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

import requests
from config import *
import re
import json
from docx import Document
class Spider(object):
    def __init__(self):
        self.url = "https://wenku.baidu.com/view/1d64ae0250d380eb6294dd88d0d233d4b14e3f4e"

    # 获取文本地址
    def getDocLink(self, articles=[]):
        r = requests.get(url=self.url, headers=HEADER_FOR_LINK)
        if r.status_code == 200:

            html = re.search(r"WkInfo.htmlUrls = '(.*)'", r.text).groups(0)[0].replace("\\x22", '"')
            result = (json.loads(html))
            for item in (result["json"]):
                data = {
                    "page_index": item["pageIndex"],
                    "download_url": eval(repr(item["pageLoadUrl"].replace("\\", "")))
                }
                # 获取文档内容
                doc = self.getDocData(data)
                # articles.append(doc.replace("\n\n", "\t"))
                articles.append(doc.replace("。\n","\n"))

            global title
            title = articles[0].split("\n")[0]

            if len(title) == 0:
                title = articles[0].split("\n")[1]
            # articles = [ art.replace(title,"") for art in articles ]
            # # 去除标题
            # words = ''.join(articles).replace("精选资料，欢迎下载\n。", "")
            # self.writeDocText( words )
    # 获取文本内容
    def getDocData(self, data=None, string=""):
        r = requests.get(url=data["download_url"], headers=HEADER_FOR_CON)
        if r.status_code == 206:
            data = re.findall(r"({.*})", r.text)[0]
            items = json.loads(data)

            for item in ((items["body"])):

                if "r" not in item or (item["r"] == [0] and len(item["c"]) == 1 ):
                    continue
                else:

                    if isinstance(item["c"], dict):
                        pass
                    else:
                        string += item["c"]
            strings = string.replace(" ", "\n")
            return strings

    # 写入doc文本
    def writeDocText(self,text):
        doc = Document( )
        doc.add_heading( title )
        doc.add_paragraph( text )
        doc.save("{0}.doc".format( title ))

if __name__ == '__main__':
    s = Spider()
    s.getDocLink()