1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
10 * New modules (see the manual for details)
12 ** `(srfi srfi-18)', multithreading support
13 ** The `(ice-9 i18n)' module provides internationalization support
15 * Changes to the distribution
17 ** Guile now uses Gnulib as a portability aid
19 * Changes to the stand-alone interpreter
20 * Changes to Scheme functions and syntax
22 ** A new 'memoize-symbol evaluator trap has been added. This trap can
23 be used for efficiently implementing a Scheme code coverage.
25 ** Duplicate bindings among used modules are resolved lazily.
26 This slightly improves program startup times.
28 ** New thread cancellation and thread cleanup API
29 See `cancel-thread', `set-thread-cleanup!', and `thread-cleanup'.
31 * Changes to the C interface
33 ** The GH interface (deprecated in version 1.6, 2001) was removed.
35 ** Internal `scm_i_' functions now have "hidden" linkage with GCC/ELF
37 This makes these internal functions technically not callable from
40 ** Functions for handling `scm_option' now no longer require an argument
41 indicating length of the `scm_t_option' array.
44 Changes in 1.8.7 (since 1.8.6)
48 ** Fix %fast-slot-ref/set!, to avoid possible segmentation fault
51 Changes in 1.8.6 (since 1.8.5)
53 * New features (see the manual for details)
55 ** New convenience function `scm_c_symbol_length ()'
57 ** Single stepping through code from Emacs
59 When you use GDS to evaluate Scheme code from Emacs, you can now use
60 `C-u' to indicate that you want to single step through that code. See
61 `Evaluating Scheme Code' in the manual for more details.
63 ** New "guile(1)" man page!
65 * Changes to the distribution
67 ** Automake's `AM_MAINTAINER_MODE' is no longer used
69 Thus, the `--enable-maintainer-mode' configure option is no longer
70 available: Guile is now always configured in "maintainer mode".
72 ** `ChangeLog' files are no longer updated
74 Instead, changes are detailed in the version control system's logs. See
75 the top-level `ChangeLog' files for details.
80 ** `symbol->string' now returns a read-only string, as per R5RS
81 ** Fix incorrect handling of the FLAGS argument of `fold-matches'
82 ** `guile-config link' now prints `-L$libdir' before `-lguile'
83 ** Fix memory corruption involving GOOPS' `class-redefinition'
84 ** Fix possible deadlock in `mutex-lock'
85 ** Fix build issue on Tru64 and ia64-hp-hpux11.23 (`SCM_UNPACK' macro)
86 ** Fix build issue on mips, mipsel, powerpc and ia64 (stack direction)
87 ** Fix build issue on hppa2.0w-hp-hpux11.11 (`dirent64' and `readdir64_r')
88 ** Fix build issue on i386-unknown-freebsd7.0 ("break strict-aliasing rules")
89 ** Fix misleading output from `(help rationalize)'
90 ** Fix build failure on Debian hppa architecture (bad stack growth detection)
91 ** Fix `gcd' when called with a single, negative argument.
92 ** Fix `Stack overflow' errors seen when building on some platforms
93 ** Fix bug when `scm_with_guile ()' was called several times from the
95 ** The handler of SRFI-34 `with-exception-handler' is now invoked in the
96 dynamic environment of the call to `raise'
97 ** Fix potential deadlock in `make-struct'
98 ** Fix compilation problem with libltdl from Libtool 2.2.x
99 ** Fix sloppy bound checking in `string-{ref,set!}' with the empty string
102 Changes in 1.8.5 (since 1.8.4)
104 * Infrastructure changes
106 ** Guile repository switched from CVS to Git
108 The new repository can be accessed using
109 "git-clone git://git.sv.gnu.org/guile.git", or can be browsed on-line at
110 http://git.sv.gnu.org/gitweb/?p=guile.git . See `README' for details.
112 ** Add support for `pkg-config'
114 See "Autoconf Support" in the manual for details.
116 * New modules (see the manual for details)
120 * New features (see the manual for details)
122 ** New `postfix' read option, for SRFI-88 keyword syntax
123 ** Some I/O primitives have been inlined, which improves I/O performance
124 ** New object-based traps infrastructure
126 This is a GOOPS-based infrastructure that builds on Guile's low-level
127 evaluator trap calls and facilitates the development of debugging
128 features like single-stepping, breakpoints, tracing and profiling.
129 See the `Traps' node of the manual for details.
131 ** New support for working on Guile code from within Emacs
133 Guile now incorporates the `GDS' library (previously distributed
134 separately) for working on Guile code from within Emacs. See the
135 `Using Guile In Emacs' node of the manual for details.
139 ** `scm_add_slot ()' no longer segfaults (fixes bug #22369)
140 ** Fixed `(ice-9 match)' for patterns like `((_ ...) ...)'
142 Previously, expressions like `(match '((foo) (bar)) (((_ ...) ...) #t))'
143 would trigger an unbound variable error for `match:andmap'.
145 ** `(oop goops describe)' now properly provides the `describe' feature
146 ** Fixed `args-fold' from `(srfi srfi-37)'
148 Previously, parsing short option names of argument-less options would
149 lead to a stack overflow.
151 ** `(srfi srfi-35)' is now visible through `cond-expand'
152 ** Fixed type-checking for the second argument of `eval'
153 ** Fixed type-checking for SRFI-1 `partition'
154 ** Fixed `struct-ref' and `struct-set!' on "light structs"
155 ** Honor struct field access rights in GOOPS
156 ** Changed the storage strategy of source properties, which fixes a deadlock
157 ** Allow compilation of Guile-using programs in C99 mode with GCC 4.3 and later
158 ** Fixed build issue for GNU/Linux on IA64
159 ** Fixed build issues on NetBSD 1.6
160 ** Fixed build issue on Solaris 2.10 x86_64
161 ** Fixed build issue with DEC/Compaq/HP's compiler
162 ** Fixed `scm_from_complex_double' build issue on FreeBSD
163 ** Fixed `alloca' build issue on FreeBSD 6
164 ** Removed use of non-portable makefile constructs
165 ** Fixed shadowing of libc's <random.h> on Tru64, which broke compilation
166 ** Make sure all tests honor `$TMPDIR'
168 * Changes to the distribution
172 We've started collecting Frequently Asked Questions (FAQ), and will
173 distribute these (with answers!) in future Guile releases.
176 Changes in 1.8.4 (since 1.8.3)
180 ** CR (ASCII 0x0d) is (again) recognized as a token delimiter by the reader
181 ** Fixed a segmentation fault which occurred when displaying the
182 backtrace of a stack with a promise object (made by `delay') in it.
183 ** Make `accept' leave guile mode while blocking
184 ** `scm_c_read ()' and `scm_c_write ()' now type-check their port argument
185 ** Fixed a build problem on AIX (use of func_data identifier)
186 ** Fixed a segmentation fault which occurred when hashx-ref or hashx-set! was
187 called with an associator proc that returns neither a pair nor #f.
188 ** Secondary threads now always return a valid module for (current-module).
189 ** Avoid MacOS build problems caused by incorrect combination of "64"
190 system and library calls.
191 ** `guile-snarf' now honors `$TMPDIR'
192 ** `guile-config compile' now reports CPPFLAGS used at compile-time
193 ** Fixed build with Sun Studio (Solaris 9)
194 ** Fixed wrong-type-arg errors when creating zero length SRFI-4
195 uniform vectors on AIX.
196 ** Fixed a deadlock that occurs upon GC with multiple threads.
197 ** Fixed compile problem with GCC on Solaris and AIX (use of _Complex_I)
198 ** Fixed autotool-derived build problems on AIX 6.1.
199 ** Fixed NetBSD/alpha support
200 ** Fixed MacOS build problem caused by use of rl_get_keymap(_name)
202 * New modules (see the manual for details)
206 * Documentation fixes and improvements
208 ** Removed premature breakpoint documentation
210 The features described are not available in the series of 1.8.x
211 releases, so the documentation was misleading and has been removed.
213 ** More about Guile's default *random-state* variable
215 ** GOOPS: more about how to use `next-method'
217 * Changes to the distribution
219 ** Corrected a few files that referred incorrectly to the old GPL + special exception licence
221 In fact Guile since 1.8.0 has been licensed with the GNU Lesser
222 General Public License, and the few incorrect files have now been
223 fixed to agree with the rest of the Guile distribution.
225 ** Removed unnecessary extra copies of COPYING*
227 The distribution now contains a single COPYING.LESSER at its top level.
230 Changes in 1.8.3 (since 1.8.2)
232 * New modules (see the manual for details)
239 ** The `(ice-9 slib)' module now works as expected
240 ** Expressions like "(set! 'x #t)" no longer yield a crash
241 ** Warnings about duplicate bindings now go to stderr
242 ** A memory leak in `make-socket-address' was fixed
243 ** Alignment issues (e.g., on SPARC) in network routines were fixed
244 ** A threading issue that showed up at least on NetBSD was fixed
245 ** Build problems on Solaris and IRIX fixed
247 * Implementation improvements
249 ** The reader is now faster, which reduces startup time
250 ** Procedures returned by `record-accessor' and `record-modifier' are faster
253 Changes in 1.8.2 (since 1.8.1):
255 * New procedures (see the manual for details)
257 ** set-program-arguments
260 * Incompatible changes
262 ** The body of a top-level `define' no longer sees the binding being created
264 In a top-level `define', the binding being created is no longer visible
265 from the `define' body. This breaks code like
266 "(define foo (begin (set! foo 1) (+ foo 1)))", where `foo' is now
267 unbound in the body. However, such code was not R5RS-compliant anyway,
272 ** Fractions were not `equal?' if stored in unreduced form.
273 (A subtle problem, since printing a value reduced it, making it work.)
274 ** srfi-60 `copy-bit' failed on 64-bit systems
275 ** "guile --use-srfi" option at the REPL can replace core functions
276 (Programs run with that option were ok, but in the interactive REPL
277 the core bindings got priority, preventing SRFI replacements or
279 ** `regexp-exec' doesn't abort() on #\nul in the input or bad flags arg
280 ** `kill' on mingw throws an error for a PID other than oneself
281 ** Procedure names are attached to procedure-with-setters
282 ** Array read syntax works with negative lower bound
283 ** `array-in-bounds?' fix if an array has different lower bounds on each index
284 ** `*' returns exact 0 for "(* inexact 0)"
285 This follows what it always did for "(* 0 inexact)".
286 ** SRFI-19: Value returned by `(current-time time-process)' was incorrect
287 ** SRFI-19: `date->julian-day' did not account for timezone offset
288 ** `ttyname' no longer crashes when passed a non-tty argument
289 ** `inet-ntop' no longer crashes on SPARC when passed an `AF_INET' address
290 ** Small memory leaks have been fixed in `make-fluid' and `add-history'
291 ** GOOPS: Fixed a bug in `method-more-specific?'
292 ** Build problems on Solaris fixed
293 ** Build problems on HP-UX IA64 fixed
294 ** Build problems on MinGW fixed
297 Changes in 1.8.1 (since 1.8.0):
299 * LFS functions are now used to access 64-bit files on 32-bit systems.
301 * New procedures (see the manual for details)
303 ** primitive-_exit - [Scheme] the-root-module
304 ** scm_primitive__exit - [C]
305 ** make-completion-function - [Scheme] (ice-9 readline)
306 ** scm_c_locale_stringn_to_number - [C]
307 ** scm_srfi1_append_reverse [C]
308 ** scm_srfi1_append_reverse_x [C]
316 ** Build problems have been fixed on MacOS, SunOS, and QNX.
318 ** `strftime' fix sign of %z timezone offset.
320 ** A one-dimensional array can now be 'equal?' to a vector.
322 ** Structures, records, and SRFI-9 records can now be compared with `equal?'.
324 ** SRFI-14 standard char sets are recomputed upon a successful `setlocale'.
326 ** `record-accessor' and `record-modifier' now have strict type checks.
328 Record accessor and modifier procedures now throw an error if the
329 record type of the record they're given is not the type expected.
330 (Previously accessors returned #f and modifiers silently did nothing).
332 ** It is now OK to use both autoload and use-modules on a given module.
334 ** `apply' checks the number of arguments more carefully on "0 or 1" funcs.
336 Previously there was no checking on primatives like make-vector that
337 accept "one or two" arguments. Now there is.
339 ** The srfi-1 assoc function now calls its equality predicate properly.
341 Previously srfi-1 assoc would call the equality predicate with the key
342 last. According to the SRFI, the key should be first.
344 ** A bug in n-par-for-each and n-for-each-par-map has been fixed.
346 ** The array-set! procedure no longer segfaults when given a bit vector.
348 ** Bugs in make-shared-array have been fixed.
350 ** string<? and friends now follow char<? etc order on 8-bit chars.
352 ** The format procedure now handles inf and nan values for ~f correctly.
354 ** exact->inexact should no longer overflow when given certain large fractions.
356 ** srfi-9 accessor and modifier procedures now have strict record type checks.
358 This matches the srfi-9 specification.
360 ** (ice-9 ftw) procedures won't ignore different files with same inode number.
362 Previously the (ice-9 ftw) procedures would ignore any file that had
363 the same inode number as a file they had already seen, even if that
364 file was on a different device.
367 Changes in 1.8.0 (changes since the 1.6.x series):
369 * Changes to the distribution
371 ** Guile is now licensed with the GNU Lesser General Public License.
373 ** The manual is now licensed with the GNU Free Documentation License.
375 ** Guile now requires GNU MP (http://swox.com/gmp).
377 Guile now uses the GNU MP library for arbitrary precision arithmetic.
379 ** Guile now has separate private and public configuration headers.
381 That is, things like HAVE_STRING_H no longer leak from Guile's
384 ** Guile now provides and uses an "effective" version number.
386 Guile now provides scm_effective_version and effective-version
387 functions which return the "effective" version number. This is just
388 the normal full version string without the final micro-version number,
389 so the current effective-version is "1.8". The effective version
390 should remain unchanged during a stable series, and should be used for
391 items like the versioned share directory name
392 i.e. /usr/share/guile/1.8.
394 Providing an unchanging version number during a stable release for
395 things like the versioned share directory can be particularly
396 important for Guile "add-on" packages, since it provides a directory
397 that they can install to that won't be changed out from under them
398 with each micro release during a stable series.
400 ** Thread implementation has changed.
402 When you configure "--with-threads=null", you will get the usual
403 threading API (call-with-new-thread, make-mutex, etc), but you can't
404 actually create new threads. Also, "--with-threads=no" is now
405 equivalent to "--with-threads=null". This means that the thread API
406 is always present, although you might not be able to create new
409 When you configure "--with-threads=pthreads" or "--with-threads=yes",
410 you will get threads that are implemented with the portable POSIX
411 threads. These threads can run concurrently (unlike the previous
412 "coop" thread implementation), but need to cooperate for things like
415 The default is "pthreads", unless your platform doesn't have pthreads,
416 in which case "null" threads are used.
418 See the manual for details, nodes "Initialization", "Multi-Threading",
419 "Blocking", and others.
421 ** There is the new notion of 'discouraged' features.
423 This is a milder form of deprecation.
425 Things that are discouraged should not be used in new code, but it is
426 OK to leave them in old code for now. When a discouraged feature is
427 used, no warning message is printed like there is for 'deprecated'
428 features. Also, things that are merely discouraged are nevertheless
429 implemented efficiently, while deprecated features can be very slow.
431 You can omit discouraged features from libguile by configuring it with
432 the '--disable-discouraged' option.
434 ** Deprecation warnings can be controlled at run-time.
436 (debug-enable 'warn-deprecated) switches them on and (debug-disable
437 'warn-deprecated) switches them off.
439 ** Support for SRFI 61, extended cond syntax for multiple values has
442 This SRFI is always available.
444 ** Support for require-extension, SRFI-55, has been added.
446 The SRFI-55 special form `require-extension' has been added. It is
447 available at startup, and provides a portable way to load Scheme
448 extensions. SRFI-55 only requires support for one type of extension,
449 "srfi"; so a set of SRFIs may be loaded via (require-extension (srfi 1
452 ** New module (srfi srfi-26) provides support for `cut' and `cute'.
454 The (srfi srfi-26) module is an implementation of SRFI-26 which
455 provides the `cut' and `cute' syntax. These may be used to specialize
456 parameters without currying.
458 ** New module (srfi srfi-31)
460 This is an implementation of SRFI-31 which provides a special form
461 `rec' for recursive evaluation.
463 ** The modules (srfi srfi-13), (srfi srfi-14) and (srfi srfi-4) have
464 been merged with the core, making their functionality always
467 The modules are still available, tho, and you could use them together
468 with a renaming import, for example.
470 ** Guile no longer includes its own version of libltdl.
472 The official version is good enough now.
474 ** The --enable-htmldoc option has been removed from 'configure'.
476 Support for translating the documentation into HTML is now always
477 provided. Use 'make html'.
479 ** New module (ice-9 serialize):
481 (serialize FORM1 ...) and (parallelize FORM1 ...) are useful when you
482 don't trust the thread safety of most of your program, but where you
483 have some section(s) of code which you consider can run in parallel to
484 other sections. See ice-9/serialize.scm for more information.
486 ** The configure option '--disable-arrays' has been removed.
488 Support for arrays and uniform numeric arrays is now always included
491 * Changes to the stand-alone interpreter
493 ** New command line option `-L'.
495 This option adds a directory to the front of the load path.
497 ** New command line option `--no-debug'.
499 Specifying `--no-debug' on the command line will keep the debugging
500 evaluator turned off, even for interactive sessions.
502 ** User-init file ~/.guile is now loaded with the debugging evaluator.
504 Previously, the normal evaluator would have been used. Using the
505 debugging evaluator gives better error messages.
507 ** The '-e' option now 'read's its argument.
509 This is to allow the new '(@ MODULE-NAME VARIABLE-NAME)' construct to
510 be used with '-e'. For example, you can now write a script like
513 exec guile -e '(@ (demo) main)' -s "$0" "$@"
516 (define-module (demo)
520 (format #t "Demo: ~a~%" args))
523 * Changes to Scheme functions and syntax
525 ** Guardians have changed back to their original semantics
527 Guardians now behave like described in the paper by Dybvig et al. In
528 particular, they no longer make guarantees about the order in which
529 they return objects, and they can no longer be greedy.
531 They no longer drop cyclic data structures.
533 The C function scm_make_guardian has been changed incompatibly and no
534 longer takes the 'greedy_p' argument.
536 ** New function hashx-remove!
538 This function completes the set of 'hashx' functions.
540 ** The concept of dynamic roots has been factored into continuation
541 barriers and dynamic states.
543 Each thread has a current dynamic state that carries the values of the
544 fluids. You can create and copy dynamic states and use them as the
545 second argument for 'eval'. See "Fluids and Dynamic States" in the
548 To restrict the influence that captured continuations can have on the
549 control flow, you can errect continuation barriers. See "Continuation
550 Barriers" in the manual.
552 The function call-with-dynamic-root now essentially temporarily
553 installs a new dynamic state and errects a continuation barrier.
555 ** The default load path no longer includes "." at the end.
557 Automatically loading modules from the current directory should not
558 happen by default. If you want to allow it in a more controlled
559 manner, set the environment variable GUILE_LOAD_PATH or the Scheme
562 ** The uniform vector and array support has been overhauled.
564 It now complies with SRFI-4 and the weird prototype based uniform
565 array creation has been deprecated. See the manual for more details.
567 Some non-compatible changes have been made:
568 - characters can no longer be stored into byte arrays.
569 - strings and bit vectors are no longer considered to be uniform numeric
571 - array-rank throws an error for non-arrays instead of returning zero.
572 - array-ref does no longer accept non-arrays when no indices are given.
574 There is the new notion of 'generalized vectors' and corresponding
575 procedures like 'generalized-vector-ref'. Generalized vectors include
576 strings, bitvectors, ordinary vectors, and uniform numeric vectors.
578 Arrays use generalized vectors as their storage, so that you still
579 have arrays of characters, bits, etc. However, uniform-array-read!
580 and uniform-array-write can no longer read/write strings and
583 ** There is now support for copy-on-write substrings, mutation-sharing
584 substrings and read-only strings.
586 Three new procedures are related to this: substring/shared,
587 substring/copy, and substring/read-only. See the manual for more
590 ** Backtraces will now highlight the value that caused the error.
592 By default, these values are enclosed in "{...}", such as in this
601 <unnamed port>:1:1: In procedure car in expression (car (quote a)):
602 <unnamed port>:1:1: Wrong type (expecting pair): a
603 ABORT: (wrong-type-arg)
605 The prefix and suffix used for highlighting can be set via the two new
606 printer options 'highlight-prefix' and 'highlight-suffix'. For
607 example, putting this into ~/.guile will output the bad value in bold
610 (print-set! highlight-prefix "\x1b[1m")
611 (print-set! highlight-suffix "\x1b[22m")
614 ** 'gettext' support for internationalization has been added.
616 See the manual for details.
618 ** New syntax '@' and '@@':
620 You can now directly refer to variables exported from a module by
623 (@ MODULE-NAME VARIABLE-NAME)
625 For example (@ (ice-9 pretty-print) pretty-print) will directly access
626 the pretty-print variable exported from the (ice-9 pretty-print)
627 module. You don't need to 'use' that module first. You can also use
628 '@' as a target of 'set!', as in (set! (@ mod var) val).
630 The related syntax (@@ MODULE-NAME VARIABLE-NAME) works just like '@',
631 but it can also access variables that have not been exported. It is
632 intended only for kluges and temporary fixes and for debugging, not
635 ** Keyword syntax has been made more disciplined.
637 Previously, the name of a keyword was read as a 'token' but printed as
638 a symbol. Now, it is read as a general Scheme datum which must be a
649 ERROR: In expression (a b c):
653 ERROR: Unbound variable: foo
658 ERROR: Wrong type (expecting symbol): 12
662 ERROR: Wrong type (expecting symbol): (a b c)
666 ** The printing of symbols that might look like keywords can be
669 The new printer option 'quote-keywordish-symbols' controls how symbols
670 are printed that have a colon as their first or last character. The
671 default now is to only quote a symbol with #{...}# when the read
672 option 'keywords' is not '#f'. Thus:
674 guile> (define foo (string->symbol ":foo"))
675 guile> (read-set! keywords #f)
678 guile> (read-set! keywords 'prefix)
681 guile> (print-set! quote-keywordish-symbols #f)
685 ** 'while' now provides 'break' and 'continue'
687 break and continue were previously bound in a while loop, but not
688 documented, and continue didn't quite work properly. The undocumented
689 parameter to break which gave a return value for the while has been
692 ** 'call-with-current-continuation' is now also available under the name
695 ** The module system now checks for duplicate bindings.
697 The module system now can check for name conflicts among imported
700 The behavior can be controlled by specifying one or more 'duplicates'
701 handlers. For example, to make Guile return an error for every name
709 The new default behavior of the module system when a name collision
710 has been detected is to
712 1. Give priority to bindings marked as a replacement.
713 2. Issue a warning (different warning if overriding core binding).
714 3. Give priority to the last encountered binding (this corresponds to
717 If you want the old behavior back without replacements or warnings you
720 (default-duplicate-binding-handler 'last)
722 to your .guile init file.
724 ** New define-module option: :replace
726 :replace works as :export, but, in addition, marks the binding as a
729 A typical example is `format' in (ice-9 format) which is a replacement
730 for the core binding `format'.
732 ** Adding prefixes to imported bindings in the module system
734 There is now a new :use-module option :prefix. It can be used to add
735 a prefix to all imported bindings.
738 :use-module ((bar) :prefix bar:))
740 will import all bindings exported from bar, but rename them by adding
743 ** Conflicting generic functions can be automatically merged.
745 When two imported bindings conflict and they are both generic
746 functions, the two functions can now be merged automatically. This is
747 activated with the 'duplicates' handler 'merge-generics'.
749 ** New function: effective-version
751 Returns the "effective" version number. This is just the normal full
752 version string without the final micro-version number. See "Changes
753 to the distribution" above.
755 ** New threading functions: parallel, letpar, par-map, and friends
757 These are convenient ways to run calculations in parallel in new
758 threads. See "Parallel forms" in the manual for details.
760 ** New function 'try-mutex'.
762 This function will attempt to lock a mutex but will return immediately
763 instead of blocking and indicate failure.
765 ** Waiting on a condition variable can have a timeout.
767 The function 'wait-condition-variable' now takes a third, optional
768 argument that specifies the point in time where the waiting should be
771 ** New function 'broadcast-condition-variable'.
773 ** New functions 'all-threads' and 'current-thread'.
775 ** Signals and system asyncs work better with threads.
777 The function 'sigaction' now takes a fourth, optional, argument that
778 specifies the thread that the handler should run in. When the
779 argument is omitted, the handler will run in the thread that called
782 Likewise, 'system-async-mark' takes a second, optional, argument that
783 specifies the thread that the async should run in. When it is
784 omitted, the async will run in the thread that called
787 C code can use the new functions scm_sigaction_for_thread and
788 scm_system_async_mark_for_thread to pass the new thread argument.
790 When a thread blocks on a mutex, a condition variable or is waiting
791 for IO to be possible, it will still execute system asyncs. This can
792 be used to interrupt such a thread by making it execute a 'throw', for
795 ** The function 'system-async' is deprecated.
797 You can now pass any zero-argument procedure to 'system-async-mark'.
798 The function 'system-async' will just return its argument unchanged
801 ** New functions 'call-with-blocked-asyncs' and
802 'call-with-unblocked-asyncs'
804 The expression (call-with-blocked-asyncs PROC) will call PROC and will
805 block execution of system asyncs for the current thread by one level
806 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
807 procedure and will unblock the execution of system asyncs by one
808 level for the current thread.
810 Only system asyncs are affected by these functions.
812 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
814 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
815 instead. Those functions are easier to use correctly and can be
818 ** New function 'unsetenv'.
820 ** New macro 'define-syntax-public'.
822 It works like 'define-syntax' and also exports the defined macro (but
825 ** There is support for Infinity and NaNs.
827 Following PLT Scheme, Guile can now work with infinite numbers, and
830 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
831 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
832 "+nan.0"). These numbers are inexact and have no exact counterpart.
834 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
835 sign of the dividend. The infinities are integers, and they answer #t
836 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
837 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
848 ERROR: Numerical overflow
850 Two new predicates 'inf?' and 'nan?' can be used to test for the
853 ** Inexact zero can have a sign.
855 Guile can now distinguish between plus and minus inexact zero, if your
856 platform supports this, too. The two zeros are equal according to
857 '=', but not according to 'eqv?'. For example
868 ** Guile now has exact rationals.
870 Guile can now represent fractions such as 1/3 exactly. Computing with
871 them is also done exactly, of course:
876 ** 'floor', 'ceiling', 'round' and 'truncate' now return exact numbers
879 For example: (floor 2) now returns an exact 2 where in the past it
880 returned an inexact 2.0. Likewise, (floor 5/4) returns an exact 1.
882 ** inexact->exact no longer returns only integers.
884 Without exact rationals, the closest exact number was always an
885 integer, but now inexact->exact returns the fraction that is exactly
886 equal to a floating point number. For example:
888 (inexact->exact 1.234)
889 => 694680242521899/562949953421312
891 When you want the old behavior, use 'round' explicitly:
893 (inexact->exact (round 1.234))
896 ** New function 'rationalize'.
898 This function finds a simple fraction that is close to a given real
899 number. For example (and compare with inexact->exact above):
901 (rationalize (inexact->exact 1.234) 1/2000)
904 Note that, as required by R5RS, rationalize returns only then an exact
905 result when both its arguments are exact.
907 ** 'odd?' and 'even?' work also for inexact integers.
909 Previously, (odd? 1.0) would signal an error since only exact integers
910 were recognized as integers. Now (odd? 1.0) returns #t, (odd? 2.0)
911 returns #f and (odd? 1.5) signals an error.
913 ** Guile now has uninterned symbols.
915 The new function 'make-symbol' will return an uninterned symbol. This
916 is a symbol that is unique and is guaranteed to remain unique.
917 However, uninterned symbols can not yet be read back in.
919 Use the new function 'symbol-interned?' to check whether a symbol is
922 ** pretty-print has more options.
924 The function pretty-print from the (ice-9 pretty-print) module can now
925 also be invoked with keyword arguments that control things like
926 maximum output width. See the manual for details.
928 ** Variables have no longer a special behavior for `equal?'.
930 Previously, comparing two variables with `equal?' would recursivly
931 compare their values. This is no longer done. Variables are now only
932 `equal?' if they are `eq?'.
934 ** `(begin)' is now valid.
936 You can now use an empty `begin' form. It will yield #<unspecified>
937 when evaluated and simply be ignored in a definition context.
939 ** Deprecated: procedure->macro
941 Change your code to use 'define-macro' or r5rs macros. Also, be aware
942 that macro expansion will not be done during evaluation, but prior to
945 ** Soft ports now allow a `char-ready?' procedure
947 The vector argument to `make-soft-port' can now have a length of
948 either 5 or 6. (Previously the length had to be 5.) The optional 6th
949 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
950 that returns the number of characters that can be read immediately
951 without the soft port blocking.
953 ** Deprecated: undefine
955 There is no replacement for undefine.
957 ** The functions make-keyword-from-dash-symbol and keyword-dash-symbol
958 have been discouraged.
