#!/usr/bin/env python
#
#usage:  get.py anyfile
#or: get.py http://anything.anywhere.com/anyfile.html
#
#we need regular expressions, ability to access files via http,
#and access to command line arguments:
import re, urllib2, sys

filename=sys.argv[1] #get filename from command line
re_http=re.compile(r"http://"); #pattern to check if file is access by http
#read file into a string:
if re_http.search(filename):
	print "connecting to: ",filename
	content=urllib2.urlopen(filename).read()
else:
	print "reading local file: ",filename
	infile=open(filename,"r")
	content=infile.read()

#make a dictionary (= perl's hash) of patterns to match and exclude
pats={} #initialize dictionary
notpats={} #ditto

#following are the patterns regular expressions to match.
#note the r"""..., this means it is a raw string, which means nothing
#special is done with the backslashes \; that specialness is preserved
#for the pattern matching operation
#note that the part of the pattern between the () is what is printed out
# see "4.2 re" at http://python.org/doc/current/lib/lib.html:
pats['http']=r"""http://(.*?)[\"\s\']"""
pats['relative url']=r"""href=\s*[\"\'](.*?)[\"\']"""
notpats['relative url']=r"""http://|mailto:"""
pats['email address']=r"""([\w\.]+\@[\w\.]+)"""

#your task: comment out the above and make the following functional:

#find url of frames, (check at wwww.ou.edu):
#pats['frame']= 

#find "Bobby noncompliant images, i.e. images without alt tag
#check at:
#   http://soonersports.ocsn.com
#   http://it.metr.ou.edu/rgraphics/radar
#   http://weather.ou.edu, etc.
#pats['bobby violation']=
#notpats['bobby violation']=

#find downloadable pdf documents.  check at http://www.ou.edu/provost/pronew/ 
#pats['pdf']=

keys=pats.keys()
keys.sort()
for key in keys:
	pat=re.compile(pats[key],re.S) # S option means .* can include newline
	notpat=""
	if notpats.has_key(key): notpat=re.compile(notpats[key],re.S)
	print "\n",key,": looking for ",pats[key],
	if notpat:
		print "without ", notpats[key]
	else:
		print ""
	count=0
	for thematch in pat.findall(content):
		if notpat:
			if notpat.search(thematch): continue
		count=count+1
		print count,": ",thematch