Mark H. Wilkinson wrote:
>
> Nathan Meyers wrote:
>
> > Java parametric polymorphism is a pretty tall order for cpp to try to
> > handle, but I'm willing to be convinced.
> >
> > This topic is quickly drifting off-topic. Bernd, if you could convince
> > your ISP to stop bouncing mail from my ISP, perhaps we could continue
> > the conversation without boring all of these bystanders.
>
> Some of these bystanders are interested too, and I don't know where else
> would be suitable for this kind of discussion. How'd you find pizza and
> gj? I haven't gone into them much myself, but I've done alot of work on
> generic programming with C++ and Sather, and a bit of ML, so I was
> interested to see the list of people involved in gj.
Sorry to exclude you. I haven't touched pizza, which is older than GJ.
I've had good success with GJ until recently... but the recent
experience has been bad enough to make me rethink relying on it. My
current problems:
1) GJ has been stuck at version 0.6i for a while, with little evidence
of activity. The pizza/GJ mailing list appears to be permanently
non-functional.
2) I recently discovered two bugs, one minor and one major. I've gotten
no response to my bug report. The bugs don't assert if GJ is used in
preprocessor mode, so I've gone to using it that way, but the whole
arrangement makes me nervous.
(In case you're curious, the minor bug is a failure under certain
circumstances to report "Can't reference this before the superclass
constructor has been called". The major bug is truly bizarre: the code
generated for the expression (true ? null : "notnull") returns a value
of "notnull".)
So, there... I've strayed off-topic for the last time :-).
Nathan Meyers
[EMAIL PROTECTED]
