Commit Graph

356 Commits

Author SHA1 Message Date
tg 2e9654ffa4 introduce “typeset -i foo[*]”, rationale see cid 1004B5DA60A56BFB604 2010-01-25 14:25:16 +00:00
tg c700693d11 Add “unset foo[*]” mksh extension, which allows to unset the *contents*
of foo[0] (but not its attributes), and the rest of the array, so that
later “set +A foo bar” will set foo[0]=bar but retain the attributes.

This is important, because, in the future, arrays will have different
attributes per element, instead of all the same (which, actually, is
not entirely true right now either, since “unset foo[0]” will not mo-
dify the attributes of a foo[1] existing at that point in time), where
foo[$newkey] will inherit from foo[0], but typeset foo will only affect
foo[0] no longer foo[*] in the future. (The rules about typeset=local
will still apply, as they affect creation of variables in a scope.)
2010-01-25 14:11:29 +00:00
tg 833810a154 While we were supposed to use Ctrl-Arrowkey for wordwise jumping,
some idiotic terminal emulators and/or people seem to use the es-
cape codes normally denoting Alt-Arrowkey instead so let's simply
bind them to the vt_hack as well... (untested)
2010-01-08 22:21:06 +00:00
tg df4137c2e2 I don't care that AT&T ksh93 fails this nameref test case, but
it's important and we'll want it.
2010-01-04 21:08:00 +00:00
tg 0611ecad1f make [[ -o ‣-U ]] = [[ -o ‣+U ]] = [[ -o ‣utf8-mode ]]
where ‘‣’ is either nothing, ‘!’ or ‘?’ as usual.
2010-01-01 18:02:01 +00:00
tg 6115f5a91c • re-implement wcswidth-like behaviour for ${%foo} even in !UTFMODE
• merge the rest of branch tg-wcswidth-behaviour
• enhance test cases for wcswidth-like behaviour
• switch hash table collision resolution algorithm to Python’s as announced
• bump vsn
2010-01-01 17:44:10 +00:00
tg c413edb34a jobs.c: apply patch from deraadt@o:
│remember to restore errno (ie. stop someone from making a mistake later)
│ok guenther
check.t, sh.h: bump vsn

I wonder though why errno must be restored even if nothing was
called after reading it… moid?
2009-12-31 14:05:44 +00:00
tg 769e222586 re-vamp __attribute__ handling; let this pass on HP-UX bundled compiler
as well as HP aCC
2009-12-12 22:27:10 +00:00
tg 8213a28403 llvm/clang scan-build is now content… 2009-12-05 17:43:50 +00:00
tg 417fd60c76 * fix 'set -A foo -- [2]=a b c' contingency (tested against GNU bash4,
which, in its latest sid incarnation, even received mksh's ability
  to produce ${!foo[*]} array keys, wow!)
* plug a memory leak while here (ATEMP only, but still)
2009-12-01 19:15:35 +00:00
tg fa3acb939b fix 'mksh -U -i' and 'mksh +U -i' 2009-11-28 15:38:30 +00:00
tg a09f05e77a /me is annoyed
I read, IIRC in the Cederqvist, that 'cvs tag' sets a sticky tag onto
the cwd… it doesn’t, apparently. (I actually like it better this way,
but one needs to know!)
2009-11-28 14:28:03 +00:00
tg 883d9d99b3 switch ${%foo} to wcswidth-like behaviour – slightly problematic, and
the “set +U” case isn’t even handled

committed to branch because I’d like to get more input on this, for now
2009-11-28 14:21:47 +00:00
tg 4cff10296b bump vsn 2009-11-22 14:25:18 +00:00
tg e240e62a55 don't tell lies 2009-11-22 14:13:38 +00:00
tg 2094e3f25c since we have wcwidth anyway, expose it as ${%strvar} to the user, iff
utf8-mode is enabled (otherwise it'll be a synonym for ${#strvar} aka
the number of octets in it)
2009-11-21 23:23:20 +00:00
tg 398b5b7a60 fix that as well 2009-11-21 22:32:08 +00:00
tg 6e67920bef improve globbing:
• a=blöd; print ${a%???} → now printf 'b' not 'bl' in UTF-8 mode
• a=hä; print ${a%?} still broken
2009-11-21 22:30:36 +00:00
tg d02e6be6cf bugfix: ${x#?} should expand to nothing not '' if x was unset 2009-11-21 22:29:12 +00:00
tg fe853065bb make print_columns() aware that octets≠characters≠columns 2009-11-09 23:35:11 +00:00
tg 6fcacf577d make tab completing filenames with ':' '=' '$' '`' work as well as
others (colon and equals sign need to be simply escaped, while dollar
sign and accent gravis need double escaping like opening square brak-
ket did back then); add = to C_QUOTE to simplify (doesn't break any-
thing) and sort these strings asciibetically while here
2009-10-30 00:57:39 +00:00
tg dc98ec23a6 do not use PATH_MAX on GNU/Hurd (even if it *was* defined), but use
some glibc-only functions that don’t require its use instead

