Commit Graph

6863 Commits

Author SHA1 Message Date
Ingo Bürk cf505eaea8
Merge pull request #3983 from orestisfl/resize-behind-fullscreen
Avoid resizing fullscreen container
2020-03-31 09:52:46 +02:00
Orestis Floros 1a2882d740
Avoid resizing fullscreen container
Fixes #3980
2020-03-31 08:52:53 +02:00
Ingo Bürk 164336099d
Merge pull request #3970 from ianyfan/ipc
ipc: always include marks property in TREE reply
2020-03-09 13:01:11 +01:00
Ian Fan d02cda4241 ipc: always include marks property in TREE reply 2020-03-07 11:12:47 +00:00
Ingo Bürk cae3b294ad
Merge pull request #3964 from ammgws/_PATH_BSHELL
Use _PATH_BSHELL in nagbar script as well
2020-03-01 08:28:28 +01:00
Jason Nader 9a3318b622
Fix error message 2020-03-01 13:25:42 +09:00
Jason Nader 191c394db8
Use _PATH_BSHELL in nagbar script as well 2020-03-01 12:46:30 +09:00
Ingo Bürk d2acdcc69f
Merge pull request #3960 from ammgws/patch-2
Remove comment referencing old source code
2020-03-01 01:29:30 +01:00
Ingo Bürk daa2ea0d05
Merge pull request #3961 from ammgws/patch-1
i3-nagbar: Use _PATH_BSHELL
2020-03-01 01:28:57 +01:00
Jason 5024a13b8b
Remove comment referencing old source code
Behaviour was changed in f691a55923
2020-03-01 05:03:41 +09:00
Jason a516bdbb92
Use _PATH_BSHELL
Possibly overlooked in f691a55923850a4d315450925fc98733d07b69c9?
2020-03-01 04:53:35 +09:00
xzfc 47be36410c
Assume xcb_cursor_context_new never fails (#3955)
According to libxcb-cursor code, the only condition in which
xcb_cursor_context_new() returns a non-zero result is a memory
allocation failure[1].  Thus, it is safe to assume that
xcursor_supported is always true, and remove dead code.

[1]: https://gitlab.freedesktop.org/xorg/lib/libxcb-cursor/blob/0.1.3/cursor/cursor.c#L131-132
2020-02-24 08:48:58 +01:00
Michael Stapelberg 0ba325c5f3
Makefile.am: respect configure --program-suffix (#3944)
related to #3838
2020-02-22 13:47:49 +01:00
Joseph 967ec2e0ea
Fix test case 180-fd-leaks when running on Fedora (#3911) 2020-02-22 10:09:11 +01:00
Orestis Floros a376d1e52f
Merge pull request #3950 from xzfc/small-fixes
Small fixes
2020-02-21 19:01:40 +01:00
Ingo Bürk c5d50ac5fd
Merge pull request #3951 from txtor/next
typo
2020-02-21 10:59:35 +01:00
Francesc Hervada-Sala 80f9bb6dbe typo 2020-02-21 10:53:02 +01:00
Albert Safin b98b055459 i3 --moreversion: erase the line before writing over
The trailing part of the line (`abort…)`) has often been appearing in
bug reports.
2020-02-21 02:07:23 +00:00
Albert Safin 83c7aff089 Limit workspace numbers within 0..INT32_MAX
Before this commit, large workspace numbers treated oddly:

   $ i3-msg 'rename workspace to 1234567890'
   # displayed in i3bar as `0`

   $ i3-msg 'rename workspace to 4294967200'
   $ i3-msg -t get_workspaces | jq '.[]|select(.focused).num'
   -96 # int32_t overflow

   $ i3-msg 'rename workspace to 99999999999999999999'
   $ i3-msg -t get_workspaces | jq '.[]|select(.focused).num'
   -1 # treated as unnumbered

This commit puts a consistent limit on workspace numbers.  Now
workspaces with numbers beyond INT32_MAX are treated as unnumbered.
2020-02-21 02:07:04 +00:00
Albert Safin d3976fee8c Code style: fix misaligned and trailing whitespaces 2020-02-21 02:06:48 +00:00
Orestis Floros 854885ea9d
Merge pull request #3949 from xzfc/overlap
Sanitize window dimensions and handle decoration overlap
2020-02-19 11:41:58 +01:00
Albert Safin 670c23600f
Decoration click/mouse move: iterate children in reverse order
In the case of decoration overlap, we should handle the one that is
drawn on top, i.e., the last one.
2020-02-19 11:33:49 +01:00
Albert Safin e6ca7ca06f
Sanitize con and window rect dimensions
Make sure they're neither zero (prohibited by X11) nor overflown during
subtraction.
2020-02-19 11:33:49 +01:00
xzfc 1f0c628cde
clang-format: bring back ForeachMacros (#3948)
* clang-format: bring back ForeachMacros

ForeachMacros was disabled in 4211274fcd
due to the breakage of include/queue.h. The currently used version,
clang-format-6.0 doesn't break it.

