alexanderbianchi opened a new pull request, #69: URL: https://github.com/apache/datafusion-iceberg/pull/69
## What changes Have `IcebergCatalogTableProvider::scan()` load the latest table and construct a per-scan `IcebergStaticTableProvider`, then delegate to its `scan()`. The catalog provider keeps ownership of catalog refresh and writes; the static provider owns the read-plan construction. This PR is based on `main` and does **not** add eager file planning or depend on #67. The generated static provider uses the catalog provider's **cached Arrow schema** instead of re-deriving a schema from the freshly loaded table. This preserves the existing contract: after the table evolves, a provider created before the schema change still exposes and scans its original columns. No public constructor or wire format changes. The existing schema-evolution test now also scans a projected column after the table changes, checking that projection indices still refer to the provider's original schema. Existing catalog insert and lazy-read tests continue to run. This is a separate refactor from #67; because both change the catalog scan method, whichever merges second may need a rebase. ## Validation - `cargo test -p iceberg-datafusion --locked --test integration_datafusion_test test_scan_after_schema_evolution_reads_provider_columns`: pass. - `cargo fmt --all -- --check`: pass. - `cargo clippy --workspace --locked --all-targets -- -D warnings`: pass. - `cargo test --workspace --locked`: pass. - `git diff --check`: pass. ## AI disclosure Developed with AI assistance and validated locally before opening as a draft. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
