GitHub user ianmcook added a comment to the discussion: Proposal: ADBC file drivers with pluggable storage
If the goal is to quickly cobble something together to expose a file's contents to ADBC clients without using a separate query engine, for your own private or internal use: yeah, just build it as a driver and don't worry about the bloat. But if the goal is to build a legible ecosystem in which a growing number of packages like these are distributed through public registries and installed by millions of users and spoken about in coherent language (which is what we're doing): please don't. GitHub link: https://github.com/apache/arrow-adbc/discussions/4877#discussioncomment-18915154 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
