萌新：使用xpath和正则表达式解析网页内容

代码如下：

import sqlite3
import re
import requests
from lxml import html

findlink = re.compile(r'<a href="(.*?)"')  # 创建正则表达式对象，表示规则（字符串的模式）
findname = re.compile(r'<a href=".*?">(.*?)</a>')
findname2 = re.compile(r'<td style="outline: 0px !important;">(.*?)</td>')
findname3 = re.compile(
    r'<td style="outline: 0px !important;"><p style="line-height: 1.8; outline: 0px !important;">(.*?)</p></td>')
findname4 = re.compile(
    r'<td style="outline: 0px !important;"><p style="line-height: 1.8; outline: 0px !important;"><a href=".*?">(.*?)</a>.*?</p></td>')
findaddres = re.compile(r'<td style="outline: 0px !important;">(.*?)</td>')
findadress1 = re.compile(r'<td style="outline: 0px !important;"><a href=".*?">(.*?)</a></td>')

'''
通过findall找到所有table里面的tr
然后对tr里面的内容进行解析，如果没有链接，则data添加信息为空，有链接调用函数来解析链接网页
再向数据库传输解析内容
'''

def main():
    basicurl = "http://www.qianmu.org/ranking/1528.htm"
    datalist = getData(basicurl)
    for data in datalist:
        print(data)
    saveDatadb(datalist,"university.db")

# 得到一个指定的网页内容
def askURL(url):
    et = html.etree
    respon = requests.get("http://www.qianmu.org/ranking/1528.htm")
    selector = et.HTML(respon.text)
    return selector

# 爬取主网页，将网页的tr提取出来进行分析
def getData(basicurl):
    datalist = []
    selector = askURL(basicurl)
    # 找出每个tr,对每个tr解析
    trs = selector.xpath('//div[@class="rankItem"]//tr[position()>1]')
    # names = selector.xpath('//div[@class="rankItem"]//tr[position()>1]/td/a/text() | //div[@class="rankItem"]//tr['
    #                        'position()>1]/td[2]/text()')
    # links = selector.xpath('//div[@class="rankItem"]//tr[position()>1]/td/a/@href')
    # 获得了每一个tr内容
    for tr in trs:
        data = []
        tr = html.tostring(tr, encoding='utf-8').decode('utf-8')
        name = re.findall(findname, tr)
        name1 = re.findall(findname2, tr)
        if len(name) == 0:
            name = name1[1]
        else:
            name = name[0]
        data.append(name)

        # 获取英文名字
        if len(re.findall(findname4, tr)) > 1 or len(re.findall(findname4, tr)) == 1:
            english = ''.join(re.findall(findname4, tr)[0])
        else:
            english = re.findall(findname3, tr)[1]
        data.append(english)

        if len(re.findall(findadress1, tr)) > 1:
            address = ''.join(re.findall(findadress1, tr)[1])
        else:
            address = re.findall(findaddres, tr)[3]
        data.append(address)
        link = re.findall(findlink, tr)
        # if len(link) > 1:
        #     link = link[0]
        # elif len(link) == 0:
        #     link = ' '
        # else:
        #     link = ''.join(link)
        # 开始对link进行分析

        if len(link) > 1:
            link = link[0]
        elif len(link) == 0:
            link = ' '
        else:
             link = ''.join(link)
        data.append(link)
        datalist.append(data)
    return datalist

# 保存数据
def saveDatadb(datalist, dbpath):
    init_db(dbpath)
    conn = sqlite3.connect(dbpath)
    cur = conn.cursor()  # 获取游标
    # print("我执行了")
    for data in datalist:
        for index in range(len(data)):
            data[index] = '"' + str(data[index]) + '"'  # '"'+data[index]+'"'
        sql = '''
            insert into university(
            name, ename, address, link) 
            values (%s)''' % ",".join(data)
        # print(sql)
        cur.execute(sql)
        conn.commit()  # 提交
    cur.close()
    conn.close()  # 关闭链接

# 创建数据库
def init_db(dbpath):
    sql = '''
        create table university(
        id integer primary key autoincrement,
        name text ,
        ename text ,
        address text ,
        link text
        );
    '''
    conn = sqlite3.connect(dbpath)  # 建表
    cursor = conn.cursor()  # 游标
    cursor.execute(sql)  # 执行sql语句建表
    conn.commit()  # 提交
    conn.close()  # 关闭

if __name__ == "__main__":  # 当程序执行时，调用函数  这样写的目的是严格控制函数执行的主流程
    main()

执行后保存的数据结果：

因有几个name正则表达式未处理妥当，可能会出现错误，可以自行调试，谅解，我懒。