// Copyright (c) 2012- PPSSPP Project.

// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, version 2.0 or later versions.

// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License 2.0 for more details.

// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/

// Official git repository and contact information can be found at
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.

#include <stdio.h>
#include <locale.h>

#if defined(_WIN32) && defined(_DEBUG)
#include "Common/CommonWindows.h"
#endif

#include "base/stringutil.h"
#include "GPU/ge_constants.h"
#include "GPU/GPUState.h"
#include "Core/Config.h"

#include "GPU/Directx9/VertexShaderGeneratorDX9.h"
#include "GPU/Common/VertexDecoderCommon.h"
#include "GPU/Common/ShaderUniforms.h"

#undef WRITE

#define WRITE p+=sprintf

namespace DX9 {

static const char * const boneWeightAttrDecl[9] = {	
	"#ERROR#",
	"float  a_w1:TEXCOORD1;\n",
	"float2 a_w1:TEXCOORD1;\n",
	"float3 a_w1:TEXCOORD1;\n",
	"float4 a_w1:TEXCOORD1;\n",
	"float4 a_w1:TEXCOORD1;\n  float  a_w2:TEXCOORD2;\n",
	"float4 a_w1:TEXCOORD1;\n  float2 a_w2:TEXCOORD2;\n",
	"float4 a_w1:TEXCOORD1;\n  float3 a_w2:TEXCOORD2;\n",
	"float4 a_w1:TEXCOORD1;\n  float4 a_w2:TEXCOORD2;\n",
};

enum DoLightComputation {
	LIGHT_OFF,
	LIGHT_SHADE,
	LIGHT_FULL,
};

void GenerateVertexShaderHLSL(const VShaderID &id, char *buffer, ShaderLanguage lang) {
	char *p = buffer;
	const u32 vertType = gstate.vertType;

	bool isModeThrough = id.Bit(VS_BIT_IS_THROUGH);
	bool lmode = id.Bit(VS_BIT_LMODE);
	bool doTexture = id.Bit(VS_BIT_DO_TEXTURE);
	bool doTextureTransform = id.Bit(VS_BIT_DO_TEXTURE_TRANSFORM);

	GETexMapMode uvGenMode = static_cast<GETexMapMode>(id.Bits(VS_BIT_UVGEN_MODE, 2));

	// this is only valid for some settings of uvGenMode
	GETexProjMapMode uvProjMode = static_cast<GETexProjMapMode>(id.Bits(VS_BIT_UVPROJ_MODE, 2));
	bool doShadeMapping = uvGenMode == GE_TEXMAP_ENVIRONMENT_MAP;
	bool doFlatShading = id.Bit(VS_BIT_FLATSHADE);

	bool useHWTransform = id.Bit(VS_BIT_USE_HW_TRANSFORM);
	bool hasColor = id.Bit(VS_BIT_HAS_COLOR) || !useHWTransform;
	bool hasNormal = id.Bit(VS_BIT_HAS_NORMAL) && useHWTransform;
	bool hasTexcoord = id.Bit(VS_BIT_HAS_TEXCOORD) || !useHWTransform;
	bool enableFog = id.Bit(VS_BIT_ENABLE_FOG);
	bool throughmode = id.Bit(VS_BIT_IS_THROUGH);
	bool flipNormal = id.Bit(VS_BIT_NORM_REVERSE);
	int ls0 = id.Bits(VS_BIT_LS0, 2);
	int ls1 = id.Bits(VS_BIT_LS1, 2);
	bool enableBones = id.Bit(VS_BIT_ENABLE_BONES);
	bool enableLighting = id.Bit(VS_BIT_LIGHTING_ENABLE);
	int matUpdate = id.Bits(VS_BIT_MATERIAL_UPDATE, 3);

	bool doBezier = id.Bit(VS_BIT_BEZIER);
	bool doSpline = id.Bit(VS_BIT_SPLINE);
	bool hasColorTess = id.Bit(VS_BIT_HAS_COLOR_TESS);
	bool hasTexcoordTess = id.Bit(VS_BIT_HAS_TEXCOORD_TESS);
	bool flipNormalTess = id.Bit(VS_BIT_NORM_REVERSE_TESS);

	DoLightComputation doLight[4] = { LIGHT_OFF, LIGHT_OFF, LIGHT_OFF, LIGHT_OFF };
	if (useHWTransform) {
		int shadeLight0 = doShadeMapping ? ls0 : -1;
		int shadeLight1 = doShadeMapping ? ls1 : -1;
		for (int i = 0; i < 4; i++) {
			if (i == shadeLight0 || i == shadeLight1)
				doLight[i] = LIGHT_SHADE;
			if (id.Bit(VS_BIT_LIGHTING_ENABLE) && id.Bit(VS_BIT_LIGHT0_ENABLE + i))
				doLight[i] = LIGHT_FULL;
		}
	}

	int numBoneWeights = 0;
	int boneWeightScale = id.Bits(VS_BIT_WEIGHT_FMTSCALE, 2);
	if (enableBones) {
		numBoneWeights = 1 + id.Bits(VS_BIT_BONES, 3);
	}

	if (lang == HLSL_DX9) {
		WRITE(p, "#pragma warning( disable : 3571 )\n");
		if (isModeThrough) {
			WRITE(p, "float4x4 u_proj_through : register(c%i);\n", CONST_VS_PROJ_THROUGH);
		} else {
			WRITE(p, "float4x4 u_proj : register(c%i);\n", CONST_VS_PROJ);
			// Add all the uniforms we'll need to transform properly.
		}

		if (enableFog) {
			WRITE(p, "float2 u_fogcoef : register(c%i);\n", CONST_VS_FOGCOEF);
		}
		if (useHWTransform || !hasColor)
			WRITE(p, "float4 u_matambientalpha : register(c%i);\n", CONST_VS_MATAMBIENTALPHA);  // matambient + matalpha

		if (useHWTransform) {
			// When transforming by hardware, we need a great deal more uniforms...
			WRITE(p, "float4x3 u_world : register(c%i);\n", CONST_VS_WORLD);
			WRITE(p, "float4x3 u_view : register(c%i);\n", CONST_VS_VIEW);
			if (doTextureTransform)
				WRITE(p, "float4x3 u_tex : register(c%i);\n", CONST_VS_TEXMTX);
			if (enableBones) {
#ifdef USE_BONE_ARRAY
				WRITE(p, "float4x3 u_bone[%i] : register(c%i);\n", numBones, CONST_VS_BONE0);
#else
				for (int i = 0; i < numBoneWeights; i++) {
					WRITE(p, "float4x3 u_bone%i : register(c%i);\n", i, CONST_VS_BONE0 + i * 3);
				}
#endif
			}
			if (doTexture) {
				WRITE(p, "float4 u_uvscaleoffset : register(c%i);\n", CONST_VS_UVSCALEOFFSET);
			}
			for (int i = 0; i < 4; i++) {
				if (doLight[i] != LIGHT_OFF) {
					// This is needed for shade mapping
					WRITE(p, "float3 u_lightpos%i : register(c%i);\n", i, CONST_VS_LIGHTPOS + i);
				}
				if (doLight[i] == LIGHT_FULL) {
					GELightType type = static_cast<GELightType>(id.Bits(VS_BIT_LIGHT0_TYPE + 4 * i, 2));
					GELightComputation comp = static_cast<GELightComputation>(id.Bits(VS_BIT_LIGHT0_COMP + 4 * i, 2));

					if (type != GE_LIGHTTYPE_DIRECTIONAL)
						WRITE(p, "float3 u_lightatt%i : register(c%i);\n", i, CONST_VS_LIGHTATT + i);

					if (type == GE_LIGHTTYPE_SPOT || type == GE_LIGHTTYPE_UNKNOWN) {
						WRITE(p, "float3 u_lightdir%i : register(c%i);\n", i, CONST_VS_LIGHTDIR + i);
						WRITE(p, "float4 u_lightangle%i : register(c%i);\n", i, CONST_VS_LIGHTANGLE + i);
						WRITE(p, "float4 u_lightspotCoef%i : register(c%i);\n", i, CONST_VS_LIGHTSPOTCOEF + i);
					}
					WRITE(p, "float3 u_lightambient%i : register(c%i);\n", i, CONST_VS_LIGHTAMBIENT + i);
					WRITE(p, "float3 u_lightdiffuse%i : register(c%i);\n", i, CONST_VS_LIGHTDIFFUSE + i);

					if (comp != GE_LIGHTCOMP_ONLYDIFFUSE) {
						WRITE(p, "float3 u_lightspecular%i : register(c%i);\n", i, CONST_VS_LIGHTSPECULAR + i);
					}
				}
			}
			if (enableLighting) {
				WRITE(p, "float4 u_ambient : register(c%i);\n", CONST_VS_AMBIENT);
				if ((gstate.materialupdate & 2) == 0 || !hasColor)
					WRITE(p, "float3 u_matdiffuse : register(c%i);\n", CONST_VS_MATDIFFUSE);
				// if ((gstate.materialupdate & 4) == 0)
				WRITE(p, "float4 u_matspecular : register(c%i);\n", CONST_VS_MATSPECULAR);  // Specular coef is contained in alpha
				WRITE(p, "float3 u_matemissive : register(c%i);\n", CONST_VS_MATEMISSIVE);
			}
		}

		if (!isModeThrough && gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
			WRITE(p, "float4 u_depthRange : register(c%i);\n", CONST_VS_DEPTHRANGE);
		}
	} else {
		WRITE(p, "cbuffer base : register(b0) {\n%s};\n", cb_baseStr);
		WRITE(p, "cbuffer lights: register(b1) {\n%s};\n", cb_vs_lightsStr);
		WRITE(p, "cbuffer bones : register(b2) {\n%s};\n", cb_vs_bonesStr);
	}

	// And the "varyings".
	bool texCoordInVec3 = false;
	if (useHWTransform) {
		WRITE(p, "struct VS_IN {                              \n");
		if ((doSpline || doBezier) && lang == HLSL_D3D11) {
			WRITE(p, "  uint instanceId : SV_InstanceID;\n");
		}
		if (enableBones) {
			WRITE(p, "  %s", boneWeightAttrDecl[numBoneWeights]);
		}
		if (doTexture && hasTexcoord) {
			WRITE(p, "  float2 texcoord : TEXCOORD0;\n");
		}
		if (hasColor)  {
			WRITE(p, "  float4 color0 : COLOR0;\n");
		}
		if (hasNormal) {
			WRITE(p, "  float3 normal : NORMAL;\n");
		}
		WRITE(p, "  float3 position : POSITION;\n");
		WRITE(p, "};\n");
		
	} else {
		WRITE(p, "struct VS_IN {\n");
		WRITE(p, "  float4 position : POSITION;\n");
		if (doTexture && hasTexcoord) {
			if (doTextureTransform && !throughmode) {
				texCoordInVec3 = true;
				WRITE(p, "  float3 texcoord : TEXCOORD0;\n");
			}
			else
				WRITE(p, "  float2 texcoord : TEXCOORD0;\n");
		}
		if (hasColor) {
			WRITE(p, "  float4 color0 : COLOR0;\n");
		}
		// only software transform supplies color1 as vertex data
		if (lmode) {
			WRITE(p, "  float4 color1 : COLOR1;\n");
		}
		WRITE(p, "};\n");
	}

	WRITE(p, "struct VS_OUT {\n");
	if (doTexture) {
		WRITE(p, "  float3 v_texcoord : TEXCOORD0;\n");
	}
	WRITE(p, "  float4 v_color0    : COLOR0;\n");
	if (lmode)
		WRITE(p, "  float3 v_color1    : COLOR1;\n");

	if (enableFog) {
		WRITE(p, "  float v_fogdepth: TEXCOORD1;\n");
	}
	if (lang == HLSL_DX9) {
		WRITE(p, "  float4 gl_Position   : POSITION;\n");
	} else {
		WRITE(p, "  float4 gl_Position   : SV_Position;\n");
	}
	WRITE(p, "};\n");

	// Confirmed: Through mode gets through exactly the same in GL and D3D in Phantasy Star: Text is 38023.0 in the test scene.

	if (!isModeThrough && gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
		// Apply the projection and viewport to get the Z buffer value, floor to integer, undo the viewport and projection.
		// The Z range in D3D is different but we compensate for that using parameters.
		WRITE(p, "\nfloat4 depthRoundZVP(float4 v) {\n");
		WRITE(p, "  float z = v.z / v.w;\n");
		WRITE(p, "  z = (z * u_depthRange.x + u_depthRange.y);\n");
		WRITE(p, "  z = floor(z);\n");
		WRITE(p, "  z = (z - u_depthRange.z) * u_depthRange.w;\n");
		WRITE(p, "  return float4(v.x, v.y, z * v.w, v.w);\n");
		WRITE(p, "}\n\n");
	}

	// Hardware tessellation
	if (doSpline || doBezier) {
		if (lang == HLSL_D3D11 || lang == HLSL_D3D11_LEVEL9) {
			WRITE(p, "Texture1D<float3> u_tess_pos_tex : register(t0);\n");
			WRITE(p, "Texture1D<float3> u_tess_tex_tex : register(t1);\n");
			WRITE(p, "Texture1D<float4> u_tess_col_tex : register(t2);\n");
		}

		const char *init[3] = { "0.0, 0.0", "0.0, 0.0, 0.0", "0.0, 0.0, 0.0, 0.0" };
		for (int i = 2; i <= 4; i++) {
			// Define 3 types float2, float3, float4
			WRITE(p, "float%d tess_sample(in float%d points[16], in float2 weights[4]) {\n", i, i);
			WRITE(p, "  float%d pos = float%d(%s);\n", i, i, init[i - 2]);
			WRITE(p, "  for (int i = 0; i < 4; ++i) {\n");
			WRITE(p, "    for (int j = 0; j < 4; ++j) {\n");
			WRITE(p, "      float f = weights[j].x * weights[i].y;\n");
			WRITE(p, "      if (f != 0.0)\n");
			WRITE(p, "        pos = pos + f * points[i * 4 + j];\n");
			WRITE(p, "    }\n");
			WRITE(p, "  }\n");
			WRITE(p, "  return pos;\n");
			WRITE(p, "}\n");
		}
		if (doSpline) {
			WRITE(p, "void spline_knot(int2 num_patches, int2 type, out float2 knot[6], int2 patch_pos) {\n");
			WRITE(p, "  for (int i = 0; i < 6; ++i) {\n");
			WRITE(p, "    knot[i] = float2(i + patch_pos.x - 2, i + patch_pos.y - 2);\n");
			WRITE(p, "  }\n");
		//	WRITE(p, "  if ((type.x & 1) != 0) {\n");
			WRITE(p, "  if ((type.x == 1) || (type.x == 3)) {\n");
			WRITE(p, "    if (patch_pos.x <= 2)\n");
			WRITE(p, "      knot[0].x = 0.0;\n");
			WRITE(p, "    if (patch_pos.x <= 1)\n");
			WRITE(p, "      knot[1].x = 0.0;\n");
			WRITE(p, "  }\n");
		//	WRITE(p, "  if ((type.x & 2) != 0) {\n");
			WRITE(p, "  if ((type.x == 2) || (type.x == 3)) {\n");
			WRITE(p, "    if (patch_pos.x >= (num_patches.x - 2))\n");
			WRITE(p, "      knot[5].x = num_patches.x;\n");
			WRITE(p, "    if (patch_pos.x == (num_patches.x - 1))\n");
			WRITE(p, "      knot[4].x = num_patches.x;\n");
			WRITE(p, "  }\n");
		//	WRITE(p, "  if ((type.y & 1) != 0) {\n");
			WRITE(p, "  if ((type.y == 1) || (type.y == 3)) {\n");
			WRITE(p, "    if (patch_pos.y <= 2)\n");
			WRITE(p, "      knot[0].y = 0.0;\n");
			WRITE(p, "    if (patch_pos.y <= 1)\n");
			WRITE(p, "      knot[1].y = 0.0;\n");
			WRITE(p, "  }\n");
		//	WRITE(p, "  if ((type.y & 2) != 0) {\n");
			WRITE(p, "  if ((type.y == 2) || (type.y == 3)) {\n");
			WRITE(p, "    if (patch_pos.y >= (num_patches.y - 2))\n");
			WRITE(p, "      knot[5].y = num_patches.y;\n");
			WRITE(p, "    if (patch_pos.y == (num_patches.y - 1))\n");
			WRITE(p, "      knot[4].y = num_patches.y;\n");
			WRITE(p, "  }\n");
			WRITE(p, "}\n");

