第一阶段（二）

日期：2021..05.04

作者：杨传伟

完成任务：爬取腾讯视频电影片库首页所有电影信息（电影名字、播放链接、评分、播放权限）并存到数据库。

爬虫源码

import requests
import json
from bs4 import BeautifulSoup       #网页解析获取数据
import sys
import re
import urllib.request,urllib.error #制定url，获取网页数据
import sqlite3
import xlwt     #excel操作
import time
import pymysql
import traceback
#连接数据库  获取游标
def get_conn():
    """
    :return: 连接，游标
    """
    # 创建连接
    conn = pymysql.connect(host="82.157.112.34",
                    user="root",
                    password="root",
                    db="MovieRankings",
                    charset="utf8")
    # 创建游标
    cursor = conn.cursor()  # 执行完毕返回的结果集默认以元组显示
    if ((conn != None) & (cursor != None)):
        print("数据库连接成功！游标创建成功！")
    else:
        print("数据库连接失败！")
    return conn, cursor
#关闭数据库连接和游标
def close_conn(conn, cursor):
    if cursor:
        cursor.close()
    if conn:
        conn.close()
    return 1

#爬取腾讯视频电影数据
def get_ten():
    conn,cursor=get_conn()
    sql="select count(*) from movieten"
    cursor.execute(sql)
    conn.commit()
    all_num=cursor.fetchall()[0][0]

    print("movieten数据库有",all_num,"条数据！")
    #   https://v.qq.com/channel/movie?listpage=1&channel=movie&sort=18&_all=1&offset=0&pagesize=30
    url="https://v.qq.com/channel/movie?listpage=1&channel=movie&sort=18&_all=1"        #链接
    param={                                                                             #参数字典
        'offset':0,
        'pagesize':30
    }
    headers={                                                                            #UA伪装
        'user-agent' : 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '+
                       'AppleWebKit/537.36 (KHTML, like Gecko) Chrome/90.0.4430.85 Safari/537.36'
    }
    # param['offset']=all_num
    offset = 0                                                                           #拼接url
    dataRes = []
    findLink = re.compile(r'href="(.*?)"')  # 链接
    findName = re.compile(r'title="(.*?)"')  # 影片名
    findScore= re.compile(r'<div class="figure_score">(.*?) </div>')        #评分
    #3*170
    for i in range(0,300):
        # res = urllib.request.urlopen(url)                             #urllib不推荐使用
        res = requests.get(url=url,params=param,headers=headers)       #编辑request请求
        # print(url)
        res.encoding='utf-8'                                           #设置返回数据的编码格式为utf-8
        html=BeautifulSoup(res.text,"html.parser")                      #BeautifulSoup解析
        part_html = html.find_all(r"a", class_="figure")               #找到整个html界面里a标签对应的html代码，返回值是一个list
        # print(part_html)
        if (len(part_html) == 0):
            print("页面返回空！")
            return dataRes
        offset = offset + 30                                            #修改参数字典+30部电影
        print("下面从第"+str(offset)+"部电影开始：")
        param['offset'] = offset
        print(param['offset'])
        for i in part_html:                                            #遍历每一个part_html
            # print(i)
            words = str(i)
            name=re.findall(findName, words)# 添加影片名
            score=re.findall(findScore, words)# 添加评分
            link=re.findall(findLink, words)# 添加链接
            findState=BeautifulSoup(words,'lxml')       #单独解析播放状态
            state=findState.select('a > img')           #找到img父级标签
            if(len(state)==1):                          #免费电影不存在播放状态的标志，所以当img长度是1的时候，需要补上一个空串
                state.append("")
            state_text=str(state[1])                    #拿到第二个img对应的内容，使用正则匹配到alt属性对应的字符串
            # print(state_text)
            temp_state=re.findall('<img alt="(.*?)"', state_text)
            if(len(temp_state)==0):
                temp_state.insert(0,"免费") # 添加播放状态---免费
            # print(temp_state[0])
            list_=[]
            if(len(score)==0):
                score.insert(0,"暂无评分")
            for i in dataRes:
                if name[0] in i[0]:
                    name.insert(0,name[0]+"（其他版本）")
            list_.append(name[0])
            list_.append(score[0])
            list_.append(link[0])
            list_.append(temp_state[0])
            # list_.append(statu)
            # print(list_)
            print(list_)
            dataRes.append(list_)
    # print(dataRes)      #打印最终结果
    # list=html.select(".figure_score")
    # for item in list:
    #     print(item)

