• Ellery Newcomer <ellery-newcomer utulsa.edu> Jun 24 2010
• Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> Jun 24 2010
• Ellery Newcomer <ellery-newcomer utulsa.edu> Jun 24 2010
• Clemens <eriatarka84 gmail.com> Jun 24 2010
• Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> Jun 24 2010

Ellery Newcomer <ellery-newcomer utulsa.edu> writes:

Hey,

in std.container, line 2623, I'm seeing

static if (is(_store.insertBack(value)))

The spec doesn't mention allowing expressions as far as I can remember, 
so what's the deal? Yes, the argument is ambiguous with type at parse 
time, but it has to get converted to an expression at some point. Is 
allowing expressions in is expression a definite keeper?

Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:

On 06/24/2010 10:01 AM, Ellery Newcomer wrote:
 Hey,

 in std.container, line 2623, I'm seeing

 static if (is(_store.insertBack(value)))

 The spec doesn't mention allowing expressions as far as I can remember,
 so what's the deal? Yes, the argument is ambiguous with type at parse
 time, but it has to get converted to an expression at some point. Is
 allowing expressions in is expression a definite keeper?


My mistake, a typeof should wrap the expression.

Andrei

Ellery Newcomer <ellery-newcomer utulsa.edu> writes:

On 06/24/2010 10:02 AM, Andrei Alexandrescu wrote:
 On 06/24/2010 10:01 AM, Ellery Newcomer wrote:
 Hey,

 in std.container, line 2623, I'm seeing

 static if (is(_store.insertBack(value)))

 The spec doesn't mention allowing expressions as far as I can remember,
 so what's the deal? Yes, the argument is ambiguous with type at parse
 time, but it has to get converted to an expression at some point. Is
 allowing expressions in is expression a definite keeper?


 My mistake, a typeof should wrap the expression.

 Andrei


Ah, very good. Maybe the compiler should bug out on expressions?

Clemens <eriatarka84 gmail.com> writes:

Andrei Alexandrescu Wrote:

 On 06/24/2010 10:01 AM, Ellery Newcomer wrote:
 Hey,

 in std.container, line 2623, I'm seeing

 static if (is(_store.insertBack(value)))

 The spec doesn't mention allowing expressions as far as I can remember,
 so what's the deal? Yes, the argument is ambiguous with type at parse
 time, but it has to get converted to an expression at some point. Is
 allowing expressions in is expression a definite keeper?


 My mistake, a typeof should wrap the expression.


...and this is exactly why we need meta.compiles(...) or something equivalent.
That proposal had such an overwhelming backing by the community; what happened
to it? There never was an official comment on it AFAIR. Such an addition should
be fine even with TDPL out since it doesn't break anything, shouldn't it?

Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:

On 06/24/2010 10:48 AM, Clemens wrote:
 Andrei Alexandrescu Wrote:

 On 06/24/2010 10:01 AM, Ellery Newcomer wrote:
 Hey,

 in std.container, line 2623, I'm seeing

 static if (is(_store.insertBack(value)))

 The spec doesn't mention allowing expressions as far as I can remember,
 so what's the deal? Yes, the argument is ambiguous with type at parse
 time, but it has to get converted to an expression at some point. Is
 allowing expressions in is expression a definite keeper?


 My mistake, a typeof should wrap the expression.


 ...and this is exactly why we need meta.compiles(...) or something equivalent.
That proposal had such an overwhelming backing by the community; what happened
to it? There never was an official comment on it AFAIR. Such an addition should
be fine even with TDPL out since it doesn't break anything, shouldn't it?


Yes, such an addition shouldn't interfere.

Andrei