class Zhihuhomepage():
def __init__(self):
self.SqlH = SqlHelper()
self.SqlH.init_db('zhihu')
self.base_url = 'https://www.zhihu.com'
def gethomepage(self,url,user_name):
req = request.Request( url=url,headers={
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36'
})
html=request.urlopen(req,timeout=5).read().decode('utf-8')
tree = etree.HTML(html)
collecter = tree.xpath("//div[@class='Profile-sideColumnItemValue']/text()")
# time.sleep(2)
if collecter:
save = collecter[0][:-3].strip()
else:
save = 0
print(user_name, collecter)
self.updateCollect(user_name,save)
#print(html)
def updateCollect(self,user_name,save):
self.SqlH.update({'user_name': user_name}, {'collect': save})
def fromdb(self):
tottal_s=self.SqlH.count(condition={'collect':'none'})
print(tottal_s)
for c_page in range(tottal_s):
time.sleep(2)
result=self.SqlH.select(conditions={'collect':'none'},count=1,page=c_page)
self.gethomepage(self.base_url+result[0]['home_page'],result[0]['user_name'])
pass
class Article():
def __init__(self):
self.black_page = 'https://www.zhihu.com/account/unhuman?type=unhuman&message=%E7%B3%BB%E7%BB%9F%E6%A3%80%E6%B5%8B%E5%88%B0%E6%82%A8%E7%9A%84%E5%B8%90%E5%8F%B7%E6%88%96IP%E5%AD%98%E5%9C%A8%E5%BC%82%E5%B8%B8%E6%B5%81%E9%87%8F%EF%BC%8C%E8%AF%B7%E8%BE%93%E5%85%A5%E4%BB%A5%E4%B8%8B%E5%AD%97%E7%AC%A6%E7%94%A8%E4%BA%8E%E7%A1%AE%E8%AE%A4%E8%BF%99%E4%BA%9B%E8%AF%B7%E6%B1%82%E4%B8%8D%E6%98%AF%E8%87%AA%E5%8A%A8%E7%A8%8B%E5%BA%8F%E5%8F%91%E5%87%BA%E7%9A%84'
self.start_url = 'https://zhuanlan.zhihu.com/yinjiaoshou886/answer'
self.browser = webdriver.Chrome(
executable_path='/home/caidong/developProgram/selenium/chromedriver'
)
self.SqlH = SqlHelper()
self.SqlH.init_db('zhihu', 'zhihu_all')
self.base_url = 'https://www.zhihu.com'
self.user_home_url = ''
self.current = 1
def crawl(self, url):
self.browser.get(url)
if self.browser.current_url == self.black_page:
print("输入验证")
sys.exit()
if self.current == 2:
time.sleep(30)
self.current = self.current + 1
time.sleep(3)
self.browser.implicitly_wait(3)
self.parse_special_column(self.browser.page_source,
self.browser.current_url)
def parse_special_column(self, html, url):
tree = etree.HTML(html)
comment_list = tree.xpath('//div[@class="ContentItem-actions"]')
if len(comment_list) > 4:
comment_list = comment_list[1:4]
article_list = []
for item in comment_list:
item = etree.ElementTree(item)
answer_comment = item.xpath(
'//button[@class="Button ContentItem-action Button--plain"]/text()'
)[0]
if str(answer_comment).startswith('添加'):
answer_comment = 0
else:
answer_comment = str(answer_comment).strip()[:-3]
print(answer_comment)
article_list.append(answer_comment)
if len(article_list) == 0:
article_list = "none"
self.SqlH.update({"user_home_url": self.user_home_url}, {
"article_comment": article_list,
})
print(article_list)
class ZHSpider():
def __init__(self):
self.black_page = 'https://www.zhihu.com/account/unhuman?type=unhuman&message=%E7%B3%BB%E7%BB%9F%E6%A3%80%E6%B5%8B%E5%88%B0%E6%82%A8%E7%9A%84%E5%B8%90%E5%8F%B7%E6%88%96IP%E5%AD%98%E5%9C%A8%E5%BC%82%E5%B8%B8%E6%B5%81%E9%87%8F%EF%BC%8C%E8%AF%B7%E8%BE%93%E5%85%A5%E4%BB%A5%E4%B8%8B%E5%AD%97%E7%AC%A6%E7%94%A8%E4%BA%8E%E7%A1%AE%E8%AE%A4%E8%BF%99%E4%BA%9B%E8%AF%B7%E6%B1%82%E4%B8%8D%E6%98%AF%E8%87%AA%E5%8A%A8%E7%A8%8B%E5%BA%8F%E5%8F%91%E5%87%BA%E7%9A%84'
