@Ariel Shaqed (Scolnicov) what’s the status on the “Return a

stat

-like object on file up-/downloads containing basic version info” plan re: the lakeFS server? I saw that the RFC was merged, and that you commented on the idea. Basically for now, I can emulate this in the frontend by grabbing the object info as normal (via

objects_api.stat_object

), and then basically doing the equivalent of

git rev-parse $ref

with the requested revision. The

is object staged or committed

use case you asked for is e.g. integrity checking: Am I working with properly tracked/committed data, or a “dirty” untracked version staged on main by my colleague without my knowledge? This would matter more when pulling data using raw branch names instead of commit SHAs, but from my experience, that’s quite common actually.

AFAIK we actually have the data in hand: on upload we literally just wrote it to staging, in download we had it. Also I'm pretty sure we already give it on upload, and also on download via the s3 gateway. So no promises but it looks like a useful bit of functionality. Thanks!