def __init__(self):
# 创建日志文件夹
nowTime = common.getNowTime()
self.nowDate = common.getNowTime()
self.logPath = common.getDirPath(nowTime, "log")
common.mkdir(self.logPath)
# 设置日志文件的文件名
self.logName = 'jsTag_' + time.strftime(
'%Y%m%d_%H_%M_%S_%M', time.localtime(time.time())) + '.log'
self.logFile = self.logPath + self.logName
# 初始化日志
# 1、设置formatter，日志的输出格式
self.logFormat = '%(asctime)s - %(filename)s[line:%(lineno)d] - %(levelname)s: %(message)s'
self.formatter = logging.Formatter(self.logFormat)
# 2、设置Handler，用于写入日志的控制。先创建Handler，然后设置Handler级别
# 级别：CRITICAL > ERROR > WARNING > INFO > DEBUG，默认级别为 WARNING
self.handler = logging.FileHandler(self.logFile, mode='a')
self.handler.setLevel(logging.DEBUG)
self.handler.setFormatter(self.formatter)
# 3、创建及配置logger
self.logger = logging.getLogger()
self.logger.addHandler(self.handler)
def test_AOL():
print("............AOL................")
# 找到前天的日期
def getDate():
d = datetime.now() + timedelta(days=-2)
d1 = d + timedelta(days=-1)
if (int(d.strftime('%Y-%m-%d %H:%M:%S')[11:13]) <= 3):
str_d = d1.strftime('%Y-%m-%d %H:%M:%S')
else:
str_d = d.strftime('%Y-%m-%d %H:%M:%S')
yyyy1 = str_d[0:4]
mmmm1 = str_d[5:7]
dddd1 = str_d[8:10]
logInURL1 = yyyy1 + "-" + mmmm1 + "-" + dddd1
return logInURL1
flag = False
for try_num in range(5):
print("AOL第" + str(try_num + 1) + "次尝试-----------")
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36')
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://onemobile.aol.com/"
dataURL = "https://onemobile.aol.com/#/seller/39625/reports"
browser.get(loginURL)
print("点击登录---------------------------------")
browser.find_element_by_id("native-login").click()
time.sleep(15)
username, password = common_mysql.selectFromTb("AOL")
print("输入用户名---------------------------------")
browser.find_element_by_xpath("//input[@placeholder='Username']").send_keys(username)
time.sleep(2)
print("确认用户名，点击下一步---------------------------------")
browser.find_element_by_xpath("//input[@name='callback_2']").click()
time.sleep(10)
print("输入密码---------------------------------")
browser.find_element_by_xpath("//input[@placeholder='Password']").send_keys(password)
time.sleep(2)
print("点击登录按钮---------------------------------")
browser.find_element_by_xpath("//input[@name='callback_2']").click()
time.sleep(10)
browser.get(dataURL)
# 等待网页刷新
t = 15
while t > 0:
print("网页加载中……倒计时" + str(t) + "秒后开始操作……")
time.sleep(1) # 等待10秒钟加载时间，网络好的话，5秒就够，但是10秒比较保守
t = t - 1
# 有一个概率性弹窗，有就点击关闭。
try:
print("点击X")
browser.find_element_by_xpath("//a[@data-dismiss='modal']").click()
print("关闭弹窗成功")
except Exception as e:
errorInfo = traceback.format_exc()
print(errorInfo)
comm_logging.myLogger.write_logger(errorInfo)
print("选择开始日期%s---------------------------------" % getDate())
browser.find_element_by_xpath("//input[@placeholder='YYYY-MM-DD'][1]").clear()
time.sleep(3)
browser.find_element_by_xpath("//input[@placeholder='YYYY-MM-DD'][1]").send_keys(getDate())
browser.find_element_by_xpath("//input[@placeholder='YYYY-MM-DD'][2]").clear()
time.sleep(3)
print("选择结束日期---------------------------------")
browser.find_element_by_xpath("//input[@placeholder='YYYY-MM-DD'][2]").send_keys(getDate())
time.sleep(3)
print("点击确认日期---------------------------------")
# class='datepicker-apply-button pendo-id-datepicker-apply-button' 的 a 有3个 为何不用 index
browser.find_element_by_xpath("//div[@class='datepicker-button-panel']/a").click()
print("点击查询---------------------------------")
browser.find_element_by_xpath("//button[@class='e-btn button-primary pendo-id-generate-report']").click()
print("点击第一行数据---------------------------------")
time.sleep(5)
browser.find_element_by_xpath("//tbody[@aria-live='polite']/tr[1]").click()
print("点击维度选择---------------------------------")
time.sleep(2)
browser.find_elements_by_xpath("//span[@class='title']")[2].click()
time.sleep(1)
print("点击tag维度---------------------------------")
browser.find_element_by_xpath("//li[@data-sid='report-dimension-adTagId']").click()
print("等待加载数据---------------------------------")
time.sleep(10)
dimensionsOfdata = ["Ad Tag", "Requests", "Served", "Delivered", "Fill Rate", "Clicks", "CTR", "Revenue",
"eCPM", "RPM"]
workbook = xlwt.Workbook(encoding='utf-8')
worksheet = workbook.add_sheet('AOL')
# 加表头
col = 0
for di in dimensionsOfdata:
worksheet.write(0, col, dimensionsOfdata[col])
col = col + 1
print("开始爬数据")
soup = BeautifulSoup(browser.page_source, "lxml")
table = soup.find("table",
{"class": "table table-body table-nexage tablesorter tablesorter-default hasResizable"})
# find() 直接返回结果 findAll()返回一个符合条件的所有tag的list
tbody = table.find("tbody")
trSum = tbody.findAll("tr")
# trSum 为表格中的所有行的list
row = 1
for tr in trSum:
# col 放在for 外面，会引起错误ValueError: column index (256) not an int in range(256)
col = 0
# tdSum 为一个tr的所有列的list
tdSum = tr.findAll("td")
# td 中存在空的<td></td>,在每个tr的最后一个td
# 不取每一行的最后一个 td
for td in tdSum:
if td.text.strip() != '':
if col == 0:
worksheet.write(row, col, td.text)
else:
worksheet.write(row, col, float(td.text.replace(",", "").replace("%", "").replace("$", "")))
col = col + 1
row = row + 1
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "AOL")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||AOL抓取完毕||||||||||||||")
flag = True
except Exception as e:
flag = False
errorInfo = traceback.format_exc()
print(errorInfo)
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num >= 4):
break
def test_Mobfox():
print("............Mobfox................")
