[boost] Re: Re: Formal Review: The Spirit Parser Framework.

From: Carl Daniel (cpdaniel_at_[hidden])
Date: 2002-10-12 08:33:37

"Joel de Guzman" <djowel_at_[hidden]> wrote in message
news:001901c271e7$caa8a830$0100a8c0_at_kim...

> Well, I am perfectly aware of that. However, if that's the only rationale,
> what you are suggesting is something I definitely would not agree on.
> First of all, the use of exceptions in Spirit is paired. Parser assertions
> are always caught by the guard. It makes no sense to write parser
> assertions without guards. Yet, if this is the intent of the programmer,
> I have to trust her 100%.

This is true :-)  I withdraw the suggestion - no one outside Spirit will
ever need to catch parse_error<>.  If an assertion is used without a
corresponding guard, the program _should_ terminate due to an unhandled
exception.

I'm less convinced about illegal_backtracking - I accept your argument that
it's intended to be an independent component and so shouldn't derive from
parse_error_base, but perhaps it should derive from std::exception.

fixed_size_queue should do _something_.  At the very least, it should
contain assert(false) in the places where a full queue is detected.

-cd