1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
7 Each release reports the NEWS in the following sections:
9 * Changes to the distribution
10 * Changes to the stand-alone interpreter
11 * Changes to Scheme functions and syntax
12 * Changes to the C interface
15 Changes since the 1.6.x series:
17 * Changes to the distribution
19 ** Guile is now licensed with the GNU Lesser General Public License.
21 ** The manual is now licensed with the GNU Free Documentation License.
23 ** Guile now requires GNU MP (http://swox.com/gmp).
25 Guile now uses the GNU MP library for arbitrary precision arithmetic.
27 ** Guile now has separate private and public configuration headers.
29 That is, things like HAVE_STRING_H no longer leak from Guile's
32 ** Guile now provides and uses an "effective" version number.
34 Guile now provides scm_effective_version and effective-version
35 functions which return the "effective" version number. This is just
36 the normal full version string without the final micro-version number,
37 so the current effective-version is "1.7". The effective version
38 should remain unchanged during a stable series, and should be used for
39 items like the versioned share directory name
40 i.e. /usr/share/guile/1.7.
42 Providing an unchanging version number during a stable release for
43 things like the versioned share directory can be particularly
44 important for Guile "add-on" packages, since it provides a directory
45 that they can install to that won't be changed out from under them
46 with each micro release during a stable series.
48 ** Thread implementation has changed.
50 When you configure "--with-threads=null", you will get the usual
51 threading API (call-with-new-thread, make-mutex, etc), but you can't
52 actually create new threads. Also, "--with-threads=no" is now
53 equivalent to "--with-threads=null". This means that the thread API
54 is always present, although you might not be able to create new
57 When you configure "--with-threads=pthreads" or "--with-threads=yes",
58 you will get threads that are implemented with the portable POSIX
59 threads. These threads can run concurrently (unlike the previous
60 "coop" thread implementation), but need to cooperate for things like
61 the GC. See the manual for details. [XXX - write this.]
63 The default is "pthreads", unless your platform doesn't have pthreads,
64 in which case "null" threads are used.
66 ** New module (ice-9 serialize):
68 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when
69 you don't trust the thread safety of most of your program, but
70 where you have some section(s) of code which you consider can run
71 in parallel to other sections.
73 ### move rest to manual
75 They "flag" (with dynamic extent) sections of code to be of
76 "serial" or "parallel" nature and have the single effect of
77 preventing a serial section from being run in parallel with any
78 serial section (including itself).
80 Both serialize and parallelize can be nested. If so, the
81 inner-most construct is in effect.
83 NOTE 1: A serial section can run in parallel with a parallel
86 NOTE 2: If a serial section S is "interrupted" by a parallel
87 section P in the following manner: S = S1 P S2, S2 is not
88 guaranteed to be resumed by the same thread that previously
91 WARNING: Spawning new threads within a serial section have
92 undefined effects. It is OK, though, to spawn threads in unflagged
93 sections of code where neither serialize or parallelize is in
96 A typical usage is when Guile is used as scripting language in some
97 application doing heavy computations. If each thread is
98 encapsulated with a serialize form, you can then put a parallelize
99 form around the code performing the heavy computations (typically a
100 C code primitive), enabling the computations to run in parallel
101 while the scripting code runs single-threadedly.
103 ** New module (srfi srfi-26)
105 This is an implementation of SRFI-26.
107 ** New module (srfi srfi-31)
109 This is an implementation of SRFI-31 which provides a special form
110 `rec' for recursive evaluation.
112 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
113 been merged with the core, making their functionality always
116 The modules are still available, tho, and you could use them together
117 with a renaming import, for example.
119 ** Guile no longer includes its own version of libltdl.
121 The official version is good enough now.
123 ** The --enable-htmldoc option has been removed from 'configure'.
125 Support for translating the documentation into HTML is now always
126 provided. Use 'make html'.
128 * Changes to the stand-alone interpreter
130 ** New command line option `-L'.
132 This option adds a directory to the front of the load path.
134 ** New command line option `--no-debug'.
136 Specifying `--no-debug' on the command line will keep the debugging
137 evaluator turned off, even for interactive sessions.
139 ** User-init file ~/.guile is now loaded with the debugging evaluator.
141 Previously, the normal evaluator would have been used. Using the
142 debugging evaluator gives better error messages.
144 ** The '-e' option now 'read's its argument.
146 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
147 be used with '-e'. For example, you can now write a script like
150 exec guile -e '(@ (demo) main)' -s "$0" "$@"
153 (define-module (demo)
157 (format #t "Demo: ~a~%" args))
160 * Changes to Scheme functions and syntax
162 ** The default load path no longer includes "." at the end.
164 Automatically loading modules from the current directory should not
165 happen by default. If you want to allow it in a more controlled
166 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
169 ** The uniform vector and array support has been overhauled.
171 It now complies with SRFI-4 and the weird prototype based uniform
172 array creation has been deprecated. See the manual for more details.
174 Some non-compatible changes have been made:
175 - characters can no longer be stored into byte arrays.
176 - strings and bit vectors are no longer considered to be uniform vectors.
177 - array-rank throws an error for non-arrays instead of returning zero.
178 - array-ref does no longer accept non-arrays when no indices are given.
180 There is the new notion of 'generalized vectors' and corresponding
181 procedures like 'generalized-vector-ref'. Generalized vectors include
182 strings bitvectors, ordinary vectors, and uniform numeric vectors.
184 Arrays use generalized vectors their storage, so that you still have
185 arrays of characters, bits, etc. However, uniform-array-read! and
186 uniform-array-write can no longer read/write strings and bitvectors.
188 ** There is now support for copy-on-write substrings, mutation-sharing
189 substrings and read-only strings.
191 Three new procedures are related to this: substring/shared,
192 substring/copy, and substring/read-only. See the manual for more
195 ** Backtraces will now highlight the value that caused the error.
197 By default, these values are enclosed in "{...}", such as in this
206 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
207 <unnamed port>:1:1: Wrong type (expecting pair): a
208 ABORT: (wrong-type-arg)
210 The prefix and suffix used for highlighting can be set via the two new
211 printer options 'highlight-prefix' and 'highlight-suffix'. For
212 example, putting this into ~/.guile will output the bad value in bold
215 (print-set! highlight-prefix "\x1b[1m")
216 (print-set! highlight-suffix "\x1b[22m")
219 ** 'gettext' support for internationalization has been added.
221 See the manual for details.
223 ** New syntax '@' and '@@':
225 You can now directly refer to variables exported from a module by
228 (@ MODULE-NAME VARIABLE-NAME)
230 For example (@ (ice-9 pretty-print) pretty-print) will directly access
231 the pretty-print variable exported from the (ice-9 pretty-print)
232 module. You don't need to 'use' that module first. You can also use
233 '@' as a target of 'set!', as in (set! (@ mod var) val).
235 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
236 but it can also access variables that have not been exported. It is
237 intended only for kluges and temporary fixes and for debugging, not
240 ** Keyword syntax has been made more disciplined.
242 Previously, the name of a keyword was read as a 'token' but printed as
243 a symbol. Now, it is read as a general Scheme datum which must be a
254 ERROR: In expression (a b c):
258 ERROR: Unbound variable: foo
263 ERROR: Wrong type (expecting symbol): 12
267 ERROR: Wrong type (expecting symbol): (a b c)
271 ** 'while' now provides 'break' and 'continue'
273 break and continue were previously bound in a while loop, but not
274 documented, and continue didn't quite work properly. The undocumented
275 parameter to break which gave a return value for the while has been
278 ** 'call-with-current-continuation' is now also available under the name
281 ** The module system now checks for duplicate bindings.
283 The module system now can check for name conflicts among imported
286 The behavior can be controlled by specifying one or more 'duplicates'
287 handlers. For example, to make Guile return an error for every name
295 The new default behavior of the module system when a name collision
296 has been detected is to
298 1. Give priority to bindings marked as a replacement.
299 2. Issue a warning (different warning if overriding core binding).
300 3. Give priority to the last encountered binding (this corresponds to
303 If you want the old behavior back without replacements or warnings you
306 (default-duplicate-binding-handler 'last)
308 to your .guile init file.
310 ### move rest to manual
312 The syntax for the :duplicates option is:
314 :duplicates HANDLER-NAME | (HANDLER1-NAME HANDLER2-NAME ...)
316 Specifying multiple handlers is useful since some handlers (such as
317 replace) can defer conflict resolution to others. Each handler is
318 tried until a binding is selected.
320 Currently available duplicates handlers are:
322 check report an error for bindings with a common name
323 warn issue a warning for bindings with a common name
324 replace replace bindings which have an imported replacement
325 warn-override-core issue a warning for imports which override core bindings
326 and accept the override
327 first select the first encountered binding (override)
328 last select the last encountered binding (override)
330 These two are provided by the (oop goops) module:
332 merge-generics merge generic functions with a common name
333 into an <extended-generic>
334 merge-accessors merge accessors with a common name
336 The default duplicates handler is:
338 (replace warn-override-core warn last)
340 A recommended handler (which is likely to correspond to future Guile
341 behavior) can be installed with:
343 (default-duplicate-binding-handler '(replace warn-override-core check))
345 ** New define-module option: :replace
347 :replace works as :export, but, in addition, marks the binding as a
350 A typical example is `format' in (ice-9 format) which is a replacement
351 for the core binding `format'.
353 ** Adding prefixes to imported bindings in the module system
355 There is now a new :use-module option :prefix. It can be used to add
356 a prefix to all imported bindings.
359 :use-module ((bar) :prefix bar:))
361 will import all bindings exported from bar, but rename them by adding
364 ** Conflicting generic functions can be automatically merged.
366 When two imported bindings conflict and they are both generic
367 functions, the two functions can now be merged automatically. This is
368 activated with the 'duplicates' handler 'merge-generics'.
370 ### move the rest to the manual
372 It is sometimes tempting to use GOOPS accessors with short names.
373 For example, it is tempting to use the name `x' for the x-coordinate
376 Assume that we work with a graphical package which needs to use two
377 independent vector packages for 2D and 3D vectors respectively. If
378 both packages export `x' we will encounter a name collision.
380 This can now be resolved automagically with the duplicates handler
381 `merge-generics' which gives the module system license to merge all
382 generic functions sharing a common name:
384 (define-module (math 2D-vectors)
385 :use-module (oop goops)
388 (define-module (math 3D-vectors)
389 :use-module (oop goops)
392 (define-module (my-module)
393 :use-module (math 2D-vectors)
394 :use-module (math 3D-vectors)
395 :duplicates merge-generics)
397 x in (my-module) will now share methods with x in both imported
400 There will, in fact, now be three distinct generic functions named
401 `x': x in (2D-vectors), x in (3D-vectors), and x in (my-module). The
402 last function will be an <extended-generic>, extending the previous
405 Let's call the imported generic functions the "ancestor functions". x
406 in (my-module) is, in turn, a "descendant function" of the imported
407 functions, extending its ancestors.
409 For any generic function G, the applicable methods are selected from
410 the union of the methods of the descendant functions, the methods of G
411 itself and the methods of the ancestor functions.
413 This, ancestor functions share methods with their descendants and vice
414 versa. This implies that x in (math 2D-vectors) can will share the
415 methods of x in (my-module) and vice versa, while x in (math 2D-vectors)
416 doesn't share the methods of x in (math 3D-vectors), thus preserving
419 Sharing is dynamic, so that adding new methods to a descendant implies
420 adding it to the ancestor.
422 If duplicates checking is desired in the above example, the following
423 form of the :duplicates option can be used instead:
425 :duplicates (merge-generics check)
427 ** New function: effective-version
429 Returns the "effective" version number. This is just the normal full
430 version string without the final micro-version number. See "Changes
431 to the distribution" above.
433 ** New feature, 'futures': future, make-future, future-ref
435 Futures are like promises, but begin execution immediately in a new
436 thread. See the "Futures" section in the reference manual.
438 ** New threading functions: parallel, letpar, par-map, and friends
440 These are convenient ways to run calculations in parallel in new
441 threads. See "Parallel forms" in the manual for details.
443 ** Fair mutexes and condition variables
445 Fair mutexes and condition variables have been added. The fairness
446 means that scheduling is arranged to give as equal time shares as
447 possible and that threads are awakened in a first-in-first-out
448 manner. This is not guaranteed with standard mutexes and condition
451 In addition, fair mutexes are recursive. Locking a fair mutex that
452 you have already locked will succeed. Every call to lock-mutex must
453 be matched with a call to unlock-mutex. Only the last call to
454 unlock-mutex will actually unlock the mutex.
456 A fair condition variable must be used together with a fair mutex,
457 just as a standard condition variable must be used together with a
460 *** New functions: make-fair-mutex, make-fair-condition-variable'
462 Make a new fair mutex and a new fair condition variable respectively.
464 ** New function 'try-mutex'.
466 This function will attempt to lock a mutex but will return immediately
467 instead if blocking and indicate failure.
469 ** Waiting on a condition variable can have a timeout.
471 The funtion 'wait-condition-variable' now takes a third, optional
472 argument that specifies the point in time where the waiting should be
475 ** New function 'broadcast-condition-variable'.
477 ** New functions 'all-threads' and 'current-thread'.
479 ** Signals and system asyncs work better with threads.
481 The function 'sigaction' now takes a fourth, optional, argument that
482 specifies the thread that the handler should run in. When the
483 argument is omitted, the handler will run in the thread that called
486 Likewise, 'system-async-mark' takes a second, optional, argument that
487 specifies the thread that the async should run in. When it is
488 omitted, the async will run in the thread that called
491 C code can use the new functions scm_sigaction_for_thread and
492 scm_system_async_mark_for_thread to pass the new thread argument.
494 ** The function 'system-async' is deprecated.
496 You can now pass any zero-argument procedure to 'system-async-mark'.
497 The function 'system-async' will just return its argument unchanged
500 ** New functions 'call-with-blocked-asyncs' and
501 'call-with-unblocked-asyncs'
503 The expression (call-with-blocked-asyncs PROC) will call PROC and will
504 block execution of system asyncs for the current thread by one level
505 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
506 procedure and will unblock the execution of system asyncs by one
507 level for the current thread.
509 Only system asyncs are affected by these functions.
511 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
513 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
514 instead. Those functions are easier to use correctly and can be
517 ** New function 'unsetenv'.
519 ** New macro 'define-syntax-public'.
521 It works like 'define-syntax' and also exports the defined macro (but
524 ** There is support for Infinity and NaNs.
526 Following PLT Scheme, Guile can now work with infinite numbers, and
529 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
530 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
531 "+nan.0"). These numbers are inexact and have no exact counterpart.
533 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
534 sign of the dividend. The infinities are integers, and they answer #t
535 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
536 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
547 ERROR: Numerical overflow
549 Two new predicates 'inf?' and 'nan?' can be used to test for the
552 ** Inexact zero can have a sign.
554 Guile can now distinguish between plus and minus inexact zero, if your
555 platform supports this, too. The two zeros are equal according to
556 '=', but not according to 'eqv?'. For example
567 ** Guile now has exact rationals.
569 Guile can now represent fractions such as 1/3 exactly. Computing with
570 them is also done exactly, of course:
575 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
578 For example: (floor 2) now returns an exact 2 where in the past it
579 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
581 ** inexact->exact no longer returns only integers.
583 Without exact rationals, the closest exact number was always an
584 integer, but now inexact->exact returns the fraction that is exactly
585 equal to a floating point number. For example:
587 (inexact->exact 1.234)
588 => 694680242521899/562949953421312
590 When you want the old behavior, use 'round' explicitely:
592 (inexact->exact (round 1.234))
595 ** New function 'rationalize'.
597 This function finds a simple fraction that is close to a given real
598 number. For example (and compare with inexact->exact above):
600 (rationalize (inexact->exact 1.234) 1/2000)
603 Note that, as required by R5RS, rationalize returns only then an exact
604 result when both its arguments are exact.
606 ** 'odd?' and 'even?' work also for inexact integers.
608 Previously, (odd? 1.0) would signal an error since only exact integers
609 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
610 returns #f and (odd? 1.5) signals an error.
612 ** Guile now has uninterned symbols.
614 The new function 'make-symbol' will return an uninterned symbol. This
615 is a symbol that is unique and is guaranteed to remain unique.
616 However, uninterned symbols can not yet be read back in.
618 Use the new function 'symbol-interned?' to check whether a symbol is
621 ** pretty-print has more options.
623 The function pretty-print from the (ice-9 pretty-print) module can now
624 also be invoked with keyword arguments that control things like
625 maximum output width. See the manual for details.
627 ** Variables have no longer a special behavior for `equal?'.
629 Previously, comparing two variables with `equal?' would recursivly
630 compare their values. This is no longer done. Variables are now only
631 `equal?' if they are `eq?'.
633 ** `(begin)' is now valid.
635 You can now use an empty `begin' form. It will yield #<unspecified>
636 when evaluated and simply be ignored in a definition context.
638 ** Deprecated: procedure->macro
640 Change your code to use 'define-macro' or r5rs macros. Also, be aware
641 that macro expansion will not be done during evaluation, but prior to
644 ** Soft ports now allow a `char-ready?' procedure
646 The vector argument to `make-soft-port' can now have a length of
647 either 5 or 6. (Previously the length had to be 5.) The optional 6th
648 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
649 that returns the number of characters that can be read immediately
650 without the soft port blocking.
652 ** New debugging feature: breakpoints.
654 Guile now has breakpoints. For details see the `Debugging Features'
655 chapter in the reference manual.
657 ** Deprecated: undefine
659 There is no replacement for undefine.
661 * The functions make-keyword-from-dash-symbol and keyword-dash-symbol
662 have been discouraged.
664 They are relics from a time where a keyword like #:foo was used
665 directly as a Tcl option "-foo" and thus keywords were internally
666 stored as a symbol with a starting dash. We now store a symbol
669 Use symbol->keyword and keyword->symbol instead.
672 * Changes to the C interface
674 ** There is the new notion of 'discouraged' features.
676 This is a milder form of deprecation.
678 Things that are discouraged should not be used in new code, but it is
679 OK to leave them in old code for now. When a discouraged feature is
680 used, no warning message is printed like there is for 'deprecated'
681 features. Also, things that are merely discouraged are nevertheless
682 implemented efficiently, while deprecated features can be very slow.
684 You can omit discouraged features from libguile by configuring it with
685 the '--disable-discouraged' option.
687 ** A new family of functions for converting between C values and
688 Scheme values has been added.
690 These functions follow a common naming scheme and are designed to be
691 easier to use, thread-safe and more future-proof than the older
696 These are predicates that return a C boolean: 1 or 0. Instead of
697 SCM_NFALSEP, you can now use scm_is_true, for example.
699 - <type> scm_to_<type> (SCM val, ...)
701 These are functions that convert a Scheme value into an appropriate
702 C value. For example, you can use scm_to_int to safely convert from
705 - SCM scm_from_<type> (<type> val, ...)
707 These functions convert from a C type to a SCM value; for example,
708 scm_from_int for ints.
710 There is a huge number of these functions, for numbers, strings,
711 symbols, vectors, etc. They are documented in the reference manual in
712 the API section together with the types that they apply to.
714 ** New functions for dealing with complex numbers in C have been added.
716 The new functions are scm_c_make_rectangular, scm_c_make_polar,
717 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
718 They work like scm_make_rectangular etc but take or return doubles
721 ** The function scm_make_complex has been discouraged.
723 Use scm_c_make_rectangular instead.
725 ** The INUM macros have been deprecated.
727 A lot of code uses these macros to do general integer conversions,
728 although the macros only work correctly with fixnums. Use the
729 following alternatives.
731 SCM_INUMP -> scm_is_integer or similar
732 SCM_NINUMP -> !scm_is_integer or similar
733 SCM_MAKINUM -> scm_from_int or similar
734 SCM_INUM -> scm_to_int or similar
736 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
737 do the validating for you.
739 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
740 have been discouraged.
742 Use the newer scm_to_<type> and scm_from_<type> functions instead for
743 new code. The functions have been discouraged since they don't fit
746 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
748 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
749 evaluates its argument twice. Use scm_is_true, etc. instead for new
752 ** The macro SCM_EQ_P has been discouraged.
754 Use scm_is_eq for new code, which fits better into the naming
757 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
760 Use the function scm_is_pair or scm_is_null instead.
762 ** The functions scm_round and scm_truncate have been deprecated and
763 are now available as scm_c_round and scm_c_truncate, respectively.
765 These functions occupy the names that scm_round_number and
766 scm_truncate_number should have.
768 ** The functions scm_c_string2str, scm_c_substring2str, and
769 scm_c_symbol2str have been deprecated.
771 Use scm_to_locale_stringbuf or similar instead, maybe together with
774 ** New functions scm_c_make_string, scm_c_string_length,
775 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
776 scm_c_substring_shared, scm_c_substring_copy.
778 These are like scm_make_string, scm_length, etc. but are slightly
779 easier to use from C.
781 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
782 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
784 They export too many assumptions about the implementation of strings
785 and symbols that are no longer true in the presence of
786 mutation-sharing substrings and when Guile switches to some form of
789 When working with strings, it is often best to use the normal string
790 functions provided by Guile, such as scm_c_string_ref,
791 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
792 manual since many more such functions are now provided than
795 When you want to convert a SCM string to a C string, use the
796 scm_to_locale_string function or similar instead. For symbols, use
797 scm_symbol_to_string and then work with that string. Because of the
798 new string representation, scm_symbol_to_string does not need to copy
799 and is thus quite efficient.
801 ** Some string, symbol and keyword functions have been discouraged.
803 They don't fit into the uniform naming scheme and are not explicit
804 about the character encoding.
806 Replace according to the following table:
808 scm_allocate_string -> scm_c_make_string
809 scm_take_str -> scm_take_locale_stringn
810 scm_take0str -> scm_take_locale_string
811 scm_mem2string -> scm_from_locale_stringn
812 scm_str2string -> scm_from_locale_string
813 scm_makfrom0str -> scm_from_locale_string
814 scm_mem2symbol -> scm_from_locale_symboln
815 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
816 scm_str2symbol -> scm_from_locale_symbol
818 SCM_SYMBOL_HASH -> scm_hashq
819 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
821 scm_c_make_keyword -> scm_from_locale_keyword
823 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
824 now also available to C code.
826 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
828 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
829 the latter returns the true name of the keyword, not the 'dash name',
830 as SCM_KEYWORDSYM used to do.
832 ** The old uniform vector and bitvector implementations have been
833 unceremoniously removed.
835 This implementation exposed the detailes of the tagging system of
836 Guile. Use the new C API explained in the manual in node "Uniform
837 Numeric Vectors" and "Bit Vector", respectively.
839 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
840 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
841 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
842 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
843 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
844 SCM_SET_BITVECTOR_LENGTH.
846 ** SCM_CELL_WORD_LOC has been deprecated.
848 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
849 to a SCM, as opposed to a pointer to a scm_t_bits.
851 This was done to allow the correct use of pointers into the Scheme
852 heap. Previously, the heap words were of type scm_t_bits and local
853 variables and function arguments were of type SCM, making it
854 non-standards-conformant to have a pointer that can point to both.
856 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
858 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
859 second and third words of double smobs. Likewise for
860 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
862 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
863 used to get and set the 16 exra bits in the zeroth word of a smob.
865 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
866 accesing the first immediate word of a smob as a SCM value, and there
867 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
868 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
870 ** New way to deal with non-local exits and re-entries.
872 There is a new set of functions that essentially do what
873 scm_internal_dynamic_wind does, but in a way that is more convenient
874 for C code in some situations. Here is a quick example of how to
875 prevent a potential memory leak:
884 mem = scm_malloc (100);
885 scm_frame_unwind_handler (free, mem, SCM_F_WIND_EXPLICITELY);
887 /* MEM would leak if BAR throws an error.
888 SCM_FRAME_UNWIND_HANDLER frees it nevertheless.
895 /* Because of SCM_F_WIND_EXPLICITELY, MEM will be freed by
896 SCM_FRAME_END as well.
900 For full documentation, see the node "Frames" in the manual.
902 ** New function scm_frame_free
904 This function calls 'free' on a given pointer when a frame is left.
905 Thus the call to scm_frame_unwind_handler above could be replaced with
906 simply scm_frame_free (mem).
908 ** New way to block and unblock asyncs
910 In addition to scm_c_call_with_blocked_asyncs you can now also use
911 scm_frame_block_asyncs in a 'frame' (see above). Likewise for
912 scm_c_call_with_unblocked_asyncs and scm_frame_unblock_asyncs.
914 ** New way to temporarily set the current input, output or error ports
916 C code can now use scm_frame_current_<foo>_port in a 'frame' (see
917 above). <foo> is one of "input", "output" or "error".
919 ** New way to temporarily set fluids
921 C code can now use scm_frame_fluid in a 'frame' (see
922 above) to temporarily set the value of a fluid.
924 ** New types scm_t_intmax and scm_t_uintmax.
926 On platforms that have them, these types are identical to intmax_t and
927 uintmax_t, respectively. On other platforms, they are identical to
928 the largest integer types that Guile knows about.
930 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
932 You should not have used them.
934 ** Many public #defines with generic names have been made private.
936 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
937 private or renamed with a more suitable public name.
939 ** The macro SCM_TYP16S has been deprecated.
941 This macro is not intended for public use.
943 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
945 Use scm_is_true (scm_inexact_p (...)) instead.
947 ** The macro SCM_SLOPPY_REALP has been deprecated.
949 Use scm_is_real instead.
951 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
953 Use scm_is_complex instead.
955 ** Some preprocessor defines have been deprecated.
957 These defines indicated whether a certain feature was present in Guile
958 or not. Going forward, assume that the features are always present.
960 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
961 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
963 The following macros have been removed completely: MEMOIZE_LOCALS,
964 SCM_RECKLESS, SCM_CAUTIOUS.
966 ** The preprocessor define STACK_DIRECTION has been deprecated.
968 There should be no need to know about the stack direction for ordinary
971 ** New function: scm_effective_version
973 Returns the "effective" version number. This is just the normal full
974 version string without the final micro-version number. See "Changes
975 to the distribution" above.
