shibboleth-dev - RE: Linux update
Subject: Shibboleth Developers
List archive
- From: Scott Cantor <>
- To: 'Derek Atkins' <>
- Cc:
- Subject: RE: Linux update
- Date: Wed, 11 Sep 2002 12:21:34 -0400
- Importance: Normal
- Organization: The Ohio State University
> So in other words I should be able to safely dynamic_cast<>
> any of my own objects, but I may need to work around times
> where I cast a SAML or Shib object.
That's about the best summation I could give you at this point.
> What about when I use a saml object's
> clone() method -- is that safe to dynamic_cast<> back to itself?
It will probably crash, but that's a safe static_cast, since you know
what you just cloned (unless you're cloning something in a really
generic way).
In fact, the C++ standard apparently now permits virtual functions to
have overridden return types, if the type is a pointer or reference to
the derived class. So the proper way to build clone() type interfaces is
by having each virtual override of clone() return a pointer of that
class type.
Unfortunately, almost no compilers support that yet, so I can't do it.
-- Scott
------------------------------------------------------mace-shib-design-+
For list utilities, archives, subscribe, unsubscribe, etc. please visit the
ListProc web interface at
http://archives.internet2.edu/
------------------------------------------------------mace-shib-design--
- Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
- RE: Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
- RE: Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
- RE: Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
- RE: Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
- RE: Linux update, Scott Cantor, 09/11/2002
- Re: Linux update, Derek Atkins, 09/11/2002
Archive powered by MHonArc 2.6.16.