applications/i3
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
2,239 Commits 3 Branches 0 Tags 9.1 MiB
72d2b2c9bd
Commit Graph

1168 Commits

Author SHA1 Message Date
Michael Stapelberg
cdeb49127f Bugfix: restore focus to the correct window when a non-focused window gets destroyed (+testcase) 2011-03-14 16:34:35 +01:00
Michael Stapelberg
f162e7efaa refactor font caching to just save the ID instead of mainting a cache with pattern→id-mapping 2011-03-10 23:20:17 +01:00
Simon Kampe
6e4a2b0b96 Fallback fonts for when requesting a erronous font with load_font (e.g. user have specified a font which does not exist in the config file). 2011-03-10 21:46:31 +01:00
Michael Stapelberg
86637d2e07 Bugfix: Make level up a noop during fullscreen mode (+testcase) (Thanks dothebart) 
Fixes #341
 2011-03-09 18:37:05 +01:00
Michael Stapelberg
74b90cd83f Bugfix: Send WM_DELETE / kill window the right way (Thanks dothebart) 
Fixes #336
 2011-03-09 18:08:26 +01:00
Michael Stapelberg
0a24057241 When leaving fullscreen, set focus to con which was opened during fullscreen (+testcase) (Thanks dothebart) 2011-03-07 00:06:27 +01:00
Michael Stapelberg
287d7f9527 Bugfix: Don’t focus new cons when there is a fullscreen con (Thanks dothebart) 
Also, remove the focus_it parameter from tree_open_con, it makes more sense to
call con_focus outside of the function.
 2011-03-06 23:26:02 +01:00
Michael Stapelberg
8ce5f2a21b Bugfix: Fix crash with transient dock clients caused by 7154fecbb 2011-03-06 22:02:02 +01:00
Michael Stapelberg
c17b3b7560 remove left-over xcb_aux_sync (debugging code) 2011-03-06 21:56:15 +01:00
Michael Stapelberg
b0b195318c rendering: bugfix: stack child windows of stacked/tabbed cons according to their focus 
fixes focusing the wrong window after closing a con
 2011-03-06 21:49:17 +01:00
Michael Stapelberg
4f26316aaa x: disable all events while re-stacking windows, prevents unwanted EnterNotifys 2011-03-06 21:48:49 +01:00
Michael Stapelberg
0689f6d8f1 Bugfix: use tree_render() instead of x_push_changes() to re-render and update the stack 2011-03-06 20:45:03 +01:00
Michael Stapelberg
7154fecbbf Implement the popup_during_fullscreen option, set default to leave_fullscreen 
Fixes #333
 2011-03-06 15:46:06 +01:00
Michael Stapelberg
51ff0f80a6 Obey minimum size when resizing floating windows 
Fixes #285
 2011-03-06 14:15:46 +01:00
Michael Stapelberg
1891a385ad Bugfix: also stop searching for resize cons when reaching a floatingcon (Thanks julien) 2011-03-06 13:35:07 +01:00
Michael Stapelberg
4514146ed9 Use the combined geometry of children when setting a split container to floating (+testcase) (Thanks mseed) 
Fixes #332
 2011-03-06 02:39:15 +01:00
Michael Stapelberg
88ab66e742 Bugfix: don’t focus dock clients with new click handling code (Thanks mseed) 2011-03-06 02:06:05 +01:00
Michael Stapelberg
24463718cc refactor the click handling completely 
Also re-implements floating modifier on tiling cons.

Fixes: #288 and a lot of headache :)
 2011-03-05 20:23:29 +01:00
Michael Stapelberg
0a2ee1d2aa Bugfix: Focus workspace after closing one of multiple dock clients (+testcase) (Thanks mseed) 2011-03-04 15:21:18 +01:00
Michael Stapelberg
f6a21994bf Re-implement focus follows mouse for outputs 
That is, moving your mouse pointer to a different workspace which does not have
any windows on it yet will correctly update the focus.
 2011-03-03 16:36:18 +01:00
Michael Stapelberg
caa1ac1a9f Use the original geometry for floating windows 2011-03-03 16:22:22 +01:00
Michael Stapelberg
b484b9ab32 Don’t create floating containers from whole workspaces when they are empty (Thanks mseed) 
This fixes #327
 2011-03-03 14:55:02 +01:00
Michael Stapelberg
a038d2674b Bugfix: 'level up' needs to stop at the workspace con + testcase (Thanks mseed) 2011-03-03 14:30:13 +01:00
Michael Stapelberg
62362a464d fix invalid memory access in xcb_reply_contains_atom (Thanks ys) 2011-03-03 14:14:35 +01:00
Michael Stapelberg
77640da9e7 Bugfix: don’t focus the dockarea when closing a dock client 2011-03-03 14:03:06 +01:00
Michael Stapelberg
e0647b7fc2 Bugfix: don’t focus dock clients (Thanks mseed, mist) 
Fixes #321 and #323
 2011-03-03 13:59:42 +01:00
Michael Stapelberg
33c2b4e582 Bugfix: Correctly fix the percent values for resizing when making a floating con tiling (Thanks mseed) (+testcase) 2011-02-28 23:59:56 +01:00
Michael Stapelberg
b595ff05c5 Fix regression: Don’t add floating nodes twice when restoring layout (+testcase) 2011-02-28 20:44:23 +01:00
Michael Stapelberg
d80105103c Bugfix: Re-attach floating cons to the right container (Thanks mseed) 
This fixes #315.
 2011-02-28 17:23:50 +01:00
Fernando Tarlá Cardoso Lemos
beaa85ceb9 Don't leak this descriptor (thanks dothebart). 
