added ability to collect DOIs of citing papers using OpenCitations #27
Conversation
|
Hey @jasongfleischer - thanks a lot for the contribution, this extension of functionality looks great! Sorry it's taken me so long to get back to you on this - been a bit distracted, but should be more on top of things now! I want to think a little about what the best organization might be here, and also check into the missing data / 0 thing, because off the top of my head I don't remember about that. When I look through this, is it cool if I just go for it and edit / push to your fork directly, or if you prefer, I can do more like a review, and write up any comments of what I'm thinking for you to potentially edit / update? |
|
Alright, I finally found some time to do some LISC updates! Thanks for this addition, @jasongfleischer - I edited together the overlapping functions to integrate you addition of collecting the DOIs. I revisited the 0 -> None thing. The reason is that a DOI that is not found in the database returns an empty page, which gets collected as '0'. This isn't a true 0 (as in "the paper was checked, and has 0 citations"), it simply means data on the paper isn't available - hence encoding as None rather than zero. |
In the end I decided to duplicate the collect_citations() infrastructure with collect_citation_dois(). If you don't like that, it could obviously be melded into a single collect_citations() using an argument about which kind of data to collect.
Also I have a question about the code in the original: I can't figure out that stuff about
"# If the return is empty, encode as None instead of zero
This is because we don't want to treat missing data as 0 citations"
I can't see the difference between missing data and 0 citations... what would the json return look like in each case? I just duped that stuff into my code, but I don't understand it TBH