Well, I forgot classic music so I was wrong :P. nmm:albumPerformer, with a cardinality 0:n is totally right. It's not handled by id3 tags but for classic music is in several cases the main difference between two music albums.
On Wed, Jun 13, 2012 at 12:18 PM, Ignacio Serantes <[email protected]> wrote: > > > On Wed, Jun 13, 2012 at 11:27 AM, Sebastian Trüg <[email protected]>wrote: > >> On 06/13/2012 11:04 AM, Ignacio Serantes wrote: >> >>> Hi, >>> >>> On Wed, Jun 13, 2012 at 10:52 AM, Sebastian Trüg <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> On 06/10/2012 08:32 PM, Ignacio Serantes wrote: >>> >>> I brief update status. >>> >>> 1) I just fix multiple performers bug in flac analyzer. >>> >>> 2) nmm:albumArtist seems to be related to DB structure and is not >>> a >>> coding bug. I just added nmm:albumArtist to SDO and now there is a >>> nmm:albumArtist in nmm:MusicAlbum. This is the code I added: >>> nmm:albumArtist >>> a rdf:Property, nrl:DefiningProperty ; >>> rdfs:subPropertyOf nco:contributor ; >>> rdfs:label "album artist" ; >>> rdfs:comment "Album artist" ; >>> rdfs:domain nmm:MusicAlbum ; >>> rdfs:range nco:Contact . >>> >>> >>> This looks good. I am only concerned about having performer and >>> producer on the tracks and artist on the album. Shouldn't that be in >>> sync somehow? After all an album is also produced, right? A >>> compilation has a producer and each track has a possibly different >>> producer... >>> >>> >>> This is right but as there is no tag for album producer this information >>> can not be obtained from the music file so this is outside my actual >>> goals. Album producers can be obtained using a query, as I'm doing in >>> Nepoogle for album performers. >>> >> >> What I mean is we need nmm:albumProducer and nmm:albumPerformer. What do >> you think? >> > > nmm:albumProducer don't bothers but I'm not sure about nmm:albumPerformer. > Performer is the "performer" and not the album artist, yes my Engrish is > bad. An album artist is a good field because and album always have an > associated artist: Queen, ELO or Various artists, but in the real world > there is not a performer associated with the album but is associated to > music piece. > > Album producer is not the same case because, sometimes, there is a general > album producer and other producers for music pieces. So album producer and > artist producer are the same case but not for performers, lyricists or > composers. It's right that many times performers, lyricist or composers are > the same for all the music album but I still consider that this are music > piece fields. > > > >> >> >>> >>> >>> I will upload a patch to Review Board for the first one but, >>> what must I >>> do with the second one? >>> >>> >>> The normal approach is to create a ticket at >>> >>> https://sourceforge.net/apps/_**_trac/oscaf/<https://sourceforge.net/apps/__trac/oscaf/> >>> >>> >>> <https://sourceforge.net/apps/**trac/oscaf/<https://sourceforge.net/apps/trac/oscaf/> >>> > >>> >>> >>> Ok, thank you. I will add a ticket. >>> >>> >>> >>> On Thu, Jun 7, 2012 at 7:59 PM, Ignacio Serantes <[email protected] >>> <mailto:[email protected]> >>> <mailto:[email protected] <mailto:[email protected]>>> wrote: >>> >>> Hi, >>> >>> I detect next issues with music albums scanning flac and mp3 >>> formats: >>> >>> 1) Performers: >>> >>> * mp3: supports it but it wrongly adding nmm:albumArtist >>> too and >>> >>> this must be related to nmm:MusicAlbum. >>> * flac: only adds one performer, the last added to the >>> file?. >>> >>> nmm:musicArtist it's not imported and an error is >>> launched: >>> "Cannot set values for abstract property >>> 'http://www.semanticdesktop.__**org/ontologies/2009/02/19/nmm#** >>> __albumArtist >>> <http://www.semanticdesktop.**org/ontologies/2009/02/19/nmm#** >>> albumArtist<http://www.semanticdesktop.org/ontologies/2009/02/19/nmm#albumArtist> >>> >' >>> <http://www.semanticdesktop.