Re: DSEL syntax, or opening the can of worms.

on Wed Aug 08 2007, John Moeller <fishcorn-AT-gmail.com> wrote:

> Rene Rivera wrote:
>> namespace_(L,"a")
>> [
>>    def("f",&a::f),
>>    class_<a::A>("A")
>>    [
>>      def("do",&a::A::do),
>>      class_<a::A::B>("B")
>>      [
>>        def("again",&a::A::B::again)
>>      ]
>>    ]
>> ]
>> 
>> Which has a one-to-one correspondence with the C++ code. So... Was this 
>> ever considered? If it was, why was it not used? Is there a discussion 
>> in the list I haven't read about this?
>
> I don't think it was ever discussed; 

It was, though it may have been a private thing between me and Daniel.

> I've been over the history at least twice now, and I don't believe
> that the DSEL was discussed to this extent.  I think that only minor
> details were covered.  I could have missed something, though.  There
> isn't any code to cover the frontend either, so it seems that the
> DSEL is open territory for discussion.
>
> Just a peanut gallery comment: I like Rene's proposed syntax
> better. 

But as Daniel points out, there are some practical disadvantages.

> Ostensibly, you should be able just to drop a class into another
> class' scope (though there may be details that would cause some
> difficulty that I haven't thought of).  It's also a lot smoother
> and, to me at least, more expressive.

Yeah, maybe... but this issue (and the one about explicit overload
selection, are really minor details that should have little impact on
the overall plan.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

The Astoria Seminar ==> http://www.astoriaseminar.com


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/