def __init__(self):
self.tpp = TaoppDt()
self.nm = NuomiDt()
self.ti = TimeDt()
class Migration:
def __init__(self):
self.tpp = TaoppDt()
self.nm = NuomiDt()
self.ti = TimeDt()
# 以淘票票数据作为比对基础
# 不做处理，直接存入
def ini_tpp(self):
count = 0
for item in self.tpp.extract():
CinemaUrl.objects.create(city=item[1],
district=item[2],
location=item[5],
cinema_name=item[3],
taopp_url=item[4],
code=item[0])
# break
# 优化 district 字段
def _optimize_district(self):
query = CinemaUrl.objects.all()
for i in query:
position = i.district
if '市' in position:
district = position.split('市')[0] + '市'
elif '县' in position:
district = position.split('县')[0] + '县'
elif '区' in position:
district = position.split('区')[0] + '区'
elif '镇' in position:
district = position.split('区')[0] + '镇'
else:
district = position
if len(district) == 1:
district = i.city + district
i.district = district
i.save()
# 合并 糯米 table 到 django
def _add_nuomi(self):
count_1 = 0
count_2 = 0
for item in self.nm.extract():
city, district, location, cinema_name, nuomi_url = item[1], item[
2], item[5], item[3], item[4]
q_res = self._query_url(city, district, location, cinema_name)
if q_res:
# print(q_res)
q_res.nuomi_url = nuomi_url
q_res.save()
count_1 += 1
else:
CinemaUrl.objects.create(city=city,
district=district,
location=location,
cinema_name=cinema_name,
nuomi_url=nuomi_url,
code=0)
count_2 += 1
print(count_1, count_2)
# break
# 查询电影 url
def _query_url(self, city, district, location, cinema_name):
# 梯次比对
query = CinemaUrl.objects.filter(city__contains=city)
if not query:
return
# 梯次一： 直接比较电影院名
query_c_name = query.filter(cinema_name__contains=cinema_name)
if len(query_c_name) == 1:
# print('梯次一比较')
return query_c_name[0]
# 梯次二：模糊比较电影院名
rates = []
for idx in range(len(query)):
rate1 = Levenshtein.ratio(query[idx].cinema_name, cinema_name)
if rate1 > 0.7:
return query[idx]
rate2 = Levenshtein.ratio(query[idx].location, location)
if rate2 > 0.7:
return query[idx]
rates.append(rate1 * rate2)
if max(rates) < 0.2:
return None
i = rates.index(max(rates))
return query[i]
# 合并 时光网 数据
def _time(self):
count_1 = 1
count_2 = 2
for item in self.ti.extract():
city, district, location, cinema_name, time_url = item[1], item[
2], item[5], item[3], item[4]
# print(city, district, location, cinema_name, time_url)
q_res = self._query_url(city, district, location, cinema_name)
if q_res:
# print(q_res)
q_res.time_url = time_url
q_res.save()
count_1 += 1
else:
CinemaUrl.objects.create(city=city,
district=district,
location=location,
cinema_name=cinema_name,
time_url=time_url,
code=0)
count_2 += 1
print(count_1, count_2)
# break
# 如果在正式环境中执行
# 就赶紧辞职跑路吧
def _delete(self):
m = CinemaUrl.objects.all()
m.delete()
def _test(self):
query = CinemaUrl.objects.filter(city__contains='随州')
print(len(query))
def _show_all(self):
query = CinemaUrl.objects.all()
# .filter(time_url__startswith='http').filter(nuomi_url__startswith='http').filter(taopp_url__startswith='http')
# filter(city__contains='上海')
# .filter(time_url__startswith='http').filter(nuomi_url__startswith='http').filter(taopp_url__startswith='http')
print(len(query))
city_lst = []