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-in-module '(+ 1 2) m) --> 3
86 (eval-in-module 'load m) --> ERROR: Unbound variable: load
88 * Changes to Scheme functions and syntax
90 ** New function `make-object-property'
92 This function returns a new `procedure with setter' P that can be used
93 to attach a property to objects. When calling P as
97 where `obj' is any kind of object, it attaches `val' to `obj' in such
98 a way that it can be retrieved by calling P as
102 This function will replace procedure properties, symbol properties and
103 source properties eventually.
105 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
107 Instead of #&optional, #&key, etc you should now use #:optional,
108 #:key, etc. Since #:optional is a keyword, you can write it as just
109 :optional when (read-set! keywords 'prefix) is active.
111 The old reader syntax `#&' is still supported, but deprecated. It
112 will be removed in the next release.
114 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
116 `eval' is now R5RS, that is it takes two arguments.
117 The second argument is an environment specifier, i.e. either
119 (scheme-report-environment 5)
121 (interaction-environment)
127 ** New define-module option: pure
129 Tells the module system not to include any bindings from the root
134 (define-module (totally-empty-module)
137 ** New define-module option: export NAME1 ...
139 Export names NAME1 ...
141 This option is required if you want to be able to export bindings from
142 a module which doesn't import one of `define-public' or `export'.
148 :use-module (ice-9 r5rs)
151 ;;; Note that we're pure R5RS below this point!
156 ** Deprecated: scm_make_shared_substring
158 Explicit shared substrings will disappear from Guile.
160 Instead, "normal" strings will be implemented using sharing
161 internally, combined with a copy-on-write strategy.
163 ** Deprecated: scm_read_only_string_p
165 The concept of read-only strings will disappear in next release of
168 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
170 Instead, use scm_memq, scm_memv, scm_member.
172 ** New function: port? X
174 Returns a boolean indicating whether X is a port. Equivalent to
175 `(or (input-port? X) (output-port? X))'.
177 ** New function: port-for-each proc
179 Apply PROC to each port in the Guile port table in turn. The
180 return value is unspecified.
182 ** New function: dup2 oldfd newfd
184 A simple wrapper for the `dup2' system call. Copies the file
185 descriptor OLDFD to descriptor number NEWFD, replacing the
186 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
187 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
188 to move away ports which are using NEWFD\n". The return value is
191 ** New function: close-fdes fd
193 A simple wrapper for the `close' system call. Close file
194 descriptor FD, which must be an integer. Unlike close (*note
195 close: Ports and File Descriptors.), the file descriptor will be
196 closed even if a port is using it. The return value is
199 ** Deprecated: close-all-ports-except. This was intended for closing
200 ports in a child process after a fork, but it has the undesirable side
201 effect of flushing buffers. port-for-each is more flexible.
203 ** The (ice-9 popen) module now attempts to set up file descriptors in
204 the child process from the current Scheme ports, instead of using the
205 current values of file descriptors 0, 1, and 2 in the parent process.
207 * Changes to the gh_ interface
209 * Changes to the scm_ interface
211 ** New function: scm_init_guile ()
213 In contrast to scm_boot_guile, scm_init_guile will return normally
214 after initializing Guile. It is not available on all systems, tho.
216 ** New functions: scm_primitive_make_property
217 scm_primitive_property_ref
218 scm_primitive_property_set_x
219 scm_primitive_property_del_x
221 These functions implement a new way to deal with object properties.
222 See libguile/properties.c for their documentation.
224 ** New function: scm_done_free (long size)
226 This function is the inverse of scm_done_malloc. Use it to report the
227 amount of smob memory you free. The previous method, which involved
228 calling scm_done_malloc with negative argument, was somewhat
229 unintuitive (and is still available, of course).
231 ** New global variable scm_gc_running_p introduced.
233 Use this variable to find out if garbage collection is being executed. Up to
234 now applications have used scm_gc_heap_lock to test if garbage collection was
235 running, which also works because of the fact that up to know only the garbage
236 collector has set this variable. But, this is an implementation detail that
237 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
238 of this variable is (and has been) not fully safe anyway.
240 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
241 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
242 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
244 Use these instead of SCM_LENGTH.
246 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_VECTOR_LENGTH
248 Use these instead of SCM_SETLENGTH
250 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
251 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
254 Use these instead of SCM_CHARS or SCM_VELTS.
256 ** New macro: SCM_BITVECTOR_P
258 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
260 Use instead of SCM_COERCE_SUBSTR.
262 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
263 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
264 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
265 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
266 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
267 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
268 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
269 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
270 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
273 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
274 Use scm_memory_error instead of SCM_NALLOC.
275 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
276 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
277 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
278 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
279 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
280 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
281 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
282 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
283 Use SCM_STRINGP instead of SCM_RWSTRINGP.
284 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
285 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
286 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
288 ** Removed function: scm_struct_init
290 ** Deprecated function: scm_call_catching_errors
292 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
294 ** Deprecated function: scm_strhash
296 Use scm_string_hash instead.
298 ** Deprecated function: scm_vector_set_length_x
300 Instead, create a fresh vector of the desired size and copy the contents.
302 ** scm_gensym has changed prototype
304 scm_gensym now only takes one argument.
306 ** New function: scm_gentemp (SCM prefix, SCM obarray)
308 The builtin `gentemp' has now become a primitive.
310 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
313 There is now only a single symbol type scm_tc7_symbol.
314 The tag scm_tc7_lvector was not used anyway.
317 Changes since Guile 1.3.4:
319 * Changes to the distribution
321 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
323 We've changed the way we handle generated files in the Guile source
324 repository. As a result, the procedure for building trees obtained
325 from the nightly FTP snapshots or via CVS has changed:
326 - You must have appropriate versions of autoconf, automake, and
327 libtool installed on your system. See README for info on how to
328 obtain these programs.
329 - Before configuring the tree, you must first run the script
330 `autogen.sh' at the top of the source tree.
332 The Guile repository used to contain not only source files, written by
333 humans, but also some generated files, like configure scripts and
334 Makefile.in files. Even though the contents of these files could be
335 derived mechanically from other files present, we thought it would
336 make the tree easier to build if we checked them into CVS.
338 However, this approach means that minor differences between
339 developer's installed tools and habits affected the whole team.
340 So we have removed the generated files from the repository, and
341 added the autogen.sh script, which will reconstruct them
345 ** configure now has experimental options to remove support for certain
348 --disable-arrays omit array and uniform array support
349 --disable-posix omit posix interfaces
350 --disable-networking omit networking interfaces
351 --disable-regex omit regular expression interfaces
353 These are likely to become separate modules some day.
355 ** New configure option --enable-debug-freelist
357 This enables a debugging version of SCM_NEWCELL(), and also registers
358 an extra primitive, the setter `gc-set-debug-check-freelist!'.
360 Configure with the --enable-debug-freelist option to enable
361 the gc-set-debug-check-freelist! primitive, and then use:
363 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
364 (gc-set-debug-check-freelist! #f) # turn off checking
366 Checking of the freelist forces a traversal of the freelist and
367 a garbage collection before each allocation of a cell. This can
368 slow down the interpreter dramatically, so the setter should be used to
369 turn on this extra processing only when necessary.
371 ** New configure option --enable-debug-malloc
373 Include code for debugging of calls to scm_must_malloc/realloc/free.
377 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
378 2. objects reallocated by scm_must_realloc has been allocated by
380 3. reallocated objects are reallocated with the same what string
382 But, most importantly, it records the number of allocated objects of
383 each kind. This is useful when searching for memory leaks.
385 A Guile compiled with this option provides the primitive
386 `malloc-stats' which returns an alist with pairs of kind and the
387 number of objects of that kind.
389 ** All includes are now referenced relative to the root directory
391 Since some users have had problems with mixups between Guile and
392 system headers, we have decided to always refer to Guile headers via
393 their parent directories. This essentially creates a "private name
394 space" for Guile headers. This means that the compiler only is given
395 -I options for the root build and root source directory.
397 ** Header files kw.h and genio.h have been removed.
399 ** The module (ice-9 getopt-gnu-style) has been removed.
401 ** New module (ice-9 documentation)
403 Implements the interface to documentation strings associated with
406 * Changes to the stand-alone interpreter
408 ** New command line option --debug
410 Start Guile with debugging evaluator and backtraces enabled.
412 This is useful when debugging your .guile init file or scripts.
416 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
417 (help REGEXP) ditto for objects with names matching REGEXP (a string)
418 (help ,EXPR) gives documentation for object returned by EXPR
419 (help) gives this text
421 `help' searches among bindings exported from loaded modules, while
422 `apropos' searches among bindings visible from the "current" module.
424 Examples: (help help)
426 (help "output-string")
428 ** `help' and `apropos' now prints full module names
430 ** Dynamic linking now uses libltdl from the libtool package.
432 The old system dependent code for doing dynamic linking has been
433 replaced with calls to the libltdl functions which do all the hairy
436 The major improvement is that you can now directly pass libtool
437 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
438 will be able to do the best shared library job you can get, via
441 The way dynamic libraries are found has changed and is not really
442 portable across platforms, probably. It is therefore recommended to
443 use absolute filenames when possible.
445 If you pass a filename without an extension to `dynamic-link', it will
446 try a few appropriate ones. Thus, the most platform ignorant way is
447 to specify a name like "libfoo", without any directories and
450 ** Guile COOP threads are now compatible with LinuxThreads
452 Previously, COOP threading wasn't possible in applications linked with
453 Linux POSIX threads due to their use of the stack pointer to find the
454 thread context. This has now been fixed with a workaround which uses
455 the pthreads to allocate the stack.
457 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
459 ** Positions of erring expression in scripts
461 With version 1.3.4, the location of the erring expression in Guile
462 scipts is no longer automatically reported. (This should have been
463 documented before the 1.3.4 release.)
465 You can get this information by enabling recording of positions of
466 source expressions and running the debugging evaluator. Put this at
467 the top of your script (or in your "site" file):
469 (read-enable 'positions)
470 (debug-enable 'debug)
472 ** Backtraces in scripts
474 It is now possible to get backtraces in scripts.
478 (debug-enable 'debug 'backtrace)
480 at the top of the script.
482 (The first options enables the debugging evaluator.
483 The second enables backtraces.)
485 ** Part of module system symbol lookup now implemented in C
487 The eval closure of most modules is now implemented in C. Since this
488 was one of the bottlenecks for loading speed, Guile now loads code
489 substantially faster than before.
491 ** Attempting to get the value of an unbound variable now produces
492 an exception with a key of 'unbound-variable instead of 'misc-error.
494 ** The initial default output port is now unbuffered if it's using a
495 tty device. Previously in this situation it was line-buffered.
497 ** gc-thunk is deprecated
499 gc-thunk will be removed in next release of Guile. It has been
500 replaced by after-gc-hook.
502 ** New hook: after-gc-hook
504 after-gc-hook takes over the role of gc-thunk. This hook is run at
505 the first SCM_TICK after a GC. (Thus, the code is run at the same
506 point during evaluation as signal handlers.)
508 Note that this hook should be used only for diagnostic and debugging
509 purposes. It is not certain that it will continue to be well-defined
510 when this hook is run in the future.
512 C programmers: Note the new C level hooks scm_before_gc_c_hook,
513 scm_before_sweep_c_hook, scm_after_gc_c_hook.
515 ** Improvements to garbage collector
517 Guile 1.4 has a new policy for triggering heap allocation and
518 determining the sizes of heap segments. It fixes a number of problems
521 1. The new policy can handle two separate pools of cells
522 (2-word/4-word) better. (The old policy would run wild, allocating
523 more and more memory for certain programs.)
525 2. The old code would sometimes allocate far too much heap so that the
526 Guile process became gigantic. The new code avoids this.
528 3. The old code would sometimes allocate too little so that few cells
529 were freed at GC so that, in turn, too much time was spent in GC.
531 4. The old code would often trigger heap allocation several times in a
532 row. (The new scheme predicts how large the segments needs to be
533 in order not to need further allocation.)
535 All in all, the new GC policy will make larger applications more
538 The new GC scheme also is prepared for POSIX threading. Threads can
539 allocate private pools of cells ("clusters") with just a single
540 function call. Allocation of single cells from such a cluster can
541 then proceed without any need of inter-thread synchronization.
543 ** New environment variables controlling GC parameters
545 GUILE_MAX_SEGMENT_SIZE Maximal segment size
548 Allocation of 2-word cell heaps:
550 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
553 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
554 GC in percent of total heap size
557 Allocation of 4-word cell heaps
558 (used for real numbers and misc other objects):
560 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
562 (See entry "Way for application to customize GC parameters" under
563 section "Changes to the scm_ interface" below.)
565 ** Guile now implements reals using 4-word cells
567 This speeds up computation with reals. (They were earlier allocated
568 with `malloc'.) There is still some room for optimizations, however.
570 ** Some further steps toward POSIX thread support have been taken
572 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
573 don't have much effect any longer, and many of them will be removed in
577 are only handled at the top of the evaluator loop, immediately after
578 I/O, and in scm_equalp.
580 *** The GC can allocate thread private pools of pairs.
582 * Changes to Scheme functions and syntax
584 ** close-input-port and close-output-port are now R5RS
586 These procedures have been turned into primitives and have R5RS behaviour.
588 ** New procedure: simple-format PORT MESSAGE ARG1 ...
590 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
591 extended by the more sophisticated version in (ice-9 format)
593 (simple-format port message . args)
594 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
595 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
596 the escapes are replaced with corresponding members of ARGS:
597 ~A formats using `display' and ~S formats using `write'.
598 If DESTINATION is #t, then use the `current-output-port',
599 if DESTINATION is #f, then return a string containing the formatted text.
600 Does not add a trailing newline."
602 ** string-ref: the second argument is no longer optional.
604 ** string, list->string: no longer accept strings in their arguments,
605 only characters, for compatibility with R5RS.
607 ** New procedure: port-closed? PORT
608 Returns #t if PORT is closed or #f if it is open.
612 The list* functionality is now provided by cons* (SRFI-1 compliant)
614 ** New procedure: cons* ARG1 ARG2 ... ARGn
616 Like `list', but the last arg provides the tail of the constructed list,
617 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
619 Requires at least one argument. If given one argument, that argument
620 is returned as result.
622 This function is called `list*' in some other Schemes and in Common LISP.
624 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
626 ** New procedure: object-documentation OBJECT
628 Returns the documentation string associated with OBJECT. The
629 procedure uses a caching mechanism so that subsequent lookups are
632 Exported by (ice-9 documentation).
634 ** module-name now returns full names of modules
636 Previously, only the last part of the name was returned (`session' for
637 `(ice-9 session)'). Ex: `(ice-9 session)'.
639 * Changes to the gh_ interface
641 ** Deprecated: gh_int2scmb
643 Use gh_bool2scm instead.
645 * Changes to the scm_ interface
647 ** Guile primitives now carry docstrings!
649 Thanks to Greg Badros!
651 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
653 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
654 macros and must contain a docstring that is extracted into foo.doc using a new
655 guile-doc-snarf script (that uses guile-doc-snarf.awk).
657 However, a major overhaul of these macros is scheduled for the next release of
660 ** Guile primitives use a new technique for validation of arguments
662 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
663 the readability of argument checking.
665 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
667 ** New macros: SCM_PACK, SCM_UNPACK
669 Compose/decompose an SCM value.
671 The SCM type is now treated as an abstract data type and may be defined as a
672 long, a void* or as a struct, depending on the architecture and compile time
673 options. This makes it easier to find several types of bugs, for example when
674 SCM values are treated as integers without conversion. Values of the SCM type
675 should be treated as "atomic" values. These macros are used when
676 composing/decomposing an SCM value, either because you want to access
677 individual bits, or because you want to treat it as an integer value.
679 E.g., in order to set bit 7 in an SCM value x, use the expression
681 SCM_PACK (SCM_UNPACK (x) | 0x80)
683 ** The name property of hooks is deprecated.
684 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
686 You can emulate this feature by using object properties.
688 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
689 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
692 These macros will be removed in a future release of Guile.
694 ** The following types, functions and macros from numbers.h are deprecated:
695 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
696 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
698 Further, it is recommended not to rely on implementation details for guile's
699 current implementation of bignums. It is planned to replace this
700 implementation with gmp in the future.
702 ** Port internals: the rw_random variable in the scm_port structure
703 must be set to non-zero in any random access port. In recent Guile
704 releases it was only set for bidirectional random-access ports.
706 ** Port internals: the seek ptob procedure is now responsible for
707 resetting the buffers if required. The change was made so that in the
708 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
709 the fport and strport ptobs can avoid resetting the buffers,
710 in particular to avoid discarding unread chars. An existing port
711 type can be fixed by adding something like the following to the
712 beginning of the ptob seek procedure:
714 if (pt->rw_active == SCM_PORT_READ)
715 scm_end_input (object);
716 else if (pt->rw_active == SCM_PORT_WRITE)
717 ptob->flush (object);
719 although to actually avoid resetting the buffers and discard unread
720 chars requires further hacking that depends on the characteristics
723 ** Deprecated functions: scm_fseek, scm_tag
725 These functions are no longer used and will be removed in a future version.
727 ** The scm_sysmissing procedure is no longer used in libguile.
728 Unless it turns out to be unexpectedly useful to somebody, it will be
729 removed in a future version.
731 ** The format of error message strings has changed
733 The two C procedures: scm_display_error and scm_error, as well as the
734 primitive `scm-error', now use scm_simple_format to do their work.
735 This means that the message strings of all code must be updated to use
736 ~A where %s was used before, and ~S where %S was used before.
738 During the period when there still are a lot of old Guiles out there,
739 you might want to support both old and new versions of Guile.
741 There are basically two methods to achieve this. Both methods use
744 AC_CHECK_FUNCS(scm_simple_format)
746 in your configure.in.
748 Method 1: Use the string concatenation features of ANSI C's
753 #ifdef HAVE_SCM_SIMPLE_FORMAT
759 Then represent each of your error messages using a preprocessor macro:
761 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
765 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
766 (define make-message string-append)
768 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
770 Method 2: Use the oldfmt function found in doc/oldfmt.c.
774 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
779 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
783 ** Deprecated: coop_mutex_init, coop_condition_variable_init
785 Don't use the functions coop_mutex_init and
786 coop_condition_variable_init. They will change.
788 Use scm_mutex_init and scm_cond_init instead.
790 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
791 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
792 COND, as `scm_cond_wait' does, but it also bounds the duration
793 of the wait. If COND has not been signaled before time ABSTIME,
794 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
795 returns the error code `ETIMEDOUT'.
797 The ABSTIME parameter specifies an absolute time, with the same
798 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
799 to 00:00:00 GMT, January 1, 1970.
801 ** New function: scm_cond_broadcast (scm_cond_t *COND)
802 `scm_cond_broadcast' restarts all the threads that are waiting
803 on the condition variable COND. Nothing happens if no threads are
806 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
807 `scm_key_create' allocates a new TSD key. The key is stored in
808 the location pointed to by KEY. There is no limit on the number
809 of keys allocated at a given time. The value initially associated
810 with the returned key is `NULL' in all currently executing threads.
812 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
813 function associated with the key. When a thread terminates,
814 DESTR_FUNCTION is called on the value associated with the key in
815 that thread. The DESTR_FUNCTION is not called if a key is deleted
816 with `scm_key_delete' or a value is changed with
817 `scm_setspecific'. The order in which destructor functions are
818 called at thread termination time is unspecified.
