summaryrefslogtreecommitdiff
path: root/usr.bin
AgeCommit message (Collapse)Author
2009-11-12Emulate the ri (reverse index) capability: this allows tmux to at least startNicholas Marriott
on Sun consoles (TERM=sun or sun-color), even if there appear to still be problems on some boxes (my Blade 100 is fine but edd's Blade 1000 shows odd screen corruption).
2009-11-12Support rxvt-style keys again, but this time: support all the variations, putNicholas Marriott
them in as raw escape sequences rather than fiddling with the values from terminfo, put them /after/ the terminfo values so the latter take precedence.
2009-11-12Use the default 4 correctly if no type size is given to od -t.Nicholas Marriott
ok deraadt miod
2009-11-12When values are changed, log the change in the ``name: old -> new''Alexandre Ratchov
format, as mixerctl does. If -n is used log the new value only. from Pawlowski Marcin Piotr <pmp.openbsd at gmail.com>, thanks. ok jakemsr
2009-11-12if a value is changed and -n is used, print the new value only.Alexandre Ratchov
from Pawlowski Marcin Piotr <pmp.openbsd at gmail.com> ok jakemsr
2009-11-11Fix memory leaks found by parfait.Nicholas Marriott
ok deraadt@ jsg@
2009-11-11fix race condition in x11/agent channel allocation: don't read afterMarkus Friedl
the end of the select read/write fdset and make sure a reused FD is not touched before the pre-handlers are called. with and ok djm@
2009-11-11Rewrite a confusing loop when freeing the arg array on exit and move the checkNicholas Marriott
for argv being NULL, prompted by parfait via deraadt. Also fix some definite brokenness when assigning multiple environment variables in arguments (such as "X=1 Y=2").
2009-11-11Add an explicit zero-length check for UTF-8 input data, prompted by a reportNicholas Marriott
from parfait via deraadt. While here, add a statement to set the width when filling with _s if not enough space (width should never be high enough at the moment anyway), and wrap some long lines.
2009-11-11memory leak with #ifdef SMALL, found by parfaitTheo de Raadt
ok jsg
2009-11-11do not walk off the prefix array, off by one; found by parfaitTheo de Raadt
ok jsg millert
2009-11-11fix a buffer overflow found by parfaitJonathan Gray
ok millert@
2009-11-11Free the pane bufferevent when the fd is closed (the signal could come beforeNicholas Marriott
the error callback).
2009-11-11Only need to chmod +x or -x the socket when a client is created, lost orNicholas Marriott
attached, rather than every event loop.
2009-11-11If this is an else-type conditional and previous corresponding oneFederico G. Schwindt
evaluated to true, skip. tested by simon@ and myself. espie@ ok.
2009-11-10There is no real standard for modifier plus function keys. Previously, tmuxNicholas Marriott
output some from rxvt but in other ways did the same as xterm or other terminals, but this is a bit inconsistent. xterm's method is fairly sensible and we already support it (xterm-keys), so enable it by default instead.
2009-11-10Don't output rxvtisms either.Nicholas Marriott
2009-11-10Twiddling the last bit is an rxvtism, so do not support it in the table byNicholas Marriott
default.
2009-11-10Whoops, this is needed for last commit as well.Nicholas Marriott
2009-11-10Lookup key as a named key (eg 'Space') before checking for single characterNicholas Marriott
keys, makes C-Space/M-Space etc resolve to the correct key code.
2009-11-10Don't return 1 unless there was actually a problem (signal/lost server) ratherNicholas Marriott
than for all events (normal exit/detach/etc).
2009-11-10fix typo: in the example use ``rate'' instead of ``sample_rate'',Alexandre Ratchov
since the later is a hidden alias of the former. from Jan Stary <hansi _at_ stare.cz>, thanks
2009-11-10Set close-on-exec on various descriptors so they don't get leaked toDarren Tucker
child processes. bz #1643, patch from jchadima at redhat, ok deraadt.
2009-11-10clarify that StrictModes does not apply to ChrootDirectory. PermissionsDamien Miller
and ownership are always checked when chrooting. bz#1532
2009-11-10explain the constraints on LocalCommand some more so people don'tDamien Miller
try to abuse it.
2009-11-09Just ignore tty fd errors rather than dying, stops the server dying if theNicholas Marriott
session is disconnected abrubtly (eg ssh ~.).
2009-11-09The input key should be a u_char. Fixes top-bit-set input problem reported byNicholas Marriott
ajacoutot@.
2009-11-09move mutually exclusive options (-c, -e, -f, -n and -u) to the right place.Igor Sobrado
ok jmc@
2009-11-09Constify buf.Nicholas Marriott
2009-11-08use the same algorithm to attach play-only, record-only andAlexandre Ratchov
full-duplex streams. Fixes, streams with different modes not starting simultaneously.
2009-11-06Don't try enable/disable the event if the window pane is dead (fd == -1), asNicholas Marriott
the event will have been freed.
2009-11-05Clear to the end of the screen from the right starting point when drawingNicholas Marriott
line-by-line (in panes or if ed not supported). Fixes problem spotted by Frank Terbeck.
2009-11-05IPv6 support for divert sockets.Michele Marchetto
tested by phessler@ pyr@ ok claudio@ "go ahead" deraadt@
2009-11-05Old xterm F1-F4 are \033O_P not \033[O_P.Nicholas Marriott
2009-11-05Unused variable. Aargh.Nicholas Marriott
2009-11-05Switch the tty key tree over to an (unbalanced) ternary tree which allowsNicholas Marriott
partial matches to be done (they wait for further data or a timer to expire, like a naked escape). Mouse and xterm-style keys still expect to be atomic.
2009-11-05key_string_lookup_key uses a static buffer, so copy its output into the workingNicholas Marriott
buffer before calling the command print function which can also use it (eg send-keys).
2009-11-05Key flags are only used for initialisation so they are not needed in the mainNicholas Marriott
tty_key struct.
2009-11-05EVLOOP_ONCE takes care of the wakeup, so no need to call event_loopexit(NULL).Nicholas Marriott
2009-11-05Now all timers are events, there is no longer any need to wake up every 50 ms -Nicholas Marriott
only wake up when an event happens.
2009-11-05Switch tty key input over to happen on a read event. This is a bit moreNicholas Marriott
complicated because of escape input, but in that case instead of processing a key immediately, schedule a timer and reprocess the bufer when it expires. This currently assumes that keys will be atomic (ie that if eg F1 is pressed the entire sequence is present in the buffer). This is usually but not always true, a change in the tree format so it can differentiate potential (partial) key sequences will happens soon and will allow this to be fixed.
2009-11-05in rpipe_done (wpipe_done), all buffers must be detachedAlexandre Ratchov
before deleting the writer (reader) otherwise the later could trigger rpipe_done again.
2009-11-05Convert the key repeat timer to an event.Nicholas Marriott
2009-11-04Change window name change to use a timer event rather than a gettimeofday()Nicholas Marriott
check every loop.
2009-11-04Move status timer check into the global once-per-second timer, this could maybeNicholas Marriott
be done better but one every second is better than once every 50 ms.
2009-11-04Use timeout events for the identify and message timers.Nicholas Marriott
2009-11-04Don't reenlist the client imsg event every loop, instead have a small functionNicholas Marriott
to it and call it after the event triggers or after a imsg is added.
2009-11-04It would help if I read my own comments... make alt keys work again by sendingNicholas Marriott
alt AND the key not alt instead of it.
2009-11-04Move some common code into a function.Nicholas Marriott
2009-11-04Tell the client to exit on configuration file error.Nicholas Marriott