1 Guile NEWS --- history of user-visible changes. -*- text -*-
2 Copyright (C) 1996, 1997, 1998, 1999, 2000 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send Guile bug reports to bug-guile@gnu.org.
7 Changes since Guile 1.4:
9 * Changes to the distribution
11 ** New modules (oop goops) etc
18 (oop goops active-slot)
19 (oop goops composite-slot)
21 plus some GOOPS utility modules have been added.
23 * Changes to the stand-alone interpreter
25 ** GOOPS has been merged into Guile
27 The Guile Object Oriented Programming System has been integrated into
32 (use-modules (oop goops))
34 access GOOPS bindings.
36 We're now ready to try some basic GOOPS functionality.
40 (define-method (+ (x <string>) (y <string>))
44 (+ "abc" "de") --> "abcde"
48 (define-class <2D-vector> ()
49 (x #:init-value 0 #:accessor x-component #:init-keyword #:x)
50 (y #:init-value 0 #:accessor y-component #:init-keyword #:y))
52 (define-method write ((obj <2D-vector>) port)
53 (display (format #f "<~S, ~S>" (x-component obj) (y-component obj))
56 (define v (make <2D-vector> #:x 3 #:y 4))
59 (define-method + ((x <2D-vector>) (y <2D-vector>))
61 #:x (+ (x-component x) (x-component y))
62 #:y (+ (y-component x) (y-component y))))
66 Asking for the type of an object
68 (class-of v) --> #<<class> <2D-vector> 40241ac0>
69 <2D-vector> --> #<<class> <2D-vector> 40241ac0>
70 (class-of 1) --> #<<class> <integer> 401b2a98>
71 <integer> --> #<<class> <integer> 401b2a98>
73 (is-a? v <2D-vector>) --> #t
75 See further in the GOOPS tutorial available in the guile-doc
76 distribution in info (goops.info) and texinfo formats.
78 ** It's now possible to create modules with controlled environments
82 (use-modules (ice-9 safe))
83 (define m (make-safe-module))
84 ;;; m will now be a module containing only a safe subset of R5RS
85 (eval '(+ 1 2) m) --> 3
86 (eval 'load m) --> ERROR: Unbound variable: load
88 * Changes to Scheme functions and syntax
90 ** The semantics of guardians have changed.
92 The changes are for the most part compatible. An important criterion
93 was to keep the typical usage of guardians as simple as before, but to
94 make the semantics safer and (as a result) more useful.
96 *** All objects returned from guardians are now properly alive.
98 It is now guaranteed that any object referenced by an object returned
99 from a guardian is alive. It's now impossible for a guardian to
100 return a "contained" object before its "containing" object.
102 One incompatible (but probably not very important) change resulting
103 from this is that it is no longer possible to guard objects that
104 indirectly reference themselves (i.e. are parts of cycles). If you do
105 so accidentally, you'll get a warning.
107 *** There are now two types of guardians: greedy and sharing.
109 If you call (make-guardian #t) or just (make-guardian), you'll get a
110 greedy guardian, and for (make-guardian #f) a sharing guardian.
112 Greedy guardians are the default because they are more "defensive".
113 You can only greedily guard an object once. If you guard an object
114 more than once, once in a greedy guardian and the rest of times in
115 sharing guardians, then it is guaranteed that the object won't be
116 returned from sharing guardians as long as it is greedily guarded
119 Guardians returned by calls to `make-guardian' can now take one more
120 optional parameter, which says whether to throw an error in case an
121 attempt is made to greedily guard an object that is already greedily
122 guarded. The default is true, i.e. throw an error. If the parameter
123 is false, the guardian invocation returns #t if guarding was
124 successful and #f if it wasn't.
126 Also, since greedy guarding is, in effect, a side-effecting operation
127 on objects, a new function is introduced: `destroy-guardian!'.
128 Invoking this function on a guardian renders it unoperative and, if
129 the guardian is greedy, clears the "greedily guarded" property of the
130 objects that were guarded by it, thus undoing the side effect.
132 Note that all this hair is hardly very important, since guardian
133 objects are usually permanent.
135 ** Escape procedures created by call-with-current-continuation now
136 accept any number of arguments, as required by R5RS.
138 ** New function `make-object-property'
140 This function returns a new `procedure with setter' P that can be used
141 to attach a property to objects. When calling P as
145 where `obj' is any kind of object, it attaches `val' to `obj' in such
146 a way that it can be retrieved by calling P as
150 This function will replace procedure properties, symbol properties and
151 source properties eventually.
153 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
155 Instead of #&optional, #&key, etc you should now use #:optional,
156 #:key, etc. Since #:optional is a keyword, you can write it as just
157 :optional when (read-set! keywords 'prefix) is active.
159 The old reader syntax `#&' is still supported, but deprecated. It
160 will be removed in the next release.
162 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
164 `eval' is now R5RS, that is it takes two arguments.
165 The second argument is an environment specifier, i.e. either
167 (scheme-report-environment 5)
169 (interaction-environment)
175 ** New define-module option: pure
177 Tells the module system not to include any bindings from the root
182 (define-module (totally-empty-module)
185 ** New define-module option: export NAME1 ...
187 Export names NAME1 ...
189 This option is required if you want to be able to export bindings from
190 a module which doesn't import one of `define-public' or `export'.
196 :use-module (ice-9 r5rs)
199 ;;; Note that we're pure R5RS below this point!
204 ** Deprecated: scm_make_shared_substring
206 Explicit shared substrings will disappear from Guile.
208 Instead, "normal" strings will be implemented using sharing
209 internally, combined with a copy-on-write strategy.
211 ** Deprecated: scm_read_only_string_p
213 The concept of read-only strings will disappear in next release of
216 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
218 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
220 ** New function: read-string!/partial str [port_or_fdes [start [end]]]
222 Read characters from an fport or file descriptor into a string
223 STR. This procedure is scsh-compatible and can efficiently read
224 large strings. It will:
226 * attempt to fill the entire string, unless the START and/or
227 END arguments are supplied. i.e., START defaults to 0 and
228 END defaults to `(string-length str)'
230 * use the current input port if PORT_OR_FDES is not supplied.
232 * read any characters that are currently available, without
233 waiting for the rest (short reads are possible).
235 * wait for as long as it needs to for the first character to
236 become available, unless the port is in non-blocking mode
238 * return `#f' if end-of-file is encountered before reading any
239 characters, otherwise return the number of characters read.
241 * return 0 if the port is in non-blocking mode and no characters
242 are immediately available.
244 * return 0 if the request is for 0 bytes, with no end-of-file
247 ** New function: port? X
249 Returns a boolean indicating whether X is a port. Equivalent to
250 `(or (input-port? X) (output-port? X))'.
252 ** New function: port-for-each proc
254 Apply PROC to each port in the Guile port table in turn. The
255 return value is unspecified.
257 ** New function: dup2 oldfd newfd
259 A simple wrapper for the `dup2' system call. Copies the file
260 descriptor OLDFD to descriptor number NEWFD, replacing the
261 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
262 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
263 to move away ports which are using NEWFD. The return value is
266 ** New function: close-fdes fd
268 A simple wrapper for the `close' system call. Close file
269 descriptor FD, which must be an integer. Unlike close (*note
270 close: Ports and File Descriptors.), the file descriptor will be
271 closed even if a port is using it. The return value is
274 ** Deprecated: close-all-ports-except. This was intended for closing
275 ports in a child process after a fork, but it has the undesirable side
276 effect of flushing buffers. port-for-each is more flexible.
278 ** The (ice-9 popen) module now attempts to set up file descriptors in
279 the child process from the current Scheme ports, instead of using the
280 current values of file descriptors 0, 1, and 2 in the parent process.
282 ** Removed function: builtin-weak-bindings
284 There is no such concept as a weak binding any more.
286 ** Removed constants: bignum-radix
288 * Changes to the gh_ interface
290 * Changes to the scm_ interface
292 ** New function: scm_init_guile ()
294 In contrast to scm_boot_guile, scm_init_guile will return normally
295 after initializing Guile. It is not available on all systems, tho.
297 ** New functions: scm_str2symbol, scm_mem2symbol
299 The function scm_str2symbol takes a const char* pointing to a zero-terminated
300 field of characters and creates a scheme symbol object from that C string.
301 The function scm_mem2symbol takes a const char* and a number of characters and
302 creates a symbol from the characters in that memory area.
304 ** New functions: scm_primitive_make_property
305 scm_primitive_property_ref
306 scm_primitive_property_set_x
307 scm_primitive_property_del_x
309 These functions implement a new way to deal with object properties.
310 See libguile/properties.c for their documentation.
312 ** New function: scm_done_free (long size)
314 This function is the inverse of scm_done_malloc. Use it to report the
315 amount of smob memory you free. The previous method, which involved
316 calling scm_done_malloc with negative argument, was somewhat
317 unintuitive (and is still available, of course).
319 ** New function: scm_c_memq (SCM obj, SCM list)
321 This function provides a fast C level alternative for scm_memq for the case
322 that the list parameter is known to be a proper list. The function is a
323 replacement for scm_sloppy_memq, but is stricter in its requirements on its
324 list input parameter, since for anything else but a proper list the function's
325 behaviour is undefined - it may even crash or loop endlessly. Further, for
326 the case that the object is not found in the list, scm_c_memq returns #f which
327 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
329 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
330 scm_remember_upto_here
332 These functions replace the function scm_remember.
334 ** Deprecated function: scm_remember
336 Use one of the new functions scm_remember_upto_here_1,
337 scm_remember_upto_here_2 or scm_remember_upto_here instead.
339 ** New global variable scm_gc_running_p introduced.
341 Use this variable to find out if garbage collection is being executed. Up to
342 now applications have used scm_gc_heap_lock to test if garbage collection was
343 running, which also works because of the fact that up to know only the garbage
344 collector has set this variable. But, this is an implementation detail that
345 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
346 of this variable is (and has been) not fully safe anyway.
348 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
350 Use these instead of SCM_LENGTH_MAX.
352 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
353 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
354 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
356 Use these instead of SCM_LENGTH.
358 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
359 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
360 SCM_SET_BITVECTOR_LENGTH
362 Use these instead of SCM_SETLENGTH
364 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
365 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
368 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
371 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
372 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
375 Use these instead of SCM_SETCHARS.
377 ** New macro: SCM_BITVECTOR_P
379 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
381 Use instead of SCM_COERCE_SUBSTR.
383 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
384 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
385 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
386 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
387 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
388 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
389 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
390 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
391 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
392 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
393 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC
395 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
396 Use scm_memory_error instead of SCM_NALLOC.
397 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
398 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
399 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
400 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
401 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
402 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
403 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
404 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
405 Use SCM_STRINGP instead of SCM_RWSTRINGP.
406 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
407 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
408 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
409 Use a type specific setter macro instead of SCM_SETLENGTH.
410 Use a type specific setter macro instead of SCM_SETCHARS.
411 Use a type specific length macro instead of SCM_LENGTH_MAX.
412 Use SCM_GCMARKP instead of SCM_GC8MARKP.
413 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
414 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
415 Use SCM_TYP16 instead of SCM_GCTYP16.
416 Use SCM_CDR instead of SCM_GCCDR.
418 ** Removed function: scm_struct_init
420 ** Removed variable: scm_symhash_dim
422 ** Renamed function: scm_make_cont has been replaced by
423 scm_make_continuation, which has a different interface.
425 ** Deprecated function: scm_call_catching_errors
427 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
429 ** Deprecated function: scm_strhash
431 Use scm_string_hash instead.
433 ** Deprecated function: scm_vector_set_length_x
435 Instead, create a fresh vector of the desired size and copy the contents.
437 ** scm_gensym has changed prototype
439 scm_gensym now only takes one argument.
441 ** New function: scm_gentemp (SCM prefix, SCM obarray)
443 The builtin `gentemp' has now become a primitive.
445 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
448 There is now only a single symbol type scm_tc7_symbol.
449 The tag scm_tc7_lvector was not used anyway.
451 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
453 Use scm_make_smob_type and scm_set_smob_XXX instead.
455 ** New function scm_set_smob_apply.
457 This can be used to set an apply function to a smob type.
460 Changes since Guile 1.3.4:
462 * Changes to the distribution
464 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
466 We've changed the way we handle generated files in the Guile source
467 repository. As a result, the procedure for building trees obtained
468 from the nightly FTP snapshots or via CVS has changed:
469 - You must have appropriate versions of autoconf, automake, and
470 libtool installed on your system. See README for info on how to
471 obtain these programs.
472 - Before configuring the tree, you must first run the script
473 `autogen.sh' at the top of the source tree.
475 The Guile repository used to contain not only source files, written by
476 humans, but also some generated files, like configure scripts and
477 Makefile.in files. Even though the contents of these files could be
478 derived mechanically from other files present, we thought it would
479 make the tree easier to build if we checked them into CVS.
481 However, this approach means that minor differences between
482 developer's installed tools and habits affected the whole team.
483 So we have removed the generated files from the repository, and
484 added the autogen.sh script, which will reconstruct them
488 ** configure now has experimental options to remove support for certain
491 --disable-arrays omit array and uniform array support
492 --disable-posix omit posix interfaces
493 --disable-networking omit networking interfaces
494 --disable-regex omit regular expression interfaces
496 These are likely to become separate modules some day.
498 ** New configure option --enable-debug-freelist
500 This enables a debugging version of SCM_NEWCELL(), and also registers
501 an extra primitive, the setter `gc-set-debug-check-freelist!'.
503 Configure with the --enable-debug-freelist option to enable
504 the gc-set-debug-check-freelist! primitive, and then use:
506 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
507 (gc-set-debug-check-freelist! #f) # turn off checking
509 Checking of the freelist forces a traversal of the freelist and
510 a garbage collection before each allocation of a cell. This can
511 slow down the interpreter dramatically, so the setter should be used to
512 turn on this extra processing only when necessary.
514 ** New configure option --enable-debug-malloc
516 Include code for debugging of calls to scm_must_malloc/realloc/free.
520 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
521 2. objects reallocated by scm_must_realloc has been allocated by
523 3. reallocated objects are reallocated with the same what string
525 But, most importantly, it records the number of allocated objects of
526 each kind. This is useful when searching for memory leaks.
528 A Guile compiled with this option provides the primitive
529 `malloc-stats' which returns an alist with pairs of kind and the
530 number of objects of that kind.
532 ** All includes are now referenced relative to the root directory
534 Since some users have had problems with mixups between Guile and
535 system headers, we have decided to always refer to Guile headers via
536 their parent directories. This essentially creates a "private name
537 space" for Guile headers. This means that the compiler only is given
538 -I options for the root build and root source directory.
540 ** Header files kw.h and genio.h have been removed.
542 ** The module (ice-9 getopt-gnu-style) has been removed.
544 ** New module (ice-9 documentation)
546 Implements the interface to documentation strings associated with
549 * Changes to the stand-alone interpreter
551 ** New command line option --debug
553 Start Guile with debugging evaluator and backtraces enabled.
555 This is useful when debugging your .guile init file or scripts.
559 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
560 (help REGEXP) ditto for objects with names matching REGEXP (a string)
561 (help ,EXPR) gives documentation for object returned by EXPR
562 (help) gives this text
564 `help' searches among bindings exported from loaded modules, while
565 `apropos' searches among bindings visible from the "current" module.
567 Examples: (help help)
569 (help "output-string")
571 ** `help' and `apropos' now prints full module names
573 ** Dynamic linking now uses libltdl from the libtool package.
575 The old system dependent code for doing dynamic linking has been
576 replaced with calls to the libltdl functions which do all the hairy
579 The major improvement is that you can now directly pass libtool
580 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
581 will be able to do the best shared library job you can get, via
584 The way dynamic libraries are found has changed and is not really
585 portable across platforms, probably. It is therefore recommended to
586 use absolute filenames when possible.
588 If you pass a filename without an extension to `dynamic-link', it will
589 try a few appropriate ones. Thus, the most platform ignorant way is
590 to specify a name like "libfoo", without any directories and
593 ** Guile COOP threads are now compatible with LinuxThreads
595 Previously, COOP threading wasn't possible in applications linked with
596 Linux POSIX threads due to their use of the stack pointer to find the
597 thread context. This has now been fixed with a workaround which uses
598 the pthreads to allocate the stack.
600 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
602 ** Positions of erring expression in scripts
604 With version 1.3.4, the location of the erring expression in Guile
605 scipts is no longer automatically reported. (This should have been
606 documented before the 1.3.4 release.)
608 You can get this information by enabling recording of positions of
609 source expressions and running the debugging evaluator. Put this at
610 the top of your script (or in your "site" file):
612 (read-enable 'positions)
613 (debug-enable 'debug)
615 ** Backtraces in scripts
617 It is now possible to get backtraces in scripts.
621 (debug-enable 'debug 'backtrace)
623 at the top of the script.
625 (The first options enables the debugging evaluator.
626 The second enables backtraces.)
628 ** Part of module system symbol lookup now implemented in C
630 The eval closure of most modules is now implemented in C. Since this
631 was one of the bottlenecks for loading speed, Guile now loads code
632 substantially faster than before.
634 ** Attempting to get the value of an unbound variable now produces
635 an exception with a key of 'unbound-variable instead of 'misc-error.
637 ** The initial default output port is now unbuffered if it's using a
638 tty device. Previously in this situation it was line-buffered.
640 ** gc-thunk is deprecated
642 gc-thunk will be removed in next release of Guile. It has been
643 replaced by after-gc-hook.
645 ** New hook: after-gc-hook
647 after-gc-hook takes over the role of gc-thunk. This hook is run at
648 the first SCM_TICK after a GC. (Thus, the code is run at the same
649 point during evaluation as signal handlers.)
651 Note that this hook should be used only for diagnostic and debugging
652 purposes. It is not certain that it will continue to be well-defined
653 when this hook is run in the future.
655 C programmers: Note the new C level hooks scm_before_gc_c_hook,
656 scm_before_sweep_c_hook, scm_after_gc_c_hook.
658 ** Improvements to garbage collector
660 Guile 1.4 has a new policy for triggering heap allocation and
661 determining the sizes of heap segments. It fixes a number of problems
664 1. The new policy can handle two separate pools of cells
665 (2-word/4-word) better. (The old policy would run wild, allocating
666 more and more memory for certain programs.)
668 2. The old code would sometimes allocate far too much heap so that the
669 Guile process became gigantic. The new code avoids this.
671 3. The old code would sometimes allocate too little so that few cells
672 were freed at GC so that, in turn, too much time was spent in GC.
674 4. The old code would often trigger heap allocation several times in a
675 row. (The new scheme predicts how large the segments needs to be
676 in order not to need further allocation.)
678 All in all, the new GC policy will make larger applications more
681 The new GC scheme also is prepared for POSIX threading. Threads can
682 allocate private pools of cells ("clusters") with just a single
683 function call. Allocation of single cells from such a cluster can
684 then proceed without any need of inter-thread synchronization.
686 ** New environment variables controlling GC parameters
688 GUILE_MAX_SEGMENT_SIZE Maximal segment size
691 Allocation of 2-word cell heaps:
693 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
696 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
697 GC in percent of total heap size
700 Allocation of 4-word cell heaps
701 (used for real numbers and misc other objects):
703 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
705 (See entry "Way for application to customize GC parameters" under
706 section "Changes to the scm_ interface" below.)
708 ** Guile now implements reals using 4-word cells
710 This speeds up computation with reals. (They were earlier allocated
711 with `malloc'.) There is still some room for optimizations, however.
713 ** Some further steps toward POSIX thread support have been taken
715 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
716 don't have much effect any longer, and many of them will be removed in
720 are only handled at the top of the evaluator loop, immediately after
721 I/O, and in scm_equalp.
723 *** The GC can allocate thread private pools of pairs.
725 * Changes to Scheme functions and syntax
727 ** close-input-port and close-output-port are now R5RS
729 These procedures have been turned into primitives and have R5RS behaviour.
731 ** New procedure: simple-format PORT MESSAGE ARG1 ...
733 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
734 extended by the more sophisticated version in (ice-9 format)
736 (simple-format port message . args)
737 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
738 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
739 the escapes are replaced with corresponding members of ARGS:
740 ~A formats using `display' and ~S formats using `write'.
