On 2017/09/16 22:28, Jeremie Courreges-Anglas wrote: > > I have proposed tweaks for rgain: > > - run tests against the currently built module, I find this less > > error-prone (no need to make reinstall/update)
yes, that's better. > > - s/py-gobject/py-gobject3/, missing dep visible when importing > > the rgain.rgcalc module ack > > - skip one test that seems to hang, no idea of the root cause Ah I see what this is now, it hangs if gstreamer1-plugins-libav is not installed. > > I can use this against ogg files, but it looks like I'm missing some > > gstreamer modules or something... > > > > ritchie /usr/ports/mystuff/audio/rgain$ pkg_info | grep gstreamer1 > > gstreamer1-1.12.2 framework for streaming media > > gstreamer1-plugins-base-1.12.2 base elements for GStreamer > > gstreamer1-plugins-good-1.12.2p0 good elements for GStreamer > > gstreamer1-plugins-ugly-1.12.2 ugly elements for GStreamer > > ritchie /usr/ports/mystuff/audio/rgain$ replaygain -r 90 a.mp3 > > Checking for Replay Gain information ... > > a.mp3: none > > Calculating Replay Gain information ... > > a.mp3: > > Error while calculating gain - GST error: Il manque un greffon dans > > votre installation de GStreamer. (gstdecodebin2.c(4600): gboolean > > gst_decode_bin_expose(GstDecodeBin *) (): > > /GstPipeline:pipeline0/GstDecodeBin:decbin: > > no suitable plugins found: > > Missing decoder: MPEG-1 Layer 3 (MP3) (audio/mpeg, mpegversion=(int)1, > > mpegaudioversion=(int)1, layer=(int)3, rate=(int)44100, channels=(int)2, > > parsed=(boolean)true) > > ) > > Installing gstreamer1-plugins-libav helps handle mp3 files, maybe > another gstreamer component would help too, I'm no expert here. It > would be nice to document this, or just to add gstreamer1-plugins-libav > to RUN_DEPENDS. What do you think? ah the joys of having semi-optional deps already installed. RUN_DEPENDS makes sense for this. > This matter can be handled later. ok to import rgain and kill mp3gain. Thanks. Importing with the dep and test changes but not the patch. In later oss-sec posts there's some defence of mp3gain as it has an interesting hack, and there seems some interest in replacing the buggy decoder fork, so maybe it should come back later, but killing it for now seems sensible to me.
