diff --git a/Performance Evaluation.bmp b/Performance Evaluation.bmp new file mode 100644 index 0000000..23a1192 Binary files /dev/null and b/Performance Evaluation.bmp differ diff --git a/Performance Evaluation.xlsx b/Performance Evaluation.xlsx new file mode 100644 index 0000000..1b01ddb Binary files /dev/null and b/Performance Evaluation.xlsx differ diff --git a/README.md b/README.md index ae6088d..57c0ccd 100644 --- a/README.md +++ b/README.md @@ -1,273 +1,251 @@ -CIS 565 Project3 : CUDA Pathtracer -=================== + + +CUDA Pathtracer +=============== Fall 2014 +========= + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_AllEffects.png) + +PROJECT DESCRIPTION +------------------- + +This is a GPU path tracing program, with following features: + +###1.Basic Features: +- Raycasting from a camera into a scene through a ray grid +- Diffuse surfaces +- Perfect specular reflective surfaces +- Cube intersection testing +- Sphere surface point sampling +- Stream compaction optimization + +#### Stream compaction +I use thrust to do stream compaction, that after each iteration in path trace, only valid rays are kept to pass on in next iteration. With stream compaction, the renderer won't run the empty thread any more and much time can be saved. To make comparison, I set blocksize as 64 and then record the renderer's run time for 100 iterations with/without stream compaction with different max trace depth and the plot shows the result:
+ +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/Performance%20Evaluation.bmp) + +We can see that the higher the max trace depth is, the more time stream compaction will save. + +###2.Extra Features: +- Super sample anti-alisasing +- Refraction, i.e. glass +- OBJ Mesh loading and rendering +- Motion blur +- Bump mapping +- Texture mapping +- Depth of field +- Interactive camera + + +####(1) Super sample anti-alisasing +- Reference: http://en.wikipedia.org/wiki/Supersampling + +- Overview write up and performance impact: + +I add super sample anti-alisasing, which makes my render result smoother. To do this, I just jitter the initial rays randomly in each iteration. And here is the comparison between the scene with SSAA and the scene without SSAA: + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/DetailWithSSAA.bmp) + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/DetailWithNoSSAA.bmp) + +- Accelerate the feature: NULL + +- Compare to a CPU version: + +In CPU version we may use grid algorithm to jitter the initial rays, but in GPU version we jitter the rays randomly. Their results are almost the same. + +- Further optimized: NULL + + +####(2) Refraction +- Reference: http://en.wikipedia.org/wiki/Fresnel_equations + +- Overview write up and performance impact: + +I add fresnel reflection and refraction. And it enables me to add transparent objects in my scene. To do this, I just use the fresnel equations to compute the reflective and refractive coefficients whenever the ray hits a refractive object, and get the reflect ray and refract ray. However, as my cuda path tracer works iteratively, I can just return one ray each time. So I generate a random number to decide which ray to return, based on the reflective and refractive coefficients. And here the upper left sphere is refractive: + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_WithRefraction.png) + +- Accelerate the feature: NULL + +- Compare to a CPU version: + +I think the main difference with CPU version is that I use a random number to decide which ray to pass on. But in CPU version(recursive), we pass both reflect ray and refract ray and add their results together. However, I think the result is the same. + +- Further optimized: NULL + + +####(3) OBJ Mesh loading and rendering +- Reference: http://www.cplusplus.com/forum/general/87738/ + +- Overview write up and performance impact: + +I add OBJ Mesh reader and render obj in my scene. To do this, I firstly learned the format of obj files and then wirte a obj reader by myself. And due to the different size of the objs I load into my scene, I scale all of the objs to the size of (1,1,1).(Maybe smaller, as the obj's length,width,height aren't always the same) After that, I load each triangle mesh as a new object in my scene to do path trace and thus the more meshes the obj file has, the slower the render will be. Here is the scene with an obj loaded(I just use a tetra obj as it has only four meshes and can be rendered faster): + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_WithOBJ.png) + +- Accelerate the feature: + +I add bounding box to the obj object to accelerate the ray intersect part. + +- Compare to a CPU version: NULL + +- Further optimized: + +If the objs are complex, I still need long time to render each frame even I add BB for them. So, I think maybe I should use more accelerate methods like kd-tree to make the renderer faster. + + +####(4) Motion blur +- Reference: http://www.cs.cmu.edu/afs/cs/academic/class/15462-s09/www/lec/13/lec13.pdf + +- Overview write up and performance impact: + +I add motion blur for objects. To do this, I add a new attribute for each object called MBV, and the object will move its position according to this velocity vector(this part is hard-coded in cudaRaytraceCore). And here the green sphere has the motion blur effect: + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_MotionBlur.png) + +- Accelerate the feature: NULL + +- Compare to a CPU version: NULL + +- Further optimized: + +The users can only set the velocity for each object, I think I can add more parameters and let users control how the object move in each frame. + + +####(5) Bump mapping +- Reference: + +http://www.paulsprojects.net/tutorials/simplebump/simplebump.html + +http://www.cs.unc.edu/~rademach/xroads-RT/RTarticle.html + +- Overview write up and performance impact: + +I try to add bump map for objects, but can only realize normal map now. To do this, I add a new attribute for each object called BUMP, and when renderer reads the scene file, it also stores the normal map's color to a buffer. When it does ray intersect, it returns the intersect normal according to the corresponding color in the buffer. And here is the scene with normal map(See the details on the sphere and floor): + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_NormalMap.png) + +- Accelerate the feature: NULL + +- Compare to a CPU version: NULL + +- Further optimized: + +Bump mapping is the combine of normal mapping and height mapping(which changes points position according to the map color). And I need to add height map, or get height map from the normal map to finish bump mapping. + + + +####(6) Texture mapping +- Reference: http://www.cs.unc.edu/~rademach/xroads-RT/RTarticle.html + +- Overview write up and performance impact: + +I add texture for cubes and spheres. To do this, I add a new attribute for each object called MAP, and when renderer reads the scene file, it also reads in the texture map's color to a buffer. When the program does path trace, it gets the color from this buffer according to the intersect point's position. And here is the scene with texture map(Much more colorful): + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_TextureMap.png) + +- Accelerate the feature: NULL + +- Compare to a CPU version: NULL + +- Further optimized: + +Right now the same texture map will be stored several times in the color buffer if it is assigned to different objects, as the map is an attribute of object. I think it will be better to give each texture map an index and read in the texrure map when a new texture index appears. In this way, much memory space can be saved. + + +####(7) Depth of field +- Reference: http://www.keithlantz.net/2013/03/path-tracer-depth-of-field/ + +- Overview write up and performance impact: + +I add depth of field in my scene. To do this, I add two new attributes for camera, DOFL and DOFR. DOFL decides the camera's distance to focal plane(on z axis) and DOFR decides the blurradius(The higher, the more blurred the scene will be). According to these two parameters, in each iteration before the path tracer begins, the camera's position and the initial rays' direction will be jittered. In this way, I get a scene picture with depth of field: + +![Alt text](https://github.com/wulinjiansheng/Project3-Pathtracer/blob/master/windows/Project3-Pathtracer/Project3-Pathtracer/Final%20Images/FinalScene_DepthOfField.png) + +- Accelerate the feature: NULL + +- Compare to a CPU version: NULL + +- Further optimized: + +Maybe I can add more focal length to let the camera focus on more focal planes. Although it is not correct physically, we may get some fantastic results. + + +####(8) Interactive camera +- Reference: NULL + +- Overview write up and performance impact: + +This one is the easiest one and I just defines more keys in keyCallback function. When the key is pressed, renderer will clear the screen and redo the path trace based on the new camera's position. See more detail in the video link. + +- Accelerate the feature: NULL + +- Compare to a CPU version: NULL + +- Further optimized: + +Using mouse to control the camera is more convenient, but to do this, I must speed my renderer up at first. + + +###3.Scene Format + +The scene format has changed due to the features I add.
+ +- Materials: Unchanged
+ +- Cameras: Add two new parameters
+DOFL(focal length) //The camera's distance to focal plane(on z axis)
+DOFR(blur radius) //The blur extent
+ +Example:
+CAMERA
+RES 800 800
+FOVY 25
+ITERATIONS 5000
+FILE test.bmp
+frame 0
+EYE 0 4.5 14
+VIEW 0 0 -1
+UP 0 1 0
+DOFL 14.0
+DOFR 0.7
+ + +- Objects: Add three new parameters
+MBV(Motion blur velocity) //The velocity the object has
+MAP(Texture map) //The object's texture map's path
+BUMP(Bump map) //The object's bump map's path
+ + +Example:
+OBJECT0
+cube
+material 0
+frame 0
+TRANS 0 0 0
+ROTAT 0 0 90
+SCALE .01 10 10
+MBV 0 0 0
+MAP texture/wood.jpg
+BUMP texture/bumpmap.jpg
+ + +###4.Scene Control +|Key | Function +|------|---------- +|Directional keys | `Move camera up/down/left/right` +|Z/C | `Zoom in/out` +|D | `Enable/Disable depth of field` +|M| `Enable/Disable motion blur` +|N| `Enable/Disable bump map` +|T| ` Enable/Disable texture map` +|Space| `Enable/Disable stream compact` +|Esc| `Exit renderer` + + +###5.Video Link
+http://youtu.be/AwxrfiRXsPQ -Due Wed, 10/8 (submit without penalty until Sun, 10/12) - -## INTRODUCTION -In this project, you will implement a CUDA based pathtracer capable of -generating pathtraced rendered images extremely quickly. Building a pathtracer can be viewed as a generalization of building a raytracer, so for those of you who have taken 460/560, the basic concept should not be very new to you. For those of you that have not taken -CIS460/560, raytracing is a technique for generating images by tracing rays of -light through pixels in an image plane out into a scene and following the way -the rays of light bounce and interact with objects in the scene. More -information can be found here: -http://en.wikipedia.org/wiki/Ray_tracing_(graphics). Pathtracing is a generalization of this technique by considering more than just the contribution of direct lighting to a surface. - -Since in this class we are concerned with working in generating actual images -and less so with mundane tasks like file I/O, this project includes basecode -for loading a scene description file format, described below, and various other -things that generally make up the render "harness" that takes care of -everything up to the rendering itself. The core renderer is left for you to -implement. Finally, note that while this basecode is meant to serve as a -strong starting point for a CUDA pathtracer, you are not required to use this -basecode if you wish, and you may also change any part of the basecode -specification as you please, so long as the final rendered result is correct. - -## CONTENTS -The Project3 root directory contains the following subdirectories: - -* src/ contains the source code for the project. Both the Windows Visual Studio - solution and the OSX and Linux makefiles reference this folder for all - source; the base source code compiles on Linux, OSX and Windows without - modification. If you are building on OSX, be sure to uncomment lines 4 & 5 of - the CMakeLists.txt in order to make sure CMake builds against clang. -* data/scenes/ contains an example scene description file. -* renders/ contains an example render of the given example scene file. -* windows/ contains a Windows Visual Studio 2010 project and all dependencies - needed for building and running on Windows 7. If you would like to create a - Visual Studio 2012 or 2013 projects, there are static libraries that you can - use for GLFW that are in external/bin/GLFW (Visual Studio 2012 uses msvc110, - and Visual Studio 2013 uses msvc120) -* external/ contains all the header, static libraries and built binaries for - 3rd party libraries (i.e. glm, GLEW, GLFW) that we use for windowing and OpenGL - extensions - -## RUNNING THE CODE -The main function requires a scene description file (that is provided in data/scenes). -The main function reads in the scene file by an argument as such : -'scene=[sceneFileName]' - -If you are using Visual Studio, you can set this in the Debugging > Command Arguments section -in the Project properties. - -## REQUIREMENTS -In this project, you are given code for: - -* Loading, reading, and storing the scene scene description format -* Example functions that can run on both the CPU and GPU for generating random - numbers, spherical intersection testing, and surface point sampling on cubes -* A class for handling image operations and saving images -* Working code for CUDA-GL interop - -You will need to implement the following features: - -* Raycasting from a camera into a scene through a pixel grid -* Diffuse surfaces -* Perfect specular reflective surfaces -* Cube intersection testing -* Sphere surface point sampling -* Stream compaction optimization - -You are also required to implement at least 2 of the following features: - -* Texture mapping -* Bump mapping -* Depth of field -* Refraction, i.e. glass -* OBJ Mesh loading and rendering -* Interactive camera -* Motion blur -* Subsurface scattering - -The 'extra features' list is not comprehensive. If you have a particular feature -you would like to implement (e.g. acceleration structures, etc.) please contact us -first! - -For each 'extra feature' you must provide the following analysis : -* overview write up of the feature -* performance impact of the feature -* if you did something to accelerate the feature, why did you do what you did -* compare your GPU version to a CPU version of this feature (you do NOT need to - implement a CPU version) -* how can this feature be further optimized (again, not necessary to implement it, but - should give a roadmap of how to further optimize and why you believe this is the next - step) - -## BASE CODE TOUR -You will be working in three files: raytraceKernel.cu, intersections.h, and -interactions.h. Within these files, areas that you need to complete are marked -with a TODO comment. Areas that are useful to and serve as hints for optional -features are marked with TODO (Optional). Functions that are useful for -reference are marked with the comment LOOK. - -* raytraceKernel.cu contains the core raytracing CUDA kernel. You will need to - complete: - * cudaRaytraceCore() handles kernel launches and memory management; this - function already contains example code for launching kernels, - transferring geometry and cameras from the host to the device, and transferring - image buffers from the host to the device and back. You will have to complete - this function to support passing materials and lights to CUDA. - * raycastFromCameraKernel() is a function that you need to implement. This - function once correctly implemented should handle camera raycasting. - * raytraceRay() is the core raytracing CUDA kernel; all of your pathtracing - logic should be implemented in this CUDA kernel. raytraceRay() should - take in a camera, image buffer, geometry, materials, and lights, and should - trace a ray through the scene and write the resultant color to a pixel in the - image buffer. - -* intersections.h contains functions for geometry intersection testing and - point generation. You will need to complete: - * boxIntersectionTest(), which takes in a box and a ray and performs an - intersection test. This function should work in the same way as - sphereIntersectionTest(). - * getRandomPointOnSphere(), which takes in a sphere and returns a random - point on the surface of the sphere with an even probability distribution. - This function should work in the same way as getRandomPointOnCube(). You can - (although do not necessarily have to) use this to generate points on a sphere - to use a point lights, or can use this for area lighting. - -* interactions.h contains functions for ray-object interactions that define how - rays behave upon hitting materials and objects. You will need to complete: - * getRandomDirectionInSphere(), which generates a random direction in a - sphere with a uniform probability. This function works in a fashion - similar to that of calculateRandomDirectionInHemisphere(), which generates a - random cosine-weighted direction in a hemisphere. - * calculateBSDF(), which takes in an incoming ray, normal, material, and - other information, and returns an outgoing ray. You can either implement - this function for ray-surface interactions, or you can replace it with your own - function(s). - -You will also want to familiarize yourself with: - -* sceneStructs.h, which contains definitions for how geometry, materials, - lights, cameras, and animation frames are stored in the renderer. -* utilities.h, which serves as a kitchen-sink of useful functions - -## NOTES ON GLM -This project uses GLM, the GL Math library, for linear algebra. You need to -know two important points on how GLM is used in this project: - -* In this project, indices in GLM vectors (such as vec3, vec4), are accessed - via swizzling. So, instead of v[0], v.x is used, and instead of v[1], v.y is - used, and so on and so forth. -* GLM Matrix operations work fine on NVIDIA Fermi cards and later, but - pre-Fermi cards do not play nice with GLM matrices. As such, in this project, - GLM matrices are replaced with a custom matrix struct, called a cudaMat4, found - in cudaMat4.h. A custom function for multiplying glm::vec4s and cudaMat4s is - provided as multiplyMV() in intersections.h. - -## SCENE FORMAT -This project uses a custom scene description format. -Scene files are flat text files that describe all geometry, materials, -lights, cameras, render settings, and animation frames inside of the scene. -Items in the format are delimited by new lines, and comments can be added at -the end of each line preceded with a double-slash. - -Materials are defined in the following fashion: - -* MATERIAL (material ID) //material header -* RGB (float r) (float g) (float b) //diffuse color -* SPECX (float specx) //specular exponent -* SPECRGB (float r) (float g) (float b) //specular color -* REFL (bool refl) //reflectivity flag, 0 for - no, 1 for yes -* REFR (bool refr) //refractivity flag, 0 for - no, 1 for yes -* REFRIOR (float ior) //index of refraction - for Fresnel effects -* SCATTER (float scatter) //scatter flag, 0 for - no, 1 for yes -* ABSCOEFF (float r) (float b) (float g) //absorption - coefficient for scattering -* RSCTCOEFF (float rsctcoeff) //reduced scattering - coefficient -* EMITTANCE (float emittance) //the emittance of the - material. Anything >0 makes the material a light source. - -Cameras are defined in the following fashion: - -* CAMERA //camera header -* RES (float x) (float y) //resolution -* FOVY (float fovy) //vertical field of - view half-angle. the horizonal angle is calculated from this and the - reslution -* ITERATIONS (float interations) //how many - iterations to refine the image, only relevant for supersampled antialiasing, - depth of field, area lights, and other distributed raytracing applications -* FILE (string filename) //file to output - render to upon completion -* frame (frame number) //start of a frame -* EYE (float x) (float y) (float z) //camera's position in - worldspace -* VIEW (float x) (float y) (float z) //camera's view - direction -* UP (float x) (float y) (float z) //camera's up vector - -Objects are defined in the following fashion: -* OBJECT (object ID) //object header -* (cube OR sphere OR mesh) //type of object, can - be either "cube", "sphere", or "mesh". Note that cubes and spheres are unit - sized and centered at the origin. -* material (material ID) //material to - assign this object -* frame (frame number) //start of a frame -* TRANS (float transx) (float transy) (float transz) //translation -* ROTAT (float rotationx) (float rotationy) (float rotationz) //rotation -* SCALE (float scalex) (float scaley) (float scalez) //scale - -An example scene file setting up two frames inside of a Cornell Box can be -found in the scenes/ directory. - -For meshes, note that the base code will only read in .obj files. For more -information on the .obj specification see http://en.wikipedia.org/wiki/Wavefront_.obj_file. - -An example of a mesh object is as follows: - -OBJECT 0 -mesh tetra.obj -material 0 -frame 0 -TRANS 0 5 -5 -ROTAT 0 90 0 -SCALE .01 10 10 - -Check the Google group for some sample .obj files of varying complexity. - -## THIRD PARTY CODE POLICY -* Use of any third-party code must be approved by asking on our Google Group. - If it is approved, all students are welcome to use it. Generally, we approve - use of third-party code that is not a core part of the project. For example, - for the ray tracer, we would approve using a third-party library for loading - models, but would not approve copying and pasting a CUDA function for doing - refraction. -* Third-party code must be credited in README.md. -* Using third-party code without its approval, including using another - student's code, is an academic integrity violation, and will result in you - receiving an F for the semester. - -## SELF-GRADING -* On the submission date, email your grade, on a scale of 0 to 100, to Harmony, - harmoli+cis565@seas.upenn.com, with a one paragraph explanation. Be concise and - realistic. Recall that we reserve 30 points as a sanity check to adjust your - grade. Your actual grade will be (0.7 * your grade) + (0.3 * our grade). We - hope to only use this in extreme cases when your grade does not realistically - reflect your work - it is either too high or too low. In most cases, we plan - to give you the exact grade you suggest. -* Projects are not weighted evenly, e.g., Project 0 doesn't count as much as - the path tracer. We will determine the weighting at the end of the semester - based on the size of each project. - -## SUBMISSION -Please change the README to reflect the answers to the questions we have posed -above. Remember: -* this is a renderer, so include images that you've made! -* be sure to back your claims for optimization with numbers and comparisons -* if you reference any other material, please provide a link to it -* you wil not e graded on how fast your path tracer runs, but getting close to - real-time is always nice -* if you have a fast GPU renderer, it is good to show case this with a video to - show interactivity. If you do so, please include a link. - -Be sure to open a pull request and to send Harmony your grade and why you -believe this is the grade you should get. diff --git a/external/lib/win/FreeImage/FreeImage.dll b/external/lib/win/FreeImage/FreeImage.dll new file mode 100644 index 0000000..3ce0a70 Binary files /dev/null and b/external/lib/win/FreeImage/FreeImage.dll differ diff --git a/external/lib/win/FreeImage/FreeImage.lib b/external/lib/win/FreeImage/FreeImage.lib new file mode 100644 index 0000000..b29ed71 Binary files /dev/null and b/external/lib/win/FreeImage/FreeImage.lib differ diff --git a/src/FreeImage.h b/src/FreeImage.h new file mode 100644 index 0000000..c281424 --- /dev/null +++ b/src/FreeImage.h @@ -0,0 +1,1104 @@ +// ========================================================== +// FreeImage 3 +// +// Design and implementation by +// - Floris van den Berg (flvdberg@wxs.nl) +// - Herv?Drolon (drolon@infonie.fr) +// +// Contributors: +// - see changes log named 'Whatsnew.txt', see header of each .h and .cpp file +// +// This file is part of FreeImage 3 +// +// COVERED CODE IS PROVIDED UNDER THIS LICENSE ON AN "AS IS" BASIS, WITHOUT WARRANTY +// OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES +// THAT THE COVERED CODE IS FREE OF DEFECTS, MERCHANTABLE, FIT FOR A PARTICULAR PURPOSE +// OR NON-INFRINGING. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE COVERED +// CODE IS WITH YOU. SHOULD ANY COVERED CODE PROVE DEFECTIVE IN ANY RESPECT, YOU (NOT +// THE INITIAL DEVELOPER OR ANY OTHER CONTRIBUTOR) ASSUME THE COST OF ANY NECESSARY +// SERVICING, REPAIR OR CORRECTION. THIS DISCLAIMER OF WARRANTY CONSTITUTES AN ESSENTIAL +// PART OF THIS LICENSE. NO USE OF ANY COVERED CODE IS AUTHORIZED HEREUNDER EXCEPT UNDER +// THIS DISCLAIMER. +// +// Use at your own risk! +// ========================================================== + +#ifndef FREEIMAGE_H +#define FREEIMAGE_H + +// Version information ------------------------------------------------------ + +#define FREEIMAGE_MAJOR_VERSION 3 +#define FREEIMAGE_MINOR_VERSION 15 +#define FREEIMAGE_RELEASE_SERIAL 4 + +// Compiler options --------------------------------------------------------- + +#include // needed for UNICODE functions + +#if defined(FREEIMAGE_LIB) + #define DLL_API + #define DLL_CALLCONV +#else + #if defined(_WIN32) || defined(__WIN32__) + #define DLL_CALLCONV __stdcall + // The following ifdef block is the standard way of creating macros which make exporting + // from a DLL simpler. All files within this DLL are compiled with the FREEIMAGE_EXPORTS + // symbol defined on the command line. this symbol should not be defined on any project + // that uses this DLL. This way any other project whose source files include this file see + // DLL_API functions as being imported from a DLL, wheras this DLL sees symbols + // defined with this macro as being exported. + #ifdef FREEIMAGE_EXPORTS + #define DLL_API __declspec(dllexport) + #else + #define DLL_API __declspec(dllimport) + #endif // FREEIMAGE_EXPORTS + #else + // try the gcc visibility support (see http://gcc.gnu.org/wiki/Visibility) + #if defined(__GNUC__) && ((__GNUC__ >= 4) || (__GNUC__ == 3 && __GNUC_MINOR__ >= 4)) + #ifndef GCC_HASCLASSVISIBILITY + #define GCC_HASCLASSVISIBILITY + #endif + #endif // __GNUC__ + #define DLL_CALLCONV + #if defined(GCC_HASCLASSVISIBILITY) + #define DLL_API __attribute__ ((visibility("default"))) + #else + #define DLL_API + #endif + #endif // WIN32 / !WIN32 +#endif // FREEIMAGE_LIB + +// Some versions of gcc may have BYTE_ORDER or __BYTE_ORDER defined +// If your big endian system isn't being detected, add an OS specific check +#if (defined(BYTE_ORDER) && BYTE_ORDER==BIG_ENDIAN) || \ + (defined(__BYTE_ORDER) && __BYTE_ORDER==__BIG_ENDIAN) || \ + defined(__BIG_ENDIAN__) +#define FREEIMAGE_BIGENDIAN +#endif // BYTE_ORDER + +// This really only affects 24 and 32 bit formats, the rest are always RGB order. +#define FREEIMAGE_COLORORDER_BGR 0 +#define FREEIMAGE_COLORORDER_RGB 1 +#if defined(FREEIMAGE_BIGENDIAN) +#define FREEIMAGE_COLORORDER FREEIMAGE_COLORORDER_RGB +#else +#define FREEIMAGE_COLORORDER FREEIMAGE_COLORORDER_BGR +#endif + +// Ensure 4-byte enums if we're using Borland C++ compilers +#if defined(__BORLANDC__) +#pragma option push -b +#endif + +// For C compatibility -------------------------------------------------------- + +#ifdef __cplusplus +#define FI_DEFAULT(x) = x +#define FI_ENUM(x) enum x +#define FI_STRUCT(x) struct x +#else +#define FI_DEFAULT(x) +#define FI_ENUM(x) typedef int x; enum x +#define FI_STRUCT(x) typedef struct x x; struct x +#endif + +// Bitmap types ------------------------------------------------------------- + +FI_STRUCT (FIBITMAP) { void *data; }; +FI_STRUCT (FIMULTIBITMAP) { void *data; }; + +// Types used in the library (directly copied from Windows) ----------------- + +#if defined(__MINGW32__) && defined(_WINDOWS_H) +#define _WINDOWS_ // prevent a bug in MinGW32 +#endif // __MINGW32__ + +#ifndef _WINDOWS_ +#define _WINDOWS_ + +#ifndef FALSE +#define FALSE 0 +#endif +#ifndef TRUE +#define TRUE 1 +#endif +#ifndef NULL +#define NULL 0 +#endif + +#ifndef SEEK_SET +#define SEEK_SET 0 +#define SEEK_CUR 1 +#define SEEK_END 2 +#endif + +#ifndef _MSC_VER +// define portable types for 32-bit / 64-bit OS +#include +typedef int32_t BOOL; +typedef uint8_t BYTE; +typedef uint16_t WORD; +typedef uint32_t DWORD; +typedef int32_t LONG; +typedef int64_t INT64; +typedef uint64_t UINT64; +#else +// MS is not C99 ISO compliant +typedef long BOOL; +typedef unsigned char BYTE; +typedef unsigned short WORD; +typedef unsigned long DWORD; +typedef long LONG; +typedef signed __int64 INT64; +typedef unsigned __int64 UINT64; +#endif // _MSC_VER + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(push, 1) +#else +#pragma pack(1) +#endif // WIN32 + +typedef struct tagRGBQUAD { +#if FREEIMAGE_COLORORDER == FREEIMAGE_COLORORDER_BGR + BYTE rgbBlue; + BYTE rgbGreen; + BYTE rgbRed; +#else + BYTE rgbRed; + BYTE rgbGreen; + BYTE rgbBlue; +#endif // FREEIMAGE_COLORORDER + BYTE rgbReserved; +} RGBQUAD; + +typedef struct tagRGBTRIPLE { +#if FREEIMAGE_COLORORDER == FREEIMAGE_COLORORDER_BGR + BYTE rgbtBlue; + BYTE rgbtGreen; + BYTE rgbtRed; +#else + BYTE rgbtRed; + BYTE rgbtGreen; + BYTE rgbtBlue; +#endif // FREEIMAGE_COLORORDER +} RGBTRIPLE; + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(pop) +#else +#pragma pack() +#endif // WIN32 + +typedef struct tagBITMAPINFOHEADER{ + DWORD biSize; + LONG biWidth; + LONG biHeight; + WORD biPlanes; + WORD biBitCount; + DWORD biCompression; + DWORD biSizeImage; + LONG biXPelsPerMeter; + LONG biYPelsPerMeter; + DWORD biClrUsed; + DWORD biClrImportant; +} BITMAPINFOHEADER, *PBITMAPINFOHEADER; + +typedef struct tagBITMAPINFO { + BITMAPINFOHEADER bmiHeader; + RGBQUAD bmiColors[1]; +} BITMAPINFO, *PBITMAPINFO; + +#endif // _WINDOWS_ + +// Types used in the library (specific to FreeImage) ------------------------ + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(push, 1) +#else +#pragma pack(1) +#endif // WIN32 + +/** 48-bit RGB +*/ +typedef struct tagFIRGB16 { + WORD red; + WORD green; + WORD blue; +} FIRGB16; + +/** 64-bit RGBA +*/ +typedef struct tagFIRGBA16 { + WORD red; + WORD green; + WORD blue; + WORD alpha; +} FIRGBA16; + +/** 96-bit RGB Float +*/ +typedef struct tagFIRGBF { + float red; + float green; + float blue; +} FIRGBF; + +/** 128-bit RGBA Float +*/ +typedef struct tagFIRGBAF { + float red; + float green; + float blue; + float alpha; +} FIRGBAF; + +/** Data structure for COMPLEX type (complex number) +*/ +typedef struct tagFICOMPLEX { + /// real part + double r; + /// imaginary part + double i; +} FICOMPLEX; + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(pop) +#else +#pragma pack() +#endif // WIN32 + +// Indexes for byte arrays, masks and shifts for treating pixels as words --- +// These coincide with the order of RGBQUAD and RGBTRIPLE ------------------- + +#ifndef FREEIMAGE_BIGENDIAN +#if FREEIMAGE_COLORORDER == FREEIMAGE_COLORORDER_BGR +// Little Endian (x86 / MS Windows, Linux) : BGR(A) order +#define FI_RGBA_RED 2 +#define FI_RGBA_GREEN 1 +#define FI_RGBA_BLUE 0 +#define FI_RGBA_ALPHA 3 +#define FI_RGBA_RED_MASK 0x00FF0000 +#define FI_RGBA_GREEN_MASK 0x0000FF00 +#define FI_RGBA_BLUE_MASK 0x000000FF +#define FI_RGBA_ALPHA_MASK 0xFF000000 +#define FI_RGBA_RED_SHIFT 16 +#define FI_RGBA_GREEN_SHIFT 8 +#define FI_RGBA_BLUE_SHIFT 0 +#define FI_RGBA_ALPHA_SHIFT 24 +#else +// Little Endian (x86 / MaxOSX) : RGB(A) order +#define FI_RGBA_RED 0 +#define FI_RGBA_GREEN 1 +#define FI_RGBA_BLUE 2 +#define FI_RGBA_ALPHA 3 +#define FI_RGBA_RED_MASK 0x000000FF +#define FI_RGBA_GREEN_MASK 0x0000FF00 +#define FI_RGBA_BLUE_MASK 0x00FF0000 +#define FI_RGBA_ALPHA_MASK 0xFF000000 +#define FI_RGBA_RED_SHIFT 0 +#define FI_RGBA_GREEN_SHIFT 8 +#define FI_RGBA_BLUE_SHIFT 16 +#define FI_RGBA_ALPHA_SHIFT 24 +#endif // FREEIMAGE_COLORORDER +#else +#if FREEIMAGE_COLORORDER == FREEIMAGE_COLORORDER_BGR +// Big Endian (PPC / none) : BGR(A) order +#define FI_RGBA_RED 2 +#define FI_RGBA_GREEN 1 +#define FI_RGBA_BLUE 0 +#define FI_RGBA_ALPHA 3 +#define FI_RGBA_RED_MASK 0x0000FF00 +#define FI_RGBA_GREEN_MASK 0x00FF0000 +#define FI_RGBA_BLUE_MASK 0xFF000000 +#define FI_RGBA_ALPHA_MASK 0x000000FF +#define FI_RGBA_RED_SHIFT 8 +#define FI_RGBA_GREEN_SHIFT 16 +#define FI_RGBA_BLUE_SHIFT 24 +#define FI_RGBA_ALPHA_SHIFT 0 +#else +// Big Endian (PPC / Linux, MaxOSX) : RGB(A) order +#define FI_RGBA_RED 0 +#define FI_RGBA_GREEN 1 +#define FI_RGBA_BLUE 2 +#define FI_RGBA_ALPHA 3 +#define FI_RGBA_RED_MASK 0xFF000000 +#define FI_RGBA_GREEN_MASK 0x00FF0000 +#define FI_RGBA_BLUE_MASK 0x0000FF00 +#define FI_RGBA_ALPHA_MASK 0x000000FF +#define FI_RGBA_RED_SHIFT 24 +#define FI_RGBA_GREEN_SHIFT 16 +#define FI_RGBA_BLUE_SHIFT 8 +#define FI_RGBA_ALPHA_SHIFT 0 +#endif // FREEIMAGE_COLORORDER +#endif // FREEIMAGE_BIGENDIAN + +#define FI_RGBA_RGB_MASK (FI_RGBA_RED_MASK|FI_RGBA_GREEN_MASK|FI_RGBA_BLUE_MASK) + +// The 16bit macros only include masks and shifts, since each color element is not byte aligned + +#define FI16_555_RED_MASK 0x7C00 +#define FI16_555_GREEN_MASK 0x03E0 +#define FI16_555_BLUE_MASK 0x001F +#define FI16_555_RED_SHIFT 10 +#define FI16_555_GREEN_SHIFT 5 +#define FI16_555_BLUE_SHIFT 0 +#define FI16_565_RED_MASK 0xF800 +#define FI16_565_GREEN_MASK 0x07E0 +#define FI16_565_BLUE_MASK 0x001F +#define FI16_565_RED_SHIFT 11 +#define FI16_565_GREEN_SHIFT 5 +#define FI16_565_BLUE_SHIFT 0 + +// ICC profile support ------------------------------------------------------ + +#define FIICC_DEFAULT 0x00 +#define FIICC_COLOR_IS_CMYK 0x01 + +FI_STRUCT (FIICCPROFILE) { + WORD flags; // info flag + DWORD size; // profile's size measured in bytes + void *data; // points to a block of contiguous memory containing the profile +}; + +// Important enums ---------------------------------------------------------- + +/** I/O image format identifiers. +*/ +FI_ENUM(FREE_IMAGE_FORMAT) { + FIF_UNKNOWN = -1, + FIF_BMP = 0, + FIF_ICO = 1, + FIF_JPEG = 2, + FIF_JNG = 3, + FIF_KOALA = 4, + FIF_LBM = 5, + FIF_IFF = FIF_LBM, + FIF_MNG = 6, + FIF_PBM = 7, + FIF_PBMRAW = 8, + FIF_PCD = 9, + FIF_PCX = 10, + FIF_PGM = 11, + FIF_PGMRAW = 12, + FIF_PNG = 13, + FIF_PPM = 14, + FIF_PPMRAW = 15, + FIF_RAS = 16, + FIF_TARGA = 17, + FIF_TIFF = 18, + FIF_WBMP = 19, + FIF_PSD = 20, + FIF_CUT = 21, + FIF_XBM = 22, + FIF_XPM = 23, + FIF_DDS = 24, + FIF_GIF = 25, + FIF_HDR = 26, + FIF_FAXG3 = 27, + FIF_SGI = 28, + FIF_EXR = 29, + FIF_J2K = 30, + FIF_JP2 = 31, + FIF_PFM = 32, + FIF_PICT = 33, + FIF_RAW = 34 +}; + +/** Image type used in FreeImage. +*/ +FI_ENUM(FREE_IMAGE_TYPE) { + FIT_UNKNOWN = 0, // unknown type + FIT_BITMAP = 1, // standard image : 1-, 4-, 8-, 16-, 24-, 32-bit + FIT_UINT16 = 2, // array of unsigned short : unsigned 16-bit + FIT_INT16 = 3, // array of short : signed 16-bit + FIT_UINT32 = 4, // array of unsigned long : unsigned 32-bit + FIT_INT32 = 5, // array of long : signed 32-bit + FIT_FLOAT = 6, // array of float : 32-bit IEEE floating point + FIT_DOUBLE = 7, // array of double : 64-bit IEEE floating point + FIT_COMPLEX = 8, // array of FICOMPLEX : 2 x 64-bit IEEE floating point + FIT_RGB16 = 9, // 48-bit RGB image : 3 x 16-bit + FIT_RGBA16 = 10, // 64-bit RGBA image : 4 x 16-bit + FIT_RGBF = 11, // 96-bit RGB float image : 3 x 32-bit IEEE floating point + FIT_RGBAF = 12 // 128-bit RGBA float image : 4 x 32-bit IEEE floating point +}; + +/** Image color type used in FreeImage. +*/ +FI_ENUM(FREE_IMAGE_COLOR_TYPE) { + FIC_MINISWHITE = 0, // min value is white + FIC_MINISBLACK = 1, // min value is black + FIC_RGB = 2, // RGB color model + FIC_PALETTE = 3, // color map indexed + FIC_RGBALPHA = 4, // RGB color model with alpha channel + FIC_CMYK = 5 // CMYK color model +}; + +/** Color quantization algorithms. +Constants used in FreeImage_ColorQuantize. +*/ +FI_ENUM(FREE_IMAGE_QUANTIZE) { + FIQ_WUQUANT = 0, // Xiaolin Wu color quantization algorithm + FIQ_NNQUANT = 1 // NeuQuant neural-net quantization algorithm by Anthony Dekker +}; + +/** Dithering algorithms. +Constants used in FreeImage_Dither. +*/ +FI_ENUM(FREE_IMAGE_DITHER) { + FID_FS = 0, // Floyd & Steinberg error diffusion + FID_BAYER4x4 = 1, // Bayer ordered dispersed dot dithering (order 2 dithering matrix) + FID_BAYER8x8 = 2, // Bayer ordered dispersed dot dithering (order 3 dithering matrix) + FID_CLUSTER6x6 = 3, // Ordered clustered dot dithering (order 3 - 6x6 matrix) + FID_CLUSTER8x8 = 4, // Ordered clustered dot dithering (order 4 - 8x8 matrix) + FID_CLUSTER16x16= 5, // Ordered clustered dot dithering (order 8 - 16x16 matrix) + FID_BAYER16x16 = 6 // Bayer ordered dispersed dot dithering (order 4 dithering matrix) +}; + +/** Lossless JPEG transformations +Constants used in FreeImage_JPEGTransform +*/ +FI_ENUM(FREE_IMAGE_JPEG_OPERATION) { + FIJPEG_OP_NONE = 0, // no transformation + FIJPEG_OP_FLIP_H = 1, // horizontal flip + FIJPEG_OP_FLIP_V = 2, // vertical flip + FIJPEG_OP_TRANSPOSE = 3, // transpose across UL-to-LR axis + FIJPEG_OP_TRANSVERSE = 4, // transpose across UR-to-LL axis + FIJPEG_OP_ROTATE_90 = 5, // 90-degree clockwise rotation + FIJPEG_OP_ROTATE_180 = 6, // 180-degree rotation + FIJPEG_OP_ROTATE_270 = 7 // 270-degree clockwise (or 90 ccw) +}; + +/** Tone mapping operators. +Constants used in FreeImage_ToneMapping. +*/ +FI_ENUM(FREE_IMAGE_TMO) { + FITMO_DRAGO03 = 0, // Adaptive logarithmic mapping (F. Drago, 2003) + FITMO_REINHARD05 = 1, // Dynamic range reduction inspired by photoreceptor physiology (E. Reinhard, 2005) + FITMO_FATTAL02 = 2 // Gradient domain high dynamic range compression (R. Fattal, 2002) +}; + +/** Upsampling / downsampling filters. +Constants used in FreeImage_Rescale. +*/ +FI_ENUM(FREE_IMAGE_FILTER) { + FILTER_BOX = 0, // Box, pulse, Fourier window, 1st order (constant) b-spline + FILTER_BICUBIC = 1, // Mitchell & Netravali's two-param cubic filter + FILTER_BILINEAR = 2, // Bilinear filter + FILTER_BSPLINE = 3, // 4th order (cubic) b-spline + FILTER_CATMULLROM = 4, // Catmull-Rom spline, Overhauser spline + FILTER_LANCZOS3 = 5 // Lanczos3 filter +}; + +/** Color channels. +Constants used in color manipulation routines. +*/ +FI_ENUM(FREE_IMAGE_COLOR_CHANNEL) { + FICC_RGB = 0, // Use red, green and blue channels + FICC_RED = 1, // Use red channel + FICC_GREEN = 2, // Use green channel + FICC_BLUE = 3, // Use blue channel + FICC_ALPHA = 4, // Use alpha channel + FICC_BLACK = 5, // Use black channel + FICC_REAL = 6, // Complex images: use real part + FICC_IMAG = 7, // Complex images: use imaginary part + FICC_MAG = 8, // Complex images: use magnitude + FICC_PHASE = 9 // Complex images: use phase +}; + +// Metadata support --------------------------------------------------------- + +/** + Tag data type information (based on TIFF specifications) + + Note: RATIONALs are the ratio of two 32-bit integer values. +*/ +FI_ENUM(FREE_IMAGE_MDTYPE) { + FIDT_NOTYPE = 0, // placeholder + FIDT_BYTE = 1, // 8-bit unsigned integer + FIDT_ASCII = 2, // 8-bit bytes w/ last byte null + FIDT_SHORT = 3, // 16-bit unsigned integer + FIDT_LONG = 4, // 32-bit unsigned integer + FIDT_RATIONAL = 5, // 64-bit unsigned fraction + FIDT_SBYTE = 6, // 8-bit signed integer + FIDT_UNDEFINED = 7, // 8-bit untyped data + FIDT_SSHORT = 8, // 16-bit signed integer + FIDT_SLONG = 9, // 32-bit signed integer + FIDT_SRATIONAL = 10, // 64-bit signed fraction + FIDT_FLOAT = 11, // 32-bit IEEE floating point + FIDT_DOUBLE = 12, // 64-bit IEEE floating point + FIDT_IFD = 13, // 32-bit unsigned integer (offset) + FIDT_PALETTE = 14, // 32-bit RGBQUAD + FIDT_LONG8 = 16, // 64-bit unsigned integer + FIDT_SLONG8 = 17, // 64-bit signed integer + FIDT_IFD8 = 18 // 64-bit unsigned integer (offset) +}; + +/** + Metadata models supported by FreeImage +*/ +FI_ENUM(FREE_IMAGE_MDMODEL) { + FIMD_NODATA = -1, + FIMD_COMMENTS = 0, // single comment or keywords + FIMD_EXIF_MAIN = 1, // Exif-TIFF metadata + FIMD_EXIF_EXIF = 2, // Exif-specific metadata + FIMD_EXIF_GPS = 3, // Exif GPS metadata + FIMD_EXIF_MAKERNOTE = 4, // Exif maker note metadata + FIMD_EXIF_INTEROP = 5, // Exif interoperability metadata + FIMD_IPTC = 6, // IPTC/NAA metadata + FIMD_XMP = 7, // Abobe XMP metadata + FIMD_GEOTIFF = 8, // GeoTIFF metadata + FIMD_ANIMATION = 9, // Animation metadata + FIMD_CUSTOM = 10, // Used to