CV7000: fix channel-wise ordering of planes

sbesson commented 3 years ago

Ports ome/bioformats#3663 to the IDR fork of Bio-Formats

sbesson commented 3 years ago

1c62bfd has been manually deployed to pilot-idr0093 where the plates of idr0093 are being reimported for validation of the fix

sbesson commented 3 years ago

Following the deployment on pilot-idr0093

for most published studies, images are loading without issue. The memo files being loaded from disk which confirms the fix does not invalidate cache file

the only notable exception is idr0088, a screen composed of CV7000 plates where the memo files were invalidated and regenerated

2021-02-23 15:28:05,554 WARN  [                   loci.formats.Memoizer] (l.Server-4) deleting invalid memo file: /data/OMERO/BioFormatsCache/data/OMERO/ManagedRepository/demo_2/Blitz-0-Ice.ThreadPool.Server-1/2020-10/23/04-07-06.213/1013608217/.1013608217.wpi.bfmemo
2021-02-23 15:28:26,415 DEBUG [                   loci.formats.Memoizer] (l.Server-4) saved to temp file: /data/OMERO/BioFormatsCache/data/OMERO/ManagedRepository/demo_2/Blitz-0-Ice.ThreadPool.Server-1/2020-10/23/04-07-06.213/1013608217/.1013608217.wpi.bfmemo13765629891496706902
2021-02-23 15:28:26,415 DEBUG [                   loci.formats.Memoizer] (l.Server-4) saved memo file: /data/OMERO/BioFormatsCache/data/OMERO/ManagedRepository/demo_2/Blitz-0-Ice.ThreadPool.Server-1/2020-10/23/04-07-06.213/1013608217/.1013608217.wpi.bfmemo (2058231 bytes)

currently loading all the plates from pilot-idr0093, the 14 plates mentioned by the submitter will be re-rendered for validation before loading into idr-next.

sbesson commented 3 years ago

Validated the re-imported idr0093 screen in a pilot with this PR included with the submitter and @francesw. Moving forward with the release of IDR/Bio-Formats and prod95

IDR / bioformats

CV7000: fix channel-wise ordering of planes #25