[BI-2028] Duplicated accession names getting assigned duplicated GIDs after Exp&Obs upload#320
[BI-2028] Duplicated accession names getting assigned duplicated GIDs after Exp&Obs upload#320
Conversation
| .filter(obsUnit -> germplasm.getAccessionNumber() != null && | ||
| germplasm.getAccessionNumber().equals(obsUnit | ||
| .getBrAPIObject() | ||
| .getAdditionalInfo().getAsJsonObject() |
There was a problem hiding this comment.
Do we need a migration to add gids to existing observation units additionalinfo?
There was a problem hiding this comment.
Most existing OUs have the correct gemplasm dbId assigned, so a migration adding the GID wouldn't matter either way for them. It's a question more for Alex and Shawn to estimate how many cases of existing OUs using germplasm there are. They haven't com across this until now, so I'm guessing it's a small number where manually fixing it in the species database would make more sense. I'm not sure what a migration would look like. It wouldn't be a case of just adding the GID to the OU additional info since the existing id in the OU germplasmDbId field could be wrong. You couldn't easily tell if it was right or wrong, and if it's wrong, it's not clear which germplasmDbId would be correct? So a migration just looks like re-importing the experiment with bug fix in place.
ffe0f60 to
0a9181c
Compare
Description
Story: BI-2028
During experiment import, existing germplasm dbid was being assigned to OU PIO by looking up the germplasm name stored in the OU PIO. When there are multiple germplasm with the same name, this was causing the same germplasm to be assigned to multiple OUs.
Dependencies
none
Testing
See the card for sample import files.
Checklist: