#version 420
#extension GL_ARB_texture_gather : enable
// shader bd8bba59e2149449
// Monochromia
// Simple Levels + Curves + LumaSharpening
// Credits to Jamie, NAVras for LumaSharpening; CeeJay.dk for Levels, Curves
// original shader dumped using cemu 1.11.c, BotW 1.3.1

// Levels controls
const int BlackPoint = 3; //[0, 255]  The black point is the new black - literally. Everything darker than this will become completely black
const int WhitePoint = 252; //[0, 255]  The new white point. Everything brighter than this becomes completely white

// Curves controls
const float Contrast = 0.25; //[-1.0, 1.0]  The amount of contrast you want

//LumaShapening
#define sharp_strength 2.5     //[0.10 to 3.00] Default 0.65 , Strength of the sharpening.
#define sharp_clamp    0.085    //[0.000 to 1.000] Default 0.035 , Limits maximum amount of sharpening a pixel recieves.
//Advanced sharpening settings
#define offset_bias 1.0 //[0.0 to 6.0] Offset bias adjusts the radius of the sampling pattern.

//Others
const float bloomFactor = 0.9;  //Default is 1.0

uniform ivec4 uf_remappedPS[1];
layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf46ac800 res 320x180x1 dim 1 tm: 4 format 0816 compSel: 0 1 2 5 mipView: 0x0 (num 0x5) sliceView: 0x0 (num 0x1) Sampler0 ClampX/Y/Z: 2 2 2 border: 1
layout(binding = 1) uniform sampler2D textureUnitPS1;// Tex1 addr 0xf5c7b800 res 1280x720x1 dim 1 tm: 4 format 0816 compSel: 0 1 2 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1
layout(location = 0) in vec4 passParameterSem0;
layout(location = 0) out vec4 passPixelColor0;
uniform vec2 uf_fragCoordScale;
int clampFI32(int v)
{
if( v == 0x7FFFFFFF )
	return floatBitsToInt(1.0);
else if( v == 0xFFFFFFFF )
	return floatBitsToInt(0.0);
return floatBitsToInt(clamp(intBitsToFloat(v), 0.0, 1.0));
}
float mul_nonIEEE(float a, float b){ if( a == 0.0 || b == 0.0 ) return 0.0; return a*b; }

/**
* Levels version 1.2
* by Christian Cann Schuldt Jensen ~ CeeJay.dk
*
* Allows you to set a new black and a white level.
* This increases contrast, but clips any colors outside the new range to either black or white
* and so some details in the shadows or highlights can be lost.
*
* The shader is very useful for expanding the 16-235 TV range to 0-255 PC range.
* You might need it if you're playing a game meant to display on a TV with an emulator that does not do this.
* But it's also a quick and easy way to uniformly increase the contrast of an image.
*
* -- Version 1.0 --
* First release
* -- Version 1.1 --
* Optimized to only use 1 instruction (down from 2 - a 100% performance increase :) )
* -- Version 1.2 --
* Added the ability to highlight clipping regions of the image with #define HighlightClipping 1
*/
vec3 LevelsPass(vec3 inputColor) {
	float black_point_float = BlackPoint / 255.0;
	float white_point_float = WhitePoint == BlackPoint ? (255.0 / 0.00025) : (255.0 / (WhitePoint - BlackPoint)); // Avoid division by zero if the white and black point are the same

	vec3 color = inputColor;
	color = color * white_point_float - (black_point_float *  white_point_float);

	return color;
}

/**
* Curves
* by Christian Cann Schuldt Jensen ~ CeeJay.dk
*
* Curves, uses S-curves to increase contrast, without clipping highlights and shadows.
*/
vec3 CurvesPass(vec3 inputColor) {
	vec3 colorInput = inputColor;
	float Contrast_blend = Contrast * 2.0; //I multiply by two to give it a strength closer to the other curves.
	vec3 x = colorInput.rgb; //if the curve should be applied to both Luma and Chroma
	x = x * (x * (1.5 - x) + 0.5); //horner form - fastest version
	vec3 color = x;  //if the curve should be applied to both Luma and Chroma
	colorInput.rgb = mix(colorInput.rgb, color, Contrast_blend); //Blend by Contrast
	return colorInput;
}