977 ** The function scm_call_with_new_thread has a new prototype.
979 Instead of taking a list with the thunk and handler, these two
980 arguments are now passed directly:
982 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
984 This is an incompatible change.
986 ** The value 'scm_mask_ints' is no longer writable.
988 Previously, you could set scm_mask_ints directly. This is no longer
989 possible. Use scm_c_call_with_blocked_asyncs and
990 scm_c_call_with_unblocked_asyncs instead.
992 ** New functions scm_c_call_with_blocked_asyncs and
993 scm_c_call_with_unblocked_asyncs
995 Like scm_call_with_blocked_asyncs etc. but for C functions.
997 ** New snarfer macro SCM_DEFINE_PUBLIC.
999 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1000 function in the init section.
1002 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1004 ** New macros SCM_VECTOR_REF and SCM_VECTOR_SET.
1006 Use these in preference to SCM_VELTS.
1008 ** The SCM_VELTS macros now returns a read-only vector. For writing,
1009 use the new macros SCM_WRITABLE_VELTS or SCM_VECTOR_SET. The use of
1010 SCM_WRITABLE_VELTS is discouraged, though.
1012 ** Garbage collector rewrite.
1014 The garbage collector is cleaned up a lot, and now uses lazy
1015 sweeping. This is reflected in the output of (gc-stats); since cells
1016 are being freed when they are allocated, the cells-allocated field
1017 stays roughly constant.
1019 For malloc related triggers, the behavior is changed. It uses the same
1020 heuristic as the cell-triggered collections. It may be tuned with the
1021 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1022 for minimum yield of malloc related triggers. The default is 40.
1023 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1026 Debugging operations for the freelist have been deprecated, along with
1027 the C variables that control garbage collection. The environment
1028 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1029 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1031 ** The function scm_definedp has been renamed to scm_defined_p
1033 The name scm_definedp is deprecated.
1035 ** The struct scm_cell type has been renamed to scm_t_cell
1037 This is in accordance to Guile's naming scheme for types. Note that
1038 the name scm_cell is now used for a function that allocates and
1039 initializes a new cell (see below).
1041 ** New functions for memory management
1043 A new set of functions for memory management has been added since the
1044 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1045 indeed, Guile itself contained some long standing bugs that could
1046 cause aborts in long running programs.
1048 The new functions are more symmetrical and do not need cooperation
1049 from smob free routines, among other improvements.
1051 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1052 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1053 scm_gc_free, scm_gc_register_collectable_memory, and
1054 scm_gc_unregister_collectable_memory. Refer to the manual for more
1055 details and for upgrading instructions.
1057 The old functions for memory management have been deprecated. They
1058 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1059 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1061 ** Declarations of exported features are marked with SCM_API.
1063 Every declaration of a feature that belongs to the exported Guile API
1064 has been marked by adding the macro "SCM_API" to the start of the
1065 declaration. This macro can expand into different things, the most
1066 common of which is just "extern" for Unix platforms. On Win32, it can
1067 be used to control which symbols are exported from a DLL.
1069 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1070 will expand into "__declspec (dllimport) extern", which is needed for
1071 linking to the Guile DLL in Windows.
1073 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1074 SCM_SRFI4_IMPORT, for the corresponding libraries.
1076 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1078 Use the new functions scm_cell and scm_double_cell instead. The old
1079 macros had problems because with them allocation and initialization
1080 was separated and the GC could sometimes observe half initialized
1081 cells. Only careful coding by the user of SCM_NEWCELL and
1082 SCM_NEWCELL2 could make this safe and efficient.
1084 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1086 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1089 ** SRCBRKP has been deprecated.
1091 Use scm_c_source_property_breakpoint_p instead.
1093 ** Deprecated: scm_makmacro
1095 Change your code to use either scm_makmmacro or to define macros in
1096 Scheme, using 'define-macro'.
1098 ** New function scm_c_port_for_each.
1100 This function is like scm_port_for_each but takes a pointer to a C
1101 function as the callback instead of a SCM value.
1103 ** Many definitions have been removed that were previously deprecated.
1105 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1106 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1107 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1108 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1109 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1110 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1111 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1112 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1113 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1114 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1115 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1116 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1117 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1118 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1119 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1120 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1121 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1122 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1123 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1124 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1125 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1126 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1127 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1128 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1129 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1130 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1131 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1132 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1133 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1134 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1135 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1138 Changes since Guile 1.4:
1140 * Changes to the distribution
1142 ** A top-level TODO file is included.
1144 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1146 Guile now always uses three numbers to represent the version,
1147 i.e. "1.6.5". The first number, 1, is the major version number, the
1148 second number, 6, is the minor version number, and the third number,
1149 5, is the micro version number. Changes in major version number
1150 indicate major changes in Guile.
1152 Minor version numbers that are even denote stable releases, and odd
1153 minor version numbers denote development versions (which may be
1154 unstable). The micro version number indicates a minor sub-revision of
1155 a given MAJOR.MINOR release.
1157 In keeping with the new scheme, (minor-version) and scm_minor_version
1158 no longer return everything but the major version number. They now
1159 just return the minor version number. Two new functions
1160 (micro-version) and scm_micro_version have been added to report the
1161 micro version number.
1163 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1165 ** New preprocessor definitions are available for checking versions.
1167 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1168 SCM_MICRO_VERSION to the appropriate integer values.
1170 ** Guile now actively warns about deprecated features.
1172 The new configure option `--enable-deprecated=LEVEL' and the
1173 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1174 See INSTALL and README for more information.
1176 ** Guile is much more likely to work on 64-bit architectures.
1178 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1179 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1180 for the use of a test machine, and thanks to Stefan Jahn for ia64
1183 ** New functions: setitimer and getitimer.
1185 These implement a fairly direct interface to the libc functions of the
1188 ** The #. reader extension is now disabled by default.
1190 For safety reasons, #. evaluation is disabled by default. To
1191 re-enable it, set the fluid read-eval? to #t. For example:
1193 (fluid-set! read-eval? #t)
1195 but make sure you realize the potential security risks involved. With
1196 read-eval? enabled, reading a data file from an untrusted source can
1199 ** New SRFI modules have been added:
1201 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1204 (srfi srfi-1) is a library containing many useful pair- and list-processing
1207 (srfi srfi-2) exports and-let*.
1209 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1211 (srfi srfi-6) is a dummy module for now, since guile already provides
1212 all of the srfi-6 procedures by default: open-input-string,
1213 open-output-string, get-output-string.
1215 (srfi srfi-8) exports receive.
1217 (srfi srfi-9) exports define-record-type.
1219 (srfi srfi-10) exports define-reader-ctor and implements the reader
1222 (srfi srfi-11) exports let-values and let*-values.
1224 (srfi srfi-13) implements the SRFI String Library.
1226 (srfi srfi-14) implements the SRFI Character-Set Library.
1228 (srfi srfi-17) implements setter and getter-with-setter and redefines
1229 some accessor procedures as procedures with getters. (such as car,
1230 cdr, vector-ref etc.)
1232 (srfi srfi-19) implements the SRFI Time/Date Library.
1234 ** New scripts / "executable modules"
1236 Subdirectory "scripts" contains Scheme modules that are packaged to
1237 also be executable as scripts. At this time, these scripts are available:
1246 See README there for more info.
1248 These scripts can be invoked from the shell with the new program
1249 "guile-tools", which keeps track of installation directory for you.
1252 $ guile-tools display-commentary srfi/*.scm
1254 guile-tools is copied to the standard $bindir on "make install".
1256 ** New module (ice-9 stack-catch):
1258 stack-catch is like catch, but saves the current state of the stack in
1259 the fluid the-last-stack. This fluid can be useful when using the
1260 debugger and when re-throwing an error.
1262 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1264 This has been done to prevent problems on lesser operating systems
1265 that can't tolerate `*'s in file names. The exported macro continues
1266 to be named `and-let*', of course.
1268 On systems that support it, there is also a compatibility module named
1269 (ice-9 and-let*). It will go away in the next release.
1271 ** New modules (oop goops) etc.:
1274 (oop goops describe)
1276 (oop goops active-slot)
1277 (oop goops composite-slot)
1279 The Guile Object Oriented Programming System (GOOPS) has been
1280 integrated into Guile. For further information, consult the GOOPS
1281 manual and tutorial in the `doc' directory.
1283 ** New module (ice-9 rdelim).
1285 This exports the following procedures which were previously defined
1286 in the default environment:
1288 read-line read-line! read-delimited read-delimited! %read-delimited!
1289 %read-line write-line
1291 For backwards compatibility the definitions are still imported into the
1292 default environment in this version of Guile. However you should add:
1294 (use-modules (ice-9 rdelim))
1296 to any program which uses the definitions, since this may change in
1299 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1300 can be used for similar functionality.
1302 ** New module (ice-9 rw)
1304 This is a subset of the (scsh rw) module from guile-scsh. Currently
1305 it defines two procedures:
1307 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1309 Read characters from a port or file descriptor into a string STR.
1310 A port must have an underlying file descriptor -- a so-called
1311 fport. This procedure is scsh-compatible and can efficiently read
1314 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1316 Write characters from a string STR to a port or file descriptor.
1317 A port must have an underlying file descriptor -- a so-called
1318 fport. This procedure is mostly compatible and can efficiently
1319 write large strings.
1321 ** New module (ice-9 match)
1323 This module includes Andrew K. Wright's pattern matcher. See
1324 ice-9/match.scm for brief description or
1326 http://www.star-lab.com/wright/code.html
1328 for complete documentation.
1330 ** New module (ice-9 buffered-input)
1332 This module provides procedures to construct an input port from an
1333 underlying source of input that reads and returns its input in chunks.
1334 The underlying input source is a Scheme procedure, specified by the
1335 caller, which the port invokes whenever it needs more input.
1337 This is useful when building an input port whose back end is Readline
1338 or a UI element such as the GtkEntry widget.
1342 The reference and tutorial documentation that was previously
1343 distributed separately, as `guile-doc', is now included in the core
1344 Guile distribution. The documentation consists of the following
1347 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1350 - The Guile Reference Manual (guile.texi) contains (or is intended to
1351 contain) reference documentation on all aspects of Guile.
1353 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1354 reference documentation for using GOOPS, Guile's Object Oriented
1357 - The Revised^5 Report on the Algorithmic Language Scheme
1360 See the README file in the `doc' directory for more details.
1362 ** There are a couple of examples in the examples/ directory now.
1364 * Changes to the stand-alone interpreter
1366 ** New command line option `--use-srfi'
1368 Using this option, SRFI modules can be loaded on startup and be
1369 available right from the beginning. This makes programming portable
1370 Scheme programs easier.
1372 The option `--use-srfi' expects a comma-separated list of numbers,
1373 each representing a SRFI number to be loaded into the interpreter
1374 before starting evaluating a script file or the REPL. Additionally,
1375 the feature identifier for the loaded SRFIs is recognized by
1376 `cond-expand' when using this option.
1379 $ guile --use-srfi=8,13
1380 guile> (receive (x z) (values 1 2) (+ 1 2))
1382 guile> (string-pad "bla" 20)
1385 ** Guile now always starts up in the `(guile-user)' module.
1387 Previously, scripts executed via the `-s' option would run in the
1388 `(guile)' module and the repl would run in the `(guile-user)' module.
1389 Now every user action takes place in the `(guile-user)' module by
1392 * Changes to Scheme functions and syntax
1394 ** Character classifiers work for non-ASCII characters.
1396 The predicates `char-alphabetic?', `char-numeric?',
1397 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1398 no longer check whether their arguments are ASCII characters.
1399 Previously, a character would only be considered alphabetic when it
1400 was also ASCII, for example.
1402 ** Previously deprecated Scheme functions have been removed:
1404 tag - no replacement.
1405 fseek - replaced by seek.
1406 list* - replaced by cons*.
1408 ** It's now possible to create modules with controlled environments
1412 (use-modules (ice-9 safe))
1413 (define m (make-safe-module))
1414 ;;; m will now be a module containing only a safe subset of R5RS
1415 (eval '(+ 1 2) m) --> 3
1416 (eval 'load m) --> ERROR: Unbound variable: load
1418 ** Evaluation of "()", the empty list, is now an error.
1420 Previously, the expression "()" evaluated to the empty list. This has
1421 been changed to signal a "missing expression" error. The correct way
1422 to write the empty list as a literal constant is to use quote: "'()".
1424 ** New concept of `Guile Extensions'.
1426 A Guile Extension is just a ordinary shared library that can be linked
1427 at run-time. We found it advantageous to give this simple concept a
1428 dedicated name to distinguish the issues related to shared libraries
1429 from the issues related to the module system.
1431 *** New function: load-extension
1433 Executing (load-extension lib init) is mostly equivalent to
1435 (dynamic-call init (dynamic-link lib))
1437 except when scm_register_extension has been called previously.
1438 Whenever appropriate, you should use `load-extension' instead of
1439 dynamic-link and dynamic-call.
1441 *** New C function: scm_c_register_extension
1443 This function registers a initialization function for use by
1444 `load-extension'. Use it when you don't want specific extensions to
1445 be loaded as shared libraries (for example on platforms that don't
1446 support dynamic linking).
1448 ** Auto-loading of compiled-code modules is deprecated.
1450 Guile used to be able to automatically find and link a shared
1451 library to satisfy requests for a module. For example, the module
1452 `(foo bar)' could be implemented by placing a shared library named
1453 "foo/libbar.so" (or with a different extension) in a directory on the
1456 This has been found to be too tricky, and is no longer supported. The
1457 shared libraries are now called "extensions". You should now write a
1458 small Scheme file that calls `load-extension' to load the shared
1459 library and initialize it explicitely.
1461 The shared libraries themselves should be installed in the usual
1462 places for shared libraries, with names like "libguile-foo-bar".
1464 For example, place this into a file "foo/bar.scm"
1466 (define-module (foo bar))
1468 (load-extension "libguile-foo-bar" "foobar_init")
1470 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1472 `eval' is now R5RS, that is it takes two arguments.
1473 The second argument is an environment specifier, i.e. either
1475 (scheme-report-environment 5)
1476 (null-environment 5)
1477 (interaction-environment)
1483 ** The module system has been made more disciplined.
1485 The function `eval' will save and restore the current module around
1486 the evaluation of the specified expression. While this expression is
1487 evaluated, `(current-module)' will now return the right module, which
1488 is the module specified as the second argument to `eval'.
1490 A consequence of this change is that `eval' is not particularly
1491 useful when you want allow the evaluated code to change what module is
1492 designated as the current module and have this change persist from one
1493 call to `eval' to the next. The read-eval-print-loop is an example
1494 where `eval' is now inadequate. To compensate, there is a new
1495 function `primitive-eval' that does not take a module specifier and
1496 that does not save/restore the current module. You should use this
1497 function together with `set-current-module', `current-module', etc
1498 when you want to have more control over the state that is carried from
1499 one eval to the next.
1501 Additionally, it has been made sure that forms that are evaluated at
1502 the top level are always evaluated with respect to the current module.
1503 Previously, subforms of top-level forms such as `begin', `case',
1504 etc. did not respect changes to the current module although these
1505 subforms are at the top-level as well.
1507 To prevent strange behavior, the forms `define-module',
1508 `use-modules', `use-syntax', and `export' have been restricted to only
1509 work on the top level. The forms `define-public' and
1510 `defmacro-public' only export the new binding on the top level. They
1511 behave just like `define' and `defmacro', respectively, when they are
1512 used in a lexical environment.
1514 Also, `export' will no longer silently re-export bindings imported
1515 from a used module. It will emit a `deprecation' warning and will
1516 cease to perform any re-export in the next version. If you actually
1517 want to re-export bindings, use the new `re-export' in place of
1518 `export'. The new `re-export' will not make copies of variables when
1519 rexporting them, as `export' did wrongly.
1521 ** Module system now allows selection and renaming of imported bindings
1523 Previously, when using `use-modules' or the `#:use-module' clause in
1524 the `define-module' form, all the bindings (association of symbols to
1525 values) for imported modules were added to the "current module" on an
1526 as-is basis. This has been changed to allow finer control through two
1527 new facilities: selection and renaming.
1529 You can now select which of the imported module's bindings are to be
1530 visible in the current module by using the `:select' clause. This
1531 clause also can be used to rename individual bindings. For example:
1533 ;; import all bindings no questions asked
1534 (use-modules (ice-9 common-list))
1536 ;; import four bindings, renaming two of them;
1537 ;; the current module sees: every some zonk-y zonk-n
1538 (use-modules ((ice-9 common-list)
1540 (remove-if . zonk-y)
1541 (remove-if-not . zonk-n))))
1543 You can also programmatically rename all selected bindings using the
1544 `:renamer' clause, which specifies a proc that takes a symbol and
1545 returns another symbol. Because it is common practice to use a prefix,
1546 we now provide the convenience procedure `symbol-prefix-proc'. For
1549 ;; import four bindings, renaming two of them specifically,
1550 ;; and all four w/ prefix "CL:";
1551 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1552 (use-modules ((ice-9 common-list)
1554 (remove-if . zonk-y)
1555 (remove-if-not . zonk-n))
1556 :renamer (symbol-prefix-proc 'CL:)))
1558 ;; import four bindings, renaming two of them specifically,
1559 ;; and all four by upcasing.
1560 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1561 (define (upcase-symbol sym)
1562 (string->symbol (string-upcase (symbol->string sym))))
1564 (use-modules ((ice-9 common-list)
1566 (remove-if . zonk-y)
1567 (remove-if-not . zonk-n))
1568 :renamer upcase-symbol))
1570 Note that programmatic renaming is done *after* individual renaming.
1571 Also, the above examples show `use-modules', but the same facilities are
1572 available for the `#:use-module' clause of `define-module'.
1574 See manual for more info.
1576 ** The semantics of guardians have changed.
1578 The changes are for the most part compatible. An important criterion
1579 was to keep the typical usage of guardians as simple as before, but to
1580 make the semantics safer and (as a result) more useful.
1582 *** All objects returned from guardians are now properly alive.
1584 It is now guaranteed that any object referenced by an object returned
1585 from a guardian is alive. It's now impossible for a guardian to
1586 return a "contained" object before its "containing" object.
1588 One incompatible (but probably not very important) change resulting
1589 from this is that it is no longer possible to guard objects that
1590 indirectly reference themselves (i.e. are parts of cycles). If you do
1591 so accidentally, you'll get a warning.
1593 *** There are now two types of guardians: greedy and sharing.
1595 If you call (make-guardian #t) or just (make-guardian), you'll get a
1596 greedy guardian, and for (make-guardian #f) a sharing guardian.
1598 Greedy guardians are the default because they are more "defensive".
1599 You can only greedily guard an object once. If you guard an object
1600 more than once, once in a greedy guardian and the rest of times in
1601 sharing guardians, then it is guaranteed that the object won't be
1602 returned from sharing guardians as long as it is greedily guarded
1605 Guardians returned by calls to `make-guardian' can now take one more
1606 optional parameter, which says whether to throw an error in case an
1607 attempt is made to greedily guard an object that is already greedily
1608 guarded. The default is true, i.e. throw an error. If the parameter
1609 is false, the guardian invocation returns #t if guarding was
1610 successful and #f if it wasn't.
1612 Also, since greedy guarding is, in effect, a side-effecting operation
1613 on objects, a new function is introduced: `destroy-guardian!'.
1614 Invoking this function on a guardian renders it unoperative and, if
1615 the guardian is greedy, clears the "greedily guarded" property of the
1616 objects that were guarded by it, thus undoing the side effect.
1618 Note that all this hair is hardly very important, since guardian
1619 objects are usually permanent.
1621 ** Continuations created by call-with-current-continuation now accept
1622 any number of arguments, as required by R5RS.
1624 ** New function `issue-deprecation-warning'
1626 This function is used to display the deprecation messages that are
1627 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1630 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1634 ;; `id' is deprecated. Use `identity' instead.
1639 ** New syntax `begin-deprecated'
1641 When deprecated features are included (as determined by the configure
1642 option --enable-deprecated), `begin-deprecated' is identical to
1643 `begin'. When deprecated features are excluded, it always evaluates
1644 to `#f', ignoring the body forms.
1646 ** New function `make-object-property'
1648 This function returns a new `procedure with setter' P that can be used
1649 to attach a property to objects. When calling P as
1653 where `obj' is any kind of object, it attaches `val' to `obj' in such
1654 a way that it can be retrieved by calling P as
1658 This function will replace procedure properties, symbol properties and
1659 source properties eventually.
1661 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1663 Instead of #&optional, #&key, etc you should now use #:optional,
1664 #:key, etc. Since #:optional is a keyword, you can write it as just
1665 :optional when (read-set! keywords 'prefix) is active.
1667 The old reader syntax `#&' is still supported, but deprecated. It
1668 will be removed in the next release.
1670 ** New define-module option: pure
1672 Tells the module system not to include any bindings from the root
1677 (define-module (totally-empty-module)
1680 ** New define-module option: export NAME1 ...
1682 Export names NAME1 ...
1684 This option is required if you want to be able to export bindings from
1685 a module which doesn't import one of `define-public' or `export'.
1689 (define-module (foo)
1691 :use-module (ice-9 r5rs)
1694 ;;; Note that we're pure R5RS below this point!
1699 ** New function: object->string OBJ
1701 Return a Scheme string obtained by printing a given object.
1703 ** New function: port? X
1705 Returns a boolean indicating whether X is a port. Equivalent to
1706 `(or (input-port? X) (output-port? X))'.
1708 ** New function: file-port?
1710 Determines whether a given object is a port that is related to a file.
1712 ** New function: port-for-each proc
1714 Apply PROC to each port in the Guile port table in turn. The return
1715 value is unspecified. More specifically, PROC is applied exactly once
1716 to every port that exists in the system at the time PORT-FOR-EACH is
1717 invoked. Changes to the port table while PORT-FOR-EACH is running
1718 have no effect as far as PORT-FOR-EACH is concerned.
1720 ** New function: dup2 oldfd newfd
1722 A simple wrapper for the `dup2' system call. Copies the file
1723 descriptor OLDFD to descriptor number NEWFD, replacing the
1724 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1725 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1726 to move away ports which are using NEWFD. The return value is
1729 ** New function: close-fdes fd
1731 A simple wrapper for the `close' system call. Close file
1732 descriptor FD, which must be an integer. Unlike close (*note
1733 close: Ports and File Descriptors.), the file descriptor will be
1734 closed even if a port is using it. The return value is
1737 ** New function: crypt password salt
1739 Encrypts `password' using the standard unix password encryption
1742 ** New function: chroot path
1744 Change the root directory of the running process to `path'.
1746 ** New functions: getlogin, cuserid
1748 Return the login name or the user name of the current effective user
1751 ** New functions: getpriority which who, setpriority which who prio
1753 Get or set the priority of the running process.
1755 ** New function: getpass prompt
1757 Read a password from the terminal, first displaying `prompt' and
1760 ** New function: flock file operation
1762 Set/remove an advisory shared or exclusive lock on `file'.
1764 ** New functions: sethostname name, gethostname
1766 Set or get the hostname of the machine the current process is running
1769 ** New function: mkstemp! tmpl
1771 mkstemp creates a new unique file in the file system and returns a
1772 new buffered port open for reading and writing to the file. TMPL
1773 is a string specifying where the file should be created: it must
1774 end with `XXXXXX' and will be changed in place to return the name
1775 of the temporary file.
1777 ** New function: open-input-string string
1779 Return an input string port which delivers the characters from
1780 `string'. This procedure, together with `open-output-string' and
1781 `get-output-string' implements SRFI-6.
1783 ** New function: open-output-string
1785 Return an output string port which collects all data written to it.
1786 The data can then be retrieved by `get-output-string'.
1788 ** New function: get-output-string
1790 Return the contents of an output string port.
1792 ** New function: identity
1794 Return the argument.
1796 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1797 are represented in Scheme as integers with normal host byte ordering.
1799 ** New function: inet-pton family address
1801 Convert a printable string network address into an integer. Note that
1802 unlike the C version of this function, the result is an integer with
1803 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1806 (inet-pton AF_INET "127.0.0.1") => 2130706433
1807 (inet-pton AF_INET6 "::1") => 1
1809 ** New function: inet-ntop family address
1811 Convert an integer network address into a printable string. Note that
1812 unlike the C version of this function, the input is an integer with
1813 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1816 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1817 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1818 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1822 Use `identity' instead.
1828 ** Deprecated: return-it
1832 ** Deprecated: string-character-length
1834 Use `string-length' instead.
1836 ** Deprecated: flags
1838 Use `logior' instead.
1840 ** Deprecated: close-all-ports-except.
1842 This was intended for closing ports in a child process after a fork,
1843 but it has the undesirable side effect of flushing buffers.
1844 port-for-each is more flexible.
1846 ** The (ice-9 popen) module now attempts to set up file descriptors in
1847 the child process from the current Scheme ports, instead of using the
1848 current values of file descriptors 0, 1, and 2 in the parent process.
1850 ** Removed function: builtin-weak-bindings
1852 There is no such concept as a weak binding any more.
1854 ** Removed constants: bignum-radix, scm-line-incrementors
1856 ** define-method: New syntax mandatory.
1858 The new method syntax is now mandatory:
1860 (define-method (NAME ARG-SPEC ...) BODY ...)
1861 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1863 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1864 REST-ARG ::= ARG-NAME
1866 If you have old code using the old syntax, import
1867 (oop goops old-define-method) before (oop goops) as in:
1869 (use-modules (oop goops old-define-method) (oop goops))
1871 ** Deprecated function: builtin-variable
1872 Removed function: builtin-bindings
1874 There is no longer a distinction between builtin or other variables.
1875 Use module system operations for all variables.
1877 ** Lazy-catch handlers are no longer allowed to return.
1879 That is, a call to `throw', `error', etc is now guaranteed to not
1882 ** Bugfixes for (ice-9 getopt-long)
1884 This module is now tested using test-suite/tests/getopt-long.test.
1885 The following bugs have been fixed:
1887 *** Parsing for options that are specified to have `optional' args now checks
1888 if the next element is an option instead of unconditionally taking it as the
1891 *** An error is now thrown for `--opt=val' when the option description
1892 does not specify `(value #t)' or `(value optional)'. This condition used to
1893 be accepted w/o error, contrary to the documentation.
1895 *** The error message for unrecognized options is now more informative.
1896 It used to be "not a record", an artifact of the implementation.
1898 *** The error message for `--opt' terminating the arg list (no value), when
1899 `(value #t)' is specified, is now more informative. It used to be "not enough
1902 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1903 The expansion used to be like so:
1905 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1907 Note that the "5d" is dropped. Now it is like so:
1909 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1911 This enables single-char options to have adjoining arguments as long as their
1912 constituent characters are not potential single-char options.
1914 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1916 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1917 property `arglist', which can be retrieved by `arity'. The result is that
1918 `arity' can give more detailed information than before:
1922 guile> (use-modules (ice-9 optargs))
1923 guile> (define* (foo #:optional a b c) a)
1925 0 or more arguments in `lambda*:G0'.
1930 3 optional arguments: `a', `b' and `c'.
1931 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
1933 2 required arguments: `a' and `b', 2 keyword arguments: `c'
1934 and `d', other keywords allowed.
1935 guile> (define* (baz a b #:optional c #:rest r) a)
1937 2 required arguments: `a' and `b', 1 optional argument: `c',
1940 * Changes to the C interface
1942 ** Types have been renamed from scm_*_t to scm_t_*.
1944 This has been done for POSIX sake. It reserves identifiers ending
1945 with "_t". What a concept.
1947 The old names are still available with status `deprecated'.
1949 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
1951 ** Deprecated features have been removed.
1955 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
1956 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
1958 *** C Functions removed
1960 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
1961 scm_fseek - replaced by scm_seek.
1962 gc-thunk - replaced by after-gc-hook.
1963 gh_int2scmb - replaced by gh_bool2scm.
1964 scm_tc_dblr - replaced by scm_tc16_real.
1965 scm_tc_dblc - replaced by scm_tc16_complex.
1966 scm_list_star - replaced by scm_cons_star.
1968 ** Deprecated: scm_makfromstr
1970 Use scm_mem2string instead.
1972 ** Deprecated: scm_make_shared_substring
1974 Explicit shared substrings will disappear from Guile.
1976 Instead, "normal" strings will be implemented using sharing
1977 internally, combined with a copy-on-write strategy.
1979 ** Deprecated: scm_read_only_string_p
1981 The concept of read-only strings will disappear in next release of
1984 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
1986 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
1988 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
1990 Call a procedure with the indicated number of arguments. See "Fly
1991 Evaluation" in the manual.
1993 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
1995 Call a procedure with the indicated number of arguments and a list of
1996 further arguments. See "Fly Evaluation" in the manual.
1998 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2000 Create a list of the given number of elements. See "List
2001 Constructors" in the manual.
2003 ** Renamed function: scm_listify has been replaced by scm_list_n.
2005 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2006 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2008 Use functions scm_list_N instead.
2010 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2012 Used by an application to read arbitrary number of bytes from a port.
2013 Same semantics as libc read, except that scm_c_read only returns less
2014 than SIZE bytes if at end-of-file.
2016 Warning: Doesn't update port line and column counts!
2018 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2020 Used by an application to write arbitrary number of bytes to an SCM
2021 port. Similar semantics as libc write. However, unlike libc
2022 write, scm_c_write writes the requested number of bytes and has no
2025 Warning: Doesn't update port line and column counts!
2027 ** New function: scm_init_guile ()
2029 In contrast to scm_boot_guile, scm_init_guile will return normally
2030 after initializing Guile. It is not available on all systems, tho.
2032 ** New functions: scm_str2symbol, scm_mem2symbol
2034 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2035 field of characters and creates a scheme symbol object from that C string.
2036 The function scm_mem2symbol takes a const char* and a number of characters and
2037 creates a symbol from the characters in that memory area.
2039 ** New functions: scm_primitive_make_property
2040 scm_primitive_property_ref
2041 scm_primitive_property_set_x
2042 scm_primitive_property_del_x
2044 These functions implement a new way to deal with object properties.
2045 See libguile/properties.c for their documentation.
2047 ** New function: scm_done_free (long size)
2049 This function is the inverse of scm_done_malloc. Use it to report the
2050 amount of smob memory you free. The previous method, which involved
2051 calling scm_done_malloc with negative argument, was somewhat
2052 unintuitive (and is still available, of course).
2054 ** New function: scm_c_memq (SCM obj, SCM list)
2056 This function provides a fast C level alternative for scm_memq for the case
2057 that the list parameter is known to be a proper list. The function is a
2058 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2059 list input parameter, since for anything else but a proper list the function's
2060 behaviour is undefined - it may even crash or loop endlessly. Further, for
2061 the case that the object is not found in the list, scm_c_memq returns #f which
2062 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2064 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2065 scm_remember_upto_here
2067 These functions replace the function scm_remember.
2069 ** Deprecated function: scm_remember
2071 Use one of the new functions scm_remember_upto_here_1,
2072 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2074 ** New function: scm_allocate_string
2076 This function replaces the function scm_makstr.
2078 ** Deprecated function: scm_makstr
2080 Use the new function scm_allocate_string instead.
2082 ** New global variable scm_gc_running_p introduced.
2084 Use this variable to find out if garbage collection is being executed. Up to
2085 now applications have used scm_gc_heap_lock to test if garbage collection was
2086 running, which also works because of the fact that up to know only the garbage
2087 collector has set this variable. But, this is an implementation detail that
2088 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2089 of this variable is (and has been) not fully safe anyway.
2091 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2093 Use these instead of SCM_LENGTH_MAX.
2095 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2096 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2097 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2099 Use these instead of SCM_LENGTH.
2101 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2102 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2103 SCM_SET_BITVECTOR_LENGTH
2105 Use these instead of SCM_SETLENGTH
2107 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2108 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2111 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2114 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2115 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2118 Use these instead of SCM_SETCHARS.
2120 ** New macro: SCM_BITVECTOR_P
2122 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2124 Use instead of SCM_COERCE_SUBSTR.
2126 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2128 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2130 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2131 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2132 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2133 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2134 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2135 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2136 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2137 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2138 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2139 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2140 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2141 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2142 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2143 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2144 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2146 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2147 Use scm_memory_error instead of SCM_NALLOC.
2148 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2149 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2150 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2151 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2152 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2153 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2154 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2155 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2156 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2157 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2158 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2159 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2160 Use a type specific setter macro instead of SCM_SETLENGTH.
2161 Use a type specific setter macro instead of SCM_SETCHARS.
2162 Use a type specific length macro instead of SCM_LENGTH_MAX.
2163 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2164 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2165 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2166 Use SCM_TYP16 instead of SCM_GCTYP16.
2167 Use SCM_CDR instead of SCM_GCCDR.
2168 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2169 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2170 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2171 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2172 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2173 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2174 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2176 ** Removed function: scm_struct_init
2178 ** Removed variable: scm_symhash_dim
2180 ** Renamed function: scm_make_cont has been replaced by
2181 scm_make_continuation, which has a different interface.
2183 ** Deprecated function: scm_call_catching_errors
2185 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2187 ** Deprecated function: scm_strhash
2189 Use scm_string_hash instead.
2191 ** Deprecated function: scm_vector_set_length_x
2193 Instead, create a fresh vector of the desired size and copy the contents.
2195 ** scm_gensym has changed prototype
2197 scm_gensym now only takes one argument.
2199 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2202 There is now only a single symbol type scm_tc7_symbol.
2203 The tag scm_tc7_lvector was not used anyway.
2205 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2207 Use scm_make_smob_type and scm_set_smob_XXX instead.
2209 ** New function scm_set_smob_apply.
2211 This can be used to set an apply function to a smob type.
2213 ** Deprecated function: scm_strprint_obj
2215 Use scm_object_to_string instead.
2217 ** Deprecated function: scm_wta
2219 Use scm_wrong_type_arg, or another appropriate error signalling function
2222 ** Explicit support for obarrays has been deprecated.
2224 Use `scm_str2symbol' and the generic hashtable functions instead.
2226 ** The concept of `vcells' has been deprecated.
2228 The data type `variable' is now used exclusively. `Vcells' have been
2229 a low-level concept so you are likely not affected by this change.
2231 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2232 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2234 Use scm_c_define or scm_c_lookup instead, as appropriate.
2236 *** New functions: scm_c_module_lookup, scm_c_lookup,
2237 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2238 scm_module_define, scm_define.
2240 These functions work with variables instead of with vcells.
2242 ** New functions for creating and defining `subr's and `gsubr's.
2244 The new functions more clearly distinguish between creating a subr (or
2245 gsubr) object and adding it to the current module.
2247 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2248 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2249 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2250 scm_c_define_gsubr_with_generic.
2252 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2253 scm_make_subr_with_generic, scm_make_gsubr,
2254 scm_make_gsubr_with_generic.
2256 Use the new ones from above instead.
2258 ** C interface to the module system has changed.
2260 While we suggest that you avoid as many explicit module system
2261 operations from C as possible for the time being, the C interface has
2262 been made more similar to the high-level Scheme module system.
2264 *** New functions: scm_c_define_module, scm_c_use_module,
2265 scm_c_export, scm_c_resolve_module.
2267 They mostly work like their Scheme namesakes. scm_c_define_module
2268 takes a function that is called a context where the new module is
2271 *** Deprecated functions: scm_the_root_module, scm_make_module,
2272 scm_ensure_user_module, scm_load_scheme_module.
2274 Use the new functions instead.
2276 ** Renamed function: scm_internal_with_fluids becomes
2279 scm_internal_with_fluids is available as a deprecated function.
2281 ** New function: scm_c_with_fluid.
2283 Just like scm_c_with_fluids, but takes one fluid and one value instead
2286 ** Deprecated typedefs: long_long, ulong_long.
2288 They are of questionable utility and they pollute the global
2291 ** Deprecated typedef: scm_sizet
2293 It is of questionable utility now that Guile requires ANSI C, and is
2296 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2297 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2298 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2300 Made more compliant with the naming policy by adding a _t at the end.
2302 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2303 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2305 With the exception of the mysterious scm_2ulong2big, they are still
2306 available under new names (scm_i_mkbig etc). These functions are not
2307 intended to be used in user code. You should avoid dealing with
2308 bignums directly, and should deal with numbers in general (which can
2311 ** Change in behavior: scm_num2long, scm_num2ulong
2313 The scm_num2[u]long functions don't any longer accept an inexact
2314 argument. This change in behavior is motivated by concordance with
2315 R5RS: It is more common that a primitive doesn't want to accept an
2316 inexact for an exact.
2318 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2319 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2320 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2323 These are conversion functions between the various ANSI C integral
2324 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2325 accept an inexact argument.
2327 ** New functions: scm_float2num, scm_double2num,
2328 scm_num2float, scm_num2double.
2330 These are conversion functions between the two ANSI C float types and
2333 ** New number validation macros:
2334 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2338 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2340 These are just nicer-named old scm_protect_object and
2341 scm_unprotect_object.
2343 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2345 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2347 These functions can be used to register pointers to locations that
2350 ** Deprecated function: scm_create_hook.
2352 Its sins are: misleading name, non-modularity and lack of general
2356 Changes since Guile 1.3.4:
2358 * Changes to the distribution
2360 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2362 We've changed the way we handle generated files in the Guile source
2363 repository. As a result, the procedure for building trees obtained
2364 from the nightly FTP snapshots or via CVS has changed:
2365 - You must have appropriate versions of autoconf, automake, and
2366 libtool installed on your system. See README for info on how to
2367 obtain these programs.
2368 - Before configuring the tree, you must first run the script
2369 `autogen.sh' at the top of the source tree.
2371 The Guile repository used to contain not only source files, written by
2372 humans, but also some generated files, like configure scripts and
2373 Makefile.in files. Even though the contents of these files could be
2374 derived mechanically from other files present, we thought it would
2375 make the tree easier to build if we checked them into CVS.
2377 However, this approach means that minor differences between
2378 developer's installed tools and habits affected the whole team.
2379 So we have removed the generated files from the repository, and
2380 added the autogen.sh script, which will reconstruct them
2384 ** configure now has experimental options to remove support for certain
2387 --disable-arrays omit array and uniform array support
2388 --disable-posix omit posix interfaces
2389 --disable-networking omit networking interfaces
2390 --disable-regex omit regular expression interfaces
2392 These are likely to become separate modules some day.
2394 ** New configure option --enable-debug-freelist
2396 This enables a debugging version of SCM_NEWCELL(), and also registers
2397 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2399 Configure with the --enable-debug-freelist option to enable
2400 the gc-set-debug-check-freelist! primitive, and then use:
2402 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2403 (gc-set-debug-check-freelist! #f) # turn off checking
2405 Checking of the freelist forces a traversal of the freelist and
2406 a garbage collection before each allocation of a cell. This can
2407 slow down the interpreter dramatically, so the setter should be used to
2408 turn on this extra processing only when necessary.
2410 ** New configure option --enable-debug-malloc
2412 Include code for debugging of calls to scm_must_malloc/realloc/free.
2416 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2417 2. objects reallocated by scm_must_realloc has been allocated by
2419 3. reallocated objects are reallocated with the same what string
2421 But, most importantly, it records the number of allocated objects of
2422 each kind. This is useful when searching for memory leaks.
2424 A Guile compiled with this option provides the primitive
2425 `malloc-stats' which returns an alist with pairs of kind and the
2426 number of objects of that kind.
2428 ** All includes are now referenced relative to the root directory
2430 Since some users have had problems with mixups between Guile and
2431 system headers, we have decided to always refer to Guile headers via
2432 their parent directories. This essentially creates a "private name
2433 space" for Guile headers. This means that the compiler only is given
2434 -I options for the root build and root source directory.
2436 ** Header files kw.h and genio.h have been removed.
2438 ** The module (ice-9 getopt-gnu-style) has been removed.
2440 ** New module (ice-9 documentation)
2442 Implements the interface to documentation strings associated with
2445 ** New module (ice-9 time)
2447 Provides a macro `time', which displays execution time of a given form.
2449 ** New module (ice-9 history)
2451 Loading this module enables value history in the repl.
2453 * Changes to the stand-alone interpreter
2455 ** New command line option --debug
2457 Start Guile with debugging evaluator and backtraces enabled.
2459 This is useful when debugging your .guile init file or scripts.
2461 ** New help facility
2463 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2464 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2465 (help 'NAME) gives documentation for NAME, even if it is not an object
2466 (help ,EXPR) gives documentation for object returned by EXPR
2467 (help (my module)) gives module commentary for `(my module)'
2468 (help) gives this text
2470 `help' searches among bindings exported from loaded modules, while
2471 `apropos' searches among bindings visible from the "current" module.
2473 Examples: (help help)
2475 (help "output-string")
2477 ** `help' and `apropos' now prints full module names
2479 ** Dynamic linking now uses libltdl from the libtool package.
2481 The old system dependent code for doing dynamic linking has been
2482 replaced with calls to the libltdl functions which do all the hairy
2485 The major improvement is that you can now directly pass libtool
2486 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2487 will be able to do the best shared library job you can get, via
2490 The way dynamic libraries are found has changed and is not really
2491 portable across platforms, probably. It is therefore recommended to
2492 use absolute filenames when possible.
2494 If you pass a filename without an extension to `dynamic-link', it will
2495 try a few appropriate ones. Thus, the most platform ignorant way is
2496 to specify a name like "libfoo", without any directories and
2499 ** Guile COOP threads are now compatible with LinuxThreads
2501 Previously, COOP threading wasn't possible in applications linked with
2502 Linux POSIX threads due to their use of the stack pointer to find the
2503 thread context. This has now been fixed with a workaround which uses
2504 the pthreads to allocate the stack.
2506 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2508 ** Positions of erring expression in scripts
2510 With version 1.3.4, the location of the erring expression in Guile
2511 scipts is no longer automatically reported. (This should have been
2512 documented before the 1.3.4 release.)
2514 You can get this information by enabling recording of positions of
2515 source expressions and running the debugging evaluator. Put this at
2516 the top of your script (or in your "site" file):
2518 (read-enable 'positions)
2519 (debug-enable 'debug)
2521 ** Backtraces in scripts
2523 It is now possible to get backtraces in scripts.
2527 (debug-enable 'debug 'backtrace)
2529 at the top of the script.
2531 (The first options enables the debugging evaluator.
2532 The second enables backtraces.)
2534 ** Part of module system symbol lookup now implemented in C
2536 The eval closure of most modules is now implemented in C. Since this
2537 was one of the bottlenecks for loading speed, Guile now loads code
2538 substantially faster than before.
2540 ** Attempting to get the value of an unbound variable now produces
2541 an exception with a key of 'unbound-variable instead of 'misc-error.
2543 ** The initial default output port is now unbuffered if it's using a
2544 tty device. Previously in this situation it was line-buffered.
2546 ** New hook: after-gc-hook
2548 after-gc-hook takes over the role of gc-thunk. This hook is run at
2549 the first SCM_TICK after a GC. (Thus, the code is run at the same
2550 point during evaluation as signal handlers.)
2552 Note that this hook should be used only for diagnostic and debugging
2553 purposes. It is not certain that it will continue to be well-defined
2554 when this hook is run in the future.
2556 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2557 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2559 ** Improvements to garbage collector
2561 Guile 1.4 has a new policy for triggering heap allocation and
2562 determining the sizes of heap segments. It fixes a number of problems
2565 1. The new policy can handle two separate pools of cells
2566 (2-word/4-word) better. (The old policy would run wild, allocating
2567 more and more memory for certain programs.)
2569 2. The old code would sometimes allocate far too much heap so that the
2570 Guile process became gigantic. The new code avoids this.
2572 3. The old code would sometimes allocate too little so that few cells
2573 were freed at GC so that, in turn, too much time was spent in GC.
2575 4. The old code would often trigger heap allocation several times in a
2576 row. (The new scheme predicts how large the segments needs to be
2577 in order not to need further allocation.)
2579 All in all, the new GC policy will make larger applications more
2582 The new GC scheme also is prepared for POSIX threading. Threads can
2583 allocate private pools of cells ("clusters") with just a single
2584 function call. Allocation of single cells from such a cluster can
2585 then proceed without any need of inter-thread synchronization.
2587 ** New environment variables controlling GC parameters
2589 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2592 Allocation of 2-word cell heaps:
2594 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2597 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2598 GC in percent of total heap size
2601 Allocation of 4-word cell heaps
2602 (used for real numbers and misc other objects):
2604 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2606 (See entry "Way for application to customize GC parameters" under
2607 section "Changes to the scm_ interface" below.)
2609 ** Guile now implements reals using 4-word cells
2611 This speeds up computation with reals. (They were earlier allocated
2612 with `malloc'.) There is still some room for optimizations, however.
2614 ** Some further steps toward POSIX thread support have been taken
2616 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2617 don't have much effect any longer, and many of them will be removed in
2621 are only handled at the top of the evaluator loop, immediately after
2622 I/O, and in scm_equalp.
2624 *** The GC can allocate thread private pools of pairs.
2626 * Changes to Scheme functions and syntax
2628 ** close-input-port and close-output-port are now R5RS
2630 These procedures have been turned into primitives and have R5RS behaviour.
2632 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2634 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2635 extended by the more sophisticated version in (ice-9 format)
2637 (simple-format port message . args)
2638 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2639 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2640 the escapes are replaced with corresponding members of ARGS:
2641 ~A formats using `display' and ~S formats using `write'.
2642 If DESTINATION is #t, then use the `current-output-port',
2643 if DESTINATION is #f, then return a string containing the formatted text.
2644 Does not add a trailing newline."
2646 ** string-ref: the second argument is no longer optional.
2648 ** string, list->string: no longer accept strings in their arguments,
2649 only characters, for compatibility with R5RS.
2651 ** New procedure: port-closed? PORT
2652 Returns #t if PORT is closed or #f if it is open.
2654 ** Deprecated: list*
2656 The list* functionality is now provided by cons* (SRFI-1 compliant)
2658 ** New procedure: cons* ARG1 ARG2 ... ARGn
2660 Like `list', but the last arg provides the tail of the constructed list,
2661 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2663 Requires at least one argument. If given one argument, that argument
2664 is returned as result.
2666 This function is called `list*' in some other Schemes and in Common LISP.
2668 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2670 ** New procedure: object-documentation OBJECT
2672 Returns the documentation string associated with OBJECT. The
2673 procedure uses a caching mechanism so that subsequent lookups are
2676 Exported by (ice-9 documentation).
2678 ** module-name now returns full names of modules
2680 Previously, only the last part of the name was returned (`session' for
2681 `(ice-9 session)'). Ex: `(ice-9 session)'.
2683 * Changes to the gh_ interface
2685 ** Deprecated: gh_int2scmb
2687 Use gh_bool2scm instead.
2689 * Changes to the scm_ interface
2691 ** Guile primitives now carry docstrings!
2693 Thanks to Greg Badros!
2695 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2697 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2698 macros and must contain a docstring that is extracted into foo.doc using a new
2699 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2701 However, a major overhaul of these macros is scheduled for the next release of
2704 ** Guile primitives use a new technique for validation of arguments
2706 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2707 the readability of argument checking.
2709 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2711 ** New macros: SCM_PACK, SCM_UNPACK
2713 Compose/decompose an SCM value.
2715 The SCM type is now treated as an abstract data type and may be defined as a
2716 long, a void* or as a struct, depending on the architecture and compile time
2717 options. This makes it easier to find several types of bugs, for example when
2718 SCM values are treated as integers without conversion. Values of the SCM type
2719 should be treated as "atomic" values. These macros are used when
2720 composing/decomposing an SCM value, either because you want to access
2721 individual bits, or because you want to treat it as an integer value.
2723 E.g., in order to set bit 7 in an SCM value x, use the expression
2725 SCM_PACK (SCM_UNPACK (x) | 0x80)
2727 ** The name property of hooks is deprecated.
2728 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2730 You can emulate this feature by using object properties.
2732 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2733 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2736 These macros will be removed in a future release of Guile.
2738 ** The following types, functions and macros from numbers.h are deprecated:
2739 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2740 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2742 ** Port internals: the rw_random variable in the scm_port structure
2743 must be set to non-zero in any random access port. In recent Guile
2744 releases it was only set for bidirectional random-access ports.
2746 ** Port internals: the seek ptob procedure is now responsible for
2747 resetting the buffers if required. The change was made so that in the
2748 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2749 the fport and strport ptobs can avoid resetting the buffers,
2750 in particular to avoid discarding unread chars. An existing port
2751 type can be fixed by adding something like the following to the
2752 beginning of the ptob seek procedure:
2754 if (pt->rw_active == SCM_PORT_READ)
2755 scm_end_input (object);
2756 else if (pt->rw_active == SCM_PORT_WRITE)
2757 ptob->flush (object);
2759 although to actually avoid resetting the buffers and discard unread
2760 chars requires further hacking that depends on the characteristics
2763 ** Deprecated functions: scm_fseek, scm_tag
2765 These functions are no longer used and will be removed in a future version.
2767 ** The scm_sysmissing procedure is no longer used in libguile.
2768 Unless it turns out to be unexpectedly useful to somebody, it will be
2769 removed in a future version.
2771 ** The format of error message strings has changed
2773 The two C procedures: scm_display_error and scm_error, as well as the
2774 primitive `scm-error', now use scm_simple_format to do their work.
2775 This means that the message strings of all code must be updated to use
2776 ~A where %s was used before, and ~S where %S was used before.
2778 During the period when there still are a lot of old Guiles out there,
2779 you might want to support both old and new versions of Guile.
2781 There are basically two methods to achieve this. Both methods use
2784 AC_CHECK_FUNCS(scm_simple_format)
2786 in your configure.in.
2788 Method 1: Use the string concatenation features of ANSI C's
2793 #ifdef HAVE_SCM_SIMPLE_FORMAT
2799 Then represent each of your error messages using a preprocessor macro:
2801 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2805 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2806 (define make-message string-append)
2808 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2810 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2814 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2819 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2823 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2825 Don't use the functions coop_mutex_init and
2826 coop_condition_variable_init. They will change.
2828 Use scm_mutex_init and scm_cond_init instead.
2830 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2831 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2832 COND, as `scm_cond_wait' does, but it also bounds the duration
2833 of the wait. If COND has not been signaled before time ABSTIME,
2834 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2835 returns the error code `ETIMEDOUT'.
2837 The ABSTIME parameter specifies an absolute time, with the same
2838 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2839 to 00:00:00 GMT, January 1, 1970.
2841 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2842 `scm_cond_broadcast' restarts all the threads that are waiting
2843 on the condition variable COND. Nothing happens if no threads are
2846 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2847 `scm_key_create' allocates a new TSD key. The key is stored in
2848 the location pointed to by KEY. There is no limit on the number
2849 of keys allocated at a given time. The value initially associated
2850 with the returned key is `NULL' in all currently executing threads.
2852 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2853 function associated with the key. When a thread terminates,
2854 DESTR_FUNCTION is called on the value associated with the key in
2855 that thread. The DESTR_FUNCTION is not called if a key is deleted
2856 with `scm_key_delete' or a value is changed with
2857 `scm_setspecific'. The order in which destructor functions are
2858 called at thread termination time is unspecified.
2860 Destructors are not yet implemented.
2862 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2863 `scm_setspecific' changes the value associated with KEY in the
2864 calling thread, storing the given POINTER instead.
2866 ** New function: scm_getspecific (scm_key_t KEY)
2867 `scm_getspecific' returns the value currently associated with
2868 KEY in the calling thread.
2870 ** New function: scm_key_delete (scm_key_t KEY)
2871 `scm_key_delete' deallocates a TSD key. It does not check
2872 whether non-`NULL' values are associated with that key in the
2873 currently executing threads, nor call the destructor function
2874 associated with the key.
