Python script that goes through an html file and prints a css file with selectors seen.

css_gen.py

#!/usr/bin/env python

"""
	css_gen.py < stdin
	
	Reads through the input (should be an html file), looking for classes and ids, and
	and then grouping them by similar name. It then prints out the text for a CSS file.
	
	The CSS file has all selectors, for each selector it has no properties.
	
	Ex:
		<div class="wrapper" id="foo"></div>
		->
		.wrapper {}
		#foo {}
"""


try:
	from pyquery import PyQuery as pq
except:
	print "You need pyquery installed"

import sys

# Writes the line of css, if type is true, it writes an i
def css(selector, id= False):
	if id:
		return '#'+ str(selector) + ' {\n\n} \n'
	else:
		return '.'+ str(selector) + ' {\n\n} \n'

# Adds an item to a list iff it is unique
def add_if_new(list, item):
	if not item in list:
		list.append(item)

# String to be printed
main_string = """
body, html { width:100%; height:!00%; margin:0; padding:0; }
img, a { outline:0; border:0; }
body {}
"""

# Gather all of the nodes into a list to be parsed later

file = sys.argv[1]
html = ''
css_classes = []
css_ids = []
html_nodes = None

html = sys.stdin.read()
	
html_nodes = pq(html)
html_nodes = html_nodes.find('*')

for node in html_nodes:
	attrib = node.attrib
	keys = attrib.keys()
	
	# Check for classes or ids, and then split them into separte
	# lists
	
	if 'class' in keys:
		_class = attrib['class']
		_class = _class.split(' ')
		
		for _c in _class:
			add_if_new(css_classes, _c)
			
	if 'id' in keys:
		_id = attrib['id']
		add_if_new(css_ids, _id)
		
css_classes = sorted(css_classes)
css_ids     = sorted(css_ids) 

css_classes_org = []
css_current = ''

# Group classes with similar names
# Not needed to be done for IDs becuase IDs are supposed to be unique
for co in css_classes:
	split_up = co.replace('.','').split('-')[0]
	if split_up != css_current:
		css_current = split_up + ''
		css_classes_org.append([])
		
	css_classes_org[len(css_classes_org)-1].append(co)

# Iterate through both classes and ids, and appaned it to the main_string	
for _cat in css_classes_org:
	main_string = main_string + '/* %s */\n' % _cat[0].replace('.', '').split('-')[0]
	for _c in _cat:
		main_string = main_string + css(_c)
	main_string += '\n'

for _id in css_ids:
	main_string = main_string + css(_id, True)

	
print main_string