- 百度文库有一些需要下载券,才能下载的资料
- 但其实文库是允许我们预览的,可是不允许我们复制内容
- 我们只是需要里面的文字内容,对内容的样式没有什么要求
windows平台运行
doc.gif
下载器实现的功能:
1.按照输入的网址,自动判断文档类型,并将下载好的资源放在相应的文件夹中
自动分类.png
将ppt类型的文档自动转换为图片,并按原本的顺序命名保存
image.png
3.pdf,word.txt类型的数据全部消除格式,以txt格式保存文本
实现效果
下载word与pdf.png
下载ppt.png
下载txt.png
下载器的数据来源
分析资源所在页面的源码,获取请求资源的接口,用requests库请求资源,然后手动实现文本的拼接规则,最后把文本内容输出到脚本同级目录下的文件夹中
word类型文档
word.png
ppt类型文档
ppt.png
txt文档
实现源码:
import os
import re
import json
import requests
from lxml import etree
# 创建文库基类
class BaiduWK(object):
def __init__(self, url):
self.title = None
self.url = url
self.docType = None
self.headers = {'User-Agent':'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/62.0.3202.75 Safari/537.36'}
self.get_response_content(self.url)
self.get_doc_type_and_title()
def get_response_content(self, url):
try:
response = requests.get(url, headers = self.headers)
return response.content
except Exception as e:
print(e)
pass
def get_doc_type_and_title(self):
# 获取源码
source_html = self.get_response_content(self.url)
# 解析源码
content = source_html.decode('gbk')
# 获取文档类型
self.docType = re.findall(r"docType.*?\:.*?\'(.*?)\'\,", content)[0]
# 获取文档标题
self.title = re.findall(r"title.*?\:.*?\'(.*?)\'\,", content)[0]
# 创建获取txt的类
class BDWKTXT(BaiduWK):
def __init__(self, url):
super().__init__(url)
self.docId = None
pass
def get_txt(self, url):
# 获取源码
source_html = self.get_response_content(url)
content = source_html.decode("gbk")
# 获取docId
self.docId = re.findall(r"docId.*?(\w{24}?)\'\,", content)[0]
# 拼接请求url
token_url = "https://wenku.baidu.com/api/doc/getdocinfo?callback=cb&doc_id=" + self.docId
# 再次请求
first_json = self.get_response_content(token_url).decode()
str_first_json = re.match(r'.*?\((\{.*?\})\).*', first_json).group(1)
# print(str_first_json)
the_first_json = json.loads(str_first_json)
md5sum = the_first_json["md5sum"]
rn = the_first_json["docInfo"]["totalPageNum"]
rsign = the_first_json["rsign"]
# print(md5sum,"-->",rn)
# 请求目标url
target_url = "https://wkretype.bdimg.com/retype/text/" + self.docId + "?" + md5sum + "&callback=cb" + "&pn=1&rn=" + rn + "&type=txt" +"&rsign=" + rsign
# https://wkretype.bdimg.com/retype/text/de53bafaf705cc1755270982?md5sum=b89f055e765e6f73db57525bcfc3c2d2&sign=7b385f9cf7&callback=cb&pn=1&rn=12&type=txt
sec_json = self.get_response_content(target_url).decode()
# print(type(sec_json),"-->")
str_sec_json = re.match(r'.*?\(\[(.*)\]\)$', sec_json).group(1)
str_sec_json += ","
str_json_list = str_sec_json.split('},')
result_txt = ""
# 截取尾部空格
str_json_list = str_json_list[:-1]
for str_json in str_json_list:
str_json += "}"
pure_dic = json.loads(str_json)
pure_txt = pure_dic["parags"][0]["c"].strip()
result_txt += pure_txt
# 创建文件目录
try:
path = "." + os.sep + self.docType
# print(type(path),"path-->",path)
os.makedirs(path)
except Exception as e:
# print("文件夹%s已存在"%(path))
pass
# 创建文件,保存信息
try:
file_name = "." + os.sep +self.docType + os.sep + self.title + ".txt"
with open(file_name, 'w') as f:
f.write(result_txt)
print("已经保存为:",self.title + '.txt')
except Exception as e:
# print(e)
pass
# 创建获取word的类
class BDWKDOC(BaiduWK):
def __init__(self, url):
super().__init__(url)
# 保存数据来源url
self.pure_addr_list = list()
# 获取数据来源url
def get_pure_addr_list(self):
# 获取页面源码
source_html = self.get_response_content(self.url).decode('gbk')
# 从源码中批量提取数据url
all_addr = re.findall(r'wkbos\.bdimg\.com.*?json.*?expire.*?\}',source_html)
pure_addr_list = list()
