mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-06 02:48:13 +00:00
07901a58ba
The check-duplicate workflow's script called gh api --paginate on the full issues endpoint and then json.loads the response. Older gh CLI versions concatenate paginated arrays without a valid separator, and even on newer versions the response is a single ~15MB blob that intermittently failed to parse in CI with 'Unterminated string' at char 902748 (truncation or buffering). Switch to jq streaming so each issue is a small self-contained JSON object per line, and only fetch the two fields the dedup logic uses (number, title) plus an is_pr marker. Payload drops to ~280KB and parsing is robust to per-line failures. Co-authored-by: Krrish Dholakia <krrish-berri-2@users.noreply.github.com> |
||
|---|---|---|
| .. | ||
| _agent_shin_actions.py | ||
| agent_shin_shared.py | ||
| close_duplicate_issues.py | ||
| close_low_quality_prs.py | ||
| scan_keywords.py | ||
| triage-requirements.txt | ||
| triage_rollout_heads_up.py | ||
| triage_with_llm.py | ||
| uv_sync_with_retries.sh | ||