bpt/guile.git
24 years ago * acconfig.h: add HAVE_ARRAYS.
Gary Houston [Fri, 19 Nov 1999 18:16:19 +0000 (18:16 +0000)]
* acconfig.h: add HAVE_ARRAYS.

* configure.in: add --disable-arrays option, probably temporary.

* the following changes allow guile to be built with the array
"module" omitted.  some of this stuff is just tc7 type support,
which wouldn't be needed if uniform array types were converted
to smobs.

* tag.c (scm_utag_bvect ... scm_utag_cvect): don't define unless
HAVE_ARRAYS.
(scm_tag): don't check array types unless HAVE_ARRAYS.

* sort.c (scm_restricted_vector_sort_x, scm_sorted_p):
remove the unused array types.
* (scm_stable_sort, scm_sort): don't support vectors if not
HAVE_ARRAYS.  a bit excessive.

* random.c (vector_scale, vector_sum_squares,
scm_random_solid_sphere_x, scm_random_hollow_sphere_x,
scm_random_normal_vector_x): don't define unless HAVE_ARRAYS.

* gh_data.c (makvect, gh_chars2byvect, gh_shorts2svect,
gh_longs2ivect, gh_ulongs2uvect, gh_floats2fvect, gh_doubles2dvect,
gh_uniform_vector_length, gh_uniform_vector_ref):
don't define unless HAVE_ARRAYS.
(gh_scm2chars, gh_scm2shorts, gh_scm2longs, gh_scm2floats,
gh_scm2doubles):
don't check vector types if not HAVE_ARRAYS.

* eq.c (scm_equal_p), eval.c (SCM_CEVAL), print.c (scm_iprin1),
gc.c (scm_gc_mark, scm_gc_sweep), objects.c (scm_class_of):
don't support the array types unless HAVE_ARRAYS is defined.

* tags.h: make nine tc7 types conditional on HAVE_ARRAYS.

* read.c (scm_lreadr): don't check for #* unless HAVE_ARRAYS is
defined (this should use read-hash-extend).

* ramap.c, unif.c: don't check whether ARRAYS is defined.

* vectors.c (scm_vector_set_length_x): moved here from unif.c.  call
scm_uniform_element_size if HAVE_ARRAYS.
vectors.h: prototype too.

* unif.c (scm_uniform_element_size): new procedure.

* init.c (scm_boot_guile_1): don't call scm_init_ramap or
scm_init_unif unless HAVE_ARRAYS is defined.

* __scm.h: don't define ARRAYS.

* Makefile.am (EXTRA_libguile_la_SOURCES): unif.c and ramap.c
moved here from libguile_la_SOURCES.

* Makefile.am (ice9_sources): add arrays.scm.

* boot-9.scm: load arrays.scm if 'array is provided.

* arrays.scm: new file with stuff from boot-9.scm.

24 years ago * configure.in: check for hstrerror.
Gary Houston [Thu, 18 Nov 1999 22:36:28 +0000 (22:36 +0000)]
* configure.in: check for hstrerror.

* socket.c (scm_htons, scm_ntohs, scm_htonl, scm_ntohl): new
functions for network data conversion.

* numbers.c (scm_num2long, scm_num2longlong):
throw out-of-range instead of wrong-type-arg if appropriate.
(scm_iint2str): handle -2^31 correctly.
(scm_num2long): handle -2^31 bignum correctly.
(scm_num2long_long): rewrite the bigdig case: basically copied
from scm_num2long.
numbers.h: (SCM_BITSPERLONGLONG): deleted.

* unif.c (rapr1): use sprintf instead of intprint for unsigned
longs: intprint can't cope with large values.

* numbers.c (scm_num2ulong): check more consistently that the
input is not negative.  if it is, throw out-of-range instead of
wrong-type-arg.

* ramap.c (scm_array_fill_int): don't limit fill to INUM for
uvect, ivect or llvect.
Check that fill doesn't overflow short uniform array.

* __scm.h: add another long to the definition of long_long and
ulong_long.

* unif.c (scm_raprin1): use 'l' instead of "long_long" in the
print representation of llvect.  read can't handle more than
one character.
(scm_dimensions_to_uniform_array): make "fill" an optional argument
instead of a rest argument.

* tags.h (scm_tc7_llvect): wasn't defined anywhere, so use the free
tag 29 for now.

* __scm.h: don't mention LONGLONGS.