2876 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2878 Initialize a C level hook HOOK with associated HOOK_DATA and type
2879 TYPE. (See scm_c_hook_run ().)
2881 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2883 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2884 is true, add it last, otherwise first. The same FUNC can be added
2885 multiple times if FUNC_DATA differ and vice versa.
2887 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2889 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2890 function is only removed if both FUNC and FUNC_DATA matches.
2892 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2894 Run hook HOOK passing DATA to the hook functions.
2896 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2897 returned is undefined.
2899 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2900 returns a non-NULL value. This value is returned as the result of
2901 scm_c_hook_run. If all functions return NULL, NULL is returned.
2903 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2904 returns a NULL value, and NULL is returned. If all functions returns
2905 a non-NULL value, the last value is returned.
2907 ** New C level GC hooks
2909 Five new C level hooks has been added to the garbage collector.
2911 scm_before_gc_c_hook
2914 are run before locking and after unlocking the heap. The system is
2915 thus in a mode where evaluation can take place. (Except that
2916 scm_before_gc_c_hook must not allocate new cells.)
2918 scm_before_mark_c_hook
2919 scm_before_sweep_c_hook
2920 scm_after_sweep_c_hook
2922 are run when the heap is locked. These are intended for extension of
2923 the GC in a modular fashion. Examples are the weaks and guardians
2926 ** Way for application to customize GC parameters
2928 The application can set up other default values for the GC heap
2929 allocation parameters
2931 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
2932 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
2933 GUILE_MAX_SEGMENT_SIZE,
2937 scm_default_init_heap_size_1, scm_default_min_yield_1,
2938 scm_default_init_heap_size_2, scm_default_min_yield_2,
2939 scm_default_max_segment_size
2941 respectively before callong scm_boot_guile.
2943 (See entry "New environment variables ..." in section
2944 "Changes to the stand-alone interpreter" above.)
2946 ** scm_protect_object/scm_unprotect_object now nest
2948 This means that you can call scm_protect_object multiple times on an
2949 object and count on the object being protected until
2950 scm_unprotect_object has been call the same number of times.
2952 The functions also have better time complexity.
2954 Still, it is usually possible to structure the application in a way
2955 that you don't need to use these functions. For example, if you use a
2956 protected standard Guile list to keep track of live objects rather
2957 than some custom data type, objects will die a natural death when they
2958 are no longer needed.
2960 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
2962 Guile does not provide the float representation for inexact real numbers any
2963 more. Now, only doubles are used to represent inexact real numbers. Further,
2964 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
2965 and scm_tc16_complex, respectively.
2967 ** Removed deprecated type scm_smobfuns
2969 ** Removed deprecated function scm_newsmob
2971 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
2973 There is an ongoing discussion among the developers whether to
2974 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
2975 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
2976 until this issue has been settled.
2978 ** Removed deprecated type tag scm_tc16_kw
2980 ** Added type tag scm_tc16_keyword
2982 (This was introduced already in release 1.3.4 but was not documented
2985 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
2987 * Changes to system call interfaces:
2989 ** The "select" procedure now tests port buffers for the ability to
2990 provide input or accept output. Previously only the underlying file
2991 descriptors were checked.
2993 ** New variable PIPE_BUF: the maximum number of bytes that can be
2994 atomically written to a pipe.
2996 ** If a facility is not available on the system when Guile is
2997 compiled, the corresponding primitive procedure will not be defined.
2998 Previously it would have been defined but would throw a system-error
2999 exception if called. Exception handlers which catch this case may
3000 need minor modification: an error will be thrown with key
3001 'unbound-variable instead of 'system-error. Alternatively it's
3002 now possible to use `defined?' to check whether the facility is
3005 ** Procedures which depend on the timezone should now give the correct
3006 result on systems which cache the TZ environment variable, even if TZ
3007 is changed without calling tzset.
3009 * Changes to the networking interfaces:
3011 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3012 long integers between network and host format. For now, it's not
3013 particularly convenient to do this kind of thing, but consider:
3015 (define write-network-long
3016 (lambda (value port)
3017 (let ((v (make-uniform-vector 1 1 0)))
3018 (uniform-vector-set! v 0 (htonl value))
3019 (uniform-vector-write v port))))
3021 (define read-network-long
3023 (let ((v (make-uniform-vector 1 1 0)))
3024 (uniform-vector-read! v port)
3025 (ntohl (uniform-vector-ref v 0)))))
3027 ** If inet-aton fails, it now throws an error with key 'misc-error
3028 instead of 'system-error, since errno is not relevant.
3030 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3031 specific keys instead of 'system-error. The latter is inappropriate
3032 since errno will not have been set. The keys are:
3033 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3035 ** sethostent, setnetent, setprotoent, setservent: now take an
3036 optional argument STAYOPEN, which specifies whether the database
3037 remains open after a database entry is accessed randomly (e.g., using
3038 gethostbyname for the hosts database.) The default is #f. Previously
3042 Changes since Guile 1.3.2:
3044 * Changes to the stand-alone interpreter
3048 An initial version of the Guile debugger written by Chris Hanson has
3049 been added. The debugger is still under development but is included
3050 in the distribution anyway since it is already quite useful.
3056 after an error to enter the debugger. Type `help' inside the debugger
3057 for a description of available commands.
3059 If you prefer to have stack frames numbered and printed in
3060 anti-chronological order and prefer up in the stack to be down on the
3061 screen as is the case in gdb, you can put
3063 (debug-enable 'backwards)
3065 in your .guile startup file. (However, this means that Guile can't
3066 use indentation to indicate stack level.)
3068 The debugger is autoloaded into Guile at the first use.
3070 ** Further enhancements to backtraces
3072 There is a new debug option `width' which controls the maximum width
3073 on the screen of printed stack frames. Fancy printing parameters
3074 ("level" and "length" as in Common LISP) are adaptively adjusted for
3075 each stack frame to give maximum information while still fitting
3076 within the bounds. If the stack frame can't be made to fit by
3077 adjusting parameters, it is simply cut off at the end. This is marked
3080 ** Some modules are now only loaded when the repl is started
3082 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3083 regex) are now loaded into (guile-user) only if the repl has been
3084 started. The effect is that the startup time for scripts has been
3085 reduced to 30% of what it was previously.
3087 Correctly written scripts load the modules they require at the top of
3088 the file and should not be affected by this change.
3090 ** Hooks are now represented as smobs
3092 * Changes to Scheme functions and syntax
3094 ** Readline support has changed again.
3096 The old (readline-activator) module is gone. Use (ice-9 readline)
3097 instead, which now contains all readline functionality. So the code
3098 to activate readline is now
3100 (use-modules (ice-9 readline))
3103 This should work at any time, including from the guile prompt.
3105 To avoid confusion about the terms of Guile's license, please only
3106 enable readline for your personal use; please don't make it the
3107 default for others. Here is why we make this rather odd-sounding
3110 Guile is normally licensed under a weakened form of the GNU General
3111 Public License, which allows you to link code with Guile without
3112 placing that code under the GPL. This exception is important to some
3115 However, since readline is distributed under the GNU General Public
3116 License, when you link Guile with readline, either statically or
3117 dynamically, you effectively change Guile's license to the strict GPL.
3118 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3119 which are normally permitted become forbidden. This is a rather
3120 non-obvious consequence of the licensing terms.
3122 So, to make sure things remain clear, please let people choose for
3123 themselves whether to link GPL'd libraries like readline with Guile.
3125 ** regexp-substitute/global has changed slightly, but incompatibly.
3127 If you include a function in the item list, the string of the match
3128 object it receives is the same string passed to
3129 regexp-substitute/global, not some suffix of that string.
3130 Correspondingly, the match's positions are relative to the entire
3131 string, not the suffix.
3133 If the regexp can match the empty string, the way matches are chosen
3134 from the string has changed. regexp-substitute/global recognizes the
3135 same set of matches that list-matches does; see below.
3137 ** New function: list-matches REGEXP STRING [FLAGS]
3139 Return a list of match objects, one for every non-overlapping, maximal
3140 match of REGEXP in STRING. The matches appear in left-to-right order.
3141 list-matches only reports matches of the empty string if there are no
3142 other matches which begin on, end at, or include the empty match's
3145 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3147 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3149 For each match of REGEXP in STRING, apply PROC to the match object,
3150 and the last value PROC returned, or INIT for the first call. Return
3151 the last value returned by PROC. We apply PROC to the matches as they
3152 appear from left to right.
3154 This function recognizes matches according to the same criteria as
3157 Thus, you could define list-matches like this:
3159 (define (list-matches regexp string . flags)
3160 (reverse! (apply fold-matches regexp string '() cons flags)))
3162 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3166 *** New function: hook? OBJ
3168 Return #t if OBJ is a hook, otherwise #f.
3170 *** New function: make-hook-with-name NAME [ARITY]
3172 Return a hook with name NAME and arity ARITY. The default value for
3173 ARITY is 0. The only effect of NAME is that it will appear when the
3174 hook object is printed to ease debugging.
3176 *** New function: hook-empty? HOOK
3178 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3180 *** New function: hook->list HOOK
3182 Return a list of the procedures that are called when run-hook is
3185 ** `map' signals an error if its argument lists are not all the same length.
3187 This is the behavior required by R5RS, so this change is really a bug
3188 fix. But it seems to affect a lot of people's code, so we're
3189 mentioning it here anyway.
3191 ** Print-state handling has been made more transparent
3193 Under certain circumstances, ports are represented as a port with an
3194 associated print state. Earlier, this pair was represented as a pair
3195 (see "Some magic has been added to the printer" below). It is now
3196 indistinguishable (almost; see `get-print-state') from a port on the
3199 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3201 Return a new port with the associated print state PRINT-STATE.
3203 *** New function: get-print-state OUTPUT-PORT
3205 Return the print state associated with this port if it exists,
3206 otherwise return #f.
3208 *** New function: directory-stream? OBJECT
3210 Returns true iff OBJECT is a directory stream --- the sort of object
3211 returned by `opendir'.
3213 ** New function: using-readline?
3215 Return #t if readline is in use in the current repl.
3217 ** structs will be removed in 1.4
3219 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3220 and use GOOPS objects as the fundamental record type.
3222 * Changes to the scm_ interface
3224 ** structs will be removed in 1.4
3226 The entire current struct interface (struct.c, struct.h) will be
3227 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3228 GOOPS objects as the fundamental record type.
3230 ** The internal representation of subr's has changed
3232 Instead of giving a hint to the subr name, the CAR field of the subr
3233 now contains an index to a subr entry in scm_subr_table.
3235 *** New variable: scm_subr_table
3237 An array of subr entries. A subr entry contains the name, properties
3238 and documentation associated with the subr. The properties and
3239 documentation slots are not yet used.
3241 ** A new scheme for "forwarding" calls to a builtin to a generic function
3243 It is now possible to extend the functionality of some Guile
3244 primitives by letting them defer a call to a GOOPS generic function on
3245 argument mismatch. This means that there is no loss of efficiency in
3250 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3251 (define-method + ((x <string>) (y <string>))
3252 (string-append x y))
3254 + will still be as efficient as usual in numerical calculations, but
3255 can also be used for concatenating strings.
3257 Who will be the first one to extend Guile's numerical tower to
3258 rationals? :) [OK, there a few other things to fix before this can
3259 be made in a clean way.]
3261 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3263 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3265 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3267 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3268 a variable GENERIC which can be used by the dispatch macros below.
3270 [This is experimental code which may change soon.]
3272 *** New macros for forwarding control to a generic on arg type error
3274 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3276 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3278 These correspond to the scm_wta function call, and have the same
3279 behaviour until the user has called the GOOPS primitive
3280 `enable-primitive-generic!'. After that, these macros will apply the
3281 generic function GENERIC to the argument(s) instead of calling
3284 [This is experimental code which may change soon.]
3286 *** New macros for argument testing with generic dispatch
3288 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3290 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3292 These correspond to the SCM_ASSERT macro, but will defer control to
3293 GENERIC on error after `enable-primitive-generic!' has been called.
3295 [This is experimental code which may change soon.]
3297 ** New function: SCM scm_eval_body (SCM body, SCM env)
3299 Evaluates the body of a special form.
3301 ** The internal representation of struct's has changed
3303 Previously, four slots were allocated for the procedure(s) of entities
3304 and operators. The motivation for this representation had to do with
3305 the structure of the evaluator, the wish to support tail-recursive
3306 generic functions, and efficiency. Since the generic function
3307 dispatch mechanism has changed, there is no longer a need for such an
3308 expensive representation, and the representation has been simplified.
3310 This should not make any difference for most users.
3312 ** GOOPS support has been cleaned up.
3314 Some code has been moved from eval.c to objects.c and code in both of
3315 these compilation units has been cleaned up and better structured.
3317 *** New functions for applying generic functions
3319 New function: SCM scm_apply_generic (GENERIC, ARGS)
3320 New function: SCM scm_call_generic_0 (GENERIC)
3321 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3322 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3323 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3325 ** Deprecated function: scm_make_named_hook
3327 It is now replaced by:
3329 ** New function: SCM scm_create_hook (const char *name, int arity)
3331 Creates a hook in the same way as make-hook above but also
3332 binds a variable named NAME to it.
3334 This is the typical way of creating a hook from C code.
3336 Currently, the variable is created in the "current" module.
3337 This might change when we get the new module system.
3339 [The behaviour is identical to scm_make_named_hook.]
3343 Changes since Guile 1.3:
3345 * Changes to mailing lists
3347 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3349 See the README file to find current addresses for all the Guile
3352 * Changes to the distribution
3354 ** Readline support is no longer included with Guile by default.
3356 Based on the different license terms of Guile and Readline, we
3357 concluded that Guile should not *by default* cause the linking of
3358 Readline into an application program. Readline support is now offered
3359 as a separate module, which is linked into an application only when
3360 you explicitly specify it.
3362 Although Guile is GNU software, its distribution terms add a special
3363 exception to the usual GNU General Public License (GPL). Guile's
3364 license includes a clause that allows you to link Guile with non-free
3365 programs. We add this exception so as not to put Guile at a
3366 disadvantage vis-a-vis other extensibility packages that support other
3369 In contrast, the GNU Readline library is distributed under the GNU
3370 General Public License pure and simple. This means that you may not
3371 link Readline, even dynamically, into an application unless it is
3372 distributed under a free software license that is compatible the GPL.
3374 Because of this difference in distribution terms, an application that
3375 can use Guile may not be able to use Readline. Now users will be
3376 explicitly offered two independent decisions about the use of these
3379 You can activate the readline support by issuing
3381 (use-modules (readline-activator))
3384 from your ".guile" file, for example.
3386 * Changes to the stand-alone interpreter
3388 ** All builtins now print as primitives.
3389 Previously builtin procedures not belonging to the fundamental subr
3390 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3391 Now, they print as #<primitive-procedure NAME>.
3393 ** Backtraces slightly more intelligible.
3394 gsubr-apply and macro transformer application frames no longer appear
3397 * Changes to Scheme functions and syntax
3399 ** Guile now correctly handles internal defines by rewriting them into
3400 their equivalent letrec. Previously, internal defines would
3401 incrementally add to the innermost environment, without checking
3402 whether the restrictions specified in RnRS were met. This lead to the
3403 correct behaviour when these restriction actually were met, but didn't
3404 catch all illegal uses. Such an illegal use could lead to crashes of
3405 the Guile interpreter or or other unwanted results. An example of
3406 incorrect internal defines that made Guile behave erratically:
3418 The problem with this example is that the definition of `c' uses the
3419 value of `b' directly. This confuses the meoization machine of Guile
3420 so that the second call of `b' (this time in a larger environment that
3421 also contains bindings for `c' and `d') refers to the binding of `c'
3422 instead of `a'. You could also make Guile crash with a variation on
3427 (define (b flag) (if flag a 1))
3428 (define c (1+ (b flag)))
3436 From now on, Guile will issue an `Unbound variable: b' error message
3441 A hook contains a list of functions which should be called on
3442 particular occasions in an existing program. Hooks are used for
3445 A window manager might have a hook before-window-map-hook. The window
3446 manager uses the function run-hooks to call all functions stored in
3447 before-window-map-hook each time a window is mapped. The user can
3448 store functions in the hook using add-hook!.
3450 In Guile, hooks are first class objects.
3452 *** New function: make-hook [N_ARGS]
3454 Return a hook for hook functions which can take N_ARGS arguments.
3455 The default value for N_ARGS is 0.
3457 (See also scm_make_named_hook below.)
3459 *** New function: add-hook! HOOK PROC [APPEND_P]
3461 Put PROC at the beginning of the list of functions stored in HOOK.
3462 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3464 PROC must be able to take the number of arguments specified when the
3467 If PROC already exists in HOOK, then remove it first.
3469 *** New function: remove-hook! HOOK PROC
3471 Remove PROC from the list of functions in HOOK.
3473 *** New function: reset-hook! HOOK
3475 Clear the list of hook functions stored in HOOK.
3477 *** New function: run-hook HOOK ARG1 ...
3479 Run all hook functions stored in HOOK with arguments ARG1 ... .
3480 The number of arguments supplied must correspond to the number given
3481 when the hook was created.
3483 ** The function `dynamic-link' now takes optional keyword arguments.
3484 The only keyword argument that is currently defined is `:global
3485 BOOL'. With it, you can control whether the shared library will be
3486 linked in global mode or not. In global mode, the symbols from the
3487 linked library can be used to resolve references from other
3488 dynamically linked libraries. In non-global mode, the linked
3489 library is essentially invisible and can only be accessed via
3490 `dynamic-func', etc. The default is now to link in global mode.
3491 Previously, the default has been non-global mode.
3493 The `#:global' keyword is only effective on platforms that support
3494 the dlopen family of functions.
3496 ** New function `provided?'
3498 - Function: provided? FEATURE
3499 Return true iff FEATURE is supported by this installation of
3500 Guile. FEATURE must be a symbol naming a feature; the global
3501 variable `*features*' is a list of available features.
3503 ** Changes to the module (ice-9 expect):
3505 *** The expect-strings macro now matches `$' in a regular expression
3506 only at a line-break or end-of-file by default. Previously it would
3507 match the end of the string accumulated so far. The old behaviour
3508 can be obtained by setting the variable `expect-strings-exec-flags'
3511 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3512 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3513 in a regular expression will still match before a line-break or
3514 end-of-file. The default is `regexp/noteol'.
3516 *** The expect-strings macro now uses a variable
3517 `expect-strings-compile-flags' for the flags to be supplied to
3518 `make-regexp'. The default is `regexp/newline', which was previously
3521 *** The expect macro now supplies two arguments to a match procedure:
3522 the current accumulated string and a flag to indicate whether
3523 end-of-file has been reached. Previously only the string was supplied.
3524 If end-of-file is reached, the match procedure will be called an
3525 additional time with the same accumulated string as the previous call
3526 but with the flag set.
3528 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3530 This code, and the documentation for it that appears here, was
3531 borrowed from SLIB, with minor adaptations for Guile.
3533 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3534 An almost complete implementation of Common LISP format description
3535 according to the CL reference book `Common LISP' from Guy L.
3536 Steele, Digital Press. Backward compatible to most of the
3537 available Scheme format implementations.
3539 Returns `#t', `#f' or a string; has side effect of printing
3540 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3541 to the current output port and `#t' is returned. If DESTINATION
3542 is `#f', a formatted string is returned as the result of the call.
3543 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3544 format string; FORMAT-STRING is then the first argument and the
3545 output is returned as a string. If DESTINATION is a number, the
3546 output is to the current error port if available by the
3547 implementation. Otherwise DESTINATION must be an output port and
3550 FORMAT-STRING must be a string. In case of a formatting error
3551 format returns `#f' and prints a message on the current output or
3552 error port. Characters are output as if the string were output by
3553 the `display' function with the exception of those prefixed by a
3554 tilde (~). For a detailed description of the FORMAT-STRING syntax
3555 please consult a Common LISP format reference manual. For a test
3556 suite to verify this format implementation load `formatst.scm'.
3557 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3559 Note: `format' is not reentrant, i.e. only one `format'-call may
3560 be executed at a time.
3563 *** Format Specification (Format version 3.0)
3565 Please consult a Common LISP format reference manual for a detailed
3566 description of the format string syntax. For a demonstration of the
3567 implemented directives see `formatst.scm'.
3569 This implementation supports directive parameters and modifiers (`:'
3570 and `@' characters). Multiple parameters must be separated by a comma
3571 (`,'). Parameters can be numerical parameters (positive or negative),
3572 character parameters (prefixed by a quote character (`''), variable
3573 parameters (`v'), number of rest arguments parameter (`#'), empty and
3574 default parameters. Directive characters are case independent. The
3575 general form of a directive is:
3577 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3579 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3581 *** Implemented CL Format Control Directives
3583 Documentation syntax: Uppercase characters represent the
3584 corresponding control directive characters. Lowercase characters
3585 represent control directive parameter descriptions.
3588 Any (print as `display' does).
3592 `~MINCOL,COLINC,MINPAD,PADCHARA'
3596 S-expression (print as `write' does).
3600 `~MINCOL,COLINC,MINPAD,PADCHARS'
3606 print number sign always.
3609 print comma separated.
3611 `~MINCOL,PADCHAR,COMMACHARD'
3617 print number sign always.
3620 print comma separated.
3622 `~MINCOL,PADCHAR,COMMACHARX'
3628 print number sign always.
3631 print comma separated.
3633 `~MINCOL,PADCHAR,COMMACHARO'
3639 print number sign always.
3642 print comma separated.
3644 `~MINCOL,PADCHAR,COMMACHARB'
3649 `~N,MINCOL,PADCHAR,COMMACHARR'
3653 print a number as a Roman numeral.
3656 print a number as an "old fashioned" Roman numeral.
3659 print a number as an ordinal English number.
3662 print a number as a cardinal English number.
3667 prints `y' and `ies'.
3670 as `~P but jumps 1 argument backward.'
3673 as `~@P but jumps 1 argument backward.'
3678 prints a character as the reader can understand it (i.e. `#\'
3682 prints a character as emacs does (eg. `^C' for ASCII 03).
3685 Fixed-format floating-point (prints a flonum like MMM.NNN).
3686 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3688 If the number is positive a plus sign is printed.
3691 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3692 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3694 If the number is positive a plus sign is printed.
3697 General floating-point (prints a flonum either fixed or
3699 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3701 If the number is positive a plus sign is printed.
3704 Dollars floating-point (prints a flonum in fixed with signs
3706 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3708 If the number is positive a plus sign is printed.
3711 A sign is always printed and appears before the padding.
3714 The sign appears before the padding.
3722 print newline if not at the beginning of the output line.
3724 prints `~&' and then N-1 newlines.
3729 print N page separators.
3739 newline is ignored, white space left.
3742 newline is left, white space ignored.
3747 relative tabulation.
3753 Indirection (expects indirect arguments as a list).
3755 extracts indirect arguments from format arguments.
3758 Case conversion (converts by `string-downcase').
3760 converts by `string-capitalize'.
3763 converts by `string-capitalize-first'.
3766 converts by `string-upcase'.
3769 Argument Jumping (jumps 1 argument forward).
3771 jumps N arguments forward.
3774 jumps 1 argument backward.
3777 jumps N arguments backward.
3780 jumps to the 0th argument.
3783 jumps to the Nth argument (beginning from 0)
3785 `~[STR0~;STR1~;...~;STRN~]'
3786 Conditional Expression (numerical clause conditional).
3788 take argument from N.
3791 true test conditional.
3794 if-else-then conditional.
3800 default clause follows.
3803 Iteration (args come from the next argument (a list)).
3805 at most N iterations.
3808 args from next arg (a list of lists).
3811 args from the rest of arguments.
3814 args from the rest args (lists).
3825 aborts if N <= M <= K
3827 *** Not Implemented CL Format Control Directives
3830 print `#f' as an empty list (see below).
3833 print `#f' as an empty list (see below).
3839 (sorry I don't understand its semantics completely)
3841 *** Extended, Replaced and Additional Control Directives
3843 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3844 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3845 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3846 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3847 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3848 COMMAWIDTH is the number of characters between two comma
3852 print a R4RS complex number as `~F~@Fi' with passed parameters for
3856 Pretty print formatting of an argument for scheme code lists.
3862 Flushes the output if format DESTINATION is a port.
3865 Print a `#\space' character
3867 print N `#\space' characters.
3870 Print a `#\tab' character
3872 print N `#\tab' characters.
3875 Takes N as an integer representation for a character. No arguments
3876 are consumed. N is converted to a character by `integer->char'. N
3877 must be a positive decimal number.
3880 Print out readproof. Prints out internal objects represented as
3881 `#<...>' as strings `"#<...>"' so that the format output can always
3882 be processed by `read'.
3885 Print out readproof. Prints out internal objects represented as
3886 `#<...>' as strings `"#<...>"' so that the format output can always
3887 be processed by `read'.
3890 Prints information and a copyright notice on the format
3893 prints format version.
3896 may also print number strings, i.e. passing a number as a string
3897 and format it accordingly.
3899 *** Configuration Variables
3901 The format module exports some configuration variables to suit the
3902 systems and users needs. There should be no modification necessary for
3903 the configuration that comes with Guile. Format detects automatically
3904 if the running scheme system implements floating point numbers and
3907 format:symbol-case-conv
3908 Symbols are converted by `symbol->string' so the case type of the
3909 printed symbols is implementation dependent.
3910 `format:symbol-case-conv' is a one arg closure which is either
3911 `#f' (no conversion), `string-upcase', `string-downcase' or
3912 `string-capitalize'. (default `#f')
3914 format:iobj-case-conv
3915 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3916 implementation internal objects. (default `#f')
3919 The character prefixing the exponent value in `~E' printing.
3922 *** Compatibility With Other Format Implementations
3928 Downward compatible except for padding support and `~A', `~S',
3929 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
3930 `printf' padding support which is completely replaced by the CL
3931 `format' padding style.
3934 Downward compatible except for `~', which is not documented
3935 (ignores all characters inside the format string up to a newline
3936 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
3937 numerical and variable parameters and `:/@' modifiers in the CL
3941 Downward compatible except for `~A' and `~S' which print in
3942 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
3943 directive parameters or modifiers)).
