Characterization - importing a characterization always generates a new characterization

[kmkostka]

Expected behavior

A user receives a Characterization JSON from another OHDSI collaborator. A user navigates to the Characterization tab in ATLAS and clicks on 'New Characterization'. A user names the Characterization with a name of their choosing (e.g. [Network Study] Characterizing endometrial cancer for Bonnie). A user saves the cohort. A user goes to the utilities tab to import the JSON file. A user pastes the Characterization JSON and hits 'Import'. A user's New Characterization is updated with the Characterization JSON shared by the other collaborator and the name of the self-specified Characterization is retained.

Actual behavior

A user receives a Characterization JSON from another OHDSI collaborator. A user navigates to the Characterization tab in ATLAS and clicks on 'New Characterization'. A user names the Characterization with a name of their choosing (e.g. [Network Study] Characterizing endometrial cancer for Bonnie). A user saves the cohort. A user goes to the utilities tab to import the JSON file. A user pastes the Characterization JSON and hits 'Import'. A user is directed to a brand new Characterization (with a different analysis #) and the same exact name as specified in the initial JSON file shared by the collaborator. A user now has 2 Characterizations created - 1) an empty Characterization that retains the name they saved but has no Characterization Design and 2) a fully specified Characterization that is named exactly what the other collaborator named their Characterization. A user can delete the first empty cohort. A user can then open and rename the shared Characterization to a name of their liking but a user is confused why he/she had to do the last step.

Steps to reproduce behavior

1. User receives a Characterization JSON from another OHDSI collaborator.
2. User opens Characterization tab in ATLAS and clicks on 'New Characterization'.
3. User types in a name of their choosing (e.g. [Network Study] Characterizing endometrial cancer for Bonnie) for Characterization.
4. User saves the cohort.
5. User goes to the utilities tab to import the JSON file
6. User pastes the Characterization JSON and hits 'Import'.
7. User is now directed to a brand new Characterization (with a different analysis #) and the same exact name as specified in the initial JSON file shared by the collaborator.
8. User now has 2 Characterizations created - 1) an empty Characterization that retains the name they saved but has no Characterization Design and 2) a fully specified Characterization that is named exactly what the other collaborator named their Characterization.
9. User can manually delete the first empty cohort.
10. User can then open and rename the shared Characterization to a name of their liking.

[chrisknoll]

I believe the issue is that when you import a characterization, it doesn't overwrite the existing one, but rather creates a new characterization using the imported design. Is that correct?

If so, I have also been confused by this behavior (that you go into a cohort characterization to do an import that will create a whole new characterization. I think if we want it to work like that, we should have a 'Import Characterization' button next to 'New Characterization' on the main CC-list page, that will create a new characterization from import. We should also change the import function inside the CC to overwrite the existing CC when you import and not create a new one.