[Tremor] ov_pcm_tell > ov_pcm_total by design?

[Kwasi Mensah]

Hi,

Checking in again to see if there's any progress. I have a work around but
it's a hack. I'd like to find out what I'm doing wrong if I am using tremor
the wrong way.

I tried debugging the tremor source code myself but I'm not uber familiar
with the specification and I don't have the time to learn it thoroughly.

Thanks,
Kwasi



On Mon, Dec 20, 2010 at 7:57 PM, Kwasi Mensah <kwasi.mensah at gmail.com>wrote:

> Hi,
>
> Has there been any progress on what could be wrong? Should I submit this as
> a formal bug (and where would I do that at?)
>
> Thanks,
> Kwasi
>
>
> On Wed, Dec 15, 2010 at 11:57 AM, Kwasi Mensah <kwasi.mensah at gmail.com>wrote:
>
>> Clarified the behavior I'm seeing in ov_read.
>>
>> On Wed, Dec 15, 2010 at 11:55 AM, Kwasi Mensah <kwasi.mensah at gmail.com>wrote:
>>
>>> It's a attached. It's one second of A220 mono at 44100hz.
>>>
>>> Also attached is the source for an example that will show the problem. I
>>> had to run it on my iPhone so its a .mm. Also, you'll have to add A220.ogg
>>> to the top level of you're .app and then recompile it to make the code
>>> signature process happy.
>>>
>>> For the last ov_read which pushes ov_pcm_tell of ov_pcm_totoal ov_read
>>> calls _fetch_and_process_packet which will call vorbis_synthesis_blockin
>>> after making sure oldsamples == 0.  vorbis_synthesis_blockin updates
>>>
>>>  v->pcm_returned to 0 and v->pcm_current to 1024.
>>>
>>> Now,  vorbis_synthesis_pcmout will always return 1024.
>>>
>>> Thanks,
>>> Kwasi
>>>
>>>
>>> On Wed, Dec 15, 2010 at 5:53 AM, <xiphmont at xiph.org> wrote:
>>>
>>>> On Wed, Dec 15, 2010 at 3:20 AM, Kwasi Mensah <kwasi.mensah at gmail.com>
>>>> wrote:
>>>> > Hey guys,
>>>> >
>>>> > I'm using Tremor cl 17618. vorbis_synthesis_blockin is doing some type
>>>> of
>>>> > overlap adding that I think is pushing ov_pcm_tell to be greater than
>>>> > ov_pcm_total when reading the last bit of audio data from an ogg file.
>>>> This
>>>> > causes a later call I make to ov_pcm_seek fail. Is this by design?
>>>> Should I
>>>> > just fix my call to ov_pcm_seek to take this into account?
>>>>
>>>> That is not by design; tell and total should agree (and do on the test
>>>> cases I have here).  Could you perhaps send me the file causing
>>>> trouble?
>>>>
>>>> Monty
>>>>
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.xiph.org/pipermail/tremor/attachments/20101229/64611087/attachment.htm