def desc_parser(self):
result = []
title = self.sel.xpath('b/text()').extract()
if title:
result.append(join_strings(title).lower().capitalize())
result.append(join_strings(self.sel.xpath('text()').extract(), ', '))
return [join_strings(result, ' ')]
def desc_parser(self):
result = []
title = self.sel.xpath('b/text()').extract()
if title:
result.append(join_strings(title).lower().capitalize())
result.append(join_strings(self.sel.xpath('text()').extract(), ', '))
return [join_strings(result, ' ')]
def prices(self):
if not self._prices:
price = join_strings(self.price_parser())
mesure = join_strings(self.mesure_parser())
price_str = ['%s %s' % (price, mesure) if price else '']
price_digit = [self.digit_price(price, mesure)]
self._prices = {'price': price_str, 'price_digit': price_digit}
return self._prices
def prices(self):
if not self._prices:
price = join_strings(self.price_parser())
mesure = join_strings(self.mesure_parser())
price_str = ['%s %s' % (price, mesure) if price else '']
price_digit = [self.digit_price(price, mesure)]
self._prices = {'price' : price_str, 'price_digit' : price_digit}
return self._prices
def process_section(self, section):
PATERN = u'Продаю'
txt = join_strings(section.xpath("text()").extract())
chains = [x.strip() for x in txt.split(u"→")]
result = {'sale': False, 'section': None}
if PATERN in chains:
result['sale'] = True
result['section'] = join_strings(chains, ' ')
return result
def process_section(self, section):
PATERN = u'Продаю'
txt = join_strings(section.xpath("text()").extract())
chains = [x.strip() for x in txt.split(u"→")]
result = {'sale':False, 'section':None}
if PATERN in chains:
result['sale'] = True
result['section'] = join_strings(chains, ' ')
return result
def process_detail_page(self, sel, response):
saler_type = join_strings(
sel.xpath('//div[@class="c_face"]/span[@class="lgrey"]/text()').
extract())
address = join_strings(
sel.xpath('//*[@id="list_sale"]/div[@class="card_block"][1]/text()'
).extract(), ', ')
name = join_strings(
sel.xpath('//div[@class="c_face"]/text()').extract())
phone_str = join_strings(
join_strings(
sel.xpath('//div[@class="c_phone"]/text()').extract()))
description = join_strings(
sel.xpath('//*[@id="list_sale"]/div[@class="card_block"]/text()').
extract(), '\n')
price = join_strings(
sel.xpath('//div[@class="card_price"]/text()').extract())
mesure = join_strings(
sel.xpath('//div[@class="card_price"]/span/text()').extract())
item = RealtyItem()
item['do_not_process'] = not self.check_saler_type(saler_type)
item['phone'] = self.filter_phone(phone_str)
item['name'] = [name]
item['desc'] = [description]
item['price'] = ['%s %s' % (price, mesure)]
item['price_digit'] = [self.get_digit_price(price, mesure)]
item['link'] = [response.url]
item['estate_type_id'] = self.estate_type_parser(sel)
item['region_id'] = self.region_parser(address)
item['locality_id'] = self.locality_parser(item['region_id'], address)
#item['microdistrict'] = ''
#item['street'] = ''
#item['estate_number'] = ''
item['room_count'] = self.room_count_parser(self.page_title(sel))
return item
def commerce_parser(self):
COMMERCE_CAT_ID = 6
full_txt = join_strings(self.sel.xpath('//div[@class="media-body"]/p[@class="text_justify"]/text()').extract())
txt = join_strings(full_txt.split()[:3], ' ')
print txt
key = 'commerce_mapper_smart'
from django.core.cache import cache
mapper = cache.get(key)
if not mapper:
types = EstateType.objects.filter(estate_type_category_id=COMMERCE_CAT_ID)
mapper = {}
for t in types:
mapper[ur'%s\s' % t.name] = t.id
mapper[ur'%s\s' % t.name_accs] = t.id
cache.set(key, mapper, 3600)
return self.re_mapper(mapper, txt) or self.ZDANIE
def desc_parser(self):
result = []
result.append(self.title())
result.append('\n')
result.append(
join_strings(
self.sel.xpath(
'//div[@itemprop="description"]//text()').extract(), ', '))
return result
def filter_phone(self):
phone = self.phone_parser()
if not phone:
return
phone_str = join_strings(phone, ',')
phones = re.split(r'[\,|\n]', phone_str)
result = []
for phone in phones:
phone = phone.strip().replace('+7', '8')
result.append(re.sub('\D', '', phone))
return result
def filter_phone(self):
phone = self.phone_parser()
if not phone:
return
phone_str = join_strings(phone,',')
phones = re.split(r'[\,|\n]', phone_str)
result = []
for phone in phones:
phone = phone.strip().replace('+7', '8')
result.append(re.sub('\D','', phone))
return result
def process_detail_page(self, sel, response):
saler_type = join_strings(sel.xpath('//div[@class="c_face"]/span[@class="lgrey"]/text()').extract())
address = join_strings(sel.xpath('//*[@id="list_sale"]/div[@class="card_block"][1]/text()').extract(), ', ')
