// zlib open source license
//
// Copyright (c) 2018 to 2019 David Forsgren Piuva
// 
// This software is provided 'as-is', without any express or implied
// warranty. In no event will the authors be held liable for any damages
// arising from the use of this software.
// 
// Permission is granted to anyone to use this software for any purpose,
// including commercial applications, and to alter it and redistribute it
// freely, subject to the following restrictions:
// 
//    1. The origin of this software must not be misrepresented; you must not
//    claim that you wrote the original software. If you use this software
//    in a product, an acknowledgment in the product documentation would be
//    appreciated but is not required.
// 
//    2. Altered source versions must be plainly marked as such, and must not be
//    misrepresented as being the original software.
// 
//    3. This notice may not be removed or altered from any source
//    distribution.

#ifndef DFPSR_API_MODEL
#define DFPSR_API_MODEL

#include "types.h"
#include "../math/FVector.h"

// TODO: How should these be exposed to the caller?
#include "../render/Camera.h"
#include "../render/ResourcePool.h"

namespace dsr {
	// Normalized texture coordinates:
	//   (0.0f, 0.0f) is the texture coordinate for the upper left corner of the upper left pixel in the 2D texture.
	//   (1.0f, 0.0f) is the texture coordinate for the upper right corner of the upper right pixel in the 2D texture.
	//   (0.0f, 1.0f) is the texture coordinate for the lower left corner of the lower left pixel in the 2D texture.
	//   (1.0f, 1.0f) is the texture coordinate for the lower right corner of the lower right pixel in the 2D texture.
	//   (0.5f, 0.5f) is the texture coordinate for the center of the 2D texture.

	// Texture sampling:
	//   By default, texture sampling is looped around the edges with bilinear with mip-maps for diffuse textures when available.
	//   In bi-linear interpolation, the center of the pixel has the full weight of the color while the sides may show pixels from the other end of the texture.
	//   When getting further away or viewing from the side, a lower resolution version of the texture will be taken from the pyramid if it was generated.
	//   Seams between resolutions will be hard seams, so avoid using hard lines in textures if you want it to look natural.
	//   Loading textures from a ResourcePool will automatically call image_generatePyramid.
	//   Images without the power of two dimensions needed to generate a pyramid can not be used as textures in models.
	//   image_isTexture can be used to know if an image has the supported dimensions for the current version of the renderer.

	// Side-effect: Creates a new empty model.
	// Post-condition: Returns a reference counted handle to the new model.
	// The model will be deleted automatically when all handles are gone.
	Model model_create();
	// Clones the geometry but refers to the same textures to save memory.
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns a reference counted handle to the clone of model.
	Model model_clone(const Model& model);

	// Assign a filter to the whole model.
	//   Assigning filters per model makes it easier to draw solid models before filtered models.
	//   Two separate models can be used if you need both solid and filtered geometry.
	// Filters:
	//   Filter::Alpha uses the alpha channel from the shader as opacity.
	//   Filter::Solid is the default setting for newly created models.
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Sets the given model's filter to the given filter argument.
	void model_setFilter(const Model& model, Filter filter);
	// Get back the filter enumeration, which was assigned to the model using model_setFilter.
	// This is useful for knowing in which pass to render your model.
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the model's filter enumeration
	Filter model_getFilter(const Model& model);
	// Post-condition: Returns true iff the model exists.
	bool model_exists(const Model& model);

	// Each part contains material settings and a list of polygons.
	//   Each polygon contains 3 or 4 vertices. (triangles and quads)
	//     Each vertex has with its own color, texture coordinates and position index.
	//     Position indices refers to the model's list of points,
	//     which is shared across multiple parts to avoid gaps between materials.
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Adds an empty part without any polygons and returns its new local part index.
	// The returned part index is relative to the model and goes from 0 to model_getNumberOfParts(model) - 1.
	int model_addEmptyPart(Model& model, const String &name);
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the number of parts in model.
	int model_getNumberOfParts(const Model& model);
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Sets the part at partIndex in model to the new name.
	void model_setPartName(Model& model, int partIndex, const String &name);
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the name of the part at partIndex in model.
	String model_getPartName(const Model& model, int partIndex);