820 Destructors are not yet implemented.
822 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
823 `scm_setspecific' changes the value associated with KEY in the
824 calling thread, storing the given POINTER instead.
826 ** New function: scm_getspecific (scm_key_t KEY)
827 `scm_getspecific' returns the value currently associated with
828 KEY in the calling thread.
830 ** New function: scm_key_delete (scm_key_t KEY)
831 `scm_key_delete' deallocates a TSD key. It does not check
832 whether non-`NULL' values are associated with that key in the
833 currently executing threads, nor call the destructor function
834 associated with the key.
836 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
838 Initialize a C level hook HOOK with associated HOOK_DATA and type
839 TYPE. (See scm_c_hook_run ().)
841 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
843 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
844 is true, add it last, otherwise first. The same FUNC can be added
845 multiple times if FUNC_DATA differ and vice versa.
847 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
849 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
850 function is only removed if both FUNC and FUNC_DATA matches.
852 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
854 Run hook HOOK passing DATA to the hook functions.
856 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
857 returned is undefined.
859 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
860 returns a non-NULL value. This value is returned as the result of
861 scm_c_hook_run. If all functions return NULL, NULL is returned.
863 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
864 returns a NULL value, and NULL is returned. If all functions returns
865 a non-NULL value, the last value is returned.
867 ** New C level GC hooks
869 Five new C level hooks has been added to the garbage collector.
874 are run before locking and after unlocking the heap. The system is
875 thus in a mode where evaluation can take place. (Except that
876 scm_before_gc_c_hook must not allocate new cells.)
878 scm_before_mark_c_hook
879 scm_before_sweep_c_hook
880 scm_after_sweep_c_hook
882 are run when the heap is locked. These are intended for extension of
883 the GC in a modular fashion. Examples are the weaks and guardians
886 ** Way for application to customize GC parameters
888 The application can set up other default values for the GC heap
889 allocation parameters
891 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
892 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
893 GUILE_MAX_SEGMENT_SIZE,
897 scm_default_init_heap_size_1, scm_default_min_yield_1,
898 scm_default_init_heap_size_2, scm_default_min_yield_2,
899 scm_default_max_segment_size
901 respectively before callong scm_boot_guile.
903 (See entry "New environment variables ..." in section
904 "Changes to the stand-alone interpreter" above.)
906 ** scm_protect_object/scm_unprotect_object now nest
908 This means that you can call scm_protect_object multiple times on an
909 object and count on the object being protected until
910 scm_unprotect_object has been call the same number of times.
912 The functions also have better time complexity.
914 Still, it is usually possible to structure the application in a way
915 that you don't need to use these functions. For example, if you use a
916 protected standard Guile list to keep track of live objects rather
917 than some custom data type, objects will die a natural death when they
918 are no longer needed.
920 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
922 Guile does not provide the float representation for inexact real numbers any
923 more. Now, only doubles are used to represent inexact real numbers. Further,
924 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
925 and scm_tc16_complex, respectively.
927 ** Removed deprecated type scm_smobfuns
929 ** Removed deprecated function scm_newsmob
931 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
933 There is an ongoing discussion among the developers whether to
934 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
935 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
936 until this issue has been settled.
938 ** Removed deprecated type tag scm_tc16_kw
940 ** Added type tag scm_tc16_keyword
942 (This was introduced already in release 1.3.4 but was not documented
945 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
947 * Changes to system call interfaces:
949 ** The "select" procedure now tests port buffers for the ability to
950 provide input or accept output. Previously only the underlying file
951 descriptors were checked.
953 ** New variable PIPE_BUF: the maximum number of bytes that can be
954 atomically written to a pipe.
956 ** If a facility is not available on the system when Guile is
957 compiled, the corresponding primitive procedure will not be defined.
958 Previously it would have been defined but would throw a system-error
959 exception if called. Exception handlers which catch this case may
960 need minor modification: an error will be thrown with key
961 'unbound-variable instead of 'system-error. Alternatively it's
962 now possible to use `defined?' to check whether the facility is
965 ** Procedures which depend on the timezone should now give the correct
966 result on systems which cache the TZ environment variable, even if TZ
967 is changed without calling tzset.
969 * Changes to the networking interfaces:
971 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
972 long integers between network and host format. For now, it's not
973 particularly convenient to do this kind of thing, but consider:
975 (define write-network-long
977 (let ((v (make-uniform-vector 1 1 0)))
978 (uniform-vector-set! v 0 (htonl value))
979 (uniform-vector-write v port))))
981 (define read-network-long
983 (let ((v (make-uniform-vector 1 1 0)))
984 (uniform-vector-read! v port)
985 (ntohl (uniform-vector-ref v 0)))))
987 ** If inet-aton fails, it now throws an error with key 'misc-error
988 instead of 'system-error, since errno is not relevant.
990 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
991 specific keys instead of 'system-error. The latter is inappropriate
992 since errno will not have been set. The keys are:
993 'host-not-found, 'try-again, 'no-recovery and 'no-data.
995 ** sethostent, setnetent, setprotoent, setservent: now take an
996 optional argument STAYOPEN, which specifies whether the database
997 remains open after a database entry is accessed randomly (e.g., using
998 gethostbyname for the hosts database.) The default is #f. Previously
1002 Changes since Guile 1.3.2:
1004 * Changes to the stand-alone interpreter
1008 An initial version of the Guile debugger written by Chris Hanson has
1009 been added. The debugger is still under development but is included
1010 in the distribution anyway since it is already quite useful.
1016 after an error to enter the debugger. Type `help' inside the debugger
1017 for a description of available commands.
1019 If you prefer to have stack frames numbered and printed in
1020 anti-chronological order and prefer up in the stack to be down on the
1021 screen as is the case in gdb, you can put
1023 (debug-enable 'backwards)
1025 in your .guile startup file. (However, this means that Guile can't
1026 use indentation to indicate stack level.)
1028 The debugger is autoloaded into Guile at the first use.
1030 ** Further enhancements to backtraces
1032 There is a new debug option `width' which controls the maximum width
1033 on the screen of printed stack frames. Fancy printing parameters
1034 ("level" and "length" as in Common LISP) are adaptively adjusted for
1035 each stack frame to give maximum information while still fitting
1036 within the bounds. If the stack frame can't be made to fit by
1037 adjusting parameters, it is simply cut off at the end. This is marked
1040 ** Some modules are now only loaded when the repl is started
1042 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
1043 regex) are now loaded into (guile-user) only if the repl has been
1044 started. The effect is that the startup time for scripts has been
1045 reduced to 30% of what it was previously.
1047 Correctly written scripts load the modules they require at the top of
1048 the file and should not be affected by this change.
1050 ** Hooks are now represented as smobs
1052 * Changes to Scheme functions and syntax
1054 ** Readline support has changed again.
1056 The old (readline-activator) module is gone. Use (ice-9 readline)
1057 instead, which now contains all readline functionality. So the code
1058 to activate readline is now
1060 (use-modules (ice-9 readline))
1063 This should work at any time, including from the guile prompt.
1065 To avoid confusion about the terms of Guile's license, please only
1066 enable readline for your personal use; please don't make it the
1067 default for others. Here is why we make this rather odd-sounding
1070 Guile is normally licensed under a weakened form of the GNU General
1071 Public License, which allows you to link code with Guile without
1072 placing that code under the GPL. This exception is important to some
1075 However, since readline is distributed under the GNU General Public
1076 License, when you link Guile with readline, either statically or
1077 dynamically, you effectively change Guile's license to the strict GPL.
1078 Whenever you link any strictly GPL'd code into Guile, uses of Guile
1079 which are normally permitted become forbidden. This is a rather
1080 non-obvious consequence of the licensing terms.
1082 So, to make sure things remain clear, please let people choose for
1083 themselves whether to link GPL'd libraries like readline with Guile.
1085 ** regexp-substitute/global has changed slightly, but incompatibly.
1087 If you include a function in the item list, the string of the match
1088 object it receives is the same string passed to
1089 regexp-substitute/global, not some suffix of that string.
1090 Correspondingly, the match's positions are relative to the entire
1091 string, not the suffix.
1093 If the regexp can match the empty string, the way matches are chosen
1094 from the string has changed. regexp-substitute/global recognizes the
1095 same set of matches that list-matches does; see below.
1097 ** New function: list-matches REGEXP STRING [FLAGS]
1099 Return a list of match objects, one for every non-overlapping, maximal
1100 match of REGEXP in STRING. The matches appear in left-to-right order.
1101 list-matches only reports matches of the empty string if there are no
1102 other matches which begin on, end at, or include the empty match's
1105 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1107 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
1109 For each match of REGEXP in STRING, apply PROC to the match object,
1110 and the last value PROC returned, or INIT for the first call. Return
1111 the last value returned by PROC. We apply PROC to the matches as they
1112 appear from left to right.
1114 This function recognizes matches according to the same criteria as
1117 Thus, you could define list-matches like this:
1119 (define (list-matches regexp string . flags)
1120 (reverse! (apply fold-matches regexp string '() cons flags)))
1122 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
1126 *** New function: hook? OBJ
1128 Return #t if OBJ is a hook, otherwise #f.
1130 *** New function: make-hook-with-name NAME [ARITY]
1132 Return a hook with name NAME and arity ARITY. The default value for
1133 ARITY is 0. The only effect of NAME is that it will appear when the
1134 hook object is printed to ease debugging.
1136 *** New function: hook-empty? HOOK
1138 Return #t if HOOK doesn't contain any procedures, otherwise #f.
1140 *** New function: hook->list HOOK
1142 Return a list of the procedures that are called when run-hook is
1145 ** `map' signals an error if its argument lists are not all the same length.
1147 This is the behavior required by R5RS, so this change is really a bug
1148 fix. But it seems to affect a lot of people's code, so we're
1149 mentioning it here anyway.
1151 ** Print-state handling has been made more transparent
1153 Under certain circumstances, ports are represented as a port with an
1154 associated print state. Earlier, this pair was represented as a pair
1155 (see "Some magic has been added to the printer" below). It is now
1156 indistinguishable (almost; see `get-print-state') from a port on the
1159 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
1161 Return a new port with the associated print state PRINT-STATE.
1163 *** New function: get-print-state OUTPUT-PORT
1165 Return the print state associated with this port if it exists,
1166 otherwise return #f.
1168 *** New function: directory-stream? OBJECT
1170 Returns true iff OBJECT is a directory stream --- the sort of object
1171 returned by `opendir'.
1173 ** New function: using-readline?
1175 Return #t if readline is in use in the current repl.
1177 ** structs will be removed in 1.4
1179 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
1180 and use GOOPS objects as the fundamental record type.
1182 * Changes to the scm_ interface
1184 ** structs will be removed in 1.4
1186 The entire current struct interface (struct.c, struct.h) will be
1187 replaced in Guile 1.4. We will merge GOOPS into libguile and use
1188 GOOPS objects as the fundamental record type.
1190 ** The internal representation of subr's has changed
1192 Instead of giving a hint to the subr name, the CAR field of the subr
1193 now contains an index to a subr entry in scm_subr_table.
1195 *** New variable: scm_subr_table
1197 An array of subr entries. A subr entry contains the name, properties
1198 and documentation associated with the subr. The properties and
1199 documentation slots are not yet used.
1201 ** A new scheme for "forwarding" calls to a builtin to a generic function
1203 It is now possible to extend the functionality of some Guile
1204 primitives by letting them defer a call to a GOOPS generic function on
1205 argument mismatch. This means that there is no loss of efficiency in
1210 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
1211 (define-method + ((x <string>) (y <string>))
1212 (string-append x y))
1214 + will still be as efficient as usual in numerical calculations, but
1215 can also be used for concatenating strings.
1217 Who will be the first one to extend Guile's numerical tower to
1218 rationals? :) [OK, there a few other things to fix before this can
1219 be made in a clean way.]
1221 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
1223 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
1225 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
1227 These do the same job as SCM_PROC and SCM_PROC1, but they also define
1228 a variable GENERIC which can be used by the dispatch macros below.
1230 [This is experimental code which may change soon.]
1232 *** New macros for forwarding control to a generic on arg type error
1234 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
1236 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
1238 These correspond to the scm_wta function call, and have the same
1239 behaviour until the user has called the GOOPS primitive
1240 `enable-primitive-generic!'. After that, these macros will apply the
1241 generic function GENERIC to the argument(s) instead of calling
1244 [This is experimental code which may change soon.]
1246 *** New macros for argument testing with generic dispatch
1248 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
1250 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
1252 These correspond to the SCM_ASSERT macro, but will defer control to
1253 GENERIC on error after `enable-primitive-generic!' has been called.
1255 [This is experimental code which may change soon.]
1257 ** New function: SCM scm_eval_body (SCM body, SCM env)
1259 Evaluates the body of a special form.
1261 ** The internal representation of struct's has changed
1263 Previously, four slots were allocated for the procedure(s) of entities
1264 and operators. The motivation for this representation had to do with
1265 the structure of the evaluator, the wish to support tail-recursive
1266 generic functions, and efficiency. Since the generic function
1267 dispatch mechanism has changed, there is no longer a need for such an
1268 expensive representation, and the representation has been simplified.
1270 This should not make any difference for most users.
1272 ** GOOPS support has been cleaned up.
1274 Some code has been moved from eval.c to objects.c and code in both of
1275 these compilation units has been cleaned up and better structured.
1277 *** New functions for applying generic functions
1279 New function: SCM scm_apply_generic (GENERIC, ARGS)
1280 New function: SCM scm_call_generic_0 (GENERIC)
1281 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
1282 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
1283 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
1285 ** Deprecated function: scm_make_named_hook
1287 It is now replaced by:
1289 ** New function: SCM scm_create_hook (const char *name, int arity)
1291 Creates a hook in the same way as make-hook above but also
1292 binds a variable named NAME to it.
1294 This is the typical way of creating a hook from C code.
1296 Currently, the variable is created in the "current" module.
1297 This might change when we get the new module system.
1299 [The behaviour is identical to scm_make_named_hook.]
1303 Changes since Guile 1.3:
1305 * Changes to mailing lists
1307 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
1309 See the README file to find current addresses for all the Guile
1312 * Changes to the distribution
1314 ** Readline support is no longer included with Guile by default.
1316 Based on the different license terms of Guile and Readline, we
1317 concluded that Guile should not *by default* cause the linking of
1318 Readline into an application program. Readline support is now offered
1319 as a separate module, which is linked into an application only when
1320 you explicitly specify it.
1322 Although Guile is GNU software, its distribution terms add a special
1323 exception to the usual GNU General Public License (GPL). Guile's
1324 license includes a clause that allows you to link Guile with non-free
1325 programs. We add this exception so as not to put Guile at a
1326 disadvantage vis-a-vis other extensibility packages that support other
1329 In contrast, the GNU Readline library is distributed under the GNU
1330 General Public License pure and simple. This means that you may not
1331 link Readline, even dynamically, into an application unless it is
1332 distributed under a free software license that is compatible the GPL.
1334 Because of this difference in distribution terms, an application that
1335 can use Guile may not be able to use Readline. Now users will be
1336 explicitly offered two independent decisions about the use of these
1339 You can activate the readline support by issuing
1341 (use-modules (readline-activator))
1344 from your ".guile" file, for example.
1346 * Changes to the stand-alone interpreter
1348 ** All builtins now print as primitives.
1349 Previously builtin procedures not belonging to the fundamental subr
1350 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
1351 Now, they print as #<primitive-procedure NAME>.
1353 ** Backtraces slightly more intelligible.
1354 gsubr-apply and macro transformer application frames no longer appear
1357 * Changes to Scheme functions and syntax
1359 ** Guile now correctly handles internal defines by rewriting them into
1360 their equivalent letrec. Previously, internal defines would
1361 incrementally add to the innermost environment, without checking
1362 whether the restrictions specified in RnRS were met. This lead to the
1363 correct behaviour when these restriction actually were met, but didn't
1364 catch all illegal uses. Such an illegal use could lead to crashes of
1365 the Guile interpreter or or other unwanted results. An example of
1366 incorrect internal defines that made Guile behave erratically:
1378 The problem with this example is that the definition of `c' uses the
1379 value of `b' directly. This confuses the meoization machine of Guile
1380 so that the second call of `b' (this time in a larger environment that
1381 also contains bindings for `c' and `d') refers to the binding of `c'
1382 instead of `a'. You could also make Guile crash with a variation on
1387 (define (b flag) (if flag a 1))
1388 (define c (1+ (b flag)))
1396 From now on, Guile will issue an `Unbound variable: b' error message
1401 A hook contains a list of functions which should be called on
1402 particular occasions in an existing program. Hooks are used for
1405 A window manager might have a hook before-window-map-hook. The window
1406 manager uses the function run-hooks to call all functions stored in
1407 before-window-map-hook each time a window is mapped. The user can
1408 store functions in the hook using add-hook!.
1410 In Guile, hooks are first class objects.
1412 *** New function: make-hook [N_ARGS]
1414 Return a hook for hook functions which can take N_ARGS arguments.
1415 The default value for N_ARGS is 0.
1417 (See also scm_make_named_hook below.)
1419 *** New function: add-hook! HOOK PROC [APPEND_P]
1421 Put PROC at the beginning of the list of functions stored in HOOK.
1422 If APPEND_P is supplied, and non-false, put PROC at the end instead.
1424 PROC must be able to take the number of arguments specified when the
1427 If PROC already exists in HOOK, then remove it first.
1429 *** New function: remove-hook! HOOK PROC
1431 Remove PROC from the list of functions in HOOK.
1433 *** New function: reset-hook! HOOK
1435 Clear the list of hook functions stored in HOOK.
1437 *** New function: run-hook HOOK ARG1 ...
1439 Run all hook functions stored in HOOK with arguments ARG1 ... .
1440 The number of arguments supplied must correspond to the number given
1441 when the hook was created.
1443 ** The function `dynamic-link' now takes optional keyword arguments.
1444 The only keyword argument that is currently defined is `:global
1445 BOOL'. With it, you can control whether the shared library will be
1446 linked in global mode or not. In global mode, the symbols from the
1447 linked library can be used to resolve references from other
1448 dynamically linked libraries. In non-global mode, the linked
1449 library is essentially invisible and can only be accessed via
1450 `dynamic-func', etc. The default is now to link in global mode.
1451 Previously, the default has been non-global mode.
1453 The `#:global' keyword is only effective on platforms that support
1454 the dlopen family of functions.
1456 ** New function `provided?'
1458 - Function: provided? FEATURE
1459 Return true iff FEATURE is supported by this installation of
1460 Guile. FEATURE must be a symbol naming a feature; the global
1461 variable `*features*' is a list of available features.
1463 ** Changes to the module (ice-9 expect):
1465 *** The expect-strings macro now matches `$' in a regular expression
1466 only at a line-break or end-of-file by default. Previously it would
1467 match the end of the string accumulated so far. The old behaviour
1468 can be obtained by setting the variable `expect-strings-exec-flags'
1471 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
1472 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
1473 in a regular expression will still match before a line-break or
1474 end-of-file. The default is `regexp/noteol'.
