Repository navigation
⚡️(backend) search users through the trigram indexes - #2906
Conversation
|
Important Review skippedReview was skipped as selected files did not have any reviewable changes. ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
WalkthroughBackend user searches now use unaccented trigram matching to find email and full-name candidates. Email searches also apply a Levenshtein distance limit. Other searches place shared-document users before same-domain matches and cap the combined results. The search sets a transaction-local trigram threshold. Tests check result ordering, index use, and threshold reset. Priority: ⬇️ Low Estimated code review effort: 3 (Moderate) | ~25 minutes Suggested reviewers: Merge Risk: ⚪ Minimal · up to
|
| Check name | Status | Explanation |
|---|---|---|
| Title check |  |
The title names the primary change: user search now uses the trigram indexes. The changeset replaces the raw-column similarity scoring with % operator queries on the f_unaccent(email) and f_unaccent(f… |
| Description check | |
The description states the problem: the search used a similarity score on raw email and full_name columns, so the trigram indexes (built on f_unaccent expressions and used only by the % operator) were… |
| Docstring Coverage | |
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 7 functions across 2 files. (1 skipped: 1 … |
| Linked Issues check | |
Check skipped because no linked issues were found for this pull request. |
| Out of Scope Changes check | |
Check skipped because no linked issues were found for this pull request. |
Full details: Title check
-
Explanation
The title names the primary change: user search now uses the trigram indexes. The changeset replaces the raw-column similarity scoring with % operator queries on the f_unaccent(email) and f_unaccent(full_name) indexed expressions, and moves shared and same-domain filtering into SQL. The title is short and specific. It does not cover the transaction-scoped threshold or the result limit, which is acceptable.
Full details: Description check
-
Explanation
The description states the problem: the search used a similarity score on raw email and full_name columns, so the trigram indexes (built on f_unaccent expressions and used only by the % operator) were not used and every search scanned the user table. It also describes the fix: the % operator on the indexed expressions, a transaction-scoped threshold, SQL-side filtering of shared and same-domain users, and a fetch limit. This matches the viewsets.py and test changes.
Full details: Docstring Coverage
-
Explanation
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 7 functions across 2 files. (1 skipped: 1 unsupported.)
✨ Finishing Touches
-
📝 Generate docstrings
-
- Commit to this branch
- Create a new PR
🧪 Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
Comment @coderabbitai help to get the list of available commands.
The user search filtered on a similarity score computed on the raw email and full_name columns. The trigram indexes are built on f_unaccent(email) and f_unaccent(full_name) and only serve the % operator, so each search scanned the whole user table, and all the matches were then filtered and sorted in Python. The search now uses the % operator on the indexed expressions, with a similarity threshold scoped to the transaction. It filters the shared and same-domain users in SQL, and fetches no more users than the response limit.
b61ab0a to
47609af
Compare
Purpose
The user search filtered on a similarity score computed on the raw email and full_name columns. The trigram indexes are built on f_unaccent(email) and f_unaccent(full_name) and only serve the % operator, so each search scanned the whole user table, and all the matches were then filtered and sorted in Python.
The search now uses the % operator on the indexed expressions, with a similarity threshold scoped to the transaction. It filters the shared and same-domain users in SQL, and fetches no more users than the response limit.
Proposal