741 If DESTINATION is #t, then use the `current-output-port',
742 if DESTINATION is #f, then return a string containing the formatted text.
743 Does not add a trailing newline."
745 ** string-ref: the second argument is no longer optional.
747 ** string, list->string: no longer accept strings in their arguments,
748 only characters, for compatibility with R5RS.
750 ** New procedure: port-closed? PORT
751 Returns #t if PORT is closed or #f if it is open.
755 The list* functionality is now provided by cons* (SRFI-1 compliant)
757 ** New procedure: cons* ARG1 ARG2 ... ARGn
759 Like `list', but the last arg provides the tail of the constructed list,
760 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
762 Requires at least one argument. If given one argument, that argument
763 is returned as result.
765 This function is called `list*' in some other Schemes and in Common LISP.
767 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
769 ** New procedure: object-documentation OBJECT
771 Returns the documentation string associated with OBJECT. The
772 procedure uses a caching mechanism so that subsequent lookups are
775 Exported by (ice-9 documentation).
777 ** module-name now returns full names of modules
779 Previously, only the last part of the name was returned (`session' for
780 `(ice-9 session)'). Ex: `(ice-9 session)'.
782 * Changes to the gh_ interface
784 ** Deprecated: gh_int2scmb
786 Use gh_bool2scm instead.
788 * Changes to the scm_ interface
790 ** Guile primitives now carry docstrings!
792 Thanks to Greg Badros!
794 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
796 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
797 macros and must contain a docstring that is extracted into foo.doc using a new
798 guile-doc-snarf script (that uses guile-doc-snarf.awk).
800 However, a major overhaul of these macros is scheduled for the next release of
803 ** Guile primitives use a new technique for validation of arguments
805 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
806 the readability of argument checking.
808 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
810 ** New macros: SCM_PACK, SCM_UNPACK
812 Compose/decompose an SCM value.
814 The SCM type is now treated as an abstract data type and may be defined as a
815 long, a void* or as a struct, depending on the architecture and compile time
816 options. This makes it easier to find several types of bugs, for example when
817 SCM values are treated as integers without conversion. Values of the SCM type
818 should be treated as "atomic" values. These macros are used when
819 composing/decomposing an SCM value, either because you want to access
820 individual bits, or because you want to treat it as an integer value.
822 E.g., in order to set bit 7 in an SCM value x, use the expression
824 SCM_PACK (SCM_UNPACK (x) | 0x80)
826 ** The name property of hooks is deprecated.
827 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
829 You can emulate this feature by using object properties.
831 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
832 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
835 These macros will be removed in a future release of Guile.
837 ** The following types, functions and macros from numbers.h are deprecated:
838 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
839 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
841 Further, it is recommended not to rely on implementation details for guile's
842 current implementation of bignums. It is planned to replace this
843 implementation with gmp in the future.
845 ** Port internals: the rw_random variable in the scm_port structure
846 must be set to non-zero in any random access port. In recent Guile
847 releases it was only set for bidirectional random-access ports.
849 ** Port internals: the seek ptob procedure is now responsible for
850 resetting the buffers if required. The change was made so that in the
851 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
852 the fport and strport ptobs can avoid resetting the buffers,
853 in particular to avoid discarding unread chars. An existing port
854 type can be fixed by adding something like the following to the
855 beginning of the ptob seek procedure:
857 if (pt->rw_active == SCM_PORT_READ)
858 scm_end_input (object);
859 else if (pt->rw_active == SCM_PORT_WRITE)
860 ptob->flush (object);
862 although to actually avoid resetting the buffers and discard unread
863 chars requires further hacking that depends on the characteristics
866 ** Deprecated functions: scm_fseek, scm_tag
868 These functions are no longer used and will be removed in a future version.
870 ** The scm_sysmissing procedure is no longer used in libguile.
871 Unless it turns out to be unexpectedly useful to somebody, it will be
872 removed in a future version.
874 ** The format of error message strings has changed
876 The two C procedures: scm_display_error and scm_error, as well as the
877 primitive `scm-error', now use scm_simple_format to do their work.
878 This means that the message strings of all code must be updated to use
879 ~A where %s was used before, and ~S where %S was used before.
881 During the period when there still are a lot of old Guiles out there,
882 you might want to support both old and new versions of Guile.
884 There are basically two methods to achieve this. Both methods use
887 AC_CHECK_FUNCS(scm_simple_format)
889 in your configure.in.
891 Method 1: Use the string concatenation features of ANSI C's
896 #ifdef HAVE_SCM_SIMPLE_FORMAT
902 Then represent each of your error messages using a preprocessor macro:
904 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
908 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
909 (define make-message string-append)
911 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
913 Method 2: Use the oldfmt function found in doc/oldfmt.c.
917 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
922 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
926 ** Deprecated: coop_mutex_init, coop_condition_variable_init
928 Don't use the functions coop_mutex_init and
929 coop_condition_variable_init. They will change.
931 Use scm_mutex_init and scm_cond_init instead.
933 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
934 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
935 COND, as `scm_cond_wait' does, but it also bounds the duration
936 of the wait. If COND has not been signaled before time ABSTIME,
937 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
938 returns the error code `ETIMEDOUT'.
940 The ABSTIME parameter specifies an absolute time, with the same
941 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
942 to 00:00:00 GMT, January 1, 1970.
944 ** New function: scm_cond_broadcast (scm_cond_t *COND)
945 `scm_cond_broadcast' restarts all the threads that are waiting
946 on the condition variable COND. Nothing happens if no threads are
949 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
950 `scm_key_create' allocates a new TSD key. The key is stored in
951 the location pointed to by KEY. There is no limit on the number
952 of keys allocated at a given time. The value initially associated
953 with the returned key is `NULL' in all currently executing threads.
955 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
956 function associated with the key. When a thread terminates,
957 DESTR_FUNCTION is called on the value associated with the key in
958 that thread. The DESTR_FUNCTION is not called if a key is deleted
959 with `scm_key_delete' or a value is changed with
960 `scm_setspecific'. The order in which destructor functions are
961 called at thread termination time is unspecified.
963 Destructors are not yet implemented.
965 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
966 `scm_setspecific' changes the value associated with KEY in the
967 calling thread, storing the given POINTER instead.
969 ** New function: scm_getspecific (scm_key_t KEY)
970 `scm_getspecific' returns the value currently associated with
971 KEY in the calling thread.
973 ** New function: scm_key_delete (scm_key_t KEY)
974 `scm_key_delete' deallocates a TSD key. It does not check
975 whether non-`NULL' values are associated with that key in the
976 currently executing threads, nor call the destructor function
977 associated with the key.
979 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
981 Initialize a C level hook HOOK with associated HOOK_DATA and type
982 TYPE. (See scm_c_hook_run ().)
984 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
986 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
987 is true, add it last, otherwise first. The same FUNC can be added
988 multiple times if FUNC_DATA differ and vice versa.
990 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
992 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
993 function is only removed if both FUNC and FUNC_DATA matches.
995 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
997 Run hook HOOK passing DATA to the hook functions.
999 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
1000 returned is undefined.
1002 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
1003 returns a non-NULL value. This value is returned as the result of
1004 scm_c_hook_run. If all functions return NULL, NULL is returned.
1006 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
1007 returns a NULL value, and NULL is returned. If all functions returns
1008 a non-NULL value, the last value is returned.
1010 ** New C level GC hooks
1012 Five new C level hooks has been added to the garbage collector.
1014 scm_before_gc_c_hook
1017 are run before locking and after unlocking the heap. The system is
1018 thus in a mode where evaluation can take place. (Except that
1019 scm_before_gc_c_hook must not allocate new cells.)
1021 scm_before_mark_c_hook
1022 scm_before_sweep_c_hook
1023 scm_after_sweep_c_hook
1025 are run when the heap is locked. These are intended for extension of
1026 the GC in a modular fashion. Examples are the weaks and guardians
1029 ** Way for application to customize GC parameters
1031 The application can set up other default values for the GC heap
1032 allocation parameters
1034 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
1035 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
1036 GUILE_MAX_SEGMENT_SIZE,
1040 scm_default_init_heap_size_1, scm_default_min_yield_1,
1041 scm_default_init_heap_size_2, scm_default_min_yield_2,
1042 scm_default_max_segment_size
1044 respectively before callong scm_boot_guile.
1046 (See entry "New environment variables ..." in section
1047 "Changes to the stand-alone interpreter" above.)
1049 ** scm_protect_object/scm_unprotect_object now nest
1051 This means that you can call scm_protect_object multiple times on an
1052 object and count on the object being protected until
1053 scm_unprotect_object has been call the same number of times.
1055 The functions also have better time complexity.
1057 Still, it is usually possible to structure the application in a way
1058 that you don't need to use these functions. For example, if you use a
1059 protected standard Guile list to keep track of live objects rather
1060 than some custom data type, objects will die a natural death when they
1061 are no longer needed.
1063 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
1065 Guile does not provide the float representation for inexact real numbers any
1066 more. Now, only doubles are used to represent inexact real numbers. Further,
1067 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
1068 and scm_tc16_complex, respectively.
1070 ** Removed deprecated type scm_smobfuns
1072 ** Removed deprecated function scm_newsmob
1074 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
1076 There is an ongoing discussion among the developers whether to
1077 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
1078 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
1079 until this issue has been settled.
1081 ** Removed deprecated type tag scm_tc16_kw
1083 ** Added type tag scm_tc16_keyword
1085 (This was introduced already in release 1.3.4 but was not documented
1088 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
1090 * Changes to system call interfaces:
1092 ** The "select" procedure now tests port buffers for the ability to
1093 provide input or accept output. Previously only the underlying file
1094 descriptors were checked.
1096 ** New variable PIPE_BUF: the maximum number of bytes that can be
1097 atomically written to a pipe.
1099 ** If a facility is not available on the system when Guile is
1100 compiled, the corresponding primitive procedure will not be defined.
1101 Previously it would have been defined but would throw a system-error
1102 exception if called. Exception handlers which catch this case may
1103 need minor modification: an error will be thrown with key
1104 'unbound-variable instead of 'system-error. Alternatively it's
1105 now possible to use `defined?' to check whether the facility is
1108 ** Procedures which depend on the timezone should now give the correct
1109 result on systems which cache the TZ environment variable, even if TZ
1110 is changed without calling tzset.
1112 * Changes to the networking interfaces:
1114 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
1115 long integers between network and host format. For now, it's not
1116 particularly convenient to do this kind of thing, but consider:
1118 (define write-network-long
1119 (lambda (value port)
1120 (let ((v (make-uniform-vector 1 1 0)))
1121 (uniform-vector-set! v 0 (htonl value))
1122 (uniform-vector-write v port))))
1124 (define read-network-long
1126 (let ((v (make-uniform-vector 1 1 0)))
1127 (uniform-vector-read! v port)
1128 (ntohl (uniform-vector-ref v 0)))))
1130 ** If inet-aton fails, it now throws an error with key 'misc-error
1131 instead of 'system-error, since errno is not relevant.
1133 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
1134 specific keys instead of 'system-error. The latter is inappropriate
1135 since errno will not have been set. The keys are:
1136 'host-not-found, 'try-again, 'no-recovery and 'no-data.
1138 ** sethostent, setnetent, setprotoent, setservent: now take an
1139 optional argument STAYOPEN, which specifies whether the database
1140 remains open after a database entry is accessed randomly (e.g., using
1141 gethostbyname for the hosts database.) The default is #f. Previously
1145 Changes since Guile 1.3.2:
1147 * Changes to the stand-alone interpreter
1151 An initial version of the Guile debugger written by Chris Hanson has
1152 been added. The debugger is still under development but is included
1153 in the distribution anyway since it is already quite useful.
1159 after an error to enter the debugger. Type `help' inside the debugger
1160 for a description of available commands.
1162 If you prefer to have stack frames numbered and printed in
1163 anti-chronological order and prefer up in the stack to be down on the
1164 screen as is the case in gdb, you can put
1166 (debug-enable 'backwards)
1168 in your .guile startup file. (However, this means that Guile can't
1169 use indentation to indicate stack level.)
1171 The debugger is autoloaded into Guile at the first use.
1173 ** Further enhancements to backtraces
1175 There is a new debug option `width' which controls the maximum width
1176 on the screen of printed stack frames. Fancy printing parameters
1177 ("level" and "length" as in Common LISP) are adaptively adjusted for
1178 each stack frame to give maximum information while still fitting
1179 within the bounds. If the stack frame can't be made to fit by
1180 adjusting parameters, it is simply cut off at the end. This is marked
1183 ** Some modules are now only loaded when the repl is started
1185 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
1186 regex) are now loaded into (guile-user) only if the repl has been
1187 started. The effect is that the startup time for scripts has been
1188 reduced to 30% of what it was previously.
1190 Correctly written scripts load the modules they require at the top of
1191 the file and should not be affected by this change.
1193 ** Hooks are now represented as smobs
1195 * Changes to Scheme functions and syntax
1197 ** Readline support has changed again.
1199 The old (readline-activator) module is gone. Use (ice-9 readline)
1200 instead, which now contains all readline functionality. So the code
1201 to activate readline is now
1203 (use-modules (ice-9 readline))
1206 This should work at any time, including from the guile prompt.
1208 To avoid confusion about the terms of Guile's license, please only
1209 enable readline for your personal use; please don't make it the
1210 default for others. Here is why we make this rather odd-sounding
1213 Guile is normally licensed under a weakened form of the GNU General
1214 Public License, which allows you to link code with Guile without
1215 placing that code under the GPL. This exception is important to some
1218 However, since readline is distributed under the GNU General Public
1219 License, when you link Guile with readline, either statically or
1220 dynamically, you effectively change Guile's license to the strict GPL.
1221 Whenever you link any strictly GPL'd code into Guile, uses of Guile
1222 which are normally permitted become forbidden. This is a rather
1223 non-obvious consequence of the licensing terms.
1225 So, to make sure things remain clear, please let people choose for
1226 themselves whether to link GPL'd libraries like readline with Guile.
1228 ** regexp-substitute/global has changed slightly, but incompatibly.
1230 If you include a function in the item list, the string of the match
1231 object it receives is the same string passed to
1232 regexp-substitute/global, not some suffix of that string.
1233 Correspondingly, the match's positions are relative to the entire
1234 string, not the suffix.
1236 If the regexp can match the empty string, the way matches are chosen
1237 from the string has changed. regexp-substitute/global recognizes the
1238 same set of matches that list-matches does; see below.
1240 ** New function: list-matches REGEXP STRING [FLAGS]
1242 Return a list of match objects, one for every non-overlapping, maximal
1243 match of REGEXP in STRING. The matches appear in left-to-right order.
1244 list-matches only reports matches of the empty string if there are no
1245 other matches which begin on, end at, or include the empty match's
1248 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1250 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
1252 For each match of REGEXP in STRING, apply PROC to the match object,
1253 and the last value PROC returned, or INIT for the first call. Return
1254 the last value returned by PROC. We apply PROC to the matches as they
1255 appear from left to right.
1257 This function recognizes matches according to the same criteria as
1260 Thus, you could define list-matches like this:
1262 (define (list-matches regexp string . flags)
1263 (reverse! (apply fold-matches regexp string '() cons flags)))
1265 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1269 *** New function: hook? OBJ
1271 Return #t if OBJ is a hook, otherwise #f.
1273 *** New function: make-hook-with-name NAME [ARITY]
1275 Return a hook with name NAME and arity ARITY. The default value for
1276 ARITY is 0. The only effect of NAME is that it will appear when the
1277 hook object is printed to ease debugging.
1279 *** New function: hook-empty? HOOK
1281 Return #t if HOOK doesn't contain any procedures, otherwise #f.
1283 *** New function: hook->list HOOK
1285 Return a list of the procedures that are called when run-hook is
1288 ** `map' signals an error if its argument lists are not all the same length.
1290 This is the behavior required by R5RS, so this change is really a bug
1291 fix. But it seems to affect a lot of people's code, so we're
1292 mentioning it here anyway.
1294 ** Print-state handling has been made more transparent
1296 Under certain circumstances, ports are represented as a port with an
1297 associated print state. Earlier, this pair was represented as a pair
1298 (see "Some magic has been added to the printer" below). It is now
1299 indistinguishable (almost; see `get-print-state') from a port on the
1302 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
1304 Return a new port with the associated print state PRINT-STATE.
1306 *** New function: get-print-state OUTPUT-PORT
1308 Return the print state associated with this port if it exists,
1309 otherwise return #f.
1311 *** New function: directory-stream? OBJECT
1313 Returns true iff OBJECT is a directory stream --- the sort of object
1314 returned by `opendir'.
1316 ** New function: using-readline?
1318 Return #t if readline is in use in the current repl.
1320 ** structs will be removed in 1.4
1322 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
1323 and use GOOPS objects as the fundamental record type.
1325 * Changes to the scm_ interface
1327 ** structs will be removed in 1.4
1329 The entire current struct interface (struct.c, struct.h) will be
1330 replaced in Guile 1.4. We will merge GOOPS into libguile and use
1331 GOOPS objects as the fundamental record type.
1333 ** The internal representation of subr's has changed
1335 Instead of giving a hint to the subr name, the CAR field of the subr
1336 now contains an index to a subr entry in scm_subr_table.
1338 *** New variable: scm_subr_table
1340 An array of subr entries. A subr entry contains the name, properties
1341 and documentation associated with the subr. The properties and
1342 documentation slots are not yet used.
1344 ** A new scheme for "forwarding" calls to a builtin to a generic function
1346 It is now possible to extend the functionality of some Guile
1347 primitives by letting them defer a call to a GOOPS generic function on
1348 argument mismatch. This means that there is no loss of efficiency in
1353 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
1354 (define-method + ((x <string>) (y <string>))
1355 (string-append x y))
1357 + will still be as efficient as usual in numerical calculations, but
1358 can also be used for concatenating strings.
1360 Who will be the first one to extend Guile's numerical tower to
1361 rationals? :) [OK, there a few other things to fix before this can
1362 be made in a clean way.]
1364 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
1366 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
1368 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
1370 These do the same job as SCM_PROC and SCM_PROC1, but they also define
1371 a variable GENERIC which can be used by the dispatch macros below.
1373 [This is experimental code which may change soon.]
1375 *** New macros for forwarding control to a generic on arg type error
1377 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
1379 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
1381 These correspond to the scm_wta function call, and have the same
1382 behaviour until the user has called the GOOPS primitive
1383 `enable-primitive-generic!'. After that, these macros will apply the
1384 generic function GENERIC to the argument(s) instead of calling
1387 [This is experimental code which may change soon.]
1389 *** New macros for argument testing with generic dispatch
1391 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
1393 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
1395 These correspond to the SCM_ASSERT macro, but will defer control to
1396 GENERIC on error after `enable-primitive-generic!' has been called.
1398 [This is experimental code which may change soon.]
1400 ** New function: SCM scm_eval_body (SCM body, SCM env)
1402 Evaluates the body of a special form.
1404 ** The internal representation of struct's has changed
1406 Previously, four slots were allocated for the procedure(s) of entities
1407 and operators. The motivation for this representation had to do with
1408 the structure of the evaluator, the wish to support tail-recursive
1409 generic functions, and efficiency. Since the generic function
1410 dispatch mechanism has changed, there is no longer a need for such an
1411 expensive representation, and the representation has been simplified.
1413 This should not make any difference for most users.
1415 ** GOOPS support has been cleaned up.
1417 Some code has been moved from eval.c to objects.c and code in both of
1418 these compilation units has been cleaned up and better structured.
1420 *** New functions for applying generic functions
1422 New function: SCM scm_apply_generic (GENERIC, ARGS)
1423 New function: SCM scm_call_generic_0 (GENERIC)
1424 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
1425 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
1426 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
1428 ** Deprecated function: scm_make_named_hook
1430 It is now replaced by:
1432 ** New function: SCM scm_create_hook (const char *name, int arity)
1434 Creates a hook in the same way as make-hook above but also
1435 binds a variable named NAME to it.
1437 This is the typical way of creating a hook from C code.
1439 Currently, the variable is created in the "current" module.
1440 This might change when we get the new module system.
1442 [The behaviour is identical to scm_make_named_hook.]