1476 *** The expect-strings macro now uses a variable
1477 `expect-strings-compile-flags' for the flags to be supplied to
1478 `make-regexp'. The default is `regexp/newline', which was previously
1481 *** The expect macro now supplies two arguments to a match procedure:
1482 the current accumulated string and a flag to indicate whether
1483 end-of-file has been reached. Previously only the string was supplied.
1484 If end-of-file is reached, the match procedure will be called an
1485 additional time with the same accumulated string as the previous call
1486 but with the flag set.
1488 ** New module (ice-9 format), implementing the Common Lisp `format' function.
1490 This code, and the documentation for it that appears here, was
1491 borrowed from SLIB, with minor adaptations for Guile.
1493 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
1494 An almost complete implementation of Common LISP format description
1495 according to the CL reference book `Common LISP' from Guy L.
1496 Steele, Digital Press. Backward compatible to most of the
1497 available Scheme format implementations.
1499 Returns `#t', `#f' or a string; has side effect of printing
1500 according to FORMAT-STRING. If DESTINATION is `#t', the output is
1501 to the current output port and `#t' is returned. If DESTINATION
1502 is `#f', a formatted string is returned as the result of the call.
1503 NEW: If DESTINATION is a string, DESTINATION is regarded as the
1504 format string; FORMAT-STRING is then the first argument and the
1505 output is returned as a string. If DESTINATION is a number, the
1506 output is to the current error port if available by the
1507 implementation. Otherwise DESTINATION must be an output port and
1510 FORMAT-STRING must be a string. In case of a formatting error
1511 format returns `#f' and prints a message on the current output or
1512 error port. Characters are output as if the string were output by
1513 the `display' function with the exception of those prefixed by a
1514 tilde (~). For a detailed description of the FORMAT-STRING syntax
1515 please consult a Common LISP format reference manual. For a test
1516 suite to verify this format implementation load `formatst.scm'.
1517 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
1519 Note: `format' is not reentrant, i.e. only one `format'-call may
1520 be executed at a time.
1523 *** Format Specification (Format version 3.0)
1525 Please consult a Common LISP format reference manual for a detailed
1526 description of the format string syntax. For a demonstration of the
1527 implemented directives see `formatst.scm'.
1529 This implementation supports directive parameters and modifiers (`:'
1530 and `@' characters). Multiple parameters must be separated by a comma
1531 (`,'). Parameters can be numerical parameters (positive or negative),
1532 character parameters (prefixed by a quote character (`''), variable
1533 parameters (`v'), number of rest arguments parameter (`#'), empty and
1534 default parameters. Directive characters are case independent. The
1535 general form of a directive is:
1537 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
1539 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
1541 *** Implemented CL Format Control Directives
1543 Documentation syntax: Uppercase characters represent the
1544 corresponding control directive characters. Lowercase characters
1545 represent control directive parameter descriptions.
1548 Any (print as `display' does).
1552 `~MINCOL,COLINC,MINPAD,PADCHARA'
1556 S-expression (print as `write' does).
1560 `~MINCOL,COLINC,MINPAD,PADCHARS'
1566 print number sign always.
1569 print comma separated.
1571 `~MINCOL,PADCHAR,COMMACHARD'
1577 print number sign always.
1580 print comma separated.
1582 `~MINCOL,PADCHAR,COMMACHARX'
1588 print number sign always.
1591 print comma separated.
1593 `~MINCOL,PADCHAR,COMMACHARO'
1599 print number sign always.
1602 print comma separated.
1604 `~MINCOL,PADCHAR,COMMACHARB'
1609 `~N,MINCOL,PADCHAR,COMMACHARR'
1613 print a number as a Roman numeral.
1616 print a number as an "old fashioned" Roman numeral.
1619 print a number as an ordinal English number.
1622 print a number as a cardinal English number.
1627 prints `y' and `ies'.
1630 as `~P but jumps 1 argument backward.'
1633 as `~@P but jumps 1 argument backward.'
1638 prints a character as the reader can understand it (i.e. `#\'
1642 prints a character as emacs does (eg. `^C' for ASCII 03).
1645 Fixed-format floating-point (prints a flonum like MMM.NNN).
1646 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
1648 If the number is positive a plus sign is printed.
1651 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
1652 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
1654 If the number is positive a plus sign is printed.
1657 General floating-point (prints a flonum either fixed or
1659 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
1661 If the number is positive a plus sign is printed.
1664 Dollars floating-point (prints a flonum in fixed with signs
1666 `~DIGITS,SCALE,WIDTH,PADCHAR$'
1668 If the number is positive a plus sign is printed.
1671 A sign is always printed and appears before the padding.
1674 The sign appears before the padding.
1682 print newline if not at the beginning of the output line.
1684 prints `~&' and then N-1 newlines.
1689 print N page separators.
1699 newline is ignored, white space left.
1702 newline is left, white space ignored.
1707 relative tabulation.
1713 Indirection (expects indirect arguments as a list).
1715 extracts indirect arguments from format arguments.
1718 Case conversion (converts by `string-downcase').
1720 converts by `string-capitalize'.
1723 converts by `string-capitalize-first'.
1726 converts by `string-upcase'.
1729 Argument Jumping (jumps 1 argument forward).
1731 jumps N arguments forward.
1734 jumps 1 argument backward.
1737 jumps N arguments backward.
1740 jumps to the 0th argument.
1743 jumps to the Nth argument (beginning from 0)
1745 `~[STR0~;STR1~;...~;STRN~]'
1746 Conditional Expression (numerical clause conditional).
1748 take argument from N.
1751 true test conditional.
1754 if-else-then conditional.
1760 default clause follows.
1763 Iteration (args come from the next argument (a list)).
1765 at most N iterations.
1768 args from next arg (a list of lists).
1771 args from the rest of arguments.
1774 args from the rest args (lists).
1785 aborts if N <= M <= K
1787 *** Not Implemented CL Format Control Directives
1790 print `#f' as an empty list (see below).
1793 print `#f' as an empty list (see below).
1799 (sorry I don't understand its semantics completely)
1801 *** Extended, Replaced and Additional Control Directives
1803 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
1804 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
1805 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
1806 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
1807 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
1808 COMMAWIDTH is the number of characters between two comma
1812 print a R4RS complex number as `~F~@Fi' with passed parameters for
1816 Pretty print formatting of an argument for scheme code lists.
1822 Flushes the output if format DESTINATION is a port.
1825 Print a `#\space' character
1827 print N `#\space' characters.
1830 Print a `#\tab' character
1832 print N `#\tab' characters.
1835 Takes N as an integer representation for a character. No arguments
1836 are consumed. N is converted to a character by `integer->char'. N
1837 must be a positive decimal number.
1840 Print out readproof. Prints out internal objects represented as
1841 `#<...>' as strings `"#<...>"' so that the format output can always
1842 be processed by `read'.
1845 Print out readproof. Prints out internal objects represented as
1846 `#<...>' as strings `"#<...>"' so that the format output can always
1847 be processed by `read'.
1850 Prints information and a copyright notice on the format
1853 prints format version.
1856 may also print number strings, i.e. passing a number as a string
1857 and format it accordingly.
1859 *** Configuration Variables
1861 The format module exports some configuration variables to suit the
1862 systems and users needs. There should be no modification necessary for
1863 the configuration that comes with Guile. Format detects automatically
1864 if the running scheme system implements floating point numbers and
1867 format:symbol-case-conv
1868 Symbols are converted by `symbol->string' so the case type of the
1869 printed symbols is implementation dependent.
1870 `format:symbol-case-conv' is a one arg closure which is either
1871 `#f' (no conversion), `string-upcase', `string-downcase' or
1872 `string-capitalize'. (default `#f')
1874 format:iobj-case-conv
1875 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
1876 implementation internal objects. (default `#f')
1879 The character prefixing the exponent value in `~E' printing.
1882 *** Compatibility With Other Format Implementations
1888 Downward compatible except for padding support and `~A', `~S',
1889 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
1890 `printf' padding support which is completely replaced by the CL
1891 `format' padding style.
1894 Downward compatible except for `~', which is not documented
1895 (ignores all characters inside the format string up to a newline
1896 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
1897 numerical and variable parameters and `:/@' modifiers in the CL
1901 Downward compatible except for `~A' and `~S' which print in
1902 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
1903 directive parameters or modifiers)).
1906 Downward compatible except for an optional destination parameter:
1907 S2C accepts a format call without a destination which returns a
1908 formatted string. This is equivalent to a #f destination in S2C.
1909 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
1910 parameters or modifiers)).
1913 ** Changes to string-handling functions.
1915 These functions were added to support the (ice-9 format) module, above.
1917 *** New function: string-upcase STRING
1918 *** New function: string-downcase STRING
1920 These are non-destructive versions of the existing string-upcase! and
1921 string-downcase! functions.
1923 *** New function: string-capitalize! STRING
1924 *** New function: string-capitalize STRING
1926 These functions convert the first letter of each word in the string to
1929 (string-capitalize "howdy there")
1932 As with the other functions, string-capitalize! modifies the string in
1933 place, while string-capitalize returns a modified copy of its argument.
1935 *** New function: string-ci->symbol STRING
1937 Return a symbol whose name is STRING, but having the same case as if
1938 the symbol had be read by `read'.
1940 Guile can be configured to be sensitive or insensitive to case
1941 differences in Scheme identifiers. If Guile is case-insensitive, all
1942 symbols are converted to lower case on input. The `string-ci->symbol'
1943 function returns a symbol whose name in STRING, transformed as Guile
1944 would if STRING were input.
1946 *** New function: substring-move! STRING1 START END STRING2 START
1948 Copy the substring of STRING1 from START (inclusive) to END
1949 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
1950 string, and the source and destination areas may overlap; in all
1951 cases, the function behaves as if all the characters were copied
1954 *** Extended functions: substring-move-left! substring-move-right!
1956 These functions now correctly copy arbitrarily overlapping substrings;
1957 they are both synonyms for substring-move!.
1960 ** New module (ice-9 getopt-long), with the function `getopt-long'.
1962 getopt-long is a function for parsing command-line arguments in a
1963 manner consistent with other GNU programs.
1965 (getopt-long ARGS GRAMMAR)
1966 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
1968 ARGS should be a list of strings. Its first element should be the
1969 name of the program; subsequent elements should be the arguments
1970 that were passed to the program on the command line. The
1971 `program-arguments' procedure returns a list of this form.
1973 GRAMMAR is a list of the form:
1974 ((OPTION (PROPERTY VALUE) ...) ...)
1976 Each OPTION should be a symbol. `getopt-long' will accept a
1977 command-line option named `--OPTION'.
1978 Each option can have the following (PROPERTY VALUE) pairs:
1980 (single-char CHAR) --- Accept `-CHAR' as a single-character
1981 equivalent to `--OPTION'. This is how to specify traditional
1983 (required? BOOL) --- If BOOL is true, the option is required.
1984 getopt-long will raise an error if it is not found in ARGS.
1985 (value BOOL) --- If BOOL is #t, the option accepts a value; if
1986 it is #f, it does not; and if it is the symbol
1987 `optional', the option may appear in ARGS with or
1989 (predicate FUNC) --- If the option accepts a value (i.e. you
1990 specified `(value #t)' for this option), then getopt
1991 will apply FUNC to the value, and throw an exception
1992 if it returns #f. FUNC should be a procedure which
1993 accepts a string and returns a boolean value; you may
1994 need to use quasiquotes to get it into GRAMMAR.
1996 The (PROPERTY VALUE) pairs may occur in any order, but each
1997 property may occur only once. By default, options do not have
1998 single-character equivalents, are not required, and do not take
2001 In ARGS, single-character options may be combined, in the usual
2002 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
2003 accepts values, then it must be the last option in the
2004 combination; the value is the next argument. So, for example, using
2005 the following grammar:
2006 ((apples (single-char #\a))
2007 (blimps (single-char #\b) (value #t))
2008 (catalexis (single-char #\c) (value #t)))
2009 the following argument lists would be acceptable:
2010 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
2011 for "blimps" and "catalexis")
2012 ("-ab" "bang" "-c" "couth") (same)
2013 ("-ac" "couth" "-b" "bang") (same)
2014 ("-abc" "couth" "bang") (an error, since `-b' is not the
2015 last option in its combination)
2017 If an option's value is optional, then `getopt-long' decides
2018 whether it has a value by looking at what follows it in ARGS. If
2019 the next element is a string, and it does not appear to be an
2020 option itself, then that string is the option's value.
2022 The value of a long option can appear as the next element in ARGS,
2023 or it can follow the option name, separated by an `=' character.
2024 Thus, using the same grammar as above, the following argument lists
2026 ("--apples" "Braeburn" "--blimps" "Goodyear")
2027 ("--apples=Braeburn" "--blimps" "Goodyear")
2028 ("--blimps" "Goodyear" "--apples=Braeburn")
2030 If the option "--" appears in ARGS, argument parsing stops there;
2031 subsequent arguments are returned as ordinary arguments, even if
2032 they resemble options. So, in the argument list:
2033 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
2034 `getopt-long' will recognize the `apples' option as having the
2035 value "Granny Smith", but it will not recognize the `blimp'
2036 option; it will return the strings "--blimp" and "Goodyear" as
2037 ordinary argument strings.
2039 The `getopt-long' function returns the parsed argument list as an
2040 assocation list, mapping option names --- the symbols from GRAMMAR
2041 --- onto their values, or #t if the option does not accept a value.
2042 Unused options do not appear in the alist.
2044 All arguments that are not the value of any option are returned
2045 as a list, associated with the empty list.
2047 `getopt-long' throws an exception if:
2048 - it finds an unrecognized option in ARGS
2049 - a required option is omitted
2050 - an option that requires an argument doesn't get one
2051 - an option that doesn't accept an argument does get one (this can
2052 only happen using the long option `--opt=value' syntax)
2053 - an option predicate fails
2058 `((lockfile-dir (required? #t)
2061 (predicate ,file-is-directory?))
2062 (verbose (required? #f)
2065 (x-includes (single-char #\x))
2066 (rnet-server (single-char #\y)
2067 (predicate ,string?))))
2069 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
2070 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
2072 => ((() "foo1" "-fred" "foo2" "foo3")
2073 (rnet-server . "lamprod")
2074 (x-includes . "/usr/include")
2075 (lockfile-dir . "/tmp")
2078 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
2080 It will be removed in a few releases.
2082 ** New syntax: lambda*
2083 ** New syntax: define*
2084 ** New syntax: define*-public
2085 ** New syntax: defmacro*
2086 ** New syntax: defmacro*-public
2087 Guile now supports optional arguments.
2089 `lambda*', `define*', `define*-public', `defmacro*' and
2090 `defmacro*-public' are identical to the non-* versions except that
2091 they use an extended type of parameter list that has the following BNF
2092 syntax (parentheses are literal, square brackets indicate grouping,
2093 and `*', `+' and `?' have the usual meaning):
2095 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
2096 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
2097 [[#&rest identifier]|[. identifier]]? ) | [identifier]
2099 ext-var-decl ::= identifier | ( identifier expression )
2101 The semantics are best illustrated with the following documentation
2102 and examples for `lambda*':
2105 lambda extended for optional and keyword arguments
2107 lambda* creates a procedure that takes optional arguments. These
2108 are specified by putting them inside brackets at the end of the
2109 paramater list, but before any dotted rest argument. For example,
2110 (lambda* (a b #&optional c d . e) '())
2111 creates a procedure with fixed arguments a and b, optional arguments c
2112 and d, and rest argument e. If the optional arguments are omitted
2113 in a call, the variables for them are unbound in the procedure. This
2114 can be checked with the bound? macro.
2116 lambda* can also take keyword arguments. For example, a procedure
2118 (lambda* (#&key xyzzy larch) '())
2119 can be called with any of the argument lists (#:xyzzy 11)
2120 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
2121 are given as keywords are bound to values.
2123 Optional and keyword arguments can also be given default values
2124 which they take on when they are not present in a call, by giving a
2125 two-item list in place of an optional argument, for example in:
2126 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
2127 foo is a fixed argument, bar is an optional argument with default
2128 value 42, and baz is a keyword argument with default value 73.
2129 Default value expressions are not evaluated unless they are needed
2130 and until the procedure is called.
2132 lambda* now supports two more special parameter list keywords.
2134 lambda*-defined procedures now throw an error by default if a
2135 keyword other than one of those specified is found in the actual
2136 passed arguments. However, specifying #&allow-other-keys
2137 immediately after the kyword argument declarations restores the
2138 previous behavior of ignoring unknown keywords. lambda* also now
2139 guarantees that if the same keyword is passed more than once, the
2140 last one passed is the one that takes effect. For example,
2141 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
2142 #:heads 37 #:tails 42 #:heads 99)
2143 would result in (99 47) being displayed.
2145 #&rest is also now provided as a synonym for the dotted syntax rest
2146 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
2147 all respects to lambda*. This is provided for more similarity to DSSSL,
2148 MIT-Scheme and Kawa among others, as well as for refugees from other
2151 Further documentation may be found in the optargs.scm file itself.
2153 The optional argument module also exports the macros `let-optional',
2154 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
2155 are not documented here because they may be removed in the future, but
2156 full documentation is still available in optargs.scm.
2158 ** New syntax: and-let*
2159 Guile now supports the `and-let*' form, described in the draft SRFI-2.
2161 Syntax: (land* (<clause> ...) <body> ...)
2162 Each <clause> should have one of the following forms:
2163 (<variable> <expression>)
2166 Each <variable> or <bound-variable> should be an identifier. Each
2167 <expression> should be a valid expression. The <body> should be a
2168 possibly empty sequence of expressions, like the <body> of a
2171 Semantics: A LAND* expression is evaluated by evaluating the
2172 <expression> or <bound-variable> of each of the <clause>s from
2173 left to right. The value of the first <expression> or
2174 <bound-variable> that evaluates to a false value is returned; the
2175 remaining <expression>s and <bound-variable>s are not evaluated.
2176 The <body> forms are evaluated iff all the <expression>s and
2177 <bound-variable>s evaluate to true values.
2179 The <expression>s and the <body> are evaluated in an environment
2180 binding each <variable> of the preceding (<variable> <expression>)
2181 clauses to the value of the <expression>. Later bindings
2182 shadow earlier bindings.
2184 Guile's and-let* macro was contributed by Michael Livshin.
2186 ** New sorting functions
2188 *** New function: sorted? SEQUENCE LESS?
2189 Returns `#t' when the sequence argument is in non-decreasing order
2190 according to LESS? (that is, there is no adjacent pair `... x y
2191 ...' for which `(less? y x)').
2193 Returns `#f' when the sequence contains at least one out-of-order
2194 pair. It is an error if the sequence is neither a list nor a
2197 *** New function: merge LIST1 LIST2 LESS?
2198 LIST1 and LIST2 are sorted lists.
2199 Returns the sorted list of all elements in LIST1 and LIST2.
2201 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
2202 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
2203 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
2204 (Here "<" should read "comes before".)
2206 *** New procedure: merge! LIST1 LIST2 LESS?
2207 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
2208 the result. If the code is compiled, and LESS? constructs no new
2209 pairs, no pairs at all will be allocated. The first pair of the
2210 result will be either the first pair of LIST1 or the first pair of
2213 *** New function: sort SEQUENCE LESS?
2214 Accepts either a list or a vector, and returns a new sequence
2215 which is sorted. The new sequence is the same type as the input.
