"""
Extract PDF from IEEE Xplore embeds and write to given path

Set PDF_DUMP_DIR to wherever PDFs should be saved (named as <ID>.pdf)

to run script -> 'python ieexplore.py <id>' (ID can be determined from the 'arnumber' argument on the article's IEEE Xplore landing page)

Please note - this script was only written because of borked PDF embeds in Ubuntu - (writing this was faster than fixing) - for most people it will be faster than nothing...

Luke Hutton - 2011
"""

import urllib2, cookielib, sys, shutil, os
from xml.dom.minidom import parseString, parse

PDF_DUMP_DIR = '%s/Documents/papers' % os.getenv('HOME')

paper_id = sys.argv[1]
embed_url = 'http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=%s' % paper_id

cj = cookielib.CookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))

response = opener.open(embed_url)

#print response.read()
for line in response:
	if ".pdf" in line:
		stripline = line.rsplit('"')
		pdf_url = stripline[1]
		pdf_response = opener.open(pdf_url)

		local_filename = os.path.join(PDF_DUMP_DIR, "%s.pdf" % paper_id)
		local_pdf = open(local_filename, "w")
		local_pdf.write(pdf_response.read())
		local_pdf.close()
		print "Successfully written %s" % local_filename
		sys.exit()

print "Could not extract PDF - check ID was correct and URL format hasn't changed"