Nicholas Cheek

LITTLE ROCK WEB DEVELOPER

import re
import urllib.request

# Retrieves URLs from the HTML source code of a website
def extractUrlsFrom(url, unique=True, sort=True, restrictToTld=None):
    # Prepend "www." if not present
    if url[0:4] != "www.":
        url = "".join(["www.",url])
    # Open a connection
    with urllib.request.urlopen("http://" + url) as h:
        # Grab the headers
        headers = h.info()
        # Default charset
        charset = "ISO-8859-1"
        # If a charset is in the headers then override the default
        for i in headers:
            match = re.search(r"charset=([\w\-]+)", headers[i], re.I)
            if match != None:
                charset = match.group(1).lower()
                break
        # Grab and decode the source code
        source = h.read().decode(charset)
        # Find all URLs in the source code
        matches = re.findall(r"http\:\/\/(www.)?([a-z0-9\-\.]+\.[a-z]{2,6})\b", source, re.I)
        # Abort if no URLs were found
        if matches == None:
            return None
        # Collect URLs
        collection = []
        # Go over URLs one by one
        for url in matches:
            url = url[1].lower()
            # If there are more than one dot then the URL contains
            # subdomain(s), which we remove
            if url.count(".") > 1:
                temp = url.split(".")
                tld = temp.pop()
                url = "".join([temp.pop(),".",tld])
            # Restrict to TLD if one is set
            if restrictToTld:
                tld = url.split(".").pop()
                if tld != restrictToTld:
                    continue
            # If only unique URLs should be returned
            if unique:
                if url not in collection:
                    collection.append(url)
            # Otherwise just add the URL to the collection
            else:
                collection.append(url)
        # Done
        return sorted(collection) if sort else collection

# Test
url = "snipplr.com"
print("From:", url)
for x in extractUrlsFrom(url):
    print("-", x)

LITTLE ROCK WEB DEVELOPER

Living the dream.

nick@nicholascheek.com