			WRITE(p, "void spline_weight(float2 t, in float2 knot[6], out float2 weights[4]) {\n");
			// TODO: Maybe compilers could be coaxed into vectorizing this code without the above explicitly...
			WRITE(p, "  float2 t0 = (t - knot[0]);\n");
			WRITE(p, "  float2 t1 = (t - knot[1]);\n");
			WRITE(p, "  float2 t2 = (t - knot[2]);\n");
			// TODO: All our knots are integers so we should be able to get rid of these divisions (How?)
			WRITE(p, "  float2 f30 = t0 / (knot[3] - knot[0]);\n");
			WRITE(p, "  float2 f41 = t1 / (knot[4] - knot[1]);\n");
			WRITE(p, "  float2 f52 = t2 / (knot[5] - knot[2]);\n");
			WRITE(p, "  float2 f31 = t1 / (knot[3] - knot[1]);\n");
			WRITE(p, "  float2 f42 = t2 / (knot[4] - knot[2]);\n");
			WRITE(p, "  float2 f32 = t2 / (knot[3] - knot[2]);\n");
			WRITE(p, "  float2 a = (1.0 - f30)*(1.0 - f31);\n");
			WRITE(p, "  float2 b = (f31*f41);\n");
			WRITE(p, "  float2 c = (1.0 - f41)*(1.0 - f42);\n");
			WRITE(p, "  float2 d = (f42*f52);\n");
			WRITE(p, "  weights[0] = a - (a*f32);\n");
			WRITE(p, "  weights[1] = 1.0 - a - b + ((a + b + c - 1.0)*f32);\n");
			WRITE(p, "  weights[2] = b + ((1.0 - b - c - d)*f32);\n");
			WRITE(p, "  weights[3] = d*f32;\n");
			WRITE(p, "}\n");
		}
	}

	WRITE(p, "VS_OUT main(VS_IN In) {\n");
	WRITE(p, "  VS_OUT Out;\n");  
	if (!useHWTransform) {
		// Simple pass-through of vertex data to fragment shader
		if (doTexture) {
			if (texCoordInVec3) {
				WRITE(p, "  Out.v_texcoord = In.texcoord;\n");
			} else {
				WRITE(p, "  Out.v_texcoord = float3(In.texcoord, 1.0);\n");
			}
		}
		if (hasColor) {
			WRITE(p, "  Out.v_color0 = In.color0;\n");
			if (lmode)
				WRITE(p, "  Out.v_color1 = In.color1.rgb;\n");
		} else {
			WRITE(p, "  Out.v_color0 = In.u_matambientalpha;\n");
			if (lmode)
				WRITE(p, "  Out.v_color1 = float3(0.0);\n");
		}
		if (enableFog) {
			WRITE(p, "  Out.v_fogdepth = In.position.w;\n");
		}
		if (lang == HLSL_D3D11 || lang == HLSL_D3D11_LEVEL9) {
			if (gstate.isModeThrough()) {
				WRITE(p, "  Out.gl_Position = mul(u_proj_through, float4(In.position.xyz, 1.0));\n");
			} else {
				if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
					WRITE(p, "  Out.gl_Position = depthRoundZVP(mul(u_proj, float4(In.position.xyz, 1.0)));\n");
				} else {
					WRITE(p, "  Out.gl_Position = mul(u_proj, float4(In.position.xyz, 1.0));\n");
				}
			}
		} else {
			if (gstate.isModeThrough()) {
				WRITE(p, "  Out.gl_Position = mul(float4(In.position.xyz, 1.0), u_proj_through);\n");
			} else {
				if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
					WRITE(p, "  Out.gl_Position = depthRoundZVP(mul(float4(In.position.xyz, 1.0), u_proj));\n");
				} else {
					WRITE(p, "  Out.gl_Position = mul(float4(In.position.xyz, 1.0), u_proj);\n");
				}
			}
		}
	}  else {
		// Step 1: World Transform / Skinning
		if (!enableBones) {
			// Hardware tessellation
			if (doSpline || doBezier) {
				WRITE(p, "  uint num_patches_u = %s;\n", doBezier ? "(u_spline_count_u - 1) / 3u" : "u_spline_count_u - 3");
				WRITE(p, "  float2 tess_pos = In.position.xy;\n");
				WRITE(p, "  int u = In.instanceId %% num_patches_u;\n");
				WRITE(p, "  int v = In.instanceId / num_patches_u;\n");
				WRITE(p, "  int2 patch_pos = int2(u, v);\n");
				WRITE(p, "  float3 _pos[16];\n");
				WRITE(p, "  float2 _tex[16];\n");
				WRITE(p, "  float4 _col[16];\n");
				WRITE(p, "  int idx;\n");
				WRITE(p, "  int2 index;\n");
				for (int i = 0; i < 4; i++) {
					for (int j = 0; j < 4; j++) {
						WRITE(p, "  idx = (%i + v%s) * u_spline_count_u + (%i + u%s);\n", i, doBezier ? " * 3" : "", j, doBezier ? " * 3" : "");
						WRITE(p, "  index = int2(idx, 0);\n");
						WRITE(p, "  _pos[%i] = u_tess_pos_tex.Load(index).xyz;\n", i * 4 + j);
						if (doTexture && hasTexcoord && hasTexcoordTess)
							WRITE(p, "  _tex[%i] = u_tess_tex_tex.Load(index).xy;\n", i * 4 + j);
						if (hasColor && hasColorTess)
							WRITE(p, "  _col[%i] = u_tess_col_tex.Load(index).rgba;\n", i * 4 + j);
					}
				}
				WRITE(p, "  float2 weights[4];\n");
				if (doBezier) {
					// Bernstein 3D
					WRITE(p, "  weights[0] = (1.0 - tess_pos) * (1.0 - tess_pos) * (1.0 - tess_pos);\n");
					WRITE(p, "  weights[1] = 3.0 * tess_pos * (1.0 - tess_pos) * (1.0 - tess_pos);\n");
					WRITE(p, "  weights[2] = 3.0 * tess_pos * tess_pos * (1.0 - tess_pos);\n");
					WRITE(p, "  weights[3] = tess_pos * tess_pos * tess_pos;\n");
				} else if (doSpline) {
					WRITE(p, "  int2 spline_num_patches = int2(u_spline_count_u - 3, u_spline_count_v - 3);\n");
					WRITE(p, "  int2 spline_type = int2(u_spline_type_u, u_spline_type_v);\n");
					WRITE(p, "  float2 knots[6];\n");
					WRITE(p, "  spline_knot(spline_num_patches, spline_type, knots, patch_pos);\n");
					WRITE(p, "  spline_weight(tess_pos + patch_pos, knots, weights);\n");
				}
				WRITE(p, "  float3 pos = tess_sample(_pos, weights);\n");
				if (doTexture && hasTexcoord) {
					if (hasTexcoordTess)
						WRITE(p, "  float2 tex = tess_sample(_tex, weights);\n");
					else
						WRITE(p, "  float2 tex = tess_pos + patch_pos;\n");
				}
				if (hasColor) {
					if (hasColorTess)
						WRITE(p, "  float4 col = tess_sample(_col, weights);\n");
					else
						WRITE(p, "  float4 col = u_tess_col_tex.Load(int2(0, 0)).rgba;\n");
				}
				if (hasNormal) {
					// Curved surface is probably always need to compute normal(not sampling from control points)
					if (doBezier) {
						// Bernstein derivative
						WRITE(p, "  float2 bernderiv[4];\n");
						WRITE(p, "  bernderiv[0] = -3.0 * (tess_pos - 1.0) * (tess_pos - 1.0); \n");
						WRITE(p, "  bernderiv[1] = 9.0 * tess_pos * tess_pos - 12.0 * tess_pos + 3.0; \n");
						WRITE(p, "  bernderiv[2] = 3.0 * (2.0 - 3.0 * tess_pos) * tess_pos; \n");
						WRITE(p, "  bernderiv[3] = 3.0 * tess_pos * tess_pos; \n");

