libguile/ChangeLog

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