Note that fclose closes the file descriptor frees the stream.
 2011-02-28 16:49:10 +01:00
Fernando Tarlá Cardoso Lemos
269d360f30 Some assertions to make the static analyzer happy. 
Assertions give hints to the static analyzer about code paths
where we make assumptions. Used the Clang Static Analyzer.
 2011-02-28 16:49:03 +01:00
Fernando Tarlá Cardoso Lemos
f4ec0bceff Very minor issues found with statical analysis. 
The Clang Static Analyzer uncovered those issues:

- The variable "changed" in handlers.c is written to, but it's
  never read since that specific write, so the write is not
  necessary.

- In util.c, "tail" may be NULL. In that case, we shouldn't pass
  it to strlen because strlen's behavior is not defined when s is
  NULL.

- In util.c, "write_index" is incremented twice. It's never used
  anymore after being incremented once, so the second increment is
  not necessary.
 2011-02-28 16:48:22 +01:00
Michael Stapelberg
ffc71859a3 Implement support for top/bottom dock clients (according to _NET_WM_STRUT_PARTIAL or requested position) 2011-02-21 14:27:32 +01:00
Michael Stapelberg
b6f81fe43c Bugfix: restore the original width/height with X11 border when restarting (Thanks Merovius) 2011-02-21 03:13:27 +01:00
Michael Stapelberg
f34b045619 Fix dock client handling for inplace restarts 2011-02-21 03:01:55 +01:00
Michael Stapelberg
3dfe5c8a9a bugfix: fix clicking on dock clients (Thanks mseed) 2011-02-21 01:58:57 +01:00
Michael Stapelberg
9719b21243 refactor some places to use output_get_content() 2011-02-21 01:55:36 +01:00
Michael Stapelberg
a92b9dca73 Bugfix: fix disabling RandR outputs 2011-02-21 01:43:39 +01:00
Michael Stapelberg
35e79c87c8 Place dock clients on the output corresponding to their geometry request 2011-02-21 01:28:29 +01:00
Michael Stapelberg
a33d869885 Bugfix: Correctly open workspaces on additional outputs 2011-02-21 01:12:22 +01:00
Michael Stapelberg
6b272fea55 Bugfix: in get_workspaces, only consider the CT_CON, not the CT_DOCKAREAs (Thanks fernandotcl) 2011-02-21 00:54:29 +01:00
Michael Stapelberg
272ab840c7 Fix fullscreen mode with dock clients 2011-02-21 00:40:35 +01:00
Michael Stapelberg
9a0bc77baf bugfix: dock clients cannot be floating 2011-02-21 00:23:07 +01:00
Michael Stapelberg
7f89c71689 Implement dock mode, update testsuite 
Currently, dock clients are only possible at the top.
 2011-02-20 23:43:03 +01:00
Michael Stapelberg
c5ab16c00d same fix, but for moving (search above the current con when moving is not possible in this direction) 2011-02-19 20:11:47 +01:00
Michael Stapelberg
86500c5b88 Skip containers which got only one child when looking for the next/previous one to focus 2011-02-19 19:50:09 +01:00
Michael Stapelberg
6a6746b967 revert the replacement of a single h/v-split with its child container 
Makes more problems than it creates. Will use a different fix suggested by
Merovius.
 2011-02-19 18:30:26 +01:00
Michael Stapelberg
53d9072ca7 implement TAILQ_SWAP (only for consecutive elements, order relevant) and use it 2011-02-15 02:21:23 +01:00
Michael Stapelberg
834f4d7bc2 add missing function documentation 2011-02-14 23:17:30 +01:00
Michael Stapelberg
26a416e016 refactor tree_move() into src/move.c, change config (!), change testcase 
Due to lots of cases which were added and added to tree_move(), the function
was not really easy to understand. For this refactoring, I wrote tree_move()
from scratch, thinking about (hopefully) all cases. The testsuite still passes.

The move command also has different parameters now. Instead of the hard to
understand 'before v' stuff, we use 'move [left|right|up|down]'.
 2011-02-14 23:05:20 +01:00
Michael Stapelberg
28dd226259 refactor code for removing children from a con 
Let’s see how this callback stuff will work out. If it doesn’t work out well,
we will remove it.
 2011-02-14 18:08:36 +01:00
Michael Stapelberg
a5e075c154 Automatically close empty parent cons when making their last child floating (Thanks mseed) 
This fixes #313
 2011-02-14 16:43:41 +01:00
Michael Stapelberg
b800555161 Bugfix: check ->layout, not ->type for L_STACKED 2011-02-14 16:35:48 +01:00
Michael Stapelberg
305eac0e71 Bugfix: Correctly check asprintf() return value 2011-02-14 16:33:42 +01:00
Fernando Tarlá Cardoso Lemos
9cf48f17bb Fix libxcursor -> xcb cursors fallback. 2011-02-06 17:47:28 +01:00
Michael Stapelberg
07381ccb7b fix a rendering problem for split cons inside tabbed cons (Thanks julien) 
This fixes #280.
 2011-02-02 17:56:29 +01:00
Michael Stapelberg
4d0106b00f bugfix: don’t crash when moving windows out of a floating con (Thanks mseed) 2011-02-01 17:53:15 +01:00
Michael Stapelberg
d855bea215 Bugfix: Correctly focus con when moving to another workspace (Thanks mseed) 
This fixes #310.