						WRITE(p, "  float2 bernderiv_u[4];\n");
						WRITE(p, "  float2 bernderiv_v[4];\n");
						WRITE(p, "  for (int i = 0; i < 4; i++) {\n");
						WRITE(p, "    bernderiv_u[i] = float2(bernderiv[i].x, weights[i].y);\n");
						WRITE(p, "    bernderiv_v[i] = float2(weights[i].x, bernderiv[i].y);\n");
						WRITE(p, "  }\n");

						WRITE(p, "  float3 du = tess_sample(_pos, bernderiv_u);\n");
						WRITE(p, "  float3 dv = tess_sample(_pos, bernderiv_v);\n");
					} else if (doSpline) {
						WRITE(p, "  float2 tess_next_u = float2(In.normal.x, 0.0);\n");
						WRITE(p, "  float2 tess_next_v = float2(0.0, In.normal.y);\n");
						// Right
						WRITE(p, "  float2 tess_pos_r = tess_pos + tess_next_u;\n");
						WRITE(p, "  spline_weight(tess_pos_r + patch_pos, knots, weights);\n");
						WRITE(p, "  float3 pos_r = tess_sample(_pos, weights);\n");
						// Left
						WRITE(p, "  float2 tess_pos_l = tess_pos - tess_next_u;\n");
						WRITE(p, "  spline_weight(tess_pos_l + patch_pos, knots, weights);\n");
						WRITE(p, "  float3 pos_l = tess_sample(_pos, weights);\n");
						// Down
						WRITE(p, "  float2 tess_pos_d = tess_pos + tess_next_v;\n");
						WRITE(p, "  spline_weight(tess_pos_d + patch_pos, knots, weights);\n");
						WRITE(p, "  float3 pos_d = tess_sample(_pos, weights);\n");
						// Up
						WRITE(p, "  float2 tess_pos_u = tess_pos - tess_next_v;\n");
						WRITE(p, "  spline_weight(tess_pos_u + patch_pos, knots, weights);\n");
						WRITE(p, "  float3 pos_u = tess_sample(_pos, weights);\n");

