def url_manage_proc(self, url_q, conn_q, root_url, page_num):
url_manager = UrlManager()
url_manager.add_new_url(root_url)
print('url_mannager is working...')
while True:
while url_manager.has_new_url():
# 从URL管理器获取新的URL
new_url = url_manager.get_new_url()
# 将新的URL发到工作节点
url_q.put(new_url)
# 加上判断， 爬满2000个链接终止爬虫并保存进度
if (url_manager.old_urls_size() > page_num):
# 通知爬虫节点结束工作
url_q.put('end')
print('控制节点发起结束通知！')
# 关闭节点同事存储状态
url_manager.save_process('new_urls.txt',
url_manager.new_urls)
url_manager.save_process('old_urls.txt',
url_manager.old_urls)
return
# 从result_solve_proc获取的URL添加到URL管理器
print('url control working..., solve result')
try:
if not conn_q.empty():
urls = conn_q.get()
url_manager.add_new_urls(urls)
except Exception as e:
time.sleep(1) # 延时休息
print('has crawl page num : ', url_manager.old_urls_size())
time.sleep(5)
class SpiderMan():
def __init__(self):
self.manager = UrlManager()
self.downloader = HtmlDownloader()
self.parser = HtmlParser()
self.output = DataOutput()
def crawl(self,root_url):
self.manager.add_new_url(root_url)
while (self.manager.has_new_url() and self.manager.old_url_size()<100):
try:
new_url =self.manager.get_new_url()
html = self.downloader.download(new_url)
new_urls,data = self.parser.parser(new_url,html)
self.manager.add_new_urls(new_urls)
self.output.store_data(data)
print("已经抓取%s个链接"%self.manager.old_url_size())
except Exception as e:
print('crawl failed')
self.output.output_html()
class Spiderman(object):
def __init__(self):
self.manage = UrlManager()
self.parser = HtmlParser()
self.downloader = Htmldownloader()
self.output = DataOutput()
def crawl(self,root_url):
self.manage.add_new_url(root_url)
print(len(self.manage.new_urls))
while(self.manage.has_new_url() and self.manage.old_url_size() < 100):
try:
new_url = self.manage.get_new_url()
html = self.downloader.download(new_url)
new_urls,data = self.parser.parser(new_url,html)
self.manage.add_new_urls(new_urls)
self.output.store_data(data=data)
print('已经抓取%s个链接' % self.manage.old_url_size())
except:
print('crawl Failed')
self.output.output_html()
def url_manager_proc(self, url_q, conn_q, root_url):
url_manager = UrlManager()
url_manager.add_new_url(root_url)
while True:
if url_manager.has_new_url():
new_url = url_manager.get_new_url()
url_q.put(new_url)
print('old_url=', url_manager.old_url_size())
if url_manager.old_url_size() > 2000:
url_q.put('end')
print('Manager notify ending!')
url_manager.save_progress('new_urls.txt',
url_manager.new_urls)
url_manager.save_progress('old_urls.txt',
url_manager.old_urls)
return
try:
if not conn_q.empty():
urls = conn_q.get()
url_manager.add_new_urls(urls)
except BaseException as e:
time.sleep(0.1)
class Anjuke(object):
def __init__(self, ):
self.count = 0
self.wcount = 0
self.mylock = Lock()
self.csvfile = file('sz.csv', 'a') #ks.csv
self.csvfile.write(codecs.BOM_UTF8)
self.item_queue = Queue()
self.headers = {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.96 Safari/537.36'
}
self.proxies = {
'http': 'https://121.61.0.33:9999',
"https": 'https://121.61.0.33:9999'
}
self.mysign = True
#https://suzhou.anjuke.com/community
self.rawurl = 'https://suzhou.anjuke.com/community/'
self.urlmanager = UrlManager()
def get_villages(self):
rep = requests.get(self.rawurl,
headers=self.headers,
verify=False,
timeout=2)
soup = BeautifulSoup(rep.text, 'lxml')
results = soup.find_all('span', attrs={'class': 'elems-l'})
items = results[0].find_all('a')
for item in items:
if item.get("title") == "全部小区":
