The name is a short for bash bibtex file or article to bibtex file (art2bib). This script will print bibtex of an article or a book from DOI, ISBN, ArXiv or PMID.
It requires:
- bash (tested with 4.4),
- wget (tested with 1.19),
- sed (tested with 4.5),
- (coreutils).
To execute the script give it execute right.
chmod +x art2bib
Some tests for fun:
./art2bib d 10.1142%2F10409
./art2bib isbn 0323030645
./art2bib a 1108.2700
./art2bib pmid 23150728
Those are some random examples.
You can use the first letter or the full name of the reference system.
To make a bib file of your own library, (if you have your pdfs with the important informations between brackets):
find -type f -iname '*.pdf' | sed 's/^.*\[//; s/\].*//'
Now, the input of the entire library are ready for the script, we will use the while loop to find all the bib files:
while read i; do art2bib $i ; done < <( find -type f -iname '*.pdf' | sed 's/^.*\[//; s/\].*//' ) > list.bib
The file called list.bib is ready !
Maybe add more reference systems, or make the program find the reference system by itself.