    #把同一部电影的信息放到一个 [ ] 里面

    return dataRes
#插入到腾讯电影数据库
def insert_ten():
    """
    插入腾讯电影数据
    :return:
    """
    cursor = None
    conn = None
    try:
        list = get_ten()
        print(f"{time.asctime()}开始插入腾讯电影数据")
        conn, cursor = get_conn()
        sql = "insert into movieten (id,name,score,path,state) values(%s,%s,%s,%s,%s)"
        for item in list:
            try:
                cursor.execute(sql,[0,item[0],item[1],item[2],item[3]])
            except pymysql.err.IntegrityError:
                print("重复！跳过！")
        conn.commit()  # 提交事务 update delete insert操作
        print(f"{time.asctime()}插入腾讯电影数据完毕")
    except:
        traceback.print_exc()
    finally:
        close_conn(conn, cursor)
    return ;
if __name__ == '__main__':
    # conn,cursor=get_conn()
    # list=[]
    # res_list=get_ten()
    # print(res_list)
    insert_ten()

html分析文件

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
</head>
<body>
<a href="https://v.qq.com/x/cover/aox91j5e2nis6sc.html" class="figure" target="scene=%E9%A2%91%E9%81%93%E9%A1%B5&amp;
    pagename=%E7%94%B5%E5%BD%B1%E9%A2%91%E9%81%93&amp;columnname=%E5%85%A8%E9%83%A8_%E5%88%97%E8%A1%A8%E9%A1%B5&amp;
    controlname=listpage&amp;cid=aox91j5e2nis6sc&amp;vid=&amp;pid=&amp;datatype=1&amp;playertype=1&amp;controlidx=0&amp;
    columnidx=0&amp;plat_bucketid=9231003&amp;cmd=1" tabindex="-1" data-float="aox91j5e2nis6sc" title="老师•好" data-floatid="8">

    <img class="figure_pic" src="//puui.qpic.cn/vcover_vt_pic/0/aox91j5e2nis6sc1568017385/220" alt="老师•好" onerror="picerr(this,'v')">

    <div class="figure_caption">01:52:59</div>

    <div class="figure_score">8.1 </div>
    <img class="mark_v mark_v_VIP" src="//vfiles-raw.gtimg.cn/vupload/20201015/tag_vip_x1.png"
         srcset="//vfiles-raw.gtimg.cn/vupload/20201015/tag_vip_x2.png 2x" alt="VIP" onerror="picerr(this)">

</a>
</body>
</html>

截图：

 

 

---

 

5.4 李楠

今日主要学习了使用正则表达式进行数据解析，成功爬取到地区与语言，下面为测试爬取的代码：

def get_tencent_data():
    url_bean = 'https://movie.douban.com/subject/26752088/'

    headers = {
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36',
    }

    text=requests.get(url=url_bean,headers=headers).text
    ex=' <span class="pl">制片国家/地区:</span> (.*?)<br/>'
    test=re.findall(ex,text,re.S)
    ex2='<span class="pl">语言:</span> (.*?)<br/>'
    test = re.findall(ex2, text, re.S)
    summary = test[0].replace(" / ", " ")
    print(summary)


之后对之前的爬取代码进行了优化，并加上了国家与语言：
def insert_data(data_beans,headers,cursor,conn):
    try:
        #外层循环（五十个拥有20个电影数据的数组）
        for data_bean in data_beans:
            #20个电影数据
            for i in data_bean:
                # 分配数据
                score = i["rate"].replace(" ", "")
                director = i["directors"]  # []
                director_str = ""
                for j in director:
                    director_str = director_str + " " + j
                name = i["title"].replace(" ", "")
                img = i["cover"].replace(" ", "")
                star = i["casts"]  # []
                star_str = ""
                for j in star:
                    star_str = star_str + " " + j
                # 分配数据