960 They are relics from a time where a keyword like #:foo was used
961 directly as a Tcl option "-foo" and thus keywords were internally
962 stored as a symbol with a starting dash. We now store a symbol
965 Use symbol->keyword and keyword->symbol instead.
967 ** The `cheap' debug option is now obsolete
969 Evaluator trap calls are now unconditionally "cheap" - in other words,
970 they pass a debug object to the trap handler rather than a full
971 continuation. The trap handler code can capture a full continuation
972 by using `call-with-current-continuation' in the usual way, if it so
975 The `cheap' option is retained for now so as not to break existing
976 code which gets or sets it, but setting it now has no effect. It will
977 be removed in the next major Guile release.
979 ** Evaluator trap calls now support `tweaking'
981 `Tweaking' means that the trap handler code can modify the Scheme
982 expression that is about to be evaluated (in the case of an
983 enter-frame trap) or the value that is being returned (in the case of
984 an exit-frame trap). The trap handler code indicates that it wants to
985 do this by returning a pair whose car is the symbol 'instead and whose
986 cdr is the modified expression or return value.
988 * Changes to the C interface
990 ** The functions scm_hash_fn_remove_x and scm_hashx_remove_x no longer
991 take a 'delete' function argument.
993 This argument makes no sense since the delete function is used to
994 remove a pair from an alist, and this must not be configurable.
996 This is an incompatible change.
998 ** The GH interface is now subject to the deprecation mechanism
1000 The GH interface has been deprecated for quite some time but now it is
1001 actually removed from Guile when it is configured with
1002 --disable-deprecated.
1004 See the manual "Transitioning away from GH" for more information.
1006 ** A new family of functions for converting between C values and
1007 Scheme values has been added.
1009 These functions follow a common naming scheme and are designed to be
1010 easier to use, thread-safe and more future-proof than the older
1013 - int scm_is_* (...)
1015 These are predicates that return a C boolean: 1 or 0. Instead of
1016 SCM_NFALSEP, you can now use scm_is_true, for example.
1018 - <type> scm_to_<type> (SCM val, ...)
1020 These are functions that convert a Scheme value into an appropriate
1021 C value. For example, you can use scm_to_int to safely convert from
1024 - SCM scm_from_<type> (<type> val, ...)
1026 These functions convert from a C type to a SCM value; for example,
1027 scm_from_int for ints.
1029 There is a huge number of these functions, for numbers, strings,
1030 symbols, vectors, etc. They are documented in the reference manual in
1031 the API section together with the types that they apply to.
1033 ** New functions for dealing with complex numbers in C have been added.
1035 The new functions are scm_c_make_rectangular, scm_c_make_polar,
1036 scm_c_real_part, scm_c_imag_part, scm_c_magnitude and scm_c_angle.
1037 They work like scm_make_rectangular etc but take or return doubles
1040 ** The function scm_make_complex has been discouraged.
1042 Use scm_c_make_rectangular instead.
1044 ** The INUM macros have been deprecated.
1046 A lot of code uses these macros to do general integer conversions,
1047 although the macros only work correctly with fixnums. Use the
1048 following alternatives.
1050 SCM_INUMP -> scm_is_integer or similar
1051 SCM_NINUMP -> !scm_is_integer or similar
1052 SCM_MAKINUM -> scm_from_int or similar
1053 SCM_INUM -> scm_to_int or similar
1055 SCM_VALIDATE_INUM_* -> Do not use these; scm_to_int, etc. will
1056 do the validating for you.
1058 ** The scm_num2<type> and scm_<type>2num functions and scm_make_real
1059 have been discouraged.
1061 Use the newer scm_to_<type> and scm_from_<type> functions instead for
1062 new code. The functions have been discouraged since they don't fit
1065 ** The 'boolean' macros SCM_FALSEP etc have been discouraged.
1067 They have strange names, especially SCM_NFALSEP, and SCM_BOOLP
1068 evaluates its argument twice. Use scm_is_true, etc. instead for new
1071 ** The macro SCM_EQ_P has been discouraged.
1073 Use scm_is_eq for new code, which fits better into the naming
1076 ** The macros SCM_CONSP, SCM_NCONSP, SCM_NULLP, and SCM_NNULLP have
1079 Use the function scm_is_pair or scm_is_null instead.
1081 ** The functions scm_round and scm_truncate have been deprecated and
1082 are now available as scm_c_round and scm_c_truncate, respectively.
1084 These functions occupy the names that scm_round_number and
1085 scm_truncate_number should have.
1087 ** The functions scm_c_string2str, scm_c_substring2str, and
1088 scm_c_symbol2str have been deprecated.
1090 Use scm_to_locale_stringbuf or similar instead, maybe together with
1093 ** New functions scm_c_make_string, scm_c_string_length,
1094 scm_c_string_ref, scm_c_string_set_x, scm_c_substring,
1095 scm_c_substring_shared, scm_c_substring_copy.
1097 These are like scm_make_string, scm_length, etc. but are slightly
1098 easier to use from C.
1100 ** The macros SCM_STRINGP, SCM_STRING_CHARS, SCM_STRING_LENGTH,
1101 SCM_SYMBOL_CHARS, and SCM_SYMBOL_LENGTH have been deprecated.
1103 They export too many assumptions about the implementation of strings
1104 and symbols that are no longer true in the presence of
1105 mutation-sharing substrings and when Guile switches to some form of
1108 When working with strings, it is often best to use the normal string
1109 functions provided by Guile, such as scm_c_string_ref,
1110 scm_c_string_set_x, scm_string_append, etc. Be sure to look in the
1111 manual since many more such functions are now provided than
1114 When you want to convert a SCM string to a C string, use the
1115 scm_to_locale_string function or similar instead. For symbols, use
1116 scm_symbol_to_string and then work with that string. Because of the
1117 new string representation, scm_symbol_to_string does not need to copy
1118 and is thus quite efficient.
1120 ** Some string, symbol and keyword functions have been discouraged.
1122 They don't fit into the uniform naming scheme and are not explicit
1123 about the character encoding.
1125 Replace according to the following table:
1127 scm_allocate_string -> scm_c_make_string
1128 scm_take_str -> scm_take_locale_stringn
1129 scm_take0str -> scm_take_locale_string
1130 scm_mem2string -> scm_from_locale_stringn
1131 scm_str2string -> scm_from_locale_string
1132 scm_makfrom0str -> scm_from_locale_string
1133 scm_mem2symbol -> scm_from_locale_symboln
1134 scm_mem2uninterned_symbol -> scm_from_locale_stringn + scm_make_symbol
1135 scm_str2symbol -> scm_from_locale_symbol
1137 SCM_SYMBOL_HASH -> scm_hashq
1138 SCM_SYMBOL_INTERNED_P -> scm_symbol_interned_p
1140 scm_c_make_keyword -> scm_from_locale_keyword
1142 ** The functions scm_keyword_to_symbol and sym_symbol_to_keyword are
1143 now also available to C code.
1145 ** SCM_KEYWORDP and SCM_KEYWORDSYM have been deprecated.
1147 Use scm_is_keyword and scm_keyword_to_symbol instead, but note that
1148 the latter returns the true name of the keyword, not the 'dash name',
1149 as SCM_KEYWORDSYM used to do.
1151 ** A new way to access arrays in a thread-safe and efficient way has
1154 See the manual, node "Accessing Arrays From C".
1156 ** The old uniform vector and bitvector implementations have been
1157 unceremoniously removed.
1159 This implementation exposed the details of the tagging system of
1160 Guile. Use the new C API explained in the manual in node "Uniform
1161 Numeric Vectors" and "Bit Vectors", respectively.
1163 The following macros are gone: SCM_UVECTOR_BASE, SCM_SET_UVECTOR_BASE,
1164 SCM_UVECTOR_MAXLENGTH, SCM_UVECTOR_LENGTH, SCM_MAKE_UVECTOR_TAG,
1165 SCM_SET_UVECTOR_LENGTH, SCM_BITVECTOR_P, SCM_BITVECTOR_BASE,
1166 SCM_SET_BITVECTOR_BASE, SCM_BITVECTOR_MAX_LENGTH,
1167 SCM_BITVECTOR_LENGTH, SCM_MAKE_BITVECTOR_TAG,
1168 SCM_SET_BITVECTOR_LENGTH, SCM_BITVEC_REF, SCM_BITVEC_SET,
1171 ** The macros dealing with vectors have been deprecated.
1173 Use the new functions scm_is_vector, scm_vector_elements,
1174 scm_vector_writable_elements, etc, or scm_is_simple_vector,
1175 SCM_SIMPLE_VECTOR_REF, SCM_SIMPLE_VECTOR_SET, etc instead. See the
1176 manual for more details.
1178 Deprecated are SCM_VECTORP, SCM_VELTS, SCM_VECTOR_MAX_LENGTH,
1179 SCM_VECTOR_LENGTH, SCM_VECTOR_REF, SCM_VECTOR_SET, SCM_WRITABLE_VELTS.
1181 The following macros have been removed: SCM_VECTOR_BASE,
1182 SCM_SET_VECTOR_BASE, SCM_MAKE_VECTOR_TAG, SCM_SET_VECTOR_LENGTH,
1183 SCM_VELTS_AS_STACKITEMS, SCM_SETVELTS, SCM_GC_WRITABLE_VELTS.
1185 ** Some C functions and macros related to arrays have been deprecated.
1187 Migrate according to the following table:
1189 scm_make_uve -> scm_make_typed_array, scm_make_u8vector etc.
1190 scm_make_ra -> scm_make_array
1191 scm_shap2ra -> scm_make_array
1192 scm_cvref -> scm_c_generalized_vector_ref
1193 scm_ra_set_contp -> do not use
1194 scm_aind -> scm_array_handle_pos
1195 scm_raprin1 -> scm_display or scm_write
1197 SCM_ARRAYP -> scm_is_array
1198 SCM_ARRAY_NDIM -> scm_c_array_rank
1199 SCM_ARRAY_DIMS -> scm_array_handle_dims
1200 SCM_ARRAY_CONTP -> do not use
1201 SCM_ARRAY_MEM -> do not use
1202 SCM_ARRAY_V -> scm_array_handle_elements or similar
1203 SCM_ARRAY_BASE -> do not use
1205 ** SCM_CELL_WORD_LOC has been deprecated.
1207 Use the new macro SCM_CELL_OBJECT_LOC instead, which returns a pointer
1208 to a SCM, as opposed to a pointer to a scm_t_bits.
1210 This was done to allow the correct use of pointers into the Scheme
1211 heap. Previously, the heap words were of type scm_t_bits and local
1212 variables and function arguments were of type SCM, making it
1213 non-standards-conformant to have a pointer that can point to both.
1215 ** New macros SCM_SMOB_DATA_2, SCM_SMOB_DATA_3, etc.
1217 These macros should be used instead of SCM_CELL_WORD_2/3 to access the
1218 second and third words of double smobs. Likewise for
1219 SCM_SET_SMOB_DATA_2 and SCM_SET_SMOB_DATA_3.
1221 Also, there is SCM_SMOB_FLAGS and SCM_SET_SMOB_FLAGS that should be
1222 used to get and set the 16 exra bits in the zeroth word of a smob.
1224 And finally, there is SCM_SMOB_OBJECT and SCM_SMOB_SET_OBJECT for
1225 accesing the first immediate word of a smob as a SCM value, and there
1226 is SCM_SMOB_OBJECT_LOC for getting a pointer to the first immediate
1227 smob word. Like wise for SCM_SMOB_OBJECT_2, etc.
1229 ** New way to deal with non-local exits and re-entries.
1231 There is a new set of functions that essentially do what
1232 scm_internal_dynamic_wind does, but in a way that is more convenient
1233 for C code in some situations. Here is a quick example of how to
1234 prevent a potential memory leak:
1241 scm_dynwind_begin (0);
1243 mem = scm_malloc (100);
1244 scm_dynwind_unwind_handler (free, mem, SCM_F_WIND_EXPLICITLY);
1246 /* MEM would leak if BAR throws an error.
1247 SCM_DYNWIND_UNWIND_HANDLER frees it nevertheless.
1254 /* Because of SCM_F_WIND_EXPLICITLY, MEM will be freed by
1255 SCM_DYNWIND_END as well.
1259 For full documentation, see the node "Dynamic Wind" in the manual.
1261 ** New function scm_dynwind_free
1263 This function calls 'free' on a given pointer when a dynwind context
1264 is left. Thus the call to scm_dynwind_unwind_handler above could be
1265 replaced with simply scm_dynwind_free (mem).
1267 ** New functions scm_c_call_with_blocked_asyncs and
1268 scm_c_call_with_unblocked_asyncs
1270 Like scm_call_with_blocked_asyncs etc. but for C functions.
1272 ** New functions scm_dynwind_block_asyncs and scm_dynwind_unblock_asyncs
1274 In addition to scm_c_call_with_blocked_asyncs you can now also use
1275 scm_dynwind_block_asyncs in a 'dynwind context' (see above). Likewise for
1276 scm_c_call_with_unblocked_asyncs and scm_dynwind_unblock_asyncs.
1278 ** The macros SCM_DEFER_INTS, SCM_ALLOW_INTS, SCM_REDEFER_INTS,
1279 SCM_REALLOW_INTS have been deprecated.
1281 They do no longer fulfill their original role of blocking signal
1282 delivery. Depending on what you want to achieve, replace a pair of
1283 SCM_DEFER_INTS and SCM_ALLOW_INTS with a dynwind context that locks a
1284 mutex, blocks asyncs, or both. See node "Critical Sections" in the
1287 ** The value 'scm_mask_ints' is no longer writable.
1289 Previously, you could set scm_mask_ints directly. This is no longer
1290 possible. Use scm_c_call_with_blocked_asyncs and
1291 scm_c_call_with_unblocked_asyncs instead.
1293 ** New way to temporarily set the current input, output or error ports
1295 C code can now use scm_dynwind_current_<foo>_port in a 'dynwind
1296 context' (see above). <foo> is one of "input", "output" or "error".
1298 ** New way to temporarily set fluids
1300 C code can now use scm_dynwind_fluid in a 'dynwind context' (see
1301 above) to temporarily set the value of a fluid.
1303 ** New types scm_t_intmax and scm_t_uintmax.
1305 On platforms that have them, these types are identical to intmax_t and
1306 uintmax_t, respectively. On other platforms, they are identical to
1307 the largest integer types that Guile knows about.
1309 ** The functions scm_unmemocopy and scm_unmemoize have been removed.
1311 You should not have used them.
1313 ** Many public #defines with generic names have been made private.
1315 #defines with generic names like HAVE_FOO or SIZEOF_FOO have been made
1316 private or renamed with a more suitable public name.
1318 ** The macro SCM_TYP16S has been deprecated.
1320 This macro is not intended for public use.
1322 ** The macro SCM_SLOPPY_INEXACTP has been deprecated.
1324 Use scm_is_true (scm_inexact_p (...)) instead.
1326 ** The macro SCM_SLOPPY_REALP has been deprecated.
1328 Use scm_is_real instead.
1330 ** The macro SCM_SLOPPY_COMPLEXP has been deprecated.
1332 Use scm_is_complex instead.
1334 ** Some preprocessor defines have been deprecated.
1336 These defines indicated whether a certain feature was present in Guile
1337 or not. Going forward, assume that the features are always present.
1339 The macros are: USE_THREADS, GUILE_ISELECT, READER_EXTENSIONS,
1340 DEBUG_EXTENSIONS, DYNAMIC_LINKING.
1342 The following macros have been removed completely: MEMOIZE_LOCALS,
1343 SCM_RECKLESS, SCM_CAUTIOUS.
1345 ** The preprocessor define STACK_DIRECTION has been deprecated.
1347 There should be no need to know about the stack direction for ordinary
1350 ** New function: scm_effective_version
1352 Returns the "effective" version number. This is just the normal full
1353 version string without the final micro-version number. See "Changes
1354 to the distribution" above.
1356 ** The function scm_call_with_new_thread has a new prototype.
1358 Instead of taking a list with the thunk and handler, these two
1359 arguments are now passed directly:
1361 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
1363 This is an incompatible change.
1365 ** New snarfer macro SCM_DEFINE_PUBLIC.
1367 This is like SCM_DEFINE, but also calls scm_c_export for the defined
1368 function in the init section.
1370 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
1372 ** Garbage collector rewrite.
1374 The garbage collector is cleaned up a lot, and now uses lazy
1375 sweeping. This is reflected in the output of (gc-stats); since cells
1376 are being freed when they are allocated, the cells-allocated field
1377 stays roughly constant.
1379 For malloc related triggers, the behavior is changed. It uses the same
1380 heuristic as the cell-triggered collections. It may be tuned with the
1381 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
1382 for minimum yield of malloc related triggers. The default is 40.
1383 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
1386 Debugging operations for the freelist have been deprecated, along with
1387 the C variables that control garbage collection. The environment
1388 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
1389 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
1391 For understanding the memory usage of a GUILE program, the routine
1392 gc-live-object-stats returns an alist containing the number of live
1393 objects for every type.
1396 ** The function scm_definedp has been renamed to scm_defined_p
1398 The name scm_definedp is deprecated.
1400 ** The struct scm_cell type has been renamed to scm_t_cell
1402 This is in accordance to Guile's naming scheme for types. Note that
1403 the name scm_cell is now used for a function that allocates and
1404 initializes a new cell (see below).
1406 ** New functions for memory management
1408 A new set of functions for memory management has been added since the
1409 old way (scm_must_malloc, scm_must_free, etc) was error prone and
1410 indeed, Guile itself contained some long standing bugs that could
1411 cause aborts in long running programs.
1413 The new functions are more symmetrical and do not need cooperation
1414 from smob free routines, among other improvements.
1416 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
1417 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
1418 scm_gc_free, scm_gc_register_collectable_memory, and
1419 scm_gc_unregister_collectable_memory. Refer to the manual for more
1420 details and for upgrading instructions.
1422 The old functions for memory management have been deprecated. They
1423 are: scm_must_malloc, scm_must_realloc, scm_must_free,
1424 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
1426 ** Declarations of exported features are marked with SCM_API.
1428 Every declaration of a feature that belongs to the exported Guile API
1429 has been marked by adding the macro "SCM_API" to the start of the
1430 declaration. This macro can expand into different things, the most
1431 common of which is just "extern" for Unix platforms. On Win32, it can
1432 be used to control which symbols are exported from a DLL.
1434 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
1435 will expand into "__declspec (dllimport) extern", which is needed for
1436 linking to the Guile DLL in Windows.
1438 There are also SCM_RL_IMPORT, SCM_SRFI1314_IMPORT, and
1439 SCM_SRFI4_IMPORT, for the corresponding libraries.
1441 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
1443 Use the new functions scm_cell and scm_double_cell instead. The old
1444 macros had problems because with them allocation and initialization
1445 was separated and the GC could sometimes observe half initialized
1446 cells. Only careful coding by the user of SCM_NEWCELL and
1447 SCM_NEWCELL2 could make this safe and efficient.
1449 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
1451 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
1454 ** SRCBRKP has been deprecated.
1456 Use scm_c_source_property_breakpoint_p instead.
1458 ** Deprecated: scm_makmacro
1460 Change your code to use either scm_makmmacro or to define macros in
1461 Scheme, using 'define-macro'.
1463 ** New function scm_c_port_for_each.
1465 This function is like scm_port_for_each but takes a pointer to a C
1466 function as the callback instead of a SCM value.
1468 ** The names scm_internal_select, scm_thread_sleep, and
1469 scm_thread_usleep have been discouraged.
1471 Use scm_std_select, scm_std_sleep, scm_std_usleep instead.
1473 ** The GC can no longer be blocked.
1475 The global flags scm_gc_heap_lock and scm_block_gc have been removed.
1476 The GC can now run (partially) concurrently with other code and thus
1477 blocking it is not well defined.
1479 ** Many definitions have been removed that were previously deprecated.
1481 scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify, s_t_ify,
1482 scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
1483 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2,
1484 scm_tc16_allocated, SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY,
1485 SCM_IM_0_COND, SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED,
1486 scm_debug_newcell, scm_debug_newcell2, SCM_HUP_SIGNAL, SCM_INT_SIGNAL,
1487 SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL,
1488 SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD, SCM_ORD_SIG,
1489 SCM_NUM_SIGS, scm_top_level_lookup_closure_var,
1490 *top-level-lookup-closure*, scm_system_transformer, scm_eval_3,
1491 scm_eval2, root_module_lookup_closure, SCM_SLOPPY_STRINGP,
1492 SCM_RWSTRINGP, scm_read_only_string_p, scm_make_shared_substring,
1493 scm_tc7_substring, sym_huh, SCM_VARVCELL, SCM_UDVARIABLEP,
1494 SCM_DEFVARIABLEP, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
1495 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
1496 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
1497 SCM_LENGTH_MAX, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
1498 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR,
1499 scm_sym2vcell, scm_intern, scm_intern0, scm_sysintern, scm_sysintern0,
1500 scm_sysintern0_no_module_lookup, scm_init_symbols_deprecated,
1501 scm_vector_set_length_x, scm_contregs, scm_debug_info,
1502 scm_debug_frame, SCM_DSIDEVAL, SCM_CONST_LONG, SCM_VCELL,
1503 SCM_GLOBAL_VCELL, SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT,
1504 SCM_HUGE_LENGTH, SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
1505 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
1506 SCM_VALIDATE_RWSTRING, DIGITS, scm_small_istr2int, scm_istr2int,
1507 scm_istr2flo, scm_istring2number, scm_istr2int, scm_istr2flo,
1508 scm_istring2number, scm_vtable_index_vcell, scm_si_vcell, SCM_ECONSP,
1509 SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
1510 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable,
1511 SCM_CHARS, SCM_LENGTH, SCM_SET_STRING_CHARS, SCM_SET_STRING_LENGTH.
1513 * Changes to bundled modules
1517 Using the (ice-9 debug) module no longer automatically switches Guile
1518 to use the debugging evaluator. If you want to switch to the
1519 debugging evaluator (which is needed for backtrace information if you
1520 hit an error), please add an explicit "(debug-enable 'debug)" to your
1521 code just after the code to use (ice-9 debug).
1524 Changes since Guile 1.4:
1526 * Changes to the distribution
1528 ** A top-level TODO file is included.
1530 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
1532 Guile now always uses three numbers to represent the version,
1533 i.e. "1.6.5". The first number, 1, is the major version number, the
1534 second number, 6, is the minor version number, and the third number,
1535 5, is the micro version number. Changes in major version number
1536 indicate major changes in Guile.
1538 Minor version numbers that are even denote stable releases, and odd
1539 minor version numbers denote development versions (which may be
1540 unstable). The micro version number indicates a minor sub-revision of
1541 a given MAJOR.MINOR release.
1543 In keeping with the new scheme, (minor-version) and scm_minor_version
1544 no longer return everything but the major version number. They now
1545 just return the minor version number. Two new functions
1546 (micro-version) and scm_micro_version have been added to report the
1547 micro version number.
1549 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
1551 ** New preprocessor definitions are available for checking versions.
1553 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
1554 SCM_MICRO_VERSION to the appropriate integer values.
1556 ** Guile now actively warns about deprecated features.
1558 The new configure option `--enable-deprecated=LEVEL' and the
1559 environment variable GUILE_WARN_DEPRECATED control this mechanism.
1560 See INSTALL and README for more information.
1562 ** Guile is much more likely to work on 64-bit architectures.
1564 Guile now compiles and passes "make check" with only two UNRESOLVED GC
1565 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
1566 for the use of a test machine, and thanks to Stefan Jahn for ia64
1569 ** New functions: setitimer and getitimer.
1571 These implement a fairly direct interface to the libc functions of the
1574 ** The #. reader extension is now disabled by default.
1576 For safety reasons, #. evaluation is disabled by default. To
1577 re-enable it, set the fluid read-eval? to #t. For example:
1579 (fluid-set! read-eval? #t)
1581 but make sure you realize the potential security risks involved. With
1582 read-eval? enabled, reading a data file from an untrusted source can
1585 ** New SRFI modules have been added:
1587 SRFI-0 `cond-expand' is now supported in Guile, without requiring
1590 (srfi srfi-1) is a library containing many useful pair- and list-processing
1593 (srfi srfi-2) exports and-let*.
1595 (srfi srfi-4) implements homogeneous numeric vector datatypes.
1597 (srfi srfi-6) is a dummy module for now, since guile already provides
1598 all of the srfi-6 procedures by default: open-input-string,
1599 open-output-string, get-output-string.
1601 (srfi srfi-8) exports receive.
1603 (srfi srfi-9) exports define-record-type.
1605 (srfi srfi-10) exports define-reader-ctor and implements the reader
1608 (srfi srfi-11) exports let-values and let*-values.
1610 (srfi srfi-13) implements the SRFI String Library.
1612 (srfi srfi-14) implements the SRFI Character-Set Library.
1614 (srfi srfi-17) implements setter and getter-with-setter and redefines
1615 some accessor procedures as procedures with getters. (such as car,
1616 cdr, vector-ref etc.)
1618 (srfi srfi-19) implements the SRFI Time/Date Library.
1620 ** New scripts / "executable modules"
1622 Subdirectory "scripts" contains Scheme modules that are packaged to
1623 also be executable as scripts. At this time, these scripts are available:
1632 See README there for more info.
1634 These scripts can be invoked from the shell with the new program
1635 "guile-tools", which keeps track of installation directory for you.
1638 $ guile-tools display-commentary srfi/*.scm
1640 guile-tools is copied to the standard $bindir on "make install".
1642 ** New module (ice-9 stack-catch):
1644 stack-catch is like catch, but saves the current state of the stack in
1645 the fluid the-last-stack. This fluid can be useful when using the
1646 debugger and when re-throwing an error.
1648 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
1650 This has been done to prevent problems on lesser operating systems
1651 that can't tolerate `*'s in file names. The exported macro continues
1652 to be named `and-let*', of course.
1654 On systems that support it, there is also a compatibility module named
1655 (ice-9 and-let*). It will go away in the next release.
1657 ** New modules (oop goops) etc.:
1660 (oop goops describe)
1662 (oop goops active-slot)
1663 (oop goops composite-slot)
1665 The Guile Object Oriented Programming System (GOOPS) has been
1666 integrated into Guile. For further information, consult the GOOPS
1667 manual and tutorial in the `doc' directory.
1669 ** New module (ice-9 rdelim).
1671 This exports the following procedures which were previously defined
1672 in the default environment:
1674 read-line read-line! read-delimited read-delimited! %read-delimited!
1675 %read-line write-line
1677 For backwards compatibility the definitions are still imported into the
1678 default environment in this version of Guile. However you should add:
1680 (use-modules (ice-9 rdelim))
1682 to any program which uses the definitions, since this may change in
1685 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
1686 can be used for similar functionality.
1688 ** New module (ice-9 rw)
1690 This is a subset of the (scsh rw) module from guile-scsh. Currently
1691 it defines two procedures:
1693 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
1695 Read characters from a port or file descriptor into a string STR.
1696 A port must have an underlying file descriptor -- a so-called
1697 fport. This procedure is scsh-compatible and can efficiently read
1700 *** New function: write-string/partial str [port_or_fdes [start [end]]]
1702 Write characters from a string STR to a port or file descriptor.
1703 A port must have an underlying file descriptor -- a so-called
1704 fport. This procedure is mostly compatible and can efficiently
1705 write large strings.
1707 ** New module (ice-9 match)
1709 This module includes Andrew K. Wright's pattern matcher. See
1710 ice-9/match.scm for brief description or
1712 http://www.star-lab.com/wright/code.html
1714 for complete documentation.
1716 ** New module (ice-9 buffered-input)
1718 This module provides procedures to construct an input port from an
1719 underlying source of input that reads and returns its input in chunks.
1720 The underlying input source is a Scheme procedure, specified by the
1721 caller, which the port invokes whenever it needs more input.
1723 This is useful when building an input port whose back end is Readline
1724 or a UI element such as the GtkEntry widget.
1728 The reference and tutorial documentation that was previously
1729 distributed separately, as `guile-doc', is now included in the core
1730 Guile distribution. The documentation consists of the following
1733 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
1736 - The Guile Reference Manual (guile.texi) contains (or is intended to
1737 contain) reference documentation on all aspects of Guile.
1739 - The GOOPS Manual (goops.texi) contains both tutorial-style and
1740 reference documentation for using GOOPS, Guile's Object Oriented
1743 - The Revised^5 Report on the Algorithmic Language Scheme
1746 See the README file in the `doc' directory for more details.
1748 ** There are a couple of examples in the examples/ directory now.
1750 * Changes to the stand-alone interpreter
1752 ** New command line option `--use-srfi'
1754 Using this option, SRFI modules can be loaded on startup and be
1755 available right from the beginning. This makes programming portable
1756 Scheme programs easier.
1758 The option `--use-srfi' expects a comma-separated list of numbers,
1759 each representing a SRFI number to be loaded into the interpreter
1760 before starting evaluating a script file or the REPL. Additionally,
1761 the feature identifier for the loaded SRFIs is recognized by
1762 `cond-expand' when using this option.
1765 $ guile --use-srfi=8,13
1766 guile> (receive (x z) (values 1 2) (+ 1 2))
1768 guile> (string-pad "bla" 20)
1771 ** Guile now always starts up in the `(guile-user)' module.
