Vít Ondruch <vondruch@xxxxxxxxxx> writes: > My typical use case is to download patches from GH, e.g.: > > ~~~ > > $ curl -OL https://github.com/rails/sprockets/pull/791.patch ... > The problem with this is that I end up with the "791.patch" file, > while I'd like have a file with similar name as if I have used the git > command: > > > ~~~ > > $ git format-patch -1 6554b6d > 0001-Fix-Minitest-constant-name-in-tests.patch > > ~~~ > > > So I wonder, is there a way to get such file name? Do you mean: GitHub should let me run this command instead $ curl -OL https://github.com/rails/sprockets/pull/0001-Fix-Minitest-constant-name-in-tests.patch It may be nice for them to give a more meaningful name to their pull request (not just the output file name) than just an integer. But that is not a question/request we can answer here (this is not a help forum for GitHub users). Something along the lines of sed -ne '/^Subject: /{ s/^Subject: *\[PATCH[^]]*\] *//; s/[^a-zA-Z0-9]/-/g; s/--*/-/g; s/$/\.patch/; p; q; }' 791.patch should be doable, but I am not sure what the benefit is. Once you get it in Git, you'd park it on a branch with a useful name and we can forget about "791", so the "The files we get from GitHub are named in a way that makes it hard to identify them" does not sound like a Git issue, at least to me.
