the width for control characters (wcwidth(wc) == -1) was hard-coded
to 2 (!) in utf_widthadj, which is true for *only* one of the two x_zotc*
functions in Emacs editing mode, and none of the other functions which
use this piece of code
change to 1, to be more correct in the general case; use of the UTF C1
control characters U+0080‥U+009F is slightly broken anyway, and this
only shifts the brokenness to different places of code
XXX maybe we want to map U+0080‥U+009F into Unicode as if they were
XXX 0x80‥0x9F in ANSI cp1252 instead, at least for displaying?
the editing code is cruel…
was hard to type and hard to fix, galloc is also hard to fix, and some
things I learned will probably improve things more but make me use the
original form as base (especially for space savings)
* let sizeofN die though, remove even more casts
* optimise, polish
* regen Makefiles
* sprinkle a few /* CONSTCOND */ while here
please pcc, prompted for by Anders “ragge” Magnusson, problem spotted
originally by Adam “replaced” Hoka
⇒ rewrote x_bs2() and utf_backch() into a combined x_bs3() function,
since these are never used in any other way
• whitespace cleanup, while here
abortion (^G – ^C is SIGINT and doesn’t work like this, but
that’s actually good IMO)
prompted by enquiry about the Emacs editing mode by <smultron:#MidnightBSD>
in a somewhat hackish way, and it’s still quite different from zsh,
but probably closer to a desired functionality
XXX this makes state by abusing 「modified」 and 「xmp」 (“the mark”).
in præfix sequences (like ANSI cursor keys), leading to annoying effects
if we forget that
this patch changes the behaviour so that another character is read/peeked
at (since this is done in the main loop after ESC anyway, no function loss
through the delay) if ESC leads in a prefix-1 sequence, and if the peeked
character leads in a prefix-1 or prefix-2 sequence when in state prefix-1,
it’s still enacted (XXX document this in manpage)
empty) appears pushed into the history, so that, when pressing cursor-up
or ^P, with a cursor-down or ^N you get it back, unless you modify a line
retrieved from the history, in which case it will overwrite the saved line
and place the current history pointer past the entered history lines
This is for Emacs mode; Vi mode had something similar already, and shares
some code and data
XXX there are several static buffers of size LINE (currently 4096) in here
use the libc functions for converting between multibyte strings and wide
strings in here any more, besides mksh has slightly different needs than
SUSv3 compliance ⇒ hand-craft optimised and unrolled functions for that
• sync the mksh-internal wcwidth function with libc
• fix vi mode (which, however, is officially orphaned) multi-line $PS1 by
using a similar algorithm for prompt skipping as emacs mode (changing
the meaning of prompt_trunc variable and using prompt_redraw, just even
more efficiently than vi mode); reported by asarch via IRC
• fix multi-line prompts if last line is “too large” by using emacs mode
algorithm of just internally appending a newline, while here ☺ this even
saves us having to re-add the prompt_skip variable…
WARNING: this is only barely tested, as almost nobody ever uses vi mode
⇒ test yourself, there may be bugs (e.g. off-by-ones); already known is
that the vi input line editing mode is NOT multibyte safe…
‣ macro afreechk() is superfluous
• get rid of macro afreechv() by re-doing the “don’t leak that much” code
• some KNF (mostly, whitespace and 80c) while here
• more int → bool
• more regression tests: check if the utf8-hack flag is really disabled
at non-interactive startup, enabled at interactive startup, if the
current locale is a UTF-8 one
• make the mksh-local multibyte handling functions globally accessible,
change their names, syntax and semantics a little (XXX more work needed)
• optimise
• utf_wctomb: src → dst, as we’re writing to that char array (pasto?)
• edit.c:x_e_getmbc(): if the second byte of a 2- or 3-byte multibyte
sequence is invalid utf-8, ungetc it (not possible for the 3rd byte yet)
• edit.c:x_zotc3(): easier (and faster) handling of UTF-8
• implement, document and test for base-1 numbers: they just get the
ASCII (8-bit) or Unicode (UTF-8) value of the octet(s) after the ‘1#’,
or do the same as print \x## or \u#### (depending on the utf8-hack flag),
plus support the PUA assignment of EF80‥EFFF for the MirBSD encoding “hack”
(print doesn’t, as it has \x## and \u#### to distinguish, but we cannot use
base-0 numbers which I had planned to use for raw octets first, as they are
used internally): http://thread.gmane.org/gmane.os.miros.general/7938
• as an application example, add a hexdumper to the regression tests ☺
this change broke abortion on failure to read input, was not
needed for gcc warnings and is the fault of Intel’s compiler
this should fix the other busy-loop problem occuring only on
GNU/Linux so far – 10x spaetzle@freewrt.o for pointing me to
the problem; reproduced on my work craptop
from netbsd via oksh
we had the NULL pointer deref already fixed
• avoid a bogus not-setting the return value of edit.c:x_file_glob()
introduced by the above change in oksh
• escape ? as well (but not ] because that’s wrong)
reminded by cbiere@netbsd via oksh
• Unsetting a non-existent variable is not an error. See
http://www.opengroup.org/onlinepubs/009695399/utilities/unset.html
report from Arkadiusz Miskiewicz; fixed based on
http://cvs.pld-linux.org diff via oksh but modified slightly
• Be more smart waiting for input for non-interactive scripts. Fix
based on a diff from debian: see their bug#296446 (via oksh)
modified slightly
this also fixes cnuke@’s “mksh busy loop” problem, for which I never
received a bug report, but the Debian bug page contains a set of two
scripts to reproduce this before (and no longer after) this commit
• some KNF
• bump version
so that archite@midnightbsd won’t have to add evil kludges to oksh again if
they switch their ksh to mksh ☺
both “clear-screen” and “error” aren’t bound; default binding for ^L stays,
as usual, “redraw” (principle of least surprise); however GNU bash converts
also might want to put “bind ^L=clear-screen” into their ~/.mkshrc.
