * Changes to the scm_ interface
5832 * Changes to system call interfaces:
5834 ** The value returned by `raise' is now unspecified. It throws an exception
5837 *** A new procedure `sigaction' can be used to install signal handlers
5839 (sigaction signum [action] [flags])
5841 signum is the signal number, which can be specified using the value
5844 If action is omitted, sigaction returns a pair: the CAR is the current
5845 signal hander, which will be either an integer with the value SIG_DFL
5846 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5847 handles the signal, or #f if a non-Scheme procedure handles the
5848 signal. The CDR contains the current sigaction flags for the handler.
5850 If action is provided, it is installed as the new handler for signum.
5851 action can be a Scheme procedure taking one argument, or the value of
5852 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5853 whatever signal handler was installed before sigaction was first used.
5854 Flags can optionally be specified for the new handler (SA_RESTART is
5855 always used if the system provides it, so need not be specified.) The
5856 return value is a pair with information about the old handler as
5859 This interface does not provide access to the "signal blocking"
5860 facility. Maybe this is not needed, since the thread support may
5861 provide solutions to the problem of consistent access to data
5864 *** A new procedure `flush-all-ports' is equivalent to running
5865 `force-output' on every port open for output.
5867 ** Guile now provides information on how it was built, via the new
5868 global variable, %guile-build-info. This variable records the values
5869 of the standard GNU makefile directory variables as an assocation
5870 list, mapping variable names (symbols) onto directory paths (strings).
5871 For example, to find out where the Guile link libraries were
5872 installed, you can say:
5874 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5877 * Changes to the scm_ interface
5879 ** The new function scm_handle_by_message_noexit is just like the
5880 existing scm_handle_by_message function, except that it doesn't call
5881 exit to terminate the process. Instead, it prints a message and just
5882 returns #f. This might be a more appropriate catch-all handler for
5883 new dynamic roots and threads.
5886 Changes in Guile 1.1 (released Friday, May 16 1997):
5888 * Changes to the distribution.
5890 The Guile 1.0 distribution has been split up into several smaller
5892 guile-core --- the Guile interpreter itself.
5893 guile-tcltk --- the interface between the Guile interpreter and
5894 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5895 is a toolkit for building graphical user interfaces.
5896 guile-rgx-ctax --- the interface between Guile and the Rx regular
5897 expression matcher, and the translator for the Ctax
5898 programming language. These are packaged together because the
5899 Ctax translator uses Rx to parse Ctax source code.
5901 This NEWS file describes the changes made to guile-core since the 1.0
5904 We no longer distribute the documentation, since it was either out of
5905 date, or incomplete. As soon as we have current documentation, we
5910 * Changes to the stand-alone interpreter
5912 ** guile now accepts command-line arguments compatible with SCSH, Olin
5913 Shivers' Scheme Shell.
5915 In general, arguments are evaluated from left to right, but there are
5916 exceptions. The following switches stop argument processing, and
5917 stash all remaining command-line arguments as the value returned by
5918 the (command-line) function.
5919 -s SCRIPT load Scheme source code from FILE, and exit
5920 -c EXPR evalute Scheme expression EXPR, and exit
5921 -- stop scanning arguments; run interactively
5923 The switches below are processed as they are encountered.
5924 -l FILE load Scheme source code from FILE
5925 -e FUNCTION after reading script, apply FUNCTION to
5926 command line arguments
5927 -ds do -s script at this point
5928 --emacs enable Emacs protocol (experimental)
5929 -h, --help display this help and exit
5930 -v, --version display version information and exit
5931 \ read arguments from following script lines
5933 So, for example, here is a Guile script named `ekko' (thanks, Olin)
5934 which re-implements the traditional "echo" command:
5936 #!/usr/local/bin/guile -s
5939 (map (lambda (arg) (display arg) (display " "))
5943 (main (command-line))
5945 Suppose we invoke this script as follows:
5947 ekko a speckled gecko
5949 Through the magic of Unix script processing (triggered by the `#!'
5950 token at the top of the file), /usr/local/bin/guile receives the
5951 following list of command-line arguments:
5953 ("-s" "./ekko" "a" "speckled" "gecko")
5955 Unix inserts the name of the script after the argument specified on
5956 the first line of the file (in this case, "-s"), and then follows that
5957 with the arguments given to the script. Guile loads the script, which
5958 defines the `main' function, and then applies it to the list of
5959 remaining command-line arguments, ("a" "speckled" "gecko").
5961 In Unix, the first line of a script file must take the following form:
5963 #!INTERPRETER ARGUMENT
5965 where INTERPRETER is the absolute filename of the interpreter
5966 executable, and ARGUMENT is a single command-line argument to pass to
5969 You may only pass one argument to the interpreter, and its length is
5970 limited. These restrictions can be annoying to work around, so Guile
5971 provides a general mechanism (borrowed from, and compatible with,
5972 SCSH) for circumventing them.
5974 If the ARGUMENT in a Guile script is a single backslash character,
5975 `\', Guile will open the script file, parse arguments from its second
5976 and subsequent lines, and replace the `\' with them. So, for example,
5977 here is another implementation of the `ekko' script:
5979 #!/usr/local/bin/guile \
5983 (for-each (lambda (arg) (display arg) (display " "))
5987 If the user invokes this script as follows:
5989 ekko a speckled gecko
5991 Unix expands this into
5993 /usr/local/bin/guile \ ekko a speckled gecko
5995 When Guile sees the `\' argument, it replaces it with the arguments
5996 read from the second line of the script, producing:
5998 /usr/local/bin/guile -e main -s ekko a speckled gecko
6000 This tells Guile to load the `ekko' script, and apply the function
6001 `main' to the argument list ("a" "speckled" "gecko").