                # 获取电影详细数据的网址
                url_details = i["url"].replace(" ", "")
                r = requests.get(url_details, headers=headers)
                soup_bean = BeautifulSoup(r.text, "lxml")
                # 获取详细数据
                span = soup_bean.find_all("span", {"property": "v:genre"})
                type = ""
                for i in span:
                    type = type + " " + i.text
                span = soup_bean.find_all("span", {"property": "v:runtime"})
                timelen = span[0].text.replace(" ", "")
                span = soup_bean.find_all("span", {"property": "v:initialReleaseDate"})
                date = span[0].text.replace(" ", "")
                span = soup_bean.find("a", {"class", "rating_people"})
                scorenum = span.text.replace(" ", "")
                span = soup_bean.find("span", {"property": "v:summary"})
                summary = span.text.replace(" ", "")  # 将空格去掉
                ex = ' <span class="pl">制片国家/地区:</span> (.*?)<br/>'
                test = re.findall(ex, r.text, re.S)
                area = test[0].replace(" ", "")
                ex2 = '<span class="pl">语言:</span> (.*?)<br/>'
                test = re.findall(ex2, r.text, re.S)
                language = test[0].replace(" / ", " ")
                print(url_details)
                # 获取详细数据
                sql = "insert into moviebean values(%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)"
                cursor.execute(sql,
                               [name, star_str, director_str, type, area, date, summary, score, language, img, scorenum,
                                timelen])
        conn.commit()  # 提交事务 update delete insert操作 //*[@id="info"]/text()[2]
        print(f"{time.asctime()}插入数据完毕")
    except:
        traceback.print_exc()

def get_tencent_data():
    #豆瓣的网址
    url_bean = 'https://movie.douban.com/j/new_search_subjects?sort=T&range=0,10&tags=%E7%94%B5%E5%BD%B1&start='

    headers = {
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36',
    }
    # num=[0,1020,2020,3020,4020,5020,6020,7020,8020,9020]
    # partial_big_loop =partial(big_loop,url_bean=url_bean,headers=headers)
    # #大循环在10个线程中进行
    # pool=Pool(10)
    # pool.map(partial_big_loop,num)
    # pool.close()
    # pool.join()
    cursor = None
    conn = None
    conn, cursor = get_conn()
    data_beans=[]
    num=6420#1440/3020/2760/3100/3180/3440/3300/3660/3860/4000/4020/4260/4380/4440(Data too long for column 'language' at row 1)
    # /4500/4680/4900/4960/5000/5140/5160/5200/5340/5440/5480/5520/5580//5600/5660/5820/5920/6000/6020/6100/6140/6320
    b=0;
    while b<=500:
        a = 1
        b=b+1
        while a <= 1:
            num_str = '%d' % num
            num = num + 20
            a = a + 1;
            # 获取豆瓣页面电影数据
            r = requests.get(url_bean + num_str, headers=headers)
            print(num_str)
            res_bean = json.loads(r.text);
            print(url_bean+num_str)
            data_beans.append(res_bean["data"])
            print(f"{time.asctime()}开始插入数据")
        insert_data(data_beans, headers,cursor,conn)
        data_beans=[]
    print(f"{time.asctime()}所有数据插入完毕")
    close_conn(conn, cursor)


初步写完了查询模块，根据类型，时间，地区查询，并可以按照评分与热度正序或倒序排列：

def find_class_order(str):
    sql="select title,star,director,score,date_time,area,type_movie,scorenum,img from moviebean where 1=1 " \
        "and (type_movie like "+"'%"+str[0]+"%'"+") and (date_time like "+"'%"+str[1]+"%'"+") and(area like "+"'%"+str[2]+"%'"+") "
    if(str[3]=="star_1"):
        sql=sql+" order by score desc "
    if(str[3]=="star_0"):
        sql=sql+" order by score "
    if(str[3]=="hot_1"):
        sql=sql+" order by scorenum desc "
    if(str[3]=="hot_0"):
        sql=sql+" order by scorenum limit "
    sql=sql+"limit "+str[4]
    print(sql)
    res = query(sql)
    print(res)
    return res

 

---

 

5.4 章英杰

任务进度： 昨天天设计页面时过于心急，工作出发点出了问题，想着直接在豆瓣网页的基础上做修改，却忽略了整个页面的分块布局问题，导致整个页面模块布局还没有设计好就去设计每一块的具体内容，今天发现页面布局不合理无法继续进行，于是重新写了页面布局。

页面布局主要分为了五部分，标题搜索框部分、电影分类部分、电影排行榜部分、广告部分和底部结尾部分。

产品页面：

 

---

 

任务看板

 

---

 

 每日照片：