GitHub user ElSaico created a discussion: Implement a custom FileFormat
Hello, I have a specific data transformation in mind: to skip the details, the input is a line-delimited JSON while the output is a list of partitioned files that follow an application's custom binary format. With those specifications in mind, an Arrow dataset feels like an *almost* perfect fit... if I could implement my own `FileFormat`, that is. Would that even be possible? GitHub link: https://github.com/apache/arrow/discussions/48069 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
