Commit Graph

3233 Commits

Author SHA1 Message Date
Giacomo Tesio 4e2ce8120c jehanne: purge syscall macros: s/rfork/sys_rfork/ 2019-11-28 00:44:31 +01:00
tg dddf3694e9 add O_MAYEXEC support for CLIP OS, zero cost otherwise
cf. https://lwn.net/Articles/768819/
2018-10-30 17:10:16 +00:00
tg 8e4829bc5f move a largish block of code around to trick a GCC warning
(the fallthrough detection is thrown off by the other comments,
and it doesn’t recognise NOTREACHED anyway…)
2018-10-20 21:04:28 +00:00
tg f95e05d511 catch up a little on test cases 2018-10-20 18:48:26 +00:00
tg 886aee865d fixup high-bit7 chars to be dumped (XXX probably want M- there) 2018-10-20 18:46:00 +00:00
tg f78e3f351e fix #1779179 (another case of signed integers sucking) 2018-10-20 18:34:14 +00:00
tg b55607d825 spelling cleanup: “programme” is distinct from “program”; even in
British English, the latter is used for computer programs, while
the former serves for things like the TV programme, or a programme
to plant trees
2018-10-07 01:10:11 +00:00
tg 58b88af62c do not repeat 0m0.00s
via https://bugzilla.redhat.com/show_bug.cgi?id=1612173
2018-08-17 20:33:08 +00:00
tg 6cea111ef1 same as in commitid 1005B6CF36E3932B560, plus assorted cleanup 2018-08-10 02:53:39 +00:00
tg d5ce724471 add ctty tests 2018-07-15 17:22:15 +00:00
tg 2e1150c8e8 against better judgement, allow unsetting COLUMNS:
• GNU coreutils’ test suite insists on it, even despite it can run
  successfully without doing it (WTF‽)…
• we already unspecial, explicitly commented as “hands-off” it,
  when imported from the environment
• it was already changeable (though that didn’t unspecial it on the
  command line)
2018-07-15 17:21:24 +00:00
tg 28a08dc7a0 fix an error message 2018-07-15 16:23:10 +00:00
tg 008f09dd09 fix a vi mode segfault when using ^W during a search; from multiplexd 2018-07-15 16:16:38 +00:00
tg 5dab09d6a8 add EU-DSGVO (GDPR) notice for the mailing list 2018-07-15 13:57:18 +00:00
tg 3d08263d69 Unicode 11.0.0 2018-06-26 21:22:24 +00:00
tg 28637e55fd fix using “-m” on the command line; spotted by G.raud <graud@gmx.com> on ML 2018-05-08 17:37:37 +00:00
tg 3560784571 fix typeset -L vs. Unicode: snprintf second arg was wrong 2018-05-07 00:07:23 +00:00
tg 2d0c795605 Coverity Scan pointed out that lseek can return -1 2018-04-28 17:16:54 +00:00
tg 09c8df8058 fix allexport for arrays; add testcase that also catches exporting arrays 2018-04-28 07:07:39 +00:00
tg 1c24f0dbc7 contrary to what I said in revision 1.453 we evidently d̲o̲ export arrays…
(XXX there’s a comment in var.c:typeset() still to the contrary, with assumptions…)
2018-04-28 07:03:17 +00:00
tg e42dc7afcb move subsection on tty below job control one, pull all stuff into it 2018-04-27 17:20:39 +00:00
tg 7a794d4dee document some more on how the terminal state is affected
From: G.raud Meyer <graud@gmx.com> (mostly)
2018-04-27 17:10:39 +00:00
tg 8d2c435f0f only ever make it possible to enter edit.c if tty_hasstate 2018-04-27 16:59:50 +00:00
tg 6818f7a136 minor spelling in comment 2018-04-27 16:54:45 +00:00
tg d10c196cf1 improve exporting documentation:
• reference “shellshock” of GNU bash for exporting functions
• mention we don’t export arrays (just their "0" members) as
  ksh93 and GNU bash don’t do it either, the latter documen‐
  ting it even
2018-04-16 00:32:35 +00:00
tg 99f6aae983 move text on ${array} => ${array[0]} implicitly, it is more global 2018-04-16 00:22:28 +00:00
tg 0096aeeed4 quote empty strings for re-entry into shell
From: Martijn Dekker <martijn@inlv.org>
2018-03-17 22:46:09 +00:00
tg 4abbaa0400 make ${foo#'bar'} in here document behave like ksh93
reported by Martijn Dekker <martijn@inlv.org>
2018-03-09 01:29:11 +00:00
tg 4df1d3abe0 Macintosh-specific updates from gecko2@
also, their sysctl(8) sucks:
13:31⎜«gecko» naja, sysctl bricht die ausgabe bei nicht vorhandenen parametern ab
13:32⎜«gecko» der listet aktuell nur bis hw.memsize auf
2018-01-25 12:33:54 +00:00
tg 1474437b57 bump; add testcase provided by @mvdan on github 2018-01-14 01:47:36 +00:00
tg 5c920fe4db clean up 2018-01-14 01:44:02 +00:00
tg cc59adb1f5 write ord’ed into c 2018-01-14 01:29:47 +00:00
tg f7b0277504 shf_getc{,har} need to return properly casted 2018-01-14 01:28:16 +00:00
tg 7977acbc72 this callback also needs to cast cleanly 2018-01-14 01:26:49 +00:00
tg 141dfaf819 move the cast earlier; the callback also needs it 2018-01-14 01:25:28 +00:00
tg d7b77a76df I think I’m onto something here… getsc_{uu,bn} never return -1 AFAICT 2018-01-14 00:57:55 +00:00
tg 7716e825ab I think the ord thing was a red herring 2018-01-14 00:51:35 +00:00
tg 9691e330b9 after literally HOURS of debugging, I got a working-ish expression… 2018-01-14 00:47:11 +00:00
tg 609f5c2256 plug the first couple of ctype bugs 2018-01-14 00:22:30 +00:00
tg 9e4868fb06 oops, casting negative char to size_t is… not nice 2018-01-14 00:09:34 +00:00
tg c763e7b0e9 revert the CORD debugging aid
(introduced so that ORD could eval its args twice in !DEBUG)
2018-01-14 00:03:05 +00:00
tg 6e486d1163 DOSPATH fix 2018-01-13 23:58:32 +00:00
tg dde8500a24 some ord() debugging; CORD for case, ORD evaluates twice, for debugging 2018-01-13 23:55:15 +00:00
tg a0b1179db3 warn when entering POSuX noncompliance 2018-01-13 21:45:07 +00:00
tg aa530d4343 implement early (mediæval) locale tracking, as a compile-time option,
for SuSE; slightly inspired by the original patch submitted by
From: Dr. Werner Fink <werner@suse.de>
2018-01-13 21:38:10 +00:00
tg 9f3c0efefc small cleanup for systems without core dumps, such as Jehanne 2018-01-05 20:08:34 +00:00
tg 82265de5d7 merge Jehanne cleanup patch
From: Giacomo Tesio <giacomo@tesio.it>
2018-01-05 20:05:26 +00:00
tg 3d522cfa27 turns out that Plan 9 does *not* have symlink(7)s
cf. https://github.com/brho/plan9/blob/master/sys/src/ape/lib/ap/plan9/dirtostat.c
despite https://github.com/brho/plan9/blob/master/sys/include/ape/sys/stat.h

