|
![](/i/fill.gif) |
Nigel Stewart wrote:
> > So why not make a commercial library to parse POV? That would solve your
> > problem
>
> No, we are talking about a way that POV can be parsed
> _without_ a special POV parser.
>
So, are you talking replacing POV's existing language, or simply
supplementing it? I was under the impression it was the latter,
but if you don't throw away backward compatibility concerns, this
solution will only parse XML format neo-POV files.
Either a POV file can or cannot contain old style code; If it
cannot, that's an end to backward compatibility. If it can, then
what is the generic XML parser going to do when it runs into non-XML
POV code? Ignore it? That's not a total solution and seems vaguely
like writing a parser that only recognizes the keywords added in
MegaPOV, and discards all standard 3.1 code. And #version won't
help, since the generic XML parser doesn't gain POV parsing
capabilities when it encounters a #version directive. You could
write a custom XML parser that also has old style POV parsing
capabilities, but then you've made more work instead of less.
If I understand other discussions around similar subjects, there
has recently been a proposal to create an external API style
language that is preprocessed into standard POV code. Perhaps an
XML style version of this proposal would work better than actually
changing the base SDL of POV-Ray. A scene code in XML style
code would then have all the advantages you indicate without any
of the encumbrance of backward compatibility concerns.
Post a reply to this message
|
![](/i/fill.gif) |