properly together. xxgdb communicates with a gdb running on a pty
that it sets to -echo,-onlcr prior to forking the gdb process.
GNU readline preserves the -echo setting while libedit was undoing it
(setting the tty to a sane state and totally confusing xxgdb's parser).
this diff simply disables libedit if both readline emulation and
"stty -echo" are used/set. that is enough to make xxgdb work once
again, but (XXX) this is not how GNU readline handles stty -echo (it
does not echo anything, but editing commands like ^A,^K, etc.
still work), so the readline emulation isn't perfect.
change reviewed by christos.
bindings and their descriptions
* manually add the output of 'sh ./makelist -m vi.c ed.c common.c' to
a new section in editrc(5) called `EDITOR COMMANDS'
history_tokenize(): avoid stepping too far if backslash is last character
on the passed string
update some comments
This makes gdb \ escaping work reliably and fixes lib/9511 by Assar Westerlund.
I wrote for tcsh(1) three years ago.
- implement EL_RPROMPT, which allows a setting/getting of a function which
returns a string to be used as the right-side prompt.
- improve HISTORY and AUTHORS sections in editline(3).
- bump shlib minor version for EL_RPROMPT.
XXX: due to an implementation issue, the rprompt has a 1 space gap before the
edge of the logical screen. editline's logical screen is 1 space less
than the full screen width, so there's a 2 space gap between the rprompt
and the right end of the physical screen. i'm not concerned about this.
re_refresh(): cast the character passed to re_addc() to unsigned char,
so we don't end up calling isprint() with negative value
when chars are signed and character value is >= 128
- don't reopen fp if it was successfully opened in a previous check
- use strlcpy()/strlcat() instead of snprintf() to build a pathname,
since the former are more portable to other systems than the latter
* whitespace fascism
plus a few bogons noted along the way:
1) Set EDIT_DISABLED if terminal type is emacs.
2) fix bug in NO_TTY mode which caused it to not notice CR or LF
3) implement EDIT_DISABLED within libedit to be somewhat like NO_TTY,
except that a prompt is printed first.
* implement el_get(EditLine *, int op, void *result), which does the
inverse of el_set()
* add EL_EDITMODE operation to el_set and el_get; if non zero editing
is enabled (the default).
* add "edit on | off" editrc command, which modifies EL_EDITMODE.
users can now add '*:edit off' in ~/.editrc as an advisory to
disable editing.
NOTE: at this time EL_EDITMODE is just an indication of the
state of the 'edit' command. It's up to the application to check
this after el_source() or el_parse() to determine if editing is still
required.
of the user's namespace. We need to do this because the readline
emulation functions call term_beep(), but readline isn't built using
the same (funky) namespace-protection method as the rest of libedit (it's
included like a normal library object).
Without this (fairly disgusting) hack, any program linked against
libedit will fail to link with an unresolved reference to term_beep()
if using an Elf toolchain (e.g. on the Alpha).
XXX Why this doesn't happen with NetBSD's a.out toolchain is a mystery
XXX to me, and I'm not sure I really want to know (given that a.out
XXX _should_ break the same way as Elf does in this case).
anywhere else.
- for now, override WARNS=0 in librpcsvc and libwrap, until they're
cleaned up
- rcsid police
lib is now clean (except for librpcsvc and libwrap) on the i386, and
this should motivate the other ports to fix any other minor problems
that their compilers pick up that the i386 version doesn't.
line (a la ^R). This is useful if the binding outputs information and
mucks up the input line. To be used in ``list-choices'' bindings (refer
to the ^D binding in csh when filec is set)
* when parsing ^char control chars, check the correct char when determining
validity (previously, ^char was a NOP interpreted as the literal string
because of this bug)
* add a man page describing editrc
* fix bugs in el_parse():
* didn't execute command when program name matched (test reversed)
* was checking against empty string instead of program name
* after checks, command to run also pointed to empty string
[christos - the author of libedit - ok-ed the man pages in general (which I
wrote from scratch by RTFS) as well as the bugfix]
is defined, V8 regular expresion functions are used if REGEXP is defined,
and BSD regular expression functions are used if neither are defined.
And defined REGEX in sys.h so that programs using libedit don't have
to link with libcompat.
mrg