【selenium】爬取cainiaojinrong后台客户管理里面的参数设置
import datetime import time import json import self as self import MessageToSlack import os import pymysql.cursors import random import sys import undetected_chromedriver.v2 as uc from decouple import config from functools import reduce from selenium.webdriver.common.by import By from selenium.webdriver.common.action_chains import ActionChains from selenium.webdriver.common.desired_capabilities import DesiredCapabilities from dateutil.relativedelta import relativedelta # 用来驱动浏览器的 from selenium import webdriver from selenium.webdriver.chrome.options import Options # from decouple import config class Cainiao_Scrape: need_customer_data = [] discount_rate_data = [] parameter_setting_data = [] page_count = 0 # 初始设置函数 def __init__(self, web_url, pro_db_host, pro_db_user, pro_db_pwd, pro_db_name, u02_db_host, u02_db_user, u02_db_pwd, u02_db_name, u01_db_host, u01_db_user, u01_db_pwd, u01_db_name, cainiao_ac, cainiao_pwd): self.vars = None self.driver = None self.web_url = web_url # production self.pro_db_host = pro_db_host self.pro_db_user = pro_db_user self.pro_db_pwd = pro_db_pwd self.pro_db_name = pro_db_name # u02 self.u02_db_host = u02_db_host self.u02_db_user = u02_db_user self.u02_db_pwd = u02_db_pwd self.u02_db_name = u02_db_name # u01 self.u01_db_host = u01_db_host self.u01_db_user = u01_db_user self.u01_db_pwd = u01_db_pwd self.u01_db_name = u01_db_name self.cainiao_ac = cainiao_ac self.cainiao_pwd = cainiao_pwd # 浏览器配置 def setup_method(self, method): port_number = "127.0.0.1:9222" # port_number = "192.168.20.24:2223" try: check_folder = os.path.isdir("ScreenCap") if not check_folder: os.makedirs("ScreenCap") chrome_options = webdriver.ChromeOptions() chrome_options.add_experimental_option( "debuggerAddress", port_number) capabilities = DesiredCapabilities.CHROME capabilities["goog:loggingPrefs"] = { 'browser': 'ALL', "performance": "ALL"} capabilities["goog:perfLoggingPrefs"] = {'enableNetwork': True} chrome_options.add_argument("--disable-extensions") chrome_options.add_argument("--disable-popup-blocking") chrome_options.add_argument("--ignore-certificate-errors") chrome_options.add_argument("--disable-plugins-discovery") chrome_options.add_argument('--no-first-run') chrome_options.add_argument('--no-service-autorun') chrome_options.add_argument('--no-default-browser-check') chrome_options.add_argument('no-startup-window') chrome_options.add_argument('--disable-dev-shm-usage') # chrome_options.add_argument('--no-sandbox') # highest right to operate # chrome_options.add_argument('--headless') #--headless是不显示浏览器启动及执行过程 # chrome_options.add_argument('--disable-gpu') # 不加载gpu,规避bug self.driver = webdriver.Chrome( options=chrome_options, desired_capabilities=capabilities) self.vars = {} # 清除浏览器cookie # self.driver.delete_all_cookies() MessageToSlack.post_message_to_slack("Start Chrome Driver") except Exception as r: self.teardown_method MessageToSlack.post_message_to_slack( "浏览器配置setup_method抛出异常:%s" % str(r)) sys.exit() # 菜鸟登录的函数 def login_cainiao(self): try: MessageToSlack.post_message_to_slack("Go to cainiao website") try: self.driver.get(self.web_url) except Exception as get2: MessageToSlack.post_message_to_slack(str(get2)) MessageToSlack.post_message_to_slack('get that page again') self.driver.get(self.web_url) link = self.web_url time.sleep(random.randint(4, 6)) # # 打开本地的浏览器的时候,先跳过登录这一块然后再测试 # self.driver.switch_to.frame(0) # time.sleep(random.randint(4, 6)) # # 点击输入账户 # self.driver.find_element(By.XPATH, "//input[@id='fm-login-id']").click() # time.sleep(2) # self.driver.find_element(By.XPATH, "//input[@id='fm-login-id']").send_keys(self.cainiao_ac) # # 