1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 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
63 The default is "pthreads", unless your platform doesn't have pthreads,
64 in which case "null" threads are used.
66 See the manual for details, nodes "Initialization", "Multi-Threading",
67 "Blocking", and others.
69 ** There is the new notion of 'discouraged' features.
71 This is a milder form of deprecation.
73 Things that are discouraged should not be used in new code, but it is
74 OK to leave them in old code for now. When a discouraged feature is
75 used, no warning message is printed like there is for 'deprecated'
76 features. Also, things that are merely discouraged are nevertheless
77 implemented efficiently, while deprecated features can be very slow.
79 You can omit discouraged features from libguile by configuring it with
80 the '--disable-discouraged' option.
82 ** Deprecation warnings can be controlled at run-time.
84 (debug-enable 'warn-deprecated) switches them on and (debug-disable
85 'warn-deprecated) switches them off.
87 ** New module (ice-9 serialize):
89 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when
90 you don't trust the thread safety of most of your program, but
91 where you have some section(s) of code which you consider can run
92 in parallel to other sections.
94 ### move rest to manual
96 They "flag" (with dynamic extent) sections of code to be of
97 "serial" or "parallel" nature and have the single effect of
98 preventing a serial section from being run in parallel with any
99 serial section (including itself).
101 Both serialize and parallelize can be nested. If so, the
102 inner-most construct is in effect.
104 NOTE 1: A serial section can run in parallel with a parallel
107 NOTE 2: If a serial section S is "interrupted" by a parallel
108 section P in the following manner: S = S1 P S2, S2 is not
109 guaranteed to be resumed by the same thread that previously
112 WARNING: Spawning new threads within a serial section have
113 undefined effects. It is OK, though, to spawn threads in unflagged
114 sections of code where neither serialize or parallelize is in
117 A typical usage is when Guile is used as scripting language in some
118 application doing heavy computations. If each thread is
119 encapsulated with a serialize form, you can then put a parallelize
120 form around the code performing the heavy computations (typically a
121 C code primitive), enabling the computations to run in parallel
122 while the scripting code runs single-threadedly.
124 ** Support for require-extension, SRFI-55, has been added.
126 The SRFI-55 special form `require-extension' has been added. It is
127 available at startup, and provides a portable way to load Scheme
128 extensions. SRFI-55 only requires support for one type of extension,
129 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
132 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
134 The (srfi srfi-26) module is an implementation of SRFI-26 which
135 provides the `cut' and `cute' syntax. These may be used to specialize
136 parameters without currying.
138 ** New module (srfi srfi-31)
140 This is an implementation of SRFI-31 which provides a special form
141 `rec' for recursive evaluation.
143 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
144 been merged with the core, making their functionality always
147 The modules are still available, tho, and you could use them together
148 with a renaming import, for example.
150 ** Guile no longer includes its own version of libltdl.
152 The official version is good enough now.
154 ** The --enable-htmldoc option has been removed from 'configure'.
156 Support for translating the documentation into HTML is now always
157 provided. Use 'make html'.
159 ** The configure option '--disable-arrays' has been removed.
161 Support for arrays and uniform numeric arrays is now always included
164 * Changes to the stand-alone interpreter
166 ** New command line option `-L'.
168 This option adds a directory to the front of the load path.
170 ** New command line option `--no-debug'.
172 Specifying `--no-debug' on the command line will keep the debugging
173 evaluator turned off, even for interactive sessions.
175 ** User-init file ~/.guile is now loaded with the debugging evaluator.
177 Previously, the normal evaluator would have been used. Using the
178 debugging evaluator gives better error messages.
180 ** The '-e' option now 'read's its argument.
182 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
183 be used with '-e'. For example, you can now write a script like
186 exec guile -e '(@ (demo) main)' -s "$0" "$@"
189 (define-module (demo)
193 (format #t "Demo: ~a~%" args))
196 * Changes to Scheme functions and syntax
198 ** New function hashx-remove!
200 This function completes the set of 'hashx' functions.
202 ** The concept of dynamic roots has been factored into continuation
203 barriers and dynamic states.
205 Each thread has a current dynamic state that carries the values of the
206 fluids. You can create and copy dynamic states and use them as the
207 second argument for 'eval'. See "Fluids and Dynamic States" in the
210 To restrict the influence that captured continuations can have on the
211 control flow, you can errect continuation barriers. See "Continuation
212 Barriers" in the manual.
214 The function call-with-dynamic-root now essentially temporarily
215 installs a new dynamic state and errects a continuation barrier.
217 ** The default load path no longer includes "." at the end.
219 Automatically loading modules from the current directory should not
220 happen by default. If you want to allow it in a more controlled
221 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
224 ** The uniform vector and array support has been overhauled.
226 It now complies with SRFI-4 and the weird prototype based uniform
227 array creation has been deprecated. See the manual for more details.
229 Some non-compatible changes have been made:
230 - characters can no longer be stored into byte arrays.
231 - strings and bit vectors are no longer considered to be uniform vectors.
232 - array-rank throws an error for non-arrays instead of returning zero.
233 - array-ref does no longer accept non-arrays when no indices are given.
235 There is the new notion of 'generalized vectors' and corresponding
236 procedures like 'generalized-vector-ref'. Generalized vectors include
237 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
239 Arrays use generalized vectors as their storage, so that you still
240 have arrays of characters, bits, etc. However, uniform-array-read!
241 and uniform-array-write can no longer read/write strings and
244 ** There is now support for copy-on-write substrings, mutation-sharing
245 substrings and read-only strings.
247 Three new procedures are related to this: substring/shared,
248 substring/copy, and substring/read-only. See the manual for more
251 ** Backtraces will now highlight the value that caused the error.
253 By default, these values are enclosed in "{...}", such as in this
262 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
263 <unnamed port>:1:1: Wrong type (expecting pair): a
264 ABORT: (wrong-type-arg)
266 The prefix and suffix used for highlighting can be set via the two new
267 printer options 'highlight-prefix' and 'highlight-suffix'. For
268 example, putting this into ~/.guile will output the bad value in bold
271 (print-set! highlight-prefix "\x1b[1m")
272 (print-set! highlight-suffix "\x1b[22m")
275 ** 'gettext' support for internationalization has been added.
277 See the manual for details.
279 ** New syntax '@' and '@@':
281 You can now directly refer to variables exported from a module by
284 (@ MODULE-NAME VARIABLE-NAME)
286 For example (@ (ice-9 pretty-print) pretty-print) will directly access
287 the pretty-print variable exported from the (ice-9 pretty-print)
288 module. You don't need to 'use' that module first. You can also use
289 '@' as a target of 'set!', as in (set! (@ mod var) val).
291 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
292 but it can also access variables that have not been exported. It is
293 intended only for kluges and temporary fixes and for debugging, not
296 ** Keyword syntax has been made more disciplined.
298 Previously, the name of a keyword was read as a 'token' but printed as
299 a symbol. Now, it is read as a general Scheme datum which must be a
310 ERROR: In expression (a b c):
314 ERROR: Unbound variable: foo
319 ERROR: Wrong type (expecting symbol): 12
323 ERROR: Wrong type (expecting symbol): (a b c)
327 ** 'while' now provides 'break' and 'continue'
329 break and continue were previously bound in a while loop, but not
330 documented, and continue didn't quite work properly. The undocumented
331 parameter to break which gave a return value for the while has been
334 ** 'call-with-current-continuation' is now also available under the name
337 ** The module system now checks for duplicate bindings.
339 The module system now can check for name conflicts among imported
342 The behavior can be controlled by specifying one or more 'duplicates'
343 handlers. For example, to make Guile return an error for every name
351 The new default behavior of the module system when a name collision
352 has been detected is to
354 1. Give priority to bindings marked as a replacement.
355 2. Issue a warning (different warning if overriding core binding).
356 3. Give priority to the last encountered binding (this corresponds to
359 If you want the old behavior back without replacements or warnings you
362 (default-duplicate-binding-handler 'last)
364 to your .guile init file.
366 ### move rest to manual
368 The syntax for the :duplicates option is:
370 :duplicates HANDLER-NAME | (HANDLER1-NAME HANDLER2-NAME ...)
372 Specifying multiple handlers is useful since some handlers (such as
373 replace) can defer conflict resolution to others. Each handler is
374 tried until a binding is selected.
376 Currently available duplicates handlers are:
378 check report an error for bindings with a common name
379 warn issue a warning for bindings with a common name
380 replace replace bindings which have an imported replacement
381 warn-override-core issue a warning for imports which override core bindings
382 and accept the override
383 first select the first encountered binding (override)
384 last select the last encountered binding (override)
386 These two are provided by the (oop goops) module:
388 merge-generics merge generic functions with a common name
389 into an <extended-generic>
390 merge-accessors merge accessors with a common name
392 The default duplicates handler is:
394 (replace warn-override-core warn last)
396 A recommended handler (which is likely to correspond to future Guile
397 behavior) can be installed with:
399 (default-duplicate-binding-handler '(replace warn-override-core check))
401 ** New define-module option: :replace
403 :replace works as :export, but, in addition, marks the binding as a
406 A typical example is `format' in (ice-9 format) which is a replacement
407 for the core binding `format'.
409 ** Adding prefixes to imported bindings in the module system
411 There is now a new :use-module option :prefix. It can be used to add
412 a prefix to all imported bindings.
415 :use-module ((bar) :prefix bar:))
417 will import all bindings exported from bar, but rename them by adding
420 ** Conflicting generic functions can be automatically merged.
422 When two imported bindings conflict and they are both generic
423 functions, the two functions can now be merged automatically. This is
424 activated with the 'duplicates' handler 'merge-generics'.
426 ### move the rest to the manual
428 It is sometimes tempting to use GOOPS accessors with short names.
429 For example, it is tempting to use the name `x' for the x-coordinate
432 Assume that we work with a graphical package which needs to use two
433 independent vector packages for 2D and 3D vectors respectively. If
434 both packages export `x' we will encounter a name collision.
436 This can now be resolved automagically with the duplicates handler
437 `merge-generics' which gives the module system license to merge all
438 generic functions sharing a common name:
440 (define-module (math 2D-vectors)
441 :use-module (oop goops)
444 (define-module (math 3D-vectors)
445 :use-module (oop goops)
448 (define-module (my-module)
449 :use-module (math 2D-vectors)
450 :use-module (math 3D-vectors)
451 :duplicates merge-generics)
453 x in (my-module) will now share methods with x in both imported
456 There will, in fact, now be three distinct generic functions named
457 `x': x in (2D-vectors), x in (3D-vectors), and x in (my-module). The
458 last function will be an <extended-generic>, extending the previous
461 Let's call the imported generic functions the "ancestor functions". x
462 in (my-module) is, in turn, a "descendant function" of the imported
463 functions, extending its ancestors.
465 For any generic function G, the applicable methods are selected from
466 the union of the methods of the descendant functions, the methods of G
467 itself and the methods of the ancestor functions.
469 This, ancestor functions share methods with their descendants and vice
470 versa. This implies that x in (math 2D-vectors) can will share the
471 methods of x in (my-module) and vice versa, while x in (math 2D-vectors)
472 doesn't share the methods of x in (math 3D-vectors), thus preserving
475 Sharing is dynamic, so that adding new methods to a descendant implies
476 adding it to the ancestor.
478 If duplicates checking is desired in the above example, the following
479 form of the :duplicates option can be used instead:
481 :duplicates (merge-generics check)
483 ** New function: effective-version
485 Returns the "effective" version number. This is just the normal full
486 version string without the final micro-version number. See "Changes
487 to the distribution" above.
489 ** New feature, 'futures': future, make-future, future-ref
491 Futures are like promises, but begin execution immediately in a new
492 thread. See the "Futures" section in the reference manual.
494 ** New threading functions: parallel, letpar, par-map, and friends
496 These are convenient ways to run calculations in parallel in new
497 threads. See "Parallel forms" in the manual for details.
499 ** New function 'try-mutex'.
501 This function will attempt to lock a mutex but will return immediately
502 instead if blocking and indicate failure.
504 ** Waiting on a condition variable can have a timeout.
506 The funtion 'wait-condition-variable' now takes a third, optional
507 argument that specifies the point in time where the waiting should be
510 ** New function 'broadcast-condition-variable'.
512 ** New functions 'all-threads' and 'current-thread'.
514 ** Signals and system asyncs work better with threads.
516 The function 'sigaction' now takes a fourth, optional, argument that
517 specifies the thread that the handler should run in. When the
518 argument is omitted, the handler will run in the thread that called
521 Likewise, 'system-async-mark' takes a second, optional, argument that
522 specifies the thread that the async should run in. When it is
523 omitted, the async will run in the thread that called
526 C code can use the new functions scm_sigaction_for_thread and
527 scm_system_async_mark_for_thread to pass the new thread argument.
529 When a thread blocks on a mutex, a condition variable or is waiting
530 for IO to be possible, it will still execute system asyncs. This can
531 be used to interrupt such a thread by making it execute a 'throw', for
534 ** The function 'system-async' is deprecated.
536 You can now pass any zero-argument procedure to 'system-async-mark'.
537 The function 'system-async' will just return its argument unchanged
540 ** New functions 'call-with-blocked-asyncs' and
541 'call-with-unblocked-asyncs'
543 The expression (call-with-blocked-asyncs PROC) will call PROC and will
544 block execution of system asyncs for the current thread by one level
545 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
546 procedure and will unblock the execution of system asyncs by one
547 level for the current thread.
549 Only system asyncs are affected by these functions.
551 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
553 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
554 instead. Those functions are easier to use correctly and can be
557 ** New function 'unsetenv'.
559 ** New macro 'define-syntax-public'.
561 It works like 'define-syntax' and also exports the defined macro (but
564 ** There is support for Infinity and NaNs.
566 Following PLT Scheme, Guile can now work with infinite numbers, and
569 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
570 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
571 "+nan.0"). These numbers are inexact and have no exact counterpart.
573 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
574 sign of the dividend. The infinities are integers, and they answer #t
575 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
576 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
587 ERROR: Numerical overflow
589 Two new predicates 'inf?' and 'nan?' can be used to test for the
592 ** Inexact zero can have a sign.
594 Guile can now distinguish between plus and minus inexact zero, if your
595 platform supports this, too. The two zeros are equal according to
596 '=', but not according to 'eqv?'. For example
607 ** Guile now has exact rationals.
609 Guile can now represent fractions such as 1/3 exactly. Computing with
610 them is also done exactly, of course:
615 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
618 For example: (floor 2) now returns an exact 2 where in the past it
619 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
621 ** inexact->exact no longer returns only integers.
623 Without exact rationals, the closest exact number was always an
624 integer, but now inexact->exact returns the fraction that is exactly
625 equal to a floating point number. For example:
627 (inexact->exact 1.234)
628 => 694680242521899/562949953421312
630 When you want the old behavior, use 'round' explicitely:
632 (inexact->exact (round 1.234))
635 ** New function 'rationalize'.
637 This function finds a simple fraction that is close to a given real
638 number. For example (and compare with inexact->exact above):
640 (rationalize (inexact->exact 1.234) 1/2000)
643 Note that, as required by R5RS, rationalize returns only then an exact
644 result when both its arguments are exact.
646 ** 'odd?' and 'even?' work also for inexact integers.
648 Previously, (odd? 1.0) would signal an error since only exact integers
649 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
650 returns #f and (odd? 1.5) signals an error.
652 ** Guile now has uninterned symbols.
654 The new function 'make-symbol' will return an uninterned symbol. This
655 is a symbol that is unique and is guaranteed to remain unique.
656 However, uninterned symbols can not yet be read back in.
658 Use the new function 'symbol-interned?' to check whether a symbol is
661 ** pretty-print has more options.
663 The function pretty-print from the (ice-9 pretty-print) module can now
664 also be invoked with keyword arguments that control things like
665 maximum output width. See the manual for details.
667 ** Variables have no longer a special behavior for `equal?'.
669 Previously, comparing two variables with `equal?' would recursivly
670 compare their values. This is no longer done. Variables are now only
671 `equal?' if they are `eq?'.
673 ** `(begin)' is now valid.
675 You can now use an empty `begin' form. It will yield #<unspecified>
676 when evaluated and simply be ignored in a definition context.
678 ** Deprecated: procedure->macro
680 Change your code to use 'define-macro' or r5rs macros. Also, be aware
681 that macro expansion will not be done during evaluation, but prior to
684 ** Soft ports now allow a `char-ready?' procedure
686 The vector argument to `make-soft-port' can now have a length of
687 either 5 or 6. (Previously the length had to be 5.) The optional 6th
688 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
689 that returns the number of characters that can be read immediately
690 without the soft port blocking.
692 ** New debugging feature: breakpoints.
694 Guile now has breakpoints. For details see the `Debugging Features'
695 chapter in the reference manual.
697 ** Deprecated: undefine
699 There is no replacement for undefine.
701 * The functions make-keyword-from-dash-symbol and keyword-dash-symbol
702 have been discouraged.
704 They are relics from a time where a keyword like #:foo was used
705 directly as a Tcl option "-foo" and thus keywords were internally
706 stored as a symbol with a starting dash. We now store a symbol
709 Use symbol->keyword and keyword->symbol instead.
712 * Changes to the C interface
714 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
715 take a 'delete' function argument.
717 This argument makes no sense since the delete function is used to
718 remove a pair from an alist, and this must not be configurable.
720 This is an incompatible change.
722 ** A new family of functions for converting between C values and
723 Scheme values has been added.
725 These functions follow a common naming scheme and are designed to be
726 easier to use, thread-safe and more future-proof than the older
731 These are predicates that return a C boolean: 1 or 0. Instead of
732 SCM_NFALSEP, you can now use scm_is_true, for example.
734 - <type> scm_to_<type> (SCM val, ...)
736 These are functions that convert a Scheme value into an appropriate
737 C value. For example, you can use scm_to_int to safely convert from
740 - SCM scm_from_<type> (<type> val, ...)
742 These functions convert from a C type to a SCM value; for example,
743 scm_from_int for ints.
745 There is a huge number of these functions, for numbers, strings,
746 symbols, vectors, etc. They are documented in the reference manual in
747 the API section together with the types that they apply to.
749 ** New functions for dealing with complex numbers in C have been added.
751 The new functions are scm_c_make_rectangular, scm_c_make_polar,
752 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
753 They work like scm_make_rectangular etc but take or return doubles
756 ** The function scm_make_complex has been discouraged.
758 Use scm_c_make_rectangular instead.
760 ** The INUM macros have been deprecated.
762 A lot of code uses these macros to do general integer conversions,
763 although the macros only work correctly with fixnums. Use the
764 following alternatives.
766 SCM_INUMP -> scm_is_integer or similar
767 SCM_NINUMP -> !scm_is_integer or similar
768 SCM_MAKINUM -> scm_from_int or similar
769 SCM_INUM -> scm_to_int or similar
771 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
772 do the validating for you.
774 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
775 have been discouraged.
777 Use the newer scm_to_<type> and scm_from_<type> functions instead for
778 new code. The functions have been discouraged since they don't fit
781 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
783 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
784 evaluates its argument twice. Use scm_is_true, etc. instead for new
787 ** The macro SCM_EQ_P has been discouraged.
789 Use scm_is_eq for new code, which fits better into the naming
792 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
795 Use the function scm_is_pair or scm_is_null instead.
797 ** The functions scm_round and scm_truncate have been deprecated and
798 are now available as scm_c_round and scm_c_truncate, respectively.
800 These functions occupy the names that scm_round_number and
801 scm_truncate_number should have.
803 ** The functions scm_c_string2str, scm_c_substring2str, and
804 scm_c_symbol2str have been deprecated.
806 Use scm_to_locale_stringbuf or similar instead, maybe together with
809 ** New functions scm_c_make_string, scm_c_string_length,
810 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
811 scm_c_substring_shared, scm_c_substring_copy.
813 These are like scm_make_string, scm_length, etc. but are slightly
814 easier to use from C.
816 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
817 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
819 They export too many assumptions about the implementation of strings
820 and symbols that are no longer true in the presence of
821 mutation-sharing substrings and when Guile switches to some form of
824 When working with strings, it is often best to use the normal string
825 functions provided by Guile, such as scm_c_string_ref,
826 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
827 manual since many more such functions are now provided than
830 When you want to convert a SCM string to a C string, use the
831 scm_to_locale_string function or similar instead. For symbols, use
832 scm_symbol_to_string and then work with that string. Because of the
833 new string representation, scm_symbol_to_string does not need to copy
834 and is thus quite efficient.
836 ** Some string, symbol and keyword functions have been discouraged.
838 They don't fit into the uniform naming scheme and are not explicit
839 about the character encoding.
841 Replace according to the following table:
843 scm_allocate_string -> scm_c_make_string
844 scm_take_str -> scm_take_locale_stringn
845 scm_take0str -> scm_take_locale_string
846 scm_mem2string -> scm_from_locale_stringn
847 scm_str2string -> scm_from_locale_string
848 scm_makfrom0str -> scm_from_locale_string
849 scm_mem2symbol -> scm_from_locale_symboln
850 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
851 scm_str2symbol -> scm_from_locale_symbol
853 SCM_SYMBOL_HASH -> scm_hashq
854 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
856 scm_c_make_keyword -> scm_from_locale_keyword
858 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
859 now also available to C code.
861 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
863 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
864 the latter returns the true name of the keyword, not the 'dash name',
865 as SCM_KEYWORDSYM used to do.
867 ** A new way to access arrays in a thread-safe and efficient way has
870 See the manual, node "Accessing Arrays From C".
872 ** The old uniform vector and bitvector implementations have been
873 unceremoniously removed.
875 This implementation exposed the details of the tagging system of
876 Guile. Use the new C API explained in the manual in node "Uniform
877 Numeric Vectors" and "Bit Vectors", respectively.
879 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
880 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
881 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
882 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
883 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
884 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
887 ** The macros dealing with vectors have been deprecated.
889 Use the new functions scm_is_vector, scm_vector_elements,
890 scm_vector_writable_elements, etc, or scm_is_simple_vector,
891 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
892 manual for more details.
894 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
895 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
897 The following macros have been removed: SCM_VECTOR_BASE,
898 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
899 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
901 ** Some C functions and macros related to arrays have been deprecated.
903 Migrate according to the following table:
905 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
906 scm_make_ra -> scm_make_array
907 scm_shap2ra -> scm_make_array
908 scm_cvref -> scm_c_generalized_vector_ref
909 scm_ra_set_contp -> do not use
910 scm_aind -> scm_array_handle_pos
911 scm_raprin1 -> scm_display or scm_write
913 SCM_ARRAYP -> scm_is_array
914 SCM_ARRAY_NDIM -> scm_c_array_rank
915 SCM_ARRAY_DIMS -> scm_array_handle_dims
916 SCM_ARRAY_CONTP -> do not use
917 SCM_ARRAY_MEM -> do not use
918 SCM_ARRAY_V -> scm_array_handle_elements or similar
919 SCM_ARRAY_BASE -> do not use
921 ** SCM_CELL_WORD_LOC has been deprecated.
923 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
924 to a SCM, as opposed to a pointer to a scm_t_bits.
926 This was done to allow the correct use of pointers into the Scheme
927 heap. Previously, the heap words were of type scm_t_bits and local
928 variables and function arguments were of type SCM, making it
929 non-standards-conformant to have a pointer that can point to both.
931 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
933 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
934 second and third words of double smobs. Likewise for
935 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
937 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
938 used to get and set the 16 exra bits in the zeroth word of a smob.
940 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
941 accesing the first immediate word of a smob as a SCM value, and there
942 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
943 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
945 ** New way to deal with non-local exits and re-entries.
947 There is a new set of functions that essentially do what
948 scm_internal_dynamic_wind does, but in a way that is more convenient
949 for C code in some situations. Here is a quick example of how to
950 prevent a potential memory leak:
959 mem = scm_malloc (100);
960 scm_frame_unwind_handler (free, mem, SCM_F_WIND_EXPLICITELY);
962 /* MEM would leak if BAR throws an error.
963 SCM_FRAME_UNWIND_HANDLER frees it nevertheless.
970 /* Because of SCM_F_WIND_EXPLICITELY, MEM will be freed by
971 SCM_FRAME_END as well.
975 For full documentation, see the node "Frames" in the manual.
977 ** New function scm_frame_free
979 This function calls 'free' on a given pointer when a frame is left.
980 Thus the call to scm_frame_unwind_handler above could be replaced with
981 simply scm_frame_free (mem).
983 ** New functions scm_c_call_with_blocked_asyncs and
984 scm_c_call_with_unblocked_asyncs
986 Like scm_call_with_blocked_asyncs etc. but for C functions.
988 ** New functions scm_frame_block_asyncs and scm_frame_unblock_asyncs
990 In addition to scm_c_call_with_blocked_asyncs you can now also use
991 scm_frame_block_asyncs in a 'frame' (see above). Likewise for
992 scm_c_call_with_unblocked_asyncs and scm_frame_unblock_asyncs.
994 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
995 SCM_REALLOW_INTS have been deprecated.
997 They do no longer fulfill their original role of blocking signal
998 delivery. Depending on what you want to achieve, replace a pair of
999 SCM_DEFER_INTS and SCM_ALLOW_INTS with a frame that locks a mutex,
1000 blocks asyncs, or both. See node "Critical Sections" in the manual.
1002 ** The value 'scm_mask_ints' is no longer writable.
1004 Previously, you could set scm_mask_ints directly. This is no longer
1005 possible. Use scm_c_call_with_blocked_asyncs and
1006 scm_c_call_with_unblocked_asyncs instead.
1008 ** New way to temporarily set the current input, output or error ports
1010 C code can now use scm_frame_current_<foo>_port in a 'frame' (see
1011 above). <foo> is one of "input", "output" or "error".
1013 ** New way to temporarily set fluids
1015 C code can now use scm_frame_fluid in a 'frame' (see
1016 above) to temporarily set the value of a fluid.
1018 ** New types scm_t_intmax and scm_t_uintmax.
1020 On platforms that have them, these types are identical to intmax_t and
1021 uintmax_t, respectively. On other platforms, they are identical to
1022 the largest integer types that Guile knows about.
