mbutrovich opened a new issue, #3648: URL: https://github.com/apache/iceberg-rust/issues/3648
### Is your feature request related to a problem or challenge? `pyiceberg-core` returns manifest entries and manifest-list entries as per-entry `#[pyclass]` objects, which PyIceberg converts into its own objects one at a time. apache/iceberg-python#4007 shows that conversion costing about as much as PyIceberg's whole Cython decode. #3388 and #3631 added `iceberg::arrow::read_manifest_entries` and `iceberg::arrow::read_manifest_list`, which decode into a `RecordBatch`. This is part of Task 3 of #3262. ### Describe the solution you'd like Add binding functions that return pyarrow tables from those readers, next to the existing ones. The binding already depends on `arrow` with the `pyarrow` feature. `read_manifest_entries` takes the manifest's `ManifestFile` to apply inheritance, so the binding needs a way to pass it in. The existing `read_manifest_list` binding hard-codes `FormatVersion::V2` ([`bindings/python/src/manifest.rs#L223-L228`](https://github.com/apache/iceberg-rust/blob/832a4eba7a732baebbbd7f76bb6fd6225f1db1e9/bindings/python/src/manifest.rs#L223-L228)). This should fix that by reading through `iceberg::arrow::read_manifest_list`, which takes the version from the file and returns an error for an unknown one. This follows #TBD-iterator, which settles the readers' return type. ### Willingness to contribute I can contribute to this feature independently -- 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]
