On 14/01/2023 17:35, Ihor Radchenko wrote:
Max Nikulin writes:
So Matt did not squashed commits before committing to the main branch
and detailed commit messages are preserved. That is why I do not
consider cgit render as a strong enough reason for reverting.
Should we then report a bug to cgit mailing list?
I do not have particular opinion. I rarely use web UI for git and mostly
for blame. I prefer to do it with local clone. I have never read cgit
docs. Perhaps it may be more apparent that it is a merge commit.
Anyway at first I would ask the savannah team. E.g. vanilla debbugs
instance is significantly more convenient than the GNU ones and the
changes in the latter are intentional. Mhonarc (list archive) at
lists.debian.org has better configuration than at lists.gnu.org as well.
So there might be specific of particular instance.
After all, it may be convenient to review cumulative changes. When I
worked with gerrit (a web application for code review) sometimes it was
inconvenient that it was impossible to see diff for whole commit series.
