Index articles

[sandbergja]

• handle wikidata timeouts
• be sure to always use the same canonical name for the journal (with same capitalization, etc. so that it can link to its facet) -- definitely need a journal model!
• only a few articles are coming through from Gazette-Times, especially with the optional subject clause
• write a test: make sure that there is no doc in solr for Q96820379, since it is available from crossref (that test will fail atm; also need to fix the deduplication!)
• links should go into three different fields: one for direct PDF download, one for other direct access, one for link resolver access; and 3 different displays accordingly
• fix all TODOs
• edit query to include enrichments
• limit by dates that we actually have full-text
• crossref map the formats (so they aren't all journal-articles) https://api.crossref.org/types
• Add an awesome display for journals that include some kind of "search within" thing
• Fix http://localhost:3000/catalog to show all results, rather than the goofy default page
• warning: already initialized constant USER_AGENT
• crossref indexing needs author
• wikidata article and journal indexing need more stuff, including authority data

[sandbergja]

One issue with the current index_articles branch: DOIs (currently used as id) include slashes and so forth -- those should be escaped when indexing

[sandbergja]

For linking to articles through link resolver:

Base url: https://linn.on.worldcat.org/atoztitles/link
Helpful ruby gem (at least for inspiration): https://github.com/openurl/openurl/