3946 Downward compatible except for an optional destination parameter:
3947 S2C accepts a format call without a destination which returns a
3948 formatted string. This is equivalent to a #f destination in S2C.
3949 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
3950 parameters or modifiers)).
3953 ** Changes to string-handling functions.
3955 These functions were added to support the (ice-9 format) module, above.
3957 *** New function: string-upcase STRING
3958 *** New function: string-downcase STRING
3960 These are non-destructive versions of the existing string-upcase! and
3961 string-downcase! functions.
3963 *** New function: string-capitalize! STRING
3964 *** New function: string-capitalize STRING
3966 These functions convert the first letter of each word in the string to
3969 (string-capitalize "howdy there")
3972 As with the other functions, string-capitalize! modifies the string in
3973 place, while string-capitalize returns a modified copy of its argument.
3975 *** New function: string-ci->symbol STRING
3977 Return a symbol whose name is STRING, but having the same case as if
3978 the symbol had be read by `read'.
3980 Guile can be configured to be sensitive or insensitive to case
3981 differences in Scheme identifiers. If Guile is case-insensitive, all
3982 symbols are converted to lower case on input. The `string-ci->symbol'
3983 function returns a symbol whose name in STRING, transformed as Guile
3984 would if STRING were input.
3986 *** New function: substring-move! STRING1 START END STRING2 START
3988 Copy the substring of STRING1 from START (inclusive) to END
3989 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
3990 string, and the source and destination areas may overlap; in all
3991 cases, the function behaves as if all the characters were copied
3994 *** Extended functions: substring-move-left! substring-move-right!
3996 These functions now correctly copy arbitrarily overlapping substrings;
3997 they are both synonyms for substring-move!.
4000 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4002 getopt-long is a function for parsing command-line arguments in a
4003 manner consistent with other GNU programs.
4005 (getopt-long ARGS GRAMMAR)
4006 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4008 ARGS should be a list of strings. Its first element should be the
4009 name of the program; subsequent elements should be the arguments
4010 that were passed to the program on the command line. The
4011 `program-arguments' procedure returns a list of this form.
4013 GRAMMAR is a list of the form:
4014 ((OPTION (PROPERTY VALUE) ...) ...)
4016 Each OPTION should be a symbol. `getopt-long' will accept a
4017 command-line option named `--OPTION'.
4018 Each option can have the following (PROPERTY VALUE) pairs:
4020 (single-char CHAR) --- Accept `-CHAR' as a single-character
4021 equivalent to `--OPTION'. This is how to specify traditional
4023 (required? BOOL) --- If BOOL is true, the option is required.
4024 getopt-long will raise an error if it is not found in ARGS.
4025 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4026 it is #f, it does not; and if it is the symbol
4027 `optional', the option may appear in ARGS with or
4029 (predicate FUNC) --- If the option accepts a value (i.e. you
4030 specified `(value #t)' for this option), then getopt
4031 will apply FUNC to the value, and throw an exception
4032 if it returns #f. FUNC should be a procedure which
4033 accepts a string and returns a boolean value; you may
4034 need to use quasiquotes to get it into GRAMMAR.
4036 The (PROPERTY VALUE) pairs may occur in any order, but each
4037 property may occur only once. By default, options do not have
4038 single-character equivalents, are not required, and do not take
4041 In ARGS, single-character options may be combined, in the usual
4042 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4043 accepts values, then it must be the last option in the
4044 combination; the value is the next argument. So, for example, using
4045 the following grammar:
4046 ((apples (single-char #\a))
4047 (blimps (single-char #\b) (value #t))
4048 (catalexis (single-char #\c) (value #t)))
4049 the following argument lists would be acceptable:
4050 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4051 for "blimps" and "catalexis")
4052 ("-ab" "bang" "-c" "couth") (same)
4053 ("-ac" "couth" "-b" "bang") (same)
4054 ("-abc" "couth" "bang") (an error, since `-b' is not the
4055 last option in its combination)
4057 If an option's value is optional, then `getopt-long' decides
4058 whether it has a value by looking at what follows it in ARGS. If
4059 the next element is a string, and it does not appear to be an
4060 option itself, then that string is the option's value.
4062 The value of a long option can appear as the next element in ARGS,
4063 or it can follow the option name, separated by an `=' character.
4064 Thus, using the same grammar as above, the following argument lists
4066 ("--apples" "Braeburn" "--blimps" "Goodyear")
4067 ("--apples=Braeburn" "--blimps" "Goodyear")
4068 ("--blimps" "Goodyear" "--apples=Braeburn")
4070 If the option "--" appears in ARGS, argument parsing stops there;
4071 subsequent arguments are returned as ordinary arguments, even if
4072 they resemble options. So, in the argument list:
4073 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4074 `getopt-long' will recognize the `apples' option as having the
4075 value "Granny Smith", but it will not recognize the `blimp'
4076 option; it will return the strings "--blimp" and "Goodyear" as
4077 ordinary argument strings.
4079 The `getopt-long' function returns the parsed argument list as an
4080 assocation list, mapping option names --- the symbols from GRAMMAR
4081 --- onto their values, or #t if the option does not accept a value.
4082 Unused options do not appear in the alist.
4084 All arguments that are not the value of any option are returned
4085 as a list, associated with the empty list.
4087 `getopt-long' throws an exception if:
4088 - it finds an unrecognized option in ARGS
4089 - a required option is omitted
4090 - an option that requires an argument doesn't get one
4091 - an option that doesn't accept an argument does get one (this can
4092 only happen using the long option `--opt=value' syntax)
4093 - an option predicate fails
4098 `((lockfile-dir (required? #t)
4101 (predicate ,file-is-directory?))
4102 (verbose (required? #f)
4105 (x-includes (single-char #\x))
4106 (rnet-server (single-char #\y)
4107 (predicate ,string?))))
4109 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4110 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4112 => ((() "foo1" "-fred" "foo2" "foo3")
4113 (rnet-server . "lamprod")
4114 (x-includes . "/usr/include")
4115 (lockfile-dir . "/tmp")
4118 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4120 It will be removed in a few releases.
4122 ** New syntax: lambda*
4123 ** New syntax: define*
4124 ** New syntax: define*-public
4125 ** New syntax: defmacro*
4126 ** New syntax: defmacro*-public
4127 Guile now supports optional arguments.
4129 `lambda*', `define*', `define*-public', `defmacro*' and
4130 `defmacro*-public' are identical to the non-* versions except that
4131 they use an extended type of parameter list that has the following BNF
4132 syntax (parentheses are literal, square brackets indicate grouping,
4133 and `*', `+' and `?' have the usual meaning):
4135 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4136 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4137 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4139 ext-var-decl ::= identifier | ( identifier expression )
4141 The semantics are best illustrated with the following documentation
4142 and examples for `lambda*':
4145 lambda extended for optional and keyword arguments
4147 lambda* creates a procedure that takes optional arguments. These
4148 are specified by putting them inside brackets at the end of the
4149 paramater list, but before any dotted rest argument. For example,
4150 (lambda* (a b #&optional c d . e) '())
4151 creates a procedure with fixed arguments a and b, optional arguments c
4152 and d, and rest argument e. If the optional arguments are omitted
4153 in a call, the variables for them are unbound in the procedure. This
4154 can be checked with the bound? macro.
4156 lambda* can also take keyword arguments. For example, a procedure
4158 (lambda* (#&key xyzzy larch) '())
4159 can be called with any of the argument lists (#:xyzzy 11)
4160 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4161 are given as keywords are bound to values.
4163 Optional and keyword arguments can also be given default values
4164 which they take on when they are not present in a call, by giving a
4165 two-item list in place of an optional argument, for example in:
4166 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4167 foo is a fixed argument, bar is an optional argument with default
4168 value 42, and baz is a keyword argument with default value 73.
4169 Default value expressions are not evaluated unless they are needed
4170 and until the procedure is called.
4172 lambda* now supports two more special parameter list keywords.
4174 lambda*-defined procedures now throw an error by default if a
4175 keyword other than one of those specified is found in the actual
4176 passed arguments. However, specifying #&allow-other-keys
4177 immediately after the kyword argument declarations restores the
4178 previous behavior of ignoring unknown keywords. lambda* also now
4179 guarantees that if the same keyword is passed more than once, the
4180 last one passed is the one that takes effect. For example,
4181 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4182 #:heads 37 #:tails 42 #:heads 99)
4183 would result in (99 47) being displayed.
4185 #&rest is also now provided as a synonym for the dotted syntax rest
4186 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4187 all respects to lambda*. This is provided for more similarity to DSSSL,
4188 MIT-Scheme and Kawa among others, as well as for refugees from other
4191 Further documentation may be found in the optargs.scm file itself.
4193 The optional argument module also exports the macros `let-optional',
4194 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4195 are not documented here because they may be removed in the future, but
4196 full documentation is still available in optargs.scm.
4198 ** New syntax: and-let*
4199 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4201 Syntax: (land* (<clause> ...) <body> ...)
4202 Each <clause> should have one of the following forms:
4203 (<variable> <expression>)
4206 Each <variable> or <bound-variable> should be an identifier. Each
4207 <expression> should be a valid expression. The <body> should be a
4208 possibly empty sequence of expressions, like the <body> of a
4211 Semantics: A LAND* expression is evaluated by evaluating the
4212 <expression> or <bound-variable> of each of the <clause>s from
4213 left to right. The value of the first <expression> or
4214 <bound-variable> that evaluates to a false value is returned; the
4215 remaining <expression>s and <bound-variable>s are not evaluated.
4216 The <body> forms are evaluated iff all the <expression>s and
4217 <bound-variable>s evaluate to true values.
4219 The <expression>s and the <body> are evaluated in an environment
4220 binding each <variable> of the preceding (<variable> <expression>)
4221 clauses to the value of the <expression>. Later bindings
4222 shadow earlier bindings.
4224 Guile's and-let* macro was contributed by Michael Livshin.
4226 ** New sorting functions
4228 *** New function: sorted? SEQUENCE LESS?
4229 Returns `#t' when the sequence argument is in non-decreasing order
4230 according to LESS? (that is, there is no adjacent pair `... x y
4231 ...' for which `(less? y x)').
4233 Returns `#f' when the sequence contains at least one out-of-order
4234 pair. It is an error if the sequence is neither a list nor a
4237 *** New function: merge LIST1 LIST2 LESS?
4238 LIST1 and LIST2 are sorted lists.
4239 Returns the sorted list of all elements in LIST1 and LIST2.
4241 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4242 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4243 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4244 (Here "<" should read "comes before".)
4246 *** New procedure: merge! LIST1 LIST2 LESS?
4247 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4248 the result. If the code is compiled, and LESS? constructs no new
4249 pairs, no pairs at all will be allocated. The first pair of the
4250 result will be either the first pair of LIST1 or the first pair of
4253 *** New function: sort SEQUENCE LESS?
4254 Accepts either a list or a vector, and returns a new sequence
4255 which is sorted. The new sequence is the same type as the input.
4256 Always `(sorted? (sort sequence less?) less?)'. The original
4257 sequence is not altered in any way. The new sequence shares its
4258 elements with the old one; no elements are copied.
4260 *** New procedure: sort! SEQUENCE LESS
4261 Returns its sorted result in the original boxes. No new storage is
4262 allocated at all. Proper usage: (set! slist (sort! slist <))
4264 *** New function: stable-sort SEQUENCE LESS?
4265 Similar to `sort' but stable. That is, if "equal" elements are
4266 ordered a < b in the original sequence, they will have the same order
4269 *** New function: stable-sort! SEQUENCE LESS?
4270 Similar to `sort!' but stable.
4271 Uses temporary storage when sorting vectors.
4273 *** New functions: sort-list, sort-list!
4274 Added for compatibility with scsh.
4276 ** New built-in random number support
4278 *** New function: random N [STATE]
4279 Accepts a positive integer or real N and returns a number of the
4280 same type between zero (inclusive) and N (exclusive). The values
4281 returned have a uniform distribution.
4283 The optional argument STATE must be of the type produced by
4284 `copy-random-state' or `seed->random-state'. It defaults to the value
4285 of the variable `*random-state*'. This object is used to maintain the
4286 state of the pseudo-random-number generator and is altered as a side
4287 effect of the `random' operation.
4289 *** New variable: *random-state*
4290 Holds a data structure that encodes the internal state of the
4291 random-number generator that `random' uses by default. The nature
4292 of this data structure is implementation-dependent. It may be
4293 printed out and successfully read back in, but may or may not
4294 function correctly as a random-number state object in another
4297 *** New function: copy-random-state [STATE]
4298 Returns a new object of type suitable for use as the value of the
4299 variable `*random-state*' and as a second argument to `random'.
4300 If argument STATE is given, a copy of it is returned. Otherwise a
4301 copy of `*random-state*' is returned.
4303 *** New function: seed->random-state SEED
4304 Returns a new object of type suitable for use as the value of the
4305 variable `*random-state*' and as a second argument to `random'.
4306 SEED is a string or a number. A new state is generated and
4307 initialized using SEED.
4309 *** New function: random:uniform [STATE]
4310 Returns an uniformly distributed inexact real random number in the
4311 range between 0 and 1.
4313 *** New procedure: random:solid-sphere! VECT [STATE]
4314 Fills VECT with inexact real random numbers the sum of whose
4315 squares is less than 1.0. Thinking of VECT as coordinates in
4316 space of dimension N = `(vector-length VECT)', the coordinates are
4317 uniformly distributed within the unit N-shere. The sum of the
4318 squares of the numbers is returned. VECT can be either a vector
4319 or a uniform vector of doubles.
4321 *** New procedure: random:hollow-sphere! VECT [STATE]
4322 Fills VECT with inexact real random numbers the sum of whose squares
4323 is equal to 1.0. Thinking of VECT as coordinates in space of
4324 dimension n = `(vector-length VECT)', the coordinates are uniformly
4325 distributed over the surface of the unit n-shere. VECT can be either
4326 a vector or a uniform vector of doubles.
4328 *** New function: random:normal [STATE]
4329 Returns an inexact real in a normal distribution with mean 0 and
4330 standard deviation 1. For a normal distribution with mean M and
4331 standard deviation D use `(+ M (* D (random:normal)))'.
4333 *** New procedure: random:normal-vector! VECT [STATE]
4334 Fills VECT with inexact real random numbers which are independent and
4335 standard normally distributed (i.e., with mean 0 and variance 1).
4336 VECT can be either a vector or a uniform vector of doubles.
4338 *** New function: random:exp STATE
4339 Returns an inexact real in an exponential distribution with mean 1.
4340 For an exponential distribution with mean U use (* U (random:exp)).
4342 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4344 These functions now operate on numbers in the range of a C unsigned
4347 These functions used to operate on numbers in the range of a C signed
4348 long; however, this seems inappropriate, because Guile integers don't
4351 ** New function: make-guardian
4352 This is an implementation of guardians as described in
4353 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4354 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4355 Programming Language Design and Implementation, June 1993
4356 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4358 ** New functions: delq1!, delv1!, delete1!
4359 These procedures behave similar to delq! and friends but delete only
4360 one object if at all.
4362 ** New function: unread-string STRING PORT
4363 Unread STRING to PORT, that is, push it back onto the port so that
4364 next read operation will work on the pushed back characters.
4366 ** unread-char can now be called multiple times
4367 If unread-char is called multiple times, the unread characters will be
4368 read again in last-in first-out order.
4370 ** the procedures uniform-array-read! and uniform-array-write! now
4371 work on any kind of port, not just ports which are open on a file.
4373 ** Now 'l' in a port mode requests line buffering.
4375 ** The procedure truncate-file now works on string ports as well
4376 as file ports. If the size argument is omitted, the current
4377 file position is used.
4379 ** new procedure: seek PORT/FDES OFFSET WHENCE
4380 The arguments are the same as for the old fseek procedure, but it
4381 works on string ports as well as random-access file ports.
4383 ** the fseek procedure now works on string ports, since it has been
4384 redefined using seek.
4386 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4387 size is not supplied.
4389 ** the newline procedure no longer flushes the port if it's not
4390 line-buffered: previously it did if it was the current output port.
4392 ** open-pipe and close-pipe are no longer primitive procedures, but
4393 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4395 ** the freopen procedure has been removed.
4397 ** new procedure: drain-input PORT
4398 Drains PORT's read buffers (including any pushed-back characters)
4399 and returns the contents as a single string.
4401 ** New function: map-in-order PROC LIST1 LIST2 ...
4402 Version of `map' which guarantees that the procedure is applied to the
4403 lists in serial order.
4405 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4406 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4407 now obsolete and will go away in release 1.5.
4409 ** New syntax: collect BODY1 ...
4410 Version of `begin' which returns a list of the results of the body
4411 forms instead of the result of the last body form. In contrast to
4412 `begin', `collect' allows an empty body.
4414 ** New functions: read-history FILENAME, write-history FILENAME
4415 Read/write command line history from/to file. Returns #t on success
4416 and #f if an error occured.
4418 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4420 These procedures return a list of definitions available in the specified
4421 argument, a relative module reference. In the case of no argument,
4422 `(current-module)' is now consulted for definitions to return, instead
4423 of simply returning #f, the former behavior.
4425 ** The #/ syntax for lists is no longer supported.
4427 Earlier versions of Scheme accepted this syntax, but printed a
4430 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4432 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4435 * Changes to the gh_ interface
4439 Now takes a second argument which is the result array. If this
4440 pointer is NULL, a new array is malloced (the old behaviour).
4442 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4443 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4447 * Changes to the scm_ interface
4449 ** Function: scm_make_named_hook (char* name, int n_args)
4451 Creates a hook in the same way as make-hook above but also
4452 binds a variable named NAME to it.
4454 This is the typical way of creating a hook from C code.
4456 Currently, the variable is created in the "current" module. This
4457 might change when we get the new module system.
4459 ** The smob interface
4461 The interface for creating smobs has changed. For documentation, see
4462 data-rep.info (made from guile-core/doc/data-rep.texi).
4464 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4466 >>> This function will be removed in 1.3.4. <<<
4470 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4471 This function adds a new smob type, named NAME, with instance size
4472 SIZE to the system. The return value is a tag that is used in
4473 creating instances of the type. If SIZE is 0, then no memory will
4474 be allocated when instances of the smob are created, and nothing
4475 will be freed by the default free function.
4477 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4478 This function sets the smob marking procedure for the smob type
4479 specified by the tag TC. TC is the tag returned by
4480 `scm_make_smob_type'.
4482 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4483 This function sets the smob freeing procedure for the smob type
4484 specified by the tag TC. TC is the tag returned by
4485 `scm_make_smob_type'.
4487 *** Function: void scm_set_smob_print (tc, print)
4489 - Function: void scm_set_smob_print (long tc,
4490 scm_sizet (*print) (SCM,
4494 This function sets the smob printing procedure for the smob type
4495 specified by the tag TC. TC is the tag returned by
4496 `scm_make_smob_type'.
4498 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4499 This function sets the smob equality-testing predicate for the
4500 smob type specified by the tag TC. TC is the tag returned by
4501 `scm_make_smob_type'.
4503 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4504 Make VALUE contain a smob instance of the type with type code TC and
4505 smob data DATA. VALUE must be previously declared as C type `SCM'.
4507 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4508 This macro expands to a block of code that creates a smob instance
4509 of the type with type code TC and smob data DATA, and returns that
4510 `SCM' value. It should be the last piece of code in a block.
4512 ** The interfaces for using I/O ports and implementing port types
4513 (ptobs) have changed significantly. The new interface is based on
4514 shared access to buffers and a new set of ptob procedures.
4516 *** scm_newptob has been removed
4520 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4522 - Function: SCM scm_make_port_type (char *type_name,
4523 int (*fill_buffer) (SCM port),
4524 void (*write_flush) (SCM port));
4526 Similarly to the new smob interface, there is a set of function
4527 setters by which the user can customize the behaviour of his port
4528 type. See ports.h (scm_set_port_XXX).
4530 ** scm_strport_to_string: New function: creates a new string from
4531 a string port's buffer.
4533 ** Plug in interface for random number generators
4534 The variable `scm_the_rng' in random.c contains a value and three
4535 function pointers which together define the current random number
4536 generator being used by the Scheme level interface and the random
4537 number library functions.
4539 The user is free to replace the default generator with the generator
4542 *** Variable: size_t scm_the_rng.rstate_size
4543 The size of the random state type used by the current RNG
4546 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4547 Given the random STATE, return 32 random bits.
4549 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4550 Seed random state STATE using string S of length N.
4552 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4553 Given random state STATE, return a malloced copy.
4556 The default RNG is the MWC (Multiply With Carry) random number
4557 generator described by George Marsaglia at the Department of
4558 Statistics and Supercomputer Computations Research Institute, The
4559 Florida State University (http://stat.fsu.edu/~geo).
4561 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4562 passes all tests in the DIEHARD test suite
4563 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4564 costs one multiply and one add on platforms which either supports long
4565 longs (gcc does this on most systems) or have 64 bit longs. The cost
4566 is four multiply on other systems but this can be optimized by writing
4567 scm_i_uniform32 in assembler.
4569 These functions are provided through the scm_the_rng interface for use
4570 by libguile and the application.
4572 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4573 Given the random STATE, return 32 random bits.
4574 Don't use this function directly. Instead go through the plugin
4575 interface (see "Plug in interface" above).
4577 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4578 Initialize STATE using SEED of length N.
4580 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4581 Return a malloc:ed copy of STATE. This function can easily be re-used
4582 in the interfaces to other RNGs.
4584 ** Random number library functions
4585 These functions use the current RNG through the scm_the_rng interface.
4586 It might be a good idea to use these functions from your C code so
4587 that only one random generator is used by all code in your program.
4589 The default random state is stored in:
4591 *** Variable: SCM scm_var_random_state
4592 Contains the vcell of the Scheme variable "*random-state*" which is
4593 used as default state by all random number functions in the Scheme
4598 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4600 *** Function: scm_rstate *scm_c_default_rstate (void)
4601 This is a convenience function which returns the value of
4602 scm_var_random_state. An error message is generated if this value
4603 isn't a random state.
4605 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4606 Make a new random state from the string SEED of length LENGTH.
4608 It is generally not a good idea to use multiple random states in a
4609 program. While subsequent random numbers generated from one random
4610 state are guaranteed to be reasonably independent, there is no such
4611 guarantee for numbers generated from different random states.
4613 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4614 Return 32 random bits.
4616 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4617 Return a sample from the uniform(0,1) distribution.
4619 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4620 Return a sample from the normal(0,1) distribution.
4622 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4623 Return a sample from the exp(1) distribution.
4625 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4626 Return a sample from the discrete uniform(0,M) distribution.
4628 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4629 Return a sample from the discrete uniform(0,M) distribution.
4630 M must be a bignum object. The returned value may be an INUM.
4634 Changes in Guile 1.3 (released Monday, October 19, 1998):
4636 * Changes to the distribution
4638 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4639 To avoid conflicts, programs should name environment variables after
4640 themselves, except when there's a common practice establishing some
4643 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4644 giving the former precedence, and printing a warning message if the
4645 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4647 ** The header files related to multi-byte characters have been removed.
4648 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4649 which referred to these explicitly will probably need to be rewritten,
4650 since the support for the variant string types has been removed; see
4653 ** The header files append.h and sequences.h have been removed. These
4654 files implemented non-R4RS operations which would encourage
4655 non-portable programming style and less easy-to-read code.
4657 * Changes to the stand-alone interpreter
4659 ** New procedures have been added to implement a "batch mode":
4661 *** Function: batch-mode?
4663 Returns a boolean indicating whether the interpreter is in batch
4666 *** Function: set-batch-mode?! ARG
4668 If ARG is true, switches the interpreter to batch mode. The `#f'
4669 case has not been implemented.
4671 ** Guile now provides full command-line editing, when run interactively.
4672 To use this feature, you must have the readline library installed.
4673 The Guile build process will notice it, and automatically include
4676 The readline library is available via anonymous FTP from any GNU
4677 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4679 ** the-last-stack is now a fluid.
4681 * Changes to the procedure for linking libguile with your programs
4683 ** You can now use the `guile-config' utility to build programs that use Guile.
4685 Guile now includes a command-line utility called `guile-config', which
4686 can provide information about how to compile and link programs that
4689 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4690 You should include this command's output on the command line you use
4691 to compile C or C++ code that #includes the Guile header files. It's
4692 usually just a `-I' flag to help the compiler find the Guile headers.
4695 *** `guile-config link' prints any linker flags necessary to link with Guile.
4697 This command writes to its standard output a list of flags which you
4698 must pass to the linker to link your code against the Guile library.
4699 The flags include '-lguile' itself, any other libraries the Guile
4700 library depends upon, and any `-L' flags needed to help the linker
4701 find those libraries.
4703 For example, here is a Makefile rule that builds a program named 'foo'
4704 from the object files ${FOO_OBJECTS}, and links them against Guile:
4707 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4709 Previous Guile releases recommended that you use autoconf to detect
4710 which of a predefined set of libraries were present on your system.
4711 It is more robust to use `guile-config', since it records exactly which
4712 libraries the installed Guile library requires.
4714 This was originally called `build-guile', but was renamed to
4715 `guile-config' before Guile 1.3 was released, to be consistent with
4716 the analogous script for the GTK+ GUI toolkit, which is called
4720 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4722 If you are using the GNU autoconf package to configure your program,
4723 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4724 (described above) and gather the necessary values for use in your
4727 The GUILE_FLAGS macro expands to configure script code which runs the
4728 `guile-config' script, to find out where Guile's header files and
4729 libraries are installed. It sets two variables, marked for
4730 substitution, as by AC_SUBST.
4732 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4733 code that uses Guile header files. This is almost always just a
4736 GUILE_LDFLAGS --- flags to pass to the linker to link a
4737 program against Guile. This includes `-lguile' for the Guile
4738 library itself, any libraries that Guile itself requires (like
4739 -lqthreads), and so on. It may also include a -L flag to tell the
4740 compiler where to find the libraries.
4742 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4743 directory of the Guile distribution. You can copy it into your
4744 package's aclocal.m4 file, and then use it in your configure.in file.
