#!/usr/bin/env python

import sys
import urllib
import zipfile
import subprocess
import HTMLParser

class SubDivXQuery:
	def __init__(self, to_search, page_number):
		self.host = "www.subdivx.com"
		self.page = "/index.php"
		self.down_page = "/bajar.php"
		self.query = dict(
			buscar = to_search,
			pg = page_number,
			accion = 5,
			masdesc = '',
			subtitulos = 1,
			realiza_b = 1,
		)
	@property
	def url(self):
		return 'http://%s%s?%s' % (self.host, self.page,
				urllib.urlencode(self.query))
	@property
	def page_uri(self):
		return self.page + '?' + urllib.urlencode(self.query)
	@property
	def down_uri(self):
		return 'http://' + self.host + self.down_page


class SubDivXHTMLParser(HTMLParser.HTMLParser):

	IDLE = 1
	HEADER = 2

	def __init__(self, down_uri):
		HTMLParser.HTMLParser.__init__(self)
		self.down_uri = down_uri
		self.depth = 0
		self.parsing = False
		self.subs = []
		self.attr = None
		self.cur = None
		self.in_script_style = False

	def handle_starttag(self, tag, attrs):
		attrs = dict(attrs)
		if tag == 'div' and attrs.get('id') == 'menu_detalle_buscador':
			self.cur = dict()
			self.subs.append(self.cur)
			self.parsing = True
		if not self.parsing:
			return
		if tag == 'script' or tag == 'style':
			self.in_script_style = True
			return
		if tag == 'div':
			if attrs.get('id') == 'buscador_detalle':
				self.parsing = True
			elif attrs.get('id') == 'buscador_detalle_sub':
				self.attr = 'desc'
		elif tag == 'a':
			if attrs.get('class') == 'titulo_menu_izq':
				self.attr = 'titulo'
			elif attrs.get('href', '').startswith(self.down_uri):
				self.cur['url'] = attrs['href']
		# br are usually not closed, so ignore them in depth calculation
		if self.parsing and tag != 'br':
			self.depth += 1

	def handle_endtag(self, tag):
		if self.parsing:
			if tag == 'script' or tag == 'style':
				self.in_script_style = False
				return
			# see comment in handle_starttag()
			if tag != 'br':
				self.depth -= 1
		if self.depth == 0:
			self.parsing = False

	def handle_data(self, data):
		if not self.parsing:
			return
		data = data.strip()
		# Hack to handle comments in <script> <style> which don't end
		# up in handle_comment(), so we just ignore the whole tags
		if self.in_script_style:
			return
		if self.attr is not None and data:
			self.cur[self.attr] = data
			self.attr = None
		elif data in ('Downloads:', 'Cds:', 'Comentarios:',
				'Formato:'):
			self.attr = data[:-1].lower()
		elif data == 'Subido por:':
			self.attr = 'autor'
		elif data == 'el':
			self.attr = 'fecha'


def subdivx_get_subs(query_str):
	page_number = 1
	subs = []
	while True:
		query = SubDivXQuery(query_str, page_number)
		url = urllib.urlopen(query.url)
		parser = SubDivXHTMLParser(query.down_uri)

		for line in url:
			parser.feed(line)

		url.close()

		if not parser.subs:
			break

		subs.extend(parser.subs)
		page_number += 1

	return sorted(subs, key=lambda s: int(s['downloads']), reverse=True)


def get_subs(query_str):
	zip_exts = ('application/zip',)
	rar_exts = ('application/rar', 'application/x-rar-compressed')

	for sub in subdivx_get_subs(query_str):
		print '''\
	- %(titulo)s (%(autor)s - %(fecha)s - %(downloads)s - %(comentarios)s)
	  %(desc)s
		DOWNLOADING ...
	''' % sub
		fname, headers = urllib.urlretrieve(sub['url'])
		if 'Content-Type' in headers:
			if headers['Content-Type'] in zip_exts:
				z = zipfile.ZipFile(fname, 'r')
				z.printdir()
				for fn in z.namelist():
					if fn.endswith('.srt') or fn.endswith('.sub'):
						if '..' in fn or fn.startswith('/'):
							print 'Dangerous file name:', fn
							continue
						print 'Extracting', fn, '...'
						z.extract(fn)
			elif headers['Content-Type'] in rar_exts:
				if subprocess.call(['rar', 'x', fname]) != 0:
					print 'Error unraring file %s' % fname
			else:
				print 'Unrecognized file type:', headers['Content-Type']
		else:
			print 'No Content-Type!'


for q in sys.argv[1:]:
	get_subs(q)