def start_crawler(self):
for i in range(1, 346):
url = self.start_url + "cp" + str(i)
res = requests.get(url, headers=self.headers)
html = etree.HTML(res.text)
comm_info_list = html.xpath("//li//div[@class='fang-info ml20 z']")
for comm_info in comm_info_list:
comm_url = comm_info.xpath("./div[@class='title']/a/@href")[0]
region = comm_info.xpath(".//a[@class='ml20']/text()")[0]
bu_id = re.search('\d+', comm_url).group(0)
data = {
"buildingId": bu_id,
'pageIndex': 1,
'pageSize': 500,
}
while True:
try:
deal_res = requests.post(
'http://hangzhou.fangtoo.com/Building/GetTradeExchange/',
data=data,
headers=self.headers)
deal_dict = json.loads(deal_res.text)
break
except:
continue
for n in deal_dict['data']:
try:
co = Comm(source)
co.city = '杭州'
size = n['Area']
area = size.replace('㎡', '')
if area:
area = float(area)
co.area = round(area, 2)
co.district_name = n['ExName']
co.total_price = int(
re.search('(\d+)', n['Price'],
re.S | re.M).group(1))
trade_date = n['ExDate']
if trade_date:
t = time.strptime(trade_date, "%Y/%m/%d %H:%M:%S")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
co.trade_date = datetime.datetime(y, m, d)
co.region = region
co.insert_db()
except Exception as e:
log.error("解析错误{}".format(e))
def get_comm_info(self, page_url):
response = requests.get(page_url, headers=self.headers)
html = response.text
comm_info_html_list = re.findall('<div class="avail_conr">.*?</li>',
html, re.S | re.M)
for i in comm_info_html_list:
comm = Comm('上海酷房网')
comm.city = '上海'
comm.district_name = re.search('class="avail_cont".*?>(.*?)<', i,
re.S | re.M).group(1).strip()
comm.region = re.search('class="avail_district".*?<a.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
comm_id = re.search(
'class="avail_cont".*?href="/xiaoqu/(.*?)\.html"', i,
re.S | re.M).group(1)
self.get_comm_detail(comm_id, comm)
def get_comm_info(self, comm_list, all_page_url):
for i in comm_list:
try:
comm = Comm('中安房')
comm.city = '合肥'
comm.district_name = re.search('zaf-nowrap.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
trade_date = re.search('zaf-fblue">(.*?)<', i,
re.S | re.M).group(1).strip()
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
comm.trade_date = datetime.datetime(y, m, d)
total_price = re.search('list-right-data.*?<span.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
comm.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
info = re.search('list-details-area.*?<span.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
area = info.split(' ')[0].replace('㎡', '')
if area:
area = float(area)
comm.area = round(area, 2)
try:
room_type = info.split(' ')[1]
except Exception as e:
room_type = None
try:
comm.room = int(
re.search('(\d)室', room_type, re.S | re.M).group(1))
except Exception as e:
comm.room = 0
try:
comm.hall = int(
re.search('(\d)厅', room_type, re.S | re.M).group(1))
except Exception as e:
comm.hall = None
try:
comm.toilet = int(
re.search('(\d)卫', room_type, re.S | re.M).group(1))
except Exception as e:
comm.toilet = None
try:
avg_price = info.split(' ')[2]
comm.avg_price = int(
re.search('(\d+)', avg_price, re.S | re.M).group(1))
except Exception as e:
comm.avg_price = None
info_2 = re.search('list-details-area.*?<span.*?<span>(.*?)<',
i, re.S | re.M).group(1).strip()
comm.direction = info_2.split(' ')[0]
try:
comm.fitment = info_2.split(' ')[1]
except Exception as e:
comm.fitment = None
info_3 = re.search('list-details-address1.*?<span>(.*?)<', i,
re.S | re.M).group(1).strip()
comm.region = info_3.split(' ')[0].strip()
comm.insert_db()
except Exception as e:
log.error('解析错误，source={},url="{}",e="{}"'.format(
'中安房', all_page_url, e))
def comm_info(self, comm_url_list, city_url):
for comm_url in comm_url_list:
url = city_url.replace('/esf/', comm_url)
re_url = url.replace('xq', 'fangjia')
res = requests.get(re_url, headers=self.headers)
con = res.text
co_name = re.search('wrap-head-name">(.*?)</div', con,
re.S | re.M).group(1)
co_name = co_name.strip()
try:
page = re.search('(\d+)">尾页', con).group(1)
except:
page = 1
for i in range(1, int(page) + 1):
page_url = re_url.rstrip('.html') + "/?n=" + str(i)
co_res = requests.get(page_url, headers=self.headers)
co_con = co_res.text
co_html = etree.HTML(co_con)
city = co_html.xpath("//span[@class='change-city']/text()")[0]
romm_info_list = co_html.xpath("//div[@class='list-cont']/div")
for room_info in romm_info_list:
try:
room = Comm(source)
room.city = city
room.district_name = co_name
floor = room_info.xpath(
".//div[@class='text']/p[2]/span[1]/text()")[0]
room.floor = int(re.search('\d+', floor).group(0))
trade_date = room_info.xpath(
".//span[@class='cj-data-num']/text()")[0]
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
room.trade_date = datetime.datetime(y, m, d)
total_price = room_info.xpath(
".//span[@class='cj-data-num c4a4a4a']/em/text()"
)[0]
room.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
avg_price = room_info.xpath(
".//span[@class='cj-data-num']/em/text()")[0]
room.avg_price = int(
re.search('(\d+)', avg_price,
re.S | re.M).group(1))
room.direction = room_info.xpath(
".//div[@class='text']/p[2]/span[2]/text()")[0]
area = room_info.xpath(".//p[1]/text()")[1]
room.region = area
size = re.search('建筑面积(.*?)平', area).group(1)
if size:
area = float(size)
room.area = round(area, 2)
room.insert_db()
except Exception as e:
log.error("{}解析房屋错误{}".format(page_url, e))
def deal_price(self):
for data in collection.find(no_cursor_timeout=True):
if 'fj_flag' in data:
if data['fj_flag'] == 1:
second_price = Comm(self.source)
second_price.city = data['fj_city']
second_price.direction = data['CJ_CX']
second_price.avg_price = float(data['CJ_CJDJ'])
second_price.area = float(data['CJ_JZMJ'])
second_price.trade_date = data['CJ_CJRQ']
second_price.total_price = float(data['CJ_CJDJ']) * float(
data['CJ_JZMJ'])
second_price.district_name = data['fj_name']
if 'CJ_ZH' in data:
second_price.house_num = data['CJ_ZH']
if 'CJ_SHBW' in data:
second_price.room_num = data['CJ_SHBW']
try:
second_price.floor = int(data['CJ_CS'])
except Exception as e:
print('楼层error', e)
second_price.region = data['fj_region']
is_success = second_price.insert_db()
def room(self, co_list, city_name):
for co in co_list:
try:
co_name = co.xpath("./div[1]/a/text()")[0]
co_url = "http:" + co.xpath("./div[1]/a/@href")[0]
region = co.xpath("./div[3]/span[1]/a[1]/text()")[0]
addr = co.xpath("./div[3]/span[3]/@title")[0]
detail = requests.get(co_url, headers=self.headers)
html = etree.HTML(detail.text)
room_url = "http:" + html.xpath("//div[@class='tab-toolbar pr']//li/a/@href")[-1]
page_index = requests.get(room_url, headers=self.headers)
except:
continue
if re.search('共(\d+)页', page_index.text):
page_num = re.search('共(\d+)页', page_index.text).group(1)
else:
log.info('小区无相关数据')
continue
for i in range(1, int(page_num) + 1):
url = re.sub('#.*', 'n', room_url) + str(i)
while True:
try:
res = requests.get(url, headers=self.headers)
break
except:
continue
con = res.text
room_html = etree.HTML(con)
room_list = room_html.xpath("//div[@class='right-information']")
for m in room_list:
try:
room = Comm(source)
room.district_name = co_name
room.city = city_name
room.region = region
room_type = m.xpath("./h3/span[2]/text()")[0]
try:
room.room = int(re.search('(\d)室', room_type, re.S | re.M).group(1))
except Exception as e:
room.room = None
try:
