# -*- coding: utf-8 -*-

from scrapy.xpath import HtmlXPathSelector
from scrapy.link.extractors import RegexLinkExtractor
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib_exp import adaptors
from eventscraper.items import EventscraperItem
from scrapy import log
from eventscraper import util
from string import join as j
from datetime import datetime, timedelta
import re


DATE_REGEX = re.compile(r'(\d+)\.(%s)\.(\d){,2}.*?([0-1]\d|2[0-4]):([0-5]\d)hs' % j([m[:3] for m in util.MONTHS], '|'), re.IGNORECASE)

def _fecha_adaptor(value):
    m = DATE_REGEX.search(value)
    if m is None:
        raise Exception("Couldn't parse: %s" % value)
    g = m.groups()
    d = datetime(2000 + int(g[2]),
                 [m[:3] for m in util.MONTHS].index(g[1].lower()) + 1,
                 int(g[0]))
    if int(g[3]) == 24: # 24 no es valido como hora, sumamos 1 dia a la fecha y seteamos hora a 0
        d = d + timedelta(1)
        d = datetime(d.year, d.month, d.day, 0, int(g[4]))
    else:
        d = datetime(d.year, d.month, d.day, int(g[3]), int(g[4]))

    return d



class ProyectoUnderSpider(CrawlSpider):
    domain_name = 'proyectounder.com'
    start_urls = ['http://www.proyectounder.com/agenda.php?agenda=1']

    rules = (
        Rule(RegexLinkExtractor(allow=(r'/evento/\d+',)), 'parse_event', follow=True),
        Rule(RegexLinkExtractor(allow=(r'/agenda.php\?action=display',)))
    )

    adaptor_pipe = [adaptors.extract,  adaptors.delist(''), adaptors.strip]

    adaptor_map = {
        'summary': adaptor_pipe,
        'description': adaptor_pipe,
        'start_date': [adaptors.extract, adaptors.delist(''), _fecha_adaptor]
    }


    def parse_event(self, response):
        i = EventscraperItem()
        i.set_adaptors(self.adaptor_map)

        xs = HtmlXPathSelector(response)

        i.attribute('guid', response.url)
        i.attribute('url', response.url)
        i.attribute('summary', xs.x('//div[contains(@class, "eventoItem")]/h1/text()'))
        i.attribute('description', xs.x('//div[contains(@class, "eventoItem")]/p/text()'))
        i.attribute('start_date', xs.x('//div[contains(@class, "eventoItem")]/span[contains(@class, "fecha")][1]/text()'))

        venue = {
            'name': xs.x('//div[contains(@class, "agendaVermas")]//a[contains(@href, "agenda.php?lugar=")][1]/text()').extract()[0].strip(),
            'city': xs.x('//div[contains(@class, "agendaVermas")]//a[contains(@href, "agenda.php?ciudad=")][1]/text()').extract()[0].strip()
        }

        log.msg(venue)

        i.attribute('meta', {'source_site': 'proyectounder.com',
                             'scraped_on': datetime.now(),
                             'source_html': response.body })

        return [i]

SPIDER = ProyectoUnderSpider()