name = join_strings(sel.xpath('//div[@class="c_face"]/text()').extract())
phone_str = join_strings(join_strings(sel.xpath('//div[@class="c_phone"]/text()').extract()))
description = join_strings(sel.xpath('//*[@id="list_sale"]/div[@class="card_block"]/text()').extract(), '\n')
price = join_strings(sel.xpath('//div[@class="card_price"]/text()').extract())
mesure = join_strings(sel.xpath('//div[@class="card_price"]/span/text()').extract())
item = RealtyItem()
item['do_not_process'] = not self.check_saler_type(saler_type)
item['phone'] = self.filter_phone(phone_str)
item['name'] = [name]
item['desc'] = [description]
item['price'] = ['%s %s' % (price, mesure)]
item['price_digit'] = [self.get_digit_price(price, mesure)]
item['link'] = [response.url]
item['estate_type_id'] = self.estate_type_parser(sel)
item['region_id'] = self.region_parser(address)
item['locality_id'] = self.locality_parser(item['region_id'], address)
#item['microdistrict'] = ''
#item['street'] = ''
#item['estate_number'] = ''
item['room_count'] = self.room_count_parser(self.page_title(sel))
return item
def desc_parser(self):
result = []
result.append(self.title())
result.append('\n')
result.append(join_strings(self.sel.xpath('//div[@class="media-body"]//text()').extract(), ', '))
return result
def room_count(self):
room_parser_result = self.room_count_parser()
if room_parser_result:
result = join_strings(room_parser_result)
return re.sub('\D','', result)
def title_parser(self):
title = join_strings(self.sel.xpath('b/text()').extract())
txt = join_strings(self.sel.xpath('text()').extract(), ' ')
txt = join_strings(txt.split()[0:3], ' ')
title = u'%s %s' % (title, txt)
return title
ur'офис': 35,
ur'склад': 53,
# 'kommercheskaya-nedvizhimost' : 93,
}
result = self.re_mapper(mapper, page_title)
if callable(result):
return result(sel)
return result or ZDANIE
def stead_parser(self, sel):
mapper = {
ur'индивидуальное жилищное строительство' : 15,
ur'сельскохозяйственного назначения': 42,
ur'коммерческое строительство': 20,
}
txt = join_strings(sel.xpath('//*[@id="list_sale"]/div[@class="card_block"]').extract())
print "stead parser %s" % txt
return self.re_mapper(mapper, txt)
def re_mapper(self, mapper, txt):
for key, value in mapper.iteritems():
matches = re.search(key, txt, re.I | re.U)
if matches:
return value
def get_digit_price(self, price, mesure):
mesures = {u'тыс. руб.':1000}
if price:
if mesure in mesures:
price_digit = float(price)
price_digit = int(price_digit * mesures[mesure])
def title(self):
if not self._title:
self._title = join_strings(self.title_parser())
return self._title
def room_count(self):
room_parser_result = self.room_count_parser()
if room_parser_result:
result = join_strings(room_parser_result)
return re.sub('\D', '', result)
def page_title(self, sel):
return join_strings(
sel.xpath('//*[@id="list_sale"]/h1/text()').extract())
def link(self):
import hashlib
id_instead_link = hashlib.md5(join_strings(self.sel.xpath('text()').extract()).encode('utf-8')).hexdigest()
return [id_instead_link]
def desc_parser(self):
result = []
result.append(self.title())
result.append('\n')
result.append(join_strings(self.sel.xpath('//div[@itemprop="description"]//text()').extract(), ', '))
return result
def page_title(self, sel):
return join_strings(sel.xpath('//*[@id="list_sale"]/h1/text()').extract())
def title_parser(self):
title = join_strings(self.sel.xpath('b/text()').extract())
txt = join_strings(self.sel.xpath('text()').extract(), ' ')
txt = join_strings(txt.split()[0:3],' ')
title = u'%s %s' % (title, txt)
return title
def link(self):
import hashlib
id_instead_link = hashlib.md5(
join_strings(self.sel.xpath('text()').extract()).encode(
'utf-8')).hexdigest()
return [id_instead_link]
ur'склад': 53,
# 'kommercheskaya-nedvizhimost' : 93,
}
result = self.re_mapper(mapper, page_title)
if callable(result):
return result(sel)
return result or ZDANIE
def stead_parser(self, sel):
mapper = {
ur'индивидуальное жилищное строительство': 15,
ur'сельскохозяйственного назначения': 42,
ur'коммерческое строительство': 20,
}
txt = join_strings(
sel.xpath(
'//*[@id="list_sale"]/div[@class="card_block"]').extract())
print "stead parser %s" % txt
return self.re_mapper(mapper, txt)
def re_mapper(self, mapper, txt):
for key, value in mapper.iteritems():
matches = re.search(key, txt, re.I | re.U)
if matches:
return value
def get_digit_price(self, price, mesure):
mesures = {u'тыс. руб.': 1000}
if price:
if mesure in mesures:
price_digit = float(price)
def locality_parser(self):
return join_strings(self.sel.re(ur'(?:п\.|ст\.|x\.|г\.)\s(\D+?)\,')) or u'Темрюк'
def title(self):
if not self._title:
self._title = join_strings(self.title_parser())
return self._title
