Is your feature request related to a problem? Please describe.
The database contains instances of multiple entries for discs with the same hash. Conceptually, one disc belongs to many releases and one release contains many discs. Currently, for some titles (at least pre-web contribution interface) one disc hash may belong to multiple discs, each in their own release. This creates circumstances where a disc that is common to multiple releases is present in the database multiple times which are not synchronized.
Describe the solution you'd like
This may be a bit complicated because it would involve some back-end work, but ultimately a given disc hash should exist uniquely in the database and each release that includes that hash should reference that unique entry instead of having it's own copy of the disc entry. I'm not sure what this looks like in the web's backend database, but in the git repo this is manifested as multiple disc0X.json files having the same disc hash but different contents.
I understand a disc hash check is performed as part of the web-based contribution process now. I'm not sure as I haven't yet had an opportunity to try to upload a duplicate disc. I suspect this may be moot for contributions going forward, but at the very least there is some cleanup to do for past contributions, in which case I guess this is really a bug issue instead of a feature request.
This might also fully break compatibility with the desktop tool as it might take significant work to adapt the tool. This would probably require that all of the discXX.json files be relocated from the movie folders to a separate discs folder, stored by hash. The release.json file would then be modified to identify the discs comprising the release by hash, which would point into the discs folder. The tool would need to perform disc ID lookups across the discs folder structure and, if found, skip import.
Describe alternatives you've considered
Additional context
An example in the current database is The Bourne Identiy. This title currently has 3 releases. Two of the releases feature a Blu-ray disc entry. Each of these discs has the same hash (F385A67FC59A9353CF6CD7D24FF1F62B) yet their disc contents are different: Bourne Ultimate Collection vs Bourne Complete Collection.
This concept is really no different than in the world of CDs and a good example of this model for CDs is MusicBrainz.org. There, a CD has a unique identifier based on its content. When a new release is made which features a disc whose ID is already in the database, that disc is linked to the release. Any edits made to that disc also show up in every other release containing that disc. This is as expected as the discs are identical between releases and fixes to the disc should be propagated to all releases incorporating the disc.
One MusicBrainz example I quickly found was U2's War, which features many published disc IDs. One of which is hHaKxuJOdf82y_IQ2nwlxPrZbTE-, which belongs to 9 different releases of various labels, countries and barcodes.
Is your feature request related to a problem? Please describe.
The database contains instances of multiple entries for discs with the same hash. Conceptually, one disc belongs to many releases and one release contains many discs. Currently, for some titles (at least pre-web contribution interface) one disc hash may belong to multiple discs, each in their own release. This creates circumstances where a disc that is common to multiple releases is present in the database multiple times which are not synchronized.
Describe the solution you'd like
This may be a bit complicated because it would involve some back-end work, but ultimately a given disc hash should exist uniquely in the database and each release that includes that hash should reference that unique entry instead of having it's own copy of the disc entry. I'm not sure what this looks like in the web's backend database, but in the git repo this is manifested as multiple
disc0X.jsonfiles having the same disc hash but different contents.I understand a disc hash check is performed as part of the web-based contribution process now. I'm not sure as I haven't yet had an opportunity to try to upload a duplicate disc. I suspect this may be moot for contributions going forward, but at the very least there is some cleanup to do for past contributions, in which case I guess this is really a bug issue instead of a feature request.
This might also fully break compatibility with the desktop tool as it might take significant work to adapt the tool. This would probably require that all of the
discXX.jsonfiles be relocated from themoviefolders to a separatediscsfolder, stored by hash. Therelease.jsonfile would then be modified to identify the discs comprising the release by hash, which would point into the discs folder. The tool would need to perform disc ID lookups across thediscsfolder structure and, if found, skip import.Describe alternatives you've considered
Additional context
An example in the current database is The Bourne Identiy. This title currently has 3 releases. Two of the releases feature a Blu-ray disc entry. Each of these discs has the same hash (F385A67FC59A9353CF6CD7D24FF1F62B) yet their disc contents are different: Bourne Ultimate Collection vs Bourne Complete Collection.
This concept is really no different than in the world of CDs and a good example of this model for CDs is MusicBrainz.org. There, a CD has a unique identifier based on its content. When a new release is made which features a disc whose ID is already in the database, that disc is linked to the release. Any edits made to that disc also show up in every other release containing that disc. This is as expected as the discs are identical between releases and fixes to the disc should be propagated to all releases incorporating the disc.
One MusicBrainz example I quickly found was U2's War, which features many published disc IDs. One of which is hHaKxuJOdf82y_IQ2nwlxPrZbTE-, which belongs to 9 different releases of various labels, countries and barcodes.