neovim will try to force the window to resize after suspend/resume.
No other terminal allows this today, so rather than log it as an
unhandled event, explicitly ignore it.
This is a port of something I found when developing some changes
in watchman. When doing something like:
`some win32 command | wsl something`
That pipe doesn't look like a normal win32 pipe and we'd end up
treating it like a socket, but it isn't really a socket either.
This commit adds some plumbing that will use the github API to
probe the currently released version of wezterm, and if it is
newer than the running version, show a window with some release
information and links to the changelog and download page.
The checks can be disabled in the config (but require a restart
to take effect!) and default to checking every 24 hours.
If running an AppImage on linux, links directly to the appimage
download. In the future I'd like to have the download button
use zsync to apply the update to the local image.
https://github.com/AppImage/AppImageKit/issues/368 introduced a way
to maintain configuration alongside an AppImage file stored on a
flash drive:
```
wez@cube-localdomain /tmp/portable-wezterm
9:06 130 $ ls -al
total 8720
drwxrwxr-x. 4 wez wez 100 May 6 08:54 ./
drwxrwxrwt. 19 root root 580 May 6 09:05 ../
-rwxr-xr-x. 1 wez wez 8929256 May 6 08:59 WezTerm-20200505-090057-31c6155f-Fedora31.AppImage*
drwxrwxr-x. 3 wez wez 60 May 6 08:15 WezTerm-20200505-090057-31c6155f-Fedora31.AppImage.config/
drwxrwxr-x. 2 wez wez 40 May 6 08:54 WezTerm-20200505-090057-31c6155f-Fedora31.AppImage.home/
```
When launched and these `<IMAGE>.config` and/or `<IMAGE>.home` dirs
exist, the appimage machinery will redirect the HOME and XDG_CONFIG_DIR
env vars to point to those locations instead of the normal place.
We weren't respecting XDG_CONFIG_DIR (fixed in this commit), but we also
need to take care: we don't want to re-export those locations to
children of wezterm so we need to load those values and then clean
up the environment.
The real problem was an inconsistency in computing the tab bar
enablement state. This makes the math the same in both places
and re-enables the `hide_tab_bar_if_only_one_tab` option.
There's some discussion about terminal identification and
there's what looks to be some consensus for adopting these
two environment variables that were pioneered by Apple.
refs: https://github.com/mintty/mintty/issues/881
tweak some container names
fixup container names and regenerate the config
Ensure that curl is installed
the rust toolchain stuff wants curl
run apt update before apt install
centralize updating apt, and install sudo
revise get-deps script for slimmer debian images
more container related tweaks
get-deps: Don't require lsb-release on debianish systems
more get-deps improvements
Explicitly ask GH actions to recursively checkout the repo
Explicitly install git
fetch tags and tweak git build for debian systems
moar adjustments
remove deb8 (its wayland is too old), fixup debian publish
Ensure git build deps are installed always, tidy up sudo usage
We need to gate with the config generation check as update_title
can be called during the configuration reloading process and
the tab bar state may not reflect the config until after the
config is reloaded!
refs: https://github.com/wez/wezterm/issues/173
Teach the window layer about window icons and implement the
plumbing for this on X11.
For Wayland there is no direct way to specify the icon; instead
the application ID is used to locate an appropriate .desktop filename.
We set the app id from the classname but that didn't match the installed
name for our desktop file which is namespaced under my domain, so change
the window class to match that and enable the window icon on Wayland.
refs: https://github.com/wez/wezterm/issues/172#issuecomment-619938047
This was a bit of a pain to track down because this behavior
isn't specified anywhere or called out explicitly.
The issue is that if you use a true color escape sequence such as
```bash
printf "\x1b[38;2;255;100;0mTRUECOLOR\n"
```
the active color would remain active when switching between the
primary and the alt screen until something (eg: `ls --color`) changed
it again.
I hadn't run into this because in my prompt, many many years ago, I had
it set to perform an SGR reset (`\x1b[0m`) as the first thing to ensure
that the shell is in a saner state.
For users that don't do this they end up with a weird looking color
bleed effect.
refs: https://github.com/wez/wezterm/issues/169
On windows, prevent console subsystem processes spawned by
lua (such as the `wsl -l` example configuration) from momentarily
popping up and stealing the focus. This was happening too fast
to see in most cases, but could cause the wezterm window to momentarily
repaint its title bar with lose focus before regaining it.
This fixes an annoyance with the configuration error window;
previously we would spawn a new window for each error that
was discovered in the config, which cluttered up the screen
and felt irritating when iterating on the config file.
This commit reuses the connection status UI infra to make a
single persistent error log window.
This commit adds some helper functions that make it possible to
dynamically discover and add WSL distributions to the launcher
menu.
refs: https://github.com/wez/wezterm/issues/159
* ctrl-R to find a line and then hit enter would cause the search
text rather than the match text to be returned and run!
* When exiting the editor, clear to end of screen to make sure
that we clean up any UI from the incremental search status
This helps us keep track of the extent and cursor position that
we render for the line editor, making it easier to make the editor
rendering more fancy.
