专题文章日益增多，想通过首页的abstract来大体了解文章内容，存储文章链接，从而有目的的阅读，提升效率。
利用Scrapy爬取专题作业，将爬取的数据存入Mongodb。

1. 确定爬取信息：

文章题目+简介+作者+时间+字数+（阅读次数+评论数量+喜欢人数）--PS：后三个用re或者Xpath都没有成功抓取，爬取到的数据都为空，目前还没找到原因。。。

image.png
image.png

item.py文件：

import scrapy

class CrazydataanalyzeItem(scrapy.Item):
    # define the fields for your item here like:
    # name = scrapy.Field()
    article_url = scrapy.Field()
    title = scrapy.Field()
    abstract = scrapy.Field()
    nickname = scrapy.Field()
    publish_time = scrapy.Field()
    wordage = scrapy.Field()
    views_count = scrapy.Field()
    comments_count = scrapy.Field()
    likes_count = scrapy.Field()
    pass

2. 编辑爬虫程序

确定title/abstract/nickname以及子网页的url：base_url+href

image.png
爬取发表日期以及字数
PS：wordage与views-count/comments-count/likes-count同样的结构，为什么可以爬取字数，但阅读数/评论数/喜欢数都爬取不到呢？？？

CrazyData.py

from scrapy.spiders import CrawlSpider  # spiders 加s
from scrapy.selector import Selector
from CrazyDataAnalyze.items import CrazydataanalyzeItem
from scrapy.http import Request
import re
import requests

class CrazyData(CrawlSpider):
    name = 'CrazyData'
    start_urls = ['https://www.jianshu.com/c/af12635a5aa3?order_by=added_at&page=1']

    def parse(self, response):
        base_url = 'https://www.jianshu.com'
        selector = Selector(response)
        infos = selector.xpath('//ul[@class="note-list"]/li')
        for info in infos:
            #print(info.xpath('div/a/@href'))
            #print(info.xpath('div/a/@href').extract())
            #print(info.xpath('div/a/@href').extract()[0])
            #print(info.xpath('div/a/@href')[0].extract())
            #print(info.xpath('div/a/@href')[0].extract()[0])
            article_url = base_url + info.xpath('div/a/@href').extract()[0]
            title = info.xpath('div/a/text()').extract()[0]
            abstract = info.xpath('div/p/text()').extract()[0]
            nickname = info.xpath('div/div/a/text()').extract()[0]
            yield Request(article_url,meta={'article_url':article_url,'title':title,'abstract':abstract,'nickname':nickname},callback=self.parse_item)
        urls = ['https://www.jianshu.com/c/af12635a5aa3?order_by=added_at&page={}'.format(str(i)) for i in range(2,20)]
        for url in urls:
            yield Request(url,callback=self.parse)


    def parse_item(self,response):
        item = CrazydataanalyzeItem()
        item['article_url'] = response.meta['article_url']
        item['title'] = response.meta['title']
        item['abstract'] = response.meta['abstract']
        item['nickname'] = response.meta['nickname']

        try:
            #url1 = response.meta['article_url']
            '''headers = {
                'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36'
            }'''
            #html = requests.get(item['article_url'], headers=headers)
            selector = Selector(response)
            publish_time = selector.xpath('//div[@class="meta"]/span[1]/text()').extract()[0]
            wordage = selector.xpath('//div[@class="meta"]/span[2]/text()').extract()[0]
            #views_count = re.findall('<span class="views-count">(.*?)</span>',html.text,re.S)
            #comments_count = re.findall('"comments-count">(.*?)</span>',html.text,re.S)
            #likes_count = re.findall('"likes-count">(.*?)</span>',html.text,re.S)
            views_count = selector.xpath('//div[@class="meta"]/span[3]/text()').extract()
            comments_count = selector.xpath('//div[@class="meta"]/span[4]/text()').extract()
            likes_count = selector.xpath('//div[@class="meta"]/span[5]/text()').extract()
            item['publish_time'] = publish_time
            item['wordage'] = wordage
            item['views_count'] = views_count
            item['comments_count'] = comments_count
            item['likes_count'] = likes_count
            yield item
        except IndexError:
            pass

3.存入MongoDB数据库（Just为了练习下数据存储）

pipelines.py

import pymongo

class CrazydataanalyzePipeline(object):
    def __init__(self):
        client = pymongo.MongoClient('localhost',27017)
        test = client['test']
        CrazyData = test['CrazyData']
        self.post = CrazyData
    def process_item(self, item, spider):
        info = dict(item)
        self.post.insert(info)
        return item

4.setting设置

settings.py

BOT_NAME = 'CrazyDataAnalyze'
SPIDER_MODULES = ['CrazyDataAnalyze.spiders']
NEWSPIDER_MODULE = 'CrazyDataAnalyze.spiders'
USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36'
ROBOTSTXT_OBEY = False
DOWNLOAD_DELAY = 3
ITEM_PIPELINES = {
    'CrazyDataAnalyze.pipelines.CrazydataanalyzePipeline': 300,
}

5.MAIN文件

main.py

from scrapy import cmdline
cmdline.execute("scrapy crawl CrazyData".split())

6.MongoDB

可以通过命令行导出CSV文件

mongoexport -d test -c CrazyData --type=csv -f article_url,title,abstract,nickname,publish_time,wordage -o crazydata.csv

并通过EXCEL中=HYPERLINK()函数，直接将url转换为超链接~

image.png

这样就可以通过简介大体了解文章内容了，也希望各位小伙伴多多采用开门见山的方法，加油加油~元旦即将来临！