My CopyMe program for Facebook Graffiti app

[a111087]

Hey,
If you are not familiar with Graffiti app on Facebook - it is a small app that allows you to draw pictures for your friends on Facebook. Pictures are posted to their wall and they can then view how you drew them, step by step.

I know it might sound silly, but I liked it. Me and one of my friends liked using it to draw all kinds of things, whenever one of us had birthday, or any other occasion. It sure was fun using it. It felt like it was better than a present you buy at the store because you put your time into it and you just know that the other person will appreciate it. (I know it silly, but keep reading. )

So, I did a lot of drawings and each one was taking more time than before. They were also more complicated and more beautiful, but I was spending up to 4 hours on one drawing. I though, I would try something new and write a program that would draw pictures for me.

Here is how it works:

1. You find a picture on the web
2. Convert it to only black and white pixels (current limitation of the program)
3. Tell the program to start drawing. (Program will look at the image and draw it in the app pixel by pixel. It might take 2-3 hours to finish. You will not be able to use your computer during that time because mouse pointer will be used to draw and click in the app.)
4. When draft is ready, it is up to you to add colors or some other interesting things.

So, what do you think?
PM me if you want to receive a copy of the program and try it out.

Gui, Add, GroupBox, x12 y10 w170 h50 , Current Mouse Pos
Gui, Add, Text, x22 y30 w50 h20 , vMouseX
Gui, Add, Text, x72 y30 w50 h20 , vMouseY
Gui, Add, Text, x122 y30 w50 h20 , vMouseColor

Gui, Add, GroupBox, x12 y70 w170 h70 , Input Mouse Coord
Gui, Add, GroupBox, x12 y170 w180 h90 , Load Mouse Pos
Gui, Add, Text, x22 y90 w50 h20 , From
Gui, Add, Text, x22 y110 w50 h20 , To
Gui, Add, Edit, x72 y90 w60 h20 , vInputBegX
Gui, Add, Edit, x132 y90 w50 h20 , vInputBegY
Gui, Add, Edit, x72 y110 w60 h20 , vInputEndX
Gui, Add, Edit, x132 y110 w50 h20 , vInputEndY
Gui, Add, Text, x22 y190 w50 h20 , From
Gui, Add, Text, x22 y210 w50 h20 , To
Gui, Add, Edit, x72 y190 w60 h20 , vInputBegX
Gui, Add, Edit, x132 y190 w50 h20 , vInputBegY
Gui, Add, Edit, x72 y210 w60 h20 , vInputEndX
Gui, Add, Edit, x132 y210 w50 h20 , vInputEndY
Gui, Add, Button, x22 y140 w50 h20 , Start
Gui, Add, Button, x82 y140 w50 h20 , Copy


Gui, Show, x524 y513 h267 w206, CopyMe
Return


StartScan:
CoordMode, Mouse, Screen
CoordMode, Pixel, Screen
MouseGetPos, xpos, ypos 
GuiControl,, Static1, %xpos%
GuiControl,, Static2, %ypos%
PixelGetColor, color, xpos-1 , ypos-1
GuiControl,, Static3, %color%
;Gui, Color, %color%
return

ButtonStart:
SetTimer, StartScan, 550
return

#z::
GuiControl,, Edit1, %xpos%
GuiControl,, Edit2, %ypos%
return

#x::
GuiControl,, Edit3, %xpos%
GuiControl,, Edit4, %ypos%
return

#c::
GuiControl,, Edit5, %xpos%
GuiControl,, Edit6, %ypos%
return

#v::
GuiControl,, Edit7, %xpos%
GuiControl,, Edit8, %ypos%
return


ButtonCopy:
SetTimer, StartScan, Off
ControlGetText, inpx, Edit1
ControlGetText, inpy, Edit2
ControlGetText, inpxx, Edit3
ControlGetText, inpyy, Edit4

ControlGetText, loadx, Edit5
ControlGetText, loady, Edit6
ControlGetText, loadxx, Edit7
ControlGetText, loadyy, Edit8

count1 := inpxx-inpx
count2 := inpyy-inpy

y = 1

Loop, %count2%
{
x = 1 
 Loop, %count1%
  {
	CoordMode, Mouse, Screen
	CoordMode, Pixel, Screen

    PixelGetColor, color, inpx + x , inpy + y
	GuiControl,, Static3, %color%    
	if (color = color2)
	    {
		MouseClick, left, loadx + x, loady + y,,0
	    }
    x:= x+1
  }
y:= y+1
}
return

#f::
CoordMode, Mouse, Screen
CoordMode, Pixel, Screen
PixelGetColor, color2, xpos-1 , ypos-1
return