6003 Here is how Guile parses the command-line arguments:
6004 - Each space character terminates an argument. This means that two
6005 spaces in a row introduce an empty-string argument.
6006 - The tab character is not permitted (unless you quote it with the
6007 backslash character, as described below), to avoid confusion.
6008 - The newline character terminates the sequence of arguments, and will
6009 also terminate a final non-empty argument. (However, a newline
6010 following a space will not introduce a final empty-string argument;
6011 it only terminates the argument list.)
6012 - The backslash character is the escape character. It escapes
6013 backslash, space, tab, and newline. The ANSI C escape sequences
6014 like \n and \t are also supported. These produce argument
6015 constituents; the two-character combination \n doesn't act like a
6016 terminating newline. The escape sequence \NNN for exactly three
6017 octal digits reads as the character whose ASCII code is NNN. As
6018 above, characters produced this way are argument constituents.
6019 Backslash followed by other characters is not allowed.
6021 * Changes to the procedure for linking libguile with your programs
6023 ** Guile now builds and installs a shared guile library, if your
6024 system support shared libraries. (It still builds a static library on
6025 all systems.) Guile automatically detects whether your system
6026 supports shared libraries. To prevent Guile from buildisg shared
6027 libraries, pass the `--disable-shared' flag to the configure script.
6029 Guile takes longer to compile when it builds shared libraries, because
6030 it must compile every file twice --- once to produce position-
6031 independent object code, and once to produce normal object code.
6033 ** The libthreads library has been merged into libguile.
6035 To link a program against Guile, you now need only link against
6036 -lguile and -lqt; -lthreads is no longer needed. If you are using
6037 autoconf to generate configuration scripts for your application, the
6038 following lines should suffice to add the appropriate libraries to
6041 ### Find quickthreads and libguile.
6042 AC_CHECK_LIB(qt, main)
6043 AC_CHECK_LIB(guile, scm_shell)
6045 * Changes to Scheme functions
6047 ** Guile Scheme's special syntax for keyword objects is now optional,
6048 and disabled by default.
6050 The syntax variation from R4RS made it difficult to port some
6051 interesting packages to Guile. The routines which accepted keyword
6052 arguments (mostly in the module system) have been modified to also
6053 accept symbols whose names begin with `:'.
6055 To change the keyword syntax, you must first import the (ice-9 debug)
6057 (use-modules (ice-9 debug))
6059 Then you can enable the keyword syntax as follows:
6060 (read-set! keywords 'prefix)
6062 To disable keyword syntax, do this:
6063 (read-set! keywords #f)
6065 ** Many more primitive functions accept shared substrings as
6066 arguments. In the past, these functions required normal, mutable
6067 strings as arguments, although they never made use of this
6070 ** The uniform array functions now operate on byte vectors. These
6071 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6072 `serial-array-map', `array-map', `array-for-each', and
6075 ** The new functions `trace' and `untrace' implement simple debugging
6076 support for Scheme functions.
6078 The `trace' function accepts any number of procedures as arguments,
6079 and tells the Guile interpreter to display each procedure's name and
6080 arguments each time the procedure is invoked. When invoked with no
6081 arguments, `trace' returns the list of procedures currently being
6084 The `untrace' function accepts any number of procedures as arguments,
6085 and tells the Guile interpreter not to trace them any more. When
6086 invoked with no arguments, `untrace' untraces all curretly traced
6089 The tracing in Guile has an advantage over most other systems: we
6090 don't create new procedure objects, but mark the procedure objects
6091 themselves. This means that anonymous and internal procedures can be
6094 ** The function `assert-repl-prompt' has been renamed to
6095 `set-repl-prompt!'. It takes one argument, PROMPT.
6096 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6097 - If PROMPT is a string, we use it as a prompt.
6098 - If PROMPT is a procedure accepting no arguments, we call it, and
6099 display the result as a prompt.
6100 - Otherwise, we display "> ".
6102 ** The new function `eval-string' reads Scheme expressions from a
6103 string and evaluates them, returning the value of the last expression
6104 in the string. If the string contains no expressions, it returns an
6107 ** The new function `thunk?' returns true iff its argument is a
6108 procedure of zero arguments.
6110 ** `defined?' is now a builtin function, instead of syntax. This
6111 means that its argument should be quoted. It returns #t iff its
6112 argument is bound in the current module.
6114 ** The new syntax `use-modules' allows you to add new modules to your
6115 environment without re-typing a complete `define-module' form. It
6116 accepts any number of module names as arguments, and imports their
6117 public bindings into the current module.
6119 ** The new function (module-defined? NAME MODULE) returns true iff
6120 NAME, a symbol, is defined in MODULE, a module object.
6122 ** The new function `builtin-bindings' creates and returns a hash
6123 table containing copies of all the root module's bindings.