1024 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1026 You should not have used them.
1028 ** Many public #defines with generic names have been made private.
1030 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1031 private or renamed with a more suitable public name.
1033 ** The macro SCM_TYP16S has been deprecated.
1035 This macro is not intended for public use.
1037 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1039 Use scm_is_true (scm_inexact_p (...)) instead.
1041 ** The macro SCM_SLOPPY_REALP has been deprecated.
1043 Use scm_is_real instead.
1045 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1047 Use scm_is_complex instead.
1049 ** Some preprocessor defines have been deprecated.
1051 These defines indicated whether a certain feature was present in Guile
1052 or not. Going forward, assume that the features are always present.
1054 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1055 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1057 The following macros have been removed completely: MEMOIZE_LOCALS,
1058 SCM_RECKLESS, SCM_CAUTIOUS.
1060 ** The preprocessor define STACK_DIRECTION has been deprecated.
1062 There should be no need to know about the stack direction for ordinary
1065 ** New function: scm_effective_version
1067 Returns the "effective" version number. This is just the normal full
1068 version string without the final micro-version number. See "Changes
1069 to the distribution" above.
1071 ** The function scm_call_with_new_thread has a new prototype.
1073 Instead of taking a list with the thunk and handler, these two
1074 arguments are now passed directly:
1076 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1078 This is an incompatible change.
1080 ** New snarfer macro SCM_DEFINE_PUBLIC.
1082 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1083 function in the init section.
1085 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1087 ** Garbage collector rewrite.
1089 The garbage collector is cleaned up a lot, and now uses lazy
1090 sweeping. This is reflected in the output of (gc-stats); since cells
1091 are being freed when they are allocated, the cells-allocated field
1092 stays roughly constant.
1094 For malloc related triggers, the behavior is changed. It uses the same
1095 heuristic as the cell-triggered collections. It may be tuned with the
1096 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1097 for minimum yield of malloc related triggers. The default is 40.
1098 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1101 Debugging operations for the freelist have been deprecated, along with
1102 the C variables that control garbage collection. The environment
1103 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1104 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1106 For understanding the memory usage of a GUILE program, the routine
1107 gc-live-object-stats returns an alist containing the number of live
1108 objects for every type.
1111 ** The function scm_definedp has been renamed to scm_defined_p
1113 The name scm_definedp is deprecated.
1115 ** The struct scm_cell type has been renamed to scm_t_cell
1117 This is in accordance to Guile's naming scheme for types. Note that
1118 the name scm_cell is now used for a function that allocates and
1119 initializes a new cell (see below).
1121 ** New functions for memory management
1123 A new set of functions for memory management has been added since the
1124 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1125 indeed, Guile itself contained some long standing bugs that could
1126 cause aborts in long running programs.
1128 The new functions are more symmetrical and do not need cooperation
1129 from smob free routines, among other improvements.
1131 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1132 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1133 scm_gc_free, scm_gc_register_collectable_memory, and
1134 scm_gc_unregister_collectable_memory. Refer to the manual for more
1135 details and for upgrading instructions.
1137 The old functions for memory management have been deprecated. They
1138 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1139 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1141 ** Declarations of exported features are marked with SCM_API.
1143 Every declaration of a feature that belongs to the exported Guile API
1144 has been marked by adding the macro "SCM_API" to the start of the
1145 declaration. This macro can expand into different things, the most
1146 common of which is just "extern" for Unix platforms. On Win32, it can
1147 be used to control which symbols are exported from a DLL.
1149 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1150 will expand into "__declspec (dllimport) extern", which is needed for
1151 linking to the Guile DLL in Windows.
1153 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1154 SCM_SRFI4_IMPORT, for the corresponding libraries.
1156 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1158 Use the new functions scm_cell and scm_double_cell instead. The old
1159 macros had problems because with them allocation and initialization
1160 was separated and the GC could sometimes observe half initialized
1161 cells. Only careful coding by the user of SCM_NEWCELL and
1162 SCM_NEWCELL2 could make this safe and efficient.
1164 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1166 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1169 ** SRCBRKP has been deprecated.
1171 Use scm_c_source_property_breakpoint_p instead.
1173 ** Deprecated: scm_makmacro
1175 Change your code to use either scm_makmmacro or to define macros in
1176 Scheme, using 'define-macro'.
1178 ** New function scm_c_port_for_each.
1180 This function is like scm_port_for_each but takes a pointer to a C
1181 function as the callback instead of a SCM value.
1183 ** The names scm_internal_select, scm_thread_sleep, and
1184 scm_thread_usleep have been discouraged.
1186 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1188 ** The GC can no longer be blocked.
1190 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1191 The GC can now run (partially) concurrently with other code and thus
1192 blocking it is not well defined.
1194 ** Many definitions have been removed that were previously deprecated.
1196 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1197 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1198 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1199 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1200 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1201 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1202 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1203 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1204 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1205 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1206 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1207 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1208 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1209 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1210 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1211 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1212 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1213 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1214 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1215 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1216 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1217 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1218 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1219 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1220 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1221 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1222 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1223 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1224 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1225 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1226 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1229 Changes since Guile 1.4:
1231 * Changes to the distribution
1233 ** A top-level TODO file is included.
1235 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1237 Guile now always uses three numbers to represent the version,
1238 i.e. "1.6.5". The first number, 1, is the major version number, the
1239 second number, 6, is the minor version number, and the third number,
1240 5, is the micro version number. Changes in major version number
1241 indicate major changes in Guile.
1243 Minor version numbers that are even denote stable releases, and odd
1244 minor version numbers denote development versions (which may be
1245 unstable). The micro version number indicates a minor sub-revision of
1246 a given MAJOR.MINOR release.
1248 In keeping with the new scheme, (minor-version) and scm_minor_version
1249 no longer return everything but the major version number. They now
1250 just return the minor version number. Two new functions
1251 (micro-version) and scm_micro_version have been added to report the
1252 micro version number.
1254 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1256 ** New preprocessor definitions are available for checking versions.
1258 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1259 SCM_MICRO_VERSION to the appropriate integer values.
1261 ** Guile now actively warns about deprecated features.
1263 The new configure option `--enable-deprecated=LEVEL' and the
1264 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1265 See INSTALL and README for more information.
1267 ** Guile is much more likely to work on 64-bit architectures.
1269 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1270 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1271 for the use of a test machine, and thanks to Stefan Jahn for ia64
1274 ** New functions: setitimer and getitimer.
1276 These implement a fairly direct interface to the libc functions of the
1279 ** The #. reader extension is now disabled by default.
1281 For safety reasons, #. evaluation is disabled by default. To
1282 re-enable it, set the fluid read-eval? to #t. For example:
1284 (fluid-set! read-eval? #t)
1286 but make sure you realize the potential security risks involved. With
1287 read-eval? enabled, reading a data file from an untrusted source can
1290 ** New SRFI modules have been added:
1292 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1295 (srfi srfi-1) is a library containing many useful pair- and list-processing
1298 (srfi srfi-2) exports and-let*.
1300 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1302 (srfi srfi-6) is a dummy module for now, since guile already provides
1303 all of the srfi-6 procedures by default: open-input-string,
1304 open-output-string, get-output-string.
1306 (srfi srfi-8) exports receive.
1308 (srfi srfi-9) exports define-record-type.
1310 (srfi srfi-10) exports define-reader-ctor and implements the reader
1313 (srfi srfi-11) exports let-values and let*-values.
1315 (srfi srfi-13) implements the SRFI String Library.
1317 (srfi srfi-14) implements the SRFI Character-Set Library.
1319 (srfi srfi-17) implements setter and getter-with-setter and redefines
1320 some accessor procedures as procedures with getters. (such as car,
1321 cdr, vector-ref etc.)
1323 (srfi srfi-19) implements the SRFI Time/Date Library.
1325 ** New scripts / "executable modules"
1327 Subdirectory "scripts" contains Scheme modules that are packaged to
1328 also be executable as scripts. At this time, these scripts are available:
1337 See README there for more info.
1339 These scripts can be invoked from the shell with the new program
1340 "guile-tools", which keeps track of installation directory for you.
1343 $ guile-tools display-commentary srfi/*.scm
1345 guile-tools is copied to the standard $bindir on "make install".
1347 ** New module (ice-9 stack-catch):
1349 stack-catch is like catch, but saves the current state of the stack in
1350 the fluid the-last-stack. This fluid can be useful when using the
1351 debugger and when re-throwing an error.
1353 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1355 This has been done to prevent problems on lesser operating systems
1356 that can't tolerate `*'s in file names. The exported macro continues
1357 to be named `and-let*', of course.
1359 On systems that support it, there is also a compatibility module named
1360 (ice-9 and-let*). It will go away in the next release.
1362 ** New modules (oop goops) etc.:
1365 (oop goops describe)
1367 (oop goops active-slot)
1368 (oop goops composite-slot)
1370 The Guile Object Oriented Programming System (GOOPS) has been
1371 integrated into Guile. For further information, consult the GOOPS
1372 manual and tutorial in the `doc' directory.
1374 ** New module (ice-9 rdelim).
1376 This exports the following procedures which were previously defined
1377 in the default environment:
1379 read-line read-line! read-delimited read-delimited! %read-delimited!
1380 %read-line write-line
1382 For backwards compatibility the definitions are still imported into the
1383 default environment in this version of Guile. However you should add:
1385 (use-modules (ice-9 rdelim))
1387 to any program which uses the definitions, since this may change in
1390 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1391 can be used for similar functionality.
1393 ** New module (ice-9 rw)
1395 This is a subset of the (scsh rw) module from guile-scsh. Currently
1396 it defines two procedures:
1398 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1400 Read characters from a port or file descriptor into a string STR.
1401 A port must have an underlying file descriptor -- a so-called
1402 fport. This procedure is scsh-compatible and can efficiently read
1405 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1407 Write characters from a string STR to a port or file descriptor.
1408 A port must have an underlying file descriptor -- a so-called
1409 fport. This procedure is mostly compatible and can efficiently
1410 write large strings.
1412 ** New module (ice-9 match)
1414 This module includes Andrew K. Wright's pattern matcher. See
1415 ice-9/match.scm for brief description or
1417 http://www.star-lab.com/wright/code.html
1419 for complete documentation.
1421 ** New module (ice-9 buffered-input)
1423 This module provides procedures to construct an input port from an
1424 underlying source of input that reads and returns its input in chunks.
1425 The underlying input source is a Scheme procedure, specified by the
1426 caller, which the port invokes whenever it needs more input.
1428 This is useful when building an input port whose back end is Readline
1429 or a UI element such as the GtkEntry widget.
1433 The reference and tutorial documentation that was previously
1434 distributed separately, as `guile-doc', is now included in the core
1435 Guile distribution. The documentation consists of the following
1438 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1441 - The Guile Reference Manual (guile.texi) contains (or is intended to
1442 contain) reference documentation on all aspects of Guile.
1444 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1445 reference documentation for using GOOPS, Guile's Object Oriented
1448 - The Revised^5 Report on the Algorithmic Language Scheme
1451 See the README file in the `doc' directory for more details.
1453 ** There are a couple of examples in the examples/ directory now.
1455 * Changes to the stand-alone interpreter
1457 ** New command line option `--use-srfi'
1459 Using this option, SRFI modules can be loaded on startup and be
1460 available right from the beginning. This makes programming portable
1461 Scheme programs easier.
1463 The option `--use-srfi' expects a comma-separated list of numbers,
1464 each representing a SRFI number to be loaded into the interpreter
1465 before starting evaluating a script file or the REPL. Additionally,
1466 the feature identifier for the loaded SRFIs is recognized by
1467 `cond-expand' when using this option.
1470 $ guile --use-srfi=8,13
1471 guile> (receive (x z) (values 1 2) (+ 1 2))
1473 guile> (string-pad "bla" 20)
1476 ** Guile now always starts up in the `(guile-user)' module.
1478 Previously, scripts executed via the `-s' option would run in the
1479 `(guile)' module and the repl would run in the `(guile-user)' module.
1480 Now every user action takes place in the `(guile-user)' module by
1483 * Changes to Scheme functions and syntax
1485 ** Character classifiers work for non-ASCII characters.
1487 The predicates `char-alphabetic?', `char-numeric?',
1488 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1489 no longer check whether their arguments are ASCII characters.
1490 Previously, a character would only be considered alphabetic when it
1491 was also ASCII, for example.
1493 ** Previously deprecated Scheme functions have been removed:
1495 tag - no replacement.
1496 fseek - replaced by seek.
1497 list* - replaced by cons*.
1499 ** It's now possible to create modules with controlled environments
1503 (use-modules (ice-9 safe))
1504 (define m (make-safe-module))
1505 ;;; m will now be a module containing only a safe subset of R5RS
1506 (eval '(+ 1 2) m) --> 3
1507 (eval 'load m) --> ERROR: Unbound variable: load
1509 ** Evaluation of "()", the empty list, is now an error.
1511 Previously, the expression "()" evaluated to the empty list. This has
1512 been changed to signal a "missing expression" error. The correct way
1513 to write the empty list as a literal constant is to use quote: "'()".
1515 ** New concept of `Guile Extensions'.
1517 A Guile Extension is just a ordinary shared library that can be linked
1518 at run-time. We found it advantageous to give this simple concept a
1519 dedicated name to distinguish the issues related to shared libraries
1520 from the issues related to the module system.
1522 *** New function: load-extension
1524 Executing (load-extension lib init) is mostly equivalent to
1526 (dynamic-call init (dynamic-link lib))
1528 except when scm_register_extension has been called previously.
1529 Whenever appropriate, you should use `load-extension' instead of
1530 dynamic-link and dynamic-call.
1532 *** New C function: scm_c_register_extension
1534 This function registers a initialization function for use by
1535 `load-extension'. Use it when you don't want specific extensions to
1536 be loaded as shared libraries (for example on platforms that don't
1537 support dynamic linking).
1539 ** Auto-loading of compiled-code modules is deprecated.
1541 Guile used to be able to automatically find and link a shared
1542 library to satisfy requests for a module. For example, the module
1543 `(foo bar)' could be implemented by placing a shared library named
1544 "foo/libbar.so" (or with a different extension) in a directory on the
1547 This has been found to be too tricky, and is no longer supported. The
1548 shared libraries are now called "extensions". You should now write a
1549 small Scheme file that calls `load-extension' to load the shared
1550 library and initialize it explicitely.
1552 The shared libraries themselves should be installed in the usual
1553 places for shared libraries, with names like "libguile-foo-bar".
1555 For example, place this into a file "foo/bar.scm"
1557 (define-module (foo bar))
1559 (load-extension "libguile-foo-bar" "foobar_init")
1561 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1563 `eval' is now R5RS, that is it takes two arguments.
1564 The second argument is an environment specifier, i.e. either
1566 (scheme-report-environment 5)
1567 (null-environment 5)
1568 (interaction-environment)
1574 ** The module system has been made more disciplined.
1576 The function `eval' will save and restore the current module around
1577 the evaluation of the specified expression. While this expression is
1578 evaluated, `(current-module)' will now return the right module, which
1579 is the module specified as the second argument to `eval'.
1581 A consequence of this change is that `eval' is not particularly
1582 useful when you want allow the evaluated code to change what module is
1583 designated as the current module and have this change persist from one
1584 call to `eval' to the next. The read-eval-print-loop is an example
1585 where `eval' is now inadequate. To compensate, there is a new
1586 function `primitive-eval' that does not take a module specifier and
1587 that does not save/restore the current module. You should use this
1588 function together with `set-current-module', `current-module', etc
1589 when you want to have more control over the state that is carried from
1590 one eval to the next.
1592 Additionally, it has been made sure that forms that are evaluated at
1593 the top level are always evaluated with respect to the current module.
1594 Previously, subforms of top-level forms such as `begin', `case',
1595 etc. did not respect changes to the current module although these
1596 subforms are at the top-level as well.
1598 To prevent strange behavior, the forms `define-module',
1599 `use-modules', `use-syntax', and `export' have been restricted to only
1600 work on the top level. The forms `define-public' and
1601 `defmacro-public' only export the new binding on the top level. They
1602 behave just like `define' and `defmacro', respectively, when they are
1603 used in a lexical environment.
1605 Also, `export' will no longer silently re-export bindings imported
1606 from a used module. It will emit a `deprecation' warning and will
1607 cease to perform any re-export in the next version. If you actually
1608 want to re-export bindings, use the new `re-export' in place of
1609 `export'. The new `re-export' will not make copies of variables when
1610 rexporting them, as `export' did wrongly.
1612 ** Module system now allows selection and renaming of imported bindings
1614 Previously, when using `use-modules' or the `#:use-module' clause in
1615 the `define-module' form, all the bindings (association of symbols to
1616 values) for imported modules were added to the "current module" on an
1617 as-is basis. This has been changed to allow finer control through two
1618 new facilities: selection and renaming.
1620 You can now select which of the imported module's bindings are to be
1621 visible in the current module by using the `:select' clause. This
1622 clause also can be used to rename individual bindings. For example:
1624 ;; import all bindings no questions asked
1625 (use-modules (ice-9 common-list))
1627 ;; import four bindings, renaming two of them;
1628 ;; the current module sees: every some zonk-y zonk-n
1629 (use-modules ((ice-9 common-list)
1631 (remove-if . zonk-y)
1632 (remove-if-not . zonk-n))))
1634 You can also programmatically rename all selected bindings using the
1635 `:renamer' clause, which specifies a proc that takes a symbol and
1636 returns another symbol. Because it is common practice to use a prefix,
1637 we now provide the convenience procedure `symbol-prefix-proc'. For
1640 ;; import four bindings, renaming two of them specifically,
1641 ;; and all four w/ prefix "CL:";
1642 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1643 (use-modules ((ice-9 common-list)
1645 (remove-if . zonk-y)
1646 (remove-if-not . zonk-n))
1647 :renamer (symbol-prefix-proc 'CL:)))
1649 ;; import four bindings, renaming two of them specifically,
1650 ;; and all four by upcasing.
1651 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1652 (define (upcase-symbol sym)
1653 (string->symbol (string-upcase (symbol->string sym))))
1655 (use-modules ((ice-9 common-list)
1657 (remove-if . zonk-y)
1658 (remove-if-not . zonk-n))
1659 :renamer upcase-symbol))
1661 Note that programmatic renaming is done *after* individual renaming.
1662 Also, the above examples show `use-modules', but the same facilities are
1663 available for the `#:use-module' clause of `define-module'.
1665 See manual for more info.
1667 ** The semantics of guardians have changed.
1669 The changes are for the most part compatible. An important criterion
1670 was to keep the typical usage of guardians as simple as before, but to
1671 make the semantics safer and (as a result) more useful.
1673 *** All objects returned from guardians are now properly alive.
1675 It is now guaranteed that any object referenced by an object returned
1676 from a guardian is alive. It's now impossible for a guardian to
1677 return a "contained" object before its "containing" object.
1679 One incompatible (but probably not very important) change resulting
1680 from this is that it is no longer possible to guard objects that
1681 indirectly reference themselves (i.e. are parts of cycles). If you do
1682 so accidentally, you'll get a warning.
1684 *** There are now two types of guardians: greedy and sharing.
1686 If you call (make-guardian #t) or just (make-guardian), you'll get a
1687 greedy guardian, and for (make-guardian #f) a sharing guardian.
1689 Greedy guardians are the default because they are more "defensive".
1690 You can only greedily guard an object once. If you guard an object
1691 more than once, once in a greedy guardian and the rest of times in
1692 sharing guardians, then it is guaranteed that the object won't be
1693 returned from sharing guardians as long as it is greedily guarded
1696 Guardians returned by calls to `make-guardian' can now take one more
1697 optional parameter, which says whether to throw an error in case an
1698 attempt is made to greedily guard an object that is already greedily
1699 guarded. The default is true, i.e. throw an error. If the parameter
1700 is false, the guardian invocation returns #t if guarding was
1701 successful and #f if it wasn't.
1703 Also, since greedy guarding is, in effect, a side-effecting operation
1704 on objects, a new function is introduced: `destroy-guardian!'.
1705 Invoking this function on a guardian renders it unoperative and, if
1706 the guardian is greedy, clears the "greedily guarded" property of the
1707 objects that were guarded by it, thus undoing the side effect.
1709 Note that all this hair is hardly very important, since guardian
1710 objects are usually permanent.
1712 ** Continuations created by call-with-current-continuation now accept
1713 any number of arguments, as required by R5RS.
1715 ** New function `issue-deprecation-warning'
1717 This function is used to display the deprecation messages that are
1718 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1721 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1725 ;; `id' is deprecated. Use `identity' instead.
1730 ** New syntax `begin-deprecated'
1732 When deprecated features are included (as determined by the configure
1733 option --enable-deprecated), `begin-deprecated' is identical to
1734 `begin'. When deprecated features are excluded, it always evaluates
1735 to `#f', ignoring the body forms.
1737 ** New function `make-object-property'
1739 This function returns a new `procedure with setter' P that can be used
1740 to attach a property to objects. When calling P as
1744 where `obj' is any kind of object, it attaches `val' to `obj' in such
1745 a way that it can be retrieved by calling P as
1749 This function will replace procedure properties, symbol properties and
1750 source properties eventually.
1752 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1754 Instead of #&optional, #&key, etc you should now use #:optional,
1755 #:key, etc. Since #:optional is a keyword, you can write it as just
1756 :optional when (read-set! keywords 'prefix) is active.
1758 The old reader syntax `#&' is still supported, but deprecated. It
1759 will be removed in the next release.
1761 ** New define-module option: pure
1763 Tells the module system not to include any bindings from the root
1768 (define-module (totally-empty-module)
1771 ** New define-module option: export NAME1 ...
1773 Export names NAME1 ...
1775 This option is required if you want to be able to export bindings from
1776 a module which doesn't import one of `define-public' or `export'.
1780 (define-module (foo)
1782 :use-module (ice-9 r5rs)
1785 ;;; Note that we're pure R5RS below this point!
1790 ** New function: object->string OBJ
1792 Return a Scheme string obtained by printing a given object.
1794 ** New function: port? X
1796 Returns a boolean indicating whether X is a port. Equivalent to
1797 `(or (input-port? X) (output-port? X))'.
1799 ** New function: file-port?
1801 Determines whether a given object is a port that is related to a file.
1803 ** New function: port-for-each proc
1805 Apply PROC to each port in the Guile port table in turn. The return
1806 value is unspecified. More specifically, PROC is applied exactly once
1807 to every port that exists in the system at the time PORT-FOR-EACH is
1808 invoked. Changes to the port table while PORT-FOR-EACH is running
1809 have no effect as far as PORT-FOR-EACH is concerned.
1811 ** New function: dup2 oldfd newfd
1813 A simple wrapper for the `dup2' system call. Copies the file
1814 descriptor OLDFD to descriptor number NEWFD, replacing the
1815 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1816 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1817 to move away ports which are using NEWFD. The return value is
1820 ** New function: close-fdes fd
1822 A simple wrapper for the `close' system call. Close file
1823 descriptor FD, which must be an integer. Unlike close (*note
1824 close: Ports and File Descriptors.), the file descriptor will be
1825 closed even if a port is using it. The return value is
1828 ** New function: crypt password salt
1830 Encrypts `password' using the standard unix password encryption
1833 ** New function: chroot path
1835 Change the root directory of the running process to `path'.
1837 ** New functions: getlogin, cuserid
1839 Return the login name or the user name of the current effective user
1842 ** New functions: getpriority which who, setpriority which who prio
1844 Get or set the priority of the running process.
1846 ** New function: getpass prompt
1848 Read a password from the terminal, first displaying `prompt' and
1851 ** New function: flock file operation
1853 Set/remove an advisory shared or exclusive lock on `file'.
1855 ** New functions: sethostname name, gethostname
1857 Set or get the hostname of the machine the current process is running
1860 ** New function: mkstemp! tmpl
1862 mkstemp creates a new unique file in the file system and returns a
1863 new buffered port open for reading and writing to the file. TMPL
1864 is a string specifying where the file should be created: it must
1865 end with `XXXXXX' and will be changed in place to return the name
1866 of the temporary file.
1868 ** New function: open-input-string string
1870 Return an input string port which delivers the characters from
1871 `string'. This procedure, together with `open-output-string' and
1872 `get-output-string' implements SRFI-6.
1874 ** New function: open-output-string
1876 Return an output string port which collects all data written to it.
1877 The data can then be retrieved by `get-output-string'.
1879 ** New function: get-output-string
1881 Return the contents of an output string port.
1883 ** New function: identity
1885 Return the argument.
1887 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1888 are represented in Scheme as integers with normal host byte ordering.
1890 ** New function: inet-pton family address
1892 Convert a printable string network address into an integer. Note that
1893 unlike the C version of this function, the result is an integer with
1894 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1897 (inet-pton AF_INET "127.0.0.1") => 2130706433
1898 (inet-pton AF_INET6 "::1") => 1
1900 ** New function: inet-ntop family address
1902 Convert an integer network address into a printable string. Note that
1903 unlike the C version of this function, the input is an integer with
1904 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1907 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1908 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1909 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1913 Use `identity' instead.
1919 ** Deprecated: return-it
1923 ** Deprecated: string-character-length
1925 Use `string-length' instead.
1927 ** Deprecated: flags
1929 Use `logior' instead.
1931 ** Deprecated: close-all-ports-except.
1933 This was intended for closing ports in a child process after a fork,
1934 but it has the undesirable side effect of flushing buffers.
1935 port-for-each is more flexible.
1937 ** The (ice-9 popen) module now attempts to set up file descriptors in
1938 the child process from the current Scheme ports, instead of using the
1939 current values of file descriptors 0, 1, and 2 in the parent process.
1941 ** Removed function: builtin-weak-bindings
1943 There is no such concept as a weak binding any more.
1945 ** Removed constants: bignum-radix, scm-line-incrementors
1947 ** define-method: New syntax mandatory.
1949 The new method syntax is now mandatory:
1951 (define-method (NAME ARG-SPEC ...) BODY ...)
