Re: Eliminating conflicts of parenthesized subexpressions

[Akim Demaille]

>>> "Hans" == Hans Aberg <address@hidden> writes:

 > I am inclined at following this line, to sort it out by action typing. This
 > makes the grammar simple. But it is not clear to me why it necessarily
 > should be viewed as semantic. One can think of grammars beyond the capacity
 > of Bison. The choice is merely dictated by practical concerns.

Semantic analysis is always due to the limited capacity of parsers.
Type-checking and other context depend analysis are 100% a matter of
_context dependent_ syntax.

But these grammars are a problem to handle (and to understand!).
That's why we do CFG and have other passes afterwards.  Indeed, "The
choice is merely *always* dictated by practical concerns."