	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the number of points in model.
	int model_getNumberOfPoints(const Model& model);
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the 3D position of the point at pointIndex in model.
	FVector3D model_getPoint(const Model& model, int pointIndex);
	// Pre-condition: model must refer to an existing model.
	void model_setPoint(Model& model, int pointIndex, const FVector3D& position);
	// Pre-condition: model must refer to an existing model.
	// Post-condition:
	//   Returns an index to the closest point in model relative to position in euclidean distance.
	//   Returns -1 if none was inside of threshold.
	// A point p is inside of threshold iff |p - position| < threshold.
	// If multiple points have the same distance approximated, the point with the lowest index will be preferred.
	int model_findPoint(const Model& model, const FVector3D &position, float threshold);
	// Add a point even if it overlaps an existing point.
	// Can be used for animation where the initial position might not always be the same.
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Adds a new point to model at position.
	// Post-condition: Returns a local index to the new point.
	int model_addPoint(const Model& model, const FVector3D &position);
	// Add a point, only if it does not overlap.
	// Can be used to seal small gaps and reduce the time needed to transform vertex positions.
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Adds a new point to model at position unless another point already exists within threshold so that model_findPoint(model, position, threshold) > -1.
	// Post-condition:
	//   If a new point was created then its new index is returned.
	//   Otherwise, if any existing point was within threshold,
	//   then the index of the closest existing point in euclidean distance is returned.
	//   If multiple existing points are within the same distance,
	//   then the point with the lowest index is preferred, just like in model_findPoint.
	int model_addPointIfNeeded(Model& model, const FVector3D &position, float threshold);
	// Get the bounding box, which expands automatically when adding or moving points in the model.
	// Side-effect: Writes model's bounding box to minimum and maximum by reference.
	void model_getBoundingBox(const Model& model, FVector3D& minimum, FVector3D& maximum);

	// Get the vertex position's index, which refers to a shared point in the model.
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the position index of the vertex. (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	int model_getVertexPointIndex(const Model& model, int partIndex, int polygonIndex, int vertexIndex);
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Sets the position index of the vertex to pointIndex. (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	void model_setVertexPointIndex(Model& model, int partIndex, int polygonIndex, int vertexIndex, int pointIndex);
	// Get the vertex position directly, without having to look it up by index using model_getPoint.
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the position of the vertex. (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	FVector3D model_getVertexPosition(const Model& model, int partIndex, int polygonIndex, int vertexIndex);
	// Get the vertex color, which is not shared with any other polygons. (Red, green, blue, alpha) channels are packed as (x, y, z, w) in FVector4D.
	// Vertex colors use a normalized scale from 0.0f to 1.0f.
	//   Transparent black is FVector4D(0.0f, 0.0f, 0.0f, 0.0f).
	//   Solid red is FVector4D(1.0f, 0.0f, 0.0f, 1.0f).
	//   Solid green is FVector4D(0.0f, 1.0f, 0.0f, 1.0f).
	//   Solid blue is FVector4D(0.0f, 0.0f, 1.0f, 1.0f).
	//   Half opaque orange is FVector4D(1.0f, 0.5f, 0.0f, 0.5f).
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the color of the vertex. (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	FVector4D model_getVertexColor(const Model& model, int partIndex, int polygonIndex, int vertexIndex);
	// Set the vertex color using the same system as model_getVertexColor.
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Sets the color of the vertex to color. (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	void model_setVertexColor(Model& model, int partIndex, int polygonIndex, int vertexIndex, const FVector4D& color);
	// Get (U1, V1, U2, V2) texture coordinates packed as (x, y, z, w) in FVector4D.
	// UV1 coordinates (x, y) refers to normalized texture sampling coordinates for the diffuse-map.
	// UV2 coordinates (z, w) refers to normalized texture sampling coordinates for the light-map.
	//   Light-maps do not use mip-map layers, which allow generating light-maps dynamically.
	// Pre-condition: model must refer to an existing model.
	FVector4D model_getTexCoord(const Model& model, int partIndex, int polygonIndex, int vertexIndex);
	// Pre-condition: model must refer to an existing model.
	// Side-effect: Sets the texture coordinates of the vertex to texCoord for both UV1 and UV2.
	//              (At vertexIndex in the polygon at polygonIndex in the part at partIndex in model.)
	void model_setTexCoord(Model& model, int partIndex, int polygonIndex, int vertexIndex, const FVector4D& texCoord);