self.start_url = 'https://www.zhihu.com/people/kaifulee/followers?page=25583'
#self.start_url = 'https://www.zhihu.com/people/ji-da-fa-37/activities'
self.base_url = 'https://www.zhihu.com'
self.SqlH = SqlHelper()
self.SqlH.init_db('zhihu','zhihu_48000')
#self.browser = webdriver.PhantomJS()
# proxy = {'address': '60.168.104.30:3128',
# 'username': '******',
# 'password': '******'
# }
# capabilities = dict(DesiredCapabilities.CHROME)
# capabilities['proxy'] = {'proxyType': 'MANUAL',
# 'httpProxy': proxy['address'],
# 'ftpProxy': proxy['address'],
# 'sslProxy': proxy['address'],
# 'noProxy': '',
# 'class': "org.openqa.selenium.Proxy",
# 'autodetect': False}
#
# capabilities['proxy']['httpUsername'] = proxy['username']
# capabilities['proxy']['httpPassword'] = proxy['password']
# chromeOptions = webdriver.ChromeOptions()
# chromeOptions.add_argument('--proxy-server=http://60.168.104.30:3128')
#self.browser = webdriver.Chrome(chrome_options=chromeOptions,executable_path='/home/caidong/developProgram/selenium/chromedriver')
#self.browser = webdriver.PhantomJS()
#cookies = ZhihuLogin().login()
#print(cookies)
self.browser = webdriver.PhantomJS()
self.browser = webdriver.Chrome(executable_path='/home/caidong/developProgram/selenium/chromedriver')
#for cookie in cookies:
# self.browser.add_cookie({cookie['name']:cookie['value']})
#self.browser.add_cookie(cookie)
time.sleep(5)
print('cookie',self.browser.get_cookies())
#print(self.browser.get_cookies())
#self.browser.add_cookie({"cookie":'_zap=b24c85f0-aae0-456a-ba87-e0919de79409; __utma=243313742.618834370.1505397831.1505397831.1505431589.2; __utmz=243313742.1505397831.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); d_c0="AJCCExnEYAyPTuiuB47mCQN_anS_LW2ZmQI=|1505432287"; q_c1=f92e81f1440d49eca643b9bd71df1d06|1505471670000|1502586350000; aliyungf_tc=AQAAABpahiv+pQIA4wmi0wpuOA0ptCdt; __utma=51854390.226003310.1505817316.1505817316.1505817316.1; __utmc=51854390; __utmz=51854390.1505817316.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); __utmv=51854390.000--|3=entry_date=20170813=1; XSRF-TOKEN=2|02bd5b9f|30893afa3ad96af92f8d3ffb67906faa338d76fe308d3fb267de6cad358569a837dc39ae|1505824255; _xsrf=24ae8d1f-0dde-4510-a20d-ec7278275ab1; l_cap_id="NDYzOWZmNjBmZDhjNDBkZWI5MDg0NjYyZDk4YTk2OTA=|1505824625|220e4527cbfe214589599d071685e4c7f62143fc"; r_cap_id="NWJhOTRmYzg2NTVlNDczY2ExZWY3YzgxNGQ2ZmRmM2I=|1505824625|b050327da2a8dedc37a8e744640b60b553f3b771"; cap_id="YjcyNGZkYjFlY2JkNDU3ZWFlYmQ0NjQ3ZDJmNDcwZjk=|1505824625|5804f3f4999cf311334c3664f2e41ad2d4d93029'})
self.start_page = 48000
self.end_page = 47000
def crawlData(self, url=None):
dcap = dict(DesiredCapabilities.PHANTOMJS)
dcap[
"phantomjs.page.settings.userAgent"] = config.get_header()
# browser = webdriver.PhantomJS(desired_capabilities=dcap)
#browser =webdriver.Firefox()
self.browser.get(url)
i=1
if i==1:
time.sleep(30)
i=i+1
# print(browser.page_source)
self.browser.implicitly_wait(3)
print('cookie',self.browser.get_cookies())
print(self.browser.page_source)
# 点击关注者
self.browser.find_element_by_xpath('//div[@class="NumberBoard FollowshipCard-counts"]').click()
self.browser.implicitly_wait(3)
more = self.browser.find_elements_by_xpath('//button[@class="Button PaginationButton Button--plain"]')
