Commit Graph

6704 Commits

Author SHA1 Message Date
Ingo Bürk 730264d9b2
Merge pull request #3065 from orestisf1993/issue-3064
Skip internal workspaces with 'move workspace to output'
2017-12-03 18:46:34 +01:00
Orestis Floros e3f8939a58 Skip internal workspaces with 'move workspace to output'
Fixes #3064.
2017-12-01 17:11:32 +02:00
Ingo Bürk f6f198c3fe
Merge pull request #3058 from d-e-s-o/fix-FREE
do not check for NULL in FREE macro
2017-11-29 08:33:27 +01:00
Daniel Mueller 3e34122de4 Perform proper cleanup for signals with 'Term' action (#3057)
Issue #3049 describes a case where terminating i3 by means of SIGTERM
causes it to leak the runtime directory and all its contents. There are
multiple issues at play: first, any cleanup handlers registered via
atexit are never invoked when a signal terminates the program (see
atexit(3)). Hence, the log SHM log cleanup performed in i3_exit is not
invoked in that case. Second, compared to the shutdown path for the
'exit' command, we do not unlink the UNIX domain socket we create,
causing it to be leaked as well. Third, a handler for SIGTERM is not
registered at all despite handle_signal claiming to be the handler for
all 'Term' signals.
This change addresses all three problems and results in a graceful exit
including cleanup to happen when we receive a signal with the default
action 'Term'. It addresses issue #3049.
2017-11-28 23:29:47 -08:00
Ingo Bürk e4d6458cc3
Merge pull request #3052 from d-e-s-o/remove-magic-numbers
i3bar: replace magic numbers with more meaningful constructs
2017-11-29 08:22:51 +01:00
Michael Stapelberg f9efc09b98
Merge pull request #3059 from stapelberg/debuglog
i3-dump-log: enable shmlog on demand
2017-11-26 23:34:10 -08:00
Michael Stapelberg 02b237b14d shmlog: remote atom when disabled 2017-11-26 18:07:13 +01:00
Michael Stapelberg 1facb450c0 i3-dump-log: enable shmlog on demand
fixes #3055
2017-11-26 18:07:02 +01:00
Michael Stapelberg c07936d91b no-op refactoring: make ipc_connect find socket path 2017-11-26 16:41:59 +01:00
Daniel Mueller 865bd462b4
do not check for NULL in FREE macro
free(3) is safe to invoke on a NULL pointer, in which case no action is
taken. This change adjusts the FREE macros to omit this unnecessary
check.
2017-11-23 15:41:33 -08:00
Ingo Bürk 2eabfc88fe
Merge pull request #3050 from d-e-s-o/fix-typo
fix typo in src/main.c
2017-11-18 19:19:25 +01:00
Daniel Mueller 2481301dfc
fix typo in src/main.c 2017-11-18 09:49:41 -08:00
Daniel Mueller 222d8210ab
i3bar: replace magic numbers with more meaningful constructs
In i3bar/src/config.c we compare string lengths agains magic numbers.
This change replaces those numbers with the lengths of the strings they
represent.
2017-11-17 06:58:12 -08:00
Ingo Bürk d1f8266df0
Merge pull request #3041 from d-e-s-o/update-userguide
docs/userguide: Correct mark/goto i3-input commands
2017-11-17 15:09:50 +01:00
Ingo Bürk a78c9ab82f
Merge pull request #3040 from jchook/next
$mod+r toggles resize mode
2017-11-13 22:01:06 +01:00
Ingo Bürk 0617b9f2a7
Merge pull request #3042 from Mayeu/kitty-terminal
add the kitty terminal in i3-sensible-terminal
2017-11-12 13:38:52 +01:00
Cast 22ba46e2a3 add the kitty terminal in i3-sensible-terminal 2017-11-12 16:48:21 +08:00
Daniel Mueller 3a22c6e765
docs/userguide: Correct mark/goto i3-input commands
The userguide still mentions an old 'goto' command which no longer
exists and will be ignored silently (when used in the i3 config) or
causes an error to be reported (when invoked from the command line).
This change updates the userguide to correct this problem. In addition
to that it also updates the i3-input command shown to no longer use the
deprecated -p flag but -F instead.
2017-11-11 17:06:43 -08:00
xzfc 103e78e04a Bugfix: avert endless loop on unexpected EOF at ipc messages (#3021)
Fix freeze on invalid ipc commands like

    echo -n $'i3-ipc\0\0\0\xa\0\0\0\0focus left' | socat - `i3 --get-socketpath`

Also, treat incomplete headers as IPC violation. Example of incomplete header:

    echo -n i3-ip | socat - `i3 --get-socketpath`
2017-11-09 11:18:23 -08:00
Ingo Bürk c207921949
Merge pull request #2980 from orestisf1993/issue-1160-pr
Use con_descend_focused for workspaces in _tree_next
2017-11-05 17:42:31 +01:00
Ingo Bürk b18f507b9b
Merge pull request #3030 from tylerbrazier/next
config: use ascii single quote
2017-11-05 17:37:53 +01:00
Ingo Bürk 01029da840
Merge pull request #3023 from orestisf1993/issue-2816
Add support to resize floating container in percentage
2017-11-05 17:27:48 +01:00
Wes Roberts 0afe714e2b $mod+r toggles resize mode 2017-10-30 21:19:12 -04:00
Tyler Brazier e0287e00d9 config: use ascii single quote
The unicode quote doesn’t render very well in terminals that don't
support unicode
2017-10-25 15:26:44 -05:00
Michael Stapelberg 787e799e3c Merge pull request #3028 from stapelberg/flakes2
Eliminate causes of flakyness
2017-10-25 08:56:02 +02:00
Michael Stapelberg 0d8b6714e3 Move XCB event handling into xcb_prepare_cb.
Previously, we used ev_check watchers, which are executed at the beginning of an
event loop iteration.

This was problematic if one of the handlers happened to fill the XCB event
queue, e.g. by reading a reply from X11 and an event happened in the meantime.

In that situation, we would hand control to the event loop, entirely ignoring
the pending event. This would manifest itself as a 1-minute hang,
reproducible (sometimes) in the i3 testsuite.

issue #2790 describes an instance of this issue in i3bar, and we fixed that by
changing the watcher priority to run last. Handling events in xcb_prepare_cb has
the same effect, as ev_prepare watchers are run just before the event loop goes
to sleep.
2017-10-23 21:06:13 +02:00
Michael Stapelberg 1946cc6cab t/525: remove flaky focus check 2017-10-23 21:06:13 +02:00
Michael Stapelberg 7a672a9d41 i3test: blockingly wait for events
…as polling the xcb file descriptor directly is not reliable.
2017-10-23 21:06:07 +02:00
Michael Stapelberg 8e528d2de8 skip ConfigureNotify events with --force_xinerama
This prevents an i3 crash under certain conditions when running the tests.
2017-10-23 21:04:20 +02:00
Michael Stapelberg 7208df2d38 i3test: reliably warp the pointer to (0, 0) before tests start 2017-10-23 21:04:20 +02:00
Michael Stapelberg 5d55f93eb3 tests: add sync_with_i3 after open_window
We need to set dont_map => 1 on the sync window to prevent an endless loop.

Further, t/219-ipc-window-focus.t made assumptions about windows being named
incrementally, and that assumption is broken by the sync window opened by the
first sync_with_i3 call from open_window, so use the more reliable ->name.
2017-10-23 21:04:20 +02:00
Michael Stapelberg 4f751610c2 _NET_ACTIVE_WINDOW: invalidate focus to force SetInputFocus call (#3027)
The sender of the _NET_ACTIVE_WINDOW client message might know better when to
set focus than i3, as i3 does not know about unmanaged (override_redirect=1)
windows.

related to https://github.com/i3/i3lock/issues/35
2017-10-22 22:16:15 +02:00
Orestis Floros b17e7b82c6 Add support to resize floating container in percentage
resize set is modified to accept both 'px' and 'ppt' height and width.