	// Create a triangle surface at given position indices.
	//   The fourth vertex is used as padding, so quads and triangles take the same amount of memory per polygon.
	//   Using two triangles instead of one quad would use twice as much memory.
	// Pre-condition: model must refer to an existing model.
	// Side-effect:
	//   Adds a new polygon in the model's part at partIndex.
	//   The new polygon contains three vertices.
	//   Each new vertex has texture coordinates set to the upper left corner using (0.0f, 0.0f, 0.0f, 0.0f).
	//   Each new vertex has the color set to solid white using (1.0f, 1.0f, 1.0f, 1.0f).
	// Post-condition:
	//   Returns the new polygon's local index within the part at partIndex in model.
	int model_addTriangle(Model& model, int partIndex, int pointA, int pointB, int pointC);
	// Create a quad surface at given position indices.
	// Pre-condition: model must refer to an existing model.
	// Side-effect:
	//   Adds a new polygon in the model's part at partIndex.
	//   The new polygon contains four vertices.
	//   Each new vertex has texture coordinates set to the upper left corner using (0.0f, 0.0f, 0.0f, 0.0f).
	//   Each new vertex has the color set to solid white using (1.0f, 1.0f, 1.0f, 1.0f).
	// Post-condition:
	//   Returns the new polygon's local index within the part at partIndex in model.
	int model_addQuad(Model& model, int partIndex, int pointA, int pointB, int pointC, int pointD);
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the number of polygons (triangles + quads) in the part at partIndex in model.
	int model_getNumberOfPolygons(const Model& model, int partIndex);
	// Pre-condition: model must refer to an existing model.
	// Post-condition: Returns the number of vertices in the polygon at polygonIndex in the part at partIndex in model.
	int model_getPolygonVertexCount(const Model& model, int partIndex, int polygonIndex);

	// Get the part's diffuse texture.
	// Pre-condition: model must refer to an existing model.
	// Post-condition:
	//   Returns an image handle to the diffuse texture in the part at partIndex in model.
	//   If the part has no diffuse image then an empth handle is returned.
	ImageRgbaU8 model_getDiffuseMap(const Model& model, int partIndex);
	// Set the part's diffuse texture.
	//   A texture is just an image fulfilling the criterias of image_isTexture to allow fast texture sampling and pyramid generation.
	// Pre-condition:
	//   model must refer to an existing model.
	//   diffuseMap must be either empty or have power-of-two dimensions accepted by image_isTexture.
	// Side-effect:
	//   Sets the diffuse texture in the part at partIndex in model to diffuseMap.
	//   If diffuseMap is an empty image handle, then the diffuse texture will be replaced by the default solid white color.
	void model_setDiffuseMap(Model& model, int partIndex, const ImageRgbaU8 &diffuseMap);
	// Automatically find the diffuse texture by name in the resource pool and assign it.
	// Pre-condition:
	//   model must refer to an existing model.
	//   pool must refer to an existing resource pool.
	//   filename must be the image's filename without any extension nor path.
	//     "Car" is accepted.
	//     "Car.png" is rejected for having an extension.
	//     "myFolder/Car" is rejected for having a path.
	//     "myFolder\\Car" is rejected for having a path.
	//     "Car_1.2" is rejected for using a dot in the actual name, just to catch more mistakes with file extensions.
	// Side-effect:
	//   Sets the diffuse texture in the part at partIndex in model to the image looked up by filename in pool.
	void model_setDiffuseMapByName(Model& model, int partIndex, ResourcePool &pool, const String &filename);
	// Get the part's light texture.
	// Pre-condition:
	//   model must refer to an existing model.
	// Post-condition:
	//   Returns an image handle to the light texture in the part at partIndex in model.
	//   If the part has no light image then an empth handle is returned.
	ImageRgbaU8 model_getLightMap(Model& model, int partIndex);
	// Set the part's light texture.
	//   A texture is just an image fulfilling the criterias of image_isTexture to allow fast texture sampling.
	//   Even though no texture-pyramid is used for light-maps, it still has to look up
	//   pixels quickly using bit-shifts with base two logarithms of power of two widths.
	// Pre-condition:
	//   model must refer to an existing model.
	//   lightMap must be either empty or have power-of-two dimensions accepted by image_isTexture.
	// Side-effect:
	//   Sets the diffuse texture in the part at partIndex in model to diffuseMap.
	//   If diffuseMap is an empty image handle, then the diffuse texture will be replaced by the default solid white color.
	void model_setLightMap(Model& model, int partIndex, const ImageRgbaU8 &lightMap);
	// Automatically find the light texture by name in the resource pool and assign it.
	// Pre-condition:
	//   model must refer to an existing model.
	//   pool must refer to an existing resource pool.
	//   filename must be the image's filename without any extension nor path.
	// Side-effect:
	//   Sets the light texture in the part at partIndex in model to the image looked up by filename in pool.
	void model_setLightMapByName(Model& model, int partIndex, ResourcePool &pool, const String &filename);

	// In order to draw two adjacent polygons without any missing pixels along the seam, they must:
	//   * Share two position indices in opposite directions.
	//     (Rounding the same value to integers twice can be rounded differently,
	//     even though it's highly unlikely to actually happen.)
	//   * Have each vertex position inside of the camera's clipping frustum.
	//     (Far outside of the view frustum, triangles must be clipped in
	//     floating-point 3D space to prevent integer overflows when converted
	//     to sub-pixel integer coordinates.)
	//   * Avoid colliding with near or far clip planes.
	//     (This would also cause clipping in floating-point 3D space, because a
	//     location behind the camera cannot be represented as a screen coordinate)
	// If your clipped polygons are fully outside of the view-frustum,
	// then you will not see the seam nor the polygons.
	// To solve this:
	//   * use model_addPointIfNeeded instead of model_addPoint when adding points.
	//   * Split polygons that are way too big and use them to produce more details.
	//     (This will also increase precision for texture coordinates by splitting up seemingly infinite planes.)
	// If this does not hold true then there is either an exception missing
	// or a bug in the renderer, which should be reported as soon as possible.