2216 Always `(sorted? (sort sequence less?) less?)'. The original
2217 sequence is not altered in any way. The new sequence shares its
2218 elements with the old one; no elements are copied.
2220 *** New procedure: sort! SEQUENCE LESS
2221 Returns its sorted result in the original boxes. No new storage is
2222 allocated at all. Proper usage: (set! slist (sort! slist <))
2224 *** New function: stable-sort SEQUENCE LESS?
2225 Similar to `sort' but stable. That is, if "equal" elements are
2226 ordered a < b in the original sequence, they will have the same order
2229 *** New function: stable-sort! SEQUENCE LESS?
2230 Similar to `sort!' but stable.
2231 Uses temporary storage when sorting vectors.
2233 *** New functions: sort-list, sort-list!
2234 Added for compatibility with scsh.
2236 ** New built-in random number support
2238 *** New function: random N [STATE]
2239 Accepts a positive integer or real N and returns a number of the
2240 same type between zero (inclusive) and N (exclusive). The values
2241 returned have a uniform distribution.
2243 The optional argument STATE must be of the type produced by
2244 `copy-random-state' or `seed->random-state'. It defaults to the value
2245 of the variable `*random-state*'. This object is used to maintain the
2246 state of the pseudo-random-number generator and is altered as a side
2247 effect of the `random' operation.
2249 *** New variable: *random-state*
2250 Holds a data structure that encodes the internal state of the
2251 random-number generator that `random' uses by default. The nature
2252 of this data structure is implementation-dependent. It may be
2253 printed out and successfully read back in, but may or may not
2254 function correctly as a random-number state object in another
2257 *** New function: copy-random-state [STATE]
2258 Returns a new object of type suitable for use as the value of the
2259 variable `*random-state*' and as a second argument to `random'.
2260 If argument STATE is given, a copy of it is returned. Otherwise a
2261 copy of `*random-state*' is returned.
2263 *** New function: seed->random-state SEED
2264 Returns a new object of type suitable for use as the value of the
2265 variable `*random-state*' and as a second argument to `random'.
2266 SEED is a string or a number. A new state is generated and
2267 initialized using SEED.
2269 *** New function: random:uniform [STATE]
2270 Returns an uniformly distributed inexact real random number in the
2271 range between 0 and 1.
2273 *** New procedure: random:solid-sphere! VECT [STATE]
2274 Fills VECT with inexact real random numbers the sum of whose
2275 squares is less than 1.0. Thinking of VECT as coordinates in
2276 space of dimension N = `(vector-length VECT)', the coordinates are
2277 uniformly distributed within the unit N-shere. The sum of the
2278 squares of the numbers is returned. VECT can be either a vector
2279 or a uniform vector of doubles.
2281 *** New procedure: random:hollow-sphere! VECT [STATE]
2282 Fills VECT with inexact real random numbers the sum of whose squares
2283 is equal to 1.0. Thinking of VECT as coordinates in space of
2284 dimension n = `(vector-length VECT)', the coordinates are uniformly
2285 distributed over the surface of the unit n-shere. VECT can be either
2286 a vector or a uniform vector of doubles.
2288 *** New function: random:normal [STATE]
2289 Returns an inexact real in a normal distribution with mean 0 and
2290 standard deviation 1. For a normal distribution with mean M and
2291 standard deviation D use `(+ M (* D (random:normal)))'.
2293 *** New procedure: random:normal-vector! VECT [STATE]
2294 Fills VECT with inexact real random numbers which are independent and
2295 standard normally distributed (i.e., with mean 0 and variance 1).
2296 VECT can be either a vector or a uniform vector of doubles.
2298 *** New function: random:exp STATE
2299 Returns an inexact real in an exponential distribution with mean 1.
2300 For an exponential distribution with mean U use (* U (random:exp)).
2302 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
2304 These functions now operate on numbers in the range of a C unsigned
2307 These functions used to operate on numbers in the range of a C signed
2308 long; however, this seems inappropriate, because Guile integers don't
2311 ** New function: make-guardian
2312 This is an implementation of guardians as described in
2313 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
2314 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
2315 Programming Language Design and Implementation, June 1993
2316 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
2318 ** New functions: delq1!, delv1!, delete1!
2319 These procedures behave similar to delq! and friends but delete only
2320 one object if at all.
2322 ** New function: unread-string STRING PORT
2323 Unread STRING to PORT, that is, push it back onto the port so that
2324 next read operation will work on the pushed back characters.
2326 ** unread-char can now be called multiple times
2327 If unread-char is called multiple times, the unread characters will be
2328 read again in last-in first-out order.
2330 ** the procedures uniform-array-read! and uniform-array-write! now
2331 work on any kind of port, not just ports which are open on a file.
2333 ** Now 'l' in a port mode requests line buffering.
2335 ** The procedure truncate-file now works on string ports as well
2336 as file ports. If the size argument is omitted, the current
2337 file position is used.
2339 ** new procedure: seek PORT/FDES OFFSET WHENCE
2340 The arguments are the same as for the old fseek procedure, but it
2341 works on string ports as well as random-access file ports.
2343 ** the fseek procedure now works on string ports, since it has been
2344 redefined using seek.
2346 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
2347 size is not supplied.
2349 ** the newline procedure no longer flushes the port if it's not
2350 line-buffered: previously it did if it was the current output port.
2352 ** open-pipe and close-pipe are no longer primitive procedures, but
2353 an emulation can be obtained using `(use-modules (ice-9 popen))'.
2355 ** the freopen procedure has been removed.
2357 ** new procedure: drain-input PORT
2358 Drains PORT's read buffers (including any pushed-back characters)
2359 and returns the contents as a single string.
2361 ** New function: map-in-order PROC LIST1 LIST2 ...
2362 Version of `map' which guarantees that the procedure is applied to the
2363 lists in serial order.
2365 ** Renamed `serial-array-copy!' and `serial-array-map!' to
2366 `array-copy-in-order!' and `array-map-in-order!'. The old names are
2367 now obsolete and will go away in release 1.5.
2369 ** New syntax: collect BODY1 ...
2370 Version of `begin' which returns a list of the results of the body
2371 forms instead of the result of the last body form. In contrast to
2372 `begin', `collect' allows an empty body.
2374 ** New functions: read-history FILENAME, write-history FILENAME
2375 Read/write command line history from/to file. Returns #t on success
2376 and #f if an error occured.
2378 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
2380 These procedures return a list of definitions available in the specified
2381 argument, a relative module reference. In the case of no argument,
2382 `(current-module)' is now consulted for definitions to return, instead
2383 of simply returning #f, the former behavior.
2385 ** The #/ syntax for lists is no longer supported.
2387 Earlier versions of Scheme accepted this syntax, but printed a
2390 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
2392 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
2395 * Changes to the gh_ interface
2399 Now takes a second argument which is the result array. If this
2400 pointer is NULL, a new array is malloced (the old behaviour).
2402 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
2403 gh_scm2shorts, gh_scm2longs, gh_scm2floats
2407 * Changes to the scm_ interface
2409 ** Function: scm_make_named_hook (char* name, int n_args)
2411 Creates a hook in the same way as make-hook above but also
2412 binds a variable named NAME to it.
2414 This is the typical way of creating a hook from C code.
2416 Currently, the variable is created in the "current" module. This
2417 might change when we get the new module system.
2419 ** The smob interface
2421 The interface for creating smobs has changed. For documentation, see
2422 data-rep.info (made from guile-core/doc/data-rep.texi).
2424 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
2426 >>> This function will be removed in 1.3.4. <<<
2430 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
2431 This function adds a new smob type, named NAME, with instance size
2432 SIZE to the system. The return value is a tag that is used in
2433 creating instances of the type. If SIZE is 0, then no memory will
2434 be allocated when instances of the smob are created, and nothing
2435 will be freed by the default free function.
2437 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
2438 This function sets the smob marking procedure for the smob type
2439 specified by the tag TC. TC is the tag returned by
2440 `scm_make_smob_type'.
2442 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
2443 This function sets the smob freeing procedure for the smob type
2444 specified by the tag TC. TC is the tag returned by
2445 `scm_make_smob_type'.
2447 *** Function: void scm_set_smob_print (tc, print)
2449 - Function: void scm_set_smob_print (long tc,
2450 scm_sizet (*print) (SCM,
2454 This function sets the smob printing procedure for the smob type
2455 specified by the tag TC. TC is the tag returned by
2456 `scm_make_smob_type'.
2458 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
2459 This function sets the smob equality-testing predicate for the
2460 smob type specified by the tag TC. TC is the tag returned by
2461 `scm_make_smob_type'.
2463 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
2464 Make VALUE contain a smob instance of the type with type code TC and
2465 smob data DATA. VALUE must be previously declared as C type `SCM'.
2467 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
2468 This macro expands to a block of code that creates a smob instance
2469 of the type with type code TC and smob data DATA, and returns that
2470 `SCM' value. It should be the last piece of code in a block.
2472 ** The interfaces for using I/O ports and implementing port types
2473 (ptobs) have changed significantly. The new interface is based on
2474 shared access to buffers and a new set of ptob procedures.
2476 *** scm_newptob has been removed
2480 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
2482 - Function: SCM scm_make_port_type (char *type_name,
2483 int (*fill_buffer) (SCM port),
2484 void (*write_flush) (SCM port));
2486 Similarly to the new smob interface, there is a set of function
2487 setters by which the user can customize the behaviour of his port
2488 type. See ports.h (scm_set_port_XXX).
2490 ** scm_strport_to_string: New function: creates a new string from
2491 a string port's buffer.
2493 ** Plug in interface for random number generators
2494 The variable `scm_the_rng' in random.c contains a value and three
2495 function pointers which together define the current random number
2496 generator being used by the Scheme level interface and the random
2497 number library functions.
2499 The user is free to replace the default generator with the generator
2502 *** Variable: size_t scm_the_rng.rstate_size
2503 The size of the random state type used by the current RNG
2506 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
2507 Given the random STATE, return 32 random bits.
2509 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
2510 Seed random state STATE using string S of length N.
2512 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
2513 Given random state STATE, return a malloced copy.
2516 The default RNG is the MWC (Multiply With Carry) random number
2517 generator described by George Marsaglia at the Department of
2518 Statistics and Supercomputer Computations Research Institute, The
2519 Florida State University (http://stat.fsu.edu/~geo).
2521 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
2522 passes all tests in the DIEHARD test suite
2523 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
2524 costs one multiply and one add on platforms which either supports long
2525 longs (gcc does this on most systems) or have 64 bit longs. The cost
2526 is four multiply on other systems but this can be optimized by writing
2527 scm_i_uniform32 in assembler.
2529 These functions are provided through the scm_the_rng interface for use
2530 by libguile and the application.
2532 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
2533 Given the random STATE, return 32 random bits.
2534 Don't use this function directly. Instead go through the plugin
2535 interface (see "Plug in interface" above).
2537 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
2538 Initialize STATE using SEED of length N.
2540 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
2541 Return a malloc:ed copy of STATE. This function can easily be re-used
2542 in the interfaces to other RNGs.
2544 ** Random number library functions
2545 These functions use the current RNG through the scm_the_rng interface.
2546 It might be a good idea to use these functions from your C code so
2547 that only one random generator is used by all code in your program.
2549 The default random state is stored in:
2551 *** Variable: SCM scm_var_random_state
2552 Contains the vcell of the Scheme variable "*random-state*" which is
2553 used as default state by all random number functions in the Scheme
2558 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
2560 *** Function: scm_rstate *scm_c_default_rstate (void)
2561 This is a convenience function which returns the value of
2562 scm_var_random_state. An error message is generated if this value
2563 isn't a random state.
2565 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
2566 Make a new random state from the string SEED of length LENGTH.
2568 It is generally not a good idea to use multiple random states in a
2569 program. While subsequent random numbers generated from one random
2570 state are guaranteed to be reasonably independent, there is no such
2571 guarantee for numbers generated from different random states.
2573 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
2574 Return 32 random bits.
2576 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
2577 Return a sample from the uniform(0,1) distribution.
2579 *** Function: double scm_c_normal01 (scm_rstate *STATE)
2580 Return a sample from the normal(0,1) distribution.
2582 *** Function: double scm_c_exp1 (scm_rstate *STATE)
2583 Return a sample from the exp(1) distribution.
2585 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
2586 Return a sample from the discrete uniform(0,M) distribution.
2588 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
2589 Return a sample from the discrete uniform(0,M) distribution.
2590 M must be a bignum object. The returned value may be an INUM.
2594 Changes in Guile 1.3 (released Monday, October 19, 1998):
2596 * Changes to the distribution
2598 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
2599 To avoid conflicts, programs should name environment variables after
2600 themselves, except when there's a common practice establishing some
2603 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
2604 giving the former precedence, and printing a warning message if the
2605 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
2607 ** The header files related to multi-byte characters have been removed.
2608 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
2609 which referred to these explicitly will probably need to be rewritten,
2610 since the support for the variant string types has been removed; see
2613 ** The header files append.h and sequences.h have been removed. These
2614 files implemented non-R4RS operations which would encourage
2615 non-portable programming style and less easy-to-read code.
2617 * Changes to the stand-alone interpreter
2619 ** New procedures have been added to implement a "batch mode":
2621 *** Function: batch-mode?
2623 Returns a boolean indicating whether the interpreter is in batch
2626 *** Function: set-batch-mode?! ARG
2628 If ARG is true, switches the interpreter to batch mode. The `#f'
2629 case has not been implemented.
2631 ** Guile now provides full command-line editing, when run interactively.
2632 To use this feature, you must have the readline library installed.
2633 The Guile build process will notice it, and automatically include
2636 The readline library is available via anonymous FTP from any GNU
2637 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
2639 ** the-last-stack is now a fluid.
2641 * Changes to the procedure for linking libguile with your programs
2643 ** You can now use the `guile-config' utility to build programs that use Guile.
2645 Guile now includes a command-line utility called `guile-config', which
2646 can provide information about how to compile and link programs that
2649 *** `guile-config compile' prints any C compiler flags needed to use Guile.
2650 You should include this command's output on the command line you use
2651 to compile C or C++ code that #includes the Guile header files. It's
2652 usually just a `-I' flag to help the compiler find the Guile headers.
2655 *** `guile-config link' prints any linker flags necessary to link with Guile.
2657 This command writes to its standard output a list of flags which you
2658 must pass to the linker to link your code against the Guile library.
2659 The flags include '-lguile' itself, any other libraries the Guile
2660 library depends upon, and any `-L' flags needed to help the linker
2661 find those libraries.
2663 For example, here is a Makefile rule that builds a program named 'foo'
2664 from the object files ${FOO_OBJECTS}, and links them against Guile:
2667 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
2669 Previous Guile releases recommended that you use autoconf to detect
2670 which of a predefined set of libraries were present on your system.
2671 It is more robust to use `guile-config', since it records exactly which
2672 libraries the installed Guile library requires.
2674 This was originally called `build-guile', but was renamed to
2675 `guile-config' before Guile 1.3 was released, to be consistent with
2676 the analogous script for the GTK+ GUI toolkit, which is called
2680 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
2682 If you are using the GNU autoconf package to configure your program,
2683 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
2684 (described above) and gather the necessary values for use in your
2687 The GUILE_FLAGS macro expands to configure script code which runs the
2688 `guile-config' script, to find out where Guile's header files and
2689 libraries are installed. It sets two variables, marked for
2690 substitution, as by AC_SUBST.
2692 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
2693 code that uses Guile header files. This is almost always just a
2696 GUILE_LDFLAGS --- flags to pass to the linker to link a
2697 program against Guile. This includes `-lguile' for the Guile
2698 library itself, any libraries that Guile itself requires (like
2699 -lqthreads), and so on. It may also include a -L flag to tell the
2700 compiler where to find the libraries.
2702 GUILE_FLAGS is defined in the file guile.m4, in the top-level
2703 directory of the Guile distribution. You can copy it into your
2704 package's aclocal.m4 file, and then use it in your configure.in file.
2706 If you are using the `aclocal' program, distributed with GNU automake,
2707 to maintain your aclocal.m4 file, the Guile installation process
2708 installs guile.m4 where aclocal will find it. All you need to do is
2709 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
2710 this will copy the definition of GUILE_FLAGS into your aclocal.m4
2714 * Changes to Scheme functions and syntax
2716 ** Multi-byte strings have been removed, as have multi-byte and wide
2717 ports. We felt that these were the wrong approach to
2718 internationalization support.
2720 ** New function: readline [PROMPT]
2721 Read a line from the terminal, and allow the user to edit it,
2722 prompting with PROMPT. READLINE provides a large set of Emacs-like
2723 editing commands, lets the user recall previously typed lines, and
2724 works on almost every kind of terminal, including dumb terminals.
2726 READLINE assumes that the cursor is at the beginning of the line when
2727 it is invoked. Thus, you can't print a prompt yourself, and then call
2728 READLINE; you need to package up your prompt as a string, pass it to
2729 the function, and let READLINE print the prompt itself. This is
2730 because READLINE needs to know the prompt's screen width.
2732 For Guile to provide this function, you must have the readline
2733 library, version 2.1 or later, installed on your system. Readline is
2734 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
2735 any GNU mirror site.
2737 See also ADD-HISTORY function.
2739 ** New function: add-history STRING
2740 Add STRING as the most recent line in the history used by the READLINE
2741 command. READLINE does not add lines to the history itself; you must
2742 call ADD-HISTORY to make previous input available to the user.
2744 ** The behavior of the read-line function has changed.
2746 This function now uses standard C library functions to read the line,
2747 for speed. This means that it doesn not respect the value of
2748 scm-line-incrementors; it assumes that lines are delimited with
2751 (Note that this is read-line, the function that reads a line of text
2752 from a port, not readline, the function that reads a line from a
2753 terminal, providing full editing capabilities.)
2755 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
2757 This module provides some simple argument parsing. It exports one
2760 Function: getopt-gnu-style ARG-LS
2761 Parse a list of program arguments into an alist of option
2764 Each item in the list of program arguments is examined to see if
2765 it meets the syntax of a GNU long-named option. An argument like
2766 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
2767 returned alist, where MUMBLE is a keyword object with the same
2768 name as the argument. An argument like `--MUMBLE=FROB' produces
2769 an element of the form (MUMBLE . FROB), where FROB is a string.
2771 As a special case, the returned alist also contains a pair whose
2772 car is the symbol `rest'. The cdr of this pair is a list
2773 containing all the items in the argument list that are not options
2774 of the form mentioned above.
2776 The argument `--' is treated specially: all items in the argument
2777 list appearing after such an argument are not examined, and are
2778 returned in the special `rest' list.
2780 This function does not parse normal single-character switches.
2781 You will need to parse them out of the `rest' list yourself.
2783 ** The read syntax for byte vectors and short vectors has changed.
2785 Instead of #bytes(...), write #y(...).
2787 Instead of #short(...), write #h(...).
2789 This may seem nutty, but, like the other uniform vectors, byte vectors
2790 and short vectors want to have the same print and read syntax (and,
2791 more basic, want to have read syntax!). Changing the read syntax to
2792 use multiple characters after the hash sign breaks with the
2793 conventions used in R5RS and the conventions used for the other
2794 uniform vectors. It also introduces complexity in the current reader,
2795 both on the C and Scheme levels. (The Right solution is probably to
2796 change the syntax and prototypes for uniform vectors entirely.)
2799 ** The new module (ice-9 session) provides useful interactive functions.