4746 If you are using the `aclocal' program, distributed with GNU automake,
4747 to maintain your aclocal.m4 file, the Guile installation process
4748 installs guile.m4 where aclocal will find it. All you need to do is
4749 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4750 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4754 * Changes to Scheme functions and syntax
4756 ** Multi-byte strings have been removed, as have multi-byte and wide
4757 ports. We felt that these were the wrong approach to
4758 internationalization support.
4760 ** New function: readline [PROMPT]
4761 Read a line from the terminal, and allow the user to edit it,
4762 prompting with PROMPT. READLINE provides a large set of Emacs-like
4763 editing commands, lets the user recall previously typed lines, and
4764 works on almost every kind of terminal, including dumb terminals.
4766 READLINE assumes that the cursor is at the beginning of the line when
4767 it is invoked. Thus, you can't print a prompt yourself, and then call
4768 READLINE; you need to package up your prompt as a string, pass it to
4769 the function, and let READLINE print the prompt itself. This is
4770 because READLINE needs to know the prompt's screen width.
4772 For Guile to provide this function, you must have the readline
4773 library, version 2.1 or later, installed on your system. Readline is
4774 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4775 any GNU mirror site.
4777 See also ADD-HISTORY function.
4779 ** New function: add-history STRING
4780 Add STRING as the most recent line in the history used by the READLINE
4781 command. READLINE does not add lines to the history itself; you must
4782 call ADD-HISTORY to make previous input available to the user.
4784 ** The behavior of the read-line function has changed.
4786 This function now uses standard C library functions to read the line,
4787 for speed. This means that it doesn not respect the value of
4788 scm-line-incrementors; it assumes that lines are delimited with
4791 (Note that this is read-line, the function that reads a line of text
4792 from a port, not readline, the function that reads a line from a
4793 terminal, providing full editing capabilities.)
4795 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4797 This module provides some simple argument parsing. It exports one
4800 Function: getopt-gnu-style ARG-LS
4801 Parse a list of program arguments into an alist of option
4804 Each item in the list of program arguments is examined to see if
4805 it meets the syntax of a GNU long-named option. An argument like
4806 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4807 returned alist, where MUMBLE is a keyword object with the same
4808 name as the argument. An argument like `--MUMBLE=FROB' produces
4809 an element of the form (MUMBLE . FROB), where FROB is a string.
4811 As a special case, the returned alist also contains a pair whose
4812 car is the symbol `rest'. The cdr of this pair is a list
4813 containing all the items in the argument list that are not options
4814 of the form mentioned above.
4816 The argument `--' is treated specially: all items in the argument
4817 list appearing after such an argument are not examined, and are
4818 returned in the special `rest' list.
4820 This function does not parse normal single-character switches.
4821 You will need to parse them out of the `rest' list yourself.
4823 ** The read syntax for byte vectors and short vectors has changed.
4825 Instead of #bytes(...), write #y(...).
4827 Instead of #short(...), write #h(...).
4829 This may seem nutty, but, like the other uniform vectors, byte vectors
4830 and short vectors want to have the same print and read syntax (and,
4831 more basic, want to have read syntax!). Changing the read syntax to
4832 use multiple characters after the hash sign breaks with the
4833 conventions used in R5RS and the conventions used for the other
4834 uniform vectors. It also introduces complexity in the current reader,
4835 both on the C and Scheme levels. (The Right solution is probably to
4836 change the syntax and prototypes for uniform vectors entirely.)
4839 ** The new module (ice-9 session) provides useful interactive functions.
4841 *** New procedure: (apropos REGEXP OPTION ...)
4843 Display a list of top-level variables whose names match REGEXP, and
4844 the modules they are imported from. Each OPTION should be one of the
4847 value --- Show the value of each matching variable.
4848 shadow --- Show bindings shadowed by subsequently imported modules.
4849 full --- Same as both `shadow' and `value'.
4853 guile> (apropos "trace" 'full)
4854 debug: trace #<procedure trace args>
4855 debug: untrace #<procedure untrace args>
4856 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4857 the-scm-module: before-backtrace-hook ()
4858 the-scm-module: backtrace #<primitive-procedure backtrace>
4859 the-scm-module: after-backtrace-hook ()
4860 the-scm-module: has-shown-backtrace-hint? #f
4863 ** There are new functions and syntax for working with macros.
4865 Guile implements macros as a special object type. Any variable whose
4866 top-level binding is a macro object acts as a macro. The macro object
4867 specifies how the expression should be transformed before evaluation.
4869 *** Macro objects now print in a reasonable way, resembling procedures.
4871 *** New function: (macro? OBJ)
4872 True iff OBJ is a macro object.
4874 *** New function: (primitive-macro? OBJ)
4875 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4876 macro transformers, implemented in eval.c rather than Scheme code.
4878 Why do we have this function?
4879 - For symmetry with procedure? and primitive-procedure?,
4880 - to allow custom print procedures to tell whether a macro is
4881 primitive, and display it differently, and
4882 - to allow compilers and user-written evaluators to distinguish
4883 builtin special forms from user-defined ones, which could be
4886 *** New function: (macro-type OBJ)
4887 Return a value indicating what kind of macro OBJ is. Possible return
4890 The symbol `syntax' --- a macro created by procedure->syntax.
4891 The symbol `macro' --- a macro created by procedure->macro.
4892 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4893 The boolean #f --- if OBJ is not a macro object.
4895 *** New function: (macro-name MACRO)
4896 Return the name of the macro object MACRO's procedure, as returned by
4899 *** New function: (macro-transformer MACRO)
4900 Return the transformer procedure for MACRO.
4902 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4904 Specify a new macro expander to use in the current module. Each
4905 MODULE is a module name, with the same meaning as in the `use-modules'
4906 form; each named module's exported bindings are added to the current
4907 top-level environment. TRANSFORMER is an expression evaluated in the
4908 resulting environment which must yield a procedure to use as the
4909 module's eval transformer: every expression evaluated in this module
4910 is passed to this function, and the result passed to the Guile
4913 *** macro-eval! is removed. Use local-eval instead.
4915 ** Some magic has been added to the printer to better handle user
4916 written printing routines (like record printers, closure printers).
4918 The problem is that these user written routines must have access to
4919 the current `print-state' to be able to handle fancy things like
4920 detection of circular references. These print-states have to be
4921 passed to the builtin printing routines (display, write, etc) to
4922 properly continue the print chain.
4924 We didn't want to change all existing print code so that it
4925 explicitly passes thru a print state in addition to a port. Instead,
4926 we extented the possible values that the builtin printing routines
4927 accept as a `port'. In addition to a normal port, they now also take
4928 a pair of a normal port and a print-state. Printing will go to the
4929 port and the print-state will be used to control the detection of
4930 circular references, etc. If the builtin function does not care for a
4931 print-state, it is simply ignored.
4933 User written callbacks are now called with such a pair as their
4934 `port', but because every function now accepts this pair as a PORT
4935 argument, you don't have to worry about that. In fact, it is probably
4936 safest to not check for these pairs.
4938 However, it is sometimes necessary to continue a print chain on a
4939 different port, for example to get a intermediate string
4940 representation of the printed value, mangle that string somehow, and
4941 then to finally print the mangled string. Use the new function
4943 inherit-print-state OLD-PORT NEW-PORT
4945 for this. It constructs a new `port' that prints to NEW-PORT but
4946 inherits the print-state of OLD-PORT.
4948 ** struct-vtable-offset renamed to vtable-offset-user
4950 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
4952 ** There is now a third optional argument to make-vtable-vtable
4953 (and fourth to make-struct) when constructing new types (vtables).
4954 This argument initializes field vtable-index-printer of the vtable.
4956 ** The detection of circular references has been extended to structs.
4957 That is, a structure that -- in the process of being printed -- prints
4958 itself does not lead to infinite recursion.
4960 ** There is now some basic support for fluids. Please read
4961 "libguile/fluid.h" to find out more. It is accessible from Scheme with
4962 the following functions and macros:
4964 Function: make-fluid
4966 Create a new fluid object. Fluids are not special variables or
4967 some other extension to the semantics of Scheme, but rather
4968 ordinary Scheme objects. You can store them into variables (that
4969 are still lexically scoped, of course) or into any other place you
4970 like. Every fluid has a initial value of `#f'.
4972 Function: fluid? OBJ
4974 Test whether OBJ is a fluid.
4976 Function: fluid-ref FLUID
4977 Function: fluid-set! FLUID VAL
4979 Access/modify the fluid FLUID. Modifications are only visible
4980 within the current dynamic root (that includes threads).
4982 Function: with-fluids* FLUIDS VALUES THUNK
4984 FLUIDS is a list of fluids and VALUES a corresponding list of
4985 values for these fluids. Before THUNK gets called the values are
4986 installed in the fluids and the old values of the fluids are
4987 saved in the VALUES list. When the flow of control leaves THUNK
4988 or reenters it, the values get swapped again. You might think of
4989 this as a `safe-fluid-excursion'. Note that the VALUES list is
4990 modified by `with-fluids*'.
4992 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
4994 The same as `with-fluids*' but with a different syntax. It looks
4995 just like `let', but both FLUID and VALUE are evaluated. Remember,
4996 fluids are not special variables but ordinary objects. FLUID
4997 should evaluate to a fluid.
4999 ** Changes to system call interfaces:
5001 *** close-port, close-input-port and close-output-port now return a
5002 boolean instead of an `unspecified' object. #t means that the port
5003 was successfully closed, while #f means it was already closed. It is
5004 also now possible for these procedures to raise an exception if an
5005 error occurs (some errors from write can be delayed until close.)
5007 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5010 *** the third argument to fcntl is now optional.
5012 *** the first argument to chown can now be a file descriptor or a port.
5014 *** the argument to stat can now be a port.
5016 *** The following new procedures have been added (most use scsh
5019 *** procedure: close PORT/FD
5020 Similar to close-port (*note close-port: Closing Ports.), but also
5021 works on file descriptors. A side effect of closing a file
5022 descriptor is that any ports using that file descriptor are moved
5023 to a different file descriptor and have their revealed counts set
5026 *** procedure: port->fdes PORT
5027 Returns the integer file descriptor underlying PORT. As a side
5028 effect the revealed count of PORT is incremented.
5030 *** procedure: fdes->ports FDES
5031 Returns a list of existing ports which have FDES as an underlying
5032 file descriptor, without changing their revealed counts.
5034 *** procedure: fdes->inport FDES
5035 Returns an existing input port which has FDES as its underlying
5036 file descriptor, if one exists, and increments its revealed count.
5037 Otherwise, returns a new input port with a revealed count of 1.
5039 *** procedure: fdes->outport FDES
5040 Returns an existing output port which has FDES as its underlying
5041 file descriptor, if one exists, and increments its revealed count.
5042 Otherwise, returns a new output port with a revealed count of 1.
5044 The next group of procedures perform a `dup2' system call, if NEWFD
5045 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5046 duplicated can be supplied as an integer or contained in a port. The
5047 type of value returned varies depending on which procedure is used.
5049 All procedures also have the side effect when performing `dup2' that
5050 any ports using NEWFD are moved to a different file descriptor and have
5051 their revealed counts set to zero.
5053 *** procedure: dup->fdes PORT/FD [NEWFD]
5054 Returns an integer file descriptor.
5056 *** procedure: dup->inport PORT/FD [NEWFD]
5057 Returns a new input port using the new file descriptor.
5059 *** procedure: dup->outport PORT/FD [NEWFD]
5060 Returns a new output port using the new file descriptor.
5062 *** procedure: dup PORT/FD [NEWFD]
5063 Returns a new port if PORT/FD is a port, with the same mode as the
5064 supplied port, otherwise returns an integer file descriptor.
5066 *** procedure: dup->port PORT/FD MODE [NEWFD]
5067 Returns a new port using the new file descriptor. MODE supplies a
5068 mode string for the port (*note open-file: File Ports.).
5070 *** procedure: setenv NAME VALUE
5071 Modifies the environment of the current process, which is also the
5072 default environment inherited by child processes.
5074 If VALUE is `#f', then NAME is removed from the environment.
5075 Otherwise, the string NAME=VALUE is added to the environment,
5076 replacing any existing string with name matching NAME.
5078 The return value is unspecified.
5080 *** procedure: truncate-file OBJ SIZE
5081 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5082 can be a string containing a file name or an integer file
5083 descriptor or port open for output on the file. The underlying
5084 system calls are `truncate' and `ftruncate'.
5086 The return value is unspecified.
5088 *** procedure: setvbuf PORT MODE [SIZE]
5089 Set the buffering mode for PORT. MODE can be:
5097 block buffered, using a newly allocated buffer of SIZE bytes.
5098 However if SIZE is zero or unspecified, the port will be made
5101 This procedure should not be used after I/O has been performed with
5104 Ports are usually block buffered by default, with a default buffer
5105 size. Procedures e.g., *Note open-file: File Ports, which accept a
5106 mode string allow `0' to be added to request an unbuffered port.
5108 *** procedure: fsync PORT/FD
5109 Copies any unwritten data for the specified output file descriptor
5110 to disk. If PORT/FD is a port, its buffer is flushed before the
5111 underlying file descriptor is fsync'd. The return value is
5114 *** procedure: open-fdes PATH FLAGS [MODES]
5115 Similar to `open' but returns a file descriptor instead of a port.
5117 *** procedure: execle PATH ENV [ARG] ...
5118 Similar to `execl', but the environment of the new process is
5119 specified by ENV, which must be a list of strings as returned by
5120 the `environ' procedure.
5122 This procedure is currently implemented using the `execve' system
5123 call, but we call it `execle' because of its Scheme calling
5126 *** procedure: strerror ERRNO
5127 Returns the Unix error message corresponding to ERRNO, an integer.
5129 *** procedure: primitive-exit [STATUS]
5130 Terminate the current process without unwinding the Scheme stack.
5131 This is would typically be useful after a fork. The exit status
5132 is STATUS if supplied, otherwise zero.
5134 *** procedure: times
5135 Returns an object with information about real and processor time.
5136 The following procedures accept such an object as an argument and
5137 return a selected component:
5140 The current real time, expressed as time units relative to an
5144 The CPU time units used by the calling process.
5147 The CPU time units used by the system on behalf of the
5151 The CPU time units used by terminated child processes of the
5152 calling process, whose status has been collected (e.g., using
5156 Similarly, the CPU times units used by the system on behalf of
5157 terminated child processes.
5159 ** Removed: list-length
5160 ** Removed: list-append, list-append!
5161 ** Removed: list-reverse, list-reverse!
5163 ** array-map renamed to array-map!
5165 ** serial-array-map renamed to serial-array-map!
5167 ** catch doesn't take #f as first argument any longer
5169 Previously, it was possible to pass #f instead of a key to `catch'.
5170 That would cause `catch' to pass a jump buffer object to the procedure
5171 passed as second argument. The procedure could then use this jump
5172 buffer objekt as an argument to throw.
5174 This mechanism has been removed since its utility doesn't motivate the
5175 extra complexity it introduces.
5177 ** The `#/' notation for lists now provokes a warning message from Guile.
5178 This syntax will be removed from Guile in the near future.
5180 To disable the warning message, set the GUILE_HUSH environment
5181 variable to any non-empty value.
5183 ** The newline character now prints as `#\newline', following the
5184 normal Scheme notation, not `#\nl'.
5186 * Changes to the gh_ interface
5188 ** The gh_enter function now takes care of loading the Guile startup files.
5189 gh_enter works by calling scm_boot_guile; see the remarks below.
5191 ** Function: void gh_write (SCM x)
5193 Write the printed representation of the scheme object x to the current
5194 output port. Corresponds to the scheme level `write'.
5196 ** gh_list_length renamed to gh_length.
5198 ** vector handling routines
5200 Several major changes. In particular, gh_vector() now resembles
5201 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5202 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5203 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5204 vector-related gh_ functions have been implemented.
5206 ** pair and list routines
5208 Implemented several of the R4RS pair and list functions that were
5211 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5213 New function. Converts double arrays back and forth between Scheme
5216 * Changes to the scm_ interface
5218 ** The function scm_boot_guile now takes care of loading the startup files.
5220 Guile's primary initialization function, scm_boot_guile, now takes
5221 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5222 Guile, define the module system, and put together some standard
5223 bindings. It also loads `init.scm', which is intended to hold
5224 site-specific initialization code.
5226 Since Guile cannot operate properly until boot-9.scm is loaded, there
5227 is no reason to separate loading boot-9.scm from Guile's other
5228 initialization processes.
5230 This job used to be done by scm_compile_shell_switches, which didn't
5231 make much sense; in particular, it meant that people using Guile for
5232 non-shell-like applications had to jump through hoops to get Guile
5233 initialized properly.
5235 ** The function scm_compile_shell_switches no longer loads the startup files.
5236 Now, Guile always loads the startup files, whenever it is initialized;
5237 see the notes above for scm_boot_guile and scm_load_startup_files.
5239 ** Function: scm_load_startup_files
5240 This new function takes care of loading Guile's initialization file
5241 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5242 this is always called by the Guile initialization process, it's
5243 probably not too useful to call this yourself, but it's there anyway.
5245 ** The semantics of smob marking have changed slightly.
5247 The smob marking function (the `mark' member of the scm_smobfuns
5248 structure) is no longer responsible for setting the mark bit on the
5249 smob. The generic smob handling code in the garbage collector will
5250 set this bit. The mark function need only ensure that any other
5251 objects the smob refers to get marked.
5253 Note that this change means that the smob's GC8MARK bit is typically
5254 already set upon entry to the mark function. Thus, marking functions
5255 which look like this:
5258 if (SCM_GC8MARKP (ptr))
5260 SCM_SETGC8MARK (ptr);
5261 ... mark objects to which the smob refers ...
5264 are now incorrect, since they will return early, and fail to mark any
5265 other objects the smob refers to. Some code in the Guile library used
5268 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5270 If you have implemented your own I/O port type, by writing the
5271 functions required by the scm_ptobfuns and then calling scm_newptob,
5272 you will need to change your functions slightly.
5274 The functions in a scm_ptobfuns structure now expect the port itself
5275 as their argument; they used to expect the `stream' member of the
5276 port's scm_port_table structure. This allows functions in an
5277 scm_ptobfuns structure to easily access the port's cell (and any flags
5278 it its CAR), and the port's scm_port_table structure.
5280 Guile now passes the I/O port itself as the `port' argument in the
5281 following scm_ptobfuns functions:
5283 int (*free) (SCM port);
5284 int (*fputc) (int, SCM port);
5285 int (*fputs) (char *, SCM port);
5286 scm_sizet (*fwrite) SCM_P ((char *ptr,
5290 int (*fflush) (SCM port);
5291 int (*fgetc) (SCM port);
5292 int (*fclose) (SCM port);
5294 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5297 If you have existing code which defines its own port types, it is easy
5298 to convert your code to the new interface; simply apply SCM_STREAM to
5299 the port argument to yield the value you code used to expect.
5301 Note that since both the port and the stream have the same type in the
5302 C code --- they are both SCM values --- the C compiler will not remind
5303 you if you forget to update your scm_ptobfuns functions.
5306 ** Function: int scm_internal_select (int fds,
5310 struct timeval *timeout);
5312 This is a replacement for the `select' function provided by the OS.
5313 It enables I/O blocking and sleeping to happen for one cooperative
5314 thread without blocking other threads. It also avoids busy-loops in
5315 these situations. It is intended that all I/O blocking and sleeping
5316 will finally go through this function. Currently, this function is
5317 only available on systems providing `gettimeofday' and `select'.
5319 ** Function: SCM scm_internal_stack_catch (SCM tag,
5320 scm_catch_body_t body,
5322 scm_catch_handler_t handler,
5325 A new sibling to the other two C level `catch' functions
5326 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5327 the stack to be saved automatically into the variable `the-last-stack'
5328 (scm_the_last_stack_var) on error. This is necessary if you want to
5329 use advanced error reporting, such as calling scm_display_error and
5330 scm_display_backtrace. (They both take a stack object as argument.)
5332 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5334 scm_catch_handler_t handler,
5337 Spawns a new thread. It does a job similar to
5338 scm_call_with_new_thread but takes arguments more suitable when
5339 spawning threads from application C code.
5341 ** The hook scm_error_callback has been removed. It was originally
5342 intended as a way for the user to install his own error handler. But
5343 that method works badly since it intervenes between throw and catch,
5344 thereby changing the semantics of expressions like (catch #t ...).
5345 The correct way to do it is to use one of the C level catch functions
5346 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5348 ** Removed functions:
5350 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5351 scm_list_reverse, scm_list_reverse_x
5353 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5355 These can be used for pretty list creation from C. The idea is taken
5356 from Erick Gallesio's STk.
5358 ** scm_array_map renamed to scm_array_map_x
5360 ** mbstrings are now removed
5362 This means that the type codes scm_tc7_mb_string and
5363 scm_tc7_mb_substring has been removed.
5365 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5367 Since we no longer support multi-byte strings, these I/O functions
5368 have been simplified, and renamed. Here are their old names, and
5369 their new names and arguments:
5371 scm_gen_putc -> void scm_putc (int c, SCM port);
5372 scm_gen_puts -> void scm_puts (char *s, SCM port);
5373 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5374 scm_gen_getc -> void scm_getc (SCM port);
5377 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5379 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5381 SCM_TYP7S now masks away the bit which distinguishes substrings from
5384 ** scm_catch_body_t: Backward incompatible change!
5386 Body functions to scm_internal_catch and friends do not any longer
5387 take a second argument. This is because it is no longer possible to
5388 pass a #f arg to catch.
5390 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5392 The function scm_protect_object protects its argument from being freed
5393 by the garbage collector. scm_unprotect_object removes that
5396 These functions now nest properly. That is, for every object O, there
5397 is a counter which scm_protect_object(O) increments and
5398 scm_unprotect_object(O) decrements, if the counter is greater than
5399 zero. Every object's counter is zero when it is first created. If an
5400 object's counter is greater than zero, the garbage collector will not
5401 reclaim its storage.
5403 This allows you to use scm_protect_object in your code without
5404 worrying that some other function you call will call
5405 scm_unprotect_object, and allow it to be freed. Assuming that the
5406 functions you call are well-behaved, and unprotect only those objects
5407 they protect, you can follow the same rule and have confidence that
5408 objects will be freed only at appropriate times.
5411 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5413 * Changes to the distribution
5415 ** Nightly snapshots are now available from ftp.red-bean.com.
5416 The old server, ftp.cyclic.com, has been relinquished to its rightful
5419 Nightly snapshots of the Guile development sources are now available via
5420 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5422 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5423 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5425 ** To run Guile without installing it, the procedure has changed a bit.
5427 If you used a separate build directory to compile Guile, you'll need
5428 to include the build directory in SCHEME_LOAD_PATH, as well as the
5429 source directory. See the `INSTALL' file for examples.
5431 * Changes to the procedure for linking libguile with your programs
5433 ** The standard Guile load path for Scheme code now includes
5434 $(datadir)/guile (usually /usr/local/share/guile). This means that
5435 you can install your own Scheme files there, and Guile will find them.
5436 (Previous versions of Guile only checked a directory whose name
5437 contained the Guile version number, so you had to re-install or move
5438 your Scheme sources each time you installed a fresh version of Guile.)
5440 The load path also includes $(datadir)/guile/site; we recommend
5441 putting individual Scheme files there. If you want to install a
5442 package with multiple source files, create a directory for them under
5445 ** Guile 1.2 will now use the Rx regular expression library, if it is
5446 installed on your system. When you are linking libguile into your own
5447 programs, this means you will have to link against -lguile, -lqt (if
5448 you configured Guile with thread support), and -lrx.
5450 If you are using autoconf to generate configuration scripts for your
5451 application, the following lines should suffice to add the appropriate
5452 libraries to your link command:
5454 ### Find Rx, quickthreads and libguile.
5455 AC_CHECK_LIB(rx, main)
5456 AC_CHECK_LIB(qt, main)
5457 AC_CHECK_LIB(guile, scm_shell)
5459 The Guile 1.2 distribution does not contain sources for the Rx
5460 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5461 retrieve it from a GNU FTP site and install it separately.
5463 * Changes to Scheme functions and syntax
5465 ** The dynamic linking features of Guile are now enabled by default.
5466 You can disable them by giving the `--disable-dynamic-linking' option
5469 (dynamic-link FILENAME)
5471 Find the object file denoted by FILENAME (a string) and link it
5472 into the running Guile application. When everything works out,
5473 return a Scheme object suitable for representing the linked object
5474 file. Otherwise an error is thrown. How object files are
5475 searched is system dependent.
5477 (dynamic-object? VAL)
5479 Determine whether VAL represents a dynamically linked object file.
5481 (dynamic-unlink DYNOBJ)
5483 Unlink the indicated object file from the application. DYNOBJ
5484 should be one of the values returned by `dynamic-link'.
5486 (dynamic-func FUNCTION DYNOBJ)
5488 Search the C function indicated by FUNCTION (a string or symbol)
5489 in DYNOBJ and return some Scheme object that can later be used
5490 with `dynamic-call' to actually call this function. Right now,
5491 these Scheme objects are formed by casting the address of the
5492 function to `long' and converting this number to its Scheme
5495 (dynamic-call FUNCTION DYNOBJ)
5497 Call the C function indicated by FUNCTION and DYNOBJ. The
5498 function is passed no arguments and its return value is ignored.
5499 When FUNCTION is something returned by `dynamic-func', call that
5500 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5501 etc.), look it up in DYNOBJ; this is equivalent to
5503 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5505 Interrupts are deferred while the C function is executing (with
5506 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5508 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5510 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5511 some arguments and return its return value. The C function is
5512 expected to take two arguments and return an `int', just like
5515 int c_func (int argc, char **argv);
5517 ARGS must be a list of strings and is converted into an array of
5518 `char *'. The array is passed in ARGV and its size in ARGC. The
5519 return value is converted to a Scheme number and returned from the
5520 call to `dynamic-args-call'.
5522 When dynamic linking is disabled or not supported on your system,
5523 the above functions throw errors, but they are still available.