	// Single-threaded rendering (Slow but easy to use for small tasks)
	//   Can be executed on different threads if targetImage and depthBuffer doesn't have overlapping memory lines between the threads
	// Pre-condition: colorBuffer and depthBuffer must have the same dimensions.
	// Side-effect: Render any model transformed by modelToWorldTransform, seen from camera, to any colorBuffer using any depthBuffer.
	//   An empty model handle will be skipped silently, which can be used instead of an model with zero polygons.
	void model_render(const Model& model, const Transform3D &modelToWorldTransform, ImageRgbaU8& colorBuffer, ImageF32& depthBuffer, const Camera &camera);
	// Simpler rendering without colorBuffer, for shadows and other depth effects
	//   Equivalent to model_render with a non-existing colorBuffer and filter forced to solid.
	//   Skip this call conditionally for filtered models (using model_getFilter) if you want full equivalence with model_render.
	// Side-effect: Render any model transformed by modelToWorldTransform, seen from camera, to any depthBuffer.
	//   An empty model handle will be skipped silently, which can be used instead of an model with zero polygons.
	void model_renderDepth(const Model& model, const Transform3D &modelToWorldTransform, ImageF32& depthBuffer, const Camera &camera);

	// Multi-threaded rendering (Huge performance boost with more CPU cores!)
	// Post-condition: Returns the handle to a new multi-threaded rendering context.
	//   It is basically a list of triangles to be drawn in parallel using a single call.
	//   After creating a renderer, you may execute a number of batches using it.
	//   Each batch may execute a number of tasks in parallel.
	//   Call pattern:
	//     renderer_create (renderer_begin renderer_giveTask* renderer_end)*
	Renderer renderer_create();
	// Post-condition: Returns true iff the renderer exists.
	bool renderer_exists(const Renderer& renderer);
	// Prepares for rendering by giving the target images to draw pixels on.
	// This step makes sure that nobody changes the target dimensions while rendering,
	// which could otherwise happen if someone requests a new canvas too often.
	// Pre-condition:
	//   renderer must refer to an existing renderer.
	//   colorBuffer and depthBuffer must have the same dimensions.
	void renderer_begin(Renderer& renderer, ImageRgbaU8& colorBuffer, ImageF32& depthBuffer);
	// Once an object passed game-specific occlusion tests, give it to the renderer using renderer_giveTask.
	// The render job will be performed during the next call to renderer_end.
	// Pre-condition: renderer must refer to an existing renderer.
	// An empty model handle will be skipped silently, which can be used instead of an model with zero polygons.
	// Side-effect: The visible triangles are queued up in the renderer.
	void renderer_giveTask(Renderer& renderer, const Model& model, const Transform3D &modelToWorldTransform, const Camera &camera);
	// Side-effect: Finishes all the jobs in the rendering context so that triangles are rasterized to the targets given to renderer_begin.
	// Pre-condition: renderer must refer to an existing renderer.
	void renderer_end(Renderer& renderer);

	// Imports a DMF model from file content.
	//   Use in combination with string_load or your own system for storing files.
	//   Example:
	//     Model crateModel = importFromContent_DMF1(string_load(mediaPath + U"Model_Crate.dmf"), pool);
	// Pre-condition:
	//   fileContent must be the content of a DMF 1.0 model file.
	//   pool must refer to an existing resource pool.
	//   0 <= detailLevel <= 2 (0 = low, 1 = medium, 2 = high)
	// Post-condition:
	//   Returns a handle to a model imported from fileContent, using pool to access resources, with parts not visible in detailLevel excluded.
	// How to import from the DMF1 format:
	//   * Only use M_Diffuse_0Tex, M_Diffuse_1Tex or M_Diffuse_2Tex as shaders.
	//       Place any diffuse texture in texture slot 0 and any lightmap in slot 1.
	//       Remove any textures that are not used by the shaders.
	//       The fixed pipeline only checks which textures are used.
	//   * Make sure that texture names are spelled case sensitive or they might not be found on some operating systems like Linux.
	// See renderer/model/format/dmf1.cpp for the implementation. (It does not exist in api/modelAPI.cpp)
	Model importFromContent_DMF1(const String &fileContent, ResourcePool &pool, int detailLevel = 2);

}

#endif