Jeffrey Yasskin <jyass...@gmail.com> added the comment:

You're definitely right that posix and java don't make these usable from the 
normal lock API. And I don't think it's essential that they be usable as 
RLocks, although it's nice for Conditions. I think what I'm actually saying is 
that, if you have an acquire() operation on RWLock, then it has to forward to 
the write lock. Forwarding to the read lock would have different semantics from 
RLock.acquire (2 threads could get in at once) and so would be incorrect.

I agree that RWLock is more obvious. As long as the docs mention that it's 
equivalent to a shared/exclusive lock, I don't care that much about the name. 
Just throwing out ideas.

----------

_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue8800>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

Reply via email to