problems. icc's warnings are bogus:
• it says int x_getc(void); is invalid (hm well, it may be static?)
• char c, d; d = c | 0x40; -> warning because (c | 0x40) is an int
(it apparently can't track value bounds)
(by dramsey again, you're DA MAN) by replicating some of the x_redraw() logic
Note that this is correct, a construct like the full-fledged
| x_e_putc2((xep > xlp) ? ((xbp > xbuf) ? '*' : '>') : (xbp > xbuf) ? '<' : ' ');
is not needed since if (xep > xlp) && (xbp > xbuf) – i.e. in the '*' case –
x_redraw() will be called anyway and because (xx_cols - 2 - x_col) == 0 the
code won't be triggered.
cf. Message-ID: <e3fded850705211623n20d2c695ke7b41d75ac439a6c@mail.gmail.com>
this one was harder to track down, additional variables coming into the play…
cf. Message-ID: <Pine.BSM.4.64L.0705211156060.16459@odem.66h.42h.de>
tested to not slow down _even_ more a 75 MHz sparc (neko.haemoglobin.org)
thanks to dramsey again for testing
Analysis:
internal_errorf(int, fmt, ...) was only a __dead function if the int argument
was non-0, which the Prevent probably was unable to follow. Change all uses of
internal_errorf(0, fmt, ...) to internal_warningf(fmt, ...); change the pro-
totype of internal_errorf() to internal_errorf(fmt, ...) and all remaining
uses remove the non-0 int argument; add __dead to internal_errorf() proto;
flesh out guts of internal_errorf() and internal_warningf() into a new local
function for optimisation purposes.
Some whitespace cleanup and dead code removal (return after internal_errorf(1))
• if it's in x_literal() (“quote” / ^V) mode, it's accepted like now
• if it's a mb sequence start, it's rejected with a beep
• if it's a mb continuation, the whole sequence is silently rejected
this makes command line editing when accidentally hitting, e.g. with
my mircvs://contrib/samples/dot.Xmodmap, Mode_switch-x much more ro-
bust.
the same as ASCII control characters (U+0001..U+001F), i.e. with a ctrl ca-
ret and its value XOR U+0040; treat their width as 2
fixes crash+cpu hog on 'meta-tab backspace'
and have it return an API-correct const char *
• enhance and stylify comments
• a little KNF and simplifications
• #ifdef DEBUG: replace strchr and strstr with ucstrchr and ucstrstr
that take and return a non-const char *, and fix the violations
• new cstrchr, cstrstr (take and give const char *)
• new vstrchr, vstrstr (take const or not, give boolean value)
• new afreechk(x) = afreechv(x,x) = if (x1) afree(x2, ATEMP)
• new ksh_isdash(str) = (str != NULL) && !strcmp(str, "-")
• replace the only use of strrchr with inlined code to shrink
• minor man page fixes
• Minix 3 signames are autogenerated with gcc
• rename strlfun.c to strlcpy.c since we don't do strlcat(3) anyway,
only strlcpy(3), and shorten it
• dot.mkshrc: move MKSH=… down to the export line
to not disturb the PS1 visual impression ☺
• dot.mkshrc: Lstripcom(): optimise
• bump version
¹) side effect from creating API-correct cstrchr, cstrstr, etc.
uses goto so it must be better ☻
tested on mirbsd-current via both Makefile and Build.sh
sequence is eaten before the command is called; cought by <TGEN>
(Thomas E. Spanjaard) via IRC
fix is to tabcomplete a newline to singlequote+newline+singlequote
* bump version
actually within a (the emacs) editing mode does not work at
all, whether forced or not, only if we leave the editing mode
(at end of input); probably due to it saving stuff from e.g.
x_cols into xx_cols, etc.
-> remove SIGWINCH handling(!)
