Fixes: https://pagure.io/koji/issue/4479
@mikem what do you think about this easy fix? Or we can do it more complex, that we can add parent-name to the data dict and work with this directly in the _writeInheritanceData, that we can set parent_id by getTagID(parent-name). What do you think?
parent-name
_writeInheritanceData
While this call can technically accept a plain dict, a list of dicts is the expected input. E.g. see edit-tag-inheritance.
I do think it would be better to handle this in the underlying call where we unpack this data.
I would not add a new field to the dict, I would just convert the value if it is not an int.
rebased onto 3999728651e1a249ce1734fbd69bb4d14c78c109
ah, yes, I missed that this is list of dicts :-). Thank you for catch. Fixed now.
The underlying call technically accepts either a list of dicts or a single dict (which it will simply wrap in a list). The current version of this patch assumes it is a list, which could break any code that relies on the alternative behavior. While I don't think any code in Koji itself does this, another client might.
The underlying call does some validation on the change list, and actually does a lookup for the parent tag to validate it. I think that is the right place to address this.
I do think it would be better to fix this in the underlying call, but I guess this works. We can always move the check later.
:thumbsup:
Metadata Update from @mikem: - Pull-request tagged with: testing-basic
Metadata Update from @mfilip: - Pull-request tagged with: testing-done
Commit 5ff8bab2 fixes this pull-request
Pull-Request has been merged by mikem
Fixes: https://pagure.io/koji/issue/4479