print '剔除(全部)这一选项'
continue
vurl = item.get('href')
print vurl
self.get_villages2(vurl)
#self.get_villages3(vurl)
print '======================================'
self.urlmanager.save_urls_process_status(self.urlmanager.new_urls,
r'new_urls.txt')
self.urlmanager.save_urls_process_status(self.urlmanager.crawled_urls,
r'crawled_urls.txt')
def get_villages2(self, vurl):
rep = requests.get(vurl, headers=self.headers, verify=False, timeout=2)
soup = BeautifulSoup(rep.text, 'lxml')
results = soup.find_all('div', attrs={'class': 'sub-items'})
items = results[0].find_all('a')
for item in items:
if item.get("title") == "全部小区":
print '剔除(全部)这一选项2'
continue
url2 = item.get('href')
print url2
self.get_villages3(url2)
def get_villages3(self, url):
while 1:
rep = requests.get(url,
headers=self.headers,
verify=False,
timeout=2)
soup = BeautifulSoup(rep.text, 'lxml')
results = soup.find_all('div', attrs={'_soj': 'xqlb'})
for result in results:
item_url = result.get('link')
self.count += 1
print 'No.', self.count, ':', item_url
self.urlmanager.add_new_url(item_url)
next_item = soup.find('a', attrs={'class': 'aNxt'})
if next_item == None:
break
else:
url = next_item.get('href')
# time.sleep(1)
def get_detail(self, c_url):
rlist = []
# item ={}
try:
rep = requests.get(c_url,
headers=self.headers,
verify=False,
timeout=4)
except:
print 'current:urls num2:', self.urlmanager.new_urls_size()
self.urlmanager.readd_new_url(c_url)
return
if rep.url.startswith('https://www.anjuke.com/captcha-verify/'):
self.urlmanager.readd_new_url(c_url)
self.urlmanager.save_urls_process_status(self.urlmanager.new_urls,
r'new_urls.txt')
self.urlmanager.save_urls_process_status(
self.urlmanager.crawled_urls, r'crawled_urls.txt')
return
print rep.url
soup = BeautifulSoup(rep.text, 'lxml')
name = soup.find('h1')
addr = name.find('span')
print name.contents[0].strip(), addr.string
rlist.append(name.contents[0].strip().decode("utf-8"))
rlist.append(addr.string.decode("utf-8"))
village = re.search("(.*?)-.*?", addr.string).group(1)
try:
y, x = geocodeG(addr.string.replace(village, '苏州')) #昆山
except:
try:
y, x = geocodeG(name.contents[0].strip().decode("utf-8"))
except:
y, x = geocodeG(u'苏州' + village) #昆山
rlist.append(y)
rlist.append(x)
price = re.search('.*comm_midprice":"(.*?)"', rep.text)
if price == None:
price = u'暂无报价'
else:
price = price.group(1)
print 'price:', price
rlist.append(price)
# item['price'] = price.group(1)
result = soup.find('dl', attrs={"class": 'basic-parms-mod'})
for a in result.find_all('dd'):
# print a.string.strip().decode('utf-8')
rlist.append(a.string.strip().decode('utf-8'))
print '----'
id = re.search('view/(\d+)', rep.url)
rent_url = 'https://ks.anjuke.com/v3/ajax/communityext/?commid=' + str(
id.group(1)) + '&useflg=onlyForAjax'
print rent_url
response = requests.get(rent_url,
headers=self.headers,
verify=False,
timeout=2)
print response.text
content = json.loads(response.text)
print content.get('comm_propnum').get('rentNum'), content.get(
'comm_propnum').get('saleNum')
rlist.append(content.get('comm_propnum').get('rentNum'))
rlist.append(content.get('comm_propnum').get('saleNum'))
# self.item_queue.put(rlist)
return rlist
