[cfe-dev] Help getting started

[David Blaikie]

Sorry, was a bit tired I guess. I meant polymorphic lambdas (template
type parameters) - I wasn't intending to suggest some other construct
you hadn't suggested.

I'm still not sure I follow the ambiguity resolution. Are you saying
this named function template syntax would diverge from lambdas? (in the
sense that a single identifier specifies an unnamed parameter in a
lambda, but an untyped on in this syntax you're proposing)
From: Dave Abrahams
Sent: 1/2/2012 5:17 AM
To: David Blaikie
Subject: Re: [cfe-dev] Help getting started

on Sun Jan 01 2012, David Blaikie <dblaikie-AT-gmail.com> wrote:

> Interesting post. One side question:
>
> You mention in your post that there might be a way to do variadic
> (essentially template) lambdas by simple omitting the type name.

Support for variadics is an interesting possibility that I hadn't
considered.

> How do you solve the ambiguity between an unnamed parameter and an
> untyped one?

By using a syntax that says "I'm leaving out the type names," i.e., by
starting with "[]" instead of "auto" or a type name.

> As for implementing your experimental terse function template
> definition syntax - somewhat naively (I'm by no means an expert in
> these matters) i expect it'll end up halfway between the two approaches
> you've mentioned.
>
> My own approach to adding features to clang so far has been to write
> the test case first, see where clang fails (break on whatever error is
> produced) & walk up Clang's stack from there looking for the spot that
> might be a reasonable place to add the functionality.

That sounds interesting, thanks.  I might take that approach.

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com