						WRITE(p, "  float3 du = pos_r - pos_l;\n");
						WRITE(p, "  float3 dv = pos_d - pos_u;\n");
					}
					WRITE(p, "  float3 nrm = cross(du, dv);\n");
					WRITE(p, "  nrm = normalize(nrm);\n");
				}
				WRITE(p, "  float3 worldpos = mul(float4(pos.xyz, 1.0), u_world);\n");
				if (hasNormal)
					WRITE(p, "  float3 worldnormal = normalize(mul(float4(%snrm, 0.0), u_world));\n", flipNormalTess ? "-" : "");
				else
					WRITE(p, "  float3 worldnormal = float3(0.0, 0.0, 1.0);\n");
			} else {
				// No skinning, just standard T&L.
				WRITE(p, "  float3 worldpos = mul(float4(In.position.xyz, 1.0), u_world);\n");
				if (hasNormal)
					WRITE(p, "  float3 worldnormal = normalize(mul(float4(%sIn.normal, 0.0), u_world));\n", flipNormal ? "-" : "");
				else
					WRITE(p, "  float3 worldnormal = float3(0.0, 0.0, 1.0);\n");
			}
		} else {
			static const char * const boneWeightAttr[8] = {
				"a_w1.x", "a_w1.y", "a_w1.z", "a_w1.w",
				"a_w2.x", "a_w2.y", "a_w2.z", "a_w2.w",
			};

#if defined(USE_FOR_LOOP) && defined(USE_BONE_ARRAY)

			// To loop through the weights, we unfortunately need to put them in a float array.
			// GLSL ES sucks - no way to directly initialize an array!
			switch (numBoneWeights) {
			case 1: WRITE(p, "  float w[1]; w[0] = a_w1;\n"); break;
			case 2: WRITE(p, "  float w[2]; w[0] = a_w1.x; w[1] = a_w1.y;\n"); break;
			case 3: WRITE(p, "  float w[3]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z;\n"); break;
			case 4: WRITE(p, "  float w[4]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z; w[3] = a_w1.w;\n"); break;
			case 5: WRITE(p, "  float w[5]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z; w[3] = a_w1.w; w[4] = a_w2;\n"); break;
			case 6: WRITE(p, "  float w[6]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z; w[3] = a_w1.w; w[4] = a_w2.x; w[5] = a_w2.y;\n"); break;
			case 7: WRITE(p, "  float w[7]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z; w[3] = a_w1.w; w[4] = a_w2.x; w[5] = a_w2.y; w[6] = a_w2.z;\n"); break;
			case 8: WRITE(p, "  float w[8]; w[0] = a_w1.x; w[1] = a_w1.y; w[2] = a_w1.z; w[3] = a_w1.w; w[4] = a_w2.x; w[5] = a_w2.y; w[6] = a_w2.z; w[7] = a_w2.w;\n"); break;
			}

			WRITE(p, "  mat4 skinMatrix = w[0] * u_bone[0];\n");
			if (numBoneWeights > 1) {
				WRITE(p, "  for (int i = 1; i < %i; i++) {\n", numBoneWeights);
				WRITE(p, "    skinMatrix += w[i] * u_bone[i];\n");
				WRITE(p, "  }\n");
			}

#else
			if (lang == HLSL_D3D11 || lang == HLSL_D3D11_LEVEL9) {
				if (numBoneWeights == 1)
					WRITE(p, "  float4x3 skinMatrix = mul(In.a_w1, u_bone[0])");
				else
					WRITE(p, "  float4x3 skinMatrix = mul(In.a_w1.x, u_bone[0])");
				for (int i = 1; i < numBoneWeights; i++) {
					const char *weightAttr = boneWeightAttr[i];
					// workaround for "cant do .x of scalar" issue
					if (numBoneWeights == 1 && i == 0) weightAttr = "a_w1";
					if (numBoneWeights == 5 && i == 4) weightAttr = "a_w2";
					WRITE(p, " + mul(In.%s, u_bone[%i])", weightAttr, i);
				}
			} else {
				if (numBoneWeights == 1)
					WRITE(p, "  float4x3 skinMatrix = mul(In.a_w1, u_bone0)");
				else
					WRITE(p, "  float4x3 skinMatrix = mul(In.a_w1.x, u_bone0)");
				for (int i = 1; i < numBoneWeights; i++) {
					const char *weightAttr = boneWeightAttr[i];
					// workaround for "cant do .x of scalar" issue
					if (numBoneWeights == 1 && i == 0) weightAttr = "a_w1";
					if (numBoneWeights == 5 && i == 4) weightAttr = "a_w2";
					WRITE(p, " + mul(In.%s, u_bone%i)", weightAttr, i);
				}
			}
#endif

			WRITE(p, ";\n");

			// Trying to simplify this results in bugs in LBP...
			WRITE(p, "  float3 skinnedpos = mul(float4(In.position.xyz, 1.0), skinMatrix);\n");
			WRITE(p, "  float3 worldpos = mul(float4(skinnedpos, 1.0), u_world);\n");

			if (hasNormal) {
				WRITE(p, "  float3 skinnednormal = mul(float4(%sIn.normal, 0.0), skinMatrix);\n", flipNormal ? "-" : "");
			} else {
				WRITE(p, "  float3 skinnednormal = mul(float4(0.0, 0.0, %s1.0, 0.0), skinMatrix);\n", flipNormal ? "-" : "");
			}
			WRITE(p, "  float3 worldnormal = normalize(mul(float4(skinnednormal, 0.0), u_world));\n");
		}

		WRITE(p, "  float4 viewPos = float4(mul(float4(worldpos, 1.0), u_view), 1.0);\n");

		if (lang == HLSL_D3D11 || lang == HLSL_D3D11_LEVEL9) {
			// Final view and projection transforms.
			if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
				WRITE(p, "  Out.gl_Position = depthRoundZVP(mul(u_proj, viewPos));\n");
			} else {
				WRITE(p, "  Out.gl_Position = mul(u_proj, viewPos);\n");
			}
		} else {
			// Final view and projection transforms.
			if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
				WRITE(p, "  Out.gl_Position = depthRoundZVP(mul(viewPos, u_proj));\n");
			} else {
				WRITE(p, "  Out.gl_Position = mul(viewPos, u_proj);\n");
			}
		}

