[FFmpeg-devel] [PATCH] ffprobe: generalize writer subsection nesting model

[Michael Niedermayer]

On Tue, Sep 18, 2012 at 01:45:33PM +0200, Stefano Sabatini wrote:
> On date Tuesday 2012-09-18 01:52:14 +0200, Michael Niedermayer encoded:
> > On Tue, Sep 18, 2012 at 12:43:55AM +0200, Stefano Sabatini wrote:
> > > On date Tuesday 2012-09-18 00:38:39 +0200, Stefano Sabatini encoded:
> > > > Discard unflexible structure based on the root/chapter/section structure
> > > > in favor of a generalized concept of section.
> > > > 
> > > > This should allow to represent sections at generic level of nesting, and
> > > > allow subsections printing selection.
> > > > 
> > > > Also, slightly simplify the code.
> > > > ---
> > > >  ffprobe.c |  515 +++++++++++++++++++++++++++++--------------------------------
> > > >  1 files changed, 242 insertions(+), 273 deletions(-)
> > > 
> > > Note: I tested the patch against a local fate test for each
> > > writer. The test file was generated with:
> > > 
> > > ffmpeg  -f lavfi -i "aevalsrc=sin(400*PI*2*t)::d=0.5[out0]; testsrc=d=0.5[out1]; testsrc=s=100x100:d=0.5[out2]" \
> > >    -flags +bitexact  -map 0:0 -map 0:1 -map 0:2 \ 
> > >    -metadata "title=ffprobe test file" \
> > >    -metadata "comment='A comment wih XML & JSON special chars': <tag value=\"x\">" \
> > >    -y tests/data/fate/ffprobe-test.nut
> > > 
> > > and I'm not sure if it is best to auto-generate it when fate is run or
> > > I should just push the generated file to the samples FATE collection.
> > 
> > some points to consider:
> > how large are the files?
> > how much time is spend by generating them?
> 
> > is it worth to test the creation of the files too?
> 
> Maybe, but this would also add a dependency on lavfi for testing
> ffprobe. On the bright side this would allow easy modifications of the
> generated file (e.g. for adding streams/metadata).
> 
> See patch attached (which needs review as I don't have deep knowledge
> of the fate system).

maybe you could reduce the duration of the files, the reference data
is kinda big

otherwise no comments from me, iam not a fate expert, maybe beastd or
jamal want to review it though ...

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

it is not once nor twice but times without number that the same ideas make
their appearance in the world. -- Aristotle
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20120918/94356afd/attachment.asc>