more[-1].click()
self.browser.implicitly_wait(3)
total_page = more[-1].text
c_page = self.browser.find_element_by_xpath('//button[@class="Button PaginationButton PaginationButton--current Button--plain"]').text
print(c_page)
for curren_page in range(int(total_page)):
c_page = self.browser.find_element_by_xpath(
'//button[@class="Button PaginationButton PaginationButton--current Button--plain"]').text
print(c_page)
print('当前页:', str(curren_page))
#点击上一页
# self.browser.find_element_by_xpath(
# '//button[@class="Button PaginationButton PaginationButton-prev Button--plain"]').click()
#点击下一页
self.browser.find_element_by_xpath(
'//Button PaginationButton PaginationButton-next Button--plain"]').click()
self.browser.implicitly_wait(3)
if int(c_page) < self.start_page and int(c_page) > self.end_page:
try:
self.loop_list()
except:
print('循环点击列表出错')
#循环解析当前列表
def loop_list(self):
items = self.browser.find_elements_by_xpath('//div[@class="ContentItem-head"]//a[@class="UserLink-link"]')
print("数目",len(items))
for item in items:
time.sleep(random.randrange(5))
c_url= item.get_attribute("href")
print("c_url",c_url)
if self.SqlH.count({"home_page":c_url+'/activities'})==0:
#不出现点击点问题
while not item.is_displayed():
time.sleep(1)
try:
item.click()
except:
print('点击错误')
self.browser.implicitly_wait(1)
handle_cnt = len(self.browser.window_handles) - 1
# print('标签数',handle_cnt)
self.browser.switch_to.window(self.browser.window_handles[handle_cnt])
print(self.browser.current_url)
if self.browser.current_url==self.black_page:
time.sleep(10*60)
try:
self.browser.implicitly_wait(3)
self.parse_home_page(self.browser.page_source, self.browser.current_url)
except:
print("页面解析错误")
if handle_cnt > 0:
self.browser.close()
self.browser.switch_to.window(self.browser.window_handles[0])
else:
print("已存在")
time.sleep(random.randrange(2))
#存储数据到mongodb
def storage_mongod(self,dic):
user_name =dic.get("user_name")
if self.SqlH.count({'user_name': user_name}) == 0:
self.SqlH.insertZhiHu(dic)
else:
self.SqlH.update({'user_name': user_name}, {'collect': dic.get('collect')})
pass
def parse_home_page(self, html,url):
tree = etree.HTML(html)
follow = tree.xpath("//div[@class='NumberBoard-value']/text()")
if follow:
flowing = follow[0]
follower = follow[1].strip()
else:
flowing = 'none'
follower = 'none'
page_header = tree.xpath("//div[@class='Card ProfileMain']//ul[@class='Tabs ProfileMain-tabs']/li[@class='Tabs-item']/a/span/text()")
answer = page_header[0]
article = page_header[2]
#print('answer_',answer,'article',article)
user_name = tree.xpath("//span[@class='ProfileHeader-name']/text()")[0]
collecter = tree.xpath("//div[@class='Profile-sideColumnItemValue']/text()")
print("收藏数",collecter)
if collecter:
for item in collecter:
if str(item).endswith("次收藏"):
save=item.strip()[:-3]
else:
save = 0
else:
save = 0
print(user_name, flowing,str(save))
zhihuObj = dict(user_name=user_name, followers=follower,
home_page=url, collect=save,article=article,
answer=answer)
try:
self.storage_mongod(zhihuObj)
except:
print("数据存储错误")
class ZHSpider():
def __init__(self):
self.start_url = 'https://www.zhihu.com/people/kaifulee/activities'
self.base_url = 'https://www.zhihu.com'
self.type = [
'hot', 'local', 'shehui', 'guonei', 'guoji', 'recomment', 'junshi',
