Compare commits

...

21 Commits

Author SHA1 Message Date
Mikayla
29611ac1cb AHHHHHHH 2024-05-31 11:16:48 -07:00
Max Brunsfeld
d12b8c3945 Simplify and improve concurrency of git status updates (#12513)
The quest for responsiveness in large git repos continues. This is a
follow-up to https://github.com/zed-industries/zed/pull/12444

Release Notes:

- N/A
2024-05-31 09:10:09 -07:00
Vitaly Slobodin
356fcec337 ruby: Add a new injection for regular expressions (#12533)
# Summary

Hello. This pull request adds a new injection to `injections.scm` for
Ruby language to highlight regular expressions. Thanks.

## Before

![CleanShot 2024-05-31 at 16 25
46@2x](https://github.com/zed-industries/zed/assets/1894248/8b88718e-8f13-4d61-b6f9-6a25b3ebcc57)

## After


![after](https://github.com/zed-industries/zed/assets/1894248/e11f6ec3-45c6-40f8-b6d9-ddbfd16a3331)

Release Notes:

- N/A

Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
2024-05-31 12:02:59 -04:00
Vitaly Slobodin
08123a270a ruby: Add proper indentation for singleton methods (#12535)
Hi. Currently, Zed uses incorrect indentation for singleton methods:

```ruby
def self.build
| # <= cursor position after hitting Enter
end
```

Handling the `singleton_method` token indentation
changes this behavior to the following:

```ruby
def self.build
  | # <= cursor position after hitting Enter
end
```

## Before


https://github.com/zed-industries/zed/assets/1894248/40fc2b37-692f-469f-9cbe-05cbb1ab4c3c

## After



https://github.com/zed-industries/zed/assets/1894248/d9ba8d27-fd17-4c74-b22c-a4de124739a3



Release Notes:

- N/A
2024-05-31 11:36:42 -04:00
张小白
6eb8e83411 docs: Update font features (#12229)
This follows up the changes in #11611 and #11898 

Release Notes:

- N/A
2024-05-31 11:34:16 -04:00
Marshall Bowers
4c51ee7816 assistant: Allow passing module paths to /rustdoc command (#12536)
This PR updates the `/rustdoc` command to accept module paths in
addition to just a crate name.

This will return the docs.rs page just for that particular module.

### Examples

```
/rustdoc bevy
/rustdoc bevy::ecs
/rustdoc bevy::ecs::component
```

<img width="641" alt="Screenshot 2024-05-31 at 11 18 25 AM"
src="https://github.com/zed-industries/zed/assets/1486634/d88af19f-5ba1-4073-8108-63cccd138db6">

<img width="641" alt="Screenshot 2024-05-31 at 11 18 35 AM"
src="https://github.com/zed-industries/zed/assets/1486634/9c414ab1-0be8-4d79-8c64-b45f19266556">


Release Notes:

- N/A
2024-05-31 11:31:22 -04:00
Vitaly Slobodin
660cf214c7 ruby: Capture the heredoc content only and downcase the language (#12532)
# Summary

Hi. Current `heredoc` injection for Ruby language captures the
`heredoc_end` token. That's a bit incorrect because we want to capture
the content only. Suppose we have the following Ruby code:

```ruby
<<~JS
  function myFunc() {
    const myConstant = [];
  }

  let a = '1'
JS
```

And this is its syntax tree:

```
[program] [0, 0] - [7, 0]
  [heredoc_beginning] [0, 0] - [0, 5]
  [heredoc_body] [0, 5] - [6, 2]
    [heredoc_content] [0, 5] - [6, 0]
    [heredoc_end] [6, 0] - [6, 2]
```

Current injection capture all content of the `heredoc_body`:

![CleanShot 2024-05-31 at 17 03
31@2x](https://github.com/zed-industries/zed/assets/1894248/ff8c5195-b532-42d2-91b1-48405a6d3b50)

But we want to capture the `heredoc_content` only and this PR resolves
that, additionally it downcases the language like Zed does in other
languages like Terraform.

![CleanShot 2024-05-31 at 17 05
17@2x](https://github.com/zed-industries/zed/assets/1894248/e81dabd0-3246-4ef2-9524-a7adcb9242ab)


Release Notes:

- N/A
2024-05-31 11:19:10 -04:00
Marshall Bowers
b2565fadfb ruby: Fix injections query location (#12534)
This PR fixes the location of the `injections.scm` query within the Ruby
extension.

Same as #12532, but without the content changes to `injections.scm`.

Release Notes:

- N/A
2024-05-31 10:42:39 -04:00
Felipe Renan
2cff075c53 elixir: Fix mix test $ZED_SYMBOL task (#11879)
$ZED_SYMBOL doesn't really work here once that will try to do something
like this:

  mix test MyModule.MyModuleTest

instead of using the path of the file:

  mix test test/my_module/my_module_test.exs
  
Release Notes:

- Fix mix test $ZED_SYMBOL to use ZED_RELATIVE_FILE instead
- Use ZED_RELATIVE_FILE instead of ZED_FILE to improve mix tasks results
on Elixir umbrella projects
2024-05-31 12:54:14 +02:00
Vladas Zakrevskis
819bb2663d Fix recent project index order (#12507)
Fixed bug introduced in:
https://github.com/zed-industries/zed/pull/12502

Filtering before `enumerate` call breaks project order and instead of
hiding current project it hides some other project.

Release Notes:
- N/A
2024-05-31 05:50:03 +03:00
moshyfawn
dc141d0f61 typescript: Fix shorthand property highlight (#12505)
Release Notes:

- Fixed Typescript shorthand property highlight
([#5239](https://github.com/zed-industries/zed/issues/5239)).

Closes: #5239
2024-05-30 18:27:03 -04:00
Bennet Bo Fenner
22cf73acec indent guides: Use primary buffer language to determine tab size (#12506)
When indent guides were still WIP, I thought it might be a good idea to
detect the tab size for every line individually, so we can handle files
with mixed indentations. However, while optimizing the performance of
indent guides I found that getting the language at a given anchor was
pretty expensive, therefore I only resolved the language for the first
visible row. However, this could lead to some weird flickering, where
the indent guides would use different tab sizes depending on the first
visible row (see #12492). This can be fixed by just using the primary
buffer language size.

So as of right now indent guides cannot handle files with mixed
indentations. Im not sure if anyone actually does/expects this, but one
use case I could imagine is something like this:
User x has a svelte file, where the tab size is set to `4`. However the
svelte code uses typescript inside a script tag, which User x wants to
use a tab size of `2`. The approach used here would not work for this,
but then again I think our formatter does not even support something
like this. Im probably overcomplicating things, so let's stick with the
simple solution for now.

Release Notes:

- Fixed an issue where indent guides would use an incorrect tab size
([#12492](https://github.com/zed-industries/zed/issues/12492)).
2024-05-30 22:55:47 +02:00
Marshall Bowers
1d46a52c62 rustdoc_to_markdown: Don't push blank space after newline (#12504)
This PR fixes a small issue in `rustdoc_to_markdown` where we could push
a blank space after a newline, leading to an unwanted leading space.

Release Notes:

- N/A
2024-05-30 16:38:01 -04:00
Max Brunsfeld
fda975fb76 Re-subscribe to channels after signing back out 2024-05-30 13:32:34 -07:00
Vladas Zakrevskis
0f32145ecb Skip current project in recent projects (#12502)
Discussion: https://github.com/zed-industries/zed/discussions/12497

Release Notes:

- Removed current project from the recent projects modals
2024-05-30 23:30:34 +03:00
Marshall Bowers
6fe665ab94 rustdoc_to_markdown: Support bold and italics (#12501)
This PR extends `rustdoc_to_markdown` with support for bold and italic
text.

Release Notes:

- N/A
2024-05-30 16:06:21 -04:00
Max Brunsfeld
279c5ab81f Reduce DB load upon initial connection due to channel loading (#12500)
#### Lazily loading channels

I've added a new RPC message called `SubscribeToChannels` that the
client now sends when it first renders the channels panel. This causes
the server to load the channels for that client and send updates to that
client as channels are updated. Previously, the server did this upon
connection.

For backwards compatibility, the server will inspect clients' version,
and continue to do this work immediately for old clients.

#### Optimizations

Running collab locally, I realized that upon connecting, we were running
two concurrent transactions that *both* queried the `channel_members`
table: one for loading your channels, and one for loading your channel
invites. I've combined these into one query. In addition, we now use a
join to load channels + members, as opposed to two separate queries.
Even though `where id in` is efficient, it adds an extra round trip to
the database, keeping the transaction open for slightly longer.

Release Notes:

- N/A
2024-05-30 13:02:55 -07:00
Marshall Bowers
99901801f4 rustdoc_to_markdown: Improve paragraph handling (#12498)
This PR improves `rustdoc_to_markdown`'s paragraph handling to produce
better output.

Specifically, there should now be fewer instances where a space is
missing between words as the result of line breaks in the source HTML.

Release Notes:

- N/A
2024-05-30 15:14:02 -04:00
Marshall Bowers
4dc98026c4 rustdoc_to_markdown: Add helper methods for checking HTML attributes (#12496)
This PR adds some helper methods to `HtmlElement` to make it easier to
interact with the element's attributes.

This cleans up a bunch of the code by a fair amount.

Release Notes:

- N/A
2024-05-30 14:15:08 -04:00
Marshall Bowers
c83d1c23d7 rustdoc_to_markdown: Handle "stabs" in item name entries (#12494)
This PR extends `rustdoc_to_markdown` with support for rustdoc's
"stabs".

These are used in item name lists to indicate that the construct is
behind a feature flag:

<img width="641" alt="Screenshot 2024-05-30 at 1 34 53 PM"
src="https://github.com/zed-industries/zed/assets/1486634/0216f325-dc4e-4302-b6db-149ace31deea">

We now treat these specially in the Markdown output:

<img width="813" alt="Screenshot 2024-05-30 at 1 35 27 PM"
src="https://github.com/zed-industries/zed/assets/1486634/96396305-123d-40b2-af49-7eed71b62971">

Release Notes:

- N/A
2024-05-30 13:46:14 -04:00
Marshall Bowers
39a2cdb13f rustdoc_to_markdown: Strip "Copy item path to clipboard" button (#12490)
This PR strips the "Copy item path to clipboard" button from the rustdoc
output.

Release Notes:

- N/A
2024-05-30 12:55:37 -04:00
23 changed files with 554 additions and 266 deletions

View File

@@ -17,10 +17,17 @@ impl RustdocSlashCommand {
async fn build_message(
http_client: Arc<HttpClientWithUrl>,
crate_name: String,
module_path: Vec<String>,
) -> Result<String> {
let version = "latest";
let path = format!(
"{crate_name}/{version}/{crate_name}/{module_path}",
module_path = module_path.join("/")
);
let mut response = http_client
.get(
&format!("https://docs.rs/{crate_name}"),
&format!("https://docs.rs/{path}"),
AsyncBody::default(),
true,
)
@@ -87,14 +94,28 @@ impl SlashCommand for RustdocSlashCommand {
};
let http_client = workspace.read(cx).client().http_client();
let crate_name = argument.to_string();
let mut path_components = argument.split("::");
let crate_name = match path_components
.next()
.ok_or_else(|| anyhow!("missing crate name"))
{
Ok(crate_name) => crate_name.to_string(),
Err(err) => return Task::ready(Err(err)),
};
let module_path = path_components.map(ToString::to_string).collect::<Vec<_>>();
let text = cx.background_executor().spawn({
let crate_name = crate_name.clone();
async move { Self::build_message(http_client, crate_name).await }
let module_path = module_path.clone();
async move { Self::build_message(http_client, crate_name, module_path).await }
});
let crate_name = SharedString::from(crate_name);
let module_path = if module_path.is_empty() {
None
} else {
Some(SharedString::from(module_path.join("::")))
};
cx.foreground_executor().spawn(async move {
let text = text.await?;
let range = 0..text.len();
@@ -107,6 +128,7 @@ impl SlashCommand for RustdocSlashCommand {
id,
unfold,
crate_name: crate_name.clone(),
module_path: module_path.clone(),
}
.into_any_element()
}),
@@ -121,17 +143,23 @@ struct RustdocPlaceholder {
pub id: ElementId,
pub unfold: Arc<dyn Fn(&mut WindowContext)>,
pub crate_name: SharedString,
pub module_path: Option<SharedString>,
}
impl RenderOnce for RustdocPlaceholder {
fn render(self, _cx: &mut WindowContext) -> impl IntoElement {
let unfold = self.unfold;
let crate_path = self
.module_path
.map(|module_path| format!("{crate_name}::{module_path}", crate_name = self.crate_name))
.unwrap_or(self.crate_name.to_string());
ButtonLike::new(self.id)
.style(ButtonStyle::Filled)
.layer(ElevationIndex::ElevatedSurface)
.child(Icon::new(IconName::FileRust))
.child(Label::new(format!("rustdoc: {}", self.crate_name)))
.child(Label::new(format!("rustdoc: {crate_path}")))
.on_click(move |_, cx| unfold(cx))
}
}

View File

@@ -62,6 +62,7 @@ pub struct ChannelStore {
opened_buffers: HashMap<ChannelId, OpenedModelHandle<ChannelBuffer>>,
opened_chats: HashMap<ChannelId, OpenedModelHandle<ChannelChat>>,
client: Arc<Client>,
did_subscribe: bool,
user_store: Model<UserStore>,
_rpc_subscriptions: [Subscription; 2],
_watch_connection_status: Task<Option<()>>,
@@ -243,6 +244,20 @@ impl ChannelStore {
.log_err();
}),
channel_states: Default::default(),
did_subscribe: false,
}
}
pub fn initialize(&mut self) {
if !self.did_subscribe {
if self
.client
.send(proto::SubscribeToChannels {})
.log_err()
.is_some()
{
self.did_subscribe = true;
}
}
}
@@ -1035,7 +1050,7 @@ impl ChannelStore {
fn handle_disconnect(&mut self, wait_for_reconnect: bool, cx: &mut ModelContext<Self>) {
cx.notify();
self.did_subscribe = false;
self.disconnect_channel_buffers_task.get_or_insert_with(|| {
cx.spawn(move |this, mut cx| async move {
if wait_for_reconnect {

View File

@@ -654,6 +654,7 @@ pub struct ChannelsForUser {
pub channel_memberships: Vec<channel_member::Model>,
pub channel_participants: HashMap<ChannelId, Vec<UserId>>,
pub hosted_projects: Vec<proto::HostedProject>,
pub invited_channels: Vec<Channel>,
pub observed_buffer_versions: Vec<proto::ChannelBufferVersion>,
pub observed_channel_messages: Vec<proto::ChannelMessageId>,

View File

@@ -416,7 +416,9 @@ impl Database {
user_id: UserId,
tx: &DatabaseTransaction,
) -> Result<MembershipUpdated> {
let new_channels = self.get_user_channels(user_id, Some(channel), tx).await?;
let new_channels = self
.get_user_channels(user_id, Some(channel), false, tx)
.await?;
let removed_channels = self
.get_channel_descendants_excluding_self([channel], tx)
.await?
@@ -481,44 +483,10 @@ impl Database {
.await
}
/// Returns all channel invites for the user with the given ID.
pub async fn get_channel_invites_for_user(&self, user_id: UserId) -> Result<Vec<Channel>> {
self.transaction(|tx| async move {
let mut role_for_channel: HashMap<ChannelId, ChannelRole> = HashMap::default();
let channel_invites = channel_member::Entity::find()
.filter(
channel_member::Column::UserId
.eq(user_id)
.and(channel_member::Column::Accepted.eq(false)),
)
.all(&*tx)
.await?;
for invite in channel_invites {
role_for_channel.insert(invite.channel_id, invite.role);
}
let channels = channel::Entity::find()
.filter(channel::Column::Id.is_in(role_for_channel.keys().copied()))
.all(&*tx)
.await?;
let channels = channels.into_iter().map(Channel::from_model).collect();
Ok(channels)
})
.await
}
/// Returns all channels for the user with the given ID.
pub async fn get_channels_for_user(&self, user_id: UserId) -> Result<ChannelsForUser> {
self.transaction(|tx| async move {
let tx = tx;
self.get_user_channels(user_id, None, &tx).await
})
.await
self.transaction(|tx| async move { self.get_user_channels(user_id, None, true, &tx).await })
.await
}
/// Returns all channels for the user with the given ID that are descendants
@@ -527,25 +495,37 @@ impl Database {
&self,
user_id: UserId,
ancestor_channel: Option<&channel::Model>,
include_invites: bool,
tx: &DatabaseTransaction,
) -> Result<ChannelsForUser> {
let mut filter = channel_member::Column::UserId
.eq(user_id)
.and(channel_member::Column::Accepted.eq(true));
let mut filter = channel_member::Column::UserId.eq(user_id);
if !include_invites {
filter = filter.and(channel_member::Column::Accepted.eq(true))
}
if let Some(ancestor) = ancestor_channel {
filter = filter.and(channel_member::Column::ChannelId.eq(ancestor.root_id()));
}
let channel_memberships = channel_member::Entity::find()
let mut channels = Vec::<channel::Model>::new();
let mut invited_channels = Vec::<Channel>::new();
let mut channel_memberships = Vec::<channel_member::Model>::new();
let mut rows = channel_member::Entity::find()
.filter(filter)
.all(tx)
.await?;
let channels = channel::Entity::find()
.filter(channel::Column::Id.is_in(channel_memberships.iter().map(|m| m.channel_id)))
.all(tx)
.inner_join(channel::Entity)
.select_also(channel::Entity)
.stream(tx)
.await?;
while let Some(row) = rows.next().await {
if let (membership, Some(channel)) = row? {
if membership.accepted {
channel_memberships.push(membership);
channels.push(channel);
} else {
invited_channels.push(Channel::from_model(channel));
}
}
}
drop(rows);
let mut descendants = self
.get_channel_descendants_excluding_self(channels.iter(), tx)
@@ -643,6 +623,7 @@ impl Database {
Ok(ChannelsForUser {
channel_memberships,
channels,
invited_channels,
hosted_projects,
channel_participants,
latest_buffer_versions,

View File

@@ -176,23 +176,23 @@ async fn test_channel_invites(db: &Arc<Database>) {
.unwrap();
let user_2_invites = db
.get_channel_invites_for_user(user_2) // -> [channel_1_1, channel_1_2]
.get_channels_for_user(user_2)
.await
.unwrap()
.invited_channels
.into_iter()
.map(|channel| channel.id)
.collect::<Vec<_>>();
assert_eq!(user_2_invites, &[channel_1_1, channel_1_2]);
let user_3_invites = db
.get_channel_invites_for_user(user_3) // -> [channel_1_1]
.get_channels_for_user(user_3)
.await
.unwrap()
.invited_channels
.into_iter()
.map(|channel| channel.id)
.collect::<Vec<_>>();
assert_eq!(user_3_invites, &[channel_1_1]);
let (mut members, _) = db

View File

@@ -557,6 +557,7 @@ impl Server {
.add_request_handler(user_handler(request_contact))
.add_request_handler(user_handler(remove_contact))
.add_request_handler(user_handler(respond_to_contact_request))
.add_message_handler(subscribe_to_channels)
.add_request_handler(user_handler(create_channel))
.add_request_handler(user_handler(delete_channel))
.add_request_handler(user_handler(invite_channel_member))
@@ -1105,34 +1106,25 @@ impl Server {
.await?;
}
let (contacts, channels_for_user, channel_invites, dev_server_projects) =
future::try_join4(
self.app_state.db.get_contacts(user.id),
self.app_state.db.get_channels_for_user(user.id),
self.app_state.db.get_channel_invites_for_user(user.id),
self.app_state.db.dev_server_projects_update(user.id),
)
.await?;
let (contacts, dev_server_projects) = future::try_join(
self.app_state.db.get_contacts(user.id),
self.app_state.db.dev_server_projects_update(user.id),
)
.await?;
{
let mut pool = self.connection_pool.lock();
pool.add_connection(connection_id, user.id, user.admin, zed_version);
for membership in &channels_for_user.channel_memberships {
pool.subscribe_to_channel(user.id, membership.channel_id, membership.role)
}
self.peer.send(
connection_id,
build_initial_contacts_update(contacts, &pool),
)?;
self.peer.send(
connection_id,
build_update_user_channels(&channels_for_user),
)?;
self.peer.send(
connection_id,
build_channels_update(channels_for_user, channel_invites),
)?;
}
if should_auto_subscribe_to_channels(zed_version) {
subscribe_user_to_channels(user.id, session).await?;
}
send_dev_server_projects_update(user.id, dev_server_projects, session).await;
if let Some(incoming_call) =
@@ -3399,6 +3391,36 @@ async fn remove_contact(
Ok(())
}
fn should_auto_subscribe_to_channels(version: ZedVersion) -> bool {
version.0.minor() < 139
}
async fn subscribe_to_channels(_: proto::SubscribeToChannels, session: Session) -> Result<()> {
subscribe_user_to_channels(
session.user_id().ok_or_else(|| anyhow!("must be a user"))?,
&session,
)
.await?;
Ok(())
}
async fn subscribe_user_to_channels(user_id: UserId, session: &Session) -> Result<(), Error> {
let channels_for_user = session.db().await.get_channels_for_user(user_id).await?;
let mut pool = session.connection_pool().await;
for membership in &channels_for_user.channel_memberships {
pool.subscribe_to_channel(user_id, membership.channel_id, membership.role)
}
session.peer.send(
session.connection_id,
build_update_user_channels(&channels_for_user),
)?;
session.peer.send(
session.connection_id,
build_channels_update(channels_for_user),
)?;
Ok(())
}
/// Creates a new channel.
async fn create_channel(
request: proto::CreateChannel,
@@ -5034,7 +5056,7 @@ fn notify_membership_updated(
..Default::default()
};
let mut update = build_channels_update(result.new_channels, vec![]);
let mut update = build_channels_update(result.new_channels);
update.delete_channels = result
.removed_channels
.into_iter()
@@ -5064,10 +5086,7 @@ fn build_update_user_channels(channels: &ChannelsForUser) -> proto::UpdateUserCh
}
}
fn build_channels_update(
channels: ChannelsForUser,
channel_invites: Vec<db::Channel>,
) -> proto::UpdateChannels {
fn build_channels_update(channels: ChannelsForUser) -> proto::UpdateChannels {
let mut update = proto::UpdateChannels::default();
for channel in channels.channels {
@@ -5086,7 +5105,7 @@ fn build_channels_update(
});
}
for channel in channel_invites {
for channel in channels.invited_channels {
update.channel_invitations.push(channel.to_proto());
}