6125 ** The new function `builtin-weak-bindings' does the same as
6126 `builtin-bindings', but creates a doubly-weak hash table.
6128 ** The `equal?' function now considers variable objects to be
6129 equivalent if they have the same name and the same value.
6131 ** The new function `command-line' returns the command-line arguments
6132 given to Guile, as a list of strings.
6134 When using guile as a script interpreter, `command-line' returns the
6135 script's arguments; those processed by the interpreter (like `-s' or
6136 `-c') are omitted. (In other words, you get the normal, expected
6137 behavior.) Any application that uses scm_shell to process its
6138 command-line arguments gets this behavior as well.
6140 ** The new function `load-user-init' looks for a file called `.guile'
6141 in the user's home directory, and loads it if it exists. This is
6142 mostly for use by the code generated by scm_compile_shell_switches,
6143 but we thought it might also be useful in other circumstances.
6145 ** The new function `log10' returns the base-10 logarithm of its
6148 ** Changes to I/O functions
6150 *** The functions `read', `primitive-load', `read-and-eval!', and
6151 `primitive-load-path' no longer take optional arguments controlling
6152 case insensitivity and a `#' parser.
6154 Case sensitivity is now controlled by a read option called
6155 `case-insensitive'. The user can add new `#' syntaxes with the
6156 `read-hash-extend' function (see below).
6158 *** The new function `read-hash-extend' allows the user to change the
6159 syntax of Guile Scheme in a somewhat controlled way.
6161 (read-hash-extend CHAR PROC)
6162 When parsing S-expressions, if we read a `#' character followed by
6163 the character CHAR, use PROC to parse an object from the stream.
6164 If PROC is #f, remove any parsing procedure registered for CHAR.
6166 The reader applies PROC to two arguments: CHAR and an input port.
6168 *** The new functions read-delimited and read-delimited! provide a
6169 general mechanism for doing delimited input on streams.
6171 (read-delimited DELIMS [PORT HANDLE-DELIM])
6172 Read until we encounter one of the characters in DELIMS (a string),
6173 or end-of-file. PORT is the input port to read from; it defaults to
6174 the current input port. The HANDLE-DELIM parameter determines how
6175 the terminating character is handled; it should be one of the
6178 'trim omit delimiter from result
6179 'peek leave delimiter character in input stream
6180 'concat append delimiter character to returned value
6181 'split return a pair: (RESULT . TERMINATOR)
6183 HANDLE-DELIM defaults to 'peek.
6185 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6186 A side-effecting variant of `read-delimited'.
6188 The data is written into the string BUF at the indices in the
6189 half-open interval [START, END); the default interval is the whole
6190 string: START = 0 and END = (string-length BUF). The values of
6191 START and END must specify a well-defined interval in BUF, i.e.
6192 0 <= START <= END <= (string-length BUF).
6194 It returns NBYTES, the number of bytes read. If the buffer filled
6195 up without a delimiter character being found, it returns #f. If the
6196 port is at EOF when the read starts, it returns the EOF object.
6198 If an integer is returned (i.e., the read is successfully terminated
6199 by reading a delimiter character), then the HANDLE-DELIM parameter
6200 determines how to handle the terminating character. It is described
6201 above, and defaults to 'peek.
6203 (The descriptions of these functions were borrowed from the SCSH
6204 manual, by Olin Shivers and Brian Carlstrom.)
6206 *** The `%read-delimited!' function is the primitive used to implement
6207 `read-delimited' and `read-delimited!'.
6209 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6211 This returns a pair of values: (TERMINATOR . NUM-READ).
6212 - TERMINATOR describes why the read was terminated. If it is a
6213 character or the eof object, then that is the value that terminated
6214 the read. If it is #f, the function filled the buffer without finding
6215 a delimiting character.
6216 - NUM-READ is the number of characters read into BUF.
6218 If the read is successfully terminated by reading a delimiter
6219 character, then the gobble? parameter determines what to do with the
6220 terminating character. If true, the character is removed from the
6221 input stream; if false, the character is left in the input stream
6222 where a subsequent read operation will retrieve it. In either case,
6223 the character is also the first value returned by the procedure call.
6225 (The descriptions of this function was borrowed from the SCSH manual,
6226 by Olin Shivers and Brian Carlstrom.)
6228 *** The `read-line' and `read-line!' functions have changed; they now
6229 trim the terminator by default; previously they appended it to the
6230 returned string. For the old behavior, use (read-line PORT 'concat).
6232 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6233 take new optional START and END arguments, specifying the region of
6234 the array to read and write.