2801 *** New procedure: (apropos REGEXP OPTION ...)
2803 Display a list of top-level variables whose names match REGEXP, and
2804 the modules they are imported from. Each OPTION should be one of the
2807 value --- Show the value of each matching variable.
2808 shadow --- Show bindings shadowed by subsequently imported modules.
2809 full --- Same as both `shadow' and `value'.
2813 guile> (apropos "trace" 'full)
2814 debug: trace #<procedure trace args>
2815 debug: untrace #<procedure untrace args>
2816 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
2817 the-scm-module: before-backtrace-hook ()
2818 the-scm-module: backtrace #<primitive-procedure backtrace>
2819 the-scm-module: after-backtrace-hook ()
2820 the-scm-module: has-shown-backtrace-hint? #f
2823 ** There are new functions and syntax for working with macros.
2825 Guile implements macros as a special object type. Any variable whose
2826 top-level binding is a macro object acts as a macro. The macro object
2827 specifies how the expression should be transformed before evaluation.
2829 *** Macro objects now print in a reasonable way, resembling procedures.
2831 *** New function: (macro? OBJ)
2832 True iff OBJ is a macro object.
2834 *** New function: (primitive-macro? OBJ)
2835 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
2836 macro transformers, implemented in eval.c rather than Scheme code.
2838 Why do we have this function?
2839 - For symmetry with procedure? and primitive-procedure?,
2840 - to allow custom print procedures to tell whether a macro is
2841 primitive, and display it differently, and
2842 - to allow compilers and user-written evaluators to distinguish
2843 builtin special forms from user-defined ones, which could be
2846 *** New function: (macro-type OBJ)
2847 Return a value indicating what kind of macro OBJ is. Possible return
2850 The symbol `syntax' --- a macro created by procedure->syntax.
2851 The symbol `macro' --- a macro created by procedure->macro.
2852 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
2853 The boolean #f --- if OBJ is not a macro object.
2855 *** New function: (macro-name MACRO)
2856 Return the name of the macro object MACRO's procedure, as returned by
2859 *** New function: (macro-transformer MACRO)
2860 Return the transformer procedure for MACRO.
2862 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
2864 Specify a new macro expander to use in the current module. Each
2865 MODULE is a module name, with the same meaning as in the `use-modules'
2866 form; each named module's exported bindings are added to the current
2867 top-level environment. TRANSFORMER is an expression evaluated in the
2868 resulting environment which must yield a procedure to use as the
2869 module's eval transformer: every expression evaluated in this module
2870 is passed to this function, and the result passed to the Guile
2873 *** macro-eval! is removed. Use local-eval instead.
2875 ** Some magic has been added to the printer to better handle user
2876 written printing routines (like record printers, closure printers).
2878 The problem is that these user written routines must have access to
2879 the current `print-state' to be able to handle fancy things like
2880 detection of circular references. These print-states have to be
2881 passed to the builtin printing routines (display, write, etc) to
2882 properly continue the print chain.
2884 We didn't want to change all existing print code so that it
2885 explicitly passes thru a print state in addition to a port. Instead,
2886 we extented the possible values that the builtin printing routines
2887 accept as a `port'. In addition to a normal port, they now also take
2888 a pair of a normal port and a print-state. Printing will go to the
2889 port and the print-state will be used to control the detection of
2890 circular references, etc. If the builtin function does not care for a
2891 print-state, it is simply ignored.
2893 User written callbacks are now called with such a pair as their
2894 `port', but because every function now accepts this pair as a PORT
2895 argument, you don't have to worry about that. In fact, it is probably
2896 safest to not check for these pairs.
2898 However, it is sometimes necessary to continue a print chain on a
2899 different port, for example to get a intermediate string
2900 representation of the printed value, mangle that string somehow, and
2901 then to finally print the mangled string. Use the new function
2903 inherit-print-state OLD-PORT NEW-PORT
2905 for this. It constructs a new `port' that prints to NEW-PORT but
2906 inherits the print-state of OLD-PORT.
2908 ** struct-vtable-offset renamed to vtable-offset-user
2910 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
2912 ** There is now a third optional argument to make-vtable-vtable
2913 (and fourth to make-struct) when constructing new types (vtables).
2914 This argument initializes field vtable-index-printer of the vtable.
2916 ** The detection of circular references has been extended to structs.
2917 That is, a structure that -- in the process of being printed -- prints
2918 itself does not lead to infinite recursion.
2920 ** There is now some basic support for fluids. Please read
2921 "libguile/fluid.h" to find out more. It is accessible from Scheme with
2922 the following functions and macros:
2924 Function: make-fluid
2926 Create a new fluid object. Fluids are not special variables or
2927 some other extension to the semantics of Scheme, but rather
2928 ordinary Scheme objects. You can store them into variables (that
2929 are still lexically scoped, of course) or into any other place you
2930 like. Every fluid has a initial value of `#f'.
2932 Function: fluid? OBJ
2934 Test whether OBJ is a fluid.
2936 Function: fluid-ref FLUID
2937 Function: fluid-set! FLUID VAL
2939 Access/modify the fluid FLUID. Modifications are only visible
2940 within the current dynamic root (that includes threads).
2942 Function: with-fluids* FLUIDS VALUES THUNK
2944 FLUIDS is a list of fluids and VALUES a corresponding list of
2945 values for these fluids. Before THUNK gets called the values are
2946 installed in the fluids and the old values of the fluids are
2947 saved in the VALUES list. When the flow of control leaves THUNK
2948 or reenters it, the values get swapped again. You might think of
2949 this as a `safe-fluid-excursion'. Note that the VALUES list is
2950 modified by `with-fluids*'.
2952 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
2954 The same as `with-fluids*' but with a different syntax. It looks
2955 just like `let', but both FLUID and VALUE are evaluated. Remember,
2956 fluids are not special variables but ordinary objects. FLUID
2957 should evaluate to a fluid.
2959 ** Changes to system call interfaces:
2961 *** close-port, close-input-port and close-output-port now return a
2962 boolean instead of an `unspecified' object. #t means that the port
2963 was successfully closed, while #f means it was already closed. It is
2964 also now possible for these procedures to raise an exception if an
2965 error occurs (some errors from write can be delayed until close.)
2967 *** the first argument to chmod, fcntl, ftell and fseek can now be a
2970 *** the third argument to fcntl is now optional.
2972 *** the first argument to chown can now be a file descriptor or a port.
2974 *** the argument to stat can now be a port.
2976 *** The following new procedures have been added (most use scsh
2979 *** procedure: close PORT/FD
2980 Similar to close-port (*note close-port: Closing Ports.), but also
2981 works on file descriptors. A side effect of closing a file
2982 descriptor is that any ports using that file descriptor are moved
2983 to a different file descriptor and have their revealed counts set
2986 *** procedure: port->fdes PORT
2987 Returns the integer file descriptor underlying PORT. As a side
2988 effect the revealed count of PORT is incremented.
2990 *** procedure: fdes->ports FDES
2991 Returns a list of existing ports which have FDES as an underlying
2992 file descriptor, without changing their revealed counts.
2994 *** procedure: fdes->inport FDES
2995 Returns an existing input port which has FDES as its underlying
2996 file descriptor, if one exists, and increments its revealed count.
2997 Otherwise, returns a new input port with a revealed count of 1.
2999 *** procedure: fdes->outport FDES
3000 Returns an existing output port which has FDES as its underlying
3001 file descriptor, if one exists, and increments its revealed count.
3002 Otherwise, returns a new output port with a revealed count of 1.
3004 The next group of procedures perform a `dup2' system call, if NEWFD
3005 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
3006 duplicated can be supplied as an integer or contained in a port. The
3007 type of value returned varies depending on which procedure is used.
3009 All procedures also have the side effect when performing `dup2' that
3010 any ports using NEWFD are moved to a different file descriptor and have
3011 their revealed counts set to zero.
3013 *** procedure: dup->fdes PORT/FD [NEWFD]
3014 Returns an integer file descriptor.
3016 *** procedure: dup->inport PORT/FD [NEWFD]
3017 Returns a new input port using the new file descriptor.
3019 *** procedure: dup->outport PORT/FD [NEWFD]
3020 Returns a new output port using the new file descriptor.
3022 *** procedure: dup PORT/FD [NEWFD]
3023 Returns a new port if PORT/FD is a port, with the same mode as the
3024 supplied port, otherwise returns an integer file descriptor.
3026 *** procedure: dup->port PORT/FD MODE [NEWFD]
3027 Returns a new port using the new file descriptor. MODE supplies a
3028 mode string for the port (*note open-file: File Ports.).
3030 *** procedure: setenv NAME VALUE
3031 Modifies the environment of the current process, which is also the
3032 default environment inherited by child processes.
3034 If VALUE is `#f', then NAME is removed from the environment.
3035 Otherwise, the string NAME=VALUE is added to the environment,
3036 replacing any existing string with name matching NAME.
3038 The return value is unspecified.
3040 *** procedure: truncate-file OBJ SIZE
3041 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
3042 can be a string containing a file name or an integer file
3043 descriptor or port open for output on the file. The underlying
3044 system calls are `truncate' and `ftruncate'.
3046 The return value is unspecified.
3048 *** procedure: setvbuf PORT MODE [SIZE]
3049 Set the buffering mode for PORT. MODE can be:
3057 block buffered, using a newly allocated buffer of SIZE bytes.
3058 However if SIZE is zero or unspecified, the port will be made
3061 This procedure should not be used after I/O has been performed with
3064 Ports are usually block buffered by default, with a default buffer
3065 size. Procedures e.g., *Note open-file: File Ports, which accept a
3066 mode string allow `0' to be added to request an unbuffered port.
3068 *** procedure: fsync PORT/FD
3069 Copies any unwritten data for the specified output file descriptor
3070 to disk. If PORT/FD is a port, its buffer is flushed before the
3071 underlying file descriptor is fsync'd. The return value is
3074 *** procedure: open-fdes PATH FLAGS [MODES]
3075 Similar to `open' but returns a file descriptor instead of a port.
3077 *** procedure: execle PATH ENV [ARG] ...
3078 Similar to `execl', but the environment of the new process is
3079 specified by ENV, which must be a list of strings as returned by
3080 the `environ' procedure.
3082 This procedure is currently implemented using the `execve' system
3083 call, but we call it `execle' because of its Scheme calling
3086 *** procedure: strerror ERRNO
3087 Returns the Unix error message corresponding to ERRNO, an integer.
3089 *** procedure: primitive-exit [STATUS]
3090 Terminate the current process without unwinding the Scheme stack.
3091 This is would typically be useful after a fork. The exit status
3092 is STATUS if supplied, otherwise zero.
3094 *** procedure: times
3095 Returns an object with information about real and processor time.
3096 The following procedures accept such an object as an argument and
3097 return a selected component:
3100 The current real time, expressed as time units relative to an
3104 The CPU time units used by the calling process.
3107 The CPU time units used by the system on behalf of the
3111 The CPU time units used by terminated child processes of the
3112 calling process, whose status has been collected (e.g., using
3116 Similarly, the CPU times units used by the system on behalf of
3117 terminated child processes.
3119 ** Removed: list-length
3120 ** Removed: list-append, list-append!
3121 ** Removed: list-reverse, list-reverse!
3123 ** array-map renamed to array-map!
3125 ** serial-array-map renamed to serial-array-map!
3127 ** catch doesn't take #f as first argument any longer
3129 Previously, it was possible to pass #f instead of a key to `catch'.
3130 That would cause `catch' to pass a jump buffer object to the procedure
3131 passed as second argument. The procedure could then use this jump
3132 buffer objekt as an argument to throw.
3134 This mechanism has been removed since its utility doesn't motivate the
3135 extra complexity it introduces.
3137 ** The `#/' notation for lists now provokes a warning message from Guile.
3138 This syntax will be removed from Guile in the near future.
3140 To disable the warning message, set the GUILE_HUSH environment
3141 variable to any non-empty value.
3143 ** The newline character now prints as `#\newline', following the
3144 normal Scheme notation, not `#\nl'.
3146 * Changes to the gh_ interface
3148 ** The gh_enter function now takes care of loading the Guile startup files.
3149 gh_enter works by calling scm_boot_guile; see the remarks below.
3151 ** Function: void gh_write (SCM x)
3153 Write the printed representation of the scheme object x to the current
3154 output port. Corresponds to the scheme level `write'.
3156 ** gh_list_length renamed to gh_length.
3158 ** vector handling routines
3160 Several major changes. In particular, gh_vector() now resembles
3161 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
3162 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
3163 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
3164 vector-related gh_ functions have been implemented.
3166 ** pair and list routines
3168 Implemented several of the R4RS pair and list functions that were
3171 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
3173 New function. Converts double arrays back and forth between Scheme
3176 * Changes to the scm_ interface
3178 ** The function scm_boot_guile now takes care of loading the startup files.
3180 Guile's primary initialization function, scm_boot_guile, now takes
3181 care of loading `boot-9.scm', in the `ice-9' module, to initialize
3182 Guile, define the module system, and put together some standard
3183 bindings. It also loads `init.scm', which is intended to hold
3184 site-specific initialization code.
3186 Since Guile cannot operate properly until boot-9.scm is loaded, there
3187 is no reason to separate loading boot-9.scm from Guile's other
3188 initialization processes.
3190 This job used to be done by scm_compile_shell_switches, which didn't
3191 make much sense; in particular, it meant that people using Guile for
3192 non-shell-like applications had to jump through hoops to get Guile
3193 initialized properly.
3195 ** The function scm_compile_shell_switches no longer loads the startup files.
3196 Now, Guile always loads the startup files, whenever it is initialized;
3197 see the notes above for scm_boot_guile and scm_load_startup_files.
3199 ** Function: scm_load_startup_files
3200 This new function takes care of loading Guile's initialization file
3201 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
3202 this is always called by the Guile initialization process, it's
3203 probably not too useful to call this yourself, but it's there anyway.
3205 ** The semantics of smob marking have changed slightly.
3207 The smob marking function (the `mark' member of the scm_smobfuns
3208 structure) is no longer responsible for setting the mark bit on the
3209 smob. The generic smob handling code in the garbage collector will
3210 set this bit. The mark function need only ensure that any other
3211 objects the smob refers to get marked.
3213 Note that this change means that the smob's GC8MARK bit is typically
3214 already set upon entry to the mark function. Thus, marking functions
3215 which look like this:
3218 if (SCM_GC8MARKP (ptr))
3220 SCM_SETGC8MARK (ptr);
3221 ... mark objects to which the smob refers ...
3224 are now incorrect, since they will return early, and fail to mark any
3225 other objects the smob refers to. Some code in the Guile library used
3228 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
3230 If you have implemented your own I/O port type, by writing the
3231 functions required by the scm_ptobfuns and then calling scm_newptob,
3232 you will need to change your functions slightly.
3234 The functions in a scm_ptobfuns structure now expect the port itself
3235 as their argument; they used to expect the `stream' member of the
3236 port's scm_port_table structure. This allows functions in an
3237 scm_ptobfuns structure to easily access the port's cell (and any flags
3238 it its CAR), and the port's scm_port_table structure.
3240 Guile now passes the I/O port itself as the `port' argument in the
3241 following scm_ptobfuns functions:
3243 int (*free) (SCM port);
3244 int (*fputc) (int, SCM port);
3245 int (*fputs) (char *, SCM port);
3246 scm_sizet (*fwrite) SCM_P ((char *ptr,
3250 int (*fflush) (SCM port);
3251 int (*fgetc) (SCM port);
3252 int (*fclose) (SCM port);
3254 The interfaces to the `mark', `print', `equalp', and `fgets' methods
3257 If you have existing code which defines its own port types, it is easy
3258 to convert your code to the new interface; simply apply SCM_STREAM to
3259 the port argument to yield the value you code used to expect.
3261 Note that since both the port and the stream have the same type in the
3262 C code --- they are both SCM values --- the C compiler will not remind
3263 you if you forget to update your scm_ptobfuns functions.
3266 ** Function: int scm_internal_select (int fds,
3270 struct timeval *timeout);
3272 This is a replacement for the `select' function provided by the OS.
3273 It enables I/O blocking and sleeping to happen for one cooperative
3274 thread without blocking other threads. It also avoids busy-loops in
3275 these situations. It is intended that all I/O blocking and sleeping
3276 will finally go through this function. Currently, this function is
3277 only available on systems providing `gettimeofday' and `select'.
3279 ** Function: SCM scm_internal_stack_catch (SCM tag,
3280 scm_catch_body_t body,
3282 scm_catch_handler_t handler,
3285 A new sibling to the other two C level `catch' functions
3286 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
3287 the stack to be saved automatically into the variable `the-last-stack'
3288 (scm_the_last_stack_var) on error. This is necessary if you want to
3289 use advanced error reporting, such as calling scm_display_error and
3290 scm_display_backtrace. (They both take a stack object as argument.)
3292 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
3294 scm_catch_handler_t handler,
3297 Spawns a new thread. It does a job similar to
3298 scm_call_with_new_thread but takes arguments more suitable when
3299 spawning threads from application C code.
3301 ** The hook scm_error_callback has been removed. It was originally
3302 intended as a way for the user to install his own error handler. But
3303 that method works badly since it intervenes between throw and catch,
3304 thereby changing the semantics of expressions like (catch #t ...).
3305 The correct way to do it is to use one of the C level catch functions
3306 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
3308 ** Removed functions:
3310 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
3311 scm_list_reverse, scm_list_reverse_x
3313 ** New macros: SCM_LISTn where n is one of the integers 0-9.
3315 These can be used for pretty list creation from C. The idea is taken
3316 from Erick Gallesio's STk.
3318 ** scm_array_map renamed to scm_array_map_x
3320 ** mbstrings are now removed
3322 This means that the type codes scm_tc7_mb_string and
3323 scm_tc7_mb_substring has been removed.
3325 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
3327 Since we no longer support multi-byte strings, these I/O functions
3328 have been simplified, and renamed. Here are their old names, and
3329 their new names and arguments:
3331 scm_gen_putc -> void scm_putc (int c, SCM port);
3332 scm_gen_puts -> void scm_puts (char *s, SCM port);
3333 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
3334 scm_gen_getc -> void scm_getc (SCM port);
3337 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
3339 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
3341 SCM_TYP7S now masks away the bit which distinguishes substrings from
3344 ** scm_catch_body_t: Backward incompatible change!
3346 Body functions to scm_internal_catch and friends do not any longer
3347 take a second argument. This is because it is no longer possible to
3348 pass a #f arg to catch.
3350 ** Calls to scm_protect_object and scm_unprotect now nest properly.
3352 The function scm_protect_object protects its argument from being freed
3353 by the garbage collector. scm_unprotect_object removes that
3356 These functions now nest properly. That is, for every object O, there
3357 is a counter which scm_protect_object(O) increments and
3358 scm_unprotect_object(O) decrements, if the counter is greater than
3359 zero. Every object's counter is zero when it is first created. If an
3360 object's counter is greater than zero, the garbage collector will not
3361 reclaim its storage.
3363 This allows you to use scm_protect_object in your code without
3364 worrying that some other function you call will call
3365 scm_unprotect_object, and allow it to be freed. Assuming that the
3366 functions you call are well-behaved, and unprotect only those objects
3367 they protect, you can follow the same rule and have confidence that
3368 objects will be freed only at appropriate times.
