aboutsummaryrefslogtreecommitdiff
path: root/degesch.c
Commit message (Collapse)AuthorAgeFilesLines
* Update NEWS, cleanupPřemysl Janouch2016-03-071-2/+1
|
* degesch: rewrite input layerPřemysl Janouch2016-03-071-446/+652
| | | | Now with less #ifdefs.
* degesch: use libffi to unify input callbacksPřemysl Janouch2016-03-061-326/+284
| | | | And fuck you both, Readline and Editline.
* degesch: fix +=/-= to null config itemsPřemysl Janouch2016-03-051-8/+11
|
* degesch: mark a problemPřemysl Janouch2016-02-291-0/+8
|
* Resolve tls_ca_{file,path} relative to config dirPřemysl Janouch2016-02-191-32/+54
|
* degesch: fix segfault on sent messages w/o bufferPřemysl Janouch2016-02-121-0/+4
|
* degesch: update screen size when terminal resumedPřemysl Janouch2016-02-101-0/+1
| | | | Apparently we don't receive the events when we give up the terminal.
* degesch: use mkstemp() with a safe umaskPřemysl Janouch2016-02-091-0/+3
|
* degesch: log messages from /quote and pluginsPřemysl Janouch2016-02-091-93/+149
| | | | That is, parse back all output messages and log based on that.
* degesch: allow changing the list of used CAPsPřemysl Janouch2016-02-011-7/+16
| | | | Bump liberty.
* degesch: optimize buffer memory usagePřemysl Janouch2016-01-311-35/+39
| | | | | | | | | | | | | We have approximately 5 formatter_items per buffer_line. Let's assume we're on a 64-bit machine. Then there were (5 * 2) + 3 useless pointers (104 bytes) as well as 5 * (4 + 4) = 40 bytes of wasted space because of needless padding. That's 144 bytes already. Compared to that, this change adds 16 bytes of overhead for an array sentinel, i.e. 128B less. With a limit of 1000 lines per buffer, we've saved ~128kB per buffer on completely useless data, and code complexity stays roughly the same. All in all, memory usage for buffers should be about 50% lower.
* kike: add support for IRCv3.2 server-timePřemysl Janouch2016-01-311-0/+5
|
* degesch: add an assertionPřemysl Janouch2016-01-181-1/+5
| | | | It should never fail with current code.
* degesch: disable TLS compressionPřemysl Janouch2016-01-181-0/+7
|
* degesch: Lua: fix memory leak on load failurePřemysl Janouch2016-01-171-2/+1
|
* degesch: lesser heap fragmentationPřemysl Janouch2016-01-171-11/+7
|
* degesch: fix backlog limitPřemysl Janouch2016-01-151-1/+1
| | | | It was effectively infinite.
* degesch: fix and simplify screen handlingPřemysl Janouch2016-01-151-18/+6
| | | | Now with less madness.
* degesch: enforce fullscreen buffersPřemysl Janouch2016-01-151-12/+4
| | | | | | Probably long overdue. Now we actually look like something resembling a regular IRC client.
* degesch: option for fullscreen buffersPřemysl Janouch2016-01-151-6/+31
|
* degesch: make backlog limit configurablePřemysl Janouch2016-01-151-11/+34
|
* degesch: Lua: add autocomplete hooksPřemysl Janouch2016-01-151-26/+156
|
* degesch: Lua: allow simulating user inputPřemysl Janouch2016-01-141-1/+23
| | | | Also added UTF-8 validation to buffer:log() while I'm at it.
* Bump liberty, shuffle some codePřemysl Janouch2016-01-141-94/+94
|
* degesch: Lua: pass hostname to on_connectedPřemysl Janouch2016-01-091-3/+3
|
* degesch: Lua: set sockets to nonblockingPřemysl Janouch2016-01-091-0/+1
|
* Bump libertyPřemysl Janouch2016-01-091-1/+0
|
* degesch: Lua: actually allow filtering out inputPřemysl Janouch2016-01-091-0/+4
|
* degesch: Lua: provide a traceback on load errorPřemysl Janouch2016-01-091-4/+9
|
* degesch: Lua: fix plugin configuration namesPřemysl Janouch2016-01-091-1/+10
|
* degesch: Lua: fix luaL_ref() usagePřemysl Janouch2016-01-091-11/+10
|
* degesch: Lua: fix configuration loadingPřemysl Janouch2016-01-091-6/+11
| | | | Not the cleanest solution but it has to do for now.
* degesch: make sure newlines are output correctlyPřemysl Janouch2016-01-081-0/+8
|
* degesch: Lua: finish implementation of connectionPřemysl Janouch2016-01-071-4/+182
|
* degesch: Lua: avoid resource leakPřemysl Janouch2016-01-071-10/+16
| | | | If a connector's on_success callback fails, we need to destroy the connection.
* Factor out socket_io_try_{read,write}()Přemysl Janouch2016-01-071-80/+45
| | | | To be reused in Lua connection API.
* degesch: fix handling of input editor deathPřemysl Janouch2016-01-071-4/+12
|
* degesch: fix switching of buffers by commandPřemysl Janouch2016-01-071-8/+27
| | | | | | Readline used to erase the new buffer's contents. Defer processing.
* degesch: Lua: fix a resource leakPřemysl Janouch2016-01-061-2/+6
|
* degesch: refactor LuaPřemysl Janouch2016-01-051-93/+62
| | | | And fix handling of nil returns from filter callbacks.
* degesch: Lua: halfplement a connector wrapperPřemysl Janouch2016-01-051-5/+237
| | | | You can't do anything reasonable with the socket now.
* degesch: Lua: allow arbitrary userdata propertiesPřemysl Janouch2016-01-041-2/+35
|
* degesch: Lua: use references for hook callbacksPřemysl Janouch2016-01-041-10/+8
| | | | | Don't associate the callback with the full userdata object, we'll need this for something else.
* degesch: add basic autocomplete for /topicPřemysl Janouch2016-01-041-1/+28
|
* Bump liberty, enable TLS SNIPřemysl Janouch2016-01-041-30/+18
| | | | | | | Involves some rewrites to fit the new APIs. SNI has been implemented Mostly just because we can, I don't think it's widely in use and kike doesn't support this feature of the protocol either.
* degesch: fix a memory leak under libeditPřemysl Janouch2016-01-011-14/+19
|
* degesch: fix segfault on /quit under libeditPřemysl Janouch2015-12-311-5/+5
|
* degesch: Alt-Tab switch to the last bufferPřemysl Janouch2015-12-311-0/+31
|
* degesch: avoid fileno() after fork()Přemysl Janouch2015-12-311-4/+4
| | | | | It's not guaranteed to be async-signal-safe, which may matter once we start using threads. And it's also cleaner to just pass the FD.