fix: improve VS Code LM token usage reporting for context window updates #6115
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Add initial usage yield with input tokens at stream start - Yield periodic token updates during streaming (every 500 chars) - Include cache token fields (set to 0) for consistency with other providers - This ensures the context window progress bar updates properly during streaming
dosubot
bot
added
size:M
hannesrudolph
added
the
Issue/PR - Triage
The PR added an initial usage yield at the start of the stream, which causes tests to receive 3 chunks instead of 2. Updated tests to: - Expect 3 chunks (initial usage + text + final usage) - Handle the new chunk ordering correctly - Fix error handling test to account for initial usage before error
hannesrudolph
added
PR - Needs Preliminary Review
and removed
Issue/PR - Triage
Member
|
This just returns 0 tokens because VSCode LM API doesn't return token info. We should probably use tiktoken instead. |
github-project-automation
bot
moved this from PR [Needs Prelim Review]
to Done
in Roo Code Roadmap
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR fixes issue #6112 where the token count and context length progress bar don't update properly when using VS Code LM API as the API Provider.
The VS Code LM provider was only yielding token usage information once at the end of the stream, which prevented the UI from updating the context window progress bar during streaming.
Changes:
Fixes #6112
Important
Improves token usage reporting in
VsCodeLmHandlerby adding initial and periodic updates during streaming, addressing issue #6112.VsCodeLmHandlerinvscode-lm.tsnow yields initial token usage at stream start and periodic updates every 500 characters.vscode-lm.spec.tsto expect initial, periodic, and final token usage chunks.getApiProtocolinprovider-settings.test.tsto return 'openai' forvscode-lmprovider.This description was created by
for 048fcf7. You can customize this summary. It will automatically update as commits are pushed.