[FFmpeg-devel] [PATCH] Document sws_scale()

[Michael Niedermayer]

On Wed, Feb 18, 2009 at 09:44:01PM +0100, Stefano Sabatini wrote:
> On date Wednesday 2009-02-18 01:13:55 +0100, Stefano Sabatini encoded:
> > On date Tuesday 2009-02-17 03:26:17 +0100, Michael Niedermayer encoded:
> > > On Tue, Feb 17, 2009 at 01:50:18AM +0100, Stefano Sabatini wrote:
> > > > On date Tuesday 2009-02-17 01:37:43 +0100, Michael Niedermayer encoded:
> > > > > On Tue, Feb 17, 2009 at 01:15:53AM +0100, Stefano Sabatini wrote:
> > > > [...]
> > > > > > +/**
> > > > > > + * Scales the image slice in \p src and puts the resulting scaled
> > > > > > + * slice in the image in \p dst. A slice is a sequence of consecutive
> > > > > > + * rows in an image.
> > > > > > + *
> > > > > > + * @param context the scaling context previously created with
> > > > > > + *                sws_getContext()
> > > > > > + * @param src     the pointer to the array of pointers to the planes
> > > > > > + *                of the source slice (just one pointer in case of
> > > > > > + *                packed image formats)
> > > > >                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > > > [...]
> > > > 
> > > > Sorry, I didn't got you meant to remove *also* those.
> > > > 
> > > > New round.
> > > 
> > > ok
> > 
> > Sorry, I read again the patch and found it quite unsatisfying, I fixed
> > some things (
> > the pointer to the array of pointers to the planes ->
> > the array containing the pointers to the planes)
> > 
> > and rearranged spacing, but there is still something I don't like.
> > 
> > The srcSliceY definition:
> >  * @param srcSliceY the position in the source image of the source
> >  *                  slice to process, that is the number in the image
> >  *                  of the first row of the slice
> > 
> > maybe s/number/offset/ may help, or to mention that the numbering
> > starts with 0.
> > 
> > Then the expression "the strides or linesizes" is ambiguous, it's not
> > clear if the two names refer to the same thing or to two distinct
> > things. I tought about something like:
> > "strides (that is linesizes)"
> > or
> > "strides (also called linesizes)"
> > 
> > but they both seem quite ugly, the third option was to remove the
> > reference to the term "linesize", but that would make the
> > documentation less helpful since linesize is a term already used at
> > least in lavc and so it may result more familiar to the user than
> > "stride" (term which is only used in libsws).
> > 
> > Suggestions are welcome, otherwise I'll try to figure out what to do
> > and will post another patch.
> 
> I believe it should be somehow better now, removed ambiguous mention
> to the term "linesize", maybe it should be documented somewhere else
> (or maybe as a @note?).

ok

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

If you really think that XML is the answer, then you definitly missunderstood
the question -- Attila Kinali
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090218/41072b96/attachment.pgp>