5525 Here is a small example that works on GNU/Linux:
5527 (define libc-obj (dynamic-link "libc.so"))
5528 (dynamic-args-call 'rand libc-obj '())
5530 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5532 ** The #/ syntax for module names is depreciated, and will be removed
5533 in a future version of Guile. Instead of
5541 The latter syntax is more consistent with existing Lisp practice.
5543 ** Guile now does fancier printing of structures. Structures are the
5544 underlying implementation for records, which in turn are used to
5545 implement modules, so all of these object now print differently and in
5546 a more informative way.
5548 The Scheme printer will examine the builtin variable *struct-printer*
5549 whenever it needs to print a structure object. When this variable is
5550 not `#f' it is deemed to be a procedure and will be applied to the
5551 structure object and the output port. When *struct-printer* is `#f'
5552 or the procedure return `#f' the structure object will be printed in
5553 the boring #<struct 80458270> form.
5555 This hook is used by some routines in ice-9/boot-9.scm to implement
5556 type specific printing routines. Please read the comments there about
5559 One of the more specific uses of structs are records. The printing
5560 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5561 called. It should behave like a *struct-printer* procedure (described
5564 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5565 token of the form #:NAME, where NAME has the same syntax as a Scheme
5566 symbol, is the external representation of the keyword named NAME.
5567 Keyword objects print using this syntax as well, so values containing
5568 keyword objects can be read back into Guile. When used in an
5569 expression, keywords are self-quoting objects.
5571 Guile suports this read syntax, and uses this print syntax, regardless
5572 of the current setting of the `keyword' read option. The `keyword'
5573 read option only controls whether Guile recognizes the `:NAME' syntax,
5574 which is incompatible with R4RS. (R4RS says such token represent
5577 ** Guile has regular expression support again. Guile 1.0 included
5578 functions for matching regular expressions, based on the Rx library.
5579 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5580 distribution, and thus Guile had no regular expression support. Guile
5581 1.2 again supports the most commonly used functions, and supports all
5582 of SCSH's regular expression functions.
5584 If your system does not include a POSIX regular expression library,
5585 and you have not linked Guile with a third-party regexp library such as
5586 Rx, these functions will not be available. You can tell whether your
5587 Guile installation includes regular expression support by checking
5588 whether the `*features*' list includes the `regex' symbol.
5590 *** regexp functions
5592 By default, Guile supports POSIX extended regular expressions. That
5593 means that the characters `(', `)', `+' and `?' are special, and must
5594 be escaped if you wish to match the literal characters.
5596 This regular expression interface was modeled after that implemented
5597 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5598 with SCSH regular expressions.
5600 **** Function: string-match PATTERN STR [START]
5601 Compile the string PATTERN into a regular expression and compare
5602 it with STR. The optional numeric argument START specifies the
5603 position of STR at which to begin matching.
5605 `string-match' returns a "match structure" which describes what,
5606 if anything, was matched by the regular expression. *Note Match
5607 Structures::. If STR does not match PATTERN at all,
5608 `string-match' returns `#f'.
5610 Each time `string-match' is called, it must compile its PATTERN
5611 argument into a regular expression structure. This operation is
5612 expensive, which makes `string-match' inefficient if the same regular
5613 expression is used several times (for example, in a loop). For better
5614 performance, you can compile a regular expression in advance and then
5615 match strings against the compiled regexp.
5617 **** Function: make-regexp STR [FLAGS]
5618 Compile the regular expression described by STR, and return the
5619 compiled regexp structure. If STR does not describe a legal
5620 regular expression, `make-regexp' throws a
5621 `regular-expression-syntax' error.
5623 FLAGS may be the bitwise-or of one or more of the following:
5625 **** Constant: regexp/extended
5626 Use POSIX Extended Regular Expression syntax when interpreting
5627 STR. If not set, POSIX Basic Regular Expression syntax is used.
5628 If the FLAGS argument is omitted, we assume regexp/extended.
5630 **** Constant: regexp/icase
5631 Do not differentiate case. Subsequent searches using the
5632 returned regular expression will be case insensitive.
5634 **** Constant: regexp/newline
5635 Match-any-character operators don't match a newline.
5637 A non-matching list ([^...]) not containing a newline matches a
5640 Match-beginning-of-line operator (^) matches the empty string
5641 immediately after a newline, regardless of whether the FLAGS
5642 passed to regexp-exec contain regexp/notbol.
5644 Match-end-of-line operator ($) matches the empty string
5645 immediately before a newline, regardless of whether the FLAGS
5646 passed to regexp-exec contain regexp/noteol.
5648 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5649 Match the compiled regular expression REGEXP against `str'. If
5650 the optional integer START argument is provided, begin matching
5651 from that position in the string. Return a match structure
5652 describing the results of the match, or `#f' if no match could be
5655 FLAGS may be the bitwise-or of one or more of the following:
5657 **** Constant: regexp/notbol
5658 The match-beginning-of-line operator always fails to match (but
5659 see the compilation flag regexp/newline above) This flag may be
5660 used when different portions of a string are passed to
5661 regexp-exec and the beginning of the string should not be
5662 interpreted as the beginning of the line.
5664 **** Constant: regexp/noteol
5665 The match-end-of-line operator always fails to match (but see the
5666 compilation flag regexp/newline above)
5668 **** Function: regexp? OBJ
5669 Return `#t' if OBJ is a compiled regular expression, or `#f'
5672 Regular expressions are commonly used to find patterns in one string
5673 and replace them with the contents of another string.
5675 **** Function: regexp-substitute PORT MATCH [ITEM...]
5676 Write to the output port PORT selected contents of the match
5677 structure MATCH. Each ITEM specifies what should be written, and
5678 may be one of the following arguments:
5680 * A string. String arguments are written out verbatim.
5682 * An integer. The submatch with that number is written.
5684 * The symbol `pre'. The portion of the matched string preceding
5685 the regexp match is written.
5687 * The symbol `post'. The portion of the matched string
5688 following the regexp match is written.
5690 PORT may be `#f', in which case nothing is written; instead,
5691 `regexp-substitute' constructs a string from the specified ITEMs
5694 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5695 Similar to `regexp-substitute', but can be used to perform global
5696 substitutions on STR. Instead of taking a match structure as an
5697 argument, `regexp-substitute/global' takes two string arguments: a
5698 REGEXP string describing a regular expression, and a TARGET string
5699 which should be matched against this regular expression.
5701 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5704 * A function may be supplied. When this function is called, it
5705 will be passed one argument: a match structure for a given
5706 regular expression match. It should return a string to be
5707 written out to PORT.
5709 * The `post' symbol causes `regexp-substitute/global' to recurse
5710 on the unmatched portion of STR. This *must* be supplied in
5711 order to perform global search-and-replace on STR; if it is
5712 not present among the ITEMs, then `regexp-substitute/global'
5713 will return after processing a single match.
5715 *** Match Structures
5717 A "match structure" is the object returned by `string-match' and
5718 `regexp-exec'. It describes which portion of a string, if any, matched
5719 the given regular expression. Match structures include: a reference to
5720 the string that was checked for matches; the starting and ending
5721 positions of the regexp match; and, if the regexp included any
5722 parenthesized subexpressions, the starting and ending positions of each
5725 In each of the regexp match functions described below, the `match'
5726 argument must be a match structure returned by a previous call to
5727 `string-match' or `regexp-exec'. Most of these functions return some
5728 information about the original target string that was matched against a
5729 regular expression; we will call that string TARGET for easy reference.
5731 **** Function: regexp-match? OBJ
5732 Return `#t' if OBJ is a match structure returned by a previous
5733 call to `regexp-exec', or `#f' otherwise.
5735 **** Function: match:substring MATCH [N]
5736 Return the portion of TARGET matched by subexpression number N.
5737 Submatch 0 (the default) represents the entire regexp match. If
5738 the regular expression as a whole matched, but the subexpression
5739 number N did not match, return `#f'.
5741 **** Function: match:start MATCH [N]
5742 Return the starting position of submatch number N.
5744 **** Function: match:end MATCH [N]
5745 Return the ending position of submatch number N.
5747 **** Function: match:prefix MATCH
5748 Return the unmatched portion of TARGET preceding the regexp match.
5750 **** Function: match:suffix MATCH
5751 Return the unmatched portion of TARGET following the regexp match.
5753 **** Function: match:count MATCH
5754 Return the number of parenthesized subexpressions from MATCH.
5755 Note that the entire regular expression match itself counts as a
5756 subexpression, and failed submatches are included in the count.
5758 **** Function: match:string MATCH
5759 Return the original TARGET string.
5761 *** Backslash Escapes
5763 Sometimes you will want a regexp to match characters like `*' or `$'
5764 exactly. For example, to check whether a particular string represents
5765 a menu entry from an Info node, it would be useful to match it against
5766 a regexp like `^* [^:]*::'. However, this won't work; because the
5767 asterisk is a metacharacter, it won't match the `*' at the beginning of
5768 the string. In this case, we want to make the first asterisk un-magic.
5770 You can do this by preceding the metacharacter with a backslash
5771 character `\'. (This is also called "quoting" the metacharacter, and
5772 is known as a "backslash escape".) When Guile sees a backslash in a
5773 regular expression, it considers the following glyph to be an ordinary
5774 character, no matter what special meaning it would ordinarily have.
5775 Therefore, we can make the above example work by changing the regexp to
5776 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5777 to match only a single asterisk in the target string.
5779 Since the backslash is itself a metacharacter, you may force a
5780 regexp to match a backslash in the target string by preceding the
5781 backslash with itself. For example, to find variable references in a
5782 TeX program, you might want to find occurrences of the string `\let\'
5783 followed by any number of alphabetic characters. The regular expression
5784 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5785 each match a single backslash in the target string.
5787 **** Function: regexp-quote STR
5788 Quote each special character found in STR with a backslash, and
5789 return the resulting string.
5791 *Very important:* Using backslash escapes in Guile source code (as
5792 in Emacs Lisp or C) can be tricky, because the backslash character has
5793 special meaning for the Guile reader. For example, if Guile encounters
5794 the character sequence `\n' in the middle of a string while processing
5795 Scheme code, it replaces those characters with a newline character.
5796 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5797 Several of these "escape sequences" are processed by the Guile reader
5798 before your code is executed. Unrecognized escape sequences are
5799 ignored: if the characters `\*' appear in a string, they will be
5800 translated to the single character `*'.
5802 This translation is obviously undesirable for regular expressions,
5803 since we want to be able to include backslashes in a string in order to
5804 escape regexp metacharacters. Therefore, to make sure that a backslash
5805 is preserved in a string in your Guile program, you must use *two*
5806 consecutive backslashes:
5808 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5810 The string in this example is preprocessed by the Guile reader before
5811 any code is executed. The resulting argument to `make-regexp' is the
5812 string `^\* [^:]*', which is what we really want.
5814 This also means that in order to write a regular expression that
5815 matches a single backslash character, the regular expression string in
5816 the source code must include *four* backslashes. Each consecutive pair
5817 of backslashes gets translated by the Guile reader to a single
5818 backslash, and the resulting double-backslash is interpreted by the
5819 regexp engine as matching a single backslash character. Hence:
5821 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5823 The reason for the unwieldiness of this syntax is historical. Both
5824 regular expression pattern matchers and Unix string processing systems
5825 have traditionally used backslashes with the special meanings described
5826 above. The POSIX regular expression specification and ANSI C standard
5827 both require these semantics. Attempting to abandon either convention
5828 would cause other kinds of compatibility problems, possibly more severe
5829 ones. Therefore, without extending the Scheme reader to support
5830 strings with different quoting conventions (an ungainly and confusing
5831 extension when implemented in other languages), we must adhere to this
5832 cumbersome escape syntax.
5834 * Changes to the gh_ interface
5836 * Changes to the scm_ interface
5838 * Changes to system call interfaces:
5840 ** The value returned by `raise' is now unspecified. It throws an exception
5843 *** A new procedure `sigaction' can be used to install signal handlers
5845 (sigaction signum [action] [flags])
5847 signum is the signal number, which can be specified using the value
5850 If action is omitted, sigaction returns a pair: the CAR is the current
5851 signal hander, which will be either an integer with the value SIG_DFL
5852 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5853 handles the signal, or #f if a non-Scheme procedure handles the
5854 signal. The CDR contains the current sigaction flags for the handler.
5856 If action is provided, it is installed as the new handler for signum.
5857 action can be a Scheme procedure taking one argument, or the value of
5858 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5859 whatever signal handler was installed before sigaction was first used.
5860 Flags can optionally be specified for the new handler (SA_RESTART is
5861 always used if the system provides it, so need not be specified.) The
5862 return value is a pair with information about the old handler as
5865 This interface does not provide access to the "signal blocking"
5866 facility. Maybe this is not needed, since the thread support may
5867 provide solutions to the problem of consistent access to data
5870 *** A new procedure `flush-all-ports' is equivalent to running
5871 `force-output' on every port open for output.
5873 ** Guile now provides information on how it was built, via the new
5874 global variable, %guile-build-info. This variable records the values
5875 of the standard GNU makefile directory variables as an assocation
5876 list, mapping variable names (symbols) onto directory paths (strings).
5877 For example, to find out where the Guile link libraries were
5878 installed, you can say:
5880 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5883 * Changes to the scm_ interface
5885 ** The new function scm_handle_by_message_noexit is just like the
5886 existing scm_handle_by_message function, except that it doesn't call
5887 exit to terminate the process. Instead, it prints a message and just
5888 returns #f. This might be a more appropriate catch-all handler for
5889 new dynamic roots and threads.
5892 Changes in Guile 1.1 (released Friday, May 16 1997):
5894 * Changes to the distribution.
5896 The Guile 1.0 distribution has been split up into several smaller
5898 guile-core --- the Guile interpreter itself.
5899 guile-tcltk --- the interface between the Guile interpreter and
5900 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5901 is a toolkit for building graphical user interfaces.
5902 guile-rgx-ctax --- the interface between Guile and the Rx regular
5903 expression matcher, and the translator for the Ctax
5904 programming language. These are packaged together because the
5905 Ctax translator uses Rx to parse Ctax source code.
5907 This NEWS file describes the changes made to guile-core since the 1.0
5910 We no longer distribute the documentation, since it was either out of
5911 date, or incomplete. As soon as we have current documentation, we
5916 * Changes to the stand-alone interpreter
5918 ** guile now accepts command-line arguments compatible with SCSH, Olin
5919 Shivers' Scheme Shell.
5921 In general, arguments are evaluated from left to right, but there are
5922 exceptions. The following switches stop argument processing, and
5923 stash all remaining command-line arguments as the value returned by
5924 the (command-line) function.
5925 -s SCRIPT load Scheme source code from FILE, and exit
5926 -c EXPR evalute Scheme expression EXPR, and exit
5927 -- stop scanning arguments; run interactively
5929 The switches below are processed as they are encountered.
5930 -l FILE load Scheme source code from FILE
5931 -e FUNCTION after reading script, apply FUNCTION to
5932 command line arguments
5933 -ds do -s script at this point
5934 --emacs enable Emacs protocol (experimental)
5935 -h, --help display this help and exit
5936 -v, --version display version information and exit
5937 \ read arguments from following script lines
5939 So, for example, here is a Guile script named `ekko' (thanks, Olin)
5940 which re-implements the traditional "echo" command:
5942 #!/usr/local/bin/guile -s
5945 (map (lambda (arg) (display arg) (display " "))
5949 (main (command-line))
5951 Suppose we invoke this script as follows:
5953 ekko a speckled gecko
5955 Through the magic of Unix script processing (triggered by the `#!'
5956 token at the top of the file), /usr/local/bin/guile receives the
5957 following list of command-line arguments:
5959 ("-s" "./ekko" "a" "speckled" "gecko")
5961 Unix inserts the name of the script after the argument specified on
5962 the first line of the file (in this case, "-s"), and then follows that
5963 with the arguments given to the script. Guile loads the script, which
5964 defines the `main' function, and then applies it to the list of
5965 remaining command-line arguments, ("a" "speckled" "gecko").
5967 In Unix, the first line of a script file must take the following form:
5969 #!INTERPRETER ARGUMENT
5971 where INTERPRETER is the absolute filename of the interpreter
5972 executable, and ARGUMENT is a single command-line argument to pass to
5975 You may only pass one argument to the interpreter, and its length is
5976 limited. These restrictions can be annoying to work around, so Guile
5977 provides a general mechanism (borrowed from, and compatible with,
5978 SCSH) for circumventing them.
5980 If the ARGUMENT in a Guile script is a single backslash character,
5981 `\', Guile will open the script file, parse arguments from its second
5982 and subsequent lines, and replace the `\' with them. So, for example,
5983 here is another implementation of the `ekko' script:
5985 #!/usr/local/bin/guile \
5989 (for-each (lambda (arg) (display arg) (display " "))
5993 If the user invokes this script as follows:
5995 ekko a speckled gecko
5997 Unix expands this into
5999 /usr/local/bin/guile \ ekko a speckled gecko
6001 When Guile sees the `\' argument, it replaces it with the arguments
6002 read from the second line of the script, producing:
6004 /usr/local/bin/guile -e main -s ekko a speckled gecko
6006 This tells Guile to load the `ekko' script, and apply the function
6007 `main' to the argument list ("a" "speckled" "gecko").
6009 Here is how Guile parses the command-line arguments:
6010 - Each space character terminates an argument. This means that two
6011 spaces in a row introduce an empty-string argument.
6012 - The tab character is not permitted (unless you quote it with the
6013 backslash character, as described below), to avoid confusion.
6014 - The newline character terminates the sequence of arguments, and will
6015 also terminate a final non-empty argument. (However, a newline
6016 following a space will not introduce a final empty-string argument;
6017 it only terminates the argument list.)
6018 - The backslash character is the escape character. It escapes
6019 backslash, space, tab, and newline. The ANSI C escape sequences
6020 like \n and \t are also supported. These produce argument
6021 constituents; the two-character combination \n doesn't act like a
6022 terminating newline. The escape sequence \NNN for exactly three
6023 octal digits reads as the character whose ASCII code is NNN. As
6024 above, characters produced this way are argument constituents.
6025 Backslash followed by other characters is not allowed.
6027 * Changes to the procedure for linking libguile with your programs
6029 ** Guile now builds and installs a shared guile library, if your
6030 system support shared libraries. (It still builds a static library on
6031 all systems.) Guile automatically detects whether your system
6032 supports shared libraries. To prevent Guile from buildisg shared
6033 libraries, pass the `--disable-shared' flag to the configure script.
6035 Guile takes longer to compile when it builds shared libraries, because
6036 it must compile every file twice --- once to produce position-
6037 independent object code, and once to produce normal object code.
6039 ** The libthreads library has been merged into libguile.
6041 To link a program against Guile, you now need only link against
6042 -lguile and -lqt; -lthreads is no longer needed. If you are using
6043 autoconf to generate configuration scripts for your application, the
6044 following lines should suffice to add the appropriate libraries to
6047 ### Find quickthreads and libguile.
6048 AC_CHECK_LIB(qt, main)
6049 AC_CHECK_LIB(guile, scm_shell)
6051 * Changes to Scheme functions
6053 ** Guile Scheme's special syntax for keyword objects is now optional,
6054 and disabled by default.
6056 The syntax variation from R4RS made it difficult to port some
6057 interesting packages to Guile. The routines which accepted keyword
6058 arguments (mostly in the module system) have been modified to also
6059 accept symbols whose names begin with `:'.
6061 To change the keyword syntax, you must first import the (ice-9 debug)
6063 (use-modules (ice-9 debug))
6065 Then you can enable the keyword syntax as follows:
6066 (read-set! keywords 'prefix)
6068 To disable keyword syntax, do this:
6069 (read-set! keywords #f)
6071 ** Many more primitive functions accept shared substrings as
6072 arguments. In the past, these functions required normal, mutable
6073 strings as arguments, although they never made use of this
6076 ** The uniform array functions now operate on byte vectors. These
6077 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6078 `serial-array-map', `array-map', `array-for-each', and
6081 ** The new functions `trace' and `untrace' implement simple debugging
6082 support for Scheme functions.
6084 The `trace' function accepts any number of procedures as arguments,
6085 and tells the Guile interpreter to display each procedure's name and
6086 arguments each time the procedure is invoked. When invoked with no
6087 arguments, `trace' returns the list of procedures currently being
6090 The `untrace' function accepts any number of procedures as arguments,
6091 and tells the Guile interpreter not to trace them any more. When
6092 invoked with no arguments, `untrace' untraces all curretly traced
6095 The tracing in Guile has an advantage over most other systems: we
6096 don't create new procedure objects, but mark the procedure objects
6097 themselves. This means that anonymous and internal procedures can be
6100 ** The function `assert-repl-prompt' has been renamed to
6101 `set-repl-prompt!'. It takes one argument, PROMPT.
6102 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6103 - If PROMPT is a string, we use it as a prompt.
6104 - If PROMPT is a procedure accepting no arguments, we call it, and
6105 display the result as a prompt.
6106 - Otherwise, we display "> ".
6108 ** The new function `eval-string' reads Scheme expressions from a
6109 string and evaluates them, returning the value of the last expression
6110 in the string. If the string contains no expressions, it returns an
6113 ** The new function `thunk?' returns true iff its argument is a
6114 procedure of zero arguments.
6116 ** `defined?' is now a builtin function, instead of syntax. This
6117 means that its argument should be quoted. It returns #t iff its
6118 argument is bound in the current module.
6120 ** The new syntax `use-modules' allows you to add new modules to your
6121 environment without re-typing a complete `define-module' form. It
6122 accepts any number of module names as arguments, and imports their
6123 public bindings into the current module.
6125 ** The new function (module-defined? NAME MODULE) returns true iff
6126 NAME, a symbol, is defined in MODULE, a module object.
6128 ** The new function `builtin-bindings' creates and returns a hash
6129 table containing copies of all the root module's bindings.
6131 ** The new function `builtin-weak-bindings' does the same as
6132 `builtin-bindings', but creates a doubly-weak hash table.
6134 ** The `equal?' function now considers variable objects to be
6135 equivalent if they have the same name and the same value.
6137 ** The new function `command-line' returns the command-line arguments
6138 given to Guile, as a list of strings.
6140 When using guile as a script interpreter, `command-line' returns the
6141 script's arguments; those processed by the interpreter (like `-s' or
6142 `-c') are omitted. (In other words, you get the normal, expected
6143 behavior.) Any application that uses scm_shell to process its
6144 command-line arguments gets this behavior as well.
6146 ** The new function `load-user-init' looks for a file called `.guile'
6147 in the user's home directory, and loads it if it exists. This is
6148 mostly for use by the code generated by scm_compile_shell_switches,
6149 but we thought it might also be useful in other circumstances.
6151 ** The new function `log10' returns the base-10 logarithm of its
6154 ** Changes to I/O functions
6156 *** The functions `read', `primitive-load', `read-and-eval!', and
6157 `primitive-load-path' no longer take optional arguments controlling
6158 case insensitivity and a `#' parser.
6160 Case sensitivity is now controlled by a read option called
6161 `case-insensitive'. The user can add new `#' syntaxes with the
6162 `read-hash-extend' function (see below).
6164 *** The new function `read-hash-extend' allows the user to change the
6165 syntax of Guile Scheme in a somewhat controlled way.
6167 (read-hash-extend CHAR PROC)
6168 When parsing S-expressions, if we read a `#' character followed by
6169 the character CHAR, use PROC to parse an object from the stream.
6170 If PROC is #f, remove any parsing procedure registered for CHAR.
6172 The reader applies PROC to two arguments: CHAR and an input port.
6174 *** The new functions read-delimited and read-delimited! provide a
6175 general mechanism for doing delimited input on streams.
6177 (read-delimited DELIMS [PORT HANDLE-DELIM])
6178 Read until we encounter one of the characters in DELIMS (a string),
6179 or end-of-file. PORT is the input port to read from; it defaults to
6180 the current input port. The HANDLE-DELIM parameter determines how
6181 the terminating character is handled; it should be one of the
6184 'trim omit delimiter from result
6185 'peek leave delimiter character in input stream
6186 'concat append delimiter character to returned value
6187 'split return a pair: (RESULT . TERMINATOR)
6189 HANDLE-DELIM defaults to 'peek.
6191 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6192 A side-effecting variant of `read-delimited'.
6194 The data is written into the string BUF at the indices in the
6195 half-open interval [START, END); the default interval is the whole
6196 string: START = 0 and END = (string-length BUF). The values of
6197 START and END must specify a well-defined interval in BUF, i.e.
6198 0 <= START <= END <= (string-length BUF).
6200 It returns NBYTES, the number of bytes read. If the buffer filled
6201 up without a delimiter character being found, it returns #f. If the
6202 port is at EOF when the read starts, it returns the EOF object.
6204 If an integer is returned (i.e., the read is successfully terminated
6205 by reading a delimiter character), then the HANDLE-DELIM parameter
6206 determines how to handle the terminating character. It is described
6207 above, and defaults to 'peek.
6209 (The descriptions of these functions were borrowed from the SCSH
6210 manual, by Olin Shivers and Brian Carlstrom.)
6212 *** The `%read-delimited!' function is the primitive used to implement
6213 `read-delimited' and `read-delimited!'.
6215 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6217 This returns a pair of values: (TERMINATOR . NUM-READ).
6218 - TERMINATOR describes why the read was terminated. If it is a
6219 character or the eof object, then that is the value that terminated
6220 the read. If it is #f, the function filled the buffer without finding
6221 a delimiting character.
6222 - NUM-READ is the number of characters read into BUF.
6224 If the read is successfully terminated by reading a delimiter
6225 character, then the gobble? parameter determines what to do with the
6226 terminating character. If true, the character is removed from the
6227 input stream; if false, the character is left in the input stream
6228 where a subsequent read operation will retrieve it. In either case,
6229 the character is also the first value returned by the procedure call.
6231 (The descriptions of this function was borrowed from the SCSH manual,
6232 by Olin Shivers and Brian Carlstrom.)