1446 Changes since Guile 1.3:
1448 * Changes to mailing lists
1450 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
1452 See the README file to find current addresses for all the Guile
1455 * Changes to the distribution
1457 ** Readline support is no longer included with Guile by default.
1459 Based on the different license terms of Guile and Readline, we
1460 concluded that Guile should not *by default* cause the linking of
1461 Readline into an application program. Readline support is now offered
1462 as a separate module, which is linked into an application only when
1463 you explicitly specify it.
1465 Although Guile is GNU software, its distribution terms add a special
1466 exception to the usual GNU General Public License (GPL). Guile's
1467 license includes a clause that allows you to link Guile with non-free
1468 programs. We add this exception so as not to put Guile at a
1469 disadvantage vis-a-vis other extensibility packages that support other
1472 In contrast, the GNU Readline library is distributed under the GNU
1473 General Public License pure and simple. This means that you may not
1474 link Readline, even dynamically, into an application unless it is
1475 distributed under a free software license that is compatible the GPL.
1477 Because of this difference in distribution terms, an application that
1478 can use Guile may not be able to use Readline. Now users will be
1479 explicitly offered two independent decisions about the use of these
1482 You can activate the readline support by issuing
1484 (use-modules (readline-activator))
1487 from your ".guile" file, for example.
1489 * Changes to the stand-alone interpreter
1491 ** All builtins now print as primitives.
1492 Previously builtin procedures not belonging to the fundamental subr
1493 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
1494 Now, they print as #<primitive-procedure NAME>.
1496 ** Backtraces slightly more intelligible.
1497 gsubr-apply and macro transformer application frames no longer appear
1500 * Changes to Scheme functions and syntax
1502 ** Guile now correctly handles internal defines by rewriting them into
1503 their equivalent letrec. Previously, internal defines would
1504 incrementally add to the innermost environment, without checking
1505 whether the restrictions specified in RnRS were met. This lead to the
1506 correct behaviour when these restriction actually were met, but didn't
1507 catch all illegal uses. Such an illegal use could lead to crashes of
1508 the Guile interpreter or or other unwanted results. An example of
1509 incorrect internal defines that made Guile behave erratically:
1521 The problem with this example is that the definition of `c' uses the
1522 value of `b' directly. This confuses the meoization machine of Guile
1523 so that the second call of `b' (this time in a larger environment that
1524 also contains bindings for `c' and `d') refers to the binding of `c'
1525 instead of `a'. You could also make Guile crash with a variation on
1530 (define (b flag) (if flag a 1))
1531 (define c (1+ (b flag)))
1539 From now on, Guile will issue an `Unbound variable: b' error message
1544 A hook contains a list of functions which should be called on
1545 particular occasions in an existing program. Hooks are used for
1548 A window manager might have a hook before-window-map-hook. The window
1549 manager uses the function run-hooks to call all functions stored in
1550 before-window-map-hook each time a window is mapped. The user can
1551 store functions in the hook using add-hook!.
1553 In Guile, hooks are first class objects.
1555 *** New function: make-hook [N_ARGS]
1557 Return a hook for hook functions which can take N_ARGS arguments.
1558 The default value for N_ARGS is 0.
1560 (See also scm_make_named_hook below.)
1562 *** New function: add-hook! HOOK PROC [APPEND_P]
1564 Put PROC at the beginning of the list of functions stored in HOOK.
1565 If APPEND_P is supplied, and non-false, put PROC at the end instead.
1567 PROC must be able to take the number of arguments specified when the
1570 If PROC already exists in HOOK, then remove it first.
1572 *** New function: remove-hook! HOOK PROC
1574 Remove PROC from the list of functions in HOOK.
1576 *** New function: reset-hook! HOOK
1578 Clear the list of hook functions stored in HOOK.
1580 *** New function: run-hook HOOK ARG1 ...
1582 Run all hook functions stored in HOOK with arguments ARG1 ... .
1583 The number of arguments supplied must correspond to the number given
1584 when the hook was created.
1586 ** The function `dynamic-link' now takes optional keyword arguments.
1587 The only keyword argument that is currently defined is `:global
1588 BOOL'. With it, you can control whether the shared library will be
1589 linked in global mode or not. In global mode, the symbols from the
1590 linked library can be used to resolve references from other
1591 dynamically linked libraries. In non-global mode, the linked
1592 library is essentially invisible and can only be accessed via
1593 `dynamic-func', etc. The default is now to link in global mode.
1594 Previously, the default has been non-global mode.
1596 The `#:global' keyword is only effective on platforms that support
1597 the dlopen family of functions.
1599 ** New function `provided?'
1601 - Function: provided? FEATURE
1602 Return true iff FEATURE is supported by this installation of
1603 Guile. FEATURE must be a symbol naming a feature; the global
1604 variable `*features*' is a list of available features.
1606 ** Changes to the module (ice-9 expect):
1608 *** The expect-strings macro now matches `$' in a regular expression
1609 only at a line-break or end-of-file by default. Previously it would
1610 match the end of the string accumulated so far. The old behaviour
1611 can be obtained by setting the variable `expect-strings-exec-flags'
1614 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
1615 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
1616 in a regular expression will still match before a line-break or
1617 end-of-file. The default is `regexp/noteol'.
1619 *** The expect-strings macro now uses a variable
1620 `expect-strings-compile-flags' for the flags to be supplied to
1621 `make-regexp'. The default is `regexp/newline', which was previously
1624 *** The expect macro now supplies two arguments to a match procedure:
1625 the current accumulated string and a flag to indicate whether
1626 end-of-file has been reached. Previously only the string was supplied.
1627 If end-of-file is reached, the match procedure will be called an
1628 additional time with the same accumulated string as the previous call
1629 but with the flag set.
1631 ** New module (ice-9 format), implementing the Common Lisp `format' function.
1633 This code, and the documentation for it that appears here, was
1634 borrowed from SLIB, with minor adaptations for Guile.
1636 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
1637 An almost complete implementation of Common LISP format description
1638 according to the CL reference book `Common LISP' from Guy L.
1639 Steele, Digital Press. Backward compatible to most of the
1640 available Scheme format implementations.
1642 Returns `#t', `#f' or a string; has side effect of printing
1643 according to FORMAT-STRING. If DESTINATION is `#t', the output is
1644 to the current output port and `#t' is returned. If DESTINATION
1645 is `#f', a formatted string is returned as the result of the call.
1646 NEW: If DESTINATION is a string, DESTINATION is regarded as the
1647 format string; FORMAT-STRING is then the first argument and the
1648 output is returned as a string. If DESTINATION is a number, the
1649 output is to the current error port if available by the
1650 implementation. Otherwise DESTINATION must be an output port and
1653 FORMAT-STRING must be a string. In case of a formatting error
1654 format returns `#f' and prints a message on the current output or
1655 error port. Characters are output as if the string were output by
1656 the `display' function with the exception of those prefixed by a
1657 tilde (~). For a detailed description of the FORMAT-STRING syntax
1658 please consult a Common LISP format reference manual. For a test
1659 suite to verify this format implementation load `formatst.scm'.
1660 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
1662 Note: `format' is not reentrant, i.e. only one `format'-call may
1663 be executed at a time.
1666 *** Format Specification (Format version 3.0)
1668 Please consult a Common LISP format reference manual for a detailed
1669 description of the format string syntax. For a demonstration of the
1670 implemented directives see `formatst.scm'.
1672 This implementation supports directive parameters and modifiers (`:'
1673 and `@' characters). Multiple parameters must be separated by a comma
1674 (`,'). Parameters can be numerical parameters (positive or negative),
1675 character parameters (prefixed by a quote character (`''), variable
1676 parameters (`v'), number of rest arguments parameter (`#'), empty and
1677 default parameters. Directive characters are case independent. The
1678 general form of a directive is:
1680 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
1682 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
1684 *** Implemented CL Format Control Directives
1686 Documentation syntax: Uppercase characters represent the
1687 corresponding control directive characters. Lowercase characters
1688 represent control directive parameter descriptions.
1691 Any (print as `display' does).
1695 `~MINCOL,COLINC,MINPAD,PADCHARA'
1699 S-expression (print as `write' does).
1703 `~MINCOL,COLINC,MINPAD,PADCHARS'
1709 print number sign always.
1712 print comma separated.
1714 `~MINCOL,PADCHAR,COMMACHARD'
1720 print number sign always.
1723 print comma separated.
1725 `~MINCOL,PADCHAR,COMMACHARX'
1731 print number sign always.
1734 print comma separated.
1736 `~MINCOL,PADCHAR,COMMACHARO'
1742 print number sign always.
1745 print comma separated.
1747 `~MINCOL,PADCHAR,COMMACHARB'
1752 `~N,MINCOL,PADCHAR,COMMACHARR'
1756 print a number as a Roman numeral.
1759 print a number as an "old fashioned" Roman numeral.
1762 print a number as an ordinal English number.
1765 print a number as a cardinal English number.
1770 prints `y' and `ies'.
1773 as `~P but jumps 1 argument backward.'
1776 as `~@P but jumps 1 argument backward.'
1781 prints a character as the reader can understand it (i.e. `#\'
1785 prints a character as emacs does (eg. `^C' for ASCII 03).
1788 Fixed-format floating-point (prints a flonum like MMM.NNN).
1789 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
1791 If the number is positive a plus sign is printed.
1794 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
1795 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
1797 If the number is positive a plus sign is printed.
1800 General floating-point (prints a flonum either fixed or
1802 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
1804 If the number is positive a plus sign is printed.
1807 Dollars floating-point (prints a flonum in fixed with signs
1809 `~DIGITS,SCALE,WIDTH,PADCHAR$'
1811 If the number is positive a plus sign is printed.
1814 A sign is always printed and appears before the padding.
1817 The sign appears before the padding.
1825 print newline if not at the beginning of the output line.
1827 prints `~&' and then N-1 newlines.
1832 print N page separators.
1842 newline is ignored, white space left.
1845 newline is left, white space ignored.
1850 relative tabulation.
1856 Indirection (expects indirect arguments as a list).
1858 extracts indirect arguments from format arguments.
1861 Case conversion (converts by `string-downcase').
1863 converts by `string-capitalize'.
1866 converts by `string-capitalize-first'.
1869 converts by `string-upcase'.
1872 Argument Jumping (jumps 1 argument forward).
1874 jumps N arguments forward.
1877 jumps 1 argument backward.
1880 jumps N arguments backward.
1883 jumps to the 0th argument.
1886 jumps to the Nth argument (beginning from 0)
1888 `~[STR0~;STR1~;...~;STRN~]'
1889 Conditional Expression (numerical clause conditional).
1891 take argument from N.
1894 true test conditional.
1897 if-else-then conditional.
1903 default clause follows.
1906 Iteration (args come from the next argument (a list)).
1908 at most N iterations.
1911 args from next arg (a list of lists).
1914 args from the rest of arguments.
1917 args from the rest args (lists).
1928 aborts if N <= M <= K
1930 *** Not Implemented CL Format Control Directives
1933 print `#f' as an empty list (see below).
1936 print `#f' as an empty list (see below).
1942 (sorry I don't understand its semantics completely)
1944 *** Extended, Replaced and Additional Control Directives
1946 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
1947 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
1948 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
1949 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
1950 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
1951 COMMAWIDTH is the number of characters between two comma
1955 print a R4RS complex number as `~F~@Fi' with passed parameters for
1959 Pretty print formatting of an argument for scheme code lists.
1965 Flushes the output if format DESTINATION is a port.
1968 Print a `#\space' character
1970 print N `#\space' characters.
1973 Print a `#\tab' character
1975 print N `#\tab' characters.
1978 Takes N as an integer representation for a character. No arguments
1979 are consumed. N is converted to a character by `integer->char'. N
1980 must be a positive decimal number.
1983 Print out readproof. Prints out internal objects represented as
1984 `#<...>' as strings `"#<...>"' so that the format output can always
1985 be processed by `read'.
1988 Print out readproof. Prints out internal objects represented as
1989 `#<...>' as strings `"#<...>"' so that the format output can always
1990 be processed by `read'.
1993 Prints information and a copyright notice on the format
1996 prints format version.
1999 may also print number strings, i.e. passing a number as a string
2000 and format it accordingly.
2002 *** Configuration Variables
2004 The format module exports some configuration variables to suit the
2005 systems and users needs. There should be no modification necessary for
2006 the configuration that comes with Guile. Format detects automatically
2007 if the running scheme system implements floating point numbers and
2010 format:symbol-case-conv
2011 Symbols are converted by `symbol->string' so the case type of the
2012 printed symbols is implementation dependent.
2013 `format:symbol-case-conv' is a one arg closure which is either
2014 `#f' (no conversion), `string-upcase', `string-downcase' or
2015 `string-capitalize'. (default `#f')
2017 format:iobj-case-conv
2018 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
2019 implementation internal objects. (default `#f')
2022 The character prefixing the exponent value in `~E' printing.
2025 *** Compatibility With Other Format Implementations
2031 Downward compatible except for padding support and `~A', `~S',
2032 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
2033 `printf' padding support which is completely replaced by the CL
2034 `format' padding style.
2037 Downward compatible except for `~', which is not documented
2038 (ignores all characters inside the format string up to a newline
2039 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
2040 numerical and variable parameters and `:/@' modifiers in the CL
2044 Downward compatible except for `~A' and `~S' which print in
2045 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
2046 directive parameters or modifiers)).
2049 Downward compatible except for an optional destination parameter:
2050 S2C accepts a format call without a destination which returns a
2051 formatted string. This is equivalent to a #f destination in S2C.
2052 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
2053 parameters or modifiers)).
2056 ** Changes to string-handling functions.
2058 These functions were added to support the (ice-9 format) module, above.
2060 *** New function: string-upcase STRING
2061 *** New function: string-downcase STRING
2063 These are non-destructive versions of the existing string-upcase! and
2064 string-downcase! functions.
2066 *** New function: string-capitalize! STRING
2067 *** New function: string-capitalize STRING
2069 These functions convert the first letter of each word in the string to
2072 (string-capitalize "howdy there")
2075 As with the other functions, string-capitalize! modifies the string in
2076 place, while string-capitalize returns a modified copy of its argument.
2078 *** New function: string-ci->symbol STRING
2080 Return a symbol whose name is STRING, but having the same case as if
2081 the symbol had be read by `read'.
2083 Guile can be configured to be sensitive or insensitive to case
2084 differences in Scheme identifiers. If Guile is case-insensitive, all
2085 symbols are converted to lower case on input. The `string-ci->symbol'
2086 function returns a symbol whose name in STRING, transformed as Guile
2087 would if STRING were input.
2089 *** New function: substring-move! STRING1 START END STRING2 START
2091 Copy the substring of STRING1 from START (inclusive) to END
2092 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
2093 string, and the source and destination areas may overlap; in all
2094 cases, the function behaves as if all the characters were copied
2097 *** Extended functions: substring-move-left! substring-move-right!
2099 These functions now correctly copy arbitrarily overlapping substrings;
2100 they are both synonyms for substring-move!.
2103 ** New module (ice-9 getopt-long), with the function `getopt-long'.
2105 getopt-long is a function for parsing command-line arguments in a
2106 manner consistent with other GNU programs.
2108 (getopt-long ARGS GRAMMAR)
2109 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
2111 ARGS should be a list of strings. Its first element should be the
2112 name of the program; subsequent elements should be the arguments
2113 that were passed to the program on the command line. The
2114 `program-arguments' procedure returns a list of this form.
2116 GRAMMAR is a list of the form:
2117 ((OPTION (PROPERTY VALUE) ...) ...)
2119 Each OPTION should be a symbol. `getopt-long' will accept a
2120 command-line option named `--OPTION'.
2121 Each option can have the following (PROPERTY VALUE) pairs:
2123 (single-char CHAR) --- Accept `-CHAR' as a single-character
2124 equivalent to `--OPTION'. This is how to specify traditional
2126 (required? BOOL) --- If BOOL is true, the option is required.
2127 getopt-long will raise an error if it is not found in ARGS.
2128 (value BOOL) --- If BOOL is #t, the option accepts a value; if
2129 it is #f, it does not; and if it is the symbol
2130 `optional', the option may appear in ARGS with or
2132 (predicate FUNC) --- If the option accepts a value (i.e. you
2133 specified `(value #t)' for this option), then getopt
2134 will apply FUNC to the value, and throw an exception
2135 if it returns #f. FUNC should be a procedure which
2136 accepts a string and returns a boolean value; you may
2137 need to use quasiquotes to get it into GRAMMAR.
2139 The (PROPERTY VALUE) pairs may occur in any order, but each
2140 property may occur only once. By default, options do not have
2141 single-character equivalents, are not required, and do not take
2144 In ARGS, single-character options may be combined, in the usual
2145 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
2146 accepts values, then it must be the last option in the
2147 combination; the value is the next argument. So, for example, using
2148 the following grammar:
2149 ((apples (single-char #\a))
2150 (blimps (single-char #\b) (value #t))
2151 (catalexis (single-char #\c) (value #t)))
2152 the following argument lists would be acceptable:
2153 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
2154 for "blimps" and "catalexis")
2155 ("-ab" "bang" "-c" "couth") (same)
2156 ("-ac" "couth" "-b" "bang") (same)
2157 ("-abc" "couth" "bang") (an error, since `-b' is not the
2158 last option in its combination)
2160 If an option's value is optional, then `getopt-long' decides
2161 whether it has a value by looking at what follows it in ARGS. If
2162 the next element is a string, and it does not appear to be an
2163 option itself, then that string is the option's value.
2165 The value of a long option can appear as the next element in ARGS,
2166 or it can follow the option name, separated by an `=' character.
2167 Thus, using the same grammar as above, the following argument lists
2169 ("--apples" "Braeburn" "--blimps" "Goodyear")
2170 ("--apples=Braeburn" "--blimps" "Goodyear")
2171 ("--blimps" "Goodyear" "--apples=Braeburn")
2173 If the option "--" appears in ARGS, argument parsing stops there;
2174 subsequent arguments are returned as ordinary arguments, even if
2175 they resemble options. So, in the argument list:
2176 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
2177 `getopt-long' will recognize the `apples' option as having the
2178 value "Granny Smith", but it will not recognize the `blimp'
2179 option; it will return the strings "--blimp" and "Goodyear" as
2180 ordinary argument strings.
2182 The `getopt-long' function returns the parsed argument list as an
2183 assocation list, mapping option names --- the symbols from GRAMMAR
2184 --- onto their values, or #t if the option does not accept a value.
2185 Unused options do not appear in the alist.
2187 All arguments that are not the value of any option are returned
2188 as a list, associated with the empty list.
2190 `getopt-long' throws an exception if:
2191 - it finds an unrecognized option in ARGS
2192 - a required option is omitted
2193 - an option that requires an argument doesn't get one
2194 - an option that doesn't accept an argument does get one (this can
2195 only happen using the long option `--opt=value' syntax)
2196 - an option predicate fails
2201 `((lockfile-dir (required? #t)
2204 (predicate ,file-is-directory?))
2205 (verbose (required? #f)
2208 (x-includes (single-char #\x))
2209 (rnet-server (single-char #\y)
2210 (predicate ,string?))))
2212 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
2213 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
2215 => ((() "foo1" "-fred" "foo2" "foo3")
2216 (rnet-server . "lamprod")
2217 (x-includes . "/usr/include")
2218 (lockfile-dir . "/tmp")
2221 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
2223 It will be removed in a few releases.
2225 ** New syntax: lambda*
2226 ** New syntax: define*
2227 ** New syntax: define*-public
2228 ** New syntax: defmacro*
2229 ** New syntax: defmacro*-public
2230 Guile now supports optional arguments.
2232 `lambda*', `define*', `define*-public', `defmacro*' and
2233 `defmacro*-public' are identical to the non-* versions except that
2234 they use an extended type of parameter list that has the following BNF
2235 syntax (parentheses are literal, square brackets indicate grouping,
2236 and `*', `+' and `?' have the usual meaning):
2238 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
2239 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
2240 [[#&rest identifier]|[. identifier]]? ) | [identifier]
2242 ext-var-decl ::= identifier | ( identifier expression )
2244 The semantics are best illustrated with the following documentation
2245 and examples for `lambda*':
2248 lambda extended for optional and keyword arguments
2250 lambda* creates a procedure that takes optional arguments. These
2251 are specified by putting them inside brackets at the end of the
2252 paramater list, but before any dotted rest argument. For example,
2253 (lambda* (a b #&optional c d . e) '())
2254 creates a procedure with fixed arguments a and b, optional arguments c
2255 and d, and rest argument e. If the optional arguments are omitted
2256 in a call, the variables for them are unbound in the procedure. This
2257 can be checked with the bound? macro.
