www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - std.algorithm.cmp doesn't seem to support numeric types?

reply DNoob <operator97 gmail.com> writes:
I'm just learning D, so it's very possible that I'm missing a 
more appropriate function that exists elsewhere, but basically I 
found today that while I could easily write a cmp function that 
worked for everything, the one in std.algorithm doesn't seem to:

import std.stdio, std.algorithm;

pure nothrow  nogc  system byte my_cmp(T1, T2)(T1 a, T2 b)
{
     if (a < b)
     {
         return -1;
     }
     else if (a == b)
     {
         return 0;
     }
     else
     {
         return 1;
     }
}

void main()
{
   //template error on the next line if uncommented
   //writeln(cmp(1, 2));

   writeln(my_cmp(1, 2));
}

Any advice would be much appreciated!
Oct 16 2019
next sibling parent Paul Backus <snarwin gmail.com> writes:
On Wednesday, 16 October 2019 at 19:25:18 UTC, DNoob wrote:

 I'm just learning D, so it's very possible that I'm missing a 
 more appropriate function that exists elsewhere, but basically 
 I found today that while I could easily write a cmp function 
 that worked for everything, the one in std.algorithm doesn't 
 seem to:


std.algorithm.cmp compares two ranges, not two values. From the 
documentation:


 Performs a lexicographical comparison on two input ranges. 
 Iterating r1 and
 r2 in lockstep, cmp compares each element e1 of r1 with the 
 corresponding
 element e2 in r2. If one of the ranges has been finished, cmp 
 returns a
 negative value if r1 has fewer elements than r2, a positive 
 value if r1 has
 more elements than r2, and 0 if the ranges have the same number 
 of elements.


Source: 
https://dlang.org/phobos/std_algorithm_comparison.html#.cmp
Oct 16 2019
prev sibling parent reply Adam D. Ruppe <destructionator gmail.com> writes:
On Wednesday, 16 October 2019 at 19:25:18 UTC, DNoob wrote:

 I'm just learning D, so it's very possible that I'm missing a 
 more appropriate function that exists elsewhere, but basically 
 I found today that while I could easily write a cmp function 
 that worked for everything, the one in std.algorithm doesn't 
 seem to:


Yeah, it is primarily for strings, to be used as a sorting 
predicate.

For numbers, simple less than (for bool comparisons) or 
subtraction operators (for the integer ones) will do it.

Notice that the docs say "a negative value" rather than -1 
specifically. That's because the implementation for integers an 
be as simple as

return a - b; // if b > a, you get a negative value

and thus doesn't really need a specific function.
Oct 16 2019
parent reply Dennis <dkorpel gmail.com> writes:
On Wednesday, 16 October 2019 at 20:07:10 UTC, Adam D. Ruppe 
wrote:

 Notice that the docs say "a negative value" rather than -1 
 specifically. That's because the implementation for integers an 
 be as simple as

 return a - b; // if b > a, you get a negative value


Except that e.g. -2 - int.max underflows to int.max suggesting 
that int.max < -2.
Oct 16 2019
next sibling parent =?UTF-8?Q?Ali_=c3=87ehreli?= <acehreli yahoo.com> writes:
On 10/16/2019 02:34 PM, Dennis wrote:


 Except that e.g. -2 - int.max underflows to int.max suggesting that
 int.max < -2.


Timon had corrected me on that point a while back, so I had added the 
following warning at

 
http://ddili.org/ders/d.en/operator_overloading.html#ix_operator_overloading.opCmp

"Warning: Using subtraction for the implementation of opCmp is a bug if 
valid values of a member can cause overflow."

However, note how I say "overflow" instead of "underflow". The reason 
is, underflow is a different concept that applies only to floating point 
values[1]. Apparently, "integer underflow" has become common usage as 
well[2].

Ali

[1] https://en.wikipedia.org/wiki/Arithmetic_underflow

[2] https://en.wikipedia.org/wiki/Integer_overflow
Oct 16 2019
prev sibling parent Adam D. Ruppe <destructionator gmail.com> writes:
On Wednesday, 16 October 2019 at 21:34:35 UTC, Dennis wrote:

 Except that e.g. -2 - int.max underflows to int.max suggesting 
 that int.max < -2.


Eh yeah, it would be something you should check in opCmp.

But this is how it is actually implemented on the processor 
itself, just the cpu happens to also check the overflow for you 
automatically...
Oct 16 2019