[asterisk-dev] Reverse Inheritance

[Tilghman Lesher]

On Wednesday 12 November 2008 16:24:26 Steve Murphy wrote:
> On Thu, 2008-11-13 at 10:50 +1300, Matt Riddell wrote:
> > On 12/11/2008 9:48 p.m., Grey Man wrote:
> > > I'd vote for +. That way next year when passing around one variable
> > > evolves to become passing around groups of variables objects could be
> > > invented and the + could become ++ and then we'd have C++ ;-).
> >
> > I guess if it can only be used in set.  I just wonder if someone might
> > read my extensions.conf and assume I'm incrementing a value before using
> > it or something.
>
> I hope it's not too late to throw in my 2 cents on this subject...
>
> I highly, highly, highly recommend NOT using syntactic indicators
> that will control inheritance. Special naming conventions like
> underscores before or after do not harm anything; but using
> what could be construed as operators will play havoc with the
> existing expression parser. Don't extend operator syntax into
> object names; it will limit what operators we will be able to use
> in anything that has to deal with those names.
>
> Like I said, using underscores doesn't extend or change the rules
> for parsing names, so that's OK, but almost anything else will
> have unintended effects that will be negative.

I don't see how this would have an effect on the names.  The ONLY time
that you refer to the prefixed characters are when you set the variable
for inheritance.  They are NOT part of the name of the variable.  So, it's
only in the Set() that the characters come into play, and only when setting
those particular variables, never when referring to them.

-- 
Tilghman