* coop.c (coop_finish): New function. Called at exit.
[bpt/guile.git] / libguile / ChangeLog
1 2000-03-29 Dirk Herrmann <D.Herrmann@tu-bs.de>
2
3 * boolean.h (SCM_TRUE_P): New macro.
4
5 * boolean.h (SCM_FALSEP, SCM_NFALSEP, SCM_BOOLP), pairs.h
6 (SCM_NULLP, SCM_NNULLP): Use SCM_EQ_P to compare SCM values.
7
8 2000-03-28 Dirk Herrmann <D.Herrmann@tu-bs.de>
9
10 * continuations.h (SCM_CONTREGS, SCM_SET_CONTREGS): New macros to
11 access continuation data.
12
13 (SCM_SETJMPBUF): Deprecated. Use SCM_SET_CONTREGS instead.
14
15 (SCM_JMPBUF, SCM_DYNENV, SCM_THROW_VALUE, SCM_BASE, SCM_SEQ,
16 SCM_DFRAME): Use SCM_CONTREGS instead of SCM_CHARS to access
17 continuation data.
18
19 * continuations.c (scm_make_cont), init.c (start_stack),
20 root.c (scm_internal_cwdr): Use SCM_SET_CONTREGS instead of
21 SCM_SETJMPBUF.
22
23 2000-03-28 Dirk Herrmann <D.Herrmann@tu-bs.de>
24
25 * symbols.h (SCM_LENGTH, SCM_SETLENGTH): Access the length field
26 of strings and symbols by using SCM_{SET_}?CELL_WORD_0.
27
28 (SCM_CHARS, SCM_UCHARS, SCM_SETCHARS): Use SCM_{SET_}?CELL_WORD_1
29 to access the char * field of strings and symbols.
30
31 2000-03-27 Dirk Herrmann <D.Herrmann@tu-bs.de>
32
33 * gc.h (SCM_NEWCELL, SCM_NEWCELL2): Use SCM_SET_CELL_TYPE to set
34 the type entry of a new cell. Added a comment about things to
35 remember when updating the list of free cells.
36
37 (SCM_FREEP, SCM_MARKEDP): Use SCM_CELL_TYPE to access the type
38 entry of a cell.
39
40 2000-03-27 Dirk Herrmann <D.Herrmann@tu-bs.de>
41
42 * pairs.h (SCM_CAR, SCM_CDR, SCM_SETCAR, SCM_SETCDR): Use
43 SCM_CELL_OBJECT and SCM_SET_CELL_OBJECT. This change implies that
44 with strict type checking enabled these macros will only work if
45 given valid SCM parameters.
46
47 (SCM_GCCDR): Moved to tags.h.
48
49 * tags.h (SCM_GCCDR): Moved here from pairs.h.
50
51 2000-03-26 Dirk Herrmann <D.Herrmann@tu-bs.de>
52
53 * tags.h (SCM2PTR, PTR2SCM): Moved to gc.h.
54
55 * pairs.h (scm_cell, SCM_CELLPTR, SCM_CELL_WORD*, SCM_CELL_OBJECT*,
56 SCM_SET_CELL_WORD*, SCM_SET_CELL_OBJECT*, SCM_CELL_TYPE,
57 SCM_SET_CELL_TYPE, SCM_PTR_LT, SCM_PTR_MASK, SCM_PTR_GT,
58 SCM_PTR_LE, SCM_PTR_GE, SCM_CELL_WORD_LOC, SCM_NEWCELL,
59 SCM_NEWCELL2): Moved to gc.h.
60
61 (SCM_CARLOC, SCM_CDRLOC, SCM_SETAND_CAR, SCM_SETAND_CDR,
62 SCM_SETOR_CAR, SCM_SETOR_CDR): Moved to gc.h. These names should
63 be changed, though, since the macros are not only pair related.
64
65 (SCMPTR): Deleted.
66
67 * gc.h (SCM2PTR, PTR2SCM, scm_cell, SCM_CELLPTR, SCM_CELL_WORD*,
68 SCM_CELL_OBJECT*, SCM_SET_CELL_WORD*, SCM_SET_CELL_OBJECT*,
69 SCM_CELL_TYPE, SCM_SET_CELL_TYPE, SCM_PTR_LT, SCM_PTR_MASK,
70 SCM_PTR_GT, SCM_PTR_LE, SCM_PTR_GE, SCM_CELL_WORD_LOC,
71 SCM_NEWCELL, SCM_NEWCELL2, SCM_CARLOC, SCM_CDRLOC, SCM_SETAND_CAR,
72 SCM_SETAND_CDR, SCM_SETOR_CAR, SCM_SETOR_CDR): Moved here from
73 tags.h and pairs.h.
74
75 2000-03-25 Dirk Herrmann <D.Herrmann@tu-bs.de>
76
77 * tags.h (SCM_STRICT_TYPING): New macro that, if defined,
78 activates strict compile time type checking for variables of
79 type SCM.
80 (SCM, SCM_PACK, SCM_UNPACK): Define according to whether
81 SCM_STRICT_TYPING or SCM_VOIDP_TEST are defined.
82 (SCM_EQ_P): Defined as a macro equivalent for eq?.
83
84 2000-03-25 Dirk Herrmann <D.Herrmann@tu-bs.de>
85
86 * tags.h (SCM_POINTERS_MUNGED): Removed.
87
88 * gc.c (scm_gc_sweep, init_heap_seg): Removed use of
89 SCM_POINTERS_MUNGED, thus fixing some illegal casts to SCM.
90
91 2000-03-24 Dirk Herrmann <D.Herrmann@tu-bs.de>
92
93 * pairs.h (SCM_CELL_OBJECT, SCM_CELL_OBJECT_[0-3],
94 SCM_SET_CELL_OBJECT, SCM_SET_CELL_OBJECT_[0-3], SCM_CELL_TYPE,
95 SCM_SET_CELL_TYPE): Added a set of low level macros for accessing
96 cell entries.
97 (SCM_CELL_WORD_[0-3]): Renamed from the SCM_CELL_WORD[0-3].
98
99 * procs.h, procs.c: Instead of SCM_{SET_}?CELL_WORD[12], use the
100 newly introduced SCM_{SET_}?CELL_OBJECT_[12] macros.
101
102 2000-03-23 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
103
104 * tags.h: Disabled definition of SCM_VOIDP_TEST.
105
106 Defining SCM as void * introduces problems which haven't been
107 handled yet. Developers who work with these issues can enable it
108 in their working copies.
109
110 Disabling this definition exposes a set of newly introduced and
111 older misuses of types which causes warning messages during
112 compilation. We'll fix this successively.
113
114 * gc.c (scm_mark_locations): Changed * (SCM **) X --> * (SCM *) X
115 in order to obtain a value of type SCM.
116 (scm_cellp): Updated with new changes to scm_mark_locations.
117
118 * continuations.h (SCM_SETJMPBUF): Cast second arg into SCM.
119
120 * continuations.c (scm_make_cont): Removed cast of size_t into
121 long.
122
123 * symbols.h (SCM_SETCHARS): Cast second arg into SCM.
124
125 2000-03-22 Dirk Herrmann <D.Herrmann@tu-bs.de>
126
127 * numbers.h (SCM_SETNUMDIGS): Use SCM_BIGSIZEFIELD macro for
128 shifting, not constant. Thanks to Dale P. Smith.
129
130 * numbers.c (scm_sum, scm_difference): Don't test a SCM value
131 for being less than zero. Decode it to a C value first. Again,
132 thank you Dale.
133
134 2000-03-22 Dirk Herrmann <D.Herrmann@tu-bs.de>
135
136 * numbers.h, ramap.c, struct.h, vectors.h: Don't use SCM2PTR for
137 non scheme values. If raw data is stored in SCM variables, it has
138 to be accessed using SCM_UNPACK until a better solution is found.
139
140 2000-03-22 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
141
142 * tags.h (SCM_ECONSP, SCM_NECONSP): More corrections of
143 pointer-arithmetic induced by the SCM_PACK/UNPACK change.
144
145 * print.c (scm_iprin1): SCM_PACK/UNPACK corrections.
146
147 * gc.c (scm_gc_sweep): SCM_PACK/UNPACK corrections.
148
149 * eval.c (SCM_CEVAL, scm_unmemocar): SCM_PACK/UNPACK corrections.
150
151 * dynwind.c (scm_swap_bindings): SCM_PACK/UNPACK corrections.
152
153 * async.c, __scm.h: Removed lots of the old async click logic. It
154 is possible to reinsert it by defining GUILE_OLD_ASYNC_CLICK in
155 __scm.h. Let's try this out and dump the old code after the
156 threads reorganization.
157 (set-tick-rate, set-switch-rate): Conditionally removed.
158
159 2000-03-21 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
160
161 * gc.c (scm_gc_mark): Bugfix 1: The recent SCM_PACK/UNPACK change
162 made SCM values into pointers. This turned an arithmetic
163 computation of the address of the vcell into a pointer-arithmetic
164 one, thereby screwing up marking of structs.
165 Bugfix 2: Removed incompletely introduced loop variable `j' used
166 when protecting the tail array of a struct.
167
168 2000-03-21 Dirk Herrmann <D.Herrmann@tu-bs.de>
169
170 * struct.h (SCM_STRUCT_DATA): Don't cast SCM values to pointers.
171
172 2000-03-21 Dirk Herrmann <D.Herrmann@tu-bs.de>
173
174 * symbols.h, symbols.c (scm_strhash): Declare the string
175 parameter as constant, since it is not modified.
176
177 * symbols.c (scm_intern_obarray_soft,
178 scm_sysintern0_no_module_lookup): Can now pass constant strings
179 to scm_strhash without need for casting.
180
181 2000-03-21 Dirk Herrmann <D.Herrmann@tu-bs.de>
182
183 * vectors.h (SCM_VELTS, SCM_VELTS_AS_STACKITEMS): Don't cast SCM
184 values to pointers. Use SCM2PTR instead.
185
186 2000-03-21 Dirk Herrmann <D.Herrmann@tu-bs.de>
187
188 * async.c (scm_set_tick_rate, scm_set_switch_rate): Don't unpack
189 results of SCM_INUM.
190
191 2000-03-21 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
192
193 * init.c (scm_boot_guile_1): Renamed GUILE_GC_TRIGGER_1 -->
194 GUILE_MIN_YIELD_1, GUILE_GC_TRIGGER_2 --> GUILE_MIN_YIELD_2.
195 GUILE_MIN_YIELD_X now take *positive* fractions of heap size.
196
197 * gc.c, gc.h (SCM_MIN_YIELD_1, SCM_MIN_YIELD_2,
198 min_yield_fraction, min_yield, adjust_min_yield): Renamed from
199 SCM_GC_TRIGGER_1, SCM_GC_TRIGGER_2, gc_trigger_fraction,
200 gc_trigger, adjust_gc_trigger.
201
202 * gc.c (alloc_some_heap): Further improvement of minimal heap size
203 prediction.
204 (SCM_MAX): New macro.
205 (scm_freelist_t): New field: collected_1. Previous amount of
206 collected cells.
207 (gc_sweep_freelist_finish): Trigger based on two last values of
208 freelist->collected to avoid unnecessary allocation due to
209 temporary peaks.
210 (SCM_INIT_HEAP_SIZE_1, SCM_GC_TRIGGER_1, SCM_GC_TRIGGER_2):
211 Adjusted to 45000 cells, 40% and 40%. Gives quick startup
212 without extra heap allocation.
213
214 2000-03-20 Dirk Herrmann <D.Herrmann@tu-bs.de>
215
216 * numbers.h (SCM_MAKINUM): The parameter to SCM_MAKINUM should
217 already be a C value. No need to unpack it.
218
219 * numbers.c (scm_long_long2num): Cast the parameter to scm_bits_t
220 if we know it fits into an inum.
221
222 * ramap.c (ramap_rp): An scm_tc7_[ui]vect object does point to a
223 field of long values. In contrast, SCM_VELTS accesses a field of
224 SCM values.
225
226 2000-03-20 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
227
228 * gc.c (scm_gc_stats): Inserted explanation of local_scm_mtrigger
229 etc.
230 (scm_gc_yield_1): New variable: Holds previous yield. Used to
231 make better judgements.
232 (gc_sweep_freelist_finish): Inserted explanation of use of
233 gc_trigger.
234
235 * print.h, stacks.h, options.c, options.h: Changed C++
236 commentaries to C.
237
238 2000-03-20 Dirk Herrmann <D.Herrmann@tu-bs.de>
239
240 * tags.h (SCM2PTR, PTR2SCM): Use SCM_PACK / SCM_UNPACK correctly.
241
242 * numbers.h (SCM_INUMP, SCM_MAKINUM, SCM_INUM0, SCM_COMPLEX_REAL,
243 SCM_COMPLEX_IMAG, SCM_NUMP, SCM_BDIGITS): Use SCM_PACK /
244 SCM_UNPACK / SCM2PTR correctly.
245
246 2000-03-20 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
247
248 * gc.c (adjust_gc_trigger): Improved documentation.
249 (alloc_some_heap): Since gc_trigger is used against
250 freelist->collected, this is the value which should be used to
251 predict minimum growth.
252
253 2000-03-20 Dirk Herrmann <D.Herrmann@tu-bs.de>
254
255 * eval.h: Fix mixup of packed/unpacked SCM values. (Thanks
256 Thien-Thi Nguyen for the patch.)
257
258 2000-03-20 Dirk Herrmann <D.Herrmann@tu-bs.de>
259
260 * numbers.c (scm_ash): Fixed typing problems with the second
261 parameter and added some documentation. (Thanks Thien-Thi Nguyen
262 for indicating the problem.)
263
264 2000-03-19 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
265
266 * gc.c, gc.h (scm_gc_yield): New variable.
267 (adjust_gc_trigger): Use scm_gc_yield.
268 (alloc_some_heap): Use scm_gc_yield instead of
269 scm_gc_cells_collected.
270
271 * coop-threads.c: Addd #include "root.h", #include "strings.h".
272
273 * debug.c: Added #include "root.h". (Thanks to Thien-Thi Nguyen.)
274
275 * gc.c (scm_gc_for_newcell, adjust_gc_trigger): Improved GC
276 trigger adjustmeant: Take yield (freed cells) for all freelists
277 into account.
278 (SCM_INIT_HEAP_SIZE_1, SCM_GC_TRIGGER_1, SCM_GC_TRIGGER_2): Tuned
279 to 50000 cell heap with 45% trigger.
280 (scm_gc_cells_collected): Reintroduced.
281 (SCM_HEAP_SIZE): New macro.
282 (scm_gc_sweep): Reintroduced correct computation of
283 scm_cells_allocated.
284 (scm_freelist_t): Corrected commentary for field `cluster_size':
285 Clustersize counts objects, not cells; New member
286 `clusters_allocated'.
287
288 2000-03-19 Michael Livshin <mlivshin@bigfoot.com>
289
290 * *.[hc]: add Emacs magic at the end of file, to ensure GNU
291 indentation style.
292
293 2000-03-19 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
294
295 * threads.h: Added #include "libguile/throw.h". (Thanks to
296 Thien-Thi Nguyen.)
297
298 2000-03-18 Michael Livshin <mlivshin@bigfoot.com>
299
300 * tags.h: (SCM_DOUBLE_CELLP, SCM_NDOUBLE_CELLP): new macros.
301
302 * gc.h: (typedef struct scm_freelist_t) remove from here.
303
304 * gc.c: (CELL_UP, CELL_DN) make these macros take additional
305 parameter (the span).
306 (CLUSTER_SIZE_IN_BYTES, ALIGNMENT_SLACK) new macros.
307 (typedef struct scm_freelist_t) move here from gc.h, it had no
308 business being externally visible.
309 (typedef struct scm_heap_seg_data_t) renamed from
310 scm_heap_seg_data, to be style-compliant.
311 (scm_mark_locations) if the possible pointer points to a
312 multy-cell, check that it's properly aligned.
313 (init_heap_seg) alighn multy-cells properly, work with the
314 assumption that the segment size divides cleanly by cluster size
315 (so that there's no spill).
316 (round_to_cluster_size) new function.
317 (alloc_some_heap, make_initial_segment) use round_to_cluster_size
318 to satisfy the new init_heap_seg invariant.
319
320 2000-03-18 Dirk Herrmann <D.Herrmann@tu-bs.de>
321
322 * _scm.h: Don't include async.h everywhere...
323
324 * eq.c eval.c iselect.c: ... only include it here.
325
326 2000-03-18 Dirk Herrmann <D.Herrmann@tu-bs.de>
327
328 * _scm.h: Don't include root.h everywhere...
329
330 * async.c continuations.c eq.c eval.c evalext.c feature.c gc.c
331 gdbint.c gsubr.c ioext.c keywords.c lang.c load.c macros.c
332 numbers.c objprop.c ports.c print.c procprop.c ramap.c read.c
333 srcprop.c stackchk.c stacks.c strports.c symbols.c unif.c
334 variable.c vectors.c vports.c: ... only include it here.
335
336 2000-03-17 Dirk Herrmann <D.Herrmann@tu-bs.de>
337
338 * _scm.h: Don't include strings.h everywhere...
339
340 * backtrace.c dynl.c error.c feature.c filesys.c fports.c gc.c
341 gdbint.c ioext.c load.c net_db.c numbers.c objects.c options.c
342 ports.c posix.c print.c procs.c random.c read.c regex-posix.c
343 simpos.c socket.c stacks.c stime.c strop.c strports.c struct.c
344 symbols.c unif.c vectors.c version.c vports.c: ... only include it
345 here.