点击输入密码 # self.driver.find_element(By.XPATH, "//input[@id='fm-login-password']").click() # time.sleep(2) # self.driver.find_element(By.XPATH, "//input[@id='fm-login-password']").send_keys(self.cainiao_pwd) # # 调取滑块验证部分 # self.slide_auth() # # 点击登陆按钮 # self.driver.find_element(By.XPATH, "//button[@type='submit']").click() # MessageToSlack.post_message_to_slack('pressed login button') # # 如果有快速进入的验证框,点击快速验证 # try: # self.driver.switch_to.frame(0) # self.driver.find_element(By.XPATH, "//button[@type='submit']").click() # MessageToSlack.post_message_to_slack('pressed quick entry button') # except Exception as gin1: # MessageToSlack.post_message_to_slack(str(gin1)) # MessageToSlack.post_message_to_slack('pressed quick entry button Unable to locate element') # pass # time.sleep(random.randint(4, 6)) # self.driver.get_log("performance") # get_log方法用于获取给定日志类型的日 get log first time, after get once log will be deleted try: self.driver.get(self.web_url) except Exception as get1: MessageToSlack.post_message_to_slack(str(get1)) MessageToSlack.post_message_to_slack('get that page again') self.driver.get(self.web_url) time.sleep(random.randint(4, 6)) MessageToSlack.post_message_to_slack("Get Data Page") MessageToSlack.post_message_to_slack( "此时浏览器打开的页面的标题:%s" % self.driver.title) # 获取cai_niao_watermark_dailylogs表中stock_central_amt>0并且日期範圍由2年前的今天到今天的customer_name数据 self.get_db_customer_data() # time.sleep(2) # 遍历客户列表 db_customer_data = self.need_customer_data MessageToSlack.post_message_to_slack('今日要更新的db_customer_data的客户数量为:%d' % len(db_customer_data)) MessageToSlack.post_message_to_slack(db_customer_data) for item in db_customer_data: customer_name = item['customer_name'] customer_id = item['customer_id'] # 调用客户查询函数 self.customer_query(customer_id, customer_name) continue self.goto_parameter_setting() time.sleep(2) MessageToSlack.post_message_to_slack("Data Scrape finished") # # to press button to log out # self.driver.switch_to.default_content() # self.driver.find_element(By.CSS_SELECTOR, ".cct-nickname").click() # time.sleep(1) # self.driver.find_element(By.LINK_TEXT, "退出").click() # MessageToSlack.post_message_to_slack("logged out") except Exception as tc: self.driver.get_screenshot_as_file( "./ScreenCap/" + datetime.datetime.now().strftime("%Y%m%d_%H%M%p") + ".png") # 异常时捕获页面截图 MessageToSlack.post_message_to_slack( "菜鸟登录的函数login_cainiao抛出异常:%s" % str(tc)) # self.teardown_method # 调用关闭浏览器函数 sys.exit() # 退出程序机制 通过引发SystemExit异常来退出Python程序 # 获取拖动按钮位置并拖动 def slide_auth(self): try: time.sleep(random.randint(4, 6)) # check have slide verify elements or not if (self.driver.find_element(By.ID, "baxia-password")).is_displayed(): self.driver.switch_to.frame('baxia-dialog-content') element = self.driver.find_element(By.ID, "nc_1_n1z") slide_bar = self.driver.find_element(By.ID, "nc_1__scale_text") action = ActionChains(self.driver) action.move_to_element(element).pause(1).click_and_hold(element).pause(1).move_by_offset( slide_bar.size['width'] - element.size['width'], 0).pause(1).release() # mobile slider action.perform() MessageToSlack.post_message_to_slack( "Slide the verification bar") self.driver.switch_to.parent_frame() time.sleep(2) if (self.driv2er.find_element(By.ID, "baxia-password")).is_displayed(): counter1 = 0 while (self.driver.find_element(By.ID, "baxia-password")).is_displayed() and counter1 < 6: self.driver.switch_to.frame('baxia-dialog-content') wrapper = self.driver.find_element(By.ID, "nocaptcha") wrapper.click() time.sleep(random.randint(2, 3)) element = self.driver.find_element(By.ID, "nc_1_n1z") slide_bar = self.driver.find_element( By.ID, "nc_1__scale_text") action = ActionChains(self.driver) action.move_to_element(element).pause(1).click_and_hold(element).pause(1).move_by_offset( slide_bar.size['width'] - element.size['width'], 0).release() # mobile slider action.perform() MessageToSlack.post_message_to_slack( "Slide the verification bar") self.driver.switch_to.parent_frame() time.sleep(random.randint(2, 3)) counter1 += 1 counter1 = 0 if (self.driver.find_element(By.ID, "baxia-password")).is_displayed(): raise Exception( "Still cannot pass the slide certification after 5 attempt") except Exception as s: self.driver.get_screenshot_as_file( "./ScreenCap/" + datetime.datetime.now().strftime("%Y%m%d_%H%M%p") + ".png") MessageToSlack.post_message_to_slack( "获取拖动按钮位置并拖动slide_auth抛出异常:%s" % str(s)) self.teardown_method sys.exit() # 点击查询客户 def customer_query(self, customer_id, customer_name): # 点击客户列表中的客户名称输入框,并赋值对应的时间段,点击查询按钮 if (self.driver.find_element(By.XPATH, '//form[@id="J_Form"]')).is_displayed(): MessageToSlack.post_message_to_slack( 'customer name search form exists') # 点击输入企业名称 self.driver.find_element( By.XPATH, "//input[@id='c-text-51']").click() time.sleep(2) self.driver.execute_script( 'document.querySelector("#c-text-51").value="";') self.driver.find_element( By.XPATH, "//input[@id='c-text-51']").send_keys(customer_name) # 日期选择范围 two_yrs_ago = datetime.datetime.now() - relativedelta(years=2) start_year = int(two_yrs_ago.strftime('%Y')) start_month = int(two_yrs_ago.strftime('%m')) start_day = int(two_yrs_ago.strftime('%d')) # 年月日的XPATH current_selection_year_start = '.c-cal-year[data-year="%s"]' % start_year current_selection_month_start = '.c-cal-month[data-month="%s"]' % start_month current_selection_day_start = '.c-cal-day[data-day="%s"]' % start_day # 点击日期选择器的DIV self.driver.find_element( By.XPATH, "//div[@id='c-calendar-select-66']").click() time.sleep(1) # 点击日期选择器的年份 self.driver.find_element( By.XPATH, '//*[@id="c-calendar-87"]/div/div[1]/div[1]').click() time.sleep(1) self.driver.find_element( By.CSS_SELECTOR, current_selection_year_start).click() # 点击日期选择器的月份 self.driver.find_element( By.XPATH, '//*[@id="c-calendar-87"]/div/div[1]/div[2]').click() time.sleep(1) self.driver.find_element( By.CSS_SELECTOR, current_selection_month_start).click() # 点击日期选择器的日期 self.driver.find_element( By.CSS_SELECTOR, current_selection_day_start).click() # 点击确定 self.driver.find_element( By.XPATH, '//*[@id="c-calendar-81"]/div[3]/button').click() # 点击查询 self.driver.find_element(By.XPATH, "//div[@id='J_search']").click() MessageToSlack.post_message_to_slack('press the search div!') time.sleep(random.randint(4, 6)) # 调取performance log分析函数 performance_log_customer_query = self.driver.get_log("performance") self.log_analysis_customer_query( performance_log_customer_query, customer_name, customer_id) else: raise Exception( "cannot find this search form in customer list page. Please check is login problem or other") # 客户列表日志分析函数 def log_analysis_customer_query(self, performance_log_customer_query, customer_name, customer_id): for row in performance_log_customer_query: log_json_customer_query = json.loads(row['message']) log_customer_query = log_json_customer_query['message'] # MessageToSlack.post_message_to_slack( # "点击查询客户的performance log在for循环中处理的结果") # MessageToSlack.post_message_to_slack(log_customer_query) method_customer_query = str(log_customer_query['method']) if method_customer_query != 'Network.responseReceived': continue url_customer_query = str( log_customer_query['params']['response']['url']) if url_customer_query != 'https://merchant.finance.cainiao.com/funds/signApproveDetail/signApproveDetailData.do': continue