[Bug 620323] mpegaudioparse: Add support for LAME tags and adjust segments based on the padding information from it

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=620323

--- Comment #27 from Christoph Reiter (lazka) <reiter.christoph at gmail.com> ---
Looked into GstAudioClippingMeta a bit.

To correctly apply the clipping to each frame we need to know in which frame we
are based on the end or start of the stream and atm we can only detect if we
are in the first (offset=0) or in the last frame (offset + bpf == xing_bytes).
A solution would be to buffer frames so that all the delay/padding samples fit
in it and if we detect a first/last frame we can attach the
GstAudioClippingMeta to the affected buffered frames.

On the mpg123 side we would need to make sure that we can link input frames
with decoded ones, but that seems possible with some work.

Any idea how to buffer frames in mpegaudioparse/GstBaseParse?

Alternative: Only support one frame of delay/padding for now, that would at
least reduce the gap.

Thoughts?

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.