Is the description of threading.Thread.is_alive a bit confusing?

Documentation
1

The is_alive() method in threading module, is described as this:

This method returns True just before the run() method starts until just after the run() method terminates.

Doesn’t that mean is_alive return True before the run() method starts? I am not a native speaker, so I am not sure whether it is indeed worded in a confusing way or just I didn’t quite understand it.

2

It says “just before” and “just after”. In colloquial English this
implies as close to the start of run() as possible and as close to the
end as possible…

I expect that the module endeavours to ensure that is_alive() returns
True for the entire period when run() is running, but as close to
that as possible. Consider how this is probably implemented
(pseudocode):

 in-the-new-thread:
     T._is_alive = True
     try:
         run()
     finally:
         T._is_alive = False

That sets the flag then immediately runs the function. But technically
the flag will be set before the function commences, and cleared
afterwards.

Cheers,
Cameron Simpson cs@cskk.id.au

3

I agree that the wording in the documentation could be better, but you got it right:

Yes. When the docs say that is_alive() returns True just before run() starts, it means that Python will first set a flag to indicate that the thread is alive, and then start executing the run() method. That is, there is a brief period of time in which the flag has been set, but the run() method has not yet been called.

1 Like
4

You missed the first statement that is key.

Return whether the thread is alive.

The logic is likely

  1. Thread starts
  2. is_alive = True
  3. run() called
  4. is_alive = False
  5. Thread exits

The purpose is to tell you that the thread is alive, run() may or may not be running. At a high level you can think of the run() being the same as the thread running.

The details are not as simple see https://github.com/python/cpython/blob/main/Lib/threading.py#L1214

1 Like