1952 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1954 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1955 REST-ARG ::= ARG-NAME
1957 If you have old code using the old syntax, import
1958 (oop goops old-define-method) before (oop goops) as in:
1960 (use-modules (oop goops old-define-method) (oop goops))
1962 ** Deprecated function: builtin-variable
1963 Removed function: builtin-bindings
1965 There is no longer a distinction between builtin or other variables.
1966 Use module system operations for all variables.
1968 ** Lazy-catch handlers are no longer allowed to return.
1970 That is, a call to `throw', `error', etc is now guaranteed to not
1973 ** Bugfixes for (ice-9 getopt-long)
1975 This module is now tested using test-suite/tests/getopt-long.test.
1976 The following bugs have been fixed:
1978 *** Parsing for options that are specified to have `optional' args now checks
1979 if the next element is an option instead of unconditionally taking it as the
1982 *** An error is now thrown for `--opt=val' when the option description
1983 does not specify `(value #t)' or `(value optional)'. This condition used to
1984 be accepted w/o error, contrary to the documentation.
1986 *** The error message for unrecognized options is now more informative.
1987 It used to be "not a record", an artifact of the implementation.
1989 *** The error message for `--opt' terminating the arg list (no value), when
1990 `(value #t)' is specified, is now more informative. It used to be "not enough
1993 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1994 The expansion used to be like so:
1996 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1998 Note that the "5d" is dropped. Now it is like so:
2000 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2002 This enables single-char options to have adjoining arguments as long as their
2003 constituent characters are not potential single-char options.
2005 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2007 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2008 property `arglist', which can be retrieved by `arity'. The result is that
2009 `arity' can give more detailed information than before:
2013 guile> (use-modules (ice-9 optargs))
2014 guile> (define* (foo #:optional a b c) a)
2016 0 or more arguments in `lambda*:G0'.
2021 3 optional arguments: `a', `b' and `c'.
2022 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2024 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2025 and `d', other keywords allowed.
2026 guile> (define* (baz a b #:optional c #:rest r) a)
2028 2 required arguments: `a' and `b', 1 optional argument: `c',
2031 * Changes to the C interface
2033 ** Types have been renamed from scm_*_t to scm_t_*.
2035 This has been done for POSIX sake. It reserves identifiers ending
2036 with "_t". What a concept.
2038 The old names are still available with status `deprecated'.
2040 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2042 ** Deprecated features have been removed.
2046 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2047 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2049 *** C Functions removed
2051 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2052 scm_fseek - replaced by scm_seek.
2053 gc-thunk - replaced by after-gc-hook.
2054 gh_int2scmb - replaced by gh_bool2scm.
2055 scm_tc_dblr - replaced by scm_tc16_real.
2056 scm_tc_dblc - replaced by scm_tc16_complex.
2057 scm_list_star - replaced by scm_cons_star.
2059 ** Deprecated: scm_makfromstr
2061 Use scm_mem2string instead.
2063 ** Deprecated: scm_make_shared_substring
2065 Explicit shared substrings will disappear from Guile.
2067 Instead, "normal" strings will be implemented using sharing
2068 internally, combined with a copy-on-write strategy.
2070 ** Deprecated: scm_read_only_string_p
2072 The concept of read-only strings will disappear in next release of
2075 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2077 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2079 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2081 Call a procedure with the indicated number of arguments. See "Fly
2082 Evaluation" in the manual.
2084 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2086 Call a procedure with the indicated number of arguments and a list of
2087 further arguments. See "Fly Evaluation" in the manual.
2089 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2091 Create a list of the given number of elements. See "List
2092 Constructors" in the manual.
2094 ** Renamed function: scm_listify has been replaced by scm_list_n.
2096 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2097 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2099 Use functions scm_list_N instead.
2101 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2103 Used by an application to read arbitrary number of bytes from a port.
2104 Same semantics as libc read, except that scm_c_read only returns less
2105 than SIZE bytes if at end-of-file.
2107 Warning: Doesn't update port line and column counts!
2109 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2111 Used by an application to write arbitrary number of bytes to an SCM
2112 port. Similar semantics as libc write. However, unlike libc
2113 write, scm_c_write writes the requested number of bytes and has no
2116 Warning: Doesn't update port line and column counts!
2118 ** New function: scm_init_guile ()
2120 In contrast to scm_boot_guile, scm_init_guile will return normally
2121 after initializing Guile. It is not available on all systems, tho.
2123 ** New functions: scm_str2symbol, scm_mem2symbol
2125 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2126 field of characters and creates a scheme symbol object from that C string.
2127 The function scm_mem2symbol takes a const char* and a number of characters and
2128 creates a symbol from the characters in that memory area.
2130 ** New functions: scm_primitive_make_property
2131 scm_primitive_property_ref
2132 scm_primitive_property_set_x
2133 scm_primitive_property_del_x
2135 These functions implement a new way to deal with object properties.
2136 See libguile/properties.c for their documentation.
2138 ** New function: scm_done_free (long size)
2140 This function is the inverse of scm_done_malloc. Use it to report the
2141 amount of smob memory you free. The previous method, which involved
2142 calling scm_done_malloc with negative argument, was somewhat
2143 unintuitive (and is still available, of course).
2145 ** New function: scm_c_memq (SCM obj, SCM list)
2147 This function provides a fast C level alternative for scm_memq for the case
2148 that the list parameter is known to be a proper list. The function is a
2149 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2150 list input parameter, since for anything else but a proper list the function's
2151 behaviour is undefined - it may even crash or loop endlessly. Further, for
2152 the case that the object is not found in the list, scm_c_memq returns #f which
2153 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2155 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2156 scm_remember_upto_here
2158 These functions replace the function scm_remember.
2160 ** Deprecated function: scm_remember
2162 Use one of the new functions scm_remember_upto_here_1,
2163 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2165 ** New function: scm_allocate_string
2167 This function replaces the function scm_makstr.
2169 ** Deprecated function: scm_makstr
2171 Use the new function scm_allocate_string instead.
2173 ** New global variable scm_gc_running_p introduced.
2175 Use this variable to find out if garbage collection is being executed. Up to
2176 now applications have used scm_gc_heap_lock to test if garbage collection was
2177 running, which also works because of the fact that up to know only the garbage
2178 collector has set this variable. But, this is an implementation detail that
2179 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2180 of this variable is (and has been) not fully safe anyway.
2182 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2184 Use these instead of SCM_LENGTH_MAX.
2186 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2187 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2188 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2190 Use these instead of SCM_LENGTH.
2192 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2193 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2194 SCM_SET_BITVECTOR_LENGTH
2196 Use these instead of SCM_SETLENGTH
2198 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2199 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2202 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2205 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2206 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2209 Use these instead of SCM_SETCHARS.
2211 ** New macro: SCM_BITVECTOR_P
2213 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2215 Use instead of SCM_COERCE_SUBSTR.
2217 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2219 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2221 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2222 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2223 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2224 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2225 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2226 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2227 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2228 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2229 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2230 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2231 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2232 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2233 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2234 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2235 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2237 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2238 Use scm_memory_error instead of SCM_NALLOC.
2239 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2240 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2241 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2242 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2243 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2244 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2245 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2246 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2247 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2248 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2249 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2250 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2251 Use a type specific setter macro instead of SCM_SETLENGTH.
2252 Use a type specific setter macro instead of SCM_SETCHARS.
2253 Use a type specific length macro instead of SCM_LENGTH_MAX.
2254 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2255 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2256 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2257 Use SCM_TYP16 instead of SCM_GCTYP16.
2258 Use SCM_CDR instead of SCM_GCCDR.
2259 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2260 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2261 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2262 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2263 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2264 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2265 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2267 ** Removed function: scm_struct_init
2269 ** Removed variable: scm_symhash_dim
2271 ** Renamed function: scm_make_cont has been replaced by
2272 scm_make_continuation, which has a different interface.
2274 ** Deprecated function: scm_call_catching_errors
2276 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2278 ** Deprecated function: scm_strhash
2280 Use scm_string_hash instead.
2282 ** Deprecated function: scm_vector_set_length_x
2284 Instead, create a fresh vector of the desired size and copy the contents.
2286 ** scm_gensym has changed prototype
2288 scm_gensym now only takes one argument.
2290 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2293 There is now only a single symbol type scm_tc7_symbol.
2294 The tag scm_tc7_lvector was not used anyway.
2296 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2298 Use scm_make_smob_type and scm_set_smob_XXX instead.
2300 ** New function scm_set_smob_apply.
2302 This can be used to set an apply function to a smob type.
2304 ** Deprecated function: scm_strprint_obj
2306 Use scm_object_to_string instead.
2308 ** Deprecated function: scm_wta
2310 Use scm_wrong_type_arg, or another appropriate error signalling function
2313 ** Explicit support for obarrays has been deprecated.
2315 Use `scm_str2symbol' and the generic hashtable functions instead.
2317 ** The concept of `vcells' has been deprecated.
2319 The data type `variable' is now used exclusively. `Vcells' have been
2320 a low-level concept so you are likely not affected by this change.
2322 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2323 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2325 Use scm_c_define or scm_c_lookup instead, as appropriate.
2327 *** New functions: scm_c_module_lookup, scm_c_lookup,
2328 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2329 scm_module_define, scm_define.
2331 These functions work with variables instead of with vcells.
2333 ** New functions for creating and defining `subr's and `gsubr's.
2335 The new functions more clearly distinguish between creating a subr (or
2336 gsubr) object and adding it to the current module.
2338 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2339 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2340 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2341 scm_c_define_gsubr_with_generic.
2343 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2344 scm_make_subr_with_generic, scm_make_gsubr,
2345 scm_make_gsubr_with_generic.
2347 Use the new ones from above instead.
2349 ** C interface to the module system has changed.
2351 While we suggest that you avoid as many explicit module system
2352 operations from C as possible for the time being, the C interface has
2353 been made more similar to the high-level Scheme module system.
2355 *** New functions: scm_c_define_module, scm_c_use_module,
2356 scm_c_export, scm_c_resolve_module.
2358 They mostly work like their Scheme namesakes. scm_c_define_module
2359 takes a function that is called a context where the new module is
2362 *** Deprecated functions: scm_the_root_module, scm_make_module,
2363 scm_ensure_user_module, scm_load_scheme_module.
2365 Use the new functions instead.
2367 ** Renamed function: scm_internal_with_fluids becomes
2370 scm_internal_with_fluids is available as a deprecated function.
2372 ** New function: scm_c_with_fluid.
2374 Just like scm_c_with_fluids, but takes one fluid and one value instead
2377 ** Deprecated typedefs: long_long, ulong_long.
2379 They are of questionable utility and they pollute the global
2382 ** Deprecated typedef: scm_sizet
2384 It is of questionable utility now that Guile requires ANSI C, and is
2387 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2388 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2389 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2391 Made more compliant with the naming policy by adding a _t at the end.
2393 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2394 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2396 With the exception of the mysterious scm_2ulong2big, they are still
2397 available under new names (scm_i_mkbig etc). These functions are not
2398 intended to be used in user code. You should avoid dealing with
2399 bignums directly, and should deal with numbers in general (which can
2402 ** Change in behavior: scm_num2long, scm_num2ulong
2404 The scm_num2[u]long functions don't any longer accept an inexact
2405 argument. This change in behavior is motivated by concordance with
2406 R5RS: It is more common that a primitive doesn't want to accept an
2407 inexact for an exact.
2409 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2410 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2411 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2414 These are conversion functions between the various ANSI C integral
2415 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2416 accept an inexact argument.
2418 ** New functions: scm_float2num, scm_double2num,
2419 scm_num2float, scm_num2double.
2421 These are conversion functions between the two ANSI C float types and
2424 ** New number validation macros:
2425 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2429 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2431 These are just nicer-named old scm_protect_object and
2432 scm_unprotect_object.
2434 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2436 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2438 These functions can be used to register pointers to locations that
2441 ** Deprecated function: scm_create_hook.
2443 Its sins are: misleading name, non-modularity and lack of general
2447 Changes since Guile 1.3.4:
2449 * Changes to the distribution
2451 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2453 We've changed the way we handle generated files in the Guile source
2454 repository. As a result, the procedure for building trees obtained
2455 from the nightly FTP snapshots or via CVS has changed:
2456 - You must have appropriate versions of autoconf, automake, and
2457 libtool installed on your system. See README for info on how to
2458 obtain these programs.
2459 - Before configuring the tree, you must first run the script
2460 `autogen.sh' at the top of the source tree.
2462 The Guile repository used to contain not only source files, written by
2463 humans, but also some generated files, like configure scripts and
2464 Makefile.in files. Even though the contents of these files could be
2465 derived mechanically from other files present, we thought it would
2466 make the tree easier to build if we checked them into CVS.
2468 However, this approach means that minor differences between
2469 developer's installed tools and habits affected the whole team.
2470 So we have removed the generated files from the repository, and
2471 added the autogen.sh script, which will reconstruct them
2475 ** configure now has experimental options to remove support for certain
2478 --disable-arrays omit array and uniform array support
2479 --disable-posix omit posix interfaces
2480 --disable-networking omit networking interfaces
2481 --disable-regex omit regular expression interfaces
2483 These are likely to become separate modules some day.
2485 ** New configure option --enable-debug-freelist
2487 This enables a debugging version of SCM_NEWCELL(), and also registers
2488 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2490 Configure with the --enable-debug-freelist option to enable
2491 the gc-set-debug-check-freelist! primitive, and then use:
2493 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2494 (gc-set-debug-check-freelist! #f) # turn off checking
2496 Checking of the freelist forces a traversal of the freelist and
2497 a garbage collection before each allocation of a cell. This can
2498 slow down the interpreter dramatically, so the setter should be used to
2499 turn on this extra processing only when necessary.
2501 ** New configure option --enable-debug-malloc
2503 Include code for debugging of calls to scm_must_malloc/realloc/free.
2507 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2508 2. objects reallocated by scm_must_realloc has been allocated by
2510 3. reallocated objects are reallocated with the same what string
2512 But, most importantly, it records the number of allocated objects of
2513 each kind. This is useful when searching for memory leaks.
2515 A Guile compiled with this option provides the primitive
2516 `malloc-stats' which returns an alist with pairs of kind and the
2517 number of objects of that kind.
2519 ** All includes are now referenced relative to the root directory
2521 Since some users have had problems with mixups between Guile and
2522 system headers, we have decided to always refer to Guile headers via
2523 their parent directories. This essentially creates a "private name
2524 space" for Guile headers. This means that the compiler only is given
2525 -I options for the root build and root source directory.
2527 ** Header files kw.h and genio.h have been removed.
2529 ** The module (ice-9 getopt-gnu-style) has been removed.
2531 ** New module (ice-9 documentation)
2533 Implements the interface to documentation strings associated with
2536 ** New module (ice-9 time)
2538 Provides a macro `time', which displays execution time of a given form.
2540 ** New module (ice-9 history)
2542 Loading this module enables value history in the repl.
2544 * Changes to the stand-alone interpreter
2546 ** New command line option --debug
2548 Start Guile with debugging evaluator and backtraces enabled.
2550 This is useful when debugging your .guile init file or scripts.
2552 ** New help facility
2554 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2555 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2556 (help 'NAME) gives documentation for NAME, even if it is not an object
2557 (help ,EXPR) gives documentation for object returned by EXPR
2558 (help (my module)) gives module commentary for `(my module)'
2559 (help) gives this text
2561 `help' searches among bindings exported from loaded modules, while
2562 `apropos' searches among bindings visible from the "current" module.
2564 Examples: (help help)
2566 (help "output-string")
2568 ** `help' and `apropos' now prints full module names
2570 ** Dynamic linking now uses libltdl from the libtool package.
2572 The old system dependent code for doing dynamic linking has been
2573 replaced with calls to the libltdl functions which do all the hairy
2576 The major improvement is that you can now directly pass libtool
2577 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2578 will be able to do the best shared library job you can get, via
2581 The way dynamic libraries are found has changed and is not really
2582 portable across platforms, probably. It is therefore recommended to
2583 use absolute filenames when possible.
2585 If you pass a filename without an extension to `dynamic-link', it will
2586 try a few appropriate ones. Thus, the most platform ignorant way is
2587 to specify a name like "libfoo", without any directories and
2590 ** Guile COOP threads are now compatible with LinuxThreads
2592 Previously, COOP threading wasn't possible in applications linked with
2593 Linux POSIX threads due to their use of the stack pointer to find the
2594 thread context. This has now been fixed with a workaround which uses
2595 the pthreads to allocate the stack.
2597 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2599 ** Positions of erring expression in scripts
2601 With version 1.3.4, the location of the erring expression in Guile
2602 scipts is no longer automatically reported. (This should have been
2603 documented before the 1.3.4 release.)
2605 You can get this information by enabling recording of positions of
2606 source expressions and running the debugging evaluator. Put this at
2607 the top of your script (or in your "site" file):
2609 (read-enable 'positions)
2610 (debug-enable 'debug)
2612 ** Backtraces in scripts
2614 It is now possible to get backtraces in scripts.
2618 (debug-enable 'debug 'backtrace)
2620 at the top of the script.
2622 (The first options enables the debugging evaluator.
2623 The second enables backtraces.)
2625 ** Part of module system symbol lookup now implemented in C
2627 The eval closure of most modules is now implemented in C. Since this
2628 was one of the bottlenecks for loading speed, Guile now loads code
2629 substantially faster than before.
2631 ** Attempting to get the value of an unbound variable now produces
2632 an exception with a key of 'unbound-variable instead of 'misc-error.
2634 ** The initial default output port is now unbuffered if it's using a
2635 tty device. Previously in this situation it was line-buffered.
2637 ** New hook: after-gc-hook
2639 after-gc-hook takes over the role of gc-thunk. This hook is run at
2640 the first SCM_TICK after a GC. (Thus, the code is run at the same
2641 point during evaluation as signal handlers.)
2643 Note that this hook should be used only for diagnostic and debugging
2644 purposes. It is not certain that it will continue to be well-defined
2645 when this hook is run in the future.
2647 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2648 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2650 ** Improvements to garbage collector
2652 Guile 1.4 has a new policy for triggering heap allocation and
2653 determining the sizes of heap segments. It fixes a number of problems
2656 1. The new policy can handle two separate pools of cells
2657 (2-word/4-word) better. (The old policy would run wild, allocating
2658 more and more memory for certain programs.)
2660 2. The old code would sometimes allocate far too much heap so that the
2661 Guile process became gigantic. The new code avoids this.
2663 3. The old code would sometimes allocate too little so that few cells
2664 were freed at GC so that, in turn, too much time was spent in GC.
2666 4. The old code would often trigger heap allocation several times in a
2667 row. (The new scheme predicts how large the segments needs to be
2668 in order not to need further allocation.)
2670 All in all, the new GC policy will make larger applications more
2673 The new GC scheme also is prepared for POSIX threading. Threads can
2674 allocate private pools of cells ("clusters") with just a single
2675 function call. Allocation of single cells from such a cluster can
2676 then proceed without any need of inter-thread synchronization.
2678 ** New environment variables controlling GC parameters
2680 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2683 Allocation of 2-word cell heaps:
2685 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2688 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2689 GC in percent of total heap size
2692 Allocation of 4-word cell heaps
2693 (used for real numbers and misc other objects):
2695 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2697 (See entry "Way for application to customize GC parameters" under
2698 section "Changes to the scm_ interface" below.)
2700 ** Guile now implements reals using 4-word cells
2702 This speeds up computation with reals. (They were earlier allocated
2703 with `malloc'.) There is still some room for optimizations, however.
2705 ** Some further steps toward POSIX thread support have been taken
2707 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2708 don't have much effect any longer, and many of them will be removed in
2712 are only handled at the top of the evaluator loop, immediately after
2713 I/O, and in scm_equalp.
2715 *** The GC can allocate thread private pools of pairs.
2717 * Changes to Scheme functions and syntax
2719 ** close-input-port and close-output-port are now R5RS
2721 These procedures have been turned into primitives and have R5RS behaviour.
2723 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2725 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2726 extended by the more sophisticated version in (ice-9 format)
2728 (simple-format port message . args)
2729 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2730 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2731 the escapes are replaced with corresponding members of ARGS:
2732 ~A formats using `display' and ~S formats using `write'.
2733 If DESTINATION is #t, then use the `current-output-port',
2734 if DESTINATION is #f, then return a string containing the formatted text.
2735 Does not add a trailing newline."
2737 ** string-ref: the second argument is no longer optional.
2739 ** string, list->string: no longer accept strings in their arguments,
2740 only characters, for compatibility with R5RS.
2742 ** New procedure: port-closed? PORT
2743 Returns #t if PORT is closed or #f if it is open.
2745 ** Deprecated: list*
2747 The list* functionality is now provided by cons* (SRFI-1 compliant)
2749 ** New procedure: cons* ARG1 ARG2 ... ARGn
2751 Like `list', but the last arg provides the tail of the constructed list,
2752 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2754 Requires at least one argument. If given one argument, that argument
2755 is returned as result.
2757 This function is called `list*' in some other Schemes and in Common LISP.
2759 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2761 ** New procedure: object-documentation OBJECT
2763 Returns the documentation string associated with OBJECT. The
2764 procedure uses a caching mechanism so that subsequent lookups are
2767 Exported by (ice-9 documentation).
2769 ** module-name now returns full names of modules
2771 Previously, only the last part of the name was returned (`session' for
2772 `(ice-9 session)'). Ex: `(ice-9 session)'.
2774 * Changes to the gh_ interface
2776 ** Deprecated: gh_int2scmb
2778 Use gh_bool2scm instead.
2780 * Changes to the scm_ interface
2782 ** Guile primitives now carry docstrings!
2784 Thanks to Greg Badros!
2786 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2788 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2789 macros and must contain a docstring that is extracted into foo.doc using a new
2790 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2792 However, a major overhaul of these macros is scheduled for the next release of
2795 ** Guile primitives use a new technique for validation of arguments
2797 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2798 the readability of argument checking.
2800 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2802 ** New macros: SCM_PACK, SCM_UNPACK
2804 Compose/decompose an SCM value.
2806 The SCM type is now treated as an abstract data type and may be defined as a
2807 long, a void* or as a struct, depending on the architecture and compile time
2808 options. This makes it easier to find several types of bugs, for example when
2809 SCM values are treated as integers without conversion. Values of the SCM type
2810 should be treated as "atomic" values. These macros are used when
2811 composing/decomposing an SCM value, either because you want to access
2812 individual bits, or because you want to treat it as an integer value.
2814 E.g., in order to set bit 7 in an SCM value x, use the expression
2816 SCM_PACK (SCM_UNPACK (x) | 0x80)
2818 ** The name property of hooks is deprecated.
2819 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2821 You can emulate this feature by using object properties.
2823 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2824 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2827 These macros will be removed in a future release of Guile.
2829 ** The following types, functions and macros from numbers.h are deprecated:
2830 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2831 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2833 ** Port internals: the rw_random variable in the scm_port structure
2834 must be set to non-zero in any random access port. In recent Guile
2835 releases it was only set for bidirectional random-access ports.
2837 ** Port internals: the seek ptob procedure is now responsible for
2838 resetting the buffers if required. The change was made so that in the
2839 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2840 the fport and strport ptobs can avoid resetting the buffers,
2841 in particular to avoid discarding unread chars. An existing port
2842 type can be fixed by adding something like the following to the
2843 beginning of the ptob seek procedure:
2845 if (pt->rw_active == SCM_PORT_READ)
2846 scm_end_input (object);
2847 else if (pt->rw_active == SCM_PORT_WRITE)
2848 ptob->flush (object);
2850 although to actually avoid resetting the buffers and discard unread
2851 chars requires further hacking that depends on the characteristics
2854 ** Deprecated functions: scm_fseek, scm_tag
2856 These functions are no longer used and will be removed in a future version.
2858 ** The scm_sysmissing procedure is no longer used in libguile.
2859 Unless it turns out to be unexpectedly useful to somebody, it will be
2860 removed in a future version.
2862 ** The format of error message strings has changed
2864 The two C procedures: scm_display_error and scm_error, as well as the
2865 primitive `scm-error', now use scm_simple_format to do their work.
2866 This means that the message strings of all code must be updated to use
2867 ~A where %s was used before, and ~S where %S was used before.
2869 During the period when there still are a lot of old Guiles out there,
2870 you might want to support both old and new versions of Guile.
2872 There are basically two methods to achieve this. Both methods use
2875 AC_CHECK_FUNCS(scm_simple_format)
2877 in your configure.in.
2879 Method 1: Use the string concatenation features of ANSI C's
2884 #ifdef HAVE_SCM_SIMPLE_FORMAT
2890 Then represent each of your error messages using a preprocessor macro:
2892 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2896 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2897 (define make-message string-append)
2899 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2901 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2905 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2910 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2914 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2916 Don't use the functions coop_mutex_init and
2917 coop_condition_variable_init. They will change.
2919 Use scm_mutex_init and scm_cond_init instead.
2921 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2922 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2923 COND, as `scm_cond_wait' does, but it also bounds the duration
2924 of the wait. If COND has not been signaled before time ABSTIME,
2925 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2926 returns the error code `ETIMEDOUT'.
2928 The ABSTIME parameter specifies an absolute time, with the same
2929 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2930 to 00:00:00 GMT, January 1, 1970.
2932 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2933 `scm_cond_broadcast' restarts all the threads that are waiting
2934 on the condition variable COND. Nothing happens if no threads are
2937 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2938 `scm_key_create' allocates a new TSD key. The key is stored in
2939 the location pointed to by KEY. There is no limit on the number
2940 of keys allocated at a given time. The value initially associated
2941 with the returned key is `NULL' in all currently executing threads.
2943 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2944 function associated with the key. When a thread terminates,
2945 DESTR_FUNCTION is called on the value associated with the key in
2946 that thread. The DESTR_FUNCTION is not called if a key is deleted
2947 with `scm_key_delete' or a value is changed with
2948 `scm_setspecific'. The order in which destructor functions are
2949 called at thread termination time is unspecified.
2951 Destructors are not yet implemented.
2953 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2954 `scm_setspecific' changes the value associated with KEY in the
2955 calling thread, storing the given POINTER instead.
