remerjohnson / resolvr Goto Github PK

Currently the file input takes a csv file, turns it into a DataFrame via pandas, and renders it in table form as an HTML page.

Much like the single query, what needs to happen is:

• a column of values will be accessed from the DataFrame
• those values will query Wikidata just like the single query
• the resulting Wikidata URIs will be captured, by adding a column result in the DataFrame rendered and/or offered as a download

Currently, the proof-of-concept for a single ID query uses only VIAF. The goal is to be able to use other IDs such as LC, Getty, FAST, etc.

Approaches

One approach would preserve the current "single page" layout for single IDs. Thus, a drop-down box would be needed to select which type of ID they have, and that would drive the type of SPARQL query run. This seems cleaner but more difficult to code up.

A different approach would be to have a page for each ID type. This would probably be best accomplished by changing the current layout so that the pages are vertical, similar to MakeTheDocs. Although potentially messier, this would be much easier to code up, as the separate pages could each have their own query similar to the VIAF query.

Flask is reporting security vulnerabilities, so would be a good time to also upgrade the project to python 3.7