6236 *** The `ungetc-char-ready?' function has been removed. We feel it's
6237 inappropriate for an interface to expose implementation details this
6240 ** Changes to the Unix library and system call interface
6242 *** The new fcntl function provides access to the Unix `fcntl' system
6245 (fcntl PORT COMMAND VALUE)
6246 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6247 Values for COMMAND are:
6249 F_DUPFD duplicate a file descriptor
6250 F_GETFD read the descriptor's close-on-exec flag
6251 F_SETFD set the descriptor's close-on-exec flag to VALUE
6252 F_GETFL read the descriptor's flags, as set on open
6253 F_SETFL set the descriptor's flags, as set on open to VALUE
6254 F_GETOWN return the process ID of a socket's owner, for SIGIO
6255 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6256 FD_CLOEXEC not sure what this is
6258 For details, see the documentation for the fcntl system call.
6260 *** The arguments to `select' have changed, for compatibility with
6261 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6262 expected behavior. The MILLISECONDS parameter has been changed to
6263 MICROSECONDS, to more closely resemble the underlying system call.
6264 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6265 corresponding return set will be the same.
6267 *** The arguments to the `mknod' system call have changed. They are
6270 (mknod PATH TYPE PERMS DEV)
6271 Create a new file (`node') in the file system. PATH is the name of
6272 the file to create. TYPE is the kind of file to create; it should
6273 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6274 permission bits to give the newly created file. If TYPE is
6275 'block-special or 'char-special, DEV specifies which device the
6276 special file refers to; its interpretation depends on the kind of
6277 special file being created.
6279 *** The `fork' function has been renamed to `primitive-fork', to avoid
6280 clashing with various SCSH forks.
6282 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6283 and `recvfrom!'. They no longer accept a size for a second argument;
6284 you must pass a string to hold the received value. They no longer
6285 return the buffer. Instead, `recv' returns the length of the message
6286 received, and `recvfrom' returns a pair containing the packet's length
6287 and originating address.
6289 *** The file descriptor datatype has been removed, as have the
6290 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6291 We plan to replace these functions with a SCSH-compatible interface.
6293 *** The `create' function has been removed; it's just a special case
6296 *** There are new functions to break down process termination status
6297 values. In the descriptions below, STATUS is a value returned by
6300 (status:exit-val STATUS)
6301 If the child process exited normally, this function returns the exit
6302 code for the child process (i.e., the value passed to exit, or
6303 returned from main). If the child process did not exit normally,
6304 this function returns #f.
6306 (status:stop-sig STATUS)
6307 If the child process was suspended by a signal, this function
6308 returns the signal that suspended the child. Otherwise, it returns
6311 (status:term-sig STATUS)
6312 If the child process terminated abnormally, this function returns
6313 the signal that terminated the child. Otherwise, this function
6316 POSIX promises that exactly one of these functions will return true on
6317 a valid STATUS value.
6319 These functions are compatible with SCSH.
6321 *** There are new accessors and setters for the broken-out time vectors
6322 returned by `localtime', `gmtime', and that ilk. They are:
6324 Component Accessor Setter
6325 ========================= ============ ============
6326 seconds tm:sec set-tm:sec
6327 minutes tm:min set-tm:min
6328 hours tm:hour set-tm:hour
6329 day of the month tm:mday set-tm:mday
6330 month tm:mon set-tm:mon
6331 year tm:year set-tm:year
6332 day of the week tm:wday set-tm:wday
6333 day in the year tm:yday set-tm:yday
6334 daylight saving time tm:isdst set-tm:isdst
6335 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6336 name of time zone tm:zone set-tm:zone
6338 *** There are new accessors for the vectors returned by `uname',
6339 describing the host system:
6342 ============================================== ================
6343 name of the operating system implementation utsname:sysname
6344 network name of this machine utsname:nodename
6345 release level of the operating system utsname:release
6346 version level of the operating system utsname:version
6347 machine hardware platform utsname:machine
6349 *** There are new accessors for the vectors returned by `getpw',
6350 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6351 system's user database:
6354 ====================== =================
6355 user name passwd:name
6356 user password passwd:passwd
6359 real name passwd:gecos
6360 home directory passwd:dir
6361 shell program passwd:shell
6363 *** There are new accessors for the vectors returned by `getgr',
6364 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6365 system's group database:
6368 ======================= ============
6369 group name group:name
6370 group password group:passwd
6372 group members group:mem
6374 *** There are new accessors for the vectors returned by `gethost',
6375 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6379 ========================= ===============
6380 official name of host hostent:name
6381 alias list hostent:aliases
6382 host address type hostent:addrtype
6383 length of address hostent:length
6384 list of addresses hostent:addr-list
6386 *** There are new accessors for the vectors returned by `getnet',
6387 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6391 ========================= ===============
6392 official name of net netent:name
6393 alias list netent:aliases
6394 net number type netent:addrtype
6395 net number netent:net
6397 *** There are new accessors for the vectors returned by `getproto',
6398 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6402 ========================= ===============
6403 official protocol name protoent:name
6404 alias list protoent:aliases
6405 protocol number protoent:proto
6407 *** There are new accessors for the vectors returned by `getserv',
6408 `getservbyname', `getservbyport', and `getservent', describing
6412 ========================= ===============
6413 official service name servent:name
6414 alias list servent:aliases
6415 port number servent:port
6416 protocol to use servent:proto
6418 *** There are new accessors for the sockaddr structures returned by
6419 `accept', `getsockname', `getpeername', `recvfrom!':
6422 ======================================== ===============
6423 address format (`family') sockaddr:fam
6424 path, for file domain addresses sockaddr:path
6425 address, for internet domain addresses sockaddr:addr
6426 TCP or UDP port, for internet sockaddr:port
6428 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6429 `getprotoent', and `getservent' functions now return #f at the end of
6430 the user database. (They used to throw an exception.)