requestId_customer_query = str( log_customer_query['params']['requestId']) try: customer_query_page_data = \ self.driver.execute_cdp_cmd('Network.getResponseBody', {'requestId': requestId_customer_query})[ 'body'] cqd = json.loads(customer_query_page_data) # MessageToSlack.post_message_to_slack('点击查询客户的cqd结果数据') # MessageToSlack.post_message_to_slack(cqd) total_customer_query = cqd['data']['total'] if total_customer_query > 0: # 调取跳转参数设置页面的函数 resultList = cqd['data']['resultList'] for item in resultList: if str(item['companyName']) != customer_name: continue approveNo = item['approveNo'] self.parameter_setting_data.append({ 'approveNo': approveNo, 'customer_id': customer_id, 'customer_name': customer_name }) else: MessageToSlack.post_message_to_slack( 'No data for this company %s was found' % customer_name) except Exception as customer_query_e: MessageToSlack.post_message_to_slack( 'Problem after press the search button! ' + str(customer_query_e)) MessageToSlack.post_message_to_slack('数据已存入parameter_setting_data中') # 调取跳转参数设置页面的函数 def goto_parameter_setting(self): parameter_setting_data = list_dict_duplicate_removal( self.parameter_setting_data) # MessageToSlack.post_message_to_slack('去重复后的parameter_setting_data') # MessageToSlack.post_message_to_slack(parameter_setting_data) for ii in parameter_setting_data: approveNo = ii['approveNo'] customer_id = ii['customer_id'] customer_name = ii['customer_name'] parameter_setting_url = 'https://merchant.finance.cainiao.com/funds/signApproveDetail/loanParameterSettings.htm?approveNo=' + approveNo try: self.driver.get(parameter_setting_url) except Exception as get_parameter_setting: MessageToSlack.post_message_to_slack( str(get_parameter_setting)) MessageToSlack.post_message_to_slack( 'get parameter setting page again') self.driver.get(parameter_setting_url) # print('跳转到参数设置的页面标题为:' + self.driver.title) time.sleep(random.randint(4, 6)) # 调取performance log分析函数 performance_log_parameter_setting = self.driver.get_log( "performance") self.log_analysis_parameter_setting( performance_log_parameter_setting, customer_name, customer_id) # 参数设置日志分析函数 def log_analysis_parameter_setting(self, performance_log_parameter_setting, customer_name, customer_id): for row in performance_log_parameter_setting: log_json_parameter_setting = json.loads(row['message']) log_parameter_setting = log_json_parameter_setting['message'] if str(log_parameter_setting['method']) != 'Network.responseReceived': continue url_parameter_setting = log_parameter_setting['params']['response']['url'] if str(url_parameter_setting) != 'https://merchant.finance.cainiao.com/funds/signApproveDetail/pledgeRateData.do': continue requestId_parameter_setting = log_parameter_setting['params']['requestId'] try: parameter_setting_page_data = self.driver.execute_cdp_cmd('Network.getResponseBody', { 'requestId': requestId_parameter_setting})['body'] psd = json.loads(parameter_setting_page_data) # MessageToSlack.post_message_to_slack('psd') # MessageToSlack.post_message_to_slack(psd) total_parameter_setting = psd['data']['resultList'] if len(total_parameter_setting) > 0: resultList = psd['data']['resultList'] for item in resultList: item['customer_id'] = customer_id item['customer_name'] = customer_name self.discount_rate_data.append(item) else: MessageToSlack.post_message_to_slack( 'No data for this company %s was found' % customer_name) except Exception as parameter_setting_e: MessageToSlack.post_message_to_slack( 'Problem after press the search button! ' + str(parameter_setting_e)) # 获取cai_niao_watermark_dailylogs表中stock_central_amt>0并且日期範圍由2年欠的今天到今天的customer_name数据 def get_db_customer_data(self): record_exist = 0 db_host = self.pro_db_host db_user = self.pro_db_user db_pwd = self.pro_db_pwd db_name = self.pro_db_name two_yrs_ago = datetime.datetime.now() - relativedelta(years=2) start_date = two_yrs_ago.strftime('%Y-%m-%d') end_date = datetime.datetime.now().strftime('%Y-%m-%d') try: connection = pymysql.connect(host=db_host, user=db_user, password=db_pwd, database=db_name, cursorclass=pymysql.cursors.DictCursor) MessageToSlack.post_message_to_slack( "The database is connected to query data") with connection: with connection.cursor() as cursor: check_sql = "SELECT EXISTS ( SELECT `customer_name` FROM `cai_niao_watermark_dailylogs` WHERE `stock_central_amt` > 0 AND `created_at` BETWEEN %s AND %s)" cursor.execute(check_sql, (start_date, end_date)) result = cursor.fetchone() if (list(result.values()))[0] == 1: record_exist = 1 if record_exist == 1: select_sql = "SELECT customer_id,customer_name FROM `cai_niao_watermark_dailylogs` WHERE `stock_central_amt` > 0 AND `created_at` BETWEEN %s AND %s" cursor.execute(select_sql, (start_date, end_date)) dbs_response = cursor.fetchall() self.need_customer_data = list_dict_duplicate_removal( dbs_response) connection.commit() except Exception as ds: MessageToSlack.post_message_to_slack( "get_db_customer_data抛出异常:%s" % str(ds)) connection.close() # 多个数据库执行数据存储 def save_to_db(self): # save to dbs self.db_save_discount_rate( self.pro_db_host, self.pro_db_user, self.pro_db_pwd, self.pro_db_name) self.db_save_discount_rate( self.u02_db_host, self.u02_db_user, self.u02_db_pwd, self.u02_db_name) self.db_save_discount_rate( self.u01_db_host, self.u01_db_user, self.u01_db_pwd, self.u01_db_name) # 将获取到的信息存储到数据库中 def db_save_discount_rate(self, db_host, db_user, db_pwd, db_name): discount_rate_data = self.discount_rate_data # MessageToSlack.post_message_to_slack("discount_rate_data:") # MessageToSlack.post_message_to_slack(discount_rate_data) if len(discount_rate_data) > 0: try: connection = pymysql.connect( host=db_host, user=db_user, password=db_pwd, database=db_name, cursorclass=pymysql.cursors.DictCursor) MessageToSlack.post_message_to_slack( "connect with db successfully") with connection: for data in discount_rate_data: download_date = datetime.datetime.now().strftime('%Y-%m-%d') created_at = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S') updated_at = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S') customer_id = data['customer_id'] customer_name = data['customer_name'] all_assets_value = data['allAssetsValue'] assets_receivable_value = data['assetsReceivableValue'] central_good_value = data['centralGoodValue'] central_pledge_rate = int(data['centralPledgeRate']) # int currency = data['currency'] discount_assets_receivable_value = data['discountAssetsReceivableValue'] goods_owner_name = data['goodsOwnerName'] goods_value = data['goodsValue'] head_trip_pledge_rate = int(data['headTripPledgeRate']) # int head_trip_value = data['headTripValue'] pledge_goods_value = data['pledgeGoodsValue'] # int pledge_rate = int(data['pledgeRate']) receivable_pledge_rate = int(data['receivablePledgeRate']) # int store_name = data['storeName'] store_t_b_id = data['storeTBId'] record_exist = 0 with connection.cursor() as cursor: check_sql = "SELECT EXISTS ( SELECT `download_date` FROM `cai_niao_company_discountrate_dailylogs` WHERE `download_date` = %s and `customer_id` = %s and `store_t_b_id` = %s)" cursor.execute( check_sql, (download_date, customer_id, store_t_b_id)) result = cursor.fetchone() if (list(result.values()))[0] == 1: record_exist = 1 if record_exist == 0: MessageToSlack.post_message_to_slack( "Not found today record, insert that record") # 