On Jul 01 17:01:41, Stuart Henderson wrote: > On 2011-07-01, Giovanni Bechis <[email protected]> wrote: > > On 07/01/11 15:39, David Coppa wrote: > >> To anyone who wants to import it, you have my ok. > >> > > Enable regression test as well, ok for me. > > The following ports are very likely to pick this up:
What does 'pick it up' mean? If they want to use opencore-amr, they will put it in their *_DEPENDS, edit their configure --options, etc. They will not just 'pick it up' by accident, right? > audio/sox I will take care of audio/sox. In fact, I have an update for audio/sox ready, as my main motivation for porting AMR was to have AMR functionality in SoX. Index: Makefile =================================================================== RCS file: /cvs/ports/audio/sox/Makefile,v retrieving revision 1.49 diff -u -p -r1.49 Makefile --- Makefile 15 Jun 2011 08:29:34 -0000 1.49 +++ Makefile 1 Jul 2011 22:13:31 -0000 @@ -3,7 +3,7 @@ COMMENT= Sound eXchange, the Swiss Army knife of audio manipulation DISTNAME= sox-14.3.2 CATEGORIES= audio HOMEPAGE= http://sox.sourceforge.net/ @@ -18,8 +18,10 @@ PERMIT_DISTFILES_FTP= Yes MODULES= converters/libiconv WANTLIB += c m ogg sndio z -WANTLIB += vorbis vorbisenc vorbisfile FLAC magic -WANTLIB += mad id3tag mp3lame wavpack png gsm +WANTLIB += vorbis vorbisenc vorbisfile FLAC +WANTLIB += opencore-amrnb opencore-amrwb +WANTLIB += mad id3tag mp3lame wavpack +WANTLIB += magic png gsm MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=sox/} @@ -31,6 +33,7 @@ LIB_DEPENDS= audio/libvorbis \ audio/libid3tag \ audio/lame \ audio/wavpack \ + audio/opencore-amr \ graphics/png \ audio/gsm @@ -58,8 +61,8 @@ CONFIGURE_ARGS+=--without-sndfile \ --with-mad \ --with-id3tag \ --with-lame \ - --without-amrwb \ - --without-amrnb \ + --with-amrnb \ + --with-amrwb \ --with-wavpack \ --with-png \ --without-ladspa > graphics/ffmpeg ffmpeg's configure recognizes --enable-libopencore-amrnb --enable-libopencore-amrwb which default to [no] (but shouldn't ffmpeg's CONFIGURE_ARGS say --disable-libopencore-amrnb --disable-libopencore-amrwb explicitly if it doesn't use it?). Currently, the output of ffmpeg's configure just says libopencore-amrnb support no libopencore-amrwb support no even with opencore-amr installed and seems to ignore it (unlike other stuff, whose --enable-* defaults to [autodetect]) Index: Makefile =================================================================== RCS file: /cvs/ports/graphics/ffmpeg/Makefile,v retrieving revision 1.73 diff -u -p -r1.73 Makefile --- Makefile 24 Jun 2011 11:16:21 -0000 1.73 +++ Makefile 1 Jul 2011 22:07:48 -0000 @@ -87,6 +87,8 @@ CONFIGURE_ARGS+= ${CONFIGURE_SHARED} \ --disable-outdev=oss \ --enable-gpl \ --enable-libgsm \ + --disable-libopencore-amrnb \ + --disable-libopencore-amrwb \ --enable-libmp3lame \ --enable-libschroedinger \ --enable-libspeex \ > multimedia/avidemux > multimedia/gstreamer-0.10 (plugins-bad and plugins-ugly) > > These need to be built with opencore-amr already installed and checked as > to whether they need additional deps listing (WANTLIB/LIB_DEPENDS as usual, > but also LIB*_EXTRALIBS in ffmpeg). Ports depending on these also need to > be checked for additional WANTLIB. What happens if some port's build chooses to use the installed opencore-amr libs, but doesn't mention it in WANTLIB etc? > I know this is a bunch of work - but I am not OK with importing this > until it's done as it risks breaking a lot of packages. Would it be OK to explicitly --disable AMR in these ports (before someone else chooses to correctly add the AMR functionality to those ports), as above for ffmpeg? Jan
