ThreadLocal
TheadLocal 用于多线程环境下，线程之间可以使用相同的变量，而这个变量只与当前线程环境有关。werkzeug中有类似的实现，使每个路由处理函数都可使用相同的request变量，而这个对象的内容只与当前请求有关。 例如:1 import threading # 创建全局ThreadLocal对象: local_school = threading.local() def process_student(): # 获取当前线程关联的student: std = local_school.student print('Hello, %s (in %s)' % (std, threading.current_thread().name)) def process_thread(name): # 绑定ThreadLocal的student: local_school.student = name process_student() t1 = threading.Thread(target= process_thread, args=('Alice',), name='Thread-A') t2 = threading.Thread(target= process_thread, args=('Bob',), name='Thread-B') t1.start() t2.start() t1.join() t2.join() 执行结果： Hello, Alice (in Thread-A) Hello, Bob (in Thread-B) 廖雪峰的官方网站 ↩︎