123456789101112131415161718192021222324252627282930313233343536373839404142434445464748 |
- 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 codecs
- import random
- def process_query(q):
- googleurl='https://www.google.com/search?q='+urllib.parse.quote(q)
- driver = webdriver.Remote(
- command_executor='http://127.0.0.1:4444/wd/hub',
- desired_capabilities=DesiredCapabilities.CHROME)
- driver.set_window_size(1400,1000)
- driver.get(googleurl)
- 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:
- webdriver.ActionChains(driver).move_to_element(elmt).perform()
- webdriver.ActionChains(driver).move_to_element(elmt).click().perform()
- break
- fw=codecs.open('c:/tmp/gg.html','w','utf-8')
- fw.write(driver.page_source)
- fw.close()
- time.sleep(10)
- # driver.save_screenshot("c:/tmp/screen.png")
- # os.startfile("c:/tmp/screen.png")
- driver.quit()
- qlist=['上築空間設計','豐聚室內裝修','富億空間設計','浩室設計','京璽國際','IS國際設計','優尼客','摩登雅舍','亞維空間設計','比沙列','富億設計','豐聚設計']
- while True:
- q=random.choice(qlist)
- print(q)
- process_query(q)
|