6234 *** The `read-line' and `read-line!' functions have changed; they now
6235 trim the terminator by default; previously they appended it to the
6236 returned string. For the old behavior, use (read-line PORT 'concat).
6238 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6239 take new optional START and END arguments, specifying the region of
6240 the array to read and write.
6242 *** The `ungetc-char-ready?' function has been removed. We feel it's
6243 inappropriate for an interface to expose implementation details this
6246 ** Changes to the Unix library and system call interface
6248 *** The new fcntl function provides access to the Unix `fcntl' system
6251 (fcntl PORT COMMAND VALUE)
6252 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6253 Values for COMMAND are:
6255 F_DUPFD duplicate a file descriptor
6256 F_GETFD read the descriptor's close-on-exec flag
6257 F_SETFD set the descriptor's close-on-exec flag to VALUE
6258 F_GETFL read the descriptor's flags, as set on open
6259 F_SETFL set the descriptor's flags, as set on open to VALUE
6260 F_GETOWN return the process ID of a socket's owner, for SIGIO
6261 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6262 FD_CLOEXEC not sure what this is
6264 For details, see the documentation for the fcntl system call.
6266 *** The arguments to `select' have changed, for compatibility with
6267 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6268 expected behavior. The MILLISECONDS parameter has been changed to
6269 MICROSECONDS, to more closely resemble the underlying system call.
6270 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6271 corresponding return set will be the same.
6273 *** The arguments to the `mknod' system call have changed. They are
6276 (mknod PATH TYPE PERMS DEV)
6277 Create a new file (`node') in the file system. PATH is the name of
6278 the file to create. TYPE is the kind of file to create; it should
6279 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6280 permission bits to give the newly created file. If TYPE is
6281 'block-special or 'char-special, DEV specifies which device the
6282 special file refers to; its interpretation depends on the kind of
6283 special file being created.
6285 *** The `fork' function has been renamed to `primitive-fork', to avoid
6286 clashing with various SCSH forks.
6288 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6289 and `recvfrom!'. They no longer accept a size for a second argument;
6290 you must pass a string to hold the received value. They no longer
6291 return the buffer. Instead, `recv' returns the length of the message
6292 received, and `recvfrom' returns a pair containing the packet's length
6293 and originating address.
6295 *** The file descriptor datatype has been removed, as have the
6296 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6297 We plan to replace these functions with a SCSH-compatible interface.
6299 *** The `create' function has been removed; it's just a special case
6302 *** There are new functions to break down process termination status
6303 values. In the descriptions below, STATUS is a value returned by
6306 (status:exit-val STATUS)
6307 If the child process exited normally, this function returns the exit
6308 code for the child process (i.e., the value passed to exit, or
6309 returned from main). If the child process did not exit normally,
6310 this function returns #f.
6312 (status:stop-sig STATUS)
6313 If the child process was suspended by a signal, this function
6314 returns the signal that suspended the child. Otherwise, it returns
6317 (status:term-sig STATUS)
6318 If the child process terminated abnormally, this function returns
6319 the signal that terminated the child. Otherwise, this function
6322 POSIX promises that exactly one of these functions will return true on
6323 a valid STATUS value.
6325 These functions are compatible with SCSH.
6327 *** There are new accessors and setters for the broken-out time vectors
6328 returned by `localtime', `gmtime', and that ilk. They are:
6330 Component Accessor Setter
6331 ========================= ============ ============
6332 seconds tm:sec set-tm:sec
6333 minutes tm:min set-tm:min
6334 hours tm:hour set-tm:hour
6335 day of the month tm:mday set-tm:mday
6336 month tm:mon set-tm:mon
6337 year tm:year set-tm:year
6338 day of the week tm:wday set-tm:wday
6339 day in the year tm:yday set-tm:yday
6340 daylight saving time tm:isdst set-tm:isdst
6341 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6342 name of time zone tm:zone set-tm:zone
6344 *** There are new accessors for the vectors returned by `uname',
6345 describing the host system:
6348 ============================================== ================
6349 name of the operating system implementation utsname:sysname
6350 network name of this machine utsname:nodename
6351 release level of the operating system utsname:release
6352 version level of the operating system utsname:version
6353 machine hardware platform utsname:machine
6355 *** There are new accessors for the vectors returned by `getpw',
6356 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6357 system's user database:
6360 ====================== =================
6361 user name passwd:name
6362 user password passwd:passwd
6365 real name passwd:gecos
6366 home directory passwd:dir
6367 shell program passwd:shell
6369 *** There are new accessors for the vectors returned by `getgr',
6370 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6371 system's group database:
6374 ======================= ============
6375 group name group:name
6376 group password group:passwd
6378 group members group:mem
6380 *** There are new accessors for the vectors returned by `gethost',
6381 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6385 ========================= ===============
6386 official name of host hostent:name
6387 alias list hostent:aliases
6388 host address type hostent:addrtype
6389 length of address hostent:length
6390 list of addresses hostent:addr-list
6392 *** There are new accessors for the vectors returned by `getnet',
6393 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6397 ========================= ===============
6398 official name of net netent:name
6399 alias list netent:aliases
6400 net number type netent:addrtype
6401 net number netent:net
6403 *** There are new accessors for the vectors returned by `getproto',
6404 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6408 ========================= ===============
6409 official protocol name protoent:name
6410 alias list protoent:aliases
6411 protocol number protoent:proto
6413 *** There are new accessors for the vectors returned by `getserv',
6414 `getservbyname', `getservbyport', and `getservent', describing
6418 ========================= ===============
6419 official service name servent:name
6420 alias list servent:aliases
6421 port number servent:port
6422 protocol to use servent:proto
6424 *** There are new accessors for the sockaddr structures returned by
6425 `accept', `getsockname', `getpeername', `recvfrom!':
6428 ======================================== ===============
6429 address format (`family') sockaddr:fam
6430 path, for file domain addresses sockaddr:path
6431 address, for internet domain addresses sockaddr:addr
6432 TCP or UDP port, for internet sockaddr:port
6434 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6435 `getprotoent', and `getservent' functions now return #f at the end of
6436 the user database. (They used to throw an exception.)
6438 Note that calling MUMBLEent function is equivalent to calling the
6439 corresponding MUMBLE function with no arguments.
6441 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6442 `setprotoent', and `setservent' routines now take no arguments.
6444 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6445 provide more useful information when they throw an exception.
6447 *** The `lnaof' function has been renamed to `inet-lnaof'.
6449 *** Guile now claims to have the `current-time' feature.
6451 *** The `mktime' function now takes an optional second argument ZONE,
6452 giving the time zone to use for the conversion. ZONE should be a
6453 string, in the same format as expected for the "TZ" environment variable.
6455 *** The `strptime' function now returns a pair (TIME . COUNT), where
6456 TIME is the parsed time as a vector, and COUNT is the number of
6457 characters from the string left unparsed. This function used to
6458 return the remaining characters as a string.
6460 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6461 The return value is now (SECONDS . MICROSECONDS); the fractional
6462 component is no longer expressed in "ticks".
6464 *** The `ticks/sec' constant has been removed, in light of the above change.
6466 * Changes to the gh_ interface
6468 ** gh_eval_str() now returns an SCM object which is the result of the
6471 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6474 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6475 and returns the array
6477 ** gh_scm2str0() is gone: there is no need to distinguish
6478 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6479 the user to interpret the data both ways.
6481 * Changes to the scm_ interface
6483 ** The new function scm_symbol_value0 provides an easy way to get a
6484 symbol's value from C code:
6486 SCM scm_symbol_value0 (char *NAME)
6487 Return the value of the symbol named by the null-terminated string
6488 NAME in the current module. If the symbol named NAME is unbound in
6489 the current module, return SCM_UNDEFINED.
6491 ** The new function scm_sysintern0 creates new top-level variables,
6492 without assigning them a value.
6494 SCM scm_sysintern0 (char *NAME)
6495 Create a new Scheme top-level variable named NAME. NAME is a
6496 null-terminated string. Return the variable's value cell.
6498 ** The function scm_internal_catch is the guts of catch. It handles
6499 all the mechanics of setting up a catch target, invoking the catch
6500 body, and perhaps invoking the handler if the body does a throw.
6502 The function is designed to be usable from C code, but is general
6503 enough to implement all the semantics Guile Scheme expects from throw.
6505 TAG is the catch tag. Typically, this is a symbol, but this function
6506 doesn't actually care about that.
6508 BODY is a pointer to a C function which runs the body of the catch;
6509 this is the code you can throw from. We call it like this:
6510 BODY (BODY_DATA, JMPBUF)
6512 BODY_DATA is just the BODY_DATA argument we received; we pass it
6513 through to BODY as its first argument. The caller can make
6514 BODY_DATA point to anything useful that BODY might need.
6515 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6516 which we have just created and initialized.
6518 HANDLER is a pointer to a C function to deal with a throw to TAG,
6519 should one occur. We call it like this:
6520 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6522 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6523 same idea as BODY_DATA above.
6524 THROWN_TAG is the tag that the user threw to; usually this is
6525 TAG, but it could be something else if TAG was #t (i.e., a
6526 catch-all), or the user threw to a jmpbuf.
6527 THROW_ARGS is the list of arguments the user passed to the THROW
6530 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6531 is just a pointer we pass through to HANDLER. We don't actually
6532 use either of those pointers otherwise ourselves. The idea is
6533 that, if our caller wants to communicate something to BODY or
6534 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6535 HANDLER can then use. Think of it as a way to make BODY and
6536 HANDLER closures, not just functions; MUMBLE_DATA points to the
6539 Of course, it's up to the caller to make sure that any data a
6540 MUMBLE_DATA needs is protected from GC. A common way to do this is
6541 to make MUMBLE_DATA a pointer to data stored in an automatic
6542 structure variable; since the collector must scan the stack for
6543 references anyway, this assures that any references in MUMBLE_DATA
6546 ** The new function scm_internal_lazy_catch is exactly like
6547 scm_internal_catch, except:
6549 - It does not unwind the stack (this is the major difference).
6550 - If handler returns, its value is returned from the throw.
6551 - BODY always receives #f as its JMPBUF argument (since there's no
6552 jmpbuf associated with a lazy catch, because we don't unwind the
6555 ** scm_body_thunk is a new body function you can pass to
6556 scm_internal_catch if you want the body to be like Scheme's `catch'
6557 --- a thunk, or a function of one argument if the tag is #f.
6559 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6560 contains the Scheme procedure to invoke as the body, and the tag
6561 we're catching. If the tag is #f, then we pass JMPBUF (created by
6562 scm_internal_catch) to the body procedure; otherwise, the body gets
6565 ** scm_handle_by_proc is a new handler function you can pass to
6566 scm_internal_catch if you want the handler to act like Scheme's catch
6567 --- call a procedure with the tag and the throw arguments.
6569 If the user does a throw to this catch, this function runs a handler
6570 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6571 variable holding the Scheme procedure object to invoke. It ought to
6572 be a pointer to an automatic variable (i.e., one living on the stack),
6573 or the procedure object should be otherwise protected from GC.
6575 ** scm_handle_by_message is a new handler function to use with
6576 `scm_internal_catch' if you want Guile to print a message and die.
6577 It's useful for dealing with throws to uncaught keys at the top level.
6579 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6580 message header to print; if zero, we use "guile" instead. That
6581 text is followed by a colon, then the message described by ARGS.
6583 ** The return type of scm_boot_guile is now void; the function does
6584 not return a value, and indeed, never returns at all.
6586 ** The new function scm_shell makes it easy for user applications to
6587 process command-line arguments in a way that is compatible with the
6588 stand-alone guile interpreter (which is in turn compatible with SCSH,
6591 To use the scm_shell function, first initialize any guile modules
6592 linked into your application, and then call scm_shell with the values
6593 of ARGC and ARGV your `main' function received. scm_shell will add
6594 any SCSH-style meta-arguments from the top of the script file to the
6595 argument vector, and then process the command-line arguments. This
6596 generally means loading a script file or starting up an interactive
6597 command interpreter. For details, see "Changes to the stand-alone
6600 ** The new functions scm_get_meta_args and scm_count_argv help you
6601 implement the SCSH-style meta-argument, `\'.
6603 char **scm_get_meta_args (int ARGC, char **ARGV)
6604 If the second element of ARGV is a string consisting of a single
6605 backslash character (i.e. "\\" in Scheme notation), open the file
6606 named by the following argument, parse arguments from it, and return
6607 the spliced command line. The returned array is terminated by a
6610 For details of argument parsing, see above, under "guile now accepts
6611 command-line arguments compatible with SCSH..."
6613 int scm_count_argv (char **ARGV)
6614 Count the arguments in ARGV, assuming it is terminated by a null
6617 For an example of how these functions might be used, see the source
6618 code for the function scm_shell in libguile/script.c.
6620 You will usually want to use scm_shell instead of calling this
6623 ** The new function scm_compile_shell_switches turns an array of
6624 command-line arguments into Scheme code to carry out the actions they
6625 describe. Given ARGC and ARGV, it returns a Scheme expression to
6626 evaluate, and calls scm_set_program_arguments to make any remaining
6627 command-line arguments available to the Scheme code. For example,
6628 given the following arguments:
6630 -e main -s ekko a speckled gecko
6632 scm_set_program_arguments will return the following expression:
6634 (begin (load "ekko") (main (command-line)) (quit))
6636 You will usually want to use scm_shell instead of calling this
6639 ** The function scm_shell_usage prints a usage message appropriate for
6640 an interpreter that uses scm_compile_shell_switches to handle its
6641 command-line arguments.
6643 void scm_shell_usage (int FATAL, char *MESSAGE)
6644 Print a usage message to the standard error output. If MESSAGE is
6645 non-zero, write it before the usage message, followed by a newline.
6646 If FATAL is non-zero, exit the process, using FATAL as the
6647 termination status. (If you want to be compatible with Guile,
6648 always use 1 as the exit status when terminating due to command-line
6651 You will usually want to use scm_shell instead of calling this
6654 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6655 expressions. It used to return SCM_EOL. Earth-shattering.
6657 ** The macros for declaring scheme objects in C code have been
6658 rearranged slightly. They are now:
6660 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6661 Declare a static SCM variable named C_NAME, and initialize it to
6662 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6663 be a C identifier, and SCHEME_NAME should be a C string.
6665 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6666 Just like SCM_SYMBOL, but make C_NAME globally visible.
6668 SCM_VCELL (C_NAME, SCHEME_NAME)
6669 Create a global variable at the Scheme level named SCHEME_NAME.
6670 Declare a static SCM variable named C_NAME, and initialize it to
6671 point to the Scheme variable's value cell.
6673 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6674 Just like SCM_VCELL, but make C_NAME globally visible.
6676 The `guile-snarf' script writes initialization code for these macros
6677 to its standard output, given C source code as input.
6679 The SCM_GLOBAL macro is gone.
6681 ** The scm_read_line and scm_read_line_x functions have been replaced
6682 by Scheme code based on the %read-delimited! procedure (known to C
6683 code as scm_read_delimited_x). See its description above for more
6686 ** The function scm_sys_open has been renamed to scm_open. It now
6687 returns a port instead of an FD object.
6689 * The dynamic linking support has changed. For more information, see
6690 libguile/DYNAMIC-LINKING.
6695 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6698 * Changes to the 'guile' program:
6700 ** Guile now loads some new files when it starts up. Guile first
6701 searches the load path for init.scm, and loads it if found. Then, if
6702 Guile is not being used to execute a script, and the user's home
6703 directory contains a file named `.guile', Guile loads that.
6705 ** You can now use Guile as a shell script interpreter.
6707 To paraphrase the SCSH manual:
6709 When Unix tries to execute an executable file whose first two
6710 characters are the `#!', it treats the file not as machine code to
6711 be directly executed by the native processor, but as source code
6712 to be executed by some interpreter. The interpreter to use is
6713 specified immediately after the #! sequence on the first line of
6714 the source file. The kernel reads in the name of the interpreter,
6715 and executes that instead. It passes the interpreter the source
6716 filename as its first argument, with the original arguments
6717 following. Consult the Unix man page for the `exec' system call
6718 for more information.
6720 Now you can use Guile as an interpreter, using a mechanism which is a
6721 compatible subset of that provided by SCSH.
6723 Guile now recognizes a '-s' command line switch, whose argument is the
6724 name of a file of Scheme code to load. It also treats the two
6725 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6726 to make a file of Scheme code directly executable by Unix, insert the
6727 following two lines at the top of the file:
6729 #!/usr/local/bin/guile -s
6732 Guile treats the argument of the `-s' command-line switch as the name
6733 of a file of Scheme code to load, and treats the sequence `#!' as the
6734 start of a block comment, terminated by `!#'.
6736 For example, here's a version of 'echo' written in Scheme:
6738 #!/usr/local/bin/guile -s
6740 (let loop ((args (cdr (program-arguments))))
6743 (display (car args))
6744 (if (pair? (cdr args))
6746 (loop (cdr args)))))
6749 Why does `#!' start a block comment terminated by `!#', instead of the
6750 end of the line? That is the notation SCSH uses, and although we
6751 don't yet support the other SCSH features that motivate that choice,
6752 we would like to be backward-compatible with any existing Guile
6753 scripts once we do. Furthermore, if the path to Guile on your system
6754 is too long for your kernel, you can start the script with this
6758 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6761 Note that some very old Unix systems don't support the `#!' syntax.
6764 ** You can now run Guile without installing it.
6766 Previous versions of the interactive Guile interpreter (`guile')
6767 couldn't start up unless Guile's Scheme library had been installed;
6768 they used the value of the environment variable `SCHEME_LOAD_PATH'
6769 later on in the startup process, but not to find the startup code
6770 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6773 To run Guile without installing it, build it in the normal way, and
6774 then set the environment variable `SCHEME_LOAD_PATH' to a
6775 colon-separated list of directories, including the top-level directory
6776 of the Guile sources. For example, if you unpacked Guile so that the
6777 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6780 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6783 ** Guile's read-eval-print loop no longer prints #<unspecified>
6784 results. If the user wants to see this, she can evaluate the
6785 expression (assert-repl-print-unspecified #t), perhaps in her startup
6788 ** Guile no longer shows backtraces by default when an error occurs;
6789 however, it does display a message saying how to get one, and how to
6790 request that they be displayed by default. After an error, evaluate
6792 to see a backtrace, and
6793 (debug-enable 'backtrace)
6794 to see them by default.
6798 * Changes to Guile Scheme:
6800 ** Guile now distinguishes between #f and the empty list.
6802 This is for compatibility with the IEEE standard, the (possibly)
6803 upcoming Revised^5 Report on Scheme, and many extant Scheme
6806 Guile used to have #f and '() denote the same object, to make Scheme's
6807 type system more compatible with Emacs Lisp's. However, the change
6808 caused too much trouble for Scheme programmers, and we found another
6809 way to reconcile Emacs Lisp with Scheme that didn't require this.
6812 ** Guile's delq, delv, delete functions, and their destructive
6813 counterparts, delq!, delv!, and delete!, now remove all matching
6814 elements from the list, not just the first. This matches the behavior
6815 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6816 functions which inspired them.
6818 I recognize that this change may break code in subtle ways, but it
6819 seems best to make the change before the FSF's first Guile release,
6823 ** The compiled-library-path function has been deleted from libguile.
6825 ** The facilities for loading Scheme source files have changed.
6827 *** The variable %load-path now tells Guile which directories to search
6828 for Scheme code. Its value is a list of strings, each of which names
6831 *** The variable %load-extensions now tells Guile which extensions to
6832 try appending to a filename when searching the load path. Its value
6833 is a list of strings. Its default value is ("" ".scm").
6835 *** (%search-load-path FILENAME) searches the directories listed in the
6836 value of the %load-path variable for a Scheme file named FILENAME,
6837 with all the extensions listed in %load-extensions. If it finds a
6838 match, then it returns its full filename. If FILENAME is absolute, it
6839 returns it unchanged. Otherwise, it returns #f.
6841 %search-load-path will not return matches that refer to directories.
6843 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6844 uses %seach-load-path to find a file named FILENAME, and loads it if
6845 it finds it. If it can't read FILENAME for any reason, it throws an
6848 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6851 *** load uses the same searching semantics as primitive-load.
6853 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6854 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6855 path, and load-module-with-path have been deleted. The functions
6856 above should serve their purposes.
6858 *** If the value of the variable %load-hook is a procedure,
6859 `primitive-load' applies its value to the name of the file being
6860 loaded (without the load path directory name prepended). If its value
6861 is #f, it is ignored. Otherwise, an error occurs.
6863 This is mostly useful for printing load notification messages.
6866 ** The function `eval!' is no longer accessible from the scheme level.
6867 We can't allow operations which introduce glocs into the scheme level,
6868 because Guile's type system can't handle these as data. Use `eval' or
6869 `read-and-eval!' (see below) as replacement.
6871 ** The new function read-and-eval! reads an expression from PORT,
6872 evaluates it, and returns the result. This is more efficient than
6873 simply calling `read' and `eval', since it is not necessary to make a
6874 copy of the expression for the evaluator to munge.
6876 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6877 for the `read' function.
6880 ** The function `int?' has been removed; its definition was identical
6881 to that of `integer?'.
6883 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6884 use the R4RS names for these functions.
6886 ** The function object-properties no longer returns the hash handle;
6887 it simply returns the object's property list.
6889 ** Many functions have been changed to throw errors, instead of
6890 returning #f on failure. The point of providing exception handling in
6891 the language is to simplify the logic of user code, but this is less
6892 useful if Guile's primitives don't throw exceptions.
6894 ** The function `fileno' has been renamed from `%fileno'.
6896 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6899 * Changes to Guile's C interface:
6901 ** The library's initialization procedure has been simplified.
6902 scm_boot_guile now has the prototype:
6904 void scm_boot_guile (int ARGC,
6906 void (*main_func) (),
6909 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6910 MAIN_FUNC should do all the work of the program (initializing other
6911 packages, reading user input, etc.) before returning. When MAIN_FUNC
6912 returns, call exit (0); this function never returns. If you want some
6913 other exit value, MAIN_FUNC may call exit itself.
6915 scm_boot_guile arranges for program-arguments to return the strings
6916 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6917 scm_set_program_arguments with the final list, so Scheme code will
6918 know which arguments have been processed.
6920 scm_boot_guile establishes a catch-all catch handler which prints an
6921 error message and exits the process. This means that Guile exits in a
6922 coherent way when system errors occur and the user isn't prepared to
6923 handle it. If the user doesn't like this behavior, they can establish
6924 their own universal catcher in MAIN_FUNC to shadow this one.
6926 Why must the caller do all the real work from MAIN_FUNC? The garbage
6927 collector assumes that all local variables of type SCM will be above
6928 scm_boot_guile's stack frame on the stack. If you try to manipulate
6929 SCM values after this function returns, it's the luck of the draw
6930 whether the GC will be able to find the objects you allocate. So,
6931 scm_boot_guile function exits, rather than returning, to discourage
6932 people from making that mistake.
6934 The IN, OUT, and ERR arguments were removed; there are other
6935 convenient ways to override these when desired.
6937 The RESULT argument was deleted; this function should never return.
6939 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
6943 ** Guile's header files should no longer conflict with your system's
6946 In order to compile code which #included <libguile.h>, previous
6947 versions of Guile required you to add a directory containing all the
6948 Guile header files to your #include path. This was a problem, since
6949 Guile's header files have names which conflict with many systems'
6952 Now only <libguile.h> need appear in your #include path; you must
6953 refer to all Guile's other header files as <libguile/mumble.h>.
6954 Guile's installation procedure puts libguile.h in $(includedir), and
6955 the rest in $(includedir)/libguile.
6958 ** Two new C functions, scm_protect_object and scm_unprotect_object,
6959 have been added to the Guile library.
6961 scm_protect_object (OBJ) protects OBJ from the garbage collector.
6962 OBJ will not be freed, even if all other references are dropped,
6963 until someone does scm_unprotect_object (OBJ). Both functions
6966 Note that calls to scm_protect_object do not nest. You can call
6967 scm_protect_object any number of times on a given object, and the
6968 next call to scm_unprotect_object will unprotect it completely.
6970 Basically, scm_protect_object and scm_unprotect_object just
6971 maintain a list of references to things. Since the GC knows about
6972 this list, all objects it mentions stay alive. scm_protect_object
6973 adds its argument to the list; scm_unprotect_object remove its
6974 argument from the list.
6977 ** scm_eval_0str now returns the value of the last expression
6980 ** The new function scm_read_0str reads an s-expression from a
6981 null-terminated string, and returns it.
6983 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
6984 to a Scheme port object.
6986 ** The new function `scm_set_program_arguments' allows C code to set
6987 the value returned by the Scheme `program-arguments' function.
6992 * Guile no longer includes sophisticated Tcl/Tk support.
6994 The old Tcl/Tk support was unsatisfying to us, because it required the
6995 user to link against the Tcl library, as well as Tk and Guile. The
6996 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
6997 referring to widgets by names, rather than exporting widgets to Scheme
6998 code as a special datatype.
7000 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7001 maintainers described some very interesting changes in progress to the
7002 Tcl/Tk internals, which would facilitate clean interfaces between lone
7003 Tk and other interpreters --- even for garbage-collected languages
7004 like Scheme. They expected the new Tk to be publicly available in the
7007 Since it seems that Guile might soon have a new, cleaner interface to
7008 lone Tk, and that the old Guile/Tk glue code would probably need to be
7009 completely rewritten, we (Jim Blandy and Richard Stallman) have
7010 decided not to support the old code. We'll spend the time instead on
7011 a good interface to the newer Tk, as soon as it is available.
7013 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7016 Copyright information:
7018 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
7020 Permission is granted to anyone to make or distribute verbatim copies
7021 of this document as received, in any medium, provided that the
7022 copyright notice and this permission notice are preserved,
7023 thus giving the recipient permission to redistribute in turn.
7025 Permission is granted to distribute modified versions
7026 of this document, or of portions of it,
7027 under the above conditions, provided also that they
7028 carry prominent notices stating who last changed them.
7033 paragraph-separate: "[
\f]*$"