346
347 2000-03-17 Dirk Herrmann <D.Herrmann@tu-bs.de>
348
349 * _scm.h: Don't include ports.h everywhere...
350
351 * arbiters.c backtrace.c debug.c dynl.c dynwind.c eval.c feature.c
352 fluids.c gc.c gdbint.c guardians.c hash.c keywords.c mallocs.c
353 numbers.c objects.c print.c read.c root.c smob.c srcprop.c
354 stackchk.c strports.c struct.c throw.c variable.c: ... only
355 include it here.
356
357 2000-03-17 Dirk Herrmann <D.Herrmann@tu-bs.de>
358
359 * _scm.h: Don't include vectors.h everywhere...
360
361 * eq.c eval.c filesys.c gc.c gsubr.c guardians.c hash.c hashtab.c
362 keywords.c net_db.c numbers.c objects.c posix.c print.c procprop.c
363 procs.c ramap.c random.c read.c scmsigs.c socket.c sort.c stime.c
364 strports.c symbols.c unif.c vports.c weaks.c: ... only include it
365 here.
366
367 2000-03-17 Dirk Herrmann <D.Herrmann@tu-bs.de>
368
369 * genio.h: removed. (Only content was '/* delete me */'.)
370
371 * Makefile.am arbiters.c backtrace.c debug.c dynl.c dynwind.c
372 error.c filesys.c fluids.c gc.c gsubr.c guardians.c keywords.c
373 libguile.h mallocs.c numbers.c print.c random.c read.c root.c
374 srcprop.c stackchk.c struct.c threads.c throw.c variable.c:
375 Removed reference to genio.h
376
377 2000-03-17 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
378
379 * gc.c, gc.h: Cleanup of the change of 2000-03-15.
380 Cluster sizes are now independent of GC trigger values.
381 GUILE_GC_TRIGGER_n can now specify a relative trigger value:
382 A negative integer gives fraction of total heap size in percent.
383 (SCM_GC_TRIGGER_1, SCM_GC_TRIGGER_2): Default values set to -40.
384
385 * init.c (scm_boot_guile_1): Introduced new variable
386 GUILE_MAX_SEGMENT_SIZE; New environment variable names:
387 GUILE_INIT_SEGMENT_SIZE_1, GUILE_GC_TRIGGER_1,
388 GUILE_INIT_SEGMENT_SIZE_2, GUILE_GC_TRIGGER_2
389
390 2000-03-16 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
391
392 * __scm.h (GC_FREE_SEGMENTS): Disable this until we have made
393 freeing of segment work with the new GC scheme. (Thanks to
394 Michael Livshin.) Oops, also happened to make GUILE_NEW_GC_SCHEME
395 the default, but I'll let this change stay in CVS Guile since this
396 code is not expected to contain serious bugs.
397
398 2000-03-16 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
399
400 * gc.c, gc.h (scm_map_free_list): Define also if GUILE_DEBUG is
401 defined.
402 (scm_free_list_length): New procedure (GUILE_DEBUG).
403 Fixed a small but serious bug introduced by the previous change.
404
405 * gc.c (scm_gc_sweep): Moved variable n_objects to inner sweep
406 loop and declare as register.
407
408 * gc.c (scm_gc_sweep): Sigh... forgot to clear private freelists
409 after GC.
410
411 Wed Mar 15 08:27:04 2000 Greg J. Badros <gjb@cs.washington.edu>
412
413 * fluids.c: Docstring patch from Keisuke Nishida. Some
414 reindentation, too, and a couple formals renamed. Should
415 fluid-set! return UNSPECIFIED instead of a value?
416
417 Wed Mar 15 08:24:58 2000 Greg J. Badros <gjb@cs.washington.edu>
418
419 * Makefile.am: Separate out DOT_X_FILES and DOT_DOC_FILES, and
420 generate the latter from the concrete listing of the former. Then
421 make guile-procedures.txt depend on DOT_DOC_FILES instead of
422 *.doc, so that rebuilding it works.
423
424 Wed Mar 15 08:12:14 2000 Greg J. Badros <gjb@cs.washington.edu>
425
426 * libguile.h: Include libguile/validate.h. Thanks Keisuke Nishida!
427
428 * guile-snarf.awk.in: Replace docstring line-ending \n" and \n\n"
429 with nothing and \n, respectively. Thanks Keisuke Nishida for
430 noticing this problem.
431
432 2000-03-15 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
433
434 * __scm.h (GUILE_NEW_GC_SCHEME): Define this if you want to test a
435 new way of allocating heap. It makes Guile fast, but still
436 contains bugs.
437
438 * gc.c, gc.h, pairs.h, init.c: Implementation of a new way of
439 allocating heap. The basic idea is to trigger GC every Nth
440 allocated cell and grow heap when free list runs out. The scheme
441 has been extended so that GC isn't triggered until all remaining
442 cells are used. The implementation is also prepared for
443 development in the direction of POSIX threads.
444
445 * gc.c (SCM_EXPHEAP): In order to grow by a factor of 1.5,
446 SCM_EXPHEAP should return half of the heap size.
447
448 2000-03-14 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
449
450 The following change to init.c is only enabled if Guile was
451 configured with --enable-guile-debug.
452
453 * init.c (scm_i_getenv_int): New function.
454 (scm_boot_guile_1): Use the environment variables
455 GUILE_INIT_HEAP_SIZE, GUILE_INIT_HEAP_SIZE2 to select heap size if
456 they exist. (This may be replaced by a Scheme level mechanism in
457 the future.)
458
459 * objprop.c (s_scm_set_object_property_x): Use scm_assq instead of
460 scm_assoc. (Thanks to Keisuke Nishida.)
461
462 2000-03-14 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
463
464 * eval.c, lang.c, lang.h (scm_lisp_nil, scm_lisp_t): Renamed from
465 scm_nil, scm_t. (Thanks to Keisuke Nishida.)
466
467 2000-03-14 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
468
469 * init.c (scm_boot_guile_1): Use same initial segment size for
470 1-word and 2-word segments. Having the smaller size causes Guile
471 to GC too often. Obviously something needs to be done to allow
472 for a smaller 2-word segment without this to happen. (The amount
473 of heap for each type should be automatically adapted to the
474 application somehow.)
475
476 [Almost all of these changes should be documented in the NEWS
477 file.]
478
479 * gc.h (scm_freelist_t): New type.
480
481 * pairs.h (SCM_NEWCELL, SCM_NEWCELL2): Use new style freelists.
482
483 * gc.c (SCM_INIT_HEAP_SIZE): Changed from 32768 --> 40000 so that
484 all of Guile basics fits into one segment and there suitable room
485 for work.
486 (SCM_EXPHEAP): Now takes an argument. Grow by a factor of 1.5
487 instead of 2.
488 (scm_freelist, scm_freelist2): Now of type scm_freelist_t.
489 Freelists now contains information about object span, cells
490 collected and amount of cells in heap segments belonging to the
491 list.
492 (scm_heap_size, scm_gc_cells_collected): Removed.
493
494 * init.c (scm_boot_guile_1): Make 2-word segment 8K (512 cells).
495
496 * Makefile.am (libguile_la_LDFLAGS): Bumped library version
497 number.
498
499 * __scm.h eq.c, eval.c, gc.c, gc.h, gh_data, hash.c, numbers.c,
500 numbers.h, objects.c, ramap.c, random.c, unif.c, unif.h: Extensive
501 rewrite of handling of real and complex numbers.
502 (SCM_FLOATS, SCM_SINGLES): These #ifdef conditionals have been
503 removed along with the support for floats. (Float vectors are
504 still supported.)
505
506 * tags.h (scm_tcs_bignums): Removed.
507 (scm_tc16_bigpos, scm_tc16_bigneg): Replaced by scm_tc16_big.
508 Use SCM_BIGSIGN(x) to test for sign!
509 (scm_tc16_big): The new bignum type.
510 (SCM_REAL_PART, SCM_IMAG_PART): Removed.
511
512 * numbers.h (SCM_BIGSIGN): Sign moved to bit 16.
513 (scm_makdbl): Deprecated.
514 (SCM_NEWREAL, SCM_NEWCOMPLEX): New macros.
515 (SCM_SINGP): Deprecated.
516 (SCM_FLO): Removed.
517 (SCM_INEXP, SCM_CPLXP): Deprecated.
518 (SCM_INEXACTP, SCM_COMPLEXP): New macros.
519 (SCM_COMPLEX_REAL, SCM_COMPLEX_IMAG): Renamed from
520 SCM_REAL, SCM_IMAG (and now only valid for complex numbers).
521 (SCM_REAL, SCM_IMAG): New, *deprecated*, selectors which work both
522 for doubles and complex numbers.
523 (SCM_REAL_VALUE): New selector for doubles.
524 (scm_double_t, scm_complex_t): New types.
525 (scm_dbl): Removed.
