comp.lang.ada
 help / color / mirror / Atom feed
From: Robert A Duff <bobduff@shell01.TheWorld.com>
Subject: Re: Q: recent Ada grammar for Ada tools ?
Date: Thu, 20 Sep 2007 14:17:36 -0400
Date: 2007-09-20T14:17:36-04:00	[thread overview]
Message-ID: <wcc3ax9dwvz.fsf@shell01.TheWorld.com> (raw)
In-Reply-To: 1190305120.14016.6.camel@sonnenregen

Georg Bauhaus <bauhaus+rm.tsoh@maps.futureapps.de> writes:

> Wouldn't the sources of the LRM have the grammar rules marked up such
> that they can be extracted?

Yes.  In fact, that's already done -- Annex P, "Syntax Summary"
is just that.

However, the RM grammar is not what you want for AYACC, which I believe
is a LALR(1) parser generator.  The RM grammar is ambiguous, for one
thing.  E.g., what is "X(Y)" in the following:

    A := X(Y);

function_call, type_conversion, indexed_component?
If it's a function_call, what is "X" -- name or
implicit_dereference?

According to the grammar, it can be any of the above,
and some I didn't mention.

Here's an interesting part of the RM grammar:

    prefix ::= name | implicit_dereference
    implicit_dereference ::= name

>...I have never seen the sources and
> I don't know whether they are publicly available at all (like the
> processed sources are).

Surely they're publicly available somewhere?
Maybe Randy can tell us.

- Bob



  reply	other threads:[~2007-09-20 18:17 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-09-16  5:17 Q: recent Ada grammar for Ada tools ? Gautier
2007-09-16  7:10 ` Pascal Obry
2007-09-16  9:58   ` Gautier
2007-09-20 16:18     ` Georg Bauhaus
2007-09-20 18:17       ` Robert A Duff [this message]
2007-09-21  0:24         ` Randy Brukardt
2007-09-21  3:36           ` Georg Bauhaus
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox