-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
None
-
None
As the query below demonstrates, there are mediums in the database that have tracks with non-contiguous positions:
SELECT B.id, B.medium, B.position, B.last_updated
FROM track A RIGHT JOIN track B
ON A.medium = B.medium AND A.position + 1 = B.position
WHERE A.position IS NULL AND B.position != 1;
id | medium | position | last_updated
----------+---------+----------+-------------------------------
1967031 | 8608 | 17 | 2011-05-16 16:08:20.288158+00
10772674 | 832328 | 21 | 2011-05-16 16:08:20.288158+00
4116654 | 867752 | 27 | 2011-05-16 16:08:20.288158+00
2140117 | 884330 | 5 | 2011-05-16 16:08:20.288158+00
16991660 | 917176 | 5 | 2014-08-16 13:28:33.974804+00
2584922 | 925385 | 5 | 2011-05-16 16:08:20.288158+00
3619375 | 928039 | 12 | 2014-08-27 13:00:38.676122+00
3500891 | 940074 | 13 | 2011-05-16 16:08:20.288158+00
2675814 | 952281 | 8 | 2011-05-16 16:08:20.288158+00
1842242 | 992507 | 12 | 2011-10-08 16:18:06.828082+00
15931786 | 1426501 | 2 | 2014-01-18 03:23:13.056228+00
15931785 | 1426501 | 18 | 2014-01-18 03:23:13.056228+00
16372652 | 1467957 | 10 | 2014-04-26 03:00:32.795528+00
16895195 | 1518218 | 93 | 2014-07-26 03:02:25.814544+00
(14 rows)
The last result (medium 1518218) belongs to this release; it's unclear how it ended up in this state, because the last edit changing the tracklist does have a track 92, and there is nothing obviously wrong in the edit data (in particular, it has a track id of null, so no id-sharing with another track).