akamai_dsa_weird_report.py

# headers
# URL,Ok Volume (MB),Error Volume (MB),Ok Hits,0xx,200,206,2xx,302,304,3xx,404,4xx,Offloaded Hits,Origin Hits,Origin OK Volume (MB)	 	,Origin Error Volume (MB)


def parse_file(file):
    lines = [l.split(',') for l in open(file) if not l.startswith('#')]
    return lines

def is_cached(line):
    # offloaded hits 
    return int(line[13]) > 0

def is_web_page_with_traffic(line):
    url = line[0]
    ext = url[-4:]
    is_page = ext in ['.jsp', '.htm', 'html'] or url.endswith('/')
    has_traffic = float(line[1]) > 0
    return is_page and has_traffic

def get_cached_objects_percentage(lines):
    cached = sum([1 for l in lines if is_cached(l)])        
    return '%.2f %%' % (100.0 / len(lines) * cached)
    
def get_avg_size_if(lines, callback):
    sizes = []
    for line in lines:
        if not callback(line):
            continue
        try:
            size = float(line[1]) / float(line[3])
        except ZeroDivisionError:
            size = 0
        sizes.append(size)
    return '%.2f KB' % (sum(sizes) / len(sizes) * 1024)

def get_avg_cached_object_size(lines):
    return get_avg_size_if(lines, lambda l: is_cached(l))

def get_avg_web_page_size(lines):
    return get_avg_size_if(lines, lambda l: is_web_page_with_traffic(l))

def main():
    file = '/Users/ernesto/Downloads/dsa_urls_volume_164077_10-01-2012-07-55-50.csv'
    lines = parse_file(file)
    print get_cached_objects_percentage(lines)
    print get_avg_cached_object_size(lines)
    print get_avg_web_page_size(lines)


if __name__ == '__main__':
    main()