//LumaSharpening

#define px (1.0/1280.0*uf_fragCoordScale.x)
#define py (1.0/720.0*uf_fragCoordScale.y)
#define CoefLuma vec3(0.2126, 0.7152, 0.0722)  

float lumaSharpening(sampler2D tex, vec2 pos){
	vec4 colorInput = texture(tex, pos);
  	
	vec3 ori = colorInput.rgb;

	// -- Combining the strength and luma multipliers --
	vec3 sharp_strength_luma = (CoefLuma * sharp_strength); 
	
	// -- Gaussian filter --
	//   [ .25, .50, .25]     [ 1 , 2 , 1 ]
	//   [ .50,   1, .50]  =  [ 2 , 4 , 2 ]
 	//   [ .25, .50, .25]     [ 1 , 2 , 1 ]


    //float px = 1.0/tex_size[0];
	//float py = 1.0/tex_size[1];

	vec3 blur_ori = texture(tex, pos + vec2(px,-py) * 0.5 * offset_bias).rgb; // South East
	blur_ori += texture(tex, pos + vec2(-px,-py) * 0.5 * offset_bias).rgb;  // South West
	blur_ori += texture(tex, pos + vec2(px,py) * 0.5 * offset_bias).rgb; // North East
	blur_ori += texture(tex, pos + vec2(-px,py) * 0.5 * offset_bias).rgb; // North West

	blur_ori *= 0.25;  // ( /= 4) Divide by the number of texture fetches

	// -- Calculate the sharpening --
	vec3 sharp = ori - blur_ori;  //Subtracting the blurred image from the original image

	// -- Adjust strength of the sharpening and clamp it--
	vec4 sharp_strength_luma_clamp = vec4(sharp_strength_luma * (0.5 / sharp_clamp),0.5); //Roll part of the clamp into the dot

	float sharp_luma = clamp((dot(vec4(sharp,1.0), sharp_strength_luma_clamp)), 0.0,1.0 ); //Calculate the luma, adjust the strength, scale up and clamp
	sharp_luma = (sharp_clamp * 2.0) * sharp_luma - sharp_clamp; //scale down


	// -- Combining the values to get the final sharpened pixel	--

	//colorInput.rgb = colorInput.rgb + sharp_luma;    // Add the sharpening to the input color.
	return sharp_luma;  
}

