#!/usr/bin/python3
"""Quick-and-dirty HTML minimizer"""
import sys
import re
import html.entities
import itertools
import os.path
_SERVERS = ['a.wearehugh.com',
'b.wearehugh.com',
'c.wearehugh.com',
'd.wearehugh.com']
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 "![](i/")
', '')
line = line.replace('', '')
# replace entities with Unicode characters
for e in re.findall('&(.+?);', line):
if e in ('lt', 'amp', 'quot', 'apos', 'nbsp'):
continue
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('#', ''))
else:
continue
line = line.replace('&' + e + ';', chr(n))
# strip leading and trailing whitespace, except inside blocks
g = line.strip()
if g.count(' in one line
in_pre = False
g = line.rstrip()
if in_pre:
_out.write(line)
else:
_out.write(g)
if g.lower() == '':
_out.write('\n\n'.format(os.path.basename(input_file)))