6432 Note that calling MUMBLEent function is equivalent to calling the
6433 corresponding MUMBLE function with no arguments.
6435 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6436 `setprotoent', and `setservent' routines now take no arguments.
6438 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6439 provide more useful information when they throw an exception.
6441 *** The `lnaof' function has been renamed to `inet-lnaof'.
6443 *** Guile now claims to have the `current-time' feature.
6445 *** The `mktime' function now takes an optional second argument ZONE,
6446 giving the time zone to use for the conversion. ZONE should be a
6447 string, in the same format as expected for the "TZ" environment variable.
6449 *** The `strptime' function now returns a pair (TIME . COUNT), where
6450 TIME is the parsed time as a vector, and COUNT is the number of
6451 characters from the string left unparsed. This function used to
6452 return the remaining characters as a string.
6454 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6455 The return value is now (SECONDS . MICROSECONDS); the fractional
6456 component is no longer expressed in "ticks".
6458 *** The `ticks/sec' constant has been removed, in light of the above change.
6460 * Changes to the gh_ interface
6462 ** gh_eval_str() now returns an SCM object which is the result of the
6465 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6468 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6469 and returns the array
6471 ** gh_scm2str0() is gone: there is no need to distinguish
6472 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6473 the user to interpret the data both ways.
6475 * Changes to the scm_ interface
6477 ** The new function scm_symbol_value0 provides an easy way to get a
6478 symbol's value from C code:
6480 SCM scm_symbol_value0 (char *NAME)
6481 Return the value of the symbol named by the null-terminated string
6482 NAME in the current module. If the symbol named NAME is unbound in
6483 the current module, return SCM_UNDEFINED.
6485 ** The new function scm_sysintern0 creates new top-level variables,
6486 without assigning them a value.
6488 SCM scm_sysintern0 (char *NAME)
6489 Create a new Scheme top-level variable named NAME. NAME is a
6490 null-terminated string. Return the variable's value cell.
6492 ** The function scm_internal_catch is the guts of catch. It handles
6493 all the mechanics of setting up a catch target, invoking the catch
6494 body, and perhaps invoking the handler if the body does a throw.
6496 The function is designed to be usable from C code, but is general
6497 enough to implement all the semantics Guile Scheme expects from throw.
6499 TAG is the catch tag. Typically, this is a symbol, but this function
6500 doesn't actually care about that.
6502 BODY is a pointer to a C function which runs the body of the catch;
6503 this is the code you can throw from. We call it like this:
6504 BODY (BODY_DATA, JMPBUF)
6506 BODY_DATA is just the BODY_DATA argument we received; we pass it
6507 through to BODY as its first argument. The caller can make
6508 BODY_DATA point to anything useful that BODY might need.
6509 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6510 which we have just created and initialized.
6512 HANDLER is a pointer to a C function to deal with a throw to TAG,
6513 should one occur. We call it like this:
6514 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6516 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6517 same idea as BODY_DATA above.
6518 THROWN_TAG is the tag that the user threw to; usually this is
6519 TAG, but it could be something else if TAG was #t (i.e., a
6520 catch-all), or the user threw to a jmpbuf.
6521 THROW_ARGS is the list of arguments the user passed to the THROW
6524 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6525 is just a pointer we pass through to HANDLER. We don't actually
6526 use either of those pointers otherwise ourselves. The idea is
6527 that, if our caller wants to communicate something to BODY or
6528 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6529 HANDLER can then use. Think of it as a way to make BODY and
6530 HANDLER closures, not just functions; MUMBLE_DATA points to the
6533 Of course, it's up to the caller to make sure that any data a
6534 MUMBLE_DATA needs is protected from GC. A common way to do this is
6535 to make MUMBLE_DATA a pointer to data stored in an automatic
6536 structure variable; since the collector must scan the stack for
6537 references anyway, this assures that any references in MUMBLE_DATA
6540 ** The new function scm_internal_lazy_catch is exactly like
6541 scm_internal_catch, except:
6543 - It does not unwind the stack (this is the major difference).
6544 - If handler returns, its value is returned from the throw.
6545 - BODY always receives #f as its JMPBUF argument (since there's no
6546 jmpbuf associated with a lazy catch, because we don't unwind the
6549 ** scm_body_thunk is a new body function you can pass to
6550 scm_internal_catch if you want the body to be like Scheme's `catch'
6551 --- a thunk, or a function of one argument if the tag is #f.
6553 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6554 contains the Scheme procedure to invoke as the body, and the tag
6555 we're catching. If the tag is #f, then we pass JMPBUF (created by
6556 scm_internal_catch) to the body procedure; otherwise, the body gets
6559 ** scm_handle_by_proc is a new handler function you can pass to
6560 scm_internal_catch if you want the handler to act like Scheme's catch
6561 --- call a procedure with the tag and the throw arguments.
6563 If the user does a throw to this catch, this function runs a handler
6564 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6565 variable holding the Scheme procedure object to invoke. It ought to
6566 be a pointer to an automatic variable (i.e., one living on the stack),
6567 or the procedure object should be otherwise protected from GC.
