Repository navigation
Conversation
Codex 0.155+ rollouts persist a token_usage_record line carrying the upstream resp_… id just before the token_count it reports. The importer attaches it as responseId when the record's usage equals that token_count's last_token_usage, so store.mergeByResponseId and the session-index _countedRids/_costByRid dedup apply to Codex imports. Entries without a matching record keep no key and the legacy id dedup. A legacy suffixed row without responseId still counts as the same turn on rescan. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Imported Codex turns now carry
responseId, sostore.mergeByResponseIdand the session-index_countedRids/_costByRiddedup apply to Codex imports (an overlapping re-import of the same response is one turn; count and cost are not doubled).token_usage_recordline (payload.response_id= upstreamresp_…) just before thetoken_countit reports.parseCodexSessionFileattaches it only when the record's usage equals thattoken_count'slast_token_usage, so a stray record cannot lend its id to another turn.token_counts) keep no key and the legacy id-based dedup.docs/wire-protocol-reference.mdand the fix(importer): Codex entries never set responseId — proxy×import duplicates can never be deduped #508 investigation note; verified by key names and pairing counts only, no transcript content copied.responseId(OpenAI HTTP/WS stay exempt per ADR 0012), so proxy×import merging remains the other half.Fixes #508
Verification