为什么在处理来自多进程的异常时.进程键盘中断异常只有在您自己抛出时才能捕获?

Question

假设我有一堆进程，我想运行它们并检查它们是否正常：

procs = [fetcher_process, checker_process, publisher_process, manager_process]
try:
for p in procs:
p.run()
except Exception as e:
print("Oopsie")
exit()
for p in procs:
p.join()

这样的代码可以完美运行，除非您按 Ctrl+C。然后你会在 p.run(( 中得到未处理KeyboardInterrupt，即使它肯定是在 try/exclude 中处理的。

Traceback (most recent call last):
File "/home/keddad/Documents/thevyshka-news-fetcher/main.py", line 39, in <module>
p.run()
File "/usr/lib/python3.8/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/home/keddad/Documents/thevyshka-news-fetcher/fetchuler.py", line 19, in fetch_news
time.sleep(5)
KeyboardInterrupt

但是，如果我创建自定义 SIGINT 处理程序来抛出相同的KeyboardInterrupt，如下所示：

def raise_keyboards_ex():
raise KeyboardInterrupt()
signal.signal(signal.SIGINT, raise_keyboards_ex)

然后 try/except 开始按预期工作。那么，为什么一个KeyboardInterrupt在这里被抓住而其他的却没有呢？

Answer 1

所以，正如@jasonharper所说，它没有被抓住的原因不是某种魔法——只是因为KeyboardInterrupt不是Exception的子类。它现在抓住它的唯一原因是，我实际上没有上升KeyboardInterrupt而是造成了另一个异常：

try:
for p in procs:
p.run()
except Exception as e:
print("Oopsie")
print(e) # prints "raise_keyboards_ex() takes 0 positional arguments but 2 were given"
exit()