Automated legal download of multiple full-text PDFs by parsing from entries in bib file

(Krishnakumar) #1

I am posting this question here after being badly bruised by Mendeley in it’s capabilities to handle PDFs gracefully.

Working from one of my university’s library computer, I had downloaded PDFs of papers for later reading and citing purposes. Thus far, I always had a good experience syncing PDFs with Mendeley’s web-importer in google chrome (on my personal machine). The library machine had an unsupported browser and furthermore did not permit any extension installation, but they did have Mendeley Desktop already installed and available. I logged on to my account and could see the entries (including all PDFs) of my entire library and this lent me confidence.

In 2-3 hours, I downloaded about 65-70 relevant papers for my research. And upon drag and dropping these files to Mendeley Desktop, it correctly recognised the metadata and created entries. I could even see the PDF icon which opened correctly upon double-clicking. I hit the sync button and left the library. The reassuring upload progress bar convinced that everything went smooth.

After coming home and opening Mendeley Desktop on my computer, I realised that although the new additions had synced, I was missing the PDFs. Now, I could manually download all 65 of them, but this does not seem like an efficient use of my time. I am able to generate a single bib file from Mendeley.

So, the question is: Can Jabref automatically parse the DOI/URL of the bibtex entries and download all the 65 full text PDFs using my university’s network which provides full, legal access to them?

(Christoph) #2


JabRef has integrated fulltext fetchers to download the fulltext for an entry.
Just select a number of entries and hit the Lookup Fulltext documents. However, be warned that you have to manually confirm the download. And JabRef is not intended for bulk downloads (we don’t want to get blacklisted by publishers), so you can currently always select bunches of 5 for automatic full text download.

The other more way is currently to open the Entry Editor -> General -> File field -Click on the magnifier icon and it will download the fulltext and rename it according to the defined filename patterns and put in the configured file dir. You can then simply navigate with Page Down/Page Up to the next entry and the current entry editor’s tab will stay visible.

Edit// I suggest you use the latest master, as there were some fixes to the IEEE Fetcher.

Hope this helps!

(Krishnakumar) #3

Amazing. That did indeed work.

However, even with the latest master (I am using the single ‘jar’ portable file on windows), the IEEE importer issue persists.

I have opened an issue on github with all relevant information to reproduce the issue. Please see here:

(Christoph) #4


yes, please and possible with errors from the Event Log.


(Krishnakumar) #5

@Siedlerchr I have to thank you wholeheartedly. Your support was immensely valuable. And saved me a lot of time …

I have opened an issue on github with all relevant information to reproduce the issue. Please see here:

I cannot post event log, since I am not using jabref.exe, but instead the jar portable. Is my understanding correct?

Thanks once again for the kind help. Much appreciate the gesture.

(Christoph) #6


I am glad I could help and thanks for creating the isuse. That helps us to track down the error.
With Event Log I meant the thing under Help -> View Event Log. Exceptions or other errors are there,

Feel free to contribute the development of JabRef through one of the many options.