#!/usr/bin/python
import sys, os, re, urllib
from HTMLParser import HTMLParser
import httplib
from subprocess import Popen
import time

def decode(string):
	return re.sub('&#[0-9]{5};', lambda x: unichr(int(x.group(0)[2:-1])),re.sub(' ', ' ', string))

def get_attr(attrs, attr):
	for key, value in attrs:
		if key == attr:
			return value
	return None

class SongParser(HTMLParser):
	song_list = []
	def parse(self, htm):
		self.feed(htm.read().decode("utf-8"))
		return self.song_list
	
	def handle_starttag(self, tag, attrs):
		if get_attr(attrs, "class") == "SongItem BottomBorder":
			self.song_list.append( get_attr(attrs, "id")[3:] )

class SongInfoParser(HTMLParser):
	# song = [ name, singer, size, format ]
	tags = {
			"td-song-name":"name",
			"td-singer":"singer",
			"td-size":"size",
			"td-format":"format"
	}
	song = {}
	attr = None
	def parse(self, htm):
		self.feed(decode(htm.read()))
		return self.song.copy()

	def handle_starttag(self, tag, attrs):
		if tag == "a":
			tag_href = get_attr(attrs, "href")
			if tag_href[0:7] == "/music/":
				self.song["address"] = "http://www.google.cn"+tag_href
		elif get_attr(attrs, "class") == "song-meta-data-table":
			self.attr = ''
		elif self.attr == '':
			tag_class = get_attr(attrs, "class")
			if tag_class in self.tags:
				self.attr = self.tags[tag_class]

	def handle_data(self, data):
		if self.attr:
			self.song[self.attr] = data

	def handle_endtag(self, tag):
		self.attr = ''

conn = httplib.HTTPConnection("www.google.cn")

if len(sys.argv) == 1:
	exit(0)

"""
query = ' '.join(sys.argv[1:])
print "send query: %s" %query 
qeuery = urllib.quote(query)
conn.request("GET", "http://www.google.cn/music/search?q="+query)
result = conn.getresponse()
"""
result = urllib.urlopen(sys.argv[1])

#print result.read().decode("utf-8")

print "Get song list..."

p = SongParser()
song_ids = p.parse(result)
songs = []
parser = SongInfoParser()

for i, id in enumerate(song_ids):
	conn.request("GET", "/music/top100/musicdownload?id="+id)
	response = conn.getresponse()
	song = parser.parse(response)
	name = song["name"]
	time.sleep(1)
	print "%d\t%s\t%s\t%s\t%s" %(i, name, song["singer"],song["format"], song["size"])
	songs.append(song)

choice = range(len(songs))

for i in choice:
	song = songs[ int(i) ]
	if isinstance(song, dict):
		print "download: \t\t%s" %(song["address"])
		Popen(['wget', '-c', '-P', song["singer"], song["address"]]).wait()
conn.close()