2259 lambda* can also take keyword arguments. For example, a procedure
2261 (lambda* (#&key xyzzy larch) '())
2262 can be called with any of the argument lists (#:xyzzy 11)
2263 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
2264 are given as keywords are bound to values.
2266 Optional and keyword arguments can also be given default values
2267 which they take on when they are not present in a call, by giving a
2268 two-item list in place of an optional argument, for example in:
2269 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
2270 foo is a fixed argument, bar is an optional argument with default
2271 value 42, and baz is a keyword argument with default value 73.
2272 Default value expressions are not evaluated unless they are needed
2273 and until the procedure is called.
2275 lambda* now supports two more special parameter list keywords.
2277 lambda*-defined procedures now throw an error by default if a
2278 keyword other than one of those specified is found in the actual
2279 passed arguments. However, specifying #&allow-other-keys
2280 immediately after the kyword argument declarations restores the
2281 previous behavior of ignoring unknown keywords. lambda* also now
2282 guarantees that if the same keyword is passed more than once, the
2283 last one passed is the one that takes effect. For example,
2284 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
2285 #:heads 37 #:tails 42 #:heads 99)
2286 would result in (99 47) being displayed.
2288 #&rest is also now provided as a synonym for the dotted syntax rest
2289 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
2290 all respects to lambda*. This is provided for more similarity to DSSSL,
2291 MIT-Scheme and Kawa among others, as well as for refugees from other
2294 Further documentation may be found in the optargs.scm file itself.
2296 The optional argument module also exports the macros `let-optional',
2297 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
2298 are not documented here because they may be removed in the future, but
2299 full documentation is still available in optargs.scm.
2301 ** New syntax: and-let*
2302 Guile now supports the `and-let*' form, described in the draft SRFI-2.
2304 Syntax: (land* (<clause> ...) <body> ...)
2305 Each <clause> should have one of the following forms:
2306 (<variable> <expression>)
2309 Each <variable> or <bound-variable> should be an identifier. Each
2310 <expression> should be a valid expression. The <body> should be a
2311 possibly empty sequence of expressions, like the <body> of a
2314 Semantics: A LAND* expression is evaluated by evaluating the
2315 <expression> or <bound-variable> of each of the <clause>s from
2316 left to right. The value of the first <expression> or
2317 <bound-variable> that evaluates to a false value is returned; the
2318 remaining <expression>s and <bound-variable>s are not evaluated.
2319 The <body> forms are evaluated iff all the <expression>s and
2320 <bound-variable>s evaluate to true values.
2322 The <expression>s and the <body> are evaluated in an environment
2323 binding each <variable> of the preceding (<variable> <expression>)
2324 clauses to the value of the <expression>. Later bindings
2325 shadow earlier bindings.
2327 Guile's and-let* macro was contributed by Michael Livshin.
2329 ** New sorting functions
2331 *** New function: sorted? SEQUENCE LESS?
2332 Returns `#t' when the sequence argument is in non-decreasing order
2333 according to LESS? (that is, there is no adjacent pair `... x y
2334 ...' for which `(less? y x)').
2336 Returns `#f' when the sequence contains at least one out-of-order
2337 pair. It is an error if the sequence is neither a list nor a
2340 *** New function: merge LIST1 LIST2 LESS?
2341 LIST1 and LIST2 are sorted lists.
2342 Returns the sorted list of all elements in LIST1 and LIST2.
2344 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
2345 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
2346 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
2347 (Here "<" should read "comes before".)
2349 *** New procedure: merge! LIST1 LIST2 LESS?
2350 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
2351 the result. If the code is compiled, and LESS? constructs no new
2352 pairs, no pairs at all will be allocated. The first pair of the
2353 result will be either the first pair of LIST1 or the first pair of
2356 *** New function: sort SEQUENCE LESS?
2357 Accepts either a list or a vector, and returns a new sequence
2358 which is sorted. The new sequence is the same type as the input.
2359 Always `(sorted? (sort sequence less?) less?)'. The original
2360 sequence is not altered in any way. The new sequence shares its
2361 elements with the old one; no elements are copied.
2363 *** New procedure: sort! SEQUENCE LESS
2364 Returns its sorted result in the original boxes. No new storage is
2365 allocated at all. Proper usage: (set! slist (sort! slist <))
2367 *** New function: stable-sort SEQUENCE LESS?
2368 Similar to `sort' but stable. That is, if "equal" elements are
2369 ordered a < b in the original sequence, they will have the same order
2372 *** New function: stable-sort! SEQUENCE LESS?
2373 Similar to `sort!' but stable.
2374 Uses temporary storage when sorting vectors.
2376 *** New functions: sort-list, sort-list!
2377 Added for compatibility with scsh.
2379 ** New built-in random number support
2381 *** New function: random N [STATE]
2382 Accepts a positive integer or real N and returns a number of the
2383 same type between zero (inclusive) and N (exclusive). The values
2384 returned have a uniform distribution.
2386 The optional argument STATE must be of the type produced by
2387 `copy-random-state' or `seed->random-state'. It defaults to the value
2388 of the variable `*random-state*'. This object is used to maintain the
2389 state of the pseudo-random-number generator and is altered as a side
2390 effect of the `random' operation.
2392 *** New variable: *random-state*
2393 Holds a data structure that encodes the internal state of the
2394 random-number generator that `random' uses by default. The nature
2395 of this data structure is implementation-dependent. It may be
2396 printed out and successfully read back in, but may or may not
2397 function correctly as a random-number state object in another
2400 *** New function: copy-random-state [STATE]
2401 Returns a new object of type suitable for use as the value of the
2402 variable `*random-state*' and as a second argument to `random'.
2403 If argument STATE is given, a copy of it is returned. Otherwise a
2404 copy of `*random-state*' is returned.
2406 *** New function: seed->random-state SEED
2407 Returns a new object of type suitable for use as the value of the
2408 variable `*random-state*' and as a second argument to `random'.
2409 SEED is a string or a number. A new state is generated and
2410 initialized using SEED.
2412 *** New function: random:uniform [STATE]
2413 Returns an uniformly distributed inexact real random number in the
2414 range between 0 and 1.
2416 *** New procedure: random:solid-sphere! VECT [STATE]
2417 Fills VECT with inexact real random numbers the sum of whose
2418 squares is less than 1.0. Thinking of VECT as coordinates in
2419 space of dimension N = `(vector-length VECT)', the coordinates are
2420 uniformly distributed within the unit N-shere. The sum of the
2421 squares of the numbers is returned. VECT can be either a vector
2422 or a uniform vector of doubles.
2424 *** New procedure: random:hollow-sphere! VECT [STATE]
2425 Fills VECT with inexact real random numbers the sum of whose squares
2426 is equal to 1.0. Thinking of VECT as coordinates in space of
2427 dimension n = `(vector-length VECT)', the coordinates are uniformly
2428 distributed over the surface of the unit n-shere. VECT can be either
2429 a vector or a uniform vector of doubles.
2431 *** New function: random:normal [STATE]
2432 Returns an inexact real in a normal distribution with mean 0 and
2433 standard deviation 1. For a normal distribution with mean M and
2434 standard deviation D use `(+ M (* D (random:normal)))'.
2436 *** New procedure: random:normal-vector! VECT [STATE]
2437 Fills VECT with inexact real random numbers which are independent and
2438 standard normally distributed (i.e., with mean 0 and variance 1).
2439 VECT can be either a vector or a uniform vector of doubles.
2441 *** New function: random:exp STATE
2442 Returns an inexact real in an exponential distribution with mean 1.
2443 For an exponential distribution with mean U use (* U (random:exp)).
2445 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
2447 These functions now operate on numbers in the range of a C unsigned
2450 These functions used to operate on numbers in the range of a C signed
2451 long; however, this seems inappropriate, because Guile integers don't
2454 ** New function: make-guardian
2455 This is an implementation of guardians as described in
2456 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
2457 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
2458 Programming Language Design and Implementation, June 1993
2459 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
2461 ** New functions: delq1!, delv1!, delete1!
2462 These procedures behave similar to delq! and friends but delete only
2463 one object if at all.
2465 ** New function: unread-string STRING PORT
2466 Unread STRING to PORT, that is, push it back onto the port so that
2467 next read operation will work on the pushed back characters.
2469 ** unread-char can now be called multiple times
2470 If unread-char is called multiple times, the unread characters will be
2471 read again in last-in first-out order.
2473 ** the procedures uniform-array-read! and uniform-array-write! now
2474 work on any kind of port, not just ports which are open on a file.
2476 ** Now 'l' in a port mode requests line buffering.
2478 ** The procedure truncate-file now works on string ports as well
2479 as file ports. If the size argument is omitted, the current
2480 file position is used.
2482 ** new procedure: seek PORT/FDES OFFSET WHENCE
2483 The arguments are the same as for the old fseek procedure, but it
2484 works on string ports as well as random-access file ports.
2486 ** the fseek procedure now works on string ports, since it has been
2487 redefined using seek.
2489 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
2490 size is not supplied.
2492 ** the newline procedure no longer flushes the port if it's not
2493 line-buffered: previously it did if it was the current output port.
2495 ** open-pipe and close-pipe are no longer primitive procedures, but
2496 an emulation can be obtained using `(use-modules (ice-9 popen))'.
2498 ** the freopen procedure has been removed.
2500 ** new procedure: drain-input PORT
2501 Drains PORT's read buffers (including any pushed-back characters)
2502 and returns the contents as a single string.
2504 ** New function: map-in-order PROC LIST1 LIST2 ...
2505 Version of `map' which guarantees that the procedure is applied to the
2506 lists in serial order.
2508 ** Renamed `serial-array-copy!' and `serial-array-map!' to
2509 `array-copy-in-order!' and `array-map-in-order!'. The old names are
2510 now obsolete and will go away in release 1.5.
2512 ** New syntax: collect BODY1 ...
2513 Version of `begin' which returns a list of the results of the body
2514 forms instead of the result of the last body form. In contrast to
2515 `begin', `collect' allows an empty body.
2517 ** New functions: read-history FILENAME, write-history FILENAME
2518 Read/write command line history from/to file. Returns #t on success
2519 and #f if an error occured.
2521 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
2523 These procedures return a list of definitions available in the specified
2524 argument, a relative module reference. In the case of no argument,
2525 `(current-module)' is now consulted for definitions to return, instead
2526 of simply returning #f, the former behavior.
2528 ** The #/ syntax for lists is no longer supported.
2530 Earlier versions of Scheme accepted this syntax, but printed a
2533 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
2535 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
2538 * Changes to the gh_ interface
2542 Now takes a second argument which is the result array. If this
2543 pointer is NULL, a new array is malloced (the old behaviour).
2545 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
2546 gh_scm2shorts, gh_scm2longs, gh_scm2floats
2550 * Changes to the scm_ interface
2552 ** Function: scm_make_named_hook (char* name, int n_args)
2554 Creates a hook in the same way as make-hook above but also
2555 binds a variable named NAME to it.
2557 This is the typical way of creating a hook from C code.
2559 Currently, the variable is created in the "current" module. This
2560 might change when we get the new module system.
2562 ** The smob interface
2564 The interface for creating smobs has changed. For documentation, see
2565 data-rep.info (made from guile-core/doc/data-rep.texi).
2567 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
2569 >>> This function will be removed in 1.3.4. <<<
2573 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
2574 This function adds a new smob type, named NAME, with instance size
2575 SIZE to the system. The return value is a tag that is used in
2576 creating instances of the type. If SIZE is 0, then no memory will
2577 be allocated when instances of the smob are created, and nothing
2578 will be freed by the default free function.
2580 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
2581 This function sets the smob marking procedure for the smob type
2582 specified by the tag TC. TC is the tag returned by
2583 `scm_make_smob_type'.
2585 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
2586 This function sets the smob freeing procedure for the smob type
2587 specified by the tag TC. TC is the tag returned by
2588 `scm_make_smob_type'.
2590 *** Function: void scm_set_smob_print (tc, print)
2592 - Function: void scm_set_smob_print (long tc,
2593 scm_sizet (*print) (SCM,
2597 This function sets the smob printing procedure for the smob type
2598 specified by the tag TC. TC is the tag returned by
2599 `scm_make_smob_type'.
2601 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
2602 This function sets the smob equality-testing predicate for the
2603 smob type specified by the tag TC. TC is the tag returned by
2604 `scm_make_smob_type'.
2606 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
2607 Make VALUE contain a smob instance of the type with type code TC and
2608 smob data DATA. VALUE must be previously declared as C type `SCM'.
2610 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
2611 This macro expands to a block of code that creates a smob instance
2612 of the type with type code TC and smob data DATA, and returns that
2613 `SCM' value. It should be the last piece of code in a block.
2615 ** The interfaces for using I/O ports and implementing port types
2616 (ptobs) have changed significantly. The new interface is based on
2617 shared access to buffers and a new set of ptob procedures.
2619 *** scm_newptob has been removed
2623 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
2625 - Function: SCM scm_make_port_type (char *type_name,
2626 int (*fill_buffer) (SCM port),
2627 void (*write_flush) (SCM port));
2629 Similarly to the new smob interface, there is a set of function
2630 setters by which the user can customize the behaviour of his port
2631 type. See ports.h (scm_set_port_XXX).
2633 ** scm_strport_to_string: New function: creates a new string from
2634 a string port's buffer.
2636 ** Plug in interface for random number generators
2637 The variable `scm_the_rng' in random.c contains a value and three
2638 function pointers which together define the current random number
2639 generator being used by the Scheme level interface and the random
2640 number library functions.
2642 The user is free to replace the default generator with the generator
2645 *** Variable: size_t scm_the_rng.rstate_size
2646 The size of the random state type used by the current RNG
2649 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
2650 Given the random STATE, return 32 random bits.
2652 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
2653 Seed random state STATE using string S of length N.
2655 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
2656 Given random state STATE, return a malloced copy.
2659 The default RNG is the MWC (Multiply With Carry) random number
2660 generator described by George Marsaglia at the Department of
2661 Statistics and Supercomputer Computations Research Institute, The
2662 Florida State University (http://stat.fsu.edu/~geo).
2664 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
2665 passes all tests in the DIEHARD test suite
2666 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
2667 costs one multiply and one add on platforms which either supports long
2668 longs (gcc does this on most systems) or have 64 bit longs. The cost
2669 is four multiply on other systems but this can be optimized by writing
2670 scm_i_uniform32 in assembler.
2672 These functions are provided through the scm_the_rng interface for use
2673 by libguile and the application.
2675 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
2676 Given the random STATE, return 32 random bits.
2677 Don't use this function directly. Instead go through the plugin
2678 interface (see "Plug in interface" above).
2680 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
2681 Initialize STATE using SEED of length N.
2683 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
2684 Return a malloc:ed copy of STATE. This function can easily be re-used
2685 in the interfaces to other RNGs.
2687 ** Random number library functions
2688 These functions use the current RNG through the scm_the_rng interface.
2689 It might be a good idea to use these functions from your C code so
2690 that only one random generator is used by all code in your program.
2692 The default random state is stored in:
2694 *** Variable: SCM scm_var_random_state
2695 Contains the vcell of the Scheme variable "*random-state*" which is
2696 used as default state by all random number functions in the Scheme
2701 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
2703 *** Function: scm_rstate *scm_c_default_rstate (void)
2704 This is a convenience function which returns the value of
2705 scm_var_random_state. An error message is generated if this value
2706 isn't a random state.
2708 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
2709 Make a new random state from the string SEED of length LENGTH.
2711 It is generally not a good idea to use multiple random states in a
2712 program. While subsequent random numbers generated from one random
2713 state are guaranteed to be reasonably independent, there is no such
2714 guarantee for numbers generated from different random states.
2716 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
2717 Return 32 random bits.
2719 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
2720 Return a sample from the uniform(0,1) distribution.
2722 *** Function: double scm_c_normal01 (scm_rstate *STATE)
2723 Return a sample from the normal(0,1) distribution.
2725 *** Function: double scm_c_exp1 (scm_rstate *STATE)
2726 Return a sample from the exp(1) distribution.
2728 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
2729 Return a sample from the discrete uniform(0,M) distribution.
2731 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
2732 Return a sample from the discrete uniform(0,M) distribution.
2733 M must be a bignum object. The returned value may be an INUM.
2737 Changes in Guile 1.3 (released Monday, October 19, 1998):
2739 * Changes to the distribution
2741 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
2742 To avoid conflicts, programs should name environment variables after
2743 themselves, except when there's a common practice establishing some
2746 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
2747 giving the former precedence, and printing a warning message if the
2748 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
2750 ** The header files related to multi-byte characters have been removed.
2751 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
2752 which referred to these explicitly will probably need to be rewritten,
2753 since the support for the variant string types has been removed; see
2756 ** The header files append.h and sequences.h have been removed. These
2757 files implemented non-R4RS operations which would encourage
2758 non-portable programming style and less easy-to-read code.
2760 * Changes to the stand-alone interpreter
2762 ** New procedures have been added to implement a "batch mode":
2764 *** Function: batch-mode?
2766 Returns a boolean indicating whether the interpreter is in batch
2769 *** Function: set-batch-mode?! ARG
2771 If ARG is true, switches the interpreter to batch mode. The `#f'
2772 case has not been implemented.
2774 ** Guile now provides full command-line editing, when run interactively.
2775 To use this feature, you must have the readline library installed.
2776 The Guile build process will notice it, and automatically include
2779 The readline library is available via anonymous FTP from any GNU
2780 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
2782 ** the-last-stack is now a fluid.
2784 * Changes to the procedure for linking libguile with your programs
2786 ** You can now use the `guile-config' utility to build programs that use Guile.
2788 Guile now includes a command-line utility called `guile-config', which
2789 can provide information about how to compile and link programs that
2792 *** `guile-config compile' prints any C compiler flags needed to use Guile.
2793 You should include this command's output on the command line you use
2794 to compile C or C++ code that #includes the Guile header files. It's
2795 usually just a `-I' flag to help the compiler find the Guile headers.
2798 *** `guile-config link' prints any linker flags necessary to link with Guile.
2800 This command writes to its standard output a list of flags which you
2801 must pass to the linker to link your code against the Guile library.
2802 The flags include '-lguile' itself, any other libraries the Guile
2803 library depends upon, and any `-L' flags needed to help the linker
2804 find those libraries.
2806 For example, here is a Makefile rule that builds a program named 'foo'
2807 from the object files ${FOO_OBJECTS}, and links them against Guile:
2810 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
2812 Previous Guile releases recommended that you use autoconf to detect
2813 which of a predefined set of libraries were present on your system.
2814 It is more robust to use `guile-config', since it records exactly which
2815 libraries the installed Guile library requires.
2817 This was originally called `build-guile', but was renamed to
2818 `guile-config' before Guile 1.3 was released, to be consistent with
2819 the analogous script for the GTK+ GUI toolkit, which is called
2823 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
2825 If you are using the GNU autoconf package to configure your program,
2826 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
2827 (described above) and gather the necessary values for use in your
2830 The GUILE_FLAGS macro expands to configure script code which runs the
2831 `guile-config' script, to find out where Guile's header files and
2832 libraries are installed. It sets two variables, marked for
2833 substitution, as by AC_SUBST.
2835 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
2836 code that uses Guile header files. This is almost always just a
2839 GUILE_LDFLAGS --- flags to pass to the linker to link a
2840 program against Guile. This includes `-lguile' for the Guile
2841 library itself, any libraries that Guile itself requires (like
2842 -lqthreads), and so on. It may also include a -L flag to tell the
2843 compiler where to find the libraries.
2845 GUILE_FLAGS is defined in the file guile.m4, in the top-level
2846 directory of the Guile distribution. You can copy it into your
2847 package's aclocal.m4 file, and then use it in your configure.in file.
