Bug#345489: [PATCH] Add support for spc plugin in EXTRA_PLUGINS

[Josh Triplett]

Loïc Minier wrote:
> On Sun, Jan 08, 2006, Josh Triplett wrote:
>>2) Anti Resonance's SPC emulator is generally considered the most
>>technically superior, both on the grounds of faithful reproduction and
>>possible enhancement; given that Anti Resonance's code and libopenspc
>>are equally non-portable, I don't think it's worth moving to a library
>>unless that library gives some other advantage.
> 
>  It seems reasonable to build against a shared library to ease security
>  upgrades.

That does indeed seem like "some other advantage". :)

>>That's a more serious concern; however, the code appears to be Freely
>>licensed by upstream.  Can you point to any particular issue or concern
>>you have, or that you've seen raised previously?
> 
>  No, I walked through the 0.8 bugs looking for things possibly obsolete
>  in 0.10, and wanted to give you some feedback on the discussions I
>  recall about that plugin.
>    This was both on IRC and in the mailing-lists IIRC.

OK.  Well, in the absence of someone pointing to some part of the code
that is non-free, I'm inclined to believe the Free licensing put in
place by upstream, for now.

>>I did notice that spc didn't seem to be present in 0.10.
> 
>  I think this is due to stricter policy upstream: plugin have to be
>  maintained actively by one person and blessed by some gstreamer hacker.
>  I think SPC lacks the active maintenance right now.

Ah, I see.  Does that include the -bad and -ugly sets of plugins?

>>> If you fixed configure.ac by:
>>> -GST_DOC()
>>> +GST_DOCBOOK_CHECK()
>>>
>>> Then this might be worthwhile to send upstream, could you explain how
>>> it break things to call GST_DOC instead of GST_DOCBOOK_CHECK?  I
>>> certainly see it is wrong, but I had no problem with it until now.
>>
>>I think it has already been fixed upstream, in newer versions than the
>>one currently in Debian.  The issue is that GST_DOC was renamed to
>>GST_DOCBOOK_CHECK, but configure.ac wasn't updated accordingly.  This
>>caused the immediately subsequent code to fail, which happened to be the
>>code which checked the target CPU to determine which arch-specific code
>>was acceptable; since SPC needs those target CPU variables set, it fails
>>unless this issue is fixed.
> 
>  Ok, I saw the rename in 0.10, but I only saw the GST_DOC() call in
>  configure today.  I'm likely to reupload a package to address that.

Cool.

>>Might it be possible to include this EXTRA_PLUGINS support until spc can
>>be sufficiently fixed to be more suitable for building by default?  It
>>would make enhancing and testing gstreamer0.8-spc significantly easier.
> 
>  That I've done in -4, but you replied faster than I uploaded.  :-P

Thanks!

- Josh Triplett

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 256 bytes
Desc: OpenPGP digital signature
Url : http://lists.alioth.debian.org/pipermail/pkg-gstreamer-maintainers/attachments/20060108/7b46f5c0/signature.pgp