On Wed, 6 Feb 2002, Simon Cozens wrote:
> Andy Dougherty:
> > Well, I know (at least) two ways to get rid of the warnings. I just don't
> > know which one's right. For example, making the bytecode be type
> > opcode_t* instead of char* gets rid of a lot of them
>
> Generally when I see char* in the source, I wonder why it's wrong and
> what it should be instead. If you just *assume* that char* is evil,
> allow yourself to be persuaded on the rare occasions that we actually do
> need it.
Oh, agreed in general. Specifically, though, is this one of those rare
occasions? Though I haven't done a complete audit of the source (and
would welcome specific pointers) all the cases I've looked at so far treat
the bytecode as a sequence of objects all of size sizeof(opcode_t).
--
Andy Dougherty [EMAIL PROTECTED]
