[FFmpeg-devel] [PATCH 2/2] movtextenc.c: Support for Bold, Italic and Underlined styles

Philip Langdale philipl at overt.org
Fri Jun 26 21:05:49 CEST 2015


On 2015-06-26 09:48, Niklesh Lalwani wrote:
> The output produced here was actually different from the expected. 
> Hence
> the test failed. I produced ASS file for both the original file and the
> encoded file. The ass file produced from the encoded mp4 file seemed to 
> be
> dropping all text after \N
> After a lot of digging, I found out that while appending '\n' into the
> buffer in encoding, it is to be considered as a single character and 
> not 2.
> Making this change in movtextenc.c solved the issue and now I have 
> matching
> ASS outputs for both the files.
> However, the test still fails even though I have exact same ASS files.
> PS - there is no style information associated here as far as I can see.
> Please correct me if I'm going wrong here.

This test produces another mp4 file - not just an ASS subtitle (which is
what the other movtextenc test does. So something else you've done has
altered the output - perhaps through metadata or adding a box that 
wasn't
there before, etc.

> Thanks,
> Niklesh
> On Tue, 23 Jun 2015 22:09:20 +0200
> Michael Niedermayer <michaelni at gmx.at> wrote:
> 
>> On Wed, Jun 24, 2015 at 01:25:53AM +0530, Niklesh Lalwani wrote:
>> > Updated patch.
>> > Compiles without any warnings. Tested too.
>> 
>> this changes the fate checksum
>> 
>> --- ./tests/ref/fate/binsub-movtextenc  2015-06-23 21:52:17.741843656
>> +0200 +++ tests/data/fate/binsub-movtextenc   2015-06-23
>> 22:06:37.389861767 +0200 @@ -1 +1 @@
>> -ef264064c522389d0cf267c4d6235561
>> +59cad033309cebac08fda1ebd01677ac
>> TEST    sub-cc
>> Test binsub-movtextenc failed. Look at
>> tests/data/fate/binsub-movtextenc.err for details. make: ***
>> [fate-binsub-movtextenc] Error 1 make: *** Waiting for unfinished
>> jobs....
>> 
>> if thats intended then the chnage should be included in the patch
>> 
> 
> Likely this is because we're now carrying over formatting that we were
> dropping before. Niklesh, you should look at the fate sample file and
> verify this is what we should expect.
> 
> --phil
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

-- 
--phil


More information about the ffmpeg-devel mailing list