Repository navigation
Review fixes - #1
Merged
Merged
Conversation
- The ONNX detector split words on a lowercased copy and returned those byte offsets for the original text, the same bug fixed earlier in the dictionary: characters whose case mapping changes length shifted every later span. Words are now located in the original text and lowercased only for the tokenizer. - Spans from any extra detector are widened to character boundaries before use, so a misaligned span redacts too much instead of panicking or too little. - Inference failures were swallowed silently, turning ML recall off unnoticed. They are now logged by stage (never text) and counted in portcullis_detector_errors_total. - /healthz read the term count under the gateway lock and queued behind scans; it now reads a lock-free gauge. - Split a six-element tuple into NerInput (clippy type_complexity under onnx). Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…st fingerprint Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
teach/unteach overwrote the file from the server's in-memory copy, silently dropping a term added by the CLI in the meantime. They now re-read the file (when it exists), apply the change, and save, serialized by a dedicated writer lock. The read and save run on the blocking pool, so Argon2 under encryption no longer stalls requests behind the gateway lock. An unreadable file fails the request rather than being overwritten. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The outbound assertion ran only over the serialized body, where JSON escaping hides a taught term containing a quote, backslash or control character. Every decoded string value and object key is now checked as well, with the placeholder check kept to the whole body. Shared by both providers. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
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
1 participant
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.
No description provided.