Python multi-threaded crawling files, how to set timeout and reconnection.

Question

When using python to crawl data, enable multi-thread crawling in a single process. After all, I don’t have multiple processes because of intensive IO.

---

code show as below

    def get_downloads_url_list(self,pageNum):
        FilePath='C:/RMDZY/h'+str(pageNum)
        os.chdir(FilePath)
        with open(FilePath+'/m3u8.txt', 'r') as f:
            m3u8_txt = f.read()
        download_ts_list = re.findall(r'ppvod' + r'\d{7}' + r'.ts', m3u8_txt)
        download_url_list = [url + str(pageNum) + '/1000kb/hls/' + download_ts_list[i] for i in
                             range(len(download_ts_list))]
        max_length=len(download_url_list)
        dat_list=['ts'+str(i)+'.ts' for i in range(max_length)]
        dat_str='+'.join(dat_list)
        ts_command='copy /b '+dat_str+' new.ts'
        with open('ts.bat','w') as f:
            f.write(ts_command)
        return download_url_list

    def download_by_m3u8(self,i,pageNum):
        download_list=self.get_downloads_url_list(pageNum)
        ts_file = requests.get(download_list[i], verify=False)
        with open('ts'+str(i)+'.ts','ab') as f:
            f.write(ts_file.content)![图片描述][1]

    def download_threading(self,pageNum):
        download_list=self.get_downloads_url_list(pageNum)
        thread_list=[]
        for i in range(len(download_list)):
            thread = threading.Thread(target=self.download_by_m3u8, args=[i,pageNum])
            thread_list.append(thread)
            thread.start()
        for thread in thread_list:
            thread.join()

---

But as long as a thread's requests do not return a value, the thread will keep waiting and will not write, so there will be a problem that the main process is not blocked.
As shown in the picture

---

How to deal with it, for example, set a timeout for requests.get, but what to do after it exceeds? After I set the timeout, it seems that the thread was killed directly, and I can continue to download the next target, but this one is not downloaded. I recorded it like this, can you catch this exception and reconnect? The main thing is writing, I'm not very good at it. Strange

Answer 1

num = 3 # 重试次数
while num > 0:
    try:
        result = requests.get(..., timeout=3) 
    except requests.exceptions.ReadTimeout:
        print 'Timeout, try again'
        num -= 1
    else:
        # 成功获取
        print 'ok'
        print result
        break
else:
    # 3次都失败
    print 'Try 3 times, But all failed'