[gst-devel] told you
hadess at hadess.net
Fri Apr 19 07:39:47 CEST 2002
On Fri, 2002-04-19 at 09:37, Thomas Vander Stichele wrote:
> well I had just woken up and read your mail and started to write a really
> pissy mail back but luckily I got interrupted.
> So now that I'm at work I'll keep it simple ;)
> > Heya,
> > I told Thomas (how many times did I tell you, don't know, but a lot of
> > times) to disable -Werror in the tarballs.
> 1) If you're not my father or my mother then don't patronise me.
> 2) If you're not my boss or alternatively paying me handsome amounts of
> cash then don't TELL me what to do. You might always try asking
Fair enough, I asked you many times still, and didn't get it.
> > It seems that the plugins for
> > 0.3.4 were distributed with -Werror enabled by default.
> 3) I seem to remember that in the end the consensus was to use -Wall and
> not -Werror for releases, while keeping both for cvs. I might have, in
> one of the tarballs, mixed up the two since in my subconsciousness it
> still feels like "Wall" is more restrictive than "Werror", even though I
> know I'm wrong. In any case, you couldn't have asked much since the
> Werror and Wall were only added by Company after the previous release.
> HOWEVER, There were TWO prereleases allowing anyone to catch these kinds
> of mistakes BEFORE the release, and maybe even help fix.
I told you about them.
> > cc1: warnings being treated as errors
> > motion.c:87: warning: `dist1_MMX' declared `static' but never defined
> > motion.c:98: warning: `dist1_SSE' declared `static' but never defined
> These are relatively easy errors to fix and if you had been a good
> developer and used regular cvs then these would have poked your eye out a
> lot sooner. I think it's up to each of the gstreamer developers to do the
> right thing and fix warnings on their archs.
These errors I pointed out to you while we were testing the pre-release
> > I hope that it won't happen next time.
> The best way to ensure that is to help out. I had announced the fact that
> we were doing a release a week beforehand, yet still people succeed in
> showing up at the VERY LAST MINUTE complaining about stuff they could have
> fixed themselves a week before. To you it might not seem like that big of
> a deal, but I end up spending big chunks of my week-end trying to make
> sure we get a good release out and waiting for everyone to chip in their
> two cents long after the due date. Most of the time the end result is
> worth it, but not when someone comes back round complaining about stuff he
> told us to do and didn't get done. That kinda takes the fun out of it.
I know, but I also know that you have high standards, and presumably a
thick skin so that you don't take the piss when somebody makes comments
about your work. I also pointed out that -Werror was evil during the
code review... And if I was frb, I'd even tell you that it only works
with gcc and should only be used with gcc.
> The irony of it all is that I spent a whole week fixing up whatever I
> could and motivating other people to help fix stuff in support of a decent
> rhythmbox release to show other people the potential ;) And when I found a
> few inconsistencies in the rhythmbox spec files, I didn't tell you to fix
> them, I sent a patch. I hope that kind of behaviour rubs off...
That's because you know that I don't use the spec files ;)
> Oh, and sorry for the still relatively high piss factor of this mail.
I love you too, and you know that.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 232 bytes
Desc: This is a digitally signed message part
More information about the gstreamer-devel