tested on gnubber, where (admittedly) sysconf(_PC_PATH_MAX) == 1024…
2009-10-27 17:00:02 +00:00
tg ceb591bf6a • better IV for the oaathash-on-binary (as if we prepended a '\x01')
• better variable name for the LCG state
2009-10-18 12:30:06 +00:00
tg eafe88aa74 • overhaul handling of special variables
• use a combination of the one-at-a-time hash and an LCG for handling
  the $RANDOM special if !HAVE_ARC4RANDOM instead of rand(3)/srand(3)
  and get rid of time(3) usage to reduce import footprint
• raise entropy state (mostly in the !HAVE_ARC4RANDOM case though…)
• simplify handling of the $RANDOM_SPECIAL generally
• tweak hash() to save a temp var for non-optimising compilers
• some int → mksh_ari_t and other type fixes
• general tweaking of code and comments
2009-10-17 21:16:05 +00:00
tg 578d4e6314 fix exit-eval-1 (Debian #550717), sync SUSv4 wording, improve comments 2009-10-15 16:36:27 +00:00
tg c7a8b367f8 remove regression-44 in favour of utilities-getopts-1 (the former checked
the AT&T ksh88 behaviour, the latter the AT&T ksh93 behaviour, which dif-
fer from each other, on the very same thing with the exact same test)
2009-10-15 16:32:23 +00:00
tg 0d1ef736c7 modify the oksh-varfunction regression test to reflect that, indeed, a
change of a variable inside a Bourne style POSIX function will affect
the current execution environment of the function caller (to be consi-
stent with exec-function-environment-1)
2009-10-15 16:24:49 +00:00
tg 7b8f207cc9 new exec-function-environment-1 replacing exec-simple-1 from posh, after
reading SUSv4 throughoutly
cf. Message-ID: <Pine.BSM.4.64L.0910151542460.4645@herc.mirbsd.org>
2009-10-15 16:15:03 +00:00
tg ce7359ffc5 posh’s utilities-echo-2 is actually incorrect; reason here
cf. Message-ID: <Pine.BSM.4.64L.0910151525270.4645@herc.mirbsd.org>
2009-10-15 15:32:34 +00:00
tg a9ca7a04c8 add two regression tests from Debian posh, adjusted to pass with ksh93 2009-10-15 15:23:41 +00:00
tg 25516ac756 another exit code in substitution questionable thing
split this one by set ±o sh
2009-10-15 14:58:51 +00:00
tg 84f8d3191a make some of these tests more posh-friendly 2009-10-15 14:12:46 +00:00
tg 620b4446d7 make exit-eval-1 an expected-fail for now, until we can fix this 2009-10-15 12:50:11 +00:00
tg b694bde402 rename regression-56 into exit-eval-1 and change expected output
from what AT&T ksh88 does to what AT&T ksh93 does and POSuX expects

Reported by Clint Adams as Debian #550717, to be fixed...
2009-10-14 18:06:57 +00:00
tg 888c7befea group exit-related tests 2009-10-14 18:04:52 +00:00
tg 549888a183 while testing haserl-0.9.26 on MirBSD, I discovered that the echo builtin
in FSH mode did in fact, contrary to POSIX and Debian Policy 10.4 behavi-
our (I think), interpret escape sequences; fix and add testsuite for echo
2009-10-10 21:17:31 +00:00
tg b4a32ff5f7 cygwin seems to force a+x for files beginning with ':\n' so prepend
a nop newline to make mksh.exe pass this test
2009-10-10 17:39:48 +00:00
tg 73147a7fe9 fix lazy evaluation of assignments in ternary ops 2009-10-04 13:19:33 +00:00
tg 5f58261287 fix long-standing parenthesēs problem: ((foo) || bar) is now parsed corr. 2009-10-04 12:45:23 +00:00
tg c2e5737fca fix dollar quotes in here strings and here documents (regression) 2009-10-04 12:44:19 +00:00
tg 40d11a2066 on the other hand, this is a regression in mksh, apparently 2009-10-04 03:13:06 +00:00
tg 33b9f8b666 * rename "set -o posix" (FPOSIX) to "set -o sh" (FSH) because it's not
just a "somewhat more POSIX" but also a "/bin/sh legacy kludge" mode
* consistently capitalise POSIX and SUSv3/SUSv4 (same as AT&T ksh) and
  Bourne shell
2009-10-02 18:08:37 +00:00
tg 2ea131ab1f make another Debian posh bug into a regression test (expected-fail) 2009-10-02 17:05:01 +00:00
tg c0589fd329 fix typos, From: Clint Adams <schizo@debian.org> 2009-10-02 16:59:53 +00:00
tg ff8b75f46f SUSv4 says that there be exit 127; the old code tried but failed, so it
was an actual programmatical bug…
2009-09-29 12:28:13 +00:00
tg a59d14b565 further optimise and simplify the handling of $RANDOM, reads and writes
to it are now either arc4random or rand/srand, but srand retains the old
state; set +o arc4random is no longer possible, but if it's there we use
arc4random(3), if not, we use rand(3) for $RANDOM reads; optimise special
variable handling too and fix a few consts and other minor things
2009-09-26 03:40:03 +00:00
tg 1dfb13b603 change undef/def MKSH_NOVI into 0/1 MKSH_S_NOVI flag (with more to come:
MKSH_S_EDIT for small (Emacs) editing mode, MKSH_S_FEAT for all the dis-
abled language features), which can be set to 0 despite MKSH_SMALL being
defined to re-enable the Vi command line editing mode (which I wouldn't,
but fits into the general mastermind scheme)
2009-09-24 17:15:33 +00:00
tg 1a28786229 * shrink MKSH_SMALL even further by removing functionality like
some GNU bash extensions (suggested by cnuke@) and bind macros
* make the random cache more efficient (and the code potentially
  smaller, although we have a new implementation of the oaat hash
  function, alongside the old one, now) and pushb only if needed
  (i.e. state has changed or user has set $RANDOM, but not onfork)
2009-09-23 18:04:58 +00:00
tg 1ea1096a4e Add a hack input function for VT100-style key sequences; support
Ctrl-CurLeft and Ctrl-CurRight (not quite ANSI, but fits the scheme)
2009-09-20 17:23:52 +00:00
tg 9fd4b9db41 $'…' functionality, documentation improvements, fixes for backslash
expansion in all modes, regression tests for both kinds of backslash
expansion; unbksl() revamp; make CTRL macro available globally
2009-09-19 21:54:46 +00:00