fix: Prevent explain_info overwrite during stream_async #1194
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
added 2 commits
May 14, 2025 08:14
… between `stream_async` and `generate_async`
There was a problem hiding this comment.
Pull Request Overview
This PR fixes the missing LLM call tracking in the explain info and refactors its initialization logic while adding tests to ensure proper behavior with pre-commit hooks.
- Added asynchronous tests to validate that the explain info correctly records LLM calls.
- Refactored llmrails.py to encapsulate explain info initialization in the _ensure_explain_info() method and removed redundant updating functions.
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| tests/test_llm_rails_context_variables.py | Introduces tests for LLM call tracking and verifies expected call counts. |
| nemoguardrails/rails/llm/llmrails.py | Refactors and unifies the logic for initializing explain info, removing redundant code. |
Comments suppressed due to low confidence (1)
tests/test_llm_rails_context_variables.py:111
- The comment above this assertion states that 6 LLM calls are expected, but the assertion checks for 5. Please update either the comment or the expected value to ensure consistency.
assert (len(chat.app.explain().llm_calls) == 5), "number of llm call not as expected. Expected 5, found {}".format(len(chat.app.explain().llm_calls))
Pouyanpi
approved these changes
May 14, 2025
This was referenced May 14, 2025
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Outdated suggestions cannot be applied.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.
Description
This PR addresses an issue where
self.explain_infocould be overwritten during calls tostream_async, potentially leading to incomplete explanation data.The root cause was that the
generate_asynctask and the_run_output_rails_in_streamingfunction (if output streaming rails were active) could independently initializeExplainInfoobjects if theexplain_info_varcontext variable was not set prior to thegenerate_asynctask's creation.The fix ensures that
explain_info_varandself.explain_infoare initialized at the beginning of thestream_asyncmethod. This guarantees that both thegenerate_asynctask (via context variable inheritance) and_run_output_rails_in_streaminguse the sameExplainInfoinstance, preserving data integrity.[same as https://github.com//pull/1193 but with pre-commit hooks done]
Related Issue(s)
#1192
Checklist