Path: csiph.com!newsfeed.hal-mli.net!feeder3.hal-mli.net!newsfeed.hal-mli.net!feeder1.hal-mli.net!news.misty.com!news.iecc.com!.POSTED!nerds-end
From: patchell@cox.net
Newsgroups: comp.compilers
Subject: LRstar
Date: Sat, 13 Oct 2012 20:11:44 -0700 (PDT)
Organization: Compilers Central
Lines: 28
Sender: johnl@iecc.com
Approved: comp.compilers@iecc.com
Message-ID: <12-10-007@comp.compilers>
NNTP-Posting-Host: news.iecc.com
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Trace: leila.iecc.com 1350250477 15766 64.57.183.58 (14 Oct 2012 21:34:37 GMT)
X-Complaints-To: abuse@iecc.com
NNTP-Posting-Date: Sun, 14 Oct 2012 21:34:37 +0000 (UTC)
Keywords: parse, tools
Posted-Date: 14 Oct 2012 17:34:37 EDT
X-submission-address: compilers@iecc.com
X-moderator-address: compilers-request@iecc.com
X-FAQ-and-archives: http://compilers.iecc.com
Xref: csiph.com comp.compilers:765

I finally got around to using LRstar (by Paul Mann). I am doing some
retro-computing.  A for the fun of it project is to re-create the Rockwell
Aim65 PL/65 compiler.  The syntax seemed almost impossible to write a grammar
for that did not have conflicts.One of the hardest ones to resolve was for
this production:

assignment ->MemRef '=' expr
           ->MemRef '+' expr
           ->MemRef '-' expr

The production with the '+' in it would be similar to += in C.  And if I
changed the '+' to += and the '-' to '-=', that would of course fix the
problem, but it would not be compatible with PL/65 syntax.

I used LRstar to first analyze the grammar with the (nd=2 switch which let it
use lalr(k) parsing.  This of course worked.  But I also dug through the
conflicts report and was easily able to figure out how to fix the grammar to
make it 'lalr(1)' with no conflicts.  One of the LRstar operators, '/->' which
is the same as the '->' will tell LRstar which rules you want it to use when a
conflict occurs.  So I easily got the grammar going.  This is the first parser
generator I have used that gave me a conflict report I could understand.  So,
despite the fact I thought I needed LALR(k), I ended up only needing LALR(1).

Now, I just have to figure out how to make LRstar work with my AST class that
I made, or just figure out the AST generation that is built into LRstar.

-Jim