GuiClose:
ExitApp

Instructions:
Ok, it's not very user friendly so far, so I will have to explain a bit.
1. When you find an image that you want to draw, covert it to only pure black and white pixels (can be done in photoshop)
2. Run program and arrange 3 windows on your display: one for facebook, one that displays edited image, and one with a program.
3. In the program click "Start". It will start tracking mouse pointer and will tell you the color of the pixel next to the mouse.
4. If you want a program to draw black pixels, then set up Graffiti app appropriately (smallest size brush, color black, opacity to max)
5. find pure black pixels in your picture or anywhere on the screen with your mouse (remember, program tells you the code for the color next to the mouse pointer) and press Windows + F (this will set a color for which program will look in the picture)
6. It is time to let program know location of the picture and graffiti app on the screen.
Locate, with your mouse pointer top left corner of the image you are going to copy and press Windows + Z
7. Locate, with your mouse pointer lower right corner of the image you are going to copy and press Windows + X
8. Locate, with your mouse pointer top left corner of the Graffiti drawing surface and press Windows + C
(there is also Windows + V to select lower right corner of graffiti app, but it is useless so far...)
9. Click copy and enjoy

[micropage7]

looks nice

[mlee49]

Your computer might take 2-3 hours to finish, and you might not be able to use it during that time.

Try me... This could be a new benchmark

[FordGT90Concept]

I think you should not use the mouse and you could really cut down on the time if it were multithreaded. XD

What language is it coded in? If you're using .NET, create a Bitmap and use the various tools to "paint" on it. You can easily target a single cell in the image and change the color.

Except for the "add color" part, I don't see what your program does. It looks like it changed black to gray.

[a111087]

Originally Posted by mlee49

Your computer might take 2-3 hours to finish, and you might not be able to use it during that time.

Try me... This could be a new benchmark

lol, i wish it was doing this faster, but it really doesn't depend on processor speed. it is just a limitation of the script language i used.

Originally Posted by FordGT90Concept

I think you should not use the mouse and you could really cut down on the time if it were multithreaded. XD

What language is it coded in? If you're using .NET, create a Bitmap and use the various tools to "paint" on it. You can easily target a single cell in the image and change the color.

Except for the "add color" part, I don't see what your program does. It looks like it changed black to gray.

I wrote this in autohotkey script language (it is very easy and can compile programs into EXE's). It is something I never even heard about until I started doing this whole thing. I first tried C++, but since I'm not an expert at it and googling for info didn't yield too much, I decided to try autohotkey and see how it works.

Heart of the program is a really small nested loop and while there are no explicit delays written into the code, autohotkey decides that you could use some delay in your program.

The reason why draft looks grey, instead of pure black is because Graffiti app can't draw single pure black pixel from just one click of a mouse (which is what my program does, it reads placement of black or white pixels, then clicks in an app to draw that image, pixel by pixel)

[de.das.dude]

can i haz it?

[FordGT90Concept]

Originally Posted by a111087

I wrote this in autohotkey script language (it is very easy and can compile programs into EXE's).

Originally Posted by a111087

Heart of the program is a really small nested loop and while there are no explicit delays written into the code, autohotkey decides that you could use some delay in your program.

That's so it doesn't overflow the user32 stack. You know how you hold down three keys on, for example, the keypad of a keyboard, it beeps at you? That's effectively the same thing it is trying to avoid. By creating a virtual image in the RAM and painting on it, you avoid that bottleneck.

Originally Posted by a111087

The reason why draft looks grey, instead of pure black is because Graffiti app can't draw single pure black pixel from just one click of a mouse (which is what my program does, it reads placement of black or white pixels, then clicks in an app to draw that image, pixel by pixel)

So all this autohotkey script is doing is copying from an image into a program (browser)? There's no way to upload an image?

[a111087]

Originally Posted by FordGT90Concept

So all this autohotkey script is doing is copying from an image into a program (browser)? There's no way to upload an image?

Yes, to put it simply, it just copies/steal an image to your browser. That's why I called it CopyMe.

There is of course a way to upload an image straight to a wall, but not to a graffiti app.

[a111087]

just added full code to the first post
and here is a link for download (click free at the bottom):
http://www.megashare.com/3666639

[a111087]

Originally Posted by FordGT90Concept

That's so it doesn't overflow the user32 stack. You know how you hold down three keys on, for example, the keypad of a keyboard, it beeps at you? That's effectively the same thing it is trying to avoid. By creating a virtual image in the RAM and painting on it, you avoid that bottleneck.

But how do you put that virtual image into graffiti app or anywhere else?

[FordGT90Concept]

Good question and I think the answer would be very complex (memory hacking). Is it not the point of a Graffitti app to paint, not copy?

Bypassing autohotkey would provide a rather large performance boost though because you could adjust the delay yourself (if it is even necessary).

[a111087]

Originally Posted by FordGT90Concept

Is it not the point of a Graffitti app to paint, not copy?

Haha, it sure is the point, but CopyMe doesn't do everything for you. It simply helps you get started on the image, then you add color and whatever else you have in mind. Just like in the picture that I posted.
So, it is cheating, but just a little bit.

[a111087]

lol, I just corrected a mistake in the instructions.
You need to use Windows key, not Control key.