Set appid/wmclass for zed window

[jakobhellermann]

fixes #9132

By setting the app id, window managers like sway can apply custom configuration like for_window [app_id="zed"] floating enable.
Tested using wlprop/hyprctl activewindow for wayland, xprop for x11.

Release Notes:

• Zed now sets the window app id / class, which can be used e.g. in window managers like sway/i3 to define custom rules

[cla-bot]

We require contributors to sign our Contributor License Agreement, and we don't have @jakobhellermann on file. You can sign our CLA at https://zed.dev/cla. Once you've signed, post a comment here that says '@cla-bot check'.

[jakobhellermann]

The x11 WM_CLASS property contains two null-terminated strings, often referred to as "instance"/"class".
In this PR I just wrote a single set_app_id function on the PlatformWindow that sets both to the same value since as far as I can tell x11 is the only platform on which this is relevant.

https://tronche.com/gui/x/icccm/sec-4.html#WM_CLASS

[jakobhellermann]

@cla-bot check

[cla-bot]

The cla-bot has been summoned, and re-checked this pull request!

[mikayla-maki]

Looks like the tests end up calling this function, I'd suggest not using unimplemented!() ever, just let it be a noop :)

[jakobhellermann]

The new test failure is in test_chunk_text_grammar (and only on mac) so nothing caused by these changes.

[mikayla-maki]

so nothing caused by these changes

That seems incorrect. The test is relying on text length checks, and this PR modifies the sample text. You should remove that change or recalculate the values. macOS is the only CI check that actually runs the tests right now, due to crashes in other implementations.

[mikayla-maki]

Mark it for review once that test is fixed!

[jakobhellermann]

That seems incorrect. The test is relying on text length checks, and this PR modifies the sample text. You should remove that change or recalculate the values. macOS is the only CI check that actually runs the tests right now, due to crashes in other implementations.

Oh right, I totally forgot I changed that test.

I've changed the identifier to dev.zed.Zed and fixed that test.

[jakobhellermann]

There's still a CI failure during cargo xtask clippy on linux that I can't seem to reproduce or understand.

It's complaining about

no field `xcb_connection` on type `&mut platform::linux::x11::window::X11Window`

on line 503, suggesting self.0.xcb_connection,

but

1. the code should be actually at line 466. Line 503 is unrelated
2. the other code like set_title also directly calls self.xcb_connection.
3. X11Window has the fields xcb_connection and x_window

not sure what's going on there...

[apricotbucket28]

@jakobhellermann It's a conflict with #11008, you just need to rebase and it should be good to go 🙂

[jakobhellermann]

rebased

[mikayla-maki]

Thank you!