Resolve external survey account deletion bug. #498
Conversation
|
@wasade When you have a moment, can you please review this PR? |
| source_id, | ||
| survey_id | ||
| ) | ||
| if not source_repo.scrub(account_id, source_id): |
There was a problem hiding this comment.
Should this check happen on line 124 above the call to has_external_surveys?
There was a problem hiding this comment.
It's less a check and more of an in-line verification that the scrub function actually works, modeled after the existing functionality to delete a source.
From that standpoint (addressing the other comment), I think it makes sense to not commit the scrub calls on the surveys, as we haven't successfully scrubbed the source.
We could avoid all of that by putting a check on line 124 that verifies the source meets the conditions to be deleted/scrubbed (i.e., the source exists and there's only a single record in the ag.source table with that ID). In that case, the scrub function could still fail if it's not a human source, but that would only happen if someone had somehow associated either a sample or external survey with a non-human source. And if that happened, I'd want it to show up in the exception logs to investigate because that would be rather concerning.
There was a problem hiding this comment.
Sounds good. I'm +1 on path of least resistance which I think is to retain the current code?
There was a problem hiding this comment.
Yep, code as written would be the simplest path. If you don't see any other issues here, I'm going to merge this, #497, and biocore/microsetta-interface#264 and get them up on staging for testing.
| survey_id | ||
| ) | ||
| if not source_repo.scrub(account_id, source_id): | ||
| return jsonify(code=404, message=SRC_NOT_FOUND_MSG), 404 |
There was a problem hiding this comment.
So I think we need to be careful with short circuit returns. If this return is hit, the calls to scrub above will not be committed
Co-authored-by: Daniel McDonald <danielmcdonald@ucsd.edu>
#492
This PR addresses a few related issues: