def parse(self, response):
sel = Selector(response)
links = sel.xpath(SalonSelectors.LIST_SALONS).extract()
next_page = self.get_property(sel, SalonSelectors.NEXT_URL)
print u'links: %s, %s' % (len(links), response.url)
if SalonSelectors.is_first_page(sel):
total = SalonSelectors.get_list_total(sel)
if total > 999:
# yahoo search can not paginate beyond 1000 items
# so need to run crawler for smaller areas
self.log_message(u'Pagination overflow: %s' % response.url)
if links:
for link in links:
canonical = link.split('?')[0]
if SalonEs.check_by_url(canonical):
self.count_skip += 1
print u'%s: skipped: %s' % (self.count_skip, link)
continue
request = Request(link, callback=self.parse_salon, errback=self.parse_err)
request.meta['page_kind'] = 'salon'
yield request
if next_page:
request = Request(next_page, callback=self.parse, errback=self.parse_err)
request.meta['page_kind'] = 'list'
yield request