room.hall = int(re.search('(\d)厅', room_type, re.S | re.M).group(1))
except Exception as e:
room.hall = None
size = m.xpath("./h3/span[3]/text()")[0]
area = size.replace('平米', '')
if area:
area = float(area)
room.area = round(area, 2)
total_price = m.xpath(".//div[@class='price fs14 ']/em/text()")[0]
room.total_price = int(re.search('(\d+)', total_price, re.S | re.M).group(1))*10000
avg_price = m.xpath(".//div[@class='size fs14']/text()")[0]
room.avg_price = int(re.search('(\d+)', avg_price, re.S | re.M).group(1))
try:
room.fitment = m.xpath(".//div[@class='t1 fs14']/text()[3]")[0]
room.direction = m.xpath(".//div[@class='t1 fs14']/text()[2]")[0]
# room.use = m.xpath(".//div[@class='t1 fs14']/text()[1]")[0]
except:
room.fitment = None
room.direction = None
# room.use = None
floor_info = m.xpath(".//div[@class='fs14']/text()[1]")[0]
try:
floor = re.search('(.*?)/', floor_info).group(1)
room.floor = int(re.search('\d+',floor).group(0))
except Exception as e:
room.floor = None
try:
room.height = int(re.search('.*?/(\d+)层', floor_info).group(1))
except:
room.height = None
trade_date = m.xpath(".//div[@class='date']/text()")[0]
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
room.trade_date = datetime.datetime(y, m, d)
room.insert_db()
except Exception as e:
log.error('房屋信息提取失败{}'.format(e))
def start_crawler(self):
page = self.get_all_page()
for i in range(1, page):
url = 'http://www.taiwu.com/building/cp' + str(i) + '/'
while True:
try:
res = requests.get(url)
if res.status_code == 200:
break
except Exception as e:
print('请求出错', e)
# print(res.content.decode())
all_info = re.search('<ul class="fang-list">.*?</ul>',
res.content.decode(), re.S | re.M).group(0)
for k in re.findall('<li>.*?</li>', all_info, re.S | re.M):
source = '太屋网'
city = '上海'
area = re.search(
'<div class="adds">.*?<a href="/building/.*?/">(.*?)</a>',
k, re.S | re.M).group(1) # 区域
building_id = re.search('<a href="/building/(.*?)/', k,
re.S | re.M).group(1)
detail_url = "http://www.taiwu.com/Building/GetHouseExchange/"
payload = "------WebKitFormBoundary7MA4YWxkTrZu0gW\r\nContent-Disposition: form-data; name=\"buildingId\"\r\n\r\n" + building_id + "\r\n------WebKitFormBoundary7MA4YWxkTrZu0gW\r\nContent-Disposition: form-data; name=\"pageIndex\"\r\n\r\n1\r\n------WebKitFormBoundary7MA4YWxkTrZu0gW\r\nContent-Disposition: form-data; name=\"pageSize\"\r\n\r\n5000\r\n------WebKitFormBoundary7MA4YWxkTrZu0gW--"
headers = {
'content-type':
"multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW",
'Cache-Control': "no-cache",
}
while True:
try:
response = requests.request("POST",
detail_url,
data=payload,
headers=headers)
if res.status_code == 200:
break
except Exception as e:
print('请求出错', e)
try:
result_json = response.json()
data_list = result_json['data']
except Exception as e:
print(e)
continue
for j in data_list:
c = Comm(source)
c.city = city
c.region = area
c.room = j['RoomCount']
c.hall = j['HollCount']
c.district_name = j['BuildingName']
c.area = j['BldArea']
trade_date = j['ExDate']
trade_date_ = int(re.search('(\d+)', trade_date).group(1))
if trade_date_:
t = time.localtime(int(trade_date_ / 1000))
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
c.trade_date = datetime.datetime(y, m, d)
c.total_price = j['ExPrice']
c.insert_db()
def new_deal_price(self):
for i in collection_new.find(no_cursor_timeout=True):
print(collection_new.database.client.address[0])
if 'fj_flag' in i:
if i['fj_flag'] == 1:
deal_price = Comm(self.new_source)
deal_price.city = i['fj_city']
deal_price.region = i['fj_region']
deal_price.district_name = i['fj_name']
deal_price.avg_price = float(i["CJDJ"])
# deal_price.total_price = float(i["CJJE"]) * 10000
deal_price.trade_date = i['CJRQ']