6569 ** scm_handle_by_message is a new handler function to use with
6570 `scm_internal_catch' if you want Guile to print a message and die.
6571 It's useful for dealing with throws to uncaught keys at the top level.
6573 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6574 message header to print; if zero, we use "guile" instead. That
6575 text is followed by a colon, then the message described by ARGS.
6577 ** The return type of scm_boot_guile is now void; the function does
6578 not return a value, and indeed, never returns at all.
6580 ** The new function scm_shell makes it easy for user applications to
6581 process command-line arguments in a way that is compatible with the
6582 stand-alone guile interpreter (which is in turn compatible with SCSH,
6585 To use the scm_shell function, first initialize any guile modules
6586 linked into your application, and then call scm_shell with the values
6587 of ARGC and ARGV your `main' function received. scm_shell will add
6588 any SCSH-style meta-arguments from the top of the script file to the
6589 argument vector, and then process the command-line arguments. This
6590 generally means loading a script file or starting up an interactive
6591 command interpreter. For details, see "Changes to the stand-alone
6594 ** The new functions scm_get_meta_args and scm_count_argv help you
6595 implement the SCSH-style meta-argument, `\'.
6597 char **scm_get_meta_args (int ARGC, char **ARGV)
6598 If the second element of ARGV is a string consisting of a single
6599 backslash character (i.e. "\\" in Scheme notation), open the file
6600 named by the following argument, parse arguments from it, and return
6601 the spliced command line. The returned array is terminated by a
6604 For details of argument parsing, see above, under "guile now accepts
6605 command-line arguments compatible with SCSH..."
6607 int scm_count_argv (char **ARGV)
6608 Count the arguments in ARGV, assuming it is terminated by a null
6611 For an example of how these functions might be used, see the source
6612 code for the function scm_shell in libguile/script.c.
6614 You will usually want to use scm_shell instead of calling this
6617 ** The new function scm_compile_shell_switches turns an array of
6618 command-line arguments into Scheme code to carry out the actions they
6619 describe. Given ARGC and ARGV, it returns a Scheme expression to
6620 evaluate, and calls scm_set_program_arguments to make any remaining
6621 command-line arguments available to the Scheme code. For example,
6622 given the following arguments:
6624 -e main -s ekko a speckled gecko
6626 scm_set_program_arguments will return the following expression:
6628 (begin (load "ekko") (main (command-line)) (quit))
6630 You will usually want to use scm_shell instead of calling this
6633 ** The function scm_shell_usage prints a usage message appropriate for
6634 an interpreter that uses scm_compile_shell_switches to handle its
6635 command-line arguments.
6637 void scm_shell_usage (int FATAL, char *MESSAGE)
6638 Print a usage message to the standard error output. If MESSAGE is
6639 non-zero, write it before the usage message, followed by a newline.
6640 If FATAL is non-zero, exit the process, using FATAL as the
6641 termination status. (If you want to be compatible with Guile,
6642 always use 1 as the exit status when terminating due to command-line
6645 You will usually want to use scm_shell instead of calling this
6648 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6649 expressions. It used to return SCM_EOL. Earth-shattering.
6651 ** The macros for declaring scheme objects in C code have been
6652 rearranged slightly. They are now:
6654 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6655 Declare a static SCM variable named C_NAME, and initialize it to
6656 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6657 be a C identifier, and SCHEME_NAME should be a C string.
6659 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6660 Just like SCM_SYMBOL, but make C_NAME globally visible.
6662 SCM_VCELL (C_NAME, SCHEME_NAME)
6663 Create a global variable at the Scheme level named SCHEME_NAME.
6664 Declare a static SCM variable named C_NAME, and initialize it to
6665 point to the Scheme variable's value cell.
6667 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6668 Just like SCM_VCELL, but make C_NAME globally visible.
6670 The `guile-snarf' script writes initialization code for these macros
6671 to its standard output, given C source code as input.
6673 The SCM_GLOBAL macro is gone.
6675 ** The scm_read_line and scm_read_line_x functions have been replaced
6676 by Scheme code based on the %read-delimited! procedure (known to C
6677 code as scm_read_delimited_x). See its description above for more
6680 ** The function scm_sys_open has been renamed to scm_open. It now
6681 returns a port instead of an FD object.
6683 * The dynamic linking support has changed. For more information, see
6684 libguile/DYNAMIC-LINKING.
6689 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6692 * Changes to the 'guile' program:
6694 ** Guile now loads some new files when it starts up. Guile first
6695 searches the load path for init.scm, and loads it if found. Then, if
6696 Guile is not being used to execute a script, and the user's home
6697 directory contains a file named `.guile', Guile loads that.
6699 ** You can now use Guile as a shell script interpreter.
6701 To paraphrase the SCSH manual:
6703 When Unix tries to execute an executable file whose first two
6704 characters are the `#!', it treats the file not as machine code to
6705 be directly executed by the native processor, but as source code
6706 to be executed by some interpreter. The interpreter to use is
6707 specified immediately after the #! sequence on the first line of
6708 the source file. The kernel reads in the name of the interpreter,
6709 and executes that instead. It passes the interpreter the source
6710 filename as its first argument, with the original arguments
6711 following. Consult the Unix man page for the `exec' system call
6712 for more information.
