[FFmpeg-devel] [CHECKLIST] what is preventing us from switch to git

Alex Converse alex.converse
Sat Jun 12 21:21:58 CEST 2010


On Sat, Jun 12, 2010 at 3:13 PM, Mans Rullgard <mans at mansr.com> wrote:
>
>
> "Stefano Sabatini" <stefano.sabatini-lala at poste.it> wrote:
>
>>On date Saturday 2010-06-12 20:23:42 +0200, Reimar D?ffinger encoded:
>>> On Sat, Jun 12, 2010 at 02:12:43PM -0400, Alex Converse wrote:
>>> > On Sat, Jun 12, 2010 at 2:05 PM, Reimar D?ffinger
>>> > <Reimar.Doeffinger at gmx.de> wrote:
>>> > > On Sat, Jun 12, 2010 at 01:49:17PM -0400, Alex Converse wrote:
>>> > >> On Sat, Jun 12, 2010 at 1:38 PM, Reimar D?ffinger
>>> > >> <Reimar.Doeffinger at gmx.de> wrote:
>>> > >> > On Sat, Jun 12, 2010 at 12:43:51PM +0100, M?ns Rullg?rd wrote:
>>> > >> >> Michael Niedermayer <michaelni at gmx.at> writes:
>>> > >> >> > * We need to document, a short cheat sheet that lists the
>>> > >> >> > ? git equivalents to the common svn commands (that of course before
>>> > >> >> > ? a switch is done)
>>> > >> >>
>>> > >> >> There are many such guides already written and available on the internet.
>>> > >> >
>>> > >> > That does not remove the need for above. Also I still have no idea how to
>>> > >> > fix commit messages on a remote repository without risk of breaking other
>>> > >> > people's checkout.
>>> > >>
>>> > >> http://www.kernel.org/pub/software/scm/git/docs/git-notes.html
>>> > >
>>> > > Sounds useless for typos or any massive messup to me...
>>> >
>>> > How so?
>>> >
>>> > If there is a commit message with something wrong add a note saying
>>> > "<wrong thing> is wrong. <right thing> is right."
>>>
>>> Well, if the commit message ended up being
>>> "Thing ist doned"
>>> Then adding
>>> "Note: meant "Thing is done""
>>> is not a real improvement IMO.
>>> To be more specific: trying to fix any typos/grammar that make things
>>> not completely impossible to understand just makes things even worse
>>> with that system AFAICT, IMHO.
>>
>>We could have a pre-commit hook which turns the commit message to a
>>commit note, then every successive change to the message would only
>>affect the note.
>
> I vehemently object to that ludicrous idea.

I agree with Mans. VLC, x264, gnome, git itself, and the linux kernel
all seem to get by without doing this.

Git formatted patches as well as messages being written at commit time
not at push time, should significantly lower the ratio of bad commit
messages.



More information about the ffmpeg-devel mailing list