deal_price.area = float(i['JZMJ'])
deal_price.room_num = i['SH']
deal_price.total_price = float(i['JZMJ']) * float(
i["CJDJ"])
try:
room = re.search('(.)室', i['FX'], re.S | re.M).group(1)
deal_price.room = check_room(room)
except Exception as e:
print('找不到室,FX={}, e={}'.format(i['FX'], e))
try:
hall = re.search('(.)厅', i['FX'], re.S | re.M).group(1)
deal_price.hall = check_room(hall)
except Exception as e:
print('找不到厅,FX={}, e={}'.format(i['FX'], e))
is_success = deal_price.insert_db()
def crawler(self, city_url, city):
res = requests.get(city_url, headers=self.headers)
con = etree.HTML(res.text)
last_page = con.xpath("//a[@class='down_page']/@href")[1]
page_num = re.search('\d+', last_page).group(0)
for i in range(1, int(page_num) + 1):
page_url = city_url + "/PG" + str(i)
page_res = requests.get(page_url, headers=self.headers)
page_con = etree.HTML(page_res.text)
temp = page_con.xpath("//h1/a/@href")
for temp_url in temp:
try:
com = Comm(source)
comm_url = city + temp_url
while True:
try:
co_res = requests.get(comm_url,
headers=self.headers,
timeout=10)
break
except:
continue
time.sleep(2)
co_con = etree.HTML(co_res.text)
com.city = co_con.xpath("//div/a[@class='show']/text()")[0]
region = co_con.xpath("//section/p/a/text()")[-1]
com.region = region
com.district_name = co_con.xpath("//cite/span/text()")[0]
info = co_con.xpath("//table/tbody/tr")
for tag in info:
size = tag.xpath("./td[2]/text()")[0]
area = size.replace('㎡', '')
area = float(area)
com.area = round(area, 2)
avg_price = tag.xpath("./td[3]/text()")[0]
com.avg_price = int(
re.search('(\d+)', avg_price,
re.S | re.M).group(1))
total_price = tag.xpath("./td/span/text()")[0]
com.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
trade_date = tag.xpath("./td/text()")[-2]
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
com.trade_date = datetime.datetime(y, m, d)
room_type = tag.xpath("./td//p/a/text()")[0]
try:
com.room = int(
re.search('(\d)室', room_type,
re.S | re.M).group(1))
except Exception as e:
com.room = None
try:
com.hall = int(
re.search('(\d)厅', room_type,
re.S | re.M).group(1))
except Exception as e:
com.hall = None
floor = tag.xpath("./td//p/span/text()")[0]
com.floor = int(re.search('(\d+)层', floor).group(1))
com.direction = re.search('层 (.*?)', floor).group(1)
com.insert_db()
except Exception as e:
log.error("{}小区信息提取错误".format(comm_url))
def comm_info(self, comm_url_list, city_url):
for comm_url in comm_url_list:
url = city_url.replace('/esf/', comm_url)
re_url = url.replace('xq', 'fangjia')
try:
res = requests.get(url=re_url,
headers=self.headers,
proxies=next(p))
except Exception as e:
log.error('请求失败, source={}, url={}, e={}'.format(
'乐有家', re_url, e))
continue
con = res.text
co_name = re.search('wrap-head-name">(.*?)</div', con,
re.S | re.M).group(1)
co_name = co_name.strip()
try:
page = re.search('(\d+)">尾页', con).group(1)
except:
page = 1
for i in range(1, int(page) + 1):
page_url = re_url.rstrip('.html') + "/?n=" + str(i)
print(page_url)
try:
co_res = requests.get(url=page_url,
headers=self.headers,
proxies=next(p))
except Exception as e:
log.error('请求失败, source={}, url={}, e={}'.format(
'乐有家', page_url, e))
continue
co_html = etree.HTML(co_res.text)
city = co_html.xpath(
"//span[@class='change-city']/text()")[0].replace(
'\t', '').replace('[', '')
romm_info_list = co_html.xpath("//div[@class='list-cont']/div")
for room_info in romm_info_list:
room = Comm(source)
# 城市
room.city = city
# 小区名称
room.district_name = co_name
try:
# 所在楼层
floor = room_info.xpath(
".//div[@class='text']/p[2]/span[1]/text()")[0]
floor = re.search('(.*?)/', floor).group(1)
room.floor = int(re.search('\d+', floor).group(0))
except:
room.floor = None
try:
# 总楼层
height = room_info.xpath(
".//div[@class='text']/p[2]/span[1]/text()")[0]
room.height = int(
re.search('/(\d+)层', height).group(1))
except:
room.height = None
try:
# 交易时间
trade_date = room_info.xpath(
".//span[@class='cj-data-num']/text()")[0]
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
room.trade_date = datetime.datetime(y, m, d)
except:
room.trade_date = None
try:
# 总价
total_price = room_info.xpath(
".//span[@class='cj-data-num c4a4a4a']/em/text()"
)[0]
if '*' in total_price:
log.error('source={}, 总价有问题 带*号'.format('乐有家'))
continue
else:
room.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
except:
room.total_price = None
try:
# 均价
avg_price = room_info.xpath(
".//span[@class='cj-data-num']/em/text()")[0]
if '*' in avg_price:
log.error('source={}, 均价有问题 带*号'.format('乐有家'))
continue
else:
room.avg_price = int(
re.search('(\d+)', avg_price,
re.S | re.M).group(1))
except:
room.avg_price = None
try:
# 朝向
room.direction = room_info.xpath(
".//div[@class='text']/p[2]/span[2]/text()"
)[0].replace('朝', '')
except:
room.direction = None
try:
region_area_info = room_info.xpath(
"./div[@class='text']/p[1]/text()")[1]
except:
return
try:
# 区域
room.region = region_area_info.split(' ')[1]
except:
room.region = None
try:
# 面积
size = re.search('建筑面积(.*?)平',
region_area_info).group(1)
if size:
area = float(size)
room.area = round(area, 2)
except:
room.area = None
room.insert_db()
def get_page_url(self, page_url, city, area_):
response = requests.get(page_url,
headers=self.headers,
proxies=self.proxy)
html = response.text
comm_html_list = re.findall('<li class=" clearfix">.*?</li>', html,
re.S | re.M)
for i in comm_html_list:
try:
comm = Comm('Q房网')
comm.city = city.strip()
comm.region = area_.strip()
comm.district_name = re.search('house-title">.*?<a.*?>(.*?)<',
i,
re.S | re.M).group(1).strip()
comm.direction = re.search(
'class="house-about clearfix".*?showKeyword">(.*?)<', i,
re.S | re.M).group(1).strip()
try:
comm.height = int(
re.search(
'class="house-about clearfix".*?showKeyword">.*?<span.*?<span>.*?/(.*?)<',
i, re.S | re.M).group(1).strip())
except Exception as e:
comm.height = None
total_price = re.search('class="show-price".*?span.*?>(.*?)<',
i, re.S | re.M).group(1).strip()
comm.total_price = int(total_price) * 10000
avg_price = re.search('class="show-price".*?<p.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
comm.avg_price = int(re.search('(\d+)', avg_price).group(1))
trade_date = re.search(
'class="show-price concluded".*?span.*?>(.*?)<', i,
re.S | re.M).group(1).strip()
if trade_date:
t = time.strptime(trade_date, "%Y.%m.%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
comm.trade_date = datetime.datetime(y, m, d)
room_type = re.search('house-title">.*?<a.*?>.*? (.*?) ', i,
re.S | re.M).group(1).strip()
try:
comm.room = int(
re.search('(\d)室', room_type, re.S | re.M).group(1))
except Exception as e:
comm.room = None
try:
comm.hall = int(
re.search('(\d)厅', room_type, re.S | re.M).group(1))
except Exception as e:
comm.hall = None
area = re.search('house-title">.*?<a.*?>.*? .*? (.*?平米)', i,
re.S | re.M).group(1).strip()
area = area.replace('㎡', '').replace('平米', '')
if area:
area = float(area)
comm.area = round(area, 2)
comm.insert_db()
except Exception as e:
log.error('解析错误，source="{}",html="{}",e="{}"'.format(
'Q房网', i, e))
def get_city_info(self, city_dict):
for city in city_dict:
city_url = city_dict[city] + 'chengjiao/'
try:
response = requests.get(city_url, headers=self.headers)
html = response.text
area_html = re.search('data-role="ershoufang".*?地铁', html,
re.S | re.M).group()
area_list_str = re.findall('<a.*?</a>', area_html, re.S | re.M)
for area_i in area_list_str:
