python爬虫——爬取古诗词

一. 概要

1.通过python爬虫循环爬取古诗词网站唐诗宋词
2.落地到本地数据库

二. 页面分析

首先通过firedebug进行页面定位：

其次源码定位：

最终生成lxml etree定位div标签源码：

# 通过 lxml进行页面分析
​response = etree.HTML(data)
# div层定位
for row in response.xpath('//div[@class="left"]/div[@class="sons"]'):
    # 标题定位
    title = row.xpath('div[@class="cont"]/p/a/b/text()')[0] if row.xpath('div[@class="cont"]/p/a/b/text()') else ''
    # 朝代定位
    dynasty = row.xpath('div[@class="cont"]/p[@class="source"]//text()')[0] if row.xpath('div[@class="cont"]/p[@class="source"]//text()') else ''
    # 诗人定位
    author = row.xpath('div[@class="cont"]/p[@class="source"]//text()')[-1] if row.xpath('div[@class="cont"]/p[@class="source"]//text()') else ''
    # 内容定位
    content = ''.join(row.xpath('div[@class="cont"]/div[@class="contson"]//text()')).replace('　　', '').replace('\n', '') if row.xpath('div[@class="cont"]/div[@class="contson"]//text()') else ''
    # 标签定位
    tag = ','.join(row.xpath('div[@class="tag"]/a/text()')) if row.xpath('div[@class="tag"]/a/text()') else ''

三. 执行结果

四. 脚本源码

#!/usr/bin/env python
# -*- coding: utf-8 -*-
'''
@Date    : 2017/12/21 11:12
@Author  : kaiqing.huang 
@File    : shigeSpider.py
'''
from utils import MySpider, MongoBase
from datetime import date
from lxml import etree
import sys

class shigeSpider():
    def __init__(self):
        self.db = MongoBase()
        self.spider = MySpider()

    def download(self, url):
        self.domain = url.split('/')[2]
        data = self.spider.get(url)
        if data:
            self.parse(data)

    def parse(self, data):
        response = etree.HTML(data)
        for row in response.xpath('//div[@class="left"]/div[@class="sons"]'):
            title = row.xpath('div[@class="cont"]/p/a/b/text()')[0] if row.xpath('div[@class="cont"]/p/a/b/text()') else ''
            dynasty = row.xpath('div[@class="cont"]/p[@class="source"]//text()')[0] if row.xpath('div[@class="cont"]/p[@class="source"]//text()') else ''
            author = row.xpath('div[@class="cont"]/p[@class="source"]//text()')[-1] if row.xpath('div[@class="cont"]/p[@class="source"]//text()') else ''
            content = ''.join(row.xpath('div[@class="cont"]/div[@class="contson"]//text()')).replace('　　', '').replace('\n', '') if row.xpath('div[@class="cont"]/div[@class="contson"]//text()') else ''
            tag = ','.join(row.xpath('div[@class="tag"]/a/text()')) if row.xpath('div[@class="tag"]/a/text()') else ''
            self.db.add_new_row('shigeSpider', { 'title': title, 'dynasty': dynasty, 'author': author, 'content': content, 'tag': tag, 'createTime': str(date.today()) })
            print 'Title: {}'.format(title)
        if response.xpath('//div[@class="pages"]/a/@href'):
            self.download('http://' + self.domain + response.xpath('//div[@class="pages"]/a/@href')[-1])

if __name__ == '__main__':
    sys.setrecursionlimit(100000)
    url = 'http://so.gushiwen.org/type.aspx'
    do = shigeSpider()
    do.download(url)

转载于:https://blog.51cto.com/8456082/2052891