If the LLM uses ellipsis in place of the REDACTED string, print the original version #614
Conversation
This file contains 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
…riginal version I was able to make the LLM to reply with `REDACTED=<...>` when summarizing code. Because our regex and the prefix marker expected the redacted string to contain the dollar sign, we never flushed the prefix buffer correctly and ended up later flushing repeatedly.
JAORMX
reviewed
Jan 16, 2025
| @@ -312,8 +312,8 @@ class SecretUnredactionStep(OutputPipelineStep): | |||
| """Pipeline step that unredacts protected content in the stream""" | |||
|
|
|||
| def __init__(self): | |||
| self.redacted_pattern = re.compile(r"REDACTED<\$([^>]+)>") | |||
| self.marker_start = "REDACTED<$" | |||
| self.redacted_pattern = re.compile(r"REDACTED<(\$?[^>]+)>") | |||
There was a problem hiding this comment.
As with all regexes, this will start getting harder to maintain with time. Mind adding a couple of examples expected patterns?
JAORMX
approved these changes
Jan 16, 2025
|
this was obsoleted by #608 |
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Outdated suggestions cannot be applied.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.
I was able to make the LLM to reply with
REDACTED=<...>whensummarizing code. Because our regex and the prefix marker expected the
redacted string to contain the dollar sign, we never flushed the prefix
buffer correctly and ended up later flushing repeatedly.