git who -l doesn't work on blobless clone

[cgoldberg]

I have a large repo that isn't working with git who -l. I suspect the issue is because it is a blobless clone (cloned with clone --filter=blob:none), so it doesn't have all commit history locally. However, most git operations work normally on such a clone and just fetch missing blobs when needed.

git who works fine, but git who -l fails with:

error running "table": error running concurrent tally: concurrent tally failed: error in worker 1: Git subprocess exited with code 128. Error output:
fatal: You are attempting to fetch e3a6119b8628ce0eccd9d881bed980dbcc7b691f, which is in the commit graph file but not in the object database.
This is probably due to repo corruption.
If you are attempting to repair this repo corruption by refetching the missing object, use 'git fetch --refetch' with the missing object.
fatal: could not fetch 7523bd5df834ba3deae64009960a2307ccf5e2ed from promisor remote

If it is unable to operate on such a repo, perhaps a nicer error message can be displayed.

[sinclairtarget]

Hey @cgoldberg , I agree that that's not a very nice error message. But I'm unsure how I'd go about checking to see if Git errored-out for this specific reason. I'm not sure there's a good way to tell.

Running git-who -l would mean turning your blobless clone into a full clone in any case, I think. git-who -l needs to diff everything reachable in the commit history. So it probably just doesn't make sense to run git-who -l on a blobless or tree-less clone.