Commit Graph

2923 Commits

Author SHA1 Message Date
Michael Stapelberg 7180dca3b2 Bugfix: Correctly free the xcb_get_geometry reply when not managing a window 2011-07-31 20:39:33 +02:00
Michael Stapelberg 22de7593c5 Bugfix: free() the grab_pointer reply 2011-07-31 19:52:58 +02:00
Michael Stapelberg 6c35a0b93b handle the xcb_grab_pointer reply 2011-07-31 19:48:39 +02:00
Michael Stapelberg ffbcfa2c97 Bugfix: also free() X11 errors in the main loop 2011-07-31 19:41:57 +02:00
Michael Stapelberg fdfa88586d Forgot free(event) 2011-07-31 19:34:55 +02:00
Michael Stapelberg fa1fe3cfed Bugfix: Same xcb_send_event fix as 1e0033bce4 2011-07-31 19:33:56 +02:00
Michael Stapelberg 5bfc89f733 Bugfix: Correctly pass the pointer to xcb_send_event 2011-07-31 19:26:38 +02:00
Michael Stapelberg 3a41aea75b add xcb_icccm_get_wm_protocols to xcb_compat.h (Thanks Merovius) 2011-07-31 18:32:28 +02:00
Axel Wagner 3302b22f71 Bugfix: Delay event-subscription until reconnect (thx aniou) 2011-07-31 18:26:52 +02:00
Michael Stapelberg 52b3646ecc xcb: don’t use the _unchecked variant for requests which have a reply 2011-07-31 18:19:41 +02:00
Michael Stapelberg 7f3f7c26b6 Bugfix: free replies after using them 2011-07-31 18:17:56 +02:00
Michael Stapelberg d9ff172d36 Bugfix: load_font: free reply after using it 2011-07-31 17:57:11 +02:00
Michael Stapelberg cbfcf5bd8b Bugfix: free font string after using it 2011-07-31 17:54:53 +02:00
Michael Stapelberg ee83df9684 Bugfix: cfgparse: correctly free parser/lexer 2011-07-31 17:50:54 +02:00
Michael Stapelberg 088681c781 Bugfix: property handlers: correctly free replies in all cases 2011-07-31 17:46:41 +02:00
Michael Stapelberg 1e0033bce4 Bugfix: Allocate 32 bytes for xcb_send_event, it always copies 32 bytes
Fixes an 'uninitialized memory' error in valgrind
2011-07-31 17:46:01 +02:00
Michael Stapelberg 4875868689 Bugfix: Free property replies in their handlers 2011-07-31 17:12:37 +02:00
Michael Stapelberg d3e3339a05 Bugfix: Fix small memory leak when initializing output Cons 2011-07-31 17:11:51 +02:00
Michael Stapelberg f2fc523de7 Bugfix: fix memory leak in src/cmdparse.y 2011-07-31 17:11:28 +02:00
Michael Stapelberg 6ef3866108 docs/hacking-howto: comment out a few sections which are not updated yet
We want to get this release out and will update the sections afterwards. Sorry!
If you need to know stuff *right now*, please ask us on IRC
2011-07-31 16:19:25 +02:00
Axel Wagner 8d09ed7bf5 Update changelog and copyright, bump version and more 2011-07-31 16:16:17 +02:00
Axel Wagner 8e275d85b5 Correct some minor mistakes in the manpage (thx sECuRE) 2011-07-31 15:58:14 +02:00
Axel Wagner 56158329b3 Merge branch 'tree' into next 2011-07-31 15:54:59 +02:00
Michael Stapelberg 033466585d docs/userguide: update the status line section, recommend i3bar
dzen2 does not get mentioned anymore. They did not release a new version for
more than three years (-dock was added 2008-04-03). I am sick of telling people
they need to check out the latest SVN trunk of dzen2.
2011-07-31 15:39:18 +02:00
Michael Stapelberg c601ac1d8e docs/userguide: little fixes by fernando 2011-07-29 13:31:37 +02:00
Michael Stapelberg 1b0d6ab202 remove unnecessary 'extern' keyword from function prototypes in include/xcursor.h 2011-07-29 13:16:26 +02:00
Michael Stapelberg 3721c598bd Bugfix: Set the cursor via the Xlib connection if XCursor is supported
This fixes a race where we created cursors on the Xlib connection, flushed,
then used the cursor on the XCB connection. Even though we flushed, the X
server did not process the requests yet and therefore returned a BadCursor
error.

