* srfi-1.scm (iota map for-each map-in-order list-index member
[bpt/guile.git] / NEWS
1 Guile NEWS --- history of user-visible changes.
2 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
3 See the end for copying conditions.
4
5 Please send Guile bug reports to bug-guile@gnu.org.
6 \f
7 Changes since the stable branch:
8
9 * Changes to the distribution
10
11 ** Guile now provide and uses an "effective" version number.
12
13 Guile now provides scm_effective_version and effective-version
14 functions which return the "effective" version number. This is just
15 the normal full version string without the final micro-version number,
16 so the current effective-version is "1.6". The effective version
17 should remain unchanged during a stable series, and should be used for
18 items like the versioned share directory name
19 i.e. /usr/share/guile/1.6.
20
21 Providing an unchanging version number during a stable release for
22 things like the versioned share directory can be particularly
23 important for Guile "add-on" packages, since it provides a directory
24 that they can install to that won't be changed out from under them
25 with each micro release during a stable series.
26
27 ** There are two new thread implementation options: "null" and "coop-pthreads".
28
29 When you configure "--with-threads=null", you will get the usual
30 threading API (call-with-new-thread, make-mutex, etc), but you can't
31 actually create new threads. Also, "--with-threads=no" is now
32 equivalent to "--with-threads=null". This means that the thread API
33 is always present, although you might not be able to create new
34 threads.
35
36 The "coop-pthread" (or shorter: "copt") thread implementation will use
37 portable POSIX threads but will restrict them so that only one thread
38 can execute 'in Guile' at any one time. This option will give you the
39 same basic behavior as the old "coop" option, but hopefully in a more
40 portable way.
41
42 The default is now "coop-pthread", unless your platform doesn't have
43 pthreads, in which case "null" threads are used.
44
45 ** Guile now includes its own version of libltdl.
46
47 We now use a modified version of libltdl that allows us to make
48 improvements to it without having to rely on libtool releases.
49
50 * Changes to the standalone interpreter
51
52 ** New command line option `--no-debug'.
53
54 Specifying `--no-debug' on the command line will keep the debugging
55 evaluator turned off, even for interactive sessions.
56
57 ** User-init file ~/.guile is now loaded with the debugging evaluator.
58
59 Previously, the normal evaluator would have been used. Using the
60 debugging evaluator gives better error messages.
61
62 * Changes to Scheme functions and syntax
63
64 ** Checking for duplicate bindings in module system
65
66 The module system now checks for duplicate imported bindings.
67
68 The behavior can be controlled by specifying one or more duplicates
69 handlers. For example, to get back the old behavior (which was to use
70 the last imported binding of a certain name), write:
71
72 (define-module (foo)
73 :use-module (bar)
74 :use-module (baz)
75 :duplicates last)
76
77 If you want the old behavior without changing your module headers, put
78 the line:
79
80 (default-module-duplicates-handler 'last)
81
82 in your .guile init file.
83
84 The syntax for the :duplicates option is:
85
86 :duplicates HANDLER-NAME | (HANDLER1-NAME HANDLER2-NAME ...)
87
88 Specifying multiple handlers is useful since some handlers (such as
89 replace) can defer conflict resolution to others. Each handler is
90 tried until a binding is selected.
91
92 Currently available duplicates handlers are:
93
94 check report an error for bindings with a common name
95 warn issue a warning for bindings with a common name
96 replace replace bindings which have an imported replacement
97 warn-override-core issue a warning for imports which override core bindings
98 first select the first encountered binding (override)
99 last select the last encountered binding (override)
100 merge-generics merge generic functions with a common name
101 into an <extended-generic>
102
103 The default duplicates handler is:
104
105 (replace warn-override-core check)
106
107 ** New define-module option: :replace
108
109 :replace works as :export, but, in addition, marks the binding as a
110 replacement.
111
112 A typical example is `format' in (ice-9 format) which is a replacement
113 for the core binding `format'.
114
115 ** Merging generic functions
116
117 It is sometimes tempting to use GOOPS accessors with short names.
118 For example, it is tempting to use the name `x' for the x-coordinate
119 in vector packages.
120
121 Assume that we work with a graphical package which needs to use two
122 independent vector packages for 2D and 3D vectors respectively. If
123 both packages export `x' we will encounter a name collision.
124
125 This can now be resolved automagically with the duplicates handler
126 `merge-generics' which gives the module system license to merge all
127 generic functions sharing a common name:
128
129 (define-module (math 2D-vectors)
130 :use-module (oop goops)
131 :export (x y ...))
132
133 (define-module (math 3D-vectors)
134 :use-module (oop goops)
135 :export (x y z ...))
136
137 (define-module (my-module)
138 :use-module (math 2D-vectors)
139 :use-module (math 3D-vectors)
140 :duplicates merge-generics)
141
142 x in (my-module) will now share methods with x in both imported
143 modules.
144
145 There will, in fact, now be three distinct generic functions named
146 `x': x in (2D-vectors), x in (3D-vectors), and x in (my-module). The
147 last function will be an <extended-generic>, extending the previous
148 two functions.
149
150 Let's call the imported generic functions the "ancestor functions". x
151 in (my-module) is, in turn, a "descendant function" of the imported
152 functions, extending its ancestors.
153
154 For any generic function G, the applicable methods are selected from
155 the union of the methods of the descendant functions, the methods of G
156 itself and the methods of the ancestor functions.
157
158 This, ancestor functions share methods with their descendants and vice
159 versa. This implies that x in (math 2D-vectors) can will share the
160 methods of x in (my-module) and vice versa, while x in (math 2D-vectors)
161 doesn't share the methods of x in (math 3D-vectors), thus preserving
162 modularity.
163
164 Sharing is dynamic, so that adding new methods to a descendant implies
165 adding it to the ancestor.
166
167 If duplicates checking is desired in the above example, the following
168 form of the :duplicates option can be used instead:
169
170 :duplicates (merge-generics check)
171
172 ** New function: effective-version
173
174 Returns the "effective" version number. This is just the normal full
175 version string without the final micro-version number. See "Changes
176 to the distribution" above.
177
178 ** Futures
179
180 Futures is a way of providing an alternative evaluation policy, very
181 similar in principle to "promises". Like promises, futures allow the
182 main process to continue instantly, but while promises postpone
183 evaluation ("lazy" evaluation) until the value is requested, futures
184 immediately starts evaluation in a parallel thread.
185
186 Futures are good when you want to express that "I'll need the value of
187 this computation sometime soon" and want to allow processing to go on
188 in the background until that time arrives.
189
190 ** New syntax: future FORM
191
192 Begin evaluation of FORM in a parallel thread and return the future
193 immediately. (Akin to 'delay'.)
194
195 ** New procedure: future-ref FUTURE
196
197 Return the computed value of the future. Wait if the computation is
198 not finished. (Akin to 'force'.)
199
200 ** New syntax: parallel FORM ...
201
202 Compute the results of FORM ... in parallel (in a separate thread for
203 each form) and return them as multiple values.
204
205 ** New syntax: letpar ((VAR EXP) ...) BODYFORM ...
206
207 Like 'let' but evaluates the binding expressions EXP ... in parallel.
208
209 ** New functions: par-map, par-for-each PROC ARGLIST ...
210
211 Like 'map' and 'for-each' but evaluate the procedure PROC in a
212 separate thread for each (set of) argument(s). All applications are
213 guaranteed to be completed before the procedure returns.
214
215 ** New functions: n-par-map, n-par-for-each N PROC ARGLIST ...
216
217 Like 'par-map' and 'par-for-each' but evaluate the procedure PROC in N
218 threads. This is useful when PROC uses large amounts of resources
219 and/or the argument list(s) is/are long so that one thread per (set
220 of) argument(s) would consume too much system resources. On a
221 dual-CPU system, N = 4 would often be a good choice.
222
223 ** Fair mutexes and condition variables
224
225 Fair mutexes and condition variables have been added. The fairness
226 means that scheduling is arranged to give as equal time shares as
227 possible and that threads are awakened in a first-in-first-out
228 manner. This is not guaranteed with standard mutexes and condition
229 variables.
230
231 In addition, fair mutexes are recursive. Locking a fair mutex that
232 you have already locked will succeed. Every call to lock-mutex must
233 be matched with a call to unlock-mutex. Only the last call to
234 unlock-mutex will actually unlock the mutex.
235
236 A fair condition variable must be used together with a fair mutex,
237 just as a standard condition variable must be used together with a
238 standard mutex.
239
240 ** New functions: make-fair-mutex, make-fair-condition-variable'
241
242 Make a new fair mutex and a new fair condition variable respectively.
243
244 ** New function 'try-mutex'.
245
246 This function will attempt to lock a mutex but will return immediately
247 instead if blocking and indicate failure.
248
249 ** Waiting on a condition variable can have a timeout.
250
251 The funtion 'wait-condition-variable' now takes a third, optional
252 argument that specifies the point in time where the waiting should be
253 aborted.
254
255 ** New function 'broadcast-condition-variable'.
256
257 ** New functions 'all-threads' and 'current-thread'.
258
259 ** Signals and system asyncs work better with threads.
260
261 The function 'sigaction' now takes a fourth, optional, argument that
262 specifies the thread that the handler should run in. When the
263 argument is omitted, the handler will run in the thread that called
264 'sigaction'.
265
266 Likewise, 'system-async-mark' takes a second, optional, argument that
267 specifies the thread that the async should run in. When it is
268 omitted, the async will run in the thread that called
269 'system-async-mark'.
270
271 C code can use the new functions scm_sigaction_for_thread and
272 scm_system_async_mark_for_thread to pass the new thread argument.
273
274 ** The function 'system-async' is deprecated.
275
276 You can now pass any zero-argument procedure to 'system-async-mark'.
277 The function 'system-async' will just return its argument unchanged
278 now.
279
280 ** New functions 'call-with-blocked-asyncs' and
281 'call-with-unblocked-asyncs'
282
283 The expression (call-with-blocked-asyncs PROC) will call PROC and will
284 block execution of system asyncs for the current thread by one level
285 while PROC runs. Likewise, call-with-unblocked-asyncs will call a
286 procedure and will unblock the execution of system asyncs by one
287 level for the current thread.
288
289 Only system asyncs are affected by these functions.
290
291 ** The functions 'mask-signals' and 'unmask-signals' are deprecated.
292
293 Use 'call-with-blocked-asyncs' or 'call-with-unblocked-asyncs'
294 instead. Those functions are easier to use correctly and can be
295 nested.
296
297 ** New function 'unsetenv'.
298
299 ** New macro 'define-syntax-public'.
300
301 It works like 'define-syntax' and also exports the defined macro (but
302 only on top-level).
303
304 ** There is support for Infinity and NaNs.
305
306 Following PLT Scheme, Guile can now work with infinite numbers, and
307 'not-a-numbers'.
308
309 There is new syntax for numbers: "+inf.0" (infinity), "-inf.0"
310 (negative infinity), "+nan.0" (not-a-number), and "-nan.0" (same as
311 "+nan.0"). These numbers are inexact and have no exact counterpart.
312
313 Dividing by an inexact zero returns +inf.0 or -inf.0, depending on the
314 sign of the dividend. The infinities are integers, and they answer #t
315 for both 'even?' and 'odd?'. The +nan.0 value is not an integer and is
316 not '=' to itself, but '+nan.0' is 'eqv?' to itself.
317
318 For example
319
320 (/ 1 0.0)
321 => +inf.0
322
323 (/ 0 0.0)
324 => +nan.0
325
326 (/ 0)
327 ERROR: Numerical overflow
328
329 Two new predicates 'inf?' and 'nan?' can be used to test for the
330 special values.
331
332 ** Inexact zero can have a sign.
333
334 Guile can now distinguish between plus and minus inexact zero, if your
335 platform supports this, too. The two zeros are equal according to
336 '=', but not according to 'eqv?'. For example
337
338 (- 0.0)
339 => -0.0
340
341 (= 0.0 (- 0.0))
342 => #t
343
344 (eqv? 0.0 (- 0.0))
345 => #f
346
347 ** We now have uninterned symbols.
348
349 The new function 'make-symbol' will return a uninterned symbol. This
350 is a symbol that is unique and is guaranteed to remain unique.
351 However, uninterned symbols can not yet be read back in.
352
353 Use the new function 'symbol-interned?' to check whether a symbol is
354 interned or not.
355
356 ** pretty-print has more options.
357
358 The function pretty-print from the (ice-9 pretty-print) module can now
359 also be invoked with keyword arguments that control things like
360 maximum output width. See its online documentation.
361
362 ** Variables have no longer a special behavior for `equal?'.
363
364 Previously, comparing two variables with `equal?' would recursivly
365 compare their values. This is no longer done. Variables are now only
366 `equal?' if they are `eq?'.
367
368 ** `(begin)' is now valid.
369
370 You can now use an empty `begin' form. It will yield #<unspecified>
371 when evaluated and simply be ignored in a definition context.
372
373 ** Removed: substring-move-left!, substring-move-right!
374
375 Use `substring-move!' instead.
376
377 ** Deprecated: procedure->macro
378
379 Change your code to use either procedure->memoizing-macro or, probably better,
380 to use r5rs macros. Also, be aware that macro expansion will not be done
381 during evaluation, but prior to evaluation.
382
383 ** Soft ports now allow a `char-ready?' procedure
384
385 The vector argument to `make-soft-port' can now have a length of
386 either 5 or 6. (Previously the length had to be 5.) The optional 6th
387 element is interpreted as an `input-waiting' thunk -- i.e. a thunk
388 that returns the number of characters that can be read immediately
389 without the soft port blocking.
390
391 ** New debugging feature: breakpoints.
392
393 Guile now has breakpoints. For details see the `Debugging Features'
394 chapter in the reference manual.
395
396 ** Deprecated: undefine
397
398 There is no replacement for undefine.
399
400 * Changes to the C interface
401
402 ** New function: scm_effective_version
403
404 Returns the "effective" version number. This is just the normal full
405 version string without the final micro-version number. See "Changes
406 to the distribution" above.
407
408 ** The function scm_call_with_new_thread has a new prototype.
409
410 Instead of taking a list with the thunk and handler, these two
411 arguments are now passed directly:
412
413 SCM scm_call_with_new_thread (SCM thunk, SCM handler);
414
415 This is an incompatible change.
416
417 ** The value 'scm_mask_ints' is no longer writable.
418
419 Previously, you could set scm_mask_ints directly. This is no longer
420 possible. Use scm_c_call_with_blocked_asyncs and
421 scm_c_call_with_unblocked_asyncs instead.
422
423 ** New functions scm_c_call_with_blocked_asyncs and
424 scm_c_call_with_unblocked_asyncs
425
426 Like scm_call_with_blocked_asyncs etc. but for C functions.
427
428 ** New snarfer macro SCM_DEFINE_PUBLIC.
429
430 This is like SCM_DEFINE, but also calls scm_c_export for the defined
431 function in the init section.
432
433 ** The snarfer macro SCM_SNARF_INIT is now officially supported.
434
435 ** New macros SCM_VECTOR_REF and SCM_VECTOR_SET.
436
437 Use these in preference to SCM_VELTS.
438
439 ** The SCM_VELTS macros now returns a read-only vector. For writing,
440 use the new macros SCM_WRITABLE_VELTS or SCM_VECTOR_SET. The use of
441 SCM_WRITABLE_VELTS is discouraged, though.
442
443 ** Garbage collector rewrite.
444
445 The garbage collector is cleaned up a lot, and now uses lazy
446 sweeping. This is reflected in the output of (gc-stats); since cells
447 are being freed when they are allocated, the cells-allocated field
448 stays roughly constant.
449
450 For malloc related triggers, the behavior is changed. It uses the same
451 heuristic as the cell-triggered collections. It may be tuned with the
452 environment variables GUILE_MIN_YIELD_MALLOC. This is the percentage
453 for minimum yield of malloc related triggers. The default is 40.
454 GUILE_INIT_MALLOC_LIMIT sets the initial trigger for doing a GC. The
455 default is 200 kb.
456
457 Debugging operations for the freelist have been deprecated, along with
458 the C variables that control garbage collection. The environment
459 variables GUILE_MAX_SEGMENT_SIZE, GUILE_INIT_SEGMENT_SIZE_2,
460 GUILE_INIT_SEGMENT_SIZE_1, and GUILE_MIN_YIELD_2 should be used.
461
462 ** The function scm_definedp has been renamed to scm_defined_p
463
464 The name scm_definedp is deprecated.
465
466 ** The struct scm_cell has been renamed to scm_t_cell
467
468 This is in accordance to Guile's naming scheme for types. Note that
469 the name scm_cell is now used for a function that allocates and
470 initializes a new cell (see below).
471
472 ** New functions for memory management
473
474 A new set of functions for memory management has been added since the
475 old way (scm_must_malloc, scm_must_free, etc) was error prone and
476 indeed, Guile itself contained some long standing bugs that could
477 cause aborts in long running programs.
478
479 The new functions are more symmetrical and do not need cooperation
480 from smob free routines, among other improvements.
481
482 The new functions are scm_malloc, scm_realloc, scm_calloc, scm_strdup,
483 scm_strndup, scm_gc_malloc, scm_gc_calloc, scm_gc_realloc,
484 scm_gc_free, scm_gc_register_collectable_memory, and
485 scm_gc_unregister_collectable_memory. Refer to the manual for more
486 details and for upgrading instructions.
487
488 The old functions for memory management have been deprecated. They
489 are: scm_must_malloc, scm_must_realloc, scm_must_free,
490 scm_must_strdup, scm_must_strndup, scm_done_malloc, scm_done_free.
491
492 ** New function: scm_str2string
493
494 This function creates a scheme string from a 0-terminated C string. The input
495 string is copied.
496
497 ** Declarations of exported features are marked with SCM_API.
498
499 Every declaration of a feature that belongs to the exported Guile API
500 has been marked by adding the macro "SCM_API" to the start of the
501 declaration. This macro can expand into different things, the most
502 common of which is just "extern" for Unix platforms. On Win32, it can
503 be used to control which symbols are exported from a DLL.
504
505 If you `#define SCM_IMPORT' before including <libguile.h>, SCM_API
506 will expand into "__declspec (dllimport) extern", which is needed for
507 linking to the Guile DLL in Windows.
508
509 There are also SCM_RL_IMPORT, QT_IMPORT, SCM_SRFI1314_IMPORT, and
510 SCM_SRFI4_IMPORT, for the corresponding libraries.
511
512 ** SCM_NEWCELL and SCM_NEWCELL2 have been deprecated.
513
514 Use the new functions scm_cell and scm_double_cell instead. The old macros
515 had problems because with them allocation and initialization was separated and
516 the GC could sometimes observe half initialized cells. Only careful coding by
517 the user of SCM_NEWCELL and SCM_NEWCELL2 could make this safe and efficient.
518
519 ** CHECK_ENTRY, CHECK_APPLY and CHECK_EXIT have been deprecated.
520
521 Use the variables scm_check_entry_p, scm_check_apply_p and scm_check_exit_p
522 instead.
523
524 ** SRCBRKP has been deprecated.
525
526 Use scm_c_source_property_breakpoint_p instead.
527
528 ** Deprecated: scm_makmacro
529
530 Change your code to use either scm_makmmacro or, probably better, to use r5rs
531 macros. Also, be aware that macro expansion will not be done during
532 evaluation, but prior to evaluation.
533
534 ** Removed from scm_root_state: def_inp, def_outp, def_errp, together
535 with corresponding macros scm_def_inp, scm_def_outp and scm_def_errp.
536 These were undocumented and unused copies of the standard ports at the
537 time that Guile was initialised. Normally the current ports should be
538 used instead, obtained from scm_current_input_port () etc. If an
539 application needs to retain earlier ports, it should save them in a
540 gc-protected location.
541
542 ** Removed compile time option MEMOIZE_LOCALS
543
544 Now, caching of local variable positions during memoization is mandatory.
545 However, the option to disable the caching has most probably not been used
546 anyway.
547
548 ** Removed compile time option SCM_RECKLESS
549
550 Full number of arguments checking of closures is mandatory now. However, the
551 option to disable the checking has most probably not been used anyway.
552
553 ** Removed compile time option SCM_CAUTIOUS
554
555 Full number of arguments checking of closures is mandatory now. However, the
556 option to disable the checking has most probably not been used anyway.
557
558 ** Deprecated configure flags USE_THREADS and GUILE_ISELECT
559
560 Previously, when the C preprocessor macro USE_THREADS was defined,
561 libguile included a thread API. This API is now always included, even
562 when threads are not really supported. Thus, you don't need to test
563 for USE_THREADS.
564
565 Analogously, GUILE_ISELECT was defined when the function
566 scm_internal_select was provided by Guile. This function is now
567 always defined, and GUILE_ISELECT with it.
568
569 ** Removed definitions: scm_lisp_nil, scm_lisp_t, s_nil_ify, scm_m_nil_ify,
570 s_t_ify, scm_m_t_ify, s_0_cond, scm_m_0_cond, s_0_ify, scm_m_0_ify, s_1_ify,
571 scm_m_1_ify, scm_debug_newcell, scm_debug_newcell2, scm_tc16_allocated,
572 SCM_SET_SYMBOL_HASH, SCM_IM_NIL_IFY, SCM_IM_T_IFY, SCM_IM_0_COND,
573 SCM_IM_0_IFY, SCM_IM_1_IFY, SCM_GC_SET_ALLOCATED, scm_debug_newcell,
574 scm_debug_newcell2, scm_substring_move_left_x, scm_substring_move_right_x,
575 long_long, ulong_long, scm_sizet, SCM_WNA, SCM_OUTOFRANGE, SCM_NALLOC,
576 SCM_HUP_SIGNAL, SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL,
577 SCM_SEGV_SIGNAL, SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL,
578 SCM_SIG_ORD, SCM_ORD_SIG, SCM_NUM_SIGS, moddata, registered_mods,
579 scm_register_module_xxx, scm_registered_modules,
580 scm_clear_registered_modules, scm_wta, *top-level-lookup-closure*,
581 scm_top_level_lookup_closure_var, scm_system_transformer, scm_eval_3,
582 scm_eval2, SCM_SETAND_CAR, SCM_SETOR_CAR, SCM_SETAND_CDR, SCM_SETOR_CDR,
583 SCM_FREEP, SCM_NFREEP, SCM_GC8MARKP, SCM_SETGC8MARK, SCM_CLRGC8MARK,
584 SCM_GCTYP16, SCM_GCCDR, scm_remember, scm_protect_object,
585 scm_unprotect_object, root_module_lookup_closure, scm_sym_app,
586 scm_sym_modules, module_prefix, make_modules_in_var,
587 beautify_user_module_x_var, try_module_autoload_var, scm_module_full_name,
588 scm_the_root_module, scm_make_module, scm_ensure_user_module,
589 scm_load_scheme_module, scm_port, scm_ptob_descriptor, scm_port_rw_active,
590 scm_close_all_ports_except, scm_rstate, scm_rng, scm_i_rstate,
591 SCM_SLOPPY_STRINGP, SCM_RWSTRINGP, SCM_STRING_UCHARS, SCM_STRING_CHARS,
592 scm_read_only_string_p, scm_makstr, scm_makfromstr,
593 scm_make_shared_substring, scm_tc7_substring, SCM_SLOPPY_CONSP,
594 SCM_SLOPPY_NCONSP, scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
595 sym_huh, scm_variable_set_name_hint, scm_builtin_variable, SCM_VARVCELL,
596 SCM_UDVARIABLEP, SCM_DEFVARIABLEP, scm_internal_with_fluids,
597 scm_make_gsubr, scm_make_gsubr_with_generic, scm_create_hook, list*,
598 SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4, SCM_LIST5,
599 SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9, scm_listify, scm_sloppy_memq,
600 scm_sloppy_memv, scm_sloppy_member, scm_end_of_file_key,
601 scm_read_and_eval_x, scm_mkbig, scm_big2inum, scm_adjbig, scm_normbig,
602 scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl, SCM_FIXNUM_BIT,
603 scm_subr_entry, SCM_SUBR_DOC, scm_make_subr_opt, scm_make_subr,
604 scm_make_subr_with_generic, setjmp_type, setjmp_type,
605 scm_call_catching_errors, scm_make_smob_type_mfpe, scm_set_smob_mfpe,
606 scm_strprint_obj, scm_read_0str, scm_eval_0str, SCM_CHARS, SCM_UCHARS,
607 SCM_SETCHARS, SCM_SLOPPY_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET,
608 SCM_LENGTH_MAX, SCM_LENGTH, SCM_SETLENGTH, SCM_ROSTRINGP, SCM_ROLENGTH,
609 SCM_ROCHARS, SCM_ROUCHARS, SCM_SUBSTRP, SCM_COERCE_SUBSTR, scm_strhash,
610 scm_sym2vcell, scm_sym2ovcell_soft, scm_sym2ovcell,
611 scm_intern_obarray_soft, scm_intern_obarray, scm_intern, scm_intern0,
612 scm_sysintern, scm_sysintern0, scm_sysintern0_no_module_lookup,
613 scm_symbol_value0, scm_string_to_obarray_symbol, scm_intern_symbol,
614 scm_unintern_symbol, scm_symbol_binding, scm_symbol_interned_p,
615 scm_symbol_bound_p, scm_symbol_set_x, scm_gentemp,
616 scm_init_symbols_deprecated, s_vector_set_length_x, scm_vector_set_length_x,
617 scm_contregs, scm_debug_info, scm_debug_frame, SCM_DSIDEVAL, SCM_OPDIRP,
618 scm_fport, scm_option, SCM_CONST_LONG, SCM_VCELL, SCM_GLOBAL_VCELL,
619 SCM_VCELL_INIT, SCM_GLOBAL_VCELL_INIT, scm_srcprops, scm_srcprops_chunk,
620 scm_info_frame, scm_stack, scm_array, scm_array_dim, SCM_ARRAY_CONTIGUOUS,
621 SCM_HUGE_LENGTH, SCM_FUNC_NAME, SCM_WTA, RETURN_SCM_WTA,
622 SCM_VALIDATE_NUMBER_COPY, SCM_VALIDATE_NUMBER_DEF_COPY,
623 SCM_VALIDATE_STRINGORSUBSTR, SCM_VALIDATE_ROSTRING,
624 SCM_VALIDATE_ROSTRING_COPY, SCM_VALIDATE_NULLORROSTRING_COPY,
625 SCM_VALIDATE_RWSTRING, SCM_VALIDATE_OPDIR, DIGITS, scm_small_istr2int,
626 scm_istr2int, scm_istr2flo, scm_istring2number, scm_istr2int,
627 scm_istr2flo, scm_istring2number, scm_vtable_index_vcell, scm_si_vcell,
628 SCM_ECONSP, SCM_NECONSP, SCM_GLOC_VAR, SCM_GLOC_VAL, SCM_GLOC_SET_VAL,
629 SCM_GLOC_VAL_LOC, scm_make_gloc, scm_gloc_p, scm_tc16_variable
630
631 Changes since Guile 1.4:
632
633 * Changes to the distribution
634
635 ** A top-level TODO file is included.
636
637 ** Guile now uses a versioning scheme similar to that of the Linux kernel.
638
639 Guile now always uses three numbers to represent the version,
640 i.e. "1.6.5". The first number, 1, is the major version number, the
641 second number, 6, is the minor version number, and the third number,
642 5, is the micro version number. Changes in major version number
643 indicate major changes in Guile.
644
645 Minor version numbers that are even denote stable releases, and odd
646 minor version numbers denote development versions (which may be
647 unstable). The micro version number indicates a minor sub-revision of
648 a given MAJOR.MINOR release.
649
650 In keeping with the new scheme, (minor-version) and scm_minor_version
651 no longer return everything but the major version number. They now
652 just return the minor version number. Two new functions
653 (micro-version) and scm_micro_version have been added to report the
654 micro version number.
655
656 In addition, ./GUILE-VERSION now defines GUILE_MICRO_VERSION.
657
658 ** New preprocessor definitions are available for checking versions.
659
660 version.h now #defines SCM_MAJOR_VERSION, SCM_MINOR_VERSION, and
661 SCM_MICRO_VERSION to the appropriate integer values.
662
663 ** Guile now actively warns about deprecated features.
664
665 The new configure option `--enable-deprecated=LEVEL' and the
666 environment variable GUILE_WARN_DEPRECATED control this mechanism.
667 See INSTALL and README for more information.
668
669 ** Guile is much more likely to work on 64-bit architectures.
670
671 Guile now compiles and passes "make check" with only two UNRESOLVED GC
672 cases on Alpha and ia64 based machines now. Thanks to John Goerzen
673 for the use of a test machine, and thanks to Stefan Jahn for ia64
674 patches.
675
676 ** New functions: setitimer and getitimer.
677
678 These implement a fairly direct interface to the libc functions of the
679 same name.
680
681 ** The #. reader extension is now disabled by default.
682
683 For safety reasons, #. evaluation is disabled by default. To
684 re-enable it, set the fluid read-eval? to #t. For example:
685
686 (fluid-set! read-eval? #t)
687
688 but make sure you realize the potential security risks involved. With
689 read-eval? enabled, reading a data file from an untrusted source can
690 be dangerous.
691
692 ** New SRFI modules have been added:
693
694 SRFI-0 `cond-expand' is now supported in Guile, without requiring
695 using a module.
696
697 (srfi srfi-1) is a library containing many useful pair- and list-processing
698 procedures.
699
700 (srfi srfi-2) exports and-let*.
701
702 (srfi srfi-4) implements homogeneous numeric vector datatypes.
703
704 (srfi srfi-6) is a dummy module for now, since guile already provides
705 all of the srfi-6 procedures by default: open-input-string,
706 open-output-string, get-output-string.
707
708 (srfi srfi-8) exports receive.
709
710 (srfi srfi-9) exports define-record-type.
711
712 (srfi srfi-10) exports define-reader-ctor and implements the reader
713 extension #,().
714
715 (srfi srfi-11) exports let-values and let*-values.
716
717 (srfi srfi-13) implements the SRFI String Library.
718
719 (srfi srfi-14) implements the SRFI Character-Set Library.
720
721 (srfi srfi-17) implements setter and getter-with-setter and redefines
722 some accessor procedures as procedures with getters. (such as car,
723 cdr, vector-ref etc.)
724
725 (srfi srfi-19) implements the SRFI Time/Date Library.
726
727 ** New scripts / "executable modules"
728
729 Subdirectory "scripts" contains Scheme modules that are packaged to
730 also be executable as scripts. At this time, these scripts are available:
731
732 display-commentary
733 doc-snarf
734 generate-autoload
735 punify
736 read-scheme-source
737 use2dot
738
739 See README there for more info.
740
741 These scripts can be invoked from the shell with the new program
742 "guile-tools", which keeps track of installation directory for you.
743 For example:
744
745 $ guile-tools display-commentary srfi/*.scm
746
747 guile-tools is copied to the standard $bindir on "make install".
748
749 ** New module (ice-9 stack-catch):
750
751 stack-catch is like catch, but saves the current state of the stack in
752 the fluid the-last-stack. This fluid can be useful when using the
753 debugger and when re-throwing an error.
754
755 ** The module (ice-9 and-let*) has been renamed to (ice-9 and-let-star)
756
757 This has been done to prevent problems on lesser operating systems
758 that can't tolerate `*'s in file names. The exported macro continues
759 to be named `and-let*', of course.
760
761 On systems that support it, there is also a compatibility module named
762 (ice-9 and-let*). It will go away in the next release.
763
764 ** New modules (oop goops) etc.:
765
766 (oop goops)
767 (oop goops describe)
768 (oop goops save)
769 (oop goops active-slot)
770 (oop goops composite-slot)
771
772 The Guile Object Oriented Programming System (GOOPS) has been
773 integrated into Guile. For further information, consult the GOOPS
774 manual and tutorial in the `doc' directory.
775
776 ** New module (ice-9 rdelim).
777
778 This exports the following procedures which were previously defined
779 in the default environment:
780
781 read-line read-line! read-delimited read-delimited! %read-delimited!
782 %read-line write-line
783
784 For backwards compatibility the definitions are still imported into the
785 default environment in this version of Guile. However you should add:
786
787 (use-modules (ice-9 rdelim))
788
789 to any program which uses the definitions, since this may change in
790 future.
791
792 Alternatively, if guile-scsh is installed, the (scsh rdelim) module
793 can be used for similar functionality.