View File

@@ -2161,6 +2161,9 @@ impl CollabPanel {
}
fn render_signed_in(&mut self, cx: &mut ViewContext<Self>) -> Div {
self.channel_store.update(cx, |channel_store, _| {
channel_store.initialize();
});
v_flex()
.size_full()
.child(list(self.list_state.clone()).size_full())

View File

@@ -11574,6 +11574,7 @@ impl ViewInputHandler for Editor {
}
fn unmark_text(&mut self, cx: &mut ViewContext<Self>) {
dbg!("unmark text");
self.clear_highlights::<InputComposition>(cx);
self.ime_transaction.take();
}
@@ -11584,6 +11585,7 @@ impl ViewInputHandler for Editor {
text: &str,
cx: &mut ViewContext<Self>,
) {
dbg!((&range_utf16, text));
if !self.input_enabled {
cx.emit(EditorEvent::InputIgnored { text: text.into() });
return;
@@ -11646,6 +11648,7 @@ impl ViewInputHandler for Editor {
new_selected_range_utf16: Option<Range<usize>>,
cx: &mut ViewContext<Self>,
) {
dbg!((&range_utf16, text, &new_selected_range_utf16));
if !self.input_enabled {
cx.emit(EditorEvent::InputIgnored { text: text.into() });
return;

View File

@@ -311,6 +311,7 @@ unsafe fn build_window_class(name: &'static str, superclass: &Class) -> *const C
decl.register()
}
#[derive(Debug)]
#[allow(clippy::enum_variant_names)]
enum ImeInput {
InsertText(String, Option<Range<usize>>),
@@ -1219,16 +1220,7 @@ extern "C" fn handle_key_event(this: &Object, native_event: id, key_equivalent:
let keydown = event.keystroke.clone();
let fn_modifier = keydown.modifiers.function;
// Ignore events from held-down keys after some of the initially-pressed keys
// were released.
if event.is_held {
if lock.last_fresh_keydown.as_ref() != Some(&keydown) {
return YES;
}
} else {
lock.last_fresh_keydown = Some(keydown.clone());
}
lock.input_during_keydown = Some(SmallVec::new());
drop(lock);
// Send the event to the input context for IME handling, unless the `fn` modifier is
@@ -1240,11 +1232,32 @@ extern "C" fn handle_key_event(this: &Object, native_event: id, key_equivalent:
let _: BOOL = msg_send![input_context, handleEvent: native_event];
}
}
let mut lock = window_state.lock();
println!("******************************************");
///// THE PROBLEM AREA ///////
// Ignore events from held-down keys after some of the initially-pressed keys
// were released.
if dbg!(event.is_held) {
if dbg!(lock.last_fresh_keydown.as_ref()) != Some(&keydown) {
return YES;
}
} else {
//????
lock.last_fresh_keydown = Some(dbg!(keydown.clone()));
}
lock.input_during_keydown = Some(SmallVec::new());
///// THE PROBLEM AREA ///////
// Send the event to the input context for IME handling, unless the `fn` modifier is
// being pressed.
// this will call back into `insert_text`, etc.
let mut handled = false;
let mut lock = window_state.lock();
let previous_keydown_inserted_text = lock.previous_keydown_inserted_text.take();
let mut input_during_keydown = lock.input_during_keydown.take().unwrap();
dbg!(&input_during_keydown);
let mut callback = lock.event_callback.take();
drop(lock);
@@ -1260,7 +1273,10 @@ extern "C" fn handle_key_event(this: &Object, native_event: id, key_equivalent:
.flatten()
.is_some();
dbg!(is_composing, &last_ime);
if let Some(ime) = last_ime {
// Problem area
if let ImeInput::InsertText(text, _) = &ime {
if !is_composing {
window_state.lock().previous_keydown_inserted_text = Some(text.clone());
@@ -1275,6 +1291,7 @@ extern "C" fn handle_key_event(this: &Object, native_event: id, key_equivalent:
handled = true;
send_to_input_handler(this, ime);
}
// Problem area
} else if !is_composing {
let is_held = event.is_held;
@@ -1655,21 +1672,27 @@ extern "C" fn valid_attributes_for_marked_text(_: &Object, _: Sel) -> id {
}
extern "C" fn has_marked_text(this: &Object, _: Sel) -> BOOL {
with_input_handler(this, |input_handler| input_handler.marked_text_range())
.flatten()
.is_some() as BOOL
dbg!("has marked range");
with_input_handler(
this,
|input_handler| dbg!(input_handler.marked_text_range()),
)
.flatten()
.is_some() as BOOL
}
extern "C" fn marked_range(this: &Object, _: Sel) -> NSRange {
dbg!("querying marked range");
with_input_handler(this, |input_handler| input_handler.marked_text_range())
.flatten()
.map_or(NSRange::invalid(), |range| range.into())
.map_or(NSRange::invalid(), |range| dbg!(range).into())
}
extern "C" fn selected_range(this: &Object, _: Sel) -> NSRange {
dbg!("asking for sel range");
with_input_handler(this, |input_handler| input_handler.selected_text_range())
.flatten()
.map_or(NSRange::invalid(), |range| range.into())
.map_or(NSRange::invalid(), |range| dbg!(range).into())
}
extern "C" fn first_rect_for_character_range(

View File

@@ -3153,10 +3153,7 @@ impl BufferSnapshot {
range: Range<Anchor>,
cx: &AppContext,
) -> Vec<IndentGuide> {
fn tab_size_for_row(this: &BufferSnapshot, row: BufferRow, cx: &AppContext) -> u32 {
let language = this.language_at(Point::new(row, 0));
language_settings(language, None, cx).tab_size.get() as u32
}
let tab_size = language_settings(self.language(), None, cx).tab_size.get() as u32;
let start_row = range.start.to_point(self).row;
let end_row = range.end.to_point(self).row;
@@ -3167,9 +3164,6 @@ impl BufferSnapshot {
let mut result_vec = Vec::new();
let mut indent_stack = SmallVec::<[IndentGuide; 8]>::new();
// TODO: This should be calculated for every row but it is pretty expensive
let tab_size = tab_size_for_row(self, start_row, cx);
while let Some((first_row, mut line_indent)) = row_indents.next() {
let current_depth = indent_stack.len() as u32;

