mirror of
https://github.com/supabase/supabase.git
synced 2026-10-08 19:05:06 +03:00
Charis
426b0183af
## I have read the [CONTRIBUTING.md](https://github.com/supabase/supabase/blob/master/CONTRIBUTING.md) file. YES ## What kind of change does this PR introduce? Feature — adds a new data-layer hook. No callers are switched over in this PR. ## What is the current behavior? The Schema Designer and Database > Tables list both call `useTablesQuery({ includeColumns: true })`, which fetches the full schema (every table, with every column, every relationship) in one round-trip. For customers with many tables this never loads. Towards FE-3428. ## What is the new behavior? Introduces `useInfiniteTablesQuery` — a cursor-paginated tables hook ready to be consumed by the two pages above in follow-up PRs. - `pg-meta`: new `getTablesPaginatedSql` SQL builder that picks a page of table OIDs first (cheap `pg_class` index scan) and constrains every enrichment CTE (primary keys, relationships, columns) to that set. Pagination is by `c.oid > $afterOid` rather than `OFFSET`, so deep pages stay O(limit). Relationships use a `UNION ALL` keyed by `table_id` so the downstream join is a plain equi-join. - `pg-meta`: `COLUMNS_SQL` is now produced by `getColumnsSql({ filter })`, letting a paginating caller push a table-OID predicate into the WHERE clause directly. The bare `COLUMNS_SQL` export is preserved for the 5 existing callers (`pg-meta-columns`, `pg-meta-tables`, `pg-meta-views`, `pg-meta-materialized-views`, `pg-meta-foreign-tables`). - `studio`: `useInfiniteTablesQuery` wires the new SQL into `useInfiniteQuery` via `executeSql`. `initialPageParam: 0`, `getNextPageParam` returns the last row's `id` or `undefined` on a short last page. - Tests: 12 new tests in `pg-meta` covering cursor invariants (no overlap / no gap), schema filtering, primary-key / relationship / column shape, and output parity against the existing `pgTableZod` schema. ## Additional context <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Enabled infinite scrolling for table listings with cursor-based pagination and optional page sizing. * Added an option to include per-table column data in paginated results. * Made column queries filterable for more targeted metadata retrieval. * **Tests** * Added comprehensive tests validating pagination, schema scoping, column inclusion, and relationship/PK shaping. <!-- review_stack_entry_start --> [](https://app.coderabbit.ai/change-stack/supabase/supabase/pull/46285?utm_source=github_walkthrough&utm_medium=github&utm_campaign=change_stack) <!-- review_stack_entry_end --> <!-- end of auto-generated comment: release notes by coderabbit.ai -->