void main()
{
vec4 R0f = vec4(0.0);
vec4 R1f = vec4(0.0);
vec4 R123f = vec4(0.0);
vec4 R125f = vec4(0.0);
vec4 R126f = vec4(0.0);
vec4 R127f = vec4(0.0);
float backupReg0f, backupReg1f, backupReg2f, backupReg3f, backupReg4f;
vec4 PV0f = vec4(0.0), PV1f = vec4(0.0);
float PS0f = 0.0, PS1f = 0.0;
vec4 tempf = vec4(0.0);
float tempResultf;
int tempResulti;
ivec4 ARi = ivec4(0);
bool predResult = true;
vec3 cubeMapSTM;
int cubeMapFaceId;
R0f = passParameterSem0;
R1f.xyz = (texture(textureUnitPS0, R0f.xy).xyz);
R1f.xyz *= bloomFactor;
R0f.xyz = texture(textureUnitPS1, passParameterSem0.xy).xyz;
float smask = lumaSharpening(textureUnitPS1, passParameterSem0.xy);
R0f.xyz += vec3(smask);
// 0
R126f.x = R1f.x + R0f.x;
PV0f.x = R126f.x;
R127f.y = R1f.y + R0f.y;
PV0f.y = R127f.y;
R126f.z = R1f.z + R0f.z;
PV0f.z = R126f.z;
R125f.w = 1.0;
// 1
tempf.x = dot(vec4(PV0f.x,PV0f.y,PV0f.z,-0.0),vec4(intBitsToFloat(0x3e99096c),intBitsToFloat(0x3f162b6b),intBitsToFloat(0x3dea4a8c),0.0));
PV1f.x = tempf.x;
PV1f.y = tempf.x;
PV1f.z = tempf.x;
PV1f.w = tempf.x;
// 2
R127f.x = -(R127f.y) * intBitsToFloat(0x3fb8aa3b);
PV0f.y = -(PV1f.x) * intBitsToFloat(0x3fb8aa3b);
R127f.z = -(R126f.x) * intBitsToFloat(0x3fb8aa3b);
R127f.w = -(R126f.z) * intBitsToFloat(0x3fb8aa3b);
R126f.w = 1.0 / PV1f.x;
PS0f = R126f.w;
// 3
PS1f = exp2(PV0f.y);
// 4
PV0f.x = -(PS1f) + 1.0;
PS0f = exp2(R127f.x);
// 5
R127f.x = -(PS0f) + 1.0;
R126f.y = mul_nonIEEE(PV0f.x, PV0f.x);
PV1f.z = PV0f.x * R126f.w;
PS1f = exp2(R127f.w);
// 6
backupReg0f = R126f.x;
backupReg1f = R127f.z;
R126f.x = mul_nonIEEE(backupReg0f, PV1f.z);
PV0f.y = -(PS1f) + 1.0;
R127f.z = mul_nonIEEE(R126f.z, PV1f.z);
PV0f.z = R127f.z;
R127f.w = mul_nonIEEE(R127f.y, PV1f.z);
PV0f.w = R127f.w;
PS0f = exp2(backupReg1f);
// 7
PV1f.x = R127f.x + -(PV0f.w);
PV1f.y = PV0f.y + -(PV0f.z);
PV1f.w = -(PS0f) + 1.0;
// 8
backupReg0f = R127f.z;
R127f.x = (mul_nonIEEE(PV1f.x,R126f.y) + R127f.w);
R127f.x = clamp(R127f.x, 0.0, 1.0);
PV0f.x = R127f.x;
PV0f.y = PV1f.w + -(R126f.x);
R127f.z = (mul_nonIEEE(PV1f.y,R126f.y) + backupReg0f);
R127f.z = clamp(R127f.z, 0.0, 1.0);
PV0f.z = R127f.z;
// 9
backupReg0f = R126f.x;
R126f.x = (mul_nonIEEE(PV0f.y,R126f.y) + backupReg0f);
R126f.x = clamp(R126f.x, 0.0, 1.0);
PV1f.x = R126f.x;
R126f.y = max(PV0f.x, PV0f.z);
PV1f.w = min(PV0f.x, PV0f.z);
// 10
tempf.x = dot(vec4(PV1f.x,R127f.x,R127f.z,R125f.w),vec4(intBitsToFloat(0x3f2aaaab),intBitsToFloat(0x3f2aaaab),intBitsToFloat(0x3f2aaaab),-(1.0)));
PV0f.x = tempf.x;
PV0f.y = tempf.x;
PV0f.z = tempf.x;
PV0f.w = tempf.x;
R126f.z = min(PV1f.x, PV1f.w);
PS0f = R126f.z;
// 11
backupReg0f = R127f.x;
backupReg1f = R127f.z;
R127f.x = max(R126f.x, R126f.y);
PV1f.x = R127f.x;
R123f.y = (mul_nonIEEE(-(PV0f.x),PV0f.x) + 1.0);
PV1f.y = R123f.y;
R127f.z = backupReg0f + -(PS0f);
R125f.w = R126f.x + -(PS0f);
R126f.y = backupReg1f + -(PS0f);
PS1f = R126f.y;
// 12
R126f.x = (mul_nonIEEE(PV1f.y,intBitsToFloat(uf_remappedPS[0].y)) + intBitsToFloat(uf_remappedPS[0].x));
PV0f.x = R126f.x;
PV0f.y = -(R126f.z) + PV1f.x;
// 13
R123f.w = (mul_nonIEEE(-(PV0f.x),PV0f.y) + R127f.x);
PV1f.w = R123f.w;
// 14
R0f.x = (mul_nonIEEE(R126f.x,R125f.w) + PV1f.w);
R0f.y = (mul_nonIEEE(R126f.x,R127f.z) + PV1f.w);
R0f.z = (mul_nonIEEE(R126f.x,R126f.y) + PV1f.w);
// export
passPixelColor0 = vec4(R0f.x, R0f.y, R0f.z, R0f.w);
vec3 color = (passPixelColor0.xyz);
color = LevelsPass(color);
color = CurvesPass(color);
passPixelColor0 = vec4(color, R0f.w);
}