/odbc2arrow

An ODBC to Apache Arrow reader, written in Rust

Primary LanguageRustApache License 2.0Apache-2.0

ODBC to Apache Arrow

This is a hopefully lean library that allows you to consume ODBC records as Arrow RecordBatches. It does not write these batches to files or streams, as that is simple enough for the implementer to do.

Origins

We based this library on the excellent work from Markus Klein's odbc2parquet crate.

However, instead of making this a binary, we made it a library. Apache Arrow is an in-memory format, so it makes more sense to read data into memory, rather than being opinionated about how to write files.

Usage

TODO