ogg dev - Jun 2008 - [PATCH] browser_plugin - kate support, build fixes, and misc

2008/4/30 ogg.k.ogg.k at googlemail.com <ogg.k.ogg.k at
googlemail.com>:
> > > Let me know if this works for you or not; eg. it may be necessary
to
>  > > also put X_EXTRA_LIBS etc. in src/Makefile.am.
>  >
>  > will try this.
>
>  It worked fine for me, with no need to add anything to Makefile.am.
>
good
>  > > >   - bad pointer dereference fix (mouse callback)
>  > >
>  > > I couldn't find that particular hunk, could you please
provide a patch
>  > > which just fixes this?
>  >
>  > Two hunks, the first one to initialize the mouse callback pointer
>  > to NULL, and the second one to test for NULL before actually using
it:
>
>  That didn't compile though, sorry, the attached patch fixes that. I
edited
>  the original patch by hand to remove the Kate stuff and it seems I ended
>  up with a stray comma.
sorry, that was my fault with the merge; I fixed it a few hours later in r3576
>  Also, I'd entirely forgotten about that, but I need to set
FIREFOX_CFLAGS
>  and FIREFOX_LIBS to (resp) "-I$(ffpath)/include" and
"-L$(ffpath)/lib" for
>  configure to not whine it can't find Firefox. I've got the Gecko
SDK, but
>  I can't see any pc files in there, so the configure.ac check's got
me
>  puzzled. Since it all builds and run if I do that, I can only guess that
>  the reason is that I'm using a prebuilt version (I did hear that FF is
a
>  real pain to build, and my dev box is still having FF 1 :)). I've got:
>  gecko-sdk-i686-pc-linux-gnu-1.8.0.4.tar.bz2
>  If you decide you don't want to support building against prebuilt
SDKs,
>  then that's fine by me, just wanted to mention it.
hmm, perhaps we should add some configure arguments like
--with-firefox-cflags= and --with-firefox-libs
>  Next, the audio code doesn't build for me, same thing about old
software.
>  My /usr/include/alsa/version.h says "1.0.9". Various functions
are being
>  undefined. I can supply the list if needed, but you may want to decide to
>  not support old versions.
>
>  Last, the building of the static lib fails, as it tries to pull liboggz,
>  liboggplay, etc, out of /usr/local/lib (well, @LIBDIR@) rather than from
>  wherever pkg-config says they are.
ok, what architecture are you on? that's been reported on x86_64 but
may occur elsewhere
>  I'd omitted these chunks from my previous patch (esp. the ALSA one,
which
>  is down to my own peculiar configuration), but that last one (which I do
>  not know how to fix) is likely to be annoying, as I suppose a distribution
>  of the plugin as a binary is going to include the static one, rather than
>  the shared one (just guessing here).
>
>  If you know how to fix it, feel free to tell me if there's a patch you
>  want me to test here.
will do, but we don't have a patch for that yet :-)

cheers,

Conrad.

> hmm, perhaps we should add some configure arguments like
> --with-firefox-cflags= and --with-firefox-libs
Yes, would make sense. I can try to add those later this week.
> >  Last, the building of the static lib fails, as it tries to pull
liboggz,
> >  liboggplay, etc, out of /usr/local/lib (well, @LIBDIR@) rather than
from
> >  wherever pkg-config says they are.
>
> ok, what architecture are you on? that's been reported on x86_64 but
> may occur elsewhere
i686-pc-linux-gnu, but I reckon the arch is irrelevant since all the libs are
pulled from @LIBDIR@, so it'll fail for every box where the libs are not
installed there.

Also, I just noticed one of your latest patches renames LIBOGGPLAY to
OGGPLAY in configure.ac - part of my earlier patch was to fixup the
CFLAGS/LIBS in Makefile.am to match, so you should now revert this
bit of my patch as your patch moots it (and restores the break I had).

Thanks

> hmm, perhaps we should add some configure arguments like
> --with-firefox-cflags= and --with-firefox-libs
I've made it a single one, --with-gecko-sdk= and also default
include path and xpidl path to that if specified.

Also included are a patch that renames LIBOGGPLAY_* vars
to OGGPLAY_* to match the latest change to configure.ac, and
one that allows the audio to be compiled out - that one's you
may want to ignore, it just doesn't compile for me (ALSA 1.0.9)
but I sent that just in case.

Patches separated this time, sorry for not doing this earlier :)

Cheers
-------------- next part --------------
A non-text attachment was scrubbed...
Name: browser_plugin-with-gecko-sdk.diff
Type: text/x-patch
Size: 1697 bytes
Desc: not available
Url :
http://lists.xiph.org/pipermail/ogg-dev/attachments/20080505/70f680aa/attachment.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: browser_plugin-oggplay-liboggplay.diff
Type: text/x-patch
Size: 1258 bytes
Desc: not available
Url :
http://lists.xiph.org/pipermail/ogg-dev/attachments/20080505/70f680aa/attachment-0001.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: browser_plugin-sa-noop.diff
Type: text/x-patch
Size: 3655 bytes
Desc: not available
Url :
http://lists.xiph.org/pipermail/ogg-dev/attachments/20080505/70f680aa/attachment-0002.bin

>> hmm, perhaps we should add some configure arguments like
>> --with-firefox-cflags= and --with-firefox-libs>
> I've made it a single one, --with-gecko-sdk= and also default
> include path and xpidl path to that if specified.
Hi,

I noticed this hasn't been applied, so I'm sending it again as it's
quite simple
and, I think, unobtrusive.
It allows building against a prebuilt Gecko SDK.
If it needs changing to be applied, feel free to tell me what you'd
like changed.

Cheers
-------------- next part --------------
A non-text attachment was scrubbed...
Name: browser_plugin-with-gecko-sdk.diff
Type: text/x-patch
Size: 1697 bytes
Desc: not available
Url :
http://lists.xiph.org/pipermail/ogg-dev/attachments/20080602/500f9bfe/attachment.bin

2008/5/5 ogg.k.ogg.k at googlemail.com <ogg.k.ogg.k at
googlemail.com>:
>> hmm, perhaps we should add some configure arguments like
>> --with-firefox-cflags= and --with-firefox-libs>
> I've made it a single one, --with-gecko-sdk= and also default
> include path and xpidl path to that if specified.
(applied after an email reminder, r3605)
> Also included are a patch that renames LIBOGGPLAY_* vars
> to OGGPLAY_* to match the latest change to configure.ac, and
the same fix was applied in r3600 and should have been attributed to
you as you'd sent this patch earlier, sorry
> one that allows the audio to be compiled out - that one's you
> may want to ignore, it just doesn't compile for me (ALSA 1.0.9)
> but I sent that just in case.
that could be useful if generalized; currently it just disables sound
for everyone ;-) Perhaps you could make it set a configure flag if
audio is explicitly disabled, or if no audio driver is detected, and
build an appropriate null audio device.
> Patches separated this time, sorry for not doing this earlier :)
Thanks, sorry for ignoring them anyway ;-)

Conrad.