Laszlo Ersek <ler...@redhat.com> writes:
> On 01/22/20 13:30, Alex Bennée wrote: >> >> Stefan Hajnoczi <stefa...@gmail.com> writes: >> >>> Around 66% of qemu.git commits since v4.1.0 include a Message-Id: tag. >>> Hooray! >>> >>> Message-Id: references the patch email that a commit was merged from. >>> This information is helpful to anyone wishing to refer back to email >>> discussions and patch series. >> >> So I guess the ones that don't are maintainer originated patches unless >> you actively rebuild your trees from a posted series? > > I *think* this should not be a huge problem process wise: > > Assuming that a maintainer does not include their own patches in a PULL > request for Peter until the same patches receive R-b/A-b/T-b feedback > from other list subscribers, the maintainer will want to rebase the > patches at least once anyway, in order to pick up those lines. Oh I always do a re-base as I apply the r-b/t-b tags. But that is working off my tree and a bunch of references to the emails with the appropriate tags in them. So which Message-Id should I use. The first time the patch was posted to the list or the last time it was? > And, in the process, the maintainer might as well add in their own > Message-Id's from the list. > > ... I realize though, that could be more burden in practice than just > running git-am against the same (known) base commit... One could always > run git-range-diff in the end, to compare the "re-pick" versus the > original local branch. I'm obviously missing out by not using patches but my own Emacs based tooling. I guess I shall have to see if I can extend it. > > Thanks > Laszlo -- Alex Bennée
