#import redis import time import traceback #import json from selenium import webdriver from selenium.webdriver.common.desired_capabilities import DesiredCapabilities import time import os import urllib.parse from selenium.webdriver.support.ui import WebDriverWait from selenium.webdriver.common.by import By from selenium.webdriver.support import expected_conditions as EC import dataset import codecs import random import requests import time import sys #import docker import codecs import random import os import time driver=None headers = { "Authorization": "Bearer " + "t35vhZtWNgvDNWHc3DJh0OKll3mcB9GvC8K2EAkBug2", "Content-Type": "application/x-www-form-urlencoded" } def send_msg(kw): params = {"message": "處理關鍵字: "+kw} r = requests.post("https://notify-api.line.me/api/notify",headers=headers, params=params) def empty_query(q): global driver googleurl='https://www.google.com/search?q='+urllib.parse.quote(q) driver.get(googleurl) time.sleep(3) def process_query(qs): q=qs[0] domain=qs[2] cnt=qs[1] global driver googleurl='https://www.google.com/search?q='+urllib.parse.quote(q) driver.get(googleurl) time.sleep(3) if cnt > 0: for i in range(cnt): elmt=driver.find_element_by_xpath("//a[@id='pnnext']") webdriver.ActionChains(driver).move_to_element(elmt).perform() webdriver.ActionChains(driver).move_to_element(elmt).click().perform() time.sleep(3) elmts=driver.find_elements_by_xpath("//div[@class='g']//div[@class='yuRUbf']//a") idx=1 ranking=-1 for elmt in elmts: href=elmt.get_attribute('href') txt=elmt.text if len(txt)>10: # if 'hhh.com.tw' in href: # if 'hhh.com.tw' in href: # if 'ai.choozmo.com' in href: if domain in href: # if 'searchome.net' in href: webdriver.ActionChains(driver).move_to_element(elmt).perform() webdriver.ActionChains(driver).move_to_element(elmt).click().perform() break def re_get_webdriver(): global driver result=[] # client = docker.from_env() # ls=client.containers.list() # print(ls) # for l in ls: # if 'p17777' in l.name: # ls[0].restart() # time.sleep(4) if driver is not None: print('closing....') driver.quit() # options = webdriver.EdgeOptions() try: # driver = webdriver.Remote( # command_executor='http://127.0.0.1:4444/wd/hub') chrome_options = webdriver.ChromeOptions() # chrome_options.add_argument('--proxy-server=host.docker.internal:'+str(proxyport)) # Specify your Kubernetes service-name here chrome_options.add_argument('--ignore-certificate-errors') chrome_options.add_argument("--no-sandbox") chrome_options.add_argument("--headless") chrome_options.add_argument("--disable-dev-shm-usage") driver = webdriver.Chrome(options=chrome_options) # command_executor='http://127.0.0.1:4444/wd/hub', # desired_capabilities=chrome_options.to_capabilities() # ) driver.set_window_size(1400,1000) return driver except: import traceback traceback.print_exc() driver=None return None driver=None def run_once(url): global driver i=random.randint(0,22) if i==0 or driver is None: # if True: time.sleep(6) re_get_webdriver() time.sleep(1) if driver is None: return try: driver.execute_script('window.open("'+url+'","_blank");') driver.execute_script("window.scrollTo(0, window.scrollY + 400)") time.sleep(0.3) except: print('exception') #client = docker.from_env() #ls=client.containers.list() #print(ls) #for l in ls: # if 'p4444' in l.name: # ls[0].restart() #time.sleep(4) fpath=__file__ #elmts=fpath.split('\\') #fpath=fpath.replace(elmts[-1],'urls.csv') os.system('docker restart p9222') print(fpath) lst=[] db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4') cursor=db.query('SELECT * FROM columnids order by rand()') for c in cursor: lst.append('https://www.hhh.com.tw/columns/detail/'+str(c['cid'])+'/index.php') #lst=['https://www.hhh.com.tw/columns/detail/3427/index.php'] #for i in range(20): while True: l=random.choice(lst) print(l) # driver.get(l) run_once(l) # time.sleep(5) # time.sleep(2)