nevi-me commented on pull request #9390: URL: https://github.com/apache/arrow/pull/9390#issuecomment-778605841
Hi @maxburke, I'm currently reviewing this; the failing tests are because the logic in this PR accounts for 1-level structs correctly (as per the schema of the attached file), but if a field is at the root of the Arrow schema, it doesn't work correctly. For example, in the schema `<schema<a: date32>>`, we don't get `a: date32` from the Arrow schema. I've opened ARROW-11618 and ARROW-11619 to make this ergonomic, as there are other issues with how we project into schemas on both Parquet and Arrow (respective to the JIRA issues). This issue is likely blocking me on other nested-read work :( ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