View File

@@ -5,6 +5,8 @@
; Properties
(property_identifier) @property
(shorthand_property_identifier) @property
(shorthand_property_identifier_pattern) @property
; Function and method calls

View File

@@ -229,6 +229,7 @@ impl PickerDelegate for RecentProjectsDelegate {
.workspaces
.iter()
.enumerate()
.filter(|(_, (id, _))| !self.is_current_workspace(*id, cx))
.map(|(id, (_, location))| {
let combined_string = match location {
SerializedWorkspaceLocation::Local(paths, _) => paths
@@ -393,8 +394,7 @@ impl PickerDelegate for RecentProjectsDelegate {
return None;
};
let (workspace_id, location) = &self.workspaces[hit.candidate_id];
let is_current_workspace = self.is_current_workspace(*workspace_id, cx);
let (_, location) = self.workspaces.get(hit.candidate_id)?;
let is_remote = matches!(location, SerializedWorkspaceLocation::DevServer(_));
let dev_server_status =
@@ -487,7 +487,7 @@ impl PickerDelegate for RecentProjectsDelegate {
highlighted.render(cx)
}),
)
.when(!is_current_workspace, |el| {
.map(|el| {
let delete_button = div()
.child(
IconButton::new("delete", IconName::Close)

View File

@@ -159,6 +159,7 @@ message Envelope {
SetChannelMemberRole set_channel_member_role = 123;
RenameChannel rename_channel = 124;
RenameChannelResponse rename_channel_response = 125;
SubscribeToChannels subscribe_to_channels = 207; // current max
JoinChannelBuffer join_channel_buffer = 126;
JoinChannelBufferResponse join_channel_buffer_response = 127;
@@ -250,7 +251,7 @@ message Envelope {
TaskContextForLocation task_context_for_location = 203;
TaskContext task_context = 204;
TaskTemplatesResponse task_templates_response = 205;
TaskTemplates task_templates = 206; // Current max
TaskTemplates task_templates = 206;
}
reserved 158 to 161;
@@ -1297,6 +1298,8 @@ message ChannelMember {
}
}
message SubscribeToChannels {}
message CreateChannel {
string name = 1;
optional uint64 parent_id = 2;

View File

@@ -277,6 +277,7 @@ messages!(
(ShareProjectResponse, Foreground),
(ShowContacts, Foreground),
(StartLanguageServer, Foreground),
(SubscribeToChannels, Foreground),
(SynchronizeBuffers, Foreground),
(SynchronizeBuffersResponse, Foreground),
(TaskContextForLocation, Background),

View File

@@ -0,0 +1,75 @@
use std::cell::RefCell;
use std::collections::HashSet;
use std::sync::OnceLock;
use html5ever::Attribute;
/// Returns a [`HashSet`] containing the HTML elements that are inline by default.
///
/// [MDN: List of "inline" elements](https://yari-demos.prod.mdn.mozit.cloud/en-US/docs/Web/HTML/Inline_elements)
fn inline_elements() -> &'static HashSet<&'static str> {
static INLINE_ELEMENTS: OnceLock<HashSet<&str>> = OnceLock::new();
&INLINE_ELEMENTS.get_or_init(|| {
HashSet::from_iter([
"a", "abbr", "acronym", "audio", "b", "bdi", "bdo", "big", "br", "button", "canvas",
"cite", "code", "data", "datalist", "del", "dfn", "em", "embed", "i", "iframe", "img",
"input", "ins", "kbd", "label", "map", "mark", "meter", "noscript", "object", "output",
"picture", "progress", "q", "ruby", "s", "samp", "script", "select", "slot", "small",
"span", "strong", "sub", "sup", "svg", "template", "textarea", "time", "tt", "u",
"var", "video", "wbr",
])
})
}
#[derive(Debug, Clone)]
pub struct HtmlElement {
pub(crate) tag: String,
pub(crate) attrs: RefCell<Vec<Attribute>>,
}
impl HtmlElement {
/// Returns whether this [`HtmlElement`] is an inline element.
pub fn is_inline(&self) -> bool {
inline_elements().contains(self.tag.as_str())
}
/// Returns the attribute with the specified name.
pub fn attr(&self, name: &str) -> Option<String> {
self.attrs
.borrow()
.iter()
.find(|attr| attr.name.local.to_string() == name)
.map(|attr| attr.value.to_string())
}
/// Returns the list of classes on this [`HtmlElement`].
pub fn classes(&self) -> Vec<String> {
self.attrs
.borrow()
.iter()
.find(|attr| attr.name.local.to_string() == "class")
.map(|attr| {
attr.value
.split(' ')
.map(|class| class.trim().to_string())
.collect::<Vec<_>>()
})
.unwrap_or_default()
}
/// Returns whether this [`HtmlElement`] has the specified class.
pub fn has_class(&self, class: &str) -> bool {
self.has_any_classes(&[class])
}
/// Returns whether this [`HtmlElement`] has any of the specified classes.
pub fn has_any_classes(&self, classes: &[&str]) -> bool {
self.attrs.borrow().iter().any(|attr| {
attr.name.local.to_string() == "class"
&& attr
.value
.split(' ')
.any(|class| classes.contains(&class.trim()))
})
}
}