* unif.c, numbers.c, eq.c, gc.c, print.c, eval.c, ramap.c:
replace LONGLONGS with HAVE_LONG_LONGS as set by configure.

* net_db.c (scm_inet_aton): throw errors using the misc-error key
instead of system-error.  inet_aton doesn't set errno.
system-error isn't right in gethost either, since it's throwing
the value of h_errno instead of errno. so:
(scm_host_not_found_key, scm_try_again_key,
scm_no_recovery_key, scm_no_data_key): new error keys.
(scm_resolv_error): new procedure, use the new keys.
(scm_gethost): call scm_resolv_error not scm_syserror_msg.

* error.c: (various): use scm_cons instead of scm_listify
to build short lists.

* boot-9.scm (read-hash-extend to set up arrays): add 'l' for
long_long uniform vectors.

* networking.scm (sethostent, setnetent, setprotoent, setservent):
take an optional argument STAYOPEN.  default is #f.

* readline.c (scm_init_readline): set rl_readline_name to Guile,
to allow conditionals in  .inputrc.

24 years ago * socket.c (scm_fill_sockaddr): zero the address structure before
Gary Houston [Thu, 4 Nov 1999 20:04:18 +0000 (20:04 +0000)]
* socket.c (scm_fill_sockaddr): zero the address structure before
use, in case it has a sin_len field and the OS doesn't like random
values (thanks to Bertrand Petit).

24 years ago1999-10-26 Mark Galassi <rosalia@lanl.gov>
Mark Galassi [Tue, 26 Oct 1999 19:38:05 +0000 (19:38 +0000)]
1999-10-26  Mark Galassi  <rosalia@lanl.gov>

* gh.h, gh_data.c (gh_symbol2scm): changed gh_symbol2scm() to take
a const char * argument, upon suggestion from Lynn Winebarger.

24 years ago * strports.c (st_end_input): avoid dubious pointer arithmetic.
Gary Houston [Tue, 26 Oct 1999 18:42:11 +0000 (18:42 +0000)]
* strports.c (st_end_input): avoid dubious pointer arithmetic.

24 years ago * tests/ports.test ("string ports"): test seeking/unreading from
Gary Houston [Sun, 24 Oct 1999 15:17:47 +0000 (15:17 +0000)]
* tests/ports.test ("string ports"): test seeking/unreading from
an input string and seeking an output string.

24 years ago * Move the responsibility for resetting port buffers from the
Gary Houston [Sun, 24 Oct 1999 11:25:31 +0000 (11:25 +0000)]
* Move the responsibility for resetting port buffers from the
caller of the ptob seek procedure to the implementation.  This
gives more control in general to the ptob seek: in particular the
change of 1999-10-20 can be made to work without breaking seek on
string ports.  There's a comment in NEWS about upgrading port
types.

* ports.c (scm_seek): don't reset the port buffers here.

* fports.c (fport_seek): reset the buffers, except for the
0 SEEK_CUR case.

* strports.c (st_end_input): (bug fix): decrement pt->read_pos by
offset.  check that it's not less than read_buf.
(st_seek): reset the buffers first, unless it's the 0 SEEK_CUR
case and currently reading.

24 years ago * ports.c (scm_seek): Add a special case for SEEK_CUR, offset 0,
Gary Houston [Wed, 20 Oct 1999 21:03:09 +0000 (21:03 +0000)]
* ports.c (scm_seek): Add a special case for SEEK_CUR, offset 0,
so that unread chars are not needlessly discarded.  (thanks to
Roland Orre).

24 years ago * tests/ports.test: in seek/tell test on input port, also test
Gary Houston [Wed, 20 Oct 1999 20:59:07 +0000 (20:59 +0000)]
* tests/ports.test: in seek/tell test on input port, also test
that ftell doesn't discard unread chars.

24 years agonotes
Gary Houston [Mon, 18 Oct 1999 20:57:47 +0000 (20:57 +0000)]
notes

24 years ago * fports.c (scm_fdes_to_port): always set rw_random if the fdes is
Gary Houston [Mon, 18 Oct 1999 20:49:29 +0000 (20:49 +0000)]
* fports.c (scm_fdes_to_port): always set rw_random if the fdes is
random access.  rw_active needs to be maintained even for single
directional ports, otherwise scm_seek and probably other things are
broken.  (thanks to Roland Orre).

* strports.c (scm_mkstrport): set rw_random to 1 unconditionally.

* ports.c (scm_add_to_port_table): initialise rw_random to 0.

* ports.h (scm_port): change the comments on rw_random and rw_active.

24 years ago * tests/ports.test: add seek/tell tests for unidirectional ports.
Gary Houston [Mon, 18 Oct 1999 20:47:37 +0000 (20:47 +0000)]
* tests/ports.test: add seek/tell tests for unidirectional ports.

24 years ago*** empty log message ***
Mikael Djurfeldt [Mon, 11 Oct 1999 17:42:01 +0000 (17:42 +0000)]
*** empty log message ***

24 years agoRegenerated
Mikael Djurfeldt [Mon, 11 Oct 1999 17:41:05 +0000 (17:41 +0000)]
Regenerated

24 years ago* Makefile.am (libguile_la_LDFLAGS): Bumped libguile version.
Mikael Djurfeldt [Mon, 11 Oct 1999 17:40:45 +0000 (17:40 +0000)]
* Makefile.am (libguile_la_LDFLAGS): Bumped libguile version.

24 years ago* tags.h (SCM_IM_DELAY): New immediate symbol.
Mikael Djurfeldt [Mon, 11 Oct 1999 17:40:32 +0000 (17:40 +0000)]
* tags.h (SCM_IM_DELAY): New immediate symbol.

24 years ago* print.c (scm_isymnames): Printed representation.
Mikael Djurfeldt [Mon, 11 Oct 1999 17:40:19 +0000 (17:40 +0000)]
* print.c (scm_isymnames): Printed representation.

24 years ago* ioext.c: Added #include "feature.h".
Mikael Djurfeldt [Mon, 11 Oct 1999 17:40:06 +0000 (17:40 +0000)]
* ioext.c: Added #include "feature.h".

24 years ago* eval.c, eval.h (scm_sym_delay): New global symbol.
Mikael Djurfeldt [Mon, 11 Oct 1999 17:39:37 +0000 (17:39 +0000)]
* eval.c, eval.h (scm_sym_delay): New global symbol.

24 years ago* eval.c, eval.h (scm_sym_delay): New global symbol.
Mikael Djurfeldt [Mon, 11 Oct 1999 17:39:24 +0000 (17:39 +0000)]
* eval.c, eval.h (scm_sym_delay): New global symbol.

* eval.c (unmemocopy, SCM_CEVAL): Handle SCM_IM_DELAY.
(scm_m_delay): Turned into a memoizing macro.

24 years ago*** empty log message ***
Jim Blandy [Mon, 11 Oct 1999 16:54:05 +0000 (16:54 +0000)]
*** empty log message ***

24 years ago* gh_data.c (gh_ints2scm, gh_doubles2scm): Make sure elements are
Jim Blandy [Mon, 11 Oct 1999 16:53:50 +0000 (16:53 +0000)]
* gh_data.c (gh_ints2scm, gh_doubles2scm): Make sure elements are
protected from GC while building the vector.  (Thanks to Bernard
Urban and Greg Harvey.)

24 years ago*** empty log message ***
Mikael Djurfeldt [Sun, 10 Oct 1999 17:49:48 +0000 (17:49 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Jim Blandy [Sun, 10 Oct 1999 06:45:29 +0000 (06:45 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Mikael Djurfeldt [Fri, 8 Oct 1999 11:51:03 +0000 (11:51 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Mikael Djurfeldt [Fri, 8 Oct 1999 10:46:38 +0000 (10:46 +0000)]
*** empty log message ***

24 years ago* throw.c (handler_message): Display backtrace if backtraces
Mikael Djurfeldt [Fri, 8 Oct 1999 10:46:26 +0000 (10:46 +0000)]
* throw.c (handler_message): Display backtrace if backtraces
enabled.

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 23:26:11 +0000 (23:26 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 23:17:56 +0000 (23:17 +0000)]
*** empty log message ***

24 years ago* autogen.sh: Don't call autoreconf at all; it's not reliable.
Jim Blandy [Tue, 5 Oct 1999 23:07:04 +0000 (23:07 +0000)]
* autogen.sh: Don't call autoreconf at all; it's not reliable.
Instead, call the various tools explicitly.  Invoke
guile-readline's autogen.sh script.