 2011-02-01 17:17:50 +01:00
Michael Stapelberg
97ab44b3d8 disable workspace-level move operations (not yet implemented) (Thanks mseed) 2011-02-01 16:59:02 +01:00
Michael Stapelberg
9b01b1a7a6 Bugfix: When the container which was just closed is focused, we *do* need to focus another one (Thanks mseed) 2011-02-01 16:42:59 +01:00
Michael Stapelberg
dad7c0da00 Don’t create a split container if no other cons are on a workspace (Thanks mseed) 
This fixes #306.
 2011-02-01 16:08:47 +01:00
Michael Stapelberg
0238ce3c73 fix some compiler warnings 2011-02-01 15:43:59 +01:00
Fernando Tarlá Cardoso Lemos
8f4b9ddaa4 Fix the choice of the right containers for resizing. 
Also fixes #311, which was an assertion failure that uncovered
this problem. Thanks mseed, Merovius.
 2011-01-30 18:52:32 +01:00
Fernando Tarlá Cardoso Lemos
ac8fb2399d Don't mess with sizes when moving to other ws. 2011-01-29 15:19:41 +01:00
Fernando Tarlá Cardoso Lemos
8be40932f2 Fix assertion when moving out of a floating container. 2011-01-28 01:25:06 +01:00
Fernando Tarlá Cardoso Lemos
f465b3c11d Don't mess with the focus if we're not killing. 
If we're not killing the mapped window and we're not killing the
parent window either in tree_close, then there's no reason to try
to change the focus. This fixes focus issues when moving a
container around another container (move up, left, bottom, right).
 2011-01-28 01:25:03 +01:00
Michael Stapelberg
5b6ef3e665 make the sighandler handle SIGABRT 2011-01-28 00:47:49 +01:00
Michael Stapelberg
f395c141c8 bugfix: initialize root variable earlier 2011-01-28 00:42:55 +01:00
Michael Stapelberg
7b01bc5eb7 Bugfix: use the global root variable, don’t get the first one (Thanks quaec) 
The case of an X11 server having multiple displays is handled correctly by the
code in src/mainx.c. However, due to some functions not being correctly
refactored and still getting the first screen (and also the first root window)
from the XCB connection, i3 was operating on the wrong root window.
 2011-01-28 00:41:53 +01:00
Michael Stapelberg
79bbde8766 add a flag to disable the signalhandler 2011-01-28 00:31:26 +01:00
Fernando Tarlá Cardoso Lemos
b29af954f6 Don't mess with the percentages in tree_flatten. 2011-01-28 00:12:26 +01:00
Fernando Tarlá Cardoso Lemos
5f4123f842 Fix some resizing issues (thanks mseed). 2011-01-27 23:42:37 +01:00
Michael Stapelberg
676afce540 bugfix: correctly move cons out of floating cons when the workspace has no other tiling cons (Thanks mseed) 2011-01-27 16:51:41 +01:00
Michael Stapelberg
334e41daa4 bugfix: don’t assume a workspace always has tiling cons when focusing (Thanks mseed) 2011-01-27 16:51:16 +01:00
Michael Stapelberg
3e08ceaff9 bugfix: resizing problem when resizing vertically on the top border (Thanks julien) 2011-01-27 16:26:19 +01:00
Michael Stapelberg
3383437705 refactor some places to use con_descend_focused instead of duplicating code 2011-01-27 16:09:14 +01:00
Michael Stapelberg
f462a9a215 re-insert floating cons next to the currently focused con of the appropriate workspace 2011-01-27 16:04:17 +01:00
Michael Stapelberg
2f5d111936 when re-inserting a floating con, start with a more fair percent value 2011-01-27 15:53:14 +01:00
Michael Stapelberg
fe851b85f0 RandR: respect primary output 2011-01-27 15:40:02 +01:00
Fernando Tarlá Cardoso Lemos
432563d6e7 Fix the percentages when moving containers. 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
07eb20851f Fix floating mode according to the new requirements. 
At all times any given non-leaf container should have the sum of
the percentages of its children == 1.0, otherwise we'll crash on
an assertion failure.
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
568cafd4ec Fix the resize algorithm I broke earlier. 
The reason it was broken was that it was ok for the sum of the
percentages to be something other than 1.0. Now this is no longer
the case, the sum of the percentages must always be 1.0 or an
assertion will fail when we render the containers.
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
bc82fc7e9f This parameter is no longer needed. 
The algorithm is now always the same, doesn't matter if we're
adding or removing a container to/from its parent.
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
45227fba54 A new logic to calculate the percentages. 
It's slower, but this way we make sure that the resulting
percentages *ALWAYS* sum up to 1.0 (or as close to that as we get
with double math).
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
a93f4643ec Only fix the percentages after we insert the container. 
This is what floating.c does and it allows us to unify the logic that
calculates those percentages.
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
89917976c7 Crash when we get the percentages wrong. 
Better to crash with an assertion than to get into an infinite loop.
We cold work around this, but there's a bug here and it's not a
rounding bug, so it's better not to conceal it.
 2011-01-27 13:00:14 +01:00
Fernando Tarlá Cardoso Lemos
485555ef72 Round up as well if needed (thanks Merovius). 2011-01-23 13:21:33 +01:00
Michael Stapelberg
d9bfd8843f Bugfix: fix restoring the orientation (Thanks Merovius/fernandotcl) 2011-01-21 22:58:22 +01:00
Michael Stapelberg
ad95d5bb1f bugfix: you can’t unfullscreen workspaces (Thanks Merovius) 2011-01-21 22:09:04 +01:00
Michael Stapelberg
cbf4fcb9b5 Bugfix: Fix focus problems when switching workspaces by pushing the window stack before mapping 2011-01-21 21:49:56 +01:00
Michael Stapelberg
81ff1f976d Bugfix: Don’t send fake configure notify with not yet rendered rect for floating windows 
This fixes a bug where opening the Xpdf find dialog when Xpdf is in fullscreen
mode would crash Xpdf due to a zero-width and zero-height ConfigureNotify rect.
