Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes: > Are we going to have the same issue with tags, c.f. parse_tag() and > there being no unparse_tag()? > > (I don't know offhand, just asking) > > I have some semi-related (test) changes locally where we do have blind > spots in tag v.s. commit parsing semi-related to this, i.e. in the whole > "unparsed" stage. > > So I wonder what happens with a tag that's pointing to a shallow object > that's parsed, but its underlying commit becomes un-parsed. > > Or maybe that's impossible, I'm not too familiar with "shallow"... As Stolee said, grafts (and shallow) are only on commits. [1] https://lore.kernel.org/git/394c054e-e1d2-41a5-a655-2ad3cb7219e0@xxxxxxxxxx/ > Nit: Can this be > e.g. s/repo-with-unreachable-upstream-shallow/repo/. The overly long > repo name makes this much harder to follow. Compare this one which would > clean up after itself too: > > test_when_finished "rm -rf repo" && > git init repo && > git -C repo submodule add ../a-submodule a-submodule && > git -C repo commit -m "added submodule" && > > SHALLOW=$(cat shallow/.git/shallow) && > git -C repo fetch --update-shallow ../shallow/.git "$SHALLOW":refs/heads/a-shallow > > (I didn't check if that test really works, i.e. do we have a "repo" > already, but you get the idea... Yes I can do that. I'll see if there are any more comments and send out a new version after the weekend.