24 years ago* Makefile.in, configure, aclocal.m4: Deleted from CVS repository.
Jim Blandy [Tue, 5 Oct 1999 23:06:49 +0000 (23:06 +0000)]
* Makefile.in, configure, aclocal.m4: Deleted from CVS repository.
Run the autogen.sh script to create generated files like these.
* autogen.sh: New script, invoked by the top-level autogen.sh.

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 23:05:55 +0000 (23:05 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 23:05:38 +0000 (23:05 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 22:36:14 +0000 (22:36 +0000)]
*** empty log message ***

24 years ago* guile.m4: Moved here from top-level directory; see the ChangeLog
Jim Blandy [Tue, 5 Oct 1999 22:35:06 +0000 (22:35 +0000)]
* guile.m4: Moved here from top-level directory; see the ChangeLog
entry there.
* Makefile.am (aclocaldir, aclocal_DATA): New variables, ensuring
that guile.m4 gets installed.

24 years ago* guile.m4: Moved to guile-config.
Jim Blandy [Tue, 5 Oct 1999 22:34:46 +0000 (22:34 +0000)]
* guile.m4: Moved to guile-config.
* Makefile.am (aclocaldir, aclocal_DATA): Delete.
(EXTRA_DIST): Move qthreads.m4 here.

24 years ago* autogen.sh: Call autoheader explicitly. autoreconf runs
Jim Blandy [Tue, 5 Oct 1999 22:34:35 +0000 (22:34 +0000)]
* autogen.sh: Call autoheader explicitly.  autoreconf runs
autoconf before autoheader, and autoconf complains if
libguile/scmconfig.h.in doesn't exist.

24 years agoShouldn't have deleted this.
Jim Blandy [Tue, 5 Oct 1999 22:33:34 +0000 (22:33 +0000)]
Shouldn't have deleted this.

24 years ago* Makefile.in, configure, aclocal.m4: Deleted from CVS repository.
Jim Blandy [Tue, 5 Oct 1999 22:32:32 +0000 (22:32 +0000)]
* Makefile.in, configure, aclocal.m4: Deleted from CVS repository.
Run the autogen.sh script to create generated files like these.

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 20:49:29 +0000 (20:49 +0000)]
*** empty log message ***

24 years ago* Makefile.in, md/Makefile.in, time/Makefile.in: Deleted from CVS
Jim Blandy [Tue, 5 Oct 1999 19:12:03 +0000 (19:12 +0000)]
* Makefile.in, md/Makefile.in, time/Makefile.in: Deleted from CVS
repository.  Run the autogen.sh script to create generated files
like this one.

24 years ago* Makefile.in, scmconfig.h.in: Deleted from CVS repository. Run
Jim Blandy [Tue, 5 Oct 1999 19:08:37 +0000 (19:08 +0000)]
* Makefile.in, scmconfig.h.in: Deleted from CVS repository.  Run
the autogen.sh script to create generated files like these.

24 years ago* Makefile.in: Deleted from CVS repository. Run the autogen.sh
Jim Blandy [Tue, 5 Oct 1999 19:08:25 +0000 (19:08 +0000)]
* Makefile.in: Deleted from CVS repository.  Run the autogen.sh
script to create generated files like this one.

24 years agoDon't store generated files in the repository any more. Instead,
Jim Blandy [Tue, 5 Oct 1999 19:07:40 +0000 (19:07 +0000)]
Don't store generated files in the repository any more.  Instead,
require people to run autogen.sh on trees from snapshots and CVS.
* Makefile.in, acconfig.h, aclocal.m4, configure: Deleted.
* autogen.sh: New file.
* ANON-CVS, SNAPSHOTS: Updated instructions.

24 years ago*** empty log message ***
Jim Blandy [Tue, 5 Oct 1999 19:00:38 +0000 (19:00 +0000)]
*** empty log message ***

24 years ago* numbers.c (scm_string_to_number): Signal an error if radix is
Jim Blandy [Tue, 5 Oct 1999 18:57:31 +0000 (18:57 +0000)]
* numbers.c (scm_string_to_number): Signal an error if radix is
less than two.  (Thanks to Jorgen Schaefer.)

* print.c (scm_write, scm_display, scm_newline, scm_write_char):
Don't assume that the current output port is valid.  Somebody
might close it.  (Thanks to Bernard Urban.)

24 years ago*** empty log message ***
Jim Blandy [Sat, 2 Oct 1999 23:53:13 +0000 (23:53 +0000)]
*** empty log message ***

24 years ago* acconfig.h (HAVE_POSIX, HAVE_NETWORKING): Add comments.
Jim Blandy [Sat, 2 Oct 1999 23:52:23 +0000 (23:52 +0000)]
* acconfig.h (HAVE_POSIX, HAVE_NETWORKING): Add comments.

24 years ago* Makefile.am (EXTRA_libguile_la_SOURCES): fix typo.
Jim Blandy [Sat, 2 Oct 1999 23:01:46 +0000 (23:01 +0000)]
* Makefile.am (EXTRA_libguile_la_SOURCES): fix typo.

24 years ago* scmconfig.h.in: Add comments for HAVE_POSIX and HAVE_NETWORKING.
Jim Blandy [Sat, 2 Oct 1999 23:01:14 +0000 (23:01 +0000)]
* scmconfig.h.in: Add comments for HAVE_POSIX and HAVE_NETWORKING.

24 years agoRevert back to text generated by automake 1.4.
Jim Blandy [Sat, 2 Oct 1999 22:57:38 +0000 (22:57 +0000)]
Revert back to text generated by automake 1.4.

24 years ago* NEWS: More complete description for --enable-debug-freelist.
Greg J. Badros [Sat, 2 Oct 1999 22:55:36 +0000 (22:55 +0000)]
* NEWS: More complete description for --enable-debug-freelist.

24 years agoFix my last commit -- DEBUG_FREELIST renamed to GUILE_DEBUG_FREELIST
Greg J. Badros [Tue, 28 Sep 1999 00:54:26 +0000 (00:54 +0000)]
Fix my last commit -- DEBUG_FREELIST renamed to GUILE_DEBUG_FREELIST

24 years ago*** empty log message ***
Greg J. Badros [Tue, 28 Sep 1999 00:39:35 +0000 (00:39 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Greg J. Badros [Tue, 28 Sep 1999 00:18:34 +0000 (00:18 +0000)]
*** empty log message ***

24 years ago* __scm.h: Fix a bunch of macros that were missing do-while(0)
Greg J. Badros [Tue, 28 Sep 1999 00:18:16 +0000 (00:18 +0000)]
* __scm.h: Fix a bunch of macros that were missing do-while(0)
sandwiches.

* debug.c, eval.c: Fix buggy uses of SCM_ALLOW_INTS (missing
semicolon) exposed by the above change.

24 years ago*** empty log message ***
Greg J. Badros [Mon, 27 Sep 1999 23:35:33 +0000 (23:35 +0000)]
*** empty log message ***

24 years ago* stacks.c: Avoid compiler warning re: unitialized var.
Greg J. Badros [Mon, 27 Sep 1999 23:30:36 +0000 (23:30 +0000)]
* stacks.c: Avoid compiler warning re: unitialized var.

* scmconfig.h.in: Added DEBUG_FREELIST

* pairs.h: Fix macro that was not do-while(0) sandwiched.

* gc.h, gc.c: Added scm_gc_set_debug_check_freelist_x,
scm_map_free_list

24 years ago* configure.in: Added --enable-debug-freelist option.
Greg J. Badros [Mon, 27 Sep 1999 23:29:07 +0000 (23:29 +0000)]
* configure.in:  Added --enable-debug-freelist option.

* acconfig.h: Added DEBUG_FREELIST.

24 years ago(One has to prove oneself after each release in order to be called
Mikael Djurfeldt [Mon, 27 Sep 1999 20:29:33 +0000 (20:29 +0000)]
(One has to prove oneself after each release in order to be called
 "dodderer" ;-)

24 years agosupport new configure options: --disable-posix, --disable-net
Gary Houston [Sun, 26 Sep 1999 16:00:36 +0000 (16:00 +0000)]
support new configure options:  --disable-posix, --disable-net
and --disable-regex.

24 years ago*** empty log message ***
Jim Blandy [Sat, 25 Sep 1999 23:41:59 +0000 (23:41 +0000)]
*** empty log message ***

24 years ago* root.c (scm_make_root): Initialize all the fields of the new
Jim Blandy [Sat, 25 Sep 1999 23:41:09 +0000 (23:41 +0000)]
* root.c (scm_make_root): Initialize all the fields of the new
root.  GC could happen any time, you know.  (Thanks to Greg
Harvey.)

