Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Tree: 85be3148de
Fetching contributors…

Cannot retrieve contributors at this time

executable file 60 lines (53 sloc) 2.228 kB
"""Quick-and-dirty HTML minimizer"""
import sys
import re
import html.entities
import itertools
import os.path
_SERVERS = ['',
available_server = itertools.cycle(_SERVERS)
input_file = sys.argv[1]
output_file = sys.argv[2]
in_pre = False
with open(output_file, 'w', encoding="utf-8") as _out, open(input_file, encoding="utf-8") as _in:
for line in _in:
# round-robin image servers
if "<img src=i/" in line:
line = line.replace("<img src=i/", "<img src=http://" + next(available_server) + "/dip3/")
if "url(i/" in line:
line = line.replace("url(i/", "url(http://" + next(available_server) + "/dip3/")
# remove selected comments (but not all comments, because some are conditional comments for IE compat)
line = line.replace('<!-- toc -->', '')
line = line.replace('<!-- /toc -->', '')
# replace entities with Unicode characters
for e in re.findall('&(.+?);', line):
if e in ('lt', 'amp', 'quot', 'apos', 'nbsp'):
n = html.entities.name2codepoint.get(e)
if not n:
if e.count('#x'):
# it's late, forgive me
n = eval(e.replace('#', '0'))
elif e.count('#'):
n = int(e.replace('#', ''))
line = line.replace('&' + e + ';', chr(n))
# strip leading and trailing whitespace, except inside <pre> blocks
g = line.strip()
if g.count('<pre'):
in_pre = True
if g.count('</pre'):
# XXX this will break if you have </pre><pre> in one line
in_pre = False
g = line.rstrip()
if in_pre:
if g.lower() == '<!doctype html>':
_out.write('\n<!-- readable source @{0} -->\n'.format(os.path.basename(input_file)))
Jump to Line
Something went wrong with that request. Please try again.