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 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 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 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 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 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))