Plain text import?

Imagine the case (as has now happened a few times) that I’ve been given a manuscript to work on (in Word) and there is a reference section at the end. Is there any way I can upload a plain text file to paperpile, and let it try to work out the references?

e.g. here is a typical reference list in plain text…

Eytan and Marom (2006)  “Dynamics and Effective Topology Underlying Synchronization in Networks of Cortical Neurons“, J. Neuroscience, 26, 8465-8476.

Grandjean P, Landrigan PJ. Developmental neurotoxicity of industrial chemicals. Lancet. 2006 368, 2167-78.

Grandjean P, Landrigan PJ. Neurobehavioural effects of developmental toxicity. Lancet Neurol. 2014. 13, 330-8.

Thanks, Stephen

1 Like

That’s not something Paperpile can do at the moment. It’s just hard because parsing citation styles can be a pain and totally fail for styles without titles.

But others have tried this. Perhaps this resource is useful, they mention a few services. If you can parse it in a BibTeX or RIS you can easily import it to Paperpile.

http://labs.crossref.org/resolving-citations-we-dont-need-no-stinkin-parser/

1 Like

Thanks Stefan. I just hacked something together with the crossref.org simple text interface. This is very neat! I’ll write up my solution tomorrow I hope.

So, thanks to Stefan’s solution, it was quite easy. All I did was to paste some references, like the following, into http://www.crossref.org/SimpleTextQuery/ (make sure you register first to use the service)

Eytan and Marom (2006)  “Dynamics and Effective Topology Underlying Synchronization in Networks of Cortical Neurons“, J. Neuroscience, 26, 8465-8476.

Grandjean P, Landrigan PJ. Developmental neurotoxicity of industrial chemicals. Lancet. 2006 368, 2167-78.

Grandjean P, Landrigan PJ. Neurobehavioural effects of developmental toxicity. Lancet Neurol. 2014. 13, 330-8.

and then grabbed the DOIs

Eytan and Marom (2006) "Dynamics and Effective Topology Underlying Synchronization in Networks of Cortical Neurons", J. Neuroscience, 26, 8465-8476.
http://dx.doi.org/10.1523/JNEUROSCI.1627-06.2006

Grandjean P, Landrigan PJ. Developmental neurotoxicity of industrial chemicals. Lancet. 2006 368, 2167-78.
http://dx.doi.org/10.1016/S0140-6736(06)69665-7

Grandjean P, Landrigan PJ. Neurobehavioural effects of developmental toxicity. Lancet Neurol. 2014. 13, 330-8.
http://dx.doi.org/10.1016/S1474-4422(13)70278-3

and then put this output through a simple script to generate the bib files:

grep '^http' op.txt | doi2bib.sh > o.bib

where doi2bib.sh is the simple script, based on crossref’s idea:

#!/bin/sh
# doi2bib.sh -- convert DOI URLs to Bibtex reference.
# stdin contains the DOIs, one per line.
while read line
do
    curl -LH "Accept: application/x-bibtex" $line
    echo
done

Hi,

I solved it in a similar manner, but using two websites instead of scripting.
Could me more accessible for people who are unfamiliar with the command line.

See my blog post here

Wow, both solutions are very creative. Do you have any experience how good the CrossRef solution works in the first place (independent how you get your papers into paperpile afterwards).

If this is something robust enough for production use, we definitely should consider adding it to Paperpile. There are actually two obvious places where this would fit: The “add manually” menu which has a field where you can paste BibTeX or RIS at the moment. And (not so well know but very useful) the Chrome context menu. It would be possible to just select any bibliography in a website and send it to Paperpile.

I only used it once, for a paper I “inherited”.
It worked perfectly though. I suppose it depends on
how complete the cross ref database is.

The “add manually” would for sure be useful, but it should allow multiple references.
I am not sure how it works right now, but an important use case is
copying the bibliography section from an existing manuscript.

Sure, the add manually supports multiple references already. The idea would be to paste a full bibliography into this box. I will have a closer look at this service and options to integrate it some day.

1 Like

I too only tried my solution so far on one round of papers: I was given a word file with a bibliography of about 20 entries; copy and paste of that bibliography into crossref seemed to catch all but one reference – and the one reference it couldn’t do I didn’t expect it to (it was a reference to a web site).

But I thought the crossref solution seemed to make sense, and is better than nothing I think even if you need to check the output manually.

I’d already be happy if paperpile accepted a bunch of DOIs, one per line, for import, but if you wanted to go the extra step and allow for plain text input for you to pass to crossref.org, even better!!!

Maybe PaperPile could integrate something like this? https://anystyle.io/

Rather than integrating, I understand we’ll develop our own plain-text detection and are getting closer to that capability. I don’t have an expectation to share at the moment, but releasing these format converters was a step in that direction (and more are currently in development).