2957 ** New function: scm_getspecific (scm_key_t KEY)
2958 `scm_getspecific' returns the value currently associated with
2959 KEY in the calling thread.
2961 ** New function: scm_key_delete (scm_key_t KEY)
2962 `scm_key_delete' deallocates a TSD key. It does not check
2963 whether non-`NULL' values are associated with that key in the
2964 currently executing threads, nor call the destructor function
2965 associated with the key.
2967 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2969 Initialize a C level hook HOOK with associated HOOK_DATA and type
2970 TYPE. (See scm_c_hook_run ().)
2972 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2974 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2975 is true, add it last, otherwise first. The same FUNC can be added
2976 multiple times if FUNC_DATA differ and vice versa.
2978 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2980 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2981 function is only removed if both FUNC and FUNC_DATA matches.
2983 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2985 Run hook HOOK passing DATA to the hook functions.
2987 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2988 returned is undefined.
2990 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2991 returns a non-NULL value. This value is returned as the result of
2992 scm_c_hook_run. If all functions return NULL, NULL is returned.
2994 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2995 returns a NULL value, and NULL is returned. If all functions returns
2996 a non-NULL value, the last value is returned.
2998 ** New C level GC hooks
3000 Five new C level hooks has been added to the garbage collector.
3002 scm_before_gc_c_hook
3005 are run before locking and after unlocking the heap. The system is
3006 thus in a mode where evaluation can take place. (Except that
3007 scm_before_gc_c_hook must not allocate new cells.)
3009 scm_before_mark_c_hook
3010 scm_before_sweep_c_hook
3011 scm_after_sweep_c_hook
3013 are run when the heap is locked. These are intended for extension of
3014 the GC in a modular fashion. Examples are the weaks and guardians
3017 ** Way for application to customize GC parameters
3019 The application can set up other default values for the GC heap
3020 allocation parameters
3022 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3023 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3024 GUILE_MAX_SEGMENT_SIZE,
3028 scm_default_init_heap_size_1, scm_default_min_yield_1,
3029 scm_default_init_heap_size_2, scm_default_min_yield_2,
3030 scm_default_max_segment_size
3032 respectively before callong scm_boot_guile.
3034 (See entry "New environment variables ..." in section
3035 "Changes to the stand-alone interpreter" above.)
3037 ** scm_protect_object/scm_unprotect_object now nest
3039 This means that you can call scm_protect_object multiple times on an
3040 object and count on the object being protected until
3041 scm_unprotect_object has been call the same number of times.
3043 The functions also have better time complexity.
3045 Still, it is usually possible to structure the application in a way
3046 that you don't need to use these functions. For example, if you use a
3047 protected standard Guile list to keep track of live objects rather
3048 than some custom data type, objects will die a natural death when they
3049 are no longer needed.
3051 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3053 Guile does not provide the float representation for inexact real numbers any
3054 more. Now, only doubles are used to represent inexact real numbers. Further,
3055 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3056 and scm_tc16_complex, respectively.
3058 ** Removed deprecated type scm_smobfuns
3060 ** Removed deprecated function scm_newsmob
3062 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3064 There is an ongoing discussion among the developers whether to
3065 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3066 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3067 until this issue has been settled.
3069 ** Removed deprecated type tag scm_tc16_kw
3071 ** Added type tag scm_tc16_keyword
3073 (This was introduced already in release 1.3.4 but was not documented
3076 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3078 * Changes to system call interfaces:
3080 ** The "select" procedure now tests port buffers for the ability to
3081 provide input or accept output. Previously only the underlying file
3082 descriptors were checked.
3084 ** New variable PIPE_BUF: the maximum number of bytes that can be
3085 atomically written to a pipe.
3087 ** If a facility is not available on the system when Guile is
3088 compiled, the corresponding primitive procedure will not be defined.
3089 Previously it would have been defined but would throw a system-error
3090 exception if called. Exception handlers which catch this case may
3091 need minor modification: an error will be thrown with key
3092 'unbound-variable instead of 'system-error. Alternatively it's
3093 now possible to use `defined?' to check whether the facility is
3096 ** Procedures which depend on the timezone should now give the correct
3097 result on systems which cache the TZ environment variable, even if TZ
3098 is changed without calling tzset.
3100 * Changes to the networking interfaces:
3102 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3103 long integers between network and host format. For now, it's not
3104 particularly convenient to do this kind of thing, but consider:
3106 (define write-network-long
3107 (lambda (value port)
3108 (let ((v (make-uniform-vector 1 1 0)))
3109 (uniform-vector-set! v 0 (htonl value))
3110 (uniform-vector-write v port))))
3112 (define read-network-long
3114 (let ((v (make-uniform-vector 1 1 0)))
3115 (uniform-vector-read! v port)
3116 (ntohl (uniform-vector-ref v 0)))))
3118 ** If inet-aton fails, it now throws an error with key 'misc-error
3119 instead of 'system-error, since errno is not relevant.
3121 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3122 specific keys instead of 'system-error. The latter is inappropriate
3123 since errno will not have been set. The keys are:
3124 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3126 ** sethostent, setnetent, setprotoent, setservent: now take an
3127 optional argument STAYOPEN, which specifies whether the database
3128 remains open after a database entry is accessed randomly (e.g., using
3129 gethostbyname for the hosts database.) The default is #f. Previously
3133 Changes since Guile 1.3.2:
3135 * Changes to the stand-alone interpreter
3139 An initial version of the Guile debugger written by Chris Hanson has
3140 been added. The debugger is still under development but is included
3141 in the distribution anyway since it is already quite useful.
3147 after an error to enter the debugger. Type `help' inside the debugger
3148 for a description of available commands.
3150 If you prefer to have stack frames numbered and printed in
3151 anti-chronological order and prefer up in the stack to be down on the
3152 screen as is the case in gdb, you can put
3154 (debug-enable 'backwards)
3156 in your .guile startup file. (However, this means that Guile can't
3157 use indentation to indicate stack level.)
3159 The debugger is autoloaded into Guile at the first use.
3161 ** Further enhancements to backtraces
3163 There is a new debug option `width' which controls the maximum width
3164 on the screen of printed stack frames. Fancy printing parameters
3165 ("level" and "length" as in Common LISP) are adaptively adjusted for
3166 each stack frame to give maximum information while still fitting
3167 within the bounds. If the stack frame can't be made to fit by
3168 adjusting parameters, it is simply cut off at the end. This is marked
3171 ** Some modules are now only loaded when the repl is started
3173 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3174 regex) are now loaded into (guile-user) only if the repl has been
3175 started. The effect is that the startup time for scripts has been
3176 reduced to 30% of what it was previously.
3178 Correctly written scripts load the modules they require at the top of
3179 the file and should not be affected by this change.
3181 ** Hooks are now represented as smobs
3183 * Changes to Scheme functions and syntax
3185 ** Readline support has changed again.
3187 The old (readline-activator) module is gone. Use (ice-9 readline)
3188 instead, which now contains all readline functionality. So the code
3189 to activate readline is now
3191 (use-modules (ice-9 readline))
3194 This should work at any time, including from the guile prompt.
3196 To avoid confusion about the terms of Guile's license, please only
3197 enable readline for your personal use; please don't make it the
3198 default for others. Here is why we make this rather odd-sounding
3201 Guile is normally licensed under a weakened form of the GNU General
3202 Public License, which allows you to link code with Guile without
3203 placing that code under the GPL. This exception is important to some
3206 However, since readline is distributed under the GNU General Public
3207 License, when you link Guile with readline, either statically or
3208 dynamically, you effectively change Guile's license to the strict GPL.
3209 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3210 which are normally permitted become forbidden. This is a rather
3211 non-obvious consequence of the licensing terms.
3213 So, to make sure things remain clear, please let people choose for
3214 themselves whether to link GPL'd libraries like readline with Guile.
3216 ** regexp-substitute/global has changed slightly, but incompatibly.
3218 If you include a function in the item list, the string of the match
3219 object it receives is the same string passed to
3220 regexp-substitute/global, not some suffix of that string.
3221 Correspondingly, the match's positions are relative to the entire
3222 string, not the suffix.
3224 If the regexp can match the empty string, the way matches are chosen
3225 from the string has changed. regexp-substitute/global recognizes the
3226 same set of matches that list-matches does; see below.
3228 ** New function: list-matches REGEXP STRING [FLAGS]
3230 Return a list of match objects, one for every non-overlapping, maximal
3231 match of REGEXP in STRING. The matches appear in left-to-right order.
3232 list-matches only reports matches of the empty string if there are no
3233 other matches which begin on, end at, or include the empty match's
3236 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3238 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3240 For each match of REGEXP in STRING, apply PROC to the match object,
3241 and the last value PROC returned, or INIT for the first call. Return
3242 the last value returned by PROC. We apply PROC to the matches as they
3243 appear from left to right.
3245 This function recognizes matches according to the same criteria as
3248 Thus, you could define list-matches like this:
3250 (define (list-matches regexp string . flags)
3251 (reverse! (apply fold-matches regexp string '() cons flags)))
3253 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3257 *** New function: hook? OBJ
3259 Return #t if OBJ is a hook, otherwise #f.
3261 *** New function: make-hook-with-name NAME [ARITY]
3263 Return a hook with name NAME and arity ARITY. The default value for
3264 ARITY is 0. The only effect of NAME is that it will appear when the
3265 hook object is printed to ease debugging.
3267 *** New function: hook-empty? HOOK
3269 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3271 *** New function: hook->list HOOK
3273 Return a list of the procedures that are called when run-hook is
3276 ** `map' signals an error if its argument lists are not all the same length.
3278 This is the behavior required by R5RS, so this change is really a bug
3279 fix. But it seems to affect a lot of people's code, so we're
3280 mentioning it here anyway.
3282 ** Print-state handling has been made more transparent
3284 Under certain circumstances, ports are represented as a port with an
3285 associated print state. Earlier, this pair was represented as a pair
3286 (see "Some magic has been added to the printer" below). It is now
3287 indistinguishable (almost; see `get-print-state') from a port on the
3290 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3292 Return a new port with the associated print state PRINT-STATE.
3294 *** New function: get-print-state OUTPUT-PORT
3296 Return the print state associated with this port if it exists,
3297 otherwise return #f.
3299 *** New function: directory-stream? OBJECT
3301 Returns true iff OBJECT is a directory stream --- the sort of object
3302 returned by `opendir'.
3304 ** New function: using-readline?
3306 Return #t if readline is in use in the current repl.
3308 ** structs will be removed in 1.4
3310 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3311 and use GOOPS objects as the fundamental record type.
3313 * Changes to the scm_ interface
3315 ** structs will be removed in 1.4
3317 The entire current struct interface (struct.c, struct.h) will be
3318 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3319 GOOPS objects as the fundamental record type.
3321 ** The internal representation of subr's has changed
3323 Instead of giving a hint to the subr name, the CAR field of the subr
3324 now contains an index to a subr entry in scm_subr_table.
3326 *** New variable: scm_subr_table
3328 An array of subr entries. A subr entry contains the name, properties
3329 and documentation associated with the subr. The properties and
3330 documentation slots are not yet used.
3332 ** A new scheme for "forwarding" calls to a builtin to a generic function
3334 It is now possible to extend the functionality of some Guile
3335 primitives by letting them defer a call to a GOOPS generic function on
3336 argument mismatch. This means that there is no loss of efficiency in
3341 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3342 (define-method + ((x <string>) (y <string>))
3343 (string-append x y))
3345 + will still be as efficient as usual in numerical calculations, but
3346 can also be used for concatenating strings.
3348 Who will be the first one to extend Guile's numerical tower to
3349 rationals? :) [OK, there a few other things to fix before this can
3350 be made in a clean way.]
3352 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3354 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3356 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3358 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3359 a variable GENERIC which can be used by the dispatch macros below.
3361 [This is experimental code which may change soon.]
3363 *** New macros for forwarding control to a generic on arg type error
3365 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3367 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3369 These correspond to the scm_wta function call, and have the same
3370 behaviour until the user has called the GOOPS primitive
3371 `enable-primitive-generic!'. After that, these macros will apply the
3372 generic function GENERIC to the argument(s) instead of calling
3375 [This is experimental code which may change soon.]
3377 *** New macros for argument testing with generic dispatch
3379 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3381 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3383 These correspond to the SCM_ASSERT macro, but will defer control to
3384 GENERIC on error after `enable-primitive-generic!' has been called.
3386 [This is experimental code which may change soon.]
3388 ** New function: SCM scm_eval_body (SCM body, SCM env)
3390 Evaluates the body of a special form.
3392 ** The internal representation of struct's has changed
3394 Previously, four slots were allocated for the procedure(s) of entities
3395 and operators. The motivation for this representation had to do with
3396 the structure of the evaluator, the wish to support tail-recursive
3397 generic functions, and efficiency. Since the generic function
3398 dispatch mechanism has changed, there is no longer a need for such an
3399 expensive representation, and the representation has been simplified.
3401 This should not make any difference for most users.
3403 ** GOOPS support has been cleaned up.
3405 Some code has been moved from eval.c to objects.c and code in both of
3406 these compilation units has been cleaned up and better structured.
3408 *** New functions for applying generic functions
3410 New function: SCM scm_apply_generic (GENERIC, ARGS)
3411 New function: SCM scm_call_generic_0 (GENERIC)
3412 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3413 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3414 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3416 ** Deprecated function: scm_make_named_hook
3418 It is now replaced by:
3420 ** New function: SCM scm_create_hook (const char *name, int arity)
3422 Creates a hook in the same way as make-hook above but also
3423 binds a variable named NAME to it.
3425 This is the typical way of creating a hook from C code.
3427 Currently, the variable is created in the "current" module.
3428 This might change when we get the new module system.
3430 [The behaviour is identical to scm_make_named_hook.]
3434 Changes since Guile 1.3:
3436 * Changes to mailing lists
3438 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3440 See the README file to find current addresses for all the Guile
3443 * Changes to the distribution
3445 ** Readline support is no longer included with Guile by default.
3447 Based on the different license terms of Guile and Readline, we
3448 concluded that Guile should not *by default* cause the linking of
3449 Readline into an application program. Readline support is now offered
3450 as a separate module, which is linked into an application only when
3451 you explicitly specify it.
3453 Although Guile is GNU software, its distribution terms add a special
3454 exception to the usual GNU General Public License (GPL). Guile's
3455 license includes a clause that allows you to link Guile with non-free
3456 programs. We add this exception so as not to put Guile at a
3457 disadvantage vis-a-vis other extensibility packages that support other
3460 In contrast, the GNU Readline library is distributed under the GNU
3461 General Public License pure and simple. This means that you may not
3462 link Readline, even dynamically, into an application unless it is
3463 distributed under a free software license that is compatible the GPL.
3465 Because of this difference in distribution terms, an application that
3466 can use Guile may not be able to use Readline. Now users will be
3467 explicitly offered two independent decisions about the use of these
3470 You can activate the readline support by issuing
3472 (use-modules (readline-activator))
3475 from your ".guile" file, for example.
3477 * Changes to the stand-alone interpreter
3479 ** All builtins now print as primitives.
3480 Previously builtin procedures not belonging to the fundamental subr
3481 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3482 Now, they print as #<primitive-procedure NAME>.
3484 ** Backtraces slightly more intelligible.
3485 gsubr-apply and macro transformer application frames no longer appear
3488 * Changes to Scheme functions and syntax
3490 ** Guile now correctly handles internal defines by rewriting them into
3491 their equivalent letrec. Previously, internal defines would
3492 incrementally add to the innermost environment, without checking
3493 whether the restrictions specified in RnRS were met. This lead to the
3494 correct behaviour when these restriction actually were met, but didn't
3495 catch all illegal uses. Such an illegal use could lead to crashes of
3496 the Guile interpreter or or other unwanted results. An example of
3497 incorrect internal defines that made Guile behave erratically:
3509 The problem with this example is that the definition of `c' uses the
3510 value of `b' directly. This confuses the meoization machine of Guile
3511 so that the second call of `b' (this time in a larger environment that
3512 also contains bindings for `c' and `d') refers to the binding of `c'
3513 instead of `a'. You could also make Guile crash with a variation on
3518 (define (b flag) (if flag a 1))
3519 (define c (1+ (b flag)))
3527 From now on, Guile will issue an `Unbound variable: b' error message
3532 A hook contains a list of functions which should be called on
3533 particular occasions in an existing program. Hooks are used for
3536 A window manager might have a hook before-window-map-hook. The window
3537 manager uses the function run-hooks to call all functions stored in
3538 before-window-map-hook each time a window is mapped. The user can
3539 store functions in the hook using add-hook!.
3541 In Guile, hooks are first class objects.
3543 *** New function: make-hook [N_ARGS]
3545 Return a hook for hook functions which can take N_ARGS arguments.
3546 The default value for N_ARGS is 0.
3548 (See also scm_make_named_hook below.)
3550 *** New function: add-hook! HOOK PROC [APPEND_P]
3552 Put PROC at the beginning of the list of functions stored in HOOK.
3553 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3555 PROC must be able to take the number of arguments specified when the
3558 If PROC already exists in HOOK, then remove it first.
3560 *** New function: remove-hook! HOOK PROC
3562 Remove PROC from the list of functions in HOOK.
3564 *** New function: reset-hook! HOOK
3566 Clear the list of hook functions stored in HOOK.
3568 *** New function: run-hook HOOK ARG1 ...
3570 Run all hook functions stored in HOOK with arguments ARG1 ... .
3571 The number of arguments supplied must correspond to the number given
3572 when the hook was created.
3574 ** The function `dynamic-link' now takes optional keyword arguments.
3575 The only keyword argument that is currently defined is `:global
3576 BOOL'. With it, you can control whether the shared library will be
3577 linked in global mode or not. In global mode, the symbols from the
3578 linked library can be used to resolve references from other
3579 dynamically linked libraries. In non-global mode, the linked
3580 library is essentially invisible and can only be accessed via
3581 `dynamic-func', etc. The default is now to link in global mode.
3582 Previously, the default has been non-global mode.
3584 The `#:global' keyword is only effective on platforms that support
3585 the dlopen family of functions.
3587 ** New function `provided?'
3589 - Function: provided? FEATURE
3590 Return true iff FEATURE is supported by this installation of
3591 Guile. FEATURE must be a symbol naming a feature; the global
3592 variable `*features*' is a list of available features.
3594 ** Changes to the module (ice-9 expect):
3596 *** The expect-strings macro now matches `$' in a regular expression
3597 only at a line-break or end-of-file by default. Previously it would
3598 match the end of the string accumulated so far. The old behaviour
3599 can be obtained by setting the variable `expect-strings-exec-flags'
3602 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3603 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3604 in a regular expression will still match before a line-break or
3605 end-of-file. The default is `regexp/noteol'.
3607 *** The expect-strings macro now uses a variable
3608 `expect-strings-compile-flags' for the flags to be supplied to
3609 `make-regexp'. The default is `regexp/newline', which was previously
3612 *** The expect macro now supplies two arguments to a match procedure:
3613 the current accumulated string and a flag to indicate whether
3614 end-of-file has been reached. Previously only the string was supplied.
3615 If end-of-file is reached, the match procedure will be called an
3616 additional time with the same accumulated string as the previous call
3617 but with the flag set.
3619 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3621 This code, and the documentation for it that appears here, was
3622 borrowed from SLIB, with minor adaptations for Guile.
3624 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3625 An almost complete implementation of Common LISP format description
3626 according to the CL reference book `Common LISP' from Guy L.
3627 Steele, Digital Press. Backward compatible to most of the
3628 available Scheme format implementations.
3630 Returns `#t', `#f' or a string; has side effect of printing
3631 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3632 to the current output port and `#t' is returned. If DESTINATION
3633 is `#f', a formatted string is returned as the result of the call.
3634 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3635 format string; FORMAT-STRING is then the first argument and the
3636 output is returned as a string. If DESTINATION is a number, the
3637 output is to the current error port if available by the
3638 implementation. Otherwise DESTINATION must be an output port and
3641 FORMAT-STRING must be a string. In case of a formatting error
3642 format returns `#f' and prints a message on the current output or
3643 error port. Characters are output as if the string were output by
3644 the `display' function with the exception of those prefixed by a
3645 tilde (~). For a detailed description of the FORMAT-STRING syntax
3646 please consult a Common LISP format reference manual. For a test
3647 suite to verify this format implementation load `formatst.scm'.
3648 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3650 Note: `format' is not reentrant, i.e. only one `format'-call may
3651 be executed at a time.
3654 *** Format Specification (Format version 3.0)
3656 Please consult a Common LISP format reference manual for a detailed
3657 description of the format string syntax. For a demonstration of the
3658 implemented directives see `formatst.scm'.
3660 This implementation supports directive parameters and modifiers (`:'
3661 and `@' characters). Multiple parameters must be separated by a comma
3662 (`,'). Parameters can be numerical parameters (positive or negative),
3663 character parameters (prefixed by a quote character (`''), variable
3664 parameters (`v'), number of rest arguments parameter (`#'), empty and
3665 default parameters. Directive characters are case independent. The
3666 general form of a directive is:
3668 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3670 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3672 *** Implemented CL Format Control Directives
3674 Documentation syntax: Uppercase characters represent the
3675 corresponding control directive characters. Lowercase characters
3676 represent control directive parameter descriptions.
3679 Any (print as `display' does).
3683 `~MINCOL,COLINC,MINPAD,PADCHARA'
3687 S-expression (print as `write' does).
3691 `~MINCOL,COLINC,MINPAD,PADCHARS'
3697 print number sign always.
3700 print comma separated.
3702 `~MINCOL,PADCHAR,COMMACHARD'
3708 print number sign always.
3711 print comma separated.
3713 `~MINCOL,PADCHAR,COMMACHARX'
3719 print number sign always.
3722 print comma separated.
3724 `~MINCOL,PADCHAR,COMMACHARO'
3730 print number sign always.
3733 print comma separated.
3735 `~MINCOL,PADCHAR,COMMACHARB'
3740 `~N,MINCOL,PADCHAR,COMMACHARR'
3744 print a number as a Roman numeral.
3747 print a number as an "old fashioned" Roman numeral.
3750 print a number as an ordinal English number.
3753 print a number as a cardinal English number.
3758 prints `y' and `ies'.
3761 as `~P but jumps 1 argument backward.'
3764 as `~@P but jumps 1 argument backward.'
3769 prints a character as the reader can understand it (i.e. `#\'
3773 prints a character as emacs does (eg. `^C' for ASCII 03).
3776 Fixed-format floating-point (prints a flonum like MMM.NNN).
3777 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3779 If the number is positive a plus sign is printed.
3782 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3783 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3785 If the number is positive a plus sign is printed.
3788 General floating-point (prints a flonum either fixed or
3790 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3792 If the number is positive a plus sign is printed.
3795 Dollars floating-point (prints a flonum in fixed with signs
3797 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3799 If the number is positive a plus sign is printed.
3802 A sign is always printed and appears before the padding.
3805 The sign appears before the padding.
3813 print newline if not at the beginning of the output line.
3815 prints `~&' and then N-1 newlines.
3820 print N page separators.
3830 newline is ignored, white space left.
3833 newline is left, white space ignored.
3838 relative tabulation.
3844 Indirection (expects indirect arguments as a list).
3846 extracts indirect arguments from format arguments.
3849 Case conversion (converts by `string-downcase').
3851 converts by `string-capitalize'.
3854 converts by `string-capitalize-first'.
3857 converts by `string-upcase'.
3860 Argument Jumping (jumps 1 argument forward).
3862 jumps N arguments forward.
3865 jumps 1 argument backward.
3868 jumps N arguments backward.
3871 jumps to the 0th argument.
3874 jumps to the Nth argument (beginning from 0)
3876 `~[STR0~;STR1~;...~;STRN~]'
3877 Conditional Expression (numerical clause conditional).
3879 take argument from N.
3882 true test conditional.
3885 if-else-then conditional.
3891 default clause follows.
3894 Iteration (args come from the next argument (a list)).
3896 at most N iterations.
3899 args from next arg (a list of lists).
3902 args from the rest of arguments.
3905 args from the rest args (lists).
3916 aborts if N <= M <= K
3918 *** Not Implemented CL Format Control Directives
3921 print `#f' as an empty list (see below).
3924 print `#f' as an empty list (see below).
3930 (sorry I don't understand its semantics completely)
3932 *** Extended, Replaced and Additional Control Directives
3934 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3935 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3936 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3937 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3938 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3939 COMMAWIDTH is the number of characters between two comma
3943 print a R4RS complex number as `~F~@Fi' with passed parameters for
3947 Pretty print formatting of an argument for scheme code lists.
3953 Flushes the output if format DESTINATION is a port.
3956 Print a `#\space' character
3958 print N `#\space' characters.
3961 Print a `#\tab' character
3963 print N `#\tab' characters.
3966 Takes N as an integer representation for a character. No arguments
3967 are consumed. N is converted to a character by `integer->char'. N
3968 must be a positive decimal number.
3971 Print out readproof. Prints out internal objects represented as
3972 `#<...>' as strings `"#<...>"' so that the format output can always
3973 be processed by `read'.
3976 Print out readproof. Prints out internal objects represented as
3977 `#<...>' as strings `"#<...>"' so that the format output can always
3978 be processed by `read'.
3981 Prints information and a copyright notice on the format
3984 prints format version.
3987 may also print number strings, i.e. passing a number as a string
3988 and format it accordingly.
3990 *** Configuration Variables
3992 The format module exports some configuration variables to suit the
3993 systems and users needs. There should be no modification necessary for
3994 the configuration that comes with Guile. Format detects automatically
3995 if the running scheme system implements floating point numbers and
3998 format:symbol-case-conv
3999 Symbols are converted by `symbol->string' so the case type of the
4000 printed symbols is implementation dependent.
4001 `format:symbol-case-conv' is a one arg closure which is either
4002 `#f' (no conversion), `string-upcase', `string-downcase' or
4003 `string-capitalize'. (default `#f')
4005 format:iobj-case-conv
4006 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4007 implementation internal objects. (default `#f')
4010 The character prefixing the exponent value in `~E' printing.
4013 *** Compatibility With Other Format Implementations
4019 Downward compatible except for padding support and `~A', `~S',
4020 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4021 `printf' padding support which is completely replaced by the CL
4022 `format' padding style.
