XiaoHongbo-Hope commented on PR #1009: URL: https://github.com/apache/paimon-rust/pull/1009#issuecomment-5980713295
> Reverified the rebased [34cf10f](https://github.com/apache/paimon-rust/commit/34cf10ffce01de1dff3e39e98375adaaf09e07f0) against main [5a81a6d](https://github.com/apache/paimon-rust/commit/5a81a6d92a1bb8a9a327c5f98c716a0c10c29645) (the integrated tree is identical to the PR head). The previous merge conflict is resolved. I rebuilt and loaded the exact native Python extension: all 113 author/Java-description interoperability read tests and 73 Rust Variant tests pass. The Java FLOAT string finding remains: the additional two actual SQL-written Parquet/native Python read probes still return 1.5 for NBSP-wrapped "1.5" under both fail_on_error=false and true, instead of NULL or a cast error. Float.parseFloat rejects NBSP; the revised variant.rs:2826 still uses Unicode trim(). Please replace this with Java-compatible string parsing/trimming in the shared conversion used by the fast path, and retain both error-policy regression tests. Thanks, fixed -- 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]