flag = False
for try_num in range(3):
print("Mobfox第" + str(try_num + 1) + "次尝试-----------")
yesterday = common.getNowTime(delta=-1, type="-")[0:10]
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://account.mobfox.com/www/cp/login.php"
dataURL = "https://account.mobfox.com/www/cp/exchange_reporting.php"
dataInnerURL = "https://account.mobfox.com/react/exchange-reporting?apikey=065f325c0728e09132ebf4cedfe10ed3&accountid=72511&hash=a305f8299d4f99e855a6a8aa26d3a221&env=prod&siteRoot=https://account.mobfox.com:443/&apiRoot=https://api-v3.mobfox.com"
dataLoadURL = "https://api-v3.mobfox.com/publisher/report?apikey=065f325c0728e09132ebf4cedfe10ed3&from=" + yesterday + "&to=" + yesterday + "&tz=Asia%2FHong_Kong&group=inventory_id&timegroup=day&totals=total_impressions%2Ctotal_served%2Ctotal_ad_source_opportunities%2Ctotal_clicks%2Ctotal_earnings&f%3Aad_source=exchange&o%3Ainclude_entities=true"
browser.get(loginURL)
time.sleep(2)
print("输入用户名密码---------------------------------")
username, password = common_mysql.selectFromTb("Mobfox")
browser.find_element_by_id("email").send_keys(username)
browser.find_element_by_id("password").send_keys(password)
# 有一个概率性弹窗，有就点击关闭。
try:
print("点击accept")
browser.find_element_by_xpath(
'//a[@class="optanon-allow-all"]').click()
print("关闭弹窗成功")
except Exception as e:
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
print(traceback.format_exc())
print("无弹窗 或 弹窗关闭失败")
time.sleep(5)
browser.find_element_by_xpath(".//*[@type='submit']").click()
# browser.find_element_by_class_name('btn btn-primary btn-myDsp ').click()
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(2)
browser.get(dataInnerURL)
print("自动选择数据维度内容---------------------------------")
browser.get(dataLoadURL)
time.sleep(2)
workbook = xlwt.Workbook(encoding='utf-8')
worksheet = workbook.add_sheet('Mobfox')
print("开始爬数据")
soup = BeautifulSoup(browser.page_source, "lxml")
dataSoup = soup.find("pre").text
#网页里提供的数据为json数据，需要json进行解析
dataJson = json.loads(dataSoup)
dataCol = dataJson["columns"]
# dataResults 的大小为 rowNum * 7
dataResults = dataJson["results"]
rowNum = dataResults.__len__()
dimensions = [
"day", "inventory", "source_opportunities", "total_served",
"total_impressions", "total_clicks", "ctr", "fillrate",
"total_earnings", "ecpm"
]
for col in range(10):
worksheet.write(0, col, dimensions[col])
try:
for row in range(0, rowNum):
for col in range(10):
if col == 0 or col == 3 or col == 5:
worksheet.write(row + 1, col,
dataResults[row][col])
elif col == 1:
InventoryID = str(dataResults[row][col])
InventoryName = dataJson["entities"][
"inventory_id"][InventoryID]["name"]
Inventory = InventoryName + " (" + InventoryID + ")"
worksheet.write(row + 1, col, Inventory)
elif col == 2:
worksheet.write(row + 1, col, dataResults[row][4])
elif col == 4:
worksheet.write(row + 1, col, dataResults[row][2])
elif col == 6:
if dataResults[row][4] != 0:
ctr = dataResults[row][5] / dataResults[row][
4] * 100
else:
ctr = "#DIV/!"
worksheet.write(row + 1, col, ctr)
elif col == 7:
if dataResults[row][2] != 0:
fillRate = dataResults[row][4] / dataResults[
row][2] * 100
else:
fillRate = "#DIV/!"
worksheet.write(row + 1, col, fillRate)
elif col == 8:
worksheet.write(row + 1, col, dataResults[row][6])
elif col == 9:
if dataResults[row][4] != 0:
ecpm = dataResults[row][6] / dataResults[row][
4] * 100
else:
ecpm = "#DIV/!"
worksheet.write(row + 1, col, ecpm)
except Exception:
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Mobfox")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Mobfox抓取完毕||||||||||||||")
flag = True
except Exception as e:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 4):
break
def test_Mobfox():
print("............Mobfox................")
