extractdoi

#!/bin/sh

# Extract DOI from PDF files
# Will take the first DOI on the first page, which usually works

# author:   François-Xavier Coudert
# e-mail:   fxcoudert@gmail.com
# updated:  2014-05-03


extract () {
  printf "%s" "$1"
  text=`pdftotext -f 1 -l 1 -raw "$2" - 2>/dev/null | grep '\<10\.'`
  echo $text | perl -pe 's@.*[ /:]10\.@10\.@ ; s@[ \t].*@@ ; s@[.,;]$@@'
}


if [ $# -eq 1 ]; then
  extract "" "$1"
else
  for i in "$@" ; do
    extract "$i: " "$i"
  done
fi