Fixes #2816.
2017-10-18 02:07:11 +03:00
Orestis Floros a05663c59e i3bar: avoid possible va_args leak 2017-10-17 04:28:07 +03:00
Orestis Floros a7bae0341d i3bar: free fontname 2017-10-17 04:28:07 +03:00
Ingo Bürk d4eaea8289 Merge pull request #2849 from Streetwalrus/next
Create a new split container when switching a workspace container to split layout
2017-10-15 22:23:40 +02:00
Michael Stapelberg 906a88e92b docs: full-size, so-called (#3019)
As per Michael Siegel, who studied the English language :).
2017-10-15 14:22:41 +02:00
Michael Stapelberg 4f03e49c3a userguide: explain why Mod4 is usually preferred (#3018) 2017-10-15 12:17:22 +02:00
Dan Elkouby 26014ca1a2 Default to L_SPLITH with toggle split when last_split_layout hasn't been initialized 2017-10-14 12:27:56 +03:00
hwangcc23 a34a0048a1 Add regression tests for #2846
1). Add one regression test in 167-workspace_layout.t:
    - Get a fresh workspace
    - Set the layout to something
    - Create windows
    - Try to switch to another layout
    - Check if successful
    - Repeat for all 12 possible transitions
2). Add another regression test in 167-workspace_layout.t:
    - Check that the command 'layout toggle split' works regardless of
      what layout we're using
2017-10-14 12:27:32 +03:00
Dan Elkouby c028f0cb17 Create a new split container when switching a workspace container to split layout
The behavior before 52ce8c8 was to do it regardless of what layout we're
switching to.

Fixes #2846
2017-10-14 12:26:47 +03:00
Michael Stapelberg 44a6efb5b0 tests: run under Xvfb by default (if available) (#2951)
This shaves off two seconds of wall-clock time (10s → 8s).
2017-10-13 00:18:49 -07:00
Michael Stapelberg 28ca1e8529 Fix AnyEvent->timer call (#3008)
And make the remaining AnyEvent->condvar and AnyEvent->timer calls
consistent.
2017-10-05 13:03:33 -07:00
Orestis 464c158d9a dump-asy.pl: use layout instead of orientation for names (#3004) 2017-09-30 10:17:20 -07:00
Michael Stapelberg ffa228e653 Merge pull request #3005 from stapelberg/tick
Implement the tick event
2017-09-30 10:16:43 -07:00
Michael Stapelberg 21cdcdb07c Fix compilation warnings on all Debian architectures. (#3007)
stbuf.st_size is of type off_t, which the standard defines as “extended signed
integral type”¹, and for which there is no correct printf format string. Hence,
we need to cast it into a hopefully-large-enough type (ugh) and use the
corresponding format string. In our case, int64_t should do it, as config files
really shouldn’t be anywhere close to those numbers.

① http://pubs.opengroup.org/onlinepubs/007908799/xsh/systypes.h.html
2017-09-30 10:16:21 -07:00
Michael Stapelberg 962750eb64 Fix memory leak when config conversion fails (#3006)
This happens on an empty config file, for example.
2017-09-30 10:15:58 -07:00
Michael Stapelberg 2bff4f1dba Synchronize with i3bar+i3, not just i3. 2017-09-30 13:04:20 +02:00
Michael Stapelberg ce21de8dde Implement the tick event
This makes our tests less flaky, shorter, and more readable.

fixes #2988
2017-09-30 11:41:57 +02:00
Michael Stapelberg 14c8cf8622 tests: implement xtest_sync_with_i3
The regular sync_with_i3 is not sufficient because i3test::XTEST uses a separate
X11 connection.
2017-09-30 11:28:20 +02:00