Date: Mon, 15 Apr 2019 09:49:08 +0300
From: Konstantin Kharlamov <hi-angel@yandex.ru>
Cc: Eli Zaretskii <eliz@gnu.org>, 35062@debbugs.gnu.org
IMO in such situations (i.e. when original changes were commited
without any modification anyway) would be nice to commit the
original
patch, and then add up further improvements as 2-nd commit.
That's true, but this is not such a situation: the original changes
were never committed without any modifications.
Sometimes committing the original and then making changes in a
followup is TRT, and sometimes it isn't; it's a judgment call. In
general, the decision depends on the percentage of the original
submission that the committer would like to change, and also on the
overall volume of the original submission. In this case, the original
patch was relatively small, and I modified it in relatively
significant ways. So it made little sense to commit something that
would be immediately modified in significant ways, it would just be
extra work for no good reason.
Ultimately what makes me sad is that if I'd want to refer to my
commits in Emacs as part of a CV, it's hard to find all suggested-by
and authored commits at the same time, and also that suggested-by
sounds kind of vague to have an influence in CV.
You can always use "git log --grep" to find references to your
contributions in the log messages. And the log message includes a
reference to the bug number, where you can refer people for your
actual contribution.