This was a regression. The previous logic would always show the cursor
if we were using a non-blinking cursor. But, if the terminal state is
explicitly requesting an invisible cursor (mode 25) then we need to hide
the cursor.
Previously, we'd send renderer screen size updates and termio sigwnch
updates on every single resize event even if the screen size or grid
sizes didn't change. This is super noisy and given how many resize
events macOS sends, its also very expensive.
This commit makes it so that we only update the renderer if the screen
changed. If the screen size didn't change, the grid size couldn't have
changed either.
If the screen size did change, its still possible the grid size didn't
change since Ghostty supports fluid pixel-level resizing. We have to
send the screen size event to the renderer so all the GPU shader vars
are right but we do not have to send a termio event.
So, only if the grid size changed do we then notify the pty that the
terminal dimensions changed. Note that the resize event for ptys does
have a pixel-level x/y but I don't think the granularity is useful
beyond grid changes.
See before/after screenshots. Before, without the space, it's wrong. I
fixed that and added a newline, since a big paragraph like that looks a
bit wrong in a dialog popup.
Feel free to ignore or close this, because this is personal and if I
could figure out the syntax, I'm sure I could overwrite the keybindings
in the config myself.
But here's my case: `Ctrl+[` sends escape and I use that instead of
`Esc` because it's easier to reach (capslock is remapped to `ctrl`, so
`ctrl+[` is homerow only).
Kitty uses it's own "kittymod" combo for a lot of keybindings and for
the equivalent of these two, it uses `Ctrl+shift`. That's already taken
by other keybindings, so I added `.super` here.
Again: feel free to ignore. Personal preference. If you close this PR,
I'll have to tweak my config on Linux.
The GTK app has a really thin, white border around the terminal
contents. It's really distracting when in fullscreen.
At first I thought it's the window that has a border, but turns out
that's not it. It's the GTK Notebook widget. Luckily for us, GTK4 has a
single API call that allows us to turn off the border.
Fixes#210
We were previously taking any `CSI <symbol> <data> m` as SGR. But SGR is
only if "symbol" is empty. There are other forms of `CSI m` that set the
intermediate symbol to `?` or `>` and we don't implement those. We
shouldn't treat that as a SGR attribute either.
See #201 for more information.
Problem is that while we fall back to a default value to pass to
`setlocale`, we don't set a `LANG` and instead reset it to `""`.
What this does here is it changes the resetting to `""` and instead sets
it to the default value.
This fixes#99 for me.
Without this fix I end up with paddings of `3.333333` because my DPI is
`125.0` on Linux. If I set it to `144.0` manually so that the `/ 72`
gives me a clean `2.0`, the blurry fonts are gone.
I do think the calculation here is correct (even though I'm not sure
whether we should use 72? Why not 96? Or another system value?), so
let's use `std.math.floor` to get us to a "clean" padding of `3.0`.
That also solves blurry fonts for me.
Fixes#201
I don't fully understand locales, but it appears that the locale
returned from NSLocale can be "valid" in general but invalid according
to libc's locale API. If you attempt to `setlocale` with this bad
locale, it defaults everything to "C", which ends up breaking a lot of
things.
This commit validates the locale, and if it is invalid, we default to
"en_US.UTF-8" so things tend to work. This behavior can be overridden
using standard environment variables (LANG, LC_ALL, etc.).
This also doesn't touch env vars so further subprocesses from the shell
see original locale env vars.
