#!/usr/bin/env python
"""
This is a simple program I wrote to check my links on a HTML page.
It does _not_ nest and check the linked pages.

Limitations: If your URL to be checked is a directory, like foo.com/bar,
you must specify it as foo.com/bar/ - with the trailing '/'.

Kumar
"""

import urllib2, sys, re, urllib, urlparse, socket
from HTMLParser import HTMLParser

class LinkParser(HTMLParser):
    files = []
    def __init__(self, base):
        HTMLParser.__init__(self)
        self.base = base
    
    def handle_starttag(self, tag, attrs):
        if tag in ("link", "a"):
            for i in attrs:
                if i[0] == "href":
                    self.files.append(urllib.basejoin(self.base, i[1]))
        elif tag == "img":
            for i in attrs:
                if i[0] == "src":
                    self.files.append(urllib.basejoin(self.base, i[1]))
          

    def getlist(self):
        HTMLParser.close(self)
        return self.files



def get_response(urlstr):
    try:
        conn = urllib2.urlopen(urlstr)
        stuff = conn.read(1)
    except urllib2.HTTPError, herr:
        return ": " + str(herr)
    except urllib2.URLError, err:
        return ": " + err.reason[1]
    conn.close()
    return ""

def print_all(data, base=None):
    lp = LinkParser(base)
    lp.feed(data)
    files = lp.getlist()
    for file in files:
        x = get_response(file)
        if not x:
            print file + ": OK!"
        else:
            print file + x


if __name__ == "__main__":
    if len(sys.argv) != 2:
        print "Usage: %s [httpurl]" % sys.argv[0]
        print "Note: If you want to specify a directory like abc.foo.com/bar, specify it as"
        print "http://abc.foo.com/bar/ - with the trailing '/'"
        sys.exit(1)
    url = urllib.urlopen(sys.argv[1])
    data = url.read()
    print_all(data, base=sys.argv[1])