`pwsh` is the newer version of `PowerShell`, while the one that comes
pre-installed on Windows is called `Windows PowerShell` and is an older
version. I have no idea why Microsoft dose this and not updated the
`Windows Powershell` on Windows.
Release Notes:
- N/A
I have my system temp dir on a different drive than the default, so this
error was spammed in the logs. This also broke Zed in many ways, one of
which was the AI system failing to work since it couldn't save settings.
```
2024-08-20T22:39:54.0660708-07:00 [ERROR] Failed to write settings to file "\\\\?\\C:\\Users\\myuser\\AppData\\Roaming\\Zed\\settings.json"
Caused by:
0: failed to persist temporary file: The system cannot move the file to a different disk drive. (os error 17)
1: The system cannot move the file to a different disk drive. (os error 17)
```
Note: This problem is probably present on MacOS due to the requirement
of the underlying api being used. I do not have Mac, so I cannot test
this. This PR only solves this issue on Windows.
Closes#16571
Release Notes:
- fix atomic_write failing on windows if destination is on a different
drive than the OS's temp dir.
Closes#14700#8164
Release Notes:
- Added `soft_wrap` value `bounded`,EditorWidth and PreferredLineLength
min value
---------
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
This PR makes it so the model's cache configuration gets passed through
from the base model when using the Zed provider.
Release Notes:
- Fixed caching for Anthropic models when using the Zed provider.
This adds an option to set the split direction for both the horizontal
splits, and the vertical splits.
A couple of things to look for when reviewing:
* The `derive` keywords on the Enums were copy pasted, no clue what they
should be
* Tried adding tests for this, but got stuck.
Co-authored with @Tobbe
Fixes: https://github.com/zed-industries/zed/issues/11342
Allows language server logs to be published prior to the completion of
the initialize request. OmniSharp is one example of an LSP that
publishes (many) messages prior to the initialization response, and this
completely floods the Zed logs.
Also adds level filtering as demonstrated below. Again, this is due to
my experience with the massive amount of log messages that OmniSharp
publishes.
Release Notes:
- Added level filtering to language server logs
---------
Co-authored-by: Thorsten Ball <mrnugget@gmail.com>
Partially addresses #8497 (namely, the occurring with `delta`)
As I mentioned in
https://github.com/zed-industries/zed/issues/8497#issuecomment-2226896371,
zed currently replies to OSC color requests (`OSC 10`, `OSC 11`, ...)
out of order when immediately followed by another request (for example
`CSI c`). All other terminals that [I have
tested](https://github.com/bash/terminal-colorsaurus/blob/main/doc/terminal-survey.md)
maintain relative order when replying to requests.
## Solution
Respond to the `ColorRequest` in `process_event` (in the same place
where other PTY writes happen) instead of queuing it up in the internal
event queue.
## Alternative
I initially thought that I could handle the color request similarly to
the `TextAreaSizeRequest` where the size is stored in `last_content` and
updated on `sync`. However this causes the terminal to report
out-of-date values when a "set color" sequence is followed by a color
request.
## Tests
1. `OSC 11; ?` (request bg color) + `CSI c` (request device attributes):
```shell
printf '\e]11;?\e\\ \e[c' && cat -v
# Expected result: ^[]11;rgb:dcdc/dcdc/dddd^[\^[[?6c
# Current result: ^[[?6c^[]11;rgb:dcdc/dcdc/dddd^[\ (❌)
# Result with this PR: ^[]11;rgb:dcdc/dcdc/dddd^[\^[[?6c (✅)
# Result with alternative: ^[]11;rgb:dcdc/dcdc/dddd^[\^[[?6c (✅)
```
2. `OSC 11; rgb:f0f0/f0f0/f0f0` (set bg color) + `OSC 11; ?` (request bg
color)
```shell
printf '\e]11;rgb:f0f0/f0f0/f0f0\e\\ \e]11;?\e\\' && cat -v
# Expected result: ^[]11;rgb:f0f0/f0f0/f0f0^[\
# Current result: ^[]11;rgb:f0f0/f0f0/f0f0^[\ (✅)
# Result with this PR: ^[]11;rgb:f0f0/f0f0/f0f0^[\ (✅)
# Result with alternative: ^[]11;rgb:OUT_OF_DATE_COLOR_HERE^[\ (❌)
```
Release Notes:
- N/A
Closes#16555
Release Notes:
- Improved the "Copy Path" and "Copy Relative Path" actions in the
project panel's context menu when selecting multiple files. All selected
files' paths will now be copied, separated by newlines.
Release Notes:
- vim: Added `gf` command to open files under the cursor.
- Filenames can now be `cmd`/`ctrl`-clicked, which opens them.
TODOs:
- [x] `main_test.go` <-- works
- [x] `./my-pkg/my_pkg.go` <-- works
- [x] `../go.mod` <-- works
- [x] `my-pkg/my_pkg.go` <-- works
- [x] `my-pkg/subpkg/subpkg_test.go` <-- works
- [x] `file\ with\ space\ in\ it.txt` <-- works
- [x] `"file\ with\ space\ in\ it.txt"` <-- works
- [x] `"main_test.go"` <-- works
- [x] `/Users/thorstenball/.vimrc` <-- works, but only locally
- [x] `~/.vimrc` <--works, but only locally
- [x] Get it working over collab
- [x] Get hover links working
Demo:
https://github.com/user-attachments/assets/26af7f3b-c392-4aaf-849a-95d6c3b00067
Collab demo:
https://github.com/user-attachments/assets/272598bd-0e82-4556-8f9c-ba53d3a95682
- Add references to locations in code for Metrics and Panic telemetry
- Remove outdated documentation (ClickhouseEvent,
ClickhouseEventWrapper, ClickhouseEventRequestBody)
- Migrate struct documentation from web docs to inline doc comments on
struct members.
This PR adds additional reporting of the active user counts as separate
logs.
We were already reporting these on individual rate limit events/logs,
but it seems like something that would be good to report on independent
of user activity.
Release Notes:
- N/A
This PR fixes an issue where the active user count spanned individual
models.
We now track the active user counts on a per-model basis.
Release Notes:
- N/A
Release Notes:
- Added switch source/header action for clangd language server (fixes
[#12801](https://github.com/zed-industries/zed/issues/12801)).
Note: I'm new to both rust and this codebase. I started my
implementation by copying how rust analyzer's "expand macro" LSP
extension is implemented. I don't yet understand some of the code I
copied (mostly the way to get the `server_to_query` in `clangd_ext.rs`
and the whole proto implementation).
---------
Co-authored-by: Kirill Bulatov <kirill@zed.dev>
This PR adds traces for when users hit LLM rate limits.
We were already emitting telemetry events for these to Clickhouse, but
it will be handy to have them available in Axiom as well.
Release Notes:
- N/A
- Implement restart kernel functionality
- Clean up shutdown process to properly drop messaging and exit status
tasks
- Refactor kernel state handling for better consistency
Closes#16037
Release Notes:
- repl: Added restart kernel action
- repl: Fixed issue with shutting down kernels that are in a failure
state
### Pull Request Title
Introduce `max_output_tokens` Field for OpenAI Models
https://platform.deepseek.com/api-docs/news/news0725/#4-8k-max_tokens-betarelease-longer-possibilities
### Description
This commit introduces a new field `max_output_tokens` to the OpenAI
models, which allows specifying the maximum number of tokens that can be
generated in the output. This field is now integrated into the request
handling across multiple crates, ensuring that the output token limit is
respected during language model completions.
Changes include:
- Adding `max_output_tokens` to the `Custom` variant of the
`open_ai::Model` enum.
- Updating the `into_open_ai` method in `LanguageModelRequest` to accept
and use `max_output_tokens`.
- Modifying the `OpenAiLanguageModel` and `CloudLanguageModel`
implementations to pass `max_output_tokens` when converting requests.
- Ensuring that the `max_output_tokens` field is correctly serialized
and deserialized in relevant structures.
This enhancement provides more control over the output length of OpenAI
model responses, improving the flexibility and accuracy of language
model interactions.
### Changes
- Added `max_output_tokens` to the `Custom` variant of the
`open_ai::Model` enum.
- Updated the `into_open_ai` method in `LanguageModelRequest` to accept
and use `max_output_tokens`.
