Add concurrency and fast-stop to generate_answers#291
Merged
Conversation
Runs generate_answer() over search results concurrently (bounded by AnswerGeneratorSettings.concurrency) and, when fast_stop is enabled, skips results that haven't started once another has already produced a good answer. Mirrors answerGenerator.ts's concurrency/fastStop behavior for the array-of-search-results case. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Pull request overview
Adds configurable concurrency and “fast stop” behavior to generate_answers() so multiple generate_answer() calls can run concurrently (bounded), and optionally stop starting new work once a good answer is found—aligning with the described TypeScript behavior. Tests are added to validate the concurrency cap and fast-stop skipping semantics.
Changes:
- Introduced
AnswerGeneratorSettingsand a concurrent implementation path forgenerate_answers(). - Implemented
_generate_answers_concurrently()using anasyncio.Semaphoreplus anasyncio.Eventto support fast-stop. - Added async tests covering concurrency limiting and fast-stop vs non-fast-stop behavior.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
src/typeagent/knowpro/answers.py |
Adds settings + concurrent/fast-stop execution for answer generation. |
tests/test_answers.py |
Adds tests validating concurrency limiting and fast-stop behavior. |
robgruen
requested changes
Jul 13, 2026
robgruen
left a comment
robgruen
left a comment
Collaborator
There was a problem hiding this comment.
Can we address the two copilot comments? I think this is the preferred behavior and if it differs from what's in the TS implementation I'll update that to match.
Address PR microsoft#291 review feedback: AnswerGeneratorSettings previously defaulted to concurrency=2/fast_stop=True, silently changing behavior for existing callers that don't pass settings. Default to concurrency=1, fast_stop=False (matching the prior sequential, run-everything behavior), overridable via TYPEAGENT_ANSWER_CONCURRENCY and TYPEAGENT_ANSWER_FAST_STOP env vars or explicit settings. Also clarify the generate_answers() return-value comment now that fast_stop can return fewer answers than search_results. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Drop the bespoke _default_concurrency/_default_fast_stop helpers with try/except in favor of the plain os.getenv(NAME, default) style already used elsewhere in the codebase (e.g. outlook_dump.py). Document the new TYPEAGENT_ANSWER_CONCURRENCY / TYPEAGENT_ANSWER_FAST_STOP vars in .env-template alongside the other optional settings. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Collaborator
Author
|
@robgruen thanks for your feedback. is fixed now |
robgruen
approved these changes
Jul 14, 2026
robgruen
left a comment
robgruen
left a comment
Collaborator
There was a problem hiding this comment.
Looks good, thanks for this change/improvement!
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
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.
Runs generate_answer() over search results concurrently (bounded by AnswerGeneratorSettings.concurrency) and, when fast_stop is enabled, skips results that haven't started once another has already produced a good answer. Mirrors answerGenerator.ts's concurrency/fastStop behavior for the array-of-search-results case.