爬取网页文章

最近写了一个爬取网易养生的一些文章,只爬取首页,暂时没有翻页。后续有空再更新吧,下面是代码:

爬虫代码

import requests
from lxml import etree
from config import db
import urllib.request
import re, os, uuid, time


def run():
    data = parse()
    insert(data)

# 解析页面
def parse():
    url = 'https://jiankang.163.com/special/yangsheng_n/?1'
    res = requests.get(url)
    if res.status_code == 200:
        print("请求成功")
    else:
        print("请求失败")
        return

    html = etree.HTML(res.content)
    news_list = html.xpath('//div[@class="news_one havepic"]')

    data = []
    if news_list:
        for news in news_list:
            # 内容链接
            content_url = news.xpath('./div[@class="news_main_info"]/h2/a/@href')[0]
            # print(content_url)
            content = source = ''
            # 内容获取
            if content_url:
                res_content = requests.get(content_url)
                # res_content.encoding = 'ISO-8859-1'
                if res_content.status_code == 200:
                    print("请求内容成功")
                else:
                    print("请求内容失败")
                    return
                content_html = etree.HTML(res_content.content)
                # 文章来源
                article_source = content_html.xpath('//div[@class="post_main"]/div[@class="post_info"]/text()')[0] \
                    .strip().encode('iso-8859-1').decode('utf-8')
                source_list = article_source.split(':')
                source = source_list[-1].strip()

                # return
                # 文章内容
                article_content = content_html.xpath('//div[@class="post_main"]/div[@class="post_content"]/div['
                                                     '@class="post_body"]/p')

                # 内容合并
                if article_content:
                    for p in article_content:
                        # 判断内容是否有图片
                        images = p.xpath('./img/@src')
                        if images is not None and len(images) > 0:
                            # pass
                            # 下载图片
                            filePaths = download_img(images)
                            if filePaths is not None and len(filePaths) > 0:
                                for file in filePaths:
                                    content += '<img src="' + file + '">'
                        # 内容获取
                        text = p.xpath('./text() | ./a/text()')
                        # 大于1,则说明有a标签
                        if len(text) > 1:
                            p_text = ''.join(text)
                        elif text is not None and len(text) == 1:
                            p_text = text[0]
                        else:
                            p_text = ''
                        content += p_text.strip().encode('iso-8859-1').decode('utf-8') + "\n"
            source_img = news.xpath('./div[@class="news_main_info"]/a/img/@src')
            local_img = download_img(source_img)[0]
            # 文章数据
            article = {
                # 标题
                'title': news.xpath('./div[@class="news_main_info"]/h2/a/text()')[0],
                # 链接
                'link': news.xpath('./div[@class="news_main_info"]/h2/a/@href')[0],
                # 图片
                'source_img': source_img[0],
                # 副标题
                'sub_title': news.xpath('./div[@class="news_main_info"]/p/text()')[0].strip(),
                # 标签
                'tags': news.xpath('./div[@class="news_sub_info"]/p[@class="keywords"]/a/text()'),
                # 发布时间
                'publish_time': news.xpath('./div[@class="news_sub_info"]/p[@class="pubtime"]/text()')[0].strip(),
                # 来源, 爬取网站的地址
                'source': news.xpath('./div[@class="news_sub_info"]/div[@class="news_share"]/ul/@data-source')[0],
                # 内容
                'content': content,
                # 真实来源, 爬取网站内容的源地址
                'real_source': source,
                # 本地地址
                'local_img': local_img,
                # 远程地址
                'oss_img': '',
            }
            data.append(article)
    return data


# 插入文章和标签
def insert(data):
    DB = db.DB()
    cur = DB.cursor()
    tags_field = {'article_id', 'name', 'status'}
    tags_value = data[0]['tags']
    del data[0]['tags']
    # 字段
    field = ', '.join('`{}`'.format(k) for k in data[0].keys())
    # 值
    values = ', '.join('%({})s'.format(k) for k in data[0].keys())
    for j, i in enumerate(data):
        # 先插入文章
        # 判断文章是否存在
        sql_exists = "select id from dl_articles where title = '%s'" % (i.get('title'))
        print(sql_exists)
        cur.execute(sql_exists)
        # 存在记录,则直接跳过
        if len(cur.fetchall()) > 0:
            print("文章已存在")
            continue
        # return
        # 插入文章记录
        insert_article_sql = "insert into %s (%s) values (%s)" % ('dl_articles', field, values)
        try:
            cur.execute(insert_article_sql, i)
            DB.commit()
            print("插入文章成功")
        except Exception as e:
            print("插入文章失败")
            print(str(e))
            DB.rollback()
            return
        # 插入文章的id
        article_id = cur.lastrowid
        # 文章id小于1,则插入失败
        if article_id < 1:
            continue
        # 插入tags