'finance', 'technology', 'sports', 'fashionbang', 'fashionbang',
'auto_moto', 'fangcan', 'technology', 'yangshengtang'
]
self.SqlH = SqlHelper()
self.SqlH.init_db('zhihu')
self.page = 2
self.totla_url_set = set()
self.wait_use_url_set = set()
self.current_type = ''
def crawlData(self, url=None):
dcap = dict(DesiredCapabilities.PHANTOMJS)
dcap[
"phantomjs.page.settings.userAgent"] = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36'
#browser = webdriver.PhantomJS(desired_capabilities=dcap)
browser = webdriver.Chrome(
'/home/caidong/developProgram/selenium/chromedriver')
browser.get(url)
#print(browser.page_source)
browser.implicitly_wait(10)
#print(browser.page_source)
#点击关注者
browser.find_element_by_xpath(
'//div[@class="NumberBoard FollowshipCard-counts"]').click()
time.sleep(2)
# for i in range(1,10):
# if i<9:
# bt_mouseover = browser.find_element_by_xpath('//li[@class="nav_item"]['+str(i)+']/a')
# actions =ActionChains(browser)
# actions.move_to_element(bt_mouseover).perform()
# browser.implicitly_wait(5)
# time.sleep(5)
# html = browser.page_source
# #print(html)
# self.current_type=self.type[i]
# self.parse(html)
# else:
# more = browser.find_elements_by_xpath('//div[@class="more_list"]/a')
# i=1
# for item in more:
# if i < 2:
# bt_mouseover = browser.find_element_by_xpath('//a[@class="more"]')
# else:
# bt_mouseover = browser.find_element_by_xpath('//a[@class="more more_current"]')
# i += 1
# actions = ActionChains(browser)
# actions.move_to_element(bt_mouseover).perform()
# time.sleep(60)
# browser.implicitly_wait(50)
# try:
# item.click()
# except:
# print ("click error")
# browser.implicitly_wait(15)
# html = browser.page_source
# self.current_type = self.type[i+6]
# print(self.current_type)
# #print(html)
# self.parse(html)
# #actions.click(item)
# time.sleep(2)
# browser.get_screenshot_as_file('1.png')
#print(browser.page_source)
#exit()
# if index <= 6:
# bt_element=('//div[@class="fieed-box"]/a[@id="%s"]'%xpath_str)
# else:
# actions = ActionChains(browser)
# more = browser.find_element_by_xpath('//div[@class="fieed-box"]/a[@id="more_anchor"]')
# actions.move_to_element(more).perform()
# bt_element=('//div[@class="tab-box-pop"]/a[@id="%s"]'%xpath_str)
# #if index > 6:
# browser.find_element_by_xpath('//div[@class="fieed-box"]/a[@id="pc_6"]').click()
# time.sleep(2)
#time.sleep(2)
#actions.move_to_element(more).perform()
# browser.find_element_by_xpath(bt_element).click()
# time.sleep(2)
#
# #browser.get_screenshot_as_file('tex.png')
# js1 = 'return document.body.scrollHeight'
# js2 = 'window.scrollTo(0, document.body.scrollHeight)'
# old_scroll_height = 0
# while(browser.execute_script(js1) > old_scroll_height):
# old_scroll_height = browser.execute_script(js1)
# browser.execute_script(js2)
# time.sleep(0.8)
# for i in range(self.page):
# load_more_xpath='//div[@class="jzgd"]/a'
# browser.find_element_by_xpath(load_more_xpath).click()
self.parse(browser.page_source, url)
print(browser.page_source)
# try:
more = browser.find_elements_by_xpath(
'//button[@class="Button PaginationButton Button--plain"]')