-> TIOCGWINSZ is now executed at end of every input line
-> use 'ESC #' if you direly need to change line size
only if we got a SIGWINCH -> interactive use speed-up on the
slower boxen (slowness courtesy of otto@obsd)
* ^L (redraw line) now (always) checks window size; use this
instead if required while editing the prompt; changes to the
variables will still be propagated after the line is entered
where we had 'noreturn' etc. but no '__noreturn__')
* Scan for __attribute__((bounded)) and __attribute__((used))
if we have __attribute__((noreturn))
* To be able to scan if certain attributes give warnings,
scan for -Werror with a simple programme which hopefully triggers none
* Convert __attribute__((unused)) to __unused, noreturn -> __dead
* Unify other attributes
* Clean up typography a little more
"set +o emacs-usemeta" and "set -o vi-show8" which are always on now,
since we have proper internationalisation (i.e. utf-8) support, and
assume the user either has a 'C' locale and can't enter 8-bit chars,
his terminal is 8bit-transparent, or he has a 'UTF-8' locale.
main.c: In function 'main':
main.c:208: warning: cast discards qualifiers from pointer target type
main.c:329: warning: cast discards qualifiers from pointer target type
no warnings at autoconf time left either; will take care of these two later
(might revisit changes from this commit), maybe change declararion for the
builtins to have their argv[] be const strings, and go through strict type
and qualifier checking again. this'll further improve stability.
XXX these changes might have introduced (more?) memory leaks,
XXX someone who knows about these tools should verify with
XXX automatic memory usage analysers (valgrind?)
still passes testsuite
XXX one of these uses a gcc extension, ok for now tho
* don't include <ctype.h> any more at all
* don't try nl_langinfo in small mode, just check locale
saves 171 .text, 4 .data, 256 .bss, 1 import
* 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
XXX should probably be screenpos, might need to use x_size* here
XXX this seems to be the tab bug I noticed which we inherited from obsd
* catch backspace if x_col == 0 (XXX better)
* integrate compat.h, version.h into sh.h (dependency trick didn't work anyway)
* mention #ksh in mksh(1) since the founder (twkm) said it's on topic too
(don't remove mention of #mksh despite it's usually empty because of control)
not to stuff, even if TIOCSTI is available
* it's transpose-chars not transpose
bug spotted by hondza who wondered that on ^T the line repeats itself forever
to keys with a trailing tilde (after prefix + control char); the
function is then invoked after prefices, control char and one more
character X are entered; if X is not a tilde, it is ungetc(3)d
with embedded newline (\n), and multi-line prompts with embedded
newlines. single-line prompts should be not affected; prompts with
embedded carriage return (\r) should be fixed by this as well.
also fix prompt_redraw comment (can't remove the variable though,
since it's required if the shell inserts a newline after the prompt
herself).
tested on cygwin and interix
(only the last line of the prompt is redrawn, keep this in mind
when doing the ANSI colour game)
TODO:
* maybe go back up one line and redraw whole prompt?
* optionally edit command on multiple lines, bash-like,
instead of scrolling horizontally
but sync RCS IDs for easier future adaption:
* Simplify savefd() by removing the "noclose" flag and make noclose
behavior the default. Almost all uses of savefd() are followed
by an implicit or explicit close.
* fix typos
* might as well make ksh_getopt() match real getopt(), ie. get rid of that
stupid EOF concept that was never true. adobriyan@gmail
* use SEEK_* for lseek()
* fix lint comments, no functional changes
* remove excessive optimization; from adobriyan@gmail
* only santa checks things twice; from adobriyan@gmail
* Interpret zero-filled numbers as decimal; PR 4213; from Alexey Dobriyan
important for the signedness of the target, to enable using
high-bit7 characters on the command line (eg. in UTF-8 locales)
* make emacs-usemeta and vi-show8 equivalent and document that
in the manual page (i.e. beware if you're in a latin1 locale)
* do not reference SUSv3 (helps GNU groff), write \&" out instead
sitory whose ChangeLog follows. mksh R21 is licenced under the MirOS li-
cence, shown in "sh.h", and a two-clause UCB-style licence by Marc Espie
as shown in "alloc.c".
This executable is a fair bit smaller and shorter than our /bin/ksh that
it is designed to eventually replace (as /bin/sh hardlink), with the old
/bin/ksh to completely vanish. It is still in beta testing though, and I
don't think it will compile on other operating systems.
mksh R21 is a completely new port, bringing together the OpenBSD-current
/bin/ksh, the MirOS-current /bin/ksh and the older mksh R20 (which still
was portable, ocvs-based).
- expat as discussed with bsiegert@ today on the phone
- ksh as announced earlier on the lists
* un-hook lib/libexpat from make includes
* remove /usr/include/{,open}ssl upgrade workaround from includes/Makefile
* nuke old bin/ksh
* nuke libexpat and xmlwf
While here,
o clean up differences where possible
o whitespace cleanup
o ifdef ./. if defined()
o '...' ./. "..."
o echo foo > bar ./. echo foo >bar
o `...` ./. $(...) ./. $$(...)
o `...' ./. '...'
o modernize "our" tree, e.g. WWW in ports
o fix some typos and brainos introduced when renaming OpenBSD to MirBSD
o use hardware 80387 by default
o migrate Apache 1.3.28 OpenBSD ./. MirBSD ./. KAME
o work around as many CVS bugs as possible (add back/delete files, ...)
Synchronize stuff, ready for ongoing changes.
tg