What is the difference between ident and pid of processes in Pythons `multiprocessing`?

Question

Processes created and started with multiprocessing.Process have an attribute pid (the process id) and ident. Isn't pid also an identifier? What is the difference between those two?

The help isn't helping either:

 |  ident
 |      Return identifier (PID) of process or `None` if it has yet to start
 |  
 |  pid
 |      Return identifier (PID) of process or `None` if it has yet to start

Example

import multiprocessing
import time


def whatever():
    print("child")
    time.sleep(10)
    print("child done")

created = multiprocessing.Process(target=whatever)
created.start()
time.sleep(1)
print("pid={}".format(created.pid))
print("ident={}".format(created.ident))

In my cases, pid == ident. Is that always the case? Is that probably OS specific? (I run it on Linux. What happens on Mac / Windows?)

Answer 1

The answer is in the source code.

cpython/Lib/multiprocessing/process.py:

class BaseProcess(object):

    ...

    @property
    def ident(self):
        '''
        Return identifier (PID) of process or `None` if it has yet to start
        '''
        self._check_closed()
        if self is _current_process:
            return os.getpid()
        else:
            return self._popen and self._popen.pid

    pid = ident  # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

    ...

Thus multiprocessing.Process.pid is just an alias for multiprocessing.Process.ident. Note that multiprocessing.Process provides the ident property so that it is compatible with threading.Thread API. According to documentation:

In addition to the threading.Thread API, Process objects also support the following attributes and methods:

pid - Return the process ID. Before the process is spawned, this will be None.

...