Closed guynir42 closed 5 months ago
In addition, and I don't know why this happened, I started to make some minor adjustments to how we merge things in the data store. That caused a whole bunch of bugs that I have been running after for three days, touching half of our code.
@rknop There are still a couple of test failures, related to object cleanup (not the actual logic of the tests). If you have time to review this, you can start, and I will fix the remaining minor issues tomorrow morning.
@rknop I've addressed all comments. Please have another look and see if it is ok to merge (use the "see changes since last review" to save time). Thanks!
Rounding up a bunch of semi-related issues into a single PR, to cleanup a bunch of open issues we collected when working on a set of recent PRs.
upstream_images
fornew_image
andref_image
with a single ID of the ref image. Issue #178Image.header
withImage.info
andImage.raw_header
withImage.header
. Issue #187section_id
. Also part of Issue #179image_id
andprovenance_id
. Issue #160provenance_id
as part of their filenames (to make them unique). Issue #146downstream_images
as a lazy-loaded relationship on Image. Issue #151recursive_merge()
in favor of regular merge and the newmerge_all()
that runs on Image, SourceList and Reference, because it needs to take special care of the products (which are not relationships!).