24 years ago* tests/reader.test: Check that number->string checks its radix
Jim Blandy [Sat, 25 Sep 1999 23:38:26 +0000 (23:38 +0000)]
* tests/reader.test: Check that number->string checks its radix
properly.

24 years ago*** empty log message ***
Jim Blandy [Sat, 25 Sep 1999 23:13:09 +0000 (23:13 +0000)]
*** empty log message ***

24 years ago* numbers.c (scm_number_to_string): Signal an error if radix is
Jim Blandy [Sat, 25 Sep 1999 23:11:23 +0000 (23:11 +0000)]
* numbers.c (scm_number_to_string): Signal an error if radix is
less than two.  (Thanks to Jorgen Schaefer.)

24 years ago*** empty log message ***
Jim Blandy [Sat, 25 Sep 1999 18:18:38 +0000 (18:18 +0000)]
*** empty log message ***

24 years agoStart a new version.
Jim Blandy [Sat, 25 Sep 1999 18:13:01 +0000 (18:13 +0000)]
Start a new version.

24 years ago*** empty log message ***
Jim Blandy [Sat, 25 Sep 1999 17:36:42 +0000 (17:36 +0000)]
*** empty log message ***

24 years ago*** empty log message ***
Jim Blandy [Sat, 25 Sep 1999 17:27:37 +0000 (17:27 +0000)]
*** empty log message ***

24 years ago* Makefile.am (EXTRA_libguile_la_SOURCES): Add memmove.c here,
Jim Blandy [Sat, 25 Sep 1999 17:26:14 +0000 (17:26 +0000)]
* Makefile.am (EXTRA_libguile_la_SOURCES): Add memmove.c here,
so automake will actually generate rules for it.
* Makefile.in: Regenerated.

24 years ago*** empty log message ***
Jim Blandy [Thu, 23 Sep 1999 06:09:57 +0000 (06:09 +0000)]
*** empty log message ***

24 years agoInclude pointer to NEWS item about readline.
Jim Blandy [Thu, 23 Sep 1999 05:37:30 +0000 (05:37 +0000)]
Include pointer to NEWS item about readline.

24 years agoAdd explanation of licensing issues to the description of the
Jim Blandy [Thu, 23 Sep 1999 05:36:00 +0000 (05:36 +0000)]
Add explanation of licensing issues to the description of the
readline interface.

24 years ago*** empty log message ***
Jim Blandy [Wed, 22 Sep 1999 19:59:11 +0000 (19:59 +0000)]
*** empty log message ***

24 years ago* configure.in: Call AM_PROG_CC_STDC.
Jim Blandy [Wed, 22 Sep 1999 19:58:54 +0000 (19:58 +0000)]
* configure.in: Call AM_PROG_CC_STDC.
* configure, aclocal.m4: Regenerated.

24 years ago* configure.in: Call AM_PROG_CC_STDC before AM_PROG_LIBTOOL, so
Jim Blandy [Wed, 22 Sep 1999 19:58:39 +0000 (19:58 +0000)]
* configure.in: Call AM_PROG_CC_STDC before AM_PROG_LIBTOOL, so
libtool knows how to get ANSI C behavior from the compiler.
* configure: Regenerated.

24 years agoAdded description of readline stuff.
Marius Vollmer [Wed, 22 Sep 1999 19:47:30 +0000 (19:47 +0000)]
Added description of readline stuff.

24 years ago*** empty log message ***
Jim Blandy [Wed, 22 Sep 1999 04:37:18 +0000 (04:37 +0000)]
*** empty log message ***

24 years ago* backtrace.c: #include "_scm.h" before testing whether
Jim Blandy [Wed, 22 Sep 1999 04:36:56 +0000 (04:36 +0000)]
* backtrace.c: #include "_scm.h" before testing whether
HAVE_UNISTD_H is #defined.

24 years ago... to injury.
Jim Blandy [Tue, 21 Sep 1999 04:21:36 +0000 (04:21 +0000)]
... to injury.

24 years agoAdd insults.
Jim Blandy [Tue, 21 Sep 1999 04:21:10 +0000 (04:21 +0000)]
Add insults.

24 years ago*** empty log message ***
Jim Blandy [Mon, 20 Sep 1999 23:57:55 +0000 (23:57 +0000)]
*** empty log message ***

24 years agoDoc fix.
Jim Blandy [Mon, 20 Sep 1999 23:57:44 +0000 (23:57 +0000)]
Doc fix.

24 years ago* tests/ports.test: Check that our input functions cope when
Jim Blandy [Mon, 20 Sep 1999 23:57:08 +0000 (23:57 +0000)]
* tests/ports.test: Check that our input functions cope when
current-input-port is closed.

24 years ago*** empty log message ***
Jim Blandy [Mon, 20 Sep 1999 23:36:16 +0000 (23:36 +0000)]
*** empty log message ***

24 years ago* read.c (scm_read): Don't assume that scm_cur_inp is always open.
Jim Blandy [Mon, 20 Sep 1999 23:34:57 +0000 (23:34 +0000)]
* read.c (scm_read): Don't assume that scm_cur_inp is always open.
* ports.c (scm_read_char): Same.
* ioext.c (scm_read_line): Same.
(Thanks to Bernard Urban.)

24 years ago*** empty log message ***
Mikael Djurfeldt [Mon, 20 Sep 1999 23:05:51 +0000 (23:05 +0000)]
*** empty log message ***

24 years ago* vectors.c (scm_vector_move_right_x): Bugfix: Remove side effect
Mikael Djurfeldt [Mon, 20 Sep 1999 23:01:33 +0000 (23:01 +0000)]
* vectors.c (scm_vector_move_right_x): Bugfix: Remove side effect
in SCM_ASSERT macro.
numbers.c: Removed #ifndef SCM_RECKLESS at places where type
checking may invoke a generic.  (Thanks to Michael Livshin.)

24 years ago*** empty log message ***
Jim Blandy [Mon, 20 Sep 1999 22:52:08 +0000 (22:52 +0000)]
*** empty log message ***

24 years ago* tests/regexp.test: Check regexp-substitute/global when there are
Jim Blandy [Mon, 20 Sep 1999 22:52:02 +0000 (22:52 +0000)]
* tests/regexp.test: Check regexp-substitute/global when there are
no matches.  (Duh.)

24 years ago*** empty log message ***
Jim Blandy [Mon, 20 Sep 1999 22:42:04 +0000 (22:42 +0000)]
*** empty log message ***

24 years agoFix
Mikael Djurfeldt [Mon, 20 Sep 1999 22:25:10 +0000 (22:25 +0000)]
Fix

24 years ago*** empty log message ***
Mikael Djurfeldt [Mon, 20 Sep 1999 22:17:14 +0000 (22:17 +0000)]
*** empty log message ***

24 years agoFix
Mikael Djurfeldt [Mon, 20 Sep 1999 22:10:38 +0000 (22:10 +0000)]
Fix

24 years ago* __scm.h (SCM_WTA_DISPATCH_0, SCM_GASSERT0): New macros.
Mikael Djurfeldt [Mon, 20 Sep 1999 22:09:21 +0000 (22:09 +0000)]
* __scm.h (SCM_WTA_DISPATCH_0, SCM_GASSERT0): New macros.

24 years agotyping correction
Gary Houston [Mon, 20 Sep 1999 22:04:51 +0000 (22:04 +0000)]
typing correction

24 years ago*** empty log message ***
Mikael Djurfeldt [Mon, 20 Sep 1999 21:52:48 +0000 (21:52 +0000)]
*** empty log message ***

24 years ago* numbers.c (scm_max, scm_min, scm_sum, scm_difference,
Mikael Djurfeldt [Mon, 20 Sep 1999 21:51:50 +0000 (21:51 +0000)]
* numbers.c (scm_max, scm_min, scm_sum, scm_difference,
scm_product, scm_divide): Bugfix: Don't pass SCM_UNDEFINED to the
generic function if the asubr is called with only one arg.

24 years ago * configure.in: check availability of siginterrupt.
Gary Houston [Mon, 20 Sep 1999 21:32:23 +0000 (21:32 +0000)]
* configure.in: check availability of siginterrupt.
* scmsigs.c (scm_sigaction): add SA_RESTART to flags only if
HAVE_RESTARTABLE_SYSCALLS.
(scm_init_scmsigs): use siginterrupt if it's available.  not
everyone who has restartable syscalls has SA_RESTART it seems.
(scm_sigaction): use scm_num2long/scm_long2num when converting
SIG_DFL/SIG_IGN, in case it doesn't fit in an INUM.  use
scm_integer_p to test the type.