zeevm commented on issue #2394: URL: https://github.com/apache/arrow-rs/issues/2394#issuecomment-1213032232
I'd think the Row level interface is central to the implementation, without it, it feels like this isn't really a proper parquet implementation library, rather a helper library mainly built to serve Arrow. Column reader and page reader (directly, not through Arrow) are also important. Arrow is well and fine, but Parquet is consumed by other in-memory columnar representations and other query engines as well. We completely disable the arrow feature when using the parquet crate. if parquet-rs design goals are specifically to serve Arrow, this should be clearly stated by the core team so folks taking dependency on it know what they're buying into. I'd think it would better serve the community to break parquet off of arrow-rs into a stand-alone project, arrow-rs can take a dependency on it. Thanks. -- 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]
