def get(self):
wikipedia.set_lang(u'ru')
try:
page = wikipedia.page(u'Проект:Города_России/Списки_улиц/Казани')
streets = []
for link in page.links:
nlink = unicode(link).encode('utf-8').strip().decode('utf-8')
norm_name = normalize(nlink)
try:
street_info = StreetInfo.get_by_norm_name(norm_name)
if not street_info:
street_info = StreetInfo()
street_page = wikipedia.page(nlink)
street_info.name = nlink
street_info.norm_name = norm_name
street_info.info = unicode(street_page.summary).encode('utf-8').strip()
street_info.images = [Image(url=x) for x in street_page.images]
street_info.city = u'Казань'.encode('utf-8').strip()
street_info.lang=u'ru'.encode('utf-8').strip()
street_info.put()
except Exception, e:
print nlink.encode('utf-8')
except DeadlineExceededError:
pass
self.response.headers['Content-Type'] = "text/html; charset=utf-8"
self.response.write(json.dumps({'success':True}))
def _search_street_infos(street, city, lang=u'ru'):
street_infos = StreetInfo.query()
for info in street_infos:
if normalize(street) in normalize(info.name):
return info
return None