1773 Previously, scripts executed via the `-s' option would run in the
1774 `(guile)' module and the repl would run in the `(guile-user)' module.
1775 Now every user action takes place in the `(guile-user)' module by
1778 * Changes to Scheme functions and syntax
1780 ** Character classifiers work for non-ASCII characters.
1782 The predicates `char-alphabetic?', `char-numeric?',
1783 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
1784 no longer check whether their arguments are ASCII characters.
1785 Previously, a character would only be considered alphabetic when it
1786 was also ASCII, for example.
1788 ** Previously deprecated Scheme functions have been removed:
1790 tag - no replacement.
1791 fseek - replaced by seek.
1792 list* - replaced by cons*.
1794 ** It's now possible to create modules with controlled environments
1798 (use-modules (ice-9 safe))
1799 (define m (make-safe-module))
1800 ;;; m will now be a module containing only a safe subset of R5RS
1801 (eval '(+ 1 2) m) --> 3
1802 (eval 'load m) --> ERROR: Unbound variable: load
1804 ** Evaluation of "()", the empty list, is now an error.
1806 Previously, the expression "()" evaluated to the empty list. This has
1807 been changed to signal a "missing expression" error. The correct way
1808 to write the empty list as a literal constant is to use quote: "'()".
1810 ** New concept of `Guile Extensions'.
1812 A Guile Extension is just a ordinary shared library that can be linked
1813 at run-time. We found it advantageous to give this simple concept a
1814 dedicated name to distinguish the issues related to shared libraries
1815 from the issues related to the module system.
1817 *** New function: load-extension
1819 Executing (load-extension lib init) is mostly equivalent to
1821 (dynamic-call init (dynamic-link lib))
1823 except when scm_register_extension has been called previously.
1824 Whenever appropriate, you should use `load-extension' instead of
1825 dynamic-link and dynamic-call.
1827 *** New C function: scm_c_register_extension
1829 This function registers a initialization function for use by
1830 `load-extension'. Use it when you don't want specific extensions to
1831 be loaded as shared libraries (for example on platforms that don't
1832 support dynamic linking).
1834 ** Auto-loading of compiled-code modules is deprecated.
1836 Guile used to be able to automatically find and link a shared
1837 library to satisfy requests for a module. For example, the module
1838 `(foo bar)' could be implemented by placing a shared library named
1839 "foo/libbar.so" (or with a different extension) in a directory on the
1842 This has been found to be too tricky, and is no longer supported. The
1843 shared libraries are now called "extensions". You should now write a
1844 small Scheme file that calls `load-extension' to load the shared
1845 library and initialize it explicitly.
1847 The shared libraries themselves should be installed in the usual
1848 places for shared libraries, with names like "libguile-foo-bar".
1850 For example, place this into a file "foo/bar.scm"
1852 (define-module (foo bar))
1854 (load-extension "libguile-foo-bar" "foobar_init")
1856 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
1858 `eval' is now R5RS, that is it takes two arguments.
1859 The second argument is an environment specifier, i.e. either
1861 (scheme-report-environment 5)
1862 (null-environment 5)
1863 (interaction-environment)
1869 ** The module system has been made more disciplined.
1871 The function `eval' will save and restore the current module around
1872 the evaluation of the specified expression. While this expression is
1873 evaluated, `(current-module)' will now return the right module, which
1874 is the module specified as the second argument to `eval'.
1876 A consequence of this change is that `eval' is not particularly
1877 useful when you want allow the evaluated code to change what module is
1878 designated as the current module and have this change persist from one
1879 call to `eval' to the next. The read-eval-print-loop is an example
1880 where `eval' is now inadequate. To compensate, there is a new
1881 function `primitive-eval' that does not take a module specifier and
1882 that does not save/restore the current module. You should use this
1883 function together with `set-current-module', `current-module', etc
1884 when you want to have more control over the state that is carried from
1885 one eval to the next.
1887 Additionally, it has been made sure that forms that are evaluated at
1888 the top level are always evaluated with respect to the current module.
1889 Previously, subforms of top-level forms such as `begin', `case',
1890 etc. did not respect changes to the current module although these
1891 subforms are at the top-level as well.
1893 To prevent strange behavior, the forms `define-module',
1894 `use-modules', `use-syntax', and `export' have been restricted to only
1895 work on the top level. The forms `define-public' and
1896 `defmacro-public' only export the new binding on the top level. They
1897 behave just like `define' and `defmacro', respectively, when they are
1898 used in a lexical environment.
1900 Also, `export' will no longer silently re-export bindings imported
1901 from a used module. It will emit a `deprecation' warning and will
1902 cease to perform any re-export in the next version. If you actually
1903 want to re-export bindings, use the new `re-export' in place of
1904 `export'. The new `re-export' will not make copies of variables when
1905 rexporting them, as `export' did wrongly.
1907 ** Module system now allows selection and renaming of imported bindings
1909 Previously, when using `use-modules' or the `#:use-module' clause in
1910 the `define-module' form, all the bindings (association of symbols to
1911 values) for imported modules were added to the "current module" on an
1912 as-is basis. This has been changed to allow finer control through two
1913 new facilities: selection and renaming.
1915 You can now select which of the imported module's bindings are to be
1916 visible in the current module by using the `:select' clause. This
1917 clause also can be used to rename individual bindings. For example:
1919 ;; import all bindings no questions asked
1920 (use-modules (ice-9 common-list))
1922 ;; import four bindings, renaming two of them;
1923 ;; the current module sees: every some zonk-y zonk-n
1924 (use-modules ((ice-9 common-list)
1926 (remove-if . zonk-y)
1927 (remove-if-not . zonk-n))))
1929 You can also programmatically rename all selected bindings using the
1930 `:renamer' clause, which specifies a proc that takes a symbol and
1931 returns another symbol. Because it is common practice to use a prefix,
1932 we now provide the convenience procedure `symbol-prefix-proc'. For
1935 ;; import four bindings, renaming two of them specifically,
1936 ;; and all four w/ prefix "CL:";
1937 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1938 (use-modules ((ice-9 common-list)
1940 (remove-if . zonk-y)
1941 (remove-if-not . zonk-n))
1942 :renamer (symbol-prefix-proc 'CL:)))
1944 ;; import four bindings, renaming two of them specifically,
1945 ;; and all four by upcasing.
1946 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1947 (define (upcase-symbol sym)
1948 (string->symbol (string-upcase (symbol->string sym))))
1950 (use-modules ((ice-9 common-list)
1952 (remove-if . zonk-y)
1953 (remove-if-not . zonk-n))
1954 :renamer upcase-symbol))
1956 Note that programmatic renaming is done *after* individual renaming.
1957 Also, the above examples show `use-modules', but the same facilities are
1958 available for the `#:use-module' clause of `define-module'.
1960 See manual for more info.
1962 ** The semantics of guardians have changed.
1964 The changes are for the most part compatible. An important criterion
1965 was to keep the typical usage of guardians as simple as before, but to
1966 make the semantics safer and (as a result) more useful.
1968 *** All objects returned from guardians are now properly alive.
1970 It is now guaranteed that any object referenced by an object returned
1971 from a guardian is alive. It's now impossible for a guardian to
1972 return a "contained" object before its "containing" object.
1974 One incompatible (but probably not very important) change resulting
1975 from this is that it is no longer possible to guard objects that
1976 indirectly reference themselves (i.e. are parts of cycles). If you do
1977 so accidentally, you'll get a warning.
1979 *** There are now two types of guardians: greedy and sharing.
1981 If you call (make-guardian #t) or just (make-guardian), you'll get a
1982 greedy guardian, and for (make-guardian #f) a sharing guardian.
1984 Greedy guardians are the default because they are more "defensive".
1985 You can only greedily guard an object once. If you guard an object
1986 more than once, once in a greedy guardian and the rest of times in
1987 sharing guardians, then it is guaranteed that the object won't be
1988 returned from sharing guardians as long as it is greedily guarded
1991 Guardians returned by calls to `make-guardian' can now take one more
1992 optional parameter, which says whether to throw an error in case an
1993 attempt is made to greedily guard an object that is already greedily
1994 guarded. The default is true, i.e. throw an error. If the parameter
1995 is false, the guardian invocation returns #t if guarding was
1996 successful and #f if it wasn't.
1998 Also, since greedy guarding is, in effect, a side-effecting operation
1999 on objects, a new function is introduced: `destroy-guardian!'.
2000 Invoking this function on a guardian renders it unoperative and, if
2001 the guardian is greedy, clears the "greedily guarded" property of the
2002 objects that were guarded by it, thus undoing the side effect.
2004 Note that all this hair is hardly very important, since guardian
2005 objects are usually permanent.
2007 ** Continuations created by call-with-current-continuation now accept
2008 any number of arguments, as required by R5RS.
2010 ** New function `issue-deprecation-warning'
2012 This function is used to display the deprecation messages that are
2013 controlled by GUILE_WARN_DEPRECATION as explained in the README.
2016 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
2020 ;; `id' is deprecated. Use `identity' instead.
2025 ** New syntax `begin-deprecated'
2027 When deprecated features are included (as determined by the configure
2028 option --enable-deprecated), `begin-deprecated' is identical to
2029 `begin'. When deprecated features are excluded, it always evaluates
2030 to `#f', ignoring the body forms.
2032 ** New function `make-object-property'
2034 This function returns a new `procedure with setter' P that can be used
2035 to attach a property to objects. When calling P as
2039 where `obj' is any kind of object, it attaches `val' to `obj' in such
2040 a way that it can be retrieved by calling P as
2044 This function will replace procedure properties, symbol properties and
2045 source properties eventually.
2047 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
2049 Instead of #&optional, #&key, etc you should now use #:optional,
2050 #:key, etc. Since #:optional is a keyword, you can write it as just
2051 :optional when (read-set! keywords 'prefix) is active.
2053 The old reader syntax `#&' is still supported, but deprecated. It
2054 will be removed in the next release.
2056 ** New define-module option: pure
2058 Tells the module system not to include any bindings from the root
2063 (define-module (totally-empty-module)
2066 ** New define-module option: export NAME1 ...
2068 Export names NAME1 ...
2070 This option is required if you want to be able to export bindings from
2071 a module which doesn't import one of `define-public' or `export'.
2075 (define-module (foo)
2077 :use-module (ice-9 r5rs)
2080 ;;; Note that we're pure R5RS below this point!
2085 ** New function: object->string OBJ
2087 Return a Scheme string obtained by printing a given object.
2089 ** New function: port? X
2091 Returns a boolean indicating whether X is a port. Equivalent to
2092 `(or (input-port? X) (output-port? X))'.
2094 ** New function: file-port?
2096 Determines whether a given object is a port that is related to a file.
2098 ** New function: port-for-each proc
2100 Apply PROC to each port in the Guile port table in turn. The return
2101 value is unspecified. More specifically, PROC is applied exactly once
2102 to every port that exists in the system at the time PORT-FOR-EACH is
2103 invoked. Changes to the port table while PORT-FOR-EACH is running
2104 have no effect as far as PORT-FOR-EACH is concerned.
2106 ** New function: dup2 oldfd newfd
2108 A simple wrapper for the `dup2' system call. Copies the file
2109 descriptor OLDFD to descriptor number NEWFD, replacing the
2110 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
2111 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
2112 to move away ports which are using NEWFD. The return value is
2115 ** New function: close-fdes fd
2117 A simple wrapper for the `close' system call. Close file
2118 descriptor FD, which must be an integer. Unlike close (*note
2119 close: Ports and File Descriptors.), the file descriptor will be
2120 closed even if a port is using it. The return value is
2123 ** New function: crypt password salt
2125 Encrypts `password' using the standard unix password encryption
2128 ** New function: chroot path
2130 Change the root directory of the running process to `path'.
2132 ** New functions: getlogin, cuserid
2134 Return the login name or the user name of the current effective user
2137 ** New functions: getpriority which who, setpriority which who prio
2139 Get or set the priority of the running process.
2141 ** New function: getpass prompt
2143 Read a password from the terminal, first displaying `prompt' and
2146 ** New function: flock file operation
2148 Set/remove an advisory shared or exclusive lock on `file'.
2150 ** New functions: sethostname name, gethostname
2152 Set or get the hostname of the machine the current process is running
2155 ** New function: mkstemp! tmpl
2157 mkstemp creates a new unique file in the file system and returns a
2158 new buffered port open for reading and writing to the file. TMPL
2159 is a string specifying where the file should be created: it must
2160 end with `XXXXXX' and will be changed in place to return the name
2161 of the temporary file.
2163 ** New function: open-input-string string
2165 Return an input string port which delivers the characters from
2166 `string'. This procedure, together with `open-output-string' and
2167 `get-output-string' implements SRFI-6.
2169 ** New function: open-output-string
2171 Return an output string port which collects all data written to it.
2172 The data can then be retrieved by `get-output-string'.
2174 ** New function: get-output-string
2176 Return the contents of an output string port.
2178 ** New function: identity
2180 Return the argument.
2182 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
2183 are represented in Scheme as integers with normal host byte ordering.
2185 ** New function: inet-pton family address
2187 Convert a printable string network address into an integer. Note that
2188 unlike the C version of this function, the result is an integer with
2189 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2192 (inet-pton AF_INET "127.0.0.1") => 2130706433
2193 (inet-pton AF_INET6 "::1") => 1
2195 ** New function: inet-ntop family address
2197 Convert an integer network address into a printable string. Note that
2198 unlike the C version of this function, the input is an integer with
2199 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
2202 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
2203 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
2204 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
2208 Use `identity' instead.
2214 ** Deprecated: return-it
2218 ** Deprecated: string-character-length
2220 Use `string-length' instead.
2222 ** Deprecated: flags
2224 Use `logior' instead.
2226 ** Deprecated: close-all-ports-except.
2228 This was intended for closing ports in a child process after a fork,
2229 but it has the undesirable side effect of flushing buffers.
2230 port-for-each is more flexible.
2232 ** The (ice-9 popen) module now attempts to set up file descriptors in
2233 the child process from the current Scheme ports, instead of using the
2234 current values of file descriptors 0, 1, and 2 in the parent process.
2236 ** Removed function: builtin-weak-bindings
2238 There is no such concept as a weak binding any more.
2240 ** Removed constants: bignum-radix, scm-line-incrementors
2242 ** define-method: New syntax mandatory.
2244 The new method syntax is now mandatory:
2246 (define-method (NAME ARG-SPEC ...) BODY ...)
2247 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
2249 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
2250 REST-ARG ::= ARG-NAME
2252 If you have old code using the old syntax, import
2253 (oop goops old-define-method) before (oop goops) as in:
2255 (use-modules (oop goops old-define-method) (oop goops))
2257 ** Deprecated function: builtin-variable
2258 Removed function: builtin-bindings
2260 There is no longer a distinction between builtin or other variables.
2261 Use module system operations for all variables.
2263 ** Lazy-catch handlers are no longer allowed to return.
2265 That is, a call to `throw', `error', etc is now guaranteed to not
2268 ** Bugfixes for (ice-9 getopt-long)
2270 This module is now tested using test-suite/tests/getopt-long.test.
2271 The following bugs have been fixed:
2273 *** Parsing for options that are specified to have `optional' args now checks
2274 if the next element is an option instead of unconditionally taking it as the
2277 *** An error is now thrown for `--opt=val' when the option description
2278 does not specify `(value #t)' or `(value optional)'. This condition used to
2279 be accepted w/o error, contrary to the documentation.
2281 *** The error message for unrecognized options is now more informative.
2282 It used to be "not a record", an artifact of the implementation.
2284 *** The error message for `--opt' terminating the arg list (no value), when
2285 `(value #t)' is specified, is now more informative. It used to be "not enough
2288 *** "Clumped" single-char args now preserve trailing string, use it as arg.
2289 The expansion used to be like so:
2291 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
2293 Note that the "5d" is dropped. Now it is like so:
2295 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
2297 This enables single-char options to have adjoining arguments as long as their
2298 constituent characters are not potential single-char options.
2300 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
2302 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
2303 property `arglist', which can be retrieved by `arity'. The result is that
2304 `arity' can give more detailed information than before:
2308 guile> (use-modules (ice-9 optargs))
2309 guile> (define* (foo #:optional a b c) a)
2311 0 or more arguments in `lambda*:G0'.
2316 3 optional arguments: `a', `b' and `c'.
2317 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
2319 2 required arguments: `a' and `b', 2 keyword arguments: `c'
2320 and `d', other keywords allowed.
2321 guile> (define* (baz a b #:optional c #:rest r) a)
2323 2 required arguments: `a' and `b', 1 optional argument: `c',
2326 * Changes to the C interface
2328 ** Types have been renamed from scm_*_t to scm_t_*.
2330 This has been done for POSIX sake. It reserves identifiers ending
2331 with "_t". What a concept.
2333 The old names are still available with status `deprecated'.
2335 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
2337 ** Deprecated features have been removed.
2341 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
2342 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
2344 *** C Functions removed
2346 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
2347 scm_fseek - replaced by scm_seek.
2348 gc-thunk - replaced by after-gc-hook.
2349 gh_int2scmb - replaced by gh_bool2scm.
2350 scm_tc_dblr - replaced by scm_tc16_real.
2351 scm_tc_dblc - replaced by scm_tc16_complex.
2352 scm_list_star - replaced by scm_cons_star.
2354 ** Deprecated: scm_makfromstr
2356 Use scm_mem2string instead.
2358 ** Deprecated: scm_make_shared_substring
2360 Explicit shared substrings will disappear from Guile.
2362 Instead, "normal" strings will be implemented using sharing
2363 internally, combined with a copy-on-write strategy.
2365 ** Deprecated: scm_read_only_string_p
2367 The concept of read-only strings will disappear in next release of
2370 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
2372 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
2374 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
2376 Call a procedure with the indicated number of arguments. See "Fly
2377 Evaluation" in the manual.
2379 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
2381 Call a procedure with the indicated number of arguments and a list of
2382 further arguments. See "Fly Evaluation" in the manual.
2384 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
2386 Create a list of the given number of elements. See "List
2387 Constructors" in the manual.
2389 ** Renamed function: scm_listify has been replaced by scm_list_n.
2391 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
2392 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
2394 Use functions scm_list_N instead.
2396 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
2398 Used by an application to read arbitrary number of bytes from a port.
2399 Same semantics as libc read, except that scm_c_read only returns less
2400 than SIZE bytes if at end-of-file.
2402 Warning: Doesn't update port line and column counts!
2404 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
2406 Used by an application to write arbitrary number of bytes to an SCM
2407 port. Similar semantics as libc write. However, unlike libc
2408 write, scm_c_write writes the requested number of bytes and has no
2411 Warning: Doesn't update port line and column counts!
2413 ** New function: scm_init_guile ()
2415 In contrast to scm_boot_guile, scm_init_guile will return normally
2416 after initializing Guile. It is not available on all systems, tho.
2418 ** New functions: scm_str2symbol, scm_mem2symbol
2420 The function scm_str2symbol takes a const char* pointing to a zero-terminated
2421 field of characters and creates a scheme symbol object from that C string.
2422 The function scm_mem2symbol takes a const char* and a number of characters and
2423 creates a symbol from the characters in that memory area.
2425 ** New functions: scm_primitive_make_property
2426 scm_primitive_property_ref
2427 scm_primitive_property_set_x
2428 scm_primitive_property_del_x
2430 These functions implement a new way to deal with object properties.
2431 See libguile/properties.c for their documentation.
2433 ** New function: scm_done_free (long size)
2435 This function is the inverse of scm_done_malloc. Use it to report the
2436 amount of smob memory you free. The previous method, which involved
2437 calling scm_done_malloc with negative argument, was somewhat
2438 unintuitive (and is still available, of course).
2440 ** New function: scm_c_memq (SCM obj, SCM list)
2442 This function provides a fast C level alternative for scm_memq for the case
2443 that the list parameter is known to be a proper list. The function is a
2444 replacement for scm_sloppy_memq, but is stricter in its requirements on its
2445 list input parameter, since for anything else but a proper list the function's
2446 behaviour is undefined - it may even crash or loop endlessly. Further, for
2447 the case that the object is not found in the list, scm_c_memq returns #f which
2448 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
2450 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
2451 scm_remember_upto_here
2453 These functions replace the function scm_remember.
2455 ** Deprecated function: scm_remember
2457 Use one of the new functions scm_remember_upto_here_1,
2458 scm_remember_upto_here_2 or scm_remember_upto_here instead.
2460 ** New function: scm_allocate_string
2462 This function replaces the function scm_makstr.
2464 ** Deprecated function: scm_makstr
2466 Use the new function scm_allocate_string instead.
2468 ** New global variable scm_gc_running_p introduced.
2470 Use this variable to find out if garbage collection is being executed. Up to
2471 now applications have used scm_gc_heap_lock to test if garbage collection was
2472 running, which also works because of the fact that up to know only the garbage
2473 collector has set this variable. But, this is an implementation detail that
2474 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
2475 of this variable is (and has been) not fully safe anyway.
2477 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
2479 Use these instead of SCM_LENGTH_MAX.
2481 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
2482 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
2483 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
2485 Use these instead of SCM_LENGTH.
2487 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
2488 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
2489 SCM_SET_BITVECTOR_LENGTH
2491 Use these instead of SCM_SETLENGTH
2493 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
2494 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
2497 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
2500 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
2501 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
2504 Use these instead of SCM_SETCHARS.
2506 ** New macro: SCM_BITVECTOR_P
2508 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
2510 Use instead of SCM_COERCE_SUBSTR.
2512 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
2514 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
2516 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
2517 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
2518 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
2519 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
2520 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
2521 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
2522 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
2523 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
2524 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
2525 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
2526 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
2527 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
2528 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
2529 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
2530 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
2532 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
2533 Use scm_memory_error instead of SCM_NALLOC.
2534 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
2535 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
2536 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
2537 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
2538 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
2539 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
2540 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
2541 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
2542 Use SCM_STRINGP instead of SCM_RWSTRINGP.
2543 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
2544 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
2545 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
2546 Use a type specific setter macro instead of SCM_SETLENGTH.
2547 Use a type specific setter macro instead of SCM_SETCHARS.
2548 Use a type specific length macro instead of SCM_LENGTH_MAX.
2549 Use SCM_GCMARKP instead of SCM_GC8MARKP.
2550 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
2551 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
2552 Use SCM_TYP16 instead of SCM_GCTYP16.
2553 Use SCM_CDR instead of SCM_GCCDR.
2554 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
2555 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
2556 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
2557 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
2558 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
2559 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
2560 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
2562 ** Removed function: scm_struct_init
2564 ** Removed variable: scm_symhash_dim
2566 ** Renamed function: scm_make_cont has been replaced by
2567 scm_make_continuation, which has a different interface.
2569 ** Deprecated function: scm_call_catching_errors
2571 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
2573 ** Deprecated function: scm_strhash
2575 Use scm_string_hash instead.
2577 ** Deprecated function: scm_vector_set_length_x
2579 Instead, create a fresh vector of the desired size and copy the contents.
2581 ** scm_gensym has changed prototype
2583 scm_gensym now only takes one argument.
2585 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
2588 There is now only a single symbol type scm_tc7_symbol.
2589 The tag scm_tc7_lvector was not used anyway.
2591 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
2593 Use scm_make_smob_type and scm_set_smob_XXX instead.
2595 ** New function scm_set_smob_apply.
2597 This can be used to set an apply function to a smob type.
2599 ** Deprecated function: scm_strprint_obj
2601 Use scm_object_to_string instead.
2603 ** Deprecated function: scm_wta
2605 Use scm_wrong_type_arg, or another appropriate error signalling function
2608 ** Explicit support for obarrays has been deprecated.
2610 Use `scm_str2symbol' and the generic hashtable functions instead.
2612 ** The concept of `vcells' has been deprecated.
2614 The data type `variable' is now used exclusively. `Vcells' have been
2615 a low-level concept so you are likely not affected by this change.
2617 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
2618 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
2620 Use scm_c_define or scm_c_lookup instead, as appropriate.
2622 *** New functions: scm_c_module_lookup, scm_c_lookup,
2623 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
2624 scm_module_define, scm_define.
2626 These functions work with variables instead of with vcells.
2628 ** New functions for creating and defining `subr's and `gsubr's.
2630 The new functions more clearly distinguish between creating a subr (or
2631 gsubr) object and adding it to the current module.
2633 These new functions are available: scm_c_make_subr, scm_c_define_subr,
2634 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
2635 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
2636 scm_c_define_gsubr_with_generic.
2638 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
2639 scm_make_subr_with_generic, scm_make_gsubr,
2640 scm_make_gsubr_with_generic.
2642 Use the new ones from above instead.
2644 ** C interface to the module system has changed.
2646 While we suggest that you avoid as many explicit module system
2647 operations from C as possible for the time being, the C interface has
2648 been made more similar to the high-level Scheme module system.
2650 *** New functions: scm_c_define_module, scm_c_use_module,
2651 scm_c_export, scm_c_resolve_module.
2653 They mostly work like their Scheme namesakes. scm_c_define_module
2654 takes a function that is called a context where the new module is
2657 *** Deprecated functions: scm_the_root_module, scm_make_module,
2658 scm_ensure_user_module, scm_load_scheme_module.
2660 Use the new functions instead.
2662 ** Renamed function: scm_internal_with_fluids becomes
2665 scm_internal_with_fluids is available as a deprecated function.
2667 ** New function: scm_c_with_fluid.
2669 Just like scm_c_with_fluids, but takes one fluid and one value instead
2672 ** Deprecated typedefs: long_long, ulong_long.
2674 They are of questionable utility and they pollute the global
2677 ** Deprecated typedef: scm_sizet
2679 It is of questionable utility now that Guile requires ANSI C, and is
2682 ** Deprecated typedefs: scm_port_rw_active, scm_port,
2683 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
2684 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
2686 Made more compliant with the naming policy by adding a _t at the end.
2688 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
2689 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
2691 With the exception of the mysterious scm_2ulong2big, they are still
2692 available under new names (scm_i_mkbig etc). These functions are not
2693 intended to be used in user code. You should avoid dealing with
2694 bignums directly, and should deal with numbers in general (which can
2697 ** Change in behavior: scm_num2long, scm_num2ulong
2699 The scm_num2[u]long functions don't any longer accept an inexact
2700 argument. This change in behavior is motivated by concordance with
2701 R5RS: It is more common that a primitive doesn't want to accept an
2702 inexact for an exact.
2704 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
2705 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
2706 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
2709 These are conversion functions between the various ANSI C integral
2710 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
2711 accept an inexact argument.
2713 ** New functions: scm_float2num, scm_double2num,
2714 scm_num2float, scm_num2double.
2716 These are conversion functions between the two ANSI C float types and
2719 ** New number validation macros:
2720 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
2724 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
2726 These are just nicer-named old scm_protect_object and
2727 scm_unprotect_object.
2729 ** Deprecated functions: scm_protect_object, scm_unprotect_object
2731 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
2733 These functions can be used to register pointers to locations that
2736 ** Deprecated function: scm_create_hook.
2738 Its sins are: misleading name, non-modularity and lack of general
2742 Changes since Guile 1.3.4:
2744 * Changes to the distribution
2746 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
2748 We've changed the way we handle generated files in the Guile source
2749 repository. As a result, the procedure for building trees obtained
2750 from the nightly FTP snapshots or via CVS has changed:
2751 - You must have appropriate versions of autoconf, automake, and
2752 libtool installed on your system. See README for info on how to
2753 obtain these programs.
2754 - Before configuring the tree, you must first run the script
2755 `autogen.sh' at the top of the source tree.
2757 The Guile repository used to contain not only source files, written by
2758 humans, but also some generated files, like configure scripts and
2759 Makefile.in files. Even though the contents of these files could be
2760 derived mechanically from other files present, we thought it would
2761 make the tree easier to build if we checked them into CVS.
2763 However, this approach means that minor differences between
2764 developer's installed tools and habits affected the whole team.
2765 So we have removed the generated files from the repository, and
2766 added the autogen.sh script, which will reconstruct them
2770 ** configure now has experimental options to remove support for certain
2773 --disable-arrays omit array and uniform array support
2774 --disable-posix omit posix interfaces
2775 --disable-networking omit networking interfaces
2776 --disable-regex omit regular expression interfaces
2778 These are likely to become separate modules some day.
2780 ** New configure option --enable-debug-freelist
2782 This enables a debugging version of SCM_NEWCELL(), and also registers
2783 an extra primitive, the setter `gc-set-debug-check-freelist!'.
2785 Configure with the --enable-debug-freelist option to enable
2786 the gc-set-debug-check-freelist! primitive, and then use:
2788 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
2789 (gc-set-debug-check-freelist! #f) # turn off checking
2791 Checking of the freelist forces a traversal of the freelist and
2792 a garbage collection before each allocation of a cell. This can
2793 slow down the interpreter dramatically, so the setter should be used to
2794 turn on this extra processing only when necessary.
2796 ** New configure option --enable-debug-malloc
2798 Include code for debugging of calls to scm_must_malloc/realloc/free.
2802 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
2803 2. objects reallocated by scm_must_realloc has been allocated by
2805 3. reallocated objects are reallocated with the same what string
2807 But, most importantly, it records the number of allocated objects of
2808 each kind. This is useful when searching for memory leaks.
2810 A Guile compiled with this option provides the primitive
2811 `malloc-stats' which returns an alist with pairs of kind and the
2812 number of objects of that kind.
