| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157 | 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.chrome.service import Servicefrom selenium.webdriver.common.keys import Keysfrom selenium.webdriver.support import expected_conditions as ECimport codecsimport randomimport requestsimport datetimeimport datasetimport timeimport tracebackimport sysimport fireimport redisdriver = Nonedef rua():    pool = [        "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:73.0) Gecko/20100101 Firefox/73.0",        "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:76.0) Gecko/20100101 Firefox/76.0",        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.88 Safari/537.36"        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.101 Safari/537.36",        "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36",        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.97 Safari/537.36",        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36 OPR/68.0.3618.125",    ]    return random.choice(pool)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):    db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')    table=db['general_log']    q=qs[0]    domain=qs[1]    global driver    googleurl = 'https://www.google.com/?num=100'    driver.get(googleurl)    time.sleep(6)    send_kw_elmt = driver.find_element(By.XPATH, '/html/body/div[1]/div[3]/form/div[1]/div[1]/div[1]/div/div[2]/input')    send_kw_elmt.send_keys(q)    time.sleep(3)    send_kw_elmt.send_keys(Keys.ENTER)    time.sleep(6)    print(driver.current_url)    elmts=driver.find_elements(By.XPATH,"//div[@class='yuRUbf']/a")    idx=1    ranking=-1    print('網頁數量',len(elmts))#    driver.save_screenshot('c:/tmp/test.png')    if 'site' in q:        href = elmts[0].get_attribute('href')        txt = elmts[0].text        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(elmts[0]).perform()        webdriver.ActionChains(driver).move_to_element(elmts[0]).click().perform()        time.sleep(5)    for elmt in elmts:        href=elmt.get_attribute('href')        txt=elmt.text        if len(txt)>10:            if domain in href:                print('clicked....')                print('點擊網址',href)                print('標題',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()                time.sleep(5)                break        idx+=1    db.close()def run_once(q):    global driver    result=[]    s = Service('/root/driver/chromedriver')    user_agent = rua()    options = webdriver.ChromeOptions()    options.add_argument('--headless')    options.add_argument('--remote-debugging-port=9222')    options.add_experimental_option("debuggerAddress", "127.0.0.1:9922")    # options.add_argument("--user-agent=" +user_agent)    options.add_argument("--incognito")    r = redis.Redis(host='db.ptt.cx', port=6379, db=2,password='choozmo9')    data=r.get('google_proxy')    jstext=data.decode('utf-8')    jsobj=json.loads(jstext)    proxy=random.choice(jsobj)    print('Freeproxy',proxy)    change_ip = ["'--proxy-server='+proxy","--proxy-server=socks5://127.0.0.1:9050","--proxy-server=socks5://192.53.174.202:8180"]    options.add_argument('--proxy-server=socks5://192.53.174.202:8180')    driver = webdriver.Chrome(    options=options,service=s)    if 'sorry' in driver.current_url:        r = redis.Redis(host='db.ptt.cx', port=6379, db=2,password='choozmo9')        data=r.get('google_proxy')        jstext=data.decode('utf-8')        jsobj=json.loads(jstext)        print('Free proxy',jsobj)        proxy=random.choice(jsobj)        change_ip_list = ['--proxy-server=%s' % proxy,"--proxy-server=socks5://127.0.0.1:9050","--proxy-server=socks5://192.53.174.202:8180"]        change_ip = random.choice(change_ip_list)        options.add_argument(change_ip)        print('使用代理ip',change_ip)            driver = webdriver.Chrome(options=options,service=s)    driver.delete_all_cookies()    driver.set_window_size(1400,1000)    process_query(q)    time.sleep(3)    driver.quit()#for c in lst:#while True:#    try:#        c=random.choice(lst)#    except:#        traceback.print_exc()#    sleepint=random.randint(320,520)#    time.sleep(sleepint)class JParams(object):  def get(self, kw,domain,port):    run_once( (kw,domain,port) )if __name__ == '__main__':  fire.Fire(JParams)
 |