“Python实现的链家网站的爬虫第四部分,最后一部分。”
本系列文将以链家南京站为例,使用Python实现链家二手房源信息的爬虫,将数据爬取,并存入数据库中,以便使用。
本系列第一部分为基础:
本系列第二部分为爬取小区信息:
本系列第三部分为爬取在售二手房信息:
本文是第四部分,爬取历史成交二手房信息并存入数据库,部分代码依赖于第一部分,同时依赖于第二部分的结果。
在前文中已经获取了小区信息,并存在了数据库中,直接读库遍历小区进行爬取:
def do_xiaoqu_chengjiao_spider(db_xq,db_cj):
"""
批量爬取小区成交记录
"""
count=0
xq_list=db_xq.fetchall()
for xq in xq_list:
xiaoqu_chengjiao_spider(db_cj,xq[0],xq[1])
count+=1
print ('have spidered %d xiaoqu %s' % (count,xq[0]))
print( 'done')
对某一个小区内的所有成交房源进行爬取,需要分页:
def xiaoqu_chengjiao_spider(db_cj, xq_url=u"https://nj.lianjia.com/xiaoqu/1411000000391/", xq_name=u"default"):
"""
爬取小区成交记录
"""
url = xq_url.replace('xiaoqu/','chengjiao/c');
try:
req = urllib.request.Request(url, headers=hds[random.randint(0, len(hds) - 1)])
source_code = urllib.request.urlopen(req, timeout=10).read()
plain_text = source_code.decode('utf-8');
soup = BeautifulSoup(plain_text,"html.parser")
except (urllib.request.HTTPError, urllib.request.URLError) as e:
print(e)
exception_write('xiaoqu_chengjiao_spider', xq_url)
return
except Exception as e:
print(e)
exception_write('xiaoqu_chengjiao_spider', xq_url)
return
content = soup.find('div', {'class': 'page-box house-lst-page-box'})
total_pages = 0
if content:
d = "d=" + content.get('page-data')
loc = {}
glb = {}
exec(d, glb, loc);
total_pages = loc['d']['totalPage']
print(u"xiaoqu %s chengjiao totalpage %d" % (xq_name,total_pages));
threads = []
for i in range(total_pages):
tmp= u'chengjiao/pg%dc'% (i + 1)
url_page = url.replace('chengjiao/c',tmp);
t = threading.Thread(target=chengjiao_spider, args=(db_cj, url_page))
threads.append(t)
for t in threads:
t.start()
for t in threads:
t.join()
爬取单个页面内的成交记录信息:
def chengjiao_spider(db_cj, url_page=u"https://nj.lianjia.com/chengjiao/pg4c1411000000142/"):
"""
爬取页面链接中的成交记录
"""
print(u"爬取页面%s成交记录" % url_page);
try:
req = urllib.request.Request(url_page, headers=hds[random.randint(0, len(hds) - 1)])
source_code = urllib.request.urlopen(req, timeout=10).read()
plain_text = source_code.decode('utf-8');
soup = BeautifulSoup(plain_text,"html.parser")
except (urllib.request.HTTPError, urllib.request.URLError) as e:
print(e)
exception_write('chengjiao_spider', url_page)
return
except Exception as e:
print(e)
exception_write('chengjiao_spider', url_page)
return
recodenum = 0;
cjs = soup.find('ul', {'class': 'listContent'});
cj_list = cjs.findAll('li', {})
for cj in cj_list:
info_dict = {}
title = cj.find('div', {'class': 'title'});
houseInfo = cj.find('div', {'class': 'houseInfo'});
dealDate = cj.find('div', {'class': 'dealDate'});
totalPrice = cj.find('div', {'class': 'totalPrice'});
positionInfo = cj.find('div', {'class': 'positionInfo'});
source = cj.find('div', {'class': 'source'});
unitPrice = cj.find('div', {'class': 'unitPrice'});
dealHouseInfo = cj.find('div', {'class': 'dealHouseInfo'});
dealCycleeInfo = cj.find('div', {'class': 'dealCycleeInfo'});
href = title.find('a')
if not href:
continue
info_dict.update({u'链接': href.attrs['href']})
content = title.text.split()
if content:
info_dict.update({u'小区名称': content[0]})
info_dict.update({u'户型': content[1]})
info_dict.update({u'面积': content[2]})
content = houseInfo.text.split('|') #unicode(cj.find('div', {'class': 'con'}).renderContents().strip())
if content:
info_dict.update({u'朝向': content[0].strip()})
if len(content) >= 2:
info_dict.update({u'装修': content[1].strip()})
if len(content) >= 3:
info_dict.update({u'电梯': content[2].strip()})
info_dict.update({u'签约时间': dealDate.text})
info_dict.update({u'签约总价': totalPrice.text}) #注意值
content = positionInfo.text.split()
if len(content) >= 2:
info_dict.update({u'楼层': content[0].strip()})
info_dict.update({u'年代楼型': content[1].strip()})
else:
info_dict.update({u'楼层': content[0].strip()})
info_dict.update({u'来源': source.text})
info_dict.update({u'签约单价': unitPrice.text}) #可能为*
#content = dealHouseInfo.text.split()
if dealHouseInfo != None:
for span in dealHouseInfo.find('span', {'class': 'dealHouseTxt'}).findAll('span'):
if span.text.find(u'房屋') != -1:
info_dict.update({u'税费': span.text}) # 满几年
elif span.text.find(u'距') != -1:
info_dict.update({u'地铁': span.text})
#content = dealCycleeInfo.text.split()
if dealCycleeInfo != None:
for span in dealCycleeInfo.find('span',{'class': 'dealCycleTxt'}).findAll('span'):
if span.text.find(u'挂牌') != -1:
info_dict.update({u'挂牌价': span.text})
elif span.text.find(u'成交周期') != -1:
info_dict.update({u'成交周期': span.text})
command = gen_chengjiao_insert_command(info_dict)
db_cj.execute(command, 1)
recodenum += 1;
print(u"爬取页面%s成交记录%d条" % (url_page,recodenum));
爬取的成交记录信息将被存储到数据库表中。
整个爬取过程还是很快的,在爬取完小区、在售、成交三类信息之后,就可以拿这些数据去进行想要做的分析了。
长按进行关注。