andrew-s-rosen / coapy Goto Github PK

Right now, a strict de-duplication scheme was implemented for the list of co-authors. However, "John Smith" and "John K Smith" and "JK Smith" will all still appear as distinct entities. Ideally, we should clean that up, but it's not a huge issue practically speaking.

NSF requests institutional affiliations for all listed collaborators. I'm not really sure the best way to fetch this at the moment. Google Scholar (and the bibtex in general) does not contain institutional affiliation. The most natural way would be to scrap using Google Scholar altogether and instead use ORCiD which contains all the necessary info, but their Python API seems unmaintained and likely broken. Web-scraping with Selenium will be too error-prone and annoying.

Right now, there is just one small functional test. We should add some real unit tests.