feat(item-tree): Parse UNIXML "relations" into direct relationships in the item tree. Previously these were special "relations" items in the tree. Convert these all into full URLs so they fit in the item graph
feat(item-tree): Parse UNIXML "relations" into direct relationships in the item tree. Previously these were special "relations" items in the tree. Convert these all into full URLs so they fit in the item graph
Don't change the parsing of inverse relationships that are calculated by Content System. These still parse into the "relations" in the item tree (they are inverted and have the "claimed-by: object" field.
Change the conversion to citeproc JSON to look in both of these places, so the output is unchanged.
Add a load of identifier types for each type found in the schema:
- pmid
- pmcid
- purl
- arxiv
- ark
- handle
- uuid
- ecli
- accession
- other
- uri
And heuristics to tidy up occasional malformed entries.
Add test file loads-of-relations.xml for the parser and convert regression test suites.
CR-339
Edited by Joe Wass