__**org/ontologies/2009/02/19/nmm#** >>> __albumArtist >>> >>> <http://www.semanticdesktop.**org/ontologies/2009/02/19/nmm#** >>> albumArtist<http://www.semanticdesktop.org/ontologies/2009/02/19/nmm#albumArtist> >>> >'>.". >>> >>> Solution: >>> >>> * 0 to n nmm:albumArtist resources must be added to >>> nmm:MusicAlbum. >>> * 0 to n nmm:performer resources must be added to >>> nmm:MusicPiece. >>> >>> >>> 2) Albums: >>> >>> * there is no url in nmm:MusicAlbum so two different >>> albums with >>> >>> same name are considered one unique resource. >>> * determine album url is tricky, you need to obtain the >>> url from a >>> >>> track, a method complicated because previous problem. >>> >>> Solution: >>> >>> * add nie:url to albums. >>> * two albums with same name but different path are different >>> >>> albums. This have a side effect with sets if sets are in >>> different paths, but I think is better that sets are >>> considered >>> different albums, than two different albums are the same >>> resource. On the other side, this could be handle with >>> some >>> smart path detection assuming some kind of organization >>> with >>> paths. Other solution could be using nmm:albumArtis to >>> determine >>> if two albums are different but, then we have the >>> problem with >>> various artists compilations. >>> * added nfo:depiction as a cover, an image that could be >>> handled >>> by Bangarang or Nepoogle. >>> * maybe more nfo:depiction for scans? >>> >>> >>> 3) Genres: >>> >>> * mp3: is not working. >>> * flac: works fine, supporting unlimited genres. >>> >>> Solution: >>> >>> * bug with mp3 files must be fixed. >>> >>> >>> 4) Sets: >>> >>> * there is no total tracks per set. >>> * the total tracks is equal to the total tracks number in >>> the last >>> track scanned. >>> >>> Solution: >>> >>> * a solution similar as seasons in tvshows. >>> * a total tracks and a total tracks per set must be handled. >>> >>> >>> I'm only using flac and mp3 so the same or other issues >>> could be in >>> other music formats. >>> >>> Sets problems is a minor issue, but the other three, must be >>> fixed. >>> >>> As practically all my music is in flac format I will try to >>> fix >>> problems 1 and 2 in flac analyzer, I'm crossing my fingers >>> because >>> is C++, but I wish to hear comments/suggestions before begin >>> to code. >>> >>> -- >>> Best wishes, >>> Ignacio >>> >>> >>> >>> >>> >>> -- >>> Best wishes, >>> Ignacio >>> >>> >>> >>> >>> ______________________________**___________________ >>> Nepomuk mailing list >>> [email protected] <mailto:[email protected]> >>> >>> https://mail.kde.org/mailman/_**_listinfo/nepomuk<https://mail.kde.org/mailman/__listinfo/nepomuk> >>> >>> <https://mail.kde.org/mailman/**listinfo/nepomuk<https://mail.kde.org/mailman/listinfo/nepomuk> >>> > >>> >>> ______________________________**___________________ >>> Nepomuk mailing list >>> [email protected] <mailto:[email protected]> >>> >>> https://mail.kde.org/mailman/_**_listinfo/nepomuk<https://mail.kde.org/mailman/__listinfo/nepomuk> >>> >>> >>> <https://mail.kde.org/mailman/**listinfo/nepomuk<https://mail.kde.org/mailman/listinfo/nepomuk> >>> > >>> >>> >>> >>> >>> -- >>> Best wishes, >>> Ignacio >>> >>> >>> >>> >>> ______________________________**_________________ >>> Nepomuk mailing list >>> [email protected] >>> https://mail.kde.org/mailman/**listinfo/nepomuk<https://mail.kde.org/mailman/listinfo/nepomuk> >>> >> ______________________________**_________________ >> Nepomuk mailing list >> [email protected] >> https://mail.kde.org/mailman/**listinfo/nepomuk<https://mail.kde.org/mailman/listinfo/nepomuk> >> > > > > -- > Best wishes, > Ignacio > > > -- Best wishes, Ignacio
_______________________________________________ Nepomuk mailing list [email protected] https://mail.kde.org/mailman/listinfo/nepomuk