 2011-01-21 21:01:02 +01:00
Michael Stapelberg
1fecbb3e5a Bugfix: also close empty split containers when the clients are moved away (Thanks mseed) 
Also update the testcase (which used only 'kill' before).
 2011-01-20 10:09:43 +01:00
Michael Stapelberg
3fe4146e24 Bugfix: fix crash in tree_flatten (Thanks mseed) 2011-01-19 09:31:31 +01:00
Michael Stapelberg
0eb5eb34cd When in stacking mode with only one child, respect border styles 1pixel and none (Thanks Merovius) 2011-01-17 14:38:16 +01:00
Michael Stapelberg
ae4331113e re-implement xkb support for detecting keyboard layout changes 2011-01-17 14:27:49 +01:00
Michael Stapelberg
d6d4c962f4 Bugfix: Call mark_unmapped() on floating nodes aswell (Thanks mseed) 
This fixes #292.
 2011-01-17 14:11:56 +01:00
Axel Wagner
92a038dd25 Save environment-variable AFTER reading the configfile 2011-01-12 10:16:11 +01:00
Axel Wagner
4caf85aa0b Use I3SOCK-environment-variable 2011-01-12 10:15:48 +01:00
Michael Stapelberg
cd2ee61ee8 fix some memory leaks when user passes command line arguments twice (Thanks Tiago) 2011-01-12 10:12:24 +01:00
Michael Stapelberg
a6f0dcd250 Fix switching to a workspace on a different output 2011-01-08 12:03:03 +01:00
Michael Stapelberg
83f6e445a0 Bugfix: Don’t use ->old_parent for floating cons (Thanks eelvex) 
Instead, we attach them to their workspace when toggling back to tiling. This
makes more sense; afterall, floating clients are always directly below a
CT_WORKSPACE container.
 2011-01-08 00:44:03 +01:00
Michael Stapelberg
0ea15ed962 fix a problem with workspace switching when the focus got to the target workspace 
This could explain some strange effects where workspaces would just stay blank.
We’ll see.
 2011-01-08 00:38:10 +01:00
Michael Stapelberg
54b9549713 Bugfix: Look for cons to focus *starting* at the ws, not beneath the ws (Thanks mseed) 
This should fix #286.
 2011-01-08 00:10:49 +01:00
Michael Stapelberg
e4bb6d859e Bugfix: Correctly maintain focus when setting a workspace to floating 2011-01-08 00:10:30 +01:00
Michael Stapelberg
b660769fe0 Bugfix: Correctly move to other workspaces when a floating window is focused on the target ws (Thanks mseed) 2011-01-07 23:56:32 +01:00
Michael Stapelberg
115462f103 Implement tree flattening to automatically solve situations of redundant chains of split containers 
This should fix the move problems. See comment of tree_flatten() for a little
example.
 2011-01-07 22:21:41 +01:00
Michael Stapelberg
228b5c51ff change many LOG/printf messages to use DLOG 2011-01-07 20:58:58 +01:00
Michael Stapelberg
186d2c7bfa ipc: change 'orientation' to human readable string instead of raw integer value 2011-01-07 20:48:01 +01:00
Michael Stapelberg
5ccd7b01e7 Bugfix: fix fullscreen mode for floating windows 2011-01-07 02:50:35 +01:00
Michael Stapelberg
55b6d31e4a Bugfix: randr: Don’t close container if it was not initialized before 2011-01-06 14:35:04 +01:00
Michael Stapelberg
2312187439 RandR: use the next unused workspace instead of fixed counting 2011-01-05 00:26:23 +01:00
Michael Stapelberg
f73c02ce92 Also initialize output->con when using neither RandR nor Xinerama 2011-01-05 00:19:51 +01:00
Michael Stapelberg
5098e45f23 Re-Implement support for RandR changes 2011-01-05 00:16:10 +01:00
Michael Stapelberg
f54ce1ddda retab! randr.c 2011-01-04 22:51:42 +01:00
Michael Stapelberg
1fb9b7c431 Bugfix: Correctly change focus after closing floating containers (Thanks litemotiv!) 2011-01-04 22:40:05 +01:00
Michael Stapelberg
0416be18df fix memleak: free struct Window members 2011-01-04 22:39:45 +01:00
Michael Stapelberg
bf2c18cc33 fix memleak: free X state structure 2011-01-04 22:39:24 +01:00
Michael Stapelberg
545566e6ba use sstrdup() instead of strdup() 2011-01-04 22:39:13 +01:00
Michael Stapelberg
80ecd157f6 fix memleak: free con->name before overwriting it 2011-01-04 22:38:33 +01:00
Michael Stapelberg
bdbda20293 cmdparse: free strings 2011-01-04 22:37:50 +01:00
Michael Stapelberg
fa44383cc6 fix small memory leak (unused ->name) 2011-01-04 22:15:52 +01:00
Michael Stapelberg
9713419327 Bugfix: Also change focus when the killed container was focused (Thanks fernandotcl) 2011-01-02 18:08:45 +01:00
Axel Wagner
aa422c07c4 Add forgetful restart to sighandler 2010-12-31 19:36:35 +01:00
Axel Wagner
6ec468ba1e Retab sighandler.c 2010-12-31 19:36:32 +01:00
Axel Wagner
fb6d117c42 Port sighandler to tree-branch 2010-12-31 19:36:29 +01:00
Michael Stapelberg
2d05c3a37d Fix dragging floating containers / click handling 2010-12-31 01:38:17 +01:00
Michael Stapelberg
daf00a932f For floating mode on workspace level, create a container around the content (Thanks Merovius) 
Like when setting a workspace to stacked, we need to create a new container
around the content and set *that* to floating.