4025 Downward compatible except for `~', which is not documented
4026 (ignores all characters inside the format string up to a newline
4027 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4028 numerical and variable parameters and `:/@' modifiers in the CL
4032 Downward compatible except for `~A' and `~S' which print in
4033 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4034 directive parameters or modifiers)).
4037 Downward compatible except for an optional destination parameter:
4038 S2C accepts a format call without a destination which returns a
4039 formatted string. This is equivalent to a #f destination in S2C.
4040 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4041 parameters or modifiers)).
4044 ** Changes to string-handling functions.
4046 These functions were added to support the (ice-9 format) module, above.
4048 *** New function: string-upcase STRING
4049 *** New function: string-downcase STRING
4051 These are non-destructive versions of the existing string-upcase! and
4052 string-downcase! functions.
4054 *** New function: string-capitalize! STRING
4055 *** New function: string-capitalize STRING
4057 These functions convert the first letter of each word in the string to
4060 (string-capitalize "howdy there")
4063 As with the other functions, string-capitalize! modifies the string in
4064 place, while string-capitalize returns a modified copy of its argument.
4066 *** New function: string-ci->symbol STRING
4068 Return a symbol whose name is STRING, but having the same case as if
4069 the symbol had be read by `read'.
4071 Guile can be configured to be sensitive or insensitive to case
4072 differences in Scheme identifiers. If Guile is case-insensitive, all
4073 symbols are converted to lower case on input. The `string-ci->symbol'
4074 function returns a symbol whose name in STRING, transformed as Guile
4075 would if STRING were input.
4077 *** New function: substring-move! STRING1 START END STRING2 START
4079 Copy the substring of STRING1 from START (inclusive) to END
4080 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4081 string, and the source and destination areas may overlap; in all
4082 cases, the function behaves as if all the characters were copied
4085 *** Extended functions: substring-move-left! substring-move-right!
4087 These functions now correctly copy arbitrarily overlapping substrings;
4088 they are both synonyms for substring-move!.
4091 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4093 getopt-long is a function for parsing command-line arguments in a
4094 manner consistent with other GNU programs.
4096 (getopt-long ARGS GRAMMAR)
4097 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4099 ARGS should be a list of strings. Its first element should be the
4100 name of the program; subsequent elements should be the arguments
4101 that were passed to the program on the command line. The
4102 `program-arguments' procedure returns a list of this form.
4104 GRAMMAR is a list of the form:
4105 ((OPTION (PROPERTY VALUE) ...) ...)
4107 Each OPTION should be a symbol. `getopt-long' will accept a
4108 command-line option named `--OPTION'.
4109 Each option can have the following (PROPERTY VALUE) pairs:
4111 (single-char CHAR) --- Accept `-CHAR' as a single-character
4112 equivalent to `--OPTION'. This is how to specify traditional
4114 (required? BOOL) --- If BOOL is true, the option is required.
4115 getopt-long will raise an error if it is not found in ARGS.
4116 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4117 it is #f, it does not; and if it is the symbol
4118 `optional', the option may appear in ARGS with or
4120 (predicate FUNC) --- If the option accepts a value (i.e. you
4121 specified `(value #t)' for this option), then getopt
4122 will apply FUNC to the value, and throw an exception
4123 if it returns #f. FUNC should be a procedure which
4124 accepts a string and returns a boolean value; you may
4125 need to use quasiquotes to get it into GRAMMAR.
4127 The (PROPERTY VALUE) pairs may occur in any order, but each
4128 property may occur only once. By default, options do not have
4129 single-character equivalents, are not required, and do not take
4132 In ARGS, single-character options may be combined, in the usual
4133 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4134 accepts values, then it must be the last option in the
4135 combination; the value is the next argument. So, for example, using
4136 the following grammar:
4137 ((apples (single-char #\a))
4138 (blimps (single-char #\b) (value #t))
4139 (catalexis (single-char #\c) (value #t)))
4140 the following argument lists would be acceptable:
4141 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4142 for "blimps" and "catalexis")
4143 ("-ab" "bang" "-c" "couth") (same)
4144 ("-ac" "couth" "-b" "bang") (same)
4145 ("-abc" "couth" "bang") (an error, since `-b' is not the
4146 last option in its combination)
4148 If an option's value is optional, then `getopt-long' decides
4149 whether it has a value by looking at what follows it in ARGS. If
4150 the next element is a string, and it does not appear to be an
4151 option itself, then that string is the option's value.
4153 The value of a long option can appear as the next element in ARGS,
4154 or it can follow the option name, separated by an `=' character.
4155 Thus, using the same grammar as above, the following argument lists
4157 ("--apples" "Braeburn" "--blimps" "Goodyear")
4158 ("--apples=Braeburn" "--blimps" "Goodyear")
4159 ("--blimps" "Goodyear" "--apples=Braeburn")
4161 If the option "--" appears in ARGS, argument parsing stops there;
4162 subsequent arguments are returned as ordinary arguments, even if
4163 they resemble options. So, in the argument list:
4164 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4165 `getopt-long' will recognize the `apples' option as having the
4166 value "Granny Smith", but it will not recognize the `blimp'
4167 option; it will return the strings "--blimp" and "Goodyear" as
4168 ordinary argument strings.
4170 The `getopt-long' function returns the parsed argument list as an
4171 assocation list, mapping option names --- the symbols from GRAMMAR
4172 --- onto their values, or #t if the option does not accept a value.
4173 Unused options do not appear in the alist.
4175 All arguments that are not the value of any option are returned
4176 as a list, associated with the empty list.
4178 `getopt-long' throws an exception if:
4179 - it finds an unrecognized option in ARGS
4180 - a required option is omitted
4181 - an option that requires an argument doesn't get one
4182 - an option that doesn't accept an argument does get one (this can
4183 only happen using the long option `--opt=value' syntax)
4184 - an option predicate fails
4189 `((lockfile-dir (required? #t)
4192 (predicate ,file-is-directory?))
4193 (verbose (required? #f)
4196 (x-includes (single-char #\x))
4197 (rnet-server (single-char #\y)
4198 (predicate ,string?))))
4200 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4201 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4203 => ((() "foo1" "-fred" "foo2" "foo3")
4204 (rnet-server . "lamprod")
4205 (x-includes . "/usr/include")
4206 (lockfile-dir . "/tmp")
4209 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4211 It will be removed in a few releases.
4213 ** New syntax: lambda*
4214 ** New syntax: define*
4215 ** New syntax: define*-public
4216 ** New syntax: defmacro*
4217 ** New syntax: defmacro*-public
4218 Guile now supports optional arguments.
4220 `lambda*', `define*', `define*-public', `defmacro*' and
4221 `defmacro*-public' are identical to the non-* versions except that
4222 they use an extended type of parameter list that has the following BNF
4223 syntax (parentheses are literal, square brackets indicate grouping,
4224 and `*', `+' and `?' have the usual meaning):
4226 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4227 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4228 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4230 ext-var-decl ::= identifier | ( identifier expression )
4232 The semantics are best illustrated with the following documentation
4233 and examples for `lambda*':
4236 lambda extended for optional and keyword arguments
4238 lambda* creates a procedure that takes optional arguments. These
4239 are specified by putting them inside brackets at the end of the
4240 paramater list, but before any dotted rest argument. For example,
4241 (lambda* (a b #&optional c d . e) '())
4242 creates a procedure with fixed arguments a and b, optional arguments c
4243 and d, and rest argument e. If the optional arguments are omitted
4244 in a call, the variables for them are unbound in the procedure. This
4245 can be checked with the bound? macro.
4247 lambda* can also take keyword arguments. For example, a procedure
4249 (lambda* (#&key xyzzy larch) '())
4250 can be called with any of the argument lists (#:xyzzy 11)
4251 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4252 are given as keywords are bound to values.
4254 Optional and keyword arguments can also be given default values
4255 which they take on when they are not present in a call, by giving a
4256 two-item list in place of an optional argument, for example in:
4257 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4258 foo is a fixed argument, bar is an optional argument with default
4259 value 42, and baz is a keyword argument with default value 73.
4260 Default value expressions are not evaluated unless they are needed
4261 and until the procedure is called.
4263 lambda* now supports two more special parameter list keywords.
4265 lambda*-defined procedures now throw an error by default if a
4266 keyword other than one of those specified is found in the actual
4267 passed arguments. However, specifying #&allow-other-keys
4268 immediately after the kyword argument declarations restores the
4269 previous behavior of ignoring unknown keywords. lambda* also now
4270 guarantees that if the same keyword is passed more than once, the
4271 last one passed is the one that takes effect. For example,
4272 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4273 #:heads 37 #:tails 42 #:heads 99)
4274 would result in (99 47) being displayed.
4276 #&rest is also now provided as a synonym for the dotted syntax rest
4277 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4278 all respects to lambda*. This is provided for more similarity to DSSSL,
4279 MIT-Scheme and Kawa among others, as well as for refugees from other
4282 Further documentation may be found in the optargs.scm file itself.
4284 The optional argument module also exports the macros `let-optional',
4285 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4286 are not documented here because they may be removed in the future, but
4287 full documentation is still available in optargs.scm.
4289 ** New syntax: and-let*
4290 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4292 Syntax: (land* (<clause> ...) <body> ...)
4293 Each <clause> should have one of the following forms:
4294 (<variable> <expression>)
4297 Each <variable> or <bound-variable> should be an identifier. Each
4298 <expression> should be a valid expression. The <body> should be a
4299 possibly empty sequence of expressions, like the <body> of a
4302 Semantics: A LAND* expression is evaluated by evaluating the
4303 <expression> or <bound-variable> of each of the <clause>s from
4304 left to right. The value of the first <expression> or
4305 <bound-variable> that evaluates to a false value is returned; the
4306 remaining <expression>s and <bound-variable>s are not evaluated.
4307 The <body> forms are evaluated iff all the <expression>s and
4308 <bound-variable>s evaluate to true values.
4310 The <expression>s and the <body> are evaluated in an environment
4311 binding each <variable> of the preceding (<variable> <expression>)
4312 clauses to the value of the <expression>. Later bindings
4313 shadow earlier bindings.
4315 Guile's and-let* macro was contributed by Michael Livshin.
4317 ** New sorting functions
4319 *** New function: sorted? SEQUENCE LESS?
4320 Returns `#t' when the sequence argument is in non-decreasing order
4321 according to LESS? (that is, there is no adjacent pair `... x y
4322 ...' for which `(less? y x)').
4324 Returns `#f' when the sequence contains at least one out-of-order
4325 pair. It is an error if the sequence is neither a list nor a
4328 *** New function: merge LIST1 LIST2 LESS?
4329 LIST1 and LIST2 are sorted lists.
4330 Returns the sorted list of all elements in LIST1 and LIST2.
4332 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4333 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4334 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4335 (Here "<" should read "comes before".)
4337 *** New procedure: merge! LIST1 LIST2 LESS?
4338 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4339 the result. If the code is compiled, and LESS? constructs no new
4340 pairs, no pairs at all will be allocated. The first pair of the
4341 result will be either the first pair of LIST1 or the first pair of
4344 *** New function: sort SEQUENCE LESS?
4345 Accepts either a list or a vector, and returns a new sequence
4346 which is sorted. The new sequence is the same type as the input.
4347 Always `(sorted? (sort sequence less?) less?)'. The original
4348 sequence is not altered in any way. The new sequence shares its
4349 elements with the old one; no elements are copied.
4351 *** New procedure: sort! SEQUENCE LESS
4352 Returns its sorted result in the original boxes. No new storage is
4353 allocated at all. Proper usage: (set! slist (sort! slist <))
4355 *** New function: stable-sort SEQUENCE LESS?
4356 Similar to `sort' but stable. That is, if "equal" elements are
4357 ordered a < b in the original sequence, they will have the same order
4360 *** New function: stable-sort! SEQUENCE LESS?
4361 Similar to `sort!' but stable.
4362 Uses temporary storage when sorting vectors.
4364 *** New functions: sort-list, sort-list!
4365 Added for compatibility with scsh.
4367 ** New built-in random number support
4369 *** New function: random N [STATE]
4370 Accepts a positive integer or real N and returns a number of the
4371 same type between zero (inclusive) and N (exclusive). The values
4372 returned have a uniform distribution.
4374 The optional argument STATE must be of the type produced by
4375 `copy-random-state' or `seed->random-state'. It defaults to the value
4376 of the variable `*random-state*'. This object is used to maintain the
4377 state of the pseudo-random-number generator and is altered as a side
4378 effect of the `random' operation.
4380 *** New variable: *random-state*
4381 Holds a data structure that encodes the internal state of the
4382 random-number generator that `random' uses by default. The nature
4383 of this data structure is implementation-dependent. It may be
4384 printed out and successfully read back in, but may or may not
4385 function correctly as a random-number state object in another
4388 *** New function: copy-random-state [STATE]
4389 Returns a new object of type suitable for use as the value of the
4390 variable `*random-state*' and as a second argument to `random'.
4391 If argument STATE is given, a copy of it is returned. Otherwise a
4392 copy of `*random-state*' is returned.
4394 *** New function: seed->random-state SEED
4395 Returns a new object of type suitable for use as the value of the
4396 variable `*random-state*' and as a second argument to `random'.
4397 SEED is a string or a number. A new state is generated and
4398 initialized using SEED.
4400 *** New function: random:uniform [STATE]
4401 Returns an uniformly distributed inexact real random number in the
4402 range between 0 and 1.
4404 *** New procedure: random:solid-sphere! VECT [STATE]
4405 Fills VECT with inexact real random numbers the sum of whose
4406 squares is less than 1.0. Thinking of VECT as coordinates in
4407 space of dimension N = `(vector-length VECT)', the coordinates are
4408 uniformly distributed within the unit N-shere. The sum of the
4409 squares of the numbers is returned. VECT can be either a vector
4410 or a uniform vector of doubles.
4412 *** New procedure: random:hollow-sphere! VECT [STATE]
4413 Fills VECT with inexact real random numbers the sum of whose squares
4414 is equal to 1.0. Thinking of VECT as coordinates in space of
4415 dimension n = `(vector-length VECT)', the coordinates are uniformly
4416 distributed over the surface of the unit n-shere. VECT can be either
4417 a vector or a uniform vector of doubles.
4419 *** New function: random:normal [STATE]
4420 Returns an inexact real in a normal distribution with mean 0 and
4421 standard deviation 1. For a normal distribution with mean M and
4422 standard deviation D use `(+ M (* D (random:normal)))'.
4424 *** New procedure: random:normal-vector! VECT [STATE]
4425 Fills VECT with inexact real random numbers which are independent and
4426 standard normally distributed (i.e., with mean 0 and variance 1).
4427 VECT can be either a vector or a uniform vector of doubles.
4429 *** New function: random:exp STATE
4430 Returns an inexact real in an exponential distribution with mean 1.
4431 For an exponential distribution with mean U use (* U (random:exp)).
4433 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4435 These functions now operate on numbers in the range of a C unsigned
4438 These functions used to operate on numbers in the range of a C signed
4439 long; however, this seems inappropriate, because Guile integers don't
4442 ** New function: make-guardian
4443 This is an implementation of guardians as described in
4444 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4445 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4446 Programming Language Design and Implementation, June 1993
4447 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4449 ** New functions: delq1!, delv1!, delete1!
4450 These procedures behave similar to delq! and friends but delete only
4451 one object if at all.
4453 ** New function: unread-string STRING PORT
4454 Unread STRING to PORT, that is, push it back onto the port so that
4455 next read operation will work on the pushed back characters.
4457 ** unread-char can now be called multiple times
4458 If unread-char is called multiple times, the unread characters will be
4459 read again in last-in first-out order.
4461 ** the procedures uniform-array-read! and uniform-array-write! now
4462 work on any kind of port, not just ports which are open on a file.
4464 ** Now 'l' in a port mode requests line buffering.
4466 ** The procedure truncate-file now works on string ports as well
4467 as file ports. If the size argument is omitted, the current
4468 file position is used.
4470 ** new procedure: seek PORT/FDES OFFSET WHENCE
4471 The arguments are the same as for the old fseek procedure, but it
4472 works on string ports as well as random-access file ports.
4474 ** the fseek procedure now works on string ports, since it has been
4475 redefined using seek.
4477 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4478 size is not supplied.
4480 ** the newline procedure no longer flushes the port if it's not
4481 line-buffered: previously it did if it was the current output port.
4483 ** open-pipe and close-pipe are no longer primitive procedures, but
4484 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4486 ** the freopen procedure has been removed.
4488 ** new procedure: drain-input PORT
4489 Drains PORT's read buffers (including any pushed-back characters)
4490 and returns the contents as a single string.
4492 ** New function: map-in-order PROC LIST1 LIST2 ...
4493 Version of `map' which guarantees that the procedure is applied to the
4494 lists in serial order.
4496 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4497 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4498 now obsolete and will go away in release 1.5.
4500 ** New syntax: collect BODY1 ...
4501 Version of `begin' which returns a list of the results of the body
4502 forms instead of the result of the last body form. In contrast to
4503 `begin', `collect' allows an empty body.
4505 ** New functions: read-history FILENAME, write-history FILENAME
4506 Read/write command line history from/to file. Returns #t on success
4507 and #f if an error occured.
4509 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4511 These procedures return a list of definitions available in the specified
4512 argument, a relative module reference. In the case of no argument,
4513 `(current-module)' is now consulted for definitions to return, instead
4514 of simply returning #f, the former behavior.
4516 ** The #/ syntax for lists is no longer supported.
4518 Earlier versions of Scheme accepted this syntax, but printed a
4521 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4523 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4526 * Changes to the gh_ interface
4530 Now takes a second argument which is the result array. If this
4531 pointer is NULL, a new array is malloced (the old behaviour).
4533 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4534 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4538 * Changes to the scm_ interface
4540 ** Function: scm_make_named_hook (char* name, int n_args)
4542 Creates a hook in the same way as make-hook above but also
4543 binds a variable named NAME to it.
4545 This is the typical way of creating a hook from C code.
4547 Currently, the variable is created in the "current" module. This
4548 might change when we get the new module system.
4550 ** The smob interface
4552 The interface for creating smobs has changed. For documentation, see
4553 data-rep.info (made from guile-core/doc/data-rep.texi).
4555 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4557 >>> This function will be removed in 1.3.4. <<<
4561 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4562 This function adds a new smob type, named NAME, with instance size
4563 SIZE to the system. The return value is a tag that is used in
4564 creating instances of the type. If SIZE is 0, then no memory will
4565 be allocated when instances of the smob are created, and nothing
4566 will be freed by the default free function.
4568 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4569 This function sets the smob marking procedure for the smob type
4570 specified by the tag TC. TC is the tag returned by
4571 `scm_make_smob_type'.
4573 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4574 This function sets the smob freeing procedure for the smob type
4575 specified by the tag TC. TC is the tag returned by
4576 `scm_make_smob_type'.
4578 *** Function: void scm_set_smob_print (tc, print)
4580 - Function: void scm_set_smob_print (long tc,
4581 scm_sizet (*print) (SCM,
4585 This function sets the smob printing procedure for the smob type
4586 specified by the tag TC. TC is the tag returned by
4587 `scm_make_smob_type'.
4589 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4590 This function sets the smob equality-testing predicate for the
4591 smob type specified by the tag TC. TC is the tag returned by
4592 `scm_make_smob_type'.
4594 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4595 Make VALUE contain a smob instance of the type with type code TC and
4596 smob data DATA. VALUE must be previously declared as C type `SCM'.
4598 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4599 This macro expands to a block of code that creates a smob instance
4600 of the type with type code TC and smob data DATA, and returns that
4601 `SCM' value. It should be the last piece of code in a block.
4603 ** The interfaces for using I/O ports and implementing port types
4604 (ptobs) have changed significantly. The new interface is based on
4605 shared access to buffers and a new set of ptob procedures.
4607 *** scm_newptob has been removed
4611 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4613 - Function: SCM scm_make_port_type (char *type_name,
4614 int (*fill_buffer) (SCM port),
4615 void (*write_flush) (SCM port));
4617 Similarly to the new smob interface, there is a set of function
4618 setters by which the user can customize the behaviour of his port
4619 type. See ports.h (scm_set_port_XXX).
4621 ** scm_strport_to_string: New function: creates a new string from
4622 a string port's buffer.
4624 ** Plug in interface for random number generators
4625 The variable `scm_the_rng' in random.c contains a value and three
4626 function pointers which together define the current random number
4627 generator being used by the Scheme level interface and the random
4628 number library functions.
4630 The user is free to replace the default generator with the generator
4633 *** Variable: size_t scm_the_rng.rstate_size
4634 The size of the random state type used by the current RNG
4637 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4638 Given the random STATE, return 32 random bits.
4640 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4641 Seed random state STATE using string S of length N.
4643 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4644 Given random state STATE, return a malloced copy.
4647 The default RNG is the MWC (Multiply With Carry) random number
4648 generator described by George Marsaglia at the Department of
4649 Statistics and Supercomputer Computations Research Institute, The
4650 Florida State University (http://stat.fsu.edu/~geo).
4652 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4653 passes all tests in the DIEHARD test suite
4654 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4655 costs one multiply and one add on platforms which either supports long
4656 longs (gcc does this on most systems) or have 64 bit longs. The cost
4657 is four multiply on other systems but this can be optimized by writing
4658 scm_i_uniform32 in assembler.
4660 These functions are provided through the scm_the_rng interface for use
4661 by libguile and the application.
4663 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4664 Given the random STATE, return 32 random bits.
4665 Don't use this function directly. Instead go through the plugin
4666 interface (see "Plug in interface" above).
4668 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4669 Initialize STATE using SEED of length N.
4671 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4672 Return a malloc:ed copy of STATE. This function can easily be re-used
4673 in the interfaces to other RNGs.
4675 ** Random number library functions
4676 These functions use the current RNG through the scm_the_rng interface.
4677 It might be a good idea to use these functions from your C code so
4678 that only one random generator is used by all code in your program.
4680 The default random state is stored in:
4682 *** Variable: SCM scm_var_random_state
4683 Contains the vcell of the Scheme variable "*random-state*" which is
4684 used as default state by all random number functions in the Scheme
4689 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4691 *** Function: scm_rstate *scm_c_default_rstate (void)
4692 This is a convenience function which returns the value of
4693 scm_var_random_state. An error message is generated if this value
4694 isn't a random state.
4696 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4697 Make a new random state from the string SEED of length LENGTH.
4699 It is generally not a good idea to use multiple random states in a
4700 program. While subsequent random numbers generated from one random
4701 state are guaranteed to be reasonably independent, there is no such
4702 guarantee for numbers generated from different random states.
4704 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4705 Return 32 random bits.
4707 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4708 Return a sample from the uniform(0,1) distribution.
4710 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4711 Return a sample from the normal(0,1) distribution.
4713 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4714 Return a sample from the exp(1) distribution.
4716 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4717 Return a sample from the discrete uniform(0,M) distribution.
4719 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4720 Return a sample from the discrete uniform(0,M) distribution.
4721 M must be a bignum object. The returned value may be an INUM.
4725 Changes in Guile 1.3 (released Monday, October 19, 1998):
4727 * Changes to the distribution
4729 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4730 To avoid conflicts, programs should name environment variables after
4731 themselves, except when there's a common practice establishing some
4734 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4735 giving the former precedence, and printing a warning message if the
4736 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4738 ** The header files related to multi-byte characters have been removed.
4739 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4740 which referred to these explicitly will probably need to be rewritten,
4741 since the support for the variant string types has been removed; see
4744 ** The header files append.h and sequences.h have been removed. These
4745 files implemented non-R4RS operations which would encourage
4746 non-portable programming style and less easy-to-read code.
4748 * Changes to the stand-alone interpreter
4750 ** New procedures have been added to implement a "batch mode":
4752 *** Function: batch-mode?
4754 Returns a boolean indicating whether the interpreter is in batch
4757 *** Function: set-batch-mode?! ARG
4759 If ARG is true, switches the interpreter to batch mode. The `#f'
4760 case has not been implemented.
4762 ** Guile now provides full command-line editing, when run interactively.
4763 To use this feature, you must have the readline library installed.
4764 The Guile build process will notice it, and automatically include
4767 The readline library is available via anonymous FTP from any GNU
4768 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4770 ** the-last-stack is now a fluid.
4772 * Changes to the procedure for linking libguile with your programs
4774 ** You can now use the `guile-config' utility to build programs that use Guile.
4776 Guile now includes a command-line utility called `guile-config', which
4777 can provide information about how to compile and link programs that
4780 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4781 You should include this command's output on the command line you use
4782 to compile C or C++ code that #includes the Guile header files. It's
4783 usually just a `-I' flag to help the compiler find the Guile headers.
4786 *** `guile-config link' prints any linker flags necessary to link with Guile.
4788 This command writes to its standard output a list of flags which you
4789 must pass to the linker to link your code against the Guile library.
4790 The flags include '-lguile' itself, any other libraries the Guile
4791 library depends upon, and any `-L' flags needed to help the linker
4792 find those libraries.
4794 For example, here is a Makefile rule that builds a program named 'foo'
4795 from the object files ${FOO_OBJECTS}, and links them against Guile:
4798 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4800 Previous Guile releases recommended that you use autoconf to detect
4801 which of a predefined set of libraries were present on your system.
4802 It is more robust to use `guile-config', since it records exactly which
4803 libraries the installed Guile library requires.
4805 This was originally called `build-guile', but was renamed to
4806 `guile-config' before Guile 1.3 was released, to be consistent with
4807 the analogous script for the GTK+ GUI toolkit, which is called
4811 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4813 If you are using the GNU autoconf package to configure your program,
4814 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4815 (described above) and gather the necessary values for use in your
4818 The GUILE_FLAGS macro expands to configure script code which runs the
4819 `guile-config' script, to find out where Guile's header files and
4820 libraries are installed. It sets two variables, marked for
4821 substitution, as by AC_SUBST.
