Zotero collects Snapshot only, not full PDF

YaelOfer · October 2, 2024

I have uninstalled and reinstalled Zotero. It's the updated version. I'm on a Mac (Sequoia 15.0) using Chrome (up to date).

I'm logged in to my university library so I can access full pdf articles. But can't seem to get Zotero to save the full pdf. I can open the pdf and save that, but then it doesn't have all the same information automatically pulled, only some of it.

https://s3.amazonaws.com/zotero.org/images/forums/u15076694/9yf1j2g31xcmu1p2gh8j.png

Any ideas why this won't work? Thank you

tim820 · October 2, 2024

Likely this issue ...
https://forums.zotero.org/discussion/109940/zotero-is-not-saving-pdfs-from-sciencedirect#latest

mjthoraval · October 2, 2024

The problem was fixed in Zotero 7 by showing captcha validation windows.
But the validation windows do not appear anymore.
I am seeing this also from yesterday. I have also tried from home and see the same problem.

URL: https://www.sciencedirect.com/science/article/pii/S0021850223001787
Zotero Connector - Debug ID: D1892424967
Zotero Desktop - Debug ID: D755254984

Also cannot get the PDF from the PDF page directly:
Zotero Connector - Debug ID: D1452394090
Zotero Desktop - Debug ID: D1693323906

Zotero 7.0.7-beta.3+20d13f5a8 (64-bit)
Windows 10

dstillman · October 3, 2024

The CAPTCHA window should now show again. Thanks.

mjthoraval · October 15, 2024

It is still the same problem for me:
Zotero Connector - Debug ID: D1807147090
Zotero Desktop - Debug ID: D2050525247

Curiously, when I try to use a proxy, it downloads a PDF file, of size 47 Ko, which is not a valid file.
Zotero Connector - Debug ID: D249656681 / D1995790426
Zotero Desktop - Debug ID: D1413065696

https://s3.amazonaws.com/zotero.org/images/forums/u265723/92iosg22unjvqy72ljqp.png

https://s3.amazonaws.com/zotero.org/images/forums/u265723/82nau0kf0dgjecqj5art.png

adomasven · October 15, 2024

@mjthoraval Are you able to load the PDF in your browser with the unproxied URL (i.e. are you authenticated with ScienceDirect via some means)? Does a valid PDF load for you in your browser when accessing it via the proxy?

mjthoraval · October 15, 2024

I forgot to mention that I was testing above on the same link:
https://www.sciencedirect.com/science/article/pii/S0021850223001787

Yes, I can load the PDF file in the browser from the unproxied URL.
It works either when I am logged in:
https://s3.amazonaws.com/zotero.org/images/forums/u265723/5jtdttd1piqnxawwkgsa.png

Or logged out and removed all cookies:
https://s3.amazonaws.com/zotero.org/images/forums/u265723/1mqdvyn5pp5hpr3lu6hh.png

I am just using the normal university access to the full text PDF.
From the unproxied PDF file page in the browser, I cannot import the PDF file to Zotero:
https://s3.amazonaws.com/zotero.org/images/forums/u265723/xug5z8d40tjq83mm2q8z.png

When using the proxy, the PDF file also loads correctly in the browser.
From the proxied PDF file page in the browser, I can import the PDF file to Zotero:
https://s3.amazonaws.com/zotero.org/images/forums/u265723/kijp0nthbbiekjgfhcb9.png
The correct PDF file is saved in this case, contrary to the invalid file downloaded from the article page.

adomasven · October 15, 2024

Hmm. There may be some weirdness due to access being provided by being on-site. Either way we are working to change the saving workflow to one that will hopefully work better and without the need to clear captchas.

DONGMarcel · 2024-10-23T18:23:31+00:00

Hello, the exact same problem here.
For PDFs that cannot be downloaded, there is at some point (after a few tries) a page with a CAPTCHA popping up. If you click it then you can download again.
But this CAPTCHA page does not appear each time a PDF cannot be downloaded.