 2010-12-30 23:09:18 +01:00
Michael Stapelberg
50914e0483 Bugfix: Correctly change focus when closing a split-container 
The problem was i3 leaving an invalid focus pointer valid (after killing the
container) because the container itself is not mapped (if it has no x11 window,
for example split containers).
 2010-12-30 23:01:58 +01:00
Michael Stapelberg
f65e4f5b16 Bugfix: Don’t redistribute resize percentage values when closing floating (!) windows (Thanks Merovius) (+testcase) 2010-12-30 02:39:14 +01:00
Michael Stapelberg
307c59bde6 Bugfix: Fix closing windows in tabbed mode with border_style == 1pixel / none 2010-12-28 19:54:56 +01:00
Michael Stapelberg
028f7d2ca7 Fix resize handling on click on borders (did not correctly use orientation) 2010-12-28 16:26:42 +01:00
Michael Stapelberg
d184587959 Bugfix: Don’t focus next window if the window was not mapped at the moment (on a different ws) 2010-12-28 02:27:11 +01:00
Michael Stapelberg
dc3c633ee4 Bugfix: store and properly load workspace order when restarting 2010-12-27 22:28:59 +01:00
Fernando Tarlá Cardoso Lemos
c88c3e3ab2 Default to a file in /tmp for the restart state. 
The file is now created in /tmp using the process PID and the
username of the user running i3. The restart state file is only
loaded when restarting (the --restart option is appended to the
command line prior to the restart). That means that renaming the
old state file with the ".old" extension is no longer needed.

This "--restart" switch is supposed to be only used by i3. The
"-L" switch can be used to load a layout (and not delete it
afterwards). We unlink the state file after we load it so that
we don't keep cruft in /tmp or try to restart from an old config
file if restart_state is set.
 2010-12-27 13:33:03 +01:00
Fernando Tarlá Cardoso Lemos
68f906f278 util.c is the proper place for those functions. 2010-12-27 13:28:17 +01:00
Fernando Tarlá Cardoso Lemos
3669bcbd5f Remove some commented out code. 2010-12-27 13:27:14 +01:00
Fernando Tarlá Cardoso Lemos
a1dd74da5a Implement default border styles (thanks litemotiv). 2010-12-27 13:10:45 +01:00
Michael Stapelberg
41eb810531 Bugfix: Don’t update focus when container is not mapped 2010-12-11 17:07:20 +01:00
Michael Stapelberg
8d5421e6a4 Bugfix: Invalidate focused_id to correctly focus new windows with the same ID 2010-12-11 17:03:53 +01:00
Fernando Tarlá Cardoso Lemos
d046fa446d Fix possible rounding errors. 2010-11-30 13:59:34 +01:00
Michael Stapelberg
3412e12602 xinerama.c: fix indention 2010-11-29 22:35:46 +01:00
Michael Stapelberg
a05b185701 xcb.c: fix indenting 2010-11-29 22:35:07 +01:00
Michael Stapelberg
1a40641462 workspace.{c,h}: remove obsolete code 2010-11-29 22:32:19 +01:00
Michael Stapelberg
5625a2f17f log.c: fix indenting 2010-11-29 22:28:23 +01:00
Michael Stapelberg
38b231b848 handlers.c: remove obsolete code 2010-11-29 22:25:55 +01:00
Michael Stapelberg
529bdf833f floating.c: remove obsolete code, fix indenting 2010-11-29 22:23:49 +01:00
Fernando Tarlá Cardoso Lemos
371ec037b8 Fix click to focus. 2010-11-29 22:09:05 +01:00
Michael Stapelberg
b0e41cb0a2 Bugfix: Don’t attach floating clients to 'nodes' when restoring (Thanks EelVex) 2010-11-29 21:46:00 +01:00
Michael Stapelberg
0f0d01336b Bugfix: Correctly attach floating Cons to a different workspace, extend testcase 2010-11-29 20:39:56 +01:00
Michael Stapelberg
ad825913b4 Bugfix: Fix crash when moving a floating Con to a different workspace, add testcase (Thanks EelVex) 2010-11-29 20:20:44 +01:00
Michael Stapelberg
b0068de3d7 Bugfix: Transfer 'percent' factor when splitting, add testcase for resizing 2010-11-29 11:24:12 +01:00
Michael Stapelberg
a0cd3c2bab Include 'percent' in tree JSON, use C-locale when dumping, update testcase 2010-11-29 11:10:17 +01:00
Michael Stapelberg
ee45c92564 Implement resizing (still buggy) 
Committing basic resizing functionality. We need testcases for the bugs and
then eliminate them.
 2010-11-29 10:56:16 +01:00
Michael Stapelberg
a86d8ab329 use con_num_children() 2010-11-28 22:45:39 +01:00
Michael Stapelberg
85b7e60bac Bugfix: Use separate child_mapped instead of checking state->initial (makes t/02-fullscreen.t pass) 
This is necessary for windows which are mapped later, for example when there is
a fullscreen window in front of everything.