flag = False
for try_num in range(3):
print("Mobfox第" + str(try_num + 1) + "次尝试-----------")
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://account.mobfox.com/www/cp/login.php"
dataURL = "https://account.mobfox.com/www/cp/exchange_reporting.php"
dataInnerURL = "https://account.mobfox.com/react/exchange-reporting?apikey=065f325c0728e09132ebf4cedfe10ed3&accountid=72511&hash=a305f8299d4f99e855a6a8aa26d3a221&env=prod&siteRoot=https://account.mobfox.com:443/&apiRoot=https://api-v3.mobfox.com"
browser.get(loginURL)
time.sleep(2)
print("输入用户名密码---------------------------------")
browser.find_element_by_id("email").send_keys(
"*****@*****.**")
browser.find_element_by_id("password").send_keys("360Security2017")
# 有一个概率性弹窗，有就点击关闭。
try:
print("点击accept")
browser.find_element_by_xpath(
'//a[@class="optanon-allow-all"]').click()
print("关闭弹窗成功")
except Exception as e:
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
print(traceback.format_exc())
print("无弹窗 或 弹窗关闭失败")
time.sleep(5)
browser.find_element_by_xpath(".//*[@type='submit']").click()
# browser.find_element_by_class_name('btn btn-primary btn-myDsp ').click()
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(2)
browser.get(dataInnerURL)
print("自动选择数据维度内容---------------------------------")
# s = browser.find_element_by_id("react-select-5--value")
# Select(s).select_by_value("inventory_id")
# s = browser.find_element_by_id("period")
# Select(s).select_by_value("yesterday")
# s = browser.find_element_by_id("timezone")
# Select(s).select_by_value("Asia/Hong_Kong")
# s = browser.find_element_by_id("timegroup")
# Select(s).select_by_value("day")
# browser.find_element_by_xpath('//span[@id="react-select-5--value"]/div[2]/input').send_keys("Daily")
# browser.find_element_by_xpath('//span[@id="react-select-6--value-item"]').text = "react-select-6--value-item"
# browser.find_element_by_class_name("text-center form-control").click()
time.sleep(10)
s = browser.find_element_by_xpath('//div[@id="content"]')
html = browser.find_element_by_xpath(
'//html[@class=" supports cssfilters"]')
browser.find_element_by_xpath('//li[text()="Yesterday"]').click()
browser.find_element_by_xpath(
'//button[@title="Download Excel"]').click()
time.sleep(2)
browser.find_element_by_xpath('//button[@type="submit"]').click()
dimensionsOfdata = [
"col-day sorting_1", "col-inventory_id",
"col-total_ad_source_opportunities", "col-total_served",
"col-total_impressions", "col-total_clicks", "col-ctr",
"col-fillrate", "col-total_earnings", "col-ecpm"
]
workbook = xlwt.Workbook(encoding='utf-8')
worksheet = workbook.add_sheet('Mobfox')
print("开始爬数据")
soup = BeautifulSoup(browser.page_source, "lxml")
tbody = soup.find("div", {
"class": "dataTables_scrollBody"
}).find("tbody")
trSum = tbody.findAll("tr")
# 加表头
col = 0
for di in dimensionsOfdata:
worksheet.write(0, col, dimensionsOfdata[col])
col = col + 1
row = 1
for tr in trSum:
# col 放在for 外面，会引起错误ValueError: column index (256) not an int in range(256)
col = 0
# tdSum 为一个tr的所有列的list
tdSum = tr.findAll("td")
for td in tdSum:
try:
worksheet.write(
row, col,
float(td.text.strip().replace(",", "").replace(
"%", "").replace("$", "")))
except ValueError:
worksheet.write(row, col, td.text.strip())
col = col + 1
row = row + 1
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Mobfox")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Mobfox抓取完毕||||||||||||||")
flag = True
except Exception as e:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 4):
break
def test_Mopub():
print("............Mopub................")
flag = False
for try_num in range(3):
print("Mopub第" + str(try_num + 1) + "次尝试-----------")
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
def getURL(str):
d = datetime.now() + timedelta(days=-1)
d1 = d + timedelta(days=-1)
d2 = d + timedelta(days=-2)
if (int(d.strftime('%Y-%m-%d %H:%M:%S')[11:13]) <= 3):
str_d1 = d1.strftime('%Y-%m-%d %H:%M:%S')
str_d2 = d2.strftime('%Y-%m-%d %H:%M:%S')
else:
str_d1 = d.strftime('%Y-%m-%d %H:%M:%S')
str_d2 = d1.strftime('%Y-%m-%d %H:%M:%S')
yyyy1 = str_d1[0:4]
mmmm1 = str_d1[5:7]
dddd1 = str_d1[8:10]
yyyy2 = str_d2[0:4]
mmmm2 = str_d2[5:7]
dddd2 = str_d2[8:10]
# 这个MopubBanner的link
logInURL1 = "https://dash.metamarkets.com/mopub-360_mobile_security/explore#ed=app_name&fs.0.k=ad_size&fs.0.v.0=300x250&fs.0.v.1=320x50&fs.1.k=timestamp&fs.1.t.0.tr.end=" + yyyy1 + "-" + mmmm1 + "-" + dddd1 + "T16&fs.1.t.0.tr.start=" + yyyy2 + "-" + mmmm2 + "-" + dddd2 + "T16&gm.0=rev_adj&gm.1=auctions&gm.2=win_rate_v2&gm.3=cleared_done&gm.4=ctr&gm.5=ecpm&gm.6=uniques&od.0=ad_size&od.1=adgroup_priority&od.2=adunit_name&od.3=app_version&od.4=app_name&od.5=country&od.6=creative_id&od.7=pub_id&sbd=0&sortBy=rev_adj&sortDim=0&sortDir=descending&td=time_day&tm.0=rev_adj&tm.1=auctions&tm.2=win_rate_v2&tm.3=cleared_done&tm.4=ctr&tm.5=ecpm&tm.6=uniques&tz=Asia~2fShanghai&zz=4"
# 这个Mopub native的link
logInURL2 = "https://dash.metamarkets.com/mopub-360_mobile_security/explore#ed=app_name&fs.0.k=ad_size&fs.0.v.0=0x0&fs.0.v.1=320x480&fs.1.k=timestamp&fs.1.t.0.tr.end=" + yyyy1 + "-" + mmmm1 + "-" + dddd1 + "T16&fs.1.t.0.tr.start=" + yyyy2 + "-" + mmmm2 + "-" + dddd2 + "T16&gm.0=rev_adj&gm.1=auctions&gm.2=win_rate_v2&gm.3=cleared_done&gm.4=ctr&gm.5=ecpm&gm.6=uniques&od.0=ad_size&od.1=adgroup_priority&od.2=adunit_name&od.3=app_version&od.4=app_name&od.5=country&od.6=creative_id&od.7=pub_id&sbd=0&sortBy=rev_adj&sortDim=0&sortDir=descending&td=time_day&tm.0=rev_adj&tm.1=auctions&tm.2=win_rate_v2&tm.3=cleared_done&tm.4=ctr&tm.5=ecpm&tm.6=uniques&tz=Asia~2fShanghai&zz=4"
if str == "Mopub_Banner":
logInURL = logInURL1
elif str == "Mopub_Native":
logInURL = logInURL2
# 传什么参数，那么就返回哪个链接
return logInURL
# dimensionsOfdata中元素 = 数据div中的colid 值
dimensionsOfdata = [
"auctions", "cleared_done", "uniques", "rev_adj",
"win_rate_v2", "ctr", "ecpm"
]
dataRes = ["Mopub_Banner", "Mopub_Native"]
workbook = xlwt.Workbook(encoding='utf-8')
for resouce in dataRes:
worksheet1 = workbook.add_sheet(resouce)
browser.get(getURL(resouce))
print("现在开始抓取" + resouce + "的数据。" + "\n链接:" + getURL(resouce))
# 因为先运行Mopub_Banner，因此只要判断第一次，第一次登陆即可
if resouce == "Mopub_Banner":
username, password = common_mysql.selectFromTb("Mopub")
browser.find_element_by_id("form-id1").send_keys(username)
browser.find_element_by_id("form-id2").send_keys(password)
browser.find_element_by_xpath(
"//button[@class='primary login']").click()
time.sleep(10)
soup = BeautifulSoup(browser.page_source, "lxml")
appNameList = soup.findAll("div", colid="app_name")
# 插入表头
m = 0
for appName in appNameList:
# 虽然网页的表头 colid="app_name" ，但是，div的下一层没有span，而数据部分的div下一层有span，因此appname的第一行空着
worksheet1.write(m, 0, appName.span.text)
m = m + 1
# 将数据按列插入excel中
j = 1
for col in dimensionsOfdata:
worksheet1.write(0, j, col)
colData = soup.findAll("div", colid=col)
i = 1
for row in colData[1:]:
try:
worksheet1.write(i, j, float(row.span["title"]))
i = i + 1
except BaseException as err:
print("Exception:", err)
j = j + 1
print(resouce + " 抓取完成----------------")
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Mopub")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
browser.quit()
print("||||||||||||Mopub抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 3):
break
def test_Adview():
print("............Adview................")