2849 If you are using the `aclocal' program, distributed with GNU automake,
2850 to maintain your aclocal.m4 file, the Guile installation process
2851 installs guile.m4 where aclocal will find it. All you need to do is
2852 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
2853 this will copy the definition of GUILE_FLAGS into your aclocal.m4
2857 * Changes to Scheme functions and syntax
2859 ** Multi-byte strings have been removed, as have multi-byte and wide
2860 ports. We felt that these were the wrong approach to
2861 internationalization support.
2863 ** New function: readline [PROMPT]
2864 Read a line from the terminal, and allow the user to edit it,
2865 prompting with PROMPT. READLINE provides a large set of Emacs-like
2866 editing commands, lets the user recall previously typed lines, and
2867 works on almost every kind of terminal, including dumb terminals.
2869 READLINE assumes that the cursor is at the beginning of the line when
2870 it is invoked. Thus, you can't print a prompt yourself, and then call
2871 READLINE; you need to package up your prompt as a string, pass it to
2872 the function, and let READLINE print the prompt itself. This is
2873 because READLINE needs to know the prompt's screen width.
2875 For Guile to provide this function, you must have the readline
2876 library, version 2.1 or later, installed on your system. Readline is
2877 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
2878 any GNU mirror site.
2880 See also ADD-HISTORY function.
2882 ** New function: add-history STRING
2883 Add STRING as the most recent line in the history used by the READLINE
2884 command. READLINE does not add lines to the history itself; you must
2885 call ADD-HISTORY to make previous input available to the user.
2887 ** The behavior of the read-line function has changed.
2889 This function now uses standard C library functions to read the line,
2890 for speed. This means that it doesn not respect the value of
2891 scm-line-incrementors; it assumes that lines are delimited with
2894 (Note that this is read-line, the function that reads a line of text
2895 from a port, not readline, the function that reads a line from a
2896 terminal, providing full editing capabilities.)
2898 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
2900 This module provides some simple argument parsing. It exports one
2903 Function: getopt-gnu-style ARG-LS
2904 Parse a list of program arguments into an alist of option
2907 Each item in the list of program arguments is examined to see if
2908 it meets the syntax of a GNU long-named option. An argument like
2909 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
2910 returned alist, where MUMBLE is a keyword object with the same
2911 name as the argument. An argument like `--MUMBLE=FROB' produces
2912 an element of the form (MUMBLE . FROB), where FROB is a string.
2914 As a special case, the returned alist also contains a pair whose
2915 car is the symbol `rest'. The cdr of this pair is a list
2916 containing all the items in the argument list that are not options
2917 of the form mentioned above.
2919 The argument `--' is treated specially: all items in the argument
2920 list appearing after such an argument are not examined, and are
2921 returned in the special `rest' list.
2923 This function does not parse normal single-character switches.
2924 You will need to parse them out of the `rest' list yourself.
2926 ** The read syntax for byte vectors and short vectors has changed.
2928 Instead of #bytes(...), write #y(...).
2930 Instead of #short(...), write #h(...).
2932 This may seem nutty, but, like the other uniform vectors, byte vectors
2933 and short vectors want to have the same print and read syntax (and,
2934 more basic, want to have read syntax!). Changing the read syntax to
2935 use multiple characters after the hash sign breaks with the
2936 conventions used in R5RS and the conventions used for the other
2937 uniform vectors. It also introduces complexity in the current reader,
2938 both on the C and Scheme levels. (The Right solution is probably to
2939 change the syntax and prototypes for uniform vectors entirely.)
2942 ** The new module (ice-9 session) provides useful interactive functions.
2944 *** New procedure: (apropos REGEXP OPTION ...)
2946 Display a list of top-level variables whose names match REGEXP, and
2947 the modules they are imported from. Each OPTION should be one of the
2950 value --- Show the value of each matching variable.
2951 shadow --- Show bindings shadowed by subsequently imported modules.
2952 full --- Same as both `shadow' and `value'.
2956 guile> (apropos "trace" 'full)
2957 debug: trace #<procedure trace args>
2958 debug: untrace #<procedure untrace args>
2959 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
2960 the-scm-module: before-backtrace-hook ()
2961 the-scm-module: backtrace #<primitive-procedure backtrace>
2962 the-scm-module: after-backtrace-hook ()
2963 the-scm-module: has-shown-backtrace-hint? #f
2966 ** There are new functions and syntax for working with macros.
2968 Guile implements macros as a special object type. Any variable whose
2969 top-level binding is a macro object acts as a macro. The macro object
2970 specifies how the expression should be transformed before evaluation.
2972 *** Macro objects now print in a reasonable way, resembling procedures.
2974 *** New function: (macro? OBJ)
2975 True iff OBJ is a macro object.
2977 *** New function: (primitive-macro? OBJ)
2978 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
2979 macro transformers, implemented in eval.c rather than Scheme code.
2981 Why do we have this function?
2982 - For symmetry with procedure? and primitive-procedure?,
2983 - to allow custom print procedures to tell whether a macro is
2984 primitive, and display it differently, and
2985 - to allow compilers and user-written evaluators to distinguish
2986 builtin special forms from user-defined ones, which could be
2989 *** New function: (macro-type OBJ)
2990 Return a value indicating what kind of macro OBJ is. Possible return
2993 The symbol `syntax' --- a macro created by procedure->syntax.
2994 The symbol `macro' --- a macro created by procedure->macro.
2995 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
2996 The boolean #f --- if OBJ is not a macro object.
2998 *** New function: (macro-name MACRO)
2999 Return the name of the macro object MACRO's procedure, as returned by
3002 *** New function: (macro-transformer MACRO)
3003 Return the transformer procedure for MACRO.
3005 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
3007 Specify a new macro expander to use in the current module. Each
3008 MODULE is a module name, with the same meaning as in the `use-modules'
3009 form; each named module's exported bindings are added to the current
3010 top-level environment. TRANSFORMER is an expression evaluated in the
3011 resulting environment which must yield a procedure to use as the
3012 module's eval transformer: every expression evaluated in this module
3013 is passed to this function, and the result passed to the Guile
3016 *** macro-eval! is removed. Use local-eval instead.
3018 ** Some magic has been added to the printer to better handle user
3019 written printing routines (like record printers, closure printers).
3021 The problem is that these user written routines must have access to
3022 the current `print-state' to be able to handle fancy things like
3023 detection of circular references. These print-states have to be
3024 passed to the builtin printing routines (display, write, etc) to
3025 properly continue the print chain.
3027 We didn't want to change all existing print code so that it
3028 explicitly passes thru a print state in addition to a port. Instead,
3029 we extented the possible values that the builtin printing routines
3030 accept as a `port'. In addition to a normal port, they now also take
3031 a pair of a normal port and a print-state. Printing will go to the
3032 port and the print-state will be used to control the detection of
3033 circular references, etc. If the builtin function does not care for a
3034 print-state, it is simply ignored.
3036 User written callbacks are now called with such a pair as their
3037 `port', but because every function now accepts this pair as a PORT
3038 argument, you don't have to worry about that. In fact, it is probably
3039 safest to not check for these pairs.
3041 However, it is sometimes necessary to continue a print chain on a
3042 different port, for example to get a intermediate string
3043 representation of the printed value, mangle that string somehow, and
3044 then to finally print the mangled string. Use the new function
3046 inherit-print-state OLD-PORT NEW-PORT
3048 for this. It constructs a new `port' that prints to NEW-PORT but
3049 inherits the print-state of OLD-PORT.
3051 ** struct-vtable-offset renamed to vtable-offset-user
3053 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
3055 ** There is now a third optional argument to make-vtable-vtable
3056 (and fourth to make-struct) when constructing new types (vtables).
3057 This argument initializes field vtable-index-printer of the vtable.
3059 ** The detection of circular references has been extended to structs.
3060 That is, a structure that -- in the process of being printed -- prints
3061 itself does not lead to infinite recursion.
3063 ** There is now some basic support for fluids. Please read
3064 "libguile/fluid.h" to find out more. It is accessible from Scheme with
3065 the following functions and macros:
3067 Function: make-fluid
3069 Create a new fluid object. Fluids are not special variables or
3070 some other extension to the semantics of Scheme, but rather
3071 ordinary Scheme objects. You can store them into variables (that
3072 are still lexically scoped, of course) or into any other place you
3073 like. Every fluid has a initial value of `#f'.
3075 Function: fluid? OBJ
3077 Test whether OBJ is a fluid.
3079 Function: fluid-ref FLUID
3080 Function: fluid-set! FLUID VAL
3082 Access/modify the fluid FLUID. Modifications are only visible
3083 within the current dynamic root (that includes threads).
3085 Function: with-fluids* FLUIDS VALUES THUNK
3087 FLUIDS is a list of fluids and VALUES a corresponding list of
3088 values for these fluids. Before THUNK gets called the values are
3089 installed in the fluids and the old values of the fluids are
3090 saved in the VALUES list. When the flow of control leaves THUNK
3091 or reenters it, the values get swapped again. You might think of
3092 this as a `safe-fluid-excursion'. Note that the VALUES list is
3093 modified by `with-fluids*'.
3095 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
3097 The same as `with-fluids*' but with a different syntax. It looks
3098 just like `let', but both FLUID and VALUE are evaluated. Remember,
3099 fluids are not special variables but ordinary objects. FLUID
3100 should evaluate to a fluid.
3102 ** Changes to system call interfaces:
3104 *** close-port, close-input-port and close-output-port now return a
3105 boolean instead of an `unspecified' object. #t means that the port
3106 was successfully closed, while #f means it was already closed. It is
3107 also now possible for these procedures to raise an exception if an
3108 error occurs (some errors from write can be delayed until close.)
3110 *** the first argument to chmod, fcntl, ftell and fseek can now be a
3113 *** the third argument to fcntl is now optional.
3115 *** the first argument to chown can now be a file descriptor or a port.
3117 *** the argument to stat can now be a port.
3119 *** The following new procedures have been added (most use scsh
3122 *** procedure: close PORT/FD
3123 Similar to close-port (*note close-port: Closing Ports.), but also
3124 works on file descriptors. A side effect of closing a file
3125 descriptor is that any ports using that file descriptor are moved
3126 to a different file descriptor and have their revealed counts set
3129 *** procedure: port->fdes PORT
3130 Returns the integer file descriptor underlying PORT. As a side
3131 effect the revealed count of PORT is incremented.
3133 *** procedure: fdes->ports FDES
3134 Returns a list of existing ports which have FDES as an underlying
3135 file descriptor, without changing their revealed counts.
3137 *** procedure: fdes->inport FDES
3138 Returns an existing input port which has FDES as its underlying
3139 file descriptor, if one exists, and increments its revealed count.
3140 Otherwise, returns a new input port with a revealed count of 1.
3142 *** procedure: fdes->outport FDES
3143 Returns an existing output port which has FDES as its underlying
3144 file descriptor, if one exists, and increments its revealed count.
3145 Otherwise, returns a new output port with a revealed count of 1.
3147 The next group of procedures perform a `dup2' system call, if NEWFD
3148 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
3149 duplicated can be supplied as an integer or contained in a port. The
3150 type of value returned varies depending on which procedure is used.
3152 All procedures also have the side effect when performing `dup2' that
3153 any ports using NEWFD are moved to a different file descriptor and have
3154 their revealed counts set to zero.
3156 *** procedure: dup->fdes PORT/FD [NEWFD]
3157 Returns an integer file descriptor.
3159 *** procedure: dup->inport PORT/FD [NEWFD]
3160 Returns a new input port using the new file descriptor.
3162 *** procedure: dup->outport PORT/FD [NEWFD]
3163 Returns a new output port using the new file descriptor.
3165 *** procedure: dup PORT/FD [NEWFD]
3166 Returns a new port if PORT/FD is a port, with the same mode as the
3167 supplied port, otherwise returns an integer file descriptor.
3169 *** procedure: dup->port PORT/FD MODE [NEWFD]
3170 Returns a new port using the new file descriptor. MODE supplies a
3171 mode string for the port (*note open-file: File Ports.).
3173 *** procedure: setenv NAME VALUE
3174 Modifies the environment of the current process, which is also the
3175 default environment inherited by child processes.
3177 If VALUE is `#f', then NAME is removed from the environment.
3178 Otherwise, the string NAME=VALUE is added to the environment,
3179 replacing any existing string with name matching NAME.
3181 The return value is unspecified.
3183 *** procedure: truncate-file OBJ SIZE
3184 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
3185 can be a string containing a file name or an integer file
3186 descriptor or port open for output on the file. The underlying
3187 system calls are `truncate' and `ftruncate'.
3189 The return value is unspecified.
3191 *** procedure: setvbuf PORT MODE [SIZE]
3192 Set the buffering mode for PORT. MODE can be:
3200 block buffered, using a newly allocated buffer of SIZE bytes.
3201 However if SIZE is zero or unspecified, the port will be made
3204 This procedure should not be used after I/O has been performed with
3207 Ports are usually block buffered by default, with a default buffer
3208 size. Procedures e.g., *Note open-file: File Ports, which accept a
3209 mode string allow `0' to be added to request an unbuffered port.
3211 *** procedure: fsync PORT/FD
3212 Copies any unwritten data for the specified output file descriptor
3213 to disk. If PORT/FD is a port, its buffer is flushed before the
3214 underlying file descriptor is fsync'd. The return value is
3217 *** procedure: open-fdes PATH FLAGS [MODES]
3218 Similar to `open' but returns a file descriptor instead of a port.
3220 *** procedure: execle PATH ENV [ARG] ...
3221 Similar to `execl', but the environment of the new process is
3222 specified by ENV, which must be a list of strings as returned by
3223 the `environ' procedure.
3225 This procedure is currently implemented using the `execve' system
3226 call, but we call it `execle' because of its Scheme calling
3229 *** procedure: strerror ERRNO
3230 Returns the Unix error message corresponding to ERRNO, an integer.
3232 *** procedure: primitive-exit [STATUS]
3233 Terminate the current process without unwinding the Scheme stack.
3234 This is would typically be useful after a fork. The exit status
3235 is STATUS if supplied, otherwise zero.
3237 *** procedure: times
3238 Returns an object with information about real and processor time.
3239 The following procedures accept such an object as an argument and
3240 return a selected component:
3243 The current real time, expressed as time units relative to an
3247 The CPU time units used by the calling process.
3250 The CPU time units used by the system on behalf of the
3254 The CPU time units used by terminated child processes of the
3255 calling process, whose status has been collected (e.g., using
3259 Similarly, the CPU times units used by the system on behalf of
3260 terminated child processes.
3262 ** Removed: list-length
3263 ** Removed: list-append, list-append!
3264 ** Removed: list-reverse, list-reverse!
3266 ** array-map renamed to array-map!
3268 ** serial-array-map renamed to serial-array-map!
3270 ** catch doesn't take #f as first argument any longer
3272 Previously, it was possible to pass #f instead of a key to `catch'.
3273 That would cause `catch' to pass a jump buffer object to the procedure
3274 passed as second argument. The procedure could then use this jump
3275 buffer objekt as an argument to throw.
3277 This mechanism has been removed since its utility doesn't motivate the
3278 extra complexity it introduces.
3280 ** The `#/' notation for lists now provokes a warning message from Guile.
3281 This syntax will be removed from Guile in the near future.
3283 To disable the warning message, set the GUILE_HUSH environment
3284 variable to any non-empty value.
3286 ** The newline character now prints as `#\newline', following the
3287 normal Scheme notation, not `#\nl'.
3289 * Changes to the gh_ interface
3291 ** The gh_enter function now takes care of loading the Guile startup files.
3292 gh_enter works by calling scm_boot_guile; see the remarks below.
3294 ** Function: void gh_write (SCM x)
3296 Write the printed representation of the scheme object x to the current
3297 output port. Corresponds to the scheme level `write'.
3299 ** gh_list_length renamed to gh_length.
3301 ** vector handling routines
3303 Several major changes. In particular, gh_vector() now resembles
3304 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
3305 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
3306 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
3307 vector-related gh_ functions have been implemented.
3309 ** pair and list routines
3311 Implemented several of the R4RS pair and list functions that were
3314 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
3316 New function. Converts double arrays back and forth between Scheme
3319 * Changes to the scm_ interface
3321 ** The function scm_boot_guile now takes care of loading the startup files.
3323 Guile's primary initialization function, scm_boot_guile, now takes
3324 care of loading `boot-9.scm', in the `ice-9' module, to initialize
3325 Guile, define the module system, and put together some standard
3326 bindings. It also loads `init.scm', which is intended to hold
3327 site-specific initialization code.
3329 Since Guile cannot operate properly until boot-9.scm is loaded, there
3330 is no reason to separate loading boot-9.scm from Guile's other
3331 initialization processes.
3333 This job used to be done by scm_compile_shell_switches, which didn't
3334 make much sense; in particular, it meant that people using Guile for
3335 non-shell-like applications had to jump through hoops to get Guile
3336 initialized properly.
3338 ** The function scm_compile_shell_switches no longer loads the startup files.
3339 Now, Guile always loads the startup files, whenever it is initialized;
3340 see the notes above for scm_boot_guile and scm_load_startup_files.
3342 ** Function: scm_load_startup_files
3343 This new function takes care of loading Guile's initialization file
3344 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
3345 this is always called by the Guile initialization process, it's
3346 probably not too useful to call this yourself, but it's there anyway.
3348 ** The semantics of smob marking have changed slightly.
3350 The smob marking function (the `mark' member of the scm_smobfuns
3351 structure) is no longer responsible for setting the mark bit on the
3352 smob. The generic smob handling code in the garbage collector will
3353 set this bit. The mark function need only ensure that any other
3354 objects the smob refers to get marked.
3356 Note that this change means that the smob's GC8MARK bit is typically
3357 already set upon entry to the mark function. Thus, marking functions
3358 which look like this:
3361 if (SCM_GC8MARKP (ptr))
3363 SCM_SETGC8MARK (ptr);
3364 ... mark objects to which the smob refers ...
3367 are now incorrect, since they will return early, and fail to mark any
3368 other objects the smob refers to. Some code in the Guile library used
3371 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
3373 If you have implemented your own I/O port type, by writing the
3374 functions required by the scm_ptobfuns and then calling scm_newptob,
3375 you will need to change your functions slightly.
3377 The functions in a scm_ptobfuns structure now expect the port itself
3378 as their argument; they used to expect the `stream' member of the
3379 port's scm_port_table structure. This allows functions in an
3380 scm_ptobfuns structure to easily access the port's cell (and any flags
3381 it its CAR), and the port's scm_port_table structure.
3383 Guile now passes the I/O port itself as the `port' argument in the
3384 following scm_ptobfuns functions:
3386 int (*free) (SCM port);
3387 int (*fputc) (int, SCM port);
3388 int (*fputs) (char *, SCM port);
3389 scm_sizet (*fwrite) SCM_P ((char *ptr,
3393 int (*fflush) (SCM port);
3394 int (*fgetc) (SCM port);
3395 int (*fclose) (SCM port);
3397 The interfaces to the `mark', `print', `equalp', and `fgets' methods
3400 If you have existing code which defines its own port types, it is easy
3401 to convert your code to the new interface; simply apply SCM_STREAM to
3402 the port argument to yield the value you code used to expect.
3404 Note that since both the port and the stream have the same type in the
3405 C code --- they are both SCM values --- the C compiler will not remind
3406 you if you forget to update your scm_ptobfuns functions.
3409 ** Function: int scm_internal_select (int fds,
3413 struct timeval *timeout);
3415 This is a replacement for the `select' function provided by the OS.
3416 It enables I/O blocking and sleeping to happen for one cooperative
3417 thread without blocking other threads. It also avoids busy-loops in
3418 these situations. It is intended that all I/O blocking and sleeping
3419 will finally go through this function. Currently, this function is
3420 only available on systems providing `gettimeofday' and `select'.
3422 ** Function: SCM scm_internal_stack_catch (SCM tag,
3423 scm_catch_body_t body,
3425 scm_catch_handler_t handler,
3428 A new sibling to the other two C level `catch' functions
3429 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
3430 the stack to be saved automatically into the variable `the-last-stack'
3431 (scm_the_last_stack_var) on error. This is necessary if you want to
3432 use advanced error reporting, such as calling scm_display_error and
3433 scm_display_backtrace. (They both take a stack object as argument.)
3435 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
3437 scm_catch_handler_t handler,
3440 Spawns a new thread. It does a job similar to
3441 scm_call_with_new_thread but takes arguments more suitable when
3442 spawning threads from application C code.
3444 ** The hook scm_error_callback has been removed. It was originally
3445 intended as a way for the user to install his own error handler. But
3446 that method works badly since it intervenes between throw and catch,
3447 thereby changing the semantics of expressions like (catch #t ...).
3448 The correct way to do it is to use one of the C level catch functions
3449 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
3451 ** Removed functions:
3453 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
3454 scm_list_reverse, scm_list_reverse_x
3456 ** New macros: SCM_LISTn where n is one of the integers 0-9.
3458 These can be used for pretty list creation from C. The idea is taken
3459 from Erick Gallesio's STk.
3461 ** scm_array_map renamed to scm_array_map_x
3463 ** mbstrings are now removed
3465 This means that the type codes scm_tc7_mb_string and
3466 scm_tc7_mb_substring has been removed.
3468 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
3470 Since we no longer support multi-byte strings, these I/O functions
3471 have been simplified, and renamed. Here are their old names, and
3472 their new names and arguments:
3474 scm_gen_putc -> void scm_putc (int c, SCM port);
3475 scm_gen_puts -> void scm_puts (char *s, SCM port);
3476 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
3477 scm_gen_getc -> void scm_getc (SCM port);
3480 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
3482 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
3484 SCM_TYP7S now masks away the bit which distinguishes substrings from
3487 ** scm_catch_body_t: Backward incompatible change!
3489 Body functions to scm_internal_catch and friends do not any longer
3490 take a second argument. This is because it is no longer possible to
3491 pass a #f arg to catch.
3493 ** Calls to scm_protect_object and scm_unprotect now nest properly.
3495 The function scm_protect_object protects its argument from being freed
3496 by the garbage collector. scm_unprotect_object removes that
3499 These functions now nest properly. That is, for every object O, there
3500 is a counter which scm_protect_object(O) increments and
3501 scm_unprotect_object(O) decrements, if the counter is greater than
3502 zero. Every object's counter is zero when it is first created. If an
3503 object's counter is greater than zero, the garbage collector will not
3504 reclaim its storage.
3506 This allows you to use scm_protect_object in your code without
3507 worrying that some other function you call will call
3508 scm_unprotect_object, and allow it to be freed. Assuming that the
3509 functions you call are well-behaved, and unprotect only those objects
3510 they protect, you can follow the same rule and have confidence that
3511 objects will be freed only at appropriate times.
3514 Changes in Guile 1.2 (released Tuesday, June 24 1997):
3516 * Changes to the distribution
3518 ** Nightly snapshots are now available from ftp.red-bean.com.
3519 The old server, ftp.cyclic.com, has been relinquished to its rightful
3522 Nightly snapshots of the Guile development sources are now available via
3523 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
3525 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
3526 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
3528 ** To run Guile without installing it, the procedure has changed a bit.
3530 If you used a separate build directory to compile Guile, you'll need
3531 to include the build directory in SCHEME_LOAD_PATH, as well as the
3532 source directory. See the `INSTALL' file for examples.
3534 * Changes to the procedure for linking libguile with your programs
3536 ** The standard Guile load path for Scheme code now includes
3537 $(datadir)/guile (usually /usr/local/share/guile). This means that
3538 you can install your own Scheme files there, and Guile will find them.
3539 (Previous versions of Guile only checked a directory whose name
3540 contained the Guile version number, so you had to re-install or move
3541 your Scheme sources each time you installed a fresh version of Guile.)
3543 The load path also includes $(datadir)/guile/site; we recommend
3544 putting individual Scheme files there. If you want to install a
3545 package with multiple source files, create a directory for them under
3548 ** Guile 1.2 will now use the Rx regular expression library, if it is
3549 installed on your system. When you are linking libguile into your own
3550 programs, this means you will have to link against -lguile, -lqt (if
3551 you configured Guile with thread support), and -lrx.
3553 If you are using autoconf to generate configuration scripts for your
3554 application, the following lines should suffice to add the appropriate
3555 libraries to your link command:
3557 ### Find Rx, quickthreads and libguile.
3558 AC_CHECK_LIB(rx, main)
3559 AC_CHECK_LIB(qt, main)
3560 AC_CHECK_LIB(guile, scm_shell)
3562 The Guile 1.2 distribution does not contain sources for the Rx
3563 library, as Guile 1.0 did. If you want to use Rx, you'll need to
3564 retrieve it from a GNU FTP site and install it separately.
3566 * Changes to Scheme functions and syntax
3568 ** The dynamic linking features of Guile are now enabled by default.
3569 You can disable them by giving the `--disable-dynamic-linking' option
3572 (dynamic-link FILENAME)
3574 Find the object file denoted by FILENAME (a string) and link it
3575 into the running Guile application. When everything works out,
3576 return a Scheme object suitable for representing the linked object
3577 file. Otherwise an error is thrown. How object files are
3578 searched is system dependent.
3580 (dynamic-object? VAL)
3582 Determine whether VAL represents a dynamically linked object file.
3584 (dynamic-unlink DYNOBJ)
3586 Unlink the indicated object file from the application. DYNOBJ
3587 should be one of the values returned by `dynamic-link'.
3589 (dynamic-func FUNCTION DYNOBJ)
3591 Search the C function indicated by FUNCTION (a string or symbol)
3592 in DYNOBJ and return some Scheme object that can later be used
3593 with `dynamic-call' to actually call this function. Right now,
3594 these Scheme objects are formed by casting the address of the
3595 function to `long' and converting this number to its Scheme
3598 (dynamic-call FUNCTION DYNOBJ)
3600 Call the C function indicated by FUNCTION and DYNOBJ. The
3601 function is passed no arguments and its return value is ignored.
3602 When FUNCTION is something returned by `dynamic-func', call that
3603 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
3604 etc.), look it up in DYNOBJ; this is equivalent to
3606 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
3608 Interrupts are deferred while the C function is executing (with
3609 SCM_DEFER_INTS/SCM_ALLOW_INTS).
3611 (dynamic-args-call FUNCTION DYNOBJ ARGS)
3613 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
3614 some arguments and return its return value. The C function is
3615 expected to take two arguments and return an `int', just like
3618 int c_func (int argc, char **argv);
3620 ARGS must be a list of strings and is converted into an array of
3621 `char *'. The array is passed in ARGV and its size in ARGC. The
3622 return value is converted to a Scheme number and returned from the
3623 call to `dynamic-args-call'.
3625 When dynamic linking is disabled or not supported on your system,
3626 the above functions throw errors, but they are still available.
3628 Here is a small example that works on GNU/Linux:
3630 (define libc-obj (dynamic-link "libc.so"))
3631 (dynamic-args-call 'rand libc-obj '())
3633 See the file `libguile/DYNAMIC-LINKING' for additional comments.
3635 ** The #/ syntax for module names is depreciated, and will be removed
3636 in a future version of Guile. Instead of
3644 The latter syntax is more consistent with existing Lisp practice.
3646 ** Guile now does fancier printing of structures. Structures are the
3647 underlying implementation for records, which in turn are used to
3648 implement modules, so all of these object now print differently and in
3649 a more informative way.
3651 The Scheme printer will examine the builtin variable *struct-printer*
3652 whenever it needs to print a structure object. When this variable is
3653 not `#f' it is deemed to be a procedure and will be applied to the
3654 structure object and the output port. When *struct-printer* is `#f'
3655 or the procedure return `#f' the structure object will be printed in
3656 the boring #<struct 80458270> form.
3658 This hook is used by some routines in ice-9/boot-9.scm to implement
3659 type specific printing routines. Please read the comments there about
3662 One of the more specific uses of structs are records. The printing
3663 procedure that could be passed to MAKE-RECORD-TYPE is now actually
3664 called. It should behave like a *struct-printer* procedure (described
3667 ** Guile now supports a new R4RS-compliant syntax for keywords. A
3668 token of the form #:NAME, where NAME has the same syntax as a Scheme
3669 symbol, is the external representation of the keyword named NAME.
3670 Keyword objects print using this syntax as well, so values containing
3671 keyword objects can be read back into Guile. When used in an
3672 expression, keywords are self-quoting objects.
3674 Guile suports this read syntax, and uses this print syntax, regardless
3675 of the current setting of the `keyword' read option. The `keyword'
3676 read option only controls whether Guile recognizes the `:NAME' syntax,
3677 which is incompatible with R4RS. (R4RS says such token represent
3680 ** Guile has regular expression support again. Guile 1.0 included
3681 functions for matching regular expressions, based on the Rx library.
3682 In Guile 1.1, the Guile/Rx interface was removed to simplify the
3683 distribution, and thus Guile had no regular expression support. Guile
3684 1.2 again supports the most commonly used functions, and supports all
3685 of SCSH's regular expression functions.
3687 If your system does not include a POSIX regular expression library,
3688 and you have not linked Guile with a third-party regexp library such as
3689 Rx, these functions will not be available. You can tell whether your
3690 Guile installation includes regular expression support by checking
3691 whether the `*features*' list includes the `regex' symbol.
3693 *** regexp functions
3695 By default, Guile supports POSIX extended regular expressions. That
3696 means that the characters `(', `)', `+' and `?' are special, and must
3697 be escaped if you wish to match the literal characters.
3699 This regular expression interface was modeled after that implemented
3700 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
3701 with SCSH regular expressions.
3703 **** Function: string-match PATTERN STR [START]
3704 Compile the string PATTERN into a regular expression and compare
3705 it with STR. The optional numeric argument START specifies the
3706 position of STR at which to begin matching.
3708 `string-match' returns a "match structure" which describes what,
3709 if anything, was matched by the regular expression. *Note Match
3710 Structures::. If STR does not match PATTERN at all,
3711 `string-match' returns `#f'.
3713 Each time `string-match' is called, it must compile its PATTERN
3714 argument into a regular expression structure. This operation is
3715 expensive, which makes `string-match' inefficient if the same regular
3716 expression is used several times (for example, in a loop). For better
3717 performance, you can compile a regular expression in advance and then
3718 match strings against the compiled regexp.
3720 **** Function: make-regexp STR [FLAGS]
3721 Compile the regular expression described by STR, and return the
3722 compiled regexp structure. If STR does not describe a legal
3723 regular expression, `make-regexp' throws a
3724 `regular-expression-syntax' error.
3726 FLAGS may be the bitwise-or of one or more of the following:
3728 **** Constant: regexp/extended
3729 Use POSIX Extended Regular Expression syntax when interpreting
3730 STR. If not set, POSIX Basic Regular Expression syntax is used.
3731 If the FLAGS argument is omitted, we assume regexp/extended.
3733 **** Constant: regexp/icase
3734 Do not differentiate case. Subsequent searches using the
3735 returned regular expression will be case insensitive.
3737 **** Constant: regexp/newline
3738 Match-any-character operators don't match a newline.
3740 A non-matching list ([^...]) not containing a newline matches a
3743 Match-beginning-of-line operator (^) matches the empty string
3744 immediately after a newline, regardless of whether the FLAGS
3745 passed to regexp-exec contain regexp/notbol.
3747 Match-end-of-line operator ($) matches the empty string
3748 immediately before a newline, regardless of whether the FLAGS
3749 passed to regexp-exec contain regexp/noteol.
3751 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
3752 Match the compiled regular expression REGEXP against `str'. If
3753 the optional integer START argument is provided, begin matching
3754 from that position in the string. Return a match structure
3755 describing the results of the match, or `#f' if no match could be
3758 FLAGS may be the bitwise-or of one or more of the following:
3760 **** Constant: regexp/notbol
3761 The match-beginning-of-line operator always fails to match (but
3762 see the compilation flag regexp/newline above) This flag may be
3763 used when different portions of a string are passed to
3764 regexp-exec and the beginning of the string should not be
3765 interpreted as the beginning of the line.
3767 **** Constant: regexp/noteol
3768 The match-end-of-line operator always fails to match (but see the
3769 compilation flag regexp/newline above)
3771 **** Function: regexp? OBJ
3772 Return `#t' if OBJ is a compiled regular expression, or `#f'
3775 Regular expressions are commonly used to find patterns in one string
3776 and replace them with the contents of another string.
3778 **** Function: regexp-substitute PORT MATCH [ITEM...]
3779 Write to the output port PORT selected contents of the match
3780 structure MATCH. Each ITEM specifies what should be written, and
3781 may be one of the following arguments:
3783 * A string. String arguments are written out verbatim.
3785 * An integer. The submatch with that number is written.
3787 * The symbol `pre'. The portion of the matched string preceding
3788 the regexp match is written.
3790 * The symbol `post'. The portion of the matched string
3791 following the regexp match is written.
3793 PORT may be `#f', in which case nothing is written; instead,
3794 `regexp-substitute' constructs a string from the specified ITEMs
3797 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
3798 Similar to `regexp-substitute', but can be used to perform global
3799 substitutions on STR. Instead of taking a match structure as an
3800 argument, `regexp-substitute/global' takes two string arguments: a
3801 REGEXP string describing a regular expression, and a TARGET string
3802 which should be matched against this regular expression.
3804 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
3807 * A function may be supplied. When this function is called, it
3808 will be passed one argument: a match structure for a given
3809 regular expression match. It should return a string to be
3810 written out to PORT.
3812 * The `post' symbol causes `regexp-substitute/global' to recurse
3813 on the unmatched portion of STR. This *must* be supplied in
3814 order to perform global search-and-replace on STR; if it is
3815 not present among the ITEMs, then `regexp-substitute/global'
3816 will return after processing a single match.
3818 *** Match Structures
3820 A "match structure" is the object returned by `string-match' and
3821 `regexp-exec'. It describes which portion of a string, if any, matched
3822 the given regular expression. Match structures include: a reference to
3823 the string that was checked for matches; the starting and ending
3824 positions of the regexp match; and, if the regexp included any
3825 parenthesized subexpressions, the starting and ending positions of each
3828 In each of the regexp match functions described below, the `match'
3829 argument must be a match structure returned by a previous call to
3830 `string-match' or `regexp-exec'. Most of these functions return some
3831 information about the original target string that was matched against a
3832 regular expression; we will call that string TARGET for easy reference.
3834 **** Function: regexp-match? OBJ
3835 Return `#t' if OBJ is a match structure returned by a previous
3836 call to `regexp-exec', or `#f' otherwise.
3838 **** Function: match:substring MATCH [N]
3839 Return the portion of TARGET matched by subexpression number N.
3840 Submatch 0 (the default) represents the entire regexp match. If
3841 the regular expression as a whole matched, but the subexpression
3842 number N did not match, return `#f'.
3844 **** Function: match:start MATCH [N]
3845 Return the starting position of submatch number N.
3847 **** Function: match:end MATCH [N]
3848 Return the ending position of submatch number N.
3850 **** Function: match:prefix MATCH
3851 Return the unmatched portion of TARGET preceding the regexp match.
3853 **** Function: match:suffix MATCH
3854 Return the unmatched portion of TARGET following the regexp match.
3856 **** Function: match:count MATCH
3857 Return the number of parenthesized subexpressions from MATCH.
3858 Note that the entire regular expression match itself counts as a
3859 subexpression, and failed submatches are included in the count.
3861 **** Function: match:string MATCH
3862 Return the original TARGET string.
3864 *** Backslash Escapes
3866 Sometimes you will want a regexp to match characters like `*' or `$'
3867 exactly. For example, to check whether a particular string represents
3868 a menu entry from an Info node, it would be useful to match it against
3869 a regexp like `^* [^:]*::'. However, this won't work; because the
3870 asterisk is a metacharacter, it won't match the `*' at the beginning of
3871 the string. In this case, we want to make the first asterisk un-magic.
3873 You can do this by preceding the metacharacter with a backslash
3874 character `\'. (This is also called "quoting" the metacharacter, and
3875 is known as a "backslash escape".) When Guile sees a backslash in a
3876 regular expression, it considers the following glyph to be an ordinary
3877 character, no matter what special meaning it would ordinarily have.
3878 Therefore, we can make the above example work by changing the regexp to
3879 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
3880 to match only a single asterisk in the target string.
3882 Since the backslash is itself a metacharacter, you may force a
3883 regexp to match a backslash in the target string by preceding the
3884 backslash with itself. For example, to find variable references in a
3885 TeX program, you might want to find occurrences of the string `\let\'
3886 followed by any number of alphabetic characters. The regular expression
3887 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
3888 each match a single backslash in the target string.
3890 **** Function: regexp-quote STR
3891 Quote each special character found in STR with a backslash, and
3892 return the resulting string.
3894 *Very important:* Using backslash escapes in Guile source code (as
3895 in Emacs Lisp or C) can be tricky, because the backslash character has
3896 special meaning for the Guile reader. For example, if Guile encounters
3897 the character sequence `\n' in the middle of a string while processing
3898 Scheme code, it replaces those characters with a newline character.
3899 Similarly, the character sequence `\t' is replaced by a horizontal tab.
3900 Several of these "escape sequences" are processed by the Guile reader
3901 before your code is executed. Unrecognized escape sequences are
3902 ignored: if the characters `\*' appear in a string, they will be
3903 translated to the single character `*'.
3905 This translation is obviously undesirable for regular expressions,
3906 since we want to be able to include backslashes in a string in order to
3907 escape regexp metacharacters. Therefore, to make sure that a backslash
3908 is preserved in a string in your Guile program, you must use *two*
3909 consecutive backslashes:
3911 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
3913 The string in this example is preprocessed by the Guile reader before
3914 any code is executed. The resulting argument to `make-regexp' is the
3915 string `^\* [^:]*', which is what we really want.
3917 This also means that in order to write a regular expression that
3918 matches a single backslash character, the regular expression string in
3919 the source code must include *four* backslashes. Each consecutive pair
3920 of backslashes gets translated by the Guile reader to a single
3921 backslash, and the resulting double-backslash is interpreted by the
3922 regexp engine as matching a single backslash character. Hence:
3924 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
3926 The reason for the unwieldiness of this syntax is historical. Both
3927 regular expression pattern matchers and Unix string processing systems
3928 have traditionally used backslashes with the special meanings described
3929 above. The POSIX regular expression specification and ANSI C standard
3930 both require these semantics. Attempting to abandon either convention
3931 would cause other kinds of compatibility problems, possibly more severe
3932 ones. Therefore, without extending the Scheme reader to support
3933 strings with different quoting conventions (an ungainly and confusing
3934 extension when implemented in other languages), we must adhere to this
3935 cumbersome escape syntax.
3937 * Changes to the gh_ interface
3939 * Changes to the scm_ interface
3941 * Changes to system call interfaces:
3943 ** The value returned by `raise' is now unspecified. It throws an exception
3946 *** A new procedure `sigaction' can be used to install signal handlers
3948 (sigaction signum [action] [flags])
3950 signum is the signal number, which can be specified using the value
3953 If action is omitted, sigaction returns a pair: the CAR is the current
3954 signal hander, which will be either an integer with the value SIG_DFL
3955 (default action) or SIG_IGN (ignore), or the Scheme procedure which
3956 handles the signal, or #f if a non-Scheme procedure handles the
3957 signal. The CDR contains the current sigaction flags for the handler.
3959 If action is provided, it is installed as the new handler for signum.
3960 action can be a Scheme procedure taking one argument, or the value of
3961 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
3962 whatever signal handler was installed before sigaction was first used.
3963 Flags can optionally be specified for the new handler (SA_RESTART is
3964 always used if the system provides it, so need not be specified.) The
3965 return value is a pair with information about the old handler as
3968 This interface does not provide access to the "signal blocking"
3969 facility. Maybe this is not needed, since the thread support may
3970 provide solutions to the problem of consistent access to data
3973 *** A new procedure `flush-all-ports' is equivalent to running
3974 `force-output' on every port open for output.
3976 ** Guile now provides information on how it was built, via the new
3977 global variable, %guile-build-info. This variable records the values
3978 of the standard GNU makefile directory variables as an assocation
3979 list, mapping variable names (symbols) onto directory paths (strings).
