• Andrej Mitrovic (7/7) Aug 02 2011 import std.algorithm;
• simendsjo (6/13) Aug 02 2011 import std.algorithm;
• Andrej Mitrovic (2/3) Aug 02 2011 Thanks!
• Andrej Mitrovic (17/17) Aug 02 2011 Does anyone know why putting this alias in module scope errors out?:
• Timon Gehr (17/34) Aug 04 2011 I believe it is because the function literal is typed as a template dele...
• Andrej Mitrovic (3/3) Aug 02 2011 Oh just realized I can use the much simpler:
• simendsjo (3/6) Aug 02 2011 Of course. I even had a lot of coffee running through my veins, so shame...
• bearophile (5/14) Aug 02 2011 See:
• Kai Meyer (3/10) Aug 04 2011 Looking at std.algorithm, I think what you really want is minCount:
• bearophile (4/6) Aug 04 2011 It's a bad design.
• Kai Meyer (2/8) Aug 05 2011 minCount is, or the usage of minCount in his particular problem?
• bearophile (6/7) Aug 05 2011 Your usage seems OK. What I meant is that I don't like the design of min...

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

import std.algorithm;

void main()
{
    auto x = min([1, 2, 3]);  // x would be 1
}

min() isn't equipped to do this on a single range. What can I use
instead? I haven't had my coffee yet. :)

simendsjo <simendsjo gmail.com> writes:

On 02.08.2011 14:03, Andrej Mitrovic wrote:
 import std.algorithm;

 void main()
 {
      auto x = min([1, 2, 3]);  // x would be 1
 }

 min() isn't equipped to do this on a single range. What can I use
 instead? I haven't had my coffee yet. :)

import std.algorithm;

void main() {
     assert(min(1, 2, 3) == 1);
     assert(reduce!"min(a,b)"([1,2,3]) == 1);
}

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

On 8/2/11, simendsjo <simendsjo gmail.com> wrote:
      assert(reduce!"min(a,b)"([1,2,3]) == 1);

Thanks!

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

Does anyone know why putting this alias in module scope errors out?:

import std.algorithm;

alias reduce!((a, b){ return 1; }) foo;

void main()
{
    foo([1, 2, 3]);
}

Error: delegate test.__dgliteral1!(int,int).__dgliteral1 is a nested
function and cannot
be accessed from reduce

But this will work:

import std.algorithm;

void main()
{
    alias reduce!((a, b){ return 1; }) foo;
    foo([1, 2, 3]);
}

Timon Gehr <timon.gehr gmx.ch> writes:

Andrei Mitrovic wrote:
 Does anyone know why putting this alias in module scope errors out?:

 import std.algorithm;

 alias reduce!((a, b){ return 1; }) foo;

 void main()
 {
     foo([1, 2, 3]);
 }

 Error: delegate test.__dgliteral1!(int,int).__dgliteral1 is a nested
 function and cannot
 be accessed from reduce

 But this will work:

 import std.algorithm;

 void main()
 {
     alias reduce!((a, b){ return 1; }) foo;
     foo([1, 2, 3]);
 }

I believe it is because the function literal is typed as a template delegate and
the compiler cannot figure out where to get the context pointer from (because at
module level, there is no need for a context pointer).
If you declare the alias inside the main function, the 'reduce' function will be
instantiated as a local function and therefore it has the context pointer. The
result is that you pass around some unnecessary context pointers (without
inlining). AFAIK Andrei wants this fixed.

You can explicitly tell the compiler to type the function literal as a function
pointer:

alias reduce!(function (a, b){ return 1; }) foo;

void main()
{
    foo([1, 2, 3]);
}

works.

-Timon

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

Oh just realized I can use the much simpler:

auto x = reduce!(min)([1, 2, 3]);

Coffee is starting to kick in!

simendsjo <simendsjo gmail.com> writes:

On 02.08.2011 14:42, Andrej Mitrovic wrote:
 Oh just realized I can use the much simpler:

 auto x = reduce!(min)([1, 2, 3]);

 Coffee is starting to kick in!

Of course. I even had a lot of coffee running through my veins, so shame 
on me!

bearophile <bearophileHUGS lycos.com> writes:

Andrej Mitrovic:

 import std.algorithm;
 
 void main()
 {
     auto x = min([1, 2, 3]);  // x would be 1
 }
 
 min() isn't equipped to do this on a single range. What can I use
 instead? I haven't had my coffee yet. :)

See:
http://d.puremagic.com/issues/show_bug.cgi?id=4705

Bye,
bearophile

Kai Meyer <kai unixlords.com> writes:

On 08/02/2011 06:03 AM, Andrej Mitrovic wrote:
 import std.algorithm;

 void main()
 {
      auto x = min([1, 2, 3]);  // x would be 1
 }

 min() isn't equipped to do this on a single range. What can I use
 instead? I haven't had my coffee yet. :)

Looking at std.algorithm, I think what you really want is minCount:
http://www.d-programming-language.org/phobos/std_algorithm.html#minCount

bearophile <bearophileHUGS lycos.com> writes:

Kai Meyer:

 Looking at std.algorithm, I think what you really want is minCount:
 http://www.d-programming-language.org/phobos/std_algorithm.html#minCount

It's a bad design.

Bye,
bearophile

Kai Meyer <kai unixlords.com> writes:

On 08/04/2011 07:54 PM, bearophile wrote:
 Kai Meyer:

 Looking at std.algorithm, I think what you really want is minCount:
 http://www.d-programming-language.org/phobos/std_algorithm.html#minCount

 It's a bad design.

 Bye,
 bearophile

minCount is, or the usage of minCount in his particular problem?

bearophile <bearophileHUGS lycos.com> writes:

Kai Meyer:

 minCount is, or the usage of minCount in his particular problem?

Your usage seems OK. What I meant is that I don't like the design of
min/max/minCount. I have expressed my ideas here:
http://d.puremagic.com/issues/show_bug.cgi?id=4705
I'd like to know what Andrei thinks about it. Then, I am probably able to
implement those ideas myself.

Bye,
bearophile