- Modified the `OpenAiLanguageModel` and `CloudLanguageModel`
implementations to pass `max_output_tokens` when converting requests.
- Ensured that the `max_output_tokens` field is correctly serialized and
deserialized in relevant structures.
### Related Issue
https://github.com/zed-industries/zed/pull/16358
### Screenshots / Media
N/A
### Checklist
- [x] Code compiles correctly.
- [x] All tests pass.
- [ ] Documentation has been updated accordingly.
- [ ] Additional tests have been added to cover new functionality.
- [ ] Relevant documentation has been updated or added.
### Release Notes
- Added `max_output_tokens` field to OpenAI models for controlling
output token length.
Closes#13579
A major painpoint in the Vim crate has been life-cycle management. We
used to have one global Vim instance that tried to track per-editor
state; this led to a number of subtle issues (e.g. #13579, the mode
indicator being global, and quick toggling between windows letting vim
mode's notion of the active editor get out of sync).
This PR changes the internal structure of the code so that there is now
one `Vim` instance per `Editor` (stored as an `Addon`); and the global
stuff is separated out. This fixes the above problems, and tidies up a
bunch of the mess in the codebase.
Release Notes:
* vim: Fixed accidental visual mode in project search and go to
references
([#13579](https://github.com/zed-industries/zed/issues/13579)).
Closes#15860
Since rust std now supports LazyLock replacing lazy_static with it
reduce the external dependency.
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR allows configuring the intensity of code fade applied to unused
code (relates to #4785).
_Note: Right now I included it as a top level config which might be a
little out of place (but is easiest to instrument). Open for suggestions
on where else it would be better suited for._
_Note 2: I am unfamiliar with the codebase. Feel free to either close
this PR and re-implement in a better way or suggest high level changes
if I'm approaching anything wrong :)._
Release Notes:
- Added `unnecesary_code_fade` setting to control how strongly to fade
unused code.
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR can improve the UX when dropping a file that cannot be opened
into the workspace pane. Previously, nothing happened without any
messages when such error occurred, which could be awkward for users.
Additionally the pane was being split even though the file failed to
open.
Here's a screen recording demonstrating the previous/updated behavior:
https://github.com/user-attachments/assets/cfdf3488-9464-4568-b16a-9b87718bd729
Changes:
- It now displays an error message if a file cannot be opened.
- Updated the logic to first try to open the file. The pane splits only
if the file opening process is successful.
Release Notes:
- Improved error handling when opening files in the workspace pane. An
error message will now be displayed if the file cannot be opened.
- Fixed an issue where unnecessary pane splitting occurred when a file
fails to open.
This PR adds two Postgrest containers—one for the app database and one
for the LLM database—to the Docker Compose cluster.
Also fixed an issue where `postgres_app.conf` and `postgres_llm.conf`
had been switched.
Release Notes:
- N/A
This PR is a small refactor to remove the leading `_` for some free
variables, as this unintentionally marks them as unused to the compiler.
While the fields on the struct _are_ unused, the free variables should
participate in usage tracking, as we want to make sure they get stored
on the struct.
Release Notes:
- N/A
Follow-up to https://github.com/zed-industries/zed/pull/16471
* Don't expand the default prompt by default, since it looks strange in
the expanded state
* Don't create two `You` headers by default. Just insert a blank line
after the default prompt.
Release Notes:
- N/A
Release Notes:
- Adds UI affordances to the assistant panel to show which messages have
been cached
- Migrate cache invalidation to be based on `has_edits_since_in_range`
to be smarter and more selective about when to invalidate the cache and
when to fetch.
<img width="310" alt="Screenshot 2024-08-16 at 11 19 23 PM"
src="https://github.com/user-attachments/assets/4ee2d111-2f55-4b0e-b944-50c4f78afc42">
<img width="580" alt="Screenshot 2024-08-18 at 10 05 16 PM"
src="https://github.com/user-attachments/assets/17630a60-7b78-421c-ae39-425246638a12">
I had originally added the lightening bolt on every message and only
added the tooltip warning about editing prior messages on the first
anchor, but thought it looked too busy, so I settled on just annotating
the last anchor.
