一条爬虫抓取一个小网站所有数据

今天闲来无事，写一个爬虫来玩玩。在网上冲浪的时候发现了一个搞笑的段子网，发现里面的内容还是比较有意思的，于是心血来潮，就想着能不能写一个Python程序，抓取几条数据下来看看，一不小心就把这个网站的所有数据都拿到了。

别逗了(biedoul.com)

这个网站主要的数据都是详情在HTML里面的，可以采用lxml模块的xpath对HTML标签的内容解析，获取到自己想要的数据，然后再保存在本地文件中，整个过程是一气呵成的。能够抓取到一页的数据之后，加一个循环就可以抓取到所有页的数据，下面的就是数据展示。

数据展示

废话少说，直接上Python代码

import requests
import csv
from lxml import etree
import time


class Page:

    def __init__(self):
        self.pre_url = "https://www.biedoul.com"
        self.start_page = 1
        self.end_page = 15233

    def askHTML(self, current_page, opportunity):
        print(
            "=============================== current page => " + str(current_page) + "===============================")
        try:

            pre_url = self.pre_url + "/index/" + str(current_page)
            page = requests.get(url=pre_url)
            html = etree.HTML(page.content)
            articles = html.xpath('/html/body/div/div/div/dl')
            return articles
        except Exception as e:
            if opportunity > 0:
                time.sleep(500)
                print(
                    "=============================== retry => " + str(opportunity) + "===============================")
                return self.askHTML(current_page, opportunity - 1)
            else:
                return None

    def analyze(self, articles):
        lines = []
        for article in articles:
            data = {}
            data["link"] = article.xpath("./span/dd/a/@href")[0]
            data["title"] = article.xpath("./span/dd/a/strong/text()")[0]
            data["content"] = self.analyze_content(article)
            picture_links = article.xpath("./dd/img/@src")
            if (picture_links is not None and len(picture_links) > 0):
                # print(picture_links)
                data["picture_links"] = picture_links
            else:
                data["picture_links"] = []

            # data["good_zan"] = article.xpath("./div/div/a[@class='pinattn good']/p/text()")[0]
            # data["bad_bs"] = article.xpath("./div/div/a[@class='pinattn bad']/p/text()")[0]
            data["good_zan"] = self.analyze_zan(article, "good")
            # article.xpath("./div/div/a[@class='pinattn good']/p/text()")[0]
            data["bad_bs"] = self.analyze_zan(article, "bad")
            # article.xpath("./div/div/a[@class='pinattn bad']/p/text()")[0]
            lines.append(data)
        return lines

    # 解析文章内容
    def analyze_content(self, article):
        # 1. 判断dd标签下是否为文本内容
        content = article.xpath("./dd/text()")
        if content is not None and len(content) > 0 and not self.is_empty_list(content):
            return content

        content = []
        p_list = article.xpath("./dd")
        for p in p_list:
            # 2. 判断dd/.../font标签下是否为文本内容
            if len(content) <= 0 or content is None:
                fonts = p.xpath(".//font")
                for font_html in fonts:
                    font_content = font_html.xpath("./text()")
                    if font_content is not None and len(font_content) > 0:
                        content.append(font_content)

            # 3. 判断dd/.../p标签下是否为文本内容
            if len(content) <= 0 or content is None:
                fonts = p.xpath(".//p")
                for font_html in fonts:
                    font_content = font_html.xpath("./text()")
                    if font_content is not None and len(font_content) > 0:
                        content.append(font_content)

        return content

    def analyze_zan(self, article, type):
        num = article.xpath("./div/div/a[@class='pinattn " + type + "']/p/text()")
        if num is not None and len(num) > 0:
            return num[0]
        return 0

    def do_word(self):
        fieldnames = ['index', 'link', 'title', 'content', 'picture_links', 'good_zan', 'bad_bs']
        with open('article.csv', 'a', encoding='UTF8', newline='') as f:
            writer = csv.DictWriter(f, fieldnames=fieldnames)
            # writer.writeheader()
            for i in range(self.start_page, self.end_page):
                articles = self.askHTML(i, 3)
                if articles is None:
                    continue
                article_list = self.analyze(articles)
                self.save(writer, article_list)

    # 保存到文件中
    def save(self, writer, lines):
        print("##### 保存中到文件中...")
        # python2可以用file替代open
        print(lines)
        writer.writerows(lines)
        print("##### 保存成功...")

    def is_empty_list(self, list):
        for l in list:
            if not self.empty(l):
                return False
        return True

    def empty(self, content):
        result = content.replace("\r", "").replace("\n", "")
        if result == "":
            return True
        return False

    # 递归解析文章内容
    def analyze_font_content(self, font_html, depth):
        content = []
        print(depth)
        font_content_list = font_html.xpath("./font/text()")
        if font_content_list is not None and len(font_content_list) > 0 and not self.is_empty_list(font_content_list):
            for font_content in font_content_list:
                content.append(font_content)
        else:
            if depth < 0:
                return []
            return self.analyze_font_content(font_html.xpath("./font"), depth - 1)

        return content


if __name__ == '__main__':
    page = Page()
    page.do_word()