[FFmpeg-devel] [PATCH] avutil/avstring: add av_strreplace API into avstring
nfxjfg at googlemail.com
Sun Apr 2 00:48:45 EEST 2017
On Sat, 1 Apr 2017 22:34:26 +0200
Alexander Strasser <eclipse7 at gmx.net> wrote:
> On 2017-04-01 13:12 +0200, wm4 wrote:
> > On Sat, 1 Apr 2017 10:01:15 +0200
> > Nicolas George <george at nsup.org> wrote:
> > > Therefore, it was not acceptable as is and should not have been pushed.
> > >
> > > Pushing while leaving only half a day to answer this was UNACCEPTABLE.
> > > Do not ever do it again please.
> > Why get upset over a simple function being added?
> > Lots of unreviewed patches or patches that were half-approved or not
> > really approved get pushed. That's how this project work. If you don't
> > like it, propose a rule change.
> I do not think a rule change is needed for this case at hand.
> While it is probably true that there were patches that didn't have enough
> review before they were pushed, it is not at all common, that patches
> which are the subject of an active discussion, as it was in this case,
> just get pushed.
> Additionally this patch adds public API which should be sufficiently
> discussed because reverting soon becomes difficult and involved.
Especially for someone new this is not obvious.
Either have clear rules, or don't hand out push rights like free
cookies to inexperienced project members (and then complain that they
make use of it).
I requested that the rules be clarified a while ago, but nobody cared,
and a patch attempting to improve the description of the existing rules
and practice was not accepted.
In the case at hand, Steven Liu might have pushed a bit too early, but
it's not like he ignored Nicolas George's post. Also I encouraged him
to separate this function and move it to libavutil, and nobody had
anything against it all that time (until Nicolas George objected later).
So I don't see a terrible offense here.
Again, if you don't like this, make the damn rules clearer.
More information about the ffmpeg-devel