Re: push parser

[Joel E. Denny]

On Sat, 16 Dec 2006, Bob Rossi wrote:

> > BTW, after knowing all of this, if I were to write the patch, I would do
> > it this way. I would leave the pure-parser option set to false. But I
> > would write a new macro, b4_pure_or_push (or something), and replace all
> > calls to b4_pure that are supposed to be b4_pure_or_push.

At the moment, that would be every call to b4_pure.  And b4_pure would 
never be used outside of b4_pure_or_push.

Let's say we one day develop another kind of modifier like %push-parser.  
I'll call it xyz and %xyz-parser.  Let's say we decide that all xyz 
parsers must also be pure parsers.  Are we then going to write 
b4_pure_or_push_or_xyz?  And then b4_pure_or_push_or_xyz_or_abc?

> > That would
> > leave bison with the ability to perform all combinations of checks,
> > which I don't think it can currently do now.

Assuming my patch, here are all 4 combinations, each with a check that 
will be true iff that combination is true:

     pure   push   check
  -----------------------------
  1  false  false  !pure_parser
  2  false  true   false
  3  true   false  pure_parser && !push_parser
  4  true   true   pure_parser && push_parser

So, assuming my patch, all checks are possible.  Notice that, since all 
push parsers are pure, #2 is a fallacy.  We can check it and we'll get the 
right answer, but what's the point?