Bug in codec L16

AR
Alicia Romero
Thu, Aug 6, 2015 2:05 PM

Hi List,
I think I have found a mistake in the L16 codec. In the function
l16_parse() the timestamp is so calculated:

frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size);

But I think instead of using the frame size, it should use samples per
frame, like in the other codecs:

frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame);

Am I right, or am I delusional??

Thanks!!

Hi List, I think I have found a mistake in the L16 codec. In the function l16_parse() the timestamp is so calculated: frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size); But I think instead of using the frame size, it should use samples per frame, like in the other codecs: frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame); Am I right, or am I delusional?? Thanks!!
M
Ming
Fri, Aug 7, 2015 9:23 AM

Hi Alicia,

You're right, of course. We've just fixed this in r5153
(https://trac.pjsip.org/repos/changeset/5153).
Thanks for spotting this!

Regards,
Ming

On Thu, Aug 6, 2015 at 10:05 PM, Alicia Romero alromor84@gmail.com wrote:

Hi List,
I think I have found a mistake in the L16 codec. In the function l16_parse()
the timestamp is so calculated:

frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size);

But I think instead of using the frame size, it should use samples per
frame, like in the other codecs:

frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame);

Am I right, or am I delusional??

Thanks!!


Visit our blog: http://blog.pjsip.org

pjsip mailing list
pjsip@lists.pjsip.org
http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org

Hi Alicia, You're right, of course. We've just fixed this in r5153 (https://trac.pjsip.org/repos/changeset/5153). Thanks for spotting this! Regards, Ming On Thu, Aug 6, 2015 at 10:05 PM, Alicia Romero <alromor84@gmail.com> wrote: > Hi List, > I think I have found a mistake in the L16 codec. In the function l16_parse() > the timestamp is so calculated: > > frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size); > > But I think instead of using the frame size, it should use samples per > frame, like in the other codecs: > > frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame); > > Am I right, or am I delusional?? > > Thanks!! > > > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip@lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >
AR
Alicia Romero
Fri, Aug 7, 2015 9:56 AM

Thanks for your answer! I had an old version and I didn't know it was
already fixed.

2015-08-07 11:23 GMT+02:00 Ming ming@teluu.com:

Hi Alicia,

You're right, of course. We've just fixed this in r5153
(https://trac.pjsip.org/repos/changeset/5153).
Thanks for spotting this!

Regards,
Ming

On Thu, Aug 6, 2015 at 10:05 PM, Alicia Romero alromor84@gmail.com
wrote:

Hi List,
I think I have found a mistake in the L16 codec. In the function

l16_parse()

the timestamp is so calculated:

frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size);

But I think instead of using the frame size, it should use samples per
frame, like in the other codecs:

frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame);

Am I right, or am I delusional??

Thanks!!


Visit our blog: http://blog.pjsip.org

pjsip mailing list
pjsip@lists.pjsip.org
http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org

Thanks for your answer! I had an old version and I didn't know it was already fixed. 2015-08-07 11:23 GMT+02:00 Ming <ming@teluu.com>: > Hi Alicia, > > You're right, of course. We've just fixed this in r5153 > (https://trac.pjsip.org/repos/changeset/5153). > Thanks for spotting this! > > Regards, > Ming > > On Thu, Aug 6, 2015 at 10:05 PM, Alicia Romero <alromor84@gmail.com> > wrote: > > Hi List, > > I think I have found a mistake in the L16 codec. In the function > l16_parse() > > the timestamp is so calculated: > > > > frames[count].timestamp.u64 = ts->u64 + (count * data->frame_size); > > > > But I think instead of using the frame size, it should use samples per > > frame, like in the other codecs: > > > > frames[count].timestamp.u64 = ts->u64 + (count * samples_per_frame); > > > > Am I right, or am I delusional?? > > > > Thanks!! > > > > > > _______________________________________________ > > Visit our blog: http://blog.pjsip.org > > > > pjsip mailing list > > pjsip@lists.pjsip.org > > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org > > > > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip@lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >