From: Markus <markus@...>
Date: 2004-10-09T12:19:26+09:00
Subject: Re: quality of error messages

     Sorry if I was not clear.  I have no reason to think it would be
hard to implement, but I haven't looked at how to do it (and am only
just learning the parser), so I don't want to declare that it would be
easy, or even possible.

    -- Markus

On Fri, 2004-10-08 at 17:58, "Pe単a, Botp" wrote:
> Markus [mailto:markus@reality.com] wrote:
> 
> >      * Augmenting 'end' (optionally) with a structure keyword and a
> >        name might be possible.  It would certainly provide enough
> >        information, but may or may not be implementable
> >
> 
> *How I wish I were as good as many of the people here (good in english and
> programming).
> 
> when I said: "make extended ends(endclass/enddef/endif) optional and make
> them  synonyms to end, thus programmer is free to use them or not",
> 
> I am _not advocating that we use all the extended ends. I am very happy w
> the simplicity and flexibility of the simple "end".
> 
> I am only suggesting that we (if possible):
> 
> 1. make endif/enddef/endclass optional and synonyms of end. Compiler does
> not care, but -c with_extended_end does (otoh, if it indeed cares, then much
> better).
> 
> 2. as a matter of style, make the use of extended ends sparingly (it is
> programmer's fault if his code looks dirty). Only use it with fear..
> 
> 3. even a simple insertion of one endif eg, can help a lot in matching
> pairs. One does not need to splat codes w endfoos...
> 
> Is #1 not implementable because of #2 & #3?
> 
> >-- Markus
> >
> 
> kind regards -botp
> 
> 
> 
> 
> 
>