Open
bjchambers
opened this issue 2 years ago
•
1 comments
Summary
It looks like Arrow supports this -- read/write. For schema inference we will likely need something like "download all or the first 10 MB of the file and do schema inference" or something like that.
kaskada-ai