4823 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4824 code that uses Guile header files. This is almost always just a
4827 GUILE_LDFLAGS --- flags to pass to the linker to link a
4828 program against Guile. This includes `-lguile' for the Guile
4829 library itself, any libraries that Guile itself requires (like
4830 -lqthreads), and so on. It may also include a -L flag to tell the
4831 compiler where to find the libraries.
4833 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4834 directory of the Guile distribution. You can copy it into your
4835 package's aclocal.m4 file, and then use it in your configure.in file.
4837 If you are using the `aclocal' program, distributed with GNU automake,
4838 to maintain your aclocal.m4 file, the Guile installation process
4839 installs guile.m4 where aclocal will find it. All you need to do is
4840 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4841 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4845 * Changes to Scheme functions and syntax
4847 ** Multi-byte strings have been removed, as have multi-byte and wide
4848 ports. We felt that these were the wrong approach to
4849 internationalization support.
4851 ** New function: readline [PROMPT]
4852 Read a line from the terminal, and allow the user to edit it,
4853 prompting with PROMPT. READLINE provides a large set of Emacs-like
4854 editing commands, lets the user recall previously typed lines, and
4855 works on almost every kind of terminal, including dumb terminals.
4857 READLINE assumes that the cursor is at the beginning of the line when
4858 it is invoked. Thus, you can't print a prompt yourself, and then call
4859 READLINE; you need to package up your prompt as a string, pass it to
4860 the function, and let READLINE print the prompt itself. This is
4861 because READLINE needs to know the prompt's screen width.
4863 For Guile to provide this function, you must have the readline
4864 library, version 2.1 or later, installed on your system. Readline is
4865 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4866 any GNU mirror site.
4868 See also ADD-HISTORY function.
4870 ** New function: add-history STRING
4871 Add STRING as the most recent line in the history used by the READLINE
4872 command. READLINE does not add lines to the history itself; you must
4873 call ADD-HISTORY to make previous input available to the user.
4875 ** The behavior of the read-line function has changed.
4877 This function now uses standard C library functions to read the line,
4878 for speed. This means that it doesn not respect the value of
4879 scm-line-incrementors; it assumes that lines are delimited with
4882 (Note that this is read-line, the function that reads a line of text
4883 from a port, not readline, the function that reads a line from a
4884 terminal, providing full editing capabilities.)
4886 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4888 This module provides some simple argument parsing. It exports one
4891 Function: getopt-gnu-style ARG-LS
4892 Parse a list of program arguments into an alist of option
4895 Each item in the list of program arguments is examined to see if
4896 it meets the syntax of a GNU long-named option. An argument like
4897 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4898 returned alist, where MUMBLE is a keyword object with the same
4899 name as the argument. An argument like `--MUMBLE=FROB' produces
4900 an element of the form (MUMBLE . FROB), where FROB is a string.
4902 As a special case, the returned alist also contains a pair whose
4903 car is the symbol `rest'. The cdr of this pair is a list
4904 containing all the items in the argument list that are not options
4905 of the form mentioned above.
4907 The argument `--' is treated specially: all items in the argument
4908 list appearing after such an argument are not examined, and are
4909 returned in the special `rest' list.
4911 This function does not parse normal single-character switches.
4912 You will need to parse them out of the `rest' list yourself.
4914 ** The read syntax for byte vectors and short vectors has changed.
4916 Instead of #bytes(...), write #y(...).
4918 Instead of #short(...), write #h(...).
4920 This may seem nutty, but, like the other uniform vectors, byte vectors
4921 and short vectors want to have the same print and read syntax (and,
4922 more basic, want to have read syntax!). Changing the read syntax to
4923 use multiple characters after the hash sign breaks with the
4924 conventions used in R5RS and the conventions used for the other
4925 uniform vectors. It also introduces complexity in the current reader,
4926 both on the C and Scheme levels. (The Right solution is probably to
4927 change the syntax and prototypes for uniform vectors entirely.)
4930 ** The new module (ice-9 session) provides useful interactive functions.
4932 *** New procedure: (apropos REGEXP OPTION ...)
4934 Display a list of top-level variables whose names match REGEXP, and
4935 the modules they are imported from. Each OPTION should be one of the
4938 value --- Show the value of each matching variable.
4939 shadow --- Show bindings shadowed by subsequently imported modules.
4940 full --- Same as both `shadow' and `value'.
4944 guile> (apropos "trace" 'full)
4945 debug: trace #<procedure trace args>
4946 debug: untrace #<procedure untrace args>
4947 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4948 the-scm-module: before-backtrace-hook ()
4949 the-scm-module: backtrace #<primitive-procedure backtrace>
4950 the-scm-module: after-backtrace-hook ()
4951 the-scm-module: has-shown-backtrace-hint? #f
4954 ** There are new functions and syntax for working with macros.
4956 Guile implements macros as a special object type. Any variable whose
4957 top-level binding is a macro object acts as a macro. The macro object
4958 specifies how the expression should be transformed before evaluation.
4960 *** Macro objects now print in a reasonable way, resembling procedures.
4962 *** New function: (macro? OBJ)
4963 True iff OBJ is a macro object.
4965 *** New function: (primitive-macro? OBJ)
4966 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4967 macro transformers, implemented in eval.c rather than Scheme code.
4969 Why do we have this function?
4970 - For symmetry with procedure? and primitive-procedure?,
4971 - to allow custom print procedures to tell whether a macro is
4972 primitive, and display it differently, and
4973 - to allow compilers and user-written evaluators to distinguish
4974 builtin special forms from user-defined ones, which could be
4977 *** New function: (macro-type OBJ)
4978 Return a value indicating what kind of macro OBJ is. Possible return
4981 The symbol `syntax' --- a macro created by procedure->syntax.
4982 The symbol `macro' --- a macro created by procedure->macro.
4983 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4984 The boolean #f --- if OBJ is not a macro object.
4986 *** New function: (macro-name MACRO)
4987 Return the name of the macro object MACRO's procedure, as returned by
4990 *** New function: (macro-transformer MACRO)
4991 Return the transformer procedure for MACRO.
4993 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4995 Specify a new macro expander to use in the current module. Each
4996 MODULE is a module name, with the same meaning as in the `use-modules'
4997 form; each named module's exported bindings are added to the current
4998 top-level environment. TRANSFORMER is an expression evaluated in the
4999 resulting environment which must yield a procedure to use as the
5000 module's eval transformer: every expression evaluated in this module
5001 is passed to this function, and the result passed to the Guile
5004 *** macro-eval! is removed. Use local-eval instead.
5006 ** Some magic has been added to the printer to better handle user
5007 written printing routines (like record printers, closure printers).
5009 The problem is that these user written routines must have access to
5010 the current `print-state' to be able to handle fancy things like
5011 detection of circular references. These print-states have to be
5012 passed to the builtin printing routines (display, write, etc) to
5013 properly continue the print chain.
5015 We didn't want to change all existing print code so that it
5016 explicitly passes thru a print state in addition to a port. Instead,
5017 we extented the possible values that the builtin printing routines
5018 accept as a `port'. In addition to a normal port, they now also take
5019 a pair of a normal port and a print-state. Printing will go to the
5020 port and the print-state will be used to control the detection of
5021 circular references, etc. If the builtin function does not care for a
5022 print-state, it is simply ignored.
5024 User written callbacks are now called with such a pair as their
5025 `port', but because every function now accepts this pair as a PORT
5026 argument, you don't have to worry about that. In fact, it is probably
5027 safest to not check for these pairs.
5029 However, it is sometimes necessary to continue a print chain on a
5030 different port, for example to get a intermediate string
5031 representation of the printed value, mangle that string somehow, and
5032 then to finally print the mangled string. Use the new function
5034 inherit-print-state OLD-PORT NEW-PORT
5036 for this. It constructs a new `port' that prints to NEW-PORT but
5037 inherits the print-state of OLD-PORT.
5039 ** struct-vtable-offset renamed to vtable-offset-user
5041 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5043 ** There is now a third optional argument to make-vtable-vtable
5044 (and fourth to make-struct) when constructing new types (vtables).
5045 This argument initializes field vtable-index-printer of the vtable.
5047 ** The detection of circular references has been extended to structs.
5048 That is, a structure that -- in the process of being printed -- prints
5049 itself does not lead to infinite recursion.
5051 ** There is now some basic support for fluids. Please read
5052 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5053 the following functions and macros:
5055 Function: make-fluid
5057 Create a new fluid object. Fluids are not special variables or
5058 some other extension to the semantics of Scheme, but rather
5059 ordinary Scheme objects. You can store them into variables (that
5060 are still lexically scoped, of course) or into any other place you
5061 like. Every fluid has a initial value of `#f'.
5063 Function: fluid? OBJ
5065 Test whether OBJ is a fluid.
5067 Function: fluid-ref FLUID
5068 Function: fluid-set! FLUID VAL
5070 Access/modify the fluid FLUID. Modifications are only visible
5071 within the current dynamic root (that includes threads).
5073 Function: with-fluids* FLUIDS VALUES THUNK
5075 FLUIDS is a list of fluids and VALUES a corresponding list of
5076 values for these fluids. Before THUNK gets called the values are
5077 installed in the fluids and the old values of the fluids are
5078 saved in the VALUES list. When the flow of control leaves THUNK
5079 or reenters it, the values get swapped again. You might think of
5080 this as a `safe-fluid-excursion'. Note that the VALUES list is
5081 modified by `with-fluids*'.
5083 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5085 The same as `with-fluids*' but with a different syntax. It looks
5086 just like `let', but both FLUID and VALUE are evaluated. Remember,
5087 fluids are not special variables but ordinary objects. FLUID
5088 should evaluate to a fluid.
5090 ** Changes to system call interfaces:
5092 *** close-port, close-input-port and close-output-port now return a
5093 boolean instead of an `unspecified' object. #t means that the port
5094 was successfully closed, while #f means it was already closed. It is
5095 also now possible for these procedures to raise an exception if an
5096 error occurs (some errors from write can be delayed until close.)
5098 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5101 *** the third argument to fcntl is now optional.
5103 *** the first argument to chown can now be a file descriptor or a port.
5105 *** the argument to stat can now be a port.
5107 *** The following new procedures have been added (most use scsh
5110 *** procedure: close PORT/FD
5111 Similar to close-port (*note close-port: Closing Ports.), but also
5112 works on file descriptors. A side effect of closing a file
5113 descriptor is that any ports using that file descriptor are moved
5114 to a different file descriptor and have their revealed counts set
5117 *** procedure: port->fdes PORT
5118 Returns the integer file descriptor underlying PORT. As a side
5119 effect the revealed count of PORT is incremented.
5121 *** procedure: fdes->ports FDES
5122 Returns a list of existing ports which have FDES as an underlying
5123 file descriptor, without changing their revealed counts.
5125 *** procedure: fdes->inport FDES
5126 Returns an existing input port which has FDES as its underlying
5127 file descriptor, if one exists, and increments its revealed count.
5128 Otherwise, returns a new input port with a revealed count of 1.
5130 *** procedure: fdes->outport FDES
5131 Returns an existing output port which has FDES as its underlying
5132 file descriptor, if one exists, and increments its revealed count.
5133 Otherwise, returns a new output port with a revealed count of 1.
5135 The next group of procedures perform a `dup2' system call, if NEWFD
5136 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5137 duplicated can be supplied as an integer or contained in a port. The
5138 type of value returned varies depending on which procedure is used.
5140 All procedures also have the side effect when performing `dup2' that
5141 any ports using NEWFD are moved to a different file descriptor and have
5142 their revealed counts set to zero.
5144 *** procedure: dup->fdes PORT/FD [NEWFD]
5145 Returns an integer file descriptor.
5147 *** procedure: dup->inport PORT/FD [NEWFD]
5148 Returns a new input port using the new file descriptor.
5150 *** procedure: dup->outport PORT/FD [NEWFD]
5151 Returns a new output port using the new file descriptor.
5153 *** procedure: dup PORT/FD [NEWFD]
5154 Returns a new port if PORT/FD is a port, with the same mode as the
5155 supplied port, otherwise returns an integer file descriptor.
5157 *** procedure: dup->port PORT/FD MODE [NEWFD]
5158 Returns a new port using the new file descriptor. MODE supplies a
5159 mode string for the port (*note open-file: File Ports.).
5161 *** procedure: setenv NAME VALUE
5162 Modifies the environment of the current process, which is also the
5163 default environment inherited by child processes.
5165 If VALUE is `#f', then NAME is removed from the environment.
5166 Otherwise, the string NAME=VALUE is added to the environment,
5167 replacing any existing string with name matching NAME.
5169 The return value is unspecified.
5171 *** procedure: truncate-file OBJ SIZE
5172 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5173 can be a string containing a file name or an integer file
5174 descriptor or port open for output on the file. The underlying
5175 system calls are `truncate' and `ftruncate'.
5177 The return value is unspecified.
5179 *** procedure: setvbuf PORT MODE [SIZE]
5180 Set the buffering mode for PORT. MODE can be:
5188 block buffered, using a newly allocated buffer of SIZE bytes.
5189 However if SIZE is zero or unspecified, the port will be made
5192 This procedure should not be used after I/O has been performed with
5195 Ports are usually block buffered by default, with a default buffer
5196 size. Procedures e.g., *Note open-file: File Ports, which accept a
5197 mode string allow `0' to be added to request an unbuffered port.
5199 *** procedure: fsync PORT/FD
5200 Copies any unwritten data for the specified output file descriptor
5201 to disk. If PORT/FD is a port, its buffer is flushed before the
5202 underlying file descriptor is fsync'd. The return value is
5205 *** procedure: open-fdes PATH FLAGS [MODES]
5206 Similar to `open' but returns a file descriptor instead of a port.
5208 *** procedure: execle PATH ENV [ARG] ...
5209 Similar to `execl', but the environment of the new process is
5210 specified by ENV, which must be a list of strings as returned by
5211 the `environ' procedure.
5213 This procedure is currently implemented using the `execve' system
5214 call, but we call it `execle' because of its Scheme calling
5217 *** procedure: strerror ERRNO
5218 Returns the Unix error message corresponding to ERRNO, an integer.
5220 *** procedure: primitive-exit [STATUS]
5221 Terminate the current process without unwinding the Scheme stack.
5222 This is would typically be useful after a fork. The exit status
5223 is STATUS if supplied, otherwise zero.
5225 *** procedure: times
5226 Returns an object with information about real and processor time.
5227 The following procedures accept such an object as an argument and
5228 return a selected component:
5231 The current real time, expressed as time units relative to an
5235 The CPU time units used by the calling process.
5238 The CPU time units used by the system on behalf of the
5242 The CPU time units used by terminated child processes of the
5243 calling process, whose status has been collected (e.g., using
5247 Similarly, the CPU times units used by the system on behalf of
5248 terminated child processes.
5250 ** Removed: list-length
5251 ** Removed: list-append, list-append!
5252 ** Removed: list-reverse, list-reverse!
5254 ** array-map renamed to array-map!
5256 ** serial-array-map renamed to serial-array-map!
5258 ** catch doesn't take #f as first argument any longer
5260 Previously, it was possible to pass #f instead of a key to `catch'.
5261 That would cause `catch' to pass a jump buffer object to the procedure
5262 passed as second argument. The procedure could then use this jump
5263 buffer objekt as an argument to throw.
5265 This mechanism has been removed since its utility doesn't motivate the
5266 extra complexity it introduces.
5268 ** The `#/' notation for lists now provokes a warning message from Guile.
5269 This syntax will be removed from Guile in the near future.
5271 To disable the warning message, set the GUILE_HUSH environment
5272 variable to any non-empty value.
5274 ** The newline character now prints as `#\newline', following the
5275 normal Scheme notation, not `#\nl'.
5277 * Changes to the gh_ interface
5279 ** The gh_enter function now takes care of loading the Guile startup files.
5280 gh_enter works by calling scm_boot_guile; see the remarks below.
5282 ** Function: void gh_write (SCM x)
5284 Write the printed representation of the scheme object x to the current
5285 output port. Corresponds to the scheme level `write'.
5287 ** gh_list_length renamed to gh_length.
5289 ** vector handling routines
5291 Several major changes. In particular, gh_vector() now resembles
5292 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5293 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5294 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5295 vector-related gh_ functions have been implemented.
5297 ** pair and list routines
5299 Implemented several of the R4RS pair and list functions that were
5302 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5304 New function. Converts double arrays back and forth between Scheme
5307 * Changes to the scm_ interface
5309 ** The function scm_boot_guile now takes care of loading the startup files.
5311 Guile's primary initialization function, scm_boot_guile, now takes
5312 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5313 Guile, define the module system, and put together some standard
5314 bindings. It also loads `init.scm', which is intended to hold
5315 site-specific initialization code.
5317 Since Guile cannot operate properly until boot-9.scm is loaded, there
5318 is no reason to separate loading boot-9.scm from Guile's other
5319 initialization processes.
5321 This job used to be done by scm_compile_shell_switches, which didn't
5322 make much sense; in particular, it meant that people using Guile for
5323 non-shell-like applications had to jump through hoops to get Guile
5324 initialized properly.
5326 ** The function scm_compile_shell_switches no longer loads the startup files.
5327 Now, Guile always loads the startup files, whenever it is initialized;
5328 see the notes above for scm_boot_guile and scm_load_startup_files.
5330 ** Function: scm_load_startup_files
5331 This new function takes care of loading Guile's initialization file
5332 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5333 this is always called by the Guile initialization process, it's
5334 probably not too useful to call this yourself, but it's there anyway.
5336 ** The semantics of smob marking have changed slightly.
5338 The smob marking function (the `mark' member of the scm_smobfuns
5339 structure) is no longer responsible for setting the mark bit on the
5340 smob. The generic smob handling code in the garbage collector will
5341 set this bit. The mark function need only ensure that any other
5342 objects the smob refers to get marked.
5344 Note that this change means that the smob's GC8MARK bit is typically
5345 already set upon entry to the mark function. Thus, marking functions
5346 which look like this:
5349 if (SCM_GC8MARKP (ptr))
5351 SCM_SETGC8MARK (ptr);
5352 ... mark objects to which the smob refers ...
5355 are now incorrect, since they will return early, and fail to mark any
5356 other objects the smob refers to. Some code in the Guile library used
5359 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5361 If you have implemented your own I/O port type, by writing the
5362 functions required by the scm_ptobfuns and then calling scm_newptob,
5363 you will need to change your functions slightly.
5365 The functions in a scm_ptobfuns structure now expect the port itself
5366 as their argument; they used to expect the `stream' member of the
5367 port's scm_port_table structure. This allows functions in an
5368 scm_ptobfuns structure to easily access the port's cell (and any flags
5369 it its CAR), and the port's scm_port_table structure.
5371 Guile now passes the I/O port itself as the `port' argument in the
5372 following scm_ptobfuns functions:
5374 int (*free) (SCM port);
5375 int (*fputc) (int, SCM port);
5376 int (*fputs) (char *, SCM port);
5377 scm_sizet (*fwrite) SCM_P ((char *ptr,
5381 int (*fflush) (SCM port);
5382 int (*fgetc) (SCM port);
5383 int (*fclose) (SCM port);
5385 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5388 If you have existing code which defines its own port types, it is easy
5389 to convert your code to the new interface; simply apply SCM_STREAM to
5390 the port argument to yield the value you code used to expect.
5392 Note that since both the port and the stream have the same type in the
5393 C code --- they are both SCM values --- the C compiler will not remind
5394 you if you forget to update your scm_ptobfuns functions.
5397 ** Function: int scm_internal_select (int fds,
5401 struct timeval *timeout);
5403 This is a replacement for the `select' function provided by the OS.
5404 It enables I/O blocking and sleeping to happen for one cooperative
5405 thread without blocking other threads. It also avoids busy-loops in
5406 these situations. It is intended that all I/O blocking and sleeping
5407 will finally go through this function. Currently, this function is
5408 only available on systems providing `gettimeofday' and `select'.
5410 ** Function: SCM scm_internal_stack_catch (SCM tag,
5411 scm_catch_body_t body,
5413 scm_catch_handler_t handler,
5416 A new sibling to the other two C level `catch' functions
5417 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5418 the stack to be saved automatically into the variable `the-last-stack'
5419 (scm_the_last_stack_var) on error. This is necessary if you want to
5420 use advanced error reporting, such as calling scm_display_error and
5421 scm_display_backtrace. (They both take a stack object as argument.)
5423 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5425 scm_catch_handler_t handler,
5428 Spawns a new thread. It does a job similar to
5429 scm_call_with_new_thread but takes arguments more suitable when
5430 spawning threads from application C code.
5432 ** The hook scm_error_callback has been removed. It was originally
5433 intended as a way for the user to install his own error handler. But
5434 that method works badly since it intervenes between throw and catch,
5435 thereby changing the semantics of expressions like (catch #t ...).
5436 The correct way to do it is to use one of the C level catch functions
5437 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5439 ** Removed functions:
5441 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5442 scm_list_reverse, scm_list_reverse_x
5444 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5446 These can be used for pretty list creation from C. The idea is taken
5447 from Erick Gallesio's STk.
5449 ** scm_array_map renamed to scm_array_map_x
5451 ** mbstrings are now removed
5453 This means that the type codes scm_tc7_mb_string and
5454 scm_tc7_mb_substring has been removed.
5456 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5458 Since we no longer support multi-byte strings, these I/O functions
5459 have been simplified, and renamed. Here are their old names, and
5460 their new names and arguments:
5462 scm_gen_putc -> void scm_putc (int c, SCM port);
5463 scm_gen_puts -> void scm_puts (char *s, SCM port);
5464 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5465 scm_gen_getc -> void scm_getc (SCM port);
5468 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5470 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5472 SCM_TYP7S now masks away the bit which distinguishes substrings from
5475 ** scm_catch_body_t: Backward incompatible change!
5477 Body functions to scm_internal_catch and friends do not any longer
5478 take a second argument. This is because it is no longer possible to
5479 pass a #f arg to catch.
5481 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5483 The function scm_protect_object protects its argument from being freed
5484 by the garbage collector. scm_unprotect_object removes that
5487 These functions now nest properly. That is, for every object O, there
5488 is a counter which scm_protect_object(O) increments and
5489 scm_unprotect_object(O) decrements, if the counter is greater than
5490 zero. Every object's counter is zero when it is first created. If an
5491 object's counter is greater than zero, the garbage collector will not
5492 reclaim its storage.
5494 This allows you to use scm_protect_object in your code without
5495 worrying that some other function you call will call
5496 scm_unprotect_object, and allow it to be freed. Assuming that the
5497 functions you call are well-behaved, and unprotect only those objects
5498 they protect, you can follow the same rule and have confidence that
5499 objects will be freed only at appropriate times.
5502 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5504 * Changes to the distribution
5506 ** Nightly snapshots are now available from ftp.red-bean.com.
5507 The old server, ftp.cyclic.com, has been relinquished to its rightful
5510 Nightly snapshots of the Guile development sources are now available via
5511 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5513 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5514 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5516 ** To run Guile without installing it, the procedure has changed a bit.
5518 If you used a separate build directory to compile Guile, you'll need
5519 to include the build directory in SCHEME_LOAD_PATH, as well as the
5520 source directory. See the `INSTALL' file for examples.
5522 * Changes to the procedure for linking libguile with your programs
5524 ** The standard Guile load path for Scheme code now includes
5525 $(datadir)/guile (usually /usr/local/share/guile). This means that
5526 you can install your own Scheme files there, and Guile will find them.
5527 (Previous versions of Guile only checked a directory whose name
5528 contained the Guile version number, so you had to re-install or move
5529 your Scheme sources each time you installed a fresh version of Guile.)
5531 The load path also includes $(datadir)/guile/site; we recommend
5532 putting individual Scheme files there. If you want to install a
5533 package with multiple source files, create a directory for them under
5536 ** Guile 1.2 will now use the Rx regular expression library, if it is
5537 installed on your system. When you are linking libguile into your own
5538 programs, this means you will have to link against -lguile, -lqt (if
5539 you configured Guile with thread support), and -lrx.
5541 If you are using autoconf to generate configuration scripts for your
5542 application, the following lines should suffice to add the appropriate
5543 libraries to your link command:
5545 ### Find Rx, quickthreads and libguile.
5546 AC_CHECK_LIB(rx, main)
5547 AC_CHECK_LIB(qt, main)
5548 AC_CHECK_LIB(guile, scm_shell)
5550 The Guile 1.2 distribution does not contain sources for the Rx
5551 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5552 retrieve it from a GNU FTP site and install it separately.
5554 * Changes to Scheme functions and syntax
5556 ** The dynamic linking features of Guile are now enabled by default.
5557 You can disable them by giving the `--disable-dynamic-linking' option
5560 (dynamic-link FILENAME)
5562 Find the object file denoted by FILENAME (a string) and link it
5563 into the running Guile application. When everything works out,
5564 return a Scheme object suitable for representing the linked object
5565 file. Otherwise an error is thrown. How object files are
5566 searched is system dependent.
5568 (dynamic-object? VAL)
5570 Determine whether VAL represents a dynamically linked object file.
5572 (dynamic-unlink DYNOBJ)
5574 Unlink the indicated object file from the application. DYNOBJ
5575 should be one of the values returned by `dynamic-link'.
5577 (dynamic-func FUNCTION DYNOBJ)
5579 Search the C function indicated by FUNCTION (a string or symbol)
5580 in DYNOBJ and return some Scheme object that can later be used
5581 with `dynamic-call' to actually call this function. Right now,
5582 these Scheme objects are formed by casting the address of the
5583 function to `long' and converting this number to its Scheme
5586 (dynamic-call FUNCTION DYNOBJ)
5588 Call the C function indicated by FUNCTION and DYNOBJ. The
5589 function is passed no arguments and its return value is ignored.
5590 When FUNCTION is something returned by `dynamic-func', call that
5591 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5592 etc.), look it up in DYNOBJ; this is equivalent to
5594 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5596 Interrupts are deferred while the C function is executing (with
5597 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5599 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5601 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5602 some arguments and return its return value. The C function is
5603 expected to take two arguments and return an `int', just like
5606 int c_func (int argc, char **argv);
5608 ARGS must be a list of strings and is converted into an array of
5609 `char *'. The array is passed in ARGV and its size in ARGC. The
5610 return value is converted to a Scheme number and returned from the
5611 call to `dynamic-args-call'.