794
795 ** New module (ice-9 rw)
796
797 This is a subset of the (scsh rw) module from guile-scsh. Currently
798 it defines two procedures:
799
800 *** New function: read-string!/partial str [port_or_fdes [start [end]]]
801
802 Read characters from a port or file descriptor into a string STR.
803 A port must have an underlying file descriptor -- a so-called
804 fport. This procedure is scsh-compatible and can efficiently read
805 large strings.
806
807 *** New function: write-string/partial str [port_or_fdes [start [end]]]
808
809 Write characters from a string STR to a port or file descriptor.
810 A port must have an underlying file descriptor -- a so-called
811 fport. This procedure is mostly compatible and can efficiently
812 write large strings.
813
814 ** New module (ice-9 match)
815
816 This module includes Andrew K. Wright's pattern matcher. See
817 ice-9/match.scm for brief description or
818
819 http://www.star-lab.com/wright/code.html
820
821 for complete documentation.
822
823 ** New module (ice-9 buffered-input)
824
825 This module provides procedures to construct an input port from an
826 underlying source of input that reads and returns its input in chunks.
827 The underlying input source is a Scheme procedure, specified by the
828 caller, which the port invokes whenever it needs more input.
829
830 This is useful when building an input port whose back end is Readline
831 or a UI element such as the GtkEntry widget.
832
833 ** Documentation
834
835 The reference and tutorial documentation that was previously
836 distributed separately, as `guile-doc', is now included in the core
837 Guile distribution. The documentation consists of the following
838 manuals.
839
840 - The Guile Tutorial (guile-tut.texi) contains a tutorial introduction
841 to using Guile.
842
843 - The Guile Reference Manual (guile.texi) contains (or is intended to
844 contain) reference documentation on all aspects of Guile.
845
846 - The GOOPS Manual (goops.texi) contains both tutorial-style and
847 reference documentation for using GOOPS, Guile's Object Oriented
848 Programming System.
849
850 - The Revised^5 Report on the Algorithmic Language Scheme
851 (r5rs.texi).
852
853 See the README file in the `doc' directory for more details.
854
855 ** There are a couple of examples in the examples/ directory now.
856
857 * Changes to the stand-alone interpreter
858
859 ** New command line option `--use-srfi'
860
861 Using this option, SRFI modules can be loaded on startup and be
862 available right from the beginning. This makes programming portable
863 Scheme programs easier.
864
865 The option `--use-srfi' expects a comma-separated list of numbers,
866 each representing a SRFI number to be loaded into the interpreter
867 before starting evaluating a script file or the REPL. Additionally,
868 the feature identifier for the loaded SRFIs is recognized by
869 `cond-expand' when using this option.
870
871 Example:
872 $ guile --use-srfi=8,13
873 guile> (receive (x z) (values 1 2) (+ 1 2))
874 3
875 guile> (string-pad "bla" 20)
876 " bla"
877
878 ** Guile now always starts up in the `(guile-user)' module.
879
880 Previously, scripts executed via the `-s' option would run in the
881 `(guile)' module and the repl would run in the `(guile-user)' module.
882 Now every user action takes place in the `(guile-user)' module by
883 default.
884
885 * Changes to Scheme functions and syntax
886
887 ** Character classifiers work for non-ASCII characters.
888
889 The predicates `char-alphabetic?', `char-numeric?',
890 `char-whitespace?', `char-lower?', `char-upper?' and `char-is-both?'
891 no longer check whether their arguments are ASCII characters.
892 Previously, a character would only be considered alphabetic when it
893 was also ASCII, for example.
894
895 ** Previously deprecated Scheme functions have been removed:
896
897 tag - no replacement.
898 fseek - replaced by seek.
899 list* - replaced by cons*.
900
901 ** It's now possible to create modules with controlled environments
902
903 Example:
904
905 (use-modules (ice-9 safe))
906 (define m (make-safe-module))
907 ;;; m will now be a module containing only a safe subset of R5RS
908 (eval '(+ 1 2) m) --> 3
909 (eval 'load m) --> ERROR: Unbound variable: load
910
911 ** Evaluation of "()", the empty list, is now an error.
912
913 Previously, the expression "()" evaluated to the empty list. This has
914 been changed to signal a "missing expression" error. The correct way
915 to write the empty list as a literal constant is to use quote: "'()".
916
917 ** New concept of `Guile Extensions'.
918
919 A Guile Extension is just a ordinary shared library that can be linked
920 at run-time. We found it advantageous to give this simple concept a
921 dedicated name to distinguish the issues related to shared libraries
922 from the issues related to the module system.
923
924 *** New function: load-extension
925
926 Executing (load-extension lib init) is mostly equivalent to
927
928 (dynamic-call init (dynamic-link lib))
929
930 except when scm_register_extension has been called previously.
931 Whenever appropriate, you should use `load-extension' instead of
932 dynamic-link and dynamic-call.
933
934 *** New C function: scm_c_register_extension
935
936 This function registers a initialization function for use by
937 `load-extension'. Use it when you don't want specific extensions to
938 be loaded as shared libraries (for example on platforms that don't
939 support dynamic linking).
940
941 ** Auto-loading of compiled-code modules is deprecated.
942
943 Guile used to be able to automatically find and link a shared
944 library to satisfy requests for a module. For example, the module
945 `(foo bar)' could be implemented by placing a shared library named
946 "foo/libbar.so" (or with a different extension) in a directory on the
947 load path of Guile.
948
949 This has been found to be too tricky, and is no longer supported. The
950 shared libraries are now called "extensions". You should now write a
951 small Scheme file that calls `load-extension' to load the shared
952 library and initialize it explicitely.
953
954 The shared libraries themselves should be installed in the usual
955 places for shared libraries, with names like "libguile-foo-bar".
956
957 For example, place this into a file "foo/bar.scm"
958
959 (define-module (foo bar))
960
961 (load-extension "libguile-foo-bar" "foobar_init")
962
963 ** Backward incompatible change: eval EXP ENVIRONMENT-SPECIFIER
964
965 `eval' is now R5RS, that is it takes two arguments.
966 The second argument is an environment specifier, i.e. either
967
968 (scheme-report-environment 5)
969 (null-environment 5)
970 (interaction-environment)
971
972 or
973
974 any module.
975
976 ** The module system has been made more disciplined.
977
978 The function `eval' will save and restore the current module around
979 the evaluation of the specified expression. While this expression is
980 evaluated, `(current-module)' will now return the right module, which
981 is the module specified as the second argument to `eval'.
982
983 A consequence of this change is that `eval' is not particularly
984 useful when you want allow the evaluated code to change what module is
985 designated as the current module and have this change persist from one
986 call to `eval' to the next. The read-eval-print-loop is an example
987 where `eval' is now inadequate. To compensate, there is a new
988 function `primitive-eval' that does not take a module specifier and
989 that does not save/restore the current module. You should use this
990 function together with `set-current-module', `current-module', etc
991 when you want to have more control over the state that is carried from
992 one eval to the next.
993
994 Additionally, it has been made sure that forms that are evaluated at
995 the top level are always evaluated with respect to the current module.
996 Previously, subforms of top-level forms such as `begin', `case',
997 etc. did not respect changes to the current module although these
998 subforms are at the top-level as well.
999
1000 To prevent strange behavior, the forms `define-module',
1001 `use-modules', `use-syntax', and `export' have been restricted to only
1002 work on the top level. The forms `define-public' and
1003 `defmacro-public' only export the new binding on the top level. They
1004 behave just like `define' and `defmacro', respectively, when they are
1005 used in a lexical environment.
1006
1007 Also, `export' will no longer silently re-export bindings imported
1008 from a used module. It will emit a `deprecation' warning and will
1009 cease to perform any re-export in the next version. If you actually
1010 want to re-export bindings, use the new `re-export' in place of
1011 `export'. The new `re-export' will not make copies of variables when
1012 rexporting them, as `export' did wrongly.
1013
1014 ** Module system now allows selection and renaming of imported bindings
1015
1016 Previously, when using `use-modules' or the `#:use-module' clause in
1017 the `define-module' form, all the bindings (association of symbols to
1018 values) for imported modules were added to the "current module" on an
1019 as-is basis. This has been changed to allow finer control through two
1020 new facilities: selection and renaming.
1021
1022 You can now select which of the imported module's bindings are to be
1023 visible in the current module by using the `:select' clause. This
1024 clause also can be used to rename individual bindings. For example:
1025
1026 ;; import all bindings no questions asked
1027 (use-modules (ice-9 common-list))
1028
1029 ;; import four bindings, renaming two of them;
1030 ;; the current module sees: every some zonk-y zonk-n
1031 (use-modules ((ice-9 common-list)
1032 :select (every some
1033 (remove-if . zonk-y)
1034 (remove-if-not . zonk-n))))
1035
1036 You can also programmatically rename all selected bindings using the
1037 `:renamer' clause, which specifies a proc that takes a symbol and
1038 returns another symbol. Because it is common practice to use a prefix,
1039 we now provide the convenience procedure `symbol-prefix-proc'. For
1040 example:
1041
1042 ;; import four bindings, renaming two of them specifically,
1043 ;; and all four w/ prefix "CL:";
1044 ;; the current module sees: CL:every CL:some CL:zonk-y CL:zonk-n
1045 (use-modules ((ice-9 common-list)
1046 :select (every some
1047 (remove-if . zonk-y)
1048 (remove-if-not . zonk-n))
1049 :renamer (symbol-prefix-proc 'CL:)))
1050
1051 ;; import four bindings, renaming two of them specifically,
1052 ;; and all four by upcasing.
1053 ;; the current module sees: EVERY SOME ZONK-Y ZONK-N
1054 (define (upcase-symbol sym)
1055 (string->symbol (string-upcase (symbol->string sym))))
1056
1057 (use-modules ((ice-9 common-list)
1058 :select (every some
1059 (remove-if . zonk-y)
1060 (remove-if-not . zonk-n))
1061 :renamer upcase-symbol))
1062
1063 Note that programmatic renaming is done *after* individual renaming.
1064 Also, the above examples show `use-modules', but the same facilities are
1065 available for the `#:use-module' clause of `define-module'.
1066
1067 See manual for more info.
1068
1069 ** The semantics of guardians have changed.
1070
1071 The changes are for the most part compatible. An important criterion
1072 was to keep the typical usage of guardians as simple as before, but to
1073 make the semantics safer and (as a result) more useful.
1074
1075 *** All objects returned from guardians are now properly alive.
1076
1077 It is now guaranteed that any object referenced by an object returned
1078 from a guardian is alive. It's now impossible for a guardian to
1079 return a "contained" object before its "containing" object.
1080
1081 One incompatible (but probably not very important) change resulting
1082 from this is that it is no longer possible to guard objects that
1083 indirectly reference themselves (i.e. are parts of cycles). If you do
1084 so accidentally, you'll get a warning.
1085
1086 *** There are now two types of guardians: greedy and sharing.
1087
1088 If you call (make-guardian #t) or just (make-guardian), you'll get a
1089 greedy guardian, and for (make-guardian #f) a sharing guardian.
1090
1091 Greedy guardians are the default because they are more "defensive".
1092 You can only greedily guard an object once. If you guard an object
1093 more than once, once in a greedy guardian and the rest of times in
1094 sharing guardians, then it is guaranteed that the object won't be
1095 returned from sharing guardians as long as it is greedily guarded
1096 and/or alive.
1097
1098 Guardians returned by calls to `make-guardian' can now take one more
1099 optional parameter, which says whether to throw an error in case an
1100 attempt is made to greedily guard an object that is already greedily
1101 guarded. The default is true, i.e. throw an error. If the parameter
1102 is false, the guardian invocation returns #t if guarding was
1103 successful and #f if it wasn't.
1104
1105 Also, since greedy guarding is, in effect, a side-effecting operation
1106 on objects, a new function is introduced: `destroy-guardian!'.
1107 Invoking this function on a guardian renders it unoperative and, if
1108 the guardian is greedy, clears the "greedily guarded" property of the
1109 objects that were guarded by it, thus undoing the side effect.
1110
1111 Note that all this hair is hardly very important, since guardian
1112 objects are usually permanent.
1113
1114 ** Continuations created by call-with-current-continuation now accept
1115 any number of arguments, as required by R5RS.
1116
1117 ** New function `issue-deprecation-warning'
1118
1119 This function is used to display the deprecation messages that are
1120 controlled by GUILE_WARN_DEPRECATION as explained in the README.
1121
1122 (define (id x)
1123 (issue-deprecation-warning "`id' is deprecated. Use `identity' instead.")
1124 (identity x))
1125
1126 guile> (id 1)
1127 ;; `id' is deprecated. Use `identity' instead.
1128 1
1129 guile> (id 1)
1130 1
1131
1132 ** New syntax `begin-deprecated'
1133
1134 When deprecated features are included (as determined by the configure
1135 option --enable-deprecated), `begin-deprecated' is identical to
1136 `begin'. When deprecated features are excluded, it always evaluates
1137 to `#f', ignoring the body forms.
1138
1139 ** New function `make-object-property'
1140
1141 This function returns a new `procedure with setter' P that can be used
1142 to attach a property to objects. When calling P as
1143
1144 (set! (P obj) val)
1145
1146 where `obj' is any kind of object, it attaches `val' to `obj' in such
1147 a way that it can be retrieved by calling P as
1148
1149 (P obj)
1150
1151 This function will replace procedure properties, symbol properties and
1152 source properties eventually.
1153
1154 ** Module (ice-9 optargs) now uses keywords instead of `#&'.
1155
1156 Instead of #&optional, #&key, etc you should now use #:optional,
1157 #:key, etc. Since #:optional is a keyword, you can write it as just
1158 :optional when (read-set! keywords 'prefix) is active.
1159
1160 The old reader syntax `#&' is still supported, but deprecated. It
1161 will be removed in the next release.
1162
1163 ** New define-module option: pure
1164
1165 Tells the module system not to include any bindings from the root
1166 module.
1167
1168 Example:
1169
1170 (define-module (totally-empty-module)
1171 :pure)
1172
1173 ** New define-module option: export NAME1 ...
1174
1175 Export names NAME1 ...
1176
1177 This option is required if you want to be able to export bindings from
1178 a module which doesn't import one of `define-public' or `export'.
1179
1180 Example:
1181
1182 (define-module (foo)
1183 :pure
1184 :use-module (ice-9 r5rs)
1185 :export (bar))
1186
1187 ;;; Note that we're pure R5RS below this point!
1188
1189 (define (bar)
1190 ...)
1191
1192 ** New function: object->string OBJ
1193
1194 Return a Scheme string obtained by printing a given object.
1195
1196 ** New function: port? X
1197
1198 Returns a boolean indicating whether X is a port. Equivalent to
1199 `(or (input-port? X) (output-port? X))'.
1200
1201 ** New function: file-port?
1202
1203 Determines whether a given object is a port that is related to a file.
1204
1205 ** New function: port-for-each proc
1206
1207 Apply PROC to each port in the Guile port table in turn. The return
1208 value is unspecified. More specifically, PROC is applied exactly once
1209 to every port that exists in the system at the time PORT-FOR-EACH is
1210 invoked. Changes to the port table while PORT-FOR-EACH is running
1211 have no effect as far as PORT-FOR-EACH is concerned.
1212
1213 ** New function: dup2 oldfd newfd
1214
1215 A simple wrapper for the `dup2' system call. Copies the file
1216 descriptor OLDFD to descriptor number NEWFD, replacing the
1217 previous meaning of NEWFD. Both OLDFD and NEWFD must be integers.
1218 Unlike for dup->fdes or primitive-move->fdes, no attempt is made
1219 to move away ports which are using NEWFD. The return value is
1220 unspecified.
1221
1222 ** New function: close-fdes fd
1223
1224 A simple wrapper for the `close' system call. Close file
1225 descriptor FD, which must be an integer. Unlike close (*note
1226 close: Ports and File Descriptors.), the file descriptor will be
1227 closed even if a port is using it. The return value is
1228 unspecified.
1229
1230 ** New function: crypt password salt
1231
1232 Encrypts `password' using the standard unix password encryption
1233 algorithm.
1234
1235 ** New function: chroot path
1236
1237 Change the root directory of the running process to `path'.
1238
1239 ** New functions: getlogin, cuserid
1240
1241 Return the login name or the user name of the current effective user
1242 id, respectively.
1243
1244 ** New functions: getpriority which who, setpriority which who prio
1245
1246 Get or set the priority of the running process.
1247
1248 ** New function: getpass prompt
1249
1250 Read a password from the terminal, first displaying `prompt' and
1251 disabling echoing.
1252
1253 ** New function: flock file operation
1254
1255 Set/remove an advisory shared or exclusive lock on `file'.
1256
1257 ** New functions: sethostname name, gethostname
1258
1259 Set or get the hostname of the machine the current process is running
1260 on.
1261
1262 ** New function: mkstemp! tmpl
1263
1264 mkstemp creates a new unique file in the file system and returns a
1265 new buffered port open for reading and writing to the file. TMPL
1266 is a string specifying where the file should be created: it must
1267 end with `XXXXXX' and will be changed in place to return the name
1268 of the temporary file.
1269
1270 ** New function: open-input-string string
1271
1272 Return an input string port which delivers the characters from
1273 `string'. This procedure, together with `open-output-string' and
1274 `get-output-string' implements SRFI-6.
1275
1276 ** New function: open-output-string
1277
1278 Return an output string port which collects all data written to it.
1279 The data can then be retrieved by `get-output-string'.
1280
1281 ** New function: get-output-string
1282
1283 Return the contents of an output string port.
1284
1285 ** New function: identity
1286
1287 Return the argument.
1288
1289 ** socket, connect, accept etc., now have support for IPv6. IPv6 addresses
1290 are represented in Scheme as integers with normal host byte ordering.
1291
1292 ** New function: inet-pton family address
1293
1294 Convert a printable string network address into an integer. Note that
1295 unlike the C version of this function, the result is an integer with
1296 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1297 e.g.,
1298
1299 (inet-pton AF_INET "127.0.0.1") => 2130706433
1300 (inet-pton AF_INET6 "::1") => 1
1301
1302 ** New function: inet-ntop family address
1303
1304 Convert an integer network address into a printable string. Note that
1305 unlike the C version of this function, the input is an integer with
1306 normal host byte ordering. FAMILY can be `AF_INET' or `AF_INET6'.
1307 e.g.,
1308
1309 (inet-ntop AF_INET 2130706433) => "127.0.0.1"
1310 (inet-ntop AF_INET6 (- (expt 2 128) 1)) =>
1311 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
1312
1313 ** Deprecated: id
1314
1315 Use `identity' instead.
1316
1317 ** Deprecated: -1+
1318
1319 Use `1-' instead.
1320
1321 ** Deprecated: return-it
1322
1323 Do without it.
1324
1325 ** Deprecated: string-character-length
1326
1327 Use `string-length' instead.
1328
1329 ** Deprecated: flags
1330
1331 Use `logior' instead.
1332
1333 ** Deprecated: close-all-ports-except.
1334
1335 This was intended for closing ports in a child process after a fork,
1336 but it has the undesirable side effect of flushing buffers.
1337 port-for-each is more flexible.
1338
1339 ** The (ice-9 popen) module now attempts to set up file descriptors in
1340 the child process from the current Scheme ports, instead of using the
1341 current values of file descriptors 0, 1, and 2 in the parent process.
1342
1343 ** Removed function: builtin-weak-bindings
1344
1345 There is no such concept as a weak binding any more.
1346
1347 ** Removed constants: bignum-radix, scm-line-incrementors
1348
1349 ** define-method: New syntax mandatory.
1350
1351 The new method syntax is now mandatory:
1352
1353 (define-method (NAME ARG-SPEC ...) BODY ...)
1354 (define-method (NAME ARG-SPEC ... . REST-ARG) BODY ...)
1355
1356 ARG-SPEC ::= ARG-NAME | (ARG-NAME TYPE)
1357 REST-ARG ::= ARG-NAME
1358
1359 If you have old code using the old syntax, import
1360 (oop goops old-define-method) before (oop goops) as in:
1361
1362 (use-modules (oop goops old-define-method) (oop goops))
1363
1364 ** Deprecated function: builtin-variable
1365 Removed function: builtin-bindings
1366
1367 There is no longer a distinction between builtin or other variables.
1368 Use module system operations for all variables.
1369
1370 ** Lazy-catch handlers are no longer allowed to return.
1371
1372 That is, a call to `throw', `error', etc is now guaranteed to not
1373 return.
1374
1375 ** Bugfixes for (ice-9 getopt-long)
1376
1377 This module is now tested using test-suite/tests/getopt-long.test.
1378 The following bugs have been fixed:
1379
1380 *** Parsing for options that are specified to have `optional' args now checks
1381 if the next element is an option instead of unconditionally taking it as the
1382 option arg.
1383
1384 *** An error is now thrown for `--opt=val' when the option description
1385 does not specify `(value #t)' or `(value optional)'. This condition used to
1386 be accepted w/o error, contrary to the documentation.
1387
1388 *** The error message for unrecognized options is now more informative.
1389 It used to be "not a record", an artifact of the implementation.
1390
1391 *** The error message for `--opt' terminating the arg list (no value), when
1392 `(value #t)' is specified, is now more informative. It used to be "not enough
1393 args".
1394
1395 *** "Clumped" single-char args now preserve trailing string, use it as arg.
1396 The expansion used to be like so:
1397
1398 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "--xyz")
1399
1400 Note that the "5d" is dropped. Now it is like so:
1401
1402 ("-abc5d" "--xyz") => ("-a" "-b" "-c" "5d" "--xyz")
1403
1404 This enables single-char options to have adjoining arguments as long as their
1405 constituent characters are not potential single-char options.
1406
1407 ** (ice-9 session) procedure `arity' now works with (ice-9 optargs) `lambda*'
1408
1409 The `lambda*' and derivative forms in (ice-9 optargs) now set a procedure
1410 property `arglist', which can be retrieved by `arity'. The result is that
1411 `arity' can give more detailed information than before:
1412
1413 Before:
1414
1415 guile> (use-modules (ice-9 optargs))
1416 guile> (define* (foo #:optional a b c) a)
1417 guile> (arity foo)
1418 0 or more arguments in `lambda*:G0'.
1419
1420 After:
1421
1422 guile> (arity foo)
1423 3 optional arguments: `a', `b' and `c'.
1424 guile> (define* (bar a b #:key c d #:allow-other-keys) a)
1425 guile> (arity bar)
1426 2 required arguments: `a' and `b', 2 keyword arguments: `c'
1427 and `d', other keywords allowed.
1428 guile> (define* (baz a b #:optional c #:rest r) a)
1429 guile> (arity baz)
1430 2 required arguments: `a' and `b', 1 optional argument: `c',
1431 the rest in `r'.
1432
1433 * Changes to the C interface
1434
1435 ** Types have been renamed from scm_*_t to scm_t_*.
1436
1437 This has been done for POSIX sake. It reserves identifiers ending
1438 with "_t". What a concept.
1439
1440 The old names are still available with status `deprecated'.
1441
1442 ** scm_t_bits (former scm_bits_t) is now a unsigned type.
1443
1444 ** Deprecated features have been removed.
1445
1446 *** Macros removed
1447
1448 SCM_INPORTP, SCM_OUTPORTP SCM_ICHRP, SCM_ICHR, SCM_MAKICHR
1449 SCM_SETJMPBUF SCM_NSTRINGP SCM_NRWSTRINGP SCM_NVECTORP SCM_DOUBLE_CELLP
1450
1451 *** C Functions removed
1452
1453 scm_sysmissing scm_tag scm_tc16_flo scm_tc_flo
1454 scm_fseek - replaced by scm_seek.
1455 gc-thunk - replaced by after-gc-hook.
1456 gh_int2scmb - replaced by gh_bool2scm.
1457 scm_tc_dblr - replaced by scm_tc16_real.
1458 scm_tc_dblc - replaced by scm_tc16_complex.
1459 scm_list_star - replaced by scm_cons_star.
1460
1461 ** Deprecated: scm_makfromstr
1462
1463 Use scm_mem2string instead.
1464
1465 ** Deprecated: scm_make_shared_substring
1466
1467 Explicit shared substrings will disappear from Guile.
1468
1469 Instead, "normal" strings will be implemented using sharing
1470 internally, combined with a copy-on-write strategy.
1471
1472 ** Deprecated: scm_read_only_string_p
1473
1474 The concept of read-only strings will disappear in next release of
1475 Guile.
1476
1477 ** Deprecated: scm_sloppy_memq, scm_sloppy_memv, scm_sloppy_member
1478
1479 Instead, use scm_c_memq or scm_memq, scm_memv, scm_member.
1480
1481 ** New functions: scm_call_0, scm_call_1, scm_call_2, scm_call_3
1482
1483 Call a procedure with the indicated number of arguments.
1484
1485 Example:
1486
1487 scm_call_1 (proc, arg1);
1488
1489 ** New functions: scm_apply_0, scm_apply_1, scm_apply_2, scm_apply_3
1490
1491 Call a procedure with the indicated number of arguments and a list
1492 of arguments.
1493
1494 Example:
1495
1496 scm_apply_1 (proc, arg1, args);
1497
1498 ** New functions: scm_list_1, scm_list_2, scm_list_3, scm_list_4, scm_list_5
1499
1500 Create a list of the given number of elements.
1501
1502 ** Renamed function: scm_listify has been replaced by scm_list_n.
1503
1504 ** Deprecated macros: SCM_LIST0, SCM_LIST1, SCM_LIST2, SCM_LIST3, SCM_LIST4,
1505 SCM_LIST5, SCM_LIST6, SCM_LIST7, SCM_LIST8, SCM_LIST9.
1506
1507 Use functions scm_list_N instead.
1508
1509 ** New function: scm_c_read (SCM port, void *buffer, scm_sizet size)
1510
1511 Used by an application to read arbitrary number of bytes from a port.
1512 Same semantics as libc read, except that scm_c_read only returns less
1513 than SIZE bytes if at end-of-file.
1514
1515 Warning: Doesn't update port line and column counts!
1516
1517 ** New function: scm_c_write (SCM port, const void *ptr, scm_sizet size)
1518
1519 Used by an application to write arbitrary number of bytes to an SCM
1520 port. Similar semantics as libc write. However, unlike libc
1521 write, scm_c_write writes the requested number of bytes and has no
1522 return value.
1523
1524 Warning: Doesn't update port line and column counts!
1525
1526 ** New function: scm_init_guile ()
1527
1528 In contrast to scm_boot_guile, scm_init_guile will return normally
1529 after initializing Guile. It is not available on all systems, tho.
1530
1531 ** New functions: scm_str2symbol, scm_mem2symbol
1532
1533 The function scm_str2symbol takes a const char* pointing to a zero-terminated
1534 field of characters and creates a scheme symbol object from that C string.
1535 The function scm_mem2symbol takes a const char* and a number of characters and
1536 creates a symbol from the characters in that memory area.
1537
1538 ** New functions: scm_primitive_make_property
1539 scm_primitive_property_ref
1540 scm_primitive_property_set_x
1541 scm_primitive_property_del_x
1542
1543 These functions implement a new way to deal with object properties.
1544 See libguile/properties.c for their documentation.
1545
1546 ** New function: scm_done_free (long size)
1547
1548 This function is the inverse of scm_done_malloc. Use it to report the
1549 amount of smob memory you free. The previous method, which involved
1550 calling scm_done_malloc with negative argument, was somewhat
1551 unintuitive (and is still available, of course).
1552
1553 ** New function: scm_c_memq (SCM obj, SCM list)
1554
1555 This function provides a fast C level alternative for scm_memq for the case
1556 that the list parameter is known to be a proper list. The function is a
1557 replacement for scm_sloppy_memq, but is stricter in its requirements on its
1558 list input parameter, since for anything else but a proper list the function's
1559 behaviour is undefined - it may even crash or loop endlessly. Further, for
1560 the case that the object is not found in the list, scm_c_memq returns #f which
1561 is similar to scm_memq, but different from scm_sloppy_memq's behaviour.
1562
1563 ** New functions: scm_remember_upto_here_1, scm_remember_upto_here_2,
1564 scm_remember_upto_here
1565
1566 These functions replace the function scm_remember.
1567
1568 ** Deprecated function: scm_remember
1569
1570 Use one of the new functions scm_remember_upto_here_1,
1571 scm_remember_upto_here_2 or scm_remember_upto_here instead.
1572
1573 ** New function: scm_allocate_string
1574
1575 This function replaces the function scm_makstr.
1576
1577 ** Deprecated function: scm_makstr
1578
1579 Use the new function scm_allocate_string instead.
1580
1581 ** New global variable scm_gc_running_p introduced.
1582
1583 Use this variable to find out if garbage collection is being executed. Up to
1584 now applications have used scm_gc_heap_lock to test if garbage collection was
1585 running, which also works because of the fact that up to know only the garbage
1586 collector has set this variable. But, this is an implementation detail that
1587 may change. Further, scm_gc_heap_lock is not set throughout gc, thus the use
1588 of this variable is (and has been) not fully safe anyway.
1589
1590 ** New macros: SCM_BITVECTOR_MAX_LENGTH, SCM_UVECTOR_MAX_LENGTH
1591
1592 Use these instead of SCM_LENGTH_MAX.
1593
1594 ** New macros: SCM_CONTINUATION_LENGTH, SCM_CCLO_LENGTH, SCM_STACK_LENGTH,
1595 SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH, SCM_UVECTOR_LENGTH,
1596 SCM_BITVECTOR_LENGTH, SCM_VECTOR_LENGTH.
1597
1598 Use these instead of SCM_LENGTH.
1599
1600 ** New macros: SCM_SET_CONTINUATION_LENGTH, SCM_SET_STRING_LENGTH,
1601 SCM_SET_SYMBOL_LENGTH, SCM_SET_VECTOR_LENGTH, SCM_SET_UVECTOR_LENGTH,
1602 SCM_SET_BITVECTOR_LENGTH
1603
1604 Use these instead of SCM_SETLENGTH
1605
1606 ** New macros: SCM_STRING_CHARS, SCM_SYMBOL_CHARS, SCM_CCLO_BASE,
1607 SCM_VECTOR_BASE, SCM_UVECTOR_BASE, SCM_BITVECTOR_BASE, SCM_COMPLEX_MEM,
1608 SCM_ARRAY_MEM
1609
1610 Use these instead of SCM_CHARS, SCM_UCHARS, SCM_ROCHARS, SCM_ROUCHARS or
1611 SCM_VELTS.
1612
1613 ** New macros: SCM_SET_BIGNUM_BASE, SCM_SET_STRING_CHARS,
1614 SCM_SET_SYMBOL_CHARS, SCM_SET_UVECTOR_BASE, SCM_SET_BITVECTOR_BASE,
1615 SCM_SET_VECTOR_BASE
1616
1617 Use these instead of SCM_SETCHARS.
1618
1619 ** New macro: SCM_BITVECTOR_P
1620
1621 ** New macro: SCM_STRING_COERCE_0TERMINATION_X
1622
1623 Use instead of SCM_COERCE_SUBSTR.
1624
1625 ** New macros: SCM_DIR_OPEN_P, SCM_DIR_FLAG_OPEN
1626
1627 For directory objects, use these instead of SCM_OPDIRP and SCM_OPN.
1628
1629 ** Deprecated macros: SCM_OUTOFRANGE, SCM_NALLOC, SCM_HUP_SIGNAL,
1630 SCM_INT_SIGNAL, SCM_FPE_SIGNAL, SCM_BUS_SIGNAL, SCM_SEGV_SIGNAL,
1631 SCM_ALRM_SIGNAL, SCM_GC_SIGNAL, SCM_TICK_SIGNAL, SCM_SIG_ORD,
1632 SCM_ORD_SIG, SCM_NUM_SIGS, SCM_SYMBOL_SLOTS, SCM_SLOTS, SCM_SLOPPY_STRINGP,
1633 SCM_VALIDATE_STRINGORSUBSTR, SCM_FREEP, SCM_NFREEP, SCM_CHARS, SCM_UCHARS,
1634 SCM_VALIDATE_ROSTRING, SCM_VALIDATE_ROSTRING_COPY,
1635 SCM_VALIDATE_NULLORROSTRING_COPY, SCM_ROLENGTH, SCM_LENGTH, SCM_HUGE_LENGTH,
1636 SCM_SUBSTRP, SCM_SUBSTR_STR, SCM_SUBSTR_OFFSET, SCM_COERCE_SUBSTR,
1637 SCM_ROSTRINGP, SCM_RWSTRINGP, SCM_VALIDATE_RWSTRING, SCM_ROCHARS,
1638 SCM_ROUCHARS, SCM_SETLENGTH, SCM_SETCHARS, SCM_LENGTH_MAX, SCM_GC8MARKP,
1639 SCM_SETGC8MARK, SCM_CLRGC8MARK, SCM_GCTYP16, SCM_GCCDR, SCM_SUBR_DOC,
1640 SCM_OPDIRP, SCM_VALIDATE_OPDIR, SCM_WTA, RETURN_SCM_WTA, SCM_CONST_LONG,
1641 SCM_WNA, SCM_FUNC_NAME, SCM_VALIDATE_NUMBER_COPY,
1642 SCM_VALIDATE_NUMBER_DEF_COPY, SCM_SLOPPY_CONSP, SCM_SLOPPY_NCONSP,
1643 SCM_SETAND_CDR, SCM_SETOR_CDR, SCM_SETAND_CAR, SCM_SETOR_CAR
1644
1645 Use SCM_ASSERT_RANGE or SCM_VALIDATE_XXX_RANGE instead of SCM_OUTOFRANGE.
1646 Use scm_memory_error instead of SCM_NALLOC.
1647 Use SCM_STRINGP instead of SCM_SLOPPY_STRINGP.
1648 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_STRINGORSUBSTR.
