doi2pdf

#!/bin/bash

# Script for finding and downloading a PDF or PS for a given 
# Digital Object Identifier (DOI), which usually points to a website 
# containing a link to the actual file.
#      
# Usage:    doi2pdf DOI [filename]
#           doi2pdf http://dx.doi.org/DOI [filename]
#      
# If filename is omitted the raw PDF/PS is written to a temporary file.
#      
# Example:  doi2pdf 10.1016/j.orl.2012.11.009
#
# Notes:
#    IEEE is not working because it uses JavaScript

function test()
{
	grep doi ../../Resources/Bibliography/World.bib | sed 's/[ \t,}]//g' | sed 's#doi={#http://dx.doi.org/#g' | while read doi
	do
		php doi2pdf.php $doi /dev/null

		if [ $? -eq 0 ]
		then
			echo *PASS* $doi
		else
			echo *FAIL* $doi
		fi
	done
}

function gettmpfile()
{
	local tmppath=${1:-/tmp}
	local tmpfile=$2${2:+-}
	tmpfile=$tmppath/$tmpfile$RANDOM$RANDOM.tmp
	if [ -e $tmpfile ]
	then
		# if file already exists, recurse and try again
		tmpfile=$(gettmpfile $1 $2)
	fi
	echo $tmpfile
}

# Issue help message if necessary
if [ $# = 0 ]; then
	echo "Usage: $(basename $0) DOI [filename]"
	echo "or:    $(basename $0) http://dx.doi.org/DOI [filename]"
	echo "will try and retrieve the PDF file for the DOI given and save it to filename or writes to stdout if filename is omitted."
	exit 1
fi

# Extract doi. Pattern from https://github.com/regexhq/doi-regex/blob/master/index.js
DOI=$(echo $1 | grep '10[.][0-9]{4,}(?:[.][0-9]+)*/(?:(?![%"#? ])\S)+' -P -o )


# Assign output file name if possible
# and call doi2pdf.php
if [ -z "$2" ]
then
	# invent a good filename
	FN=$(doi2pdfname.sh "$DOI")
	if [ -z "$FN" ]
	then
		FN=$(gettmpfile);
	fi
	php $(dirname $0)/doi2pdf.php $DOI $FN || exit 1
else
	php $(dirname $0)/doi2pdf.php $@ || exit 1
fi

exit 0