6714 Now you can use Guile as an interpreter, using a mechanism which is a
6715 compatible subset of that provided by SCSH.
6717 Guile now recognizes a '-s' command line switch, whose argument is the
6718 name of a file of Scheme code to load. It also treats the two
6719 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6720 to make a file of Scheme code directly executable by Unix, insert the
6721 following two lines at the top of the file:
6723 #!/usr/local/bin/guile -s
6726 Guile treats the argument of the `-s' command-line switch as the name
6727 of a file of Scheme code to load, and treats the sequence `#!' as the
6728 start of a block comment, terminated by `!#'.
6730 For example, here's a version of 'echo' written in Scheme:
6732 #!/usr/local/bin/guile -s
6734 (let loop ((args (cdr (program-arguments))))
6737 (display (car args))
6738 (if (pair? (cdr args))
6740 (loop (cdr args)))))
6743 Why does `#!' start a block comment terminated by `!#', instead of the
6744 end of the line? That is the notation SCSH uses, and although we
6745 don't yet support the other SCSH features that motivate that choice,
6746 we would like to be backward-compatible with any existing Guile
6747 scripts once we do. Furthermore, if the path to Guile on your system
6748 is too long for your kernel, you can start the script with this
6752 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6755 Note that some very old Unix systems don't support the `#!' syntax.
6758 ** You can now run Guile without installing it.
6760 Previous versions of the interactive Guile interpreter (`guile')
6761 couldn't start up unless Guile's Scheme library had been installed;
6762 they used the value of the environment variable `SCHEME_LOAD_PATH'
6763 later on in the startup process, but not to find the startup code
6764 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6767 To run Guile without installing it, build it in the normal way, and
6768 then set the environment variable `SCHEME_LOAD_PATH' to a
6769 colon-separated list of directories, including the top-level directory
6770 of the Guile sources. For example, if you unpacked Guile so that the
6771 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6774 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6777 ** Guile's read-eval-print loop no longer prints #<unspecified>
6778 results. If the user wants to see this, she can evaluate the
6779 expression (assert-repl-print-unspecified #t), perhaps in her startup
6782 ** Guile no longer shows backtraces by default when an error occurs;
6783 however, it does display a message saying how to get one, and how to
6784 request that they be displayed by default. After an error, evaluate
6786 to see a backtrace, and
6787 (debug-enable 'backtrace)
6788 to see them by default.
6792 * Changes to Guile Scheme:
6794 ** Guile now distinguishes between #f and the empty list.
6796 This is for compatibility with the IEEE standard, the (possibly)
6797 upcoming Revised^5 Report on Scheme, and many extant Scheme
6800 Guile used to have #f and '() denote the same object, to make Scheme's
6801 type system more compatible with Emacs Lisp's. However, the change
6802 caused too much trouble for Scheme programmers, and we found another
6803 way to reconcile Emacs Lisp with Scheme that didn't require this.
6806 ** Guile's delq, delv, delete functions, and their destructive
6807 counterparts, delq!, delv!, and delete!, now remove all matching
6808 elements from the list, not just the first. This matches the behavior
6809 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6810 functions which inspired them.
6812 I recognize that this change may break code in subtle ways, but it
6813 seems best to make the change before the FSF's first Guile release,
6817 ** The compiled-library-path function has been deleted from libguile.
6819 ** The facilities for loading Scheme source files have changed.
6821 *** The variable %load-path now tells Guile which directories to search
6822 for Scheme code. Its value is a list of strings, each of which names
6825 *** The variable %load-extensions now tells Guile which extensions to
6826 try appending to a filename when searching the load path. Its value
6827 is a list of strings. Its default value is ("" ".scm").
6829 *** (%search-load-path FILENAME) searches the directories listed in the
6830 value of the %load-path variable for a Scheme file named FILENAME,
6831 with all the extensions listed in %load-extensions. If it finds a
6832 match, then it returns its full filename. If FILENAME is absolute, it
6833 returns it unchanged. Otherwise, it returns #f.
6835 %search-load-path will not return matches that refer to directories.
6837 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6838 uses %seach-load-path to find a file named FILENAME, and loads it if
6839 it finds it. If it can't read FILENAME for any reason, it throws an
6842 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6845 *** load uses the same searching semantics as primitive-load.
6847 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6848 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6849 path, and load-module-with-path have been deleted. The functions
6850 above should serve their purposes.
6852 *** If the value of the variable %load-hook is a procedure,
6853 `primitive-load' applies its value to the name of the file being
6854 loaded (without the load path directory name prepended). If its value
6855 is #f, it is ignored. Otherwise, an error occurs.
6857 This is mostly useful for printing load notification messages.
6860 ** The function `eval!' is no longer accessible from the scheme level.
6861 We can't allow operations which introduce glocs into the scheme level,
6862 because Guile's type system can't handle these as data. Use `eval' or
6863 `read-and-eval!' (see below) as replacement.
6865 ** The new function read-and-eval! reads an expression from PORT,
6866 evaluates it, and returns the result. This is more efficient than
6867 simply calling `read' and `eval', since it is not necessary to make a
6868 copy of the expression for the evaluator to munge.