 2010-11-28 22:09:26 +01:00
Michael Stapelberg
17caaf1159 Correctly update the _NET_WM_STATE hint when *not* going into fullscreen (when already in fullscreen) 2010-11-28 22:07:54 +01:00
Michael Stapelberg
49308d5026 Bugfix: don’t update focus when moving mouse over stacked/tabbed decorations (Thanks fernandotcl) 2010-11-28 21:07:08 +01:00
Michael Stapelberg
61e3415ddc Bugfix: Add deco_height to bsr.y (to configure the floatingcon correctly). Fixes t/12-floating-resize.t 2010-11-28 20:43:35 +01:00
Michael Stapelberg
2dba7ec1ec Bugfix for the last commit (broke some moving situations), update testcase 2010-11-28 20:15:47 +01:00
Michael Stapelberg
780b0ddbbc look & feel: when moving, descend if the container in target direction is a split-container 2010-11-28 18:35:11 +01:00
Michael Stapelberg
4fcd2f6e7e Bugfix: Fix focus when moving Cons 
When having two v-splits on a horizontal desktop:
----------------
|  t1   |  t3  |
|-------|------|
|  t2   |  t4  |
----------------
…focus is on t2, and you move it into the right v-split (move after h), the
focus was not properly updated. That is, inside the right v-split, focus was
correct, but the workspace focus was still pointing to the left v-split.
 2010-11-28 18:05:53 +01:00
Michael Stapelberg
4f1260ffe8 Also store/load window_rect when restarting 2010-11-28 17:20:29 +01:00
Michael Stapelberg
871da48b56 Bugfix: Only set to_focus when focused is actually 1 2010-11-28 17:20:16 +01:00
Michael Stapelberg
511cbec49b look and feel: when moving a Con is not possible, split a workspace level 
This allows you to open three cons, then move the last one to the right (like
in previous i3 releases).
 2010-11-28 14:45:14 +01:00
Michael Stapelberg
81044a7104 Correctly restore focus when restarting (Thanks fernandotcl) 2010-11-28 14:27:44 +01:00
Michael Stapelberg
f7fff5cec1 update indenting of src/util.c 2010-11-28 14:14:34 +01:00
Michael Stapelberg
32cc7134aa re-enable ipc_shutdown() when restarting 2010-11-28 14:11:44 +01:00
Michael Stapelberg
ab8400fff9 Bugfix: Use setsid() to avoid SIGINT for child processes 2010-11-28 14:03:43 +01:00
Michael Stapelberg
02b786509b Upon ConfigureRequests, send a generated ConfigureNotify (Thanks fernandotcl) 
This fixes problems with GVim in stacking mode for example.
 2010-11-28 13:19:49 +01:00
Michael Stapelberg
d60e8c56dc little style fixes, add vim modeline 2010-11-28 13:02:29 +01:00
Fernando Tarlá Cardoso Lemos
69fc6449dc libXcursor support (themed cursors). 2010-11-28 13:01:45 +01:00
Michael Stapelberg
3bab222aa7 Bugfix: Re-attach windows in correct order when switching layout (Thanks fernandotcl) 2010-11-28 01:51:16 +01:00
Michael Stapelberg
d0de3f403d Bugfix: Restore focus after changing layout (Thanks fernandotcl) 2010-11-28 01:39:47 +01:00
Michael Stapelberg
2c157283ea fix third argument to strncat(), use smalloc(), use strlen(".old")+1 2010-11-28 01:22:10 +01:00
Fernando Tarlá Cardoso Lemos
0a17fe973c Make the restart path configurable. 2010-11-28 01:18:10 +01:00
Michael Stapelberg
622b51a1ea Fix switching containers by moving the mouse over their decorations when in the same container 2010-11-28 01:13:18 +01:00
Michael Stapelberg
6fe0e58a64 Bugfix: Also render decorations of nearby cons when getting an ExposeEvent (Thanks fernandotcl) 2010-11-28 00:52:24 +01:00
Michael Stapelberg
0cfebcb5b6 remove some debug messages 2010-11-27 18:05:45 +01:00
Michael Stapelberg
2c3e5dbc65 Bugfix: unmap windows in a separate step to avoid focus problems with fullscreen windows 
This fixes an ugly bug with Adobe Flash in fullscreen mode, for example on
YouTube. See comments in the diff for some explanation.
 2010-11-27 17:45:23 +01:00
Michael Stapelberg
d47a1edf22 Bugfix: configure windows before mapping, correctly store window_rect instead of rect 2010-11-27 16:44:45 +01:00
Michael Stapelberg
3f3fa08b15 more debug for ClientMessages 2010-11-27 16:44:32 +01:00
Michael Stapelberg
6431d3d187 fix indention 2010-11-27 16:44:20 +01:00
Fernando Tarlá Cardoso Lemos
1d52bf179b Run the autostart commands. 2010-11-26 23:30:27 +01:00
Fernando Tarlá Cardoso Lemos
2faac65237 Always add to the focus list, fixes crash. 2010-11-26 23:26:53 +01:00
Michael Stapelberg
77d0d42ed2 look and feel: create split container when switching workspace layout 
Quote from the source:
  When the container type is CT_WORKSPACE, the user wants to change the
  whole workspace into stacked/tabbed mode. To do this and still allow
  intuitive operations (like level-up and then opening a new window), we
  need to create a new split container. */
 2010-11-26 23:08:12 +01:00
Michael Stapelberg
c3981e12d3 Bugfix: use bufcopy instead of buf (Thanks fernando) 2010-11-26 22:38:05 +01:00
Michael Stapelberg
65bd71b213 remove some obsolete code 2010-11-26 22:15:09 +01:00
Michael Stapelberg
b2db9ac797 Bugfix: always reset state->initial to false, not only for different stacking order 
As initial may be set to true again after initializing the window, we need to
ensure that it is properly cleared. Otherwise, this leads to ghost windows when
unmapping (such as the Xpdf about dialog) due to i3 issuing MapWindow after an
Unmap happened but before i3 actually received/handled the UnmapNotify.