View File

@@ -1,12 +1,12 @@
use std::cell::RefCell;
use std::collections::VecDeque;
use std::sync::OnceLock;
use anyhow::Result;
use html5ever::Attribute;
use markup5ever_rcdom::{Handle, NodeData};
use regex::Regex;
use crate::html_element::HtmlElement;
fn empty_line_regex() -> &'static Regex {
static REGEX: OnceLock<Regex> = OnceLock::new();
REGEX.get_or_init(|| Regex::new(r"^\s*$").unwrap())
@@ -17,11 +17,7 @@ fn more_than_three_newlines_regex() -> &'static Regex {
REGEX.get_or_init(|| Regex::new(r"\n{3,}").unwrap())
}
#[derive(Debug, Clone)]
struct HtmlElement {
tag: String,
attrs: RefCell<Vec<Attribute>>,
}
const RUSTDOC_ITEM_NAME_CLASS: &str = "item-name";
enum StartTagOutcome {
Continue,
@@ -135,6 +131,16 @@ impl MarkdownWriter {
}
fn start_tag(&mut self, tag: &HtmlElement) -> StartTagOutcome {
if tag.is_inline() && self.is_inside("p") {
if let Some(parent) = self.current_element_stack.iter().last() {
if !parent.is_inline() {
if !(self.markdown.ends_with(' ') || self.markdown.ends_with('\n')) {
self.push_str(" ");
}
}
}
}
match tag.tag.as_str() {
"head" | "script" | "nav" => return StartTagOutcome::Skip,
"h1" => self.push_str("\n\n# "),
@@ -143,24 +149,17 @@ impl MarkdownWriter {
"h4" => self.push_str("\n\n#### "),
"h5" => self.push_str("\n\n##### "),
"h6" => self.push_str("\n\n###### "),
"p" => self.push_blank_line(),
"strong" => self.push_str("**"),
"em" => self.push_str("_"),
"code" => {
if !self.is_inside("pre") {
self.push_str("`")
self.push_str("`");
}
}
"pre" => {
let attrs = tag.attrs.borrow();
let classes = attrs
.iter()
.find(|attr| attr.name.local.to_string() == "class")
.map(|attr| {
attr.value
.split(' ')
.map(|class| class.trim())
.collect::<Vec<_>>()
})
.unwrap_or_default();
let is_rust = classes.iter().any(|class| class == &"rust");
let classes = tag.classes();
let is_rust = classes.iter().any(|class| class == "rust");
let language = is_rust
.then(|| "rs")
.or_else(|| {
@@ -174,7 +173,7 @@ impl MarkdownWriter {
})
.unwrap_or("");
self.push_str(&format!("\n\n```{language}\n"))
self.push_str(&format!("\n\n```{language}\n"));
}
"ul" | "ol" => self.push_newline(),
"li" => self.push_str("- "),
@@ -198,29 +197,23 @@ impl MarkdownWriter {
self.push_str("| ");
}
"summary" => {
if tag.attrs.borrow().iter().any(|attr| {
attr.name.local.to_string() == "class" && attr.value.to_string() == "hideme"
}) {
if tag.has_class("hideme") {
return StartTagOutcome::Skip;
}
}
"button" => {
if tag.attr("id").as_deref() == Some("copy-path") {
return StartTagOutcome::Skip;
}
}
"div" | "span" => {
let classes_to_skip = ["nav-container", "sidebar-elems", "out-of-band"];
if tag.attrs.borrow().iter().any(|attr| {
attr.name.local.to_string() == "class"
&& attr
.value
.split(' ')
.any(|class| classes_to_skip.contains(&class.trim()))
}) {
if tag.has_any_classes(&classes_to_skip) {
return StartTagOutcome::Skip;
}
if tag.attrs.borrow().iter().any(|attr| {
attr.name.local.to_string() == "class" && attr.value.to_string() == "item-name"
}) {
self.push_str("`");
if self.is_inside_item_name() && tag.has_class("stab") {
self.push_str(" [");
}
}
_ => {}
@@ -232,9 +225,11 @@ impl MarkdownWriter {
fn end_tag(&mut self, tag: &HtmlElement) {
match tag.tag.as_str() {
"h1" | "h2" | "h3" | "h4" | "h5" | "h6" => self.push_str("\n\n"),
"strong" => self.push_str("**"),
"em" => self.push_str("_"),
"code" => {
if !self.is_inside("pre") {
self.push_str("`")
self.push_str("`");
}
}
"pre" => self.push_str("\n```\n"),
@@ -258,11 +253,13 @@ impl MarkdownWriter {
"table" => {
self.current_table_columns = 0;
}
"div" => {
if tag.attrs.borrow().iter().any(|attr| {
attr.name.local.to_string() == "class" && attr.value.to_string() == "item-name"
}) {
self.push_str("`: ");
"div" | "span" => {
if tag.has_class(RUSTDOC_ITEM_NAME_CLASS) {
self.push_str(": ");
}
if self.is_inside_item_name() && tag.has_class("stab") {
self.push_str("]");
}
}
_ => {}
@@ -275,9 +272,25 @@ impl MarkdownWriter {
return Ok(());
}
let trimmed_text = text.trim_matches(|char| char == '\n' || char == '\r' || char == '§');
self.push_str(trimmed_text);
let text = text
.trim_matches(|char| char == '\n' || char == '\r' || char == '§')
.replace('\n', " ");
if self.is_inside_item_name() && !self.is_inside("span") && !self.is_inside("code") {
self.push_str(&format!("`{text}`"));
return Ok(());
}
self.push_str(&text);
Ok(())
}
/// Returns whether we're currently inside of an `.item-name` element, which
/// rustdoc uses to display Rust items in a list.
fn is_inside_item_name(&self) -> bool {
self.current_element_stack
.iter()
.any(|element| element.has_class(RUSTDOC_ITEM_NAME_CLASS))
}
}

