# MySpace parser
# Tom Dyson, July 2006
# http://www.throwingbeans.org/

# Requires BeautifulSoup, tested on version 3:
# http://www.crummy.com/software/BeautifulSoup/

# TODO: distinguish band sites from personal sites

# usage: 
# from MySpace import *
# n = MySpace('soundofnorthwood')
# total_friends = n.friend_count
# last_comment = n.comments[0]

__version__ = "0.2"
__license__ = "GPL 2"
__author__ = "Tom Dyson"
__contributors__ = "Tim Hatch (http://timhatch.com/)"

__all__ = ["MySpace"]

import BeautifulSoup
import re
import urllib

# Entity decoding from SBP's getlinks.py
import htmlentitydefs
htmlentitydefs.name2codepoint['apos'] = 0x27

r_entity = re.compile(r'&(#x[0-9A-Fa-f]+|#[0-9]+|[A-Za-z]+);')

def entity(m):
	name = m.group(1)
	if name.startswith('#x'): 
		return unichr(int(name[2:].lstrip('0'), 16))
	elif name.startswith('#'): 
		return unichr(int(name[1:].lstrip('0')))
	elif htmlentitydefs.name2codepoint.has_key(name): 
		return unichr(htmlentitydefs.name2codepoint[name])
	else: return '&' + name + ';'

def decode_entities(s):
	return r_entity.sub(entity, s)

SPACE_STUB = 'http://www.myspace.com/'
FRIEND_STUB = 'http://profile.myspace.com/index.cfm?fuseaction=user.viewprofile&friendid='
FRIEND_HREF_ID = re.compile('ctl00_Main_ctl00_UserFriends1_F')
ADD_FRIEND_ID = 'ctl00_Main_ctl00_UserContactLinks1_AddFriendLink'
COMMENT_TD = {'align':'center', 'valign':'top', 'width':'150', 'bgcolor':'FF9933', 'style':'word-wrap: break-word'}

class MySpace(object):

	def __init__(self, myspace_name):
		# get the HTML for MySpace URLs or numerical IDs
		if myspace_name.isdigit():
			html = urllib.urlopen(FRIEND_STUB + myspace_name).read()
		else:	
			html = urllib.urlopen(SPACE_STUB + myspace_name).read()
		self.soup = BeautifulSoup.BeautifulSoup(html)
		
		# find MySpace name
		self.name = self.soup.first('span','nametext').string
		
		# find MySpace URL
		title = self.soup.title.string.strip()
		title = title.replace('www.myspace.com/','')
		if len(title): self.url = title
		else: self.url = None
			
		# find MySpace numerical ID
		add_friend_href = self.soup.first('a',id=ADD_FRIEND_ID)['href']
		self.id = re.search('friendID=([0-9]+)', add_friend_href).group(1)
		
		# find number of friends
		self.friend_count = int(self.soup.first('span','redbtext').string)
		
		# find Friends
		friend_links = self.soup.findAll('a', id=FRIEND_HREF_ID)
		self.top_friends = []
		for link in friend_links:
			if link.string and len(link.string.strip()):
				friend_name = decode_entities(link.string.strip())
				friend_id = link['href'].replace(FRIEND_STUB, '')
				self.top_friends.append((friend_name, friend_id))
				
		# find influences
		try:
			self.influences = self.soup.first('td',id='ProfileInfluences').string
		except:
			self.influences = None
			
		# get comments
		self.comments=[]
		for td in self.soup.findAll("td", attrs = COMMENT_TD):
			comment_link = td.first('a')
			commenter = decode_entities(comment_link.string.strip())
			comment_href = comment_link['href']
			commenter_id = re.search('friendID=([0-9]+)', comment_href, re.IGNORECASE).group(1)
			comment_td = td.nextSibling.nextSibling
			comment_date = comment_td.first('span','blacktext10').string.strip()
			comment_lines = []
			for node in comment_td.contents:
				if len(node):
					try: # look into why this is failing
						comment_line = node.string.strip()
						if comment_date not in comment_line:
							comment_lines.append(comment_line)
					except: pass
			comment = ' '.join(comment_lines).strip()
			self.comments.append((commenter, commenter_id, comment_date, comment))