rishav394 commented on issue #4363: URL: https://github.com/apache/arrow-adbc/issues/4363#issuecomment-4632713386
You're right, the crash is inside `pyarrow.RecordBatchReader._import_from_c()` (apache/arrow C++), not this repo. `_reader.pyx`'s `_import_from_c` just calls into it. Please transfer to apache/arrow. For a self-contained repro: I considered the SQLite ADBC driver but it's written in C and only maps SQLite's primitive storage types to basic Arrow types - it has no path to produce Decimal32/64 or any type that would be unrecognized. The bug requires a Go-based driver (from adbc-drivers/ org) since those use arrow-go, which can adopt newer Arrow spec types independently of PyArrow. Simplest repro is `SELECT CAST(10.1 AS DECIMAL(10,4))` via any driverbase-go ADBC driver (Trino, MySQL, etc.) - no tables needed, just a running instance. -- 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]
