news_seo1/from_news.py

#import redis
import time
import traceback
#import json
from selenium import webdriver
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
import time
import os
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
from selenium.webdriver.common.keys import Keys
import json
import random
import time
import redis
import sys
import codecs
import random
import os
import time
import requests
from time import sleep
#from wrapt_timeout_decorator import *
import pymysql
pymysql.install_as_MySQLdb()

driver=None

# db = dataset.connect('mysql://choozmo:pAssw0rd@db.ptt.cx:3306/seo?charset=utf8mb4')

headers = {
        "Authorization": "Bearer " + "6SDULL1Ebklduc6TFxa97AFto5Sj21kyJ30CxiLiSoi",
        "Content-Type": "application/x-www-form-urlencoded"
}

def send_msg(kw):
    params = {"message":kw}  
    # r = requests.post("https://notify-api.line.me/api/notify",headers=headers, params=params)




#@timeout(20)
def re_get_webdriver():
    global port
    global driver
    result=[]
    if driver is not None:
        print('closing....')
        driver.quit()
        os.system('killall chrome')
        print('quit....')
        driver=None
    try:
        options = webdriver.ChromeOptions()
        options.add_argument("--no-sandbox")
        options.add_argument("--disable-dev-shm-usage")
        # options.add_argument("--headless")
#        options.add_argument('--proxy-server='+proxy)
#        options.add_argument('--proxy-server=socks5://172.104.93.163:41800')

#        options.add_argument('--proxy-server=socks5://127.0.0.1:9050')
#        options.add_argument('--proxy-server=%s' % proxy)
#        print('--proxy-server=%s' % proxy)
#        options.add_experimental_option("debuggerAddress", '127.0.0.1:9922')
#        options.add_argument("--user-agent=" +user_agent)
        options.add_argument("--incognito")
        driver=None
        try:
#            driver = webdriver.Chrome(options=options)
            if os.name=='nt':
                driver = webdriver.Chrome(executable_path='C:/portable/webdriver/chrome102/chromedriver.exe',options=options)
            else:
                # driver = webdriver.Chrome(options=options)
                driver = webdriver.Chrome(options=options,executable_path='/Users/zooeytsai/Downloads/chromedriver 4')
        except:
            traceback.print_exc()

            driver.quit()
            sys.exit()
            return
        driver.set_window_size(1400,1000)
        return
    except:
        traceback.print_exc()
        driver=None
        return None



def run_once(jsobj):

    # table=db['rank_detection']
    print(jsobj)
    global driver
    kw=jsobj['kw']

    i=100
    if driver is None:
        re_get_webdriver()
    if driver is None:
        return

    driver.get('https://news.google.com/topstories?hl=zh-TW&gl=TW&ceid=TW:zh-Hant')

    time.sleep(5)

    elmt = driver.find_element(By.XPATH, "//input[contains(@aria-label,'搜尋')]")
    print(elmt)

    time.sleep(1)
    elmt.send_keys(kw)
    elmt.send_keys(Keys.ENTER)
    time.sleep(7)


    elmts = driver.find_elements(By.XPATH, "//main//div[@class='xrnccd']//article/a[contains(@href,'./articles/')]")
    print(len(elmts))
    elmt=elmts[-1]
    print(elmt)
    webdriver.ActionChains(driver).move_to_element(elmt).perform()
    webdriver.ActionChains(driver).move_to_element(elmt).click().perform()
    time.sleep(8)
    driver.quit()
    sys.exit()
par1='tiny3'
port='9942'
os.system('killall chrome')
#os.system('docker container restart proxy1')
time.sleep(8)
#os.system('curl --socks5 choozmo:choozmo9@172.104.93.163:41800 http://www.google.com')
time.sleep(3)

# r=random.randint(0,1)
r=1
if r==0:
    prefix="site:hhh.com.tw "
    domain=['hhh.com.tw']
    positive=['北歐 風格 定義','','房 間隔 局','小 房間 設計','裝潢 費用','室內 裝修','設計 公司','透 天 裝潢','鄉村 風 裝潢']
elif r==1:
    prefix="site:gs-rack.com "
    domain=['gs-rack.com']
    positive=['神助','移動模組','移動貨架','無軌道','創新設計','專利移動櫃','重型架','物流設備','客戶實績','儲存方案','客戶服務','平台移動','重型電動','分離式','重型移動','冷庫重型移動','防爆倉','專利移動模組','輕中型移動貨架','傳統後推','少量多樣','大量少樣','重型貨品']
elif r==2:
    prefix="基金 "
    postfix=' site:ltn.com.tw'
    domain=['ltn.com.tw']
    positive=['對沖基金','俄烏戰爭','美國升息']


kw=random.choice(positive)
run_once({'domain':domain,'kw':kw+prefix})