Re: new feature -> with()

Rasmus Lerdorf skrev:

I think that is one of the strongest reasons not to implement something actually. If there is a way to do something in a clear and concise syntax, adding an alternate less clear syntax that isn't immediately obvious to everyone simply obfuscates the language. The other fatal blow to a feature like this is that it would be quite a bit slower to compile and execute than the longer form because we now can't simply do a single hash lookup in the function table when we see a function. We first have to walk up the tree of objects to see if there is a method with this name, and if there is no method, then we assume it is a straight function call, unless of course there is a __call() method, then what? Way too many gotchas here.

May I also add that the use of "with" is generally considered bad practice by leading JavaScript gurus like Douglas Crockford, David Flanagan, John Resig and PPK. I even have a vague memory of Brendan Eich commenting upon the feature negatively as well. Crockford speaks at: http://yuiblog.com/blog/2006/04/11/with-statement-considered-harmful/ Pro s& Cons are listed at: http://developer.mozilla.org/en/docs/Core_JavaScript_1.5_Reference:Statements:with Erik Arvidsson warns against the use in JS at: http://erik.eae.net/archives/2005/04/11/18.39.51/ Lars Gunther