# coding=utf-8
import asyncio
import multiprocessing
import os
import re
import time
from math import floor
from multiprocessing import Manager
import aiohttp
import requests
from lxml import html
import threading
from src.my_lib import retry
from src.my_lib import time_statistics


class M3U8Download:
 _path = "./resource\\" # 本地文件路徑
 _url_seed = None # 資源所在鏈接前綴
 _target_url = {} # 資源任務目標字典
 _mode = ""
 _headers = {"User-agent": "Mozilla/5.0"} # 瀏覽器代理
 _target_num = 100

 def __init__(self):
 self._ml = Manager().list() # 進程通信列表
 if not os.path.exists(self._path): # 檢測本地目錄存在否
  os.makedirs(self._path)
 exec_str = r'chcp 65001'
 os.system(exec_str) # 先切換utf-8輸出,防止控制臺亂碼

 def sniffing(self, url):
 self._url = url
 print("開始嗅探...")
 try:
  r = requests.get(self._url) # 訪問嗅探網址,獲取網頁信息
 except:
  print("嗅探失敗,網址不正確")
  os.system("pause")
 else:
  tree = html.fromstring(r.content)
  try:
  source_url = tree.xpath('//video//source/@src')[0] # 嗅探資源控制文件鏈接,這里只針對一個資源控制文件
  # self._url_seed = re.split("/\w+\.m3u8", source_url)[0] # 從資源控制文件鏈接解析域名
  except:
  print("嗅探失敗,未發現資源")
  os.system("pause")
  else:
  self.analysis(source_url)

 def analysis(self, source_url):
 try:
  self._url_seed = re.split("/\w+\.m3u8", source_url)[0] # 從資源控制文件鏈接解析域名
  with requests.get(source_url) as r: # 訪問資源控制文件,獲得資源信息
  src = re.split("\n*#.+\n", r.text) # 解析資源信息
  for sub_src in src: # 將資源地址儲存到任務字典
   if sub_src:
   self._target_url[sub_src] = self._url_seed + "/" + sub_src
 except Exception as e:
  print("資源無法成功解析", e)
  os.system("pause")
 else:
  self._target_num = len(self._target_url)
  print("sniffing success!!!,found", self._target_num, "url.")
  self._mode = input(
  "1:-> 單進程(Low B)\n2:-> 多進程+多線程(網速開始biubiu飛起!)\n3:-> 多進程+協程(最先進的并發!!!)\n")
  if self._mode == "1":
  for path, url in self._target_url.items():
   self._download(path, url)
  elif self._mode == "2" or self._mode == "3":
  self._multiprocessing()

 def _multiprocessing(self, processing_num=4): # 多進程,多線程
 target_list = {} # 進程任務字典,儲存每個進程分配的任務
 pool = multiprocessing.Pool(processes=processing_num) # 開啟進程池
 i = 0 # 任務分配標識
 for path, url in self._target_url.items(): # 分配進程任務
  target_list[path] = url
  i += 1
  if i % 10 == 0 or i == len(self._target_url): # 每個進程分配十個任務
  if self._mode == "2":
   pool.apply_async(self._sub_multithreading, kwds=target_list) # 使用多線程驅動方法
  else:
   pool.apply_async(self._sub_coroutine, kwds=target_list) # 使用協程驅動方法
  target_list = {}
 pool.close() # join函數等待所有子進程結束
 pool.join() # 調用join之前，先調用close函數，否則會出錯。執行完close后不會有新的進程加入到pool
 while True:
  if self._judge_over():
  self._combine()
  break

 def _sub_multithreading(self, **kwargs):
 for path, url in kwargs.items(): # 根據進程任務開啟線程
  t = threading.Thread(target=self._download, args=(path, url,))
  t.start()

 @retry()
 def _download(self, path, url): # 同步下載方法
 with requests.get(url, headers=self._headers) as r:
  if r.status_code == 200:
  with open(self._path + path, "wb")as file:
   file.write(r.content)
  self._ml.append(0) # 每成功一個就往進程通信列表增加一個值
  percent = '%.2f' % (len(self._ml) / self._target_num * 100)
  print(len(self._ml), ": ", path, "->OK", "\tcomplete:", percent, "%") # 顯示下載進度
  else:
  print(path, r.status_code, r.reason)

 def _sub_coroutine(self, **kwargs):
 tasks = []
 for path, url in kwargs.items(): # 根據進程任務創建協程任務列表
  tasks.append(asyncio.ensure_future(self._async_download(path, url)))
 loop = asyncio.get_event_loop() # 創建異步事件循環
 loop.run_until_complete(asyncio.wait(tasks)) # 注冊任務列表

 async def _async_download(self, path, url): # 異步下載方法
 async with aiohttp.ClientSession() as session:
  async with session.get(url, headers=self._headers) as resp:
  try:
   assert resp.status == 200, "E" # 斷言狀態碼為200,否則拋異常,觸發重試裝飾器
   with open(self._path + path, "wb")as file:
   file.write(await resp.read())
  except Exception as e:
   print(e)
  else:
   self._ml.append(0) # 每成功一個就往進程通信列表增加一個值
   percent = '%.2f' % (len(self._ml) / self._target_num * 100)
   print(len(self._ml), ": ", path, "->OK", "\tcomplete:", percent, "%") # 顯示下載進度

 def _combine(self): # 組合資源方法
 try:
  print("開始組合資源...")
  identification = str(floor(time.time()))
  exec_str = r'copy /b "' + self._path + r'*.ts" "' + self._path + 'video' + identification + '.mp4"'
  os.system(exec_str) # 使用cmd命令將資源整合
  exec_str = r'del "' + self._path + r'*.ts"'
  os.system(exec_str) # 刪除原來的文件
 except:
  print("資源組合失敗")
 else:
  print("資源組合成功!")

 def _judge_over(self): # 判斷是否全部下載完成
 if len(self._ml) == len(self._target_url):
  return True
 return False


@time_statistics
def app():
 multiprocessing.freeze_support()
 url = input("輸入嗅探網址:\n")
 m3u8 = M3U8Download()
 m3u8.sniffing(url)
 # m3u8.analysis(url)


if __name__ == "__main__":
 app()

import time


def time_statistics(fun):
 def function_timer(*args, **kwargs):
 t0 = time.time()
 result = fun(*args, **kwargs)
 t1 = time.time()
 print("Total time running %s: %s seconds" % (fun.__name__, str(t1 - t0)))
 return result

 return function_timer


def retry(retries=3):
 def _retry(fun):
 def wrapper(*args, **kwargs):
  for _ in range(retries):
  try:
   return fun(*args, **kwargs)
  except Exception as e:
   print("@", fun.__name__, "->", e)

 return wrapper

 return _retry