[r6rs-discuss] Better diagnostic messages for unmatched parentheses?

From: Matthias Felleisen <matthias>
Date: Thu, 14 Jun 2007 09:26:16 -0400

Take a look at DrScheme's (Module) behavior for these cases. -- Matthias



On Jun 14, 2007, at 3:53 AM, r6rs-discuss-request at lists.r6rs.org wrote:

> Message: 6
> Date: Wed, 13 Jun 2007 21:02:04 -1000 (HST)
> From: Shiro Kawai <shiro at lava.net>
> Subject: [r6rs-discuss] Better diagnostic messages for unmatched
> parentheses?
> To: r6rs-discuss at lists.r6rs.org
> Message-ID: <20070613.210204.586262237.shiro at lava.net>
> Content-Type: Text/Plain; charset=us-ascii
>
> One of the most frequent complaints from entry-level Scheme
> programmers is that, when they miss one close parenthesis of
> a definition, a naive implementation reports missing parenthesis
> at the end of file, giving no clue of which definition has
> the problem.
>
> I made Gauche's reader report the position of opening
> parenthesis which misses the closing one, and it works
> well for most cases to identify the toplevel definition
> that actually have missing closing paren. However, with
> R6RS library syntax this strategy becomes useless, for the
> reader will report the beginning of the 'library' form.
>
> I just wonder if those who have tried the proposed library
> form have any idea to address this issue.
>
> (Forcing users to use Emacs isn't a solution).
>
> --shiro
Received on Thu Jun 14 2007 - 09:26:16 UTC