 2010-11-26 20:15:08 +01:00
Michael Stapelberg
97bc8f4b86 Bugfix: Don’t draw decorations of CT_FLOATING_CONs 2010-11-26 19:24:14 +01:00
Michael Stapelberg
db3002fce0 Implement click on stack/tab decoration to focus 2010-11-26 18:57:15 +01:00
Michael Stapelberg
df7788386d re-indent click function 2010-11-26 18:48:27 +01:00
Michael Stapelberg
962fe075b9 Bugfix: Upon ExposEvents, redraw decoration also for the window itself 
In the meantime, windows can have decorations (borders) on their own frame, too.
 2010-11-26 18:37:34 +01:00
Michael Stapelberg
e744b48b30 Bugfix: When handling an EnterNotify for a child window, access the correct con->layout 2010-11-26 18:18:40 +01:00
Michael Stapelberg
4d7c24b92c Bugfix: Correctly count variables when parsing the configfile (Thanks dbp) 2010-11-26 11:25:51 +01:00
Michael Stapelberg
1de97a1f1f correctly sort numbered workspaces (+testcase) 
Numbered workspaces (workspaces with a name containing only digits) will be
inserted in the correct order now. Named workspaces are always sorted after
numbered workspaces and in the order of creation.
 2010-11-21 23:35:49 +01:00
Michael Stapelberg
fab8b84db7 ipc: fix current_workspace 2010-11-21 22:12:34 +01:00
Michael Stapelberg
a71d782da9 fix indention of src/ipc.c 2010-11-21 22:03:55 +01:00
Fernando Tarlá Cardoso Lemos
ae22fe065f Send change:empty too. 2010-11-21 21:36:33 +01:00
Fernando Tarlá Cardoso Lemos
e2517bd48c Unlink the resolved path. 2010-11-21 21:36:30 +01:00
Fernando Tarlá Cardoso Lemos
767dcea7ba Send the workspace focus change event. 2010-11-21 21:36:28 +01:00
Fernando Tarlá Cardoso Lemos
5d830e7a27 Ported over some message types from -next. 2010-11-21 21:36:19 +01:00
Michael Stapelberg
2d280469af Bugfix: Don’t draw borders for fullscreen windows 2010-11-21 17:00:10 +01:00
Michael Stapelberg
09b5b17830 Bugfix: Don’t attach tiling containers to floating containers 
This bug happened when there were only floating containers on a workspace and a
new tiling window was to be opened.
 2010-11-21 16:49:59 +01:00
Michael Stapelberg
f53fafe100 ipc: s/floating-nodes/floating_nodes for consistency 2010-11-21 16:34:45 +01:00
Michael Stapelberg
39fa1d724a bugfix: don’t treat workspace as empty if they only have floating windows (+testcase) 2010-11-20 20:16:15 +01:00
Michael Stapelberg
db651679c5 Bugfix: Properly ignore UnmapNotify events (especially for floating windows) 
This fixes the bug which caused floating windows to be visible even when
switching to a different workspace.

Instead of ignoring a specific sequence, we now set an ignore_unmap counter for
each container. (So, should containers be closed too early or stay open even if
they should be closed, we probably need to have a closer look at the counter.
At the moment, it is increased by one on reparenting and unmapping (for
workspace changes) and decremented by one on each UnmapNotify event).

This system is better because a sequence does not describe a single unmap or
reparent request but a request to X11 on the network layer -- which can contain
multiple requests.
 2010-11-20 19:11:43 +01:00
Michael Stapelberg
131a2f8765 debugging: implement a NOP command which just spits out its argument 
You can use this in testcases to mark specific sections:
  $i3->command('nop before trying to crash')->recv;
leads to the following output in the i3 logfile:
  -------------------------------------------------
    NOP: before trying to crash
  -------------------------------------------------
 2010-11-20 18:40:25 +01:00
Michael Stapelberg
576de246d8 Bugfix: when no windows to focus, return the workspace instead of NULL 2010-11-20 18:38:24 +01:00
Fernando Tarlá Cardoso Lemos
5ebb3b4832 Port the IPC socket creation code from -next. 2010-11-15 14:00:48 +01:00
Fernando Tarlá Cardoso Lemos
bfa12a5819 Port the path resolution and config loading code from -next. 2010-11-15 14:00:46 +01:00
Michael Stapelberg
c0c7d04264 When in tabbed mode, nail the orientation to horizontal 
This is intuitive for the user as he can use the left/right keys to switch.
Also, we need to nail the orientation because you can be in either vertical or
horizontal mode when you enter the tabbed layout (like with the stacking
layout).