1649 Use SCM_FREE_CELL_P instead of SCM_FREEP/SCM_NFREEP
1650 Use a type specific accessor macro instead of SCM_CHARS/SCM_UCHARS.
1651 Use a type specific accessor instead of SCM(_|_RO|_HUGE_)LENGTH.
1652 Use SCM_VALIDATE_(SYMBOL|STRING) instead of SCM_VALIDATE_ROSTRING.
1653 Use SCM_STRING_COERCE_0TERMINATION_X instead of SCM_COERCE_SUBSTR.
1654 Use SCM_STRINGP or SCM_SYMBOLP instead of SCM_ROSTRINGP.
1655 Use SCM_STRINGP instead of SCM_RWSTRINGP.
1656 Use SCM_VALIDATE_STRING instead of SCM_VALIDATE_RWSTRING.
1657 Use SCM_STRING_CHARS instead of SCM_ROCHARS.
1658 Use SCM_STRING_UCHARS instead of SCM_ROUCHARS.
1659 Use a type specific setter macro instead of SCM_SETLENGTH.
1660 Use a type specific setter macro instead of SCM_SETCHARS.
1661 Use a type specific length macro instead of SCM_LENGTH_MAX.
1662 Use SCM_GCMARKP instead of SCM_GC8MARKP.
1663 Use SCM_SETGCMARK instead of SCM_SETGC8MARK.
1664 Use SCM_CLRGCMARK instead of SCM_CLRGC8MARK.
1665 Use SCM_TYP16 instead of SCM_GCTYP16.
1666 Use SCM_CDR instead of SCM_GCCDR.
1667 Use SCM_DIR_OPEN_P instead of SCM_OPDIRP.
1668 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of SCM_WTA.
1669 Use SCM_MISC_ERROR or SCM_WRONG_TYPE_ARG instead of RETURN_SCM_WTA.
1670 Use SCM_VCELL_INIT instead of SCM_CONST_LONG.
1671 Use SCM_WRONG_NUM_ARGS instead of SCM_WNA.
1672 Use SCM_CONSP instead of SCM_SLOPPY_CONSP.
1673 Use !SCM_CONSP instead of SCM_SLOPPY_NCONSP.
1674
1675 ** Removed function: scm_struct_init
1676
1677 ** Removed variable: scm_symhash_dim
1678
1679 ** Renamed function: scm_make_cont has been replaced by
1680 scm_make_continuation, which has a different interface.
1681
1682 ** Deprecated function: scm_call_catching_errors
1683
1684 Use scm_catch or scm_lazy_catch from throw.[ch] instead.
1685
1686 ** Deprecated function: scm_strhash
1687
1688 Use scm_string_hash instead.
1689
1690 ** Deprecated function: scm_vector_set_length_x
1691
1692 Instead, create a fresh vector of the desired size and copy the contents.
1693
1694 ** scm_gensym has changed prototype
1695
1696 scm_gensym now only takes one argument.
1697
1698 ** Deprecated type tags: scm_tc7_ssymbol, scm_tc7_msymbol, scm_tcs_symbols,
1699 scm_tc7_lvector
1700
1701 There is now only a single symbol type scm_tc7_symbol.
1702 The tag scm_tc7_lvector was not used anyway.
1703
1704 ** Deprecated function: scm_make_smob_type_mfpe, scm_set_smob_mfpe.
1705
1706 Use scm_make_smob_type and scm_set_smob_XXX instead.
1707
1708 ** New function scm_set_smob_apply.
1709
1710 This can be used to set an apply function to a smob type.
1711
1712 ** Deprecated function: scm_strprint_obj
1713
1714 Use scm_object_to_string instead.
1715
1716 ** Deprecated function: scm_wta
1717
1718 Use scm_wrong_type_arg, or another appropriate error signalling function
1719 instead.
1720
1721 ** Explicit support for obarrays has been deprecated.
1722
1723 Use `scm_str2symbol' and the generic hashtable functions instead.
1724
1725 ** The concept of `vcells' has been deprecated.
1726
1727 The data type `variable' is now used exclusively. `Vcells' have been
1728 a low-level concept so you are likely not affected by this change.
1729
1730 *** Deprecated functions: scm_sym2vcell, scm_sysintern,
1731 scm_sysintern0, scm_symbol_value0, scm_intern, scm_intern0.
1732
1733 Use scm_c_define or scm_c_lookup instead, as appropriate.
1734
1735 *** New functions: scm_c_module_lookup, scm_c_lookup,
1736 scm_c_module_define, scm_c_define, scm_module_lookup, scm_lookup,
1737 scm_module_define, scm_define.
1738
1739 These functions work with variables instead of with vcells.
1740
1741 ** New functions for creating and defining `subr's and `gsubr's.
1742
1743 The new functions more clearly distinguish between creating a subr (or
1744 gsubr) object and adding it to the current module.
1745
1746 These new functions are available: scm_c_make_subr, scm_c_define_subr,
1747 scm_c_make_subr_with_generic, scm_c_define_subr_with_generic,
1748 scm_c_make_gsubr, scm_c_define_gsubr, scm_c_make_gsubr_with_generic,
1749 scm_c_define_gsubr_with_generic.
1750
1751 ** Deprecated functions: scm_make_subr, scm_make_subr_opt,
1752 scm_make_subr_with_generic, scm_make_gsubr,
1753 scm_make_gsubr_with_generic.
1754
1755 Use the new ones from above instead.
1756
1757 ** C interface to the module system has changed.
1758
1759 While we suggest that you avoid as many explicit module system
1760 operations from C as possible for the time being, the C interface has
1761 been made more similar to the high-level Scheme module system.
1762
1763 *** New functions: scm_c_define_module, scm_c_use_module,
1764 scm_c_export, scm_c_resolve_module.
1765
1766 They mostly work like their Scheme namesakes. scm_c_define_module
1767 takes a function that is called a context where the new module is
1768 current.
1769
1770 *** Deprecated functions: scm_the_root_module, scm_make_module,
1771 scm_ensure_user_module, scm_load_scheme_module.
1772
1773 Use the new functions instead.
1774
1775 ** Renamed function: scm_internal_with_fluids becomes
1776 scm_c_with_fluids.
1777
1778 scm_internal_with_fluids is available as a deprecated function.
1779
1780 ** New function: scm_c_with_fluid.
1781
1782 Just like scm_c_with_fluids, but takes one fluid and one value instead
1783 of lists of same.
1784
1785 ** Deprecated typedefs: long_long, ulong_long.
1786
1787 They are of questionable utility and they pollute the global
1788 namespace.
1789
1790 ** Deprecated typedef: scm_sizet
1791
1792 It is of questionable utility now that Guile requires ANSI C, and is
1793 oddly named.
1794
1795 ** Deprecated typedefs: scm_port_rw_active, scm_port,
1796 scm_ptob_descriptor, scm_debug_info, scm_debug_frame, scm_fport,
1797 scm_option, scm_rstate, scm_rng, scm_array, scm_array_dim.
1798
1799 Made more compliant with the naming policy by adding a _t at the end.
1800
1801 ** Deprecated functions: scm_mkbig, scm_big2num, scm_adjbig,
1802 scm_normbig, scm_copybig, scm_2ulong2big, scm_dbl2big, scm_big2dbl
1803
1804 With the exception of the mysterious scm_2ulong2big, they are still
1805 available under new names (scm_i_mkbig etc). These functions are not
1806 intended to be used in user code. You should avoid dealing with
1807 bignums directly, and should deal with numbers in general (which can
1808 be bignums).
1809
1810 ** Change in behavior: scm_num2long, scm_num2ulong
1811
1812 The scm_num2[u]long functions don't any longer accept an inexact
1813 argument. This change in behavior is motivated by concordance with
1814 R5RS: It is more common that a primitive doesn't want to accept an
1815 inexact for an exact.
1816
1817 ** New functions: scm_short2num, scm_ushort2num, scm_int2num,
1818 scm_uint2num, scm_size2num, scm_ptrdiff2num, scm_num2short,
1819 scm_num2ushort, scm_num2int, scm_num2uint, scm_num2ptrdiff,
1820 scm_num2size.
1821
1822 These are conversion functions between the various ANSI C integral
1823 types and Scheme numbers. NOTE: The scm_num2xxx functions don't
1824 accept an inexact argument.
1825
1826 ** New functions: scm_float2num, scm_double2num,
1827 scm_num2float, scm_num2double.
1828
1829 These are conversion functions between the two ANSI C float types and
1830 Scheme numbers.
1831
1832 ** New number validation macros:
1833 SCM_NUM2{SIZE,PTRDIFF,SHORT,USHORT,INT,UINT}[_DEF]
1834
1835 See above.
1836
1837 ** New functions: scm_gc_protect_object, scm_gc_unprotect_object
1838
1839 These are just nicer-named old scm_protect_object and
1840 scm_unprotect_object.
1841
1842 ** Deprecated functions: scm_protect_object, scm_unprotect_object
1843
1844 ** New functions: scm_gc_[un]register_root, scm_gc_[un]register_roots
1845
1846 These functions can be used to register pointers to locations that
1847 hold SCM values.
1848
1849 ** Deprecated function: scm_create_hook.
1850
1851 Its sins are: misleading name, non-modularity and lack of general
1852 usefulness.
1853
1854 \f
1855 Changes since Guile 1.3.4:
1856
1857 * Changes to the distribution
1858
1859 ** Trees from nightly snapshots and CVS now require you to run autogen.sh.
1860
1861 We've changed the way we handle generated files in the Guile source
1862 repository. As a result, the procedure for building trees obtained
1863 from the nightly FTP snapshots or via CVS has changed:
1864 - You must have appropriate versions of autoconf, automake, and
1865 libtool installed on your system. See README for info on how to
1866 obtain these programs.
1867 - Before configuring the tree, you must first run the script
1868 `autogen.sh' at the top of the source tree.
1869
1870 The Guile repository used to contain not only source files, written by
1871 humans, but also some generated files, like configure scripts and
1872 Makefile.in files. Even though the contents of these files could be
1873 derived mechanically from other files present, we thought it would
1874 make the tree easier to build if we checked them into CVS.
1875
1876 However, this approach means that minor differences between
1877 developer's installed tools and habits affected the whole team.
1878 So we have removed the generated files from the repository, and
1879 added the autogen.sh script, which will reconstruct them
1880 appropriately.
1881
1882
1883 ** configure now has experimental options to remove support for certain
1884 features:
1885
1886 --disable-arrays omit array and uniform array support
1887 --disable-posix omit posix interfaces
1888 --disable-networking omit networking interfaces
1889 --disable-regex omit regular expression interfaces
1890
1891 These are likely to become separate modules some day.
1892
1893 ** New configure option --enable-debug-freelist
1894
1895 This enables a debugging version of SCM_NEWCELL(), and also registers
1896 an extra primitive, the setter `gc-set-debug-check-freelist!'.
1897
1898 Configure with the --enable-debug-freelist option to enable
1899 the gc-set-debug-check-freelist! primitive, and then use:
1900
1901 (gc-set-debug-check-freelist! #t) # turn on checking of the freelist
1902 (gc-set-debug-check-freelist! #f) # turn off checking
1903
1904 Checking of the freelist forces a traversal of the freelist and
1905 a garbage collection before each allocation of a cell. This can
1906 slow down the interpreter dramatically, so the setter should be used to
1907 turn on this extra processing only when necessary.
1908
1909 ** New configure option --enable-debug-malloc
1910
1911 Include code for debugging of calls to scm_must_malloc/realloc/free.
1912
1913 Checks that
1914
1915 1. objects freed by scm_must_free has been mallocated by scm_must_malloc
1916 2. objects reallocated by scm_must_realloc has been allocated by
1917 scm_must_malloc
1918 3. reallocated objects are reallocated with the same what string
1919
1920 But, most importantly, it records the number of allocated objects of
1921 each kind. This is useful when searching for memory leaks.
1922
1923 A Guile compiled with this option provides the primitive
1924 `malloc-stats' which returns an alist with pairs of kind and the
1925 number of objects of that kind.
1926
1927 ** All includes are now referenced relative to the root directory
1928
1929 Since some users have had problems with mixups between Guile and
1930 system headers, we have decided to always refer to Guile headers via
1931 their parent directories. This essentially creates a "private name
1932 space" for Guile headers. This means that the compiler only is given
1933 -I options for the root build and root source directory.
1934
1935 ** Header files kw.h and genio.h have been removed.
1936
1937 ** The module (ice-9 getopt-gnu-style) has been removed.
1938
1939 ** New module (ice-9 documentation)
1940
1941 Implements the interface to documentation strings associated with
1942 objects.
1943
1944 ** New module (ice-9 time)
1945
1946 Provides a macro `time', which displays execution time of a given form.
1947
1948 ** New module (ice-9 history)
1949
1950 Loading this module enables value history in the repl.
1951
1952 * Changes to the stand-alone interpreter
1953
1954 ** New command line option --debug
1955
1956 Start Guile with debugging evaluator and backtraces enabled.
1957
1958 This is useful when debugging your .guile init file or scripts.
1959
1960 ** New help facility
1961
1962 Usage: (help NAME) gives documentation about objects named NAME (a symbol)
1963 (help REGEXP) ditto for objects with names matching REGEXP (a string)
1964 (help 'NAME) gives documentation for NAME, even if it is not an object
1965 (help ,EXPR) gives documentation for object returned by EXPR
1966 (help (my module)) gives module commentary for `(my module)'
1967 (help) gives this text
1968
1969 `help' searches among bindings exported from loaded modules, while
1970 `apropos' searches among bindings visible from the "current" module.
1971
1972 Examples: (help help)
1973 (help cons)
1974 (help "output-string")
1975
1976 ** `help' and `apropos' now prints full module names
1977
1978 ** Dynamic linking now uses libltdl from the libtool package.
1979
1980 The old system dependent code for doing dynamic linking has been
1981 replaced with calls to the libltdl functions which do all the hairy
1982 details for us.
1983
1984 The major improvement is that you can now directly pass libtool
1985 library names like "libfoo.la" to `dynamic-link' and `dynamic-link'
1986 will be able to do the best shared library job you can get, via
1987 libltdl.
1988
1989 The way dynamic libraries are found has changed and is not really
1990 portable across platforms, probably. It is therefore recommended to
1991 use absolute filenames when possible.
1992
1993 If you pass a filename without an extension to `dynamic-link', it will
1994 try a few appropriate ones. Thus, the most platform ignorant way is
1995 to specify a name like "libfoo", without any directories and
1996 extensions.
1997
1998 ** Guile COOP threads are now compatible with LinuxThreads
1999
2000 Previously, COOP threading wasn't possible in applications linked with
2001 Linux POSIX threads due to their use of the stack pointer to find the
2002 thread context. This has now been fixed with a workaround which uses
2003 the pthreads to allocate the stack.
2004
2005 ** New primitives: `pkgdata-dir', `site-dir', `library-dir'
2006
2007 ** Positions of erring expression in scripts
2008
2009 With version 1.3.4, the location of the erring expression in Guile
2010 scipts is no longer automatically reported. (This should have been
2011 documented before the 1.3.4 release.)
2012
2013 You can get this information by enabling recording of positions of
2014 source expressions and running the debugging evaluator. Put this at
2015 the top of your script (or in your "site" file):
2016
2017 (read-enable 'positions)
2018 (debug-enable 'debug)
2019
2020 ** Backtraces in scripts
2021
2022 It is now possible to get backtraces in scripts.
2023
2024 Put
2025
2026 (debug-enable 'debug 'backtrace)
2027
2028 at the top of the script.
2029
2030 (The first options enables the debugging evaluator.
2031 The second enables backtraces.)
2032
2033 ** Part of module system symbol lookup now implemented in C
2034
2035 The eval closure of most modules is now implemented in C. Since this
2036 was one of the bottlenecks for loading speed, Guile now loads code
2037 substantially faster than before.
2038
2039 ** Attempting to get the value of an unbound variable now produces
2040 an exception with a key of 'unbound-variable instead of 'misc-error.
2041
2042 ** The initial default output port is now unbuffered if it's using a
2043 tty device. Previously in this situation it was line-buffered.
2044
2045 ** New hook: after-gc-hook
2046
2047 after-gc-hook takes over the role of gc-thunk. This hook is run at
2048 the first SCM_TICK after a GC. (Thus, the code is run at the same
2049 point during evaluation as signal handlers.)
2050
2051 Note that this hook should be used only for diagnostic and debugging
2052 purposes. It is not certain that it will continue to be well-defined
2053 when this hook is run in the future.
2054
2055 C programmers: Note the new C level hooks scm_before_gc_c_hook,
2056 scm_before_sweep_c_hook, scm_after_gc_c_hook.
2057
2058 ** Improvements to garbage collector
2059
2060 Guile 1.4 has a new policy for triggering heap allocation and
2061 determining the sizes of heap segments. It fixes a number of problems
2062 in the old GC.
2063
2064 1. The new policy can handle two separate pools of cells
2065 (2-word/4-word) better. (The old policy would run wild, allocating
2066 more and more memory for certain programs.)
2067
2068 2. The old code would sometimes allocate far too much heap so that the
2069 Guile process became gigantic. The new code avoids this.
2070
2071 3. The old code would sometimes allocate too little so that few cells
2072 were freed at GC so that, in turn, too much time was spent in GC.
2073
2074 4. The old code would often trigger heap allocation several times in a
2075 row. (The new scheme predicts how large the segments needs to be
2076 in order not to need further allocation.)
2077
2078 All in all, the new GC policy will make larger applications more
2079 efficient.
2080
2081 The new GC scheme also is prepared for POSIX threading. Threads can
2082 allocate private pools of cells ("clusters") with just a single
2083 function call. Allocation of single cells from such a cluster can
2084 then proceed without any need of inter-thread synchronization.
2085
2086 ** New environment variables controlling GC parameters
2087
2088 GUILE_MAX_SEGMENT_SIZE Maximal segment size
2089 (default = 2097000)
2090
2091 Allocation of 2-word cell heaps:
2092
2093 GUILE_INIT_SEGMENT_SIZE_1 Size of initial heap segment in bytes
2094 (default = 360000)
2095
2096 GUILE_MIN_YIELD_1 Minimum number of freed cells at each
2097 GC in percent of total heap size
2098 (default = 40)
2099
2100 Allocation of 4-word cell heaps
2101 (used for real numbers and misc other objects):
2102
2103 GUILE_INIT_SEGMENT_SIZE_2, GUILE_MIN_YIELD_2
2104
2105 (See entry "Way for application to customize GC parameters" under
2106 section "Changes to the scm_ interface" below.)
2107
2108 ** Guile now implements reals using 4-word cells
2109
2110 This speeds up computation with reals. (They were earlier allocated
2111 with `malloc'.) There is still some room for optimizations, however.
2112
2113 ** Some further steps toward POSIX thread support have been taken
2114
2115 *** Guile's critical sections (SCM_DEFER/ALLOW_INTS)
2116 don't have much effect any longer, and many of them will be removed in
2117 next release.
2118
2119 *** Signals
2120 are only handled at the top of the evaluator loop, immediately after
2121 I/O, and in scm_equalp.
2122
2123 *** The GC can allocate thread private pools of pairs.
2124
2125 * Changes to Scheme functions and syntax
2126
2127 ** close-input-port and close-output-port are now R5RS
2128
2129 These procedures have been turned into primitives and have R5RS behaviour.
2130
2131 ** New procedure: simple-format PORT MESSAGE ARG1 ...
2132
2133 (ice-9 boot) makes `format' an alias for `simple-format' until possibly
2134 extended by the more sophisticated version in (ice-9 format)
2135
2136 (simple-format port message . args)
2137 Write MESSAGE to DESTINATION, defaulting to `current-output-port'.
2138 MESSAGE can contain ~A (was %s) and ~S (was %S) escapes. When printed,
2139 the escapes are replaced with corresponding members of ARGS:
2140 ~A formats using `display' and ~S formats using `write'.
2141 If DESTINATION is #t, then use the `current-output-port',
2142 if DESTINATION is #f, then return a string containing the formatted text.
2143 Does not add a trailing newline."
2144
2145 ** string-ref: the second argument is no longer optional.
2146
2147 ** string, list->string: no longer accept strings in their arguments,
2148 only characters, for compatibility with R5RS.
2149
2150 ** New procedure: port-closed? PORT
2151 Returns #t if PORT is closed or #f if it is open.
2152
2153 ** Deprecated: list*
2154
2155 The list* functionality is now provided by cons* (SRFI-1 compliant)
2156
2157 ** New procedure: cons* ARG1 ARG2 ... ARGn
2158
2159 Like `list', but the last arg provides the tail of the constructed list,
2160 returning (cons ARG1 (cons ARG2 (cons ... ARGn))).
2161
2162 Requires at least one argument. If given one argument, that argument
2163 is returned as result.
2164
2165 This function is called `list*' in some other Schemes and in Common LISP.
2166
2167 ** Removed deprecated: serial-map, serial-array-copy!, serial-array-map!
2168
2169 ** New procedure: object-documentation OBJECT
2170
2171 Returns the documentation string associated with OBJECT. The
2172 procedure uses a caching mechanism so that subsequent lookups are
2173 faster.
2174
2175 Exported by (ice-9 documentation).
2176
2177 ** module-name now returns full names of modules
2178
2179 Previously, only the last part of the name was returned (`session' for
2180 `(ice-9 session)'). Ex: `(ice-9 session)'.
2181
2182 * Changes to the gh_ interface
2183
2184 ** Deprecated: gh_int2scmb
2185
2186 Use gh_bool2scm instead.
2187
2188 * Changes to the scm_ interface
2189
2190 ** Guile primitives now carry docstrings!
2191
2192 Thanks to Greg Badros!
2193
2194 ** Guile primitives are defined in a new way: SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2195
2196 Now Guile primitives are defined using the SCM_DEFINE/SCM_DEFINE1/SCM_PROC
2197 macros and must contain a docstring that is extracted into foo.doc using a new
2198 guile-doc-snarf script (that uses guile-doc-snarf.awk).
2199
2200 However, a major overhaul of these macros is scheduled for the next release of
2201 guile.
2202
2203 ** Guile primitives use a new technique for validation of arguments
2204
2205 SCM_VALIDATE_* macros are defined to ease the redundancy and improve
2206 the readability of argument checking.
2207
2208 ** All (nearly?) K&R prototypes for functions replaced with ANSI C equivalents.
2209
2210 ** New macros: SCM_PACK, SCM_UNPACK
2211
2212 Compose/decompose an SCM value.
2213
2214 The SCM type is now treated as an abstract data type and may be defined as a
2215 long, a void* or as a struct, depending on the architecture and compile time
2216 options. This makes it easier to find several types of bugs, for example when
2217 SCM values are treated as integers without conversion. Values of the SCM type
2218 should be treated as "atomic" values. These macros are used when
2219 composing/decomposing an SCM value, either because you want to access
2220 individual bits, or because you want to treat it as an integer value.
2221
2222 E.g., in order to set bit 7 in an SCM value x, use the expression
2223
2224 SCM_PACK (SCM_UNPACK (x) | 0x80)
2225
2226 ** The name property of hooks is deprecated.
2227 Thus, the use of SCM_HOOK_NAME and scm_make_hook_with_name is deprecated.
2228
2229 You can emulate this feature by using object properties.
2230
2231 ** Deprecated macros: SCM_INPORTP, SCM_OUTPORTP, SCM_CRDY, SCM_ICHRP,
2232 SCM_ICHR, SCM_MAKICHR, SCM_SETJMPBUF, SCM_NSTRINGP, SCM_NRWSTRINGP,
2233 SCM_NVECTORP
2234
2235 These macros will be removed in a future release of Guile.
2236
2237 ** The following types, functions and macros from numbers.h are deprecated:
2238 scm_dblproc, SCM_UNEGFIXABLE, SCM_FLOBUFLEN, SCM_INEXP, SCM_CPLXP, SCM_REAL,
2239 SCM_IMAG, SCM_REALPART, scm_makdbl, SCM_SINGP, SCM_NUM2DBL, SCM_NO_BIGDIG
2240
2241 Further, it is recommended not to rely on implementation details for guile's
2242 current implementation of bignums. It is planned to replace this
2243 implementation with gmp in the future.
2244
2245 ** Port internals: the rw_random variable in the scm_port structure
2246 must be set to non-zero in any random access port. In recent Guile
2247 releases it was only set for bidirectional random-access ports.
2248
2249 ** Port internals: the seek ptob procedure is now responsible for
2250 resetting the buffers if required. The change was made so that in the
2251 special case of reading the current position (i.e., seek p 0 SEEK_CUR)
2252 the fport and strport ptobs can avoid resetting the buffers,
2253 in particular to avoid discarding unread chars. An existing port
2254 type can be fixed by adding something like the following to the
2255 beginning of the ptob seek procedure:
2256
2257 if (pt->rw_active == SCM_PORT_READ)
2258 scm_end_input (object);
2259 else if (pt->rw_active == SCM_PORT_WRITE)
2260 ptob->flush (object);
2261
2262 although to actually avoid resetting the buffers and discard unread
2263 chars requires further hacking that depends on the characteristics
2264 of the ptob.
2265
2266 ** Deprecated functions: scm_fseek, scm_tag
2267
2268 These functions are no longer used and will be removed in a future version.
2269
2270 ** The scm_sysmissing procedure is no longer used in libguile.
2271 Unless it turns out to be unexpectedly useful to somebody, it will be
2272 removed in a future version.
2273
2274 ** The format of error message strings has changed
2275
2276 The two C procedures: scm_display_error and scm_error, as well as the
2277 primitive `scm-error', now use scm_simple_format to do their work.
2278 This means that the message strings of all code must be updated to use
2279 ~A where %s was used before, and ~S where %S was used before.
2280
2281 During the period when there still are a lot of old Guiles out there,
2282 you might want to support both old and new versions of Guile.
2283
2284 There are basically two methods to achieve this. Both methods use
2285 autoconf. Put
2286
2287 AC_CHECK_FUNCS(scm_simple_format)
2288
2289 in your configure.in.
2290
2291 Method 1: Use the string concatenation features of ANSI C's
2292 preprocessor.
2293
2294 In C:
2295
2296 #ifdef HAVE_SCM_SIMPLE_FORMAT
2297 #define FMT_S "~S"
2298 #else
2299 #define FMT_S "%S"
2300 #endif
2301
2302 Then represent each of your error messages using a preprocessor macro:
2303
2304 #define E_SPIDER_ERROR "There's a spider in your " ## FMT_S ## "!!!"
2305
2306 In Scheme:
2307
2308 (define fmt-s (if (defined? 'simple-format) "~S" "%S"))
2309 (define make-message string-append)
2310
2311 (define e-spider-error (make-message "There's a spider in your " fmt-s "!!!"))
2312
2313 Method 2: Use the oldfmt function found in doc/oldfmt.c.
2314
2315 In C:
2316
2317 scm_misc_error ("picnic", scm_c_oldfmt0 ("There's a spider in your ~S!!!"),
2318 ...);
2319
2320 In Scheme:
2321
2322 (scm-error 'misc-error "picnic" (oldfmt "There's a spider in your ~S!!!")
2323 ...)
2324
2325
2326 ** Deprecated: coop_mutex_init, coop_condition_variable_init
2327
2328 Don't use the functions coop_mutex_init and
2329 coop_condition_variable_init. They will change.
2330
2331 Use scm_mutex_init and scm_cond_init instead.
2332
2333 ** New function: int scm_cond_timedwait (scm_cond_t *COND, scm_mutex_t *MUTEX, const struct timespec *ABSTIME)
2334 `scm_cond_timedwait' atomically unlocks MUTEX and waits on
2335 COND, as `scm_cond_wait' does, but it also bounds the duration
2336 of the wait. If COND has not been signaled before time ABSTIME,
2337 the mutex MUTEX is re-acquired and `scm_cond_timedwait'
2338 returns the error code `ETIMEDOUT'.
2339
2340 The ABSTIME parameter specifies an absolute time, with the same
2341 origin as `time' and `gettimeofday': an ABSTIME of 0 corresponds
2342 to 00:00:00 GMT, January 1, 1970.
2343
2344 ** New function: scm_cond_broadcast (scm_cond_t *COND)
2345 `scm_cond_broadcast' restarts all the threads that are waiting
2346 on the condition variable COND. Nothing happens if no threads are
2347 waiting on COND.
2348
2349 ** New function: scm_key_create (scm_key_t *KEY, void (*destr_function) (void *))
2350 `scm_key_create' allocates a new TSD key. The key is stored in
2351 the location pointed to by KEY. There is no limit on the number
2352 of keys allocated at a given time. The value initially associated
2353 with the returned key is `NULL' in all currently executing threads.
2354
2355 The DESTR_FUNCTION argument, if not `NULL', specifies a destructor
2356 function associated with the key. When a thread terminates,
2357 DESTR_FUNCTION is called on the value associated with the key in
2358 that thread. The DESTR_FUNCTION is not called if a key is deleted
2359 with `scm_key_delete' or a value is changed with
2360 `scm_setspecific'. The order in which destructor functions are
2361 called at thread termination time is unspecified.
2362
2363 Destructors are not yet implemented.
2364
2365 ** New function: scm_setspecific (scm_key_t KEY, const void *POINTER)
2366 `scm_setspecific' changes the value associated with KEY in the
2367 calling thread, storing the given POINTER instead.
2368
2369 ** New function: scm_getspecific (scm_key_t KEY)
2370 `scm_getspecific' returns the value currently associated with
2371 KEY in the calling thread.
2372
2373 ** New function: scm_key_delete (scm_key_t KEY)
2374 `scm_key_delete' deallocates a TSD key. It does not check
2375 whether non-`NULL' values are associated with that key in the
2376 currently executing threads, nor call the destructor function
2377 associated with the key.
2378
2379 ** New function: scm_c_hook_init (scm_c_hook_t *HOOK, void *HOOK_DATA, scm_c_hook_type_t TYPE)
2380
2381 Initialize a C level hook HOOK with associated HOOK_DATA and type
2382 TYPE. (See scm_c_hook_run ().)
2383
2384 ** New function: scm_c_hook_add (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA, int APPENDP)
2385
2386 Add hook function FUNC with associated FUNC_DATA to HOOK. If APPENDP
2387 is true, add it last, otherwise first. The same FUNC can be added
2388 multiple times if FUNC_DATA differ and vice versa.
2389
2390 ** New function: scm_c_hook_remove (scm_c_hook_t *HOOK, scm_c_hook_function_t FUNC, void *FUNC_DATA)
2391
2392 Remove hook function FUNC with associated FUNC_DATA from HOOK. A
2393 function is only removed if both FUNC and FUNC_DATA matches.
2394
2395 ** New function: void *scm_c_hook_run (scm_c_hook_t *HOOK, void *DATA)
2396
2397 Run hook HOOK passing DATA to the hook functions.
2398
2399 If TYPE is SCM_C_HOOK_NORMAL, all hook functions are run. The value
2400 returned is undefined.