3371 Changes in Guile 1.2 (released Tuesday, June 24 1997):
3373 * Changes to the distribution
3375 ** Nightly snapshots are now available from ftp.red-bean.com.
3376 The old server, ftp.cyclic.com, has been relinquished to its rightful
3379 Nightly snapshots of the Guile development sources are now available via
3380 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
3382 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
3383 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
3385 ** To run Guile without installing it, the procedure has changed a bit.
3387 If you used a separate build directory to compile Guile, you'll need
3388 to include the build directory in SCHEME_LOAD_PATH, as well as the
3389 source directory. See the `INSTALL' file for examples.
3391 * Changes to the procedure for linking libguile with your programs
3393 ** The standard Guile load path for Scheme code now includes
3394 $(datadir)/guile (usually /usr/local/share/guile). This means that
3395 you can install your own Scheme files there, and Guile will find them.
3396 (Previous versions of Guile only checked a directory whose name
3397 contained the Guile version number, so you had to re-install or move
3398 your Scheme sources each time you installed a fresh version of Guile.)
3400 The load path also includes $(datadir)/guile/site; we recommend
3401 putting individual Scheme files there. If you want to install a
3402 package with multiple source files, create a directory for them under
3405 ** Guile 1.2 will now use the Rx regular expression library, if it is
3406 installed on your system. When you are linking libguile into your own
3407 programs, this means you will have to link against -lguile, -lqt (if
3408 you configured Guile with thread support), and -lrx.
3410 If you are using autoconf to generate configuration scripts for your
3411 application, the following lines should suffice to add the appropriate
3412 libraries to your link command:
3414 ### Find Rx, quickthreads and libguile.
3415 AC_CHECK_LIB(rx, main)
3416 AC_CHECK_LIB(qt, main)
3417 AC_CHECK_LIB(guile, scm_shell)
3419 The Guile 1.2 distribution does not contain sources for the Rx
3420 library, as Guile 1.0 did. If you want to use Rx, you'll need to
3421 retrieve it from a GNU FTP site and install it separately.
3423 * Changes to Scheme functions and syntax
3425 ** The dynamic linking features of Guile are now enabled by default.
3426 You can disable them by giving the `--disable-dynamic-linking' option
3429 (dynamic-link FILENAME)
3431 Find the object file denoted by FILENAME (a string) and link it
3432 into the running Guile application. When everything works out,
3433 return a Scheme object suitable for representing the linked object
3434 file. Otherwise an error is thrown. How object files are
3435 searched is system dependent.
3437 (dynamic-object? VAL)
3439 Determine whether VAL represents a dynamically linked object file.
3441 (dynamic-unlink DYNOBJ)
3443 Unlink the indicated object file from the application. DYNOBJ
3444 should be one of the values returned by `dynamic-link'.
3446 (dynamic-func FUNCTION DYNOBJ)
3448 Search the C function indicated by FUNCTION (a string or symbol)
3449 in DYNOBJ and return some Scheme object that can later be used
3450 with `dynamic-call' to actually call this function. Right now,
3451 these Scheme objects are formed by casting the address of the
3452 function to `long' and converting this number to its Scheme
3455 (dynamic-call FUNCTION DYNOBJ)
3457 Call the C function indicated by FUNCTION and DYNOBJ. The
3458 function is passed no arguments and its return value is ignored.
3459 When FUNCTION is something returned by `dynamic-func', call that
3460 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
3461 etc.), look it up in DYNOBJ; this is equivalent to
3463 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
3465 Interrupts are deferred while the C function is executing (with
3466 SCM_DEFER_INTS/SCM_ALLOW_INTS).
3468 (dynamic-args-call FUNCTION DYNOBJ ARGS)
3470 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
3471 some arguments and return its return value. The C function is
3472 expected to take two arguments and return an `int', just like
3475 int c_func (int argc, char **argv);
3477 ARGS must be a list of strings and is converted into an array of
3478 `char *'. The array is passed in ARGV and its size in ARGC. The
3479 return value is converted to a Scheme number and returned from the
3480 call to `dynamic-args-call'.
3482 When dynamic linking is disabled or not supported on your system,
3483 the above functions throw errors, but they are still available.
3485 Here is a small example that works on GNU/Linux:
3487 (define libc-obj (dynamic-link "libc.so"))
3488 (dynamic-args-call 'rand libc-obj '())
3490 See the file `libguile/DYNAMIC-LINKING' for additional comments.
3492 ** The #/ syntax for module names is depreciated, and will be removed
3493 in a future version of Guile. Instead of
3501 The latter syntax is more consistent with existing Lisp practice.
3503 ** Guile now does fancier printing of structures. Structures are the
3504 underlying implementation for records, which in turn are used to
3505 implement modules, so all of these object now print differently and in
3506 a more informative way.
3508 The Scheme printer will examine the builtin variable *struct-printer*
3509 whenever it needs to print a structure object. When this variable is
3510 not `#f' it is deemed to be a procedure and will be applied to the
3511 structure object and the output port. When *struct-printer* is `#f'
3512 or the procedure return `#f' the structure object will be printed in
3513 the boring #<struct 80458270> form.
3515 This hook is used by some routines in ice-9/boot-9.scm to implement
3516 type specific printing routines. Please read the comments there about
3519 One of the more specific uses of structs are records. The printing
3520 procedure that could be passed to MAKE-RECORD-TYPE is now actually
3521 called. It should behave like a *struct-printer* procedure (described
3524 ** Guile now supports a new R4RS-compliant syntax for keywords. A
3525 token of the form #:NAME, where NAME has the same syntax as a Scheme
3526 symbol, is the external representation of the keyword named NAME.
3527 Keyword objects print using this syntax as well, so values containing
3528 keyword objects can be read back into Guile. When used in an
3529 expression, keywords are self-quoting objects.
3531 Guile suports this read syntax, and uses this print syntax, regardless
3532 of the current setting of the `keyword' read option. The `keyword'
3533 read option only controls whether Guile recognizes the `:NAME' syntax,
3534 which is incompatible with R4RS. (R4RS says such token represent
3537 ** Guile has regular expression support again. Guile 1.0 included
3538 functions for matching regular expressions, based on the Rx library.
3539 In Guile 1.1, the Guile/Rx interface was removed to simplify the
3540 distribution, and thus Guile had no regular expression support. Guile
3541 1.2 again supports the most commonly used functions, and supports all
3542 of SCSH's regular expression functions.
3544 If your system does not include a POSIX regular expression library,
3545 and you have not linked Guile with a third-party regexp library such as
3546 Rx, these functions will not be available. You can tell whether your
3547 Guile installation includes regular expression support by checking
3548 whether the `*features*' list includes the `regex' symbol.
3550 *** regexp functions
3552 By default, Guile supports POSIX extended regular expressions. That
3553 means that the characters `(', `)', `+' and `?' are special, and must
3554 be escaped if you wish to match the literal characters.
3556 This regular expression interface was modeled after that implemented
3557 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
3558 with SCSH regular expressions.
3560 **** Function: string-match PATTERN STR [START]
3561 Compile the string PATTERN into a regular expression and compare
3562 it with STR. The optional numeric argument START specifies the
3563 position of STR at which to begin matching.
3565 `string-match' returns a "match structure" which describes what,
3566 if anything, was matched by the regular expression. *Note Match
3567 Structures::. If STR does not match PATTERN at all,
3568 `string-match' returns `#f'.
3570 Each time `string-match' is called, it must compile its PATTERN
3571 argument into a regular expression structure. This operation is
3572 expensive, which makes `string-match' inefficient if the same regular
3573 expression is used several times (for example, in a loop). For better
3574 performance, you can compile a regular expression in advance and then
3575 match strings against the compiled regexp.
3577 **** Function: make-regexp STR [FLAGS]
3578 Compile the regular expression described by STR, and return the
3579 compiled regexp structure. If STR does not describe a legal
3580 regular expression, `make-regexp' throws a
3581 `regular-expression-syntax' error.
3583 FLAGS may be the bitwise-or of one or more of the following:
3585 **** Constant: regexp/extended
3586 Use POSIX Extended Regular Expression syntax when interpreting
3587 STR. If not set, POSIX Basic Regular Expression syntax is used.
3588 If the FLAGS argument is omitted, we assume regexp/extended.
3590 **** Constant: regexp/icase
3591 Do not differentiate case. Subsequent searches using the
3592 returned regular expression will be case insensitive.
3594 **** Constant: regexp/newline
3595 Match-any-character operators don't match a newline.
3597 A non-matching list ([^...]) not containing a newline matches a
3600 Match-beginning-of-line operator (^) matches the empty string
3601 immediately after a newline, regardless of whether the FLAGS
3602 passed to regexp-exec contain regexp/notbol.
3604 Match-end-of-line operator ($) matches the empty string
3605 immediately before a newline, regardless of whether the FLAGS
3606 passed to regexp-exec contain regexp/noteol.
3608 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
3609 Match the compiled regular expression REGEXP against `str'. If
3610 the optional integer START argument is provided, begin matching
3611 from that position in the string. Return a match structure
3612 describing the results of the match, or `#f' if no match could be
3615 FLAGS may be the bitwise-or of one or more of the following:
3617 **** Constant: regexp/notbol
3618 The match-beginning-of-line operator always fails to match (but
3619 see the compilation flag regexp/newline above) This flag may be
3620 used when different portions of a string are passed to
3621 regexp-exec and the beginning of the string should not be
3622 interpreted as the beginning of the line.
3624 **** Constant: regexp/noteol
3625 The match-end-of-line operator always fails to match (but see the
3626 compilation flag regexp/newline above)
3628 **** Function: regexp? OBJ
3629 Return `#t' if OBJ is a compiled regular expression, or `#f'
3632 Regular expressions are commonly used to find patterns in one string
3633 and replace them with the contents of another string.
3635 **** Function: regexp-substitute PORT MATCH [ITEM...]
3636 Write to the output port PORT selected contents of the match
3637 structure MATCH. Each ITEM specifies what should be written, and
3638 may be one of the following arguments:
3640 * A string. String arguments are written out verbatim.
3642 * An integer. The submatch with that number is written.
3644 * The symbol `pre'. The portion of the matched string preceding
3645 the regexp match is written.
3647 * The symbol `post'. The portion of the matched string
3648 following the regexp match is written.
3650 PORT may be `#f', in which case nothing is written; instead,
3651 `regexp-substitute' constructs a string from the specified ITEMs
3654 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
3655 Similar to `regexp-substitute', but can be used to perform global
3656 substitutions on STR. Instead of taking a match structure as an
3657 argument, `regexp-substitute/global' takes two string arguments: a
3658 REGEXP string describing a regular expression, and a TARGET string
3659 which should be matched against this regular expression.
3661 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
3664 * A function may be supplied. When this function is called, it
3665 will be passed one argument: a match structure for a given
3666 regular expression match. It should return a string to be
3667 written out to PORT.
3669 * The `post' symbol causes `regexp-substitute/global' to recurse
3670 on the unmatched portion of STR. This *must* be supplied in
3671 order to perform global search-and-replace on STR; if it is
3672 not present among the ITEMs, then `regexp-substitute/global'
3673 will return after processing a single match.
3675 *** Match Structures
3677 A "match structure" is the object returned by `string-match' and
3678 `regexp-exec'. It describes which portion of a string, if any, matched
3679 the given regular expression. Match structures include: a reference to
3680 the string that was checked for matches; the starting and ending
3681 positions of the regexp match; and, if the regexp included any
3682 parenthesized subexpressions, the starting and ending positions of each
3685 In each of the regexp match functions described below, the `match'
3686 argument must be a match structure returned by a previous call to
3687 `string-match' or `regexp-exec'. Most of these functions return some
3688 information about the original target string that was matched against a
3689 regular expression; we will call that string TARGET for easy reference.
3691 **** Function: regexp-match? OBJ
3692 Return `#t' if OBJ is a match structure returned by a previous
3693 call to `regexp-exec', or `#f' otherwise.
3695 **** Function: match:substring MATCH [N]
3696 Return the portion of TARGET matched by subexpression number N.
3697 Submatch 0 (the default) represents the entire regexp match. If
3698 the regular expression as a whole matched, but the subexpression
3699 number N did not match, return `#f'.
3701 **** Function: match:start MATCH [N]
3702 Return the starting position of submatch number N.
3704 **** Function: match:end MATCH [N]
3705 Return the ending position of submatch number N.
3707 **** Function: match:prefix MATCH
3708 Return the unmatched portion of TARGET preceding the regexp match.
3710 **** Function: match:suffix MATCH
3711 Return the unmatched portion of TARGET following the regexp match.
3713 **** Function: match:count MATCH
3714 Return the number of parenthesized subexpressions from MATCH.
3715 Note that the entire regular expression match itself counts as a
3716 subexpression, and failed submatches are included in the count.
3718 **** Function: match:string MATCH
3719 Return the original TARGET string.
3721 *** Backslash Escapes
3723 Sometimes you will want a regexp to match characters like `*' or `$'
3724 exactly. For example, to check whether a particular string represents
3725 a menu entry from an Info node, it would be useful to match it against
3726 a regexp like `^* [^:]*::'. However, this won't work; because the
3727 asterisk is a metacharacter, it won't match the `*' at the beginning of
3728 the string. In this case, we want to make the first asterisk un-magic.
3730 You can do this by preceding the metacharacter with a backslash
3731 character `\'. (This is also called "quoting" the metacharacter, and
3732 is known as a "backslash escape".) When Guile sees a backslash in a
3733 regular expression, it considers the following glyph to be an ordinary
3734 character, no matter what special meaning it would ordinarily have.
3735 Therefore, we can make the above example work by changing the regexp to
3736 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
3737 to match only a single asterisk in the target string.
3739 Since the backslash is itself a metacharacter, you may force a
3740 regexp to match a backslash in the target string by preceding the
3741 backslash with itself. For example, to find variable references in a
3742 TeX program, you might want to find occurrences of the string `\let\'
3743 followed by any number of alphabetic characters. The regular expression
3744 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
3745 each match a single backslash in the target string.
3747 **** Function: regexp-quote STR
3748 Quote each special character found in STR with a backslash, and
3749 return the resulting string.
3751 *Very important:* Using backslash escapes in Guile source code (as
3752 in Emacs Lisp or C) can be tricky, because the backslash character has
3753 special meaning for the Guile reader. For example, if Guile encounters
3754 the character sequence `\n' in the middle of a string while processing
3755 Scheme code, it replaces those characters with a newline character.
3756 Similarly, the character sequence `\t' is replaced by a horizontal tab.
3757 Several of these "escape sequences" are processed by the Guile reader
3758 before your code is executed. Unrecognized escape sequences are
3759 ignored: if the characters `\*' appear in a string, they will be
3760 translated to the single character `*'.
3762 This translation is obviously undesirable for regular expressions,
3763 since we want to be able to include backslashes in a string in order to
3764 escape regexp metacharacters. Therefore, to make sure that a backslash
3765 is preserved in a string in your Guile program, you must use *two*
3766 consecutive backslashes:
3768 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
3770 The string in this example is preprocessed by the Guile reader before
3771 any code is executed. The resulting argument to `make-regexp' is the
3772 string `^\* [^:]*', which is what we really want.
3774 This also means that in order to write a regular expression that
3775 matches a single backslash character, the regular expression string in
3776 the source code must include *four* backslashes. Each consecutive pair
3777 of backslashes gets translated by the Guile reader to a single
3778 backslash, and the resulting double-backslash is interpreted by the
3779 regexp engine as matching a single backslash character. Hence:
3781 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
3783 The reason for the unwieldiness of this syntax is historical. Both
3784 regular expression pattern matchers and Unix string processing systems
3785 have traditionally used backslashes with the special meanings described
3786 above. The POSIX regular expression specification and ANSI C standard
3787 both require these semantics. Attempting to abandon either convention
3788 would cause other kinds of compatibility problems, possibly more severe
3789 ones. Therefore, without extending the Scheme reader to support
3790 strings with different quoting conventions (an ungainly and confusing
3791 extension when implemented in other languages), we must adhere to this
3792 cumbersome escape syntax.
3794 * Changes to the gh_ interface
3796 * Changes to the scm_ interface
3798 * Changes to system call interfaces:
3800 ** The value returned by `raise' is now unspecified. It throws an exception
3803 *** A new procedure `sigaction' can be used to install signal handlers
3805 (sigaction signum [action] [flags])
3807 signum is the signal number, which can be specified using the value
3810 If action is omitted, sigaction returns a pair: the CAR is the current
3811 signal hander, which will be either an integer with the value SIG_DFL
3812 (default action) or SIG_IGN (ignore), or the Scheme procedure which
3813 handles the signal, or #f if a non-Scheme procedure handles the
3814 signal. The CDR contains the current sigaction flags for the handler.
3816 If action is provided, it is installed as the new handler for signum.
3817 action can be a Scheme procedure taking one argument, or the value of
3818 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
3819 whatever signal handler was installed before sigaction was first used.
3820 Flags can optionally be specified for the new handler (SA_RESTART is
3821 always used if the system provides it, so need not be specified.) The
3822 return value is a pair with information about the old handler as
3825 This interface does not provide access to the "signal blocking"
3826 facility. Maybe this is not needed, since the thread support may
3827 provide solutions to the problem of consistent access to data
3830 *** A new procedure `flush-all-ports' is equivalent to running
3831 `force-output' on every port open for output.
3833 ** Guile now provides information on how it was built, via the new
3834 global variable, %guile-build-info. This variable records the values
3835 of the standard GNU makefile directory variables as an assocation
3836 list, mapping variable names (symbols) onto directory paths (strings).
3837 For example, to find out where the Guile link libraries were
3838 installed, you can say:
3840 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
3843 * Changes to the scm_ interface
3845 ** The new function scm_handle_by_message_noexit is just like the
3846 existing scm_handle_by_message function, except that it doesn't call
3847 exit to terminate the process. Instead, it prints a message and just
3848 returns #f. This might be a more appropriate catch-all handler for
3849 new dynamic roots and threads.
3852 Changes in Guile 1.1 (released Friday, May 16 1997):
3854 * Changes to the distribution.
3856 The Guile 1.0 distribution has been split up into several smaller
3858 guile-core --- the Guile interpreter itself.
3859 guile-tcltk --- the interface between the Guile interpreter and
3860 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
3861 is a toolkit for building graphical user interfaces.
3862 guile-rgx-ctax --- the interface between Guile and the Rx regular
3863 expression matcher, and the translator for the Ctax
3864 programming language. These are packaged together because the
3865 Ctax translator uses Rx to parse Ctax source code.
3867 This NEWS file describes the changes made to guile-core since the 1.0
3870 We no longer distribute the documentation, since it was either out of
3871 date, or incomplete. As soon as we have current documentation, we
3876 * Changes to the stand-alone interpreter
3878 ** guile now accepts command-line arguments compatible with SCSH, Olin
3879 Shivers' Scheme Shell.
3881 In general, arguments are evaluated from left to right, but there are
3882 exceptions. The following switches stop argument processing, and
3883 stash all remaining command-line arguments as the value returned by
3884 the (command-line) function.