if 'ershoufang' in area_i:
continue
area_url = re.search('href="(.*?)"', area_i,
re.S | re.M).group(1)
area = re.search('<a.*?>(.*?)<', area_i,
re.S | re.M).group(1)
for i in range(1, 101):
city_url_ = city_url.replace(
'/chengjiao/', '') + area_url + 'pg' + str(i)
try:
result = requests.get(city_url_,
headers=self.headers)
content = result.text
comm_str_list = re.findall(
'class="info".*?</div></div></li>', content,
re.S | re.M)
for i in comm_str_list:
comm = Comm('链家在线')
comm.region = area.strip()
comm.city = city.strip()
comm.district_name = re.search(
'target="_blank">(.*?)<', i,
re.S | re.M).group(1).strip()
comm.direction = re.search(
'class="houseIcon"></span>(.*?) \|', i,
re.S | re.M).group(1).strip()
try:
comm.fitment = re.search(
'class="houseIcon"></span>.*? \|(.*?)\| ',
i, re.S | re.M).group(1).strip()
except Exception as e:
comm.fitment = None
try:
height = re.search(
'class="positionIcon"></span>.*?\((.*?)\)',
i, re.S | re.M).group(1).strip()
comm.height = int(
re.search('(\d+)', height,
re.S | re.M).group(1))
except Exception as e:
comm.height = None
total_price = re.search(
"class='number'>(.*?)<", i,
re.S | re.M).group(1).strip()
if "*" in total_price:
continue
comm.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
room_type = re.search(
'arget="_blank">.*? (.*?) ', i,
re.S | re.M).group(1).strip()
try:
comm.room = int(
re.search('(\d)室', room_type,
re.S | re.M).group(1))
except Exception as e:
comm.room = 0
try:
comm.hall = int(
re.search('(\d)厅', room_type,
re.S | re.M).group(1))
except Exception as e:
comm.hall = None
area_ = re.search(
'target="_blank">.*? .*? (.*?平米)', i,
re.S | re.M).group(1).strip()
if area_:
area_ = area_.replace('㎡', '').replace(
'平米', '')
try:
area_ = float(area_)
comm.area = round(area_, 2)
except Exception as e:
comm.area = None
trade_date = re.search(
'dealDate">(.*?)<', i,
re.S | re.M).group(1).strip()
if trade_date:
t = time.strptime(trade_date, "%Y.%m.%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
comm.trade_date = datetime.datetime(
y, m, d)
try:
comm.avg_price = int(i['total_price'] /
i['area'])
except Exception as e:
comm.avg_price = None
comm.insert_db()
except Exception as e:
log.error(
'解析错误，source="{}",html="{}",e="{}"'.format(
'链家在线', html, e))
except Exception as e:
log.error('请求错误，source="{}",url="{}",e="{}"'.format(
'链家在线', city_url, e))
def into_mongo(coll):
com = Comm('澜斯')
results = coll.find(no_cursor_timeout=True)
for result in results:
# 这个地方写一个try是因为我再测试的时候发现有的木有fj_city
try:
com.city = result['fj_city'] # 城市
com.region = result['fj_region'] # 区域
except Exception as e:
log.error('城市或者区域没有')
com.m_date = result['updatedate'] # 更新日期
com.create_date = datetime.datetime.now() # 创建时间
com.fitment = result['newdiskdecoration'] # 装修
com.floor = result['flevel'] # 所在楼层
# try是因为在插入数据库中这几个如果不符合，就不会插入
try:
com.district_name = result['fj_name'] # 小区名称
com.avg_price = result['unitprice'] # 单价
com.total_price = result['usd'] # 总价
com.area = result['acreage'] # 面积=建筑面积
t = time.strptime(result['signingdate'].split('T')[0], "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
com.trade_date = datetime.datetime(y, m, d)
except Exception as e:
log.error(e)
# 这一部分我写了正则从地址中匹配单元号和室号，如果组长感觉不对，，直接注释掉就好
houseaddress = result['houseaddress']
try:
res = re.search('(\d+)号(\d+)', houseaddress)
com.unit_num = res.group(1) # 单元号
com.room_num = res.group(2) # 室号
except Exception as e:
print('无法匹配大盘单元号和室号，houseaddress={}'.find(houseaddress))
# 以下数据库确定无法匹配,写上是为了让您看看
# com.direction = None # 朝向
# com.room = None # 室数
# com.hall = None # 厅数
# com.toilet = None # 卫数
# com.height = None # 总楼层
# com.house_num = None # 楼栋号
# 执行插入操作
com.insert_db()