6870 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6871 for the `read' function.
6874 ** The function `int?' has been removed; its definition was identical
6875 to that of `integer?'.
6877 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6878 use the R4RS names for these functions.
6880 ** The function object-properties no longer returns the hash handle;
6881 it simply returns the object's property list.
6883 ** Many functions have been changed to throw errors, instead of
6884 returning #f on failure. The point of providing exception handling in
6885 the language is to simplify the logic of user code, but this is less
6886 useful if Guile's primitives don't throw exceptions.
6888 ** The function `fileno' has been renamed from `%fileno'.
6890 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6893 * Changes to Guile's C interface:
6895 ** The library's initialization procedure has been simplified.
6896 scm_boot_guile now has the prototype:
6898 void scm_boot_guile (int ARGC,
6900 void (*main_func) (),
6903 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6904 MAIN_FUNC should do all the work of the program (initializing other
6905 packages, reading user input, etc.) before returning. When MAIN_FUNC
6906 returns, call exit (0); this function never returns. If you want some
6907 other exit value, MAIN_FUNC may call exit itself.
6909 scm_boot_guile arranges for program-arguments to return the strings
6910 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6911 scm_set_program_arguments with the final list, so Scheme code will
6912 know which arguments have been processed.
6914 scm_boot_guile establishes a catch-all catch handler which prints an
6915 error message and exits the process. This means that Guile exits in a
6916 coherent way when system errors occur and the user isn't prepared to
6917 handle it. If the user doesn't like this behavior, they can establish
6918 their own universal catcher in MAIN_FUNC to shadow this one.
6920 Why must the caller do all the real work from MAIN_FUNC? The garbage
6921 collector assumes that all local variables of type SCM will be above
6922 scm_boot_guile's stack frame on the stack. If you try to manipulate
6923 SCM values after this function returns, it's the luck of the draw
6924 whether the GC will be able to find the objects you allocate. So,
6925 scm_boot_guile function exits, rather than returning, to discourage
6926 people from making that mistake.
6928 The IN, OUT, and ERR arguments were removed; there are other
6929 convenient ways to override these when desired.
6931 The RESULT argument was deleted; this function should never return.
6933 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
6937 ** Guile's header files should no longer conflict with your system's
6940 In order to compile code which #included <libguile.h>, previous
6941 versions of Guile required you to add a directory containing all the
6942 Guile header files to your #include path. This was a problem, since
6943 Guile's header files have names which conflict with many systems'
6946 Now only <libguile.h> need appear in your #include path; you must
6947 refer to all Guile's other header files as <libguile/mumble.h>.
6948 Guile's installation procedure puts libguile.h in $(includedir), and
6949 the rest in $(includedir)/libguile.
6952 ** Two new C functions, scm_protect_object and scm_unprotect_object,
6953 have been added to the Guile library.
6955 scm_protect_object (OBJ) protects OBJ from the garbage collector.
6956 OBJ will not be freed, even if all other references are dropped,
6957 until someone does scm_unprotect_object (OBJ). Both functions
6960 Note that calls to scm_protect_object do not nest. You can call
6961 scm_protect_object any number of times on a given object, and the
6962 next call to scm_unprotect_object will unprotect it completely.
6964 Basically, scm_protect_object and scm_unprotect_object just
6965 maintain a list of references to things. Since the GC knows about
6966 this list, all objects it mentions stay alive. scm_protect_object
6967 adds its argument to the list; scm_unprotect_object remove its
6968 argument from the list.
6971 ** scm_eval_0str now returns the value of the last expression
6974 ** The new function scm_read_0str reads an s-expression from a
6975 null-terminated string, and returns it.
6977 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
6978 to a Scheme port object.
6980 ** The new function `scm_set_program_arguments' allows C code to set
6981 the value returned by the Scheme `program-arguments' function.
6986 * Guile no longer includes sophisticated Tcl/Tk support.
6988 The old Tcl/Tk support was unsatisfying to us, because it required the
6989 user to link against the Tcl library, as well as Tk and Guile. The
6990 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
6991 referring to widgets by names, rather than exporting widgets to Scheme
6992 code as a special datatype.
6994 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
6995 maintainers described some very interesting changes in progress to the
6996 Tcl/Tk internals, which would facilitate clean interfaces between lone
6997 Tk and other interpreters --- even for garbage-collected languages
6998 like Scheme. They expected the new Tk to be publicly available in the
7001 Since it seems that Guile might soon have a new, cleaner interface to
7002 lone Tk, and that the old Guile/Tk glue code would probably need to be
7003 completely rewritten, we (Jim Blandy and Richard Stallman) have
7004 decided not to support the old code. We'll spend the time instead on
7005 a good interface to the newer Tk, as soon as it is available.
7007 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7010 Copyright information:
7012 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
7014 Permission is granted to anyone to make or distribute verbatim copies
7015 of this document as received, in any medium, provided that the
7016 copyright notice and this permission notice are preserved,
7017 thus giving the recipient permission to redistribute in turn.
7019 Permission is granted to distribute modified versions
7020 of this document, or of portions of it,
7021 under the above conditions, provided also that they
7022 carry prominent notices stating who last changed them.
7027 paragraph-separate: "[
\f]*$"