* Add curly braces

Co-authored-by: Orestis Floros <orestisflo@gmail.com>
2020-02-19 11:31:09 +01:00
Ingo Bürk e3f120c0b6
Merge pull request #3946 from phy1729/i3bar_command-exec
Prepend "exec " to default i3bar_command
2020-02-18 09:16:27 +01:00
Ingo Bürk 13ca970bfe
Merge pull request #3945 from stapelberg/deps
Makefile.am: i3-config-wizard: add dependency on config parser
2020-02-18 09:16:06 +01:00
Michael Stapelberg 12095103fa Makefile.am: i3-config-wizard: add dependency on config parser
fixes #3670
2020-02-17 21:49:47 +01:00
Michael Stapelberg 30e886b031 debian: update changelog 2020-02-17 18:31:40 +01:00
Michael Stapelberg 0dd60756af Merge branch 'release-4.18' 2020-02-17 18:27:46 +01:00
Michael Stapelberg 949aff3016 Set non-git version to 4.18-non-git. 2020-02-17 18:27:46 +01:00
Michael Stapelberg 9929949d26 release i3 4.18 2020-02-17 18:27:38 +01:00
Ingo Bürk 2914c7fb43
Merge pull request #3942 from stapelberg/iconv
configure.ac: test for iconv_open with #include <iconv.h>
2020-02-16 18:50:54 +01:00
Michael Stapelberg 91b00c5605 configure.ac: test for iconv_open with #include <iconv.h>
The previously used AC_SEARCH_LIBS uses AC_LANG_CALL, which is discouraged
because it intentionally violates the C type system: it has no way of specifying
function parameters, so it does not include the required headers to suppress
type warnings:
https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.69/html_node/Generating-Sources.html

The new code does what AC_SEARCH_LIBS does, but uses AC_LANG_PROGRAM instead of
AC_LANG_CALL. It explicitly includes iconv.h and calls iconv_open with the
correct number and type of arguments.

This fixes compilation on systems where libiconv’s iconv.h is discovered first,
but glibc also provides iconv. Previously, this would result in configure
discovering glibc’s iconv, and make failing to compile because -liconv wasn’t
specified, but iconv.h pulled in libiconv.
2020-02-16 18:38:44 +01:00
Michael Stapelberg f517b5aa57
Merge pull request #3918 from orestisfl/fno-common
Fix fno-common problems with gcc10
2020-02-15 12:25:58 +01:00
Ingo Bürk c3bf7288b3
Merge pull request #3928 from Iskustvo/improve_window_properties_documentation
Improved documentation for "window_properties" field in GET_TREE reply.
2020-02-11 09:06:55 +01:00
Iskustvo 36a8bc6463 Improved documentation for "window_properties" field in GET_TREE reply. 2020-02-08 11:41:48 +01:00
Ingo Bürk 4338c3ea55
Merge pull request #3931 from acheronfail/chore/add-docs-for-window-type
docs: add window_type to ipc documentation
2020-02-07 10:44:42 +01:00
acheronfail a1736f4df6 docs: add window_type to ipc documentation 2020-02-07 19:50:23 +11:00
Orestis Floros b9b0774733
Merge pull request #3926 from stapelberg/next
configure: deal with git worktree checkouts, where .git is a file
2020-02-02 13:12:20 +01:00
Michael Stapelberg 2edae71570 configure: deal with git worktree checkouts, where .git is a file
fixes #3013
2020-02-02 11:17:02 +01:00
Orestis Floros cb1012e608
Travis: Default to -fno-common
See #3914
2020-02-01 17:41:19 +01:00
Orestis Floros 2eac53fada
atoms: Properly declare as global variables
See #3914
2020-02-01 17:34:01 +01:00
Orestis Floros d14e59b3a7
i3bar: Make header declarations extern
See #3914
2020-02-01 17:33:53 +01:00
Orestis Floros e9f011397a
libi3: Make visual_type extern
See #3914
2020-02-01 17:32:54 +01:00
Orestis Floros d11e862919
Delete duplicate definition of ewmh_window
See #3914
2020-02-01 17:32:53 +01:00
Matthew Martin a2cba79f9c Prepend "exec " to default i3bar_command
Avoids leaving around a useless shell process.
2020-01-30 21:02:50 -06:00
Ingo Bürk fc01ece365
Merge pull request #3909 from orestisfl/bug-3903
Do not error on exec not matching a container
2020-01-18 17:42:17 +01:00
Orestis Floros f76b11327f
Do not error on exec not matching a container
See #3903, #3905
2020-01-18 11:09:14 +01:00
Ingo Bürk 110bdcbee5
Merge pull request #3905 from Airblader/bug-3903
Respect match criteria for exec command
2020-01-16 22:19:25 +01:00
Ingo Bürk f002584509 Respect match criteria for exec command
We currently do not evaluate match criteria for the exec command
since generally executing the same command multiple times is
unlikely to make sense.

However, it does make sense when the match is empty and this should
prevent the command from running, which currently does not happen.

For consisteny we execute the command as many times as there are
matched criteria, but print a warning if it matches more than one
container.

fixes #3903
2020-01-16 22:07:37 +01:00