fix: strip double-quotes from extracted sequence name in SERIAL ownership detection#469
Merged
tianzhou merged 4 commits intoJun 10, 2026
Conversation
…ship detection
When a sequence has a mixed-case quoted name, the nextval() default expression
stores it as nextval('schema."SeqName"'::regclass). The existing two-pass
REGEXP_REPLACE extracted 'SeqName' with double-quotes still present, causing
the col_table JOIN to compare '"SeqName"' against 'SeqName' (as stored in
pg_sequences) — the join always failed for such sequences.
Add a third REGEXP_REPLACE pass that strips leading/trailing double-quotes
from the extracted sequence name so the col_table JOIN matches correctly.
With this fix, OwnedByTable is correctly populated for quoted SERIAL sequences,
the skip condition fires, and pgschema no longer emits a standalone
CREATE SEQUENCE alongside the SERIAL-generated one — eliminating the _seq1
drift on fresh database deploys.
Greptile SummaryThis PR updates sequence ownership detection for quoted SERIAL sequence defaults. The main changes are:
Confidence Score: 3/5This should be fixed before merging.
Important Files Changed
Reviews (1): Last reviewed commit: "fix: strip double-quotes from extracted ..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
After stripping outer quotes, a sequence name like My"Seq is stored in column_default as My""Seq (PostgreSQL SQL identifier escaping). Without unescaping, the col_table JOIN still fails for such names. Add REPLACE(..., '""', '"') as a final pass to unescape doubled quotes, as suggested in code review.
The previous regex '^[^.]*\.' splits at the first dot, which breaks for quoted schemas containing a dot (e.g. "foo.bar"). Replace with a quote-aware alternative that matches either a full quoted identifier followed by a dot, or an unquoted prefix followed by a dot.
Contributor
|
Pls add a test case to validate this fix |
Contributor
There was a problem hiding this comment.
Pull request overview
Fixes SERIAL sequence ownership detection for mixed-case / quoted sequence names so pgschema doesn’t emit redundant standalone CREATE SEQUENCE statements (which can lead to _seq1 duplicates on fresh deploys).
Changes:
- Updates
GetSequencesForSchema’snextval(...)parsing to strip schema prefixes that may be quoted and to remove surrounding double-quotes from the extracted sequence name. - Normalizes escaped quotes (
""→") to match the unquotedpg_sequences.sequencenamejoin key. - Applies the same SQL change to both the source SQL and the sqlc-generated Go query constant.
Reviewed changes
Copilot reviewed 1 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
ir/queries/queries.sql |
Adjusts the sequence-name extraction logic from column_default to correctly join mixed-case quoted sequence names to pg_sequences. |
ir/queries/queries.sql.go |
Regenerates/updates the sqlc output so the runtime query matches the updated SQL source. |
Files not reviewed (1)
- ir/queries/queries.sql.go: Language not supported
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
…tection
Adds a query-level test that directly calls GetSequencesForSchema and
asserts OwnedByTable/OwnedByColumn are populated for a SERIAL column
with a mixed-case name ("orderId"). The pg_depend ownership edge is
removed via OWNED BY NONE to force the column_default parsing fallback
— the path that was broken before this fix.
Without the fix, the test fails with:
OwnedByTable = "", want "orders"
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Contributor
|
LGTM |
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
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.
Problem
When a sequence has a mixed-case quoted name (e.g.
"MyTable_myCol_seq"), thenextval()column default stores it as:nextval('schema."MyTable_myCol_seq"'::regclass)
The existing two-pass
REGEXP_REPLACEingetSequencesForSchemaextracted the name with double-quotes still attached —"MyTable_myCol_seq"— and thecol_tableJOIN compared this againstMyTable_myCol_seq(as stored unquoted inpg_sequences). The join always failed for such sequences, leavingOwnedByTableempty.With
OwnedByTableempty, the skip condition indiff.gonever fired, so pgschema emitted a standaloneCREATE SEQUENCEfor sequences already handled bySERIAL, causing PostgreSQL to create a duplicate with a_seq1suffix on every fresh database deploy.Fix
Add a third
REGEXP_REPLACEpass that strips leading/trailing double-quotes from the extracted sequence name before the JOIN. This makes the JOIN match correctly,OwnedByTableis populated, the skip condition fires, and no standaloneCREATE SEQUENCEis emitted for SERIAL-backed sequences.Testing
Verified on a schema with ~90 mixed-case SERIAL sequences across multiple tables:
DROP SEQUENCEoperations on every second apply