[FFmpeg-devel] [RFC] replace some static with asm_visibility or?so
Wed Jan 30 13:13:04 CET 2008
On Wed, Jan 30, 2008 at 12:28:05PM +0100, Balatoni Denes wrote:
> Wednesday 30 January 2008 09:11-kor Benoit Fouet ezt ?rta:
> > > ps: there is one thing about the high coding standards - I understand the
> > > need, however it would be great if in the first reply (or at least soon)
> > > it was made clear what is the needed minimum for svn acceptance.
> > could you be clearer ?
> > isn't the 'coding rules' section enough ?
> I think I was pretty clear. You know, in a review checking the indent size and
> doxy comments is a very small part, the showstoppers are usually something
> completely different (e.g. in the mms patch, the problem right now is the
> FSM). These issues should be pointed out early, so that the contributor has
> an estimate of much work he has to do still. Otherwise - among other things -
> it could happen (and seems like it does happen), that the contributor fixes
> things for some time - he believes he will get somewhere with the fixes - but
> eventually he finds that every time there is a new issue, and gives up
> (because of fustration, or simply lack of time, as sometimes people are
> working on ffmpeg as a hobby).
There are 2 problems ...
1. If he knew the amount of time at the begin chances are he wouldnt even
start fixing anything.
2. Reviews take time, and spoting missing doxygen and such is VERY easy.
Realizing that there are fundamental design issues or otherwise that
something can be redesigned so it is much simpler and still otherwise
equivalent takes much more time and understanding of the
code. So these things are not always found in the first review iteration.
Now one could argue that the first review should be more complete and
more time should be spend to not miss anything. But that would cause
significantly longer review cycles. And if the author choose not to
fix anything due to the amount of work, the review time would be
And please dont forget reviewing is about checking that the code and its
design is (near) optimal, much more than it is about checking that
indention and doxygen matches the rules. So skiping that would definitly
not do any good. It would be like spellchecking a document whos content
makes no sense.
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
Democracy is the form of government in which you can choose your dictator
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: Digital signature
More information about the ffmpeg-devel