Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.c++ > #121004
| From | Michael S <already5chosen@yahoo.com> |
|---|---|
| Newsgroups | comp.lang.c++ |
| Subject | Re: Interesting results |
| Date | 2024-11-25 01:04 +0200 |
| Organization | A noiseless patient Spider |
| Message-ID | <20241125010452.00006c08@yahoo.com> (permalink) |
| References | <vhvk6i$2a5nc$1@raubtier-asyl.eternal-september.org> |
On Sun, 24 Nov 2024 17:29:00 +0100
Bonita Montero <Bonita.Montero@gmail.com> wrote:
> This code combines finite values with NaN-values and Inf-values and
> compares them with all basic comparison operators.
>
> #include <iostream>
> #include <limits>
>
> using namespace std;
>
> int main()
> {
> constexpr double
> NaN = numeric_limits<double>::quiet_NaN(),
> Inf = numeric_limits<double>::infinity(),
> Finite = 1.0;
> static struct value_t
> {
> char const *what;
> double value;
> } const values[] =
> {
> { "Finite", Finite },
> { "Inf", Inf },
> { "NaN", NaN }
> };
> static struct op_t
> {
> char const *what;
> bool (*op)( double, double );
> } const ops[] =
> {
> { "<", +[]( double a, double b ) { return a < b; } },
> { "<=", +[]( double a, double b ) { return a <= b; }
> }, { "==", +[]( double a, double b ) { return a == b; } },
> { ">=", +[]( double a, double b ) { return a >= b; }
> }, { ">", +[]( double a, double b ) { return a > b; } }
> };
> for( value_t const &vA : values )
> {
> cout << vA.what << endl;
> for( value_t const &vB : values )
> {
> if( &vA == &vB && &vA == values )
> continue;
> cout << "\t" << vB.what << endl;
> for( op_t const &op : ops )
> {
> cout << "\t\t" << op.what << ": ";
> cout << (ops->op( vA.value, vB.value
> ) ? "true" : "false") << endl; }
> }
> }
> }
>
> This are the results:
>
> Finite
> Inf
> <: true
> <=: true
> ==: true
> >=: true
> >: true
> NaN
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> Inf
> Finite
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> Inf
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> NaN
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> NaN
> Finite
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> Inf
> <: false
> <=: false
> ==: false
> >=: false
> >: false
> NaN
> <: false
> <=: false
> ==: false
> >=: false
> >: false
>
> Interesting results !
>
Nothing of interest here, except for another proof that combination of
borderline-sane programming language with insane programming style
causes bugs.
Here is the same test written in more sane language and style. Results
are 100% as expected:
#include <stdio.h>
#include <math.h>
int main()
{
static const double vals[3] = { 1.0, INFINITY, NAN };
_Bool results[3][3][5];
for (int i = 0; i < 3; ++i) {
for (int j = 0; j < 3; ++j) {
results[i][j][0] = vals[i] < vals[j];
results[i][j][1] = vals[i] <= vals[j];
results[i][j][2] = vals[i] == vals[j];
results[i][j][3] = vals[i] >= vals[j];
results[i][j][4] = vals[i] > vals[j];
}
}
static const char* ops[5] = { "<", "<=", "==", ">=", ">" };
for (int i = 0; i < 3; ++i) {
for (int k = 0; k < 5; ++k) {
for (int j = 0; j < 3; ++j)
printf(" %3g %2s %3g : %-5s ;",
vals[i], ops[k], vals[j],
results[i][j][k] ? "true" : "false");
printf("\n");
}
}
return 0;
}
Results:
1 < 1 : false ; 1 < inf : true ; 1 < nan : false ;
1 <= 1 : true ; 1 <= inf : true ; 1 <= nan : false ;
1 == 1 : true ; 1 == inf : false ; 1 == nan : false ;
1 >= 1 : true ; 1 >= inf : false ; 1 >= nan : false ;
1 > 1 : false ; 1 > inf : false ; 1 > nan : false ;
inf < 1 : false ; inf < inf : false ; inf < nan : false ;
inf <= 1 : false ; inf <= inf : true ; inf <= nan : false ;
inf == 1 : false ; inf == inf : true ; inf == nan : false ;
inf >= 1 : true ; inf >= inf : true ; inf >= nan : false ;
inf > 1 : true ; inf > inf : false ; inf > nan : false ;
nan < 1 : false ; nan < inf : false ; nan < nan : false ;
nan <= 1 : false ; nan <= inf : false ; nan <= nan : false ;
nan == 1 : false ; nan == inf : false ; nan == nan : false ;
nan >= 1 : false ; nan >= inf : false ; nan >= nan : false ;
nan > 1 : false ; nan > inf : false ; nan > nan : false ;
Back to comp.lang.c++ | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
Interesting results Bonita Montero <Bonita.Montero@gmail.com> - 2024-11-24 17:29 +0100
Re: Interesting results Paavo Helde <eesnimi@osa.pri.ee> - 2024-11-25 00:25 +0200
Re: Interesting results Bonita Montero <Bonita.Montero@gmail.com> - 2024-11-25 06:32 +0100
Re: Interesting results Michael S <already5chosen@yahoo.com> - 2024-11-25 01:04 +0200
Re: Interesting results Bonita Montero <Bonita.Montero@gmail.com> - 2024-11-25 06:32 +0100
Re: Interesting results wij <wyniijj5@gmail.com> - 2024-11-26 03:32 +0800
Re: Interesting results wij <wyniijj5@gmail.com> - 2024-11-26 03:41 +0800
csiph-web