slachiewicz opened a new issue, #4463: URL: https://github.com/apache/iceberg-python/issues/4463
Body: PyIceberg's `_HasIds` check treats a file as fully ID-bearing only when every field, including list elements and map keys/values, has an ID. Otherwise it resolves the whole file through `schema.name-mapping.default`. The spec now says the opposite: before projection, readers check the full file schema; if any field ID is present at any depth, readers use stored IDs and treat ID-less fields, lists missing an element ID and maps missing a key or value ID as absent. Absent fields resolve through partition metadata, `initial-default`, then null — never through the name mapping, which applies only to files with no field IDs. Reference fixture: apache/iceberg-verification#27 (`v2/partial-field-ids-parquet-data`), with expected values. On that file all projections currently fail in PyIceberg 0.12.0. -- 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]