2814 ** All includes are now referenced relative to the root directory
2816 Since some users have had problems with mixups between Guile and
2817 system headers, we have decided to always refer to Guile headers via
2818 their parent directories. This essentially creates a "private name
2819 space" for Guile headers. This means that the compiler only is given
2820 -I options for the root build and root source directory.
2822 ** Header files kw.h and genio.h have been removed.
2824 ** The module (ice-9 getopt-gnu-style) has been removed.
2826 ** New module (ice-9 documentation)
2828 Implements the interface to documentation strings associated with
2831 ** New module (ice-9 time)
2833 Provides a macro `time', which displays execution time of a given form.
2835 ** New module (ice-9 history)
2837 Loading this module enables value history in the repl.
2839 * Changes to the stand-alone interpreter
2841 ** New command line option --debug
2843 Start Guile with debugging evaluator and backtraces enabled.
2845 This is useful when debugging your .guile init file or scripts.
2847 ** New help facility
2849 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
2850 (help REGEXP) ditto for objects with names matching REGEXP (a string)
2851 (help 'NAME) gives documentation for NAME, even if it is not an object
2852 (help ,EXPR) gives documentation for object returned by EXPR
2853 (help (my module)) gives module commentary for `(my module)'
2854 (help) gives this text
2856 `help' searches among bindings exported from loaded modules, while
2857 `apropos' searches among bindings visible from the "current" module.
2859 Examples: (help help)
2861 (help "output-string")
2863 ** `help' and `apropos' now prints full module names
2865 ** Dynamic linking now uses libltdl from the libtool package.
2867 The old system dependent code for doing dynamic linking has been
2868 replaced with calls to the libltdl functions which do all the hairy
2871 The major improvement is that you can now directly pass libtool
2872 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
2873 will be able to do the best shared library job you can get, via
2876 The way dynamic libraries are found has changed and is not really
2877 portable across platforms, probably. It is therefore recommended to
2878 use absolute filenames when possible.
2880 If you pass a filename without an extension to `dynamic-link', it will
2881 try a few appropriate ones. Thus, the most platform ignorant way is
2882 to specify a name like "libfoo", without any directories and
2885 ** Guile COOP threads are now compatible with LinuxThreads
2887 Previously, COOP threading wasn't possible in applications linked with
2888 Linux POSIX threads due to their use of the stack pointer to find the
2889 thread context. This has now been fixed with a workaround which uses
2890 the pthreads to allocate the stack.
2892 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2894 ** Positions of erring expression in scripts
2896 With version 1.3.4, the location of the erring expression in Guile
2897 scipts is no longer automatically reported. (This should have been
2898 documented before the 1.3.4 release.)
2900 You can get this information by enabling recording of positions of
2901 source expressions and running the debugging evaluator. Put this at
2902 the top of your script (or in your "site" file):
2904 (read-enable 'positions)
2905 (debug-enable 'debug)
2907 ** Backtraces in scripts
2909 It is now possible to get backtraces in scripts.
2913 (debug-enable 'debug 'backtrace)
2915 at the top of the script.
2917 (The first options enables the debugging evaluator.
2918 The second enables backtraces.)
2920 ** Part of module system symbol lookup now implemented in C
2922 The eval closure of most modules is now implemented in C. Since this
2923 was one of the bottlenecks for loading speed, Guile now loads code
2924 substantially faster than before.
2926 ** Attempting to get the value of an unbound variable now produces
2927 an exception with a key of 'unbound-variable instead of 'misc-error.
2929 ** The initial default output port is now unbuffered if it's using a
2930 tty device. Previously in this situation it was line-buffered.
2932 ** New hook: after-gc-hook
2934 after-gc-hook takes over the role of gc-thunk. This hook is run at
2935 the first SCM_TICK after a GC. (Thus, the code is run at the same
2936 point during evaluation as signal handlers.)
2938 Note that this hook should be used only for diagnostic and debugging
2939 purposes. It is not certain that it will continue to be well-defined
2940 when this hook is run in the future.
2942 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2943 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2945 ** Improvements to garbage collector
2947 Guile 1.4 has a new policy for triggering heap allocation and
2948 determining the sizes of heap segments. It fixes a number of problems
2951 1. The new policy can handle two separate pools of cells
2952 (2-word/4-word) better. (The old policy would run wild, allocating
2953 more and more memory for certain programs.)
2955 2. The old code would sometimes allocate far too much heap so that the
2956 Guile process became gigantic. The new code avoids this.
2958 3. The old code would sometimes allocate too little so that few cells
2959 were freed at GC so that, in turn, too much time was spent in GC.
2961 4. The old code would often trigger heap allocation several times in a
2962 row. (The new scheme predicts how large the segments needs to be
2963 in order not to need further allocation.)
2965 All in all, the new GC policy will make larger applications more
2968 The new GC scheme also is prepared for POSIX threading. Threads can
2969 allocate private pools of cells ("clusters") with just a single
2970 function call. Allocation of single cells from such a cluster can
2971 then proceed without any need of inter-thread synchronization.
2973 ** New environment variables controlling GC parameters
2975 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2978 Allocation of 2-word cell heaps:
2980 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2983 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2984 GC in percent of total heap size
2987 Allocation of 4-word cell heaps
2988 (used for real numbers and misc other objects):
2990 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2992 (See entry "Way for application to customize GC parameters" under
2993 section "Changes to the scm_ interface" below.)
2995 ** Guile now implements reals using 4-word cells
2997 This speeds up computation with reals. (They were earlier allocated
2998 with `malloc'.) There is still some room for optimizations, however.
3000 ** Some further steps toward POSIX thread support have been taken
3002 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
3003 don't have much effect any longer, and many of them will be removed in
3007 are only handled at the top of the evaluator loop, immediately after
3008 I/O, and in scm_equalp.
3010 *** The GC can allocate thread private pools of pairs.
3012 * Changes to Scheme functions and syntax
3014 ** close-input-port and close-output-port are now R5RS
3016 These procedures have been turned into primitives and have R5RS behaviour.
3018 ** New procedure: simple-format PORT MESSAGE ARG1 ...
3020 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
3021 extended by the more sophisticated version in (ice-9 format)
3023 (simple-format port message . args)
3024 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
3025 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
3026 the escapes are replaced with corresponding members of ARGS:
3027 ~A formats using `display' and ~S formats using `write'.
3028 If DESTINATION is #t, then use the `current-output-port',
3029 if DESTINATION is #f, then return a string containing the formatted text.
3030 Does not add a trailing newline."
3032 ** string-ref: the second argument is no longer optional.
3034 ** string, list->string: no longer accept strings in their arguments,
3035 only characters, for compatibility with R5RS.
3037 ** New procedure: port-closed? PORT
3038 Returns #t if PORT is closed or #f if it is open.
3040 ** Deprecated: list*
3042 The list* functionality is now provided by cons* (SRFI-1 compliant)
3044 ** New procedure: cons* ARG1 ARG2 ... ARGn
3046 Like `list', but the last arg provides the tail of the constructed list,
3047 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
3049 Requires at least one argument. If given one argument, that argument
3050 is returned as result.
3052 This function is called `list*' in some other Schemes and in Common LISP.
3054 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
3056 ** New procedure: object-documentation OBJECT
3058 Returns the documentation string associated with OBJECT. The
3059 procedure uses a caching mechanism so that subsequent lookups are
3062 Exported by (ice-9 documentation).
3064 ** module-name now returns full names of modules
3066 Previously, only the last part of the name was returned (`session' for
3067 `(ice-9 session)'). Ex: `(ice-9 session)'.
3069 * Changes to the gh_ interface
3071 ** Deprecated: gh_int2scmb
3073 Use gh_bool2scm instead.
3075 * Changes to the scm_ interface
3077 ** Guile primitives now carry docstrings!
3079 Thanks to Greg Badros!
3081 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3083 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
3084 macros and must contain a docstring that is extracted into foo.doc using a new
3085 guile-doc-snarf script (that uses guile-doc-snarf.awk).
3087 However, a major overhaul of these macros is scheduled for the next release of
3090 ** Guile primitives use a new technique for validation of arguments
3092 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
3093 the readability of argument checking.
3095 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
3097 ** New macros: SCM_PACK, SCM_UNPACK
3099 Compose/decompose an SCM value.
3101 The SCM type is now treated as an abstract data type and may be defined as a
3102 long, a void* or as a struct, depending on the architecture and compile time
3103 options. This makes it easier to find several types of bugs, for example when
3104 SCM values are treated as integers without conversion. Values of the SCM type
3105 should be treated as "atomic" values. These macros are used when
3106 composing/decomposing an SCM value, either because you want to access
3107 individual bits, or because you want to treat it as an integer value.
3109 E.g., in order to set bit 7 in an SCM value x, use the expression
3111 SCM_PACK (SCM_UNPACK (x) | 0x80)
3113 ** The name property of hooks is deprecated.
3114 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
3116 You can emulate this feature by using object properties.
3118 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
3119 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
3122 These macros will be removed in a future release of Guile.
3124 ** The following types, functions and macros from numbers.h are deprecated:
3125 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
3126 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
3128 ** Port internals: the rw_random variable in the scm_port structure
3129 must be set to non-zero in any random access port. In recent Guile
3130 releases it was only set for bidirectional random-access ports.
3132 ** Port internals: the seek ptob procedure is now responsible for
3133 resetting the buffers if required. The change was made so that in the
3134 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
3135 the fport and strport ptobs can avoid resetting the buffers,
3136 in particular to avoid discarding unread chars. An existing port
3137 type can be fixed by adding something like the following to the
3138 beginning of the ptob seek procedure:
3140 if (pt->rw_active == SCM_PORT_READ)
3141 scm_end_input (object);
3142 else if (pt->rw_active == SCM_PORT_WRITE)
3143 ptob->flush (object);
3145 although to actually avoid resetting the buffers and discard unread
3146 chars requires further hacking that depends on the characteristics
3149 ** Deprecated functions: scm_fseek, scm_tag
3151 These functions are no longer used and will be removed in a future version.
3153 ** The scm_sysmissing procedure is no longer used in libguile.
3154 Unless it turns out to be unexpectedly useful to somebody, it will be
3155 removed in a future version.
3157 ** The format of error message strings has changed
3159 The two C procedures: scm_display_error and scm_error, as well as the
3160 primitive `scm-error', now use scm_simple_format to do their work.
3161 This means that the message strings of all code must be updated to use
3162 ~A where %s was used before, and ~S where %S was used before.
3164 During the period when there still are a lot of old Guiles out there,
3165 you might want to support both old and new versions of Guile.
3167 There are basically two methods to achieve this. Both methods use
3170 AC_CHECK_FUNCS(scm_simple_format)
3172 in your configure.in.
3174 Method 1: Use the string concatenation features of ANSI C's
3179 #ifdef HAVE_SCM_SIMPLE_FORMAT
3185 Then represent each of your error messages using a preprocessor macro:
3187 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
3191 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
3192 (define make-message string-append)
3194 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
3196 Method 2: Use the oldfmt function found in doc/oldfmt.c.
3200 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
3205 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
3209 ** Deprecated: coop_mutex_init, coop_condition_variable_init
3211 Don't use the functions coop_mutex_init and
3212 coop_condition_variable_init. They will change.
3214 Use scm_mutex_init and scm_cond_init instead.
3216 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
3217 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
3218 COND, as `scm_cond_wait' does, but it also bounds the duration
3219 of the wait. If COND has not been signaled before time ABSTIME,
3220 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
3221 returns the error code `ETIMEDOUT'.
3223 The ABSTIME parameter specifies an absolute time, with the same
3224 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
3225 to 00:00:00 GMT, January 1, 1970.
3227 ** New function: scm_cond_broadcast (scm_cond_t *COND)
3228 `scm_cond_broadcast' restarts all the threads that are waiting
3229 on the condition variable COND. Nothing happens if no threads are
3232 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
3233 `scm_key_create' allocates a new TSD key. The key is stored in
3234 the location pointed to by KEY. There is no limit on the number
3235 of keys allocated at a given time. The value initially associated
3236 with the returned key is `NULL' in all currently executing threads.
3238 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
3239 function associated with the key. When a thread terminates,
3240 DESTR_FUNCTION is called on the value associated with the key in
3241 that thread. The DESTR_FUNCTION is not called if a key is deleted
3242 with `scm_key_delete' or a value is changed with
3243 `scm_setspecific'. The order in which destructor functions are
3244 called at thread termination time is unspecified.
3246 Destructors are not yet implemented.
3248 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
3249 `scm_setspecific' changes the value associated with KEY in the
3250 calling thread, storing the given POINTER instead.
3252 ** New function: scm_getspecific (scm_key_t KEY)
3253 `scm_getspecific' returns the value currently associated with
3254 KEY in the calling thread.
3256 ** New function: scm_key_delete (scm_key_t KEY)
3257 `scm_key_delete' deallocates a TSD key. It does not check
3258 whether non-`NULL' values are associated with that key in the
3259 currently executing threads, nor call the destructor function
3260 associated with the key.
3262 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
3264 Initialize a C level hook HOOK with associated HOOK_DATA and type
3265 TYPE. (See scm_c_hook_run ().)
3267 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
3269 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
3270 is true, add it last, otherwise first. The same FUNC can be added
3271 multiple times if FUNC_DATA differ and vice versa.
3273 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
3275 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
3276 function is only removed if both FUNC and FUNC_DATA matches.
3278 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
3280 Run hook HOOK passing DATA to the hook functions.
3282 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
3283 returned is undefined.
3285 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
3286 returns a non-NULL value. This value is returned as the result of
3287 scm_c_hook_run. If all functions return NULL, NULL is returned.
3289 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
3290 returns a NULL value, and NULL is returned. If all functions returns
3291 a non-NULL value, the last value is returned.
3293 ** New C level GC hooks
3295 Five new C level hooks has been added to the garbage collector.
3297 scm_before_gc_c_hook
3300 are run before locking and after unlocking the heap. The system is
3301 thus in a mode where evaluation can take place. (Except that
3302 scm_before_gc_c_hook must not allocate new cells.)
3304 scm_before_mark_c_hook
3305 scm_before_sweep_c_hook
3306 scm_after_sweep_c_hook
3308 are run when the heap is locked. These are intended for extension of
3309 the GC in a modular fashion. Examples are the weaks and guardians
3312 ** Way for application to customize GC parameters
3314 The application can set up other default values for the GC heap
3315 allocation parameters
3317 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
3318 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
3319 GUILE_MAX_SEGMENT_SIZE,
3323 scm_default_init_heap_size_1, scm_default_min_yield_1,
3324 scm_default_init_heap_size_2, scm_default_min_yield_2,
3325 scm_default_max_segment_size
3327 respectively before callong scm_boot_guile.
3329 (See entry "New environment variables ..." in section
3330 "Changes to the stand-alone interpreter" above.)
3332 ** scm_protect_object/scm_unprotect_object now nest
3334 This means that you can call scm_protect_object multiple times on an
3335 object and count on the object being protected until
3336 scm_unprotect_object has been call the same number of times.
3338 The functions also have better time complexity.
3340 Still, it is usually possible to structure the application in a way
3341 that you don't need to use these functions. For example, if you use a
3342 protected standard Guile list to keep track of live objects rather
3343 than some custom data type, objects will die a natural death when they
3344 are no longer needed.
3346 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
3348 Guile does not provide the float representation for inexact real numbers any
3349 more. Now, only doubles are used to represent inexact real numbers. Further,
3350 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
3351 and scm_tc16_complex, respectively.
3353 ** Removed deprecated type scm_smobfuns
3355 ** Removed deprecated function scm_newsmob
3357 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
3359 There is an ongoing discussion among the developers whether to
3360 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
3361 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
3362 until this issue has been settled.
3364 ** Removed deprecated type tag scm_tc16_kw
3366 ** Added type tag scm_tc16_keyword
3368 (This was introduced already in release 1.3.4 but was not documented
3371 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
3373 * Changes to system call interfaces:
3375 ** The "select" procedure now tests port buffers for the ability to
3376 provide input or accept output. Previously only the underlying file
3377 descriptors were checked.
3379 ** New variable PIPE_BUF: the maximum number of bytes that can be
3380 atomically written to a pipe.
3382 ** If a facility is not available on the system when Guile is
3383 compiled, the corresponding primitive procedure will not be defined.
3384 Previously it would have been defined but would throw a system-error
3385 exception if called. Exception handlers which catch this case may
3386 need minor modification: an error will be thrown with key
3387 'unbound-variable instead of 'system-error. Alternatively it's
3388 now possible to use `defined?' to check whether the facility is
3391 ** Procedures which depend on the timezone should now give the correct
3392 result on systems which cache the TZ environment variable, even if TZ
3393 is changed without calling tzset.
3395 * Changes to the networking interfaces:
3397 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
3398 long integers between network and host format. For now, it's not
3399 particularly convenient to do this kind of thing, but consider:
3401 (define write-network-long
3402 (lambda (value port)
3403 (let ((v (make-uniform-vector 1 1 0)))
3404 (uniform-vector-set! v 0 (htonl value))
3405 (uniform-vector-write v port))))
3407 (define read-network-long
3409 (let ((v (make-uniform-vector 1 1 0)))
3410 (uniform-vector-read! v port)
3411 (ntohl (uniform-vector-ref v 0)))))
3413 ** If inet-aton fails, it now throws an error with key 'misc-error
3414 instead of 'system-error, since errno is not relevant.
3416 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
3417 specific keys instead of 'system-error. The latter is inappropriate
3418 since errno will not have been set. The keys are:
3419 'host-not-found, 'try-again, 'no-recovery and 'no-data.
3421 ** sethostent, setnetent, setprotoent, setservent: now take an
3422 optional argument STAYOPEN, which specifies whether the database
3423 remains open after a database entry is accessed randomly (e.g., using
3424 gethostbyname for the hosts database.) The default is #f. Previously
3428 Changes since Guile 1.3.2:
3430 * Changes to the stand-alone interpreter
3434 An initial version of the Guile debugger written by Chris Hanson has
3435 been added. The debugger is still under development but is included
3436 in the distribution anyway since it is already quite useful.
3442 after an error to enter the debugger. Type `help' inside the debugger
3443 for a description of available commands.
3445 If you prefer to have stack frames numbered and printed in
3446 anti-chronological order and prefer up in the stack to be down on the
3447 screen as is the case in gdb, you can put
3449 (debug-enable 'backwards)
3451 in your .guile startup file. (However, this means that Guile can't
3452 use indentation to indicate stack level.)
3454 The debugger is autoloaded into Guile at the first use.
3456 ** Further enhancements to backtraces
3458 There is a new debug option `width' which controls the maximum width
3459 on the screen of printed stack frames. Fancy printing parameters
3460 ("level" and "length" as in Common LISP) are adaptively adjusted for
3461 each stack frame to give maximum information while still fitting
3462 within the bounds. If the stack frame can't be made to fit by
3463 adjusting parameters, it is simply cut off at the end. This is marked
3466 ** Some modules are now only loaded when the repl is started
3468 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
3469 regex) are now loaded into (guile-user) only if the repl has been
3470 started. The effect is that the startup time for scripts has been
3471 reduced to 30% of what it was previously.
3473 Correctly written scripts load the modules they require at the top of
3474 the file and should not be affected by this change.
3476 ** Hooks are now represented as smobs
3478 * Changes to Scheme functions and syntax
3480 ** Readline support has changed again.
3482 The old (readline-activator) module is gone. Use (ice-9 readline)
3483 instead, which now contains all readline functionality. So the code
3484 to activate readline is now
3486 (use-modules (ice-9 readline))
3489 This should work at any time, including from the guile prompt.
3491 To avoid confusion about the terms of Guile's license, please only
3492 enable readline for your personal use; please don't make it the
3493 default for others. Here is why we make this rather odd-sounding
3496 Guile is normally licensed under a weakened form of the GNU General
3497 Public License, which allows you to link code with Guile without
3498 placing that code under the GPL. This exception is important to some
3501 However, since readline is distributed under the GNU General Public
3502 License, when you link Guile with readline, either statically or
3503 dynamically, you effectively change Guile's license to the strict GPL.
3504 Whenever you link any strictly GPL'd code into Guile, uses of Guile
3505 which are normally permitted become forbidden. This is a rather
3506 non-obvious consequence of the licensing terms.
3508 So, to make sure things remain clear, please let people choose for
3509 themselves whether to link GPL'd libraries like readline with Guile.
3511 ** regexp-substitute/global has changed slightly, but incompatibly.
3513 If you include a function in the item list, the string of the match
3514 object it receives is the same string passed to
3515 regexp-substitute/global, not some suffix of that string.
3516 Correspondingly, the match's positions are relative to the entire
3517 string, not the suffix.
3519 If the regexp can match the empty string, the way matches are chosen
3520 from the string has changed. regexp-substitute/global recognizes the
3521 same set of matches that list-matches does; see below.
3523 ** New function: list-matches REGEXP STRING [FLAGS]
3525 Return a list of match objects, one for every non-overlapping, maximal
3526 match of REGEXP in STRING. The matches appear in left-to-right order.
3527 list-matches only reports matches of the empty string if there are no
3528 other matches which begin on, end at, or include the empty match's
3531 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3533 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
3535 For each match of REGEXP in STRING, apply PROC to the match object,
3536 and the last value PROC returned, or INIT for the first call. Return
3537 the last value returned by PROC. We apply PROC to the matches as they
3538 appear from left to right.
3540 This function recognizes matches according to the same criteria as
3543 Thus, you could define list-matches like this:
3545 (define (list-matches regexp string . flags)
3546 (reverse! (apply fold-matches regexp string '() cons flags)))
3548 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
3552 *** New function: hook? OBJ
3554 Return #t if OBJ is a hook, otherwise #f.
3556 *** New function: make-hook-with-name NAME [ARITY]
3558 Return a hook with name NAME and arity ARITY. The default value for
3559 ARITY is 0. The only effect of NAME is that it will appear when the
3560 hook object is printed to ease debugging.
3562 *** New function: hook-empty? HOOK
3564 Return #t if HOOK doesn't contain any procedures, otherwise #f.
3566 *** New function: hook->list HOOK
3568 Return a list of the procedures that are called when run-hook is
3571 ** `map' signals an error if its argument lists are not all the same length.
3573 This is the behavior required by R5RS, so this change is really a bug
3574 fix. But it seems to affect a lot of people's code, so we're
3575 mentioning it here anyway.
3577 ** Print-state handling has been made more transparent
3579 Under certain circumstances, ports are represented as a port with an
3580 associated print state. Earlier, this pair was represented as a pair
3581 (see "Some magic has been added to the printer" below). It is now
3582 indistinguishable (almost; see `get-print-state') from a port on the
3585 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
3587 Return a new port with the associated print state PRINT-STATE.
3589 *** New function: get-print-state OUTPUT-PORT
3591 Return the print state associated with this port if it exists,
3592 otherwise return #f.
3594 *** New function: directory-stream? OBJECT
3596 Returns true iff OBJECT is a directory stream --- the sort of object
3597 returned by `opendir'.
3599 ** New function: using-readline?
3601 Return #t if readline is in use in the current repl.
3603 ** structs will be removed in 1.4
3605 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
3606 and use GOOPS objects as the fundamental record type.
3608 * Changes to the scm_ interface
3610 ** structs will be removed in 1.4
3612 The entire current struct interface (struct.c, struct.h) will be
3613 replaced in Guile 1.4. We will merge GOOPS into libguile and use
3614 GOOPS objects as the fundamental record type.
3616 ** The internal representation of subr's has changed
3618 Instead of giving a hint to the subr name, the CAR field of the subr
3619 now contains an index to a subr entry in scm_subr_table.
3621 *** New variable: scm_subr_table
3623 An array of subr entries. A subr entry contains the name, properties
3624 and documentation associated with the subr. The properties and
3625 documentation slots are not yet used.
3627 ** A new scheme for "forwarding" calls to a builtin to a generic function
3629 It is now possible to extend the functionality of some Guile
3630 primitives by letting them defer a call to a GOOPS generic function on
3631 argument mismatch. This means that there is no loss of efficiency in
3636 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
3637 (define-method + ((x <string>) (y <string>))
3638 (string-append x y))
3640 + will still be as efficient as usual in numerical calculations, but
3641 can also be used for concatenating strings.
3643 Who will be the first one to extend Guile's numerical tower to
3644 rationals? :) [OK, there a few other things to fix before this can
3645 be made in a clean way.]
3647 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
3649 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
3651 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
3653 These do the same job as SCM_PROC and SCM_PROC1, but they also define
3654 a variable GENERIC which can be used by the dispatch macros below.
3656 [This is experimental code which may change soon.]
3658 *** New macros for forwarding control to a generic on arg type error
3660 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
3662 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
3664 These correspond to the scm_wta function call, and have the same
3665 behaviour until the user has called the GOOPS primitive
3666 `enable-primitive-generic!'. After that, these macros will apply the
3667 generic function GENERIC to the argument(s) instead of calling
3670 [This is experimental code which may change soon.]
3672 *** New macros for argument testing with generic dispatch
3674 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
3676 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
3678 These correspond to the SCM_ASSERT macro, but will defer control to
3679 GENERIC on error after `enable-primitive-generic!' has been called.
3681 [This is experimental code which may change soon.]
3683 ** New function: SCM scm_eval_body (SCM body, SCM env)
3685 Evaluates the body of a special form.
3687 ** The internal representation of struct's has changed
3689 Previously, four slots were allocated for the procedure(s) of entities
3690 and operators. The motivation for this representation had to do with
3691 the structure of the evaluator, the wish to support tail-recursive
3692 generic functions, and efficiency. Since the generic function
3693 dispatch mechanism has changed, there is no longer a need for such an
3694 expensive representation, and the representation has been simplified.
3696 This should not make any difference for most users.
3698 ** GOOPS support has been cleaned up.
3700 Some code has been moved from eval.c to objects.c and code in both of
3701 these compilation units has been cleaned up and better structured.
3703 *** New functions for applying generic functions
3705 New function: SCM scm_apply_generic (GENERIC, ARGS)
3706 New function: SCM scm_call_generic_0 (GENERIC)
3707 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
3708 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
3709 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
3711 ** Deprecated function: scm_make_named_hook
3713 It is now replaced by:
3715 ** New function: SCM scm_create_hook (const char *name, int arity)
3717 Creates a hook in the same way as make-hook above but also
3718 binds a variable named NAME to it.
3720 This is the typical way of creating a hook from C code.
3722 Currently, the variable is created in the "current" module.
3723 This might change when we get the new module system.
3725 [The behaviour is identical to scm_make_named_hook.]
3729 Changes since Guile 1.3:
3731 * Changes to mailing lists
3733 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
3735 See the README file to find current addresses for all the Guile
3738 * Changes to the distribution
3740 ** Readline support is no longer included with Guile by default.
3742 Based on the different license terms of Guile and Readline, we
3743 concluded that Guile should not *by default* cause the linking of
3744 Readline into an application program. Readline support is now offered
3745 as a separate module, which is linked into an application only when
3746 you explicitly specify it.
3748 Although Guile is GNU software, its distribution terms add a special
3749 exception to the usual GNU General Public License (GPL). Guile's
3750 license includes a clause that allows you to link Guile with non-free
3751 programs. We add this exception so as not to put Guile at a
3752 disadvantage vis-a-vis other extensibility packages that support other
3755 In contrast, the GNU Readline library is distributed under the GNU
3756 General Public License pure and simple. This means that you may not
3757 link Readline, even dynamically, into an application unless it is
3758 distributed under a free software license that is compatible the GPL.
3760 Because of this difference in distribution terms, an application that
3761 can use Guile may not be able to use Readline. Now users will be
3762 explicitly offered two independent decisions about the use of these
3765 You can activate the readline support by issuing
3767 (use-modules (readline-activator))
3770 from your ".guile" file, for example.
3772 * Changes to the stand-alone interpreter
3774 ** All builtins now print as primitives.
3775 Previously builtin procedures not belonging to the fundamental subr
3776 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
3777 Now, they print as #<primitive-procedure NAME>.
3779 ** Backtraces slightly more intelligible.
3780 gsubr-apply and macro transformer application frames no longer appear
3783 * Changes to Scheme functions and syntax
3785 ** Guile now correctly handles internal defines by rewriting them into
3786 their equivalent letrec. Previously, internal defines would
3787 incrementally add to the innermost environment, without checking
3788 whether the restrictions specified in RnRS were met. This lead to the
3789 correct behaviour when these restriction actually were met, but didn't
3790 catch all illegal uses. Such an illegal use could lead to crashes of
3791 the Guile interpreter or or other unwanted results. An example of
3792 incorrect internal defines that made Guile behave erratically:
3804 The problem with this example is that the definition of `c' uses the
3805 value of `b' directly. This confuses the meoization machine of Guile
3806 so that the second call of `b' (this time in a larger environment that
3807 also contains bindings for `c' and `d') refers to the binding of `c'
3808 instead of `a'. You could also make Guile crash with a variation on
3813 (define (b flag) (if flag a 1))
3814 (define c (1+ (b flag)))
3822 From now on, Guile will issue an `Unbound variable: b' error message
3827 A hook contains a list of functions which should be called on
3828 particular occasions in an existing program. Hooks are used for
3831 A window manager might have a hook before-window-map-hook. The window
3832 manager uses the function run-hooks to call all functions stored in
3833 before-window-map-hook each time a window is mapped. The user can
3834 store functions in the hook using add-hook!.
