Path: csiph.com!eeepc.pasdenom.info!news.pasdenom.info!news.dougwise.org!nntpfeed.proxad.net!proxad.net!feeder1-1.proxad.net!198.186.194.250.MISMATCH!news-out.readnews.com!news-xxxfer.readnews.com!news.misty.com!news.iecc.com!nerds-end
From: torbenm@diku.dk (Torben Ægidius Mogensen)
Newsgroups: comp.compilers
Subject: Re: how to parse this?
Date: Tue, 01 Feb 2011 13:35:34 +0100
Organization: SunSITE.dk - Supporting Open source
Lines: 40
Sender: news@iecc.com
Approved: comp.compilers@iecc.com
Message-ID: <11-02-002@comp.compilers>
References: <11-01-095@comp.compilers>
NNTP-Posting-Host: news.iecc.com
X-Trace: gal.iecc.com 1296583611 82879 64.57.183.58 (1 Feb 2011 18:06:51 GMT)
X-Complaints-To: abuse@iecc.com
NNTP-Posting-Date: Tue, 1 Feb 2011 18:06:51 +0000 (UTC)
Keywords: parse, yacc
Posted-Date: 01 Feb 2011 13:06:50 EST
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:976

nojb <n.oje.bar@gmail.com> writes:

> Hi,
>
> My language has function applications like ML, i.e. f g h means f
> applied to two parameters g and h. But the rule
>
> exp: IDENT list(exp)
>
> in Yacc will parse this as f (g h). How can I fix this?

You can either fix it by post-processing the parse tree, or you can
write the production using explicit left-recursion:

exp: application

application: IDENT exp
           | application exp

If applications are really ML-like, they don't need to start with an
identifier, so you should rather have something like

exp: exp exp

which, however, is ambiguous.  You can fix this by adding a %prec
declaration or by dividing expressions into simple and complex
expressions and only allowing simple expressions as arguments to
function applications, i.e.,

exp: exp simple
   | simple

simple: IDENT
      | LPAR exp RPAR
      | ...

where LPAR and RPAR are the token names for parentheses.

	Torben