3885 -s SCRIPT load Scheme source code from FILE, and exit
3886 -c EXPR evalute Scheme expression EXPR, and exit
3887 -- stop scanning arguments; run interactively
3889 The switches below are processed as they are encountered.
3890 -l FILE load Scheme source code from FILE
3891 -e FUNCTION after reading script, apply FUNCTION to
3892 command line arguments
3893 -ds do -s script at this point
3894 --emacs enable Emacs protocol (experimental)
3895 -h, --help display this help and exit
3896 -v, --version display version information and exit
3897 \ read arguments from following script lines
3899 So, for example, here is a Guile script named `ekko' (thanks, Olin)
3900 which re-implements the traditional "echo" command:
3902 #!/usr/local/bin/guile -s
3905 (map (lambda (arg) (display arg) (display " "))
3909 (main (command-line))
3911 Suppose we invoke this script as follows:
3913 ekko a speckled gecko
3915 Through the magic of Unix script processing (triggered by the `#!'
3916 token at the top of the file), /usr/local/bin/guile receives the
3917 following list of command-line arguments:
3919 ("-s" "./ekko" "a" "speckled" "gecko")
3921 Unix inserts the name of the script after the argument specified on
3922 the first line of the file (in this case, "-s"), and then follows that
3923 with the arguments given to the script. Guile loads the script, which
3924 defines the `main' function, and then applies it to the list of
3925 remaining command-line arguments, ("a" "speckled" "gecko").
3927 In Unix, the first line of a script file must take the following form:
3929 #!INTERPRETER ARGUMENT
3931 where INTERPRETER is the absolute filename of the interpreter
3932 executable, and ARGUMENT is a single command-line argument to pass to
3935 You may only pass one argument to the interpreter, and its length is
3936 limited. These restrictions can be annoying to work around, so Guile
3937 provides a general mechanism (borrowed from, and compatible with,
3938 SCSH) for circumventing them.
3940 If the ARGUMENT in a Guile script is a single backslash character,
3941 `\', Guile will open the script file, parse arguments from its second
3942 and subsequent lines, and replace the `\' with them. So, for example,
3943 here is another implementation of the `ekko' script:
3945 #!/usr/local/bin/guile \
3949 (for-each (lambda (arg) (display arg) (display " "))
3953 If the user invokes this script as follows:
3955 ekko a speckled gecko
3957 Unix expands this into
3959 /usr/local/bin/guile \ ekko a speckled gecko
3961 When Guile sees the `\' argument, it replaces it with the arguments
3962 read from the second line of the script, producing:
3964 /usr/local/bin/guile -e main -s ekko a speckled gecko
3966 This tells Guile to load the `ekko' script, and apply the function
3967 `main' to the argument list ("a" "speckled" "gecko").
3969 Here is how Guile parses the command-line arguments:
3970 - Each space character terminates an argument. This means that two
3971 spaces in a row introduce an empty-string argument.
3972 - The tab character is not permitted (unless you quote it with the
3973 backslash character, as described below), to avoid confusion.
3974 - The newline character terminates the sequence of arguments, and will
3975 also terminate a final non-empty argument. (However, a newline
3976 following a space will not introduce a final empty-string argument;
3977 it only terminates the argument list.)
3978 - The backslash character is the escape character. It escapes
3979 backslash, space, tab, and newline. The ANSI C escape sequences
3980 like \n and \t are also supported. These produce argument
3981 constituents; the two-character combination \n doesn't act like a
3982 terminating newline. The escape sequence \NNN for exactly three
3983 octal digits reads as the character whose ASCII code is NNN. As
3984 above, characters produced this way are argument constituents.
3985 Backslash followed by other characters is not allowed.
3987 * Changes to the procedure for linking libguile with your programs
3989 ** Guile now builds and installs a shared guile library, if your
3990 system support shared libraries. (It still builds a static library on
3991 all systems.) Guile automatically detects whether your system
3992 supports shared libraries. To prevent Guile from buildisg shared
3993 libraries, pass the `--disable-shared' flag to the configure script.
3995 Guile takes longer to compile when it builds shared libraries, because
3996 it must compile every file twice --- once to produce position-
3997 independent object code, and once to produce normal object code.
3999 ** The libthreads library has been merged into libguile.
4001 To link a program against Guile, you now need only link against
4002 -lguile and -lqt; -lthreads is no longer needed. If you are using
4003 autoconf to generate configuration scripts for your application, the
4004 following lines should suffice to add the appropriate libraries to
4007 ### Find quickthreads and libguile.
4008 AC_CHECK_LIB(qt, main)
4009 AC_CHECK_LIB(guile, scm_shell)
4011 * Changes to Scheme functions
4013 ** Guile Scheme's special syntax for keyword objects is now optional,
4014 and disabled by default.
4016 The syntax variation from R4RS made it difficult to port some
4017 interesting packages to Guile. The routines which accepted keyword
4018 arguments (mostly in the module system) have been modified to also
4019 accept symbols whose names begin with `:'.
4021 To change the keyword syntax, you must first import the (ice-9 debug)
4023 (use-modules (ice-9 debug))
4025 Then you can enable the keyword syntax as follows:
4026 (read-set! keywords 'prefix)
4028 To disable keyword syntax, do this:
4029 (read-set! keywords #f)
4031 ** Many more primitive functions accept shared substrings as
4032 arguments. In the past, these functions required normal, mutable
4033 strings as arguments, although they never made use of this
4036 ** The uniform array functions now operate on byte vectors. These
4037 functions are `array-fill!', `serial-array-copy!', `array-copy!',
4038 `serial-array-map', `array-map', `array-for-each', and
4041 ** The new functions `trace' and `untrace' implement simple debugging
4042 support for Scheme functions.
4044 The `trace' function accepts any number of procedures as arguments,
4045 and tells the Guile interpreter to display each procedure's name and
4046 arguments each time the procedure is invoked. When invoked with no
4047 arguments, `trace' returns the list of procedures currently being
4050 The `untrace' function accepts any number of procedures as arguments,
4051 and tells the Guile interpreter not to trace them any more. When
4052 invoked with no arguments, `untrace' untraces all curretly traced
4055 The tracing in Guile has an advantage over most other systems: we
4056 don't create new procedure objects, but mark the procedure objects
4057 themselves. This means that anonymous and internal procedures can be
4060 ** The function `assert-repl-prompt' has been renamed to
4061 `set-repl-prompt!'. It takes one argument, PROMPT.
4062 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
4063 - If PROMPT is a string, we use it as a prompt.
4064 - If PROMPT is a procedure accepting no arguments, we call it, and
4065 display the result as a prompt.
4066 - Otherwise, we display "> ".
4068 ** The new function `eval-string' reads Scheme expressions from a
4069 string and evaluates them, returning the value of the last expression
4070 in the string. If the string contains no expressions, it returns an
4073 ** The new function `thunk?' returns true iff its argument is a
4074 procedure of zero arguments.
4076 ** `defined?' is now a builtin function, instead of syntax. This
4077 means that its argument should be quoted. It returns #t iff its
4078 argument is bound in the current module.
4080 ** The new syntax `use-modules' allows you to add new modules to your
4081 environment without re-typing a complete `define-module' form. It
4082 accepts any number of module names as arguments, and imports their
4083 public bindings into the current module.
4085 ** The new function (module-defined? NAME MODULE) returns true iff
4086 NAME, a symbol, is defined in MODULE, a module object.
4088 ** The new function `builtin-bindings' creates and returns a hash
4089 table containing copies of all the root module's bindings.
4091 ** The new function `builtin-weak-bindings' does the same as
4092 `builtin-bindings', but creates a doubly-weak hash table.
4094 ** The `equal?' function now considers variable objects to be
4095 equivalent if they have the same name and the same value.
4097 ** The new function `command-line' returns the command-line arguments
4098 given to Guile, as a list of strings.
4100 When using guile as a script interpreter, `command-line' returns the
4101 script's arguments; those processed by the interpreter (like `-s' or
4102 `-c') are omitted. (In other words, you get the normal, expected
4103 behavior.) Any application that uses scm_shell to process its
4104 command-line arguments gets this behavior as well.
4106 ** The new function `load-user-init' looks for a file called `.guile'
4107 in the user's home directory, and loads it if it exists. This is
4108 mostly for use by the code generated by scm_compile_shell_switches,
4109 but we thought it might also be useful in other circumstances.
4111 ** The new function `log10' returns the base-10 logarithm of its
4114 ** Changes to I/O functions
4116 *** The functions `read', `primitive-load', `read-and-eval!', and
4117 `primitive-load-path' no longer take optional arguments controlling
4118 case insensitivity and a `#' parser.
4120 Case sensitivity is now controlled by a read option called
4121 `case-insensitive'. The user can add new `#' syntaxes with the
4122 `read-hash-extend' function (see below).
4124 *** The new function `read-hash-extend' allows the user to change the
4125 syntax of Guile Scheme in a somewhat controlled way.
4127 (read-hash-extend CHAR PROC)
4128 When parsing S-expressions, if we read a `#' character followed by
4129 the character CHAR, use PROC to parse an object from the stream.
4130 If PROC is #f, remove any parsing procedure registered for CHAR.
4132 The reader applies PROC to two arguments: CHAR and an input port.
4134 *** The new functions read-delimited and read-delimited! provide a
4135 general mechanism for doing delimited input on streams.
4137 (read-delimited DELIMS [PORT HANDLE-DELIM])
4138 Read until we encounter one of the characters in DELIMS (a string),
4139 or end-of-file. PORT is the input port to read from; it defaults to
4140 the current input port. The HANDLE-DELIM parameter determines how
4141 the terminating character is handled; it should be one of the
4144 'trim omit delimiter from result
4145 'peek leave delimiter character in input stream
4146 'concat append delimiter character to returned value
4147 'split return a pair: (RESULT . TERMINATOR)
4149 HANDLE-DELIM defaults to 'peek.
4151 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
4152 A side-effecting variant of `read-delimited'.
4154 The data is written into the string BUF at the indices in the
4155 half-open interval [START, END); the default interval is the whole
4156 string: START = 0 and END = (string-length BUF). The values of
4157 START and END must specify a well-defined interval in BUF, i.e.
4158 0 <= START <= END <= (string-length BUF).
4160 It returns NBYTES, the number of bytes read. If the buffer filled
4161 up without a delimiter character being found, it returns #f. If the
4162 port is at EOF when the read starts, it returns the EOF object.
4164 If an integer is returned (i.e., the read is successfully terminated
4165 by reading a delimiter character), then the HANDLE-DELIM parameter
4166 determines how to handle the terminating character. It is described
4167 above, and defaults to 'peek.
4169 (The descriptions of these functions were borrowed from the SCSH
4170 manual, by Olin Shivers and Brian Carlstrom.)
4172 *** The `%read-delimited!' function is the primitive used to implement
4173 `read-delimited' and `read-delimited!'.
4175 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
4177 This returns a pair of values: (TERMINATOR . NUM-READ).
4178 - TERMINATOR describes why the read was terminated. If it is a
4179 character or the eof object, then that is the value that terminated
4180 the read. If it is #f, the function filled the buffer without finding
4181 a delimiting character.
4182 - NUM-READ is the number of characters read into BUF.
4184 If the read is successfully terminated by reading a delimiter
4185 character, then the gobble? parameter determines what to do with the
4186 terminating character. If true, the character is removed from the
4187 input stream; if false, the character is left in the input stream
4188 where a subsequent read operation will retrieve it. In either case,
4189 the character is also the first value returned by the procedure call.
4191 (The descriptions of this function was borrowed from the SCSH manual,
4192 by Olin Shivers and Brian Carlstrom.)
4194 *** The `read-line' and `read-line!' functions have changed; they now
4195 trim the terminator by default; previously they appended it to the
4196 returned string. For the old behavior, use (read-line PORT 'concat).
4198 *** The functions `uniform-array-read!' and `uniform-array-write!' now
4199 take new optional START and END arguments, specifying the region of
4200 the array to read and write.
4202 *** The `ungetc-char-ready?' function has been removed. We feel it's
4203 inappropriate for an interface to expose implementation details this
4206 ** Changes to the Unix library and system call interface
4208 *** The new fcntl function provides access to the Unix `fcntl' system
4211 (fcntl PORT COMMAND VALUE)
4212 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
4213 Values for COMMAND are:
4215 F_DUPFD duplicate a file descriptor
4216 F_GETFD read the descriptor's close-on-exec flag
4217 F_SETFD set the descriptor's close-on-exec flag to VALUE
4218 F_GETFL read the descriptor's flags, as set on open
4219 F_SETFL set the descriptor's flags, as set on open to VALUE
4220 F_GETOWN return the process ID of a socket's owner, for SIGIO
4221 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
4222 FD_CLOEXEC not sure what this is
4224 For details, see the documentation for the fcntl system call.
4226 *** The arguments to `select' have changed, for compatibility with
4227 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
4228 expected behavior. The MILLISECONDS parameter has been changed to
4229 MICROSECONDS, to more closely resemble the underlying system call.
4230 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
4231 corresponding return set will be the same.
4233 *** The arguments to the `mknod' system call have changed. They are
4236 (mknod PATH TYPE PERMS DEV)
4237 Create a new file (`node') in the file system. PATH is the name of
4238 the file to create. TYPE is the kind of file to create; it should
4239 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
4240 permission bits to give the newly created file. If TYPE is
4241 'block-special or 'char-special, DEV specifies which device the
4242 special file refers to; its interpretation depends on the kind of
4243 special file being created.
4245 *** The `fork' function has been renamed to `primitive-fork', to avoid
4246 clashing with various SCSH forks.
4248 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
4249 and `recvfrom!'. They no longer accept a size for a second argument;
4250 you must pass a string to hold the received value. They no longer
4251 return the buffer. Instead, `recv' returns the length of the message
4252 received, and `recvfrom' returns a pair containing the packet's length
4253 and originating address.
4255 *** The file descriptor datatype has been removed, as have the
4256 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
4257 We plan to replace these functions with a SCSH-compatible interface.
4259 *** The `create' function has been removed; it's just a special case
4262 *** There are new functions to break down process termination status
4263 values. In the descriptions below, STATUS is a value returned by
4266 (status:exit-val STATUS)
4267 If the child process exited normally, this function returns the exit
4268 code for the child process (i.e., the value passed to exit, or
4269 returned from main). If the child process did not exit normally,
4270 this function returns #f.
4272 (status:stop-sig STATUS)
4273 If the child process was suspended by a signal, this function
4274 returns the signal that suspended the child. Otherwise, it returns
4277 (status:term-sig STATUS)
4278 If the child process terminated abnormally, this function returns
4279 the signal that terminated the child. Otherwise, this function
4282 POSIX promises that exactly one of these functions will return true on
4283 a valid STATUS value.
4285 These functions are compatible with SCSH.
4287 *** There are new accessors and setters for the broken-out time vectors
4288 returned by `localtime', `gmtime', and that ilk. They are:
4290 Component Accessor Setter
4291 ========================= ============ ============
4292 seconds tm:sec set-tm:sec
4293 minutes tm:min set-tm:min
4294 hours tm:hour set-tm:hour
4295 day of the month tm:mday set-tm:mday
4296 month tm:mon set-tm:mon
4297 year tm:year set-tm:year
4298 day of the week tm:wday set-tm:wday
4299 day in the year tm:yday set-tm:yday
4300 daylight saving time tm:isdst set-tm:isdst
4301 GMT offset, seconds tm:gmtoff set-tm:gmtoff
4302 name of time zone tm:zone set-tm:zone
4304 *** There are new accessors for the vectors returned by `uname',
4305 describing the host system:
4308 ============================================== ================
4309 name of the operating system implementation utsname:sysname
4310 network name of this machine utsname:nodename
4311 release level of the operating system utsname:release
4312 version level of the operating system utsname:version
4313 machine hardware platform utsname:machine
4315 *** There are new accessors for the vectors returned by `getpw',
4316 `getpwnam', `getpwuid', and `getpwent', describing entries from the
4317 system's user database:
4320 ====================== =================
4321 user name passwd:name
4322 user password passwd:passwd
4325 real name passwd:gecos
4326 home directory passwd:dir
4327 shell program passwd:shell
4329 *** There are new accessors for the vectors returned by `getgr',
4330 `getgrnam', `getgrgid', and `getgrent', describing entries from the
4331 system's group database:
4334 ======================= ============
4335 group name group:name
4336 group password group:passwd
4338 group members group:mem
4340 *** There are new accessors for the vectors returned by `gethost',
4341 `gethostbyaddr', `gethostbyname', and `gethostent', describing
4345 ========================= ===============
4346 official name of host hostent:name
4347 alias list hostent:aliases
4348 host address type hostent:addrtype
4349 length of address hostent:length
4350 list of addresses hostent:addr-list
4352 *** There are new accessors for the vectors returned by `getnet',
4353 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
4357 ========================= ===============
4358 official name of net netent:name
4359 alias list netent:aliases
4360 net number type netent:addrtype
4361 net number netent:net
4363 *** There are new accessors for the vectors returned by `getproto',
4364 `getprotobyname', `getprotobynumber', and `getprotoent', describing
4368 ========================= ===============
4369 official protocol name protoent:name
4370 alias list protoent:aliases
4371 protocol number protoent:proto
4373 *** There are new accessors for the vectors returned by `getserv',
4374 `getservbyname', `getservbyport', and `getservent', describing
4378 ========================= ===============
4379 official service name servent:name
4380 alias list servent:aliases
4381 port number servent:port
4382 protocol to use servent:proto
4384 *** There are new accessors for the sockaddr structures returned by
4385 `accept', `getsockname', `getpeername', `recvfrom!':
4388 ======================================== ===============
4389 address format (`family') sockaddr:fam
4390 path, for file domain addresses sockaddr:path
4391 address, for internet domain addresses sockaddr:addr
4392 TCP or UDP port, for internet sockaddr:port
4394 *** The `getpwent', `getgrent', `gethostent', `getnetent',
4395 `getprotoent', and `getservent' functions now return #f at the end of
4396 the user database. (They used to throw an exception.)
4398 Note that calling MUMBLEent function is equivalent to calling the
4399 corresponding MUMBLE function with no arguments.
4401 *** The `setpwent', `setgrent', `sethostent', `setnetent',
4402 `setprotoent', and `setservent' routines now take no arguments.
4404 *** The `gethost', `getproto', `getnet', and `getserv' functions now
4405 provide more useful information when they throw an exception.
4407 *** The `lnaof' function has been renamed to `inet-lnaof'.
4409 *** Guile now claims to have the `current-time' feature.
4411 *** The `mktime' function now takes an optional second argument ZONE,
4412 giving the time zone to use for the conversion. ZONE should be a
4413 string, in the same format as expected for the "TZ" environment variable.
4415 *** The `strptime' function now returns a pair (TIME . COUNT), where
4416 TIME is the parsed time as a vector, and COUNT is the number of
4417 characters from the string left unparsed. This function used to
4418 return the remaining characters as a string.
4420 *** The `gettimeofday' function has replaced the old `time+ticks' function.
4421 The return value is now (SECONDS . MICROSECONDS); the fractional
4422 component is no longer expressed in "ticks".
4424 *** The `ticks/sec' constant has been removed, in light of the above change.
4426 * Changes to the gh_ interface
4428 ** gh_eval_str() now returns an SCM object which is the result of the
4431 ** gh_scm2str() now copies the Scheme data to a caller-provided C
4434 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
4435 and returns the array
4437 ** gh_scm2str0() is gone: there is no need to distinguish
4438 null-terminated from non-null-terminated, since gh_scm2newstr() allows
4439 the user to interpret the data both ways.
