Antoine suggested me to add these functions in the issue #22042.

Here is a patch to implement these functions and use them in various places.

Oops, test_asyncio was failing. Fixed patched.

Can someone please review my patch?

Maybe is_something() is a better name than get_something()
if something is a boolean?

> Maybe is_something() is a better name than get_something()
> if something is a boolean?

os and socket modules already use getxxx() / setxxx() names:

- os.get_inheritable() / os.set_inheritable()
- os.getuid() / os.setuid()
- os.getgroups() / os.setgroups()
- os.getxattr() os.setxattr()
- socket.gettimeout() / socket.settimeout()
- socket.get_inheritable() / socket.set_inheritable()
- etc.

There is only one "is" function: os.isatty(), but it does not have a "os.settty" method.

> os and socket modules already use getxxx() / setxxx() names:
>
> - os.get_inheritable() / os.set_inheritable()
> - os.getuid() / os.setuid()
> - os.getgroups() / os.setgroups()
> - os.getxattr() os.setxattr()
> - socket.gettimeout() / socket.settimeout()
> - socket.get_inheritable() / socket.set_inheritable()
> - etc.

egid, euid, gid, groups, pgid, pgrp, priority, resgid, sid, uid, xatrr,
hostname, timeout, sockopt are not boolean. Though
get_inheritable() spoils the pot.

Compare:

  if os.get_blocking(fd): # is it a bug? Does it return some mode bits?
     # Do I need stat.IS_BLK(os.get_blocking(fd)) here instead?

And:

  if os.is_blocking(fd):
     # it is clear that fd is blocking

There could be an argument that get_inheritable should be renamed to
is_inheritable instead.

I agree with Akira, although it's probably too late now to rename.

Guido van Rossum told me that he's ok for "get": "I'd go with get, since that seems the tradition in the os module."

New changeset 8f0b8ddbb66b by Victor Stinner in branch 'default':
Issue #22054: Add os.get_blocking() and os.set_blocking() functions to get and
http://hg.python.org/cpython/rev/8f0b8ddbb66b