| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121 | import timeimport jsonfrom selenium import webdriverfrom selenium.webdriver.common.desired_capabilities import DesiredCapabilitiesimport timeimport osimport urllib.parsefrom selenium.webdriver.support.ui import WebDriverWaitfrom selenium.webdriver.common.by import Byfrom selenium.webdriver.support import expected_conditions as ECimport codecsimport randomimport requestsimport datetimeimport datasetimport timeimport tracebackimport sysimport fire#from INNNews import notifytestdb = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')table=db['general_log']  driver=Noneheaders = {        "Authorization": "Bearer " + "6SDULL1Ebklduc6TFxa97AFto5Sj21kyJ30CxiLiSoi",        "Content-Type": "application/x-www-form-urlencoded"}def send_msg(kw):    params = {"message": "error: "+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[1]    global driver    googleurl = 'https://www.google.com/search?q={}&num={}&hl={}'.format(urllib.parse.quote(q), 100,'zh-TW')    print(googleurl)    driver.get(googleurl)    time.sleep(6)    elmts=driver.find_elements(By.XPATH,"//div[@class='yuRUbf']/a")    idx=1    ranking=-1    print(len(elmts))#    driver.save_screenshot('c:/tmp/test.png')    for elmt in elmts:        href=elmt.get_attribute('href')        txt=elmt.text        if len(txt)>10:            if domain in href:                print('clicked....')                print(href)                print(txt)                print("ranking", idx)                table.insert({'kw':q,'domain':domain,'ranking':idx,'title':txt,'url':href,'dt':datetime.datetime.now()})                webdriver.ActionChains(driver).move_to_element(elmt).perform()                webdriver.ActionChains(driver).move_to_element(elmt).click().perform()                break        idx+=1def run_once(q):    global driver    result=[]    options = webdriver.ChromeOptions()    options.add_argument('--headless')    options.add_argument('--disable-dev-shm-usage')    options.add_argument('--no-sandbox')#    options.add_argument('--remote-debugging-port=9222')    epath=os.environ['WEBDRIVER']#    options.add_experimental_option("debuggerAddress", "192.168.192.45:9922")    options.add_experimental_option("debuggerAddress", q[2])#    options.add_argument("--user-agent=" +user_agent)    options.add_argument("--incognito")#    driver = webdriver.Chrome(executable_path=r'C:\portable\webdriver\chrome98\chromedriver.exe',options=options)    driver = webdriver.Chrome(executable_path=epath,options=options)    driver.delete_all_cookies()    driver.set_window_size(1400,1000)    print(q)    process_query(q)    time.sleep(3)    driver.quit()class JParams(object):  def get(self, kw,domain,address):    try:        print(kw)        print(domain)        run_once( (kw,domain,address)   )    except:        traceback.print_exc()        send_msg('SEO docker exception... stop')if __name__ == '__main__':  fire.Fire(JParams)
 |