		// TODO: Declare variables for dots for shade mapping if needed.

		const char *ambientStr = (gstate.materialupdate & 1) && hasColor ? "In.color0" : "u_matambientalpha";
		const char *diffuseStr = (gstate.materialupdate & 2) && hasColor ? "In.color0.rgb" : "u_matdiffuse";
		const char *specularStr = (gstate.materialupdate & 4) && hasColor ? "In.color0.rgb" : "u_matspecular.rgb";
		if (doBezier || doSpline) {
			ambientStr = (matUpdate & 1) && hasColor ? "col" : "u_matambientalpha";
			diffuseStr = (matUpdate & 2) && hasColor ? "col.rgb" : "u_matdiffuse";
			specularStr = (matUpdate & 4) && hasColor ? "col.rgb" : "u_matspecular.rgb";
		}

		bool diffuseIsZero = true;
		bool specularIsZero = true;
		bool distanceNeeded = false;

		if (enableLighting) {
			WRITE(p, "  float4 lightSum0 = u_ambient * %s + float4(u_matemissive, 0.0);\n", ambientStr);

			for (int i = 0; i < 4; i++) {
				GELightType type = static_cast<GELightType>(id.Bits(VS_BIT_LIGHT0_TYPE + 4 * i, 2));
				GELightComputation comp = static_cast<GELightComputation>(id.Bits(VS_BIT_LIGHT0_COMP + 4 * i, 2));
				if (doLight[i] != LIGHT_FULL)
					continue;
				diffuseIsZero = false;
				if (comp != GE_LIGHTCOMP_ONLYDIFFUSE)
					specularIsZero = false;
				if (type != GE_LIGHTTYPE_DIRECTIONAL)
					distanceNeeded = true;
			}

			if (!specularIsZero) {
				WRITE(p, "  float3 lightSum1 = 0;\n");
			}
			if (!diffuseIsZero) {
				WRITE(p, "  float3 toLight;\n");
				WRITE(p, "  float3 diffuse;\n");
			}
			if (distanceNeeded) {
				WRITE(p, "  float distance;\n");
				WRITE(p, "  float lightScale;\n");
			}
		}

		// Calculate lights if needed. If shade mapping is enabled, lights may need to be
		// at least partially calculated.
		for (int i = 0; i < 4; i++) {
			if (doLight[i] != LIGHT_FULL)
				continue;

			GELightType type = static_cast<GELightType>(id.Bits(VS_BIT_LIGHT0_TYPE + 4 * i, 2));
			GELightComputation comp = static_cast<GELightComputation>(id.Bits(VS_BIT_LIGHT0_COMP + 4 * i, 2));

			if (type == GE_LIGHTTYPE_DIRECTIONAL) {
				// We prenormalize light positions for directional lights.
				WRITE(p, "  toLight = u_lightpos%i;\n", i);
			} else {
				WRITE(p, "  toLight = u_lightpos%i - worldpos;\n", i);
				WRITE(p, "  distance = length(toLight);\n");
				WRITE(p, "  toLight /= distance;\n");
			}

			bool doSpecular = comp != GE_LIGHTCOMP_ONLYDIFFUSE;
			bool poweredDiffuse = comp == GE_LIGHTCOMP_BOTHWITHPOWDIFFUSE;

			if (poweredDiffuse) {
				WRITE(p, "  float dot%i = pow(dot(toLight, worldnormal), u_matspecular.a);\n", i);
				// TODO: Somehow the NaN check from GLES seems unnecessary here?
				// If it returned 0, it'd be wrong, so that's strange.
			} else {
				WRITE(p, "  float dot%i = dot(toLight, worldnormal);\n", i);
			}

			const char *timesLightScale = " * lightScale";

			// Attenuation
			switch (type) {
			case GE_LIGHTTYPE_DIRECTIONAL:
				timesLightScale = "";
				break;
			case GE_LIGHTTYPE_POINT:
				WRITE(p, "  lightScale = clamp(1.0 / dot(u_lightatt%i, float3(1.0, distance, distance*distance)), 0.0, 1.0);\n", i);
				break;
			case GE_LIGHTTYPE_SPOT:
			case GE_LIGHTTYPE_UNKNOWN:
				WRITE(p, "  float angle%i = dot(normalize(u_lightdir%i), toLight);\n", i, i);
				WRITE(p, "  if (angle%i >= u_lightangle%i.x) {\n", i, i);
				WRITE(p, "    lightScale = clamp(1.0 / dot(u_lightatt%i, float3(1.0, distance, distance*distance)), 0.0, 1.0) * pow(angle%i, u_lightspotCoef%i.x);\n", i, i, i);
				WRITE(p, "  } else {\n");
				WRITE(p, "    lightScale = 0.0;\n");
				WRITE(p, "  }\n");
				break;
			default:
				// ILLEGAL
				break;
			}

