Re: arrays in srucs

[Andy Balba via Digitalmars-d-learn]

On Friday, 31 July 2020 at 17:26:17 UTC, Paul Backus wrote:

On Friday, 31 July 2020 at 17:02:46 UTC, Andy Balba wrote:
The above code, compiles and runs ok  .. but sometimes I get 
run runtime errors using the same paradym, which disappear 
when I substitute (img.p)[i]


Any explanation for this ?


Can you show an example where it doesn't work?


Copy the code block below into your editor then
Look down at second statement marked "// printf "

when I use () it works fine, but without () it gives run-time 
errors


import image_BMP ;
import colorMaps : hsv_colormap, jet_colormap, prism_colormap, 
vga_colormap ;


import std.math ;
import std.random ;
import std.stdio ;

int main()
{  int cmapM= 2, cmapJ= 3; // cmap:  0=hsv_colormap, 
1=jet_colormap, 2=prism_colormap, 3= vga_colormap
   // cmapM, cmapJ = colormap for 
Mendelbrot, Julia
   const uint W = 1200, H =  800, Bp=3, row_size= W*Bp ; // 
fractal width, height, Bytes per pixel

   IFImage img ;  // fractal image
   img.w = W; img.h = H; img.c = ColFmt.RGB; img.pixels = new 
ubyte[H*W*3];   // c= color format: 1=Gray 2=Gray+Alpha, 3=RGB  
4=RGB +Alpha

  // 1200 * 800 *3 = 2.8 MB per image 4 images= 11.2 MB
  int max_iterations= 1000;
  for (int Y=0; Y< H; ++Y)
  {  for (int X=0; X< W; ++X)
 {
double cr= 1.5 *(2.0*X /W-1.0) -0.5,  ci= (2.0*Y /H 
-1.0); // Mendelbrot Fractal constant= (cr, ci)

double nr=0, ni= 0,  pr=0, pi = 0;
for (int i= 0; i< max_iterations; i++)
{  pr= nr; pi = ni;  nr= pr*pr -pi*pi +cr;  ni= 
2*pr*pi +ci;  // z(real, imag)= (pr,pi),  z^2+c (real, imag)= 
(nr,ni)

   if (((nr*nr) +(ni*ni)) > 4)
   {  if (max_iterations != i)
  {  double z= sqrt(nr*nr +ni*ni);
 int cid = cast(int) (1000.0*log2(1.75 +i 
-log2(log2(z)))/ log2(max_iterations));
 ubyte[3][4] c= [ hsv_colormap[cid], 
jet_colormap[cid], prism_colormap[cid], vga_colormap[cid] ];

//printf("Y%i X%i :\n", Y, X);  stdout.flush();
 foreach ( p; 0..3 )
 {
//printf(">%i %i %i", p, cmapM, Y*row_size +X*Bp +p );   
stdout.flush();
 img.pixels [ Y*row_size +X*Bp +p ] = 
c[p][cmapM] ;

 }
  }
  break;
  }  }  }  }
  write_image ( "M_fractal.bmp", cast(long) img.w, cast(long) 
img.h, img.pixels, cast(long)img.c ) ;

Re: arrays in srucs

[Paul Backus via Digitalmars-d-learn]

On Friday, 31 July 2020 at 17:02:46 UTC, Andy Balba wrote:
The above code, compiles and runs ok  .. but sometimes I get 
run runtime errors using the same paradym, which disappear when 
I substitute (img.p)[i]


Any explanation for this ?


Can you show an example where it doesn't work?

arrays in srucs

[Andy Balba via Digitalmars-d-learn]

I frequently use this paradigm for my image processing apps :

import std.stdio ;
struct S { int a, b; ubyte[] p;
}
void main()
{ S img= S(11,22) ;  img.p = new ubyte[5];

foreach (i; 0 .. 5) {
  img.p[i]=cast(ubyte)(10+i) ; printf(" sa %d sb %d : %d\n", 
img.a, img.b, img.p[i] );

}
}

The above code, compiles and runs ok  .. but sometimes I get run 
runtime errors using the same paradym, which disappear when I 
substitute (img.p)[i]


Any explanation for this ?