Frontiers sites won't import PDFs automatically

mdt_

I looked for another topic with this but it seems to not be listed. For the "Frontiers in ..." journal family, there appears to be a bug in the site translator.

I have had many failures in using the "frontiers" site translator--the reference data imports automatically with the "Save to Zotero" button, but the PDF downloads fail.

Here are two examples that fail for me:

http://journal.frontiersin.org/article/10.3389/fgene.2012.00190/full
http://journal.frontiersin.org/article/10.3389/fpsyg.2016.00073/full

In both cases the PDF fails to download when the site translator runs.

The Zotero website recommends reporting errors like this, I can provide more info as needed. Here is the system data:

version => 4.0.29.10, platform => MacIntel, oscpu => Intel Mac OS X 10.11, locale => en-US, appName => Firefox, appVersion => 40.0.3, extensions => Zotero Word for Mac Integration (3.5.14, extension), OneTab (1.17.0, extension), OpenH264 Video Codec provided by Cisco Systems, Inc. (1.5.3, plugin), Default Browser Helper (601, plugin, disabled)

Cheers,
Matt

vpolito

I have the same issue with frontiers journals. Is it working ok for some people or malfunctioning for everyone?

mdt_

@vpolito It is still happening to me, but not for every article. :-(

zuphilip

I looked at the example and the problem is that the url of the pdf http://journal.frontiersin.org/article/10.3389/fgene.2012.00190/pdf does not work when called directly, but it works when one calls it from the Frontiers website. Maybe they are requiring some cookies settings... However, you can try to reach out for the publisher and ask them if they can fix their pdf urls, such that they also work outside the frontier context. Because these are Open Access journals and therefore everybody can access the articles, I don't see any reason for the current behavior.

mdt_

Interestingly, the URL gets translated to:

http://journal-cdn.frontiersin.org/article/31294/files/pubmed-zip/versions/1/pdf

when I type it in.

Actually things have gotten worse--they've apparently broken the "save PDF to Zotero" function when downloading PDFs.

zuphilip

Actually things have gotten worse--they've apparently broken the "save PDF to Zotero" function when downloading PDFs.

Yeah, this is kind of the same as calling the url for the PDF directly.

Can you reach out to frontiers? I guess you have more weight as reader and maybe author of their papers.

mdt_

It would be good to have someone associated with Zotero involved as it is an issue between the two systems.

Ok, I have actually written to their web/IT address. It took some effort to write up a reproducible error case so that they can see the issue.

We'll see what comes of it. I may know an editor to hit up for more if this doesn't lead to a solution.

zuphilip

Thank you! Looking forward to hear from them. Keep us informed.

mdt_

I am happy to report that the problem appears to have been temporary due to some site updates on the server-side. Frontiers IT department reported to me that the site is working again and now allows downloads from outside, including Zotero.

I have tested this just now and everything seems to be working fine. So I suspect this matter is closed.

zuphilip

Thanks for the update and I am happy that it works again.

tmchartrand

Re-opening this issue: currently can't pull any pdf attachments from frontiers.
I dug into the code a little and it seems that for single articles the Frontiers translator is just directly calling the Embedded Metadata, which is then failing to find the pdf url.
I would dig a bit further, but having trouble reverting to 4.0 to use Scaffold.

adamsmith

Yeah, they have removed the PDF from the metadata, for whatever reason.

To replace this, I think we'll simply want to construct the PDF URL by replacing /\/full([#?].+)?/ with "/pdf" You should be able to do that without scaffold if you're in a hurry.

Note that Scaffold does run in Zotero 5 and provides a browser pane there if you install the latest version built from github
(I'm assuming here this is not over your head, given that your first impulse was to look at the translator code yourself).

If you don't get to it, we'll fix this soon, hopefully.

tmchartrand

Thanks for the tip on Scaffold, I didn't think to look. Will look into a fix today.