def city_info(self, index_url, city):
for i in range(1, 101):
index_url_ = index_url + 'i3' + str(i) + '/'
if i == 1:
index_url_ = index_url
try:
response = requests.get(index_url_, headers=self.headers)
html = response.text
try:
city_real = re.search('city = "(.*?)"', html, re.S | re.M).group(1)
if city != city_real:
break
house_num = re.search('class="org">(.*?)</b>', html, re.S | re.M).group(1)
if house_num == '0':
break
comm_info_paper_list = re.findall('class="info rel floatr".*?</dd>', html, re.S | re.M)
for comm_info_paper in comm_info_paper_list:
comm = Comm('房天下')
comm.city = city
comm.district_name = re.search('<a.*?>(.*?)<', comm_info_paper, re.S | re.M).group(1).strip()
if '�' in comm.district_name:
log.error('网页出现繁体字, url={}'.format(index_url_))
continue
comm.direction = re.search('class="mt18">(.*?)<', comm_info_paper, re.S | re.M).group(1)
try:
comm.height = int(re.search('共(.*?)层', comm_info_paper, re.S | re.M).group(1))
except Exception as e:
comm.height = None
comm.region = re.search('class="mt15">.*?<a.*?chengjiao.*?>(.*?)<', comm_info_paper,
re.S | re.M).group(
1)
total_price = re.search('class="price">(.*?)<', comm_info_paper, re.S | re.M).group(1)
if '*' in total_price:
continue
comm.total_price = int(total_price) * 10000
comm.room = int(re.search('(\d+)室', comm.district_name, re.S | re.M).group(1))
comm.hall = int(re.search('(\d+)厅', comm.district_name, re.S | re.M).group(1))
try:
comm.area = float(re.search('(\d+\.\d+)平米', comm.district_name, re.S | re.M).group(1))
except Exception as e:
comm.area = None
trade_date = re.search('class="time".*?>(.*?)<', comm_info_paper, re.S | re.M).group(1)
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
comm.trade_date = datetime.datetime(y, m, d)
try:
comm.avg_price = int(comm.total_price / comm.area)
except Exception as e:
comm.avg_price = None
comm.insert_db()
except Exception as e:
log.error('解析错误,source="{}",html="{}",e="{}"'.format('房天下', html, e))
except Exception as e:
log.error('请求错误,source="{}"，url="{}",e="{}"'.format('房天下', index_url_, e))
def get_comm_detail(self, comm_url, region, city):
comm = Comm('购房网')
comm.url = comm_url
comm.region = region.strip()
comm.city = city
try:
response = requests.get(url=comm_url,
headers=self.headers,
proxies=next(p))
except Exception as e:
log.error('请求错误,source="{}"，url="{}",e="{}"'.format(
'购房网', comm_url, e))
return
html = response.text
comm.district_name = re.search('title fl.*?<h1>(.*?)</h1>', html,
re.S | re.M).group(1).strip()
comm_info_html = re.search('<ul class="lscjlist">.*?</ul>', html,
re.S | re.M).group()
comm_info_list = re.findall('<li>(.*?)</li>', comm_info_html,
re.S | re.M)
if not comm_info_list:
log.info('source={}, 此小区没有数据，url="{}"'.format('购房网', comm_url))
for i in comm_info_list:
trade_date = re.search('<span>(.*?)</span>', i,
re.S | re.M).group(1).strip()
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
comm.trade_date = datetime.datetime(y, m, d)
room_type = re.search('<span>.*?<span>(.*?)</span>', i,
re.S | re.M).group(1).strip()
try:
comm.room = int(
re.search('(\d)室', room_type, re.S | re.M).group(1))
comm.hall = int(
re.search('(\d)厅', room_type, re.S | re.M).group(1))
except Exception as e:
comm.room = None
comm.hall = None
area = re.search('<span>.*?<span>.*?<span>(.*?)</span>', i,
re.S | re.M).group(1).strip().replace('㎡',
'').replace(
'平', '')
if area:
area = float(area)
comm.area = round(area, 2)
try:
height = re.search(
'<span>.*?<span>.*?<span>.*?<span>.*?/(.*?)</span>', i,
re.S | re.M).group(1).strip()
comm.height = int(re.search('(\d+)', height).group(1))
except Exception as e:
comm.height = None
comm.fitment = re.search(
'<span>.*?<span>.*?<span>.*?<span>.*?<span>(.*?)</span>', i,
