WUGC_AUscorebot

wugcbot.py

'''
Created on 08/07/2012

@author: roger
'''
import requests
from lxml import etree
import lxml. html
import memcache
from twitter.oauth import OAuth
from twitter.api import Twitter
import re

mc = memcache.Client(['127.0.0.1:11211'], debug=0)

url="http://wugc2012.org/fdsys/pscore/scv_search_tn/search"
t = Twitter(auth=OAuth('secret', 'secret', 'secret', 'secret'))

all_text = lambda element: lxml.html.tostring(element, method="text", encoding=unicode)

team_mapping = {
    "Women's": "@Firetails2012",
    "Women Masters": "@FlyingFoxes2012",
    "Open": "@Dingoes2012",
    "Masters": "Wombats",
    "Mixed": "@Barramundis2012",
    "Guts": "@Ironbarks2012",
}

def wugcbot(url, parms):

    key = 'wugcsearch'
    page = mc.get(key)
    if not page:
        searchpage = requests.post(url, parms, headers={ 'User-Agent': 'Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.26 Safari/535.11' })
        page = searchpage.content
        mc.set(key, page, time=5*60)
    
    html = lxml.html.fromstring(page)
    '''<a href="#" class="Select_Link" onclick="document.location = 'http://wugc2012.org/fdsys/pscore/scv_match/open/556?lang=en'; return false;">[NowPlaying]</a>'''
    current_games = html.xpath("//a[contains(text(),'NowPlaying')]")

    game_set = set()

    for game in current_games:
        game_url = game.get('onclick').split("'")[1]
        game_id = game_url.split('/')[-2] + '/' + game_url.split('/')[-1].split('?')[0]
        game_set.add(game_id)
    
    for game in game_set:
        gamepage = requests.get("http://wugc2012.org/fdsys/pscore/scv_match/%s?lang=en" % game, headers={ 'User-Agent': 'Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.26 Safari/535.11' })
        page = gamepage.content
        html = lxml.html.fromstring(page)
        try:
            titlecell = html.cssselect('td.CM.TG_VSTITLE')[0]
        except IndexError:
            print page
            raise
        titlerowtextlist = titlecell.getparent().text_content().split()
        division = html.cssselect('#breadcrumb a')[1].text_content().split(':')[1][:-1]
        gutsscore = [el.text for el in html.cssselect('.TG_GUTS_SET_SEL')]
        if len(gutsscore):
            game_text = ' '.join(titlerowtextlist + ["[Guts]"]).replace('0 VS 0', gutsscore[0][1:-1].split()[0])
            division = "Guts"
        else:
            game_text = ' '.join(titlerowtextlist + ["[%s]" % division])
        event_rows = html.cssselect('#events tr')
        for row in reversed(event_rows[1:6]):
            cells = [cell.text_content() for cell in row.cssselect('td')]
            if cells[4] == "Goal":
                event_text = ' '.join((cells[4], cells[0] or '', cells[1] or cells[7] or '', "to", cells[2] or cells[6] or ''))
            elif cells[4] in ["GoodThrow", "ThrowMiss"]:
                event_text = ' '.join((cells[4], cells[1] or cells[7] or ''))
            elif cells[4] in ["FollowCatch", "CleanCatch"]:
                continue
                c1 = cells[2] if not '---' in cells[2] else ''
                c2 = cells[6] if not '---' in cells[6] else ''
                event_text = ' '.join((cells[4], cells[1] or cells[7] or '', "caught by", c1 or c2 or ''))
            else:
                event_text = ' '.join((cells[4], cells[0] or ''))
            tweet = ' '.join((event_text, '-', game_text, '#wugc2012'))
            if division in team_mapping and cells[4] in ["Goal", "GoodThrow"]:
                if ((titlerowtextlist[0] == "Australia" and cells[1]) or
                    (titlerowtextlist[4] == "Australia" and cells[7])):
                    tweet = tweet + ' Go ' + team_mapping[division] + '!'
                
            key = ''.join(event_text.split() + re.sub('[0-9]+', 'x', game_text).split())
            doneit = mc.get(key)
            if not doneit:
                t.statuses.update(status=tweet)
            mc.set(key, True, 12*60*60)

if __name__ == '__main__':
    parms={
       'tid':2,
       'lang':'en',
       'team':'Australia',
       'ymd':'',
       'field':'',
    }
    wugcbot(url, parms)