Help finding a plugin for importing references from in-text citations

txw1001
Hi everyone,
I used to have a plugin that allowed me to import citations directly from within the text. For example, when I clicked on a citation like “Becker, 1975” in a sentence, it would display the full reference and give me the option to import it into Zotero. Does anyone know which plugin this might be?

Many thanks in advance!
  • tim820
    Might have been the Reference plugin ?
    https://github.com/MuiseDestiny/zotero-reference

    It possibly had some interaction with Zotero v7's reference popup hover window, like what you describe. There is a setting in the plugin's settings that refers to that, but it may not be working in the latest version ...
    https://github.com/MuiseDestiny/zotero-reference/issues/368
    (Chrome does a decent job of translating the Chinese text at github.)

    The Reference plugin adds a right pane tab that lists a PDF's references, showing you which you already have (relies on DOI search). So you should see that in the right panel (whether or not you hover over the in-text citation and see Zotero's popup); clicking on a reference in that list will show you the cited paper's information. It can also take you to that paper in your library if you have it, or to the URL of the paper if you don't. It can also list papers that have cited the current PDF paper. It relies on online databases (Symantic Scholar, Open Alex, Research Rabbit), so a paper's cited references and cited-by references have to be listed there (sometimes one database may work better than others).

    (The documentation has not been updated for Zotero v7, but you learn how it works mostly by using it.)
  • rdiaz02
    edited 11 days ago
    @tim820 : I've seen several threads where you comment on the use of the Zotero Reference plugin, so I am resurrecting this thread with some questions:

    I must be doing something wrong, or the functionality has changed (using Zotero 8.0.3).
    - Often, it seems the plugin stops right after the parsing of references, without actually showing them; for example as in here (using OpenAlex)

    https://s3.amazonaws.com/zotero.org/images/forums/u301830/zbtja2igv0jymcivl4qt.png

    or here (using Semantic Scholar, same arXiv paper)

    https://s3.amazonaws.com/zotero.org/images/forums/u301830/gi7lxacpm0qlbuwoeeux.png

    Sometimes, changing between OpenAlex and Semantic Scholar as "Source" fixes it, but often it does not.

    (Edit: I've added my Semantic Scholar API key in the Config Editor, in extensions.zotero.zoteroreference.semanticsScholar.apiKey, but it does not seem to improve things).


    - When it works and I go to "Import to Zotero" it does get the reference in Zotero, but so far it does not always grab the PDF too and this includes from sources that have PDFs readily available, such as bioRxiv (I solve it by going to the reference, clicking on the URL, downloading the PDF, and adding it to the entry).
    Edit: it seems things improve if I go to the "Config Editor" and I change the setting of "extensions.zotero.zoteroreference.savePDFReferences" to true. (So far, it seems to be getting the PDFs of arXiv preprints).

    - To change the "Source" between OpenAlex and Semantic Scholar I need to open the "Settings"; is there a simpler/faster way? Is it possible to add other providers?
  • tim820
    edited 10 days ago
    The Reference plugin still seems to be a "work in progress". But it's so good when it works that I put up with it when it's not perfect.

    It changed a lot between Zotero v6 and v7, which may explain some of the issues. And the documentation still shows the v6 screens - so you have to figure it out by using it. I'm not using v8 as I am still waiting on one or two other plugins to be compatible, so I haven't tried the Reference plugin in v8.

    Currently it seems to me most often reliant on extracting and parsing a paper's reference list. And some common formats it still messes up. Like missing the last half of the reference list. Under Zotero v6 you could choose whether it used parsing or online API to get the reference list. Now it just seems to do both. I am not sure I've seen anything that I can tell is better or worse with OpenAlex or Semantic Scholar chosen in Settings (it did have others under v6 IIRC). Although I haven't looked at "cited by" information for a while - it obviously can't get that from parsing the paper. I too had a Semantic Scholar API key but then its entry field disappeared from the Settings. And the developer said it wasn't really necessary (but could be added via the config preference as you have). I'm actually not sure if I still have mine in there.

    I usually rely on it taking me to the website to get a cited paper/PDF that I want (so that's then down to Zotero). I had noticed too that expecting more than that from the plugin itself wasn't reliable. Also asking it to take me to a paper I have in my library doesn't always work. And that's for cited papers it *says* are in my library, but that test is not always accurate (possibly due to older items with missing DOIs in my library). Also when it pops up the abstract screen it sometimes gets the cited paper wrong.

    I'm hoping everything will fall into place one day ! I have logged issues at github, as I see you have done. But it still works well enough often enough for me to use it often.
  • rdiaz02
    edited 8 days ago
    Thank you very much for the details. Those changes between versions I was not aware of explain a lot of my feeling disconcerted (or the not understanding what I was doing wrong that did not allow me to replicate the help :-) ).

    I haven't been able to figure out whether setting OpenAlex or Semantic Scholar is "the overall better choice" and I wonder why not use both or even a bunch of possible ones, in order, and stop on first hit.

    As for your case, I am interested in the references in the paper (not much in the "cited by"), so solid parsing of the paper to, as you say, "taking me to the website to get a cited paper/PDF that I want (so that's then down to Zotero)." is what really matters to me. That is why the glitches with the arXiv papers are a bit of a pain; it seems most of the examples in the screenshots and videos use "the usual papers in biomedical or general scientific journals" so maybe my situation is not that common.

    Taking me to papers already in my library, though, works very well (although it might say that what is in my library is the preprint, when I have both preprint and paper, it will take me to the published one). But then, this is not my primary use.

    Yes, I've logged an issue, hopefully it will be solved. Though I've just seen https://forums.zotero.org/discussion/129786/agpl-security-concerns-zotero-reference-gpt-style-plugins-ship-obfuscated-code-not-matching-repos, so I might put on hold for now further use of the plugin until these concerns get addressed or I have a clear idea that using the plugin is both safe and compliant.


    Edit: Note there is a related discussion here: https://forums.zotero.org/discussion/124564/how-can-i-extract-references-from-papers .
Sign In or Register to comment.