# -*- coding: utf-8 -*-

import re
from time import sleep
from bs4 import BeautifulSoup
import requests
from requests.cookies import RequestsCookieJar
import hashlib
import pymysql
import oss2
import config
from math import ceil


class GetAriticle(object):
    def __init__(self, token, cookie, query_name):
        self.auth = oss2.Auth(config.oss_appid, config.oss_secretkey)
        self.bucket = oss2.Bucket(self.auth, config.oss_url, config.oss_bucket)
        self.conn = pymysql.connect(host=config.mysql_host,
                                    user=config.mysql_username,
                                    password=config.mysql_password,
                                    database=config.mysql_database)
        self.time_gap = 30
        self.token = token
        self.query_name = query_name
        self.headers = {
            'Host': 'mp.weixin.qq.com',
            'User-Agent': r'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:65.0) Gecko/20100101 Firefox/65.0',
        }
        self.sess = requests.Session()
        self.c = RequestsCookieJar()
        for i in cookie:
            self.c.set(i["name"], i["value"])
            self.sess.cookies.update(self.c)

    def Get_WeChat_Subscription(self):
        url = r'https://mp.weixin.qq.com/cgi-bin/searchbiz?action=search_biz&token={0}&lang=zh_CN&f=json&ajax=1&rando' \
              r'm=0.5182749224035845&query={1}&begin=0&count=5'.format(self.token, self.query_name)
        html_json = self.sess.get(url, headers=self.headers).json()
        print(html_json)
        fakeid = html_json['list'][0]['fakeid']
        nickname = html_json['list'][0]['nickname']
        print("fakeid:", fakeid)
        print("nickname: ", nickname)
        return fakeid

    def Get_Articles(self, fake_id):
        title_buf = []
        link_buf = []
        img_buf = []

        Total_buf = []
        url = r'https://mp.weixin.qq.com/cgi-bin/appmsg?token={0}&lang=zh_CN&f=json&ajax=1&random=0.977467295649225&' \
              r'action=list_ex&begin=0&count=5&query=&fakeid={1}&type=9'.format(self.token, fake_id)
        html_json = self.sess.get(url, headers=self.headers).json()
        try:
            Total_Page = ceil(int(html_json['app_msg_cnt'])/5)
        except Exception as e:
            print(e)
            print("!! 失败信息：", html_json['base_resp']['err_msg'])
            return
        for i in range(Total_Page):
            print("第[%d/%d]页" % (i+1, Total_Page))
            begin = i * 5
            url = r'https://mp.weixin.qq.com/cgi-bin/appmsg?token={0}&lang=zh_CN&f=json&ajax=1&' \
                  r'random=0.977467295649225&action=list_ex&begin={1}&count=5&query=&fakeid={2}&type=9'.\
                format(self.token, begin, fake_id)
            html_json = self.sess.get(url, headers=self.headers).json()
            # print(html_json)
            app_msg_list = html_json['app_msg_list']
            if str(app_msg_list) == '[]':
                break
            for j in range(20):
                try:
                    if app_msg_list[j]['title'] in Total_buf:
                        print("本条已存在，跳过")
                        continue
                    title_buf.append(app_msg_list[j]['title'])
                    Total_buf.append(app_msg_list[j]['title'])
                    link_buf.append(app_msg_list[j]['link'])
                    img_buf.append(app_msg_list[j]['cover'])
                    with open('/spider.txt', 'a+') as fp:
                        fp.write('*'*60+'\nTitle: '+title_buf[j]+'\nLink: '+link_buf[j]+'\nImg: '+img_buf[j]+'\r\n')
                        fp.close()
                        print(">> 第%d条写入完成：%s" % (j+1, title_buf[j]))
                except Exception as e:
                    print(">> 本页抓取结束")
                    print(e)
                    break
            print(">> 一页抓取结束，开始下载")
            self.get_content(title_buf, link_buf)
            title_buf.clear()  # 清除缓存
            print(">> 休息 %d s" % self.time_gap)
            sleep(self.time_gap)
        print(">> 抓取结束")

    def get_content(self, title_buf, link_buf):  # 获取地址对应的文章内容
        length = len(title_buf)

        for index in range(length):
            each_title = re.sub(r'[\|\/\<\>\:\*\?\\\"]', "_", title_buf[index])  # 剔除不合法字符
            html = self.sess.get(link_buf[index], headers=self.headers)
            soup = BeautifulSoup(html.text, 'lxml')
            article = soup.find(class_="rich_media_content").find_all("p")  # 查找文章内容位置
            img_urls = soup.find(class_="rich_media_content").find_all("img")  # 获得文章图片URL集

            print("*" * 60)
            print(each_title)
            print(">> 保存文档 - ", end="")
            content_item = ''
            for i in article:
                line_content = i.get_text()  # 获取标签内的文本
                # print(line_content)
                if line_content is not None:
                    content_item = content_item + line_content + '\n'
            content_result = str(content_item).replace("'", "")
            print("完毕!")
            print(">> 保存图片 - %d张" % len(img_urls), end="")
            pic_item = []
            for i in range(len(img_urls)):
                pic_down = requests.get(img_urls[i]["data-src"])
                pic_md5 = hashlib.md5(str(pic_down.content).encode('utf-8')).hexdigest()
                pic_item.append(pic_md5)
                self.bucket.put_object('{}/{}.jpeg'.format(config.oss_save_dir, pic_md5), pic_down)
                print('\n' + '*' * 60 + '\n{}/{}.jpeg'.format(config.oss_save_path, pic_md5))
            pic_result = str(pic_item).replace("'", "").replace('[', '').replace(']', '')
            cursor = self.conn.cursor()
            sql = "INSERT INTO wechat(title, content, from_type, img, status) VALUES('{}', '{}', '{}', '{}', '{}')"\
                .format(each_title, content_result, self.query_name, pic_result, '1')
            print(sql)
            cursor.execute(sql)
            self.conn.commit()
            print("完毕!\r\n")