3836 In Guile, hooks are first class objects.
3838 *** New function: make-hook [N_ARGS]
3840 Return a hook for hook functions which can take N_ARGS arguments.
3841 The default value for N_ARGS is 0.
3843 (See also scm_make_named_hook below.)
3845 *** New function: add-hook! HOOK PROC [APPEND_P]
3847 Put PROC at the beginning of the list of functions stored in HOOK.
3848 If APPEND_P is supplied, and non-false, put PROC at the end instead.
3850 PROC must be able to take the number of arguments specified when the
3853 If PROC already exists in HOOK, then remove it first.
3855 *** New function: remove-hook! HOOK PROC
3857 Remove PROC from the list of functions in HOOK.
3859 *** New function: reset-hook! HOOK
3861 Clear the list of hook functions stored in HOOK.
3863 *** New function: run-hook HOOK ARG1 ...
3865 Run all hook functions stored in HOOK with arguments ARG1 ... .
3866 The number of arguments supplied must correspond to the number given
3867 when the hook was created.
3869 ** The function `dynamic-link' now takes optional keyword arguments.
3870 The only keyword argument that is currently defined is `:global
3871 BOOL'. With it, you can control whether the shared library will be
3872 linked in global mode or not. In global mode, the symbols from the
3873 linked library can be used to resolve references from other
3874 dynamically linked libraries. In non-global mode, the linked
3875 library is essentially invisible and can only be accessed via
3876 `dynamic-func', etc. The default is now to link in global mode.
3877 Previously, the default has been non-global mode.
3879 The `#:global' keyword is only effective on platforms that support
3880 the dlopen family of functions.
3882 ** New function `provided?'
3884 - Function: provided? FEATURE
3885 Return true iff FEATURE is supported by this installation of
3886 Guile. FEATURE must be a symbol naming a feature; the global
3887 variable `*features*' is a list of available features.
3889 ** Changes to the module (ice-9 expect):
3891 *** The expect-strings macro now matches `$' in a regular expression
3892 only at a line-break or end-of-file by default. Previously it would
3893 match the end of the string accumulated so far. The old behaviour
3894 can be obtained by setting the variable `expect-strings-exec-flags'
3897 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3898 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3899 in a regular expression will still match before a line-break or
3900 end-of-file. The default is `regexp/noteol'.
3902 *** The expect-strings macro now uses a variable
3903 `expect-strings-compile-flags' for the flags to be supplied to
3904 `make-regexp'. The default is `regexp/newline', which was previously
3907 *** The expect macro now supplies two arguments to a match procedure:
3908 the current accumulated string and a flag to indicate whether
3909 end-of-file has been reached. Previously only the string was supplied.
3910 If end-of-file is reached, the match procedure will be called an
3911 additional time with the same accumulated string as the previous call
3912 but with the flag set.
3914 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3916 This code, and the documentation for it that appears here, was
3917 borrowed from SLIB, with minor adaptations for Guile.
3919 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3920 An almost complete implementation of Common LISP format description
3921 according to the CL reference book `Common LISP' from Guy L.
3922 Steele, Digital Press. Backward compatible to most of the
3923 available Scheme format implementations.
3925 Returns `#t', `#f' or a string; has side effect of printing
3926 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3927 to the current output port and `#t' is returned. If DESTINATION
3928 is `#f', a formatted string is returned as the result of the call.
3929 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3930 format string; FORMAT-STRING is then the first argument and the
3931 output is returned as a string. If DESTINATION is a number, the
3932 output is to the current error port if available by the
3933 implementation. Otherwise DESTINATION must be an output port and
3936 FORMAT-STRING must be a string. In case of a formatting error
3937 format returns `#f' and prints a message on the current output or
3938 error port. Characters are output as if the string were output by
3939 the `display' function with the exception of those prefixed by a
3940 tilde (~). For a detailed description of the FORMAT-STRING syntax
3941 please consult a Common LISP format reference manual. For a test
3942 suite to verify this format implementation load `formatst.scm'.
3943 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3945 Note: `format' is not reentrant, i.e. only one `format'-call may
3946 be executed at a time.
3949 *** Format Specification (Format version 3.0)
3951 Please consult a Common LISP format reference manual for a detailed
3952 description of the format string syntax. For a demonstration of the
3953 implemented directives see `formatst.scm'.
3955 This implementation supports directive parameters and modifiers (`:'
3956 and `@' characters). Multiple parameters must be separated by a comma
3957 (`,'). Parameters can be numerical parameters (positive or negative),
3958 character parameters (prefixed by a quote character (`''), variable
3959 parameters (`v'), number of rest arguments parameter (`#'), empty and
3960 default parameters. Directive characters are case independent. The
3961 general form of a directive is:
3963 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3965 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3967 *** Implemented CL Format Control Directives
3969 Documentation syntax: Uppercase characters represent the
3970 corresponding control directive characters. Lowercase characters
3971 represent control directive parameter descriptions.
3974 Any (print as `display' does).
3978 `~MINCOL,COLINC,MINPAD,PADCHARA'
3982 S-expression (print as `write' does).
3986 `~MINCOL,COLINC,MINPAD,PADCHARS'
3992 print number sign always.
3995 print comma separated.
3997 `~MINCOL,PADCHAR,COMMACHARD'
4003 print number sign always.
4006 print comma separated.
4008 `~MINCOL,PADCHAR,COMMACHARX'
4014 print number sign always.
4017 print comma separated.
4019 `~MINCOL,PADCHAR,COMMACHARO'
4025 print number sign always.
4028 print comma separated.
4030 `~MINCOL,PADCHAR,COMMACHARB'
4035 `~N,MINCOL,PADCHAR,COMMACHARR'
4039 print a number as a Roman numeral.
4042 print a number as an "old fashioned" Roman numeral.
4045 print a number as an ordinal English number.
4048 print a number as a cardinal English number.
4053 prints `y' and `ies'.
4056 as `~P but jumps 1 argument backward.'
4059 as `~@P but jumps 1 argument backward.'
4064 prints a character as the reader can understand it (i.e. `#\'
4068 prints a character as emacs does (eg. `^C' for ASCII 03).
4071 Fixed-format floating-point (prints a flonum like MMM.NNN).
4072 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
4074 If the number is positive a plus sign is printed.
4077 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
4078 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
4080 If the number is positive a plus sign is printed.
4083 General floating-point (prints a flonum either fixed or
4085 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
4087 If the number is positive a plus sign is printed.
4090 Dollars floating-point (prints a flonum in fixed with signs
4092 `~DIGITS,SCALE,WIDTH,PADCHAR$'
4094 If the number is positive a plus sign is printed.
4097 A sign is always printed and appears before the padding.
4100 The sign appears before the padding.
4108 print newline if not at the beginning of the output line.
4110 prints `~&' and then N-1 newlines.
4115 print N page separators.
4125 newline is ignored, white space left.
4128 newline is left, white space ignored.
4133 relative tabulation.
4139 Indirection (expects indirect arguments as a list).
4141 extracts indirect arguments from format arguments.
4144 Case conversion (converts by `string-downcase').
4146 converts by `string-capitalize'.
4149 converts by `string-capitalize-first'.
4152 converts by `string-upcase'.
4155 Argument Jumping (jumps 1 argument forward).
4157 jumps N arguments forward.
4160 jumps 1 argument backward.
4163 jumps N arguments backward.
4166 jumps to the 0th argument.
4169 jumps to the Nth argument (beginning from 0)
4171 `~[STR0~;STR1~;...~;STRN~]'
4172 Conditional Expression (numerical clause conditional).
4174 take argument from N.
4177 true test conditional.
4180 if-else-then conditional.
4186 default clause follows.
4189 Iteration (args come from the next argument (a list)).
4191 at most N iterations.
4194 args from next arg (a list of lists).
4197 args from the rest of arguments.
4200 args from the rest args (lists).
4211 aborts if N <= M <= K
4213 *** Not Implemented CL Format Control Directives
4216 print `#f' as an empty list (see below).
4219 print `#f' as an empty list (see below).
4225 (sorry I don't understand its semantics completely)
4227 *** Extended, Replaced and Additional Control Directives
4229 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
4230 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
4231 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
4232 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
4233 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
4234 COMMAWIDTH is the number of characters between two comma
4238 print a R4RS complex number as `~F~@Fi' with passed parameters for
4242 Pretty print formatting of an argument for scheme code lists.
4248 Flushes the output if format DESTINATION is a port.
4251 Print a `#\space' character
4253 print N `#\space' characters.
4256 Print a `#\tab' character
4258 print N `#\tab' characters.
4261 Takes N as an integer representation for a character. No arguments
4262 are consumed. N is converted to a character by `integer->char'. N
4263 must be a positive decimal number.
4266 Print out readproof. Prints out internal objects represented as
4267 `#<...>' as strings `"#<...>"' so that the format output can always
4268 be processed by `read'.
4271 Print out readproof. Prints out internal objects represented as
4272 `#<...>' as strings `"#<...>"' so that the format output can always
4273 be processed by `read'.
4276 Prints information and a copyright notice on the format
4279 prints format version.
4282 may also print number strings, i.e. passing a number as a string
4283 and format it accordingly.
4285 *** Configuration Variables
4287 The format module exports some configuration variables to suit the
4288 systems and users needs. There should be no modification necessary for
4289 the configuration that comes with Guile. Format detects automatically
4290 if the running scheme system implements floating point numbers and
4293 format:symbol-case-conv
4294 Symbols are converted by `symbol->string' so the case type of the
4295 printed symbols is implementation dependent.
4296 `format:symbol-case-conv' is a one arg closure which is either
4297 `#f' (no conversion), `string-upcase', `string-downcase' or
4298 `string-capitalize'. (default `#f')
4300 format:iobj-case-conv
4301 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
4302 implementation internal objects. (default `#f')
4305 The character prefixing the exponent value in `~E' printing.
4308 *** Compatibility With Other Format Implementations
4314 Downward compatible except for padding support and `~A', `~S',
4315 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
4316 `printf' padding support which is completely replaced by the CL
4317 `format' padding style.
4320 Downward compatible except for `~', which is not documented
4321 (ignores all characters inside the format string up to a newline
4322 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
4323 numerical and variable parameters and `:/@' modifiers in the CL
4327 Downward compatible except for `~A' and `~S' which print in
4328 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
4329 directive parameters or modifiers)).
4332 Downward compatible except for an optional destination parameter:
4333 S2C accepts a format call without a destination which returns a
4334 formatted string. This is equivalent to a #f destination in S2C.
4335 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
4336 parameters or modifiers)).
4339 ** Changes to string-handling functions.
4341 These functions were added to support the (ice-9 format) module, above.
4343 *** New function: string-upcase STRING
4344 *** New function: string-downcase STRING
4346 These are non-destructive versions of the existing string-upcase! and
4347 string-downcase! functions.
4349 *** New function: string-capitalize! STRING
4350 *** New function: string-capitalize STRING
4352 These functions convert the first letter of each word in the string to
4355 (string-capitalize "howdy there")
4358 As with the other functions, string-capitalize! modifies the string in
4359 place, while string-capitalize returns a modified copy of its argument.
4361 *** New function: string-ci->symbol STRING
4363 Return a symbol whose name is STRING, but having the same case as if
4364 the symbol had be read by `read'.
4366 Guile can be configured to be sensitive or insensitive to case
4367 differences in Scheme identifiers. If Guile is case-insensitive, all
4368 symbols are converted to lower case on input. The `string-ci->symbol'
4369 function returns a symbol whose name in STRING, transformed as Guile
4370 would if STRING were input.
4372 *** New function: substring-move! STRING1 START END STRING2 START
4374 Copy the substring of STRING1 from START (inclusive) to END
4375 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
4376 string, and the source and destination areas may overlap; in all
4377 cases, the function behaves as if all the characters were copied
4380 *** Extended functions: substring-move-left! substring-move-right!
4382 These functions now correctly copy arbitrarily overlapping substrings;
4383 they are both synonyms for substring-move!.
4386 ** New module (ice-9 getopt-long), with the function `getopt-long'.
4388 getopt-long is a function for parsing command-line arguments in a
4389 manner consistent with other GNU programs.
4391 (getopt-long ARGS GRAMMAR)
4392 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
4394 ARGS should be a list of strings. Its first element should be the
4395 name of the program; subsequent elements should be the arguments
4396 that were passed to the program on the command line. The
4397 `program-arguments' procedure returns a list of this form.
4399 GRAMMAR is a list of the form:
4400 ((OPTION (PROPERTY VALUE) ...) ...)
4402 Each OPTION should be a symbol. `getopt-long' will accept a
4403 command-line option named `--OPTION'.
4404 Each option can have the following (PROPERTY VALUE) pairs:
4406 (single-char CHAR) --- Accept `-CHAR' as a single-character
4407 equivalent to `--OPTION'. This is how to specify traditional
4409 (required? BOOL) --- If BOOL is true, the option is required.
4410 getopt-long will raise an error if it is not found in ARGS.
4411 (value BOOL) --- If BOOL is #t, the option accepts a value; if
4412 it is #f, it does not; and if it is the symbol
4413 `optional', the option may appear in ARGS with or
4415 (predicate FUNC) --- If the option accepts a value (i.e. you
4416 specified `(value #t)' for this option), then getopt
4417 will apply FUNC to the value, and throw an exception
4418 if it returns #f. FUNC should be a procedure which
4419 accepts a string and returns a boolean value; you may
4420 need to use quasiquotes to get it into GRAMMAR.
4422 The (PROPERTY VALUE) pairs may occur in any order, but each
4423 property may occur only once. By default, options do not have
4424 single-character equivalents, are not required, and do not take
4427 In ARGS, single-character options may be combined, in the usual
4428 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
4429 accepts values, then it must be the last option in the
4430 combination; the value is the next argument. So, for example, using
4431 the following grammar:
4432 ((apples (single-char #\a))
4433 (blimps (single-char #\b) (value #t))
4434 (catalexis (single-char #\c) (value #t)))
4435 the following argument lists would be acceptable:
4436 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
4437 for "blimps" and "catalexis")
4438 ("-ab" "bang" "-c" "couth") (same)
4439 ("-ac" "couth" "-b" "bang") (same)
4440 ("-abc" "couth" "bang") (an error, since `-b' is not the
4441 last option in its combination)
4443 If an option's value is optional, then `getopt-long' decides
4444 whether it has a value by looking at what follows it in ARGS. If
4445 the next element is a string, and it does not appear to be an
4446 option itself, then that string is the option's value.
4448 The value of a long option can appear as the next element in ARGS,
4449 or it can follow the option name, separated by an `=' character.
4450 Thus, using the same grammar as above, the following argument lists
4452 ("--apples" "Braeburn" "--blimps" "Goodyear")
4453 ("--apples=Braeburn" "--blimps" "Goodyear")
4454 ("--blimps" "Goodyear" "--apples=Braeburn")
4456 If the option "--" appears in ARGS, argument parsing stops there;
4457 subsequent arguments are returned as ordinary arguments, even if
4458 they resemble options. So, in the argument list:
4459 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
4460 `getopt-long' will recognize the `apples' option as having the
4461 value "Granny Smith", but it will not recognize the `blimp'
4462 option; it will return the strings "--blimp" and "Goodyear" as
4463 ordinary argument strings.
4465 The `getopt-long' function returns the parsed argument list as an
4466 assocation list, mapping option names --- the symbols from GRAMMAR
4467 --- onto their values, or #t if the option does not accept a value.
4468 Unused options do not appear in the alist.
4470 All arguments that are not the value of any option are returned
4471 as a list, associated with the empty list.
4473 `getopt-long' throws an exception if:
4474 - it finds an unrecognized option in ARGS
4475 - a required option is omitted
4476 - an option that requires an argument doesn't get one
4477 - an option that doesn't accept an argument does get one (this can
4478 only happen using the long option `--opt=value' syntax)
4479 - an option predicate fails
4484 `((lockfile-dir (required? #t)
4487 (predicate ,file-is-directory?))
4488 (verbose (required? #f)
4491 (x-includes (single-char #\x))
4492 (rnet-server (single-char #\y)
4493 (predicate ,string?))))
4495 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
4496 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
4498 => ((() "foo1" "-fred" "foo2" "foo3")
4499 (rnet-server . "lamprod")
4500 (x-includes . "/usr/include")
4501 (lockfile-dir . "/tmp")
4504 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
4506 It will be removed in a few releases.
4508 ** New syntax: lambda*
4509 ** New syntax: define*
4510 ** New syntax: define*-public
4511 ** New syntax: defmacro*
4512 ** New syntax: defmacro*-public
4513 Guile now supports optional arguments.
4515 `lambda*', `define*', `define*-public', `defmacro*' and
4516 `defmacro*-public' are identical to the non-* versions except that
4517 they use an extended type of parameter list that has the following BNF
4518 syntax (parentheses are literal, square brackets indicate grouping,
4519 and `*', `+' and `?' have the usual meaning):
4521 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
4522 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
4523 [[#&rest identifier]|[. identifier]]? ) | [identifier]
4525 ext-var-decl ::= identifier | ( identifier expression )
4527 The semantics are best illustrated with the following documentation
4528 and examples for `lambda*':
4531 lambda extended for optional and keyword arguments
4533 lambda* creates a procedure that takes optional arguments. These
4534 are specified by putting them inside brackets at the end of the
4535 paramater list, but before any dotted rest argument. For example,
4536 (lambda* (a b #&optional c d . e) '())
4537 creates a procedure with fixed arguments a and b, optional arguments c
4538 and d, and rest argument e. If the optional arguments are omitted
4539 in a call, the variables for them are unbound in the procedure. This
4540 can be checked with the bound? macro.
4542 lambda* can also take keyword arguments. For example, a procedure
4544 (lambda* (#&key xyzzy larch) '())
4545 can be called with any of the argument lists (#:xyzzy 11)
4546 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
4547 are given as keywords are bound to values.
4549 Optional and keyword arguments can also be given default values
4550 which they take on when they are not present in a call, by giving a
4551 two-item list in place of an optional argument, for example in:
4552 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
4553 foo is a fixed argument, bar is an optional argument with default
4554 value 42, and baz is a keyword argument with default value 73.
4555 Default value expressions are not evaluated unless they are needed
4556 and until the procedure is called.
4558 lambda* now supports two more special parameter list keywords.
4560 lambda*-defined procedures now throw an error by default if a
4561 keyword other than one of those specified is found in the actual
4562 passed arguments. However, specifying #&allow-other-keys
4563 immediately after the kyword argument declarations restores the
4564 previous behavior of ignoring unknown keywords. lambda* also now
4565 guarantees that if the same keyword is passed more than once, the
4566 last one passed is the one that takes effect. For example,
4567 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
4568 #:heads 37 #:tails 42 #:heads 99)
4569 would result in (99 47) being displayed.
4571 #&rest is also now provided as a synonym for the dotted syntax rest
4572 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
4573 all respects to lambda*. This is provided for more similarity to DSSSL,
4574 MIT-Scheme and Kawa among others, as well as for refugees from other
4577 Further documentation may be found in the optargs.scm file itself.
4579 The optional argument module also exports the macros `let-optional',
4580 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
4581 are not documented here because they may be removed in the future, but
4582 full documentation is still available in optargs.scm.
4584 ** New syntax: and-let*
4585 Guile now supports the `and-let*' form, described in the draft SRFI-2.
4587 Syntax: (land* (<clause> ...) <body> ...)
4588 Each <clause> should have one of the following forms:
4589 (<variable> <expression>)
4592 Each <variable> or <bound-variable> should be an identifier. Each
4593 <expression> should be a valid expression. The <body> should be a
4594 possibly empty sequence of expressions, like the <body> of a
4597 Semantics: A LAND* expression is evaluated by evaluating the
4598 <expression> or <bound-variable> of each of the <clause>s from
4599 left to right. The value of the first <expression> or
4600 <bound-variable> that evaluates to a false value is returned; the
4601 remaining <expression>s and <bound-variable>s are not evaluated.
4602 The <body> forms are evaluated iff all the <expression>s and
4603 <bound-variable>s evaluate to true values.
4605 The <expression>s and the <body> are evaluated in an environment
4606 binding each <variable> of the preceding (<variable> <expression>)
4607 clauses to the value of the <expression>. Later bindings
4608 shadow earlier bindings.
4610 Guile's and-let* macro was contributed by Michael Livshin.
4612 ** New sorting functions
4614 *** New function: sorted? SEQUENCE LESS?
4615 Returns `#t' when the sequence argument is in non-decreasing order
4616 according to LESS? (that is, there is no adjacent pair `... x y
4617 ...' for which `(less? y x)').
4619 Returns `#f' when the sequence contains at least one out-of-order
4620 pair. It is an error if the sequence is neither a list nor a
4623 *** New function: merge LIST1 LIST2 LESS?
4624 LIST1 and LIST2 are sorted lists.
4625 Returns the sorted list of all elements in LIST1 and LIST2.
4627 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
4628 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
4629 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
4630 (Here "<" should read "comes before".)
4632 *** New procedure: merge! LIST1 LIST2 LESS?
4633 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
4634 the result. If the code is compiled, and LESS? constructs no new
4635 pairs, no pairs at all will be allocated. The first pair of the
4636 result will be either the first pair of LIST1 or the first pair of
4639 *** New function: sort SEQUENCE LESS?
4640 Accepts either a list or a vector, and returns a new sequence
4641 which is sorted. The new sequence is the same type as the input.
4642 Always `(sorted? (sort sequence less?) less?)'. The original
4643 sequence is not altered in any way. The new sequence shares its
4644 elements with the old one; no elements are copied.
4646 *** New procedure: sort! SEQUENCE LESS
4647 Returns its sorted result in the original boxes. No new storage is
4648 allocated at all. Proper usage: (set! slist (sort! slist <))
4650 *** New function: stable-sort SEQUENCE LESS?
4651 Similar to `sort' but stable. That is, if "equal" elements are
4652 ordered a < b in the original sequence, they will have the same order
4655 *** New function: stable-sort! SEQUENCE LESS?
4656 Similar to `sort!' but stable.
4657 Uses temporary storage when sorting vectors.
4659 *** New functions: sort-list, sort-list!
4660 Added for compatibility with scsh.
4662 ** New built-in random number support
4664 *** New function: random N [STATE]
4665 Accepts a positive integer or real N and returns a number of the
4666 same type between zero (inclusive) and N (exclusive). The values
4667 returned have a uniform distribution.
4669 The optional argument STATE must be of the type produced by
4670 `copy-random-state' or `seed->random-state'. It defaults to the value
4671 of the variable `*random-state*'. This object is used to maintain the
4672 state of the pseudo-random-number generator and is altered as a side
4673 effect of the `random' operation.
4675 *** New variable: *random-state*
4676 Holds a data structure that encodes the internal state of the
4677 random-number generator that `random' uses by default. The nature
4678 of this data structure is implementation-dependent. It may be
4679 printed out and successfully read back in, but may or may not
4680 function correctly as a random-number state object in another
4683 *** New function: copy-random-state [STATE]
4684 Returns a new object of type suitable for use as the value of the
4685 variable `*random-state*' and as a second argument to `random'.
4686 If argument STATE is given, a copy of it is returned. Otherwise a
4687 copy of `*random-state*' is returned.
4689 *** New function: seed->random-state SEED
4690 Returns a new object of type suitable for use as the value of the
4691 variable `*random-state*' and as a second argument to `random'.
4692 SEED is a string or a number. A new state is generated and
4693 initialized using SEED.
4695 *** New function: random:uniform [STATE]
4696 Returns an uniformly distributed inexact real random number in the
4697 range between 0 and 1.
4699 *** New procedure: random:solid-sphere! VECT [STATE]
4700 Fills VECT with inexact real random numbers the sum of whose
4701 squares is less than 1.0. Thinking of VECT as coordinates in
4702 space of dimension N = `(vector-length VECT)', the coordinates are
4703 uniformly distributed within the unit N-shere. The sum of the
4704 squares of the numbers is returned. VECT can be either a vector
4705 or a uniform vector of doubles.
4707 *** New procedure: random:hollow-sphere! VECT [STATE]
4708 Fills VECT with inexact real random numbers the sum of whose squares
4709 is equal to 1.0. Thinking of VECT as coordinates in space of
4710 dimension n = `(vector-length VECT)', the coordinates are uniformly
4711 distributed over the surface of the unit n-shere. VECT can be either
4712 a vector or a uniform vector of doubles.
4714 *** New function: random:normal [STATE]
4715 Returns an inexact real in a normal distribution with mean 0 and
4716 standard deviation 1. For a normal distribution with mean M and
4717 standard deviation D use `(+ M (* D (random:normal)))'.
4719 *** New procedure: random:normal-vector! VECT [STATE]
4720 Fills VECT with inexact real random numbers which are independent and
4721 standard normally distributed (i.e., with mean 0 and variance 1).
4722 VECT can be either a vector or a uniform vector of doubles.
4724 *** New function: random:exp STATE
4725 Returns an inexact real in an exponential distribution with mean 1.
4726 For an exponential distribution with mean U use (* U (random:exp)).
4728 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
4730 These functions now operate on numbers in the range of a C unsigned
4733 These functions used to operate on numbers in the range of a C signed
4734 long; however, this seems inappropriate, because Guile integers don't
4737 ** New function: make-guardian
4738 This is an implementation of guardians as described in
4739 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
4740 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
4741 Programming Language Design and Implementation, June 1993
4742 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
4744 ** New functions: delq1!, delv1!, delete1!
4745 These procedures behave similar to delq! and friends but delete only
4746 one object if at all.
4748 ** New function: unread-string STRING PORT
4749 Unread STRING to PORT, that is, push it back onto the port so that
4750 next read operation will work on the pushed back characters.
4752 ** unread-char can now be called multiple times
4753 If unread-char is called multiple times, the unread characters will be
4754 read again in last-in first-out order.
4756 ** the procedures uniform-array-read! and uniform-array-write! now
4757 work on any kind of port, not just ports which are open on a file.
4759 ** Now 'l' in a port mode requests line buffering.
4761 ** The procedure truncate-file now works on string ports as well
4762 as file ports. If the size argument is omitted, the current
4763 file position is used.
4765 ** new procedure: seek PORT/FDES OFFSET WHENCE
4766 The arguments are the same as for the old fseek procedure, but it
4767 works on string ports as well as random-access file ports.
4769 ** the fseek procedure now works on string ports, since it has been
4770 redefined using seek.
4772 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
4773 size is not supplied.
4775 ** the newline procedure no longer flushes the port if it's not
4776 line-buffered: previously it did if it was the current output port.
4778 ** open-pipe and close-pipe are no longer primitive procedures, but
4779 an emulation can be obtained using `(use-modules (ice-9 popen))'.
4781 ** the freopen procedure has been removed.
4783 ** new procedure: drain-input PORT
4784 Drains PORT's read buffers (including any pushed-back characters)
4785 and returns the contents as a single string.
4787 ** New function: map-in-order PROC LIST1 LIST2 ...
4788 Version of `map' which guarantees that the procedure is applied to the
4789 lists in serial order.
4791 ** Renamed `serial-array-copy!' and `serial-array-map!' to
4792 `array-copy-in-order!' and `array-map-in-order!'. The old names are
4793 now obsolete and will go away in release 1.5.
4795 ** New syntax: collect BODY1 ...
4796 Version of `begin' which returns a list of the results of the body
4797 forms instead of the result of the last body form. In contrast to
4798 `begin', `collect' allows an empty body.
4800 ** New functions: read-history FILENAME, write-history FILENAME
4801 Read/write command line history from/to file. Returns #t on success
4802 and #f if an error occured.
4804 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
4806 These procedures return a list of definitions available in the specified
4807 argument, a relative module reference. In the case of no argument,
4808 `(current-module)' is now consulted for definitions to return, instead
4809 of simply returning #f, the former behavior.
4811 ** The #/ syntax for lists is no longer supported.
4813 Earlier versions of Scheme accepted this syntax, but printed a
4816 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
4818 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
4821 * Changes to the gh_ interface
4825 Now takes a second argument which is the result array. If this
4826 pointer is NULL, a new array is malloced (the old behaviour).
4828 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
4829 gh_scm2shorts, gh_scm2longs, gh_scm2floats
4833 * Changes to the scm_ interface
4835 ** Function: scm_make_named_hook (char* name, int n_args)
4837 Creates a hook in the same way as make-hook above but also
4838 binds a variable named NAME to it.
4840 This is the typical way of creating a hook from C code.
4842 Currently, the variable is created in the "current" module. This
4843 might change when we get the new module system.
4845 ** The smob interface
4847 The interface for creating smobs has changed. For documentation, see
4848 data-rep.info (made from guile-core/doc/data-rep.texi).
4850 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
4852 >>> This function will be removed in 1.3.4. <<<
4856 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
4857 This function adds a new smob type, named NAME, with instance size
4858 SIZE to the system. The return value is a tag that is used in
4859 creating instances of the type. If SIZE is 0, then no memory will
4860 be allocated when instances of the smob are created, and nothing
4861 will be freed by the default free function.
4863 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
4864 This function sets the smob marking procedure for the smob type
4865 specified by the tag TC. TC is the tag returned by
4866 `scm_make_smob_type'.
4868 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
4869 This function sets the smob freeing procedure for the smob type
4870 specified by the tag TC. TC is the tag returned by
4871 `scm_make_smob_type'.
4873 *** Function: void scm_set_smob_print (tc, print)
4875 - Function: void scm_set_smob_print (long tc,
4876 scm_sizet (*print) (SCM,
4880 This function sets the smob printing procedure for the smob type
4881 specified by the tag TC. TC is the tag returned by
4882 `scm_make_smob_type'.
4884 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4885 This function sets the smob equality-testing predicate for the
4886 smob type specified by the tag TC. TC is the tag returned by
4887 `scm_make_smob_type'.
4889 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4890 Make VALUE contain a smob instance of the type with type code TC and
4891 smob data DATA. VALUE must be previously declared as C type `SCM'.
4893 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4894 This macro expands to a block of code that creates a smob instance
4895 of the type with type code TC and smob data DATA, and returns that
4896 `SCM' value. It should be the last piece of code in a block.
4898 ** The interfaces for using I/O ports and implementing port types
4899 (ptobs) have changed significantly. The new interface is based on
4900 shared access to buffers and a new set of ptob procedures.
4902 *** scm_newptob has been removed
4906 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4908 - Function: SCM scm_make_port_type (char *type_name,
4909 int (*fill_buffer) (SCM port),
4910 void (*write_flush) (SCM port));
4912 Similarly to the new smob interface, there is a set of function
4913 setters by which the user can customize the behaviour of his port
4914 type. See ports.h (scm_set_port_XXX).
4916 ** scm_strport_to_string: New function: creates a new string from
4917 a string port's buffer.
4919 ** Plug in interface for random number generators
4920 The variable `scm_the_rng' in random.c contains a value and three
4921 function pointers which together define the current random number
4922 generator being used by the Scheme level interface and the random
4923 number library functions.
4925 The user is free to replace the default generator with the generator
4928 *** Variable: size_t scm_the_rng.rstate_size
4929 The size of the random state type used by the current RNG
4932 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4933 Given the random STATE, return 32 random bits.
4935 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4936 Seed random state STATE using string S of length N.
4938 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4939 Given random state STATE, return a malloced copy.
4942 The default RNG is the MWC (Multiply With Carry) random number
4943 generator described by George Marsaglia at the Department of
4944 Statistics and Supercomputer Computations Research Institute, The
4945 Florida State University (http://stat.fsu.edu/~geo).
4947 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4948 passes all tests in the DIEHARD test suite
4949 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4950 costs one multiply and one add on platforms which either supports long
4951 longs (gcc does this on most systems) or have 64 bit longs. The cost
4952 is four multiply on other systems but this can be optimized by writing
4953 scm_i_uniform32 in assembler.
4955 These functions are provided through the scm_the_rng interface for use
4956 by libguile and the application.
4958 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4959 Given the random STATE, return 32 random bits.
4960 Don't use this function directly. Instead go through the plugin
4961 interface (see "Plug in interface" above).
4963 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4964 Initialize STATE using SEED of length N.
4966 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4967 Return a malloc:ed copy of STATE. This function can easily be re-used
4968 in the interfaces to other RNGs.
4970 ** Random number library functions
4971 These functions use the current RNG through the scm_the_rng interface.
4972 It might be a good idea to use these functions from your C code so
4973 that only one random generator is used by all code in your program.
4975 The default random state is stored in:
4977 *** Variable: SCM scm_var_random_state
4978 Contains the vcell of the Scheme variable "*random-state*" which is
4979 used as default state by all random number functions in the Scheme
4984 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4986 *** Function: scm_rstate *scm_c_default_rstate (void)
4987 This is a convenience function which returns the value of
4988 scm_var_random_state. An error message is generated if this value
4989 isn't a random state.
4991 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4992 Make a new random state from the string SEED of length LENGTH.
4994 It is generally not a good idea to use multiple random states in a
4995 program. While subsequent random numbers generated from one random
4996 state are guaranteed to be reasonably independent, there is no such
4997 guarantee for numbers generated from different random states.
4999 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
5000 Return 32 random bits.
5002 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
5003 Return a sample from the uniform(0,1) distribution.
5005 *** Function: double scm_c_normal01 (scm_rstate *STATE)
5006 Return a sample from the normal(0,1) distribution.
5008 *** Function: double scm_c_exp1 (scm_rstate *STATE)
5009 Return a sample from the exp(1) distribution.
5011 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
5012 Return a sample from the discrete uniform(0,M) distribution.
5014 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
5015 Return a sample from the discrete uniform(0,M) distribution.
5016 M must be a bignum object. The returned value may be an INUM.
5020 Changes in Guile 1.3 (released Monday, October 19, 1998):
5022 * Changes to the distribution
5024 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
5025 To avoid conflicts, programs should name environment variables after
5026 themselves, except when there's a common practice establishing some
5029 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
5030 giving the former precedence, and printing a warning message if the
5031 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
5033 ** The header files related to multi-byte characters have been removed.
5034 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
5035 which referred to these explicitly will probably need to be rewritten,
5036 since the support for the variant string types has been removed; see
5039 ** The header files append.h and sequences.h have been removed. These
5040 files implemented non-R4RS operations which would encourage
5041 non-portable programming style and less easy-to-read code.
5043 * Changes to the stand-alone interpreter
5045 ** New procedures have been added to implement a "batch mode":
5047 *** Function: batch-mode?
5049 Returns a boolean indicating whether the interpreter is in batch
5052 *** Function: set-batch-mode?! ARG
5054 If ARG is true, switches the interpreter to batch mode. The `#f'
5055 case has not been implemented.
5057 ** Guile now provides full command-line editing, when run interactively.
5058 To use this feature, you must have the readline library installed.
5059 The Guile build process will notice it, and automatically include
5062 The readline library is available via anonymous FTP from any GNU
5063 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
5065 ** the-last-stack is now a fluid.
5067 * Changes to the procedure for linking libguile with your programs
5069 ** You can now use the `guile-config' utility to build programs that use Guile.
5071 Guile now includes a command-line utility called `guile-config', which
5072 can provide information about how to compile and link programs that
5075 *** `guile-config compile' prints any C compiler flags needed to use Guile.
5076 You should include this command's output on the command line you use
5077 to compile C or C++ code that #includes the Guile header files. It's
5078 usually just a `-I' flag to help the compiler find the Guile headers.
5081 *** `guile-config link' prints any linker flags necessary to link with Guile.
5083 This command writes to its standard output a list of flags which you
5084 must pass to the linker to link your code against the Guile library.
5085 The flags include '-lguile' itself, any other libraries the Guile
5086 library depends upon, and any `-L' flags needed to help the linker
5087 find those libraries.
5089 For example, here is a Makefile rule that builds a program named 'foo'
5090 from the object files ${FOO_OBJECTS}, and links them against Guile:
5093 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
5095 Previous Guile releases recommended that you use autoconf to detect
5096 which of a predefined set of libraries were present on your system.
5097 It is more robust to use `guile-config', since it records exactly which
5098 libraries the installed Guile library requires.
5100 This was originally called `build-guile', but was renamed to
5101 `guile-config' before Guile 1.3 was released, to be consistent with
5102 the analogous script for the GTK+ GUI toolkit, which is called
5106 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
5108 If you are using the GNU autoconf package to configure your program,
5109 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
5110 (described above) and gather the necessary values for use in your
5113 The GUILE_FLAGS macro expands to configure script code which runs the
5114 `guile-config' script, to find out where Guile's header files and
5115 libraries are installed. It sets two variables, marked for
5116 substitution, as by AC_SUBST.
5118 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
5119 code that uses Guile header files. This is almost always just a
5122 GUILE_LDFLAGS --- flags to pass to the linker to link a
5123 program against Guile. This includes `-lguile' for the Guile
5124 library itself, any libraries that Guile itself requires (like
5125 -lqthreads), and so on. It may also include a -L flag to tell the
5126 compiler where to find the libraries.
5128 GUILE_FLAGS is defined in the file guile.m4, in the top-level
5129 directory of the Guile distribution. You can copy it into your
5130 package's aclocal.m4 file, and then use it in your configure.in file.
5132 If you are using the `aclocal' program, distributed with GNU automake,
5133 to maintain your aclocal.m4 file, the Guile installation process
5134 installs guile.m4 where aclocal will find it. All you need to do is
5135 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
5136 this will copy the definition of GUILE_FLAGS into your aclocal.m4
5140 * Changes to Scheme functions and syntax
5142 ** Multi-byte strings have been removed, as have multi-byte and wide
5143 ports. We felt that these were the wrong approach to
5144 internationalization support.
5146 ** New function: readline [PROMPT]
5147 Read a line from the terminal, and allow the user to edit it,
5148 prompting with PROMPT. READLINE provides a large set of Emacs-like
5149 editing commands, lets the user recall previously typed lines, and
5150 works on almost every kind of terminal, including dumb terminals.
5152 READLINE assumes that the cursor is at the beginning of the line when
5153 it is invoked. Thus, you can't print a prompt yourself, and then call
5154 READLINE; you need to package up your prompt as a string, pass it to
5155 the function, and let READLINE print the prompt itself. This is
5156 because READLINE needs to know the prompt's screen width.
5158 For Guile to provide this function, you must have the readline
5159 library, version 2.1 or later, installed on your system. Readline is
5160 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
5161 any GNU mirror site.
5163 See also ADD-HISTORY function.
5165 ** New function: add-history STRING
5166 Add STRING as the most recent line in the history used by the READLINE
5167 command. READLINE does not add lines to the history itself; you must
5168 call ADD-HISTORY to make previous input available to the user.
5170 ** The behavior of the read-line function has changed.
5172 This function now uses standard C library functions to read the line,
5173 for speed. This means that it doesn not respect the value of
5174 scm-line-incrementors; it assumes that lines are delimited with
5177 (Note that this is read-line, the function that reads a line of text
5178 from a port, not readline, the function that reads a line from a
5179 terminal, providing full editing capabilities.)
5181 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
5183 This module provides some simple argument parsing. It exports one
5186 Function: getopt-gnu-style ARG-LS
5187 Parse a list of program arguments into an alist of option
5190 Each item in the list of program arguments is examined to see if
5191 it meets the syntax of a GNU long-named option. An argument like
5192 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
5193 returned alist, where MUMBLE is a keyword object with the same
5194 name as the argument. An argument like `--MUMBLE=FROB' produces
5195 an element of the form (MUMBLE . FROB), where FROB is a string.
5197 As a special case, the returned alist also contains a pair whose
5198 car is the symbol `rest'. The cdr of this pair is a list
5199 containing all the items in the argument list that are not options
5200 of the form mentioned above.
5202 The argument `--' is treated specially: all items in the argument
5203 list appearing after such an argument are not examined, and are
5204 returned in the special `rest' list.
5206 This function does not parse normal single-character switches.
5207 You will need to parse them out of the `rest' list yourself.
5209 ** The read syntax for byte vectors and short vectors has changed.
5211 Instead of #bytes(...), write #y(...).
5213 Instead of #short(...), write #h(...).
5215 This may seem nutty, but, like the other uniform vectors, byte vectors
5216 and short vectors want to have the same print and read syntax (and,
5217 more basic, want to have read syntax!). Changing the read syntax to
5218 use multiple characters after the hash sign breaks with the
5219 conventions used in R5RS and the conventions used for the other
5220 uniform vectors. It also introduces complexity in the current reader,
5221 both on the C and Scheme levels. (The Right solution is probably to
5222 change the syntax and prototypes for uniform vectors entirely.)
5225 ** The new module (ice-9 session) provides useful interactive functions.
5227 *** New procedure: (apropos REGEXP OPTION ...)
5229 Display a list of top-level variables whose names match REGEXP, and
5230 the modules they are imported from. Each OPTION should be one of the
5233 value --- Show the value of each matching variable.
5234 shadow --- Show bindings shadowed by subsequently imported modules.
5235 full --- Same as both `shadow' and `value'.
5239 guile> (apropos "trace" 'full)
5240 debug: trace #<procedure trace args>
5241 debug: untrace #<procedure untrace args>
5242 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
5243 the-scm-module: before-backtrace-hook ()
5244 the-scm-module: backtrace #<primitive-procedure backtrace>
5245 the-scm-module: after-backtrace-hook ()
5246 the-scm-module: has-shown-backtrace-hint? #f
5249 ** There are new functions and syntax for working with macros.
5251 Guile implements macros as a special object type. Any variable whose
5252 top-level binding is a macro object acts as a macro. The macro object
5253 specifies how the expression should be transformed before evaluation.
5255 *** Macro objects now print in a reasonable way, resembling procedures.
5257 *** New function: (macro? OBJ)
5258 True iff OBJ is a macro object.
5260 *** New function: (primitive-macro? OBJ)
5261 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
5262 macro transformers, implemented in eval.c rather than Scheme code.
5264 Why do we have this function?
5265 - For symmetry with procedure? and primitive-procedure?,
5266 - to allow custom print procedures to tell whether a macro is
5267 primitive, and display it differently, and
5268 - to allow compilers and user-written evaluators to distinguish
5269 builtin special forms from user-defined ones, which could be
5272 *** New function: (macro-type OBJ)
5273 Return a value indicating what kind of macro OBJ is. Possible return
5276 The symbol `syntax' --- a macro created by procedure->syntax.
5277 The symbol `macro' --- a macro created by procedure->macro.
5278 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
5279 The boolean #f --- if OBJ is not a macro object.
5281 *** New function: (macro-name MACRO)
5282 Return the name of the macro object MACRO's procedure, as returned by
5285 *** New function: (macro-transformer MACRO)
5286 Return the transformer procedure for MACRO.
5288 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
5290 Specify a new macro expander to use in the current module. Each
5291 MODULE is a module name, with the same meaning as in the `use-modules'
5292 form; each named module's exported bindings are added to the current
5293 top-level environment. TRANSFORMER is an expression evaluated in the
5294 resulting environment which must yield a procedure to use as the
5295 module's eval transformer: every expression evaluated in this module
5296 is passed to this function, and the result passed to the Guile
5299 *** macro-eval! is removed. Use local-eval instead.
5301 ** Some magic has been added to the printer to better handle user
5302 written printing routines (like record printers, closure printers).
5304 The problem is that these user written routines must have access to
5305 the current `print-state' to be able to handle fancy things like
5306 detection of circular references. These print-states have to be
5307 passed to the builtin printing routines (display, write, etc) to
5308 properly continue the print chain.
5310 We didn't want to change all existing print code so that it
5311 explicitly passes thru a print state in addition to a port. Instead,
5312 we extented the possible values that the builtin printing routines
5313 accept as a `port'. In addition to a normal port, they now also take
5314 a pair of a normal port and a print-state. Printing will go to the
5315 port and the print-state will be used to control the detection of
5316 circular references, etc. If the builtin function does not care for a
5317 print-state, it is simply ignored.
5319 User written callbacks are now called with such a pair as their
5320 `port', but because every function now accepts this pair as a PORT
5321 argument, you don't have to worry about that. In fact, it is probably
5322 safest to not check for these pairs.
5324 However, it is sometimes necessary to continue a print chain on a
5325 different port, for example to get a intermediate string
5326 representation of the printed value, mangle that string somehow, and
5327 then to finally print the mangled string. Use the new function
5329 inherit-print-state OLD-PORT NEW-PORT
5331 for this. It constructs a new `port' that prints to NEW-PORT but
5332 inherits the print-state of OLD-PORT.
5334 ** struct-vtable-offset renamed to vtable-offset-user
5336 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
5338 ** There is now a third optional argument to make-vtable-vtable
5339 (and fourth to make-struct) when constructing new types (vtables).
5340 This argument initializes field vtable-index-printer of the vtable.
5342 ** The detection of circular references has been extended to structs.
5343 That is, a structure that -- in the process of being printed -- prints
5344 itself does not lead to infinite recursion.
5346 ** There is now some basic support for fluids. Please read
5347 "libguile/fluid.h" to find out more. It is accessible from Scheme with
5348 the following functions and macros:
5350 Function: make-fluid
5352 Create a new fluid object. Fluids are not special variables or
5353 some other extension to the semantics of Scheme, but rather
5354 ordinary Scheme objects. You can store them into variables (that
5355 are still lexically scoped, of course) or into any other place you
5356 like. Every fluid has a initial value of `#f'.
5358 Function: fluid? OBJ
5360 Test whether OBJ is a fluid.
5362 Function: fluid-ref FLUID
5363 Function: fluid-set! FLUID VAL
5365 Access/modify the fluid FLUID. Modifications are only visible
5366 within the current dynamic root (that includes threads).
5368 Function: with-fluids* FLUIDS VALUES THUNK
5370 FLUIDS is a list of fluids and VALUES a corresponding list of
5371 values for these fluids. Before THUNK gets called the values are
5372 installed in the fluids and the old values of the fluids are
5373 saved in the VALUES list. When the flow of control leaves THUNK
5374 or reenters it, the values get swapped again. You might think of
5375 this as a `safe-fluid-excursion'. Note that the VALUES list is
5376 modified by `with-fluids*'.
5378 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
5380 The same as `with-fluids*' but with a different syntax. It looks
5381 just like `let', but both FLUID and VALUE are evaluated. Remember,
5382 fluids are not special variables but ordinary objects. FLUID
5383 should evaluate to a fluid.
5385 ** Changes to system call interfaces:
5387 *** close-port, close-input-port and close-output-port now return a
5388 boolean instead of an `unspecified' object. #t means that the port
5389 was successfully closed, while #f means it was already closed. It is
5390 also now possible for these procedures to raise an exception if an
5391 error occurs (some errors from write can be delayed until close.)
5393 *** the first argument to chmod, fcntl, ftell and fseek can now be a
5396 *** the third argument to fcntl is now optional.
5398 *** the first argument to chown can now be a file descriptor or a port.
5400 *** the argument to stat can now be a port.
5402 *** The following new procedures have been added (most use scsh
5405 *** procedure: close PORT/FD
5406 Similar to close-port (*note close-port: Closing Ports.), but also
5407 works on file descriptors. A side effect of closing a file
5408 descriptor is that any ports using that file descriptor are moved
5409 to a different file descriptor and have their revealed counts set
5412 *** procedure: port->fdes PORT
5413 Returns the integer file descriptor underlying PORT. As a side
5414 effect the revealed count of PORT is incremented.
5416 *** procedure: fdes->ports FDES
5417 Returns a list of existing ports which have FDES as an underlying
5418 file descriptor, without changing their revealed counts.
5420 *** procedure: fdes->inport FDES
5421 Returns an existing input port which has FDES as its underlying
5422 file descriptor, if one exists, and increments its revealed count.
5423 Otherwise, returns a new input port with a revealed count of 1.
5425 *** procedure: fdes->outport FDES
5426 Returns an existing output port which has FDES as its underlying
5427 file descriptor, if one exists, and increments its revealed count.
5428 Otherwise, returns a new output port with a revealed count of 1.
5430 The next group of procedures perform a `dup2' system call, if NEWFD
5431 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
5432 duplicated can be supplied as an integer or contained in a port. The
5433 type of value returned varies depending on which procedure is used.
5435 All procedures also have the side effect when performing `dup2' that
5436 any ports using NEWFD are moved to a different file descriptor and have
5437 their revealed counts set to zero.
5439 *** procedure: dup->fdes PORT/FD [NEWFD]
5440 Returns an integer file descriptor.
5442 *** procedure: dup->inport PORT/FD [NEWFD]
5443 Returns a new input port using the new file descriptor.
5445 *** procedure: dup->outport PORT/FD [NEWFD]
5446 Returns a new output port using the new file descriptor.
5448 *** procedure: dup PORT/FD [NEWFD]
5449 Returns a new port if PORT/FD is a port, with the same mode as the
5450 supplied port, otherwise returns an integer file descriptor.
5452 *** procedure: dup->port PORT/FD MODE [NEWFD]
5453 Returns a new port using the new file descriptor. MODE supplies a
5454 mode string for the port (*note open-file: File Ports.).
5456 *** procedure: setenv NAME VALUE
5457 Modifies the environment of the current process, which is also the
5458 default environment inherited by child processes.
5460 If VALUE is `#f', then NAME is removed from the environment.
5461 Otherwise, the string NAME=VALUE is added to the environment,
5462 replacing any existing string with name matching NAME.
5464 The return value is unspecified.
5466 *** procedure: truncate-file OBJ SIZE
5467 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
5468 can be a string containing a file name or an integer file
5469 descriptor or port open for output on the file. The underlying
5470 system calls are `truncate' and `ftruncate'.
5472 The return value is unspecified.
5474 *** procedure: setvbuf PORT MODE [SIZE]
5475 Set the buffering mode for PORT. MODE can be:
5483 block buffered, using a newly allocated buffer of SIZE bytes.
5484 However if SIZE is zero or unspecified, the port will be made
5487 This procedure should not be used after I/O has been performed with
5490 Ports are usually block buffered by default, with a default buffer
5491 size. Procedures e.g., *Note open-file: File Ports, which accept a
5492 mode string allow `0' to be added to request an unbuffered port.
5494 *** procedure: fsync PORT/FD
5495 Copies any unwritten data for the specified output file descriptor
5496 to disk. If PORT/FD is a port, its buffer is flushed before the
5497 underlying file descriptor is fsync'd. The return value is
5500 *** procedure: open-fdes PATH FLAGS [MODES]
5501 Similar to `open' but returns a file descriptor instead of a port.
5503 *** procedure: execle PATH ENV [ARG] ...
5504 Similar to `execl', but the environment of the new process is
5505 specified by ENV, which must be a list of strings as returned by
5506 the `environ' procedure.
5508 This procedure is currently implemented using the `execve' system
5509 call, but we call it `execle' because of its Scheme calling
5512 *** procedure: strerror ERRNO
5513 Returns the Unix error message corresponding to ERRNO, an integer.
5515 *** procedure: primitive-exit [STATUS]
5516 Terminate the current process without unwinding the Scheme stack.
5517 This is would typically be useful after a fork. The exit status
5518 is STATUS if supplied, otherwise zero.
5520 *** procedure: times
5521 Returns an object with information about real and processor time.
5522 The following procedures accept such an object as an argument and
5523 return a selected component:
5526 The current real time, expressed as time units relative to an
5530 The CPU time units used by the calling process.
5533 The CPU time units used by the system on behalf of the
5537 The CPU time units used by terminated child processes of the
5538 calling process, whose status has been collected (e.g., using
5542 Similarly, the CPU times units used by the system on behalf of
5543 terminated child processes.
5545 ** Removed: list-length
5546 ** Removed: list-append, list-append!
5547 ** Removed: list-reverse, list-reverse!
5549 ** array-map renamed to array-map!
5551 ** serial-array-map renamed to serial-array-map!
5553 ** catch doesn't take #f as first argument any longer
5555 Previously, it was possible to pass #f instead of a key to `catch'.
5556 That would cause `catch' to pass a jump buffer object to the procedure
5557 passed as second argument. The procedure could then use this jump
5558 buffer objekt as an argument to throw.
5560 This mechanism has been removed since its utility doesn't motivate the
5561 extra complexity it introduces.
5563 ** The `#/' notation for lists now provokes a warning message from Guile.
5564 This syntax will be removed from Guile in the near future.
5566 To disable the warning message, set the GUILE_HUSH environment
5567 variable to any non-empty value.
5569 ** The newline character now prints as `#\newline', following the
5570 normal Scheme notation, not `#\nl'.
5572 * Changes to the gh_ interface
5574 ** The gh_enter function now takes care of loading the Guile startup files.
5575 gh_enter works by calling scm_boot_guile; see the remarks below.
5577 ** Function: void gh_write (SCM x)
5579 Write the printed representation of the scheme object x to the current
5580 output port. Corresponds to the scheme level `write'.
5582 ** gh_list_length renamed to gh_length.
5584 ** vector handling routines
5586 Several major changes. In particular, gh_vector() now resembles
5587 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
5588 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
5589 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
5590 vector-related gh_ functions have been implemented.
5592 ** pair and list routines
5594 Implemented several of the R4RS pair and list functions that were
5597 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
5599 New function. Converts double arrays back and forth between Scheme
5602 * Changes to the scm_ interface
5604 ** The function scm_boot_guile now takes care of loading the startup files.
5606 Guile's primary initialization function, scm_boot_guile, now takes
5607 care of loading `boot-9.scm', in the `ice-9' module, to initialize
5608 Guile, define the module system, and put together some standard
5609 bindings. It also loads `init.scm', which is intended to hold
5610 site-specific initialization code.
5612 Since Guile cannot operate properly until boot-9.scm is loaded, there
5613 is no reason to separate loading boot-9.scm from Guile's other
5614 initialization processes.
5616 This job used to be done by scm_compile_shell_switches, which didn't
5617 make much sense; in particular, it meant that people using Guile for
5618 non-shell-like applications had to jump through hoops to get Guile
5619 initialized properly.
5621 ** The function scm_compile_shell_switches no longer loads the startup files.
5622 Now, Guile always loads the startup files, whenever it is initialized;
5623 see the notes above for scm_boot_guile and scm_load_startup_files.
5625 ** Function: scm_load_startup_files
5626 This new function takes care of loading Guile's initialization file
5627 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
5628 this is always called by the Guile initialization process, it's
5629 probably not too useful to call this yourself, but it's there anyway.
5631 ** The semantics of smob marking have changed slightly.
5633 The smob marking function (the `mark' member of the scm_smobfuns
5634 structure) is no longer responsible for setting the mark bit on the
5635 smob. The generic smob handling code in the garbage collector will
5636 set this bit. The mark function need only ensure that any other
5637 objects the smob refers to get marked.
5639 Note that this change means that the smob's GC8MARK bit is typically
5640 already set upon entry to the mark function. Thus, marking functions
5641 which look like this:
5644 if (SCM_GC8MARKP (ptr))
5646 SCM_SETGC8MARK (ptr);
5647 ... mark objects to which the smob refers ...
5650 are now incorrect, since they will return early, and fail to mark any
5651 other objects the smob refers to. Some code in the Guile library used
5654 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
5656 If you have implemented your own I/O port type, by writing the
5657 functions required by the scm_ptobfuns and then calling scm_newptob,
5658 you will need to change your functions slightly.
5660 The functions in a scm_ptobfuns structure now expect the port itself
5661 as their argument; they used to expect the `stream' member of the
5662 port's scm_port_table structure. This allows functions in an
5663 scm_ptobfuns structure to easily access the port's cell (and any flags
5664 it its CAR), and the port's scm_port_table structure.
5666 Guile now passes the I/O port itself as the `port' argument in the
5667 following scm_ptobfuns functions:
5669 int (*free) (SCM port);
5670 int (*fputc) (int, SCM port);
5671 int (*fputs) (char *, SCM port);
5672 scm_sizet (*fwrite) SCM_P ((char *ptr,
5676 int (*fflush) (SCM port);
5677 int (*fgetc) (SCM port);
5678 int (*fclose) (SCM port);
5680 The interfaces to the `mark', `print', `equalp', and `fgets' methods
5683 If you have existing code which defines its own port types, it is easy
5684 to convert your code to the new interface; simply apply SCM_STREAM to
5685 the port argument to yield the value you code used to expect.
5687 Note that since both the port and the stream have the same type in the
5688 C code --- they are both SCM values --- the C compiler will not remind
5689 you if you forget to update your scm_ptobfuns functions.
5692 ** Function: int scm_internal_select (int fds,
5696 struct timeval *timeout);
5698 This is a replacement for the `select' function provided by the OS.
5699 It enables I/O blocking and sleeping to happen for one cooperative
5700 thread without blocking other threads. It also avoids busy-loops in
5701 these situations. It is intended that all I/O blocking and sleeping
5702 will finally go through this function. Currently, this function is
5703 only available on systems providing `gettimeofday' and `select'.
5705 ** Function: SCM scm_internal_stack_catch (SCM tag,
5706 scm_catch_body_t body,
5708 scm_catch_handler_t handler,
5711 A new sibling to the other two C level `catch' functions
5712 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
5713 the stack to be saved automatically into the variable `the-last-stack'
5714 (scm_the_last_stack_var) on error. This is necessary if you want to
5715 use advanced error reporting, such as calling scm_display_error and
5716 scm_display_backtrace. (They both take a stack object as argument.)
5718 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
5720 scm_catch_handler_t handler,
5723 Spawns a new thread. It does a job similar to
5724 scm_call_with_new_thread but takes arguments more suitable when
5725 spawning threads from application C code.
5727 ** The hook scm_error_callback has been removed. It was originally
5728 intended as a way for the user to install his own error handler. But
5729 that method works badly since it intervenes between throw and catch,
5730 thereby changing the semantics of expressions like (catch #t ...).
5731 The correct way to do it is to use one of the C level catch functions
5732 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
5734 ** Removed functions:
5736 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
5737 scm_list_reverse, scm_list_reverse_x
5739 ** New macros: SCM_LISTn where n is one of the integers 0-9.
5741 These can be used for pretty list creation from C. The idea is taken
5742 from Erick Gallesio's STk.
5744 ** scm_array_map renamed to scm_array_map_x
5746 ** mbstrings are now removed
5748 This means that the type codes scm_tc7_mb_string and
5749 scm_tc7_mb_substring has been removed.
