Skip to content

Conversation

@mysticmind
Copy link
Member

@mysticmind mysticmind changed the title Fix blind replace of d.in TypedLocator causing issues when schema name ends with d Fix blind replace of d. in TypedLocator causing issues when schema name ends with d Jun 4, 2025
@mysticmind mysticmind marked this pull request as draft June 4, 2025 12:57
@mysticmind mysticmind marked this pull request as ready for review June 4, 2025 14:18
@mysticmind
Copy link
Member Author

The CI failure DocumentDbTests.Bugs.Bug_2169_concurrency_error.there_should_be_no_error [FAIL] is unrelated to this PR.

@jeremydmiller jeremydmiller merged commit 39ad637 into master Jun 4, 2025
1 of 5 checks passed
@jeremydmiller jeremydmiller deleted the fix_3778 branch June 4, 2025 18:09
@aarearnold
Copy link
Contributor

If the index contains a reference to the additional column (for example, a column used as a foreign key), the SQL becomes invalid because d. no longer gets replaced.

output prior this PR:
CREATE UNIQUE INDEX mt_doc_user_accounts_uidx_provider_nameuser_id ON public.mt_doc_user_accounts USING btree ((data ->> 'providerName'), (user_id));

and from now on:
CREATE UNIQUE INDEX mt_doc_user_accounts_uidx_provider_nameuser_id ON public.mt_doc_user_accounts USING btree ((data ->> 'providerName'), (d.user_id));

@mysticmind
Copy link
Member Author

mysticmind commented Jun 8, 2025

This requires more fiddling, will take a look and fix the same. Sorry about this regression.

@mysticmind
Copy link
Member Author

@aare-arnold this PR has been reverted and v8.2.1 released.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Schema name ending with d results in issues with generated index definitions

3 participants