526
527 * numbers.c (scm_floprint, scm_floequal): Removed.
528 (scm_print_real, scm_print_complex, scm_real_equalp,
529 scm_complex_equalp): New functions.
530
531 * unif.c (scm_makflo): Removed.
532
533 * smob.h (SCM_SMOB_PREDICATE): New macro.
534 (SCM_NEWSMOB2, SCM_RETURN_NEWSMOB2, SCM_NEWSMOB3,
535 SCM_RETURN_NEWSMOB3): New macros.
536
537 1999-11-21 Michael Livshin <mlivshin@bigfoot.com>
538
539 The following changes implement primitive support for double cells
540 (i.e. four-word cells) and change the representation of some
541 things to multi-cells instead of cons+malloc. (Applied and
542 modified by mdj.)
543
544 * pairs.h (SCM_NEWCELL2): double-cell variants of SCM_NEWCELL.
545 (SCM_CELL_WORD, SCM_CELL_WORDLOC, SCM_SET_CELL_WORD): primitive
546 multi-cell access macros (used by the ones below).
547 (SCM_CELL_WORD[0-3], SCM_SET_CELL_WORD[0-3]): multi-cell access
548 macros.
549
550 * gc.c (scm_freelist2): multi-cell freelists.
551 (inner_map_free_list): map_free_list, parameterized on ncells.
552 "nn cells in segment mm" was misleading for ncells > 1; changed to
553 "objects". still print cells too, though.
554 (scm_map_free_list): rewritten using inner_map_free_list.
555 (scm_check_freelist): get freelist as parameter, since now we have
556 more than one.
557 (scm_debug_newcell2): multi-cell variants of
558 scm_debug_newcell.
559 (scm_gc_for_newcell): take ncells and freelist pointer as
560 parameters.
561 (scm_gc_mark): add case for tc7_pws (procedures with setters are
562 now double cells).
563 (scm_gc_sweep): don't free the float data, since it's not malloced
564 anymore.
565 (init_heap_seg): didn't understand what n_new_objects stood for,
566 so changed to n_new_cells.
567 (make_initial_segment): new function, makes an initial segment
568 according to given ncells.
569 (scm_init_storage): call make_initial_segment, for ncells={1,2,3}.
570
571 * numbers.c (scm_makdbl): no malloc'ing needed, so the
572 {DEFER,ALLOW}_INTS thing removed.
573
574 * numbers.h (struct scm_dbl): changed to represent a double cell,
575 with the number in the second half.
576
577 * dynwind.c: changed the wind-guards representation to double
578 cell.
579
580 * procs.c, procs.h: changed the procedure-with-setter representation
581 to double cell.
582
583 * async.c, async.h: made async representation a double cell.
584
585 * dynl.c: made dynamic_obj representation a double cell.
586
587 2000-03-13 Gary Houston <ghouston@arglist.com>
588
589 * ports.c (flush_void_port): renamed to flush_port_default.
590 (end_input_void_port): renamed to end_input_default.
591
592 * init.c (scm_standard_stream_to_port): create a void port instead
593 of opening /dev/null if the standard file descriptors are bad.
594 advantages: no portability problems, doesn't waste a file descriptor,
595 simplifies the code (thanks to Marius for the idea).
596
597 * vports.c (s_scm_make_soft_port): call scm_port_non_buffer.
598
599 * void ports: make reading from a void port give EOF instead of
600 segv:
601 * ports.c (s_scm_sys_make_void_port): modified docstring.
602 (fill_input_void_port): new proc.
603 (scm_init_ports): set up fill_input_void_port.
604 * ports.c (scm_port_non_buffer): new proc.
605 (scm_void_port): call scm_port_non_buffer.
606
607 * fports.c (scm_setvbuf): docstring: remove the fcntl documentation
608 which was incorrectly appended.
609
610 2000-03-13 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
611
612 * guile-doc-snarf.in: Don't use absolute path for `sed'. (Note
613 that we can't use autoconf for this. Autoconf itself relies on
614 the existence of `sed' somewhere on your path.) (Thanks to Dirk
615 Herrman.)
616
617 2000-03-13 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
618
619 * Makefile.am (libguile_la_SOURCES): Moved iselect.c here from
620 EXTRA_libguile_la_SOURCES.
621
622 * iselect.h: Always declare scm_internal_select.
623
624 * iselect.c (scm_internal_select): Added SCM_ASYNC_TICK at the
625 end. Also let scm_internal_select be a real function also when
626 not using threads.
627
628 * __scm.h (SCM_TICK): Oops! Forgot to put SCM_ASYNC_TICK here...
629
630 2000-03-13 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
631
632 * __scm.h (SCM_ALLOW_INTS, SCM_REALLOW_INTS): Removed call to
633 SCM_ASYNC_TICK. (This is a preparation for POSIX threads support,
634 and kind of an experiment: Will this cause problems?)
635
636 Sun Mar 12 13:26:30 2000 Greg J. Badros <gjb@cs.washington.edu>
637
638 * Makefile.am: Added *.doc to DISTCLEANFILES.
639
640 2000-03-12 Gary Houston <ghouston@arglist.com>
641
642 * fports.c (scm_fdes_to_port): call fcntl F_GETFL to test that
643 the fdes is valid before doing anything else. check that
644 the file descriptor supports the modes required.
645 (scm_fport_buffer_add): don't throw an error if fstat doesn't
646 work: just use the default buffer size.
647
648 * throw.c: change an outdated comment about scm_internal_catch
649 BODY: it doesn't take a jumpbuf arg.
650
651 * init.c (scm_standard_stream_to_port): install a handler in case
652 scm_fdes_to_port throws an error. don't check here whether the
653 file descriptor is valid, since scm_fdes_to_port will do that.
654 set the revealed count depending on whether the port got the
655 standard file descriptor.
656 (stream_body_data): new type.
657 (stream_body, stream_handler): new procs.
658
659 2000-03-12 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
660
661 * stacks.c, stacks.h, struct.c, tags.h, unif.c (scm_bits_t):
662 Renamed from SCMWORD.
663
664 * tags.h (SCM_NCELLP): Removed (SCMWORD).
665
666 * arbiters.c (SCM_ARB_LOCKED): Use SCM_UNPACK_CAR.
667
668 * async.c, boolean.h, debug.c, dynl.c, dynwind.c, eval.c, eval.h,
669 feature.h, filesys.h, fluids.h, fports.c, fports.h, gc.c, gc.h,
670 hash.c, keywords.h, macros.c, numbers.c, numbers.h, objects.c,
671 objects.h, options.c, pairs.h, ports.c, ports.h, print.c,
672 procs.h, ramap.c, read.c, smob.c, smob.h, srcprop.h, stacks.c,
673 stacks.h, strports.c, struct.c, struct.h, tag.c, tags.h,
674 throw.c, unif.c, unif.h, variable.h, vectors.h, weaks.c,
675 weaks.h (SCM_PACK, SCM_UNPACK, SCM_UNPACK_CAR): Renamed from
676 SCM_ASSCM, SCM_ASWORD, SCM_CARW).
677
678 * numbers.h (SCM_SRS, SCM_INUM): Corrected SCM_ASSCM/ASWORD fixes.
679
680 * alist.c, eval.c, net_db.c, posix.c, print.c, snarf.h, struct.c,
681 tags.h: Fixed copyright notices.
682
683 * struct.c, coop-threads.c: SCM_ASSCM/ASWORD fixes.
684
685 2000-03-12 Marius Vollmer <mvo@zagadka.ping.de>
686
687 * init.c (scm_standard_stream_to_port): Check whether the file
688 descriptor is valid and substitute "/dev/null" when not.
689
690 2000-03-12 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
691
692 * coop-defs.h (struct timespec): Conditionally defined.
693
694 * coop.c (coop_condition_variable_timed_wait_mutex): Use ETIMEDOUT
695 instead of ETIME.
696
697 * readline.c (match_paren): Bugfix: First arg to select is not
698 number of descriptors but the number of the highest descriptor +
699 1.
700
701 This is a preliminary attempt at a cleanup of the threads support
702 code. It moves things to better places, makes arguments more
703 consistent with the POSIX API (which is used in GNOME's glib), and
704 adds new functionality.
705
706 * readline.c (scm_init_readline): Added new arg to scm_init_mutex.
707
708 * coop-defs.h (scm_mutex_trylock): New macro: alias for
709 coop_mutex_trylock.
710 (scm_cond_init): Changed definition to
711 coop_new_condition_variable_init.
712
713 * coop.c: #include <errno.h>
714 (coop_timeout_qinsert): Moved here from iselect.c
715 (coop_new_mutex_init, coop_new_condition_variable_init): New
716 functions. The strange names are temporary. Use scm_mutex_init
717 and scm_cond_init instead.
718 (coop_mutex_trylock): New function. Uses errno.h:EBUSY. errno.h
719 is ANSI C, but should we check for individual error codes in
720 configure.in?