View File

@@ -2,6 +2,7 @@
#![deny(missing_docs)]
mod html_element;
mod markdown_writer;
use std::io::Read;
@@ -44,6 +45,112 @@ mod tests {
use super::*;
#[test]
fn test_main_heading_buttons_get_removed() {
let html = indoc! {r##"
<div class="main-heading">
<h1>Crate <a class="mod" href="#">serde</a><button id="copy-path" title="Copy item path to clipboard">Copy item path</button></h1>
<span class="out-of-band">
<a class="src" href="../src/serde/lib.rs.html#1-340">source</a> · <button id="toggle-all-docs" title="collapse all docs">[<span></span>]</button>
</span>
</div>
"##};
let expected = indoc! {"
# Crate serde
"}
.trim();
assert_eq!(
convert_rustdoc_to_markdown(html.as_bytes()).unwrap(),
expected
)
}
#[test]
fn test_single_paragraph() {
let html = indoc! {r#"
<p>In particular, the last point is what sets <code>axum</code> apart from other frameworks.
<code>axum</code> doesnt have its own middleware system but instead uses
<a href="https://docs.rs/tower-service/0.3.2/x86_64-unknown-linux-gnu/tower_service/trait.Service.html" title="trait tower_service::Service"><code>tower::Service</code></a>. This means <code>axum</code> gets timeouts, tracing, compression,
authorization, and more, for free. It also enables you to share middleware with
applications written using <a href="http://crates.io/crates/hyper"><code>hyper</code></a> or <a href="http://crates.io/crates/tonic"><code>tonic</code></a>.</p>
"#};
let expected = indoc! {"
In particular, the last point is what sets `axum` apart from other frameworks. `axum` doesnt have its own middleware system but instead uses `tower::Service`. This means `axum` gets timeouts, tracing, compression, authorization, and more, for free. It also enables you to share middleware with applications written using `hyper` or `tonic`.
"}
.trim();
assert_eq!(
convert_rustdoc_to_markdown(html.as_bytes()).unwrap(),
expected
)
}
#[test]
fn test_multiple_paragraphs() {
let html = indoc! {r##"
<h2 id="serde"><a class="doc-anchor" href="#serde">§</a>Serde</h2>
<p>Serde is a framework for <em><strong>ser</strong></em>ializing and <em><strong>de</strong></em>serializing Rust data
structures efficiently and generically.</p>
<p>The Serde ecosystem consists of data structures that know how to serialize
and deserialize themselves along with data formats that know how to
serialize and deserialize other things. Serde provides the layer by which
these two groups interact with each other, allowing any supported data
structure to be serialized and deserialized using any supported data format.</p>
<p>See the Serde website <a href="https://serde.rs/">https://serde.rs/</a> for additional documentation and
usage examples.</p>
<h3 id="design"><a class="doc-anchor" href="#design">§</a>Design</h3>
<p>Where many other languages rely on runtime reflection for serializing data,
Serde is instead built on Rusts powerful trait system. A data structure
that knows how to serialize and deserialize itself is one that implements
Serdes <code>Serialize</code> and <code>Deserialize</code> traits (or uses Serdes derive
attribute to automatically generate implementations at compile time). This
avoids any overhead of reflection or runtime type information. In fact in
many situations the interaction between data structure and data format can
be completely optimized away by the Rust compiler, leaving Serde
serialization to perform the same speed as a handwritten serializer for the
specific selection of data structure and data format.</p>
"##};
let expected = indoc! {"
## Serde
Serde is a framework for _**ser**_ializing and _**de**_serializing Rust data structures efficiently and generically.
The Serde ecosystem consists of data structures that know how to serialize and deserialize themselves along with data formats that know how to serialize and deserialize other things. Serde provides the layer by which these two groups interact with each other, allowing any supported data structure to be serialized and deserialized using any supported data format.
See the Serde website https://serde.rs/ for additional documentation and usage examples.
### Design
Where many other languages rely on runtime reflection for serializing data, Serde is instead built on Rusts powerful trait system. A data structure that knows how to serialize and deserialize itself is one that implements Serdes `Serialize` and `Deserialize` traits (or uses Serdes derive attribute to automatically generate implementations at compile time). This avoids any overhead of reflection or runtime type information. In fact in many situations the interaction between data structure and data format can be completely optimized away by the Rust compiler, leaving Serde serialization to perform the same speed as a handwritten serializer for the specific selection of data structure and data format.
"}
.trim();
assert_eq!(
convert_rustdoc_to_markdown(html.as_bytes()).unwrap(),
expected
)
}
#[test]
fn test_styled_text() {
let html = indoc! {r#"
<p>This text is <strong>bolded</strong>.</p>
<p>This text is <em>italicized</em>.</p>
"#};
let expected = indoc! {"
This text is **bolded**.
This text is _italicized_.
"}
.trim();
assert_eq!(
convert_rustdoc_to_markdown(html.as_bytes()).unwrap(),
expected
)
}
#[test]
fn test_rust_code_block() {
let html = indoc! {r#"
@@ -118,6 +225,42 @@ mod tests {
)
}
#[test]
fn test_item_table() {
let html = indoc! {r##"
<h2 id="structs" class="section-header">Structs<a href="#structs" class="anchor">§</a></h2>
<ul class="item-table">
<li><div class="item-name"><a class="struct" href="struct.Error.html" title="struct axum::Error">Error</a></div><div class="desc docblock-short">Errors that can happen when using axum.</div></li>
<li><div class="item-name"><a class="struct" href="struct.Extension.html" title="struct axum::Extension">Extension</a></div><div class="desc docblock-short">Extractor and response for extensions.</div></li>
<li><div class="item-name"><a class="struct" href="struct.Form.html" title="struct axum::Form">Form</a><span class="stab portability" title="Available on crate feature `form` only"><code>form</code></span></div><div class="desc docblock-short">URL encoded extractor and response.</div></li>
<li><div class="item-name"><a class="struct" href="struct.Json.html" title="struct axum::Json">Json</a><span class="stab portability" title="Available on crate feature `json` only"><code>json</code></span></div><div class="desc docblock-short">JSON Extractor / Response.</div></li>
<li><div class="item-name"><a class="struct" href="struct.Router.html" title="struct axum::Router">Router</a></div><div class="desc docblock-short">The router type for composing handlers and services.</div></li></ul>
<h2 id="functions" class="section-header">Functions<a href="#functions" class="anchor">§</a></h2>
<ul class="item-table">
<li><div class="item-name"><a class="fn" href="fn.serve.html" title="fn axum::serve">serve</a><span class="stab portability" title="Available on crate feature `tokio` and (crate features `http1` or `http2`) only"><code>tokio</code> and (<code>http1</code> or <code>http2</code>)</span></div><div class="desc docblock-short">Serve the service with the supplied listener.</div></li>
</ul>
"##};
let expected = indoc! {r#"
## Structs
- `Error`: Errors that can happen when using axum.
- `Extension`: Extractor and response for extensions.
- `Form` [`form`]: URL encoded extractor and response.
- `Json` [`json`]: JSON Extractor / Response.
- `Router`: The router type for composing handlers and services.
## Functions
- `serve` [`tokio` and (`http1` or `http2`)]: Serve the service with the supplied listener.
"#}
.trim();
assert_eq!(
convert_rustdoc_to_markdown(html.as_bytes()).unwrap(),
expected
)
}
#[test]
fn test_table() {
let html = indoc! {r##"
@@ -144,8 +287,9 @@ mod tests {
let expected = indoc! {r#"
## Feature flags
axum uses a set of feature flags to reduce the amount of compiled and
optional dependencies.The following optional features are available:
axum uses a set of feature flags to reduce the amount of compiled and optional dependencies.
The following optional features are available:
| Name | Description | Default? |
| --- | --- | --- |