4441 * Changes to the scm_ interface
4443 ** The new function scm_symbol_value0 provides an easy way to get a
4444 symbol's value from C code:
4446 SCM scm_symbol_value0 (char *NAME)
4447 Return the value of the symbol named by the null-terminated string
4448 NAME in the current module. If the symbol named NAME is unbound in
4449 the current module, return SCM_UNDEFINED.
4451 ** The new function scm_sysintern0 creates new top-level variables,
4452 without assigning them a value.
4454 SCM scm_sysintern0 (char *NAME)
4455 Create a new Scheme top-level variable named NAME. NAME is a
4456 null-terminated string. Return the variable's value cell.
4458 ** The function scm_internal_catch is the guts of catch. It handles
4459 all the mechanics of setting up a catch target, invoking the catch
4460 body, and perhaps invoking the handler if the body does a throw.
4462 The function is designed to be usable from C code, but is general
4463 enough to implement all the semantics Guile Scheme expects from throw.
4465 TAG is the catch tag. Typically, this is a symbol, but this function
4466 doesn't actually care about that.
4468 BODY is a pointer to a C function which runs the body of the catch;
4469 this is the code you can throw from. We call it like this:
4470 BODY (BODY_DATA, JMPBUF)
4472 BODY_DATA is just the BODY_DATA argument we received; we pass it
4473 through to BODY as its first argument. The caller can make
4474 BODY_DATA point to anything useful that BODY might need.
4475 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
4476 which we have just created and initialized.
4478 HANDLER is a pointer to a C function to deal with a throw to TAG,
4479 should one occur. We call it like this:
4480 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
4482 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
4483 same idea as BODY_DATA above.
4484 THROWN_TAG is the tag that the user threw to; usually this is
4485 TAG, but it could be something else if TAG was #t (i.e., a
4486 catch-all), or the user threw to a jmpbuf.
4487 THROW_ARGS is the list of arguments the user passed to the THROW
4490 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
4491 is just a pointer we pass through to HANDLER. We don't actually
4492 use either of those pointers otherwise ourselves. The idea is
4493 that, if our caller wants to communicate something to BODY or
4494 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
4495 HANDLER can then use. Think of it as a way to make BODY and
4496 HANDLER closures, not just functions; MUMBLE_DATA points to the
4499 Of course, it's up to the caller to make sure that any data a
4500 MUMBLE_DATA needs is protected from GC. A common way to do this is
4501 to make MUMBLE_DATA a pointer to data stored in an automatic
4502 structure variable; since the collector must scan the stack for
4503 references anyway, this assures that any references in MUMBLE_DATA
4506 ** The new function scm_internal_lazy_catch is exactly like
4507 scm_internal_catch, except:
4509 - It does not unwind the stack (this is the major difference).
4510 - If handler returns, its value is returned from the throw.
4511 - BODY always receives #f as its JMPBUF argument (since there's no
4512 jmpbuf associated with a lazy catch, because we don't unwind the
4515 ** scm_body_thunk is a new body function you can pass to
4516 scm_internal_catch if you want the body to be like Scheme's `catch'
4517 --- a thunk, or a function of one argument if the tag is #f.
4519 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
4520 contains the Scheme procedure to invoke as the body, and the tag
4521 we're catching. If the tag is #f, then we pass JMPBUF (created by
4522 scm_internal_catch) to the body procedure; otherwise, the body gets
4525 ** scm_handle_by_proc is a new handler function you can pass to
4526 scm_internal_catch if you want the handler to act like Scheme's catch
4527 --- call a procedure with the tag and the throw arguments.
4529 If the user does a throw to this catch, this function runs a handler
4530 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
4531 variable holding the Scheme procedure object to invoke. It ought to
4532 be a pointer to an automatic variable (i.e., one living on the stack),
4533 or the procedure object should be otherwise protected from GC.
4535 ** scm_handle_by_message is a new handler function to use with
4536 `scm_internal_catch' if you want Guile to print a message and die.
4537 It's useful for dealing with throws to uncaught keys at the top level.
4539 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
4540 message header to print; if zero, we use "guile" instead. That
4541 text is followed by a colon, then the message described by ARGS.
4543 ** The return type of scm_boot_guile is now void; the function does
4544 not return a value, and indeed, never returns at all.
4546 ** The new function scm_shell makes it easy for user applications to
4547 process command-line arguments in a way that is compatible with the
4548 stand-alone guile interpreter (which is in turn compatible with SCSH,
4551 To use the scm_shell function, first initialize any guile modules
4552 linked into your application, and then call scm_shell with the values
4553 of ARGC and ARGV your `main' function received. scm_shell will add
4554 any SCSH-style meta-arguments from the top of the script file to the
4555 argument vector, and then process the command-line arguments. This
4556 generally means loading a script file or starting up an interactive
4557 command interpreter. For details, see "Changes to the stand-alone
4560 ** The new functions scm_get_meta_args and scm_count_argv help you
4561 implement the SCSH-style meta-argument, `\'.
4563 char **scm_get_meta_args (int ARGC, char **ARGV)
4564 If the second element of ARGV is a string consisting of a single
4565 backslash character (i.e. "\\" in Scheme notation), open the file
4566 named by the following argument, parse arguments from it, and return
4567 the spliced command line. The returned array is terminated by a
4570 For details of argument parsing, see above, under "guile now accepts
4571 command-line arguments compatible with SCSH..."
4573 int scm_count_argv (char **ARGV)
4574 Count the arguments in ARGV, assuming it is terminated by a null
4577 For an example of how these functions might be used, see the source
4578 code for the function scm_shell in libguile/script.c.
4580 You will usually want to use scm_shell instead of calling this
4583 ** The new function scm_compile_shell_switches turns an array of
4584 command-line arguments into Scheme code to carry out the actions they
4585 describe. Given ARGC and ARGV, it returns a Scheme expression to
4586 evaluate, and calls scm_set_program_arguments to make any remaining
4587 command-line arguments available to the Scheme code. For example,
4588 given the following arguments:
4590 -e main -s ekko a speckled gecko
4592 scm_set_program_arguments will return the following expression:
4594 (begin (load "ekko") (main (command-line)) (quit))
4596 You will usually want to use scm_shell instead of calling this
4599 ** The function scm_shell_usage prints a usage message appropriate for
4600 an interpreter that uses scm_compile_shell_switches to handle its
4601 command-line arguments.
4603 void scm_shell_usage (int FATAL, char *MESSAGE)
4604 Print a usage message to the standard error output. If MESSAGE is
4605 non-zero, write it before the usage message, followed by a newline.
4606 If FATAL is non-zero, exit the process, using FATAL as the
4607 termination status. (If you want to be compatible with Guile,
4608 always use 1 as the exit status when terminating due to command-line
4611 You will usually want to use scm_shell instead of calling this
4614 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
4615 expressions. It used to return SCM_EOL. Earth-shattering.
4617 ** The macros for declaring scheme objects in C code have been
4618 rearranged slightly. They are now:
4620 SCM_SYMBOL (C_NAME, SCHEME_NAME)
4621 Declare a static SCM variable named C_NAME, and initialize it to
4622 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
4623 be a C identifier, and SCHEME_NAME should be a C string.
4625 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
4626 Just like SCM_SYMBOL, but make C_NAME globally visible.
4628 SCM_VCELL (C_NAME, SCHEME_NAME)
4629 Create a global variable at the Scheme level named SCHEME_NAME.
4630 Declare a static SCM variable named C_NAME, and initialize it to
4631 point to the Scheme variable's value cell.
4633 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
4634 Just like SCM_VCELL, but make C_NAME globally visible.
4636 The `guile-snarf' script writes initialization code for these macros
4637 to its standard output, given C source code as input.
4639 The SCM_GLOBAL macro is gone.
4641 ** The scm_read_line and scm_read_line_x functions have been replaced
4642 by Scheme code based on the %read-delimited! procedure (known to C
4643 code as scm_read_delimited_x). See its description above for more
4646 ** The function scm_sys_open has been renamed to scm_open. It now
4647 returns a port instead of an FD object.
4649 * The dynamic linking support has changed. For more information, see
4650 libguile/DYNAMIC-LINKING.
4655 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
4658 * Changes to the 'guile' program:
4660 ** Guile now loads some new files when it starts up. Guile first
4661 searches the load path for init.scm, and loads it if found. Then, if
4662 Guile is not being used to execute a script, and the user's home
4663 directory contains a file named `.guile', Guile loads that.
4665 ** You can now use Guile as a shell script interpreter.
4667 To paraphrase the SCSH manual:
4669 When Unix tries to execute an executable file whose first two
4670 characters are the `#!', it treats the file not as machine code to
4671 be directly executed by the native processor, but as source code
4672 to be executed by some interpreter. The interpreter to use is
4673 specified immediately after the #! sequence on the first line of
4674 the source file. The kernel reads in the name of the interpreter,
4675 and executes that instead. It passes the interpreter the source
4676 filename as its first argument, with the original arguments
4677 following. Consult the Unix man page for the `exec' system call
4678 for more information.
4680 Now you can use Guile as an interpreter, using a mechanism which is a
4681 compatible subset of that provided by SCSH.
4683 Guile now recognizes a '-s' command line switch, whose argument is the
4684 name of a file of Scheme code to load. It also treats the two
4685 characters `#!' as the start of a comment, terminated by `!#'. Thus,
4686 to make a file of Scheme code directly executable by Unix, insert the
4687 following two lines at the top of the file:
4689 #!/usr/local/bin/guile -s
4692 Guile treats the argument of the `-s' command-line switch as the name
4693 of a file of Scheme code to load, and treats the sequence `#!' as the
4694 start of a block comment, terminated by `!#'.
4696 For example, here's a version of 'echo' written in Scheme:
4698 #!/usr/local/bin/guile -s
4700 (let loop ((args (cdr (program-arguments))))
4703 (display (car args))
4704 (if (pair? (cdr args))
4706 (loop (cdr args)))))
4709 Why does `#!' start a block comment terminated by `!#', instead of the
4710 end of the line? That is the notation SCSH uses, and although we
4711 don't yet support the other SCSH features that motivate that choice,
4712 we would like to be backward-compatible with any existing Guile
4713 scripts once we do. Furthermore, if the path to Guile on your system
4714 is too long for your kernel, you can start the script with this
4718 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
4721 Note that some very old Unix systems don't support the `#!' syntax.
4724 ** You can now run Guile without installing it.
4726 Previous versions of the interactive Guile interpreter (`guile')
4727 couldn't start up unless Guile's Scheme library had been installed;
4728 they used the value of the environment variable `SCHEME_LOAD_PATH'
4729 later on in the startup process, but not to find the startup code
4730 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
4733 To run Guile without installing it, build it in the normal way, and
4734 then set the environment variable `SCHEME_LOAD_PATH' to a
4735 colon-separated list of directories, including the top-level directory
4736 of the Guile sources. For example, if you unpacked Guile so that the
4737 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
4740 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
4743 ** Guile's read-eval-print loop no longer prints #<unspecified>
4744 results. If the user wants to see this, she can evaluate the
4745 expression (assert-repl-print-unspecified #t), perhaps in her startup
4748 ** Guile no longer shows backtraces by default when an error occurs;
4749 however, it does display a message saying how to get one, and how to
4750 request that they be displayed by default. After an error, evaluate
4752 to see a backtrace, and
4753 (debug-enable 'backtrace)
4754 to see them by default.
4758 * Changes to Guile Scheme:
4760 ** Guile now distinguishes between #f and the empty list.
4762 This is for compatibility with the IEEE standard, the (possibly)
4763 upcoming Revised^5 Report on Scheme, and many extant Scheme
4766 Guile used to have #f and '() denote the same object, to make Scheme's
4767 type system more compatible with Emacs Lisp's. However, the change
4768 caused too much trouble for Scheme programmers, and we found another
4769 way to reconcile Emacs Lisp with Scheme that didn't require this.
4772 ** Guile's delq, delv, delete functions, and their destructive
4773 counterparts, delq!, delv!, and delete!, now remove all matching
4774 elements from the list, not just the first. This matches the behavior
4775 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
4776 functions which inspired them.
4778 I recognize that this change may break code in subtle ways, but it
4779 seems best to make the change before the FSF's first Guile release,
4783 ** The compiled-library-path function has been deleted from libguile.
4785 ** The facilities for loading Scheme source files have changed.
4787 *** The variable %load-path now tells Guile which directories to search
4788 for Scheme code. Its value is a list of strings, each of which names
4791 *** The variable %load-extensions now tells Guile which extensions to
4792 try appending to a filename when searching the load path. Its value
4793 is a list of strings. Its default value is ("" ".scm").
4795 *** (%search-load-path FILENAME) searches the directories listed in the
4796 value of the %load-path variable for a Scheme file named FILENAME,
4797 with all the extensions listed in %load-extensions. If it finds a
4798 match, then it returns its full filename. If FILENAME is absolute, it
4799 returns it unchanged. Otherwise, it returns #f.
4801 %search-load-path will not return matches that refer to directories.
4803 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
4804 uses %seach-load-path to find a file named FILENAME, and loads it if
4805 it finds it. If it can't read FILENAME for any reason, it throws an
4808 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
4811 *** load uses the same searching semantics as primitive-load.
4813 *** The functions %try-load, try-load-with-path, %load, load-with-path,
4814 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
4815 path, and load-module-with-path have been deleted. The functions
4816 above should serve their purposes.
4818 *** If the value of the variable %load-hook is a procedure,
4819 `primitive-load' applies its value to the name of the file being
4820 loaded (without the load path directory name prepended). If its value
4821 is #f, it is ignored. Otherwise, an error occurs.
4823 This is mostly useful for printing load notification messages.
4826 ** The function `eval!' is no longer accessible from the scheme level.
4827 We can't allow operations which introduce glocs into the scheme level,
4828 because Guile's type system can't handle these as data. Use `eval' or
4829 `read-and-eval!' (see below) as replacement.
4831 ** The new function read-and-eval! reads an expression from PORT,
4832 evaluates it, and returns the result. This is more efficient than
4833 simply calling `read' and `eval', since it is not necessary to make a
4834 copy of the expression for the evaluator to munge.
4836 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
4837 for the `read' function.
4840 ** The function `int?' has been removed; its definition was identical
4841 to that of `integer?'.
4843 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
4844 use the R4RS names for these functions.
4846 ** The function object-properties no longer returns the hash handle;
4847 it simply returns the object's property list.
4849 ** Many functions have been changed to throw errors, instead of
4850 returning #f on failure. The point of providing exception handling in
4851 the language is to simplify the logic of user code, but this is less
4852 useful if Guile's primitives don't throw exceptions.
4854 ** The function `fileno' has been renamed from `%fileno'.
4856 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
4859 * Changes to Guile's C interface:
4861 ** The library's initialization procedure has been simplified.
4862 scm_boot_guile now has the prototype:
4864 void scm_boot_guile (int ARGC,
4866 void (*main_func) (),
4869 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
4870 MAIN_FUNC should do all the work of the program (initializing other
4871 packages, reading user input, etc.) before returning. When MAIN_FUNC
4872 returns, call exit (0); this function never returns. If you want some
4873 other exit value, MAIN_FUNC may call exit itself.
4875 scm_boot_guile arranges for program-arguments to return the strings
4876 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
4877 scm_set_program_arguments with the final list, so Scheme code will
4878 know which arguments have been processed.
4880 scm_boot_guile establishes a catch-all catch handler which prints an
4881 error message and exits the process. This means that Guile exits in a
4882 coherent way when system errors occur and the user isn't prepared to
4883 handle it. If the user doesn't like this behavior, they can establish
4884 their own universal catcher in MAIN_FUNC to shadow this one.
4886 Why must the caller do all the real work from MAIN_FUNC? The garbage
4887 collector assumes that all local variables of type SCM will be above
4888 scm_boot_guile's stack frame on the stack. If you try to manipulate
4889 SCM values after this function returns, it's the luck of the draw
4890 whether the GC will be able to find the objects you allocate. So,
4891 scm_boot_guile function exits, rather than returning, to discourage
4892 people from making that mistake.
4894 The IN, OUT, and ERR arguments were removed; there are other
4895 convenient ways to override these when desired.
4897 The RESULT argument was deleted; this function should never return.
4899 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
4903 ** Guile's header files should no longer conflict with your system's
4906 In order to compile code which #included <libguile.h>, previous
4907 versions of Guile required you to add a directory containing all the
4908 Guile header files to your #include path. This was a problem, since
4909 Guile's header files have names which conflict with many systems'
4912 Now only <libguile.h> need appear in your #include path; you must
4913 refer to all Guile's other header files as <libguile/mumble.h>.
4914 Guile's installation procedure puts libguile.h in $(includedir), and
4915 the rest in $(includedir)/libguile.
4918 ** Two new C functions, scm_protect_object and scm_unprotect_object,
4919 have been added to the Guile library.
4921 scm_protect_object (OBJ) protects OBJ from the garbage collector.
4922 OBJ will not be freed, even if all other references are dropped,
4923 until someone does scm_unprotect_object (OBJ). Both functions
4926 Note that calls to scm_protect_object do not nest. You can call
4927 scm_protect_object any number of times on a given object, and the
4928 next call to scm_unprotect_object will unprotect it completely.
4930 Basically, scm_protect_object and scm_unprotect_object just
4931 maintain a list of references to things. Since the GC knows about
4932 this list, all objects it mentions stay alive. scm_protect_object
4933 adds its argument to the list; scm_unprotect_object remove its
4934 argument from the list.
4937 ** scm_eval_0str now returns the value of the last expression
4940 ** The new function scm_read_0str reads an s-expression from a
4941 null-terminated string, and returns it.
4943 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
4944 to a Scheme port object.
4946 ** The new function `scm_set_program_arguments' allows C code to set
4947 the value returned by the Scheme `program-arguments' function.
4952 * Guile no longer includes sophisticated Tcl/Tk support.
4954 The old Tcl/Tk support was unsatisfying to us, because it required the
4955 user to link against the Tcl library, as well as Tk and Guile. The
4956 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
4957 referring to widgets by names, rather than exporting widgets to Scheme
4958 code as a special datatype.
4960 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
4961 maintainers described some very interesting changes in progress to the
4962 Tcl/Tk internals, which would facilitate clean interfaces between lone
4963 Tk and other interpreters --- even for garbage-collected languages
4964 like Scheme. They expected the new Tk to be publicly available in the
4967 Since it seems that Guile might soon have a new, cleaner interface to
4968 lone Tk, and that the old Guile/Tk glue code would probably need to be
4969 completely rewritten, we (Jim Blandy and Richard Stallman) have
4970 decided not to support the old code. We'll spend the time instead on
4971 a good interface to the newer Tk, as soon as it is available.
4973 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
4976 Copyright information:
4978 Copyright (C) 1996,1997 Free Software Foundation, Inc.
4980 Permission is granted to anyone to make or distribute verbatim copies
4981 of this document as received, in any medium, provided that the
4982 copyright notice and this permission notice are preserved,
4983 thus giving the recipient permission to redistribute in turn.
4985 Permission is granted to distribute modified versions
4986 of this document, or of portions of it,
4987 under the above conditions, provided also that they
4988 carry prominent notices stating who last changed them.
4993 paragraph-separate: "[
\f]*$"