If a user tries to upload a paper with a url or DOI, @dhimmel is building a system to grab the metadata from that paper: https://github.com/quantfive/researchhub-backend/pull/132.
Currently his endpoint doesn't work on actual PDF's, but an example url that would work is: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1007128
He's returning a list of papers that are similar to the uploaded paper as well as a dictionary of metadata.
If there are similar papers, we want to show them to the user and ask them if they mean this paper, otherwise let the user move forward in the flow.
We want to take the metadata given back from this endpoint and fill the metadata on our next page.
- [x] Use CSL Item Endpoint, fetching metadata works
- [x] If CSL Item cannot get the information and the URL links directly to a PDF, use our old pdf endpoint upload flow
- [x] If CSL Item cannot get the information and the URL does not link to a PDF, make the user link another URL or upload a PDF.
- [x] Allow a user to enter JUST a doi -- if the CSL endpoint returns metadata that's useable then use that, otherwise make the user again link another URL / DOI or upload a PDF.
enhancement