# 获取文档标题
self.title = etree.HTML(source_html).xpath("//title/text()")[0].strip()
# 净化数据来源url列表
for addr in all_addr:
addr = "https://" + addr.replace("\\\\\\/", "/")
addr = addr[:-5]
pure_addr_list.append(addr)
# 将处理好的url列表保存为全局属性
self.pure_addr_list = pure_addr_list
return pure_addr_list
# 从数据来源的url列表中提取数据
def get_json_content(self, url_list):
content = ''
result = ''
sum = len(url_list)
i = 1
for pure_addr in url_list:
print("正在下载第%d条数据, 剩余%d条"%(i, sum-i))
i += 1
try:
# 获取json数据
content = self.get_response_content(pure_addr).decode()
# 处理json数据
content = re.match(r'.*?\((.*)\)$', content).group(1)
# 将json数据中需要的内容提取出来
all_body_info = json.loads(content)["body"]
# 遍历获取所有信息,并将信息拼接到一起
for body_info in all_body_info:
try:
result = result + body_info["c"].strip()
except Exception as e:
# print(e)
pass
except Exception as e:
# print(e)
pass
# 创建文件目录
try:
path = "." + os.sep + self.docType
# print(type(path),"path-->",path)
os.makedirs(path)
except Exception as e:
# print("文件夹%s已存在"%(path))
pass
# 创建文件,保存信息
try:
file_name = "." + os.sep +self.docType + os.sep + self.title + ".txt"
with open(file_name, 'w') as f:
f.write(result)
print("已经保存为:",self.title + '.txt')
except Exception as e:
# print(e)
pass
# 创建获取ppt的类
class BDWKPPT(BaiduWK):
def __init__(self, url):
self.all_img_url = list()
super().__init__(url)
# 获取json数据保存文件
def get_ppt_json_info(self):
# 获取源文件
ppt_source_html = self.get_response_content(self.url)
# 解析源文件
content = ppt_source_html.decode('gbk')
# print("-->",len(content))
# 测试
with open("test.html", "w") as f:
f.write(content)
# 获取文档Id
self.docId = re.findall(r"docId.*?(\w{24}?)\'\,", content)[0]
# 拼接请求json的接口
source_json_url = 'https://wenku.baidu.com/browse/getbcsurl?doc_id=%s&type=ppt&callback=zhaozhao'%self.docId
# 获取字符串类型的json数据
str_source_json = self.get_response_content(source_json_url).decode()
# 处理字符串类型的json数据,使其成为标准格式
pure_str_source_json = re.match(r'.*?\((.*?)\)', str_source_json).group(1)
# 将字符串json转为可处理的正式json
source_json = json.loads(pure_str_source_json)
# 遍历字典中的数据类型list
for j in source_json['list']:
# 创建临时列表
temp_num_url = list()
# 将url和page拼接到列表中
temp_num_url.append(j["zoom"])
temp_num_url.append(j["page"])
# 将列表信息添加到全局变量中
self.all_img_url.append(temp_num_url)
# 建立文件夹
try:
os.makedirs("./ppt/%s"%(self.title))
except Exception as e:
# print("---->>",e)
pass
for img_url in self.all_img_url:
# print(img_url)
print("正在获取第%d页资源(剩余%d页)"%(img_url[1], len(self.all_img_url)-img_url[1]))
data = self.get_response_content(img_url[0])
path = "./ppt/%s/%s"%(self.title, str(img_url[1])+'.jpg')
with open (path, 'wb') as f:
f.write(data)
print("写入完毕")
# 运行主程序
def main():
try:
url = input("请输入资源所在的网址:")
docType = BaiduWK(url).docType
except:
print("您输入的url,有误请重新输入!")
os.exit()
print("类型为","-->",docType)
if docType == "ppt":
ppt = BDWKPPT(url)
print("您将要获取的演示文稿(ppt)名称为:", ppt.title)
ppt.get_ppt_json_info()
elif docType == "doc":
word = BDWKDOC(url)
print("您将要获取的文档(word)名称为", word.title)
pure_addr_list = word.get_pure_addr_list()
word.get_json_content(pure_addr_list)
elif docType == "pdf":
pdf = BDWKDOC(url)
print("您将要获取的PDF名称为:", pdf.title)
pure_addr_list = pdf.get_pure_addr_list()
pdf.get_json_content(pure_addr_list)
elif docType == "txt":
txt = BDWKTXT(url)
print("您将要下载的文本文档(txt)名称为:", txt.title)
txt.get_txt(url)
else:
other = BDWKPPT(url)
print("暂不支持下载%s类型"%(other.docType))
pass
if __name__ == '__main__':
main()
我把这个脚本编译成了exe文件,windows用户从下面的资源帖子里按文章名自取: http://www.jianshu.com/p/4f28e1ae08b1