OpenSAML user discussion

["Scott Cantor" <>]

> Though still in development, we are interested in its 
> capabilities and like to be able to use it for testing. 
> Basically Ive tried to get the C++ version of OpenSAML2 to 
> work using the Trunk checkout of SVN. Im using Visual Studio 
> 2005 Pro but am getting a lot of errors about not being able 
> to instantiate a abstract class 
> opensaml2\saml\saml2\metadata\metadata.h(161) : error C2259: 
> 'xmltooling::AttributeExtensibleXMLObject' : cannot 
> instantiate abstract class.

The code currently builds out of trunk. That's not true all of the time
since it's still under development, but it's usually true. Occasionally
there are Unix issues if I haven't done a build there in a while, but mostly
with makefiles. Windows definitely builds.

I can't really do much to help resolve errors that are local. I've tried to
minimally document the build process for the time being, but there will be
more later once I firm up the library versions.

That particular error doesn't make much sense. I would assume you've got the
wrong xmltooling code, maybe, but I can't tell.

> Now I know this is of course still an alpha/beta version, but 
> is it workable yet?

Workable, yes. Supported, no, and the APIs may still change. And there still
is no actual release of XML-Security that works fully with this code. 1.3.1
is due out shortly.

> So far, this is the only implementation 
> of SAML2 Ive seen out there, unless anybody knows of any 
> other, Id like to know to.

There are few, if any, that are designed as libraries for others to build
on. There are projects that have code in them in the service of whatever
they're delivering themselves, like zxid.org, which is in C. There's also
lasso, from entrouvert, also in C.

I'm not aware of any other C++, anywhere.

> If possible Id like to be able to use the OpenSAML2 library 
> in our research/proof of concept project. Could you shed some 
> light on the possibilities of the library in the near future?

I expect to be nearly feature complete by end of year, and probably API
stable shortly after.

There is at least one other group using the code, but two projects doesn't
incent me very strongly to spend the time on an official beta. If I get more
interest, I'll consider it, but I have to wait for the Apache code to
finalize.

-- Scott