flag = False
for try_num in range(5):
print("Adview第" + str(try_num + 1) + "次尝试-----------")
try:
# adview取前天
yesterday = common.getNowTime(-2)
yyyy = yesterday[0:4]
mm = yesterday[5:7]
dd = yesterday[8:10]
# beforeYerter = common.getNowTime(-2)
# yyyyBefore = beforeYerter[0:4]
# mmBefore = beforeYerter[5:7]
# ddBefore = beforeYerter[8:10]
nowTime = common.getNowTime()
yearNow = nowTime[0:4]
monthNow = nowTime[5:7]
dayNow = nowTime[8:10]
# match_Str 为要保留的日期，但是pandas的日期格式为 yyyy-mm-dd
match_Str = yyyy + "-" + mm + "-" + dd
# match_Str_before = yyyyBefore + "-" + mmBefore + "-" + ddBefore
# Adview下载的文件名的前缀
prefix = "AdView_" + yearNow + monthNow + dayNow
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Adview")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36')
# 修改默认下载地址
prefs = {'profile.default_content_settings.popups': 0, 'download.default_directory': path}
chrome_options.add_experimental_option('prefs', prefs)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "http://www.adview.cn/web/overseas/login"
dataURL = "http://www.adview.cn/user/bid/income"
browser.get(loginURL)
time.sleep(5)
print("输入用户名密码---------------------------------")
browser.find_element_by_xpath('//span[text()="Publishers"]').click()
username,password = common_mysql.selectFromTb("Adview")
browser.find_element_by_id("email").send_keys(username)
browser.find_element_by_id("pwd").send_keys(password)
ctypes.windll.user32.MessageBoxA(0, u"点击确定后，请在15秒内输入验证码，不要点击登录!!!".encode('gb2312'),
u' 信息'.encode('gb2312'), 0)
print("请输入验证码，等待 15秒")
for i in range(1,15):
print("倒计时：",15-i)
time.sleep(1)
print("点击登录按钮------")
browser.find_element_by_xpath('//button[@class="form-control btn btn-blue blue submitBtn"]').click()
time.sleep(5)
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(5)
try:
browser.find_element_by_xpath('//a[text()="English"]').click()
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
time.sleep(2)
print("点击下载")
browser.find_element_by_xpath('//input[@value="Export detail CSV"]').click()
time.sleep(2)
# 扫描文件夹，获取 Adview 的文件列表
print("正在扫描下载的 csv 文件---------")
fileList = common.scan_File(path, prefix)
common.mkdir(path)
# 读取下载的文件，删除无用数据并 重新保存
print("正在剔除其他日期的数据 并另存为 excel ---------")
common.turnToXls_ByPandas(path + fileList[-1], path + excelName, 'Adview', match_Str)
common.remove_File(path, prefix)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Adview抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 5):
break
def test_Pubnative():
print("............Pubnative................")
flag = False
for try_num in range(3):
print("Pubnative第" + str(try_num + 1) + "次尝试-----------")
try:
yesterday = common.getNowTime(-1)
yyyy = yesterday[0:4]
mm = yesterday[5:7]
dd = yesterday[8:10]
# downloadFile = "C:/Users/Administrator/Downloads" + "/" + "Publisher App_" + dd + "." + mm + "." + yyyy + ".csv"
# 下载的文件名
downloadFile = "Publisher App_" + dd + "." + mm + "." + yyyy + "-" + dd + "." + mm + "." + yyyy + ".csv"
# 用于删除 匹配的字符串
prefix = "Publisher App_" + dd + "." + mm + "." + yyyy + "-" + dd + "." + mm + "." + yyyy
# print(downloadFile)
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
# print(path)
excelName = common.getExcelName(nowTime, "Pubnative")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 修改默认下载地址
prefs = {
'profile.default_content_settings.popups': 0,
'download.default_directory': path
}
chrome_options.add_experimental_option('prefs', prefs)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://dashboard.pubnative.net"
dataURL = "https://dashboard.pubnative.net/partner/#!/api"
browser.get(loginURL)
time.sleep(5)
print("输入用户名密码---------------------------------")
from public import common_mysql
username, password = common_mysql.selectFromTb("Pubnative")
browser.find_element_by_id("email").send_keys(username)
browser.find_element_by_id("password").send_keys(password)
# browser.find_element_by_xpath('//form[@action="/sessions"]/div[0]/div/input').send_keys("*****@*****.**")
# browser.find_element_by_xpath('//form[@action="/sessions"]/div[1]/div/input').send_keys("360Security2017666")
time.sleep(2)
browser.find_element_by_xpath('//input[@value="LOGIN"]').click()
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(15)
print("自动选择数据维度内容---------------------------------")
browser.find_element_by_xpath('//div[@class="col-md-7"]').click()
browser.find_element_by_xpath(
'//div[@class="daterange"]/ul/li[2]').click()
s = browser.find_element_by_tag_name("select")
# Select(s).select_by_value("0")
Select(s).select_by_index(12)
browser.find_element_by_xpath(
'//button[@class="btn btn-secondary ng-star-inserted"]').click(
)
time.sleep(2)
browser.find_element_by_xpath(
'//div[@class="reports-filters row"]/div[3]').click()
print("准备下载...")