5751 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
5753 Since we no longer support multi-byte strings, these I/O functions
5754 have been simplified, and renamed. Here are their old names, and
5755 their new names and arguments:
5757 scm_gen_putc -> void scm_putc (int c, SCM port);
5758 scm_gen_puts -> void scm_puts (char *s, SCM port);
5759 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
5760 scm_gen_getc -> void scm_getc (SCM port);
5763 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
5765 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
5767 SCM_TYP7S now masks away the bit which distinguishes substrings from
5770 ** scm_catch_body_t: Backward incompatible change!
5772 Body functions to scm_internal_catch and friends do not any longer
5773 take a second argument. This is because it is no longer possible to
5774 pass a #f arg to catch.
5776 ** Calls to scm_protect_object and scm_unprotect now nest properly.
5778 The function scm_protect_object protects its argument from being freed
5779 by the garbage collector. scm_unprotect_object removes that
5782 These functions now nest properly. That is, for every object O, there
5783 is a counter which scm_protect_object(O) increments and
5784 scm_unprotect_object(O) decrements, if the counter is greater than
5785 zero. Every object's counter is zero when it is first created. If an
5786 object's counter is greater than zero, the garbage collector will not
5787 reclaim its storage.
5789 This allows you to use scm_protect_object in your code without
5790 worrying that some other function you call will call
5791 scm_unprotect_object, and allow it to be freed. Assuming that the
5792 functions you call are well-behaved, and unprotect only those objects
5793 they protect, you can follow the same rule and have confidence that
5794 objects will be freed only at appropriate times.
5797 Changes in Guile 1.2 (released Tuesday, June 24 1997):
5799 * Changes to the distribution
5801 ** Nightly snapshots are now available from ftp.red-bean.com.
5802 The old server, ftp.cyclic.com, has been relinquished to its rightful
5805 Nightly snapshots of the Guile development sources are now available via
5806 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
5808 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
5809 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
5811 ** To run Guile without installing it, the procedure has changed a bit.
5813 If you used a separate build directory to compile Guile, you'll need
5814 to include the build directory in SCHEME_LOAD_PATH, as well as the
5815 source directory. See the `INSTALL' file for examples.
5817 * Changes to the procedure for linking libguile with your programs
5819 ** The standard Guile load path for Scheme code now includes
5820 $(datadir)/guile (usually /usr/local/share/guile). This means that
5821 you can install your own Scheme files there, and Guile will find them.
5822 (Previous versions of Guile only checked a directory whose name
5823 contained the Guile version number, so you had to re-install or move
5824 your Scheme sources each time you installed a fresh version of Guile.)
5826 The load path also includes $(datadir)/guile/site; we recommend
5827 putting individual Scheme files there. If you want to install a
5828 package with multiple source files, create a directory for them under
5831 ** Guile 1.2 will now use the Rx regular expression library, if it is
5832 installed on your system. When you are linking libguile into your own
5833 programs, this means you will have to link against -lguile, -lqt (if
5834 you configured Guile with thread support), and -lrx.
5836 If you are using autoconf to generate configuration scripts for your
5837 application, the following lines should suffice to add the appropriate
5838 libraries to your link command:
5840 ### Find Rx, quickthreads and libguile.
5841 AC_CHECK_LIB(rx, main)
5842 AC_CHECK_LIB(qt, main)
5843 AC_CHECK_LIB(guile, scm_shell)
5845 The Guile 1.2 distribution does not contain sources for the Rx
5846 library, as Guile 1.0 did. If you want to use Rx, you'll need to
5847 retrieve it from a GNU FTP site and install it separately.
5849 * Changes to Scheme functions and syntax
5851 ** The dynamic linking features of Guile are now enabled by default.
5852 You can disable them by giving the `--disable-dynamic-linking' option
5855 (dynamic-link FILENAME)
5857 Find the object file denoted by FILENAME (a string) and link it
5858 into the running Guile application. When everything works out,
5859 return a Scheme object suitable for representing the linked object
5860 file. Otherwise an error is thrown. How object files are
5861 searched is system dependent.
5863 (dynamic-object? VAL)
5865 Determine whether VAL represents a dynamically linked object file.
5867 (dynamic-unlink DYNOBJ)
5869 Unlink the indicated object file from the application. DYNOBJ
5870 should be one of the values returned by `dynamic-link'.
5872 (dynamic-func FUNCTION DYNOBJ)
5874 Search the C function indicated by FUNCTION (a string or symbol)
5875 in DYNOBJ and return some Scheme object that can later be used
5876 with `dynamic-call' to actually call this function. Right now,
5877 these Scheme objects are formed by casting the address of the
5878 function to `long' and converting this number to its Scheme
5881 (dynamic-call FUNCTION DYNOBJ)
5883 Call the C function indicated by FUNCTION and DYNOBJ. The
5884 function is passed no arguments and its return value is ignored.
5885 When FUNCTION is something returned by `dynamic-func', call that
5886 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5887 etc.), look it up in DYNOBJ; this is equivalent to
5889 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5891 Interrupts are deferred while the C function is executing (with
5892 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5894 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5896 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5897 some arguments and return its return value. The C function is
5898 expected to take two arguments and return an `int', just like
5901 int c_func (int argc, char **argv);
5903 ARGS must be a list of strings and is converted into an array of
5904 `char *'. The array is passed in ARGV and its size in ARGC. The
5905 return value is converted to a Scheme number and returned from the
5906 call to `dynamic-args-call'.
5908 When dynamic linking is disabled or not supported on your system,
5909 the above functions throw errors, but they are still available.
5911 Here is a small example that works on GNU/Linux:
5913 (define libc-obj (dynamic-link "libc.so"))
5914 (dynamic-args-call 'rand libc-obj '())
5916 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5918 ** The #/ syntax for module names is depreciated, and will be removed
5919 in a future version of Guile. Instead of
5927 The latter syntax is more consistent with existing Lisp practice.
5929 ** Guile now does fancier printing of structures. Structures are the
5930 underlying implementation for records, which in turn are used to
5931 implement modules, so all of these object now print differently and in
5932 a more informative way.
5934 The Scheme printer will examine the builtin variable *struct-printer*
5935 whenever it needs to print a structure object. When this variable is
5936 not `#f' it is deemed to be a procedure and will be applied to the
5937 structure object and the output port. When *struct-printer* is `#f'
5938 or the procedure return `#f' the structure object will be printed in
5939 the boring #<struct 80458270> form.
5941 This hook is used by some routines in ice-9/boot-9.scm to implement
5942 type specific printing routines. Please read the comments there about
5945 One of the more specific uses of structs are records. The printing
5946 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5947 called. It should behave like a *struct-printer* procedure (described
5950 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5951 token of the form #:NAME, where NAME has the same syntax as a Scheme
5952 symbol, is the external representation of the keyword named NAME.
5953 Keyword objects print using this syntax as well, so values containing
5954 keyword objects can be read back into Guile. When used in an
5955 expression, keywords are self-quoting objects.
5957 Guile suports this read syntax, and uses this print syntax, regardless
5958 of the current setting of the `keyword' read option. The `keyword'
5959 read option only controls whether Guile recognizes the `:NAME' syntax,
5960 which is incompatible with R4RS. (R4RS says such token represent
5963 ** Guile has regular expression support again. Guile 1.0 included
5964 functions for matching regular expressions, based on the Rx library.
5965 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5966 distribution, and thus Guile had no regular expression support. Guile
5967 1.2 again supports the most commonly used functions, and supports all
5968 of SCSH's regular expression functions.
5970 If your system does not include a POSIX regular expression library,
5971 and you have not linked Guile with a third-party regexp library such as
5972 Rx, these functions will not be available. You can tell whether your
5973 Guile installation includes regular expression support by checking
5974 whether the `*features*' list includes the `regex' symbol.
5976 *** regexp functions
5978 By default, Guile supports POSIX extended regular expressions. That
5979 means that the characters `(', `)', `+' and `?' are special, and must
5980 be escaped if you wish to match the literal characters.
5982 This regular expression interface was modeled after that implemented
5983 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5984 with SCSH regular expressions.
5986 **** Function: string-match PATTERN STR [START]
5987 Compile the string PATTERN into a regular expression and compare
5988 it with STR. The optional numeric argument START specifies the
5989 position of STR at which to begin matching.
5991 `string-match' returns a "match structure" which describes what,
5992 if anything, was matched by the regular expression. *Note Match
5993 Structures::. If STR does not match PATTERN at all,
5994 `string-match' returns `#f'.
5996 Each time `string-match' is called, it must compile its PATTERN
5997 argument into a regular expression structure. This operation is
5998 expensive, which makes `string-match' inefficient if the same regular
5999 expression is used several times (for example, in a loop). For better
6000 performance, you can compile a regular expression in advance and then
6001 match strings against the compiled regexp.
6003 **** Function: make-regexp STR [FLAGS]
6004 Compile the regular expression described by STR, and return the
6005 compiled regexp structure. If STR does not describe a legal
6006 regular expression, `make-regexp' throws a
6007 `regular-expression-syntax' error.
6009 FLAGS may be the bitwise-or of one or more of the following:
6011 **** Constant: regexp/extended
6012 Use POSIX Extended Regular Expression syntax when interpreting
6013 STR. If not set, POSIX Basic Regular Expression syntax is used.
6014 If the FLAGS argument is omitted, we assume regexp/extended.
6016 **** Constant: regexp/icase
6017 Do not differentiate case. Subsequent searches using the
6018 returned regular expression will be case insensitive.
6020 **** Constant: regexp/newline
6021 Match-any-character operators don't match a newline.
6023 A non-matching list ([^...]) not containing a newline matches a
6026 Match-beginning-of-line operator (^) matches the empty string
6027 immediately after a newline, regardless of whether the FLAGS
6028 passed to regexp-exec contain regexp/notbol.
6030 Match-end-of-line operator ($) matches the empty string
6031 immediately before a newline, regardless of whether the FLAGS
6032 passed to regexp-exec contain regexp/noteol.
6034 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
6035 Match the compiled regular expression REGEXP against `str'. If
6036 the optional integer START argument is provided, begin matching
6037 from that position in the string. Return a match structure
6038 describing the results of the match, or `#f' if no match could be
6041 FLAGS may be the bitwise-or of one or more of the following:
6043 **** Constant: regexp/notbol
6044 The match-beginning-of-line operator always fails to match (but
6045 see the compilation flag regexp/newline above) This flag may be
6046 used when different portions of a string are passed to
6047 regexp-exec and the beginning of the string should not be
6048 interpreted as the beginning of the line.
6050 **** Constant: regexp/noteol
6051 The match-end-of-line operator always fails to match (but see the
6052 compilation flag regexp/newline above)
6054 **** Function: regexp? OBJ
6055 Return `#t' if OBJ is a compiled regular expression, or `#f'
6058 Regular expressions are commonly used to find patterns in one string
6059 and replace them with the contents of another string.
6061 **** Function: regexp-substitute PORT MATCH [ITEM...]
6062 Write to the output port PORT selected contents of the match
6063 structure MATCH. Each ITEM specifies what should be written, and
6064 may be one of the following arguments:
6066 * A string. String arguments are written out verbatim.
6068 * An integer. The submatch with that number is written.
6070 * The symbol `pre'. The portion of the matched string preceding
6071 the regexp match is written.
6073 * The symbol `post'. The portion of the matched string
6074 following the regexp match is written.
6076 PORT may be `#f', in which case nothing is written; instead,
6077 `regexp-substitute' constructs a string from the specified ITEMs
6080 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
6081 Similar to `regexp-substitute', but can be used to perform global
6082 substitutions on STR. Instead of taking a match structure as an
6083 argument, `regexp-substitute/global' takes two string arguments: a
6084 REGEXP string describing a regular expression, and a TARGET string
6085 which should be matched against this regular expression.
6087 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
6090 * A function may be supplied. When this function is called, it
6091 will be passed one argument: a match structure for a given
6092 regular expression match. It should return a string to be
6093 written out to PORT.
6095 * The `post' symbol causes `regexp-substitute/global' to recurse
6096 on the unmatched portion of STR. This *must* be supplied in
6097 order to perform global search-and-replace on STR; if it is
6098 not present among the ITEMs, then `regexp-substitute/global'
6099 will return after processing a single match.
6101 *** Match Structures
6103 A "match structure" is the object returned by `string-match' and
6104 `regexp-exec'. It describes which portion of a string, if any, matched
6105 the given regular expression. Match structures include: a reference to
6106 the string that was checked for matches; the starting and ending
6107 positions of the regexp match; and, if the regexp included any
6108 parenthesized subexpressions, the starting and ending positions of each
6111 In each of the regexp match functions described below, the `match'
6112 argument must be a match structure returned by a previous call to
6113 `string-match' or `regexp-exec'. Most of these functions return some
6114 information about the original target string that was matched against a
6115 regular expression; we will call that string TARGET for easy reference.
6117 **** Function: regexp-match? OBJ
6118 Return `#t' if OBJ is a match structure returned by a previous
6119 call to `regexp-exec', or `#f' otherwise.
6121 **** Function: match:substring MATCH [N]
6122 Return the portion of TARGET matched by subexpression number N.
6123 Submatch 0 (the default) represents the entire regexp match. If
6124 the regular expression as a whole matched, but the subexpression
6125 number N did not match, return `#f'.
6127 **** Function: match:start MATCH [N]
6128 Return the starting position of submatch number N.
6130 **** Function: match:end MATCH [N]
6131 Return the ending position of submatch number N.
6133 **** Function: match:prefix MATCH
6134 Return the unmatched portion of TARGET preceding the regexp match.
6136 **** Function: match:suffix MATCH
6137 Return the unmatched portion of TARGET following the regexp match.
6139 **** Function: match:count MATCH
6140 Return the number of parenthesized subexpressions from MATCH.
6141 Note that the entire regular expression match itself counts as a
6142 subexpression, and failed submatches are included in the count.
6144 **** Function: match:string MATCH
6145 Return the original TARGET string.
6147 *** Backslash Escapes
6149 Sometimes you will want a regexp to match characters like `*' or `$'
6150 exactly. For example, to check whether a particular string represents
6151 a menu entry from an Info node, it would be useful to match it against
6152 a regexp like `^* [^:]*::'. However, this won't work; because the
6153 asterisk is a metacharacter, it won't match the `*' at the beginning of
6154 the string. In this case, we want to make the first asterisk un-magic.
6156 You can do this by preceding the metacharacter with a backslash
6157 character `\'. (This is also called "quoting" the metacharacter, and
6158 is known as a "backslash escape".) When Guile sees a backslash in a
6159 regular expression, it considers the following glyph to be an ordinary
6160 character, no matter what special meaning it would ordinarily have.
6161 Therefore, we can make the above example work by changing the regexp to
6162 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
6163 to match only a single asterisk in the target string.
6165 Since the backslash is itself a metacharacter, you may force a
6166 regexp to match a backslash in the target string by preceding the
6167 backslash with itself. For example, to find variable references in a
6168 TeX program, you might want to find occurrences of the string `\let\'
6169 followed by any number of alphabetic characters. The regular expression
6170 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
6171 each match a single backslash in the target string.
6173 **** Function: regexp-quote STR
6174 Quote each special character found in STR with a backslash, and
6175 return the resulting string.
6177 *Very important:* Using backslash escapes in Guile source code (as
6178 in Emacs Lisp or C) can be tricky, because the backslash character has
6179 special meaning for the Guile reader. For example, if Guile encounters
6180 the character sequence `\n' in the middle of a string while processing
6181 Scheme code, it replaces those characters with a newline character.
6182 Similarly, the character sequence `\t' is replaced by a horizontal tab.
6183 Several of these "escape sequences" are processed by the Guile reader
6184 before your code is executed. Unrecognized escape sequences are
6185 ignored: if the characters `\*' appear in a string, they will be
6186 translated to the single character `*'.
6188 This translation is obviously undesirable for regular expressions,
6189 since we want to be able to include backslashes in a string in order to
6190 escape regexp metacharacters. Therefore, to make sure that a backslash
6191 is preserved in a string in your Guile program, you must use *two*
6192 consecutive backslashes:
6194 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
6196 The string in this example is preprocessed by the Guile reader before
6197 any code is executed. The resulting argument to `make-regexp' is the
6198 string `^\* [^:]*', which is what we really want.
6200 This also means that in order to write a regular expression that
6201 matches a single backslash character, the regular expression string in
6202 the source code must include *four* backslashes. Each consecutive pair
6203 of backslashes gets translated by the Guile reader to a single
6204 backslash, and the resulting double-backslash is interpreted by the
6205 regexp engine as matching a single backslash character. Hence:
6207 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
6209 The reason for the unwieldiness of this syntax is historical. Both
6210 regular expression pattern matchers and Unix string processing systems
6211 have traditionally used backslashes with the special meanings described
6212 above. The POSIX regular expression specification and ANSI C standard
6213 both require these semantics. Attempting to abandon either convention
6214 would cause other kinds of compatibility problems, possibly more severe
6215 ones. Therefore, without extending the Scheme reader to support
6216 strings with different quoting conventions (an ungainly and confusing
6217 extension when implemented in other languages), we must adhere to this
6218 cumbersome escape syntax.
6220 * Changes to the gh_ interface
6222 * Changes to the scm_ interface
6224 * Changes to system call interfaces:
6226 ** The value returned by `raise' is now unspecified. It throws an exception
6229 *** A new procedure `sigaction' can be used to install signal handlers
6231 (sigaction signum [action] [flags])
6233 signum is the signal number, which can be specified using the value
6236 If action is omitted, sigaction returns a pair: the CAR is the current
6237 signal hander, which will be either an integer with the value SIG_DFL
6238 (default action) or SIG_IGN (ignore), or the Scheme procedure which
6239 handles the signal, or #f if a non-Scheme procedure handles the
6240 signal. The CDR contains the current sigaction flags for the handler.
6242 If action is provided, it is installed as the new handler for signum.
6243 action can be a Scheme procedure taking one argument, or the value of
6244 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
6245 whatever signal handler was installed before sigaction was first used.
6246 Flags can optionally be specified for the new handler (SA_RESTART is
6247 always used if the system provides it, so need not be specified.) The
6248 return value is a pair with information about the old handler as
6251 This interface does not provide access to the "signal blocking"
6252 facility. Maybe this is not needed, since the thread support may
6253 provide solutions to the problem of consistent access to data
6256 *** A new procedure `flush-all-ports' is equivalent to running
6257 `force-output' on every port open for output.
6259 ** Guile now provides information on how it was built, via the new
6260 global variable, %guile-build-info. This variable records the values
6261 of the standard GNU makefile directory variables as an assocation
6262 list, mapping variable names (symbols) onto directory paths (strings).
6263 For example, to find out where the Guile link libraries were
6264 installed, you can say:
6266 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
6269 * Changes to the scm_ interface
6271 ** The new function scm_handle_by_message_noexit is just like the
6272 existing scm_handle_by_message function, except that it doesn't call
6273 exit to terminate the process. Instead, it prints a message and just
6274 returns #f. This might be a more appropriate catch-all handler for
6275 new dynamic roots and threads.
6278 Changes in Guile 1.1 (released Friday, May 16 1997):
6280 * Changes to the distribution.
6282 The Guile 1.0 distribution has been split up into several smaller
6284 guile-core --- the Guile interpreter itself.
6285 guile-tcltk --- the interface between the Guile interpreter and
6286 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
6287 is a toolkit for building graphical user interfaces.
6288 guile-rgx-ctax --- the interface between Guile and the Rx regular
6289 expression matcher, and the translator for the Ctax
6290 programming language. These are packaged together because the
6291 Ctax translator uses Rx to parse Ctax source code.
6293 This NEWS file describes the changes made to guile-core since the 1.0
6296 We no longer distribute the documentation, since it was either out of
6297 date, or incomplete. As soon as we have current documentation, we
6302 * Changes to the stand-alone interpreter
6304 ** guile now accepts command-line arguments compatible with SCSH, Olin
6305 Shivers' Scheme Shell.
6307 In general, arguments are evaluated from left to right, but there are
6308 exceptions. The following switches stop argument processing, and
6309 stash all remaining command-line arguments as the value returned by
6310 the (command-line) function.
6311 -s SCRIPT load Scheme source code from FILE, and exit
6312 -c EXPR evalute Scheme expression EXPR, and exit
6313 -- stop scanning arguments; run interactively
6315 The switches below are processed as they are encountered.
6316 -l FILE load Scheme source code from FILE
6317 -e FUNCTION after reading script, apply FUNCTION to
6318 command line arguments
6319 -ds do -s script at this point
6320 --emacs enable Emacs protocol (experimental)
6321 -h, --help display this help and exit
6322 -v, --version display version information and exit
6323 \ read arguments from following script lines
6325 So, for example, here is a Guile script named `ekko' (thanks, Olin)
6326 which re-implements the traditional "echo" command:
6328 #!/usr/local/bin/guile -s
6331 (map (lambda (arg) (display arg) (display " "))
6335 (main (command-line))
6337 Suppose we invoke this script as follows:
6339 ekko a speckled gecko
6341 Through the magic of Unix script processing (triggered by the `#!'
6342 token at the top of the file), /usr/local/bin/guile receives the
6343 following list of command-line arguments:
6345 ("-s" "./ekko" "a" "speckled" "gecko")
6347 Unix inserts the name of the script after the argument specified on
6348 the first line of the file (in this case, "-s"), and then follows that
6349 with the arguments given to the script. Guile loads the script, which
6350 defines the `main' function, and then applies it to the list of
6351 remaining command-line arguments, ("a" "speckled" "gecko").
6353 In Unix, the first line of a script file must take the following form:
6355 #!INTERPRETER ARGUMENT
6357 where INTERPRETER is the absolute filename of the interpreter
6358 executable, and ARGUMENT is a single command-line argument to pass to
6361 You may only pass one argument to the interpreter, and its length is
6362 limited. These restrictions can be annoying to work around, so Guile
6363 provides a general mechanism (borrowed from, and compatible with,
6364 SCSH) for circumventing them.
6366 If the ARGUMENT in a Guile script is a single backslash character,
6367 `\', Guile will open the script file, parse arguments from its second
6368 and subsequent lines, and replace the `\' with them. So, for example,
6369 here is another implementation of the `ekko' script:
6371 #!/usr/local/bin/guile \
6375 (for-each (lambda (arg) (display arg) (display " "))
6379 If the user invokes this script as follows:
6381 ekko a speckled gecko
6383 Unix expands this into
6385 /usr/local/bin/guile \ ekko a speckled gecko
6387 When Guile sees the `\' argument, it replaces it with the arguments
6388 read from the second line of the script, producing:
6390 /usr/local/bin/guile -e main -s ekko a speckled gecko
6392 This tells Guile to load the `ekko' script, and apply the function
6393 `main' to the argument list ("a" "speckled" "gecko").
6395 Here is how Guile parses the command-line arguments:
6396 - Each space character terminates an argument. This means that two
6397 spaces in a row introduce an empty-string argument.
6398 - The tab character is not permitted (unless you quote it with the
6399 backslash character, as described below), to avoid confusion.
6400 - The newline character terminates the sequence of arguments, and will
6401 also terminate a final non-empty argument. (However, a newline
6402 following a space will not introduce a final empty-string argument;
6403 it only terminates the argument list.)
6404 - The backslash character is the escape character. It escapes
6405 backslash, space, tab, and newline. The ANSI C escape sequences
6406 like \n and \t are also supported. These produce argument
6407 constituents; the two-character combination \n doesn't act like a
6408 terminating newline. The escape sequence \NNN for exactly three
6409 octal digits reads as the character whose ASCII code is NNN. As
6410 above, characters produced this way are argument constituents.
6411 Backslash followed by other characters is not allowed.
6413 * Changes to the procedure for linking libguile with your programs
6415 ** Guile now builds and installs a shared guile library, if your
6416 system support shared libraries. (It still builds a static library on
6417 all systems.) Guile automatically detects whether your system
6418 supports shared libraries. To prevent Guile from buildisg shared
6419 libraries, pass the `--disable-shared' flag to the configure script.
6421 Guile takes longer to compile when it builds shared libraries, because
6422 it must compile every file twice --- once to produce position-
6423 independent object code, and once to produce normal object code.
6425 ** The libthreads library has been merged into libguile.
6427 To link a program against Guile, you now need only link against
6428 -lguile and -lqt; -lthreads is no longer needed. If you are using
6429 autoconf to generate configuration scripts for your application, the
6430 following lines should suffice to add the appropriate libraries to
6433 ### Find quickthreads and libguile.
6434 AC_CHECK_LIB(qt, main)
6435 AC_CHECK_LIB(guile, scm_shell)
6437 * Changes to Scheme functions
6439 ** Guile Scheme's special syntax for keyword objects is now optional,
6440 and disabled by default.
6442 The syntax variation from R4RS made it difficult to port some
6443 interesting packages to Guile. The routines which accepted keyword
6444 arguments (mostly in the module system) have been modified to also
6445 accept symbols whose names begin with `:'.
6447 To change the keyword syntax, you must first import the (ice-9 debug)
6449 (use-modules (ice-9 debug))
6451 Then you can enable the keyword syntax as follows:
6452 (read-set! keywords 'prefix)
6454 To disable keyword syntax, do this:
6455 (read-set! keywords #f)
6457 ** Many more primitive functions accept shared substrings as
6458 arguments. In the past, these functions required normal, mutable
6459 strings as arguments, although they never made use of this
6462 ** The uniform array functions now operate on byte vectors. These
6463 functions are `array-fill!', `serial-array-copy!', `array-copy!',
6464 `serial-array-map', `array-map', `array-for-each', and
6467 ** The new functions `trace' and `untrace' implement simple debugging
6468 support for Scheme functions.
6470 The `trace' function accepts any number of procedures as arguments,
6471 and tells the Guile interpreter to display each procedure's name and
6472 arguments each time the procedure is invoked. When invoked with no
6473 arguments, `trace' returns the list of procedures currently being
6476 The `untrace' function accepts any number of procedures as arguments,
6477 and tells the Guile interpreter not to trace them any more. When
6478 invoked with no arguments, `untrace' untraces all curretly traced
6481 The tracing in Guile has an advantage over most other systems: we
6482 don't create new procedure objects, but mark the procedure objects
6483 themselves. This means that anonymous and internal procedures can be
6486 ** The function `assert-repl-prompt' has been renamed to
6487 `set-repl-prompt!'. It takes one argument, PROMPT.
6488 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
6489 - If PROMPT is a string, we use it as a prompt.
6490 - If PROMPT is a procedure accepting no arguments, we call it, and
6491 display the result as a prompt.
6492 - Otherwise, we display "> ".
6494 ** The new function `eval-string' reads Scheme expressions from a
6495 string and evaluates them, returning the value of the last expression
6496 in the string. If the string contains no expressions, it returns an
6499 ** The new function `thunk?' returns true iff its argument is a
6500 procedure of zero arguments.
6502 ** `defined?' is now a builtin function, instead of syntax. This
6503 means that its argument should be quoted. It returns #t iff its
6504 argument is bound in the current module.
6506 ** The new syntax `use-modules' allows you to add new modules to your
6507 environment without re-typing a complete `define-module' form. It
6508 accepts any number of module names as arguments, and imports their
6509 public bindings into the current module.
6511 ** The new function (module-defined? NAME MODULE) returns true iff
6512 NAME, a symbol, is defined in MODULE, a module object.
6514 ** The new function `builtin-bindings' creates and returns a hash
6515 table containing copies of all the root module's bindings.
6517 ** The new function `builtin-weak-bindings' does the same as
6518 `builtin-bindings', but creates a doubly-weak hash table.
6520 ** The `equal?' function now considers variable objects to be
6521 equivalent if they have the same name and the same value.
6523 ** The new function `command-line' returns the command-line arguments
6524 given to Guile, as a list of strings.
6526 When using guile as a script interpreter, `command-line' returns the
6527 script's arguments; those processed by the interpreter (like `-s' or
6528 `-c') are omitted. (In other words, you get the normal, expected
6529 behavior.) Any application that uses scm_shell to process its
6530 command-line arguments gets this behavior as well.
6532 ** The new function `load-user-init' looks for a file called `.guile'
6533 in the user's home directory, and loads it if it exists. This is
6534 mostly for use by the code generated by scm_compile_shell_switches,
6535 but we thought it might also be useful in other circumstances.
6537 ** The new function `log10' returns the base-10 logarithm of its
6540 ** Changes to I/O functions
6542 *** The functions `read', `primitive-load', `read-and-eval!', and
6543 `primitive-load-path' no longer take optional arguments controlling
6544 case insensitivity and a `#' parser.
6546 Case sensitivity is now controlled by a read option called
6547 `case-insensitive'. The user can add new `#' syntaxes with the
6548 `read-hash-extend' function (see below).
6550 *** The new function `read-hash-extend' allows the user to change the
6551 syntax of Guile Scheme in a somewhat controlled way.
6553 (read-hash-extend CHAR PROC)
6554 When parsing S-expressions, if we read a `#' character followed by
6555 the character CHAR, use PROC to parse an object from the stream.
6556 If PROC is #f, remove any parsing procedure registered for CHAR.
6558 The reader applies PROC to two arguments: CHAR and an input port.
6560 *** The new functions read-delimited and read-delimited! provide a
6561 general mechanism for doing delimited input on streams.