721 (coop_condition_variable_timed_wait_mutex): New function.
722 (coop_key_create, coop_setspecific, coop_getspecific,
723 coop_key_delete): New functions.
724
725 * iselect.c (coop_timout_qinsert): Moved to coop.c
726
727 2000-03-11 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
728
729 * pairs.h (SCM_SETAND_CAR, SCM_SETAND_CDR, SCM_SETOR_CAR,
730 SCM_SETOR_CDR): Corrected SCM_ASSCM/WORD fixes.
731
732 * tags.h (SCM_VOIDP_TEST): Renamed from VOIDP_TEST.
733 Layout cleanups.
734
735 * objects.h (SCM_CLASS_FLAGS, SCM_OBJ_CLASS_FLAGS,
736 SCM_SET_CLASS_INSTANCE_SIZE), struct.h (SCM_STRUCT_VTABLE_DATA),
737 proc.h (SCM_CLOSCAR): SCM_ASSCM/WORD fixes.
738
739 * eval.c (scm_lookupcar1): Inserted SCM_ASWORD in expressions
740 dealing with ilocs.
741
742 2000-03-11 Dale P. Smith <dpsm@en.com>, applied by Greg J. Badros, <gjb@cs.washington.edu>
743
744 * numbers.c (scm_copy_big_dec, scm_copy_smaller, scm_big_ior,
745 scm_big_xor, scm_big_and, scm_big_test): Added new lowlevel bignum
746 logical functions from SCM.
747
748 (logand, logior, logxor, logtest, logbit?): Extended scheme
749 logical functions to use bignums from SCM.
750
751 (lognot): Removed call to `SCM_VALIDATE_INUM' that prevented
752 lognot from using bignums.
753
754 Thu Mar 9 11:33:25 2000 Greg J. Badros <gjb@cs.washington.edu>
755
756 * vectors.h (SCM_VELTS_AS_STACKITEMS): Added this macro to help in
757 eliminating some warnings.
758
759 * unif.c, strports.c, print.c, options.c: Fix some warnings on
760 mis-use of SCM/long
761
762 * gc.c, gc.h: Added scm_return_first_int(), and added comment re:
763 what the scm_return_first* functions do.
764
765 2000-03-09 Han-Wen Nienhuys <hanwen@cs.uu.nl>, applied by Greg J. Badros, <gjb@cs.washington.edu>
766
767 * libguile/*.[ch]: make a distinction between SCM as a generic
768 name for a Scheme object (now a void*), and SCM as 32 bit word for
769 storing tags and immediates (now a long int). Introduced
770 SCM_ASWORD and SCM_ASSCM for conversion. Fixed various dubious
771 code in the process: arbiter.c (use macros), unif.c (scm_array_p),
772
773 Wed Mar 8 10:15:59 2000 Greg J. Badros <gjb@cs.washington.edu>
774
775 * numbers.c: Use SCM_VALIDATE_LONG_COPY, and longs, not ints, in
776 various logXXX primitives. Thanks Eric Moore!
777
778 Tue Mar 7 08:05:22 2000 Greg J. Badros <gjb@cs.washington.edu>
779
780 * run-test, remaining-docs-needed: Added these scripts. The
781 second one is only temporary until the docstring additions are
782 complete. run-test may best live on, but is here mostly for
783 convenience and awareness for now.
784
785 * hash.c: Docs, minor cleanup patch from Dirk Herrman.
786
787 Thu Mar 2 16:06:58 2000 Greg J. Badros <gjb@cs.washington.edu>
788
789 * error.h, error.c: Added `scm_wrong_type_arg_msg' to support
790 displaying the expected type. Use SCM_LISTn in a couple places
791 instead of scm_cons-ing by hand.
792
793 * __scm.h: Added SCM_ASSERT_TYPE macro.
794
795 * validate.h, scm_validate.h: Added the former, as a renamed
796 version of the latter with SCM_ASSERT_TYPE used in
797 SCM_MAKE_VALIDATE (instead of just SCM_ASSERT)
798
799 * Makefile.am: Rename scm_validate.h to validate.h.
800
801 * *.c, *.h: Include validate.h, not scm_validate.h (old name's
802 prefix was superfluous).
803
804 Thu Mar 2 15:33:12 2000 Greg J. Badros <gjb@cs.washington.edu>
805
806 * hashtab.c: Improved documentation for lots of functions. Added
807 handwritten docs for `hash-fold'.
808
809 Thu Mar 2 15:13:25 2000 Greg J. Badros <gjb@cs.washington.edu>
810
811 * list.c: Added hand-written docs for `del{q,v,ete}1!'.
812
813 Thu Mar 2 12:38:30 2000 Greg J. Badros <gjb@cs.washington.edu>
814
815 * list.c: Moved append docs to append! Thanks Dirk Hermann. Also,
816 added append docs from R4RS.
817
818 * strings.c: Docstring typo fix, + eliminate unneeded IMP tests.
819 Thanks Dirk Hermann!
820
821 * chars.h: Provide SCM_CHARP, SCM_CHAR, SCM_MAKE_CHAR and
822 deprecate SCM_ICHRP, SCM_ICHR, SCM_MAKICHR. Thanks Dirk Hermann!
823
824 * *.h, *.c: Use SCM_CHARP, SCM_CHAR, SCM_MAKE_CHAR throughout.
825 Drop use of SCM_P for function prototypes... assume an ANSI C
826 compiler. Thanks Dirk Hermann!
827
828 Sat Feb 19 12:20:12 2000 Greg J. Badros <gjb@cs.washington.edu>
829
830 * ports.c: Made `set-port-column!' and `set-port-line!' each
831 return SCM_UNSPECIFIED instead of a (not-scheme-object) integer
832 that caused a seg fault. Also fixed `set-port-column!'s
833 docstring. Thanks Han-Wen Nienhuys for finding the bug!
834
835 Sun Feb 13 19:11:42 2000 Greg J. Badros <gjb@cs.washington.edu>
836
837 * arbiters.c, eq.c, gc.c, guardians.c, list.c, ports.c, print.c,
838 regex-posix.c, scmsigs.c, stime.c, strings.c, variable.c, stime.c,
839 strings.c, variable.c: Added lots of documentation, cleaned up
840 some existing documentation. Occasionally changed formal params
841 to match docs. Also folded an #ifdef into the inners of a
842 primitive instead of having two copies of the primitive
843 (`get-internal-real-time', from stime.c)
844
845 Sun Feb 13 18:12:19 2000 Greg J. Badros <gjb@cs.washington.edu>
846
847 * ports.c: Added docs for primitives missing them. Written by
848 hand.
849
850 Sun Feb 13 09:40:36 2000 Greg J. Badros <gjb@cs.washington.edu>
851
852 * guile-doc-snarf.in: Use ${AWK} -f guile-func-name-check, not
853 just execing guile-func-name-check. Thanks Michael Livshin!
854
855 Thu Feb 10 11:43:23 2000 Greg J. Badros <gjb@cs.washington.edu>
856
857 * guile-snarf.awk.in: Tweak to work with Sun/HP awk, removed some
858 dead code. Patch from Michael Livshin.
859
860 * guile-doc-snarf.in: Tweak to work with Sun/HP sh. Patch from
861 Michael Livshin.
862
863 2000-02-09 Gary Houston <ghouston@arglist.com>
864
865 * init.c (scm_init_standard_ports): when stdout is a tty, make the
866 current-output-port unbuffered by default. this is less confusing
867 for interactive use. it was line-buffered because of a
868 performance problem with unbuffered ports, but I think it will be
869 OK now.
870
871 2000-02-08 Gary Houston <ghouston@arglist.com>
872
873 * __scm.h: don't define long_long or ulong_long if HAVE_LONG_LONGS
874 is not defined.
875
876 * stime.c (scm_localtime, scm_mktime): if neither HAVE_TM_ZONE nor
877 HAVE_TZNAME are defined, use an empty string instead of giving two
878 spurious compile-time errors.
879
880 Tue Feb 8 13:57:46 2000 Greg J. Badros <gjb@cs.washington.edu>
881
882 * ports.c: Doc patches from Richard Kim. Pasted from MIT Scheme.
883 Thanks Richard!
884
885 Mon Feb 7 09:07:31 2000 Greg J. Badros <gjb@cs.washington.edu>
886
887 * macros.c: Doc patches from Richard Kim. Pasted from scm.texi.
888
889 Sun Feb 6 20:26:21 2000 Greg J. Badros <gjb@cs.washington.edu>
890
891 * pairs.c: Doc patches from Richard Kim. Pasted from MIT Scheme
892 (which is GNU GPL'd).
893
894 2000-01-31 Gary Houston <ghouston@arglist.com>
895
896 * strings.h: don't use SCM_P. don't include <string.h>.
897 * error.c, gh_data.c, ports.c, script.c, strop.c: include <string.h>.
