Please provide us with a solution on how to query zbib.org

kirk86

Dear devs and zotero community,
I was wondering if it would be possible to provide us with an api to query zbib.org service programmaticcally. Thanks!

adamsmith

My understanding is that the absence of an API is on purpose, since many of the services queried have explicit or implicit query limits. You can set up your own instance of the Zotero translation server or query the Wikipedia Citoid API (which uses a mostly Zotero backend). (no time for links but should be easy to find)

kirk86

@adamsmith thanks for the reply. From a quick look it seems that you can only query with doi, url, isbn, etc. But I'm more interested in being able to do full text search query. If Isetup the zotero translation server will I be able to do that?

adamsmithpresentatio

What do you mean by a full text query? Requesting full text (i.e. the equivalent of "find available PDF") or sending full text (i.e. the equivalent of retrieve PDF metadata)?

Neither is possible with any Zotero server infrastructure by itself - both run significant parts of the service locally. The code for the PDF recognizer is on github. The find available PDF I think may run entirely locally (with API calls to UnPaywall as needed)

(same person as above; just on a different account right now)

kirk86

Thanks, I was referring to sending full text (i.e. the equivalent of retrieve PDF metadata, here I suppose that by pdf metadata you mean the bib info retrieved from different services such as crossref, scholar, etc.).

I just wanted a way to accurately get bib info from the pdf metadata that I extract, by sending that metadata as query (for instance title, or text from the pdf document) to zbib.org.

adamsmith

zbib doesn't do anything with the text from the PDF document, nor does any other Zotero server (the recognizer extracts specific JSON elements & then sends them to the server for analysis & recognition)