Re: %destructor feedback

[Joel E. Denny]

On Wed, 19 Oct 2005, Wolfgang Spraul wrote:

> > Are you now talking about a YYCLEANUP macro or YYABORT_CLEANUP and
> > YYERROR_CLEANUP macros?
> >
> > If you mean YYCLEANUP, it would need to return.  Would a function call be
> > too inefficient?
>
> Sorry, I made a mistake here. You are right - if YYCLEANUP is a goto, then it
> cannot return. Surely a single function call would not be too much overhead
> in my opinion.

Great.

> I would prefer something like YYERROR_CLEANUP and
> YYABORT_CLEANUP in _addition_ to the regular YYERROR/YYABORT macros.

After reading the ChangeLog, I've noticed there would also need to be a 
YYACCEPT_CLEANUP.  More encouragement to go the single YYCLEANUP route.

> My vote is in - YYERROR_CLEANUP+YYABORT_CLEANUP, but I would happily accept to
> be the minority, and I wouldn't even have the time to create a patch for my
> two cleanup macros anyway, so I depend on someone else's generosity to offer
> or not offer 'more cleanup flexibility'.

Unless someone (hopefully) beats me to it, I may be the one to implement 
this.  If so, I'll go for a single YYCLEANUP function.  It may not be your 
strongest preference, but it seems it'll satisfy your needs.

A few warnings:

1. I believe there's a feature freeze until 2.2 is released.

2. I may rename YYCLEANUP to something better: yycleanup_rhs() or 
something like that.  Just watch the mailing lists.

Thanks for the discussion.

Joel