2401
2402 If TYPE is SCM_C_HOOK_OR, hook functions are run until a function
2403 returns a non-NULL value. This value is returned as the result of
2404 scm_c_hook_run. If all functions return NULL, NULL is returned.
2405
2406 If TYPE is SCM_C_HOOK_AND, hook functions are run until a function
2407 returns a NULL value, and NULL is returned. If all functions returns
2408 a non-NULL value, the last value is returned.
2409
2410 ** New C level GC hooks
2411
2412 Five new C level hooks has been added to the garbage collector.
2413
2414 scm_before_gc_c_hook
2415 scm_after_gc_c_hook
2416
2417 are run before locking and after unlocking the heap. The system is
2418 thus in a mode where evaluation can take place. (Except that
2419 scm_before_gc_c_hook must not allocate new cells.)
2420
2421 scm_before_mark_c_hook
2422 scm_before_sweep_c_hook
2423 scm_after_sweep_c_hook
2424
2425 are run when the heap is locked. These are intended for extension of
2426 the GC in a modular fashion. Examples are the weaks and guardians
2427 modules.
2428
2429 ** Way for application to customize GC parameters
2430
2431 The application can set up other default values for the GC heap
2432 allocation parameters
2433
2434 GUILE_INIT_HEAP_SIZE_1, GUILE_MIN_YIELD_1,
2435 GUILE_INIT_HEAP_SIZE_2, GUILE_MIN_YIELD_2,
2436 GUILE_MAX_SEGMENT_SIZE,
2437
2438 by setting
2439
2440 scm_default_init_heap_size_1, scm_default_min_yield_1,
2441 scm_default_init_heap_size_2, scm_default_min_yield_2,
2442 scm_default_max_segment_size
2443
2444 respectively before callong scm_boot_guile.
2445
2446 (See entry "New environment variables ..." in section
2447 "Changes to the stand-alone interpreter" above.)
2448
2449 ** scm_protect_object/scm_unprotect_object now nest
2450
2451 This means that you can call scm_protect_object multiple times on an
2452 object and count on the object being protected until
2453 scm_unprotect_object has been call the same number of times.
2454
2455 The functions also have better time complexity.
2456
2457 Still, it is usually possible to structure the application in a way
2458 that you don't need to use these functions. For example, if you use a
2459 protected standard Guile list to keep track of live objects rather
2460 than some custom data type, objects will die a natural death when they
2461 are no longer needed.
2462
2463 ** Deprecated type tags: scm_tc16_flo, scm_tc_flo, scm_tc_dblr, scm_tc_dblc
2464
2465 Guile does not provide the float representation for inexact real numbers any
2466 more. Now, only doubles are used to represent inexact real numbers. Further,
2467 the tag names scm_tc_dblr and scm_tc_dblc have been changed to scm_tc16_real
2468 and scm_tc16_complex, respectively.
2469
2470 ** Removed deprecated type scm_smobfuns
2471
2472 ** Removed deprecated function scm_newsmob
2473
2474 ** Warning: scm_make_smob_type_mfpe might become deprecated in a future release
2475
2476 There is an ongoing discussion among the developers whether to
2477 deprecate `scm_make_smob_type_mfpe' or not. Please use the current
2478 standard interface (scm_make_smob_type, scm_set_smob_XXX) in new code
2479 until this issue has been settled.
2480
2481 ** Removed deprecated type tag scm_tc16_kw
2482
2483 ** Added type tag scm_tc16_keyword
2484
2485 (This was introduced already in release 1.3.4 but was not documented
2486 until now.)
2487
2488 ** gdb_print now prints "*** Guile not initialized ***" until Guile initialized
2489
2490 * Changes to system call interfaces:
2491
2492 ** The "select" procedure now tests port buffers for the ability to
2493 provide input or accept output. Previously only the underlying file
2494 descriptors were checked.
2495
2496 ** New variable PIPE_BUF: the maximum number of bytes that can be
2497 atomically written to a pipe.
2498
2499 ** If a facility is not available on the system when Guile is
2500 compiled, the corresponding primitive procedure will not be defined.
2501 Previously it would have been defined but would throw a system-error
2502 exception if called. Exception handlers which catch this case may
2503 need minor modification: an error will be thrown with key
2504 'unbound-variable instead of 'system-error. Alternatively it's
2505 now possible to use `defined?' to check whether the facility is
2506 available.
2507
2508 ** Procedures which depend on the timezone should now give the correct
2509 result on systems which cache the TZ environment variable, even if TZ
2510 is changed without calling tzset.
2511
2512 * Changes to the networking interfaces:
2513
2514 ** New functions: htons, ntohs, htonl, ntohl: for converting short and
2515 long integers between network and host format. For now, it's not
2516 particularly convenient to do this kind of thing, but consider:
2517
2518 (define write-network-long
2519 (lambda (value port)
2520 (let ((v (make-uniform-vector 1 1 0)))
2521 (uniform-vector-set! v 0 (htonl value))
2522 (uniform-vector-write v port))))
2523
2524 (define read-network-long
2525 (lambda (port)
2526 (let ((v (make-uniform-vector 1 1 0)))
2527 (uniform-vector-read! v port)
2528 (ntohl (uniform-vector-ref v 0)))))
2529
2530 ** If inet-aton fails, it now throws an error with key 'misc-error
2531 instead of 'system-error, since errno is not relevant.
2532
2533 ** Certain gethostbyname/gethostbyaddr failures now throw errors with
2534 specific keys instead of 'system-error. The latter is inappropriate
2535 since errno will not have been set. The keys are:
2536 'host-not-found, 'try-again, 'no-recovery and 'no-data.
2537
2538 ** sethostent, setnetent, setprotoent, setservent: now take an
2539 optional argument STAYOPEN, which specifies whether the database
2540 remains open after a database entry is accessed randomly (e.g., using
2541 gethostbyname for the hosts database.) The default is #f. Previously
2542 #t was always used.
2543
2544 \f
2545 Changes since Guile 1.3.2:
2546
2547 * Changes to the stand-alone interpreter
2548
2549 ** Debugger
2550
2551 An initial version of the Guile debugger written by Chris Hanson has
2552 been added. The debugger is still under development but is included
2553 in the distribution anyway since it is already quite useful.
2554
2555 Type
2556
2557 (debug)
2558
2559 after an error to enter the debugger. Type `help' inside the debugger
2560 for a description of available commands.
2561
2562 If you prefer to have stack frames numbered and printed in
2563 anti-chronological order and prefer up in the stack to be down on the
2564 screen as is the case in gdb, you can put
2565
2566 (debug-enable 'backwards)
2567
2568 in your .guile startup file. (However, this means that Guile can't
2569 use indentation to indicate stack level.)
2570
2571 The debugger is autoloaded into Guile at the first use.
2572
2573 ** Further enhancements to backtraces
2574
2575 There is a new debug option `width' which controls the maximum width
2576 on the screen of printed stack frames. Fancy printing parameters
2577 ("level" and "length" as in Common LISP) are adaptively adjusted for
2578 each stack frame to give maximum information while still fitting
2579 within the bounds. If the stack frame can't be made to fit by
2580 adjusting parameters, it is simply cut off at the end. This is marked
2581 with a `$'.
2582
2583 ** Some modules are now only loaded when the repl is started
2584
2585 The modules (ice-9 debug), (ice-9 session), (ice-9 threads) and (ice-9
2586 regex) are now loaded into (guile-user) only if the repl has been
2587 started. The effect is that the startup time for scripts has been
2588 reduced to 30% of what it was previously.
2589
2590 Correctly written scripts load the modules they require at the top of
2591 the file and should not be affected by this change.
2592
2593 ** Hooks are now represented as smobs
2594
2595 * Changes to Scheme functions and syntax
2596
2597 ** Readline support has changed again.
2598
2599 The old (readline-activator) module is gone. Use (ice-9 readline)
2600 instead, which now contains all readline functionality. So the code
2601 to activate readline is now
2602
2603 (use-modules (ice-9 readline))
2604 (activate-readline)
2605
2606 This should work at any time, including from the guile prompt.
2607
2608 To avoid confusion about the terms of Guile's license, please only
2609 enable readline for your personal use; please don't make it the
2610 default for others. Here is why we make this rather odd-sounding
2611 request:
2612
2613 Guile is normally licensed under a weakened form of the GNU General
2614 Public License, which allows you to link code with Guile without
2615 placing that code under the GPL. This exception is important to some
2616 people.
2617
2618 However, since readline is distributed under the GNU General Public
2619 License, when you link Guile with readline, either statically or
2620 dynamically, you effectively change Guile's license to the strict GPL.
2621 Whenever you link any strictly GPL'd code into Guile, uses of Guile
2622 which are normally permitted become forbidden. This is a rather
2623 non-obvious consequence of the licensing terms.
2624
2625 So, to make sure things remain clear, please let people choose for
2626 themselves whether to link GPL'd libraries like readline with Guile.
2627
2628 ** regexp-substitute/global has changed slightly, but incompatibly.
2629
2630 If you include a function in the item list, the string of the match
2631 object it receives is the same string passed to
2632 regexp-substitute/global, not some suffix of that string.
2633 Correspondingly, the match's positions are relative to the entire
2634 string, not the suffix.
2635
2636 If the regexp can match the empty string, the way matches are chosen
2637 from the string has changed. regexp-substitute/global recognizes the
2638 same set of matches that list-matches does; see below.
2639
2640 ** New function: list-matches REGEXP STRING [FLAGS]
2641
2642 Return a list of match objects, one for every non-overlapping, maximal
2643 match of REGEXP in STRING. The matches appear in left-to-right order.
2644 list-matches only reports matches of the empty string if there are no
2645 other matches which begin on, end at, or include the empty match's
2646 position.
2647
2648 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2649
2650 ** New function: fold-matches REGEXP STRING INIT PROC [FLAGS]
2651
2652 For each match of REGEXP in STRING, apply PROC to the match object,
2653 and the last value PROC returned, or INIT for the first call. Return
2654 the last value returned by PROC. We apply PROC to the matches as they
2655 appear from left to right.
2656
2657 This function recognizes matches according to the same criteria as
2658 list-matches.
2659
2660 Thus, you could define list-matches like this:
2661
2662 (define (list-matches regexp string . flags)
2663 (reverse! (apply fold-matches regexp string '() cons flags)))
2664
2665 If present, FLAGS is passed as the FLAGS argument to regexp-exec.
2666
2667 ** Hooks
2668
2669 *** New function: hook? OBJ
2670
2671 Return #t if OBJ is a hook, otherwise #f.
2672
2673 *** New function: make-hook-with-name NAME [ARITY]
2674
2675 Return a hook with name NAME and arity ARITY. The default value for
2676 ARITY is 0. The only effect of NAME is that it will appear when the
2677 hook object is printed to ease debugging.
2678
2679 *** New function: hook-empty? HOOK
2680
2681 Return #t if HOOK doesn't contain any procedures, otherwise #f.
2682
2683 *** New function: hook->list HOOK
2684
2685 Return a list of the procedures that are called when run-hook is
2686 applied to HOOK.
2687
2688 ** `map' signals an error if its argument lists are not all the same length.
2689
2690 This is the behavior required by R5RS, so this change is really a bug
2691 fix. But it seems to affect a lot of people's code, so we're
2692 mentioning it here anyway.
2693
2694 ** Print-state handling has been made more transparent
2695
2696 Under certain circumstances, ports are represented as a port with an
2697 associated print state. Earlier, this pair was represented as a pair
2698 (see "Some magic has been added to the printer" below). It is now
2699 indistinguishable (almost; see `get-print-state') from a port on the
2700 user level.
2701
2702 *** New function: port-with-print-state OUTPUT-PORT PRINT-STATE
2703
2704 Return a new port with the associated print state PRINT-STATE.
2705
2706 *** New function: get-print-state OUTPUT-PORT
2707
2708 Return the print state associated with this port if it exists,
2709 otherwise return #f.
2710
2711 *** New function: directory-stream? OBJECT
2712
2713 Returns true iff OBJECT is a directory stream --- the sort of object
2714 returned by `opendir'.
2715
2716 ** New function: using-readline?
2717
2718 Return #t if readline is in use in the current repl.
2719
2720 ** structs will be removed in 1.4
2721
2722 Structs will be replaced in Guile 1.4. We will merge GOOPS into Guile
2723 and use GOOPS objects as the fundamental record type.
2724
2725 * Changes to the scm_ interface
2726
2727 ** structs will be removed in 1.4
2728
2729 The entire current struct interface (struct.c, struct.h) will be
2730 replaced in Guile 1.4. We will merge GOOPS into libguile and use
2731 GOOPS objects as the fundamental record type.
2732
2733 ** The internal representation of subr's has changed
2734
2735 Instead of giving a hint to the subr name, the CAR field of the subr
2736 now contains an index to a subr entry in scm_subr_table.
2737
2738 *** New variable: scm_subr_table
2739
2740 An array of subr entries. A subr entry contains the name, properties
2741 and documentation associated with the subr. The properties and
2742 documentation slots are not yet used.
2743
2744 ** A new scheme for "forwarding" calls to a builtin to a generic function
2745
2746 It is now possible to extend the functionality of some Guile
2747 primitives by letting them defer a call to a GOOPS generic function on
2748 argument mismatch. This means that there is no loss of efficiency in
2749 normal evaluation.
2750
2751 Example:
2752
2753 (use-modules (oop goops)) ; Must be GOOPS version 0.2.
2754 (define-method + ((x <string>) (y <string>))
2755 (string-append x y))
2756
2757 + will still be as efficient as usual in numerical calculations, but
2758 can also be used for concatenating strings.
2759
2760 Who will be the first one to extend Guile's numerical tower to
2761 rationals? :) [OK, there a few other things to fix before this can
2762 be made in a clean way.]
2763
2764 *** New snarf macros for defining primitives: SCM_GPROC, SCM_GPROC1
2765
2766 New macro: SCM_GPROC (CNAME, SNAME, REQ, OPT, VAR, CFUNC, GENERIC)
2767
2768 New macro: SCM_GPROC1 (CNAME, SNAME, TYPE, CFUNC, GENERIC)
2769
2770 These do the same job as SCM_PROC and SCM_PROC1, but they also define
2771 a variable GENERIC which can be used by the dispatch macros below.
2772
2773 [This is experimental code which may change soon.]
2774
2775 *** New macros for forwarding control to a generic on arg type error
2776
2777 New macro: SCM_WTA_DISPATCH_1 (GENERIC, ARG1, POS, SUBR)
2778
2779 New macro: SCM_WTA_DISPATCH_2 (GENERIC, ARG1, ARG2, POS, SUBR)
2780
2781 These correspond to the scm_wta function call, and have the same
2782 behaviour until the user has called the GOOPS primitive
2783 `enable-primitive-generic!'. After that, these macros will apply the
2784 generic function GENERIC to the argument(s) instead of calling
2785 scm_wta.
2786
2787 [This is experimental code which may change soon.]
2788
2789 *** New macros for argument testing with generic dispatch
2790
2791 New macro: SCM_GASSERT1 (COND, GENERIC, ARG1, POS, SUBR)
2792
2793 New macro: SCM_GASSERT2 (COND, GENERIC, ARG1, ARG2, POS, SUBR)
2794
2795 These correspond to the SCM_ASSERT macro, but will defer control to
2796 GENERIC on error after `enable-primitive-generic!' has been called.
2797
2798 [This is experimental code which may change soon.]
2799
2800 ** New function: SCM scm_eval_body (SCM body, SCM env)
2801
2802 Evaluates the body of a special form.
2803
2804 ** The internal representation of struct's has changed
2805
2806 Previously, four slots were allocated for the procedure(s) of entities
2807 and operators. The motivation for this representation had to do with
2808 the structure of the evaluator, the wish to support tail-recursive
2809 generic functions, and efficiency. Since the generic function
2810 dispatch mechanism has changed, there is no longer a need for such an
2811 expensive representation, and the representation has been simplified.
2812
2813 This should not make any difference for most users.
2814
2815 ** GOOPS support has been cleaned up.
2816
2817 Some code has been moved from eval.c to objects.c and code in both of
2818 these compilation units has been cleaned up and better structured.
2819
2820 *** New functions for applying generic functions
2821
2822 New function: SCM scm_apply_generic (GENERIC, ARGS)
2823 New function: SCM scm_call_generic_0 (GENERIC)
2824 New function: SCM scm_call_generic_1 (GENERIC, ARG1)
2825 New function: SCM scm_call_generic_2 (GENERIC, ARG1, ARG2)
2826 New function: SCM scm_call_generic_3 (GENERIC, ARG1, ARG2, ARG3)
2827
2828 ** Deprecated function: scm_make_named_hook
2829
2830 It is now replaced by:
2831
2832 ** New function: SCM scm_create_hook (const char *name, int arity)
2833
2834 Creates a hook in the same way as make-hook above but also
2835 binds a variable named NAME to it.
2836
2837 This is the typical way of creating a hook from C code.
2838
2839 Currently, the variable is created in the "current" module.
2840 This might change when we get the new module system.
2841
2842 [The behaviour is identical to scm_make_named_hook.]
2843
2844
2845 \f
2846 Changes since Guile 1.3:
2847
2848 * Changes to mailing lists
2849
2850 ** Some of the Guile mailing lists have moved to sourceware.cygnus.com.
2851
2852 See the README file to find current addresses for all the Guile
2853 mailing lists.
2854
2855 * Changes to the distribution
2856
2857 ** Readline support is no longer included with Guile by default.
2858
2859 Based on the different license terms of Guile and Readline, we
2860 concluded that Guile should not *by default* cause the linking of
2861 Readline into an application program. Readline support is now offered
2862 as a separate module, which is linked into an application only when
2863 you explicitly specify it.
2864
2865 Although Guile is GNU software, its distribution terms add a special
2866 exception to the usual GNU General Public License (GPL). Guile's
2867 license includes a clause that allows you to link Guile with non-free
2868 programs. We add this exception so as not to put Guile at a
2869 disadvantage vis-a-vis other extensibility packages that support other
2870 languages.
2871
2872 In contrast, the GNU Readline library is distributed under the GNU
2873 General Public License pure and simple. This means that you may not
2874 link Readline, even dynamically, into an application unless it is
2875 distributed under a free software license that is compatible the GPL.
2876
2877 Because of this difference in distribution terms, an application that
2878 can use Guile may not be able to use Readline. Now users will be
2879 explicitly offered two independent decisions about the use of these
2880 two packages.
2881
2882 You can activate the readline support by issuing
2883
2884 (use-modules (readline-activator))
2885 (activate-readline)
2886
2887 from your ".guile" file, for example.
2888
2889 * Changes to the stand-alone interpreter
2890
2891 ** All builtins now print as primitives.
2892 Previously builtin procedures not belonging to the fundamental subr
2893 types printed as #<compiled closure #<primitive-procedure gsubr-apply>>.
2894 Now, they print as #<primitive-procedure NAME>.
2895
2896 ** Backtraces slightly more intelligible.
2897 gsubr-apply and macro transformer application frames no longer appear
2898 in backtraces.
2899
2900 * Changes to Scheme functions and syntax
2901
2902 ** Guile now correctly handles internal defines by rewriting them into
2903 their equivalent letrec. Previously, internal defines would
2904 incrementally add to the innermost environment, without checking
2905 whether the restrictions specified in RnRS were met. This lead to the
2906 correct behaviour when these restriction actually were met, but didn't
2907 catch all illegal uses. Such an illegal use could lead to crashes of
2908 the Guile interpreter or or other unwanted results. An example of
2909 incorrect internal defines that made Guile behave erratically:
2910
2911 (let ()
2912 (define a 1)
2913 (define (b) a)
2914 (define c (1+ (b)))
2915 (define d 3)
2916
2917 (b))
2918
2919 => 2
2920
2921 The problem with this example is that the definition of `c' uses the
2922 value of `b' directly. This confuses the meoization machine of Guile
2923 so that the second call of `b' (this time in a larger environment that
2924 also contains bindings for `c' and `d') refers to the binding of `c'
2925 instead of `a'. You could also make Guile crash with a variation on
2926 this theme:
2927
2928 (define (foo flag)
2929 (define a 1)
2930 (define (b flag) (if flag a 1))
2931 (define c (1+ (b flag)))
2932 (define d 3)
2933
2934 (b #t))
2935
2936 (foo #f)
2937 (foo #t)
2938
2939 From now on, Guile will issue an `Unbound variable: b' error message
2940 for both examples.
2941
2942 ** Hooks
2943
2944 A hook contains a list of functions which should be called on
2945 particular occasions in an existing program. Hooks are used for
2946 customization.
2947
2948 A window manager might have a hook before-window-map-hook. The window
2949 manager uses the function run-hooks to call all functions stored in
2950 before-window-map-hook each time a window is mapped. The user can
2951 store functions in the hook using add-hook!.
2952
2953 In Guile, hooks are first class objects.
2954
2955 *** New function: make-hook [N_ARGS]
2956
2957 Return a hook for hook functions which can take N_ARGS arguments.
2958 The default value for N_ARGS is 0.
2959
2960 (See also scm_make_named_hook below.)
2961
2962 *** New function: add-hook! HOOK PROC [APPEND_P]
2963
2964 Put PROC at the beginning of the list of functions stored in HOOK.
2965 If APPEND_P is supplied, and non-false, put PROC at the end instead.
2966
2967 PROC must be able to take the number of arguments specified when the
2968 hook was created.
2969
2970 If PROC already exists in HOOK, then remove it first.
2971
2972 *** New function: remove-hook! HOOK PROC
2973
2974 Remove PROC from the list of functions in HOOK.
2975
2976 *** New function: reset-hook! HOOK
2977
2978 Clear the list of hook functions stored in HOOK.
2979
2980 *** New function: run-hook HOOK ARG1 ...
2981
2982 Run all hook functions stored in HOOK with arguments ARG1 ... .
2983 The number of arguments supplied must correspond to the number given
2984 when the hook was created.
2985
2986 ** The function `dynamic-link' now takes optional keyword arguments.
2987 The only keyword argument that is currently defined is `:global
2988 BOOL'. With it, you can control whether the shared library will be
2989 linked in global mode or not. In global mode, the symbols from the
2990 linked library can be used to resolve references from other
2991 dynamically linked libraries. In non-global mode, the linked
2992 library is essentially invisible and can only be accessed via
2993 `dynamic-func', etc. The default is now to link in global mode.
2994 Previously, the default has been non-global mode.
2995
2996 The `#:global' keyword is only effective on platforms that support
2997 the dlopen family of functions.
2998
2999 ** New function `provided?'
3000
3001 - Function: provided? FEATURE
3002 Return true iff FEATURE is supported by this installation of
3003 Guile. FEATURE must be a symbol naming a feature; the global
3004 variable `*features*' is a list of available features.
3005
3006 ** Changes to the module (ice-9 expect):
3007
3008 *** The expect-strings macro now matches `$' in a regular expression
3009 only at a line-break or end-of-file by default. Previously it would
3010 match the end of the string accumulated so far. The old behaviour
3011 can be obtained by setting the variable `expect-strings-exec-flags'
3012 to 0.
3013
3014 *** The expect-strings macro now uses a variable `expect-strings-exec-flags'
3015 for the regexp-exec flags. If `regexp/noteol' is included, then `$'
3016 in a regular expression will still match before a line-break or
3017 end-of-file. The default is `regexp/noteol'.
3018
3019 *** The expect-strings macro now uses a variable
3020 `expect-strings-compile-flags' for the flags to be supplied to
3021 `make-regexp'. The default is `regexp/newline', which was previously
3022 hard-coded.
3023
3024 *** The expect macro now supplies two arguments to a match procedure:
3025 the current accumulated string and a flag to indicate whether
3026 end-of-file has been reached. Previously only the string was supplied.
3027 If end-of-file is reached, the match procedure will be called an
3028 additional time with the same accumulated string as the previous call
3029 but with the flag set.
3030
3031 ** New module (ice-9 format), implementing the Common Lisp `format' function.
3032
3033 This code, and the documentation for it that appears here, was
3034 borrowed from SLIB, with minor adaptations for Guile.
3035
3036 - Function: format DESTINATION FORMAT-STRING . ARGUMENTS
3037 An almost complete implementation of Common LISP format description
3038 according to the CL reference book `Common LISP' from Guy L.
3039 Steele, Digital Press. Backward compatible to most of the
3040 available Scheme format implementations.
3041
3042 Returns `#t', `#f' or a string; has side effect of printing
3043 according to FORMAT-STRING. If DESTINATION is `#t', the output is
3044 to the current output port and `#t' is returned. If DESTINATION
3045 is `#f', a formatted string is returned as the result of the call.
3046 NEW: If DESTINATION is a string, DESTINATION is regarded as the
3047 format string; FORMAT-STRING is then the first argument and the
3048 output is returned as a string. If DESTINATION is a number, the
3049 output is to the current error port if available by the
3050 implementation. Otherwise DESTINATION must be an output port and
3051 `#t' is returned.
3052
3053 FORMAT-STRING must be a string. In case of a formatting error
3054 format returns `#f' and prints a message on the current output or
3055 error port. Characters are output as if the string were output by
3056 the `display' function with the exception of those prefixed by a
3057 tilde (~). For a detailed description of the FORMAT-STRING syntax
3058 please consult a Common LISP format reference manual. For a test
3059 suite to verify this format implementation load `formatst.scm'.
3060 Please send bug reports to `lutzeb@cs.tu-berlin.de'.
3061
3062 Note: `format' is not reentrant, i.e. only one `format'-call may
3063 be executed at a time.
3064
3065
3066 *** Format Specification (Format version 3.0)
3067
3068 Please consult a Common LISP format reference manual for a detailed
3069 description of the format string syntax. For a demonstration of the
3070 implemented directives see `formatst.scm'.
3071
3072 This implementation supports directive parameters and modifiers (`:'
3073 and `@' characters). Multiple parameters must be separated by a comma
3074 (`,'). Parameters can be numerical parameters (positive or negative),
3075 character parameters (prefixed by a quote character (`''), variable
3076 parameters (`v'), number of rest arguments parameter (`#'), empty and
3077 default parameters. Directive characters are case independent. The
3078 general form of a directive is:
3079
3080 DIRECTIVE ::= ~{DIRECTIVE-PARAMETER,}[:][@]DIRECTIVE-CHARACTER
3081
3082 DIRECTIVE-PARAMETER ::= [ [-|+]{0-9}+ | 'CHARACTER | v | # ]
3083
3084 *** Implemented CL Format Control Directives
3085
3086 Documentation syntax: Uppercase characters represent the
3087 corresponding control directive characters. Lowercase characters
3088 represent control directive parameter descriptions.
3089
3090 `~A'
3091 Any (print as `display' does).
3092 `~@A'
3093 left pad.
3094
3095 `~MINCOL,COLINC,MINPAD,PADCHARA'
3096 full padding.
3097
3098 `~S'
3099 S-expression (print as `write' does).
3100 `~@S'
3101 left pad.
3102
3103 `~MINCOL,COLINC,MINPAD,PADCHARS'
3104 full padding.
3105
3106 `~D'
3107 Decimal.
3108 `~@D'
3109 print number sign always.
3110
3111 `~:D'
3112 print comma separated.
3113
3114 `~MINCOL,PADCHAR,COMMACHARD'
3115 padding.
3116
3117 `~X'
3118 Hexadecimal.
3119 `~@X'
3120 print number sign always.
3121
3122 `~:X'
3123 print comma separated.
3124
3125 `~MINCOL,PADCHAR,COMMACHARX'
3126 padding.
3127
3128 `~O'
3129 Octal.
3130 `~@O'
3131 print number sign always.
3132
3133 `~:O'
3134 print comma separated.
3135
3136 `~MINCOL,PADCHAR,COMMACHARO'
3137 padding.
3138
3139 `~B'
3140 Binary.
3141 `~@B'
3142 print number sign always.
3143
3144 `~:B'
3145 print comma separated.
3146
3147 `~MINCOL,PADCHAR,COMMACHARB'
3148 padding.
3149
3150 `~NR'
3151 Radix N.
3152 `~N,MINCOL,PADCHAR,COMMACHARR'
3153 padding.
3154
3155 `~@R'
3156 print a number as a Roman numeral.
3157
3158 `~:@R'
3159 print a number as an "old fashioned" Roman numeral.
3160
3161 `~:R'
3162 print a number as an ordinal English number.
3163
3164 `~:@R'
3165 print a number as a cardinal English number.
3166
3167 `~P'
3168 Plural.
3169 `~@P'
3170 prints `y' and `ies'.
3171
3172 `~:P'
3173 as `~P but jumps 1 argument backward.'
3174
3175 `~:@P'
3176 as `~@P but jumps 1 argument backward.'
3177
3178 `~C'
3179 Character.
3180 `~@C'
3181 prints a character as the reader can understand it (i.e. `#\'
3182 prefixing).
3183
3184 `~:C'
3185 prints a character as emacs does (eg. `^C' for ASCII 03).
3186
3187 `~F'
3188 Fixed-format floating-point (prints a flonum like MMM.NNN).
3189 `~WIDTH,DIGITS,SCALE,OVERFLOWCHAR,PADCHARF'
3190 `~@F'
3191 If the number is positive a plus sign is printed.
3192
3193 `~E'
3194 Exponential floating-point (prints a flonum like MMM.NNN`E'EE).
3195 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARE'
3196 `~@E'
3197 If the number is positive a plus sign is printed.
3198
3199 `~G'
3200 General floating-point (prints a flonum either fixed or
3201 exponential).
3202 `~WIDTH,DIGITS,EXPONENTDIGITS,SCALE,OVERFLOWCHAR,PADCHAR,EXPONENTCHARG'
3203 `~@G'
3204 If the number is positive a plus sign is printed.
3205
3206 `~$'
3207 Dollars floating-point (prints a flonum in fixed with signs
3208 separated).
3209 `~DIGITS,SCALE,WIDTH,PADCHAR$'
3210 `~@$'
3211 If the number is positive a plus sign is printed.
3212
3213 `~:@$'
3214 A sign is always printed and appears before the padding.
3215
3216 `~:$'
3217 The sign appears before the padding.
3218
3219 `~%'
3220 Newline.
3221 `~N%'
3222 print N newlines.
3223
3224 `~&'
3225 print newline if not at the beginning of the output line.
3226 `~N&'
3227 prints `~&' and then N-1 newlines.
3228
3229 `~|'
3230 Page Separator.
3231 `~N|'
3232 print N page separators.
3233
3234 `~~'
3235 Tilde.
3236 `~N~'
3237 print N tildes.
3238
3239 `~'<newline>
3240 Continuation Line.
3241 `~:'<newline>
3242 newline is ignored, white space left.
3243
3244 `~@'<newline>
3245 newline is left, white space ignored.
3246
3247 `~T'
3248 Tabulation.
3249 `~@T'
3250 relative tabulation.
3251
3252 `~COLNUM,COLINCT'
3253 full tabulation.
3254
3255 `~?'
3256 Indirection (expects indirect arguments as a list).
3257 `~@?'
3258 extracts indirect arguments from format arguments.
3259
3260 `~(STR~)'
3261 Case conversion (converts by `string-downcase').
3262 `~:(STR~)'
3263 converts by `string-capitalize'.
3264
3265 `~@(STR~)'
3266 converts by `string-capitalize-first'.
3267
3268 `~:@(STR~)'
3269 converts by `string-upcase'.
3270
3271 `~*'
3272 Argument Jumping (jumps 1 argument forward).
3273 `~N*'
3274 jumps N arguments forward.
3275
3276 `~:*'
3277 jumps 1 argument backward.
3278
3279 `~N:*'
3280 jumps N arguments backward.
3281
3282 `~@*'
3283 jumps to the 0th argument.
3284
3285 `~N@*'
3286 jumps to the Nth argument (beginning from 0)
3287
3288 `~[STR0~;STR1~;...~;STRN~]'
3289 Conditional Expression (numerical clause conditional).
3290 `~N['
3291 take argument from N.
3292
3293 `~@['
3294 true test conditional.
3295
3296 `~:['
3297 if-else-then conditional.
3298
3299 `~;'
3300 clause separator.
3301
3302 `~:;'
3303 default clause follows.
3304
3305 `~{STR~}'
3306 Iteration (args come from the next argument (a list)).
3307 `~N{'
3308 at most N iterations.
3309
3310 `~:{'
3311 args from next arg (a list of lists).
3312
3313 `~@{'
3314 args from the rest of arguments.
