#375 Autodownload PDF

[Steven de Jong]

Subject: From a computer scientist: idea and prototype for inclusion in JabREF

---

Hi!

I think I have a truly useful addition for JabREF.

You see, I am writing a survey paper of an entire field (what a job) and was downloading all those PDF documents mentioned in the BibTeX file by hand. After 5 files, I decided to spend my time differently and wrote a small Java program that takes the title and authors of the entries, then searches for PDFs on Google and downloads them (sometimes the user needs to specify which one to download). For the 100 references in my file, I managed to retrieve around 70 PDF files. These are neatly placed in a single folder and renamed to the BibTeX key.

Obviously, it's really just a small prototype, based on the command-line, with naïve parsing of Google results and BibTeX files, but I think the idea is easily transferrable to JabREF and truly great.

I am willing to give you my prototype, or even to develop a module for JabREF. Please let me know.

Regards,

Steven de Jong
Maastricht University, NL

[Steven de Jong]

The main Java file I developed

[Steven de Jong]

Logged In: YES
user_id=1182980
Originator: YES

File Added: PaperFinder.java

[Morten Omholt Alver]

Logged In: YES
user_id=704995
Originator: NO

Steven,

thank you for this addition. Sorry I haven't answered before, but a new feature like this wasn't to be targeted for version 2.3 anyway. Now I'll have a look at it. The code will need to be integrated into JabRef, or course, but I like your idea, and hope we can make something out of it.

[Steven de Jong]

Source files of the paper finder and bibtex merger

[Steven de Jong]

Logged In: YES
user_id=1182980
Originator: YES

File Added: PaperFinder.zip

[Steven de Jong]

Logged In: YES
user_id=1182980
Originator: YES

Thanks for the reply.
I attach my latest version of the sources for this idea.

What's there:

* GUI frontend (unfinished)
* Merge multiple bibtex files into a big one, remove duplicates or near-duplicates
* Retrieve all documents mentioned in a bibtex file using Google
* Give user option to select desired document in case of multiple candidates (command line only)

What's not there:

* Retrieve documents from other sources than Google
* GUI for selecting which documents to keep
* Multi-threaded searching in Google (now searches one by one, rather slow)

I hope at least some of my code could be useful for JabRef. Please do let me know.

[warakurna]

Hej,

I like this idea! Is it still under development?
Together with a citation fetcher for ISI WOS or Science Direct this would be very powerful, since you can do most of your paper search from JabRef without leaving the program and using a web browser. If this can be integrated, wow!
So far, I reaaly like to have ISI WOS and Science Direct search support includinfg a possible import of DOIs. To have the DOI would would be the first step...

[Stefan Kolb]

See https://github.com/JabRef/jabref/pull/101