Re: user destructor for unresolved GLR semantic value

[Joel E. Denny]

On Thu, 25 Aug 2005, Paul Eggert wrote:

> Thanks for mentioning this.  However, I'd rather not complicate
> all the skeletons for the sake of this GLR-specific problem.

In my approach, "c.m4" required only a two line addition that made
yydestruct() safer anyway, but that was about it.  As I pointed out, the
one line change to yysymprint() wasn't really required.  However, I like
your patch because...

> I installed the following patch instead.  It outputs a bit more
> information than the patch you proposed, since it mentions which
> popped items were unresolved.  Please let me know how it works for you.

The extra information is helpful.

> On Wed, 24 Aug 2005, Joel E. Denny wrote:
>
> 3. Although my patch prevents the potential seg fault, I believe there's
> still a potential memory leak that it does not address.  That is, if
> !yyresolved, the yyGLRState stores a list of yySemanticOption's.  Each
> yySemanticOption has a yystate member.  That yystate is the last state in
> a RHS potentially containing many semantic values.  For this case of an
> error during a split stack, I cannot find where the user destructors are
> called for those semantic values.  Based on the `well-formed' comment at
> `yyreturn:', I wonder if this might be a conscious decision.  I'll
> continue to explore this, but does anyone already know what's going on?

I was almost done with a patch to this problem when I saw your patch. I'm
glad to see you created yydestroyStackItem().  I had already created a
similar function in my new patch. Once I update it for your version of the
function, I'll post it.  It'll probably be later today.

Joel