3315
3316 `~:@{'
3317 args from the rest args (lists).
3318
3319 `~^'
3320 Up and out.
3321 `~N^'
3322 aborts if N = 0
3323
3324 `~N,M^'
3325 aborts if N = M
3326
3327 `~N,M,K^'
3328 aborts if N <= M <= K
3329
3330 *** Not Implemented CL Format Control Directives
3331
3332 `~:A'
3333 print `#f' as an empty list (see below).
3334
3335 `~:S'
3336 print `#f' as an empty list (see below).
3337
3338 `~<~>'
3339 Justification.
3340
3341 `~:^'
3342 (sorry I don't understand its semantics completely)
3343
3344 *** Extended, Replaced and Additional Control Directives
3345
3346 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHD'
3347 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHX'
3348 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHO'
3349 `~MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHB'
3350 `~N,MINCOL,PADCHAR,COMMACHAR,COMMAWIDTHR'
3351 COMMAWIDTH is the number of characters between two comma
3352 characters.
3353
3354 `~I'
3355 print a R4RS complex number as `~F~@Fi' with passed parameters for
3356 `~F'.
3357
3358 `~Y'
3359 Pretty print formatting of an argument for scheme code lists.
3360
3361 `~K'
3362 Same as `~?.'
3363
3364 `~!'
3365 Flushes the output if format DESTINATION is a port.
3366
3367 `~_'
3368 Print a `#\space' character
3369 `~N_'
3370 print N `#\space' characters.
3371
3372 `~/'
3373 Print a `#\tab' character
3374 `~N/'
3375 print N `#\tab' characters.
3376
3377 `~NC'
3378 Takes N as an integer representation for a character. No arguments
3379 are consumed. N is converted to a character by `integer->char'. N
3380 must be a positive decimal number.
3381
3382 `~:S'
3383 Print out readproof. Prints out internal objects represented as
3384 `#<...>' as strings `"#<...>"' so that the format output can always
3385 be processed by `read'.
3386
3387 `~:A'
3388 Print out readproof. Prints out internal objects represented as
3389 `#<...>' as strings `"#<...>"' so that the format output can always
3390 be processed by `read'.
3391
3392 `~Q'
3393 Prints information and a copyright notice on the format
3394 implementation.
3395 `~:Q'
3396 prints format version.
3397
3398 `~F, ~E, ~G, ~$'
3399 may also print number strings, i.e. passing a number as a string
3400 and format it accordingly.
3401
3402 *** Configuration Variables
3403
3404 The format module exports some configuration variables to suit the
3405 systems and users needs. There should be no modification necessary for
3406 the configuration that comes with Guile. Format detects automatically
3407 if the running scheme system implements floating point numbers and
3408 complex numbers.
3409
3410 format:symbol-case-conv
3411 Symbols are converted by `symbol->string' so the case type of the
3412 printed symbols is implementation dependent.
3413 `format:symbol-case-conv' is a one arg closure which is either
3414 `#f' (no conversion), `string-upcase', `string-downcase' or
3415 `string-capitalize'. (default `#f')
3416
3417 format:iobj-case-conv
3418 As FORMAT:SYMBOL-CASE-CONV but applies for the representation of
3419 implementation internal objects. (default `#f')
3420
3421 format:expch
3422 The character prefixing the exponent value in `~E' printing.
3423 (default `#\E')
3424
3425 *** Compatibility With Other Format Implementations
3426
3427 SLIB format 2.x:
3428 See `format.doc'.
3429
3430 SLIB format 1.4:
3431 Downward compatible except for padding support and `~A', `~S',
3432 `~P', `~X' uppercase printing. SLIB format 1.4 uses C-style
3433 `printf' padding support which is completely replaced by the CL
3434 `format' padding style.
3435
3436 MIT C-Scheme 7.1:
3437 Downward compatible except for `~', which is not documented
3438 (ignores all characters inside the format string up to a newline
3439 character). (7.1 implements `~a', `~s', ~NEWLINE, `~~', `~%',
3440 numerical and variable parameters and `:/@' modifiers in the CL
3441 sense).
3442
3443 Elk 1.5/2.0:
3444 Downward compatible except for `~A' and `~S' which print in
3445 uppercase. (Elk implements `~a', `~s', `~~', and `~%' (no
3446 directive parameters or modifiers)).
3447
3448 Scheme->C 01nov91:
3449 Downward compatible except for an optional destination parameter:
3450 S2C accepts a format call without a destination which returns a
3451 formatted string. This is equivalent to a #f destination in S2C.
3452 (S2C implements `~a', `~s', `~c', `~%', and `~~' (no directive
3453 parameters or modifiers)).
3454
3455
3456 ** Changes to string-handling functions.
3457
3458 These functions were added to support the (ice-9 format) module, above.
3459
3460 *** New function: string-upcase STRING
3461 *** New function: string-downcase STRING
3462
3463 These are non-destructive versions of the existing string-upcase! and
3464 string-downcase! functions.
3465
3466 *** New function: string-capitalize! STRING
3467 *** New function: string-capitalize STRING
3468
3469 These functions convert the first letter of each word in the string to
3470 upper case. Thus:
3471
3472 (string-capitalize "howdy there")
3473 => "Howdy There"
3474
3475 As with the other functions, string-capitalize! modifies the string in
3476 place, while string-capitalize returns a modified copy of its argument.
3477
3478 *** New function: string-ci->symbol STRING
3479
3480 Return a symbol whose name is STRING, but having the same case as if
3481 the symbol had be read by `read'.
3482
3483 Guile can be configured to be sensitive or insensitive to case
3484 differences in Scheme identifiers. If Guile is case-insensitive, all
3485 symbols are converted to lower case on input. The `string-ci->symbol'
3486 function returns a symbol whose name in STRING, transformed as Guile
3487 would if STRING were input.
3488
3489 *** New function: substring-move! STRING1 START END STRING2 START
3490
3491 Copy the substring of STRING1 from START (inclusive) to END
3492 (exclusive) to STRING2 at START. STRING1 and STRING2 may be the same
3493 string, and the source and destination areas may overlap; in all
3494 cases, the function behaves as if all the characters were copied
3495 simultanously.
3496
3497 *** Extended functions: substring-move-left! substring-move-right!
3498
3499 These functions now correctly copy arbitrarily overlapping substrings;
3500 they are both synonyms for substring-move!.
3501
3502
3503 ** New module (ice-9 getopt-long), with the function `getopt-long'.
3504
3505 getopt-long is a function for parsing command-line arguments in a
3506 manner consistent with other GNU programs.
3507
3508 (getopt-long ARGS GRAMMAR)
3509 Parse the arguments ARGS according to the argument list grammar GRAMMAR.
3510
3511 ARGS should be a list of strings. Its first element should be the
3512 name of the program; subsequent elements should be the arguments
3513 that were passed to the program on the command line. The
3514 `program-arguments' procedure returns a list of this form.
3515
3516 GRAMMAR is a list of the form:
3517 ((OPTION (PROPERTY VALUE) ...) ...)
3518
3519 Each OPTION should be a symbol. `getopt-long' will accept a
3520 command-line option named `--OPTION'.
3521 Each option can have the following (PROPERTY VALUE) pairs:
3522
3523 (single-char CHAR) --- Accept `-CHAR' as a single-character
3524 equivalent to `--OPTION'. This is how to specify traditional
3525 Unix-style flags.
3526 (required? BOOL) --- If BOOL is true, the option is required.
3527 getopt-long will raise an error if it is not found in ARGS.
3528 (value BOOL) --- If BOOL is #t, the option accepts a value; if
3529 it is #f, it does not; and if it is the symbol
3530 `optional', the option may appear in ARGS with or
3531 without a value.
3532 (predicate FUNC) --- If the option accepts a value (i.e. you
3533 specified `(value #t)' for this option), then getopt
3534 will apply FUNC to the value, and throw an exception
3535 if it returns #f. FUNC should be a procedure which
3536 accepts a string and returns a boolean value; you may
3537 need to use quasiquotes to get it into GRAMMAR.
3538
3539 The (PROPERTY VALUE) pairs may occur in any order, but each
3540 property may occur only once. By default, options do not have
3541 single-character equivalents, are not required, and do not take
3542 values.
3543
3544 In ARGS, single-character options may be combined, in the usual
3545 Unix fashion: ("-x" "-y") is equivalent to ("-xy"). If an option
3546 accepts values, then it must be the last option in the
3547 combination; the value is the next argument. So, for example, using
3548 the following grammar:
3549 ((apples (single-char #\a))
3550 (blimps (single-char #\b) (value #t))
3551 (catalexis (single-char #\c) (value #t)))
3552 the following argument lists would be acceptable:
3553 ("-a" "-b" "bang" "-c" "couth") ("bang" and "couth" are the values
3554 for "blimps" and "catalexis")
3555 ("-ab" "bang" "-c" "couth") (same)
3556 ("-ac" "couth" "-b" "bang") (same)
3557 ("-abc" "couth" "bang") (an error, since `-b' is not the
3558 last option in its combination)
3559
3560 If an option's value is optional, then `getopt-long' decides
3561 whether it has a value by looking at what follows it in ARGS. If
3562 the next element is a string, and it does not appear to be an
3563 option itself, then that string is the option's value.
3564
3565 The value of a long option can appear as the next element in ARGS,
3566 or it can follow the option name, separated by an `=' character.
3567 Thus, using the same grammar as above, the following argument lists
3568 are equivalent:
3569 ("--apples" "Braeburn" "--blimps" "Goodyear")
3570 ("--apples=Braeburn" "--blimps" "Goodyear")
3571 ("--blimps" "Goodyear" "--apples=Braeburn")
3572
3573 If the option "--" appears in ARGS, argument parsing stops there;
3574 subsequent arguments are returned as ordinary arguments, even if
3575 they resemble options. So, in the argument list:
3576 ("--apples" "Granny Smith" "--" "--blimp" "Goodyear")
3577 `getopt-long' will recognize the `apples' option as having the
3578 value "Granny Smith", but it will not recognize the `blimp'
3579 option; it will return the strings "--blimp" and "Goodyear" as
3580 ordinary argument strings.
3581
3582 The `getopt-long' function returns the parsed argument list as an
3583 assocation list, mapping option names --- the symbols from GRAMMAR
3584 --- onto their values, or #t if the option does not accept a value.
3585 Unused options do not appear in the alist.
3586
3587 All arguments that are not the value of any option are returned
3588 as a list, associated with the empty list.
3589
3590 `getopt-long' throws an exception if:
3591 - it finds an unrecognized option in ARGS
3592 - a required option is omitted
3593 - an option that requires an argument doesn't get one
3594 - an option that doesn't accept an argument does get one (this can
3595 only happen using the long option `--opt=value' syntax)
3596 - an option predicate fails
3597
3598 So, for example:
3599
3600 (define grammar
3601 `((lockfile-dir (required? #t)
3602 (value #t)
3603 (single-char #\k)
3604 (predicate ,file-is-directory?))
3605 (verbose (required? #f)
3606 (single-char #\v)
3607 (value #f))
3608 (x-includes (single-char #\x))
3609 (rnet-server (single-char #\y)
3610 (predicate ,string?))))
3611
3612 (getopt-long '("my-prog" "-vk" "/tmp" "foo1" "--x-includes=/usr/include"
3613 "--rnet-server=lamprod" "--" "-fred" "foo2" "foo3")
3614 grammar)
3615 => ((() "foo1" "-fred" "foo2" "foo3")
3616 (rnet-server . "lamprod")
3617 (x-includes . "/usr/include")
3618 (lockfile-dir . "/tmp")
3619 (verbose . #t))
3620
3621 ** The (ice-9 getopt-gnu-style) module is obsolete; use (ice-9 getopt-long).
3622
3623 It will be removed in a few releases.
3624
3625 ** New syntax: lambda*
3626 ** New syntax: define*
3627 ** New syntax: define*-public
3628 ** New syntax: defmacro*
3629 ** New syntax: defmacro*-public
3630 Guile now supports optional arguments.
3631
3632 `lambda*', `define*', `define*-public', `defmacro*' and
3633 `defmacro*-public' are identical to the non-* versions except that
3634 they use an extended type of parameter list that has the following BNF
3635 syntax (parentheses are literal, square brackets indicate grouping,
3636 and `*', `+' and `?' have the usual meaning):
3637
3638 ext-param-list ::= ( [identifier]* [#&optional [ext-var-decl]+]?
3639 [#&key [ext-var-decl]+ [#&allow-other-keys]?]?
3640 [[#&rest identifier]|[. identifier]]? ) | [identifier]
3641
3642 ext-var-decl ::= identifier | ( identifier expression )
3643
3644 The semantics are best illustrated with the following documentation
3645 and examples for `lambda*':
3646
3647 lambda* args . body
3648 lambda extended for optional and keyword arguments
3649
3650 lambda* creates a procedure that takes optional arguments. These
3651 are specified by putting them inside brackets at the end of the
3652 paramater list, but before any dotted rest argument. For example,
3653 (lambda* (a b #&optional c d . e) '())
3654 creates a procedure with fixed arguments a and b, optional arguments c
3655 and d, and rest argument e. If the optional arguments are omitted
3656 in a call, the variables for them are unbound in the procedure. This
3657 can be checked with the bound? macro.
3658
3659 lambda* can also take keyword arguments. For example, a procedure
3660 defined like this:
3661 (lambda* (#&key xyzzy larch) '())
3662 can be called with any of the argument lists (#:xyzzy 11)
3663 (#:larch 13) (#:larch 42 #:xyzzy 19) (). Whichever arguments
3664 are given as keywords are bound to values.
3665
3666 Optional and keyword arguments can also be given default values
3667 which they take on when they are not present in a call, by giving a
3668 two-item list in place of an optional argument, for example in:
3669 (lambda* (foo #&optional (bar 42) #&key (baz 73)) (list foo bar baz))
3670 foo is a fixed argument, bar is an optional argument with default
3671 value 42, and baz is a keyword argument with default value 73.
3672 Default value expressions are not evaluated unless they are needed
3673 and until the procedure is called.
3674
3675 lambda* now supports two more special parameter list keywords.
3676
3677 lambda*-defined procedures now throw an error by default if a
3678 keyword other than one of those specified is found in the actual
3679 passed arguments. However, specifying #&allow-other-keys
3680 immediately after the kyword argument declarations restores the
3681 previous behavior of ignoring unknown keywords. lambda* also now
3682 guarantees that if the same keyword is passed more than once, the
3683 last one passed is the one that takes effect. For example,
3684 ((lambda* (#&key (heads 0) (tails 0)) (display (list heads tails)))
3685 #:heads 37 #:tails 42 #:heads 99)
3686 would result in (99 47) being displayed.
3687
3688 #&rest is also now provided as a synonym for the dotted syntax rest
3689 argument. The argument lists (a . b) and (a #&rest b) are equivalent in
3690 all respects to lambda*. This is provided for more similarity to DSSSL,
3691 MIT-Scheme and Kawa among others, as well as for refugees from other
3692 Lisp dialects.
3693
3694 Further documentation may be found in the optargs.scm file itself.
3695
3696 The optional argument module also exports the macros `let-optional',
3697 `let-optional*', `let-keywords', `let-keywords*' and `bound?'. These
3698 are not documented here because they may be removed in the future, but
3699 full documentation is still available in optargs.scm.
3700
3701 ** New syntax: and-let*
3702 Guile now supports the `and-let*' form, described in the draft SRFI-2.
3703
3704 Syntax: (land* (<clause> ...) <body> ...)
3705 Each <clause> should have one of the following forms:
3706 (<variable> <expression>)
3707 (<expression>)
3708 <bound-variable>
3709 Each <variable> or <bound-variable> should be an identifier. Each
3710 <expression> should be a valid expression. The <body> should be a
3711 possibly empty sequence of expressions, like the <body> of a
3712 lambda form.
3713
3714 Semantics: A LAND* expression is evaluated by evaluating the
3715 <expression> or <bound-variable> of each of the <clause>s from
3716 left to right. The value of the first <expression> or
3717 <bound-variable> that evaluates to a false value is returned; the
3718 remaining <expression>s and <bound-variable>s are not evaluated.
3719 The <body> forms are evaluated iff all the <expression>s and
3720 <bound-variable>s evaluate to true values.
3721
3722 The <expression>s and the <body> are evaluated in an environment
3723 binding each <variable> of the preceding (<variable> <expression>)
3724 clauses to the value of the <expression>. Later bindings
3725 shadow earlier bindings.
3726
3727 Guile's and-let* macro was contributed by Michael Livshin.
3728
3729 ** New sorting functions
3730
3731 *** New function: sorted? SEQUENCE LESS?
3732 Returns `#t' when the sequence argument is in non-decreasing order
3733 according to LESS? (that is, there is no adjacent pair `... x y
3734 ...' for which `(less? y x)').
3735
3736 Returns `#f' when the sequence contains at least one out-of-order
3737 pair. It is an error if the sequence is neither a list nor a
3738 vector.
3739
3740 *** New function: merge LIST1 LIST2 LESS?
3741 LIST1 and LIST2 are sorted lists.
3742 Returns the sorted list of all elements in LIST1 and LIST2.
3743
3744 Assume that the elements a and b1 in LIST1 and b2 in LIST2 are "equal"
3745 in the sense that (LESS? x y) --> #f for x, y in {a, b1, b2},
3746 and that a < b1 in LIST1. Then a < b1 < b2 in the result.
3747 (Here "<" should read "comes before".)
3748
3749 *** New procedure: merge! LIST1 LIST2 LESS?
3750 Merges two lists, re-using the pairs of LIST1 and LIST2 to build
3751 the result. If the code is compiled, and LESS? constructs no new
3752 pairs, no pairs at all will be allocated. The first pair of the
3753 result will be either the first pair of LIST1 or the first pair of
3754 LIST2.
3755
3756 *** New function: sort SEQUENCE LESS?
3757 Accepts either a list or a vector, and returns a new sequence
3758 which is sorted. The new sequence is the same type as the input.
3759 Always `(sorted? (sort sequence less?) less?)'. The original
3760 sequence is not altered in any way. The new sequence shares its
3761 elements with the old one; no elements are copied.
3762
3763 *** New procedure: sort! SEQUENCE LESS
3764 Returns its sorted result in the original boxes. No new storage is
3765 allocated at all. Proper usage: (set! slist (sort! slist <))
3766
3767 *** New function: stable-sort SEQUENCE LESS?
3768 Similar to `sort' but stable. That is, if "equal" elements are
3769 ordered a < b in the original sequence, they will have the same order
3770 in the result.
3771
3772 *** New function: stable-sort! SEQUENCE LESS?
3773 Similar to `sort!' but stable.
3774 Uses temporary storage when sorting vectors.
3775
3776 *** New functions: sort-list, sort-list!
3777 Added for compatibility with scsh.
3778
3779 ** New built-in random number support
3780
3781 *** New function: random N [STATE]
3782 Accepts a positive integer or real N and returns a number of the
3783 same type between zero (inclusive) and N (exclusive). The values
3784 returned have a uniform distribution.
3785
3786 The optional argument STATE must be of the type produced by
3787 `copy-random-state' or `seed->random-state'. It defaults to the value
3788 of the variable `*random-state*'. This object is used to maintain the
3789 state of the pseudo-random-number generator and is altered as a side
3790 effect of the `random' operation.
3791
3792 *** New variable: *random-state*
3793 Holds a data structure that encodes the internal state of the
3794 random-number generator that `random' uses by default. The nature
3795 of this data structure is implementation-dependent. It may be
3796 printed out and successfully read back in, but may or may not
3797 function correctly as a random-number state object in another
3798 implementation.
3799
3800 *** New function: copy-random-state [STATE]
3801 Returns a new object of type suitable for use as the value of the
3802 variable `*random-state*' and as a second argument to `random'.
3803 If argument STATE is given, a copy of it is returned. Otherwise a
3804 copy of `*random-state*' is returned.
3805
3806 *** New function: seed->random-state SEED
3807 Returns a new object of type suitable for use as the value of the
3808 variable `*random-state*' and as a second argument to `random'.
3809 SEED is a string or a number. A new state is generated and
3810 initialized using SEED.
3811
3812 *** New function: random:uniform [STATE]
3813 Returns an uniformly distributed inexact real random number in the
3814 range between 0 and 1.
3815
3816 *** New procedure: random:solid-sphere! VECT [STATE]
3817 Fills VECT with inexact real random numbers the sum of whose
3818 squares is less than 1.0. Thinking of VECT as coordinates in
3819 space of dimension N = `(vector-length VECT)', the coordinates are
3820 uniformly distributed within the unit N-shere. The sum of the
3821 squares of the numbers is returned. VECT can be either a vector
3822 or a uniform vector of doubles.
3823
3824 *** New procedure: random:hollow-sphere! VECT [STATE]
3825 Fills VECT with inexact real random numbers the sum of whose squares
3826 is equal to 1.0. Thinking of VECT as coordinates in space of
3827 dimension n = `(vector-length VECT)', the coordinates are uniformly
3828 distributed over the surface of the unit n-shere. VECT can be either
3829 a vector or a uniform vector of doubles.
3830
3831 *** New function: random:normal [STATE]
3832 Returns an inexact real in a normal distribution with mean 0 and
3833 standard deviation 1. For a normal distribution with mean M and
3834 standard deviation D use `(+ M (* D (random:normal)))'.
3835
3836 *** New procedure: random:normal-vector! VECT [STATE]
3837 Fills VECT with inexact real random numbers which are independent and
3838 standard normally distributed (i.e., with mean 0 and variance 1).
3839 VECT can be either a vector or a uniform vector of doubles.
3840
3841 *** New function: random:exp STATE
3842 Returns an inexact real in an exponential distribution with mean 1.
3843 For an exponential distribution with mean U use (* U (random:exp)).
3844
3845 ** The range of logand, logior, logxor, logtest, and logbit? have changed.
3846
3847 These functions now operate on numbers in the range of a C unsigned
3848 long.
3849
3850 These functions used to operate on numbers in the range of a C signed
3851 long; however, this seems inappropriate, because Guile integers don't
3852 overflow.
3853
3854 ** New function: make-guardian
3855 This is an implementation of guardians as described in
3856 R. Kent Dybvig, Carl Bruggeman, and David Eby (1993) "Guardians in a
3857 Generation-Based Garbage Collector" ACM SIGPLAN Conference on
3858 Programming Language Design and Implementation, June 1993
3859 ftp://ftp.cs.indiana.edu/pub/scheme-repository/doc/pubs/guardians.ps.gz
3860
3861 ** New functions: delq1!, delv1!, delete1!
3862 These procedures behave similar to delq! and friends but delete only
3863 one object if at all.
3864
3865 ** New function: unread-string STRING PORT
3866 Unread STRING to PORT, that is, push it back onto the port so that
3867 next read operation will work on the pushed back characters.
3868
3869 ** unread-char can now be called multiple times
3870 If unread-char is called multiple times, the unread characters will be
3871 read again in last-in first-out order.
3872
3873 ** the procedures uniform-array-read! and uniform-array-write! now
3874 work on any kind of port, not just ports which are open on a file.
3875
3876 ** Now 'l' in a port mode requests line buffering.
3877
3878 ** The procedure truncate-file now works on string ports as well
3879 as file ports. If the size argument is omitted, the current
3880 file position is used.
3881
3882 ** new procedure: seek PORT/FDES OFFSET WHENCE
3883 The arguments are the same as for the old fseek procedure, but it
3884 works on string ports as well as random-access file ports.
3885
3886 ** the fseek procedure now works on string ports, since it has been
3887 redefined using seek.
3888
3889 ** the setvbuf procedure now uses a default size if mode is _IOFBF and
3890 size is not supplied.
3891
3892 ** the newline procedure no longer flushes the port if it's not
3893 line-buffered: previously it did if it was the current output port.
3894
3895 ** open-pipe and close-pipe are no longer primitive procedures, but
3896 an emulation can be obtained using `(use-modules (ice-9 popen))'.
3897
3898 ** the freopen procedure has been removed.
3899
3900 ** new procedure: drain-input PORT
3901 Drains PORT's read buffers (including any pushed-back characters)
3902 and returns the contents as a single string.
3903
3904 ** New function: map-in-order PROC LIST1 LIST2 ...
3905 Version of `map' which guarantees that the procedure is applied to the
3906 lists in serial order.
3907
3908 ** Renamed `serial-array-copy!' and `serial-array-map!' to
3909 `array-copy-in-order!' and `array-map-in-order!'. The old names are
3910 now obsolete and will go away in release 1.5.
3911
3912 ** New syntax: collect BODY1 ...
3913 Version of `begin' which returns a list of the results of the body
3914 forms instead of the result of the last body form. In contrast to
3915 `begin', `collect' allows an empty body.
3916
3917 ** New functions: read-history FILENAME, write-history FILENAME
3918 Read/write command line history from/to file. Returns #t on success
3919 and #f if an error occured.
3920
3921 ** `ls' and `lls' in module (ice-9 ls) now handle no arguments.
3922
3923 These procedures return a list of definitions available in the specified
3924 argument, a relative module reference. In the case of no argument,
3925 `(current-module)' is now consulted for definitions to return, instead
3926 of simply returning #f, the former behavior.
3927
3928 ** The #/ syntax for lists is no longer supported.
3929
3930 Earlier versions of Scheme accepted this syntax, but printed a
3931 warning.
3932
3933 ** Guile no longer consults the SCHEME_LOAD_PATH environment variable.
3934
3935 Instead, you should set GUILE_LOAD_PATH to tell Guile where to find
3936 modules.
3937
3938 * Changes to the gh_ interface
3939
3940 ** gh_scm2doubles
3941
3942 Now takes a second argument which is the result array. If this
3943 pointer is NULL, a new array is malloced (the old behaviour).
3944
3945 ** gh_chars2byvect, gh_shorts2svect, gh_floats2fvect, gh_scm2chars,
3946 gh_scm2shorts, gh_scm2longs, gh_scm2floats
3947
3948 New functions.
3949
3950 * Changes to the scm_ interface
3951
3952 ** Function: scm_make_named_hook (char* name, int n_args)
3953
3954 Creates a hook in the same way as make-hook above but also
3955 binds a variable named NAME to it.
3956
3957 This is the typical way of creating a hook from C code.
3958
3959 Currently, the variable is created in the "current" module. This
3960 might change when we get the new module system.
3961
3962 ** The smob interface
3963
3964 The interface for creating smobs has changed. For documentation, see
3965 data-rep.info (made from guile-core/doc/data-rep.texi).
3966
3967 *** Deprecated function: SCM scm_newsmob (scm_smobfuns *)
3968
3969 >>> This function will be removed in 1.3.4. <<<
3970
3971 It is replaced by:
3972
3973 *** Function: SCM scm_make_smob_type (const char *name, scm_sizet size)
3974 This function adds a new smob type, named NAME, with instance size
3975 SIZE to the system. The return value is a tag that is used in
3976 creating instances of the type. If SIZE is 0, then no memory will
3977 be allocated when instances of the smob are created, and nothing
3978 will be freed by the default free function.
3979
3980 *** Function: void scm_set_smob_mark (long tc, SCM (*mark) (SCM))
3981 This function sets the smob marking procedure for the smob type
3982 specified by the tag TC. TC is the tag returned by
3983 `scm_make_smob_type'.
3984
3985 *** Function: void scm_set_smob_free (long tc, SCM (*mark) (SCM))
3986 This function sets the smob freeing procedure for the smob type
3987 specified by the tag TC. TC is the tag returned by
3988 `scm_make_smob_type'.
3989
3990 *** Function: void scm_set_smob_print (tc, print)
3991
3992 - Function: void scm_set_smob_print (long tc,
3993 scm_sizet (*print) (SCM,
3994 SCM,
3995 scm_print_state *))
3996
3997 This function sets the smob printing procedure for the smob type
3998 specified by the tag TC. TC is the tag returned by
3999 `scm_make_smob_type'.
4000
4001 *** Function: void scm_set_smob_equalp (long tc, SCM (*equalp) (SCM, SCM))
4002 This function sets the smob equality-testing predicate for the
4003 smob type specified by the tag TC. TC is the tag returned by
4004 `scm_make_smob_type'.
4005
4006 *** Macro: void SCM_NEWSMOB (SCM var, long tc, void *data)
4007 Make VALUE contain a smob instance of the type with type code TC and
4008 smob data DATA. VALUE must be previously declared as C type `SCM'.
4009
4010 *** Macro: fn_returns SCM_RETURN_NEWSMOB (long tc, void *data)
4011 This macro expands to a block of code that creates a smob instance
4012 of the type with type code TC and smob data DATA, and returns that
4013 `SCM' value. It should be the last piece of code in a block.
4014
4015 ** The interfaces for using I/O ports and implementing port types
4016 (ptobs) have changed significantly. The new interface is based on
4017 shared access to buffers and a new set of ptob procedures.
4018
4019 *** scm_newptob has been removed
4020
4021 It is replaced by:
4022
4023 *** Function: SCM scm_make_port_type (type_name, fill_buffer, write_flush)
4024
4025 - Function: SCM scm_make_port_type (char *type_name,
4026 int (*fill_buffer) (SCM port),
4027 void (*write_flush) (SCM port));
4028
4029 Similarly to the new smob interface, there is a set of function
4030 setters by which the user can customize the behaviour of his port
4031 type. See ports.h (scm_set_port_XXX).
4032
4033 ** scm_strport_to_string: New function: creates a new string from
4034 a string port's buffer.
4035
4036 ** Plug in interface for random number generators
4037 The variable `scm_the_rng' in random.c contains a value and three
4038 function pointers which together define the current random number
4039 generator being used by the Scheme level interface and the random
4040 number library functions.
4041
4042 The user is free to replace the default generator with the generator
4043 of his own choice.
4044
4045 *** Variable: size_t scm_the_rng.rstate_size
4046 The size of the random state type used by the current RNG
4047 measured in chars.
4048
4049 *** Function: unsigned long scm_the_rng.random_bits (scm_rstate *STATE)
4050 Given the random STATE, return 32 random bits.
4051
4052 *** Function: void scm_the_rng.init_rstate (scm_rstate *STATE, chars *S, int N)
4053 Seed random state STATE using string S of length N.
4054
4055 *** Function: scm_rstate *scm_the_rng.copy_rstate (scm_rstate *STATE)
4056 Given random state STATE, return a malloced copy.
4057
4058 ** Default RNG
4059 The default RNG is the MWC (Multiply With Carry) random number
4060 generator described by George Marsaglia at the Department of
4061 Statistics and Supercomputer Computations Research Institute, The
4062 Florida State University (http://stat.fsu.edu/~geo).
4063
4064 It uses 64 bits, has a period of 4578426017172946943 (4.6e18), and
4065 passes all tests in the DIEHARD test suite
4066 (http://stat.fsu.edu/~geo/diehard.html). The generation of 32 bits
4067 costs one multiply and one add on platforms which either supports long
4068 longs (gcc does this on most systems) or have 64 bit longs. The cost
4069 is four multiply on other systems but this can be optimized by writing
4070 scm_i_uniform32 in assembler.
4071
4072 These functions are provided through the scm_the_rng interface for use
4073 by libguile and the application.
4074
4075 *** Function: unsigned long scm_i_uniform32 (scm_i_rstate *STATE)
4076 Given the random STATE, return 32 random bits.
4077 Don't use this function directly. Instead go through the plugin
4078 interface (see "Plug in interface" above).
4079
4080 *** Function: void scm_i_init_rstate (scm_i_rstate *STATE, char *SEED, int N)
4081 Initialize STATE using SEED of length N.
4082
4083 *** Function: scm_i_rstate *scm_i_copy_rstate (scm_i_rstate *STATE)
4084 Return a malloc:ed copy of STATE. This function can easily be re-used
4085 in the interfaces to other RNGs.
4086
4087 ** Random number library functions
4088 These functions use the current RNG through the scm_the_rng interface.
4089 It might be a good idea to use these functions from your C code so
4090 that only one random generator is used by all code in your program.
4091
4092 The default random state is stored in:
4093
4094 *** Variable: SCM scm_var_random_state
4095 Contains the vcell of the Scheme variable "*random-state*" which is
4096 used as default state by all random number functions in the Scheme
4097 level interface.
4098
4099 Example:
4100
4101 double x = scm_c_uniform01 (SCM_RSTATE (SCM_CDR (scm_var_random_state)));
4102
4103 *** Function: scm_rstate *scm_c_default_rstate (void)
4104 This is a convenience function which returns the value of
4105 scm_var_random_state. An error message is generated if this value
4106 isn't a random state.
