python 类变量在多线程下的共享与释放问题-爱站

最近被多线程给坑了下，没意识到类变量在多线程下是共享的，还有一个就是没意识到内存释放问题，导致越累越大

1.python 类变量在多线程情况下的是共享的

2.python 类变量在多线程情况下的释放是不完全的

3.python 类变量在多线程情况下没释放的那部分内存是可以重复利用的

import threading
 import time
  
 class Test:
  
     cache= {}
      
     @classmethod
     def get_value(self, key):
         value= Test.cache.get(key, [])
         return len(value)
  
     @classmethod
     def store_value(self, key, value):
         if not Test.cache.has_key(key):
             Test.cache[key]= range(value)
         else:
             Test.cache[key].extend(range(value))
         return len(Test.cache[key])
  
     @classmethod
     def release_value(self, key):
         if Test.cache.has_key(key):
             Test.cache.pop(key)
         return True
  
     @classmethod
     def print_cache(self):
         print 'print_cache:'
         for keyin Test.cache:
             print 'key: %d, value:%d' % (key,len(Test.cache[key]))
  
 def worker(number, value):
     key= number% 5
     print 'threading: %d, store_value: %d' % (number, Test.store_value(key, value))
     time.sleep(10)
     print 'threading: %d, release_value: %s' % (number, Test.release_value(key))
  
 if __name__== '__main__':
     thread_num= 10
      
     thread_pool= []
     for iin range(thread_num):
         th= threading.Thread(target=worker,args=[i,1000000])
         thread_pool.append(th)
         thread_pool[i].start()
  
     for threadin thread_pool:
         threading.Thread.join(thread)
      
     Test.print_cache()
     time.sleep(10)
      
     thread_pool= []
     for iin range(thread_num):
         th= threading.Thread(target=worker,args=[i,100000])
         thread_pool.append(th)
         thread_pool[i].start()
  
     for threadin thread_pool:
         threading.Thread.join(thread)
      
     Test.print_cache()
     time.sleep(10)