time.sleep(2)
dimensionsOfdata = [
"Publisher App", "Impressions", "Requests", "Fill Rate",
"eCPM", "Clicks", "CTR", "Conversions", "Payout"
]
# workbook = xlwt.Workbook(encoding='utf-8')
# worksheet = workbook.add_sheet('Pubnative')
# print("开始爬数据")
# soup = BeautifulSoup(browser.page_source, "lxml")
# tbody = soup.find("tbody")
# time.sleep(2)
# trSum = tbody.findAll("tr")
# time.sleep(2)
# # 加表头
# col = 0
# for di in dimensionsOfdata:
# worksheet.write(0, col, dimensionsOfdata[col])
# col = col + 1
#
# row = 1
# for tr in trSum[3:]:
# # col 放在for 外面，会引起错误ValueError: column index (256) not an int in range(256)
# col = 0
# # tdSum 为一个tr的所有列的list
# tdSum = tr.findAll("td")
# for td in tdSum:
# worksheet.write(row, col, td.text.strip())
# col = col + 1
# row = row + 1
print("点击下载")
browser.find_element_by_xpath(
'//div[@class="col-md order-disabled optional-buttons"]/a[2]'
).click()
time.sleep(2)
common.mkdir(path)
# workbook.save(path + excelName)
print("读取 csv 另存为 excel")
common.turnToXls_ByPandas(path + downloadFile, path + excelName,
'Pubnative')
print("删除多余的 csv")
common.remove_File(path, prefix)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Pubnative抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 3):
break
def test_OpenX():
print("............OpenX................")
nowTime = common.getNowTime()
yearNow = nowTime[0:4]
monthNow = nowTime[5:7]
dayNow = nowTime[8:10]
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
# 用于删除 匹配的字符串
prefix = "Workspace 1-" + monthNow + "-" + dayNow + "-" + yearNow
excelName = common.getExcelName(nowTime, "openx")
# 尝试5次，错误继续
flag = False
for try_num in range(2):
print("OpenX第" + str(try_num + 1) + "次尝试-----------")
try:
print("打开浏览器")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
prefs = {
'profile.default_content_settings.popups': 0,
'download.default_directory': path
}
chrome_options.add_experimental_option('prefs', prefs)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
# 不最大化 某些按钮按不到会报错
browser.maximize_window()
time.sleep(2)
logInURL = "https://sso.openx.com/login/login"
dataURL = "http://mobimagic-ui.openx.net/app.html#/reports/pie"
# 前天
yesterday = common.getNowTime(-2)
yyyy = yesterday[0:4]
mm = yesterday[5:7]
dd = yesterday[8:10]
# print(yyyy,mm,dd)
print("打开登录网址")
browser.get(logInURL)
time.sleep(5)
# try 是正常不需要验证码登录，如果try里面失败了，相当于登录失败，需要验证码，然后就跑到了except中，按照需要验证码的方式进行登录。
username, password = common_mysql.selectFromTb("OpenX")
try:
browser.find_element_by_id("email").send_keys(username)
time.sleep(3)
browser.find_element_by_id('password').send_keys(password)
time.sleep(2)
browser.find_element_by_id("submit").click()
time.sleep(5)
browser.find_element_by_xpath(
'//a[text()="http:// mobimagic-ui.openx.net/"]').click()
except:
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
# 不能再重新载入界面，重新载入后不会出现验证码，然后继续失败
# browser.get(logInURL)
time.sleep(2)
browser.find_element_by_id("email").clear()
browser.find_element_by_id("email").send_keys(username)
time.sleep(3)
browser.find_element_by_id('password').clear()
browser.find_element_by_id('password').send_keys(password)
ctypes.windll.user32.MessageBoxA(
0, u"点击确定后，请在15秒内输入验证码，不要点击登录!!!".encode('gb2312'),
u' 信息'.encode('gb2312'), 0)
print("请输入验证码，等待 15秒")
for i in range(1, 15):
print("倒计时：", 15 - i)
time.sleep(1)
browser.find_element_by_id("submit").click()
print("模拟登录成功")
print("打开数据网址")
browser.get(dataURL)
time.sleep(10)
browser.find_element_by_xpath(
'//div[@class="date-range-filter__input"]').click()
browser.find_element_by_xpath(
'//li[@class="date-range-filter__filter date-range-filter__filter--custom date-range-filter__filter__label"]'
).click()
time.sleep(5)
# startElements = browser.find_element_by_xpath('//input[@placeholder="Enter Date"]').send_keys(mm + "/" + dd + "/" + yyyy)
# 各种方法试了一通，根本不行，下面可以list[0] 是开始日期，list[1]是结束日期
print("开始选择日期和维度")
inputElements = browser.find_elements_by_css_selector(
'[placeholder="Enter Date"]')
inputElements[0].clear()
inputElements[0].send_keys(mm + "/" + dd + "/" + yyyy)
inputElements[1].clear()
inputElements[1].send_keys(mm + "/" + dd + "/" + yyyy)
browser.find_element_by_xpath(
'//div[@class="date-range-filter__custom-date-range-menu__buttons"]/div/button[2]'
).click()
time.sleep(5)
# 隐藏日期栏、折线图 不隐藏点不到下载按钮
browser.find_element_by_xpath(
'//div[@class="reports-pie-data-collection-container__collapse-trigger"]'
).click()
time.sleep(1)
browser.find_element_by_xpath(
'//div[@class="reports-pie-collapse-toggle"]').click()
time.sleep(1)
# 点击下载 选择xlsx
browser.find_element_by_xpath(
'//div[@class="reports-pie-chart-toolkit__export"]/ox-dropdown'
).click()
time.sleep(1)
# browser.find_element_by_xpath('//button[@class="ox-btn ox-btn--tertiary"]/div').click()
browser.find_element_by_xpath('//span[text()="Excel "]').click()
time.sleep(10)
common.mkdir(path)
# 扫描文件夹，获取 openx 的文件列表
print("正在扫描下载的 xlsx 文件---------")
fileList = common.scan_File(path, prefix)
common.turnToXls_ByPandas(path + fileList[-1], path + excelName,
'openx')
common.remove_File(path, prefix)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||OpenX抓取完毕||||||||||||||")
flag = True
except Exception as e:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 4):
break
def test_cm():
print("............cm................")
