Book and chapter metadata

Book- and chapter-level metadata is extracted from an external file rather than from the epub (see documents and templates/book-chapter-metadata-TEMPLATE.xml). In fact, metadata recorded in the epub is often too scarce, and of too variable quality, to be used to populate the TEI <header>.

This external file needs to follow the CrossRef schema, version 4.3.5 or newer (see https://support.crossref.org/hc/en-us/articles/214530063). The same identical file can therefore be used to supply metadata to the converter as well as to register content to the CrossRef database (see https://support.crossref.org/hc/en-us/articles/215577783-Creating-content-registration-XML). Each of these files must gather basic information at the book level (listed within the <book_metadata> element) and, where possible, should also contain chapter-level information (within <content_item component_type="chapter">).

The script governing conversion to XML plucks each DOI included in the book content and tries to match it with one of the DOIs recorded in the metadata file. If the matching is successful, information such as title, contributors and license are extracted from the metadata file and used to populate the XML header. This method proves quite useful to spot any errors or mismatch that may have occurred between each DOI as printed in the book and as deposited to CrossRef.

The following book-level data must be provided:

• <contributors> (<person_name> must contain both <given_name> and <surname>)
• <title>
• <publication_date> (must contain both <month> and <year>)
• <isbn> (at least one)
• <publisher_name>
• <ai:license_ref>
• <doi>
• <resource>