898
899 * strings.c (scm_string_ref): make the 2nd argument compulsory.
900 previously it defaulted to zero for no good reason that I can see.
901 use a local variable for SCM_INUM (k). replace
902 SCM_VALIDATE_INUM_DEF with SCM_VALIDATE_INUM_COPY.
903
904 (scm_makfromstr): cosmetic changes.
905
906 (scm_string): Accept only chars in the list, not strings, for
907 conformance to R5RS (particularly for list->string, which is
908 supposed to be the inverse of string->list.) remove
909 SCM_DEFER_INTS/SCM_ALLOW_INTS, which is unnecessary since
910 scm_makstr handles the cell allocation. when reporting wrong-type
911 arg, don't report the position as 1.
912
913 * posix.c (scm_init_posix): intern PIPE_BUF if it's defined.
914
915 2000-01-29 Gary Houston <ghouston@arglist.com>
916
917 * posix.c (scm_pipe): rewrote the docstring.
918
919 * filesys.c (scm_select, retrieve_select_type, get_element,
920 fill_select_type, set_element): modified so that Scheme
921 "select" tests port buffers for the ability to provide input
922 or accept output. Previously only the underlying file descriptors
923 were checked. Rewrote the docstring.
924
925 Thu Jan 27 10:14:25 2000 Greg J. Badros <gjb@cs.washington.edu>
926
927 * vectors.c, symbols.c, strorder.c: Documentation cut and pasted
928 from Gregg Reynolds. Thanks Gregg!
929
930 Thu Jan 27 09:59:38 2000 Greg J. Badros <gjb@cs.washington.edu>
931
932 * strop.c (scm_i_index): Obfuscated commented-out SCM_DEFINE by
933 adding "x" prefix to the line so that guile-func-name-check
934 doesn't complain unnecessarily.
935
936 Wed Jan 26 17:33:52 2000 Greg J. Badros <gjb@cs.washington.edu>
937
938 * throw.c: Factor out an #ifdef/#else/#endif choice more finely
939 for maintainability.
940
941 * strop.c: Documentation added by Gregg A. Reynolds. Pasted in
942 from qdocs, RnRs.
943
944 Wed Jan 26 10:02:11 2000 Greg J. Badros <gjb@cs.washington.edu>
945
946 * tag.c: Added doc for `tag', but mark as deprecated since Mikael
947 suggests removing tag.c altogether (and using a new `class-of'
948 instead).
949
950 * strings.c: Added documentation from Gregg A. Reynolds. Edited
951 a bit by me to use FOO instead of @var{foo} and to have the
952 summary come before preconditions on input. Also dropped trailing
953 (rnrs) note.
954
955 * gsubr.c: Do not use SCM_DEFINE for `gsubr-apply'. Register the
956 function with scm_make_subr_opt w/ last arg of 0 so it is not
957 visible at the Scheme level. Mikael says that this is the right
958 thing because the first arg to the proc is the guts of a compiled
959 closure and shouldn't be exposed to the Scheme level.
960
961 Tue Jan 25 17:15:47 2000 Greg J. Badros <gjb@cs.washington.edu>
962
963 * sort.c: typo in comment fixed.
964
965 * keywords.c: Added documentation.
966
967 * guardians.c: Added documentation (could be better).
968
969 * gc.c: Added docs for gc-set-debug-check-freelist.
970
971 * eq.c: Added docs for eq?, eqv? equal? abridged from R4RS.
972
973 * boolean.c: Added docs for `not', `boolean?' (by hand).
974
975 Tue Jan 25 13:28:56 2000 Greg J. Badros <gjb@cs.washington.edu>
976
977 * random.c: Added documentation, from SLIB page:
978 http://angela.ctrl-c.liu.se/~calle/scheme/slib_toc.html
979
980 Mon Jan 24 17:50:20 2000 Greg J. Badros <gjb@cs.washington.edu>
981
982 * variable.c, version.c: Added documentation, written by hand
983 since I could not find anything already written that was
984 relevant.
985
986 2000-01-23 Gary Houston <ghouston@arglist.com>
987
988 * filesys.c (scm_chown): omit port/fdes support if HAVE_FCHOWN is
989 not defined (thanks to Richard Y. Kim).
990
991 Thu Jan 20 13:00:38 2000 Greg J. Badros <gjb@cs.washington.edu>
992
993 * Makefile.in: Removed, this is auto-generated.
994
995 Thu Jan 20 11:33:47 2000 Dirk Hermann <dirk@ida.ing.tu-bs.de> --applied 01/20/00 gjb
996
997 * list.c: Put some variable initialization code at the point of
998 declaration; Added a comment for list*; Formatting changes.
999
1000 * load.c: use SCM_NNULLP to make sure the end of a list is not
1001 reached yet.
1002
1003 2000-01-18 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
1004
1005 * backtrace.c (scm_display_error_message): Bugfix: Don't use
1006 result of scm_list_p as C boolean.
1007 (scm_display_error_message, scm_set_print_params_x): Use new
1008 validation macros. (Thanks to Dirk Herrmann.)
1009
1010 * net_db.c (scm_resolv_error): Cast result from hstrerror.
1011
1012 * strports.c (st_end_input): Inserted parenthesis to get operator
1013 grouping correct.
1014
1015 * list.h (scm_init_list): Removed SCM_P around prototypes.
1016
1017 * fports.c, list.c, numbers.c, ports.c, stime.c, symbols.c,
1018 filesys.c, posix.c: Converted docstrings to ANSI C format and
1019 escaped " occurring inside string literals.
1020
1021 Tue Jan 18 13:21:08 2000 Mikael Djurfeldt <mdj@r11n07-s.pdc.kth.se>
1022
1023 * posix.c (scm_mknod): Escape " occuring inside docstring.
1024
1025 2000-01-18 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
1026
1027 * alist.c, chars.c, debug.c, dynl.c, dynwind.c, error.c, eval.c,
1028 evalext.c, filesys.c, gc.c, hash.c, hashtab.c, ioext.c,
1029 keywords.c, list.c, load.c, macros.c, net_db.c, numbers.c,
1030 objprop.c, ports.c, posix.c, print.c, procprop.c, procs.c,
1031 ramap.c, regex-posix.c, root.c, scmsigs.c, simpos.c, socket.c,
1032 stacks.c, stime.c, strings.c, strop.c, strports.c, struct.c,
1033 symbols.c, throw.c, unif.c, vectors.c, version.c, vports.c,
1034 weaks.c: Converted docstrings to ANSI C format.
1035
1036 * filesys.c (scm_chmod), simpos.c (scm_system), version
1037 (scm_version), vports (scm_make_soft_port): Escape " occuring
1038 inside docstring.
1039
1040 Mon Jan 17 11:41:22 2000 Greg J. Badros <gjb@cs.washington.edu>
1041
1042 * scm_validate.h: Added SCM_VALIDATE_ULONG_COPY,
1043 SCM_VALIDATE_LONG_COPY
1044
1045 * numbers.c: Use SCM_VALIDATE_ULONG_COPY, instead of
1046 SCM_VALIDATE_INUM_COPY to let bigger numbers be used. Rename a
1047 couple of formal arguments (and fix their uses) to make arguments
1048 match the documentation.
1049
1050 2000-01-14 <mstachow@alum.mit.edu>
1051
1052 * Makefile.am: Augment path when running guile-doc-snarf so
1053 guile-func-name-check is found.
1054
1055 Fri Jan 14 09:34:55 2000 Greg J. Badros <gjb@cs.washington.edu>
1056
1057 * scm_validate.h (SCM_NUM2LONG_DEF): Fix this macro to just use
1058 def, not SCM_MAKINUM(def); thanks Janis Bzerins!
1059
1060 Wed Jan 12 00:06:53 2000 Greg J. Badros <gjb@cs.washington.edu>
1061
1062 * net_db.c (s_scm_inet_makeaddr): Use SCM_NUM2ULONG since that's
1063 the way guile-1.3.4 worked, but #if 0 out the version using
1064 SCM_VALIDATE_INUM_COPY for stricter testing.
1065
1066 Tue Jan 11 18:24:18 2000 Greg J. Badros <gjb@cs.washington.edu>
1067
1068 * guile-doc-snarf.in: Use new $fullfilename for running
1069 guile-func-name-check, and put "$fullfilename" and "$filename" in
1070 quotes at uses to make sure re-splitting on whitespace does not
1071 occur (so filenames w/ embedded whitespace would work okay, though
1072 I sure hope we never have to deal with that! :-) ). Thanks to
1073 Mikael for pointing out the source_dir != build_dir was broken.
1074
1075 Tue Jan 11 17:42:40 2000 Greg J. Badros <gjb@cs.washington.edu>
1076
1077 * scm_validate.h: Added SCM_NUM2LONG_DEF macro. Make
1078 SCM_OUT_OF_RANGE use SCM_MAKINUM, not scm_long2num. Added
1079 SCM_COERCE_ROSTRING macro. Added SCM_VALIDATE_NONEMPTYLIST
1080 macro. Fix SCM_VALIDATE_STRINGORSUBSTR macro to not use SLOPPY
1081 variants.
1082
1083 * ports.c (scm_port_closed_p): Validate that the arg is a PORT,
1084 then return whether it's an open port (was validating that it was
1085 an open port -- this was a bug I introduced back in December, but
1086 my careful reading of diffs caught it).
1087
1088 * numbers.c: Recombine the two conditional-compilation paths for
1089 all the log* primitives -- they were split based on #ifndef
1090 scm_long2num; factored out a SCM_LOGOP_RETURN macro, and fixed
1091 some bugs and inconsistencies in the two sets of implementations.
1092 (scm_lognot) Fixed *atrocious* re-use of a SCM as an integer!
1093
1094 * ioext.c: Use SCM_ASSERT_RANGE in a couple places, and
1095 SCM_VALIDATE_INUM_COPY once where it should've been used.
1096
1097 * fluids.c (scm_internal_with_fluids): Use
1098 SCM_VALIDATE_LIST_COPYLEN.
1099
1100 * filesys.c: Use SCM_NUM2LONG instead of SCM_VALIDATE_INUM_COPY;
1101 this is questionable as it relaxes type safety, but other changes
1102 were useful and all SCM_NUM2LONG's should probably be
1103 revisited. Use SCM_OUT_OF_RANGE, SCM_WRONG_TYPE_ARG.
1104
1105 * evalext.c: line-break change on 1 line.
1106
1107 * eval.c (nconc2last): Takes a non-empty list as its first
1108 argument, not just a list.
1109
1110 * dynl.c: Use new SCM_COERCE_ROSTRING macro.
1111
1112 Tue Jan 11 15:44:23 2000 Greg J. Badros <gjb@cs.washington.edu>
1113
1114 * dynl.c, feature.c, filesys.c, fports.c, list.c, load.c,
1115 net_db.c, sort.c, stacks.c, unif.c: Use SCM_WTA, SCM_MISC_ERROR
1116 where possible.
1117
1118 * symbols.c (scm_sysintern0): Fixed the function name in a
1119 scm_misc_error invocation.
1120
1121 * print.c (scm_simple_format): Do not need SCM_COERCE_SUBSTR, and
1122 use scm_return_first to ward off latent GC bug that Mikael caught.
1123
1124 * async.c: Use SCM_VALIDATE_ASYNC_COPY one place where it wasn't
1125 used before but should've been.
1126
1127 2000-01-12 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
1128
1129 * snarf.h (SCM_PROC1): Replaced SCM (*) (...) with
1130 SCM_FUNC_CAST_ARBITRARY_ARGS.
1131
1132 Tue Jan 11 13:44:07 2000 Greg J. Badros <gjb@cs.washington.edu>
1133
1134 * guile-func-name-check.in: Added this script to statically check
1135 #define FUNC_NAME, #undef FUNC_NAME in the source.
1136
1137 * sort.c, posix.c: Fix #undef FUNC_NAME lines to not have trailing
1138 redundant comment, semicolon; caught by new guile-func-name-check
1139 script.
1140
1141 * debug.c: Fix mistaken #define FUNC_NAME for scm_make_iloc.
1142 Caught by new guile-func-name-check-script.
1143
1144 * Makefile.am: Added guile-func-name-check to bin_SCRIPTS
1145
1146 * ramap.c: Fix #if 0'd out code to be syntactically acceptable to
1147 guile-func-name-check.
1148
1149 * guile-doc-snarf.in: Run guile-func-name-check on the file before
1150 doing the snarf.
1151
1152 Tue Jan 11 11:31:10 2000 Greg J. Badros <gjb@cs.washington.edu>
1153
1154 * fports.c, ports.c, ports.h, strports.c, vports.c: Make write
1155 port function take const void*, not void*.
1156
1157 Tue Jan 11 11:18:07 2000 Greg J. Badros <gjb@cs.washington.edu>
1158
1159 * scm_validate.h, chars.c, ports.c, print.c, read.c, strings.c,
1160 strop.c: Use SCM_VALIDATE_ICHR, SCM_VALIDATE_ICHR_COPY instead of
1161 SCM_VALIDATE_CHAR, SCM_VALIDATE_CHAR_COPY. Change made for
1162 consistency with the other macros dealing with immediate
1163 characters. (Similar to INT -> INUM change a week or so ago).
1164
1165 Tue Jan 11 10:41:46 2000 Greg J. Badros <gjb@cs.washington.edu>
1166
1167 * dynl.c, error.c, eval.c, feature.c, filesys.c, fports.c, list.c, load.c,
1168 net_db.c, read.c, socket.c: Update error messages to use ~A for
1169 %s, ~S for %S to work with new `simple-format' format and be
1170 standardized better.
1171
1172 * print.h, print.c (scm_simple_format): Added `simple-format'
1173 primitive. It's the old scm_display_error, with ARGS now a rest
1174 parameter, and the destination first instead of last (and a couple
1175 new capabilities inspired by `format' -- #t as destination means
1176 current-output-port, #f means return the formatted text as a
1177 string.
1178
1179 * gh.h, gh_data.c, ports.h, ports.c: Added some missing const specifications.
1180
1181 * backtrace.c (scm_display_error_message): Rewrote to use
1182 scm_simple_format() procedure.
1183
1184 * __scm.h: Added commented-out #define of GUILE_DEBUG_FREELIST
1185
1186 2000-01-09 Marius Vollmer <mvo@zagadka.ping.de>
1187
1188 Finally applied the libltdl patch from Thomas Tanner, with slight
1189 modifications.
1190
1191 * DYNAMIC-LINKING: Removed because it is obsolete.
1192 * dynl.c: Use ANSI prototypes.
1193 (sysdep_dynl_link): Use lt_dlopenext instead of lt_dlopen.
1194 * scmconfig.h.in: Do not change, as it is automatically generated.
1195
1196 1999-07-25 Thomas Tanner <tanner@ffii.org>
1197
1198 * dynl-dl.c, dynl-dld.c, dynl-shl.c, dynl-vms.c: deleted
1199 (obsolete)
1200 * Makefile.am: likewise, add INCLTDL (libltdl headers) to
1201 INCLUDES, set dlpreopened files in LDFLAGS, link libguile
1202 against libltdl
1203 * dynl.c: use libltdl if DYNAMIC_LINKING is enabled,
1204 * guile.c: register preloaded modules
1205 * scmconfig.h.in: remove obsolete symbols
1206
1207 2000-01-09 Gary Houston <ghouston@arglist.com>
1208
1209 * These changes should make it unnecessary to call tzset from
1210 Scheme after modifying the TZ environment variable, even if the
1211 system date facilities cache the value.
1212 * stime.c (setzone, scm_localtime): added comments.
1213 (tzset): don't define a noop tzset macro if HAVE_TZSET not defined.
1214 (setzone): don't call tzset.
1215 (restorezone): call tzset only if HAVE_TZSET is defined.
1216 (scm_tzset): don't define if HAVE_TZSET not defined. Change the
1217 doc string to indicate that this procedure isn't likely to do
1218 anything useful.
1219 (scm_localtime, scm_strftime, scm_mktime): call tzset if
1220 LOCALTIME_CACHE is defined.
1221
1222 2000-01-09 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
1223
1224 * posix.c (scm_sync): Return SCM_UNSPECIFIED.
1225
1226 2000-01-09 Gary Houston <ghouston@arglist.com>
1227
1228 * eval.c: define scm_unbound_variable_key ('unbound-variable).
1229 scm_lookupcar1: throw an error with key 'unbound-variable instead
1230 of 'misc-error when an unbound variable is encountered.
1231
1232 * filesys.c (scm_mkdir, scm_rmdir, scm_getcwd, scm_select,
1233 scm_symlink, scm_readlink, scm_lstat),
1234 posix.c (scm_setpgid, scm_setsid, scm_ctermid, scm_tcgetpgrp,
1235 scm_tcsetpgrp, scm_uname, scm_setlocale, scm_mknod, scm_nice,
1236 scm_sync),
1237 simpos.c (scm_system),
1238 stime.c (scm_times, scm_strptime):
1239 move the HAVE_XXX feature tests out of the procedure bodies.
1240 don't use SCM_SYSMISSING.
1241 scm_validate.h (SCM_SYSMISSING): removed.
1242 error.h, error.c (scm_sysmissing): comment that this is deprecated.
1243 see ChangeLog entry for 1999-12-28.
1244
1245 Sat Jan 8 19:52:04 2000 Greg J. Badros <gjb@cs.washington.edu>
1246
1247 * scm_validate.h (SCM_VALIDATE_BOOL_COPY): Fix typo.
1248
1249 Sat Jan 8 17:06:46 2000 Greg J. Badros <gjb@cs.washington.edu>
1250
1251 * backtrace.c: Fix spelling typo in a comment.
1252
1253 * snarf.h: Use new SCM_DOCS macro to encapsulate the non SCM_INIT
1254 text. Reformatted some of the expansions.
1255
1256 Fri Jan 7 15:50:46 2000 Greg J. Badros <gjb@cs.washington.edu>
1257
1258 * scm_validate.h (SCM_OUT_OF_RANGE): Use scm_out_of_range_pos to
1259 report the position of the argument.
1260
1261 * error.h, error.c (scm_out_of_range_pos): Added this function to
1262 take extra "pos" argument, the position number of the errant
1263 argument.
1264
1265 * debug.c: Use SCM_OUT_OF_RANGE instead of scm_out_of_range.
1266
1267 * snarf.h: Use SCM_HERE and SCM_INIT as names, not SCM_NOTSNARF
1268 and SCM_SNARFING. Also put the %%% in the SCM_INIT since Mikael
1269 prefers that and I'm reasonably indifferent.
1270
1271 Fri Jan 7 15:03:32 2000 Greg J. Badros <gjb@cs.washington.edu>
1272
1273 * snarf.h: Factor out differences between C++ and non-C++ into
1274 SCM_FUNC_CAST_ARBITRARY_ARGS macro. Modify all the snarf macro
1275 definitions to use SCM_NOTSNARF and SCM_SNARFING macros (like
1276 Mikael's macros, below, but changed names and SCM_SNARFING no
1277 longer expands to include %%% -- that must appear in the argument
1278 so that the token appears at the call-site as a reminder).
1279
1280 2000-01-07 Mikael Djurfeldt <mdj@thalamus.nada.kth.se>
1281
1282 * snarf.h (SCM_INSITU, SCM_INIT): New snarf macros for use in user
1283 snarf macro definitions.
1284
1285 2000-01-06 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
1286
1287 * chars.c (scm_integer_to_char): Use Greg's nice
1288 SCM_VALIDATE_INUM_RANGE macro for argument checking for closer
1289 adherence to R5RS.
1290
1291 Thu Jan 6 11:48:49 2000 Greg J. Badros <gjb@cs.washington.edu>
1292
1293 * *.c, snarf.h: Replace GUILE_PROC1 with SCM_DEFINE1 throughout.
1294
1295 Thu Jan 6 11:22:53 2000 Greg J. Badros <gjb@cs.washington.edu>
1296
1297 * Makefile.am (ETAGS_ARGS): Switch to SCM_DEFINE, SCM_DEFINE1
1298 instead of GUILE_PROC.
1299
1300 Thu Jan 6 11:21:49 2000 Greg J. Badros <gjb@cs.washington.edu>
1301
1302 * alist.c: Do not report mismatch errors on some uses of `tmp' (do
1303 this by using SCM_ARG2 instead of `2' in the SCM_VALIDATE_CONS
1304 macro call.
1305
1306 Thu Jan 6 09:54:33 2000 Dirk Herrmann <dirk@ida.ing.tu-bs.de> --gjb applied
1307
1308 * scm_validate.h: Remove some redundant NIMP tests.
1309
1310 * alist.c: minimize scope of the tmp variables, and initialize
1311 them when declared. The strange SCM_NIMP tests are replaced by
1312 SCM_CONSP tests that more closely reflect the intended semantics.
1313 However, we don't get a performance penalty here, because the
1314 SCM_CONSP test was performed by the ALISTCELL test anyway. * The
1315 extremely ugly use of ASRTGO macros was removed: The calls to
1316 ASRTGO were not encapsulated by "#ifndef SCM_RECKLESS", but got a
1317 label parameter that only exists when SCM_RECKLESS is not defined.
1318 This works, because ASRTGO itself is defined in a way that it only
1319 makes use of the label parameter if SCM_RECKLESS is not defined
1320 (shudder!). Does guile make at all use of the possibility to
1321 define SCM_RECKLESS? * Codesize is likely to be reduced, since
1322 instead of two calls to SCM_ASSERT performed by the ALISTCELL test
1323 we now only get one test.
1324
1325 * list.c: Use SCM_NNULLP, not SCM_NIMP as appropriate. Also use
1326 SCM_NULLP instead of SCM_IMP. Drop use of "register" keyword on
1327 some variables in `list?'. Fix `reverse' and `reverse!'
1328 primitives to handle improper lists better.
1329
1330 Wed Jan 5 11:24:53 2000 Greg J. Badros <gjb@cs.washington.edu>
1331
1332 * *.[ch]: Whitespace changes -- added space after SCM_VALIDATE_*
1333 macros and SCM_DEFINE macros to match GNU coding standards.
1334
1335 Wed Jan 5 11:04:24 2000 Greg J. Badros <gjb@cs.washington.edu>
1336
1337 * *.[ch]: Replace GUILE_PROC w/ SCM_DEFINE.
1338
1339 Wed Jan 5 10:59:06 2000 Greg J. Badros <gjb@cs.washington.edu>
1340
1341 * *.[ch]: Replace SCM_VALIDATE_INT w/ SCM_VALIDATE_INUM for
1342 better consistency with the names of other SCM_VALIDATE_ macros
1343 and better conformance to guile naming policy.
1344
1345 Wed Jan 5 10:50:39 2000 Greg J. Badros <gjb@cs.washington.edu>
1346
1347 * ports.c (s_scm_close_all_ports_except): Use SCM_ARG1 in a
1348 SCM_VALIDATE instead of 1 to avoid a check on the argument (since
1349 it's not the actual name of the formal).
1350
1351 * guile-snarf.awk.in: Do argument/number mismatch checking and
1352 print warnings in an Emacs compile-mode parseable format.
1353
1354 * struct.c: Use SCM_ASSERT_RANGE instead of SCM_ASSERT w/
1355 SCM_OUTOFRANGE as 3rd argument.
1356
1357 * random.c: Fix argument/number mismatch (that I introduced :-( ).
1358
1359 * __scm.h: Do not #define SCM_ARG* when snarfing;
1360 lets us distinguish between 1 and SCM_ARG1 when snarfing as only
1361 the former (using the number) requires the argument to match the
1362 formal in the current argument snarfing check.
1363
1364 * snarf.h: Give new definition of SCM_ASSERT when in
1365 snarfing mode to output a lexically-identifiable sequence that the
1366 guile-snarf.awk script uses to verify argument/position matching.
1367
1368 * ramap.c: Remove extraneous #undef FUNC_NAME.
1369
1370 Wed Jan 5 08:36:38 2000 Greg J. Badros <gjb@cs.washington.edu>
1371
1372 * guile-doc-snarf.awk.in: Removed -- guile-snarf.awk.in is the
1373 current version of the same functionality; it writes the .x output
1374 to stdout instead of directly into the file.
1375
1376 Wed Jan 5 08:15:04 2000 Greg J. Badros <gjb@cs.washington.edu>
1377
1378 * unif.c, symbols.c, strings.c, stacks.c, random.c, print.c,
1379 posix.c: Eliminated a bunch of SCM_NIMP(..)s that are now
1380 redundant with the safer macros. Patch from Dirk Hermann applied
1381 by hand. Thanks Dirk!
1382
1383 * scm_validate.h: Added SCM_VALIDATE_VECTOR_OR_DVECTOR for some
1384 uses in random.c.
1385
1386 * ramap.c: whitespace change.
1387
1388 Tue Jan 4 14:21:35 2000 Greg J. Badros <gjb@cs.washington.edu>
1389
1390 * options.c, objects.c, keywords.c, gc.c: Some redundant SCM_NIMP
1391 removals from Dirk Hermann.
1392
1393 * alist.c: Rename formals to match the parameter names in the
1394 documentation, updates to documentation. Thanks Dirk Hermann!
1395
1396 2000-01-04 Mikael Djurfeldt <mdj@mdj.nada.kth.se>
1397
1398 * eval.c (SCM_CEVAL): Reverse order of
1399 scm_stack_checking_enabled_p and SCM_STACK_OVERFLOW_P
1400 (Thanks to Brad Knotwell.)
1401
1402 Mon Jan 3 08:30:02 2000 Greg Harvey <Greg.Harvey@thezone.net> (applied --01/03/00 gjb)
1403
1404 * gc.c (scm_debug_newcell): Added SCM_SETCAR of the newly
1405 allocated cell.
1406
1407 * pairs.h: Added a comment about the need for the SCM_SETCAR in
1408 SCM_NEWCELL macro.
1409
1410 Mon Jan 3 08:25:19 2000 Greg J. Badros <gjb@cs.washington.edu>
1411
1412 * dynl-vms.c, debug.c, coop-threads.c, backtrace.c, eval.c: More
1413 SCM_NIMP tests that were redundant are now eliminated. Patches
1414 from Dirk Hermann applied by hand.
1415
1416 The ChangeLog continues in the file: "ChangeLog-1996-1999"