import difflib, io, json, sys

import requests
from bs4 import BeautifulSoup as bs


with open(sys.argv[1]) as config_file:
    config = json.load(config_file)


def get_file(urlpart, prefixed=True):
    url = config['source']['path'].strip('/') + '/'
    if prefixed:
        url += config['source']['prefix']
    url += urlpart
    r = requests.get(url)
    r.raise_for_status()
    r.encoding = 'utf8'
    return r.text


def check_round_date(round_no, round_date):
    round_html = get_file('runda%d.html' % round_no)
    round_content = bs(round_html, 'lxml')
    date_cell = round_content.select('td.bdnt12')[0].text.split('\xa0\xa0')
    if date_cell[1] != round_date:
        print('\tRound date for round %d differs: expected "%s" got "%s"' % (
            round_no, round_date, date_cell[1]))


def check_dates():
    for i in range(0, len(config['round_dates'])):
        check_round_date(i+1, config['round_dates'][i])


def print_diff(str_1, str_2):
    d = difflib.Differ()
    for diff in d.compare(
            [l.strip() for l in str_1.splitlines()],
            [l.strip() for l in str_2.splitlines()]
    ):
        if not diff.startswith(' '):
            print('\t' + diff)


def check_logoh():
    if 'custom_file' in config['logoh']:
        html_logoh = get_file(config['logoh']['custom_file'], prefixed=False)
    else:
        html_logoh = get_file('logo.html')
    with open('config/' + config['logoh']['template']) as logoh_file:
        template_logoh = logoh_file.read()
    for var, val in config['logoh'].get('variables', {}).items():
        template_logoh = template_logoh.replace('%' + var + '%', val)
    print_diff(template_logoh, html_logoh)


def check_kolorki():
    server_kolorki = get_file('css/kolorki.css', prefixed=False)
    with io.open('config/' + config['kolorki'], encoding='iso-8859-2') as kolorki_file:
        template_kolorki = kolorki_file.read()
    print_diff(template_kolorki, server_kolorki)


def check_vp_table():
    with open('config/' + config['vp_table']) as vp_file:
        vp_table = [l.strip().split(' - ') for l in vp_file.readlines() if l.strip()]
    vp_html = bs(get_file('vptable.html'), 'lxml')
    vp_row = 0
    for row in vp_html.select('tr'):
        imp_cells = row.select('td.bdc1')
        vp_cells = row.select('td.bdc')
        if len(imp_cells) == len(vp_cells) and len(imp_cells) > 0:
            if imp_cells[0].parent == row:
                for i in range(0, len(imp_cells)):
                    if (imp_cells[i].text.strip().replace('\xa0', ' ') != vp_table[vp_row][0].strip()) or \
                       (vp_cells[i].text.strip().replace('\xa0', ' ')  != vp_table[vp_row][1].strip()):
                        print('\tVP table differs: expected (%s - %s), got (%s - %s)' % (
                            *vp_table[vp_row],
                            imp_cells[i].text.strip(), vp_cells[i].text.strip()))
                    vp_row += 1


def check_page_language():
    leaderboard = bs(get_file('leaderb.html'), 'lxml')
    place_str = leaderboard.select('tr > td.bdcc12')[0].text
    if place_str != config['language']:
        print('\tPage language does not match: expected "%s", got "%s"' % (
            config['language'],
            place_str))


def get_cezar_data(team_id):
    r = requests.get('https://msc.com.pl/cezar/?p=213&action=1&id=%d' % (team_id))
    r.raise_for_status()
    r.encoding = 'utf8'
    cezar_html = bs(r.text, 'lxml')
    title = cezar_html.select('p.msc_pagetitle')[0].contents[0].strip().replace('Drużyna ', '').replace('\xa0', ' ').replace('  ', ' ')
    player_ids = []
    for link in cezar_html.select('a[href]'):
        if link['href'].startswith('?p=21&pid='):
            player_ids.append(int(link['href'].replace('?p=21&pid=', '')))
    return (title, player_ids)


def get_cezar_names():
    with open(config['team_names']) as team_file:
        teams_lines = team_file.readlines()
    teams = {}
    for t in teams_lines:
        t = t.strip()
        if t:
            t = t.split()
            teams[int(t[0])] = int(t[1])
    return { team_no: get_cezar_data(team_id) for team_no, team_id in teams.items() }


def get_html_names():
    leaderboard = bs(get_file('leaderb.html'), 'lxml')
    cells = leaderboard.select('td.bd')
    teams = {}
    for cell in cells:
        links = cell.select('a[onmouseover][href]')
        for link in links:
            if link['href'].startswith(config['source']['prefix'] + 'T'):
                team_id = int(link['href'].replace(config['source']['prefix'] + 'T', '').split('.')[0])
                team_name = link.text.strip()
                team_page = bs(get_file('T%d.html' % (team_id)), 'lxml')
                team_players = []
                for link in team_page.select('a[href]'):
                    if link['href'].startswith('http://www.msc.com.pl/cezar/?mycl=1&p=21&r='):
                        team_players.append(int(link['href'].replace('http://www.msc.com.pl/cezar/?mycl=1&p=21&r=', '')))
                teams[team_id] = (team_name, team_players)
    return teams


def check_team_names():
    cezar_names = get_cezar_names()
    html_names = get_html_names()
    for team_id, team_name in cezar_names.items():
        if html_names[team_id][0] != cezar_names[team_id][0]:
            print('\tTeam name differs: expected "%s", got "%s"' % (cezar_names[team_id][0], html_names[team_id][0]))
        if not html_names[team_id][1]:
            print('\tNo player links found for team "%s"' % (html_names[team_id][0]))
        else:
            for pid in html_names[team_id][1]:
                if pid not in cezar_names[team_id][1]:
                    print('\tPlayer #%d not a member of team "%s"' % (pid, html_names[team_id][0]))



print('Checking %s' % (sys.argv[1]))
for check in sys.argv[2:]:
    print(' ' + check)
    globals()['check_' + check]()