This middleware strips unnessicery characters from a HTML document in order to save bytes. It will removes things like CR, LR, TAB's and multiples spaces.

html.py

"""
The idea of this was to strip responses of any unnecessary
characters to save bandwidth, after some testing the savings
were only up-to 8% on rare occasion.
"""

import re

re_is_html = re.compile(r'(*)</html>(*)')

class HTMLOptimizerMiddleware(object):
    """
    This middleware strips unnessicery characters from a HTML
    document in order to save bytes. It will removes things
    like CR, LR, TAB's and multiples spaces.
    
    This middleware should go before any CacheMiddleware or GZipMiddleware
    """
    def process_response(self, request, response):
        # It's not worth compressing non-OK or really short responses.
        if response.status_code != 200 or len(response.content) < 1000:
            return response
        
        # Check if it is html.
        if not re_is_html.search(response.content):
            return response
        
        # Strip CR, LR, TAB, and more than one space
        response.content = response.content.replace('\r', '')
        response.content = response.content.replace('\n', '')
        response.content = response.content.replace('\t', '')
        response.content = response.content.replace('  ', '')
        
        return response