3980 For example, to find out where the Guile link libraries were
3981 installed, you can say:
3983 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
3986 * Changes to the scm_ interface
3988 ** The new function scm_handle_by_message_noexit is just like the
3989 existing scm_handle_by_message function, except that it doesn't call
3990 exit to terminate the process. Instead, it prints a message and just
3991 returns #f. This might be a more appropriate catch-all handler for
3992 new dynamic roots and threads.
3995 Changes in Guile 1.1 (released Friday, May 16 1997):
3997 * Changes to the distribution.
3999 The Guile 1.0 distribution has been split up into several smaller
4001 guile-core --- the Guile interpreter itself.
4002 guile-tcltk --- the interface between the Guile interpreter and
4003 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
4004 is a toolkit for building graphical user interfaces.
4005 guile-rgx-ctax --- the interface between Guile and the Rx regular
4006 expression matcher, and the translator for the Ctax
4007 programming language. These are packaged together because the
4008 Ctax translator uses Rx to parse Ctax source code.
4010 This NEWS file describes the changes made to guile-core since the 1.0
4013 We no longer distribute the documentation, since it was either out of
4014 date, or incomplete. As soon as we have current documentation, we
4019 * Changes to the stand-alone interpreter
4021 ** guile now accepts command-line arguments compatible with SCSH, Olin
4022 Shivers' Scheme Shell.
4024 In general, arguments are evaluated from left to right, but there are
4025 exceptions. The following switches stop argument processing, and
4026 stash all remaining command-line arguments as the value returned by
4027 the (command-line) function.
4028 -s SCRIPT load Scheme source code from FILE, and exit
4029 -c EXPR evalute Scheme expression EXPR, and exit
4030 -- stop scanning arguments; run interactively
4032 The switches below are processed as they are encountered.
4033 -l FILE load Scheme source code from FILE
4034 -e FUNCTION after reading script, apply FUNCTION to
4035 command line arguments
4036 -ds do -s script at this point
4037 --emacs enable Emacs protocol (experimental)
4038 -h, --help display this help and exit
4039 -v, --version display version information and exit
4040 \ read arguments from following script lines
4042 So, for example, here is a Guile script named `ekko' (thanks, Olin)
4043 which re-implements the traditional "echo" command:
4045 #!/usr/local/bin/guile -s
4048 (map (lambda (arg) (display arg) (display " "))
4052 (main (command-line))
4054 Suppose we invoke this script as follows:
4056 ekko a speckled gecko
4058 Through the magic of Unix script processing (triggered by the `#!'
4059 token at the top of the file), /usr/local/bin/guile receives the
4060 following list of command-line arguments:
4062 ("-s" "./ekko" "a" "speckled" "gecko")
4064 Unix inserts the name of the script after the argument specified on
4065 the first line of the file (in this case, "-s"), and then follows that
4066 with the arguments given to the script. Guile loads the script, which
4067 defines the `main' function, and then applies it to the list of
4068 remaining command-line arguments, ("a" "speckled" "gecko").
4070 In Unix, the first line of a script file must take the following form:
4072 #!INTERPRETER ARGUMENT
4074 where INTERPRETER is the absolute filename of the interpreter
4075 executable, and ARGUMENT is a single command-line argument to pass to
4078 You may only pass one argument to the interpreter, and its length is
4079 limited. These restrictions can be annoying to work around, so Guile
4080 provides a general mechanism (borrowed from, and compatible with,
4081 SCSH) for circumventing them.
4083 If the ARGUMENT in a Guile script is a single backslash character,
4084 `\', Guile will open the script file, parse arguments from its second
4085 and subsequent lines, and replace the `\' with them. So, for example,
4086 here is another implementation of the `ekko' script:
4088 #!/usr/local/bin/guile \
4092 (for-each (lambda (arg) (display arg) (display " "))
4096 If the user invokes this script as follows:
4098 ekko a speckled gecko
4100 Unix expands this into
4102 /usr/local/bin/guile \ ekko a speckled gecko
4104 When Guile sees the `\' argument, it replaces it with the arguments
4105 read from the second line of the script, producing:
4107 /usr/local/bin/guile -e main -s ekko a speckled gecko
4109 This tells Guile to load the `ekko' script, and apply the function
4110 `main' to the argument list ("a" "speckled" "gecko").
4112 Here is how Guile parses the command-line arguments:
4113 - Each space character terminates an argument. This means that two
4114 spaces in a row introduce an empty-string argument.
4115 - The tab character is not permitted (unless you quote it with the
4116 backslash character, as described below), to avoid confusion.
4117 - The newline character terminates the sequence of arguments, and will
4118 also terminate a final non-empty argument. (However, a newline
4119 following a space will not introduce a final empty-string argument;
4120 it only terminates the argument list.)
4121 - The backslash character is the escape character. It escapes
4122 backslash, space, tab, and newline. The ANSI C escape sequences
4123 like \n and \t are also supported. These produce argument
4124 constituents; the two-character combination \n doesn't act like a
4125 terminating newline. The escape sequence \NNN for exactly three
4126 octal digits reads as the character whose ASCII code is NNN. As
4127 above, characters produced this way are argument constituents.
4128 Backslash followed by other characters is not allowed.
4130 * Changes to the procedure for linking libguile with your programs
4132 ** Guile now builds and installs a shared guile library, if your
4133 system support shared libraries. (It still builds a static library on
4134 all systems.) Guile automatically detects whether your system
4135 supports shared libraries. To prevent Guile from buildisg shared
4136 libraries, pass the `--disable-shared' flag to the configure script.
4138 Guile takes longer to compile when it builds shared libraries, because
4139 it must compile every file twice --- once to produce position-
4140 independent object code, and once to produce normal object code.
4142 ** The libthreads library has been merged into libguile.
4144 To link a program against Guile, you now need only link against
4145 -lguile and -lqt; -lthreads is no longer needed. If you are using
4146 autoconf to generate configuration scripts for your application, the
4147 following lines should suffice to add the appropriate libraries to
4150 ### Find quickthreads and libguile.
4151 AC_CHECK_LIB(qt, main)
4152 AC_CHECK_LIB(guile, scm_shell)
4154 * Changes to Scheme functions
4156 ** Guile Scheme's special syntax for keyword objects is now optional,
4157 and disabled by default.
4159 The syntax variation from R4RS made it difficult to port some
4160 interesting packages to Guile. The routines which accepted keyword
4161 arguments (mostly in the module system) have been modified to also
4162 accept symbols whose names begin with `:'.
4164 To change the keyword syntax, you must first import the (ice-9 debug)
4166 (use-modules (ice-9 debug))
4168 Then you can enable the keyword syntax as follows:
4169 (read-set! keywords 'prefix)
4171 To disable keyword syntax, do this:
4172 (read-set! keywords #f)
4174 ** Many more primitive functions accept shared substrings as
4175 arguments. In the past, these functions required normal, mutable
4176 strings as arguments, although they never made use of this
4179 ** The uniform array functions now operate on byte vectors. These
4180 functions are `array-fill!', `serial-array-copy!', `array-copy!',
4181 `serial-array-map', `array-map', `array-for-each', and
4184 ** The new functions `trace' and `untrace' implement simple debugging
4185 support for Scheme functions.
4187 The `trace' function accepts any number of procedures as arguments,
4188 and tells the Guile interpreter to display each procedure's name and
4189 arguments each time the procedure is invoked. When invoked with no
4190 arguments, `trace' returns the list of procedures currently being
4193 The `untrace' function accepts any number of procedures as arguments,
4194 and tells the Guile interpreter not to trace them any more. When
4195 invoked with no arguments, `untrace' untraces all curretly traced
4198 The tracing in Guile has an advantage over most other systems: we
4199 don't create new procedure objects, but mark the procedure objects
4200 themselves. This means that anonymous and internal procedures can be
4203 ** The function `assert-repl-prompt' has been renamed to
4204 `set-repl-prompt!'. It takes one argument, PROMPT.
4205 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
4206 - If PROMPT is a string, we use it as a prompt.
4207 - If PROMPT is a procedure accepting no arguments, we call it, and
4208 display the result as a prompt.
4209 - Otherwise, we display "> ".
4211 ** The new function `eval-string' reads Scheme expressions from a
4212 string and evaluates them, returning the value of the last expression
4213 in the string. If the string contains no expressions, it returns an
4216 ** The new function `thunk?' returns true iff its argument is a
4217 procedure of zero arguments.
4219 ** `defined?' is now a builtin function, instead of syntax. This
4220 means that its argument should be quoted. It returns #t iff its
4221 argument is bound in the current module.
4223 ** The new syntax `use-modules' allows you to add new modules to your
4224 environment without re-typing a complete `define-module' form. It
4225 accepts any number of module names as arguments, and imports their
4226 public bindings into the current module.
4228 ** The new function (module-defined? NAME MODULE) returns true iff
4229 NAME, a symbol, is defined in MODULE, a module object.
4231 ** The new function `builtin-bindings' creates and returns a hash
4232 table containing copies of all the root module's bindings.
4234 ** The new function `builtin-weak-bindings' does the same as
4235 `builtin-bindings', but creates a doubly-weak hash table.
4237 ** The `equal?' function now considers variable objects to be
4238 equivalent if they have the same name and the same value.
4240 ** The new function `command-line' returns the command-line arguments
4241 given to Guile, as a list of strings.
4243 When using guile as a script interpreter, `command-line' returns the
4244 script's arguments; those processed by the interpreter (like `-s' or
4245 `-c') are omitted. (In other words, you get the normal, expected
4246 behavior.) Any application that uses scm_shell to process its
4247 command-line arguments gets this behavior as well.
4249 ** The new function `load-user-init' looks for a file called `.guile'
4250 in the user's home directory, and loads it if it exists. This is
4251 mostly for use by the code generated by scm_compile_shell_switches,
4252 but we thought it might also be useful in other circumstances.
4254 ** The new function `log10' returns the base-10 logarithm of its
4257 ** Changes to I/O functions
4259 *** The functions `read', `primitive-load', `read-and-eval!', and
4260 `primitive-load-path' no longer take optional arguments controlling
4261 case insensitivity and a `#' parser.
4263 Case sensitivity is now controlled by a read option called
4264 `case-insensitive'. The user can add new `#' syntaxes with the
4265 `read-hash-extend' function (see below).
4267 *** The new function `read-hash-extend' allows the user to change the
4268 syntax of Guile Scheme in a somewhat controlled way.
4270 (read-hash-extend CHAR PROC)
4271 When parsing S-expressions, if we read a `#' character followed by
4272 the character CHAR, use PROC to parse an object from the stream.
4273 If PROC is #f, remove any parsing procedure registered for CHAR.
4275 The reader applies PROC to two arguments: CHAR and an input port.
4277 *** The new functions read-delimited and read-delimited! provide a
4278 general mechanism for doing delimited input on streams.
4280 (read-delimited DELIMS [PORT HANDLE-DELIM])
4281 Read until we encounter one of the characters in DELIMS (a string),
4282 or end-of-file. PORT is the input port to read from; it defaults to
4283 the current input port. The HANDLE-DELIM parameter determines how
4284 the terminating character is handled; it should be one of the
4287 'trim omit delimiter from result
4288 'peek leave delimiter character in input stream
4289 'concat append delimiter character to returned value
4290 'split return a pair: (RESULT . TERMINATOR)
4292 HANDLE-DELIM defaults to 'peek.
4294 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
4295 A side-effecting variant of `read-delimited'.
4297 The data is written into the string BUF at the indices in the
4298 half-open interval [START, END); the default interval is the whole
4299 string: START = 0 and END = (string-length BUF). The values of
4300 START and END must specify a well-defined interval in BUF, i.e.
4301 0 <= START <= END <= (string-length BUF).
4303 It returns NBYTES, the number of bytes read. If the buffer filled
4304 up without a delimiter character being found, it returns #f. If the
4305 port is at EOF when the read starts, it returns the EOF object.
4307 If an integer is returned (i.e., the read is successfully terminated
4308 by reading a delimiter character), then the HANDLE-DELIM parameter
4309 determines how to handle the terminating character. It is described
4310 above, and defaults to 'peek.
4312 (The descriptions of these functions were borrowed from the SCSH
4313 manual, by Olin Shivers and Brian Carlstrom.)
4315 *** The `%read-delimited!' function is the primitive used to implement
4316 `read-delimited' and `read-delimited!'.
4318 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
4320 This returns a pair of values: (TERMINATOR . NUM-READ).
4321 - TERMINATOR describes why the read was terminated. If it is a
4322 character or the eof object, then that is the value that terminated
4323 the read. If it is #f, the function filled the buffer without finding
4324 a delimiting character.
4325 - NUM-READ is the number of characters read into BUF.
4327 If the read is successfully terminated by reading a delimiter
4328 character, then the gobble? parameter determines what to do with the
4329 terminating character. If true, the character is removed from the
4330 input stream; if false, the character is left in the input stream
4331 where a subsequent read operation will retrieve it. In either case,
4332 the character is also the first value returned by the procedure call.
4334 (The descriptions of this function was borrowed from the SCSH manual,
4335 by Olin Shivers and Brian Carlstrom.)
4337 *** The `read-line' and `read-line!' functions have changed; they now
4338 trim the terminator by default; previously they appended it to the
4339 returned string. For the old behavior, use (read-line PORT 'concat).
4341 *** The functions `uniform-array-read!' and `uniform-array-write!' now
4342 take new optional START and END arguments, specifying the region of
4343 the array to read and write.
4345 *** The `ungetc-char-ready?' function has been removed. We feel it's
4346 inappropriate for an interface to expose implementation details this
4349 ** Changes to the Unix library and system call interface
4351 *** The new fcntl function provides access to the Unix `fcntl' system
4354 (fcntl PORT COMMAND VALUE)
4355 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
4356 Values for COMMAND are:
4358 F_DUPFD duplicate a file descriptor
4359 F_GETFD read the descriptor's close-on-exec flag
4360 F_SETFD set the descriptor's close-on-exec flag to VALUE
4361 F_GETFL read the descriptor's flags, as set on open
4362 F_SETFL set the descriptor's flags, as set on open to VALUE
4363 F_GETOWN return the process ID of a socket's owner, for SIGIO
4364 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
4365 FD_CLOEXEC not sure what this is
4367 For details, see the documentation for the fcntl system call.
4369 *** The arguments to `select' have changed, for compatibility with
4370 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
4371 expected behavior. The MILLISECONDS parameter has been changed to
4372 MICROSECONDS, to more closely resemble the underlying system call.
4373 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
4374 corresponding return set will be the same.
4376 *** The arguments to the `mknod' system call have changed. They are
4379 (mknod PATH TYPE PERMS DEV)
4380 Create a new file (`node') in the file system. PATH is the name of
4381 the file to create. TYPE is the kind of file to create; it should
4382 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
4383 permission bits to give the newly created file. If TYPE is
4384 'block-special or 'char-special, DEV specifies which device the
4385 special file refers to; its interpretation depends on the kind of
4386 special file being created.
4388 *** The `fork' function has been renamed to `primitive-fork', to avoid
4389 clashing with various SCSH forks.
4391 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
4392 and `recvfrom!'. They no longer accept a size for a second argument;
4393 you must pass a string to hold the received value. They no longer
4394 return the buffer. Instead, `recv' returns the length of the message
4395 received, and `recvfrom' returns a pair containing the packet's length
4396 and originating address.
4398 *** The file descriptor datatype has been removed, as have the
4399 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
4400 We plan to replace these functions with a SCSH-compatible interface.
4402 *** The `create' function has been removed; it's just a special case
4405 *** There are new functions to break down process termination status
4406 values. In the descriptions below, STATUS is a value returned by
4409 (status:exit-val STATUS)
4410 If the child process exited normally, this function returns the exit
4411 code for the child process (i.e., the value passed to exit, or
4412 returned from main). If the child process did not exit normally,
4413 this function returns #f.
4415 (status:stop-sig STATUS)
4416 If the child process was suspended by a signal, this function
4417 returns the signal that suspended the child. Otherwise, it returns
4420 (status:term-sig STATUS)
4421 If the child process terminated abnormally, this function returns
4422 the signal that terminated the child. Otherwise, this function
4425 POSIX promises that exactly one of these functions will return true on
4426 a valid STATUS value.
4428 These functions are compatible with SCSH.
4430 *** There are new accessors and setters for the broken-out time vectors
4431 returned by `localtime', `gmtime', and that ilk. They are:
4433 Component Accessor Setter
4434 ========================= ============ ============
4435 seconds tm:sec set-tm:sec
4436 minutes tm:min set-tm:min
4437 hours tm:hour set-tm:hour
4438 day of the month tm:mday set-tm:mday
4439 month tm:mon set-tm:mon
4440 year tm:year set-tm:year
4441 day of the week tm:wday set-tm:wday
4442 day in the year tm:yday set-tm:yday
4443 daylight saving time tm:isdst set-tm:isdst
4444 GMT offset, seconds tm:gmtoff set-tm:gmtoff
4445 name of time zone tm:zone set-tm:zone
4447 *** There are new accessors for the vectors returned by `uname',
4448 describing the host system:
4451 ============================================== ================
4452 name of the operating system implementation utsname:sysname
4453 network name of this machine utsname:nodename
4454 release level of the operating system utsname:release
4455 version level of the operating system utsname:version
4456 machine hardware platform utsname:machine
4458 *** There are new accessors for the vectors returned by `getpw',
4459 `getpwnam', `getpwuid', and `getpwent', describing entries from the
4460 system's user database:
4463 ====================== =================
4464 user name passwd:name
4465 user password passwd:passwd
4468 real name passwd:gecos
4469 home directory passwd:dir
4470 shell program passwd:shell
4472 *** There are new accessors for the vectors returned by `getgr',
4473 `getgrnam', `getgrgid', and `getgrent', describing entries from the
4474 system's group database:
4477 ======================= ============
4478 group name group:name
4479 group password group:passwd
4481 group members group:mem
4483 *** There are new accessors for the vectors returned by `gethost',
4484 `gethostbyaddr', `gethostbyname', and `gethostent', describing
4488 ========================= ===============
4489 official name of host hostent:name
4490 alias list hostent:aliases
4491 host address type hostent:addrtype
4492 length of address hostent:length
4493 list of addresses hostent:addr-list
4495 *** There are new accessors for the vectors returned by `getnet',
4496 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
4500 ========================= ===============
4501 official name of net netent:name
4502 alias list netent:aliases
4503 net number type netent:addrtype
4504 net number netent:net
4506 *** There are new accessors for the vectors returned by `getproto',
4507 `getprotobyname', `getprotobynumber', and `getprotoent', describing
4511 ========================= ===============
4512 official protocol name protoent:name
4513 alias list protoent:aliases
4514 protocol number protoent:proto
4516 *** There are new accessors for the vectors returned by `getserv',
4517 `getservbyname', `getservbyport', and `getservent', describing
4521 ========================= ===============
4522 official service name servent:name
4523 alias list servent:aliases
4524 port number servent:port
4525 protocol to use servent:proto
4527 *** There are new accessors for the sockaddr structures returned by
4528 `accept', `getsockname', `getpeername', `recvfrom!':
4531 ======================================== ===============
4532 address format (`family') sockaddr:fam
4533 path, for file domain addresses sockaddr:path
4534 address, for internet domain addresses sockaddr:addr
4535 TCP or UDP port, for internet sockaddr:port
4537 *** The `getpwent', `getgrent', `gethostent', `getnetent',
4538 `getprotoent', and `getservent' functions now return #f at the end of
4539 the user database. (They used to throw an exception.)
4541 Note that calling MUMBLEent function is equivalent to calling the
4542 corresponding MUMBLE function with no arguments.
4544 *** The `setpwent', `setgrent', `sethostent', `setnetent',
4545 `setprotoent', and `setservent' routines now take no arguments.
4547 *** The `gethost', `getproto', `getnet', and `getserv' functions now
4548 provide more useful information when they throw an exception.
4550 *** The `lnaof' function has been renamed to `inet-lnaof'.
4552 *** Guile now claims to have the `current-time' feature.
4554 *** The `mktime' function now takes an optional second argument ZONE,
4555 giving the time zone to use for the conversion. ZONE should be a
4556 string, in the same format as expected for the "TZ" environment variable.
