[theora-dev] further debugging of my ogg/theora decoder

[salsaman]

Aha ! Solved it, it was a single line of code:

while (ogg_stream_packet_out() && !frame_out) {
  ..
}

should have been:

while (!frame_out && ogg_stream_packet_out() ) {

...

}


Doh !

:-)


Gabriel.



On Tue, Nov 11, 2008 at 10:31 AM, salsaman <salsaman at gmail.com> wrote:

>
>
> On Tue, Nov 11, 2008 at 5:10 AM, salsaman <salsaman at gmail.com> wrote:
>
>>
>>
>> On Tue, Nov 11, 2008 at 4:48 AM, Ralph Giles <giles at xiph.org> wrote:
>>
>>> On 10-Nov-08, at 11:30 PM, salsaman wrote:
>>>
>>>  get page at 1773554
>>>> got packet
>>>> frame++, skip is 0
>>>> frame decoded
>>>> page processed, kframe=444 : offs=444
>>>>
>>>
>>> Why doesn't offs reset to zero here?
>>>
>>>  -r
>>>
>>
>>
>> My mistake - it actually outputs the frame number rather than the offset.
>> Subtract kframe from offs to get the real offset.
>>
>> Gabriel.
>>
>>
>
> Now I am not so sure this has anything to do with the audio packets
> (although that may be a problem too).
>
> What I am seeing in another file is, for example:
>
> first set of pages decodes to 5 frames (last granulepos is 1,5)
>
> - decoding in sequence without returning anything results in 5 packets
> being decoded - good
>
> - decoding in sequence, returning a frame each time and then re-entering
> the plugin to get the next frame results in only 3 packets being decoded
>
> I don't understand this at all, because after returning the processing
> continues where it left off, with ogg_stream_packetout(), using the same
> stream data...
>
> Gabriel.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.xiph.org/pipermail/theora-dev/attachments/20081111/05e05c89/attachment-0001.htm