re.S | re.M).group(1).strip()
comm.direction = re.search(
'<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span>(.*?)</span>',
i, re.S | re.M).group(1).strip()
avg_price = re.search(
'<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span>(.*?)</span>',
i, re.S | re.M).group(1)
comm.avg_price = int(
re.search('(\d+)', avg_price, re.S | re.M).group(1))
total_price = re.search(
'<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span>.*?<span.*?>(.*?)</span>',
i, re.S | re.M).group(1)
comm.total_price = int(
re.search('(\d+)', total_price, re.S | re.M).group(1)) * 10000
comm.insert_db()
def info_parse(self,ro_html,co_name,region,city_name):
room_list = ro_html.xpath("//ul[@class='pList zu']/li")
for room in room_list:
try:
ro = Comm(source)
ro.city = city_name
ro.district_name = co_name
ro.region = region
room_type = room.xpath(".//p[@class='sTit']/strong/text()")[0]
try:
ro.room = int(re.search('(\d)室', room_type, re.S | re.M).group(1))
except Exception as e:
ro.room = None
try:
ro.hall = int(re.search('(\d)厅', room_type, re.S | re.M).group(1))
except Exception as e:
ro.hall = None
total_price = room.xpath(".//div[@class='jiage']/strong/text()")[0]
ro.total_price = int(re.search('(\d+)', total_price, re.S | re.M).group(1))*10000
avg_price = room.xpath(".//div[@class='jiage']/p/text()")[0]
ro.avg_price = int(re.search('(\d+)', avg_price, re.S | re.M).group(1))
info = room.xpath(".//div/p[2]/text()")[0]
area = re.search('·(.*?)平米',info).group(1)
area = float(area)
ro.area = round(area, 2)
direction = re.search('平米 · (.*)',info).group(1)
ro.direction = direction.strip()
trade_date = room.xpath(".//div/p[3]/text()")[0]
trade_date = trade_date.strip()
if trade_date:
t = time.strptime(trade_date, "成交日期：%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
ro.trade_date = datetime.datetime(y, m, d)
ro.insert_db()
except Exception as e:
log.error("房间信息提取错误{}".format(e))
def comm_detail(self, comm_url_list, city):
for comm_url in comm_url_list[1:]:
try:
com_url = city.replace('/xiaoqu/', comm_url)
statecode = re.search('xq-(.*)', comm_url).group(1)
code = statecode.upper()
comm_detail_url = 'http://sh.centanet.com/apipost/GetDealRecord?estateCode=' + code + '&posttype=S&pageindex=1&pagesize=10000'
com_res = requests.get(com_url, headers=self.headers)
res = requests.get(comm_detail_url, headers=self.headers)
time.sleep(2)
html = etree.HTML(com_res.text)
data_dict = json.loads(res.text)
district_name = html.xpath("//div/h3/text()")[0]
city_name = html.xpath("//div[@class='idx-city']/text()")[0]
region = html.xpath("//a[@class='f000']/text()")[0]
for data in data_dict["result"]:
try:
co = Comm(source)
co.district_name = district_name.strip()
co.region = region
co.city = city_name
try:
room_type = data["houseType"]
co.room = int(
re.search('(\d)室', room_type,
re.S | re.M).group(1))
co.hall = int(
re.search('(\d)厅', room_type,
re.S | re.M).group(1))
except Exception as e:
log.error('roomtype为空'.format(e))
area = data['areaSize'].replace('平', '')
if area:
area = float(area)
co.area = round(area, 2)
co.direction = data['direction']
trade_date = '20' + data['dealTime']
if trade_date:
t = time.strptime(trade_date, "%Y-%m-%d")
y = t.tm_year
m = t.tm_mon
d = t.tm_mday
co.trade_date = datetime.datetime(y, m, d)
total_price = data['dealPrice']
co.total_price = int(
re.search('(\d+)', total_price,
re.S | re.M).group(1)) * 10000
avg_price = data['unitPrice']
try:
co.avg_price = int(
re.search('(\d+)', avg_price,
re.S | re.M).group(1))
except Exception as e:
co.avg_price = None
co.insert_db()
except Exception as e:
log.error('解析失败{}'.format(e))
except Exception as e:
log.error("小区成交信息错误{}".format(e))