493ad928e3
if MKSH_SMALL, disable persistent history code as well (like on Solaris,
...
where it can't work anyway), and fix that code to shrink
saves 2071t 20d 5i
2006-11-10 05:23:14 +00:00
7c4f8d7e16
oops
2006-11-10 05:21:38 +00:00
d5e4c261db
simplify. and, a few byte compares are cheaper than a structure
...
copy and a structure compare
gain: 103t 4d 1i
loss: had to move some code even though no diff (cvs issue)
2006-11-10 04:31:04 +00:00
208c45da25
more cleanup; incidentally, replacing libc function with syscall maybe speedup
...
9t 4d 1i
2006-11-10 04:22:13 +00:00
5dff5df5e5
get rid of the 'mksh -T' functionality (chvt) if MKSH_SMALL
...
848 text 12 data 3 imports
2006-11-10 04:07:59 +00:00
a1ff719ba4
thanks to the nice property of my reimplementation of getn to also
...
return a value in case of an error (0 or the partial result, which
is the full result in case of trailing junk even), using it to rid
atoi() is possible, saving 9t 4d 1i
2006-11-10 04:03:59 +00:00
d91d7feff0
exit not _exit, 50 text 4 data 1 import
2006-11-10 03:54:38 +00:00
ecc08e9681
use chown not fchown, and move the ownership check to earlier
...
saves 45 text, 4 data and security peoples' nerves
2006-11-10 03:50:05 +00:00
d10dd31b9a
correct a few protos and comments
...
saves 4 .text bytes and my nerves
2006-11-10 03:45:57 +00:00
f8e7fdbb71
use qsort(3) instead of rolling our own
...
saves 284 in .text, no added import since we already use qsort(3) once
2006-11-10 03:23:50 +00:00
f103c6b2a0
use errorf not bi_errorf if we want to abort
...
saves 16 .text ;)
2006-11-10 02:10:45 +00:00
658dc2b407
rewrite getn() which is some kind of atoi with return code
...
costs 1 .text, saves 4 .data, 1 libc import
2006-11-10 02:05:06 +00:00
242d2c086d
overhaul parse_args - don't use strlcpy and strlen when overkill
...
also fixes off-by-one
saves 48 .text
2006-11-10 01:52:23 +00:00
c60dbdc6cb
* change EXECSHELL to /bin/sh (we can override it via environment anyway,
...
and this is a sensible choice instead of ourselves)
* move this stuff from sh.h into exec.c where it belongs
* simplify set -o stuff
saves 8 bytes
2006-11-10 01:44:40 +00:00
cd9202835c
cosmetics
2006-11-10 01:25:22 +00:00
f1e29cb1a1
use NELEM
2006-11-10 01:24:38 +00:00
ebf58f6b42
if !MKSH_SMALL, str_save() can be a macro
2006-11-10 01:19:17 +00:00
273ca89019
* check.t: add new regression test "typeset-padding-1" according to TFM
...
* edit.c: remove debug stuff again; next time better use shl.c functions ;)
* sh.h: add format attributes to a few shf functions
* histrap.c, var.c: fix format string mistakes
* main.c, sh.h: error_prefix and warningf take bool not int
* misc.c: make chvt() stuff use shf_* functions
* misc.c: rewrite the TIOCSTTY stuff to be better integrated in mksh,
since it originally was an external patch
* misc.c: chvt() no longer fails if e.g. chown fails due to e.g. R/O / fs
* var.c: fix typeset padding for right-justified zero-filled
2006-11-10 01:13:52 +00:00
c2aec39358
save 176 text (92 text 32 bss relative to before last commit)
...
by making correct use of shf_* functions
2006-11-10 00:09:27 +00:00
320f503391
get rid of ulton() - a joke
...
saves 32 bss, but adds 84 text oO
2006-11-09 23:55:52 +00:00
88bf6422d9
shrink chtypes array down to 8 bit
...
saves 72 text and a lot bss
2006-11-09 23:39:16 +00:00
6334091d86
omit usage msg if small
2006-11-09 23:19:52 +00:00
625928f9f1
make autoconf checks more verbose as well
2006-11-09 23:09:05 +00:00
e49b50f8f1
check for a few more things
2006-11-09 23:04:34 +00:00
7148a691e0
simplify
2006-11-09 22:56:55 +00:00
ee55f29a31
new function ac_testn() does the same as ac_test() but doesn't amend CPPFLAGS
2006-11-09 22:56:09 +00:00
2bdcd71ef0
unbreak -d (was broken in R28)
2006-11-09 22:53:21 +00:00
a48a880c00
ac_test for MKSH_SMALL presence in CPPFLAGS
...
if present, don't even look for setmode as it's only used by mknod builtin
2006-11-09 22:51:49 +00:00
68caa9afdd
save 464 text with restructuring
2006-11-09 22:38:31 +00:00
3c33cbcecc
don't compile mknod(1) builtin if MKSH_SMALL
...
saves 1037 text, 20 data (on i386)
2006-11-09 22:18:10 +00:00
2345d8b72b
more getconf/confstr fixes
2006-11-09 22:11:08 +00:00
d4ca2c9141
simplify getconf and confstr stuff
2006-11-09 22:08:08 +00:00
bc215df211
make a few functions static that should be, and optimise a little
2006-11-09 21:20:49 +00:00
c43e701025
optional (#ifdef MKSH_SMALL) saving of 364 more bytes of .text
...
intended e.g. for FreeWRT
2006-11-09 21:00:13 +00:00
35776a9d4d
using a function instead of a macro saves 896 bytes of .text
2006-11-09 20:53:42 +00:00
617950554f
missed these
2006-11-09 15:03:56 +00:00
36ac8dc0f7
get rid of the need of strlcat() altogether (only one use was left)
2006-11-09 15:02:31 +00:00
2b375abef1
fix SIGSEGV, we are passed NULL sometimes
2006-11-09 14:58:27 +00:00
a1bfc17aa0
optimise and, ahem… plug…
2006-11-09 14:19:31 +00:00
39057dd592
add the new function utf_ptradj() which does the same as utf_widthadj()
...
except it doesn't return a value and is much cheaper (no internal con-
version to UCS-2, just range checking on the multibytes)
2006-11-09 00:39:27 +00:00
fb9c6e30bb
fix another warning
...
now builds on MirOS, DragonFly, NetBSD, Debian stale (i386), Solaris 8,
Debian testing (s390x) - Solaris /bin/sh, bash, ksh93, mksh
2006-11-09 00:28:36 +00:00
1ffa722d73
optic sugar
2006-11-09 00:13:27 +00:00
0307ecb492
* instead of including <sys/cdefs.h> (not in Solaris), we hope <sys/types.h>
...
includes it for us
* autoconf-test for <sys/param.h> (not in SUSv3) existence
2006-11-09 00:11:39 +00:00
9b0dbaddbf
oops, missed syntactic change
2006-11-09 00:08:25 +00:00
1c7e58a9d9
use \u0060...\u0060 instead of $(...)
...
demanded by solaris /bin/sh
2006-11-09 00:06:33 +00:00
84088b91d0
there is no compat.c any more
2006-11-09 00:03:37 +00:00
dc9f457f24
don't use __predict_false() here either
2006-11-09 00:01:36 +00:00
e97ddf1eef
some build issues
2006-11-09 00:01:04 +00:00
d8de18df87
can use short form here
2006-11-08 23:50:47 +00:00
1edc9c071a
fix description of setlocale stuff a little
2006-11-08 23:49:01 +00:00
tg