[FFmpeg-devel] [VOTE] Multiple inclusion guards in headers

[Diego Biurrun]

On Sun, Aug 17, 2008 at 09:40:46PM +0200, Michael Niedermayer wrote:
> On Sun, Aug 17, 2008 at 09:15:30PM +0200, Diego Biurrun wrote:
> > On Sun, Aug 17, 2008 at 07:37:02PM +0200, Michael Niedermayer wrote:
> > > On Sun, Aug 17, 2008 at 06:27:45PM +0100, Robert Swain wrote:
> > > > 2008/8/17 Michael Niedermayer <michaelni at gmx.at>:
> > > > [The beginning of another potential flame war]
> > > > 
> > > > Michael, do you really want to have the burden of flame wars on top of
> > > > reviewing code right now? I was hoping nothing would flare up for your
> > > > sake so you didn't feel compelled to deal with such but you seem to
> > > > now be starting it yourself...
> > > 
> > > Sorry but if i commit code with a clear notice that it will be cleaned
> > > up ASAP and diego replies with nothing more than 
> > > "(sigh)"
> > 
> > I replied because of the license headers, not because of the inclusion
> > guards, the lack of which I did in fact overlook.
> 
> > 
> > Nothing is gained by committing rubbish intermediate versions to the
> > repository
> 
> it is not rubbish, it is the hisorically correct version.

Nothing is gained by preserving a history of

r1: fix typo
r2: forgot ;
r3: compile fix
r4: real compile fix
r5: this time, real real compile fix

By the same token, some history *may* be worth preserving, but things
like

#if 1 --> #if TEST

are just clutter and add no value at all.  It's better to commit the
correct version right away.

> > > then i really think you are now complaining to the wrong person about
> > > starting something
> > 
> > Look, you threatened Vladimir with stopping to review his patches, when
> > he submitted *working* but suboptimal code.  If you want to be elitist,
> > perfectionist, nitpickish
> 
> I commited code with history. The first revission had some known issues that
> where cleaned up immedeatly.
> I did mention some rather concrete problems in respect to vladimirs patches
> and did review many iteratons before i flamed him.

And I have commented many times on your same mistakes, fixed them myself
even before I just said *sigh*.

> Here with pca.c/h i posted the files and after ramiro the next days said he
> could use them, i commited and then cleaned them up. Where where you all
> the time, you did not consider reviewing the patches it seems, but are
> rather quick at adding meaningless single word replies to commits of
> patches that HAD been posted previously.

Yes, I missed that it seems.  I'm plenty busy with other things already,
helping SoC students among other things.

It's not like I have done anything that would warrant you throwing around
sarcasm in commit messages.  Which is extremely annoying because commit
messages are an important part of the project history.

Diego