4558 *** The `strptime' function now returns a pair (TIME . COUNT), where
4559 TIME is the parsed time as a vector, and COUNT is the number of
4560 characters from the string left unparsed. This function used to
4561 return the remaining characters as a string.
4563 *** The `gettimeofday' function has replaced the old `time+ticks' function.
4564 The return value is now (SECONDS . MICROSECONDS); the fractional
4565 component is no longer expressed in "ticks".
4567 *** The `ticks/sec' constant has been removed, in light of the above change.
4569 * Changes to the gh_ interface
4571 ** gh_eval_str() now returns an SCM object which is the result of the
4574 ** gh_scm2str() now copies the Scheme data to a caller-provided C
4577 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
4578 and returns the array
4580 ** gh_scm2str0() is gone: there is no need to distinguish
4581 null-terminated from non-null-terminated, since gh_scm2newstr() allows
4582 the user to interpret the data both ways.
4584 * Changes to the scm_ interface
4586 ** The new function scm_symbol_value0 provides an easy way to get a
4587 symbol's value from C code:
4589 SCM scm_symbol_value0 (char *NAME)
4590 Return the value of the symbol named by the null-terminated string
4591 NAME in the current module. If the symbol named NAME is unbound in
4592 the current module, return SCM_UNDEFINED.
4594 ** The new function scm_sysintern0 creates new top-level variables,
4595 without assigning them a value.
4597 SCM scm_sysintern0 (char *NAME)
4598 Create a new Scheme top-level variable named NAME. NAME is a
4599 null-terminated string. Return the variable's value cell.
4601 ** The function scm_internal_catch is the guts of catch. It handles
4602 all the mechanics of setting up a catch target, invoking the catch
4603 body, and perhaps invoking the handler if the body does a throw.
4605 The function is designed to be usable from C code, but is general
4606 enough to implement all the semantics Guile Scheme expects from throw.
4608 TAG is the catch tag. Typically, this is a symbol, but this function
4609 doesn't actually care about that.
4611 BODY is a pointer to a C function which runs the body of the catch;
4612 this is the code you can throw from. We call it like this:
4613 BODY (BODY_DATA, JMPBUF)
4615 BODY_DATA is just the BODY_DATA argument we received; we pass it
4616 through to BODY as its first argument. The caller can make
4617 BODY_DATA point to anything useful that BODY might need.
4618 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
4619 which we have just created and initialized.
4621 HANDLER is a pointer to a C function to deal with a throw to TAG,
4622 should one occur. We call it like this:
4623 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
4625 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
4626 same idea as BODY_DATA above.
4627 THROWN_TAG is the tag that the user threw to; usually this is
4628 TAG, but it could be something else if TAG was #t (i.e., a
4629 catch-all), or the user threw to a jmpbuf.
4630 THROW_ARGS is the list of arguments the user passed to the THROW
4633 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
4634 is just a pointer we pass through to HANDLER. We don't actually
4635 use either of those pointers otherwise ourselves. The idea is
4636 that, if our caller wants to communicate something to BODY or
4637 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
4638 HANDLER can then use. Think of it as a way to make BODY and
4639 HANDLER closures, not just functions; MUMBLE_DATA points to the
4642 Of course, it's up to the caller to make sure that any data a
4643 MUMBLE_DATA needs is protected from GC. A common way to do this is
4644 to make MUMBLE_DATA a pointer to data stored in an automatic
4645 structure variable; since the collector must scan the stack for
4646 references anyway, this assures that any references in MUMBLE_DATA
4649 ** The new function scm_internal_lazy_catch is exactly like
4650 scm_internal_catch, except:
4652 - It does not unwind the stack (this is the major difference).
4653 - If handler returns, its value is returned from the throw.
4654 - BODY always receives #f as its JMPBUF argument (since there's no
4655 jmpbuf associated with a lazy catch, because we don't unwind the
4658 ** scm_body_thunk is a new body function you can pass to
4659 scm_internal_catch if you want the body to be like Scheme's `catch'
4660 --- a thunk, or a function of one argument if the tag is #f.
4662 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
4663 contains the Scheme procedure to invoke as the body, and the tag
4664 we're catching. If the tag is #f, then we pass JMPBUF (created by
4665 scm_internal_catch) to the body procedure; otherwise, the body gets
4668 ** scm_handle_by_proc is a new handler function you can pass to
4669 scm_internal_catch if you want the handler to act like Scheme's catch
4670 --- call a procedure with the tag and the throw arguments.
4672 If the user does a throw to this catch, this function runs a handler
4673 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
4674 variable holding the Scheme procedure object to invoke. It ought to
4675 be a pointer to an automatic variable (i.e., one living on the stack),
4676 or the procedure object should be otherwise protected from GC.
4678 ** scm_handle_by_message is a new handler function to use with
4679 `scm_internal_catch' if you want Guile to print a message and die.
4680 It's useful for dealing with throws to uncaught keys at the top level.
4682 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
4683 message header to print; if zero, we use "guile" instead. That
4684 text is followed by a colon, then the message described by ARGS.
4686 ** The return type of scm_boot_guile is now void; the function does
4687 not return a value, and indeed, never returns at all.
4689 ** The new function scm_shell makes it easy for user applications to
4690 process command-line arguments in a way that is compatible with the
4691 stand-alone guile interpreter (which is in turn compatible with SCSH,
4694 To use the scm_shell function, first initialize any guile modules
4695 linked into your application, and then call scm_shell with the values
4696 of ARGC and ARGV your `main' function received. scm_shell will add
4697 any SCSH-style meta-arguments from the top of the script file to the
4698 argument vector, and then process the command-line arguments. This
4699 generally means loading a script file or starting up an interactive
4700 command interpreter. For details, see "Changes to the stand-alone
4703 ** The new functions scm_get_meta_args and scm_count_argv help you
4704 implement the SCSH-style meta-argument, `\'.
4706 char **scm_get_meta_args (int ARGC, char **ARGV)
4707 If the second element of ARGV is a string consisting of a single
4708 backslash character (i.e. "\\" in Scheme notation), open the file
4709 named by the following argument, parse arguments from it, and return
4710 the spliced command line. The returned array is terminated by a
4713 For details of argument parsing, see above, under "guile now accepts
4714 command-line arguments compatible with SCSH..."
4716 int scm_count_argv (char **ARGV)
4717 Count the arguments in ARGV, assuming it is terminated by a null
4720 For an example of how these functions might be used, see the source
4721 code for the function scm_shell in libguile/script.c.
4723 You will usually want to use scm_shell instead of calling this
4726 ** The new function scm_compile_shell_switches turns an array of
4727 command-line arguments into Scheme code to carry out the actions they
4728 describe. Given ARGC and ARGV, it returns a Scheme expression to
4729 evaluate, and calls scm_set_program_arguments to make any remaining
4730 command-line arguments available to the Scheme code. For example,
4731 given the following arguments:
4733 -e main -s ekko a speckled gecko
4735 scm_set_program_arguments will return the following expression:
4737 (begin (load "ekko") (main (command-line)) (quit))
4739 You will usually want to use scm_shell instead of calling this
4742 ** The function scm_shell_usage prints a usage message appropriate for
4743 an interpreter that uses scm_compile_shell_switches to handle its
4744 command-line arguments.
4746 void scm_shell_usage (int FATAL, char *MESSAGE)
4747 Print a usage message to the standard error output. If MESSAGE is
4748 non-zero, write it before the usage message, followed by a newline.
4749 If FATAL is non-zero, exit the process, using FATAL as the
4750 termination status. (If you want to be compatible with Guile,
4751 always use 1 as the exit status when terminating due to command-line
4754 You will usually want to use scm_shell instead of calling this
4757 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
4758 expressions. It used to return SCM_EOL. Earth-shattering.
4760 ** The macros for declaring scheme objects in C code have been
4761 rearranged slightly. They are now:
4763 SCM_SYMBOL (C_NAME, SCHEME_NAME)
4764 Declare a static SCM variable named C_NAME, and initialize it to
4765 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
4766 be a C identifier, and SCHEME_NAME should be a C string.
4768 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
4769 Just like SCM_SYMBOL, but make C_NAME globally visible.
4771 SCM_VCELL (C_NAME, SCHEME_NAME)
4772 Create a global variable at the Scheme level named SCHEME_NAME.
4773 Declare a static SCM variable named C_NAME, and initialize it to
4774 point to the Scheme variable's value cell.
4776 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
4777 Just like SCM_VCELL, but make C_NAME globally visible.
4779 The `guile-snarf' script writes initialization code for these macros
4780 to its standard output, given C source code as input.
4782 The SCM_GLOBAL macro is gone.
4784 ** The scm_read_line and scm_read_line_x functions have been replaced
4785 by Scheme code based on the %read-delimited! procedure (known to C
4786 code as scm_read_delimited_x). See its description above for more
4789 ** The function scm_sys_open has been renamed to scm_open. It now
4790 returns a port instead of an FD object.
4792 * The dynamic linking support has changed. For more information, see
4793 libguile/DYNAMIC-LINKING.
4798 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
4801 * Changes to the 'guile' program:
4803 ** Guile now loads some new files when it starts up. Guile first
4804 searches the load path for init.scm, and loads it if found. Then, if
4805 Guile is not being used to execute a script, and the user's home
4806 directory contains a file named `.guile', Guile loads that.
4808 ** You can now use Guile as a shell script interpreter.
4810 To paraphrase the SCSH manual:
4812 When Unix tries to execute an executable file whose first two
4813 characters are the `#!', it treats the file not as machine code to
4814 be directly executed by the native processor, but as source code
4815 to be executed by some interpreter. The interpreter to use is
4816 specified immediately after the #! sequence on the first line of
4817 the source file. The kernel reads in the name of the interpreter,
4818 and executes that instead. It passes the interpreter the source
4819 filename as its first argument, with the original arguments
4820 following. Consult the Unix man page for the `exec' system call
4821 for more information.
4823 Now you can use Guile as an interpreter, using a mechanism which is a
4824 compatible subset of that provided by SCSH.
4826 Guile now recognizes a '-s' command line switch, whose argument is the
4827 name of a file of Scheme code to load. It also treats the two
4828 characters `#!' as the start of a comment, terminated by `!#'. Thus,
4829 to make a file of Scheme code directly executable by Unix, insert the
4830 following two lines at the top of the file:
4832 #!/usr/local/bin/guile -s
4835 Guile treats the argument of the `-s' command-line switch as the name
4836 of a file of Scheme code to load, and treats the sequence `#!' as the
4837 start of a block comment, terminated by `!#'.
4839 For example, here's a version of 'echo' written in Scheme:
4841 #!/usr/local/bin/guile -s
4843 (let loop ((args (cdr (program-arguments))))
4846 (display (car args))
4847 (if (pair? (cdr args))
4849 (loop (cdr args)))))
4852 Why does `#!' start a block comment terminated by `!#', instead of the
4853 end of the line? That is the notation SCSH uses, and although we
4854 don't yet support the other SCSH features that motivate that choice,
4855 we would like to be backward-compatible with any existing Guile
4856 scripts once we do. Furthermore, if the path to Guile on your system
4857 is too long for your kernel, you can start the script with this
4861 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
4864 Note that some very old Unix systems don't support the `#!' syntax.
4867 ** You can now run Guile without installing it.
4869 Previous versions of the interactive Guile interpreter (`guile')
4870 couldn't start up unless Guile's Scheme library had been installed;
4871 they used the value of the environment variable `SCHEME_LOAD_PATH'
4872 later on in the startup process, but not to find the startup code
4873 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
4876 To run Guile without installing it, build it in the normal way, and
4877 then set the environment variable `SCHEME_LOAD_PATH' to a
4878 colon-separated list of directories, including the top-level directory
4879 of the Guile sources. For example, if you unpacked Guile so that the
4880 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
4883 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
4886 ** Guile's read-eval-print loop no longer prints #<unspecified>
4887 results. If the user wants to see this, she can evaluate the
4888 expression (assert-repl-print-unspecified #t), perhaps in her startup
4891 ** Guile no longer shows backtraces by default when an error occurs;
4892 however, it does display a message saying how to get one, and how to
4893 request that they be displayed by default. After an error, evaluate
4895 to see a backtrace, and
4896 (debug-enable 'backtrace)
4897 to see them by default.
4901 * Changes to Guile Scheme:
4903 ** Guile now distinguishes between #f and the empty list.
4905 This is for compatibility with the IEEE standard, the (possibly)
4906 upcoming Revised^5 Report on Scheme, and many extant Scheme
4909 Guile used to have #f and '() denote the same object, to make Scheme's
4910 type system more compatible with Emacs Lisp's. However, the change
4911 caused too much trouble for Scheme programmers, and we found another
4912 way to reconcile Emacs Lisp with Scheme that didn't require this.
4915 ** Guile's delq, delv, delete functions, and their destructive
4916 counterparts, delq!, delv!, and delete!, now remove all matching
4917 elements from the list, not just the first. This matches the behavior
4918 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
4919 functions which inspired them.
4921 I recognize that this change may break code in subtle ways, but it
4922 seems best to make the change before the FSF's first Guile release,
4926 ** The compiled-library-path function has been deleted from libguile.
4928 ** The facilities for loading Scheme source files have changed.
4930 *** The variable %load-path now tells Guile which directories to search
4931 for Scheme code. Its value is a list of strings, each of which names
4934 *** The variable %load-extensions now tells Guile which extensions to
4935 try appending to a filename when searching the load path. Its value
4936 is a list of strings. Its default value is ("" ".scm").
4938 *** (%search-load-path FILENAME) searches the directories listed in the
4939 value of the %load-path variable for a Scheme file named FILENAME,
4940 with all the extensions listed in %load-extensions. If it finds a
4941 match, then it returns its full filename. If FILENAME is absolute, it
4942 returns it unchanged. Otherwise, it returns #f.
4944 %search-load-path will not return matches that refer to directories.
4946 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
4947 uses %seach-load-path to find a file named FILENAME, and loads it if
4948 it finds it. If it can't read FILENAME for any reason, it throws an
4951 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
4954 *** load uses the same searching semantics as primitive-load.
4956 *** The functions %try-load, try-load-with-path, %load, load-with-path,
4957 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
4958 path, and load-module-with-path have been deleted. The functions
4959 above should serve their purposes.
4961 *** If the value of the variable %load-hook is a procedure,
4962 `primitive-load' applies its value to the name of the file being
4963 loaded (without the load path directory name prepended). If its value
4964 is #f, it is ignored. Otherwise, an error occurs.
4966 This is mostly useful for printing load notification messages.
4969 ** The function `eval!' is no longer accessible from the scheme level.
4970 We can't allow operations which introduce glocs into the scheme level,
4971 because Guile's type system can't handle these as data. Use `eval' or
4972 `read-and-eval!' (see below) as replacement.
4974 ** The new function read-and-eval! reads an expression from PORT,
4975 evaluates it, and returns the result. This is more efficient than
4976 simply calling `read' and `eval', since it is not necessary to make a
4977 copy of the expression for the evaluator to munge.
4979 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
4980 for the `read' function.
4983 ** The function `int?' has been removed; its definition was identical
4984 to that of `integer?'.
4986 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
4987 use the R4RS names for these functions.
4989 ** The function object-properties no longer returns the hash handle;
4990 it simply returns the object's property list.
4992 ** Many functions have been changed to throw errors, instead of
4993 returning #f on failure. The point of providing exception handling in
4994 the language is to simplify the logic of user code, but this is less
4995 useful if Guile's primitives don't throw exceptions.
4997 ** The function `fileno' has been renamed from `%fileno'.
4999 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
5002 * Changes to Guile's C interface:
5004 ** The library's initialization procedure has been simplified.
5005 scm_boot_guile now has the prototype:
5007 void scm_boot_guile (int ARGC,
5009 void (*main_func) (),
5012 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
5013 MAIN_FUNC should do all the work of the program (initializing other
5014 packages, reading user input, etc.) before returning. When MAIN_FUNC
5015 returns, call exit (0); this function never returns. If you want some
5016 other exit value, MAIN_FUNC may call exit itself.
5018 scm_boot_guile arranges for program-arguments to return the strings
5019 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
5020 scm_set_program_arguments with the final list, so Scheme code will
5021 know which arguments have been processed.
5023 scm_boot_guile establishes a catch-all catch handler which prints an
5024 error message and exits the process. This means that Guile exits in a
5025 coherent way when system errors occur and the user isn't prepared to
5026 handle it. If the user doesn't like this behavior, they can establish
5027 their own universal catcher in MAIN_FUNC to shadow this one.
5029 Why must the caller do all the real work from MAIN_FUNC? The garbage
5030 collector assumes that all local variables of type SCM will be above
5031 scm_boot_guile's stack frame on the stack. If you try to manipulate
5032 SCM values after this function returns, it's the luck of the draw
5033 whether the GC will be able to find the objects you allocate. So,
5034 scm_boot_guile function exits, rather than returning, to discourage
5035 people from making that mistake.
5037 The IN, OUT, and ERR arguments were removed; there are other
5038 convenient ways to override these when desired.
5040 The RESULT argument was deleted; this function should never return.
5042 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
5046 ** Guile's header files should no longer conflict with your system's
5049 In order to compile code which #included <libguile.h>, previous
5050 versions of Guile required you to add a directory containing all the
5051 Guile header files to your #include path. This was a problem, since
5052 Guile's header files have names which conflict with many systems'
5055 Now only <libguile.h> need appear in your #include path; you must
5056 refer to all Guile's other header files as <libguile/mumble.h>.
5057 Guile's installation procedure puts libguile.h in $(includedir), and
5058 the rest in $(includedir)/libguile.
5061 ** Two new C functions, scm_protect_object and scm_unprotect_object,
5062 have been added to the Guile library.
5064 scm_protect_object (OBJ) protects OBJ from the garbage collector.
5065 OBJ will not be freed, even if all other references are dropped,
5066 until someone does scm_unprotect_object (OBJ). Both functions
5069 Note that calls to scm_protect_object do not nest. You can call
5070 scm_protect_object any number of times on a given object, and the
5071 next call to scm_unprotect_object will unprotect it completely.
5073 Basically, scm_protect_object and scm_unprotect_object just
5074 maintain a list of references to things. Since the GC knows about
5075 this list, all objects it mentions stay alive. scm_protect_object
5076 adds its argument to the list; scm_unprotect_object remove its
5077 argument from the list.
5080 ** scm_eval_0str now returns the value of the last expression
5083 ** The new function scm_read_0str reads an s-expression from a
5084 null-terminated string, and returns it.
5086 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
5087 to a Scheme port object.
5089 ** The new function `scm_set_program_arguments' allows C code to set
5090 the value returned by the Scheme `program-arguments' function.
5095 * Guile no longer includes sophisticated Tcl/Tk support.
5097 The old Tcl/Tk support was unsatisfying to us, because it required the
5098 user to link against the Tcl library, as well as Tk and Guile. The
5099 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
5100 referring to widgets by names, rather than exporting widgets to Scheme
5101 code as a special datatype.
5103 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
5104 maintainers described some very interesting changes in progress to the
5105 Tcl/Tk internals, which would facilitate clean interfaces between lone
5106 Tk and other interpreters --- even for garbage-collected languages
5107 like Scheme. They expected the new Tk to be publicly available in the
5110 Since it seems that Guile might soon have a new, cleaner interface to
5111 lone Tk, and that the old Guile/Tk glue code would probably need to be
5112 completely rewritten, we (Jim Blandy and Richard Stallman) have
5113 decided not to support the old code. We'll spend the time instead on
5114 a good interface to the newer Tk, as soon as it is available.
5116 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
5119 Copyright information:
5121 Copyright (C) 1996,1997 Free Software Foundation, Inc.
5123 Permission is granted to anyone to make or distribute verbatim copies
5124 of this document as received, in any medium, provided that the
5125 copyright notice and this permission notice are preserved,
5126 thus giving the recipient permission to redistribute in turn.
5128 Permission is granted to distribute modified versions
5129 of this document, or of portions of it,
5130 under the above conditions, provided also that they
5131 carry prominent notices stating who last changed them.
5136 paragraph-separate: "[
\f]*$"