[Logo] Jaikoz and SongKong Forums
  [Search] Search   [Recent Topics] Recent Topics   [Members]  Member Listing   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
More MusicBrainz tagging issues  XML
Forum Index -> SongKong Issues
Author Message
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

Sorry to keep raising questions like this, but I did say I'd road-test the classical functionality!
Just ripped https://musicbrainz.org/release/42bb2ee7-2102-3db9-9f0b-1bb43055683f
Mostly it's fine, but SongKong has not picked up the MusicBrainz_Work and Work_composition for the Variations on a Theme of Paganini. The MB page seems to indicate they are all present and correct (not partial etc.). Any ideas?
Also Rachmaninov is applied to the Arranger tag by SK, as well as to composer. This is arguably correct for Piano Concerto No.1 where MB lists him as composer and "additional composer" (reworked it later) but for all the other pieces, MB does not show an arranger - just Rachmaninov as composer. There is no Discogs data to confuse things and it is clear that SK added Arranger as a new tag, so I am puzzled.
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

MDE wrote:
Sorry to keep raising questions like this, but I did say I'd road-test the classical functionality!
 

Not a problem at all, I cannot afford to employ a tester so its great you are taking the time to do this and giving me such useful feedback !

MDE wrote:

Just ripped https://musicbrainz.org/release/42bb2ee7-2102-3db9-9f0b-1bb43055683f
Mostly it's fine, but SongKong has not picked up the MusicBrainz_Work and Work_composition for the Variations on a Theme of Paganini. The MB page seems to indicate they are all present and correct (not partial etc.). Any ideas?
 

Okay, I think its related to a work-work relationship that the work is part of, this work has an attribute of (op 43) o its work relationship https://musicbrainz.org/work/17eea09b-0497-370c-a0a2-4f7f2d49300f/edit and that may inadvertently be filtering out this recording/work. I'llook into it in more detail after lunch.

MDE wrote:

Also Rachmaninov is applied to the Arranger tag by SK, as well as to composer. This is arguably correct for Piano Concerto No.1 where MB lists him as composer and "additional composer" (reworked it later) but for all the other pieces, MB does not show an arranger - just Rachmaninov as composer. There is no Discogs data to confuse things and it is clear that SK added Arranger as a new tag, so I am puzzled. 

Yep, unfortunately I introduced a bug in the databae trying to fix the piano arranger issue, it caused some composers to be added as arranger as well. Its now fixed as of 5 minutes ago, if you remove the metadata and retry the issue should not recoccur.

thanks Paul (Administrator)
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

Found the issue, its due to a work higher up the change having a based on relationship to another work

https://jthink.atlassian.net/browse/SONGKONG-1190

thanks Paul (Administrator)
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

if you remove the metadata and retry the issue should not recoccur. 

Did that and it re-added arranger for every track
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

I turned off server cache a few minutes ago but I think problem may be you need to empty SongKong cache by doing File:Empty Database and then retry, which reminds me https://jthink.atlassian.net/browse/SONGKONG-119

thanks Paul (Administrator)
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

That's fixed the arranger issue. I assume the work issue will take a little longer.
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

Re the work issue, this one: https://musicbrainz.org/release/ca55b3d0-5a2a-4f1b-88a8-ae124f28258b also had missing MUSICBRAINZ_WORK... tags for tracks 4 - 14. Again, the problem seems to be that the higher-level work "is based on" another work. This will frequently be the case for classical music, where composers often create orchestral suites based on larger works. The MB style guideline is (quite correctly) to treat the suite as a separate work. I agree that SongKong should ignore the "is based on" relationship. (Muso users can easily see the MB page and navigate wherever they wish!)
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

And this one too: https://musicbrainz.org/release/d40d3284-996d-4b29-a472-55ed7a6c5c3e
No MUSICBRAINZ_WORK.. tags at all. The first track is a bit odd (but still should get a MUSIC_BRAINZ_WORK_COMPOSITION I think), while the others look pretty regular to me.
I'll stop ripping for a bit until we get to the bottom of this!
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

So the based on issue is quite simple to fix but requires a database rebuild, will get this done this week.

Regarding the latest issue, the work for first track is not added because the work the work is linked to has a revision of relationship with another work, this is essentially the same issue (has a not part of relationship with a work and will be fixed as above)

I cant see why the others dont have a work added, can you send support files please.

thanks Paul (Administrator)
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

The issue with works that are linked to a work that have a based on or revision of relationship is now fixed, please run File:Empty Database and retry.

I havent yet solved the issue with https://musicbrainz.org/release/d40d3284-996d-4b29-a472-55ed7a6c5c3e it is now showing a work for track 1 but none of the others.

thanks Paul (Administrator)
MDE

Pro

Joined: 11/03/2017 05:19:12
Messages: 58
Offline

Tested & seems OK. I'll assume that the missing data for https://musicbrainz.org/release/d40d3284-996d-4b29-a472-55ed7a6c5c3e is a bit of a one-off and resume ripping now.
paultaylor

Pro
[Avatar]

Joined: 21/08/2006 09:21:27
Messages: 7297
Offline

The issue with https://musicbrainz.org/release/d40d3284-996d-4b29-a472-55ed7a6c5c3e has now been fixed in Albunack database.

thanks Paul (Administrator)
 
Forum Index -> SongKong Issues
Go to:   
Powered by JForum 2.1.6 © JForum Team