fix - in run_local_search(), avoid re-reading create_final_nodes.parquet file
Description
The file create_final_nodes.parquet is being read twice in run_local_search function in query component
Checklist
- [x] I have tested these changes locally.
- [x] I have reviewed the code changes.
- [ ] I have updated the documentation (if necessary).
- [ ] I have added appropriate unit tests (if applicable).
LGTM
Please run poetry run semversioner add-change -t patch -d "remove duplicate file read" and add/commit the generated json file to the repo. Feel free to modify the message commit message if you wish.
This will allow the PR to pass the Semver Check.
Looks like there is some trailing whitespace. Please run ruff format . --check --preview and re-comit the cli.py file. I tried to do it myself but did not have permissions to push the change to your fork.
I think you just ran ruff format on the file. It needs the --preview flag for proper formatting or else our ci/cd pipeline will flag it. ruff format graphrag/query/cli.py --preview should fix the file.
Thank you @ksachdeva for your contribution and your patience! We have a minor issue with the cicd pipeline but it did pass the formatting checks so I will go ahead and merge.