Re: Regressions in the syntax

[Pavel Roskin <proski gnu org>]

> > context linestart === \n brown/12
> 
> I tested it with sh.syntax and c.syntax where "whole linestart" is ok.

"whole linestart" was OK, "context linestart" wasn't.

> > Please test your changes.  I don't even have time to apply patches, so I
> > really don't have time to spend on regressions.
> 
> OK. I am too discouraged to fix syntax files in the nearest time.

I don't want to discourage you from fixing syntax files.  Let's learn from
the errors and prevent others from repeating them.  Keywords starting with
"\[" don't work - let's see if it can be fixed.  It it cannot be fixed
easily, let's add a warning.

I think that it should be OK to improve the existing syntax code.  If
Cooledit users want to use our rules - fine.  Our changes to the code are
open and documented.  Port them to Cooledit and use our syntax files.

In particular, I think that we should remove color numbers, i.e. convert
"yellow/24" to "yellow" and so on.  Cooledit can interpret those names
with a trivial patch.  Cooledit users can also add better colors for X.  

On the other hand, it's unreasonable to expect mc contributors to get
those numbers right, since they would have to install Cooledit just to
test them.

-- 
Regards,
Pavel Roskin