4107
4108 *** Function: scm_rstate *scm_c_make_rstate (char *SEED, int LENGTH)
4109 Make a new random state from the string SEED of length LENGTH.
4110
4111 It is generally not a good idea to use multiple random states in a
4112 program. While subsequent random numbers generated from one random
4113 state are guaranteed to be reasonably independent, there is no such
4114 guarantee for numbers generated from different random states.
4115
4116 *** Macro: unsigned long scm_c_uniform32 (scm_rstate *STATE)
4117 Return 32 random bits.
4118
4119 *** Function: double scm_c_uniform01 (scm_rstate *STATE)
4120 Return a sample from the uniform(0,1) distribution.
4121
4122 *** Function: double scm_c_normal01 (scm_rstate *STATE)
4123 Return a sample from the normal(0,1) distribution.
4124
4125 *** Function: double scm_c_exp1 (scm_rstate *STATE)
4126 Return a sample from the exp(1) distribution.
4127
4128 *** Function: unsigned long scm_c_random (scm_rstate *STATE, unsigned long M)
4129 Return a sample from the discrete uniform(0,M) distribution.
4130
4131 *** Function: SCM scm_c_random_bignum (scm_rstate *STATE, SCM M)
4132 Return a sample from the discrete uniform(0,M) distribution.
4133 M must be a bignum object. The returned value may be an INUM.
4134
4135
4136 \f
4137 Changes in Guile 1.3 (released Monday, October 19, 1998):
4138
4139 * Changes to the distribution
4140
4141 ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH.
4142 To avoid conflicts, programs should name environment variables after
4143 themselves, except when there's a common practice establishing some
4144 other convention.
4145
4146 For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH,
4147 giving the former precedence, and printing a warning message if the
4148 latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all.
4149
4150 ** The header files related to multi-byte characters have been removed.
4151 They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code
4152 which referred to these explicitly will probably need to be rewritten,
4153 since the support for the variant string types has been removed; see
4154 below.
4155
4156 ** The header files append.h and sequences.h have been removed. These
4157 files implemented non-R4RS operations which would encourage
4158 non-portable programming style and less easy-to-read code.
4159
4160 * Changes to the stand-alone interpreter
4161
4162 ** New procedures have been added to implement a "batch mode":
4163
4164 *** Function: batch-mode?
4165
4166 Returns a boolean indicating whether the interpreter is in batch
4167 mode.
4168
4169 *** Function: set-batch-mode?! ARG
4170
4171 If ARG is true, switches the interpreter to batch mode. The `#f'
4172 case has not been implemented.
4173
4174 ** Guile now provides full command-line editing, when run interactively.
4175 To use this feature, you must have the readline library installed.
4176 The Guile build process will notice it, and automatically include
4177 support for it.
4178
4179 The readline library is available via anonymous FTP from any GNU
4180 mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu".
4181
4182 ** the-last-stack is now a fluid.
4183
4184 * Changes to the procedure for linking libguile with your programs
4185
4186 ** You can now use the `guile-config' utility to build programs that use Guile.
4187
4188 Guile now includes a command-line utility called `guile-config', which
4189 can provide information about how to compile and link programs that
4190 use Guile.
4191
4192 *** `guile-config compile' prints any C compiler flags needed to use Guile.
4193 You should include this command's output on the command line you use
4194 to compile C or C++ code that #includes the Guile header files. It's
4195 usually just a `-I' flag to help the compiler find the Guile headers.
4196
4197
4198 *** `guile-config link' prints any linker flags necessary to link with Guile.
4199
4200 This command writes to its standard output a list of flags which you
4201 must pass to the linker to link your code against the Guile library.
4202 The flags include '-lguile' itself, any other libraries the Guile
4203 library depends upon, and any `-L' flags needed to help the linker
4204 find those libraries.
4205
4206 For example, here is a Makefile rule that builds a program named 'foo'
4207 from the object files ${FOO_OBJECTS}, and links them against Guile:
4208
4209 foo: ${FOO_OBJECTS}
4210 ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo
4211
4212 Previous Guile releases recommended that you use autoconf to detect
4213 which of a predefined set of libraries were present on your system.
4214 It is more robust to use `guile-config', since it records exactly which
4215 libraries the installed Guile library requires.
4216
4217 This was originally called `build-guile', but was renamed to
4218 `guile-config' before Guile 1.3 was released, to be consistent with
4219 the analogous script for the GTK+ GUI toolkit, which is called
4220 `gtk-config'.
4221
4222
4223 ** Use the GUILE_FLAGS macro in your configure.in file to find Guile.
4224
4225 If you are using the GNU autoconf package to configure your program,
4226 you can use the GUILE_FLAGS autoconf macro to call `guile-config'
4227 (described above) and gather the necessary values for use in your
4228 Makefiles.
4229
4230 The GUILE_FLAGS macro expands to configure script code which runs the
4231 `guile-config' script, to find out where Guile's header files and
4232 libraries are installed. It sets two variables, marked for
4233 substitution, as by AC_SUBST.
4234
4235 GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build
4236 code that uses Guile header files. This is almost always just a
4237 -I flag.
4238
4239 GUILE_LDFLAGS --- flags to pass to the linker to link a
4240 program against Guile. This includes `-lguile' for the Guile
4241 library itself, any libraries that Guile itself requires (like
4242 -lqthreads), and so on. It may also include a -L flag to tell the
4243 compiler where to find the libraries.
4244
4245 GUILE_FLAGS is defined in the file guile.m4, in the top-level
4246 directory of the Guile distribution. You can copy it into your
4247 package's aclocal.m4 file, and then use it in your configure.in file.
4248
4249 If you are using the `aclocal' program, distributed with GNU automake,
4250 to maintain your aclocal.m4 file, the Guile installation process
4251 installs guile.m4 where aclocal will find it. All you need to do is
4252 use GUILE_FLAGS in your configure.in file, and then run `aclocal';
4253 this will copy the definition of GUILE_FLAGS into your aclocal.m4
4254 file.
4255
4256
4257 * Changes to Scheme functions and syntax
4258
4259 ** Multi-byte strings have been removed, as have multi-byte and wide
4260 ports. We felt that these were the wrong approach to
4261 internationalization support.
4262
4263 ** New function: readline [PROMPT]
4264 Read a line from the terminal, and allow the user to edit it,
4265 prompting with PROMPT. READLINE provides a large set of Emacs-like
4266 editing commands, lets the user recall previously typed lines, and
4267 works on almost every kind of terminal, including dumb terminals.
4268
4269 READLINE assumes that the cursor is at the beginning of the line when
4270 it is invoked. Thus, you can't print a prompt yourself, and then call
4271 READLINE; you need to package up your prompt as a string, pass it to
4272 the function, and let READLINE print the prompt itself. This is
4273 because READLINE needs to know the prompt's screen width.
4274
4275 For Guile to provide this function, you must have the readline
4276 library, version 2.1 or later, installed on your system. Readline is
4277 available via anonymous FTP from prep.ai.mit.edu in pub/gnu, or from
4278 any GNU mirror site.
4279
4280 See also ADD-HISTORY function.
4281
4282 ** New function: add-history STRING
4283 Add STRING as the most recent line in the history used by the READLINE
4284 command. READLINE does not add lines to the history itself; you must
4285 call ADD-HISTORY to make previous input available to the user.
4286
4287 ** The behavior of the read-line function has changed.
4288
4289 This function now uses standard C library functions to read the line,
4290 for speed. This means that it doesn not respect the value of
4291 scm-line-incrementors; it assumes that lines are delimited with
4292 #\newline.
4293
4294 (Note that this is read-line, the function that reads a line of text
4295 from a port, not readline, the function that reads a line from a
4296 terminal, providing full editing capabilities.)
4297
4298 ** New module (ice-9 getopt-gnu-style): Parse command-line arguments.
4299
4300 This module provides some simple argument parsing. It exports one
4301 function:
4302
4303 Function: getopt-gnu-style ARG-LS
4304 Parse a list of program arguments into an alist of option
4305 descriptions.
4306
4307 Each item in the list of program arguments is examined to see if
4308 it meets the syntax of a GNU long-named option. An argument like
4309 `--MUMBLE' produces an element of the form (MUMBLE . #t) in the
4310 returned alist, where MUMBLE is a keyword object with the same
4311 name as the argument. An argument like `--MUMBLE=FROB' produces
4312 an element of the form (MUMBLE . FROB), where FROB is a string.
4313
4314 As a special case, the returned alist also contains a pair whose
4315 car is the symbol `rest'. The cdr of this pair is a list
4316 containing all the items in the argument list that are not options
4317 of the form mentioned above.
4318
4319 The argument `--' is treated specially: all items in the argument
4320 list appearing after such an argument are not examined, and are
4321 returned in the special `rest' list.
4322
4323 This function does not parse normal single-character switches.
4324 You will need to parse them out of the `rest' list yourself.
4325
4326 ** The read syntax for byte vectors and short vectors has changed.
4327
4328 Instead of #bytes(...), write #y(...).
4329
4330 Instead of #short(...), write #h(...).
4331
4332 This may seem nutty, but, like the other uniform vectors, byte vectors
4333 and short vectors want to have the same print and read syntax (and,
4334 more basic, want to have read syntax!). Changing the read syntax to
4335 use multiple characters after the hash sign breaks with the
4336 conventions used in R5RS and the conventions used for the other
4337 uniform vectors. It also introduces complexity in the current reader,
4338 both on the C and Scheme levels. (The Right solution is probably to
4339 change the syntax and prototypes for uniform vectors entirely.)
4340
4341
4342 ** The new module (ice-9 session) provides useful interactive functions.
4343
4344 *** New procedure: (apropos REGEXP OPTION ...)
4345
4346 Display a list of top-level variables whose names match REGEXP, and
4347 the modules they are imported from. Each OPTION should be one of the
4348 following symbols:
4349
4350 value --- Show the value of each matching variable.
4351 shadow --- Show bindings shadowed by subsequently imported modules.
4352 full --- Same as both `shadow' and `value'.
4353
4354 For example:
4355
4356 guile> (apropos "trace" 'full)
4357 debug: trace #<procedure trace args>
4358 debug: untrace #<procedure untrace args>
4359 the-scm-module: display-backtrace #<compiled-closure #<primitive-procedure gsubr-apply>>
4360 the-scm-module: before-backtrace-hook ()
4361 the-scm-module: backtrace #<primitive-procedure backtrace>
4362 the-scm-module: after-backtrace-hook ()
4363 the-scm-module: has-shown-backtrace-hint? #f
4364 guile>
4365
4366 ** There are new functions and syntax for working with macros.
4367
4368 Guile implements macros as a special object type. Any variable whose
4369 top-level binding is a macro object acts as a macro. The macro object
4370 specifies how the expression should be transformed before evaluation.
4371
4372 *** Macro objects now print in a reasonable way, resembling procedures.
4373
4374 *** New function: (macro? OBJ)
4375 True iff OBJ is a macro object.
4376
4377 *** New function: (primitive-macro? OBJ)
4378 Like (macro? OBJ), but true only if OBJ is one of the Guile primitive
4379 macro transformers, implemented in eval.c rather than Scheme code.
4380
4381 Why do we have this function?
4382 - For symmetry with procedure? and primitive-procedure?,
4383 - to allow custom print procedures to tell whether a macro is
4384 primitive, and display it differently, and
4385 - to allow compilers and user-written evaluators to distinguish
4386 builtin special forms from user-defined ones, which could be
4387 compiled.
4388
4389 *** New function: (macro-type OBJ)
4390 Return a value indicating what kind of macro OBJ is. Possible return
4391 values are:
4392
4393 The symbol `syntax' --- a macro created by procedure->syntax.
4394 The symbol `macro' --- a macro created by procedure->macro.
4395 The symbol `macro!' --- a macro created by procedure->memoizing-macro.
4396 The boolean #f --- if OBJ is not a macro object.
4397
4398 *** New function: (macro-name MACRO)
4399 Return the name of the macro object MACRO's procedure, as returned by
4400 procedure-name.
4401
4402 *** New function: (macro-transformer MACRO)
4403 Return the transformer procedure for MACRO.
4404
4405 *** New syntax: (use-syntax MODULE ... TRANSFORMER)
4406
4407 Specify a new macro expander to use in the current module. Each
4408 MODULE is a module name, with the same meaning as in the `use-modules'
4409 form; each named module's exported bindings are added to the current
4410 top-level environment. TRANSFORMER is an expression evaluated in the
4411 resulting environment which must yield a procedure to use as the
4412 module's eval transformer: every expression evaluated in this module
4413 is passed to this function, and the result passed to the Guile
4414 interpreter.
4415
4416 *** macro-eval! is removed. Use local-eval instead.
4417
4418 ** Some magic has been added to the printer to better handle user
4419 written printing routines (like record printers, closure printers).
4420
4421 The problem is that these user written routines must have access to
4422 the current `print-state' to be able to handle fancy things like
4423 detection of circular references. These print-states have to be
4424 passed to the builtin printing routines (display, write, etc) to
4425 properly continue the print chain.
4426
4427 We didn't want to change all existing print code so that it
4428 explicitly passes thru a print state in addition to a port. Instead,
4429 we extented the possible values that the builtin printing routines
4430 accept as a `port'. In addition to a normal port, they now also take
4431 a pair of a normal port and a print-state. Printing will go to the
4432 port and the print-state will be used to control the detection of
4433 circular references, etc. If the builtin function does not care for a
4434 print-state, it is simply ignored.
4435
4436 User written callbacks are now called with such a pair as their
4437 `port', but because every function now accepts this pair as a PORT
4438 argument, you don't have to worry about that. In fact, it is probably
4439 safest to not check for these pairs.
4440
4441 However, it is sometimes necessary to continue a print chain on a
4442 different port, for example to get a intermediate string
4443 representation of the printed value, mangle that string somehow, and
4444 then to finally print the mangled string. Use the new function
4445
4446 inherit-print-state OLD-PORT NEW-PORT
4447
4448 for this. It constructs a new `port' that prints to NEW-PORT but
4449 inherits the print-state of OLD-PORT.
4450
4451 ** struct-vtable-offset renamed to vtable-offset-user
4452
4453 ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer
4454
4455 ** There is now a third optional argument to make-vtable-vtable
4456 (and fourth to make-struct) when constructing new types (vtables).
4457 This argument initializes field vtable-index-printer of the vtable.
4458
4459 ** The detection of circular references has been extended to structs.
4460 That is, a structure that -- in the process of being printed -- prints
4461 itself does not lead to infinite recursion.
4462
4463 ** There is now some basic support for fluids. Please read
4464 "libguile/fluid.h" to find out more. It is accessible from Scheme with
4465 the following functions and macros:
4466
4467 Function: make-fluid
4468
4469 Create a new fluid object. Fluids are not special variables or
4470 some other extension to the semantics of Scheme, but rather
4471 ordinary Scheme objects. You can store them into variables (that
4472 are still lexically scoped, of course) or into any other place you
4473 like. Every fluid has a initial value of `#f'.
4474
4475 Function: fluid? OBJ
4476
4477 Test whether OBJ is a fluid.
4478
4479 Function: fluid-ref FLUID
4480 Function: fluid-set! FLUID VAL
4481
4482 Access/modify the fluid FLUID. Modifications are only visible
4483 within the current dynamic root (that includes threads).
4484
4485 Function: with-fluids* FLUIDS VALUES THUNK
4486
4487 FLUIDS is a list of fluids and VALUES a corresponding list of
4488 values for these fluids. Before THUNK gets called the values are
4489 installed in the fluids and the old values of the fluids are
4490 saved in the VALUES list. When the flow of control leaves THUNK
4491 or reenters it, the values get swapped again. You might think of
4492 this as a `safe-fluid-excursion'. Note that the VALUES list is
4493 modified by `with-fluids*'.
4494
4495 Macro: with-fluids ((FLUID VALUE) ...) FORM ...
4496
4497 The same as `with-fluids*' but with a different syntax. It looks
4498 just like `let', but both FLUID and VALUE are evaluated. Remember,
4499 fluids are not special variables but ordinary objects. FLUID
4500 should evaluate to a fluid.
4501
4502 ** Changes to system call interfaces:
4503
4504 *** close-port, close-input-port and close-output-port now return a
4505 boolean instead of an `unspecified' object. #t means that the port
4506 was successfully closed, while #f means it was already closed. It is
4507 also now possible for these procedures to raise an exception if an
4508 error occurs (some errors from write can be delayed until close.)
4509
4510 *** the first argument to chmod, fcntl, ftell and fseek can now be a
4511 file descriptor.
4512
4513 *** the third argument to fcntl is now optional.
4514
4515 *** the first argument to chown can now be a file descriptor or a port.
4516
4517 *** the argument to stat can now be a port.
4518
4519 *** The following new procedures have been added (most use scsh
4520 interfaces):
4521
4522 *** procedure: close PORT/FD
4523 Similar to close-port (*note close-port: Closing Ports.), but also
4524 works on file descriptors. A side effect of closing a file
4525 descriptor is that any ports using that file descriptor are moved
4526 to a different file descriptor and have their revealed counts set
4527 to zero.
4528
4529 *** procedure: port->fdes PORT
4530 Returns the integer file descriptor underlying PORT. As a side
4531 effect the revealed count of PORT is incremented.
4532
4533 *** procedure: fdes->ports FDES
4534 Returns a list of existing ports which have FDES as an underlying
4535 file descriptor, without changing their revealed counts.
4536
4537 *** procedure: fdes->inport FDES
4538 Returns an existing input port which has FDES as its underlying
4539 file descriptor, if one exists, and increments its revealed count.
4540 Otherwise, returns a new input port with a revealed count of 1.
4541
4542 *** procedure: fdes->outport FDES
4543 Returns an existing output port which has FDES as its underlying
4544 file descriptor, if one exists, and increments its revealed count.
4545 Otherwise, returns a new output port with a revealed count of 1.
4546
4547 The next group of procedures perform a `dup2' system call, if NEWFD
4548 (an integer) is supplied, otherwise a `dup'. The file descriptor to be
4549 duplicated can be supplied as an integer or contained in a port. The
4550 type of value returned varies depending on which procedure is used.
4551
4552 All procedures also have the side effect when performing `dup2' that
4553 any ports using NEWFD are moved to a different file descriptor and have
4554 their revealed counts set to zero.
4555
4556 *** procedure: dup->fdes PORT/FD [NEWFD]
4557 Returns an integer file descriptor.
4558
4559 *** procedure: dup->inport PORT/FD [NEWFD]
4560 Returns a new input port using the new file descriptor.
4561
4562 *** procedure: dup->outport PORT/FD [NEWFD]
4563 Returns a new output port using the new file descriptor.
4564
4565 *** procedure: dup PORT/FD [NEWFD]
4566 Returns a new port if PORT/FD is a port, with the same mode as the
4567 supplied port, otherwise returns an integer file descriptor.
4568
4569 *** procedure: dup->port PORT/FD MODE [NEWFD]
4570 Returns a new port using the new file descriptor. MODE supplies a
4571 mode string for the port (*note open-file: File Ports.).
4572
4573 *** procedure: setenv NAME VALUE
4574 Modifies the environment of the current process, which is also the
4575 default environment inherited by child processes.
4576
4577 If VALUE is `#f', then NAME is removed from the environment.
4578 Otherwise, the string NAME=VALUE is added to the environment,
4579 replacing any existing string with name matching NAME.
4580
4581 The return value is unspecified.
4582
4583 *** procedure: truncate-file OBJ SIZE
4584 Truncates the file referred to by OBJ to at most SIZE bytes. OBJ
4585 can be a string containing a file name or an integer file
4586 descriptor or port open for output on the file. The underlying
4587 system calls are `truncate' and `ftruncate'.
4588
4589 The return value is unspecified.
4590
4591 *** procedure: setvbuf PORT MODE [SIZE]
4592 Set the buffering mode for PORT. MODE can be:
4593 `_IONBF'
4594 non-buffered
4595
4596 `_IOLBF'
4597 line buffered
4598
4599 `_IOFBF'
4600 block buffered, using a newly allocated buffer of SIZE bytes.
4601 However if SIZE is zero or unspecified, the port will be made
4602 non-buffered.
4603
4604 This procedure should not be used after I/O has been performed with
4605 the port.
4606
4607 Ports are usually block buffered by default, with a default buffer
4608 size. Procedures e.g., *Note open-file: File Ports, which accept a
4609 mode string allow `0' to be added to request an unbuffered port.
4610
4611 *** procedure: fsync PORT/FD
4612 Copies any unwritten data for the specified output file descriptor
4613 to disk. If PORT/FD is a port, its buffer is flushed before the
4614 underlying file descriptor is fsync'd. The return value is
4615 unspecified.
4616
4617 *** procedure: open-fdes PATH FLAGS [MODES]
4618 Similar to `open' but returns a file descriptor instead of a port.
4619
4620 *** procedure: execle PATH ENV [ARG] ...
4621 Similar to `execl', but the environment of the new process is
4622 specified by ENV, which must be a list of strings as returned by
4623 the `environ' procedure.
4624
4625 This procedure is currently implemented using the `execve' system
4626 call, but we call it `execle' because of its Scheme calling
4627 interface.
4628
4629 *** procedure: strerror ERRNO
4630 Returns the Unix error message corresponding to ERRNO, an integer.
4631
4632 *** procedure: primitive-exit [STATUS]
4633 Terminate the current process without unwinding the Scheme stack.
4634 This is would typically be useful after a fork. The exit status
4635 is STATUS if supplied, otherwise zero.
4636
4637 *** procedure: times
4638 Returns an object with information about real and processor time.
4639 The following procedures accept such an object as an argument and
4640 return a selected component:
4641
4642 `tms:clock'
4643 The current real time, expressed as time units relative to an
4644 arbitrary base.
4645
4646 `tms:utime'
4647 The CPU time units used by the calling process.
4648
4649 `tms:stime'
4650 The CPU time units used by the system on behalf of the
4651 calling process.
4652
4653 `tms:cutime'
4654 The CPU time units used by terminated child processes of the
4655 calling process, whose status has been collected (e.g., using
4656 `waitpid').
4657
4658 `tms:cstime'
4659 Similarly, the CPU times units used by the system on behalf of
4660 terminated child processes.
4661
4662 ** Removed: list-length
4663 ** Removed: list-append, list-append!
4664 ** Removed: list-reverse, list-reverse!
4665
4666 ** array-map renamed to array-map!
4667
4668 ** serial-array-map renamed to serial-array-map!
4669
4670 ** catch doesn't take #f as first argument any longer
4671
4672 Previously, it was possible to pass #f instead of a key to `catch'.
4673 That would cause `catch' to pass a jump buffer object to the procedure
4674 passed as second argument. The procedure could then use this jump
4675 buffer objekt as an argument to throw.
4676
4677 This mechanism has been removed since its utility doesn't motivate the
4678 extra complexity it introduces.
4679
4680 ** The `#/' notation for lists now provokes a warning message from Guile.
4681 This syntax will be removed from Guile in the near future.
4682
4683 To disable the warning message, set the GUILE_HUSH environment
4684 variable to any non-empty value.
4685
4686 ** The newline character now prints as `#\newline', following the
4687 normal Scheme notation, not `#\nl'.
4688
4689 * Changes to the gh_ interface
4690
4691 ** The gh_enter function now takes care of loading the Guile startup files.
4692 gh_enter works by calling scm_boot_guile; see the remarks below.
4693
4694 ** Function: void gh_write (SCM x)
4695
4696 Write the printed representation of the scheme object x to the current
4697 output port. Corresponds to the scheme level `write'.
4698
4699 ** gh_list_length renamed to gh_length.
4700
4701 ** vector handling routines
4702
4703 Several major changes. In particular, gh_vector() now resembles
4704 (vector ...) (with a caveat -- see manual), and gh_make_vector() now
4705 exists and behaves like (make-vector ...). gh_vset() and gh_vref()
4706 have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing
4707 vector-related gh_ functions have been implemented.
4708
4709 ** pair and list routines
4710
4711 Implemented several of the R4RS pair and list functions that were
4712 missing.
4713
4714 ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect
4715
4716 New function. Converts double arrays back and forth between Scheme
4717 and C.
4718
4719 * Changes to the scm_ interface
4720
4721 ** The function scm_boot_guile now takes care of loading the startup files.
4722
4723 Guile's primary initialization function, scm_boot_guile, now takes
4724 care of loading `boot-9.scm', in the `ice-9' module, to initialize
4725 Guile, define the module system, and put together some standard
4726 bindings. It also loads `init.scm', which is intended to hold
4727 site-specific initialization code.
4728
4729 Since Guile cannot operate properly until boot-9.scm is loaded, there
4730 is no reason to separate loading boot-9.scm from Guile's other
4731 initialization processes.
4732
4733 This job used to be done by scm_compile_shell_switches, which didn't
4734 make much sense; in particular, it meant that people using Guile for
4735 non-shell-like applications had to jump through hoops to get Guile
4736 initialized properly.
4737
4738 ** The function scm_compile_shell_switches no longer loads the startup files.
4739 Now, Guile always loads the startup files, whenever it is initialized;
4740 see the notes above for scm_boot_guile and scm_load_startup_files.
4741
4742 ** Function: scm_load_startup_files
4743 This new function takes care of loading Guile's initialization file
4744 (`boot-9.scm'), and the site initialization file, `init.scm'. Since
4745 this is always called by the Guile initialization process, it's
4746 probably not too useful to call this yourself, but it's there anyway.
4747
4748 ** The semantics of smob marking have changed slightly.
4749
4750 The smob marking function (the `mark' member of the scm_smobfuns
4751 structure) is no longer responsible for setting the mark bit on the
4752 smob. The generic smob handling code in the garbage collector will
4753 set this bit. The mark function need only ensure that any other
4754 objects the smob refers to get marked.
4755
4756 Note that this change means that the smob's GC8MARK bit is typically
4757 already set upon entry to the mark function. Thus, marking functions
4758 which look like this:
4759
4760 {
4761 if (SCM_GC8MARKP (ptr))
4762 return SCM_BOOL_F;
4763 SCM_SETGC8MARK (ptr);
4764 ... mark objects to which the smob refers ...
4765 }
4766
4767 are now incorrect, since they will return early, and fail to mark any
4768 other objects the smob refers to. Some code in the Guile library used
4769 to work this way.
4770
4771 ** The semantics of the I/O port functions in scm_ptobfuns have changed.
4772
4773 If you have implemented your own I/O port type, by writing the
4774 functions required by the scm_ptobfuns and then calling scm_newptob,
4775 you will need to change your functions slightly.
4776
4777 The functions in a scm_ptobfuns structure now expect the port itself
4778 as their argument; they used to expect the `stream' member of the
4779 port's scm_port_table structure. This allows functions in an
4780 scm_ptobfuns structure to easily access the port's cell (and any flags
4781 it its CAR), and the port's scm_port_table structure.
4782
4783 Guile now passes the I/O port itself as the `port' argument in the
4784 following scm_ptobfuns functions:
4785
4786 int (*free) (SCM port);
4787 int (*fputc) (int, SCM port);
4788 int (*fputs) (char *, SCM port);
4789 scm_sizet (*fwrite) SCM_P ((char *ptr,
4790 scm_sizet size,
4791 scm_sizet nitems,
4792 SCM port));
4793 int (*fflush) (SCM port);
4794 int (*fgetc) (SCM port);
4795 int (*fclose) (SCM port);
4796
4797 The interfaces to the `mark', `print', `equalp', and `fgets' methods
4798 are unchanged.
4799
4800 If you have existing code which defines its own port types, it is easy
4801 to convert your code to the new interface; simply apply SCM_STREAM to
4802 the port argument to yield the value you code used to expect.
4803
4804 Note that since both the port and the stream have the same type in the
4805 C code --- they are both SCM values --- the C compiler will not remind
4806 you if you forget to update your scm_ptobfuns functions.
4807
4808
4809 ** Function: int scm_internal_select (int fds,
4810 SELECT_TYPE *rfds,
4811 SELECT_TYPE *wfds,
4812 SELECT_TYPE *efds,
4813 struct timeval *timeout);
4814
4815 This is a replacement for the `select' function provided by the OS.
4816 It enables I/O blocking and sleeping to happen for one cooperative
4817 thread without blocking other threads. It also avoids busy-loops in
4818 these situations. It is intended that all I/O blocking and sleeping
4819 will finally go through this function. Currently, this function is
4820 only available on systems providing `gettimeofday' and `select'.
4821
4822 ** Function: SCM scm_internal_stack_catch (SCM tag,
4823 scm_catch_body_t body,
4824 void *body_data,
4825 scm_catch_handler_t handler,
4826 void *handler_data)
4827
4828 A new sibling to the other two C level `catch' functions
4829 scm_internal_catch and scm_internal_lazy_catch. Use it if you want
4830 the stack to be saved automatically into the variable `the-last-stack'
4831 (scm_the_last_stack_var) on error. This is necessary if you want to
4832 use advanced error reporting, such as calling scm_display_error and
4833 scm_display_backtrace. (They both take a stack object as argument.)
4834
4835 ** Function: SCM scm_spawn_thread (scm_catch_body_t body,
4836 void *body_data,
4837 scm_catch_handler_t handler,
4838 void *handler_data)
4839
4840 Spawns a new thread. It does a job similar to
4841 scm_call_with_new_thread but takes arguments more suitable when
4842 spawning threads from application C code.
4843
4844 ** The hook scm_error_callback has been removed. It was originally
4845 intended as a way for the user to install his own error handler. But
4846 that method works badly since it intervenes between throw and catch,
4847 thereby changing the semantics of expressions like (catch #t ...).
4848 The correct way to do it is to use one of the C level catch functions
4849 in throw.c: scm_internal_catch/lazy_catch/stack_catch.
4850
4851 ** Removed functions:
4852
4853 scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x,
4854 scm_list_reverse, scm_list_reverse_x
4855
4856 ** New macros: SCM_LISTn where n is one of the integers 0-9.
4857
4858 These can be used for pretty list creation from C. The idea is taken
4859 from Erick Gallesio's STk.
4860
4861 ** scm_array_map renamed to scm_array_map_x
4862
4863 ** mbstrings are now removed
4864
4865 This means that the type codes scm_tc7_mb_string and
4866 scm_tc7_mb_substring has been removed.
4867
4868 ** scm_gen_putc, scm_gen_puts, scm_gen_write, and scm_gen_getc have changed.
4869
4870 Since we no longer support multi-byte strings, these I/O functions
4871 have been simplified, and renamed. Here are their old names, and
4872 their new names and arguments:
4873
4874 scm_gen_putc -> void scm_putc (int c, SCM port);
4875 scm_gen_puts -> void scm_puts (char *s, SCM port);
4876 scm_gen_write -> void scm_lfwrite (char *ptr, scm_sizet size, SCM port);
4877 scm_gen_getc -> void scm_getc (SCM port);
4878
4879
4880 ** The macros SCM_TYP7D and SCM_TYP7SD has been removed.
4881
4882 ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D
4883
4884 SCM_TYP7S now masks away the bit which distinguishes substrings from
4885 strings.
4886
4887 ** scm_catch_body_t: Backward incompatible change!
4888
4889 Body functions to scm_internal_catch and friends do not any longer
4890 take a second argument. This is because it is no longer possible to
4891 pass a #f arg to catch.
4892
4893 ** Calls to scm_protect_object and scm_unprotect now nest properly.
4894
4895 The function scm_protect_object protects its argument from being freed
4896 by the garbage collector. scm_unprotect_object removes that
4897 protection.
4898
4899 These functions now nest properly. That is, for every object O, there
4900 is a counter which scm_protect_object(O) increments and
4901 scm_unprotect_object(O) decrements, if the counter is greater than
4902 zero. Every object's counter is zero when it is first created. If an
4903 object's counter is greater than zero, the garbage collector will not
4904 reclaim its storage.
4905
4906 This allows you to use scm_protect_object in your code without
4907 worrying that some other function you call will call
4908 scm_unprotect_object, and allow it to be freed. Assuming that the
4909 functions you call are well-behaved, and unprotect only those objects
4910 they protect, you can follow the same rule and have confidence that
4911 objects will be freed only at appropriate times.
4912
4913 \f
4914 Changes in Guile 1.2 (released Tuesday, June 24 1997):
4915
4916 * Changes to the distribution
4917
4918 ** Nightly snapshots are now available from ftp.red-bean.com.