see http://static.usenix.org/events/usenix2000/general/full_papers/pikelex/pikelex.pdf though…
2017-12-22 18:36:21 +00:00
tg 942035921c Pass arguments via a resonse file if executing a child fails
Command line length limit of OS/2 is 32KiB. If the total length of
all arguments is larger than this limit, it's needed to use a
response file.

Previously, the total length calculation was simply to add length of
all arguments. However, this result was not match with real length of
arguments, which are passed to child processes with OS/2 APIs.

Because conversion methods of arguments from libc to OS/2 APIs are
different depending on libc.

For example, kLIBC inserts its signature to an argument list. In
addition, it passes arguments with a leading space like:

    arg0
     kLIBC signature
     arg1
     arg2
     ...

Whereas, EMX just distinguishes arg0 and others like:

    arg0
    arg1 arg2 arg3 ...

After all, simple sum of a length of arguments are not correct.

The better way is to try to execute a child process, and to retry with
a response file if it fails due to arguments-too-long.

This has been found while doing 'bootstrap', especially 'autoreconf'
in coreutils git repo. It stops with:

    autom4te: /usr/local/bin/m4: Invalid argument

From: KO Myung-Hun <komh@chollian.net>
2017-12-22 16:41:42 +00:00
tg d93bd77f28 Set stdin to text mode before executing child processes
Normal OS/2 programs expect that standard IOs, especially stdin,
are opened in text mode at the startup. By the way, on OS/2 kLIBC
child processes inherit a translation mode of a parent process.
As a result, if stdin is set to binary mode in a parent process,
stdin of child processes is opened in binary mode as well at the
startup. In this case, some programs such as sed suffer from CR.

This is the regression fix of commit 20dbf6.

From: KO Myung-Hun <komh@chollian.net>
2017-12-22 16:37:14 +00:00