# except:
# print('没有下一页')
# for page in range(len(more)):
# browser.find_elements_by_xpath('//button[@class="Button PaginationButton Button--plain"]')[page].click()
# time.sleep(2)
# self.parse_page(browser.page_source)
#browser.find_element_by_xpath('//button[@class="Button PaginationButton PaginationButton-next Button--plain"]').click()
######每一次执行下一页
total_page = more[-1].text
print("tot", total_page)
for curren_page in range(int(total_page)):
try:
browser.find_element_by_xpath(
'//button[@class="Button PaginationButton PaginationButton-next Button--plain"]'
).click()
time.sleep(2)
except:
print('没有下一页')
self.parse_page(browser.page_source)
#exit()
######end
browser.quit()
def parse_page(self, html):
tree = etree.HTML(html)
followerList = tree.xpath('//div[@class="List-item"]')
# print(followerList)
for item in followerList:
followerInfo = etree.ElementTree(item)
name = followerInfo.xpath("//a[@class='UserLink-link']/text()")[0]
home_page = followerInfo.xpath(
"//a[@class='UserLink-link']/@href")[0] # 主页
follower_c = followerInfo.xpath(
"//span[@class='ContentItem-statusItem']/text()")[2]
# print('---------',home_page)
if home_page and self.base_url + home_page not in self.totla_url_set:
self.wait_use_url_set.add(self.base_url + home_page)
self.totla_url_set.add(self.base_url + home_page)
zhihuObj = dict(user_name=name,
followers=follower_c[:-3].strip(),
home_page=home_page,
collect='none')
self.saveDB(zhihuObj, name)
print(name, home_page, follower_c)
def parse(self, html, url):
tree = etree.HTML(html)
follow = tree.xpath("//div[@class='NumberBoard-value']/text()")
follower = follow[1]
#print('====',follower,type(int(follower)))
if int(follower) > 0:
followerList = tree.xpath('//div[@class="List-item"]')
#print(followerList)
for item in followerList:
followerInfo = etree.ElementTree(item)
name = followerInfo.xpath(
"//a[@class='UserLink-link']/text()")[0]
home_page = followerInfo.xpath(
"//a[@class='UserLink-link']/@href")[0] #主页
follower_c = followerInfo.xpath(
"//span[@class='ContentItem-statusItem']/text()")[2]
#print('---------',home_page)
if home_page and self.base_url + home_page not in self.totla_url_set:
self.wait_use_url_set.add(self.base_url + home_page)
self.totla_url_set.add(self.base_url + home_page)
zhihuObj = dict(user_name=name,
followers=follower_c[:-3].strip(),
home_page=home_page,
collect='none')
self.saveDB(zhihuObj, name)
print(name, home_page, follower_c)
user_name = tree.xpath("//span[@class='ProfileHeader-name']/text()")[0]
collecter = tree.xpath(
"//div[@class='Profile-sideColumnItemValue']/text()")
#time.sleep(2)
flowing = follow[0]
if collecter:
save = collecter[2][:-3].strip()
else:
save = 0
#print(save)
#print(html)
print(user_name, flowing, save)
zhihuObj = dict(user_name=user_name,
followers=follower,
flowing=flowing,
collect=save,
home_page=url)
#zhihuContent = {'user_name':user_name,'followers':follower,"flowing":flowing,"save":save,}
if self.SqlH.count({'user_name': user_name}) == 0:
self.SqlH.insertZhiHu(zhihuObj)
elif self.SqlH.count({'user_name': user_name, 'collect': 'none'}):
self.SqlH.update({'user_name': user_name}, {'collect': save})
#print(zhihuContent)
def saveDB(self, content, user_name):
if self.SqlH.count({'user_name': user_name}) == 0:
self.SqlH.insertZhiHu(content)
elif self.SqlH.count({'user_name': user_name, 'collect': 'none'}):
self.SqlH.update({'user_name': user_name}, {'collect': 'none'})
pass