Hello, André Batista <[email protected]> skribis:
> Well, I guess ENOENT != GIT_ENOTFOUND and, in that case, I think we > are better off gracefully failing to update it than trying to "branch" > our local repo. The patch below tests for the directory's existence > before proceding with the update. With this patch applied I've managed > to refresh pytorch, its submodules and got the following code: But you got a wrong hash I presume, because submodule weren’t actually checked out, right? > + (if (file-exists? directory) > + ((lambda () > + (submodule-update submodule > + #:fetch-options fetch-options) The problem is that it papers over an actual libgit2 bug. I think we should instead report it upstream. Do you feel like doing it? I guess we’d need to give them the C version of the three-line snippet I gave earlier. WDYT? Thanks, Ludo’.