 2010-11-15 13:55:10 +01:00
Fernando Tarlá Cardoso Lemos
8048cb2e12 Render the tabbed mode correctly. 2010-11-15 12:27:16 +01:00
Fernando Tarlá Cardoso Lemos
ea1e9b20be Get rid of the remaining hardcoded height. 2010-11-15 12:25:06 +01:00
Fernando Tarlá Cardoso Lemos
69f29b2b8e Use the configured font to draw the decorations. 2010-11-15 12:24:16 +01:00
Michael Stapelberg
cbd53e8a7f bugfix: check if the client leader is a managed window (Thanks fernandotcl) 2010-11-14 23:55:53 +01:00
Michael Stapelberg
76c07900c2 take into account x11 border_width settings (fixes uxterm border issue) 2010-11-14 23:44:13 +01:00
Michael Stapelberg
39b378b0a4 don’t allow useless splits, change orientation of existing split container instead 2010-11-14 23:18:39 +01:00
Michael Stapelberg
e8b5a802e2 Bugfix: only set clients to floating which have a leader that is not their own window (Thanks fernandotcl) 2010-11-14 22:49:05 +01:00
Michael Stapelberg
a27af52706 fix floating focus behaviour, extend testcase 2010-11-14 22:35:44 +01:00
Michael Stapelberg
d401703135 Bugfix: use rectangles instead of a polygon to avoid strange rendering errors 
With the polygon, when using pidgin, having the buddy list in the middle of the
screen, 200 px width, full screen high, then opening the manage accounts
window, the decorations of the buddy list were visible on the accounts window.
 2010-11-14 21:17:27 +01:00
Michael Stapelberg
33eb00d6ae automatically set windows with client_leader to floating 2010-11-14 20:15:51 +01:00
Michael Stapelberg
055bd18142 Bugfix: after the first UnmapNotify, unignore the event 2010-11-14 20:15:12 +01:00
Michael Stapelberg
d760a1c7b2 Bugfix: don’t kill parent when currently in tree_close() for a child of this parent 2010-11-14 20:14:09 +01:00
Michael Stapelberg
f0efb3737e don’t remove floating container twice (it’s already removed in con_detach) 2010-11-14 20:11:46 +01:00
Michael Stapelberg
5d7344af8a resize/unmap container x11 windows on demand (makes background images visible again) 2010-11-14 18:52:40 +01:00
Michael Stapelberg
945632ddcb Implement setting the WM_NAME of i3 container windows for debugging 2010-11-14 16:41:46 +01:00
Michael Stapelberg
e85bb09017 set withdrawn/normal state when unmapping/mapping (for xprop/java) 2010-11-14 13:53:47 +01:00
Michael Stapelberg
a415d56048 parser: return a proper JSON reply on parse errors 2010-11-14 01:45:05 +01:00
Michael Stapelberg
7c6f2dbfc6 Rendering fixes for stacking mode 2010-11-13 22:39:59 +01:00
Michael Stapelberg
4cd6dd0303 port fernando’s custom background color patch 2010-11-13 20:07:49 +01:00
Fernando Tarlá Cardoso Lemos
53b98fdc7e Paint the window decorations using the theme. 2010-11-13 19:17:08 +01:00
Michael Stapelberg
dc10c67060 Bugfix: Close containers which are empty due to a move (Thanks fernando) 2010-11-13 14:55:11 +01:00
Michael Stapelberg
432073dbe5 implement support for WM_TRANSIENT_FOR, expand testcase 2010-11-13 01:19:21 +01:00
Michael Stapelberg
ad9be5402a Implement support for WM_CLIENT_LEADER 2010-11-12 23:46:03 +01:00
Baptiste Daroussin
1bb6906c56 Fix build on freebsd (uint32_t unknown if stdint.h not 
included)
 2010-11-12 23:11:36 +01:00
Michael Stapelberg
8f7bd538d8 implement configure requests, adapt testcase 
testcase does not pass 100% due to clients not being reparented correctly yet.
 2010-11-12 21:41:10 +01:00
Michael Stapelberg
0723876429 only re-position floating clients mapped to (0, 0), add pixels for decoration 2010-11-12 20:55:43 +01:00
Michael Stapelberg
57e602a97c respect position in geometry of floating windows 2010-11-12 20:55:26 +01:00
Michael Stapelberg
e07fee4472 parser: don’t exit(1) on invalid command, use better error message 2010-11-12 20:22:55 +01:00
Michael Stapelberg
0e264cb5c4 implement different border styles 
Wow, that actually was easy :).
 2010-11-12 19:37:43 +01:00
Michael Stapelberg
66dc8883fd make borders around every container configurable for debugging 2010-11-12 19:37:32 +01:00
Michael Stapelberg
d248f35267 include border style in ipc tree dump 2010-11-12 19:16:38 +01:00
Michael Stapelberg
ea4e3e7682 parser: bugfix: initialize match when parsing new command 2010-11-12 19:07:48 +01:00
Michael Stapelberg
6eb7f2a01d lexer/parser: implement 'border' command 2010-11-12 18:41:54 +01:00
Michael Stapelberg
c6c084faa5 don’t close workspaces which are still visible (multi-monitor) 2010-11-12 17:34:13 +01:00
Michael Stapelberg
c1d574f84e enumerate workspaces when initializing outputs 2010-11-12 17:33:59 +01:00
Michael Stapelberg
5c2758af26 Implement support for size hints (including test case) 2010-10-11 21:32:29 +02:00
Michael Stapelberg
0925e8b7dc Implement sticky windows 
The implementation works like this:
Containers can have a 'sticky-group' attribute. Imagine two different
containers (on two different workspaces) which have the same sticky-group.
Now you open a window in the first container. When you switch to the
other workspace, the window will be re-assigned to the other container.

An obvious problem which is not covered with the code at the moment is
having two containers with the same sticky-group visible at the same time.
 2010-09-01 18:11:01 +02:00
Michael Stapelberg
4d12e18571 remove proof-of-concept code in main.c 2010-08-21 18:36:51 +02:00
Lourens Rozema
8159ffaa06 Bugfix for the kill command (used a.o. in test 18). Call to match_init is to be used i.o. a simple memset() to zero. Otherwise the boolean dock field doesn't get initalized to -1. 2010-08-21 18:28:37 +02:00