Commit Graph

1967 Commits

Author SHA1 Message Date
tg
84096a0a86 Message-ID: <CAOw7k5gBmt+WqzQxgK8xrjMvCh3iBHcQ7cU78cy1WbzRf8d_Hg@mail.gmail.com>
just to make sure, got a copyright licence for this
2011-09-13 18:46:46 +00:00
tg
9782f6b4d1 • access(2) is broken in at least kFreeBSD 9.0 as “modern” OS, so bring
back the wrapper code as well as refactor most other code calling it
• apparently, names can’t end in ‘_’ or contain ‘__’ anywhere…
2011-09-07 15:24:22 +00:00
tg
577c918beb patch most of Jerker Bäck’s concerns out, unless not applicable 2011-08-27 18:06:52 +00:00
tg
5d9c88ceeb jg71 reported -DMKSH_ASSUME_UTF8=* breaks defining stristr 2011-08-27 17:30:07 +00:00
tg
da80c56acb 12:23⎜<cnuke:#bosng> angefangen damit, dass es kein uname unter OPENSTEP gibt :) 2011-08-26 10:24:51 +00:00
tg
6f76d7f2fe fix… pasto, I think, from some time ago, spotted by cnuke@
12:17⎜<cnuke:#bosng> OPENSTEP hat kein dirname und $srcdir/sh.h wird dann /sh.h
2011-08-26 10:20:28 +00:00
tg
308290e2b6 dict.leo.org says this is correct 2011-08-13 22:19:41 +00:00
tg
80223417d7 followup fix for 1004D90467358D6B13C regarding escaped substitutions
found in, out of all things, Debian/m68k, by Wouter Verhelst (Yoe)
fix hacked at DebConf 11, Бања Лука, Република Српска, Босна и Херцеговина
2011-07-26 16:57:28 +00:00
tg
a7566387cf optimise an if away, and possibly even the function bodies… 2011-07-20 23:47:29 +00:00
tg
6ce68e906f now that I got Hash.cs compiled and run, I’m amazed; make NZAAT (and NZAT!) all green with a small tweak; bit distribution and χ2 look OK as well 2011-07-18 00:35:46 +00:00
tg
f51883e350 eek, only be a ksh_nice if we HAVE_NICE 2011-07-16 23:37:58 +00:00
tg
ac814611f1 somebody (who is probably inlining) complains about memset with the
third argument being 0; maybe (I think can only be this) fix that
2011-07-16 18:15:45 +00:00
tg
1b9bee46fc check later whether the final link may succeed, and include "sh.h" → catch inlining bugs in system headers fighting with -c combine 2011-07-16 18:03:04 +00:00
tg
0168f6dc48 some distributions declare nice(3) with warn-unused-result; annoy mksh users running into this so they complain to their distros 2011-07-16 17:41:50 +00:00
tg
09db3d67a2 finish merging from R40-stable; set version number to yesterday though 2011-07-16 17:08:52 +00:00
tg
620faa3dc9 ‣ merge from mksh R40-stable
drop heredoc-weird-3 while here, it’s not deterministic by design(?)
2011-07-16 17:08:35 +00:00
tg
e03b485d75 ‣ merge from mksh R40-stable
move /etc/{,suid_}profile to /system/etc/ for Android (but do not make
the location of /etc configurable); rewrite manpage section about
/etc/{,suid_}profile, .profile, .mkshrc
2011-07-16 17:08:21 +00:00
tg
b0ee967d00 ‣ merge from mksh R40-stable
first part of long-overdue manpage improvements
2011-07-16 17:08:04 +00:00
tg
d6445aa007 ‣ merge from mksh R40-stable
document the unportability of the echo utility;
arekm's non-bug report inspired this
2011-07-16 17:07:48 +00:00
tg
8d1edbd95e ‣ merge from mksh R40-stable
use common macro bodies shf_{get,put}c_ for definition
of shf_{put,get}c as functions (MKSH_SMALL) or macros, respectively
2011-07-16 17:07:35 +00:00
tg
851f8fb7d8 introduce MKSH_NO_EXTERNAL_CAT skipping the “call /bin/cat if we have an
option” nonsense, intended for Android, Plan 9, and other systems which
honour that cat -v is deemed harmful
2011-07-07 21:24:53 +00:00
tg
3f014c8daf utf8opt-3 needs to be split in half, part of it needs-ctty
10x arekm again
2011-07-07 21:02:39 +00:00
tg
a4a000eafb take down the entire xterm (or GNU screen tab), hard, with:
mksh-R40 -c 'x=$COLUMNS; awk "BEGIN{exit(1)}"'
10x to the PLD Linux guys for spotting _and_ bisecting this!
2011-07-07 20:24:53 +00:00
tg
3ef67e3296 replace uses of OAAT hash in all its variants by NZAT (mksh internal) or NZAAT (all others) 2011-07-06 22:22:02 +00:00
tg
fe2c0221b3 replace typeset with local, as style(9) now mandates
(keep globals for a while, until it’s widespread, though)

