When testing Zed locally, it's rarely necessary to log in the real with,
via Zed.dev and GitHub. We usually use `zed-local`. Since zed.dev is not
going to be open source (at least right away), this PR removes it from
our local development workflow.
* Remove zed.dev from the Procfile
* Change the `seed` script to not create an admin user for your
signed-in github user
* Instead have both `zed-local` and the `seed` script read from an
`.admins.json` file, which the user can create in order to customize who
they sign in as when running `zed-local`.
* Update all of the docs for building and developing zed.
When the `List` element's state is `ListState::reset()`, it eagerly
trashes it's cached element heights in anticipation of a prompt render.
But, due to the recent `display_layer` changes, that re-render is not
always forthcoming. This is a problem for `ListState::scroll()`, which
depends on these cached elements to correctly calculate the new logical
scroll offset.
Solutions we attempted:
- Cache the element heights and continue the scroll calculation
- This was conceptually incorrect, reset should only be called when the
underlying data has been changed, making any calculation with the old
results meaningless.
- Lazily re-compute the element heights in scroll
- Beyond being a non-trivial refactor, this would probably also cause us
to double-render the list in a single frame, which is bad.
- Cache the scroll offset and only calculate it in paint
- This solution felt awkward to implement and meant we can't supply
synchronous list scroll events.
- Delay resetting until paint
- This means that all of the other APIs that `ListState` supplies would
give temporarily incorrect results, worsening the problem
Given these issues, we settled on the solution with the least
compromises: drop scroll events if the state has been `reset()` between
`paint()` and `scroll()`. This shifts the responsibility for the problem
out of the List element and into consumers of `List`, if you want
perfectly smooth scrolling then you need to use `reset()` judiciously
and prefer `splice()`.
That said, I tested this by aggressively scrolling the Collab panel, and
it seems to work as well as it did before.
This PR also includes some initial testing infrastructure for working
with input from the platform and rendered elements.
Release Notes:
- N/A
Instead, admins are specified using a JSON file, 'admins.json'. This file is
gitignored. If it is not present, there is a default list of admins in
'admins.default.json'.
This PR decreases the size of the timestamps in the assistant's
conversation editor.
Ideally we'd want to align the baseline of the timestamp text with the
text in the sender button. I spent a while trying to do this, but it
seems like it may be pretty tricky.
Release Notes:
- Decreased the size of timestamps in the assistant panel conversation
editor.
There's a deadlock that we're seeing when joining a room, which we think
is a bug in the LiveKit client.
Also, we're still getting crashes when leaving calls:
https://github.com/livekit/client-sdk-swift/issues/299.
At this point, we believe both problems are due to recent changes to the
LiveKit swift sdk.
This happens a lot in guest workflows where they open the call with a
link and are jumped straight to a shared workspace.
Release Notes:
- Fixed opening the chat automatically in new windows
This PR adds the `color` crate, which will be the home of a number of
color-related utilities, and also acts as an interface between
[`palette`](https://crates.io/crates/palette) and the way `gpui` colors
work.
The goal of this crate is to centralize color utilities like mixing and
blending, building color ramps and sets of colors for state and more.
## Todo:
- [x] hex -> Color
- [x] Color mixing
- [x] Color blending using blend modes (overlay, multiply, etc)
- [ ] ~~Build color ramp from color~~
- [x] Build state set from color
- [ ] ~~Update Theme to use the color crate~~
Release Notes:
- None (Internal changes: Adds the `color` crate for working with
colors.)
In #3910 we made the welcome screen use the same background color as the
editor.
However, this later regressed in
cdd5cb16ed.
This PR fixes that regression and restores the correct color for the
welcome page.
Release Notes:
- Fixed the background color of the welcome screen.
Before this change we wouldn't submit all possible primitives of the
same kind that are less-than the max order.
Result was that we would submit, say, 10 paths each in a separate batch
instead of actually batching them.
This was overly strict because even if the order of two different
primitives was the same, we could have still batched the 1st primitive
kind, if its implicit ordering was less than 2nd kind.
Example: say we have the following primitives and these orders
5x paths, order 3
2x sprites, order 3
Previously, we would submit 1 path, 1 path, 1 path, 1 path, 1 path, then
the sprites.
With this changes, we batch the 5 paths into one batch.
Co-authored-by: Antonio <antonio@zed.dev>
This "adds" the keybindings I was missing in Vim mode (e.g. `Ctrl-[` to
cancel a selection) by fixing the definitions in the keymap from
`Ctrl+[` to `Ctrl-[`.
Previously, we were using `size_of` but passing the wrong type in
(`MonochromeSprite` instead of `PathSprite`). This caused us to read
outside of the `sprites` smallvec and triggered the segfault. This
reverts #4078 because I don't think using a `SmallVec` was the issue (it
might have masked this problem though, because we would most of the time
copy from the stack and not from the heap).
With this pull request we are also fixing another potential source of
segfaults, due to checking if we exhausted the instance buffer too late
when drawing underlines.
Release Notes:
- Fixed a crash that could happen during rendering.
Previously, we were using `size_of` but passing the wrong type in
(MonochromeSprite instead of PathSprite). This caused us to read outside
of the `sprites` smallvec and triggered the segfault.
We're occasionally seeing a crash in MetalRenderer::draw.
Looking at the backtrace, it seems almost certainly to be happening in
the call to `ptr::copy_nonoverlapping` on line 604 (see `#Don't Panic!`
channel notes)
As we already have added bounds checking to the destination, it seems
most
likely (however improbable) that somehow we're getting an invalid Ptr
and
length from the SmallVec.
To try and make progress on this, let's try a Vec for a bit lest there
is a subtle issue in SmallVec (though I couldn't spot one).
Release Notes:
- (maybe) Fixes SEGFAULT in MetalRenderer::draw
