Re: virtual method inheritance question

[jinxidoru]

Paul Pluzhnikov wrote:
>
> There is a more fundamental reason why it would not work:
> a non-static method can not be a signal handler, since the kernel
> will not supply 'this' parameter to it.
>
Please just stick to the actual question.  I already have my class
handling signals just fine.  I have a workaround for the above issue.

> Besides, all kinds of things (operator new; cout; exceptions,
> etc.) do not work in the (async)signal handling context, so putting
> any C++ into the body of a signal handler will likely only itroduce
> hard to replicate bugs.
>
When did I say I was going to put C++ code into the signal handler?  I
know the restrictions of putting C++ code into a signal handler.  Once
again, please just answer the actual question or don't respond.  The
question is about addressing virtual method addresses, not the wisdom
of using signal handlers in C++.

> What's so cool about it? Sounds pretty lame to me.
>
Go to *#&@.  If you have nothing useful to say, please control yourself
and don't respond at all.  It's a lot easier than allowing everyone on
the internet to see how big of a jerk you are.