# 尝试5次，错误继续
flag = False
for try_num in range(5):
print("cmcm第" + str(try_num + 1) + "次尝试-----------")
try:
print("打开浏览器")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
logInURL = "http://console.cmcm.com"
d = datetime.now() + timedelta(days=-1)
d1 = d + timedelta(days=-1)
if (int(d.strftime('%Y-%m-%d %H:%M:%S')[11:13]) <= 3):
str_d1 = d1.strftime('%Y-%m-%d %H:%M:%S')
else:
str_d1 = d.strftime('%Y-%m-%d %H:%M:%S')
yyyy1 = str_d1[0:4]
mmmm1 = str_d1[5:7]
dddd1 = str_d1[8:10]
dataURL = "http://console.cmcm.com/report/placement?by=day&from=" + yyyy1 + "-" + mmmm1 + "-" + dddd1 + "&to=" + yyyy1 + "-" + mmmm1 + "-" + dddd1
print("打开登录网址")
browser.get(logInURL)
time.sleep(2)
print("模拟输入用户名密码中")
username, password = common_mysql.selectFromTb("cm")
browser.find_element_by_name('email').send_keys(username)
browser.find_element_by_name('password').send_keys(password)
browser.find_element_by_xpath(
'//*[@id="login-form"]/div[3]/div/button').click()
print("模拟登录成功")
print("打开数据网址")
browser.get(dataURL)
time.sleep(20)
dimensionsOfdata = [
"datekey", "item_placement", "backfill", "wins", "impressions",
"requests", "fillrate", "winrate", "clicks", "ctr", "ecpm",
"money"
]
workbook = xlwt.Workbook(encoding='utf-8')
worksheet = workbook.add_sheet('cmcm')
pageNum = browser.find_elements_by_xpath(
'//li[@data-page]').__len__()
# 如果不进行此判断，会出现空表
if pageNum >= 1:
row = 1
for page in range(pageNum):
print("-------读取第" + str(page + 1) + "页----------")
browser.find_element_by_xpath('//li[@data-page=' +
str(page + 1) + ']').click()
soup = BeautifulSoup(browser.page_source, "lxml")
tbody = soup.find(
"table", {
"class": "bordered highlighted scrolling-table"
}).find("tbody")
trSum = tbody.findAll("tr")
for tr in trSum:
col = 0
tdSum = tr.findAll("td")
for td in tdSum:
if col == 0 or col == 1:
worksheet.write(row, col, td.text)
else:
worksheet.write(
row, col, float(td.text.replace(",", "")))
col = col + 1
row = row + 1
time.sleep(2)
else:
continue
# 处理表头
col = 0
for d in dimensionsOfdata:
worksheet.write(0, col, dimensionsOfdata[col])
col = col + 1
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "cmcm")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||cmcm抓取完毕||||||||||||||")
flag = True
except Exception as e:
flag = False
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 4):
break
def test_Smaato():
print("............Smaato................")
# 尝试5次，错误继续
flag = False
for try_num in range(5):
print("Smaato第" + str(try_num + 1) + "次尝试-----------")
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://spx.smaato.com/publisherportal/pages/login.xhtml"
username, password = common_mysql.selectFromTb("Smaato")
dataURL = "https://spx.smaato.com/publisherportal/pages/reporting/reporting.xhtml"
print("进入登陆页---------------------------------")
browser.get(loginURL)
print(" 输入用户名密码-------------------------")
browser.find_element_by_id("j_username").send_keys(username)
browser.find_element_by_id("j_password").send_keys(password)
browser.find_element_by_xpath("//button[@type='submit']").click()
time.sleep(5)
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(10)
print(" 点击日历表------------------")
browser.find_element_by_xpath(
"//span[@id='reporting:popup']").click()
print(" 选择yesterday----------------")
browser.find_element_by_xpath(
"//div[@class='drp_shortcuts-block1']/span[2]").click()
print(" 点击update----------------")
browser.find_element_by_xpath(
"//input[@class='apply-btn']").click()
time.sleep(2)
print(" 点击display by：----------------")
browser.find_element_by_xpath(
"//label[@id='reporting:displayByMenu_label']").click()
print(" 点击Adspace----------------")
# data-label 是 li 的数据部分
browser.find_element_by_xpath(
"//li[@data-label='Adspace']").click()
# 需要一个刷新延时
time.sleep(5)
# 表头信息列表
dimensionsOfdata = [
"Adspace", "Adspace ID", "Net Revenue", "Gross Revenue",
"Ad Requests", "Served Ads", "Fillrate", "Impressions",
"Viewrate", "Net eCPM", "Gross eCPM", "Clicks"
]
# 新建表格对象
workbook = xlwt.Workbook(encoding='utf-8')
# 新建sheet对象
worksheet = workbook.add_sheet('Smaato')
soup = BeautifulSoup(browser.page_source, "lxml")
print("开始爬数据------------------")
table = soup.find("tbody",
{"id": "reporting:reportingSummaryTable_data"})
trSum = table.findAll("tr")
time.sleep(2)
# 处理数据部分
# 第一行为表头，从第二行开始填充数据
row = 1
# 对所有行进行循环
for tr in trSum:
col = 0
tdSum = tr.findAll("td")
# 对每一行的所有列进行循环
for td in tdSum:
try:
if col == 2 or col == 3 or col == 9 or col == 10:
worksheet.write(
row, col,
float(
td.text.replace("$", "").replace(",", "")))
elif col == 4 or col == 5 or col == 7 or col == 11:
worksheet.write(row, col,
float(td.text.replace(",", "")))
elif col == 6 or col == 8:
worksheet.write(row, col,
float(td.text.replace("%", "")))
elif col == 1:
worksheet.write(row, col, float(td.text))
else:
worksheet.write(row, col, td.text)
except Exception as e:
print(traceback.format_exc())
# continue
col = col + 1
row = row + 1
# 处理表头
col = 0
for d in dimensionsOfdata:
worksheet.write(0, col, dimensionsOfdata[col])
col = col + 1
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Smaato")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Smaato抓取完毕||||||||||||||")
flag = True
# read_workbook = xlrd.open_workbook(path + excelName)
# read_sheetName = read_workbook.sheet_names()[0]
# read_sheet = read_workbook.sheet_by_name(read_sheetName)
# read_numRow, read_numCol = read_sheet.nrows, read_sheet.ncols
# if read_sheet.cell(read_numRow - 1,1) is '':
# flag = False
except Exception as e:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 4):
break
def test_NewCM():
print("............NewCM................")