6563 (read-delimited DELIMS [PORT HANDLE-DELIM])
6564 Read until we encounter one of the characters in DELIMS (a string),
6565 or end-of-file. PORT is the input port to read from; it defaults to
6566 the current input port. The HANDLE-DELIM parameter determines how
6567 the terminating character is handled; it should be one of the
6570 'trim omit delimiter from result
6571 'peek leave delimiter character in input stream
6572 'concat append delimiter character to returned value
6573 'split return a pair: (RESULT . TERMINATOR)
6575 HANDLE-DELIM defaults to 'peek.
6577 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
6578 A side-effecting variant of `read-delimited'.
6580 The data is written into the string BUF at the indices in the
6581 half-open interval [START, END); the default interval is the whole
6582 string: START = 0 and END = (string-length BUF). The values of
6583 START and END must specify a well-defined interval in BUF, i.e.
6584 0 <= START <= END <= (string-length BUF).
6586 It returns NBYTES, the number of bytes read. If the buffer filled
6587 up without a delimiter character being found, it returns #f. If the
6588 port is at EOF when the read starts, it returns the EOF object.
6590 If an integer is returned (i.e., the read is successfully terminated
6591 by reading a delimiter character), then the HANDLE-DELIM parameter
6592 determines how to handle the terminating character. It is described
6593 above, and defaults to 'peek.
6595 (The descriptions of these functions were borrowed from the SCSH
6596 manual, by Olin Shivers and Brian Carlstrom.)
6598 *** The `%read-delimited!' function is the primitive used to implement
6599 `read-delimited' and `read-delimited!'.
6601 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
6603 This returns a pair of values: (TERMINATOR . NUM-READ).
6604 - TERMINATOR describes why the read was terminated. If it is a
6605 character or the eof object, then that is the value that terminated
6606 the read. If it is #f, the function filled the buffer without finding
6607 a delimiting character.
6608 - NUM-READ is the number of characters read into BUF.
6610 If the read is successfully terminated by reading a delimiter
6611 character, then the gobble? parameter determines what to do with the
6612 terminating character. If true, the character is removed from the
6613 input stream; if false, the character is left in the input stream
6614 where a subsequent read operation will retrieve it. In either case,
6615 the character is also the first value returned by the procedure call.
6617 (The descriptions of this function was borrowed from the SCSH manual,
6618 by Olin Shivers and Brian Carlstrom.)
6620 *** The `read-line' and `read-line!' functions have changed; they now
6621 trim the terminator by default; previously they appended it to the
6622 returned string. For the old behavior, use (read-line PORT 'concat).
6624 *** The functions `uniform-array-read!' and `uniform-array-write!' now
6625 take new optional START and END arguments, specifying the region of
6626 the array to read and write.
6628 *** The `ungetc-char-ready?' function has been removed. We feel it's
6629 inappropriate for an interface to expose implementation details this
6632 ** Changes to the Unix library and system call interface
6634 *** The new fcntl function provides access to the Unix `fcntl' system
6637 (fcntl PORT COMMAND VALUE)
6638 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
6639 Values for COMMAND are:
6641 F_DUPFD duplicate a file descriptor
6642 F_GETFD read the descriptor's close-on-exec flag
6643 F_SETFD set the descriptor's close-on-exec flag to VALUE
6644 F_GETFL read the descriptor's flags, as set on open
6645 F_SETFL set the descriptor's flags, as set on open to VALUE
6646 F_GETOWN return the process ID of a socket's owner, for SIGIO
6647 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
6648 FD_CLOEXEC not sure what this is
6650 For details, see the documentation for the fcntl system call.
6652 *** The arguments to `select' have changed, for compatibility with
6653 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
6654 expected behavior. The MILLISECONDS parameter has been changed to
6655 MICROSECONDS, to more closely resemble the underlying system call.
6656 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
6657 corresponding return set will be the same.
6659 *** The arguments to the `mknod' system call have changed. They are
6662 (mknod PATH TYPE PERMS DEV)
6663 Create a new file (`node') in the file system. PATH is the name of
6664 the file to create. TYPE is the kind of file to create; it should
6665 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
6666 permission bits to give the newly created file. If TYPE is
6667 'block-special or 'char-special, DEV specifies which device the
6668 special file refers to; its interpretation depends on the kind of
6669 special file being created.
6671 *** The `fork' function has been renamed to `primitive-fork', to avoid
6672 clashing with various SCSH forks.
6674 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
6675 and `recvfrom!'. They no longer accept a size for a second argument;
6676 you must pass a string to hold the received value. They no longer
6677 return the buffer. Instead, `recv' returns the length of the message
6678 received, and `recvfrom' returns a pair containing the packet's length
6679 and originating address.
6681 *** The file descriptor datatype has been removed, as have the
6682 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
6683 We plan to replace these functions with a SCSH-compatible interface.
6685 *** The `create' function has been removed; it's just a special case
6688 *** There are new functions to break down process termination status
6689 values. In the descriptions below, STATUS is a value returned by
6692 (status:exit-val STATUS)
6693 If the child process exited normally, this function returns the exit
6694 code for the child process (i.e., the value passed to exit, or
6695 returned from main). If the child process did not exit normally,
6696 this function returns #f.
6698 (status:stop-sig STATUS)
6699 If the child process was suspended by a signal, this function
6700 returns the signal that suspended the child. Otherwise, it returns
6703 (status:term-sig STATUS)
6704 If the child process terminated abnormally, this function returns
6705 the signal that terminated the child. Otherwise, this function
6708 POSIX promises that exactly one of these functions will return true on
6709 a valid STATUS value.
6711 These functions are compatible with SCSH.
6713 *** There are new accessors and setters for the broken-out time vectors
6714 returned by `localtime', `gmtime', and that ilk. They are:
6716 Component Accessor Setter
6717 ========================= ============ ============
6718 seconds tm:sec set-tm:sec
6719 minutes tm:min set-tm:min
6720 hours tm:hour set-tm:hour
6721 day of the month tm:mday set-tm:mday
6722 month tm:mon set-tm:mon
6723 year tm:year set-tm:year
6724 day of the week tm:wday set-tm:wday
6725 day in the year tm:yday set-tm:yday
6726 daylight saving time tm:isdst set-tm:isdst
6727 GMT offset, seconds tm:gmtoff set-tm:gmtoff
6728 name of time zone tm:zone set-tm:zone
6730 *** There are new accessors for the vectors returned by `uname',
6731 describing the host system:
6734 ============================================== ================
6735 name of the operating system implementation utsname:sysname
6736 network name of this machine utsname:nodename
6737 release level of the operating system utsname:release
6738 version level of the operating system utsname:version
6739 machine hardware platform utsname:machine
6741 *** There are new accessors for the vectors returned by `getpw',
6742 `getpwnam', `getpwuid', and `getpwent', describing entries from the
6743 system's user database:
6746 ====================== =================
6747 user name passwd:name
6748 user password passwd:passwd
6751 real name passwd:gecos
6752 home directory passwd:dir
6753 shell program passwd:shell
6755 *** There are new accessors for the vectors returned by `getgr',
6756 `getgrnam', `getgrgid', and `getgrent', describing entries from the
6757 system's group database:
6760 ======================= ============
6761 group name group:name
6762 group password group:passwd
6764 group members group:mem
6766 *** There are new accessors for the vectors returned by `gethost',
6767 `gethostbyaddr', `gethostbyname', and `gethostent', describing
6771 ========================= ===============
6772 official name of host hostent:name
6773 alias list hostent:aliases
6774 host address type hostent:addrtype
6775 length of address hostent:length
6776 list of addresses hostent:addr-list
6778 *** There are new accessors for the vectors returned by `getnet',
6779 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
6783 ========================= ===============
6784 official name of net netent:name
6785 alias list netent:aliases
6786 net number type netent:addrtype
6787 net number netent:net
6789 *** There are new accessors for the vectors returned by `getproto',
6790 `getprotobyname', `getprotobynumber', and `getprotoent', describing
6794 ========================= ===============
6795 official protocol name protoent:name
6796 alias list protoent:aliases
6797 protocol number protoent:proto
6799 *** There are new accessors for the vectors returned by `getserv',
6800 `getservbyname', `getservbyport', and `getservent', describing
6804 ========================= ===============
6805 official service name servent:name
6806 alias list servent:aliases
6807 port number servent:port
6808 protocol to use servent:proto
6810 *** There are new accessors for the sockaddr structures returned by
6811 `accept', `getsockname', `getpeername', `recvfrom!':
6814 ======================================== ===============
6815 address format (`family') sockaddr:fam
6816 path, for file domain addresses sockaddr:path
6817 address, for internet domain addresses sockaddr:addr
6818 TCP or UDP port, for internet sockaddr:port
6820 *** The `getpwent', `getgrent', `gethostent', `getnetent',
6821 `getprotoent', and `getservent' functions now return #f at the end of
6822 the user database. (They used to throw an exception.)
6824 Note that calling MUMBLEent function is equivalent to calling the
6825 corresponding MUMBLE function with no arguments.
6827 *** The `setpwent', `setgrent', `sethostent', `setnetent',
6828 `setprotoent', and `setservent' routines now take no arguments.
6830 *** The `gethost', `getproto', `getnet', and `getserv' functions now
6831 provide more useful information when they throw an exception.
6833 *** The `lnaof' function has been renamed to `inet-lnaof'.
6835 *** Guile now claims to have the `current-time' feature.
6837 *** The `mktime' function now takes an optional second argument ZONE,
6838 giving the time zone to use for the conversion. ZONE should be a
6839 string, in the same format as expected for the "TZ" environment variable.
6841 *** The `strptime' function now returns a pair (TIME . COUNT), where
6842 TIME is the parsed time as a vector, and COUNT is the number of
6843 characters from the string left unparsed. This function used to
6844 return the remaining characters as a string.
6846 *** The `gettimeofday' function has replaced the old `time+ticks' function.
6847 The return value is now (SECONDS . MICROSECONDS); the fractional
6848 component is no longer expressed in "ticks".
6850 *** The `ticks/sec' constant has been removed, in light of the above change.
6852 * Changes to the gh_ interface
6854 ** gh_eval_str() now returns an SCM object which is the result of the
6857 ** gh_scm2str() now copies the Scheme data to a caller-provided C
6860 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
6861 and returns the array
6863 ** gh_scm2str0() is gone: there is no need to distinguish
6864 null-terminated from non-null-terminated, since gh_scm2newstr() allows
6865 the user to interpret the data both ways.
6867 * Changes to the scm_ interface
6869 ** The new function scm_symbol_value0 provides an easy way to get a
6870 symbol's value from C code:
6872 SCM scm_symbol_value0 (char *NAME)
6873 Return the value of the symbol named by the null-terminated string
6874 NAME in the current module. If the symbol named NAME is unbound in
6875 the current module, return SCM_UNDEFINED.
6877 ** The new function scm_sysintern0 creates new top-level variables,
6878 without assigning them a value.
6880 SCM scm_sysintern0 (char *NAME)
6881 Create a new Scheme top-level variable named NAME. NAME is a
6882 null-terminated string. Return the variable's value cell.
6884 ** The function scm_internal_catch is the guts of catch. It handles
6885 all the mechanics of setting up a catch target, invoking the catch
6886 body, and perhaps invoking the handler if the body does a throw.
6888 The function is designed to be usable from C code, but is general
6889 enough to implement all the semantics Guile Scheme expects from throw.
6891 TAG is the catch tag. Typically, this is a symbol, but this function
6892 doesn't actually care about that.
6894 BODY is a pointer to a C function which runs the body of the catch;
6895 this is the code you can throw from. We call it like this:
6896 BODY (BODY_DATA, JMPBUF)
6898 BODY_DATA is just the BODY_DATA argument we received; we pass it
6899 through to BODY as its first argument. The caller can make
6900 BODY_DATA point to anything useful that BODY might need.
6901 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6902 which we have just created and initialized.
6904 HANDLER is a pointer to a C function to deal with a throw to TAG,
6905 should one occur. We call it like this:
6906 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6908 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6909 same idea as BODY_DATA above.
6910 THROWN_TAG is the tag that the user threw to; usually this is
6911 TAG, but it could be something else if TAG was #t (i.e., a
6912 catch-all), or the user threw to a jmpbuf.
6913 THROW_ARGS is the list of arguments the user passed to the THROW
6916 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6917 is just a pointer we pass through to HANDLER. We don't actually
6918 use either of those pointers otherwise ourselves. The idea is
6919 that, if our caller wants to communicate something to BODY or
6920 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6921 HANDLER can then use. Think of it as a way to make BODY and
6922 HANDLER closures, not just functions; MUMBLE_DATA points to the
6925 Of course, it's up to the caller to make sure that any data a
6926 MUMBLE_DATA needs is protected from GC. A common way to do this is
6927 to make MUMBLE_DATA a pointer to data stored in an automatic
6928 structure variable; since the collector must scan the stack for
6929 references anyway, this assures that any references in MUMBLE_DATA
6932 ** The new function scm_internal_lazy_catch is exactly like
6933 scm_internal_catch, except:
6935 - It does not unwind the stack (this is the major difference).
6936 - If handler returns, its value is returned from the throw.
6937 - BODY always receives #f as its JMPBUF argument (since there's no
6938 jmpbuf associated with a lazy catch, because we don't unwind the
6941 ** scm_body_thunk is a new body function you can pass to
6942 scm_internal_catch if you want the body to be like Scheme's `catch'
6943 --- a thunk, or a function of one argument if the tag is #f.
6945 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6946 contains the Scheme procedure to invoke as the body, and the tag
6947 we're catching. If the tag is #f, then we pass JMPBUF (created by
6948 scm_internal_catch) to the body procedure; otherwise, the body gets
6951 ** scm_handle_by_proc is a new handler function you can pass to
6952 scm_internal_catch if you want the handler to act like Scheme's catch
6953 --- call a procedure with the tag and the throw arguments.
6955 If the user does a throw to this catch, this function runs a handler
6956 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6957 variable holding the Scheme procedure object to invoke. It ought to
6958 be a pointer to an automatic variable (i.e., one living on the stack),
6959 or the procedure object should be otherwise protected from GC.
6961 ** scm_handle_by_message is a new handler function to use with
6962 `scm_internal_catch' if you want Guile to print a message and die.
6963 It's useful for dealing with throws to uncaught keys at the top level.
6965 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6966 message header to print; if zero, we use "guile" instead. That
6967 text is followed by a colon, then the message described by ARGS.
6969 ** The return type of scm_boot_guile is now void; the function does
6970 not return a value, and indeed, never returns at all.
6972 ** The new function scm_shell makes it easy for user applications to
6973 process command-line arguments in a way that is compatible with the
6974 stand-alone guile interpreter (which is in turn compatible with SCSH,
6977 To use the scm_shell function, first initialize any guile modules
6978 linked into your application, and then call scm_shell with the values
6979 of ARGC and ARGV your `main' function received. scm_shell will add
6980 any SCSH-style meta-arguments from the top of the script file to the
6981 argument vector, and then process the command-line arguments. This
6982 generally means loading a script file or starting up an interactive
6983 command interpreter. For details, see "Changes to the stand-alone
6986 ** The new functions scm_get_meta_args and scm_count_argv help you
6987 implement the SCSH-style meta-argument, `\'.
6989 char **scm_get_meta_args (int ARGC, char **ARGV)
6990 If the second element of ARGV is a string consisting of a single
6991 backslash character (i.e. "\\" in Scheme notation), open the file
6992 named by the following argument, parse arguments from it, and return
6993 the spliced command line. The returned array is terminated by a
6996 For details of argument parsing, see above, under "guile now accepts
6997 command-line arguments compatible with SCSH..."
6999 int scm_count_argv (char **ARGV)
7000 Count the arguments in ARGV, assuming it is terminated by a null
7003 For an example of how these functions might be used, see the source
7004 code for the function scm_shell in libguile/script.c.
7006 You will usually want to use scm_shell instead of calling this
7009 ** The new function scm_compile_shell_switches turns an array of
7010 command-line arguments into Scheme code to carry out the actions they
7011 describe. Given ARGC and ARGV, it returns a Scheme expression to
7012 evaluate, and calls scm_set_program_arguments to make any remaining
7013 command-line arguments available to the Scheme code. For example,
7014 given the following arguments:
7016 -e main -s ekko a speckled gecko
7018 scm_set_program_arguments will return the following expression:
7020 (begin (load "ekko") (main (command-line)) (quit))
7022 You will usually want to use scm_shell instead of calling this
7025 ** The function scm_shell_usage prints a usage message appropriate for
7026 an interpreter that uses scm_compile_shell_switches to handle its
7027 command-line arguments.
7029 void scm_shell_usage (int FATAL, char *MESSAGE)
7030 Print a usage message to the standard error output. If MESSAGE is
7031 non-zero, write it before the usage message, followed by a newline.
7032 If FATAL is non-zero, exit the process, using FATAL as the
7033 termination status. (If you want to be compatible with Guile,
7034 always use 1 as the exit status when terminating due to command-line
7037 You will usually want to use scm_shell instead of calling this
7040 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
7041 expressions. It used to return SCM_EOL. Earth-shattering.
7043 ** The macros for declaring scheme objects in C code have been
7044 rearranged slightly. They are now:
7046 SCM_SYMBOL (C_NAME, SCHEME_NAME)
7047 Declare a static SCM variable named C_NAME, and initialize it to
7048 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
7049 be a C identifier, and SCHEME_NAME should be a C string.
7051 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
7052 Just like SCM_SYMBOL, but make C_NAME globally visible.
7054 SCM_VCELL (C_NAME, SCHEME_NAME)
7055 Create a global variable at the Scheme level named SCHEME_NAME.
7056 Declare a static SCM variable named C_NAME, and initialize it to
7057 point to the Scheme variable's value cell.
7059 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
7060 Just like SCM_VCELL, but make C_NAME globally visible.
7062 The `guile-snarf' script writes initialization code for these macros
7063 to its standard output, given C source code as input.
7065 The SCM_GLOBAL macro is gone.
7067 ** The scm_read_line and scm_read_line_x functions have been replaced
7068 by Scheme code based on the %read-delimited! procedure (known to C
7069 code as scm_read_delimited_x). See its description above for more
7072 ** The function scm_sys_open has been renamed to scm_open. It now
7073 returns a port instead of an FD object.
7075 * The dynamic linking support has changed. For more information, see
7076 libguile/DYNAMIC-LINKING.
7081 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
7084 * Changes to the 'guile' program:
7086 ** Guile now loads some new files when it starts up. Guile first
7087 searches the load path for init.scm, and loads it if found. Then, if
7088 Guile is not being used to execute a script, and the user's home
7089 directory contains a file named `.guile', Guile loads that.
7091 ** You can now use Guile as a shell script interpreter.
7093 To paraphrase the SCSH manual:
7095 When Unix tries to execute an executable file whose first two
7096 characters are the `#!', it treats the file not as machine code to
7097 be directly executed by the native processor, but as source code
7098 to be executed by some interpreter. The interpreter to use is
7099 specified immediately after the #! sequence on the first line of
7100 the source file. The kernel reads in the name of the interpreter,
7101 and executes that instead. It passes the interpreter the source
7102 filename as its first argument, with the original arguments
7103 following. Consult the Unix man page for the `exec' system call
7104 for more information.
7106 Now you can use Guile as an interpreter, using a mechanism which is a
7107 compatible subset of that provided by SCSH.
7109 Guile now recognizes a '-s' command line switch, whose argument is the
7110 name of a file of Scheme code to load. It also treats the two
7111 characters `#!' as the start of a comment, terminated by `!#'. Thus,
7112 to make a file of Scheme code directly executable by Unix, insert the
7113 following two lines at the top of the file:
7115 #!/usr/local/bin/guile -s
7118 Guile treats the argument of the `-s' command-line switch as the name
7119 of a file of Scheme code to load, and treats the sequence `#!' as the
7120 start of a block comment, terminated by `!#'.
7122 For example, here's a version of 'echo' written in Scheme:
7124 #!/usr/local/bin/guile -s
7126 (let loop ((args (cdr (program-arguments))))
7129 (display (car args))
7130 (if (pair? (cdr args))
7132 (loop (cdr args)))))
7135 Why does `#!' start a block comment terminated by `!#', instead of the
7136 end of the line? That is the notation SCSH uses, and although we
7137 don't yet support the other SCSH features that motivate that choice,
7138 we would like to be backward-compatible with any existing Guile
7139 scripts once we do. Furthermore, if the path to Guile on your system
7140 is too long for your kernel, you can start the script with this
7144 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
7147 Note that some very old Unix systems don't support the `#!' syntax.
7150 ** You can now run Guile without installing it.
7152 Previous versions of the interactive Guile interpreter (`guile')
7153 couldn't start up unless Guile's Scheme library had been installed;
7154 they used the value of the environment variable `SCHEME_LOAD_PATH'
7155 later on in the startup process, but not to find the startup code
7156 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
7159 To run Guile without installing it, build it in the normal way, and
7160 then set the environment variable `SCHEME_LOAD_PATH' to a
7161 colon-separated list of directories, including the top-level directory
7162 of the Guile sources. For example, if you unpacked Guile so that the
7163 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
7166 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
7169 ** Guile's read-eval-print loop no longer prints #<unspecified>
7170 results. If the user wants to see this, she can evaluate the
7171 expression (assert-repl-print-unspecified #t), perhaps in her startup
7174 ** Guile no longer shows backtraces by default when an error occurs;
7175 however, it does display a message saying how to get one, and how to
7176 request that they be displayed by default. After an error, evaluate
7178 to see a backtrace, and
7179 (debug-enable 'backtrace)
7180 to see them by default.
7184 * Changes to Guile Scheme:
7186 ** Guile now distinguishes between #f and the empty list.
7188 This is for compatibility with the IEEE standard, the (possibly)
7189 upcoming Revised^5 Report on Scheme, and many extant Scheme
7192 Guile used to have #f and '() denote the same object, to make Scheme's
7193 type system more compatible with Emacs Lisp's. However, the change
7194 caused too much trouble for Scheme programmers, and we found another
7195 way to reconcile Emacs Lisp with Scheme that didn't require this.
7198 ** Guile's delq, delv, delete functions, and their destructive
7199 counterparts, delq!, delv!, and delete!, now remove all matching
7200 elements from the list, not just the first. This matches the behavior
7201 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
7202 functions which inspired them.
7204 I recognize that this change may break code in subtle ways, but it
7205 seems best to make the change before the FSF's first Guile release,
7209 ** The compiled-library-path function has been deleted from libguile.
7211 ** The facilities for loading Scheme source files have changed.
7213 *** The variable %load-path now tells Guile which directories to search
7214 for Scheme code. Its value is a list of strings, each of which names
7217 *** The variable %load-extensions now tells Guile which extensions to
7218 try appending to a filename when searching the load path. Its value
7219 is a list of strings. Its default value is ("" ".scm").
7221 *** (%search-load-path FILENAME) searches the directories listed in the
7222 value of the %load-path variable for a Scheme file named FILENAME,
7223 with all the extensions listed in %load-extensions. If it finds a
7224 match, then it returns its full filename. If FILENAME is absolute, it
7225 returns it unchanged. Otherwise, it returns #f.
7227 %search-load-path will not return matches that refer to directories.
7229 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
7230 uses %seach-load-path to find a file named FILENAME, and loads it if
7231 it finds it. If it can't read FILENAME for any reason, it throws an
7234 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
7237 *** load uses the same searching semantics as primitive-load.
7239 *** The functions %try-load, try-load-with-path, %load, load-with-path,
7240 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
7241 path, and load-module-with-path have been deleted. The functions
7242 above should serve their purposes.
7244 *** If the value of the variable %load-hook is a procedure,
7245 `primitive-load' applies its value to the name of the file being
7246 loaded (without the load path directory name prepended). If its value
7247 is #f, it is ignored. Otherwise, an error occurs.
7249 This is mostly useful for printing load notification messages.
7252 ** The function `eval!' is no longer accessible from the scheme level.
7253 We can't allow operations which introduce glocs into the scheme level,
7254 because Guile's type system can't handle these as data. Use `eval' or
7255 `read-and-eval!' (see below) as replacement.
7257 ** The new function read-and-eval! reads an expression from PORT,
7258 evaluates it, and returns the result. This is more efficient than
7259 simply calling `read' and `eval', since it is not necessary to make a
7260 copy of the expression for the evaluator to munge.
7262 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
7263 for the `read' function.
7266 ** The function `int?' has been removed; its definition was identical
7267 to that of `integer?'.
7269 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
7270 use the R4RS names for these functions.
7272 ** The function object-properties no longer returns the hash handle;
7273 it simply returns the object's property list.
7275 ** Many functions have been changed to throw errors, instead of
7276 returning #f on failure. The point of providing exception handling in
7277 the language is to simplify the logic of user code, but this is less
7278 useful if Guile's primitives don't throw exceptions.
7280 ** The function `fileno' has been renamed from `%fileno'.
7282 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
7285 * Changes to Guile's C interface:
7287 ** The library's initialization procedure has been simplified.
7288 scm_boot_guile now has the prototype:
7290 void scm_boot_guile (int ARGC,
7292 void (*main_func) (),
7295 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
7296 MAIN_FUNC should do all the work of the program (initializing other
7297 packages, reading user input, etc.) before returning. When MAIN_FUNC
7298 returns, call exit (0); this function never returns. If you want some
7299 other exit value, MAIN_FUNC may call exit itself.
7301 scm_boot_guile arranges for program-arguments to return the strings
7302 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
7303 scm_set_program_arguments with the final list, so Scheme code will
7304 know which arguments have been processed.
7306 scm_boot_guile establishes a catch-all catch handler which prints an
7307 error message and exits the process. This means that Guile exits in a
7308 coherent way when system errors occur and the user isn't prepared to
7309 handle it. If the user doesn't like this behavior, they can establish
7310 their own universal catcher in MAIN_FUNC to shadow this one.
7312 Why must the caller do all the real work from MAIN_FUNC? The garbage
7313 collector assumes that all local variables of type SCM will be above
7314 scm_boot_guile's stack frame on the stack. If you try to manipulate
7315 SCM values after this function returns, it's the luck of the draw
7316 whether the GC will be able to find the objects you allocate. So,
7317 scm_boot_guile function exits, rather than returning, to discourage
7318 people from making that mistake.
7320 The IN, OUT, and ERR arguments were removed; there are other
7321 convenient ways to override these when desired.
7323 The RESULT argument was deleted; this function should never return.
7325 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
7329 ** Guile's header files should no longer conflict with your system's
7332 In order to compile code which #included <libguile.h>, previous
7333 versions of Guile required you to add a directory containing all the
7334 Guile header files to your #include path. This was a problem, since
7335 Guile's header files have names which conflict with many systems'
7338 Now only <libguile.h> need appear in your #include path; you must
7339 refer to all Guile's other header files as <libguile/mumble.h>.
7340 Guile's installation procedure puts libguile.h in $(includedir), and
7341 the rest in $(includedir)/libguile.
7344 ** Two new C functions, scm_protect_object and scm_unprotect_object,
7345 have been added to the Guile library.
7347 scm_protect_object (OBJ) protects OBJ from the garbage collector.
7348 OBJ will not be freed, even if all other references are dropped,
7349 until someone does scm_unprotect_object (OBJ). Both functions
7352 Note that calls to scm_protect_object do not nest. You can call
7353 scm_protect_object any number of times on a given object, and the
7354 next call to scm_unprotect_object will unprotect it completely.
7356 Basically, scm_protect_object and scm_unprotect_object just
7357 maintain a list of references to things. Since the GC knows about
7358 this list, all objects it mentions stay alive. scm_protect_object
7359 adds its argument to the list; scm_unprotect_object remove its
7360 argument from the list.
7363 ** scm_eval_0str now returns the value of the last expression
7366 ** The new function scm_read_0str reads an s-expression from a
7367 null-terminated string, and returns it.
7369 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
7370 to a Scheme port object.
7372 ** The new function `scm_set_program_arguments' allows C code to set
7373 the value returned by the Scheme `program-arguments' function.
7378 * Guile no longer includes sophisticated Tcl/Tk support.
7380 The old Tcl/Tk support was unsatisfying to us, because it required the
7381 user to link against the Tcl library, as well as Tk and Guile. The
7382 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
7383 referring to widgets by names, rather than exporting widgets to Scheme
7384 code as a special datatype.
7386 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
7387 maintainers described some very interesting changes in progress to the
7388 Tcl/Tk internals, which would facilitate clean interfaces between lone
7389 Tk and other interpreters --- even for garbage-collected languages
7390 like Scheme. They expected the new Tk to be publicly available in the
7393 Since it seems that Guile might soon have a new, cleaner interface to
7394 lone Tk, and that the old Guile/Tk glue code would probably need to be
7395 completely rewritten, we (Jim Blandy and Richard Stallman) have
7396 decided not to support the old code. We'll spend the time instead on
7397 a good interface to the newer Tk, as soon as it is available.
7399 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
7402 Copyright information:
7404 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2006 Free Software Foundation, Inc.
7406 Permission is granted to anyone to make or distribute verbatim copies
7407 of this document as received, in any medium, provided that the
7408 copyright notice and this permission notice are preserved,
7409 thus giving the recipient permission to redistribute in turn.
7411 Permission is granted to distribute modified versions
7412 of this document, or of portions of it,
7413 under the above conditions, provided also that they
7414 carry prominent notices stating who last changed them.
7419 paragraph-separate: "[
\f]*$"