To remain consistent with the rest of the artifacts:id tells one artifact of the same type from another, therefore must be unique. It cannot be unique within a repository/project, because that leads to a huge pile of the same type and the same id, 1 (since many projects one day get their first pull request). It makes no sense to not use unique value for id property, that would require all consumers to add extra exception when decoding artifact type and id: for all artifacts, type+id pair identifies the particular artifact, just pull requests are the exception and one has to examine uid field...
Therefore renamed uid to id to provide the unique ID value, and id renamed to pr_id to carry the project-specific ID of the pull request.
To remain consistent with the rest of the artifacts:
id
tells one artifact of the sametype
from another, therefore must be unique. It cannot be unique within a repository/project, because that leads to a huge pile of the same type and the same id,1
(since many projects one day get their first pull request). It makes no sense to not use unique value forid
property, that would require all consumers to add extra exception when decoding artifact type and id: for all artifacts,type
+id
pair identifies the particular artifact, just pull requests are the exception and one has to examineuid
field...Therefore renamed
uid
toid
to provide the unique ID value, andid
renamed topr_id
to carry the project-specific ID of the pull request.