Subj : Re: E4X
To   : Jiri Znamenacek
From : Brendan Eich
Date : Thu Jul 22 2004 12:11 pm

Jiri Znamenacek wrote:
> I don't know about any better place so I'm going to ask here. And 
> despite the comments I like this extension to ECMAScript very much.
> 
> Why "XPath" on objects is written in this way
>   e.employee.(name == "Joe")
> e.g. object->object->special_kind_of_object?


Because that's a syntactic extension, which is allowed by ECMA-262 
Edition 3 Chapter 16.


>  From the JS point of view (erm, MY point of view) I would expect rather
>   e.employee[name == "Joe"]
> which at first sigh returns object(s) of "type" employee.


That's not a syntactic extension, it is an incompatible revision to 
existing syntax and semantics: a[name == "Joe"] today should evaluate 
the index expression and convert its boolean result to a string, 
referencing a property identified by either 'true' or 'false', depending 
on whether name == "Joe".

/be

.