        if j == 0:
            tags = tags_value
        else:
            tags = i.get('tags')
        if len(tags) > 0:
            for t in tags:
                select_tags_sql = "select id from dl_tags where `article_id` = %s and `name` = '%s'" % (article_id, t)
                print(select_tags_sql)
                # return

                try:
                    cur.execute(select_tags_sql)
                    tag_ids = cur.fetchall()
                except Exception as e:
                    print("查询标签失败")
                    print(str(e))
                    return
                # 已存在文章对应的标签
                if len(tag_ids) > 0:
                    continue
                # 插入标签
                insert_tags_sql = "insert into %s (`article_id`, `name`, `status`) values %s" % (
                'dl_tags', (article_id, t, 0))
                print(insert_tags_sql)
                # return
                try:
                    cur.execute(insert_tags_sql)
                    DB.commit()
                    print("插入标签成功")
                except Exception as e:
                    print("插入标签失败")
                    print(str(e))
                    DB.rollback()


# 下载图片
def download_img(images):
    data = []
    ym = time.strftime("%Y%m", time.localtime())
    d = time.strftime("%d", time.localtime())
    path = './download/images/' + ym + '/' + d
    if not os.path.exists(path):
        os.makedirs(path)
    suffix_list = ['jpg', 'png', 'jpeg']
    suffix = 'jpg'
    if type(images) == list and len(images) > 0:
        for img in images:
            print(img)
            for s in suffix_list:
                if re.search(s, img):
                    suffix = s
                    break
            uid = str(uuid.uuid4())
            filename = uid + '.' + suffix
            filepath = path + '/' + filename
            urllib.request.urlretrieve(img, filepath)
            data.append(filepath)
        return data

执行代码

> python .\main.py
请求成功
请求内容成功
http://cms-bucket.ws.126.net/2021/0929/0128808cj00r0647z000mc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
https://nimg.ws.126.net/?url=http%3A%2F%2Fcms-bucket.ws.126.net%2F2021%2F0928%2F6734e010j00r04eeu00kpc000k0032dc.jpg&thumbnail=660x2147483647&quality=80&type=jpg
http://cms-bucket.ws.126.net/2021/0928/9027cb39j00r04egc000qc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0926/b982d3aaj00r01tgi000nc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0925/a96809c3j00qzzq84000pc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0924/6cff9112j00qzx3np001ac0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0923/7c8f7b5bj00qzvkee000sc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0923/86cdbf99j00qzv050000jc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0922/11309378j00qzt3ny002yc0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
https://nimg.ws.126.net/?url=http%3A%2F%2Fcms-bucket.ws.126.net%2F2021%2F0917%2F02364e55j00qzl25900dec000k001vgc.jpg&thumbnail=660x2147483647&quality=80&type=jpg
http://cms-bucket.ws.126.net/2021/0917/b6c35977j00qzl26s0011c0009c0070c.jpg?imageView&thumbnail=150y100
请求内容成功
http://cms-bucket.ws.126.net/2021/0916/0679e91fp00qzj54e001rc0009c0070c.png?imageView&thumbnail=150y100
select id from dl_articles where title = '记好这些要诀,不吃安眠药也能睡个好觉'
插入文章成功
select id from dl_tags where `article_id` = 1 and `name` = '安眠药'
insert into dl_tags (`article_id`, `name`, `status`) values (1, '安眠药', 0)
插入标签成功
select id from dl_tags where `article_id` = 1 and `name` = '失眠'
insert into dl_tags (`article_id`, `name`, `status`) values (1, '失眠', 0)
插入标签成功
...

图片存储

文章存储

标签存储

本作品采用《CC 协议》,转载必须注明作者和本文链接
charliecen
讨论数量: 0
(= ̄ω ̄=)··· 暂无内容!

讨论应以学习和精进为目的。请勿发布不友善或者负能量的内容,与人为善,比聪明更重要!