[gstreamer-bugs] [Bug 377280] [cdiocddasrc] issue if drive endianness != machine endianness

[GStreamer (bugzilla.gnome.org)]

Do not reply to this via email (we are currently unable to handle email
responses and they get discarded).  You can add comments to this bug at
http://bugzilla.gnome.org/show_bug.cgi?id=377280

  GStreamer | gst-plugins-good | Ver: 0.10.4





------- Comment #11 from Tim-Philipp Müller  2006-11-21 11:45 UTC -------
> I tried looking at cdiocddasrc
> http://webcvs.freedesktop.org/gstreamer/gst-plugins-good/ext/cdio/ but I don't
> see the paranoia part let alone the part where the CPU endianness is taken 
> into account (as it is in libcdio/src/cd-paranoia/cd-paranoia.c).

That's correct. The cdio plugin doesn't use any of the cdparanoia API. That's
partly because it has been written before that API made it into libcdio. That's
also why I was asking for a solution to this problem without involving the
cd-paranoia API.

The CPU endianness isn't "taken into account" in the code, it is simply
assumed. The assumption was simply that cdio delivers samples in cpu endianness
(made in the absence of any hints that this might not be so). I see now that
this assumption is incorrect.

>From your comments and the header files I conclude then that when reading audio
using the 'raw' cdio API, the endianness is undefined and there is no API to
find out either, nor will there be. This means we'll have to up the cdio
requirements and use cdio-cdparanoia. That's fine with me, I was just trying to
solve this without involving cdio-cdparanoia if possible first (since up'ing
the requirements obviously means the plugin won't be available any longer on
certain older distributions).


-- 
Configure bugmail: http://bugzilla.gnome.org/userprefs.cgi?tab=email