This bugfix uses the Xlib connection for setting the root window cursor which
will ensure that the requests are properly serialized.

An easy test for this (on my machine) is the following ~/.xsession:
    xsetroot -cursor_name cross
    exec i3
If you see a cross cursor instead of the pointer, the race happens. You’ll see
a error_code=6 error in your ~/.xsession-errors.
2011-07-29 13:13:51 +02:00
Michael Stapelberg bc7dec6a4c X11 errors: also log the error_code (see X Protocol, Appendix B, Errors) 2011-07-29 13:11:02 +02:00
Fernando Tarlá Cardoso Lemos ae9c90ba19 Don't leak the cursor_id.
Fixes a server-side leak. The cursor_id is not needed when we're
using themed cursors (we never freed it either).
2011-07-29 12:38:44 +02:00
Fernando Tarlá Cardoso Lemos 7a18dcc89e Set the root window cursor. Fixes #442. 2011-07-29 12:38:24 +02:00
Michael Stapelberg 20c68b1869 docs/userguide: explain the split command 2011-07-29 01:12:19 +02:00
Michael Stapelberg 7e6aa72664 docs/userguide: explain default_orientation config directive 2011-07-29 01:12:06 +02:00
Michael Stapelberg 85b01df11b userguide: add popup_during_fullscreen configuration directive 2011-07-29 00:26:30 +02:00
Michael Stapelberg 8ef51a7cfa update docs/userguide (various little fixes/updates) 2011-07-28 22:51:48 +02:00
Michael Stapelberg ed30043950 Bugfix: Also abort drag_pointer after XCB_KEY_PRESS / XCB_KEY_RELEASE (Thanks eeemsi) 2011-07-28 21:01:34 +02:00
Michael Stapelberg 700f5bd443 docs/hacking-howto: a few little updates 2011-07-27 17:05:28 +02:00
Michael Stapelberg d1fae4c0f7 docs/hacking-howto: update for RandR instead of Xinerama 2011-07-27 16:37:19 +02:00
Michael Stapelberg e8335bbb89 docs/hacking-howto: update the file description 2011-07-27 16:32:24 +02:00
Michael Stapelberg 118a37510d docs/debugging: make the difference between logging and verbose logging more clear, little style fixes 2011-07-27 16:13:15 +02:00
Michael Stapelberg eea804cd94 migrate-config: convert old 'with container' commands to 'focus parent; $command' (+test) (Thanks Marcus) 2011-07-27 14:36:22 +02:00
Michael Stapelberg 2b940296a8 tests: t/17-workspace.t: keep workspace 1 open
When not running i3 in a terminal from workspace 1, this would fail otherwise
2011-07-27 14:35:40 +02:00
Michael Stapelberg 26c1a463db tests: keep an X11 connection open to every display, fixes races
Otherwise, i3 is the only client and thus the X server shuts down after i3
disconnects. Xdummy restarts an X server, but that takes some time and
therefore sometimes the next testcase was unable to connect.
2011-07-27 14:34:35 +02:00
Michael Stapelberg 86de02ca8c default config: start i3bar -d by default
like i3-migrate-config-to-v4.pl does it
2011-07-27 13:32:27 +02:00
Michael Stapelberg 0d77ca9a3d i3-nagbar: correctly check for errors when connecting to X11 2011-07-27 13:30:12 +02:00
Michael Stapelberg ec88f6501d i3-config-wizard: use a managed, floating (dialog) window. fixes focus problems
Probably only when another window is already opened, but people can do that
(start i3 from within a terminal in their X session)
2011-07-27 13:28:32 +02:00
Michael Stapelberg 2a43a49595 i3-config-wizard: correctly check for errors when connecting to X11 2011-07-27 13:05:40 +02:00
Michael Stapelberg f84264ae41 remove _NET_WM_WORKAREA from _NET_SUPPORTED for now
See ticket #301, this is a problem with Qt apps on your second monitor.

We should implement better EWMH support, but not before the release :).
2011-07-27 12:51:38 +02:00
Michael Stapelberg 44be907197 add Xdummy script to start a headless X11 server 2011-07-25 15:37:46 +02:00
Michael Stapelberg 4dde5bb863 testsuite: allow parallel runs. decreases the test time by 60% when using 4 displays 2011-07-25 15:37:13 +02:00
Michael Stapelberg 49ea805905 tests: get the socket path from the X11 atom I3_SOCKET_PATH
The hard-coded socket paths are not useful when running the test suite in
parallel.
2011-07-25 13:35:56 +02:00