since we don’t export PS1 any longer, no need to stay AT&T ksh compatible
2011-07-05 20:19:18 +00:00
tg
f2405b7dde • bump patchlevel to today’s
• while here, reformat 'struct tbl' comment-wise and placement-wise
  and drop the Tflag typedef
• while here, write regression test for the "global" built-in, which
  does what typeset is supposed to do except that it doubles as "local"
2011-07-05 20:12:20 +00:00
tg
71fa2ca232 allow ^Cing the built-in cat, spotted by Johannes (jal) in IRC 2011-07-05 19:56:24 +00:00
tg
02b76603f8 new function for dot.mkshrc:
23:19⎜<laffer1:#midnightbsd> oddly enough the only cshism i like is setting environment variables at this point.
2011-07-03 23:26:47 +00:00
tg
0739fa74a0 address "env RANDOM=1=2=3 mksh" DoS by integrifying more 2011-07-02 17:57:41 +00:00
tg
64aade1a52 add new experimental "global" command 2011-07-02 17:57:24 +00:00
tg
0785c2d71c fix tab completion of "~/M↹" 2011-07-02 17:57:12 +00:00
tg
b89c144344 fix vstrchr like use of cstrchr 2011-07-02 17:57:02 +00:00
tg
257021e510 fix, even removing code, good… 2011-06-30 14:09:04 +00:00
tg
edaab2cafe tabcomplete ~ and ~foo like $FOO (idea by yofuh) 2011-06-30 13:48:13 +00:00
tg
535649cc60 sync clog 2011-06-21 22:02:09 +00:00
tg
43c8d8765f pretty tricky thing, IBM’s curses library does nameref foo=foo… 2011-06-21 21:50:26 +00:00
tg
c57f8ede17 make set_refflag an enum to get easier to understand code 2011-06-21 21:11:21 +00:00
tg
fab3be3584 re-indent comments 2011-06-21 21:10:12 +00:00
tg
f4e22bc4c8 factor out some common code 2011-06-21 21:08:50 +00:00
tg
6524812f6b drop more long-deprecated things 2011-06-12 15:54:42 +00:00
tg
6807ce3100 drop long-deprecated 'set ±o arc4random' 2011-06-12 15:37:10 +00:00
tg
463dc26db4 drop the RC 2011-06-12 15:34:19 +00:00
tg
7ce3a2ad54 HP-UX aCC suggested more “≠ 0” and parenthesēs 2011-06-12 14:58:45 +00:00
tg
c133d423df ACK found a missing "static", gcc on Minix found a missing cast; add more type size checks 2011-06-12 14:45:34 +00:00
tg
3589604dea Minix 3 has no _setrlimit (XXX mirtoconf that in R41) 2011-06-12 14:37:00 +00:00
tg
6571d66386 add missing shf_flush() before prompting 2011-06-11 03:14:50 +00:00
tg
5aa22cff70 • fix empty input
• fix input without trailing newline for integer-base-one-3{A,W}r
2011-06-09 21:53:12 +00:00
tg
1fbd3e5ee0 bugfix for when the last line was full 2011-06-09 21:26:21 +00:00
tg
60ccad3dfc check for embedded NULs work 2011-06-09 21:10:50 +00:00
tg
e0fb8dc431 • rework hash table interna to avoid gcc-4.1 on Debian etch bug
• also improve behaviour with _a lot_ (>2²⁸) entries
• while here, improve comments and whitespace
2011-06-05 19:58:21 +00:00