kdev-pg: Token definitions

[Jakob Petsovits]

On Wednesday, 8. February 2006 18:29, Roberto Raggi wrote:
> > Hmm... to me this looks very much like a lexer replacement. ANTLR has a
>
> it's a bit more complicated than that :-) of course you can use LL(K) to
> tokenize the buffer, but my feeling is it will make the things very slow
> :-) you know.. recursive function call, and so on..
>
> of course in future we can integrate a scanner generator in kdev-pg. I'd
> love to specify the regular expression that recognize the token in kdev-pg,
> but to do that we have to provide:
>  (...)

erm, no, I didn't mean we should implement an own lexer, I just meant that the 
proposed syntax looks like it would trigger the lexer (even if it doesn't).
I just wanted to say that if we put the "error" strings in parentheses,
it's less likely that the user confuses them with lexer capabilities.

sorry for the riots,
  Jakob