0eb62ab5c5
Why am I even in the repository?
1997-04-08 21:45:50 +00:00
0b4cfa8d21
Added warning about opening the scsh structure before the scheme structure
...
in modules.
1997-04-08 21:44:41 +00:00
0b90a5a030
Added warning about opening scsh structure before scheme structure
...
in modules.
1997-04-08 21:42:23 +00:00
33e3bc8c62
Was not properly testing for error in fdport_write_substring's call to
...
fwrite().
1997-04-08 20:07:30 +00:00
18e8530076
SIGPIPE now runs with a no-op sighandler, instead of SIG_IGN. This is
...
so that if scsh exec's a sleazy C program that needs the default
kill-me-I'm-a-loser action of SIGPIPE, the handler will be reset to SIG_DFL
on exec, rather than continue to be ignored.
1997-04-08 20:06:06 +00:00
459658efa6
floatnums now available
1997-04-05 22:16:21 +00:00
c89fb29390
more elegant signal ifdef
1997-04-05 22:16:05 +00:00
0225bb0bf3
user contributed file
1997-04-05 22:15:38 +00:00
bf28f0aa86
Documented static heap linker. Such as it is.
1997-04-04 22:41:06 +00:00
4c26efe136
No more draft status. Version 0.4.5.
1997-04-04 22:40:44 +00:00
494787705d
Took out "draft" code. Upgraded release to 0.4.5
1997-04-04 22:39:55 +00:00
8233556820
Random update to regex doc.
1997-04-04 22:36:16 +00:00
db69cd677b
Some functions that were passed string "out" paramters (char**)'s were
...
neglecting to assign them in error cases. We'd return from the C function
into the cig C stub with random garbage still in these pointers. The
string rep converters would then treat this garbage as a real pointer to
be copied into malloc'd storage, and havoc would result.
I fixed the code to assign 0 to the pointers in error cases. I should fix
cig to pre-initialise out parameters. The only real problem code was in
userinfo1.c, but I scanned all the sources, and fixed a few suspicious cases
here and there as well.
1997-04-04 20:49:55 +00:00
1a1950d2f5
Fixed typo bug in set-file-group.
1997-04-02 20:12:33 +00:00
70679cd247
You don't have to be root to call set-file-group.
1997-04-02 20:11:50 +00:00
934905ed57
Hacked code so that you don't get [for-syntax] noise during loads when
...
in batch mode.
1997-04-02 18:47:06 +00:00
ddddd1d31b
More stuff.
1997-03-31 22:33:33 +00:00
8c65247c28
New stuff.
1997-03-31 22:20:22 +00:00
0e28399b12
Changed regexp-substring so that unmatched submatches return #f instead of
...
raising an error.
1997-03-28 20:55:23 +00:00
5e85d401b9
Updated manual.
1997-03-11 02:49:52 +00:00
230263ee0d
Fixed an arity bug in %set-unix-signal-handler's call to
...
%%set-unix-signal-handler.
1997-03-10 23:46:42 +00:00
8d7dee0bfa
*** empty log message ***
1997-03-10 04:57:18 +00:00
417b6046f8
Early autoreaping now done via SIGCHLD signal handler.
...
Ignored signals were done by putting a noop handler on the signal.
This could screw up passing ignores across execs.
Now, when you set an S48 interrupt handler to #f, we pass this through
to Unix.
1997-03-09 07:28:20 +00:00
8d6805a733
Rewrote SLEEP (and added SLEEP-UNTIL) so that it doesn't hang up signal
...
delivery in C code.
We now ,open floatnums in ,user so that users won't lose on flonum code.
1997-03-09 07:25:06 +00:00
2773441f05
Rewrote SLEEP (and added SLEEP-UNTIL) so that it doesn't hang up signal
...
delivery in C code.
1997-03-09 07:22:41 +00:00
101a61082f
solaris 2.5 tested
1997-03-09 02:37:06 +00:00
e989a744e4
Had to shuffle packages due to broken package dependencies.
...
The whole thing is a disaster and should be completely re-factorised.
1997-02-27 04:50:57 +00:00
6a7ecc38a1
Added #ifdef linux conditionalisation to the linux-specific fixup code.
1997-02-27 04:49:28 +00:00
b980922f0d
Fixed bug in stdio->stdports -- it was using fdes->inport to make the
...
output ports.
1997-02-26 22:21:48 +00:00
0229b80a6c
Linux-specific support.
1997-02-26 22:18:29 +00:00
b02ac4dc83
hacked in a gross fix for linux bug. The problem is that
...
linux stdio functions retry when interrupted, instead of returning
errno=EINTR. linux/libansi.c now has a #include "fix_stdio.c"
in it to get the fixup code included. This is very bogus.
1997-02-26 20:38:39 +00:00
66735d9c9e
fdports1.c: hacked in a gross fix for linux bug. The problem is that
...
linux stdio functions retry when interrupted, instead of returning
errno=EINTR.
regexp: New regexp-substitute function.
scsh-interfaces: Fixed a typo in a type decl.
1997-02-26 20:37:53 +00:00
9ccd709a52
spelling error
1997-02-21 08:51:06 +00:00
630f678236
Added a few bits.
1997-02-19 23:29:41 +00:00
bad528299e
Ripped out now-useless cloexec_unrevealed.
1997-02-19 23:28:38 +00:00
bf449b53b7
Added types to decls, possibly other stuff.
1997-02-19 23:24:53 +00:00
06eb60980d
Upgraded regexp system.
1997-02-19 23:23:48 +00:00
25aa2b845a
SIGIO and SIGPOLL have value 29, not 23.
1997-02-19 23:19:37 +00:00
a8e25a6827
Corrected spelling error in comment.
1997-02-19 23:15:27 +00:00
14559b2b4a
new portability problem
1996-12-10 21:49:16 +00:00
ae81b06f85
host-info bug
1996-12-02 22:18:25 +00:00
a620b7592a
new bug report
1996-11-18 16:48:20 +00:00
4ba01266de
Patches for BSDI dlsym stupidity
1996-11-17 18:49:56 +00:00
9f7d624413
updated bug list
1996-11-15 18:37:07 +00:00
ae7f18480d
updated bug list
1996-11-15 18:36:03 +00:00
e945b99adf
Replaced _POSIX_PATH_MAX with its value, 255, because HP-UX didn't
...
have this POSIX-required symbol. ???
1996-11-12 11:44:20 +00:00
faa08159d0
The cwd code nows starts out with the filename buffer initialised
...
to length _POSIX_PATH_MAX.
1996-11-11 20:56:52 +00:00
1cdcaaee53
Made FD/PORT FDES/PORT PORT/FD PORT/FDES parameter names consistently FD/PORT.
1996-11-11 19:38:18 +00:00
2f7c39be3f
-s<num> is now -sfd <num> -- more consistent with general switch design.
1996-11-11 19:21:25 +00:00
47c2eea2dd
Changed OPEN/NONBLOCKING to OPEN/NON-BLOCKING and
...
SET-RECORD-LOCK-NOBLOCK to SET-RECORD-LOCK-NO-BLOCK.
The former was always documented to be
OPEN/NON-BLOCKING
and the latter isn't exported in any event.
1996-11-11 18:52:19 +00:00
shivers