Easiest way to retrieve metadata for PDFs without OCRed text

gabrielhand
Hi all,

I am currently attempting to move from Mendeley to zotero due to various issues with Mendeley.

I thought the best way to preserve my library of ~1200 journal articles would be to take my folder of PDFs generated by Mendeley, which are named like "Afari, Buchwald - 2004 - On Chronic Fatigue Syndrome.pdf" select all, and drag them into zotero then let it retrieve the metadata for them, then manually sort them all again.

Of these, about 80 PDFs give me "Could not read text from PDF".

In Mendeley one can simply enter the DOI to retrieve metadata for files that can't be read by the software.

Also, in Mendeley you can search for metadata by filename and if I google the filename of these PDF in Scholar, they always give me the correct result.

Can someone tell me how to do either of these things in zotero or another simple way to add metadata for ~80 pdfs that cannot be read?
  • dstillman
    edited March 27, 2019
    In Mendeley one can simply enter the DOI to retrieve metadata for files that can't be read by the software.
    Paste the DOI into Add Item by Identifier in the Zotero toolbar, and then drag the PDF item onto the item that was created.
    Also, in Mendeley you can search for metadata by filename and if I google the filename of these PDF in Scholar, they always give me the correct result.
    Sorry, not sure what you mean by this one. If you're asking how you can copy the filename to paste into a search, you can click the filename in the right-hand pane, copy from the dialog that pops up, paste that into Google Scholar, and then save to Zotero normally using the Save to Zotero button, and then drag the PDF onto the created item (if Zotero doesn't save a PDF already).
  • gabrielhand
    Thanks dstillman, I'll try that tomorrow when I get into the lab.

    As for the second thing, I just meant that in Mendeley another way of retrieving metadata was to click a button that would search Google Scholar by whatever title was currently there to attempt to retrieve the rest of the metadata from the first result.

    Thanks again for your help.
  • dstillman
    We're working on a function to update partial metadata, though if you have the DOI, ISBN, etc., it's generally more precise to just paste it into Add Item by Identifier.

    But to approximate what you're describing, you can also click on the PDF item and select Google Scholar Search from the Locate menu (green arrow above the item pane), and then possibly remove ".pdf" from the search. Assuming that finds a result, save that to Zotero and drag the PDF onto it.
  • gabrielhand
    Thanks again for your help dstillman, I think I'm on top of it now.

    One simple suggestion — if the two functions "Add item(s) by identifier" and "Create Parent Item" were combined into the new function "Create Parent Item by Identifier" when right clicking on a PDF, it would have saved me somewhere between one and two hours of work in this process of moving over from Mendeley to zotero. And I'm sure that's something it'd be better to encourage :)

    Cheers
  • dstillman
    We're planning to add the ability to update metadata for items, so you'll be able to create a parent item, enter a DOI, and then update the metadata to fill in other fields.
Sign In or Register to comment.