# item['property-type'] = value_list[0].string.strip().replace("：",'')
# item['property-cost'] = value_list[1].string.strip().replace("：",'')
# item['area'] = value_list[2].string.strip().replace("：",'')
# item['households'] = value_list[3].string.strip().replace("：",'')
# item['build-years'] = value_list[4].string.strip().replace("：",'')
# item['parking-nums'] = value_list[5].string.strip().replace("：",'')
# item['cap-rate'] = value_list[6].string.strip().replace("：",'')
# item['greeening-rate'] = value_list[7].string.strip().replace("：",'')
# item['developer'] = value_list[8].string.strip().replace("：",'')
# item['property-management'] = value_list[9].string.strip().replace("：",'')
# print j.string.strip().replace("：",'')
# for (k,v) in item.items():
# print "dict[%s]=" % k,v
def write_to_csv(self, item):
csv_write = unicodecsv.writer(self.csvfile,
encoding='utf-8-sig',
dialect='excel')
csv_write.writerow(item)
def write_to_csv2(self):
if not self.item_queue.empty():
self.mylock.acquire(10)
with open('ks.csv', 'a') as csvfile:
item = self.item_queue.get()
csv_write = unicodecsv.writer(csvfile,
encoding='utf-8-sig',
dialect='excel')
csv_write.writerows(item)
self.mylock.release()
def write_to_csv3(self):
while not self.item_queue.empty():
item = self.item_queue.get()
csv_write = unicodecsv.writer(self.csvfile,
encoding='utf-8-sig',
dialect='excel')
self.wcount += 1
print 'write No.', self.wcount, 'url'
csv_write.writerow(item)
def start2(self):
num = 0
self.get_villages()
print 'current:urls num1:', self.urlmanager.new_urls_size()
while self.urlmanager.has_new_url():
num += 1
new_url = self.urlmanager.get_new_url()
try:
print 'get No.', num, 'url'
url_process = Process(target=self.get_detail,
args=(new_url, self.item_queue))
url_process.start()
except:
with open("anjuke.log", 'w+') as f:
f.write('current:urls num2:')
f.write(str(self.urlmanager.new_urls_size()))
self.urlmanager.readd_new_url(new_url)
self.urlmanager.save_urls_process_status(
self.urlmanager.new_urls, r'new_urls.txt')
self.urlmanager.save_urls_process_status(
self.urlmanager.crawled_urls, r'crawled_urls.txt')
while not self.item_queue.empty():
print 'write No.', num, 'url'
self.write_to_csv(item)
# write_process = Process(target=self.write_to_csv2)
# write_process.start()
def start3(self):
# self.get_villages()
fo = open("anjuke.log", 'w+')
s = sys.stdout
sys.stdout = fo
print 'current:urls num1:', self.urlmanager.new_urls_size()
while self.urlmanager.has_new_url():
newlist = []
flag = 100
while flag:
if self.urlmanager.has_new_url():
new_url = self.urlmanager.get_new_url()
newlist.append(new_url)
flag -= 1
else:
break
pool = threadpool.ThreadPool(12)
requests = threadpool.makeRequests(self.get_detail, newlist)
[pool.putRequest(req) for req in requests]
pool.wait()
self.urlmanager.save_urls_process_status(self.urlmanager.new_urls,
r'new_urls.txt')
self.urlmanager.save_urls_process_status(
self.urlmanager.crawled_urls, r'crawled_urls.txt')
self.write_to_csv3()
sys.stdout = s
def start(self):
fo = open("anjuke.log", 'w+')
s = sys.stdout
sys.stdout = fo
num = 0
# self.get_villages()
print 'current:urls num1:', self.urlmanager.new_urls_size()
while self.urlmanager.has_new_url():
new_url = self.urlmanager.get_new_url()
self.urlmanager.save_urls_process_status(self.urlmanager.new_urls,
r'new_urls.txt')
self.urlmanager.save_urls_process_status(
self.urlmanager.crawled_urls, r'crawled_urls.txt')
try:
item = self.get_detail(new_url)
num += 1
print 'write No.', num, 'url'
self.write_to_csv(item)
except:
print 'current:urls num2:', self.urlmanager.new_urls_size()
self.urlmanager.readd_new_url(new_url)
self.urlmanager.save_urls_process_status(
self.urlmanager.new_urls, r'new_urls.txt')
self.urlmanager.save_urls_process_status(
self.urlmanager.crawled_urls, r'crawled_urls.txt')
sys.stdout = s