Re-fetching an existing item

I would like to see a button to update (re-fetch) from PMID or doi to get the last up-to-date information.
The problem is generally occuring because I bookmark a lot of article "In press", which doesn't have a volume:no. If I cited them, I can't, a couple of months later, update easily my word document. I have to create a new item, so the link is lost between Zotero and Word. Furthermore, I have now in my librairy many "orphan" citations which I can use as-it, which if I need them, I will have to re-do all the process to re-save them, this time with a volume:no.
This feature exist in Mendeley.
  • Because of the way the search mechanism that powers DOI and Pubmed searches it built, this ought to be pretty easy, and you're right that this would be useful.

    But note that you can just update the volume and issue numbers manually in Zotero in the meantime to avoid creating orphans and duplicates. You can click on the DOI label in the info tab to go to the page, then just type in the missing information-- that should take less time than creating a new item and fixing the citation.
  • I second how useful this would be. I really want this feature.

    I'm somebody who switched to Zotero for the promise of a single library synchronized across multiple PCs. I'd used other citation managers previously, and did not want to lose my library. So after importing a bunch of references from a different citation manager, I noticed some persistent errors. Page numbers were wrong (for some reason a lot were written as last, first instead of first-last), journal names were capitalized differently, extra words were added to the journal titles for some articles, abbreviations used for others, etc. Lots of sloppiness. Not to mention missing entries in the Journal Abbreviation field. So needless to say, when I tried to actually create a bibliography it was a mess.

    I've also noticed occasional formatting inconsistencies with Zotero's translators. Usually these are small things, like page numbers being first -last (with a space) instead of first-last (no space), or differences with journal name or missing abbreviations depending on where I am when I choose to add the article (PubMed vs. Google Scholar vs. ScienceDirect, etc.).

    While these are essentially bugs in other software or small problems in the translators themselves, their effective damage to the Zotero library can be quite persistent. If I could click a button and magically fix these problems, I would be thrilled.

    My solution so far is the one suggested by ajlyon: I go in and hand-correct every broken reference. I've found well over 100 "broken" items in my library, so fixing them has provided hours of productive procrastination. Should somebody less stubborn than me make a bibliography and find it riddled with small errors, though, they may just conclude that Zotero stinks and give up on it entirely.
  • I don't have a timeline on this specific feature, but please do provide details on the import and translator issues, so they can be addressed. Nits like that can be fixed rather easily, but first they need to be reported.
  • Unfortunately I don't have any useful details to provide. I can't reproduce any translator issues, which is good, since it suggests that they've been fixed. My specific library import issue is not going to be of interest to anyone. I imported into an older version of Zotero, and I suspect the errors were at least in part due to a faulty export from my previous library, or even a series of faulty imports into that library.

    Fixing the translators is great for the next guy who tries to import an article, but in my case I didn't notice the errors for a number of months. I'd imagine that happens quite often, so a tool or plugin to fix existing errors in a library would surely be helpful for a lot of users.
  • I too have thought several times about how useful such a feature would be. It is pretty common for me to have in press references tied to different documents, and though I could download the new reference and delete the old, not only would I have to manually update all the papers that link to it, I have to worry that I've missed one. I could leave the old one in the database to be on the safe side, but over time that will add up to a lot of clutter.

    My two cents on the subject is that I would be 95% as happy with the ability to overwrite one record with another, or to reassign the pointer of one record to another. So, if I have an existing record and download a new version of it, I could tell zotero to reassign the pointer for the existing record to the new record (and then transfer my notes, pdfs, etc, manually). I imagine this approach would be a lot easier to implement, not mention more robust. I worry about creating a feature that just grabs whatever the DOI currently links to--I find in a lot of cases that I'm better off downloading a citation from one database or another, or directly from the journal web page because those may be more complete than what comes up off the DOI. So in those cases, a DOI based refetch would be less useful, and in some cases, actually dangerous.

    Another approach (that may be completely impractical) would be for zotero to (somehow, and I don't know enough to know how) remember where the citation came from, and then refetch from that page when requested.
  • So, if I have an existing record and download a new version of it, I could tell zotero to reassign the pointer for the existing record to the new record
    Zotero 3.0 has duplicate detection that will let you do that.
  • Well, color me thrilled. I clearly haven't paid enough attention to the features in the new release--time to upgrade.
  • Yes, the duplicate merging system makes this much less urgent, but it would still be a fun feature to add.
  • Here's the relevant ticket, for reference.
  • I would like to join the request! The reference manager citavi has a function like this and it is really easy and very comfortable.
  • This is definitely a must feature. I can think of several scenarios that make it useful:

    1. I often download citations for article "in press" or "online first" and would like to be able to right click later on and fetch the complete citation once it is published (this would include downloading the PDF attachment if I have access to the journal or another online source exists).

    2. Citations downloaded from some online sources are incomplete, but once they are downloaded it would be nice to be able to right click on the citation and have it updated from other more complete sources (plus fetch the doi and PDF as above).
  • This feature would be really useful. I just switched from Mendele (for obvious reasons) and there I always used the DOI or PMID lookup to correct entries.

    I like the Mendeley solution very much, ther you just have a little update icon nextto the DOI and PMID field.
  • This is generally planned, though no specific timeline.
  • I absolutely support this request and it seams that it is still pending.
    If you come from another lit. management software as I do, you can't really do without this feature. A simple update button next to the doi etc. would do the job. Additionally, right-click on multiple items of your database with the same "update from doi" would be very helpful to manage large databases and keep them up to date.
  • I just can believe it. A task which is easy 7 years ago still not realized in 2018.1.15 !!! And weird enough, it is already planned in 2013!! A must have feature!! Please add it. Thank you so much
  • Just adding my +1 to this as a feature request, as it would save me a heck of a lot of time. Does the 'conflict resolution pane' mentioned in refer to an adaptation of the UI used for managing, at a field level, the merging of duplicate items? If so, that would be marvellous!
  • We're working on the ability to re-fetch metadata, but you can already use merging for this, which is why it hasn't been a priority. You can just save the item again — most easily by pasting the DOI/PMID into Add Item by Identifier — and then select both items, right-click, and merge.
  • Great to hear it's still being worked on :) I have indeed been using the merge function to update from re-added records, but I hadn't thought of using add by identifier to speed up getting those records into my library - thanks for the tip. An update/refresh button would still be good but I'll use that method in the meantime. Cheers!
  • I really hope you include the feature soon. In the meantime, why there is not a way to automatize this process. Like a function or sync button that actually does that, it takes the DOI and readds and object and merge the objects with the same DOI.
  • I agree that this feature is needed. I currently have ~80 items in my library that are missing page numbers because I added them in 'just accepted' or 'early view' format (Angewandte Chemie accounts for many). It would be very tedious to add by DOI and merge one by one. All you would have to do is write some basic code to automate the process as the previous commenter mentioned.
  • Definitely need this feature, too. This is the thing I'm most uncomfortable with switching from Mendeley.
  • I highly recommend having this much-needed feature. So many times I have to go back to the papers and manual add the missed page, date, etc.
Sign In or Register to comment.