attach other metadata types to a dib + FIMD_EXIF_RAW = 11 // Exif metadata as a raw buffer +}; + +/** + Handle to a metadata model +*/ +FI_STRUCT (FIMETADATA) { void *data; }; + +/** + Handle to a FreeImage tag +*/ +FI_STRUCT (FITAG) { void *data; }; + +// File IO routines --------------------------------------------------------- + +#ifndef FREEIMAGE_IO +#define FREEIMAGE_IO + +typedef void* fi_handle; +typedef unsigned (DLL_CALLCONV *FI_ReadProc) (void *buffer, unsigned size, unsigned count, fi_handle handle); +typedef unsigned (DLL_CALLCONV *FI_WriteProc) (void *buffer, unsigned size, unsigned count, fi_handle handle); +typedef int (DLL_CALLCONV *FI_SeekProc) (fi_handle handle, long offset, int origin); +typedef long (DLL_CALLCONV *FI_TellProc) (fi_handle handle); + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(push, 1) +#else +#pragma pack(1) +#endif // WIN32 + +FI_STRUCT(FreeImageIO) { + FI_ReadProc read_proc; // pointer to the function used to read data + FI_WriteProc write_proc; // pointer to the function used to write data + FI_SeekProc seek_proc; // pointer to the function used to seek + FI_TellProc tell_proc; // pointer to the function used to aquire the current position +}; + +#if (defined(_WIN32) || defined(__WIN32__)) +#pragma pack(pop) +#else +#pragma pack() +#endif // WIN32 + +/** +Handle to a memory I/O stream +*/ +FI_STRUCT (FIMEMORY) { void *data; }; + +#endif // FREEIMAGE_IO + +// Plugin routines ---------------------------------------------------------- + +#ifndef PLUGINS +#define PLUGINS + +typedef const char *(DLL_CALLCONV *FI_FormatProc)(void); +typedef const char *(DLL_CALLCONV *FI_DescriptionProc)(void); +typedef const char *(DLL_CALLCONV *FI_ExtensionListProc)(void); +typedef const char *(DLL_CALLCONV *FI_RegExprProc)(void); +typedef void *(DLL_CALLCONV *FI_OpenProc)(FreeImageIO *io, fi_handle handle, BOOL read); +typedef void (DLL_CALLCONV *FI_CloseProc)(FreeImageIO *io, fi_handle handle, void *data); +typedef int (DLL_CALLCONV *FI_PageCountProc)(FreeImageIO *io, fi_handle handle, void *data); +typedef int (DLL_CALLCONV *FI_PageCapabilityProc)(FreeImageIO *io, fi_handle handle, void *data); +typedef FIBITMAP *(DLL_CALLCONV *FI_LoadProc)(FreeImageIO *io, fi_handle handle, int page, int flags, void *data); +typedef BOOL (DLL_CALLCONV *FI_SaveProc)(FreeImageIO *io, FIBITMAP *dib, fi_handle handle, int page, int flags, void *data); +typedef BOOL (DLL_CALLCONV *FI_ValidateProc)(FreeImageIO *io, fi_handle handle); +typedef const char *(DLL_CALLCONV *FI_MimeProc)(void); +typedef BOOL (DLL_CALLCONV *FI_SupportsExportBPPProc)(int bpp); +typedef BOOL (DLL_CALLCONV *FI_SupportsExportTypeProc)(FREE_IMAGE_TYPE type); +typedef BOOL (DLL_CALLCONV *FI_SupportsICCProfilesProc)(void); +typedef BOOL (DLL_CALLCONV *FI_SupportsNoPixelsProc)(void); + +FI_STRUCT (Plugin) { + FI_FormatProc format_proc; + FI_DescriptionProc description_proc; + FI_ExtensionListProc extension_proc; + FI_RegExprProc regexpr_proc; + FI_OpenProc open_proc; + FI_CloseProc close_proc; + FI_PageCountProc pagecount_proc; + FI_PageCapabilityProc pagecapability_proc; + FI_LoadProc load_proc; + FI_SaveProc save_proc; + FI_ValidateProc validate_proc; + FI_MimeProc mime_proc; + FI_SupportsExportBPPProc supports_export_bpp_proc; + FI_SupportsExportTypeProc supports_export_type_proc; + FI_SupportsICCProfilesProc supports_icc_profiles_proc; + FI_SupportsNoPixelsProc supports_no_pixels_proc; +}; + +typedef void (DLL_CALLCONV *FI_InitProc)(Plugin *plugin, int format_id); + +#endif // PLUGINS + + +// Load / Save flag constants ----------------------------------------------- + +#define FIF_LOAD_NOPIXELS 0x8000 // loading: load the image header only (not supported by all plugins, default to full loading) + +#define BMP_DEFAULT 0 +#define BMP_SAVE_RLE 1 +#define CUT_DEFAULT 0 +#define DDS_DEFAULT 0 +#define EXR_DEFAULT 0 // save data as half with piz-based wavelet compression +#define EXR_FLOAT 0x0001 // save data as float instead of as half (not recommended) +#define EXR_NONE 0x0002 // save with no compression +#define EXR_ZIP 0x0004 // save with zlib compression, in blocks of 16 scan lines +#define EXR_PIZ 0x0008 // save with piz-based wavelet compression +#define EXR_PXR24 0x0010 // save with lossy 24-bit float compression +#define EXR_B44 0x0020 // save with lossy 44% float compression - goes to 22% when combined with EXR_LC +#define EXR_LC 0x0040 // save images with one luminance and two chroma channels, rather than as RGB (lossy compression) +#define FAXG3_DEFAULT 0 +#define GIF_DEFAULT 0 +#define GIF_LOAD256 1 // Load the image as a 256 color image with ununsed palette entries, if it's 16 or 2 color +#define GIF_PLAYBACK 2 // 'Play' the GIF to generate each frame (as 32bpp) instead of returning raw frame data when loading +#define HDR_DEFAULT 0 +#define ICO_DEFAULT 0 +#define ICO_MAKEALPHA 1 // convert to 32bpp and create an alpha channel from the AND-mask when loading +#define IFF_DEFAULT 0 +#define J2K_DEFAULT 0 // save with a 16:1 rate +#define JP2_DEFAULT 0 // save with a 16:1 rate +#define JPEG_DEFAULT 0 // loading (see JPEG_FAST); saving (see JPEG_QUALITYGOOD|JPEG_SUBSAMPLING_420) +#define JPEG_FAST 0x0001 // load the file as fast as possible, sacrificing some quality +#define JPEG_ACCURATE 0x0002 // load the file with the best quality, sacrificing some speed +#define JPEG_CMYK 0x0004 // load separated CMYK "as is" (use | to combine with other load flags) +#define JPEG_EXIFROTATE 0x0008 // load and rotate according to Exif 'Orientation' tag if available +#define JPEG_GREYSCALE 0x0010 // load and convert to a 8-bit greyscale image +#define JPEG_QUALITYSUPERB 0x80 // save with superb quality (100:1) +#define JPEG_QUALITYGOOD 0x0100 // save with good quality (75:1) +#define JPEG_QUALITYNORMAL 0x0200 // save with normal quality (50:1) +#define JPEG_QUALITYAVERAGE 0x0400 // save with average quality (25:1) +#define JPEG_QUALITYBAD 0x0800 // save with bad quality (10:1) +#define JPEG_PROGRESSIVE 0x2000 // save as a progressive-JPEG (use | to combine with other save flags) +#define JPEG_SUBSAMPLING_411 0x1000 // save with high 4x1 chroma subsampling (4:1:1) +#define JPEG_SUBSAMPLING_420 0x4000 // save with medium 2x2 medium chroma subsampling (4:2:0) - default value +#define JPEG_SUBSAMPLING_422 0x8000 // save with low 2x1 chroma subsampling (4:2:2) +#define JPEG_SUBSAMPLING_444 0x10000 // save with no chroma subsampling (4:4:4) +#define JPEG_OPTIMIZE 0x20000 // on saving, compute optimal Huffman coding tables (can reduce a few percent of file size) +#define JPEG_BASELINE 0x40000 // save basic JPEG, without metadata or any markers +#define KOALA_DEFAULT 0 +#define LBM_DEFAULT 0 +#define MNG_DEFAULT 0 +#define PCD_DEFAULT 0 +#define PCD_BASE 1 // load the bitmap sized 768 x 512 +#define PCD_BASEDIV4 2 // load the bitmap sized 384 x 256 +#define PCD_BASEDIV16 3 // load the bitmap sized 192 x 128 +#define PCX_DEFAULT 0 +#define PFM_DEFAULT 0 +#define PICT_DEFAULT 0 +#define PNG_DEFAULT 0 +#define PNG_IGNOREGAMMA 1 // loading: avoid gamma correction +#define PNG_Z_BEST_SPEED 0x0001 // save using ZLib level 1 compression flag (default value is 6) +#define PNG_Z_DEFAULT_COMPRESSION 0x0006 // save using ZLib level 6 compression flag (default recommended value) +#define PNG_Z_BEST_COMPRESSION 0x0009 // save using ZLib level 9 compression flag (default value is 6) +#define PNG_Z_NO_COMPRESSION 0x0100 // save without ZLib compression +#define PNG_INTERLACED 0x0200 // save using Adam7 interlacing (use | to combine with other save flags) +#define PNM_DEFAULT 0 +#define PNM_SAVE_RAW 0 // If set the writer saves in RAW format (i.e. P4, P5 or P6) +#define PNM_SAVE_ASCII 1 // If set the writer saves in ASCII format (i.e. P1, P2 or P3) +#define PSD_DEFAULT 0 +#define PSD_CMYK 1 // reads tags for separated CMYK (default is conversion to RGB) +#define PSD_LAB 2 // reads tags for CIELab (default is conversion to RGB) +#define RAS_DEFAULT 0 +#define RAW_DEFAULT 0 // load the file as linear RGB 48-bit +#define RAW_PREVIEW 1 // try to load the embedded JPEG preview with included Exif Data or default to RGB 24-bit +#define RAW_DISPLAY 2 // load the file as RGB 24-bit +#define RAW_HALFSIZE 4 // output a half-size color image +#define SGI_DEFAULT 0 +#define TARGA_DEFAULT 0 +#define TARGA_LOAD_RGB888 1 // If set the loader converts RGB555 and ARGB8888 -> RGB888. +#define TARGA_SAVE_RLE 2 // If set, the writer saves with RLE compression +#define TIFF_DEFAULT 0 +#define TIFF_CMYK 0x0001 // reads/stores tags for separated CMYK (use | to combine with compression flags) +#define TIFF_PACKBITS 0x0100 // save using PACKBITS compression +#define TIFF_DEFLATE 0x0200 // save using DEFLATE compression (a.k.a. ZLIB compression) +#define TIFF_ADOBE_DEFLATE 0x0400 // save using ADOBE DEFLATE compression +#define TIFF_NONE 0x0800 // save without any compression +#define TIFF_CCITTFAX3 0x1000 // save using CCITT Group 3 fax encoding +#define TIFF_CCITTFAX4 0x2000 // save using CCITT Group 4 fax encoding +#define TIFF_LZW 0x4000 // save using LZW compression +#define TIFF_JPEG 0x8000 // save using JPEG compression +#define TIFF_LOGLUV 0x10000 // save using LogLuv compression +#define WBMP_DEFAULT 0 +#define XBM_DEFAULT 0 +#define XPM_DEFAULT 0 + +// Background filling options --------------------------------------------------------- +// Constants used in FreeImage_FillBackground and FreeImage_EnlargeCanvas + +#define FI_COLOR_IS_RGB_COLOR 0x00 // RGBQUAD color is a RGB color (contains no valid alpha channel) +#define FI_COLOR_IS_RGBA_COLOR 0x01 // RGBQUAD color is a RGBA color (contains a valid alpha channel) +#define FI_COLOR_FIND_EQUAL_COLOR 0x02 // For palettized images: lookup equal RGB color from palette +#define FI_COLOR_ALPHA_IS_INDEX 0x04 // The color's rgbReserved member (alpha) contains the palette index to be used +#define FI_COLOR_PALETTE_SEARCH_MASK (FI_COLOR_FIND_EQUAL_COLOR | FI_COLOR_ALPHA_IS_INDEX) // No color lookup is performed + + +#ifdef __cplusplus +extern "C" { +#endif + +// Init / Error routines ---------------------------------------------------- + +DLL_API void DLL_CALLCONV FreeImage_Initialise(BOOL load_local_plugins_only FI_DEFAULT(FALSE)); +DLL_API void DLL_CALLCONV FreeImage_DeInitialise(void); + +// Version routines --------------------------------------------------------- + +DLL_API const char *DLL_CALLCONV FreeImage_GetVersion(void); +DLL_API const char *DLL_CALLCONV FreeImage_GetCopyrightMessage(void); + +// Message output functions ------------------------------------------------- + +typedef void (*FreeImage_OutputMessageFunction)(FREE_IMAGE_FORMAT fif, const char *msg); +typedef void (DLL_CALLCONV *FreeImage_OutputMessageFunctionStdCall)(FREE_IMAGE_FORMAT fif, const char *msg); + +DLL_API void DLL_CALLCONV FreeImage_SetOutputMessageStdCall(FreeImage_OutputMessageFunctionStdCall omf); +DLL_API void DLL_CALLCONV FreeImage_SetOutputMessage(FreeImage_OutputMessageFunction omf); +DLL_API void DLL_CALLCONV FreeImage_OutputMessageProc(int fif, const char *fmt, ...); + +// Allocate / Clone / Unload routines --------------------------------------- + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Allocate(int width, int height, int bpp, unsigned red_mask FI_DEFAULT(0), unsigned green_mask FI_DEFAULT(0), unsigned blue_mask FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_AllocateT(FREE_IMAGE_TYPE type, int width, int height, int bpp FI_DEFAULT(8), unsigned red_mask FI_DEFAULT(0), unsigned green_mask FI_DEFAULT(0), unsigned blue_mask FI_DEFAULT(0)); +DLL_API FIBITMAP * DLL_CALLCONV FreeImage_Clone(FIBITMAP *dib); +DLL_API void DLL_CALLCONV FreeImage_Unload(FIBITMAP *dib); + +// Header loading routines +DLL_API BOOL DLL_CALLCONV FreeImage_HasPixels(FIBITMAP *dib); + +// Load / Save routines ----------------------------------------------------- + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Load(FREE_IMAGE_FORMAT fif, const char *filename, int flags FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_LoadU(FREE_IMAGE_FORMAT fif, const wchar_t *filename, int flags FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_LoadFromHandle(FREE_IMAGE_FORMAT fif, FreeImageIO *io, fi_handle handle, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_Save(FREE_IMAGE_FORMAT fif, FIBITMAP *dib, const char *filename, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_SaveU(FREE_IMAGE_FORMAT fif, FIBITMAP *dib, const wchar_t *filename, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_SaveToHandle(FREE_IMAGE_FORMAT fif, FIBITMAP *dib, FreeImageIO *io, fi_handle handle, int flags FI_DEFAULT(0)); + +// Memory I/O stream routines ----------------------------------------------- + +DLL_API FIMEMORY *DLL_CALLCONV FreeImage_OpenMemory(BYTE *data FI_DEFAULT(0), DWORD size_in_bytes FI_DEFAULT(0)); +DLL_API void DLL_CALLCONV FreeImage_CloseMemory(FIMEMORY *stream); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_LoadFromMemory(FREE_IMAGE_FORMAT fif, FIMEMORY *stream, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_SaveToMemory(FREE_IMAGE_FORMAT fif, FIBITMAP *dib, FIMEMORY *stream, int flags FI_DEFAULT(0)); +DLL_API long DLL_CALLCONV FreeImage_TellMemory(FIMEMORY *stream); +DLL_API BOOL DLL_CALLCONV FreeImage_SeekMemory(FIMEMORY *stream, long offset, int origin); +DLL_API BOOL DLL_CALLCONV FreeImage_AcquireMemory(FIMEMORY *stream, BYTE **data, DWORD *size_in_bytes); +DLL_API unsigned DLL_CALLCONV FreeImage_ReadMemory(void *buffer, unsigned size, unsigned count, FIMEMORY *stream); +DLL_API unsigned DLL_CALLCONV FreeImage_WriteMemory(const void *buffer, unsigned size, unsigned