4919 The old server, ftp.cyclic.com, has been relinquished to its rightful
4920 owner.
4921
4922 Nightly snapshots of the Guile development sources are now available via
4923 anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz.
4924
4925 Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz
4926 For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz
4927
4928 ** To run Guile without installing it, the procedure has changed a bit.
4929
4930 If you used a separate build directory to compile Guile, you'll need
4931 to include the build directory in SCHEME_LOAD_PATH, as well as the
4932 source directory. See the `INSTALL' file for examples.
4933
4934 * Changes to the procedure for linking libguile with your programs
4935
4936 ** The standard Guile load path for Scheme code now includes
4937 $(datadir)/guile (usually /usr/local/share/guile). This means that
4938 you can install your own Scheme files there, and Guile will find them.
4939 (Previous versions of Guile only checked a directory whose name
4940 contained the Guile version number, so you had to re-install or move
4941 your Scheme sources each time you installed a fresh version of Guile.)
4942
4943 The load path also includes $(datadir)/guile/site; we recommend
4944 putting individual Scheme files there. If you want to install a
4945 package with multiple source files, create a directory for them under
4946 $(datadir)/guile.
4947
4948 ** Guile 1.2 will now use the Rx regular expression library, if it is
4949 installed on your system. When you are linking libguile into your own
4950 programs, this means you will have to link against -lguile, -lqt (if
4951 you configured Guile with thread support), and -lrx.
4952
4953 If you are using autoconf to generate configuration scripts for your
4954 application, the following lines should suffice to add the appropriate
4955 libraries to your link command:
4956
4957 ### Find Rx, quickthreads and libguile.
4958 AC_CHECK_LIB(rx, main)
4959 AC_CHECK_LIB(qt, main)
4960 AC_CHECK_LIB(guile, scm_shell)
4961
4962 The Guile 1.2 distribution does not contain sources for the Rx
4963 library, as Guile 1.0 did. If you want to use Rx, you'll need to
4964 retrieve it from a GNU FTP site and install it separately.
4965
4966 * Changes to Scheme functions and syntax
4967
4968 ** The dynamic linking features of Guile are now enabled by default.
4969 You can disable them by giving the `--disable-dynamic-linking' option
4970 to configure.
4971
4972 (dynamic-link FILENAME)
4973
4974 Find the object file denoted by FILENAME (a string) and link it
4975 into the running Guile application. When everything works out,
4976 return a Scheme object suitable for representing the linked object
4977 file. Otherwise an error is thrown. How object files are
4978 searched is system dependent.
4979
4980 (dynamic-object? VAL)
4981
4982 Determine whether VAL represents a dynamically linked object file.
4983
4984 (dynamic-unlink DYNOBJ)
4985
4986 Unlink the indicated object file from the application. DYNOBJ
4987 should be one of the values returned by `dynamic-link'.
4988
4989 (dynamic-func FUNCTION DYNOBJ)
4990
4991 Search the C function indicated by FUNCTION (a string or symbol)
4992 in DYNOBJ and return some Scheme object that can later be used
4993 with `dynamic-call' to actually call this function. Right now,
4994 these Scheme objects are formed by casting the address of the
4995 function to `long' and converting this number to its Scheme
4996 representation.
4997
4998 (dynamic-call FUNCTION DYNOBJ)
4999
5000 Call the C function indicated by FUNCTION and DYNOBJ. The
5001 function is passed no arguments and its return value is ignored.
5002 When FUNCTION is something returned by `dynamic-func', call that
5003 function and ignore DYNOBJ. When FUNCTION is a string (or symbol,
5004 etc.), look it up in DYNOBJ; this is equivalent to
5005
5006 (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f)
5007
5008 Interrupts are deferred while the C function is executing (with
5009 SCM_DEFER_INTS/SCM_ALLOW_INTS).
5010
5011 (dynamic-args-call FUNCTION DYNOBJ ARGS)
5012
5013 Call the C function indicated by FUNCTION and DYNOBJ, but pass it
5014 some arguments and return its return value. The C function is
5015 expected to take two arguments and return an `int', just like
5016 `main':
5017
5018 int c_func (int argc, char **argv);
5019
5020 ARGS must be a list of strings and is converted into an array of
5021 `char *'. The array is passed in ARGV and its size in ARGC. The
5022 return value is converted to a Scheme number and returned from the
5023 call to `dynamic-args-call'.
5024
5025 When dynamic linking is disabled or not supported on your system,
5026 the above functions throw errors, but they are still available.
5027
5028 Here is a small example that works on GNU/Linux:
5029
5030 (define libc-obj (dynamic-link "libc.so"))
5031 (dynamic-args-call 'rand libc-obj '())
5032
5033 See the file `libguile/DYNAMIC-LINKING' for additional comments.
5034
5035 ** The #/ syntax for module names is depreciated, and will be removed
5036 in a future version of Guile. Instead of
5037
5038 #/foo/bar/baz
5039
5040 instead write
5041
5042 (foo bar baz)
5043
5044 The latter syntax is more consistent with existing Lisp practice.
5045
5046 ** Guile now does fancier printing of structures. Structures are the
5047 underlying implementation for records, which in turn are used to
5048 implement modules, so all of these object now print differently and in
5049 a more informative way.
5050
5051 The Scheme printer will examine the builtin variable *struct-printer*
5052 whenever it needs to print a structure object. When this variable is
5053 not `#f' it is deemed to be a procedure and will be applied to the
5054 structure object and the output port. When *struct-printer* is `#f'
5055 or the procedure return `#f' the structure object will be printed in
5056 the boring #<struct 80458270> form.
5057
5058 This hook is used by some routines in ice-9/boot-9.scm to implement
5059 type specific printing routines. Please read the comments there about
5060 "printing structs".
5061
5062 One of the more specific uses of structs are records. The printing
5063 procedure that could be passed to MAKE-RECORD-TYPE is now actually
5064 called. It should behave like a *struct-printer* procedure (described
5065 above).
5066
5067 ** Guile now supports a new R4RS-compliant syntax for keywords. A
5068 token of the form #:NAME, where NAME has the same syntax as a Scheme
5069 symbol, is the external representation of the keyword named NAME.
5070 Keyword objects print using this syntax as well, so values containing
5071 keyword objects can be read back into Guile. When used in an
5072 expression, keywords are self-quoting objects.
5073
5074 Guile suports this read syntax, and uses this print syntax, regardless
5075 of the current setting of the `keyword' read option. The `keyword'
5076 read option only controls whether Guile recognizes the `:NAME' syntax,
5077 which is incompatible with R4RS. (R4RS says such token represent
5078 symbols.)
5079
5080 ** Guile has regular expression support again. Guile 1.0 included
5081 functions for matching regular expressions, based on the Rx library.
5082 In Guile 1.1, the Guile/Rx interface was removed to simplify the
5083 distribution, and thus Guile had no regular expression support. Guile
5084 1.2 again supports the most commonly used functions, and supports all
5085 of SCSH's regular expression functions.
5086
5087 If your system does not include a POSIX regular expression library,
5088 and you have not linked Guile with a third-party regexp library such as
5089 Rx, these functions will not be available. You can tell whether your
5090 Guile installation includes regular expression support by checking
5091 whether the `*features*' list includes the `regex' symbol.
5092
5093 *** regexp functions
5094
5095 By default, Guile supports POSIX extended regular expressions. That
5096 means that the characters `(', `)', `+' and `?' are special, and must
5097 be escaped if you wish to match the literal characters.
5098
5099 This regular expression interface was modeled after that implemented
5100 by SCSH, the Scheme Shell. It is intended to be upwardly compatible
5101 with SCSH regular expressions.
5102
5103 **** Function: string-match PATTERN STR [START]
5104 Compile the string PATTERN into a regular expression and compare
5105 it with STR. The optional numeric argument START specifies the
5106 position of STR at which to begin matching.
5107
5108 `string-match' returns a "match structure" which describes what,
5109 if anything, was matched by the regular expression. *Note Match
5110 Structures::. If STR does not match PATTERN at all,
5111 `string-match' returns `#f'.
5112
5113 Each time `string-match' is called, it must compile its PATTERN
5114 argument into a regular expression structure. This operation is
5115 expensive, which makes `string-match' inefficient if the same regular
5116 expression is used several times (for example, in a loop). For better
5117 performance, you can compile a regular expression in advance and then
5118 match strings against the compiled regexp.
5119
5120 **** Function: make-regexp STR [FLAGS]
5121 Compile the regular expression described by STR, and return the
5122 compiled regexp structure. If STR does not describe a legal
5123 regular expression, `make-regexp' throws a
5124 `regular-expression-syntax' error.
5125
5126 FLAGS may be the bitwise-or of one or more of the following:
5127
5128 **** Constant: regexp/extended
5129 Use POSIX Extended Regular Expression syntax when interpreting
5130 STR. If not set, POSIX Basic Regular Expression syntax is used.
5131 If the FLAGS argument is omitted, we assume regexp/extended.
5132
5133 **** Constant: regexp/icase
5134 Do not differentiate case. Subsequent searches using the
5135 returned regular expression will be case insensitive.
5136
5137 **** Constant: regexp/newline
5138 Match-any-character operators don't match a newline.
5139
5140 A non-matching list ([^...]) not containing a newline matches a
5141 newline.
5142
5143 Match-beginning-of-line operator (^) matches the empty string
5144 immediately after a newline, regardless of whether the FLAGS
5145 passed to regexp-exec contain regexp/notbol.
5146
5147 Match-end-of-line operator ($) matches the empty string
5148 immediately before a newline, regardless of whether the FLAGS
5149 passed to regexp-exec contain regexp/noteol.
5150
5151 **** Function: regexp-exec REGEXP STR [START [FLAGS]]
5152 Match the compiled regular expression REGEXP against `str'. If
5153 the optional integer START argument is provided, begin matching
5154 from that position in the string. Return a match structure
5155 describing the results of the match, or `#f' if no match could be
5156 found.
5157
5158 FLAGS may be the bitwise-or of one or more of the following:
5159
5160 **** Constant: regexp/notbol
5161 The match-beginning-of-line operator always fails to match (but
5162 see the compilation flag regexp/newline above) This flag may be
5163 used when different portions of a string are passed to
5164 regexp-exec and the beginning of the string should not be
5165 interpreted as the beginning of the line.
5166
5167 **** Constant: regexp/noteol
5168 The match-end-of-line operator always fails to match (but see the
5169 compilation flag regexp/newline above)
5170
5171 **** Function: regexp? OBJ
5172 Return `#t' if OBJ is a compiled regular expression, or `#f'
5173 otherwise.
5174
5175 Regular expressions are commonly used to find patterns in one string
5176 and replace them with the contents of another string.
5177
5178 **** Function: regexp-substitute PORT MATCH [ITEM...]
5179 Write to the output port PORT selected contents of the match
5180 structure MATCH. Each ITEM specifies what should be written, and
5181 may be one of the following arguments:
5182
5183 * A string. String arguments are written out verbatim.
5184
5185 * An integer. The submatch with that number is written.
5186
5187 * The symbol `pre'. The portion of the matched string preceding
5188 the regexp match is written.
5189
5190 * The symbol `post'. The portion of the matched string
5191 following the regexp match is written.
5192
5193 PORT may be `#f', in which case nothing is written; instead,
5194 `regexp-substitute' constructs a string from the specified ITEMs
5195 and returns that.
5196
5197 **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...]
5198 Similar to `regexp-substitute', but can be used to perform global
5199 substitutions on STR. Instead of taking a match structure as an
5200 argument, `regexp-substitute/global' takes two string arguments: a
5201 REGEXP string describing a regular expression, and a TARGET string
5202 which should be matched against this regular expression.
5203
5204 Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following
5205 exceptions:
5206
5207 * A function may be supplied. When this function is called, it
5208 will be passed one argument: a match structure for a given
5209 regular expression match. It should return a string to be
5210 written out to PORT.
5211
5212 * The `post' symbol causes `regexp-substitute/global' to recurse
5213 on the unmatched portion of STR. This *must* be supplied in
5214 order to perform global search-and-replace on STR; if it is
5215 not present among the ITEMs, then `regexp-substitute/global'
5216 will return after processing a single match.
5217
5218 *** Match Structures
5219
5220 A "match structure" is the object returned by `string-match' and
5221 `regexp-exec'. It describes which portion of a string, if any, matched
5222 the given regular expression. Match structures include: a reference to
5223 the string that was checked for matches; the starting and ending
5224 positions of the regexp match; and, if the regexp included any
5225 parenthesized subexpressions, the starting and ending positions of each
5226 submatch.
5227
5228 In each of the regexp match functions described below, the `match'
5229 argument must be a match structure returned by a previous call to
5230 `string-match' or `regexp-exec'. Most of these functions return some
5231 information about the original target string that was matched against a
5232 regular expression; we will call that string TARGET for easy reference.
5233
5234 **** Function: regexp-match? OBJ
5235 Return `#t' if OBJ is a match structure returned by a previous
5236 call to `regexp-exec', or `#f' otherwise.
5237
5238 **** Function: match:substring MATCH [N]
5239 Return the portion of TARGET matched by subexpression number N.
5240 Submatch 0 (the default) represents the entire regexp match. If
5241 the regular expression as a whole matched, but the subexpression
5242 number N did not match, return `#f'.
5243
5244 **** Function: match:start MATCH [N]
5245 Return the starting position of submatch number N.
5246
5247 **** Function: match:end MATCH [N]
5248 Return the ending position of submatch number N.
5249
5250 **** Function: match:prefix MATCH
5251 Return the unmatched portion of TARGET preceding the regexp match.
5252
5253 **** Function: match:suffix MATCH
5254 Return the unmatched portion of TARGET following the regexp match.
5255
5256 **** Function: match:count MATCH
5257 Return the number of parenthesized subexpressions from MATCH.
5258 Note that the entire regular expression match itself counts as a
5259 subexpression, and failed submatches are included in the count.
5260
5261 **** Function: match:string MATCH
5262 Return the original TARGET string.
5263
5264 *** Backslash Escapes
5265
5266 Sometimes you will want a regexp to match characters like `*' or `$'
5267 exactly. For example, to check whether a particular string represents
5268 a menu entry from an Info node, it would be useful to match it against
5269 a regexp like `^* [^:]*::'. However, this won't work; because the
5270 asterisk is a metacharacter, it won't match the `*' at the beginning of
5271 the string. In this case, we want to make the first asterisk un-magic.
5272
5273 You can do this by preceding the metacharacter with a backslash
5274 character `\'. (This is also called "quoting" the metacharacter, and
5275 is known as a "backslash escape".) When Guile sees a backslash in a
5276 regular expression, it considers the following glyph to be an ordinary
5277 character, no matter what special meaning it would ordinarily have.
5278 Therefore, we can make the above example work by changing the regexp to
5279 `^\* [^:]*::'. The `\*' sequence tells the regular expression engine
5280 to match only a single asterisk in the target string.
5281
5282 Since the backslash is itself a metacharacter, you may force a
5283 regexp to match a backslash in the target string by preceding the
5284 backslash with itself. For example, to find variable references in a
5285 TeX program, you might want to find occurrences of the string `\let\'
5286 followed by any number of alphabetic characters. The regular expression
5287 `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp
5288 each match a single backslash in the target string.
5289
5290 **** Function: regexp-quote STR
5291 Quote each special character found in STR with a backslash, and
5292 return the resulting string.
5293
5294 *Very important:* Using backslash escapes in Guile source code (as
5295 in Emacs Lisp or C) can be tricky, because the backslash character has
5296 special meaning for the Guile reader. For example, if Guile encounters
5297 the character sequence `\n' in the middle of a string while processing
5298 Scheme code, it replaces those characters with a newline character.
5299 Similarly, the character sequence `\t' is replaced by a horizontal tab.
5300 Several of these "escape sequences" are processed by the Guile reader
5301 before your code is executed. Unrecognized escape sequences are
5302 ignored: if the characters `\*' appear in a string, they will be
5303 translated to the single character `*'.
5304
5305 This translation is obviously undesirable for regular expressions,
5306 since we want to be able to include backslashes in a string in order to
5307 escape regexp metacharacters. Therefore, to make sure that a backslash
5308 is preserved in a string in your Guile program, you must use *two*
5309 consecutive backslashes:
5310
5311 (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*"))
5312
5313 The string in this example is preprocessed by the Guile reader before
5314 any code is executed. The resulting argument to `make-regexp' is the
5315 string `^\* [^:]*', which is what we really want.
5316
5317 This also means that in order to write a regular expression that
5318 matches a single backslash character, the regular expression string in
5319 the source code must include *four* backslashes. Each consecutive pair
5320 of backslashes gets translated by the Guile reader to a single
5321 backslash, and the resulting double-backslash is interpreted by the
5322 regexp engine as matching a single backslash character. Hence:
5323
5324 (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*"))
5325
5326 The reason for the unwieldiness of this syntax is historical. Both
5327 regular expression pattern matchers and Unix string processing systems
5328 have traditionally used backslashes with the special meanings described
5329 above. The POSIX regular expression specification and ANSI C standard
5330 both require these semantics. Attempting to abandon either convention
5331 would cause other kinds of compatibility problems, possibly more severe
5332 ones. Therefore, without extending the Scheme reader to support
5333 strings with different quoting conventions (an ungainly and confusing
5334 extension when implemented in other languages), we must adhere to this
5335 cumbersome escape syntax.
5336
5337 * Changes to the gh_ interface
5338
5339 * Changes to the scm_ interface
5340
5341 * Changes to system call interfaces:
5342
5343 ** The value returned by `raise' is now unspecified. It throws an exception
5344 if an error occurs.
5345
5346 *** A new procedure `sigaction' can be used to install signal handlers
5347
5348 (sigaction signum [action] [flags])
5349
5350 signum is the signal number, which can be specified using the value
5351 of SIGINT etc.
5352
5353 If action is omitted, sigaction returns a pair: the CAR is the current
5354 signal hander, which will be either an integer with the value SIG_DFL
5355 (default action) or SIG_IGN (ignore), or the Scheme procedure which
5356 handles the signal, or #f if a non-Scheme procedure handles the
5357 signal. The CDR contains the current sigaction flags for the handler.
5358
5359 If action is provided, it is installed as the new handler for signum.
5360 action can be a Scheme procedure taking one argument, or the value of
5361 SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore
5362 whatever signal handler was installed before sigaction was first used.
5363 Flags can optionally be specified for the new handler (SA_RESTART is
5364 always used if the system provides it, so need not be specified.) The
5365 return value is a pair with information about the old handler as
5366 described above.
5367
5368 This interface does not provide access to the "signal blocking"
5369 facility. Maybe this is not needed, since the thread support may
5370 provide solutions to the problem of consistent access to data
5371 structures.
5372
5373 *** A new procedure `flush-all-ports' is equivalent to running
5374 `force-output' on every port open for output.
5375
5376 ** Guile now provides information on how it was built, via the new
5377 global variable, %guile-build-info. This variable records the values
5378 of the standard GNU makefile directory variables as an assocation
5379 list, mapping variable names (symbols) onto directory paths (strings).
5380 For example, to find out where the Guile link libraries were
5381 installed, you can say:
5382
5383 guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)"
5384
5385
5386 * Changes to the scm_ interface
5387
5388 ** The new function scm_handle_by_message_noexit is just like the
5389 existing scm_handle_by_message function, except that it doesn't call
5390 exit to terminate the process. Instead, it prints a message and just
5391 returns #f. This might be a more appropriate catch-all handler for
5392 new dynamic roots and threads.
5393
5394 \f
5395 Changes in Guile 1.1 (released Friday, May 16 1997):
5396
5397 * Changes to the distribution.
5398
5399 The Guile 1.0 distribution has been split up into several smaller
5400 pieces:
5401 guile-core --- the Guile interpreter itself.
5402 guile-tcltk --- the interface between the Guile interpreter and
5403 Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk
5404 is a toolkit for building graphical user interfaces.
5405 guile-rgx-ctax --- the interface between Guile and the Rx regular
5406 expression matcher, and the translator for the Ctax
5407 programming language. These are packaged together because the
5408 Ctax translator uses Rx to parse Ctax source code.
5409
5410 This NEWS file describes the changes made to guile-core since the 1.0
5411 release.
5412
5413 We no longer distribute the documentation, since it was either out of
5414 date, or incomplete. As soon as we have current documentation, we
5415 will distribute it.
5416
5417
5418
5419 * Changes to the stand-alone interpreter
5420
5421 ** guile now accepts command-line arguments compatible with SCSH, Olin
5422 Shivers' Scheme Shell.
5423
5424 In general, arguments are evaluated from left to right, but there are
5425 exceptions. The following switches stop argument processing, and
5426 stash all remaining command-line arguments as the value returned by
5427 the (command-line) function.
5428 -s SCRIPT load Scheme source code from FILE, and exit
5429 -c EXPR evalute Scheme expression EXPR, and exit
5430 -- stop scanning arguments; run interactively
5431
5432 The switches below are processed as they are encountered.
5433 -l FILE load Scheme source code from FILE
5434 -e FUNCTION after reading script, apply FUNCTION to
5435 command line arguments
5436 -ds do -s script at this point
5437 --emacs enable Emacs protocol (experimental)
5438 -h, --help display this help and exit
5439 -v, --version display version information and exit
5440 \ read arguments from following script lines
5441
5442 So, for example, here is a Guile script named `ekko' (thanks, Olin)
5443 which re-implements the traditional "echo" command:
5444
5445 #!/usr/local/bin/guile -s
5446 !#
5447 (define (main args)
5448 (map (lambda (arg) (display arg) (display " "))
5449 (cdr args))
5450 (newline))
5451
5452 (main (command-line))
5453
5454 Suppose we invoke this script as follows:
5455
5456 ekko a speckled gecko
5457
5458 Through the magic of Unix script processing (triggered by the `#!'
5459 token at the top of the file), /usr/local/bin/guile receives the
5460 following list of command-line arguments:
5461
5462 ("-s" "./ekko" "a" "speckled" "gecko")
5463
5464 Unix inserts the name of the script after the argument specified on
5465 the first line of the file (in this case, "-s"), and then follows that
5466 with the arguments given to the script. Guile loads the script, which
5467 defines the `main' function, and then applies it to the list of
5468 remaining command-line arguments, ("a" "speckled" "gecko").
5469
5470 In Unix, the first line of a script file must take the following form:
5471
5472 #!INTERPRETER ARGUMENT
5473
5474 where INTERPRETER is the absolute filename of the interpreter
5475 executable, and ARGUMENT is a single command-line argument to pass to
5476 the interpreter.
5477
5478 You may only pass one argument to the interpreter, and its length is
5479 limited. These restrictions can be annoying to work around, so Guile
5480 provides a general mechanism (borrowed from, and compatible with,
5481 SCSH) for circumventing them.
5482
5483 If the ARGUMENT in a Guile script is a single backslash character,
5484 `\', Guile will open the script file, parse arguments from its second
5485 and subsequent lines, and replace the `\' with them. So, for example,
5486 here is another implementation of the `ekko' script:
5487
5488 #!/usr/local/bin/guile \
5489 -e main -s
5490 !#
5491 (define (main args)
5492 (for-each (lambda (arg) (display arg) (display " "))
5493 (cdr args))
5494 (newline))
5495
5496 If the user invokes this script as follows:
5497
5498 ekko a speckled gecko
5499
5500 Unix expands this into
5501
5502 /usr/local/bin/guile \ ekko a speckled gecko
5503
5504 When Guile sees the `\' argument, it replaces it with the arguments
5505 read from the second line of the script, producing:
5506
5507 /usr/local/bin/guile -e main -s ekko a speckled gecko
5508
5509 This tells Guile to load the `ekko' script, and apply the function
5510 `main' to the argument list ("a" "speckled" "gecko").
5511
5512 Here is how Guile parses the command-line arguments:
5513 - Each space character terminates an argument. This means that two
5514 spaces in a row introduce an empty-string argument.
5515 - The tab character is not permitted (unless you quote it with the
5516 backslash character, as described below), to avoid confusion.
5517 - The newline character terminates the sequence of arguments, and will
5518 also terminate a final non-empty argument. (However, a newline
5519 following a space will not introduce a final empty-string argument;
5520 it only terminates the argument list.)
5521 - The backslash character is the escape character. It escapes
5522 backslash, space, tab, and newline. The ANSI C escape sequences
5523 like \n and \t are also supported. These produce argument
5524 constituents; the two-character combination \n doesn't act like a
5525 terminating newline. The escape sequence \NNN for exactly three
5526 octal digits reads as the character whose ASCII code is NNN. As
5527 above, characters produced this way are argument constituents.
5528 Backslash followed by other characters is not allowed.
5529
5530 * Changes to the procedure for linking libguile with your programs
5531
5532 ** Guile now builds and installs a shared guile library, if your
5533 system support shared libraries. (It still builds a static library on
5534 all systems.) Guile automatically detects whether your system
5535 supports shared libraries. To prevent Guile from buildisg shared
5536 libraries, pass the `--disable-shared' flag to the configure script.
5537
5538 Guile takes longer to compile when it builds shared libraries, because
5539 it must compile every file twice --- once to produce position-
5540 independent object code, and once to produce normal object code.
5541
5542 ** The libthreads library has been merged into libguile.
5543
5544 To link a program against Guile, you now need only link against
5545 -lguile and -lqt; -lthreads is no longer needed. If you are using
5546 autoconf to generate configuration scripts for your application, the
5547 following lines should suffice to add the appropriate libraries to
5548 your link command:
5549
5550 ### Find quickthreads and libguile.
5551 AC_CHECK_LIB(qt, main)
5552 AC_CHECK_LIB(guile, scm_shell)
5553
5554 * Changes to Scheme functions
5555
5556 ** Guile Scheme's special syntax for keyword objects is now optional,
5557 and disabled by default.
5558
5559 The syntax variation from R4RS made it difficult to port some
5560 interesting packages to Guile. The routines which accepted keyword
5561 arguments (mostly in the module system) have been modified to also
5562 accept symbols whose names begin with `:'.
5563
5564 To change the keyword syntax, you must first import the (ice-9 debug)
5565 module:
5566 (use-modules (ice-9 debug))
5567
5568 Then you can enable the keyword syntax as follows:
5569 (read-set! keywords 'prefix)
5570
5571 To disable keyword syntax, do this:
5572 (read-set! keywords #f)
5573
5574 ** Many more primitive functions accept shared substrings as
5575 arguments. In the past, these functions required normal, mutable
5576 strings as arguments, although they never made use of this
5577 restriction.
5578
5579 ** The uniform array functions now operate on byte vectors. These
5580 functions are `array-fill!', `serial-array-copy!', `array-copy!',
5581 `serial-array-map', `array-map', `array-for-each', and
5582 `array-index-map!'.
5583
5584 ** The new functions `trace' and `untrace' implement simple debugging
5585 support for Scheme functions.
5586
5587 The `trace' function accepts any number of procedures as arguments,
5588 and tells the Guile interpreter to display each procedure's name and
5589 arguments each time the procedure is invoked. When invoked with no
5590 arguments, `trace' returns the list of procedures currently being
5591 traced.
5592
5593 The `untrace' function accepts any number of procedures as arguments,
5594 and tells the Guile interpreter not to trace them any more. When
5595 invoked with no arguments, `untrace' untraces all curretly traced
5596 procedures.
5597
5598 The tracing in Guile has an advantage over most other systems: we
5599 don't create new procedure objects, but mark the procedure objects
5600 themselves. This means that anonymous and internal procedures can be
5601 traced.
5602
5603 ** The function `assert-repl-prompt' has been renamed to
5604 `set-repl-prompt!'. It takes one argument, PROMPT.
5605 - If PROMPT is #f, the Guile read-eval-print loop will not prompt.
5606 - If PROMPT is a string, we use it as a prompt.
5607 - If PROMPT is a procedure accepting no arguments, we call it, and
5608 display the result as a prompt.
5609 - Otherwise, we display "> ".
5610
5611 ** The new function `eval-string' reads Scheme expressions from a
5612 string and evaluates them, returning the value of the last expression
5613 in the string. If the string contains no expressions, it returns an
5614 unspecified value.
5615
5616 ** The new function `thunk?' returns true iff its argument is a
5617 procedure of zero arguments.
5618
5619 ** `defined?' is now a builtin function, instead of syntax. This
5620 means that its argument should be quoted. It returns #t iff its
5621 argument is bound in the current module.
5622
5623 ** The new syntax `use-modules' allows you to add new modules to your
5624 environment without re-typing a complete `define-module' form. It
5625 accepts any number of module names as arguments, and imports their
5626 public bindings into the current module.
5627
5628 ** The new function (module-defined? NAME MODULE) returns true iff
5629 NAME, a symbol, is defined in MODULE, a module object.
5630
5631 ** The new function `builtin-bindings' creates and returns a hash
5632 table containing copies of all the root module's bindings.
5633
5634 ** The new function `builtin-weak-bindings' does the same as
5635 `builtin-bindings', but creates a doubly-weak hash table.
5636
5637 ** The `equal?' function now considers variable objects to be
5638 equivalent if they have the same name and the same value.
5639
5640 ** The new function `command-line' returns the command-line arguments
5641 given to Guile, as a list of strings.
5642
5643 When using guile as a script interpreter, `command-line' returns the
5644 script's arguments; those processed by the interpreter (like `-s' or
5645 `-c') are omitted. (In other words, you get the normal, expected
5646 behavior.) Any application that uses scm_shell to process its
5647 command-line arguments gets this behavior as well.
5648
5649 ** The new function `load-user-init' looks for a file called `.guile'
5650 in the user's home directory, and loads it if it exists. This is
5651 mostly for use by the code generated by scm_compile_shell_switches,
5652 but we thought it might also be useful in other circumstances.
5653
5654 ** The new function `log10' returns the base-10 logarithm of its
5655 argument.
5656
5657 ** Changes to I/O functions
5658
5659 *** The functions `read', `primitive-load', `read-and-eval!', and
5660 `primitive-load-path' no longer take optional arguments controlling
5661 case insensitivity and a `#' parser.
5662
5663 Case sensitivity is now controlled by a read option called
5664 `case-insensitive'. The user can add new `#' syntaxes with the
5665 `read-hash-extend' function (see below).
5666
5667 *** The new function `read-hash-extend' allows the user to change the
5668 syntax of Guile Scheme in a somewhat controlled way.
5669
5670 (read-hash-extend CHAR PROC)
5671 When parsing S-expressions, if we read a `#' character followed by
5672 the character CHAR, use PROC to parse an object from the stream.
5673 If PROC is #f, remove any parsing procedure registered for CHAR.
5674
5675 The reader applies PROC to two arguments: CHAR and an input port.
5676
5677 *** The new functions read-delimited and read-delimited! provide a
5678 general mechanism for doing delimited input on streams.
5679
5680 (read-delimited DELIMS [PORT HANDLE-DELIM])
5681 Read until we encounter one of the characters in DELIMS (a string),
5682 or end-of-file. PORT is the input port to read from; it defaults to
5683 the current input port. The HANDLE-DELIM parameter determines how
5684 the terminating character is handled; it should be one of the
5685 following symbols:
5686
5687 'trim omit delimiter from result
5688 'peek leave delimiter character in input stream
5689 'concat append delimiter character to returned value
5690 'split return a pair: (RESULT . TERMINATOR)
5691
5692 HANDLE-DELIM defaults to 'peek.
5693
5694 (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END])
5695 A side-effecting variant of `read-delimited'.
5696
5697 The data is written into the string BUF at the indices in the
5698 half-open interval [START, END); the default interval is the whole
5699 string: START = 0 and END = (string-length BUF). The values of
5700 START and END must specify a well-defined interval in BUF, i.e.
5701 0 <= START <= END <= (string-length BUF).
5702
5703 It returns NBYTES, the number of bytes read. If the buffer filled
5704 up without a delimiter character being found, it returns #f. If the
5705 port is at EOF when the read starts, it returns the EOF object.
5706
5707 If an integer is returned (i.e., the read is successfully terminated
5708 by reading a delimiter character), then the HANDLE-DELIM parameter
5709 determines how to handle the terminating character. It is described
5710 above, and defaults to 'peek.
5711
5712 (The descriptions of these functions were borrowed from the SCSH
5713 manual, by Olin Shivers and Brian Carlstrom.)
