In article <3c95fd1b@news.povray.org>, Warp <war### [at] tagpovrayorg> 
wrote:

>   So we reduce the number of reserved keywords by 0 and we force the parser
> to make a much more complicated task (it starts by creating a union, but
> in the middle of parsing it suddenly has to convert it to merge).

The goal is not to reduce the number of keywords, where did you get that 
idea? I'm talking about simplifying the syntax.
And it wouldn't be nearly as complex as you imply...actually, it would 
probably be simpler, they already share most of their code. You wouldn't 
want to convert from one type of object to another, that *would* be a 
stupid way to do it.


>   Same problem. Number of reserved keywords are reduced by 0 and the parser
> has to change the lathe to a sor (which might be a bit complicated).

Not complicated at all, and I'm not trying to reduce the number of 
keywords.


>   You said that this would simplify the syntax. I don't see how. If all of
> them are made into one sweep primitive, we end up getting one reserved
> keyword more than currently. I don't see how this makes anything simpler.

Stop counting reserved keywords, it is useless.
One shape rather than three. Easier to maintain, easier to learn, easier 
to extend with additional sweep types.


> > Smooth triangles could be made part of ordinary 
> > triangles.
> 
>   And how do we differentiate between them?

Smooth triangles have normals, obviously.

-- 
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/

Post a reply to this message