Skip to content

fix(parser): apply Claude queue links incrementally - #1714

Open
naveenspark wants to merge 1 commit into
kenn-io:mainfrom
naveenspark:fix/claude-queue-link-churn
Open

fix(parser): apply Claude queue links incrementally#1714
naveenspark wants to merge 1 commit into
kenn-io:mainfrom
naveenspark:fix/claude-queue-link-churn

Conversation

@naveenspark

Copy link
Copy Markdown
Contributor

Summary

  • Apply Claude queue and progress subagent links incrementally.
  • Preserve deterministic ordering and existing first-link precedence.
  • Keep link-only appends on the incremental sync path.

Why

Active Claude sessions append queue and progress events. Replacing the stored session for each event repeatedly parses large JSONL files and drives sustained CPU use.

Tests

  • go test ./internal/parser -count=1
  • focused incremental sync tests for Claude queue and progress links
  • focused database incremental-link tests
  • go vet ./internal/parser

@roborev-ci

roborev-ci Bot commented Sep 10, 2026

Copy link
Copy Markdown

roborev: Combined Review (3411f0e)

Verdict: Changes require fixes for 1 finding.

Medium

  • internal/parser/claude.go:946-947: claudeSubagentMapLinks entries have HasResult == false but are passed as agent links, so a queue/progress mapping and late tool_result in the same append can cause the result to be skipped and never copied to the existing call. Only treat result-bearing links as suppressing generic late-result links, or pass map-only links separately; add an integration test covering both updates persisting.

    Reported by: codex


Reviewers: 2 done | Synthesis: codex, 8s | Total: 2m51s

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

1 participant