View File

@@ -78,8 +78,6 @@ pub const FS_WATCH_LATENCY: Duration = Duration::from_millis(100);
#[cfg(not(feature = "test-support"))]
pub const FS_WATCH_LATENCY: Duration = Duration::from_millis(100);
const GIT_STATUS_UPDATE_BATCH_SIZE: usize = 1024;
#[derive(Copy, Clone, PartialEq, Eq, Debug, Hash, PartialOrd, Ord)]
pub struct WorktreeId(usize);
@@ -4293,7 +4291,7 @@ impl BackgroundScanner {
async fn update_git_repositories(&self, dot_git_paths: Vec<PathBuf>) {
log::debug!("reloading repositories: {dot_git_paths:?}");
let (update_job_tx, update_job_rx) = channel::unbounded();
let mut repo_updates = Vec::new();
{
let mut state = self.state.lock();
let scan_id = state.snapshot.scan_id;
@@ -4308,7 +4306,7 @@ impl BackgroundScanner {
.then(|| (*entry_id, repo.clone()))
});
let (work_dir, repository) = match existing_repository_entry {
let (work_directory, repository) = match existing_repository_entry {
None => {
match state.build_git_repository(dot_git_dir.into(), self.fs.as_ref()) {
Some(output) => output,
@@ -4327,7 +4325,6 @@ impl BackgroundScanner {
continue;
};
log::info!("reload git repository {dot_git_dir:?}");
let repo = &repository.repo_ptr;
let branch = repo.branch_name();
repo.reload_index();
@@ -4345,41 +4342,16 @@ impl BackgroundScanner {
}
};
let statuses = repository
.statuses(Path::new(""))
.log_err()
.unwrap_or_default();
let entries = state.snapshot.entries_by_path.clone();
let location_in_repo = state
.snapshot
.repository_entries
.get(&work_dir)
.and_then(|repo| repo.location_in_repo.clone());
let mut files =
state
repo_updates.push(UpdateGitStatusesJob {
location_in_repo: state
.snapshot
.traverse_from_path(true, false, false, work_dir.0.as_ref());
let mut start_path = work_dir.0.clone();
while start_path.starts_with(&work_dir.0) {
files.advance_by(GIT_STATUS_UPDATE_BATCH_SIZE);
let end_path = files.entry().map(|e| e.path.clone());
smol::block_on(update_job_tx.send(UpdateGitStatusesJob {
start_path: start_path.clone(),
end_path: end_path.clone(),
entries: entries.clone(),
location_in_repo: location_in_repo.clone(),
containing_repository: ScanJobContainingRepository {
work_directory: work_dir.clone(),
statuses: statuses.clone(),
},
}))
.unwrap();
if let Some(end_path) = end_path {
start_path = end_path;
} else {
break;
}
}
.repository_entries
.get(&work_directory)
.and_then(|repo| repo.location_in_repo.clone())
.clone(),
work_directory,
repository,
});
}
// Remove any git repositories whose .git entry no longer exists.
@@ -4414,87 +4386,92 @@ impl BackgroundScanner {
.repository_entries
.retain(|_, entry| ids_to_preserve.contains(&entry.work_directory.0));
}
drop(update_job_tx);
let (mut updates_done_tx, mut updates_done_rx) = barrier::channel();
self.executor
.scoped(|scope| {
for _ in 0..self.executor.num_cpus() {
scope.spawn(async {
loop {
select_biased! {
// Process any path refresh requests before moving on to process
// the queue of git statuses.
request = self.scan_requests_rx.recv().fuse() => {
let Ok(request) = request else { break };
if !self.process_scan_request(request, true).await {
return;
}
}
scope.spawn(async {
for repo_update in repo_updates {
self.update_git_statuses(repo_update);
}
updates_done_tx.blocking_send(()).ok();
});
// Process git status updates in batches.
job = update_job_rx.recv().fuse() => {
let Ok(job) = job else { break };
self.update_git_statuses(job);
scope.spawn(async {
loop {
select_biased! {
// Process any path refresh requests before moving on to process
// the queue of git statuses.
request = self.scan_requests_rx.recv().fuse() => {
let Ok(request) = request else { break };
if !self.process_scan_request(request, true).await {
return;
}
}
_ = updates_done_rx.recv().fuse() => break,
}
});
}
}
});
})
.await;
}
/// Update the git statuses for a given batch of entries.
fn update_git_statuses(&self, job: UpdateGitStatusesJob) {
// Determine which entries in this batch have changed their git status.
log::trace!("updating git statuses for repo {:?}", job.work_directory.0);
let t0 = Instant::now();
let mut edits = Vec::new();
for entry in Traversal::new(&job.entries, true, false, false, &job.start_path) {
if job
.end_path
.as_ref()
.map_or(false, |end| &entry.path >= end)
{
let Some(statuses) = job.repository.statuses(Path::new("")).log_err() else {
return;
};
log::trace!(
"computed git statuses for repo {:?} in {:?}",
job.work_directory.0,
t0.elapsed()
);
let t0 = Instant::now();
let mut changes = Vec::new();
let snapshot = self.state.lock().snapshot.snapshot.clone();
for file in snapshot.traverse_from_path(true, false, false, job.work_directory.0.as_ref()) {
let Ok(repo_path) = file.path.strip_prefix(&job.work_directory.0) else {
break;
}
let Ok(repo_path) = entry
.path
.strip_prefix(&job.containing_repository.work_directory)
else {
continue;
};
let repo_path = RepoPath(if let Some(location) = &job.location_in_repo {
location.join(repo_path)
let git_status = if let Some(location) = &job.location_in_repo {
statuses.get(&location.join(repo_path))
} else {
repo_path.to_path_buf()
});
let git_status = job.containing_repository.statuses.get(&repo_path);
if entry.git_status != git_status {
let mut entry = entry.clone();
statuses.get(&repo_path)
};
if file.git_status != git_status {
let mut entry = file.clone();
entry.git_status = git_status;
edits.push(Edit::Insert(entry));
changes.push((entry.path, git_status));
}
}
let mut state = self.state.lock();
let edits = changes
.iter()
.filter_map(|(path, git_status)| {
let entry = state.snapshot.entry_for_path(path)?.clone();
Some(Edit::Insert(Entry {
git_status: *git_status,
..entry.clone()
}))
})
.collect();
// Apply the git status changes.
if edits.len() > 0 {
let mut state = self.state.lock();
let path_changes = edits.iter().map(|edit| {
if let Edit::Insert(entry) = edit {
entry.path.clone()
} else {
unreachable!()
}
});
util::extend_sorted(&mut state.changed_paths, path_changes, usize::MAX, Ord::cmp);
state.snapshot.entries_by_path.edit(edits, &());
}
util::extend_sorted(
&mut state.changed_paths,
changes.iter().map(|p| p.0.clone()),
usize::MAX,
Ord::cmp,
);
state.snapshot.entries_by_path.edit(edits, &());
log::trace!(
"refreshed git status of entries starting with {} in {:?}",
// entries.len(),
job.start_path.display(),
t0.elapsed()
"applied git status updates for repo {:?} in {:?}",
job.work_directory.0,
t0.elapsed(),
);
}
@@ -4664,11 +4641,9 @@ struct UpdateIgnoreStatusJob {
}
struct UpdateGitStatusesJob {
entries: SumTree<Entry>,
start_path: Arc<Path>,
end_path: Option<Arc<Path>>,
containing_repository: ScanJobContainingRepository,
work_directory: RepositoryWorkDirectory,
location_in_repo: Option<Arc<Path>>,
repository: Arc<dyn GitRepository>,
}
pub trait WorktreeModelHandle {

View File

@@ -110,14 +110,15 @@ The name of any font family installed on the user's system
**Options**
Zed supports a subset of OpenType features that can be enabled or disabled for a given buffer or terminal font. The following [OpenType features](https://en.wikipedia.org/wiki/List_of_typographic_features) can be enabled or disabled too: `calt`, `case`, `cpsp`, `frac`, `liga`, `onum`, `ordn`, `pnum`, `ss01`, `ss02`, `ss03`, `ss04`, `ss05`, `ss06`, `ss07`, `ss08`, `ss09`, `ss10`, `ss11`, `ss12`, `ss13`, `ss14`, `ss15`, `ss16`, `ss17`, `ss18`, `ss19`, `ss20`, `subs`, `sups`, `swsh`, `titl`, `tnum`, `zero`.
Zed supports all OpenType features that can be enabled, disabled or set a value to a font feature for a given buffer or terminal font.
For example, to disable ligatures for a given font you can add the following to your settings:
For example, to disable ligatures and set `7` to `cv01` for a given font you can add the following to your settings:
```json
{
"buffer_font_features": {
"calt": false
"calt": false,
"cv01": 7
}
}
```

View File

@@ -11,18 +11,18 @@
"args": ["test", "--failed"]
},
{
"label": "mix test $ZED_SYMBOL",
"label": "mix test $ZED_RELATIVE_FILE",
"command": "mix",
"args": ["test", "$ZED_SYMBOL"]
"args": ["test", "$ZED_RELATIVE_FILE"]
},
{
"label": "mix test $ZED_FILE:$ZED_ROW",
"label": "mix test $ZED_RELATIVE_FILE:$ZED_ROW",
"command": "mix",
"args": ["test", "$ZED_FILE:$ZED_ROW"]
"args": ["test", "$ZED_RELATIVE_FILE:$ZED_ROW"]
},
{
"label": "Elixir: break line",
"command": "iex",
"args": ["-S", "mix", "test", "-b", "$ZED_FILE:$ZED_ROW"]
"args": ["-S", "mix", "test", "-b", "$ZED_RELATIVE_FILE:$ZED_ROW"]
}
]

View File

@@ -2,6 +2,7 @@
(class "end" @end) @indent
(module "end" @end) @indent
(begin "end" @end) @indent
(singleton_method "end" @end) @indent
(do_block "end" @end) @indent
(then) @indent

View File

@@ -0,0 +1,8 @@
(heredoc_body
(heredoc_content) @content
(heredoc_end) @language
(#downcase! @language))
((regex
(string_content) @content)
(#set! "language" "regex"))

View File

@@ -1,2 +0,0 @@
(heredoc_body
(heredoc_end) @language) @content