执行插入数据 # 新增语句 insert_sql = f"Insert INTO `cai_niao_company_discountrate_dailylogs` (`download_date`,`created_at`,`customer_id`,`customer_name`,`all_assets_value`,`assets_receivable_value`,`central_good_value`,`central_pledge_rate`,`currency`,`discount_assets_receivable_value`,`goods_owner_name`,`goods_value`,`head_trip_pledge_rate`,`head_trip_value`,`pledge_goods_value`,`pledge_rate`,`receivable_pledge_rate`,`store_name`,`store_t_b_id`) VALUES ('{download_date}', '{created_at}', '{customer_id}', '{customer_name}', '{all_assets_value}', '{assets_receivable_value}', '{central_good_value}', {central_pledge_rate}, '{currency}', '{discount_assets_receivable_value}', '{goods_owner_name}', '{goods_value}', {head_trip_pledge_rate}, '{head_trip_value}', '{pledge_goods_value}', {pledge_rate}, {receivable_pledge_rate}, '{store_name}', '{store_t_b_id}')" MessageToSlack.post_message_to_slack("insert_sql: %s" % insert_sql) with connection.cursor() as cursor: cursor.execute(insert_sql) elif record_exist == 1: MessageToSlack.post_message_to_slack( "Today record is existed, update that record") # 执行更新数据 # 更新语句 update_sql = f"UPDATE `cai_niao_company_discountrate_dailylogs` SET `updated_at` = '{updated_at}',`all_assets_value` = '{all_assets_value}',`assets_receivable_value` = '{assets_receivable_value}',`central_good_value` = '{central_good_value}',`central_pledge_rate` = {central_pledge_rate},`currency` = '{currency}',`discount_assets_receivable_value` = '{discount_assets_receivable_value}',`goods_owner_name` = '{goods_owner_name}',`goods_value` = '{goods_value}',`head_trip_pledge_rate` = {head_trip_pledge_rate},`head_trip_value` = '{head_trip_value}',`pledge_goods_value` = '{pledge_goods_value}',`pledge_rate` = {pledge_rate},`receivable_pledge_rate` = {receivable_pledge_rate},`store_name` = '{store_name}',`store_t_b_id` = '{store_t_b_id}' WHERE `download_date` = '{download_date}' and `customer_id` = '{customer_id}' and `store_t_b_id` = '{store_t_b_id}'" MessageToSlack.post_message_to_slack("update_sql: %s" % update_sql) with connection.cursor() as cursor: cursor.execute(update_sql) connection.commit() MessageToSlack.post_message_to_slack( "Inserted into %s" % db_name) except Exception as ds: MessageToSlack.post_message_to_slack( "db_save_discount_rate抛出异常:%s" % str(ds)) connection.close() else: MessageToSlack.post_message_to_slack( 'No data need to store as no data grabbed') MessageToSlack.post_message_to_slack("Bye") # 关闭所有的浏览器窗口 def teardown_method(self): # self.driver.quit() MessageToSlack.post_message_to_slack( "浏览器要关闭时打开的页面的标题: %s" % self.driver.title) MessageToSlack.post_message_to_slack("Browser Close") # 列表里的字典元素去重复 def list_dict_duplicate_removal(data_list): def run_function(x, y): return x if y in x else x + [y] return reduce(run_function, [[], ] + data_list) if __name__ == '__main__': scrape_cainiao = Cainiao_Scrape( 'https://merchant.finance.cainiao.com/funds/signApproveDetail/SignApproveDetailManage.htm', config ('PRO_DB_HOST'), config ('PRO_DB_USERNAME'), config ('PRO_DB_PASSWORD'), config ('PRO_DB_DATABASE'), config ('U02_DB_HOST'), config ('U02_DB_USERNAME'), config ('U02_DB_PASSWORD'), config ('U02_DB_DATABASE'), config ('U01_DB_HOST'), config ('U01_DB_USERNAME'), config ('U01_DB_PASSWORD'), config ('U01_DB_DATABASE'), config ('CAINIAO_AC'), config ('CAINIAO_PWD') ) # 获取cai_niao_watermark_dailylogs表中stock_central_amt>0并且日期範圍由2年前的今天到今天的customer_name数据 scrape_cainiao.get_db_customer_data() # 浏览器配置函数 scrape_cainiao.setup_method('') # 链接菜鸟后台网址,并scrape相关数据 scrape_cainiao.login_cainiao () # 关闭所有的浏览器窗口 scrape_cainiao.teardown_method () # 数据存储到数据库的函数 scrape_cainiao.save_to_db ()