from selenium import webdriver from selenium.webdriver.common.desired_capabilities import DesiredCapabilities import csv #设置UA标识 dcap = dict(DesiredCapabilities.PHANTOMJS) dcap["phantomjs.page.settings.userAgent"] = ( "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36 " ) driver = webdriver.PhantomJS(executable_path=‘phantomjs‘, desired_capabilities=dcap) #构造百度音乐歌单首页 pageNum = 1 url = ‘%s%d%s‘ % (‘http://music.baidu.com/songlist/tag/%E5%85%A8%E9%83%A8?orderType=1&offset=‘, 20*pageNum, ‘&third_type=‘) #创建歌单csv文件 csv_file = open("songlist.csv", "w", newline=‘‘, encoding=‘utf_8_sig‘) write = csv.writer(csv_file) write.writerow([‘标题‘, ‘播放数‘, ‘封面链接‘, ‘歌单链接‘]) while 1: #找到歌单内容存放位置 driver.get(url) data = driver.find_element_by_class_name("songlist-list").find_element_by_tag_name("ul"). find_elements_by_tag_name("li") #最后一页跳出循环 if len(data) == 0: break #解析当前页歌单列表 for i in range(len(data)): num = data[i].find_element_by_class_name("num").text #获取播放数大于500万的歌单 if ‘万‘ in num and int(num.split("万")[0]) > 200: wrap = data[i].find_element_by_class_name("img-wrap"). find_element_by_tag_name("img") songListTitle = data[i].find_element_by_class_name("text-title"). find_element_by_tag_name("a") #输出歌单名等信息 print(songListTitle.get_attribute(‘title‘), num, wrap.get_attribute(‘src‘), songListTitle.get_attribute(‘href‘)) #写入csv文件 write.writerow([songListTitle.get_attribute(‘title‘), num, wrap.get_attribute(‘src‘), songListTitle.get_attribute(‘href‘)]) #构造下一页url pageNum += 1 url = ‘%s%d%s‘ % (‘http://music.baidu.com/songlist/tag/%E5%85%A8%E9%83%A8?orderType=1&offset=‘, 20*pageNum, ‘&third_type=‘) csv_file.close()
1 加载浏览器驱动: webdriver.PhantomJS() 2 打开页面:get() 3 关闭浏览器:quit() 4 最大化窗口: maximize_window() 5 设置窗口参数:set_window_size(600,800) 6 后退到前一页: back() 7 前进到后一页: forward() 8 刷新页面: refresh()
id定位:find_element_by_id() name定位:find_element_by_name() class定位:find_element_by_class() tag定位:find_element_by_tag_name() link定位:find_element_by_link_text() partial link 定位: find_element_by_partial_link_text()
绝对路径:find_element_by_xpath("绝对路径") 元素属性:find_element_by_xpath("//unput[@id=‘kw‘]") 层级与属性结合:find_element_by_xpath("//form[@id=‘loginForm‘]/ul/input[1]") 逻辑运算符:find_element_by_xpath("//input[@id=‘kw‘ and@class=‘s_ipt‘]")
find_element_by_css_selector()
find_elements_by_id() find_elements_by_name() find_elements_by_class() find_elements_by_tag_name() find_elements_by_link_text() find_elements_by_partial_link_text() find_elements_by_xpath() find_elements_by_css_selector()
原文:https://www.cnblogs.com/dmpang/p/9195919.html