Skip to content

Latest commit

 

History

History
42 lines (23 loc) · 2.69 KB

README.md

File metadata and controls

42 lines (23 loc) · 2.69 KB

citation-parser

XML/XSLT for processing a list of plain text references into a) their constituent DOIs, b) a valid resource list for Crossref deposit

Crossref requires that members use resource DOIs wherever possible. When assigning a DOI to an article, the list of references must also have DOIs. These tools help identify those DOIs and create the properly formed <citation_list> XML for inclusion in a Metadata deposit https://support.crossref.org/hc/en-us/articles/214169586 or Resource deposit https://support.crossref.org/hc/en-us/articles/216009323-Resource-deposit-schema-4-3-6.

Required:

XML editor/parser (Oxygen)
Text Editor (Sublime Text, Textwrangler)

citation.parser.plus.0.2.php citations.processed.xsl (+ crossref.results.xml)

ARTICLE-LEVEL WORK

In Oxygen, open citations.processed.xsl. Create crossref.results.xml

Import all references, in order, to a text editor. QC: Review this list to ensure that each line contains only one reference. One reference to a line, very important.

Paste this list in the input box in citation.parser.plus.0.2.php (you'll need to mount this somewhere it can make http requests and return data). Include the doi of the article you're processing. Submit. Wait (can take a good long time, failed at 100 citations, worked for up to 40. Break up the list accordingly.)

When results post, open up the page source and copy all the XML output (browser may not display the XML tags-- you need them). Paste entirely over the contents of crossref.results.xml. Save crossref.results.xml.

For each <citation> in crossref.results.xml.

  • Compare <raw_string> to the two <reference> elements.

  • If one of these is a match, change the flag attribute in the associated <result> element to TRUE. Otherwise, leave both alone.

  • You might need to test a doi at https://doi.org/[doi] to see if it goes where you think

  • You can edit fields directly here; add an <edition_number>, <series_title>, or <issue>, change <raw_string> to <unstructured_citation> if necessary

  • Wiley, Springer, and Elsevier imprints may sometimes have a DOI that can be found through their respective sites

  • Sometimes both <reference>s will match <raw_string>, and the task is to determine the better doi

After you've gone entirely through each <citation>, save crossref.results.xml.

Transform crossref.results.xml in an XML parser using citations.processed.xsl

Copy all <citation> elements (minus the header). Paste into your working file (wherever you're building your XML deposit document).