5613 When dynamic linking is disabled or not supported on your system,
5614 the above functions throw errors, but they are still available.
5616 Here is a small example that works on GNU/Linux:
5618 (define libc-obj (dynamic-link "libc.so"))
5619 (dynamic-args-call 'rand libc-obj '())
5621 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5623 ** The #/ syntax for module names is depreciated, and will be removed
5624 in a future version of Guile. Instead of
5632 The latter syntax is more consistent with existing Lisp practice.
5634 ** Guile now does fancier printing of structures. Structures are the
5635 underlying implementation for records, which in turn are used to
5636 implement modules, so all of these object now print differently and in
5637 a more informative way.
5639 The Scheme printer will examine the builtin variable *struct-printer*
5640 whenever it needs to print a structure object. When this variable is
5641 not `#f' it is deemed to be a procedure and will be applied to the
5642 structure object and the output port. When *struct-printer* is `#f'
5643 or the procedure return `#f' the structure object will be printed in
5644 the boring #<struct 80458270> form.
5646 This hook is used by some routines in ice-9/boot-9.scm to implement
5647 type specific printing routines. Please read the comments there about
5650 One of the more specific uses of structs are records. The printing
5651 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5652 called. It should behave like a *struct-printer* procedure (described
5655 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5656 token of the form #:NAME, where NAME has the same syntax as a Scheme
5657 symbol, is the external representation of the keyword named NAME.
5658 Keyword objects print using this syntax as well, so values containing
5659 keyword objects can be read back into Guile. When used in an
5660 expression, keywords are self-quoting objects.
5662 Guile suports this read syntax, and uses this print syntax, regardless
5663 of the current setting of the `keyword' read option. The `keyword'
5664 read option only controls whether Guile recognizes the `:NAME' syntax,
5665 which is incompatible with R4RS. (R4RS says such token represent
5668 ** Guile has regular expression support again. Guile 1.0 included
5669 functions for matching regular expressions, based on the Rx library.
5670 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5671 distribution, and thus Guile had no regular expression support. Guile
5672 1.2 again supports the most commonly used functions, and supports all
5673 of SCSH's regular expression functions.
5675 If your system does not include a POSIX regular expression library,
5676 and you have not linked Guile with a third-party regexp library such as
5677 Rx, these functions will not be available. You can tell whether your
5678 Guile installation includes regular expression support by checking
5679 whether the `*features*' list includes the `regex' symbol.
5681 *** regexp functions
5683 By default, Guile supports POSIX extended regular expressions. That
5684 means that the characters `(', `)', `+' and `?' are special, and must
5685 be escaped if you wish to match the literal characters.
5687 This regular expression interface was modeled after that implemented
5688 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5689 with SCSH regular expressions.
5691 **** Function: string-match PATTERN STR [START]
5692 Compile the string PATTERN into a regular expression and compare
5693 it with STR. The optional numeric argument START specifies the
5694 position of STR at which to begin matching.
5696 `string-match' returns a "match structure" which describes what,
5697 if anything, was matched by the regular expression. *Note Match
5698 Structures::. If STR does not match PATTERN at all,
5699 `string-match' returns `#f'.
5701 Each time `string-match' is called, it must compile its PATTERN
5702 argument into a regular expression structure. This operation is
5703 expensive, which makes `string-match' inefficient if the same regular
5704 expression is used several times (for example, in a loop). For better
5705 performance, you can compile a regular expression in advance and then
5706 match strings against the compiled regexp.
5708 **** Function: make-regexp STR [FLAGS]
5709 Compile the regular expression described by STR, and return the
5710 compiled regexp structure. If STR does not describe a legal
5711 regular expression, `make-regexp' throws a
5712 `regular-expression-syntax' error.
5714 FLAGS may be the bitwise-or of one or more of the following:
5716 **** Constant: regexp/extended
5717 Use POSIX Extended Regular Expression syntax when interpreting
5718 STR. If not set, POSIX Basic Regular Expression syntax is used.
5719 If the FLAGS argument is omitted, we assume regexp/extended.
5721 **** Constant: regexp/icase
5722 Do not differentiate case. Subsequent searches using the
5723 returned regular expression will be case insensitive.
5725 **** Constant: regexp/newline
5726 Match-any-character operators don't match a newline.
5728 A non-matching list ([^...]) not containing a newline matches a
5731 Match-beginning-of-line operator (^) matches the empty string
5732 immediately after a newline, regardless of whether the FLAGS
5733 passed to regexp-exec contain regexp/notbol.
5735 Match-end-of-line operator ($) matches the empty string
5736 immediately before a newline, regardless of whether the FLAGS
5737 passed to regexp-exec contain regexp/noteol.
5739 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5740 Match the compiled regular expression REGEXP against `str'. If
5741 the optional integer START argument is provided, begin matching
5742 from that position in the string. Return a match structure
5743 describing the results of the match, or `#f' if no match could be
5746 FLAGS may be the bitwise-or of one or more of the following:
5748 **** Constant: regexp/notbol
5749 The match-beginning-of-line operator always fails to match (but
5750 see the compilation flag regexp/newline above) This flag may be
5751 used when different portions of a string are passed to
5752 regexp-exec and the beginning of the string should not be
5753 interpreted as the beginning of the line.
5755 **** Constant: regexp/noteol
5756 The match-end-of-line operator always fails to match (but see the
5757 compilation flag regexp/newline above)
5759 **** Function: regexp? OBJ
5760 Return `#t' if OBJ is a compiled regular expression, or `#f'
5763 Regular expressions are commonly used to find patterns in one string
5764 and replace them with the contents of another string.
5766 **** Function: regexp-substitute PORT MATCH [ITEM...]
5767 Write to the output port PORT selected contents of the match
5768 structure MATCH. Each ITEM specifies what should be written, and
5769 may be one of the following arguments:
5771 * A string. String arguments are written out verbatim.
5773 * An integer. The submatch with that number is written.
5775 * The symbol `pre'. The portion of the matched string preceding
5776 the regexp match is written.
5778 * The symbol `post'. The portion of the matched string
5779 following the regexp match is written.
5781 PORT may be `#f', in which case nothing is written; instead,
5782 `regexp-substitute' constructs a string from the specified ITEMs
5785 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5786 Similar to `regexp-substitute', but can be used to perform global
5787 substitutions on STR. Instead of taking a match structure as an
5788 argument, `regexp-substitute/global' takes two string arguments: a
5789 REGEXP string describing a regular expression, and a TARGET string
5790 which should be matched against this regular expression.
5792 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5795 * A function may be supplied. When this function is called, it
5796 will be passed one argument: a match structure for a given
5797 regular expression match. It should return a string to be
5798 written out to PORT.
5800 * The `post' symbol causes `regexp-substitute/global' to recurse
5801 on the unmatched portion of STR. This *must* be supplied in
5802 order to perform global search-and-replace on STR; if it is
5803 not present among the ITEMs, then `regexp-substitute/global'
5804 will return after processing a single match.
5806 *** Match Structures
5808 A "match structure" is the object returned by `string-match' and
5809 `regexp-exec'. It describes which portion of a string, if any, matched
5810 the given regular expression. Match structures include: a reference to
5811 the string that was checked for matches; the starting and ending
5812 positions of the regexp match; and, if the regexp included any
5813 parenthesized subexpressions, the starting and ending positions of each
5816 In each of the regexp match functions described below, the `match'
5817 argument must be a match structure returned by a previous call to
5818 `string-match' or `regexp-exec'. Most of these functions return some
5819 information about the original target string that was matched against a
5820 regular expression; we will call that string TARGET for easy reference.
5822 **** Function: regexp-match? OBJ
5823 Return `#t' if OBJ is a match structure returned by a previous
5824 call to `regexp-exec', or `#f' otherwise.
5826 **** Function: match:substring MATCH [N]
5827 Return the portion of TARGET matched by subexpression number N.
5828 Submatch 0 (the default) represents the entire regexp match. If
5829 the regular expression as a whole matched, but the subexpression
5830 number N did not match, return `#f'.
5832 **** Function: match:start MATCH [N]
5833 Return the starting position of submatch number N.
5835 **** Function: match:end MATCH [N]
5836 Return the ending position of submatch number N.
5838 **** Function: match:prefix MATCH
5839 Return the unmatched portion of TARGET preceding the regexp match.
5841 **** Function: match:suffix MATCH
5842 Return the unmatched portion of TARGET following the regexp match.
5844 **** Function: match:count MATCH
5845 Return the number of parenthesized subexpressions from MATCH.
5846 Note that the entire regular expression match itself counts as a
5847 subexpression, and failed submatches are included in the count.
5849 **** Function: match:string MATCH
5850 Return the original TARGET string.
5852 *** Backslash Escapes
5854 Sometimes you will want a regexp to match characters like `*' or `$'
5855 exactly. For example, to check whether a particular string represents
5856 a menu entry from an Info node, it would be useful to match it against
5857 a regexp like `^* [^:]*::'. However, this won't work; because the
5858 asterisk is a metacharacter, it won't match the `*' at the beginning of
5859 the string. In this case, we want to make the first asterisk un-magic.
5861 You can do this by preceding the metacharacter with a backslash
5862 character `\'. (This is also called "quoting" the metacharacter, and
5863 is known as a "backslash escape".) When Guile sees a backslash in a
5864 regular expression, it considers the following glyph to be an ordinary
5865 character, no matter what special meaning it would ordinarily have.
5866 Therefore, we can make the above example work by changing the regexp to
5867 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5868 to match only a single asterisk in the target string.
5870 Since the backslash is itself a metacharacter, you may force a
5871 regexp to match a backslash in the target string by preceding the
5872 backslash with itself. For example, to find variable references in a
5873 TeX program, you might want to find occurrences of the string `\let\'
5874 followed by any number of alphabetic characters. The regular expression
5875 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5876 each match a single backslash in the target string.
5878 **** Function: regexp-quote STR
5879 Quote each special character found in STR with a backslash, and
5880 return the resulting string.
5882 *Very important:* Using backslash escapes in Guile source code (as
5883 in Emacs Lisp or C) can be tricky, because the backslash character has
5884 special meaning for the Guile reader. For example, if Guile encounters
5885 the character sequence `\n' in the middle of a string while processing
5886 Scheme code, it replaces those characters with a newline character.
5887 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5888 Several of these "escape sequences" are processed by the Guile reader
5889 before your code is executed. Unrecognized escape sequences are
5890 ignored: if the characters `\*' appear in a string, they will be
5891 translated to the single character `*'.
5893 This translation is obviously undesirable for regular expressions,
5894 since we want to be able to include backslashes in a string in order to
5895 escape regexp metacharacters. Therefore, to make sure that a backslash
5896 is preserved in a string in your Guile program, you must use *two*
5897 consecutive backslashes:
5899 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5901 The string in this example is preprocessed by the Guile reader before
5902 any code is executed. The resulting argument to `make-regexp' is the
5903 string `^\* [^:]*', which is what we really want.
5905 This also means that in order to write a regular expression that
5906 matches a single backslash character, the regular expression string in
5907 the source code must include *four* backslashes. Each consecutive pair
5908 of backslashes gets translated by the Guile reader to a single
5909 backslash, and the resulting double-backslash is interpreted by the
5910 regexp engine as matching a single backslash character. Hence:
5912 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5914 The reason for the unwieldiness of this syntax is historical. Both
5915 regular expression pattern matchers and Unix string processing systems
5916 have traditionally used backslashes with the special meanings described
5917 above. The POSIX regular expression specification and ANSI C standard
5918 both require these semantics. Attempting to abandon either convention
5919 would cause other kinds of compatibility problems, possibly more severe
5920 ones. Therefore, without extending the Scheme reader to support
5921 strings with different quoting conventions (an ungainly and confusing
5922 extension when implemented in other languages), we must adhere to this
5923 cumbersome escape syntax.
5925 * Changes to the gh_ interface
5927 * Changes to the scm_ interface
5929 * Changes to system call interfaces:
5931 ** The value returned by `raise' is now unspecified. It throws an exception
5934 *** A new procedure `sigaction' can be used to install signal handlers
5936 (sigaction signum [action] [flags])
5938 signum is the signal number, which can be specified using the value
5941 If action is omitted, sigaction returns a pair: the CAR is the current
5942 signal hander, which will be either an integer with the value SIG_DFL
5943 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5944 handles the signal, or #f if a non-Scheme procedure handles the
5945 signal. The CDR contains the current sigaction flags for the handler.
5947 If action is provided, it is installed as the new handler for signum.
5948 action can be a Scheme procedure taking one argument, or the value of
5949 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5950 whatever signal handler was installed before sigaction was first used.
5951 Flags can optionally be specified for the new handler (SA_RESTART is
5952 always used if the system provides it, so need not be specified.) The
5953 return value is a pair with information about the old handler as
5956 This interface does not provide access to the "signal blocking"
5957 facility. Maybe this is not needed, since the thread support may
5958 provide solutions to the problem of consistent access to data
5961 *** A new procedure `flush-all-ports' is equivalent to running
5962 `force-output' on every port open for output.
5964 ** Guile now provides information on how it was built, via the new
5965 global variable, %guile-build-info. This variable records the values
5966 of the standard GNU makefile directory variables as an assocation
5967 list, mapping variable names (symbols) onto directory paths (strings).
5968 For example, to find out where the Guile link libraries were
5969 installed, you can say:
5971 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5974 * Changes to the scm_ interface
5976 ** The new function scm_handle_by_message_noexit is just like the
5977 existing scm_handle_by_message function, except that it doesn't call
5978 exit to terminate the process. Instead, it prints a message and just
5979 returns #f. This might be a more appropriate catch-all handler for
5980 new dynamic roots and threads.
5983 Changes in Guile 1.1 (released Friday, May 16 1997):
5985 * Changes to the distribution.
5987 The Guile 1.0 distribution has been split up into several smaller
5989 guile-core --- the Guile interpreter itself.
5990 guile-tcltk --- the interface between the Guile interpreter and
5991 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5992 is a toolkit for building graphical user interfaces.
5993 guile-rgx-ctax --- the interface between Guile and the Rx regular
5994 expression matcher, and the translator for the Ctax
5995 programming language. These are packaged together because the
5996 Ctax translator uses Rx to parse Ctax source code.
5998 This NEWS file describes the changes made to guile-core since the 1.0
6001 We no longer distribute the documentation, since it was either out of
6002 date, or incomplete. As soon as we have current documentation, we
6007 * Changes to the stand-alone interpreter
6009 ** guile now accepts command-line arguments compatible with SCSH, Olin
6010 Shivers' Scheme Shell.
6012 In general, arguments are evaluated from left to right, but there are
6013 exceptions. The following switches stop argument processing, and
6014 stash all remaining command-line arguments as the value returned by
6015 the (command-line) function.
6016 -s SCRIPT load Scheme source code from FILE, and exit
6017 -c EXPR evalute Scheme expression EXPR, and exit
6018 -- stop scanning arguments; run interactively
6020 The switches below are processed as they are encountered.
6021 -l FILE load Scheme source code from FILE
6022 -e FUNCTION after reading script, apply FUNCTION to
6023 command line arguments
6024 -ds do -s script at this point
6025 --emacs enable Emacs protocol (experimental)
6026 -h, --help display this help and exit
6027 -v, --version display version information and exit
6028 \ read arguments from following script lines
6030 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6031 which re-implements the traditional "echo" command:
6033 #!/usr/local/bin/guile -s
6036 (map (lambda (arg) (display arg) (display " "))
6040 (main (command-line))
6042 Suppose we invoke this script as follows:
6044 ekko a speckled gecko
6046 Through the magic of Unix script processing (triggered by the `#!'
6047 token at the top of the file), /usr/local/bin/guile receives the
6048 following list of command-line arguments:
6050 ("-s" "./ekko" "a" "speckled" "gecko")
6052 Unix inserts the name of the script after the argument specified on
6053 the first line of the file (in this case, "-s"), and then follows that
6054 with the arguments given to the script. Guile loads the script, which
6055 defines the `main' function, and then applies it to the list of
6056 remaining command-line arguments, ("a" "speckled" "gecko").
6058 In Unix, the first line of a script file must take the following form:
6060 #!INTERPRETER ARGUMENT
6062 where INTERPRETER is the absolute filename of the interpreter
6063 executable, and ARGUMENT is a single command-line argument to pass to
6066 You may only pass one argument to the interpreter, and its length is
6067 limited. These restrictions can be annoying to work around, so Guile
6068 provides a general mechanism (borrowed from, and compatible with,
6069 SCSH) for circumventing them.
6071 If the ARGUMENT in a Guile script is a single backslash character,
6072 `\', Guile will open the script file, parse arguments from its second
6073 and subsequent lines, and replace the `\' with them. So, for example,
6074 here is another implementation of the `ekko' script:
6076 #!/usr/local/bin/guile \
6080 (for-each (lambda (arg) (display arg) (display " "))
6084 If the user invokes this script as follows:
6086 ekko a speckled gecko
6088 Unix expands this into
6090 /usr/local/bin/guile \ ekko a speckled gecko
6092 When Guile sees the `\' argument, it replaces it with the arguments
6093 read from the second line of the script, producing:
6095 /usr/local/bin/guile -e main -s ekko a speckled gecko
6097 This tells Guile to load the `ekko' script, and apply the function
6098 `main' to the argument list ("a" "speckled" "gecko").
6100 Here is how Guile parses the command-line arguments:
6101 - Each space character terminates an argument. This means that two
6102 spaces in a row introduce an empty-string argument.
6103 - The tab character is not permitted (unless you quote it with the
6104 backslash character, as described below), to avoid confusion.
6105 - The newline character terminates the sequence of arguments, and will
6106 also terminate a final non-empty argument. (However, a newline
6107 following a space will not introduce a final empty-string argument;
6108 it only terminates the argument list.)
6109 - The backslash character is the escape character. It escapes
6110 backslash, space, tab, and newline. The ANSI C escape sequences
6111 like \n and \t are also supported. These produce argument
6112 constituents; the two-character combination \n doesn't act like a
6113 terminating newline. The escape sequence \NNN for exactly three
6114 octal digits reads as the character whose ASCII code is NNN. As
6115 above, characters produced this way are argument constituents.
6116 Backslash followed by other characters is not allowed.
6118 * Changes to the procedure for linking libguile with your programs
6120 ** Guile now builds and installs a shared guile library, if your
6121 system support shared libraries. (It still builds a static library on
6122 all systems.) Guile automatically detects whether your system
6123 supports shared libraries. To prevent Guile from buildisg shared
6124 libraries, pass the `--disable-shared' flag to the configure script.
6126 Guile takes longer to compile when it builds shared libraries, because
6127 it must compile every file twice --- once to produce position-
6128 independent object code, and once to produce normal object code.
6130 ** The libthreads library has been merged into libguile.
6132 To link a program against Guile, you now need only link against
6133 -lguile and -lqt; -lthreads is no longer needed. If you are using
6134 autoconf to generate configuration scripts for your application, the
6135 following lines should suffice to add the appropriate libraries to
6138 ### Find quickthreads and libguile.
6139 AC_CHECK_LIB(qt, main)
6140 AC_CHECK_LIB(guile, scm_shell)
6142 * Changes to Scheme functions
6144 ** Guile Scheme's special syntax for keyword objects is now optional,
6145 and disabled by default.
6147 The syntax variation from R4RS made it difficult to port some
6148 interesting packages to Guile. The routines which accepted keyword
6149 arguments (mostly in the module system) have been modified to also
6150 accept symbols whose names begin with `:'.
6152 To change the keyword syntax, you must first import the (ice-9 debug)
6154 (use-modules (ice-9 debug))
6156 Then you can enable the keyword syntax as follows:
6157 (read-set! keywords 'prefix)
6159 To disable keyword syntax, do this:
6160 (read-set! keywords #f)
6162 ** Many more primitive functions accept shared substrings as
6163 arguments. In the past, these functions required normal, mutable
6164 strings as arguments, although they never made use of this
6167 ** The uniform array functions now operate on byte vectors. These
6168 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6169 `serial-array-map', `array-map', `array-for-each', and
6172 ** The new functions `trace' and `untrace' implement simple debugging
6173 support for Scheme functions.
6175 The `trace' function accepts any number of procedures as arguments,
6176 and tells the Guile interpreter to display each procedure's name and
6177 arguments each time the procedure is invoked. When invoked with no
6178 arguments, `trace' returns the list of procedures currently being
6181 The `untrace' function accepts any number of procedures as arguments,
6182 and tells the Guile interpreter not to trace them any more. When
6183 invoked with no arguments, `untrace' untraces all curretly traced
6186 The tracing in Guile has an advantage over most other systems: we
6187 don't create new procedure objects, but mark the procedure objects
6188 themselves. This means that anonymous and internal procedures can be
6191 ** The function `assert-repl-prompt' has been renamed to
6192 `set-repl-prompt!'. It takes one argument, PROMPT.
6193 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6194 - If PROMPT is a string, we use it as a prompt.
6195 - If PROMPT is a procedure accepting no arguments, we call it, and
6196 display the result as a prompt.
6197 - Otherwise, we display "> ".
6199 ** The new function `eval-string' reads Scheme expressions from a
6200 string and evaluates them, returning the value of the last expression
6201 in the string. If the string contains no expressions, it returns an
6204 ** The new function `thunk?' returns true iff its argument is a
6205 procedure of zero arguments.
6207 ** `defined?' is now a builtin function, instead of syntax. This
6208 means that its argument should be quoted. It returns #t iff its
6209 argument is bound in the current module.
6211 ** The new syntax `use-modules' allows you to add new modules to your
6212 environment without re-typing a complete `define-module' form. It
6213 accepts any number of module names as arguments, and imports their
6214 public bindings into the current module.
6216 ** The new function (module-defined? NAME MODULE) returns true iff
6217 NAME, a symbol, is defined in MODULE, a module object.
6219 ** The new function `builtin-bindings' creates and returns a hash
6220 table containing copies of all the root module's bindings.
6222 ** The new function `builtin-weak-bindings' does the same as
6223 `builtin-bindings', but creates a doubly-weak hash table.
6225 ** The `equal?' function now considers variable objects to be
6226 equivalent if they have the same name and the same value.
6228 ** The new function `command-line' returns the command-line arguments
6229 given to Guile, as a list of strings.
6231 When using guile as a script interpreter, `command-line' returns the
6232 script's arguments; those processed by the interpreter (like `-s' or
6233 `-c') are omitted. (In other words, you get the normal, expected
6234 behavior.) Any application that uses scm_shell to process its
6235 command-line arguments gets this behavior as well.
6237 ** The new function `load-user-init' looks for a file called `.guile'
6238 in the user's home directory, and loads it if it exists. This is
6239 mostly for use by the code generated by scm_compile_shell_switches,
6240 but we thought it might also be useful in other circumstances.
6242 ** The new function `log10' returns the base-10 logarithm of its
6245 ** Changes to I/O functions
6247 *** The functions `read', `primitive-load', `read-and-eval!', and
6248 `primitive-load-path' no longer take optional arguments controlling
6249 case insensitivity and a `#' parser.
6251 Case sensitivity is now controlled by a read option called
6252 `case-insensitive'. The user can add new `#' syntaxes with the
6253 `read-hash-extend' function (see below).
6255 *** The new function `read-hash-extend' allows the user to change the
6256 syntax of Guile Scheme in a somewhat controlled way.
6258 (read-hash-extend CHAR PROC)
6259 When parsing S-expressions, if we read a `#' character followed by
6260 the character CHAR, use PROC to parse an object from the stream.
6261 If PROC is #f, remove any parsing procedure registered for CHAR.
6263 The reader applies PROC to two arguments: CHAR and an input port.
6265 *** The new functions read-delimited and read-delimited! provide a
6266 general mechanism for doing delimited input on streams.
6268 (read-delimited DELIMS [PORT HANDLE-DELIM])
6269 Read until we encounter one of the characters in DELIMS (a string),
6270 or end-of-file. PORT is the input port to read from; it defaults to
6271 the current input port. The HANDLE-DELIM parameter determines how
6272 the terminating character is handled; it should be one of the
6275 'trim omit delimiter from result
6276 'peek leave delimiter character in input stream
6277 'concat append delimiter character to returned value
6278 'split return a pair: (RESULT . TERMINATOR)
6280 HANDLE-DELIM defaults to 'peek.
6282 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6283 A side-effecting variant of `read-delimited'.
6285 The data is written into the string BUF at the indices in the
6286 half-open interval [START, END); the default interval is the whole
6287 string: START = 0 and END = (string-length BUF). The values of
6288 START and END must specify a well-defined interval in BUF, i.e.
6289 0 <= START <= END <= (string-length BUF).
6291 It returns NBYTES, the number of bytes read. If the buffer filled
6292 up without a delimiter character being found, it returns #f. If the
6293 port is at EOF when the read starts, it returns the EOF object.
6295 If an integer is returned (i.e., the read is successfully terminated
6296 by reading a delimiter character), then the HANDLE-DELIM parameter
6297 determines how to handle the terminating character. It is described
6298 above, and defaults to 'peek.
6300 (The descriptions of these functions were borrowed from the SCSH
6301 manual, by Olin Shivers and Brian Carlstrom.)
6303 *** The `%read-delimited!' function is the primitive used to implement
6304 `read-delimited' and `read-delimited!'.
6306 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6308 This returns a pair of values: (TERMINATOR . NUM-READ).
6309 - TERMINATOR describes why the read was terminated. If it is a
6310 character or the eof object, then that is the value that terminated
6311 the read. If it is #f, the function filled the buffer without finding
6312 a delimiting character.
6313 - NUM-READ is the number of characters read into BUF.
6315 If the read is successfully terminated by reading a delimiter
6316 character, then the gobble? parameter determines what to do with the
6317 terminating character. If true, the character is removed from the
6318 input stream; if false, the character is left in the input stream
6319 where a subsequent read operation will retrieve it. In either case,
6320 the character is also the first value returned by the procedure call.
