- Split cmd_bar into 2 functions, improving errors and reducing strcmps
- Only update barconfig when something has changed
- Only update barconfig for the specific bar that has changed
Fixes#3958
- Includes updated bigpicture.png from dump-asy.pl script
- The X11 root window is not a container
- Adds some extra information
bigpicture.png is created by converting the .asy to an .eps with `asy
bigpicture .eps` and then using the following gs command:
gs -dSAFER -dBATCH -dNOPAUSE -dEPSCrop -r600 -sDEVICE=pngalpha -sOutputFile=bigpicture.png bigpicture.eps
Instead of sorting alphanumerically, use a more opinionated order,
listing more "important" files towards the beginning.
Info in this section was a bit outdated. First of all, i3 contains many
more files that were not mentioned. Instead of trying to include
everything, I deleted some files with very obvious descriptions. We can
always re-add something if we want to add more details to it.
This still had some leftovers from the patch era. Since git and
specifically GitHub-like developing is much more mainstream now we don't
need to link introductions or even mention the idea of "patching".
The deleted "them" was referencing an old sentence referring to patches,
generated by the git cli. As emailing patches is not common at all for
GitHub repos, I removed the sentence altogether.
Also simplifies the 'branches' subsection a bit. Asking people to verify
their patch on master seems too much preliminary work and I doubt that
anyone does it anyway.
The previous fix when using _DEPENDENCIES was wrong because that
dependency is only created for the final executable. However, the build
fails when building the object file. The manual explicitly mentions that
using _DEPENDENCIES is wrong for source files:
> In rare cases you may need to add other kinds of files such as linker
> scripts, but listing a source file in _DEPENDENCIES is wrong. If some
> source file needs to be built before all the components of a program
> are built, consider using the BUILT_SOURCES variable instead (see
> Sources).
https://www.gnu.org/software/automake/manual/automake.html#Linking
Instead, using BUILT_SOURCES works, as mentioned in the manual.
https://www.gnu.org/software/automake/manual/automake.html#Sources
I have also removed the dependencies from i3_SOURCES since AFAIK
dependencies to header files don't do anything. I have verified that
modifying the header correctly re-triggers the build for i3 &
i3-config-wizard.
> Header files listed in a _SOURCES definition will be included in the
> distribution but otherwise ignored. In case it isn’t obvious, you
> should not include the header file generated by configure in a
> _SOURCES variable; this file should not be distributed. Lex (.l) and
> Yacc (.y) files can also be listed; see Yacc and Lex.
https://www.gnu.org/software/automake/manual/automake.html#Program-Sources
An alternative instead of BUILT_SOURCES that should also work in our
case is found in this section:
https://www.gnu.org/software/automake/manual/automake.html#Built-Sources-Example
see "Recording Dependencies manually". The syntax would be:
foo.$(OBJEXT): $(config_parser_SOURCES) $(command_parser_SOURCES)
The benefit of this over BUILT_SOURCES is that it will work for targets
other than 'all', 'check' and 'install'. However, since we don't really
have such targets we don't need to do this right now.
Tested extensively using this script:
#!/bin/bash
set -x
autoreconf -fi
while mkdir build && cd build && ../configure && make -j; do
cd ..
rm -rf build
done
Fixes#3670
i3-config-wizard uses libi3.a as part of its build process. When
parallel build is enabled, build of i3-config-wizard may start before
libi3.a finished building. Fix this by adding dependency on libi3.a
Fixes: #4020
* Add RELEASE-NOTES template for next
The placeholder items are useful instead of having to find the •
character for every new release.
Closes#4040
* Remove leftover bugfix release notes from tree
Fixes#4039
Crash with docking clients where the floating field is set even though
floating_enable refuses to make them floating.
See issue for example with logs.
Following the reproduction instructions from
https://github.com/i3/i3/issues/3242#issuecomment-436175346
For me, #3242 happened when the following sequence executed:
1. Fullscreening window correctly calls `stop_child()` in
6e24e2ad6f/i3bar/src/xcb.c (L685)
2. Xrandr change, `reconfig_windows()` is called and `output->visible` is
set to `true` in this line:
6e24e2ad6f/i3bar/src/xcb.c (L1791)
3. When the window's fullscreen is disabled,
`handle_visibility_notify()` returns in this line:
6e24e2ad6f/i3bar/src/xcb.c (L677)
because previously `output->visible` was set to `true`
To fix this, I call `cont_child()` more leniently since it is a no-op
when the child is not stopped.
Fixes#3242Closes#3761
nixo