Unable to save many PDFs from Chrome

This discussion was created from comments split from: Watch folder.
  • I just want to clarify why so many people are requesting a "watch folder" feature:

    1. Clicking the "Save to Zotero" button on the Chrome toolbar frequently fails to get the pdf. This problem seems to be increasing rapidly, perhaps because more publishers are using their own pdf viewing solutions instead of the browser solution? Recently, I've seen Zotero getting the "author's submitted version" instead of the publisher version, even though I'm viewing the publisher version in the browser!

    2. Due to #1, I now just download the pdf and drag it into Zotero, and let Zotero fetch the metadata. But this involves lots of additional clicks for each article.

    3. I like to use Safari (because tracking), and the Zotero Safari connector is not out of beta, so downloading and dragging to Zotero is the only option here.

    I understand a watch folder feature is a low priority, but there is a real use case for it. if the Zotero team ever finds time to implement it, I think it would be popular.
  • Clicking the "Save to Zotero" button on the Chrome toolbar frequently fails to get the pdf. This problem seems to be increasing rapidly, perhaps because more publishers are using their own pdf viewing solutions instead of the browser solution?
    That's something you'd want to report. It's not something we're seeing or aware of. Afaik, PDFs are imported from all major publishers (and many smaller ones), definitely including Elsevier, Springer/Nature, Wiley, Oxford, Sage, Taylor&Francis, OUP and CUP as well as most others using the most widely used academic CMS systems (Atypon, Silverchair, and OJS)
  • Thanks for the response. Where is the place to report? Here?

    https://github.com/zotero/zotero/issues

    It's definitely a frequent but inconsistent problem. What info would be helpful to report?
  • Please report here, in a new thread for each case, and we need the URL exactly as you see it and a debug ID from Zotero (not the connector)
  • Due to #1, I now just download the pdf and drag it into Zotero, and let Zotero fetch the metadata. But this involves lots of additional clicks for each article.
    @hagenlab: Note that you don't need to follow these steps either way. Unless you've changed your browser settings not to show the PDF in the browser, you can just click the save button while viewing the PDF in your browser to save it directly to Zotero, which will kick off metadata retrieval. You don't need to save to disk and drag it in.
  • edited June 15, 2021
    In practice, clicking the save button when the pdf is displayed in the browser often fails to save the pdf to Zotero: I get the meta-data, but not the pdf. The only way I've found to reliably get the pdf is to download it and then drag it into Zotero.

    For example, I just now picked one article to download at random:

    https://ntserver1.wsulibs.wsu.edu:2118/doi/full/10.1111/brv.12374

    Instead of Zotero saving the publisher version, it saved the "open source" author submitted version. Yet when I click the pdf link, the publisher version shows up in my browser. When I then click the Zotero save button, it *still* saves the open source version. Here is a screenshot:

    https://i.imgur.com/41bjdhH.png

    But, I was able to download the publisher version of the pdf and drag it into Zotero.

    Sometimes, I don't get any pdf at all when I click the Zotero save button, even though it's displayed in the browser (but again, I can download the pdf just fine).

    EDIT: here is a screenshot of Zotero failing to grab the pdf, even though I can download it (second image):

    https://imgur.com/a/hwnDovS

    The article url: https://www.sciencedirect.com/science/article/pii/S0065308X15000299?casa_token=J3-kH52-KagAAAAA:QfPtIdiJAfa5NB_Dyt0PVnSXSpLcN15_FkoTniFEtVCcQzbnFQQ7U1Rv_IsWPOwqLKkbEQ8S1uaj
  • @hagenlab: Your university is using the old port-based EZproxy configuration, which generally works less well with Zotero than the newer hyphenated-domain-based method, so that's likely the problem for you. Note that on Wiley you're only getting the Embedded Metadata translator rather than the Wiley translator. If you start separate threads with the info adamsmith requests above, we may be able to help you configure your proxy settings manually.
    Yet when I click the pdf link, the publisher version shows up in my browser. When I then click the Zotero save button, it *still* saves the open source version.
    That's actually not the PDF — it's some obnoxious cloud-based reader that Wiley is using. But Zotero still handles it fine when it can detect the site, and that's just not working for you because of your proxy setup. (E.g., Zotero will save article data, a snapshot, and the PDF while on this Wiley PDF URL, even though it's the cloud-based PDF reader.)

    In an upcoming version, we also plan to make it possible to open a PDF with Zotero in order to add the file, which would help in a case where the article page fails to download the PDF and the site prevents you from viewing it directly in the browser.

    For the ScienceDirect page, please start a new thread, and we'll take a look.
Sign In or Register to comment.