6322 (The descriptions of this function was borrowed from the SCSH manual,
6323 by Olin Shivers and Brian Carlstrom.)
6325 *** The `read-line' and `read-line!' functions have changed; they now
6326 trim the terminator by default; previously they appended it to the
6327 returned string. For the old behavior, use (read-line PORT 'concat).
6329 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6330 take new optional START and END arguments, specifying the region of
6331 the array to read and write.
6333 *** The `ungetc-char-ready?' function has been removed. We feel it's
6334 inappropriate for an interface to expose implementation details this
6337 ** Changes to the Unix library and system call interface
6339 *** The new fcntl function provides access to the Unix `fcntl' system
6342 (fcntl PORT COMMAND VALUE)
6343 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6344 Values for COMMAND are:
6346 F_DUPFD duplicate a file descriptor
6347 F_GETFD read the descriptor's close-on-exec flag
6348 F_SETFD set the descriptor's close-on-exec flag to VALUE
6349 F_GETFL read the descriptor's flags, as set on open
6350 F_SETFL set the descriptor's flags, as set on open to VALUE
6351 F_GETOWN return the process ID of a socket's owner, for SIGIO
6352 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6353 FD_CLOEXEC not sure what this is
6355 For details, see the documentation for the fcntl system call.
6357 *** The arguments to `select' have changed, for compatibility with
6358 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6359 expected behavior. The MILLISECONDS parameter has been changed to
6360 MICROSECONDS, to more closely resemble the underlying system call.
6361 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6362 corresponding return set will be the same.
6364 *** The arguments to the `mknod' system call have changed. They are
6367 (mknod PATH TYPE PERMS DEV)
6368 Create a new file (`node') in the file system. PATH is the name of
6369 the file to create. TYPE is the kind of file to create; it should
6370 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6371 permission bits to give the newly created file. If TYPE is
6372 'block-special or 'char-special, DEV specifies which device the
6373 special file refers to; its interpretation depends on the kind of
6374 special file being created.
6376 *** The `fork' function has been renamed to `primitive-fork', to avoid
6377 clashing with various SCSH forks.
6379 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6380 and `recvfrom!'. They no longer accept a size for a second argument;
6381 you must pass a string to hold the received value. They no longer
6382 return the buffer. Instead, `recv' returns the length of the message
6383 received, and `recvfrom' returns a pair containing the packet's length
6384 and originating address.
6386 *** The file descriptor datatype has been removed, as have the
6387 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6388 We plan to replace these functions with a SCSH-compatible interface.
6390 *** The `create' function has been removed; it's just a special case
6393 *** There are new functions to break down process termination status
6394 values. In the descriptions below, STATUS is a value returned by
6397 (status:exit-val STATUS)
6398 If the child process exited normally, this function returns the exit
6399 code for the child process (i.e., the value passed to exit, or
6400 returned from main). If the child process did not exit normally,
6401 this function returns #f.
6403 (status:stop-sig STATUS)
6404 If the child process was suspended by a signal, this function
6405 returns the signal that suspended the child. Otherwise, it returns
6408 (status:term-sig STATUS)
6409 If the child process terminated abnormally, this function returns
6410 the signal that terminated the child. Otherwise, this function
6413 POSIX promises that exactly one of these functions will return true on
6414 a valid STATUS value.
6416 These functions are compatible with SCSH.
6418 *** There are new accessors and setters for the broken-out time vectors
6419 returned by `localtime', `gmtime', and that ilk. They are:
6421 Component Accessor Setter
6422 ========================= ============ ============
6423 seconds tm:sec set-tm:sec
6424 minutes tm:min set-tm:min
6425 hours tm:hour set-tm:hour
6426 day of the month tm:mday set-tm:mday
6427 month tm:mon set-tm:mon
6428 year tm:year set-tm:year
6429 day of the week tm:wday set-tm:wday
6430 day in the year tm:yday set-tm:yday
6431 daylight saving time tm:isdst set-tm:isdst
6432 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6433 name of time zone tm:zone set-tm:zone
6435 *** There are new accessors for the vectors returned by `uname',
6436 describing the host system:
6439 ============================================== ================
6440 name of the operating system implementation utsname:sysname
6441 network name of this machine utsname:nodename
6442 release level of the operating system utsname:release
6443 version level of the operating system utsname:version
6444 machine hardware platform utsname:machine
6446 *** There are new accessors for the vectors returned by `getpw',
6447 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6448 system's user database:
6451 ====================== =================
6452 user name passwd:name
6453 user password passwd:passwd
6456 real name passwd:gecos
6457 home directory passwd:dir
6458 shell program passwd:shell
6460 *** There are new accessors for the vectors returned by `getgr',
6461 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6462 system's group database:
6465 ======================= ============
6466 group name group:name
6467 group password group:passwd
6469 group members group:mem
6471 *** There are new accessors for the vectors returned by `gethost',
6472 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6476 ========================= ===============
6477 official name of host hostent:name
6478 alias list hostent:aliases
6479 host address type hostent:addrtype
6480 length of address hostent:length
6481 list of addresses hostent:addr-list
6483 *** There are new accessors for the vectors returned by `getnet',
6484 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6488 ========================= ===============
6489 official name of net netent:name
6490 alias list netent:aliases
6491 net number type netent:addrtype
6492 net number netent:net
6494 *** There are new accessors for the vectors returned by `getproto',
6495 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6499 ========================= ===============
6500 official protocol name protoent:name
6501 alias list protoent:aliases
6502 protocol number protoent:proto
6504 *** There are new accessors for the vectors returned by `getserv',
6505 `getservbyname', `getservbyport', and `getservent', describing
6509 ========================= ===============
6510 official service name servent:name
6511 alias list servent:aliases
6512 port number servent:port
6513 protocol to use servent:proto
6515 *** There are new accessors for the sockaddr structures returned by
6516 `accept', `getsockname', `getpeername', `recvfrom!':
6519 ======================================== ===============
6520 address format (`family') sockaddr:fam
6521 path, for file domain addresses sockaddr:path
6522 address, for internet domain addresses sockaddr:addr
6523 TCP or UDP port, for internet sockaddr:port
6525 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6526 `getprotoent', and `getservent' functions now return #f at the end of
6527 the user database. (They used to throw an exception.)
6529 Note that calling MUMBLEent function is equivalent to calling the
6530 corresponding MUMBLE function with no arguments.
6532 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6533 `setprotoent', and `setservent' routines now take no arguments.
6535 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6536 provide more useful information when they throw an exception.
6538 *** The `lnaof' function has been renamed to `inet-lnaof'.
6540 *** Guile now claims to have the `current-time' feature.
6542 *** The `mktime' function now takes an optional second argument ZONE,
6543 giving the time zone to use for the conversion. ZONE should be a
6544 string, in the same format as expected for the "TZ" environment variable.
6546 *** The `strptime' function now returns a pair (TIME . COUNT), where
6547 TIME is the parsed time as a vector, and COUNT is the number of
6548 characters from the string left unparsed. This function used to
6549 return the remaining characters as a string.
6551 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6552 The return value is now (SECONDS . MICROSECONDS); the fractional
6553 component is no longer expressed in "ticks".
6555 *** The `ticks/sec' constant has been removed, in light of the above change.
6557 * Changes to the gh_ interface
6559 ** gh_eval_str() now returns an SCM object which is the result of the
6562 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6565 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6566 and returns the array
6568 ** gh_scm2str0() is gone: there is no need to distinguish
6569 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6570 the user to interpret the data both ways.
6572 * Changes to the scm_ interface
6574 ** The new function scm_symbol_value0 provides an easy way to get a
6575 symbol's value from C code:
6577 SCM scm_symbol_value0 (char *NAME)
6578 Return the value of the symbol named by the null-terminated string
6579 NAME in the current module. If the symbol named NAME is unbound in
6580 the current module, return SCM_UNDEFINED.
6582 ** The new function scm_sysintern0 creates new top-level variables,
6583 without assigning them a value.
6585 SCM scm_sysintern0 (char *NAME)
6586 Create a new Scheme top-level variable named NAME. NAME is a
6587 null-terminated string. Return the variable's value cell.
6589 ** The function scm_internal_catch is the guts of catch. It handles
6590 all the mechanics of setting up a catch target, invoking the catch
6591 body, and perhaps invoking the handler if the body does a throw.
6593 The function is designed to be usable from C code, but is general
6594 enough to implement all the semantics Guile Scheme expects from throw.
6596 TAG is the catch tag. Typically, this is a symbol, but this function
6597 doesn't actually care about that.
6599 BODY is a pointer to a C function which runs the body of the catch;
6600 this is the code you can throw from. We call it like this:
6601 BODY (BODY_DATA, JMPBUF)
6603 BODY_DATA is just the BODY_DATA argument we received; we pass it
6604 through to BODY as its first argument. The caller can make
6605 BODY_DATA point to anything useful that BODY might need.
6606 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6607 which we have just created and initialized.
6609 HANDLER is a pointer to a C function to deal with a throw to TAG,
6610 should one occur. We call it like this:
6611 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6613 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6614 same idea as BODY_DATA above.
6615 THROWN_TAG is the tag that the user threw to; usually this is
6616 TAG, but it could be something else if TAG was #t (i.e., a
6617 catch-all), or the user threw to a jmpbuf.
6618 THROW_ARGS is the list of arguments the user passed to the THROW
6621 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6622 is just a pointer we pass through to HANDLER. We don't actually
6623 use either of those pointers otherwise ourselves. The idea is
6624 that, if our caller wants to communicate something to BODY or
6625 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6626 HANDLER can then use. Think of it as a way to make BODY and
6627 HANDLER closures, not just functions; MUMBLE_DATA points to the
6630 Of course, it's up to the caller to make sure that any data a
6631 MUMBLE_DATA needs is protected from GC. A common way to do this is
6632 to make MUMBLE_DATA a pointer to data stored in an automatic
6633 structure variable; since the collector must scan the stack for
6634 references anyway, this assures that any references in MUMBLE_DATA
6637 ** The new function scm_internal_lazy_catch is exactly like
6638 scm_internal_catch, except:
6640 - It does not unwind the stack (this is the major difference).
6641 - If handler returns, its value is returned from the throw.
6642 - BODY always receives #f as its JMPBUF argument (since there's no
6643 jmpbuf associated with a lazy catch, because we don't unwind the
6646 ** scm_body_thunk is a new body function you can pass to
6647 scm_internal_catch if you want the body to be like Scheme's `catch'
6648 --- a thunk, or a function of one argument if the tag is #f.
6650 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6651 contains the Scheme procedure to invoke as the body, and the tag
6652 we're catching. If the tag is #f, then we pass JMPBUF (created by
6653 scm_internal_catch) to the body procedure; otherwise, the body gets
6656 ** scm_handle_by_proc is a new handler function you can pass to
6657 scm_internal_catch if you want the handler to act like Scheme's catch
6658 --- call a procedure with the tag and the throw arguments.
6660 If the user does a throw to this catch, this function runs a handler
6661 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6662 variable holding the Scheme procedure object to invoke. It ought to
6663 be a pointer to an automatic variable (i.e., one living on the stack),
6664 or the procedure object should be otherwise protected from GC.
6666 ** scm_handle_by_message is a new handler function to use with
6667 `scm_internal_catch' if you want Guile to print a message and die.
6668 It's useful for dealing with throws to uncaught keys at the top level.
6670 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6671 message header to print; if zero, we use "guile" instead. That
6672 text is followed by a colon, then the message described by ARGS.
6674 ** The return type of scm_boot_guile is now void; the function does
6675 not return a value, and indeed, never returns at all.
6677 ** The new function scm_shell makes it easy for user applications to
6678 process command-line arguments in a way that is compatible with the
6679 stand-alone guile interpreter (which is in turn compatible with SCSH,
6682 To use the scm_shell function, first initialize any guile modules
6683 linked into your application, and then call scm_shell with the values
6684 of ARGC and ARGV your `main' function received. scm_shell will add
6685 any SCSH-style meta-arguments from the top of the script file to the
6686 argument vector, and then process the command-line arguments. This
6687 generally means loading a script file or starting up an interactive
6688 command interpreter. For details, see "Changes to the stand-alone
6691 ** The new functions scm_get_meta_args and scm_count_argv help you
6692 implement the SCSH-style meta-argument, `\'.
6694 char **scm_get_meta_args (int ARGC, char **ARGV)
6695 If the second element of ARGV is a string consisting of a single
6696 backslash character (i.e. "\\" in Scheme notation), open the file
6697 named by the following argument, parse arguments from it, and return
6698 the spliced command line. The returned array is terminated by a
6701 For details of argument parsing, see above, under "guile now accepts
6702 command-line arguments compatible with SCSH..."
6704 int scm_count_argv (char **ARGV)
6705 Count the arguments in ARGV, assuming it is terminated by a null
6708 For an example of how these functions might be used, see the source
6709 code for the function scm_shell in libguile/script.c.
6711 You will usually want to use scm_shell instead of calling this
6714 ** The new function scm_compile_shell_switches turns an array of
6715 command-line arguments into Scheme code to carry out the actions they
6716 describe. Given ARGC and ARGV, it returns a Scheme expression to
6717 evaluate, and calls scm_set_program_arguments to make any remaining
6718 command-line arguments available to the Scheme code. For example,
6719 given the following arguments:
6721 -e main -s ekko a speckled gecko
6723 scm_set_program_arguments will return the following expression:
6725 (begin (load "ekko") (main (command-line)) (quit))
6727 You will usually want to use scm_shell instead of calling this
6730 ** The function scm_shell_usage prints a usage message appropriate for
6731 an interpreter that uses scm_compile_shell_switches to handle its
6732 command-line arguments.
6734 void scm_shell_usage (int FATAL, char *MESSAGE)
6735 Print a usage message to the standard error output. If MESSAGE is
6736 non-zero, write it before the usage message, followed by a newline.
6737 If FATAL is non-zero, exit the process, using FATAL as the
6738 termination status. (If you want to be compatible with Guile,
6739 always use 1 as the exit status when terminating due to command-line
6742 You will usually want to use scm_shell instead of calling this
6745 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6746 expressions. It used to return SCM_EOL. Earth-shattering.
6748 ** The macros for declaring scheme objects in C code have been
6749 rearranged slightly. They are now:
6751 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6752 Declare a static SCM variable named C_NAME, and initialize it to
6753 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6754 be a C identifier, and SCHEME_NAME should be a C string.
6756 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6757 Just like SCM_SYMBOL, but make C_NAME globally visible.
6759 SCM_VCELL (C_NAME, SCHEME_NAME)
6760 Create a global variable at the Scheme level named SCHEME_NAME.
6761 Declare a static SCM variable named C_NAME, and initialize it to
6762 point to the Scheme variable's value cell.
6764 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6765 Just like SCM_VCELL, but make C_NAME globally visible.
6767 The `guile-snarf' script writes initialization code for these macros
6768 to its standard output, given C source code as input.
6770 The SCM_GLOBAL macro is gone.
6772 ** The scm_read_line and scm_read_line_x functions have been replaced
6773 by Scheme code based on the %read-delimited! procedure (known to C
6774 code as scm_read_delimited_x). See its description above for more
6777 ** The function scm_sys_open has been renamed to scm_open. It now
6778 returns a port instead of an FD object.
6780 * The dynamic linking support has changed. For more information, see
6781 libguile/DYNAMIC-LINKING.
6786 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6789 * Changes to the 'guile' program:
6791 ** Guile now loads some new files when it starts up. Guile first
6792 searches the load path for init.scm, and loads it if found. Then, if
6793 Guile is not being used to execute a script, and the user's home
6794 directory contains a file named `.guile', Guile loads that.
6796 ** You can now use Guile as a shell script interpreter.
6798 To paraphrase the SCSH manual:
6800 When Unix tries to execute an executable file whose first two
6801 characters are the `#!', it treats the file not as machine code to
6802 be directly executed by the native processor, but as source code
6803 to be executed by some interpreter. The interpreter to use is
6804 specified immediately after the #! sequence on the first line of
6805 the source file. The kernel reads in the name of the interpreter,
6806 and executes that instead. It passes the interpreter the source
6807 filename as its first argument, with the original arguments
6808 following. Consult the Unix man page for the `exec' system call
6809 for more information.
6811 Now you can use Guile as an interpreter, using a mechanism which is a
6812 compatible subset of that provided by SCSH.
6814 Guile now recognizes a '-s' command line switch, whose argument is the
6815 name of a file of Scheme code to load. It also treats the two
6816 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6817 to make a file of Scheme code directly executable by Unix, insert the
6818 following two lines at the top of the file:
6820 #!/usr/local/bin/guile -s
6823 Guile treats the argument of the `-s' command-line switch as the name
6824 of a file of Scheme code to load, and treats the sequence `#!' as the
6825 start of a block comment, terminated by `!#'.
6827 For example, here's a version of 'echo' written in Scheme:
6829 #!/usr/local/bin/guile -s
6831 (let loop ((args (cdr (program-arguments))))
6834 (display (car args))
6835 (if (pair? (cdr args))
6837 (loop (cdr args)))))
6840 Why does `#!' start a block comment terminated by `!#', instead of the
6841 end of the line? That is the notation SCSH uses, and although we
6842 don't yet support the other SCSH features that motivate that choice,
6843 we would like to be backward-compatible with any existing Guile
6844 scripts once we do. Furthermore, if the path to Guile on your system
6845 is too long for your kernel, you can start the script with this
6849 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6852 Note that some very old Unix systems don't support the `#!' syntax.
6855 ** You can now run Guile without installing it.
6857 Previous versions of the interactive Guile interpreter (`guile')
6858 couldn't start up unless Guile's Scheme library had been installed;
6859 they used the value of the environment variable `SCHEME_LOAD_PATH'
6860 later on in the startup process, but not to find the startup code
6861 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6864 To run Guile without installing it, build it in the normal way, and
6865 then set the environment variable `SCHEME_LOAD_PATH' to a
6866 colon-separated list of directories, including the top-level directory
6867 of the Guile sources. For example, if you unpacked Guile so that the
6868 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6871 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6874 ** Guile's read-eval-print loop no longer prints #<unspecified>
6875 results. If the user wants to see this, she can evaluate the
6876 expression (assert-repl-print-unspecified #t), perhaps in her startup
6879 ** Guile no longer shows backtraces by default when an error occurs;
6880 however, it does display a message saying how to get one, and how to
6881 request that they be displayed by default. After an error, evaluate
6883 to see a backtrace, and
6884 (debug-enable 'backtrace)
6885 to see them by default.
6889 * Changes to Guile Scheme:
6891 ** Guile now distinguishes between #f and the empty list.
6893 This is for compatibility with the IEEE standard, the (possibly)
6894 upcoming Revised^5 Report on Scheme, and many extant Scheme
6897 Guile used to have #f and '() denote the same object, to make Scheme's
6898 type system more compatible with Emacs Lisp's. However, the change
6899 caused too much trouble for Scheme programmers, and we found another
6900 way to reconcile Emacs Lisp with Scheme that didn't require this.
6903 ** Guile's delq, delv, delete functions, and their destructive
6904 counterparts, delq!, delv!, and delete!, now remove all matching
6905 elements from the list, not just the first. This matches the behavior
6906 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6907 functions which inspired them.
6909 I recognize that this change may break code in subtle ways, but it
6910 seems best to make the change before the FSF's first Guile release,
6914 ** The compiled-library-path function has been deleted from libguile.
6916 ** The facilities for loading Scheme source files have changed.
6918 *** The variable %load-path now tells Guile which directories to search
6919 for Scheme code. Its value is a list of strings, each of which names
6922 *** The variable %load-extensions now tells Guile which extensions to
6923 try appending to a filename when searching the load path. Its value
6924 is a list of strings. Its default value is ("" ".scm").
6926 *** (%search-load-path FILENAME) searches the directories listed in the
6927 value of the %load-path variable for a Scheme file named FILENAME,
6928 with all the extensions listed in %load-extensions. If it finds a
6929 match, then it returns its full filename. If FILENAME is absolute, it
6930 returns it unchanged. Otherwise, it returns #f.
6932 %search-load-path will not return matches that refer to directories.
6934 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6935 uses %seach-load-path to find a file named FILENAME, and loads it if
6936 it finds it. If it can't read FILENAME for any reason, it throws an
6939 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6942 *** load uses the same searching semantics as primitive-load.
6944 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6945 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6946 path, and load-module-with-path have been deleted. The functions
6947 above should serve their purposes.
6949 *** If the value of the variable %load-hook is a procedure,
6950 `primitive-load' applies its value to the name of the file being
6951 loaded (without the load path directory name prepended). If its value
6952 is #f, it is ignored. Otherwise, an error occurs.
6954 This is mostly useful for printing load notification messages.
6957 ** The function `eval!' is no longer accessible from the scheme level.
6958 We can't allow operations which introduce glocs into the scheme level,
6959 because Guile's type system can't handle these as data. Use `eval' or
6960 `read-and-eval!' (see below) as replacement.
6962 ** The new function read-and-eval! reads an expression from PORT,
6963 evaluates it, and returns the result. This is more efficient than
6964 simply calling `read' and `eval', since it is not necessary to make a
6965 copy of the expression for the evaluator to munge.
6967 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6968 for the `read' function.
6971 ** The function `int?' has been removed; its definition was identical
6972 to that of `integer?'.
6974 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6975 use the R4RS names for these functions.
6977 ** The function object-properties no longer returns the hash handle;
6978 it simply returns the object's property list.
6980 ** Many functions have been changed to throw errors, instead of
6981 returning #f on failure. The point of providing exception handling in
6982 the language is to simplify the logic of user code, but this is less
6983 useful if Guile's primitives don't throw exceptions.
6985 ** The function `fileno' has been renamed from `%fileno'.
6987 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6990 * Changes to Guile's C interface:
6992 ** The library's initialization procedure has been simplified.
6993 scm_boot_guile now has the prototype:
6995 void scm_boot_guile (int ARGC,
6997 void (*main_func) (),
7000 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7001 MAIN_FUNC should do all the work of the program (initializing other
7002 packages, reading user input, etc.) before returning. When MAIN_FUNC
7003 returns, call exit (0); this function never returns. If you want some
7004 other exit value, MAIN_FUNC may call exit itself.
7006 scm_boot_guile arranges for program-arguments to return the strings
7007 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7008 scm_set_program_arguments with the final list, so Scheme code will
7009 know which arguments have been processed.
7011 scm_boot_guile establishes a catch-all catch handler which prints an
7012 error message and exits the process. This means that Guile exits in a
7013 coherent way when system errors occur and the user isn't prepared to
7014 handle it. If the user doesn't like this behavior, they can establish
7015 their own universal catcher in MAIN_FUNC to shadow this one.
7017 Why must the caller do all the real work from MAIN_FUNC? The garbage
7018 collector assumes that all local variables of type SCM will be above
7019 scm_boot_guile's stack frame on the stack. If you try to manipulate
7020 SCM values after this function returns, it's the luck of the draw
7021 whether the GC will be able to find the objects you allocate. So,
7022 scm_boot_guile function exits, rather than returning, to discourage
7023 people from making that mistake.
7025 The IN, OUT, and ERR arguments were removed; there are other
7026 convenient ways to override these when desired.
7028 The RESULT argument was deleted; this function should never return.
7030 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7034 ** Guile's header files should no longer conflict with your system's
7037 In order to compile code which #included <libguile.h>, previous
7038 versions of Guile required you to add a directory containing all the
7039 Guile header files to your #include path. This was a problem, since
7040 Guile's header files have names which conflict with many systems'
7043 Now only <libguile.h> need appear in your #include path; you must
7044 refer to all Guile's other header files as <libguile/mumble.h>.
7045 Guile's installation procedure puts libguile.h in $(includedir), and
7046 the rest in $(includedir)/libguile.
7049 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7050 have been added to the Guile library.
7052 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7053 OBJ will not be freed, even if all other references are dropped,
7054 until someone does scm_unprotect_object (OBJ). Both functions
7057 Note that calls to scm_protect_object do not nest. You can call
7058 scm_protect_object any number of times on a given object, and the
7059 next call to scm_unprotect_object will unprotect it completely.
7061 Basically, scm_protect_object and scm_unprotect_object just
7062 maintain a list of references to things. Since the GC knows about
7063 this list, all objects it mentions stay alive. scm_protect_object
7064 adds its argument to the list; scm_unprotect_object remove its
7065 argument from the list.
7068 ** scm_eval_0str now returns the value of the last expression
7071 ** The new function scm_read_0str reads an s-expression from a
7072 null-terminated string, and returns it.
7074 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7075 to a Scheme port object.
7077 ** The new function `scm_set_program_arguments' allows C code to set
7078 the value returned by the Scheme `program-arguments' function.
7083 * Guile no longer includes sophisticated Tcl/Tk support.
7085 The old Tcl/Tk support was unsatisfying to us, because it required the
7086 user to link against the Tcl library, as well as Tk and Guile. The
7087 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7088 referring to widgets by names, rather than exporting widgets to Scheme
7089 code as a special datatype.
7091 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7092 maintainers described some very interesting changes in progress to the
7093 Tcl/Tk internals, which would facilitate clean interfaces between lone
7094 Tk and other interpreters --- even for garbage-collected languages
7095 like Scheme. They expected the new Tk to be publicly available in the
7098 Since it seems that Guile might soon have a new, cleaner interface to
7099 lone Tk, and that the old Guile/Tk glue code would probably need to be
7100 completely rewritten, we (Jim Blandy and Richard Stallman) have
7101 decided not to support the old code. We'll spend the time instead on
7102 a good interface to the newer Tk, as soon as it is available.
7104 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7107 Copyright information:
7109 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
7111 Permission is granted to anyone to make or distribute verbatim copies
7112 of this document as received, in any medium, provided that the
7113 copyright notice and this permission notice are preserved,
7114 thus giving the recipient permission to redistribute in turn.
7116 Permission is granted to distribute modified versions
7117 of this document, or of portions of it,
7118 under the above conditions, provided also that they
7119 carry prominent notices stating who last changed them.
7124 paragraph-separate: "[
\f]*$"