flag = False
for try_num in range(5):
print("NewCM第" + str(try_num + 1) + "次尝试-----------")
try:
yesterday = common.getNowTime(-1)
yyyy = yesterday[0:4]
mm = yesterday[5:7]
dd = yesterday[8:10]
nowTime = common.getNowTime()
yearNow = nowTime[0:4]
monthNow = nowTime[5:7]
dayNow = nowTime[8:10]
# NewCM下载的文件名的前缀
prefix = "Brand+Details_" + yearNow + monthNow + dayNow
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "NewCM")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 修改默认下载地址
prefs = {
'profile.default_content_settings.popups': 0,
'download.default_directory': path
}
chrome_options.add_experimental_option('prefs', prefs)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://peg.cmcm.com/login"
dataURL = "https://peg.cmcm.com/frontReport/default"
browser.get(loginURL)
time.sleep(5)
print("输入用户名密码---------------------------------")
username, password = common_mysql.selectFromTb("NewCM")
browser.find_element_by_name("username").send_keys(username)
browser.find_element_by_name("password").send_keys(password)
ctypes.windll.user32.MessageBoxA(
0, u"点击确定后，请在15秒内输入验证码，不要点击登录!!!".encode('gb2312'),
u' 信息'.encode('gb2312'), 0)
print("请输入验证码，等待 15秒")
for i in range(1, 15):
print("倒计时：", 15 - i)
time.sleep(1)
print("点击登录按钮------")
browser.find_element_by_xpath('//button[@type="submit"]').click()
time.sleep(5)
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(5)
# 选择 AD unit
print("选择 AD unit")
browser.find_element_by_xpath(
'//button[text()="Dimension & Metric"]').click()
browser.find_element_by_xpath('//span[text()="Ad Unit"]').click()
time.sleep(2)
browser.find_element_by_xpath('//button[text()="Submit"]').click()
time.sleep(2)
# 选择日期
print("选择日期")
browser.find_element_by_xpath(
'//button[@class="btn btn-default ng-binding"]').click()
browser.find_element_by_xpath('//span[text()="Yesterday"]').click()
time.sleep(2)
browser.find_element_by_xpath('//button[text()="Submit"]').click()
time.sleep(2)
# 点击search
print("# 点击search")
browser.find_element_by_xpath('//input[@value="Search"]').click()
time.sleep(5)
print("点击下载")
browser.find_element_by_xpath(
'//a[@class="btn btn-outline"]').click()
time.sleep(5)
# 扫描文件夹，获取 NewCM 的文件列表
print("正在扫描下载的 csv 文件---------")
fileList = common.scan_File(path, prefix)
common.mkdir(path)
# 读取下载的文件，删除无用数据并 重新保存
print("正在剔除其他日期的数据 并另存为 excel ---------")
common.turnToXls_ByPandas(path + fileList[-1], path + excelName,
'NewCM')
common.remove_File(path, prefix)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||NewCM抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 5):
break
def test_Tappx():
print("............Tappx................")
flag = False
for try_num in range(3):
print("Tappx第" + str(try_num + 1) + "次尝试-----------")
try:
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://www.tappx.com/en/admin/login/"
dataURL = "https://www.tappx.com/en/admin/monetize/"
browser.get(loginURL)
print("输入用户名密码---------------------------------")
username, password = common_mysql.selectFromTb("Tappx")
browser.find_element_by_id("username").send_keys(username)
browser.find_element_by_id("password").send_keys(password)
browser.find_element_by_xpath("//button[@type='submit']").click()
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
print("自动选择数据维度内容---------------------------------")
time.sleep(5)
print("选择货币形式为美元---------------------------------")
browser.find_element_by_xpath(
"//div[@id='currency-selector']/a[2]").click()
time.sleep(1)
print("取消APP维度---------------------------------")
browser.find_element_by_xpath(
"//i[@class='tappxicon tappxicon-close']").click()
time.sleep(1)
print("展开维度---------------------------------")
browser.find_element_by_xpath(
"//button[@class='btn-block btn btn-default dropdown-toggle']"
).click()
time.sleep(1)
print("点击显示下拉维度---------------------------------")
browser.find_element_by_xpath(
"//a[@class='dropdown-opener']").click()
time.sleep(1)
print("点击day维度---------------------------------")
browser.find_element_by_xpath(
"//li[@class='model-g_time_day no-image']").click()
time.sleep(1)
print("展开维度---------------------------------")
browser.find_element_by_xpath(
"//button[@class='btn-block btn btn-default dropdown-toggle']"
).click()
time.sleep(1)
print("点击app维度---------------------------------")
browser.find_element_by_xpath(
"//ul[@class='list-unstyled']/li[5]").click()
time.sleep(1)
print("展开维度---------------------------------")
browser.find_element_by_xpath(
"//button[@class='btn-block btn btn-default dropdown-toggle']"
).click()
time.sleep(1)
print("点击format维度---------------------------------")
browser.find_element_by_xpath(
"//ul[@class='list-unstyled']/li[2]").click()
time.sleep(1)
print("展开维度---------------------------------")
browser.find_element_by_xpath(
"//button[@class='btn btn-default btn-plus dropdown-toggle']"
).click()
time.sleep(1)
print("点击时间下拉维度---------------------------------")
browser.find_element_by_xpath(
"//ul[@id='default-options']/li[4]").click()
time.sleep(1)
print("点击昨天维度---------------------------------")
browser.find_element_by_xpath(
"//ul[@id='default-options']/li[4]/ul/li[2]").click()
time.sleep(3)
dimensionsOfdata = [
"Date", "App", "Format", "Requests", "Deliveries",
"Impressions", "Clicks", "CPM", "CTR", "Fill Rate",
"Render Rate", "Benefits"
]
workbook = xlwt.Workbook(encoding='utf-8')
worksheet = workbook.add_sheet('Tappx')
soup = BeautifulSoup(browser.page_source, "lxml")
print("开始爬数据")
tbody = soup.find("tbody", {"id": "data-raw-table"})
trSum = tbody.findAll("tr")
row = 1
for tr in trSum:
col = 0
tdSum = tr.findAll("td")
for td in tdSum:
# print(td.text)
try:
if col >= 3:
worksheet.write(
row, col,
float(
td.text.replace(",", "").replace(
"%", "").replace("$", "")))
elif col == 1:
x = td.text.replace("\n", "").strip()
worksheet.write(row, col,
td.text.replace("\n", "").strip())
else:
worksheet.write(row, col, float(td.text))
except ValueError:
worksheet.write(row, col, td.text)
col = col + 1
row = row + 1
# 加表头
col = 0
for di in dimensionsOfdata:
worksheet.write(0, col, dimensionsOfdata[col])
col = col + 1
nowTime = common.getNowTime()
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Tappx")
common.mkdir(path)
workbook.save(path + excelName)
print(" excel保存成功，路径：" + path + "-----------")
browser.quit()
print("||||||||||||Tappx抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 3):
break
def test_Solo():
print("............Solo................")
flag = False
for try_num in range(3):
print("Solo第" + str(try_num + 1) + "次尝试-----------")
try:
dayBeforeYester = common.getNowTime(-2)
yyyy = dayBeforeYester[0:4]
mm = dayBeforeYester[5:7]
dd = dayBeforeYester[8:10]
nowTime = common.getNowTime()
yearNow = nowTime[0:4]
monthNow = nowTime[5:7]
dayNow = nowTime[8:10]
# Solo下载的文件名的前缀
prefix = "ReportTable"
# 路径只取nowTime的日期部分
path = common.getDirPath(nowTime)
excelName = common.getExcelName(nowTime, "Solo")
chrome_options = webdriver.ChromeOptions()
# 使用headless无界面浏览器模式
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
chrome_options.add_argument(
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36'
)
# 修改默认下载地址
prefs = {
'profile.default_content_settings.popups': 0,
'download.default_directory': path
}
chrome_options.add_experimental_option('prefs', prefs)
# 启动浏览器，获取网页源代码
browser = webdriver.Chrome(chrome_options=chrome_options)
loginURL = "https://portal.newborntown.com/logout"
dataURL = "https://portal.newborntown.com/report"
browser.get(loginURL)
time.sleep(5)
print("输入用户名密码---------------------------------")
username, password = common_mysql.selectFromTb("Solo")
browser.find_element_by_name("email").send_keys(username)
browser.find_element_by_name("password").send_keys(password)
ctypes.windll.user32.MessageBoxA(
0, u"点击确定后，请在15秒内输入验证码，不要点击登录!!!".encode('gb2312'),
u' 信息'.encode('gb2312'), 0)
print("请输入验证码，等待 15秒")
for i in range(1, 15):
print("倒计时：", 15 - i)
time.sleep(1)
print("点击登录按钮------")
browser.find_element_by_xpath('//button[@id="btn_login"]').click()
time.sleep(5)
print("跳转到数据页面---------------------------------")
browser.get(dataURL)
time.sleep(5)
# 选择stats页
print("选择stats页")
browser.find_element_by_xpath('//h4[text()="Stats"]').click()
time.sleep(5)
# 开始选择日期
print("开始选择日期")
browser.find_element_by_id("reportrange").click()
time.sleep(2)
browser.find_element_by_xpath(
'//li[text()="Custom Range"]').click()
time.sleep(2)
browser.find_element_by_name("daterangepicker_start").clear()
browser.find_element_by_name("daterangepicker_start").send_keys(
mm + "/" + dd + "/" + yyyy)
time.sleep(2)
browser.find_element_by_name("daterangepicker_end").clear()
browser.find_element_by_name("daterangepicker_end").send_keys(mm +
"/" +
dd +
"/" +
yyyy)
time.sleep(2)
browser.find_element_by_xpath('//button[text()="Apply"]').click()
time.sleep(5)
# 点击选择 slot 页
print("点击选择 slot 页")
browser.find_element_by_id("slot").click()
# browser.maximize_window()
time.sleep(5)
print("点击下载")
browser.find_element_by_id("export").click()
time.sleep(2)
# 扫描文件夹，获取 Solo 的文件列表
print("正在扫描下载的 csv 文件---------")
fileList = common.scan_File(path, prefix)
common.mkdir(path)
# 读取下载的文件，删除无用数据并 重新保存
print("正在剔除其他日期的数据 并另存为 excel ---------")
common.turnToXls_ByPandas(path + fileList[0], path + excelName,
'Solo')
common.remove_File(path, prefix)
print(" excel保存成功，路径：" + path + "-----------")
print("||||||||||||Solo抓取完毕||||||||||||||")
flag = True
except:
flag = False
print(traceback.format_exc())
errorInfo = traceback.format_exc()
comm_logging.myLogger.write_logger(errorInfo)
continue
finally:
browser.quit()
if (flag or try_num == 3):
break