Alix Pexton wrote:
On 31/03/2011 18:37, Jonathan M Davis wrote:
On 2011-03-31 06:35, Kagamin wrote:
Ary Manzana Wrote:
I just hate it when you have to write too much
hasMember!(S, "m")
is only 1 character longer than
S::hasMember("m")
not too much for me
And hasMember!(S, "m") is actually consistent with the rest of the
language
and straight forward to read for those who know the language.
S::hasMember("m") just adds more syntax where there's no need for it, and
makes it so that there's that much more syntax to learn and keep
straight.
- Jonathan M Davis
I thought the :: syntax seemed pretty intuitive (keystroke savings never
entered into my evaluation), but I seem to be in a minority on that.
After re-reading the previous discussion of __traits and its proposed
replacement, however, I am reminded of an important point: language
features should not have special powers that can't be replicated in
client code.
So, with that in mind, I have to agree that using templates for things
like hasMember has to be the way forward.
Having templates that are internal to the complier or that simply
forward to __traits seems a bit odd to me. I think I'd rather see
__traits simplified and more of work moved into the library, which may
make it more versatile and easier to extend (pure speculation ^^).
Right, I'm off to go and squirrel all my uses or traits away inside
templates, I have a sneaking suspicion that it is going to make some big
improvements in readability ^^
A...
The chief problem is that there is no way to hide __traits(compiles).
It's just impossible. The whole idea of the meta namespace was to make
it possible to hide the __traits.
Don't forget that an important part of that proposal was to clean up
is(), as well.