python歌词解析成类 python爬歌词

首先导入的库当然是selenium，为了设置超时，需要导入time

from selenium import webdriver
import time
from selenium.webdriver.common.keys import Keys

导入之后可以打开网易主页

driver = webdriver.Chrome()
driver.get(u"http://music.163.com/")
print "正在打开网页",driver.title

等待页面加载，选择爬陈粒的网易歌词第一页

time.sleep(2)
name = u"陈粒"
driver.find_element_by_css_selector("input[class=\"txt j-flag\"]").send_keys(name)
driver.find_element_by_css_selector("input[class=\"txt j-flag\"]").send_keys(Keys.ENTER)

转到名称为contengframe的框架上来

driver.switch_to_frame("contentFrame")

爬前三十首歌，定义一下fun函数

def fun(a):
    print "正在打开歌曲页面"
    time.sleep(2)
    try:
        driver.find_element_by_css_selector("a[href^='javascript:void(0)']").click()
    except:   
        js="var q=document.body.scrollTop=500"
        driver.execute_script(js)
        time.sleep(1)
        driver.find_element_by_css_selector("a[href^='javascript:void(0)']").click()
    else:
        driver.implicitly_wait(10) # seconds
        driver.execute_script('window.stop()')
        
    print "正在获取歌词"
    text = driver.find_element_by_id("lyric-content").text
    text_all = text[:-2]
    name = driver.find_element_by_class_name("tit").text
    print "正在新建文件保存"
    f=open('C:/Users/user/Desktop/lyric/%s.txt'%name.replace("\n", "@@").split("@@")[0],'w')
    f.write(text_all.encode('utf8'))
    print "正在保存歌词",name
    f.close()
    driver.back()
    driver.switch_to_frame("contentFrame")
    time.sleep(2)

此函数为打开歌曲页面之后的操作，然而在打开之前会有很多其他的操作

首先确认该歌曲歌手是不是我们搜索的名字，试图点击是在界面刷新之后的界面，当歌曲链接不在此界面上，则下拉，另外还会出现网易的登录界面如下，此时必须回到主窗口关闭掉再回到刚才的contentframe窗口继续爬取，具体代码如下，另外中间可能会有一些错误出现，可能是由于页面加载不及时等原因，此时重新运行即可

for i in range(0,30):
    a=i+1
    print "第",a,"首"
    if name in driver.find_element_by_css_selector('div#m-search>div:nth-child(3)>div>div>div:nth-child(%d)>div:nth-child(4)'%a).text:
        try:
            print "试图点击"
            driver.find_element_by_css_selector('div#m-search>div:nth-child(3)>div>div>div:nth-child(%d)>div:nth-child(2)>div>div>a'%a).click()
            fun(a)
        except:
            try:
                print "试图下拉点击"
                js="var q=document.body.scrollTop=600"
                driver.execute_script(js)
                time.sleep(1)
                driver.find_element_by_css_selector('div#m-search>div:nth-child(3)>div>div>div:nth-child(%d)>div:nth-child(2)>div>div>a'%a).click()
                fun(a)
            except:
                try:
                    print "试图二次下拉点击"
                    js="var q=document.body.scrollTop=1000"
                    driver.execute_script(js)
                    time.sleep(1)
                    driver.find_element_by_css_selector('div#m-search>div:nth-child(3)>div>div>div:nth-child(%d)>div:nth-child(2)>div>div>a'%a).click()
                    fun(a)
                except:
                    print "试图回到主页面关闭窗口"
                    driver.switch_to.default_content()
                    driver.find_element_by_class_name('zcls').click()
                    driver.switch_to_frame("contentFrame")
                    js="var q=document.body.scrollTop=1000"
                    driver.execute_script(js)
                    time.sleep(1)
                    driver.find_element_by_css_selector('div#m-search>div:nth-child(3)>div>div>div:nth-child(%d)>div:nth-child(2)>div>div>a'%a).click()
                    fun(a)                   
    else:
        pass