5714
5715 *** The `%read-delimited!' function is the primitive used to implement
5716 `read-delimited' and `read-delimited!'.
5717
5718 (%read-delimited! DELIMS BUF GOBBLE? [PORT START END])
5719
5720 This returns a pair of values: (TERMINATOR . NUM-READ).
5721 - TERMINATOR describes why the read was terminated. If it is a
5722 character or the eof object, then that is the value that terminated
5723 the read. If it is #f, the function filled the buffer without finding
5724 a delimiting character.
5725 - NUM-READ is the number of characters read into BUF.
5726
5727 If the read is successfully terminated by reading a delimiter
5728 character, then the gobble? parameter determines what to do with the
5729 terminating character. If true, the character is removed from the
5730 input stream; if false, the character is left in the input stream
5731 where a subsequent read operation will retrieve it. In either case,
5732 the character is also the first value returned by the procedure call.
5733
5734 (The descriptions of this function was borrowed from the SCSH manual,
5735 by Olin Shivers and Brian Carlstrom.)
5736
5737 *** The `read-line' and `read-line!' functions have changed; they now
5738 trim the terminator by default; previously they appended it to the
5739 returned string. For the old behavior, use (read-line PORT 'concat).
5740
5741 *** The functions `uniform-array-read!' and `uniform-array-write!' now
5742 take new optional START and END arguments, specifying the region of
5743 the array to read and write.
5744
5745 *** The `ungetc-char-ready?' function has been removed. We feel it's
5746 inappropriate for an interface to expose implementation details this
5747 way.
5748
5749 ** Changes to the Unix library and system call interface
5750
5751 *** The new fcntl function provides access to the Unix `fcntl' system
5752 call.
5753
5754 (fcntl PORT COMMAND VALUE)
5755 Apply COMMAND to PORT's file descriptor, with VALUE as an argument.
5756 Values for COMMAND are:
5757
5758 F_DUPFD duplicate a file descriptor
5759 F_GETFD read the descriptor's close-on-exec flag
5760 F_SETFD set the descriptor's close-on-exec flag to VALUE
5761 F_GETFL read the descriptor's flags, as set on open
5762 F_SETFL set the descriptor's flags, as set on open to VALUE
5763 F_GETOWN return the process ID of a socket's owner, for SIGIO
5764 F_SETOWN set the process that owns a socket to VALUE, for SIGIO
5765 FD_CLOEXEC not sure what this is
5766
5767 For details, see the documentation for the fcntl system call.
5768
5769 *** The arguments to `select' have changed, for compatibility with
5770 SCSH. The TIMEOUT parameter may now be non-integral, yielding the
5771 expected behavior. The MILLISECONDS parameter has been changed to
5772 MICROSECONDS, to more closely resemble the underlying system call.
5773 The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the
5774 corresponding return set will be the same.
5775
5776 *** The arguments to the `mknod' system call have changed. They are
5777 now:
5778
5779 (mknod PATH TYPE PERMS DEV)
5780 Create a new file (`node') in the file system. PATH is the name of
5781 the file to create. TYPE is the kind of file to create; it should
5782 be 'fifo, 'block-special, or 'char-special. PERMS specifies the
5783 permission bits to give the newly created file. If TYPE is
5784 'block-special or 'char-special, DEV specifies which device the
5785 special file refers to; its interpretation depends on the kind of
5786 special file being created.
5787
5788 *** The `fork' function has been renamed to `primitive-fork', to avoid
5789 clashing with various SCSH forks.
5790
5791 *** The `recv' and `recvfrom' functions have been renamed to `recv!'
5792 and `recvfrom!'. They no longer accept a size for a second argument;
5793 you must pass a string to hold the received value. They no longer
5794 return the buffer. Instead, `recv' returns the length of the message
5795 received, and `recvfrom' returns a pair containing the packet's length
5796 and originating address.
5797
5798 *** The file descriptor datatype has been removed, as have the
5799 `read-fd', `write-fd', `close', `lseek', and `dup' functions.
5800 We plan to replace these functions with a SCSH-compatible interface.
5801
5802 *** The `create' function has been removed; it's just a special case
5803 of `open'.
5804
5805 *** There are new functions to break down process termination status
5806 values. In the descriptions below, STATUS is a value returned by
5807 `waitpid'.
5808
5809 (status:exit-val STATUS)
5810 If the child process exited normally, this function returns the exit
5811 code for the child process (i.e., the value passed to exit, or
5812 returned from main). If the child process did not exit normally,
5813 this function returns #f.
5814
5815 (status:stop-sig STATUS)
5816 If the child process was suspended by a signal, this function
5817 returns the signal that suspended the child. Otherwise, it returns
5818 #f.
5819
5820 (status:term-sig STATUS)
5821 If the child process terminated abnormally, this function returns
5822 the signal that terminated the child. Otherwise, this function
5823 returns false.
5824
5825 POSIX promises that exactly one of these functions will return true on
5826 a valid STATUS value.
5827
5828 These functions are compatible with SCSH.
5829
5830 *** There are new accessors and setters for the broken-out time vectors
5831 returned by `localtime', `gmtime', and that ilk. They are:
5832
5833 Component Accessor Setter
5834 ========================= ============ ============
5835 seconds tm:sec set-tm:sec
5836 minutes tm:min set-tm:min
5837 hours tm:hour set-tm:hour
5838 day of the month tm:mday set-tm:mday
5839 month tm:mon set-tm:mon
5840 year tm:year set-tm:year
5841 day of the week tm:wday set-tm:wday
5842 day in the year tm:yday set-tm:yday
5843 daylight saving time tm:isdst set-tm:isdst
5844 GMT offset, seconds tm:gmtoff set-tm:gmtoff
5845 name of time zone tm:zone set-tm:zone
5846
5847 *** There are new accessors for the vectors returned by `uname',
5848 describing the host system:
5849
5850 Component Accessor
5851 ============================================== ================
5852 name of the operating system implementation utsname:sysname
5853 network name of this machine utsname:nodename
5854 release level of the operating system utsname:release
5855 version level of the operating system utsname:version
5856 machine hardware platform utsname:machine
5857
5858 *** There are new accessors for the vectors returned by `getpw',
5859 `getpwnam', `getpwuid', and `getpwent', describing entries from the
5860 system's user database:
5861
5862 Component Accessor
5863 ====================== =================
5864 user name passwd:name
5865 user password passwd:passwd
5866 user id passwd:uid
5867 group id passwd:gid
5868 real name passwd:gecos
5869 home directory passwd:dir
5870 shell program passwd:shell
5871
5872 *** There are new accessors for the vectors returned by `getgr',
5873 `getgrnam', `getgrgid', and `getgrent', describing entries from the
5874 system's group database:
5875
5876 Component Accessor
5877 ======================= ============
5878 group name group:name
5879 group password group:passwd
5880 group id group:gid
5881 group members group:mem
5882
5883 *** There are new accessors for the vectors returned by `gethost',
5884 `gethostbyaddr', `gethostbyname', and `gethostent', describing
5885 internet hosts:
5886
5887 Component Accessor
5888 ========================= ===============
5889 official name of host hostent:name
5890 alias list hostent:aliases
5891 host address type hostent:addrtype
5892 length of address hostent:length
5893 list of addresses hostent:addr-list
5894
5895 *** There are new accessors for the vectors returned by `getnet',
5896 `getnetbyaddr', `getnetbyname', and `getnetent', describing internet
5897 networks:
5898
5899 Component Accessor
5900 ========================= ===============
5901 official name of net netent:name
5902 alias list netent:aliases
5903 net number type netent:addrtype
5904 net number netent:net
5905
5906 *** There are new accessors for the vectors returned by `getproto',
5907 `getprotobyname', `getprotobynumber', and `getprotoent', describing
5908 internet protocols:
5909
5910 Component Accessor
5911 ========================= ===============
5912 official protocol name protoent:name
5913 alias list protoent:aliases
5914 protocol number protoent:proto
5915
5916 *** There are new accessors for the vectors returned by `getserv',
5917 `getservbyname', `getservbyport', and `getservent', describing
5918 internet protocols:
5919
5920 Component Accessor
5921 ========================= ===============
5922 official service name servent:name
5923 alias list servent:aliases
5924 port number servent:port
5925 protocol to use servent:proto
5926
5927 *** There are new accessors for the sockaddr structures returned by
5928 `accept', `getsockname', `getpeername', `recvfrom!':
5929
5930 Component Accessor
5931 ======================================== ===============
5932 address format (`family') sockaddr:fam
5933 path, for file domain addresses sockaddr:path
5934 address, for internet domain addresses sockaddr:addr
5935 TCP or UDP port, for internet sockaddr:port
5936
5937 *** The `getpwent', `getgrent', `gethostent', `getnetent',
5938 `getprotoent', and `getservent' functions now return #f at the end of
5939 the user database. (They used to throw an exception.)
5940
5941 Note that calling MUMBLEent function is equivalent to calling the
5942 corresponding MUMBLE function with no arguments.
5943
5944 *** The `setpwent', `setgrent', `sethostent', `setnetent',
5945 `setprotoent', and `setservent' routines now take no arguments.
5946
5947 *** The `gethost', `getproto', `getnet', and `getserv' functions now
5948 provide more useful information when they throw an exception.
5949
5950 *** The `lnaof' function has been renamed to `inet-lnaof'.
5951
5952 *** Guile now claims to have the `current-time' feature.
5953
5954 *** The `mktime' function now takes an optional second argument ZONE,
5955 giving the time zone to use for the conversion. ZONE should be a
5956 string, in the same format as expected for the "TZ" environment variable.
5957
5958 *** The `strptime' function now returns a pair (TIME . COUNT), where
5959 TIME is the parsed time as a vector, and COUNT is the number of
5960 characters from the string left unparsed. This function used to
5961 return the remaining characters as a string.
5962
5963 *** The `gettimeofday' function has replaced the old `time+ticks' function.
5964 The return value is now (SECONDS . MICROSECONDS); the fractional
5965 component is no longer expressed in "ticks".
5966
5967 *** The `ticks/sec' constant has been removed, in light of the above change.
5968
5969 * Changes to the gh_ interface
5970
5971 ** gh_eval_str() now returns an SCM object which is the result of the
5972 evaluation
5973
5974 ** gh_scm2str() now copies the Scheme data to a caller-provided C
5975 array
5976
5977 ** gh_scm2newstr() now makes a C array, copies the Scheme data to it,
5978 and returns the array
5979
5980 ** gh_scm2str0() is gone: there is no need to distinguish
5981 null-terminated from non-null-terminated, since gh_scm2newstr() allows
5982 the user to interpret the data both ways.
5983
5984 * Changes to the scm_ interface
5985
5986 ** The new function scm_symbol_value0 provides an easy way to get a
5987 symbol's value from C code:
5988
5989 SCM scm_symbol_value0 (char *NAME)
5990 Return the value of the symbol named by the null-terminated string
5991 NAME in the current module. If the symbol named NAME is unbound in
5992 the current module, return SCM_UNDEFINED.
5993
5994 ** The new function scm_sysintern0 creates new top-level variables,
5995 without assigning them a value.
5996
5997 SCM scm_sysintern0 (char *NAME)
5998 Create a new Scheme top-level variable named NAME. NAME is a
5999 null-terminated string. Return the variable's value cell.
6000
6001 ** The function scm_internal_catch is the guts of catch. It handles
6002 all the mechanics of setting up a catch target, invoking the catch
6003 body, and perhaps invoking the handler if the body does a throw.
6004
6005 The function is designed to be usable from C code, but is general
6006 enough to implement all the semantics Guile Scheme expects from throw.
6007
6008 TAG is the catch tag. Typically, this is a symbol, but this function
6009 doesn't actually care about that.
6010
6011 BODY is a pointer to a C function which runs the body of the catch;
6012 this is the code you can throw from. We call it like this:
6013 BODY (BODY_DATA, JMPBUF)
6014 where:
6015 BODY_DATA is just the BODY_DATA argument we received; we pass it
6016 through to BODY as its first argument. The caller can make
6017 BODY_DATA point to anything useful that BODY might need.
6018 JMPBUF is the Scheme jmpbuf object corresponding to this catch,
6019 which we have just created and initialized.
6020
6021 HANDLER is a pointer to a C function to deal with a throw to TAG,
6022 should one occur. We call it like this:
6023 HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS)
6024 where
6025 HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the
6026 same idea as BODY_DATA above.
6027 THROWN_TAG is the tag that the user threw to; usually this is
6028 TAG, but it could be something else if TAG was #t (i.e., a
6029 catch-all), or the user threw to a jmpbuf.
6030 THROW_ARGS is the list of arguments the user passed to the THROW
6031 function.
6032
6033 BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA
6034 is just a pointer we pass through to HANDLER. We don't actually
6035 use either of those pointers otherwise ourselves. The idea is
6036 that, if our caller wants to communicate something to BODY or
6037 HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and
6038 HANDLER can then use. Think of it as a way to make BODY and
6039 HANDLER closures, not just functions; MUMBLE_DATA points to the
6040 enclosed variables.
6041
6042 Of course, it's up to the caller to make sure that any data a
6043 MUMBLE_DATA needs is protected from GC. A common way to do this is
6044 to make MUMBLE_DATA a pointer to data stored in an automatic
6045 structure variable; since the collector must scan the stack for
6046 references anyway, this assures that any references in MUMBLE_DATA
6047 will be found.
6048
6049 ** The new function scm_internal_lazy_catch is exactly like
6050 scm_internal_catch, except:
6051
6052 - It does not unwind the stack (this is the major difference).
6053 - If handler returns, its value is returned from the throw.
6054 - BODY always receives #f as its JMPBUF argument (since there's no
6055 jmpbuf associated with a lazy catch, because we don't unwind the
6056 stack.)
6057
6058 ** scm_body_thunk is a new body function you can pass to
6059 scm_internal_catch if you want the body to be like Scheme's `catch'
6060 --- a thunk, or a function of one argument if the tag is #f.
6061
6062 BODY_DATA is a pointer to a scm_body_thunk_data structure, which
6063 contains the Scheme procedure to invoke as the body, and the tag
6064 we're catching. If the tag is #f, then we pass JMPBUF (created by
6065 scm_internal_catch) to the body procedure; otherwise, the body gets
6066 no arguments.
6067
6068 ** scm_handle_by_proc is a new handler function you can pass to
6069 scm_internal_catch if you want the handler to act like Scheme's catch
6070 --- call a procedure with the tag and the throw arguments.
6071
6072 If the user does a throw to this catch, this function runs a handler
6073 procedure written in Scheme. HANDLER_DATA is a pointer to an SCM
6074 variable holding the Scheme procedure object to invoke. It ought to
6075 be a pointer to an automatic variable (i.e., one living on the stack),
6076 or the procedure object should be otherwise protected from GC.
6077
6078 ** scm_handle_by_message is a new handler function to use with
6079 `scm_internal_catch' if you want Guile to print a message and die.
6080 It's useful for dealing with throws to uncaught keys at the top level.
6081
6082 HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a
6083 message header to print; if zero, we use "guile" instead. That
6084 text is followed by a colon, then the message described by ARGS.
6085
6086 ** The return type of scm_boot_guile is now void; the function does
6087 not return a value, and indeed, never returns at all.
6088
6089 ** The new function scm_shell makes it easy for user applications to
6090 process command-line arguments in a way that is compatible with the
6091 stand-alone guile interpreter (which is in turn compatible with SCSH,
6092 the Scheme shell).
6093
6094 To use the scm_shell function, first initialize any guile modules
6095 linked into your application, and then call scm_shell with the values
6096 of ARGC and ARGV your `main' function received. scm_shell will add
6097 any SCSH-style meta-arguments from the top of the script file to the
6098 argument vector, and then process the command-line arguments. This
6099 generally means loading a script file or starting up an interactive
6100 command interpreter. For details, see "Changes to the stand-alone
6101 interpreter" above.
6102
6103 ** The new functions scm_get_meta_args and scm_count_argv help you
6104 implement the SCSH-style meta-argument, `\'.
6105
6106 char **scm_get_meta_args (int ARGC, char **ARGV)
6107 If the second element of ARGV is a string consisting of a single
6108 backslash character (i.e. "\\" in Scheme notation), open the file
6109 named by the following argument, parse arguments from it, and return
6110 the spliced command line. The returned array is terminated by a
6111 null pointer.
6112
6113 For details of argument parsing, see above, under "guile now accepts
6114 command-line arguments compatible with SCSH..."
6115
6116 int scm_count_argv (char **ARGV)
6117 Count the arguments in ARGV, assuming it is terminated by a null
6118 pointer.
6119
6120 For an example of how these functions might be used, see the source
6121 code for the function scm_shell in libguile/script.c.
6122
6123 You will usually want to use scm_shell instead of calling this
6124 function yourself.
6125
6126 ** The new function scm_compile_shell_switches turns an array of
6127 command-line arguments into Scheme code to carry out the actions they
6128 describe. Given ARGC and ARGV, it returns a Scheme expression to
6129 evaluate, and calls scm_set_program_arguments to make any remaining
6130 command-line arguments available to the Scheme code. For example,
6131 given the following arguments:
6132
6133 -e main -s ekko a speckled gecko
6134
6135 scm_set_program_arguments will return the following expression:
6136
6137 (begin (load "ekko") (main (command-line)) (quit))
6138
6139 You will usually want to use scm_shell instead of calling this
6140 function yourself.
6141
6142 ** The function scm_shell_usage prints a usage message appropriate for
6143 an interpreter that uses scm_compile_shell_switches to handle its
6144 command-line arguments.
6145
6146 void scm_shell_usage (int FATAL, char *MESSAGE)
6147 Print a usage message to the standard error output. If MESSAGE is
6148 non-zero, write it before the usage message, followed by a newline.
6149 If FATAL is non-zero, exit the process, using FATAL as the
6150 termination status. (If you want to be compatible with Guile,
6151 always use 1 as the exit status when terminating due to command-line
6152 usage problems.)
6153
6154 You will usually want to use scm_shell instead of calling this
6155 function yourself.
6156
6157 ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no
6158 expressions. It used to return SCM_EOL. Earth-shattering.
6159
6160 ** The macros for declaring scheme objects in C code have been
6161 rearranged slightly. They are now:
6162
6163 SCM_SYMBOL (C_NAME, SCHEME_NAME)
6164 Declare a static SCM variable named C_NAME, and initialize it to
6165 point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should
6166 be a C identifier, and SCHEME_NAME should be a C string.
6167
6168 SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME)
6169 Just like SCM_SYMBOL, but make C_NAME globally visible.
6170
6171 SCM_VCELL (C_NAME, SCHEME_NAME)
6172 Create a global variable at the Scheme level named SCHEME_NAME.
6173 Declare a static SCM variable named C_NAME, and initialize it to
6174 point to the Scheme variable's value cell.
6175
6176 SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME)
6177 Just like SCM_VCELL, but make C_NAME globally visible.
6178
6179 The `guile-snarf' script writes initialization code for these macros
6180 to its standard output, given C source code as input.
6181
6182 The SCM_GLOBAL macro is gone.
6183
6184 ** The scm_read_line and scm_read_line_x functions have been replaced
6185 by Scheme code based on the %read-delimited! procedure (known to C
6186 code as scm_read_delimited_x). See its description above for more
6187 information.
6188
6189 ** The function scm_sys_open has been renamed to scm_open. It now
6190 returns a port instead of an FD object.
6191
6192 * The dynamic linking support has changed. For more information, see
6193 libguile/DYNAMIC-LINKING.
6194
6195 \f
6196 Guile 1.0b3
6197
6198 User-visible changes from Thursday, September 5, 1996 until Guile 1.0
6199 (Sun 5 Jan 1997):
6200
6201 * Changes to the 'guile' program:
6202
6203 ** Guile now loads some new files when it starts up. Guile first
6204 searches the load path for init.scm, and loads it if found. Then, if
6205 Guile is not being used to execute a script, and the user's home
6206 directory contains a file named `.guile', Guile loads that.
6207
6208 ** You can now use Guile as a shell script interpreter.
6209
6210 To paraphrase the SCSH manual:
6211
6212 When Unix tries to execute an executable file whose first two
6213 characters are the `#!', it treats the file not as machine code to
6214 be directly executed by the native processor, but as source code
6215 to be executed by some interpreter. The interpreter to use is
6216 specified immediately after the #! sequence on the first line of
6217 the source file. The kernel reads in the name of the interpreter,
6218 and executes that instead. It passes the interpreter the source
6219 filename as its first argument, with the original arguments
6220 following. Consult the Unix man page for the `exec' system call
6221 for more information.
6222
6223 Now you can use Guile as an interpreter, using a mechanism which is a
6224 compatible subset of that provided by SCSH.
6225
6226 Guile now recognizes a '-s' command line switch, whose argument is the
6227 name of a file of Scheme code to load. It also treats the two
6228 characters `#!' as the start of a comment, terminated by `!#'. Thus,
6229 to make a file of Scheme code directly executable by Unix, insert the
6230 following two lines at the top of the file:
6231
6232 #!/usr/local/bin/guile -s
6233 !#
6234
6235 Guile treats the argument of the `-s' command-line switch as the name
6236 of a file of Scheme code to load, and treats the sequence `#!' as the
6237 start of a block comment, terminated by `!#'.
6238
6239 For example, here's a version of 'echo' written in Scheme:
6240
6241 #!/usr/local/bin/guile -s
6242 !#
6243 (let loop ((args (cdr (program-arguments))))
6244 (if (pair? args)
6245 (begin
6246 (display (car args))
6247 (if (pair? (cdr args))
6248 (display " "))
6249 (loop (cdr args)))))
6250 (newline)
6251
6252 Why does `#!' start a block comment terminated by `!#', instead of the
6253 end of the line? That is the notation SCSH uses, and although we
6254 don't yet support the other SCSH features that motivate that choice,
6255 we would like to be backward-compatible with any existing Guile
6256 scripts once we do. Furthermore, if the path to Guile on your system
6257 is too long for your kernel, you can start the script with this
6258 horrible hack:
6259
6260 #!/bin/sh
6261 exec /really/long/path/to/guile -s "$0" ${1+"$@"}
6262 !#
6263
6264 Note that some very old Unix systems don't support the `#!' syntax.
6265
6266
6267 ** You can now run Guile without installing it.
6268
6269 Previous versions of the interactive Guile interpreter (`guile')
6270 couldn't start up unless Guile's Scheme library had been installed;
6271 they used the value of the environment variable `SCHEME_LOAD_PATH'
6272 later on in the startup process, but not to find the startup code
6273 itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme
6274 code.
6275
6276 To run Guile without installing it, build it in the normal way, and
6277 then set the environment variable `SCHEME_LOAD_PATH' to a
6278 colon-separated list of directories, including the top-level directory
6279 of the Guile sources. For example, if you unpacked Guile so that the
6280 full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then
6281 you might say
6282
6283 export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3
6284
6285
6286 ** Guile's read-eval-print loop no longer prints #<unspecified>
6287 results. If the user wants to see this, she can evaluate the
6288 expression (assert-repl-print-unspecified #t), perhaps in her startup
6289 file.
6290
6291 ** Guile no longer shows backtraces by default when an error occurs;
6292 however, it does display a message saying how to get one, and how to
6293 request that they be displayed by default. After an error, evaluate
6294 (backtrace)
6295 to see a backtrace, and
6296 (debug-enable 'backtrace)
6297 to see them by default.
6298
6299
6300
6301 * Changes to Guile Scheme:
6302
6303 ** Guile now distinguishes between #f and the empty list.
6304
6305 This is for compatibility with the IEEE standard, the (possibly)
6306 upcoming Revised^5 Report on Scheme, and many extant Scheme
6307 implementations.
6308
6309 Guile used to have #f and '() denote the same object, to make Scheme's
6310 type system more compatible with Emacs Lisp's. However, the change
6311 caused too much trouble for Scheme programmers, and we found another
6312 way to reconcile Emacs Lisp with Scheme that didn't require this.
6313
6314
6315 ** Guile's delq, delv, delete functions, and their destructive
6316 counterparts, delq!, delv!, and delete!, now remove all matching
6317 elements from the list, not just the first. This matches the behavior
6318 of the corresponding Emacs Lisp functions, and (I believe) the Maclisp
6319 functions which inspired them.
6320
6321 I recognize that this change may break code in subtle ways, but it
6322 seems best to make the change before the FSF's first Guile release,
6323 rather than after.
6324
6325
6326 ** The compiled-library-path function has been deleted from libguile.
6327
6328 ** The facilities for loading Scheme source files have changed.
6329
6330 *** The variable %load-path now tells Guile which directories to search
6331 for Scheme code. Its value is a list of strings, each of which names
6332 a directory.
6333
6334 *** The variable %load-extensions now tells Guile which extensions to
6335 try appending to a filename when searching the load path. Its value
6336 is a list of strings. Its default value is ("" ".scm").
6337
6338 *** (%search-load-path FILENAME) searches the directories listed in the
6339 value of the %load-path variable for a Scheme file named FILENAME,
6340 with all the extensions listed in %load-extensions. If it finds a
6341 match, then it returns its full filename. If FILENAME is absolute, it
6342 returns it unchanged. Otherwise, it returns #f.
6343
6344 %search-load-path will not return matches that refer to directories.
6345
6346 *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP)
6347 uses %seach-load-path to find a file named FILENAME, and loads it if
6348 it finds it. If it can't read FILENAME for any reason, it throws an
6349 error.
6350
6351 The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the
6352 `read' function.
6353
6354 *** load uses the same searching semantics as primitive-load.
6355
6356 *** The functions %try-load, try-load-with-path, %load, load-with-path,
6357 basic-try-load-with-path, basic-load-with-path, try-load-module-with-
6358 path, and load-module-with-path have been deleted. The functions
6359 above should serve their purposes.
6360
6361 *** If the value of the variable %load-hook is a procedure,
6362 `primitive-load' applies its value to the name of the file being
6363 loaded (without the load path directory name prepended). If its value
6364 is #f, it is ignored. Otherwise, an error occurs.
6365
6366 This is mostly useful for printing load notification messages.
6367
6368
6369 ** The function `eval!' is no longer accessible from the scheme level.
6370 We can't allow operations which introduce glocs into the scheme level,
6371 because Guile's type system can't handle these as data. Use `eval' or
6372 `read-and-eval!' (see below) as replacement.
6373
6374 ** The new function read-and-eval! reads an expression from PORT,
6375 evaluates it, and returns the result. This is more efficient than
6376 simply calling `read' and `eval', since it is not necessary to make a
6377 copy of the expression for the evaluator to munge.
6378
6379 Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as
6380 for the `read' function.
6381
6382
6383 ** The function `int?' has been removed; its definition was identical
6384 to that of `integer?'.
6385
6386 ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should
6387 use the R4RS names for these functions.
6388
6389 ** The function object-properties no longer returns the hash handle;
6390 it simply returns the object's property list.
6391
6392 ** Many functions have been changed to throw errors, instead of
6393 returning #f on failure. The point of providing exception handling in
6394 the language is to simplify the logic of user code, but this is less
6395 useful if Guile's primitives don't throw exceptions.
6396
6397 ** The function `fileno' has been renamed from `%fileno'.
6398
6399 ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0.
6400
6401
6402 * Changes to Guile's C interface:
6403
6404 ** The library's initialization procedure has been simplified.
6405 scm_boot_guile now has the prototype:
6406
6407 void scm_boot_guile (int ARGC,
6408 char **ARGV,
6409 void (*main_func) (),
6410 void *closure);
6411
6412 scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV.
6413 MAIN_FUNC should do all the work of the program (initializing other
6414 packages, reading user input, etc.) before returning. When MAIN_FUNC
6415 returns, call exit (0); this function never returns. If you want some
6416 other exit value, MAIN_FUNC may call exit itself.
6417
6418 scm_boot_guile arranges for program-arguments to return the strings
6419 given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call
6420 scm_set_program_arguments with the final list, so Scheme code will
6421 know which arguments have been processed.
6422
6423 scm_boot_guile establishes a catch-all catch handler which prints an
6424 error message and exits the process. This means that Guile exits in a
6425 coherent way when system errors occur and the user isn't prepared to
6426 handle it. If the user doesn't like this behavior, they can establish
6427 their own universal catcher in MAIN_FUNC to shadow this one.
6428
6429 Why must the caller do all the real work from MAIN_FUNC? The garbage
6430 collector assumes that all local variables of type SCM will be above
6431 scm_boot_guile's stack frame on the stack. If you try to manipulate
6432 SCM values after this function returns, it's the luck of the draw
6433 whether the GC will be able to find the objects you allocate. So,
6434 scm_boot_guile function exits, rather than returning, to discourage
6435 people from making that mistake.
6436
6437 The IN, OUT, and ERR arguments were removed; there are other
6438 convenient ways to override these when desired.
6439
6440 The RESULT argument was deleted; this function should never return.
6441
6442 The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more
6443 general.
6444
6445
6446 ** Guile's header files should no longer conflict with your system's
6447 header files.
6448
6449 In order to compile code which #included <libguile.h>, previous
6450 versions of Guile required you to add a directory containing all the
6451 Guile header files to your #include path. This was a problem, since
6452 Guile's header files have names which conflict with many systems'
6453 header files.
6454
6455 Now only <libguile.h> need appear in your #include path; you must
6456 refer to all Guile's other header files as <libguile/mumble.h>.
6457 Guile's installation procedure puts libguile.h in $(includedir), and
6458 the rest in $(includedir)/libguile.
6459
6460
6461 ** Two new C functions, scm_protect_object and scm_unprotect_object,
6462 have been added to the Guile library.
6463
6464 scm_protect_object (OBJ) protects OBJ from the garbage collector.
6465 OBJ will not be freed, even if all other references are dropped,
6466 until someone does scm_unprotect_object (OBJ). Both functions
6467 return OBJ.
6468
6469 Note that calls to scm_protect_object do not nest. You can call
6470 scm_protect_object any number of times on a given object, and the
6471 next call to scm_unprotect_object will unprotect it completely.
6472
6473 Basically, scm_protect_object and scm_unprotect_object just
6474 maintain a list of references to things. Since the GC knows about
6475 this list, all objects it mentions stay alive. scm_protect_object
6476 adds its argument to the list; scm_unprotect_object remove its
6477 argument from the list.
6478
6479
6480 ** scm_eval_0str now returns the value of the last expression
6481 evaluated.
6482
6483 ** The new function scm_read_0str reads an s-expression from a
6484 null-terminated string, and returns it.
6485
6486 ** The new function `scm_stdio_to_port' converts a STDIO file pointer
6487 to a Scheme port object.
6488
6489 ** The new function `scm_set_program_arguments' allows C code to set
6490 the value returned by the Scheme `program-arguments' function.
6491
6492 \f
6493 Older changes:
6494
6495 * Guile no longer includes sophisticated Tcl/Tk support.
6496
6497 The old Tcl/Tk support was unsatisfying to us, because it required the
6498 user to link against the Tcl library, as well as Tk and Guile. The
6499 interface was also un-lispy, in that it preserved Tcl/Tk's practice of
6500 referring to widgets by names, rather than exporting widgets to Scheme
6501 code as a special datatype.
6502
6503 In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk
6504 maintainers described some very interesting changes in progress to the
6505 Tcl/Tk internals, which would facilitate clean interfaces between lone
6506 Tk and other interpreters --- even for garbage-collected languages
6507 like Scheme. They expected the new Tk to be publicly available in the
6508 fall of 1996.
6509
6510 Since it seems that Guile might soon have a new, cleaner interface to
6511 lone Tk, and that the old Guile/Tk glue code would probably need to be
6512 completely rewritten, we (Jim Blandy and Richard Stallman) have
6513 decided not to support the old code. We'll spend the time instead on
6514 a good interface to the newer Tk, as soon as it is available.
6515
6516 Until then, gtcltk-lib provides trivial, low-maintenance functionality.
6517
6518 \f
6519 Copyright information:
6520
6521 Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
6522
6523 Permission is granted to anyone to make or distribute verbatim copies
6524 of this document as received, in any medium, provided that the
6525 copyright notice and this permission notice are preserved,
6526 thus giving the recipient permission to redistribute in turn.
6527
6528 Permission is granted to distribute modified versions
6529 of this document, or of portions of it,
6530 under the above conditions, provided also that they
6531 carry prominent notices stating who last changed them.
6532
6533 \f
6534 Local variables:
6535 mode: outline
6536 paragraph-separate: "[ \f]*$"
6537 end: