newbie effect question...

[drakaan]

Hi all, I'm trying (unsuccessfully, so far) to use codelab to create a plugin that splits an image in two. Just one horizontal split.

The code I have so far is:

void Render(Surface dst, Surface src, Rectangle rect)
{
   PdnRegion selectionRegion = EnvironmentParameters.GetSelection(src.Bounds);

   // Delete any of these lines you don't need
   Rectangle selection = this.EnvironmentParameters.GetSelection(src.Bounds).GetBoundsInt();
   int offsetx = 0;
   int V1 = 10;
   int H1 = 10;
   int spacing = 10;

               for (int y = rect.Bottom; y > rect.Top; --y)
               {
                   for (int x = rect.Right; x > rect.Left; --x)
                   {
                       if (x > rect.Left + V1)
                       {
                           offsetx = spacing * 2;
                       } else {
               	    offsetx = 0;
                       }

                       if (x < rect.Right - offsetx)
                       {
                           dst[x + offsetx, y] = src[x, y];
                       }
                   }
               }
}

I had some errors with x being beyond the bounds of the selection which seem handled, but there's no obvious movement of the ol' pixels.

Can somebody hit me with a clue-by-four?

[BoltBait]

Show me an example of a "before" and "after" image.

The first issue I see is in the following line of code:

dst[x + offsetx, y] = src[x, y];

it should look more like this:

dst[x, y] = src[ ... ];

Remember, in your loops, you should be looping through the destination canvas NOT the source canvas.

[drakaan]

Sure...it's a simple idea...

Before:

After:

Iterating over the destination canvas sounded counterintuitive to me when I read it, but after thinking about it, I believe I grok it in fullness.

I should be using the current X and Y for the destination surface and for that point, set the pixel's values according to my desired value based on calculations of the source surface's geometry (instead of bass-ackwards, like I just attempted to do).

Does that mean that effects can never alter pixels outside of the selection?

It seems like an obvious question, but I can think of some situations where having the selection not be a mask for the effect would be handy.

[BoltBait]

Does that mean that effects can never alter pixels outside of the selection?

Effects are not allowed to modify pixels outside of the ROI* that has been passed to it.

You can always read pixels outside of your selection, you just can't modify them.

There is some discussion of this here: http://paintdotnet.12.forumer.com/viewtopic.php?t=5281

*ROI=Rectangle of Interest

[drakaan]

Okay...modified it a bit. Now I see copied pixels, but I guess I have to clear the source pixels for the locations where I moved them?

EDIT: Now it's clipping correctly, I just don't know how to make the right pixels transparent...

void Render(Surface dst, Surface src, Rectangle rect) 
{ 
   PdnRegion selectionRegion = EnvironmentParameters.GetSelection(src.Bounds); 

   // Delete any of these lines you don't need 
   Rectangle selection = this.EnvironmentParameters.GetSelection(src.Bounds).GetBoundsInt(); 
   int offsetx = 0; 
   int V1 = 10; 
   int H1 = 10; 
   int spacing = 10; 

               for (int y = rect.Top; y < rect.Bottom; ++y) 
               { 
                   for (int x = rect.Left; x < rect.Right; ++x) 
                   { 
                       if (x > rect.Left + V1) 
                       { 
                           offsetx = spacing; 
                       } else { 
                      offsetx = 0; 
                       } 

                       if (x > rect.Left + offsetx) 
                       { 
               	    if (x > rect.Left + (offsetx * 2)) {
                   		dst[x, y] = src[x - offsetx, y];
                           } else {

               ColorBgra CurrentPixel = src[x,y];
               // TODO: Add pixel processing code here
               // Access RGBA values this way, for example:
                CurrentPixel.R = 0;
                CurrentPixel.G = 0;
                CurrentPixel.B = 0;
                CurrentPixel.A = 0;
                           }
                       } 
                   } 
               } 
}

[drakaan]

Woohoo! Figured it out!

void Render(Surface dst, Surface src, Rectangle rect) 
{ 
   PdnRegion selectionRegion = EnvironmentParameters.GetSelection(src.Bounds); 

   // Delete any of these lines you don't need 
   Rectangle selection = this.EnvironmentParameters.GetSelection(src.Bounds).GetBoundsInt(); 
   int offsetx = 0; 
   int V1 = 10; 
   int H1 = 10; 
   int spacing = 10; 

               for (int y = rect.Top; y < rect.Bottom; ++y) 
               { 
                   for (int x = rect.Left; x < rect.Right; ++x) 
                   { 
                       if (x > rect.Left + V1) 
                       { 
                           offsetx = spacing; 
                       } else { 
                      offsetx = 0; 
                       } 

                       if (x > rect.Left + offsetx) 
                       { 
               	    if (x > rect.Left + (offsetx * 2)) {
                   		dst[x, y] = src[x - offsetx, y];
                           } else {
                           	ColorBgra dpix = src[x - offsetx, y];
                   		dpix.A = 0;
               		dst[x,y] = dpix;
                           }
                       } 
                   } 
               } 
} 

Learned a few things, too...

[BoltBait]

Cool.

Now, you need to read part 2 of my plugin tutorials so you can put some sliders on the UI:

http://paintdotnet.12.forumer.com/viewtopic.php?t=5308

[drakaan]

Cool.

Now, you need to read part 2 of my plugin tutorials so you can put some sliders on the UI:

http://paintdotnet.12.forumer.com/viewtopic.php?t=5308

I plan on more than that, eventually (draggable dividers on a thumbnail image)...I definitely need to read that article, though.

Final goal is a basic image splitter for cutting images up to use in web forms. You'll have a choice of one or two horizontal or vertical dividers and their placement on the image...maybe a "make symmetrical" button, too.