You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 8663a84
Browse filesBrowse the repository at this point in the historyBrowse files
fix(schema): batch DuckDB introspection and bound indexing
- `duckdb`: restrict schema/table/column introspection to the current catalog and de-duplicate schemas, so attached catalogs no longer produce repeated schemas, duplicate tables or merged columns
- `duckdb`: add optional `introspectSchema(signal)` connector capability returning schemas, tables and columns in two queries independent of table count
- `schema cache`: prefer bulk metadata when available, dedupe schema/table names for other connectors, and replace cache rows plus the warehouse summary in a single transaction
- `schema_index`: forward session cancellation, apply a 30s default deadline (including connection acquisition), report partial results and stopping reason, and prevent late cache writes after cancellation
- tests: driver, cache, tool and real-DuckDB integration regressions (native tests gated on `ALTIMATE_DUCKDB_E2E=1`)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
`SELECT column_name, data_type, is_nullable FROM information_schema.columns WHERE table_schema = $1 AND table_name = $2 ORDER BY ordinal_position`,
325
+
`SELECT column_name, data_type, is_nullable FROM information_schema.columns WHERE table_catalog = current_database() AND table_schema = $1 AND table_name = $2 ORDER BY ordinal_position`,
0 commit comments