Most pages won't get any metadata
I get one of a couple errors when trying to add a citation for most papers.
- Big red "X" : An error occurred saving with [e.g.] ACM Digital Library. Attempting to Save as Webpage instead
- An error occurred while saving this item. See Troubleshooting Translator issues for details.
A new record is created in Zotero (in the correct folder) with no metadata, only the URL.
I have reset translators in Zotero and for the Zotero Extension, to no avail. It doesn't seem to be attached to the particular source (i.e. some ACM Digital Library resources save just fine)
- Big red "X" : An error occurred saving with [e.g.] ACM Digital Library. Attempting to Save as Webpage instead
- An error occurred while saving this item. See Troubleshooting Translator issues for details.
A new record is created in Zotero (in the correct folder) with no metadata, only the URL.
I have reset translators in Zotero and for the Zotero Extension, to no avail. It doesn't seem to be attached to the particular source (i.e. some ACM Digital Library resources save just fine)
ex. URL: https://dl-acm-org.proxy-um.researchport.umd.edu/doi/pdf/10.1145/1651343.1651356
"An error occurred saving with ACM Digital Library. Attempting to save using Save as Webpage instead.
An error occurred while saving this item. Try again, and if the issue persists see Getting Help for more information."
But most likely Zotero hasn't detected your proxy server. If you go the Proxies tab of the Zotero Connector preferences, do you see an entry for your proxy?
There are two proxies listed. One is for my old institution and the other for the new one. If I click on them, the old institution has dozens of hostnames listed; the new institution has three.
What's the scheme for the new one?
If you restart your browser and log in to ACM through your university library, does Zotero detect that you're connecting through the proxy and add dl.acm.org to the host list? If not, can you provide a Debug ID from the Zotero Connector for that process?
On the webpage itself, are you able to generate BibTeX from the "Export Citation" button when connected via the proxy?
2. I don't know what you mean by these things:
"What's the scheme for the new one?" (I don't understand what you mean by scheme)
"On the webpage itself, are you able to generate a citation?"
I have deleted the old proxy and will try to log in with the new university log in. Right now, going to scholar.google.com goes through my old univ proxy, so I realize I might need to change that.
To clarify, Zotero's proxy functionality does two things: 1) it lets Zotero detect the actual site you're on when connected through a proxy (e.g., dl-acm-org.proxy.school.edu → dl.acm.org) and 2) it will redirect you via the proxy if you load an unproxied URL (e.g., dl.acm.org → dl-acm-org.proxy.school.edu). But the latter is no different from just loading the site via the proxy yourself. There's an Export Citation button on the page. You should click that and see if you can generate the BibTeX output. In your save attempt, your UMD proxy is rejecting a request for that, which is preventing saving from working.
UMD-- old
UMich -- new.
I deleted the UMD proxy from Zotero Connector and set up Google Scholar to go through UMich so that they both match.
There's a webpage on ACM listing the information about the paper and with a PDF button (now that I've sorted out Google Scholar, the PDF button is available on this webpage and on alongside the Google Scholar search results).
Now I can get a proper citation through the ACM listing page, but not on the PDF. (so if I search as I normally would and select the PDF from the Google Scholar results page, I will not be able to get the citation, but I can get it if I go through the ACM webpage first, create the citation with Connector, and then open the PDF.)
This seems to be the case for other PDFs as well,
e.g. https://www-tandfonline-com.proxy.lib.umich.edu/doi/pdf/10.1080/10803548.2003.11076589
won't give me a citation, but
https://www-tandfonline-com.proxy.lib.umich.edu/doi/abs/10.1080/10803548.2003.11076589
will.
This is not ideal, but it's much better than it was when we started. Thanks for your help so far. Do you have advice about getting citations from PDFs/is this a known issue?
You should always save via the article page when one is available rather than trying to save the PDF directly.
On some sites with clear URL schemes containing identifiers, Zotero can save normally when on the PDF, but that's less common, either because the site doesn't make it possible or because it hasn't been implemented.
When that's not supported, the save button will just show a PDF icon. You can save the PDF directly to Zotero and it will attempt to retrieve metadata for the PDF, but how well that works depends on the PDF, so it's usually better to go back to the article page and save from there. Zotero will save the PDF as well as long as you have access to it — you don't need to download the PDF separately.
I normally don't have trouble with PDFs, but if I do, I go to the application and ask it to retrieve metadata.
I can adjust my workflow to visit article pages if necessary. I know it gets the PDF, I just don't get any value out of the article page and would prefer to go directly to the PDF to save time. Also, sometimes I'll get a PDF through other means, like email, so searching it to find the article page is a bit of an annoyance. If that's my solution, though, I can do it.