			WRITE(p, "  diffuse = (u_lightdiffuse%i * %s) * max(dot%i, 0.0);\n", i, diffuseStr, i);
			if (doSpecular) {
				WRITE(p, "  dot%i = dot(normalize(toLight + float3(0.0, 0.0, 1.0)), worldnormal);\n", i);
				WRITE(p, "  if (dot%i > 0.0)\n", i);
				WRITE(p, "    lightSum1 += u_lightspecular%i * %s * (pow(dot%i, u_matspecular.a) %s);\n", i, specularStr, i, timesLightScale);
			}
			WRITE(p, "  lightSum0.rgb += (u_lightambient%i * %s.rgb + diffuse)%s;\n", i, ambientStr, timesLightScale);
		}

		if (enableLighting) {
			// Sum up ambient, emissive here.
			if (lmode) {
				WRITE(p, "  Out.v_color0 = clamp(lightSum0, 0.0, 1.0);\n");
				// v_color1 only exists when lmode = 1.
				if (specularIsZero) {
					WRITE(p, "  Out.v_color1 = float3(0, 0, 0);\n");
				} else {
					WRITE(p, "  Out.v_color1 = clamp(lightSum1, 0.0, 1.0);\n");
				}
			} else {
				if (specularIsZero) {
					WRITE(p, "  Out.v_color0 = clamp(lightSum0, 0.0, 1.0);\n");
				} else {
					WRITE(p, "  Out.v_color0 = clamp(clamp(lightSum0, 0.0, 1.0) + float4(lightSum1, 0.0), 0.0, 1.0);\n");
				}
			}
		} else {
			// Lighting doesn't affect color.
			if (hasColor) {
				if (doBezier || doSpline)
					WRITE(p, "  Out.v_color0 = col;\n");
				else
					WRITE(p, "  Out.v_color0 = In.color0;\n");
			} else {
				WRITE(p, "  Out.v_color0 = u_matambientalpha;\n");
			}
			if (lmode)
				WRITE(p, "  Out.v_color1 = float3(0, 0, 0);\n");
		}

		bool scaleUV = !throughmode && (uvGenMode == GE_TEXMAP_TEXTURE_COORDS || uvGenMode == GE_TEXMAP_UNKNOWN);


		// Step 3: UV generation
		if (doTexture) {
			switch (uvGenMode) {
			case GE_TEXMAP_TEXTURE_COORDS:  // Scale-offset. Easy.
			case GE_TEXMAP_UNKNOWN: // Not sure what this is, but Riviera uses it.  Treating as coords works.
				if (scaleUV) {
					if (hasTexcoord) {
						if (doBezier || doSpline)
							WRITE(p, "  Out.v_texcoord = float3(tex.xy * u_uvscaleoffset.xy + u_uvscaleoffset.zw, 0.0);\n");
						else
							WRITE(p, "  Out.v_texcoord = float3(In.texcoord.xy * u_uvscaleoffset.xy, 0.0);\n");
					} else {
						WRITE(p, "  Out.v_texcoord = float3(0.0, 0.0, 0.0);\n");
					}
				} else {
					if (hasTexcoord) {
						if (doBezier || doSpline)
							WRITE(p, "  Out.v_texcoord = float3(tex.xy * u_uvscaleoffset.xy + u_uvscaleoffset.zw, 0.0);\n");
						else
							WRITE(p, "  Out.v_texcoord = float3(In.texcoord.xy * u_uvscaleoffset.xy + u_uvscaleoffset.zw, 0.0);\n");
					} else {
						WRITE(p, "  Out.v_texcoord = float3(u_uvscaleoffset.zw, 0.0);\n");
					}
				}
				break;

			case GE_TEXMAP_TEXTURE_MATRIX:  // Projection mapping.
				{
					std::string temp_tc;
					switch (uvProjMode) {
					case GE_PROJMAP_POSITION:  // Use model space XYZ as source
						temp_tc = "float4(In.position.xyz, 1.0)";
						break;
					case GE_PROJMAP_UV:  // Use unscaled UV as source
						{
							if (hasTexcoord) {
								temp_tc = StringFromFormat("float4(In.texcoord.xy, 0.0, 1.0)");
							} else {
								temp_tc = "float4(0.0, 0.0, 0.0, 1.0)";
							}
						}
						break;
					case GE_PROJMAP_NORMALIZED_NORMAL:  // Use normalized transformed normal as source
						if (hasNormal)
							temp_tc = flipNormal ? "float4(normalize(-In.normal), 1.0)" : "float4(normalize(In.normal), 1.0)";
						else
							temp_tc = "float4(0.0, 0.0, 1.0, 1.0)";
						break;
					case GE_PROJMAP_NORMAL:  // Use non-normalized transformed normal as source
						if (hasNormal)
							temp_tc =  flipNormal ? "float4(-In.normal, 1.0)" : "float4(In.normal, 1.0)";
						else
							temp_tc = "float4(0.0, 0.0, 1.0, 1.0)";
						break;
					}
					// Transform by texture matrix. XYZ as we are doing projection mapping.
					WRITE(p, "  Out.v_texcoord.xyz = mul(%s, u_tex) * float3(u_uvscaleoffset.xy, 1.0);\n", temp_tc.c_str());
				}
				break;

			case GE_TEXMAP_ENVIRONMENT_MAP:  // Shade mapping - use dots from light sources.
				WRITE(p, "  Out.v_texcoord = float3(u_uvscaleoffset.xy * float2(1.0 + dot(normalize(u_lightpos%i), worldnormal), 1.0 + dot(normalize(u_lightpos%i), worldnormal)) * 0.5, 1.0);\n", ls0, ls1);
				break;

			default:
				// ILLEGAL
				break;
			}
		}

		// Compute fogdepth
		if (enableFog) {
			WRITE(p, "  Out.v_fogdepth = (viewPos.z + u_fogcoef.x) * u_fogcoef.y;\n");
		}
	}

	WRITE(p, "  return Out;\n");
	WRITE(p, "}\n");
}

};