count, FIMEMORY *stream); + +DLL_API FIMULTIBITMAP *DLL_CALLCONV FreeImage_LoadMultiBitmapFromMemory(FREE_IMAGE_FORMAT fif, FIMEMORY *stream, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_SaveMultiBitmapToMemory(FREE_IMAGE_FORMAT fif, FIMULTIBITMAP *bitmap, FIMEMORY *stream, int flags); + +// Plugin Interface --------------------------------------------------------- + +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_RegisterLocalPlugin(FI_InitProc proc_address, const char *format FI_DEFAULT(0), const char *description FI_DEFAULT(0), const char *extension FI_DEFAULT(0), const char *regexpr FI_DEFAULT(0)); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_RegisterExternalPlugin(const char *path, const char *format FI_DEFAULT(0), const char *description FI_DEFAULT(0), const char *extension FI_DEFAULT(0), const char *regexpr FI_DEFAULT(0)); +DLL_API int DLL_CALLCONV FreeImage_GetFIFCount(void); +DLL_API int DLL_CALLCONV FreeImage_SetPluginEnabled(FREE_IMAGE_FORMAT fif, BOOL enable); +DLL_API int DLL_CALLCONV FreeImage_IsPluginEnabled(FREE_IMAGE_FORMAT fif); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFIFFromFormat(const char *format); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFIFFromMime(const char *mime); +DLL_API const char *DLL_CALLCONV FreeImage_GetFormatFromFIF(FREE_IMAGE_FORMAT fif); +DLL_API const char *DLL_CALLCONV FreeImage_GetFIFExtensionList(FREE_IMAGE_FORMAT fif); +DLL_API const char *DLL_CALLCONV FreeImage_GetFIFDescription(FREE_IMAGE_FORMAT fif); +DLL_API const char *DLL_CALLCONV FreeImage_GetFIFRegExpr(FREE_IMAGE_FORMAT fif); +DLL_API const char *DLL_CALLCONV FreeImage_GetFIFMimeType(FREE_IMAGE_FORMAT fif); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFIFFromFilename(const char *filename); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFIFFromFilenameU(const wchar_t *filename); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsReading(FREE_IMAGE_FORMAT fif); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsWriting(FREE_IMAGE_FORMAT fif); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsExportBPP(FREE_IMAGE_FORMAT fif, int bpp); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsExportType(FREE_IMAGE_FORMAT fif, FREE_IMAGE_TYPE type); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsICCProfiles(FREE_IMAGE_FORMAT fif); +DLL_API BOOL DLL_CALLCONV FreeImage_FIFSupportsNoPixels(FREE_IMAGE_FORMAT fif); + +// Multipaging interface ---------------------------------------------------- + +DLL_API FIMULTIBITMAP * DLL_CALLCONV FreeImage_OpenMultiBitmap(FREE_IMAGE_FORMAT fif, const char *filename, BOOL create_new, BOOL read_only, BOOL keep_cache_in_memory FI_DEFAULT(FALSE), int flags FI_DEFAULT(0)); +DLL_API FIMULTIBITMAP * DLL_CALLCONV FreeImage_OpenMultiBitmapFromHandle(FREE_IMAGE_FORMAT fif, FreeImageIO *io, fi_handle handle, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_SaveMultiBitmapToHandle(FREE_IMAGE_FORMAT fif, FIMULTIBITMAP *bitmap, FreeImageIO *io, fi_handle handle, int flags FI_DEFAULT(0)); +DLL_API BOOL DLL_CALLCONV FreeImage_CloseMultiBitmap(FIMULTIBITMAP *bitmap, int flags FI_DEFAULT(0)); +DLL_API int DLL_CALLCONV FreeImage_GetPageCount(FIMULTIBITMAP *bitmap); +DLL_API void DLL_CALLCONV FreeImage_AppendPage(FIMULTIBITMAP *bitmap, FIBITMAP *data); +DLL_API void DLL_CALLCONV FreeImage_InsertPage(FIMULTIBITMAP *bitmap, int page, FIBITMAP *data); +DLL_API void DLL_CALLCONV FreeImage_DeletePage(FIMULTIBITMAP *bitmap, int page); +DLL_API FIBITMAP * DLL_CALLCONV FreeImage_LockPage(FIMULTIBITMAP *bitmap, int page); +DLL_API void DLL_CALLCONV FreeImage_UnlockPage(FIMULTIBITMAP *bitmap, FIBITMAP *data, BOOL changed); +DLL_API BOOL DLL_CALLCONV FreeImage_MovePage(FIMULTIBITMAP *bitmap, int target, int source); +DLL_API BOOL DLL_CALLCONV FreeImage_GetLockedPageNumbers(FIMULTIBITMAP *bitmap, int *pages, int *count); + +// Filetype request routines ------------------------------------------------ + +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFileType(const char *filename, int size FI_DEFAULT(0)); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFileTypeU(const wchar_t *filename, int size FI_DEFAULT(0)); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFileTypeFromHandle(FreeImageIO *io, fi_handle handle, int size FI_DEFAULT(0)); +DLL_API FREE_IMAGE_FORMAT DLL_CALLCONV FreeImage_GetFileTypeFromMemory(FIMEMORY *stream, int size FI_DEFAULT(0)); + +// Image type request routine ----------------------------------------------- + +DLL_API FREE_IMAGE_TYPE DLL_CALLCONV FreeImage_GetImageType(FIBITMAP *dib); + +// FreeImage helper routines ------------------------------------------------ + +DLL_API BOOL DLL_CALLCONV FreeImage_IsLittleEndian(void); +DLL_API BOOL DLL_CALLCONV FreeImage_LookupX11Color(const char *szColor, BYTE *nRed, BYTE *nGreen, BYTE *nBlue); +DLL_API BOOL DLL_CALLCONV FreeImage_LookupSVGColor(const char *szColor, BYTE *nRed, BYTE *nGreen, BYTE *nBlue); + +// Pixel access routines ---------------------------------------------------- + +DLL_API BYTE *DLL_CALLCONV FreeImage_GetBits(FIBITMAP *dib); +DLL_API BYTE *DLL_CALLCONV FreeImage_GetScanLine(FIBITMAP *dib, int scanline); + +DLL_API BOOL DLL_CALLCONV FreeImage_GetPixelIndex(FIBITMAP *dib, unsigned x, unsigned y, BYTE *value); +DLL_API BOOL DLL_CALLCONV FreeImage_GetPixelColor(FIBITMAP *dib, unsigned x, unsigned y, RGBQUAD *value); +DLL_API BOOL DLL_CALLCONV FreeImage_SetPixelIndex(FIBITMAP *dib, unsigned x, unsigned y, BYTE *value); +DLL_API BOOL DLL_CALLCONV FreeImage_SetPixelColor(FIBITMAP *dib, unsigned x, unsigned y, RGBQUAD *value); + +// DIB info routines -------------------------------------------------------- + +DLL_API unsigned DLL_CALLCONV FreeImage_GetColorsUsed(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetBPP(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetWidth(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetHeight(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetLine(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetPitch(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetDIBSize(FIBITMAP *dib); +DLL_API RGBQUAD *DLL_CALLCONV FreeImage_GetPalette(FIBITMAP *dib); + +DLL_API unsigned DLL_CALLCONV FreeImage_GetDotsPerMeterX(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetDotsPerMeterY(FIBITMAP *dib); +DLL_API void DLL_CALLCONV FreeImage_SetDotsPerMeterX(FIBITMAP *dib, unsigned res); +DLL_API void DLL_CALLCONV FreeImage_SetDotsPerMeterY(FIBITMAP *dib, unsigned res); + +DLL_API BITMAPINFOHEADER *DLL_CALLCONV FreeImage_GetInfoHeader(FIBITMAP *dib); +DLL_API BITMAPINFO *DLL_CALLCONV FreeImage_GetInfo(FIBITMAP *dib); +DLL_API FREE_IMAGE_COLOR_TYPE DLL_CALLCONV FreeImage_GetColorType(FIBITMAP *dib); + +DLL_API unsigned DLL_CALLCONV FreeImage_GetRedMask(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetGreenMask(FIBITMAP *dib); +DLL_API unsigned DLL_CALLCONV FreeImage_GetBlueMask(FIBITMAP *dib); + +DLL_API unsigned DLL_CALLCONV FreeImage_GetTransparencyCount(FIBITMAP *dib); +DLL_API BYTE * DLL_CALLCONV FreeImage_GetTransparencyTable(FIBITMAP *dib); +DLL_API void DLL_CALLCONV FreeImage_SetTransparent(FIBITMAP *dib, BOOL enabled); +DLL_API void DLL_CALLCONV FreeImage_SetTransparencyTable(FIBITMAP *dib, BYTE *table, int count); +DLL_API BOOL DLL_CALLCONV FreeImage_IsTransparent(FIBITMAP *dib); +DLL_API void DLL_CALLCONV FreeImage_SetTransparentIndex(FIBITMAP *dib, int index); +DLL_API int DLL_CALLCONV FreeImage_GetTransparentIndex(FIBITMAP *dib); + +DLL_API BOOL DLL_CALLCONV FreeImage_HasBackgroundColor(FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_GetBackgroundColor(FIBITMAP *dib, RGBQUAD *bkcolor); +DLL_API BOOL DLL_CALLCONV FreeImage_SetBackgroundColor(FIBITMAP *dib, RGBQUAD *bkcolor); + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_GetThumbnail(FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_SetThumbnail(FIBITMAP *dib, FIBITMAP *thumbnail); + +// ICC profile routines ----------------------------------------------------- + +DLL_API FIICCPROFILE *DLL_CALLCONV FreeImage_GetICCProfile(FIBITMAP *dib); +DLL_API FIICCPROFILE *DLL_CALLCONV FreeImage_CreateICCProfile(FIBITMAP *dib, void *data, long size); +DLL_API void DLL_CALLCONV FreeImage_DestroyICCProfile(FIBITMAP *dib); + +// Line conversion routines ------------------------------------------------- + +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To4(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine8To4(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To4_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To4_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine24To4(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine32To4(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To8(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine4To8(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To8_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To8_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine24To8(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine32To8(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To16_555(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine4To16_555(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine8To16_555(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16_565_To16_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine24To16_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine32To16_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To16_565(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine4To16_565(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine8To16_565(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16_555_To16_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine24To16_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine32To16_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To24(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine4To24(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine8To24(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To24_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To24_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine32To24(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine1To32(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine4To32(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine8To32(BYTE *target, BYTE *source, int width_in_pixels, RGBQUAD *palette); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To32_555(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine16To32_565(BYTE *target, BYTE *source, int width_in_pixels); +DLL_API void DLL_CALLCONV FreeImage_ConvertLine24To32(BYTE *target, BYTE *source, int width_in_pixels); + +// Smart conversion routines ------------------------------------------------ + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo4Bits(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo8Bits(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToGreyscale(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo16Bits555(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo16Bits565(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo24Bits(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertTo32Bits(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ColorQuantize(FIBITMAP *dib, FREE_IMAGE_QUANTIZE quantize); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ColorQuantizeEx(FIBITMAP *dib, FREE_IMAGE_QUANTIZE quantize FI_DEFAULT(FIQ_WUQUANT), int PaletteSize FI_DEFAULT(256), int ReserveSize FI_DEFAULT(0), RGBQUAD *ReservePalette FI_DEFAULT(NULL)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Threshold(FIBITMAP *dib, BYTE T); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Dither(FIBITMAP *dib, FREE_IMAGE_DITHER algorithm); + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertFromRawBits(BYTE *bits, int width, int height, int pitch, unsigned bpp, unsigned red_mask, unsigned green_mask, unsigned blue_mask, BOOL topdown FI_DEFAULT(FALSE)); +DLL_API void DLL_CALLCONV FreeImage_ConvertToRawBits(BYTE *bits, FIBITMAP *dib, int pitch, unsigned bpp, unsigned red_mask, unsigned green_mask, unsigned blue_mask, BOOL topdown FI_DEFAULT(FALSE)); + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToFloat(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToRGBF(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToUINT16(FIBITMAP *dib); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToRGB16(FIBITMAP *dib); + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToStandardType(FIBITMAP *src, BOOL scale_linear FI_DEFAULT(TRUE)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ConvertToType(FIBITMAP *src, FREE_IMAGE_TYPE dst_type, BOOL scale_linear FI_DEFAULT(TRUE)); + +// tone mapping operators +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_ToneMapping(FIBITMAP *dib, FREE_IMAGE_TMO tmo, double first_param FI_DEFAULT(0), double second_param FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_TmoDrago03(FIBITMAP *src, double gamma FI_DEFAULT(2.2), double exposure FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_TmoReinhard05(FIBITMAP *src, double intensity FI_DEFAULT(0), double contrast FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_TmoReinhard05Ex(FIBITMAP *src, double intensity FI_DEFAULT(0), double contrast FI_DEFAULT(0), double adaptation FI_DEFAULT(1), double color_correction FI_DEFAULT(0)); + +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_TmoFattal02(FIBITMAP *src, double color_saturation FI_DEFAULT(0.5), double attenuation FI_DEFAULT(0.85)); + +// ZLib interface ----------------------------------------------------------- + +DLL_API DWORD DLL_CALLCONV FreeImage_ZLibCompress(BYTE *target, DWORD target_size, BYTE *source, DWORD source_size); +DLL_API DWORD DLL_CALLCONV FreeImage_ZLibUncompress(BYTE *target, DWORD target_size, BYTE *source, DWORD source_size); +DLL_API DWORD DLL_CALLCONV FreeImage_ZLibGZip(BYTE *target, DWORD target_size, BYTE *source, DWORD source_size); +DLL_API DWORD DLL_CALLCONV FreeImage_ZLibGUnzip(BYTE *target, DWORD target_size, BYTE *source, DWORD source_size); +DLL_API DWORD DLL_CALLCONV FreeImage_ZLibCRC32(DWORD crc, BYTE *source, DWORD source_size); + +// -------------------------------------------------------------------------- +// Metadata routines -------------------------------------------------------- +// -------------------------------------------------------------------------- + +// tag creation / destruction +DLL_API FITAG *DLL_CALLCONV FreeImage_CreateTag(void); +DLL_API void DLL_CALLCONV FreeImage_DeleteTag(FITAG *tag); +DLL_API FITAG *DLL_CALLCONV FreeImage_CloneTag(FITAG *tag); + +// tag getters and setters +DLL_API const char *DLL_CALLCONV FreeImage_GetTagKey(FITAG *tag); +DLL_API const char *DLL_CALLCONV FreeImage_GetTagDescription(FITAG *tag); +DLL_API WORD DLL_CALLCONV FreeImage_GetTagID(FITAG *tag); +DLL_API FREE_IMAGE_MDTYPE DLL_CALLCONV FreeImage_GetTagType(FITAG *tag); +DLL_API DWORD DLL_CALLCONV FreeImage_GetTagCount(FITAG *tag); +DLL_API DWORD DLL_CALLCONV FreeImage_GetTagLength(FITAG *tag); +DLL_API const void *DLL_CALLCONV FreeImage_GetTagValue(FITAG *tag); + +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagKey(FITAG *tag, const char *key); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagDescription(FITAG *tag, const char *description); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagID(FITAG *tag, WORD id); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagType(FITAG *tag, FREE_IMAGE_MDTYPE type); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagCount(FITAG *tag, DWORD count); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagLength(FITAG *tag, DWORD length); +DLL_API BOOL DLL_CALLCONV FreeImage_SetTagValue(FITAG *tag, const void *value); + +// iterator +DLL_API FIMETADATA *DLL_CALLCONV FreeImage_FindFirstMetadata(FREE_IMAGE_MDMODEL model, FIBITMAP *dib, FITAG **tag); +DLL_API BOOL DLL_CALLCONV FreeImage_FindNextMetadata(FIMETADATA *mdhandle, FITAG **tag); +DLL_API void DLL_CALLCONV FreeImage_FindCloseMetadata(FIMETADATA *mdhandle); + +// metadata setter and getter +DLL_API BOOL DLL_CALLCONV FreeImage_SetMetadata(FREE_IMAGE_MDMODEL model, FIBITMAP *dib, const char *key, FITAG *tag); +DLL_API BOOL DLL_CALLCONV FreeImage_GetMetadata(FREE_IMAGE_MDMODEL model, FIBITMAP *dib, const char *key, FITAG **tag); + +// helpers +DLL_API unsigned DLL_CALLCONV FreeImage_GetMetadataCount(FREE_IMAGE_MDMODEL model, FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_CloneMetadata(FIBITMAP *dst, FIBITMAP *src); + +// tag to C string conversion +DLL_API const char* DLL_CALLCONV FreeImage_TagToString(FREE_IMAGE_MDMODEL model, FITAG *tag, char *Make FI_DEFAULT(NULL)); + +// -------------------------------------------------------------------------- +// Image manipulation toolkit ----------------------------------------------- +// -------------------------------------------------------------------------- + +// rotation and flipping +/// @deprecated see FreeImage_Rotate +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_RotateClassic(FIBITMAP *dib, double angle); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Rotate(FIBITMAP *dib, double angle, const void *bkcolor FI_DEFAULT(NULL)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_RotateEx(FIBITMAP *dib, double angle, double x_shift, double y_shift, double x_origin, double y_origin, BOOL use_mask); +DLL_API BOOL DLL_CALLCONV FreeImage_FlipHorizontal(FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_FlipVertical(FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_JPEGTransform(const char *src_file, const char *dst_file, FREE_IMAGE_JPEG_OPERATION operation, BOOL perfect FI_DEFAULT(FALSE)); +DLL_API BOOL DLL_CALLCONV FreeImage_JPEGTransformU(const wchar_t *src_file, const wchar_t *dst_file, FREE_IMAGE_JPEG_OPERATION operation, BOOL perfect FI_DEFAULT(FALSE)); + +// upsampling / downsampling +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Rescale(FIBITMAP *dib, int dst_width, int dst_height, FREE_IMAGE_FILTER filter); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_MakeThumbnail(FIBITMAP *dib, int max_pixel_size, BOOL convert FI_DEFAULT(TRUE)); + +// color manipulation routines (point operations) +DLL_API BOOL DLL_CALLCONV FreeImage_AdjustCurve(FIBITMAP *dib, BYTE *LUT, FREE_IMAGE_COLOR_CHANNEL channel); +DLL_API BOOL DLL_CALLCONV FreeImage_AdjustGamma(FIBITMAP *dib, double gamma); +DLL_API BOOL DLL_CALLCONV FreeImage_AdjustBrightness(FIBITMAP *dib, double percentage); +DLL_API BOOL DLL_CALLCONV FreeImage_AdjustContrast(FIBITMAP *dib, double percentage); +DLL_API BOOL DLL_CALLCONV FreeImage_Invert(FIBITMAP *dib); +DLL_API BOOL DLL_CALLCONV FreeImage_GetHistogram(FIBITMAP *dib, DWORD *histo, FREE_IMAGE_COLOR_CHANNEL channel FI_DEFAULT(FICC_BLACK)); +DLL_API int DLL_CALLCONV FreeImage_GetAdjustColorsLookupTable(BYTE *LUT, double brightness, double contrast, double gamma, BOOL invert); +DLL_API BOOL DLL_CALLCONV FreeImage_AdjustColors(FIBITMAP *dib, double brightness, double contrast, double gamma, BOOL invert FI_DEFAULT(FALSE)); +DLL_API unsigned DLL_CALLCONV FreeImage_ApplyColorMapping(FIBITMAP *dib, RGBQUAD *srccolors, RGBQUAD *dstcolors, unsigned count, BOOL ignore_alpha, BOOL swap); +DLL_API unsigned DLL_CALLCONV FreeImage_SwapColors(FIBITMAP *dib, RGBQUAD *color_a, RGBQUAD *color_b, BOOL ignore_alpha); +DLL_API unsigned DLL_CALLCONV FreeImage_ApplyPaletteIndexMapping(FIBITMAP *dib, BYTE *srcindices, BYTE *dstindices, unsigned count, BOOL swap); +DLL_API unsigned DLL_CALLCONV FreeImage_SwapPaletteIndices(FIBITMAP *dib, BYTE *index_a, BYTE *index_b); + +// channel processing routines +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_GetChannel(FIBITMAP *dib, FREE_IMAGE_COLOR_CHANNEL channel); +DLL_API BOOL DLL_CALLCONV FreeImage_SetChannel(FIBITMAP *dst, FIBITMAP *src, FREE_IMAGE_COLOR_CHANNEL channel); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_GetComplexChannel(FIBITMAP *src, FREE_IMAGE_COLOR_CHANNEL channel); +DLL_API BOOL DLL_CALLCONV FreeImage_SetComplexChannel(FIBITMAP *dst, FIBITMAP *src, FREE_IMAGE_COLOR_CHANNEL channel); + +// copy / paste / composite routines +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Copy(FIBITMAP *dib, int left, int top, int right, int bottom); +DLL_API BOOL DLL_CALLCONV FreeImage_Paste(FIBITMAP *dst, FIBITMAP *src, int left, int top, int alpha); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_Composite(FIBITMAP *fg, BOOL useFileBkg FI_DEFAULT(FALSE), RGBQUAD *appBkColor FI_DEFAULT(NULL), FIBITMAP *bg FI_DEFAULT(NULL)); +DLL_API BOOL DLL_CALLCONV FreeImage_JPEGCrop(const char *src_file, const char *dst_file, int left, int top, int right, int bottom); +DLL_API BOOL DLL_CALLCONV FreeImage_JPEGCropU(const wchar_t *src_file, const wchar_t *dst_file, int left, int top, int right, int bottom); +DLL_API BOOL DLL_CALLCONV FreeImage_PreMultiplyWithAlpha(FIBITMAP *dib); + +// background filling routines +DLL_API BOOL DLL_CALLCONV FreeImage_FillBackground(FIBITMAP *dib, const void *color, int options FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_EnlargeCanvas(FIBITMAP *src, int left, int top, int right, int bottom, const void *color, int options FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_AllocateEx(int width, int height, int bpp, const RGBQUAD *color, int options FI_DEFAULT(0), const RGBQUAD *palette FI_DEFAULT(NULL), unsigned red_mask FI_DEFAULT(0), unsigned green_mask FI_DEFAULT(0), unsigned blue_mask FI_DEFAULT(0)); +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_AllocateExT(FREE_IMAGE_TYPE type, int width, int height, int bpp, const void *color, int options FI_DEFAULT(0), const RGBQUAD *palette FI_DEFAULT(NULL), unsigned red_mask FI_DEFAULT(0), unsigned green_mask FI_DEFAULT(0), unsigned blue_mask FI_DEFAULT(0)); + +// miscellaneous algorithms +DLL_API FIBITMAP *DLL_CALLCONV FreeImage_MultigridPoissonSolver(FIBITMAP *Laplacian, int ncycle FI_DEFAULT(3)); + +// restore the borland-specific enum size option +#if defined(__BORLANDC__) +#pragma option pop +#endif + +#ifdef __cplusplus +} +#endif + +#endif // FREEIMAGE_H diff --git a/src/FreeImage.lib b/src/FreeImage.lib new file mode 100644 index 0000000..b29ed71 Binary files /dev/null and b/src/FreeImage.lib differ diff --git a/src/clock.h b/src/clock.h new file mode 100644 index 0000000..ca13984 --- /dev/null +++ b/src/clock.h @@ -0,0 +1,113 @@ + +#ifndef __MMC_CLOCK_H__ +#define __MMC_CLOCK_H__ + +#include "timer.h" + +namespace mmc +{ + + /*! \brief High precision clock for timing events. + + This class is implemented on top of the Timer class. Usually it is + not necessary to use the Timer class directly; instead, use this + class. */ + class Clock + { + public: + /*! Construct a clock. */ + inline Clock (); + + /*! Reset the clock (0 time elapsed). */ + inline void reset (); + + /*! Call once per frame to update the internal clock state. */ + inline void inc (); + + /*! Returns the amount of time (in ms) elapsed between last + two calls to \p inc(). */ + inline long queryInc () const; + + /*! Returns the amount of time (in ms) elapsed since clock creation or + reset() was called. */ + inline long queryTime () const; + + /*! Pause the clock. */ + inline void pauseToggle (); + + private: + Timer *timer_; + MMC_TIME_TYPE inc_; + MMC_TIME_TYPE curTime_; + MMC_TIME_TYPE start_; + bool paused_; + double invFreq_; + }; + + ////////////////////////////////////////////////////////////////////// + // Implementation. + // + + inline + Clock::Clock () + : paused_(false) + { + timer_ = new Timer; + invFreq_ = timer_->getInvFreq(); + timer_->start(); + reset(); + } + + void + Clock::reset () + { + start_ = timer_->queryElapsed(); + inc_ = curTime_ = 0; + } + + void + Clock::inc () + { + timer_->inc(); + if (!paused_) + { + inc_ = timer_->queryInc(); + curTime_ += inc_; + } + else + { + inc_ = 0; + } + } + + // Converts internal clock increment to long in milliseconds. + long + Clock::queryInc () const + { + return (long) (1000.0 * inc_ * invFreq_); + } + + // Converts internal clock time to long in milliseconds. + long + Clock::queryTime () const + { + return (long) (1000.0 * curTime_ * invFreq_); + } + + void + Clock::pauseToggle () + { + if (paused_) + { + paused_ = false; + } + else + { + paused_ = true; + inc_ = 0; + } + } + +} // namespace mmc + +#endif // __MMC_CLOCK_H__ \ No newline at end of file diff --git a/src/fps.cpp b/src/fps.cpp new file mode 100644 index 0000000..b1d29ba --- /dev/null +++ b/src/fps.cpp @@ -0,0 +1,57 @@ +#include "fps.h" +#include "clock.h" + +using namespace mmc; + +FpsTracker::FpsTracker (int smoothSteps) + : steps_(smoothSteps), nSnaps_(0) +{ + clock_ = new Clock; + snaps_ = new long[steps_]; + memset(snaps_, 0, sizeof(long) * steps_); +} + +FpsTracker::~FpsTracker () +{ + delete snaps_; + delete clock_; +} + +void + FpsTracker::timestamp () +{ + clock_->inc(); + snaps_[nSnaps_ % steps_] = clock_->queryInc(); + nSnaps_++; +} + +float + FpsTracker::fpsAverage () const +{ + int count = (nSnaps_ < steps_) ? nSnaps_ : steps_; + long sum = 0L; + for (int i = 0; i < count; ++i) + { + sum += snaps_[i]; + } + if (!sum) sum = 1L; // prevent div-by-zero + return 1000.0f * count / (float) sum; +} + +float + FpsTracker::fpsInstant () const +{ + long inc = clock_->queryInc(); + if (!inc) inc = 1L; // prevent div-by-zero + return 1000.0f / (float) inc; +} + +void + FpsTracker::setNumSteps (int smoothSteps) +{ + steps_ = smoothSteps; + delete snaps_; + snaps_ = new long[steps_]; + memset(snaps_, 0, sizeof(long) * steps_); + nSnaps_ = 0; +} \ No newline at end of file diff --git a/src/fps.h b/src/fps.h new file mode 100644 index 0000000..4088c6a --- /dev/null +++ b/src/fps.h @@ -0,0 +1,54 @@ + +/*! \file fps.h + \brief Defines a class for keeping track of frames/sec. + \author Eric Chan + \date July 2003 +*/ + +#ifndef __MMC_FPS_H__ +#define __MMC_FPS_H__ + +#include + +namespace mmc +{ + class Clock; + + /*! \brief Utility class for measuring framerate. */ + class FpsTracker + { + public: + /*! Create a FpsTracker. \p smoothSteps is the window size + over which to average the time measurements. */ + FpsTracker (int smoothSteps = 4); + + /*! Deallocate a FpsTracker. */ + ~FpsTracker (); + + /*! Specify the window size \p smoothSteps over which the time + measurements will be averaged. */ + void setNumSteps (int smoothSteps); + + /*! Makes a timestamp (i.e. takes a snapshot); measures + time intervals between successful calls. Usually, you want + to call this function once per rendering loop. */ + void timestamp (); + + /*! Get the average FPS (averaged over \p smoothSteps + interval). */ + float fpsAverage () const; + + /*! Get the instantaneous FPS (estimated from the last frame + only). */ + float fpsInstant () const; + + private: + Clock *clock_; + int steps_; + int nSnaps_; + long *snaps_; + }; + +} // namespace mmc + +#endif // __MMC_FPS_H__ \ No newline at end of file diff --git a/src/interactions.h b/src/interactions.h index 7bf6fab..507bfa4 100644 --- a/src/interactions.h +++ b/src/interactions.h @@ -38,24 +38,50 @@ __host__ __device__ bool calculateScatterAndAbsorption(ray& r, float& depth, Abs return false; } -// TODO (OPTIONAL): IMPLEMENT THIS FUNCTION +// TODO (OPTIONAL): IMPLEMENT THIS FUNCTION +//Useless as I caculate ray in/out of object in intersection test and just use glm::refract __host__ __device__ glm::vec3 calculateTransmissionDirection(glm::vec3 normal, glm::vec3 incident, float incidentIOR, float transmittedIOR) { return glm::vec3(0,0,0); } // TODO (OPTIONAL): IMPLEMENT THIS FUNCTION +//Useless as I caculate ray in/out of object in intersection test and just use glm::reflect __host__ __device__ glm::vec3 calculateReflectionDirection(glm::vec3 normal, glm::vec3 incident) { //nothing fancy here return glm::vec3(0,0,0); } // TODO (OPTIONAL): IMPLEMENT THIS FUNCTION -__host__ __device__ Fresnel calculateFresnel(glm::vec3 normal, glm::vec3 incident, float incidentIOR, float transmittedIOR, glm::vec3 reflectionDirection, glm::vec3 transmissionDirection) { - Fresnel fresnel; +__host__ __device__ Fresnel calculateFresnel(glm::vec3 normal, glm::vec3 rdirect, float incidentIOR, float transmittedIOR, glm::vec3 reflectionDirection, glm::vec3 transmissionDirection) { + Fresnel fresnel; + float n1 = incidentIOR; + float n2 = transmittedIOR; + float n = n1 / n2; + + rdirect = glm::normalize(rdirect); + normal = glm::normalize(normal); + float c1 = glm::dot(-rdirect, normal); + float c2 = 1 - (n*n)*(1 - c1*c1); + + if(c2>=0) + { + c2 = sqrt(c2); + float R1 = glm::abs( (n1*c1 - n2*c2) / (n1*c1 + n2*c2) ) * glm::abs( (n1*c1 - n2*c2) / (n1*c1 + n2*c2) ); + float R2 = glm::abs( (n1*c2 - n2*c1) / (n1*c2 + n2*c1) ) * glm::abs( (n1*c2 - n2*c1) / (n1*c2 + n2*c1) ); + + float R = (R1 + R2) / 2.0f; + float T = 1.0 - R; + + fresnel.reflectionCoefficient = R; + fresnel.transmissionCoefficient = T; + } + else + { + fresnel.reflectionCoefficient = 1.0f; + fresnel.transmissionCoefficient = 0; + } - fresnel.reflectionCoefficient = 1; - fresnel.transmissionCoefficient = 0; - return fresnel; + return fresnel; } // LOOK: This function demonstrates cosine weighted random direction generation in a sphere! @@ -91,11 +117,77 @@ __host__ __device__ glm::vec3 calculateRandomDirectionInHemisphere(glm::vec3 nor // non-cosine (uniform) weighted random direction generation. // This should be much easier than if you had to implement calculateRandomDirectionInHemisphere. __host__ __device__ glm::vec3 getRandomDirectionInSphere(float xi1, float xi2) { - return glm::vec3(0,0,0); + float theta = 2 * TWO_PI * xi1; + float phi = acos(2 * xi2 - 1); + float u = cos(phi); + + return glm::vec3(sqrt(1 - u*u) * cos(theta), sqrt(1 - u*u) * sin(theta), u); } // TODO (PARTIALLY OPTIONAL): IMPLEMENT THIS FUNCTION // Returns 0 if diffuse scatter, 1 if reflected, 2 if transmitted. +#define MINNUM 0.001f +__host__ __device__ int calculateBSDF(ray& r, glm::vec3 InterSectP,glm::vec3 InterSectN, material m,float randomSeed,int depth){ + + thrust::uniform_real_distribution u01(0,1); + thrust::default_random_engine rng(hash(randomSeed)); + float russianRoulette = (float)u01(rng); + //if(depth>5 && russianRoulette < 0.2){ //Stop the recursion randomly based on the surface reflectivity + // r.exist = false; + //} + + + + //Only diffuse + if(m.hasReflectiveMINNUM) + { + float reflect_range = -1; + float IOR12 = oldIOR/newIOR; + refractR = glm::refract(r.direction,InterSectN,IOR12); + fresnel = calculateFresnel(InterSectN,r.direction,oldIOR,newIOR,reflectR,refractR); + + //return reflect ray,refract ray randomly depends on their Coefficients + reflect_range = fresnel.reflectionCoefficient; + + if((float)u01(rng) 0) + u = theta; + else + u = 1 - theta; + + int bi = v * bw; + int bj = u * bh; + int bIndex = bi * bh + bj; + if(geoms[geomIndex].texindex>=1) + bIndex += tnums[geoms[geomIndex].texindex-1]; + + m.color.x = tcolors[bIndex].x/255.0f; + m.color.y = tcolors[bIndex].y/255.0f; + m.color.z = tcolors[bIndex].z/255.0f; + return; + } + else if(geoms[geomIndex].type==CUBE) + { + glm::vec3 orginP = multiplyMV(geoms[geomIndex].inverseTransform, glm::vec4(InterSectP, 1.0)); + float u,v; + if(orginP[0]-0.5=1) + bIndex += tnums[geoms[geomIndex].texindex-1]; + + m.color.x = tcolors[bIndex].x/255.0f; + m.color.y = tcolors[bIndex].y/255.0f; + m.color.z = tcolors[bIndex].z/255.0f; + return; + } + else + return; + } +} + + +//Added to read bump map +__host__ __device__ void bumpMap( staticGeom* geoms,int geomIndex,glm::vec3 &InterSectN,glm::vec3 &InterSectP,uint3* bcolors,int* bnums) +{ + if(geoms[geomIndex].bumpindex != -1) + { + int bw = geoms[geomIndex].bwidth; + int bh = geoms[geomIndex].bheight; + if(geoms[geomIndex].type==SPHERE) + { + glm::vec3 NorthPole = multiplyMV(geoms[geomIndex].transform, glm::vec4(0, 0.5, 0, 1.0)); + glm::vec3 Equator = multiplyMV(geoms[geomIndex].transform, glm::vec4(0.5, 0, 0, 1.0)); + glm::vec3 Origin = multiplyMV(geoms[geomIndex].transform, glm::vec4(0,0,0,1)); + + glm::vec3 vn = glm::normalize(NorthPole - Origin); + glm::vec3 ve = glm::normalize(Equator - Origin); + + float phi = std::acos(-glm::dot(vn, InterSectN)); + float v = phi / PI; + + float theta = (std::acos(glm::dot(ve, InterSectN) / sin(phi))) / (2 * PI); + float u; + if(glm::dot(glm::cross(vn, ve), InterSectN) > 0) + u = theta; + else + u = 1 - theta; + + int bi = v * bw; + int bj = u * bh; + int bIndex = bi * bh + bj; + if(geoms[geomIndex].bumpindex>=1) + bIndex += bnums[geoms[geomIndex].bumpindex-1]; + + glm::vec3 Bumpuv = glm::vec3(2.0f*bcolors[bIndex].x/255.0f-1.0f,2.0f*bcolors[bIndex].y/255.0f-1.0f,2.0f*bcolors[bIndex].z/255.0f-1.0f); + InterSectN = glm::normalize(multiplyMV(geoms[geomIndex].transinverseTransform, glm::vec4(Bumpuv,0))); + return; + } + else if(geoms[geomIndex].type==CUBE) + { + glm::vec3 orginP = multiplyMV(geoms[geomIndex].inverseTransform, glm::vec4(InterSectP, 1.0)); + float u,v; + if(orginP[0]-0.5=1) + bIndex += bnums[geoms[geomIndex].bumpindex-1]; + + + glm::vec3 Bumpuv = glm::vec3(2.0f*bcolors[bIndex].x/255.0f-1.0f,2.0f*bcolors[bIndex].y/255.0f-1.0f,2.0f*bcolors[bIndex].z/255.0f-1.0f); + InterSectN = glm::normalize(multiplyMV(geoms[geomIndex].transinverseTransform, glm::vec4(Bumpuv,0))); + + return; + } + else + return; + } +} + #endif diff --git a/src/intersections.h b/src/intersections.h index c9eafb6..51eed8a 100644 --- a/src/intersections.h +++ b/src/intersections.h @@ -21,6 +21,50 @@ __host__ __device__ glm::vec3 getInverseDirectionOfRay(ray r); __host__ __device__ float boxIntersectionTest(staticGeom sphere, ray r, glm::vec3& intersectionPoint, glm::vec3& normal); __host__ __device__ float sphereIntersectionTest(staticGeom sphere, ray r, glm::vec3& intersectionPoint, glm::vec3& normal); __host__ __device__ glm::vec3 getRandomPointOnCube(staticGeom cube, float randomSeed); +//Added +__host__ __device__ bool Intersecttest(ray r, glm::vec3& intersectp, glm::vec3& intersectn, staticGeom* geoms, int numberOfGeoms, int& geomId); +__host__ __device__ bool epsilonCheck(float a, float b); +__host__ __device__ float triIntersectionTest(staticGeom mesh, ray r, glm::vec3& intersectionPoint, glm::vec3& normal); +__host__ __device__ bool BBIntersectionTest(staticGeom cube, ray r); + +__host__ __device__ bool Intersecttest(ray r, glm::vec3& intersectp, glm::vec3& intersectn, staticGeom* geoms, int numberOfGeoms, int& geomId) +{ + float tempdist = -1.0f,dist = FLT_MAX; + glm::vec3 temp_intersectp, temp_intersectn; + bool intersected = false; + + bool inbb = false; + for(int i = 0; i < numberOfGeoms; i++){ + if(geoms[i].type==SPHERE) + tempdist = sphereIntersectionTest(geoms[i], r, temp_intersectp,temp_intersectn); + else if (geoms[i].type==CUBE) + tempdist = boxIntersectionTest(geoms[i], r, temp_intersectp,temp_intersectn); + else if(geoms[i].type==MESH) + { + //Judge whether intersect with bounding box + if(inbb == false) + inbb = BBIntersectionTest(geoms[i],r); + + if(inbb == true) + tempdist = triIntersectionTest(geoms[i], r, temp_intersectp,temp_intersectn); + else + i = i + geoms[i].trinum -1; //if not intersected go to next geometry + } + + if(!epsilonCheck(tempdist, -1.0f)&&dist>tempdist) + { + dist = tempdist; + intersectn = temp_intersectn; + intersectp = temp_intersectp; + geomId = i; + intersected = true; + } + } + + return intersected; +} + + // Handy dandy little hashing function that provides seeds for random number generation __host__ __device__ unsigned int hash(unsigned int a){ @@ -73,47 +117,216 @@ __host__ __device__ glm::vec3 getSignOfRay(ray r){ // Cube intersection test, return -1 if no intersection, otherwise, distance to intersection __host__ __device__ float boxIntersectionTest(staticGeom box, ray r, glm::vec3& intersectionPoint, glm::vec3& normal){ - return -1; + glm::vec3 ro = multiplyMV(box.inverseTransform, glm::vec4(r.origin,1.0f)); + glm::vec3 rd = glm::normalize(multiplyMV(box.inverseTransform, glm::vec4(r.direction,0.0f))); + + ray rt; rt.origin = ro; rt.direction = rd; + + float sign=1.0f; + if(abs(rt.origin.x)-0.5<0&&abs(rt.origin.y)-0.5<0&&abs(rt.origin.z)-0.5<0) + sign=-1.0f; + + double tnear = -999999; + double tfar = 999999; + double t1,t2,temp,t; + for (int i = 0; i < 3; i++) { + if (rd[i] ==0 ) { + if (ro[i] > 0.5 || ro[i] < -0.5) { + return -1; + } + } + t1 = (-0.5 - ro[i])/rd[i]; + t2 = (0.5 - ro[i])/rd[i]; + if (t1 > t2) { + temp = t1; + t1 = t2; + t2 = temp; + } + if (t1 > tnear) { + tnear = t1; + } + if (t2 < tfar) { + tfar = t2; + } + if (tnear > tfar) { + return -1; + } + if (tfar < 0) { + return -1; + } + } + + if (tnear < -0.0001) + t=tfar; + else + t=tnear; + + + glm::vec3 P = getPointOnRay(rt, t); + if(abs(P[0]-0.5)<0.001) + normal = glm::vec3(1,0,0); + else if(abs(P[0]+0.5)<0.001) + normal = glm::vec3(-1,0,0); + else if(abs(P[1]-0.5)<0.001) + normal = glm::vec3(0,1,0); + else if(abs(P[1]+0.5)<0.001) + normal = glm::vec3(0,-1,0); + else if(abs(P[2]-0.5)<0.001) + normal = glm::vec3(0,0,1); + else if(abs(P[2]+0.5)<0.001) + normal = glm::vec3(0,0,-1); + + glm::vec3 realIntersectionPoint = multiplyMV(box.transform, glm::vec4(P, 1.0)); + + intersectionPoint = realIntersectionPoint; + normal = glm::normalize(sign * multiplyMV(box.transform, glm::vec4(normal,0))); + + return glm::length(r.origin - realIntersectionPoint); } // LOOK: Here's an intersection test example from a sphere. Now you just need to figure out cube and, optionally, triangle. // Sphere intersection test, return -1 if no intersection, otherwise, distance to intersection __host__ __device__ float sphereIntersectionTest(staticGeom sphere, ray r, glm::vec3& intersectionPoint, glm::vec3& normal){ - float radius = .5; - - glm::vec3 ro = multiplyMV(sphere.inverseTransform, glm::vec4(r.origin,1.0f)); - glm::vec3 rd = glm::normalize(multiplyMV(sphere.inverseTransform, glm::vec4(r.direction,0.0f))); + float radius = .5; - ray rt; rt.origin = ro; rt.direction = rd; - - float vDotDirection = glm::dot(rt.origin, rt.direction); - float radicand = vDotDirection * vDotDirection - (glm::dot(rt.origin, rt.origin) - pow(radius, 2)); - if (radicand < 0){ - return -1; - } - - float squareRoot = sqrt(radicand); - float firstTerm = -vDotDirection; - float t1 = firstTerm + squareRoot; - float t2 = firstTerm - squareRoot; - - float t = 0; - if (t1 < 0 && t2 < 0) { - return -1; - } else if (t1 > 0 && t2 > 0) { - t = min(t1, t2); - } else { - t = max(t1, t2); - } - - glm::vec3 realIntersectionPoint = multiplyMV(sphere.transform, glm::vec4(getPointOnRay(rt, t), 1.0)); - glm::vec3 realOrigin = multiplyMV(sphere.transform, glm::vec4(0,0,0,1)); - - intersectionPoint = realIntersectionPoint; - normal = glm::normalize(realIntersectionPoint - realOrigin); - - return glm::length(r.origin - realIntersectionPoint); + glm::vec3 ro = multiplyMV(sphere.inverseTransform, glm::vec4(r.origin,1.0f)); + glm::vec3 rd = glm::normalize(multiplyMV(sphere.inverseTransform, glm::vec4(r.direction,0.0f))); + + float sign=1.0f; + if(sqrt(glm::dot(ro,ro)) 0 && t2 > 0) { + t = min(t1, t2); + } else { + t = max(t1, t2); + } + + glm::vec3 realIntersectionPoint = multiplyMV(sphere.transform, glm::vec4(getPointOnRay(rt, t), 1.0)); + glm::vec3 realOrigin = multiplyMV(sphere.transform, glm::vec4(0,0,0,1)); + + intersectionPoint = realIntersectionPoint; + normal = glm::normalize(sign*(realIntersectionPoint - realOrigin)); + + return glm::length(r.origin - realIntersectionPoint); +} + + +//Triangle intersection test, return -1 if no intersection, otherwise, distance to intersection +__host__ __device__ float triIntersectionTest(staticGeom mesh, ray r, glm::vec3& intersectionPoint, glm::vec3& normal){ + + glm::vec3 ro = multiplyMV(mesh.inverseTransform, glm::vec4(r.origin,1.0f)); + glm::vec3 rd = glm::normalize(multiplyMV(mesh.inverseTransform, glm::vec4(r.direction,0.0f))); + + ray rt; rt.origin = ro; rt.direction = rd; + + glm::vec3 p = glm::vec3(rt.origin.x, rt.origin.y, rt.origin.z); + glm::vec3 d = glm::vec3(rt.direction.x, rt.direction.y, rt.direction.z); + glm::vec3 v0 = mesh.tri.p1; + glm::vec3 v1 = mesh.tri.p2; + glm::vec3 v2 = mesh.tri.p3; + glm::vec3 e1 = v1 - v0; + glm::vec3 e2 = v2 - v0; + + glm::vec3 h = glm::cross(d, e2); + float a = glm::dot(e1, h); + + if (a > -0.00001 && a < 0.00001) { + return -1; + } + + double f = 1.0/a; + glm::vec3 s = p - v0; + double u = f * (glm::dot(s, h)); + if (u < -0.00001 || u > 1.00001) { + return -1; + } + + glm::vec3 q = glm::cross(s, e1); + double v = f * (glm::dot(d, q)); + + if (v < -0.00001 || u + v > 1.00001) { + return -1; + } + + double t = f * glm::dot(e2, q); + + if (t > 0.001) { + glm::vec3 realIntersectionPoint = multiplyMV(mesh.transform, glm::vec4(getPointOnRay(rt, t), 1.0)); + intersectionPoint = realIntersectionPoint; + + normal = glm::normalize(multiplyMV(mesh.transinverseTransform, glm::vec4(mesh.tri.normal,0))); + if(glm::dot(r.direction,normal) > 0) + normal = -1.0f * normal; + return t; + } + + return -1; +} + +//Accelerate OBJ scan +__host__ __device__ bool BBIntersectionTest(staticGeom box, ray r){ + + glm::vec3 ro = multiplyMV(box.inverseTransform, glm::vec4(r.origin,1.0f)); + glm::vec3 rd = glm::normalize(multiplyMV(box.inverseTransform, glm::vec4(r.direction,0.0f))); + + ray rt; rt.origin = ro; rt.direction = rd; + + //The init size of objs are set to be smaller than 1*1*1, + //thus they must in area (-0.5,-0.5,-0.5) to (0.5,0.5,0.5) + double tnear = -999999; + double tfar = 999999; + double t1,t2,temp,t; + for (int i = 0; i < 3; i++) { + if (rd[i] ==0 ) { + if (ro[i] > 0.5 || ro[i] < -0.5) { + return false; + } + } + t1 = (-0.5 - ro[i])/rd[i]; + t2 = (0.5 - ro[i])/rd[i]; + if (t1 > t2) { + temp = t1; + t1 = t2; + t2 = temp; + } + if (t1 > tnear) { + tnear = t1; + } + if (t2 < tfar) { + tfar = t2; + } + if (tnear > tfar) { + return false; + } + if (tfar < 0) { + return false; + } + } + + if (tnear < -0.0001) + t=tfar; + else + t=tnear; + + return true; } // Returns x,y,z half-dimensions of tightest bounding box @@ -132,9 +345,9 @@ __host__ __device__ glm::vec3 getRadiuses(staticGeom geom){ // Generates a random point on a given cube __host__ __device__ glm::vec3 getRandomPointOnCube(staticGeom cube, float randomSeed){ - thrust::default_random_engine rng(hash(randomSeed)); - thrust::uniform_real_distribution u01(0,1); - thrust::uniform_real_distribution u02(-0.5,0.5); + thrust::default_random_engine rng(hash(randomSeed)); + thrust::uniform_real_distribution u01(0,1); + thrust::uniform_real_distribution u02(-0.5,0.5); // Get surface areas of sides glm::vec3 radii = getRadiuses(cube); @@ -178,7 +391,24 @@ __host__ __device__ glm::vec3 getRandomPointOnCube(staticGeom cube, float random // Generates a random point on a given sphere __host__ __device__ glm::vec3 getRandomPointOnSphere(staticGeom sphere, float randomSeed){ - return glm::vec3(0,0,0); + thrust::default_random_engine rng(hash(randomSeed)); + thrust::uniform_real_distribution u01(0,180); + thrust::uniform_real_distribution u02(0,360); + + float radius = .5f; + + glm::vec3 point; + + float theta, phi; + theta = glm::radians((float)u01(rng)); + phi = glm::radians((float)u02(rng)); + + point.x = radius * sin(theta) * cos(phi); + point.y = radius * sin(theta) * sin(phi); + point.z = radius * cos(theta); + + glm::vec3 randPoint = multiplyMV(sphere.transform, glm::vec4(point,1.0f)); + return randPoint; } #endif diff --git a/src/main.cpp b/src/main.cpp index b002500..e571535 100755 --- a/src/main.cpp +++ b/src/main.cpp @@ -25,6 +25,8 @@ int main(int argc, char** argv){ string header; string data; istringstream liness(argv[i]); getline(liness, header, '='); getline(liness, data, '='); + scenename = data.substr(0,data.length()-4); + scenename = scenename.substr(7,scenename.length()); if(strcmp(header.c_str(), "scene")==0){ renderScene = new scene(data); loadedScene = true; @@ -42,8 +44,8 @@ int main(int argc, char** argv){ // Set up camera stuff from loaded pathtracer settings iterations = 0; renderCam = &renderScene->renderCam; - width = renderCam->resolution[0]; - height = renderCam->resolution[1]; + width = (int)renderCam->resolution[0]; + height = (int)renderCam->resolution[1]; if(targetFrame >= renderCam->frames){ cout << "Warning: Specified target frame is out of range, defaulting to frame 0." << endl; @@ -63,8 +65,12 @@ void mainLoop() { while(!glfwWindowShouldClose(window)){ glfwPollEvents(); runCuda(); - - string title = "CIS565 Render | " + utilityCore::convertIntToString(iterations) + " Iterations"; + theFpsTracker.timestamp(); + string FPS; + stringstream ss; + ss<>FPS; + string title = "CIS565 Render | " + utilityCore::convertIntToString(iterations) + " Iterations" + " AverageFPS:" + FPS; glfwSetWindowTitle(window, title.c_str()); glBindBuffer(GL_PIXEL_UNPACK_BUFFER, pbo); @@ -88,8 +94,9 @@ void runCuda(){ // Map OpenGL buffer object for writing from CUDA on a single GPU // No data is moved (Win & Linux). When mapped to CUDA, OpenGL should not use this buffer - - if(iterations < renderCam->iterations){ + if(iterations < (int)renderCam->iterations){ + if(isRecording||iterations == (int)renderCam->iterations-1) + grabScreen(); uchar4 *dptr=NULL; iterations++; cudaGLMapBufferObject((void**)&dptr, pbo); @@ -98,15 +105,17 @@ void runCuda(){ geom* geoms = new geom[renderScene->objects.size()]; material* materials = new material[renderScene->materials.size()]; - for (int i=0; i < renderScene->objects.size(); i++) { + + for (int i=0; i < (int)renderScene->objects.size(); i++) { geoms[i] = renderScene->objects[i]; } - for (int i=0; i < renderScene->materials.size(); i++) { + for (int i=0; i < (int)renderScene->materials.size(); i++) { materials[i] = renderScene->materials[i]; } - + // execute the kernel - cudaRaytraceCore(dptr, renderCam, targetFrame, iterations, materials, renderScene->materials.size(), geoms, renderScene->objects.size() ); + cudaRaytraceCore(dptr, renderCam, targetFrame, iterations, materials, renderScene->materials.size(), + geoms, renderScene->objects.size() ,renderScene->colors,renderScene->lastnum,renderScene->bump_colors,renderScene->bump_lastnum); // unmap buffer object cudaGLUnmapBufferObject(pbo); @@ -114,19 +123,19 @@ void runCuda(){ if (!finishedRender) { // output image file - image outputImage(renderCam->resolution.x, renderCam->resolution.y); + image outputImage((int)renderCam->resolution.x, (int)renderCam->resolution.y); for (int x=0; x < renderCam->resolution.x; x++) { for (int y=0; y < renderCam->resolution.y; y++) { - int index = x + (y * renderCam->resolution.x); - outputImage.writePixelRGB(renderCam->resolution.x-1-x,y,renderCam->image[index]); + int index = x + (y * (int)renderCam->resolution.x); + outputImage.writePixelRGB((int)renderCam->resolution.x-1-x,y,renderCam->image[index]); } } gammaSettings gamma; gamma.applyGamma = true; - gamma.gamma = 1.0; - gamma.divisor = 1.0; + gamma.gamma = (int)1.0; + gamma.divisor = (int)1.0; outputImage.setGammaSettings(gamma); string filename = renderCam->imageName; string s; @@ -146,13 +155,13 @@ void runCuda(){ if (targetFrame < renderCam->frames-1) { // clear image buffer and move onto next frame - targetFrame++; - iterations = 0; - for(int i=0; iresolution.x*renderCam->resolution.y; i++){ - renderCam->image[i] = glm::vec3(0,0,0); - } - cudaDeviceReset(); - finishedRender = false; + targetFrame++; + iterations = 0; + for(int i=0; iresolution.x*renderCam->resolution.y; i++){ + renderCam->image[i] = glm::vec3(0,0,0); + } + cudaDeviceReset(); + finishedRender = false; } } } @@ -316,8 +325,125 @@ void errorCallback(int error, const char* description){ fputs(description, stderr); } +void ClearScreen() +{ + iterations = 0; + for(int i=0; iresolution.x*renderCam->resolution.y; i++){ + renderCam->image[i] = glm::vec3(0,0,0); + } +} + void keyCallback(GLFWwindow* window, int key, int scancode, int action, int mods){ if(key == GLFW_KEY_ESCAPE && action == GLFW_PRESS){ glfwSetWindowShouldClose(window, GL_TRUE); } + + if (key == GLFW_KEY_R&& action == GLFW_PRESS) + isRecording = !isRecording; + + + if(key==GLFW_KEY_T && action == GLFW_PRESS) + { + ClearScreen(); + texturemap_b =!texturemap_b; + if(texturemap_b) + cout<<"Texture Map Enabled!"<positions[0]=glm::vec3(renderCam->positions[0].x,renderCam->positions[0].y+0.1f,renderCam->positions[0].z); + } + else if(key==GLFW_KEY_DOWN) + { + ClearScreen(); + renderCam->positions[0]= glm::vec3(renderCam->positions[0].x,renderCam->positions[0].y-0.1f,renderCam->positions[0].z); + } + else if(key==GLFW_KEY_LEFT) + { + ClearScreen(); + renderCam->positions[0]= glm::vec3(renderCam->positions[0].x+0.1f,renderCam->positions[0].y,renderCam->positions[0].z); + } + else if(key==GLFW_KEY_RIGHT) + { + ClearScreen(); + renderCam->positions[0]= glm::vec3(renderCam->positions[0].x-0.1f,renderCam->positions[0].y,renderCam->positions->z); + } + else if(key==GLFW_KEY_Z) + { + ClearScreen(); + renderCam->positions[0]= glm::vec3(renderCam->positions[0].x,renderCam->positions[0].y,renderCam->positions[0].z-0.1f); + } + else if(key==GLFW_KEY_C) + { + ClearScreen(); + renderCam->positions[0]= glm::vec3(renderCam->positions[0].x,renderCam->positions[0].y,renderCam->positions[0].z+0.1f); + } } + + +//Added +void grabScreen(void) +{ + int window_width = 800; + int window_height = 800; + unsigned char* bitmapData = new unsigned char[3 * window_width * window_height]; + + for (int i=0; i < window_height; i++) + { + glReadPixels(0, i, window_width, 1, GL_RGB, GL_UNSIGNED_BYTE, + bitmapData + (window_width * 3 * ((window_height - 1) - i))); + } + + char anim_filename[2048]; + string f1 = "output/" + scenename; + f1 = f1 + "_%04d.png"; + char* filename = (char*)f1.c_str(); + sprintf_s(anim_filename, 2048, filename, iterations); + stbi_write_png(anim_filename, window_width, window_height, 3, bitmapData, window_width * 3); + + delete [] bitmapData; +} \ No newline at end of file diff --git a/src/main.h b/src/main.h index 8bd2aed..89201a5 100644 --- a/src/main.h +++ b/src/main.h @@ -27,6 +27,12 @@ #include "utilities.h" #include "scene.h" +//Add screen shot headers +#include "stb_image.h" +#include "stb_image_write.h" + +#include "fps.h" +mmc::FpsTracker theFpsTracker; using namespace std; //------------------------------- @@ -102,4 +108,15 @@ void mainLoop(); void errorCallback(int error, const char *description); void keyCallback(GLFWwindow *window, int key, int scancode, int action, int mods); + +//GrabScreen +bool isRecording=false; +void grabScreen(void); +string scenename; +bool streamcompact_b = true; +bool texturemap_b = false;//true; +bool bumpmap_b = false;//true; +bool DOF_b = false;//true; +bool MB_b = false;//true; + #endif diff --git a/src/objreader.cpp b/src/objreader.cpp new file mode 100644 index 0000000..64e6bbf --- /dev/null +++ b/src/objreader.cpp @@ -0,0 +1,217 @@ +#include "objreader.h" + +//vv for vertices positions +//vi for each triangle point's index +//fn for each triangle's normal +void OBJreader(vector &vv,vector &fn,vector &vi,string file,vec3 &maxpoint,vec3 &minpoint) +{ + vi.clear(); + vv.clear(); + fn.clear(); + string line; + ifstream infile(file,ios::in); + vector vn; + vector fnindex; + + float tempmax0,tempmax1,tempmax2,tempmin0,tempmin1,tempmin2; + maxpoint[0]=-1e20f; + minpoint[0]=1e20f; + maxpoint[1]=-1e20f; + minpoint[1]=1e20f; + maxpoint[2]=-1e20f; + minpoint[2]=1e20f; + + /*******************/ + + + while(getline(infile, line)) + { + if (line.substr(0,2) == "v ") + { + istringstream s(line.substr(2)); + vec3 v; + s >> v[0]; s >> v[1]; s >> v[2]; + v = v; + vv.push_back(v); + tempmax0=(float)v[0]; + tempmin0=(float)v[0]; + tempmax1=(float)v[1]; + tempmin1=(float)v[1]; + tempmax2=(float)v[2]; + tempmin2=(float)v[2]; + maxpoint[0]=std::max((float)maxpoint[0],tempmax0); + minpoint[0]=std::min((float)minpoint[0],tempmin0); + maxpoint[1]=std::max((float)maxpoint[1],tempmax1); + minpoint[1]=std::min((float)minpoint[1],tempmin1); + maxpoint[2]=std::max((float)maxpoint[2],tempmax2); + minpoint[2]=std::min((float)minpoint[2],tempmin2); + } + else if (line.substr(0,3) == "vn ") + { + istringstream s(line.substr(3)); + vec3 v; + s >> v[0]; s >> v[1]; s >> v[2]; + v = v; + vn.push_back(v); + } + else if (line.substr(0,2) == "f ") + { + istringstream s(line.substr(2)); + vec3 v; + vec3 vf; + string a,b,c; + string tempa,tempb,tempc; + int cut; + s >> a; s >> b; s >> c; + for(int i=0;i<(int)a.length();i++) + { + if(a.at(i)=='/') + { + cut=i; + tempa=a.substr(0,cut); + break; + } + + if(i==(int)a.length()-1) + { + cut=i; + tempa=a.substr(0,cut+1); + } + } + + for(int i=0;i<(int)b.length();i++) + { + if(b.at(i)=='/') + { + cut=i; + tempb=b.substr(0,cut); + break; + } + + if(i==(int)b.length()-1) + { + cut=i; + tempb=b.substr(0,cut+1); + } + } + + for(int i=0;i<(int)c.length();i++) + { + if(c.at(i)=='/') + { + cut=i; + tempc=c.substr(0,cut); + break; + } + + if(i==(int)c.length()-1) + { + cut=i; + tempc=c.substr(0,cut+1); + } + } + + + v[0]=atoi(tempa.c_str()); + v[1]=atoi(tempb.c_str()); + v[2]=atoi(tempc.c_str()); + + v[0]--; + v[1]--; + v[2]--; + + vi.push_back(v); + + if(vn.size()!=0) + { + //vnormal index + for(int i=(int)a.length()-1;i>=0;--i) + { + if(a.at(i)=='/') + { + cut=i; + tempa=a.substr(i+1,(int)a.length()); + break; + } + } + + for(int i=(int)b.length()-1;i>=0;--i) + { + if(b.at(i)=='/') + { + cut=i; + tempb=b.substr(i+1,(int)b.length()); + break; + } + } + + for(int i=(int)c.length()-1;i>=0;--i) + { + if(c.at(i)=='/') + { + cut=i; + tempc=c.substr(i+1,(int)c.length()); + break; + } + } + + v[0]=atoi(tempa.c_str()); + v[1]=atoi(tempb.c_str()); + v[2]=atoi(tempc.c_str()); + + v[0]--; + v[1]--; + v[2]--; + fnindex.push_back(v); + } + } + } + + //Scale all obj files to the size of 1(bounding box's longest edge is 1) + float x = maxpoint[0]-minpoint[0]; + float y = maxpoint[1]-minpoint[1]; + float z = maxpoint[2]-minpoint[2]; + float maxdim = std::max(std::max(x,y),z); + x = x/maxdim; + y = y/maxdim; + z = z/maxdim; + for(int i=0;i<(int)vv.size();i++) + { + vv[i] = vv[i]-minpoint; + vv[i][0] = (vv[i][0]/(maxpoint[0]-minpoint[0]) - 0.5f) * x; + vv[i][1] = (vv[i][1]/(maxpoint[1]-minpoint[1]) - 0.5f) * y; + vv[i][2] = (vv[i][2]/(maxpoint[2]-minpoint[2]) - 0.5f) * z; + } + + //for container + maxpoint[0] = 0.5f * x; + maxpoint[1] = 0.5f * y; + maxpoint[2] = 0.5f * z; + minpoint[0] = -0.5f * x; + minpoint[1] = -0.5f * y; + minpoint[2] = -0.5f * z; + + + //If the obj file has vn + if(vn.size()!=0) + { + for(unsigned int i = 0; i < fnindex.size() ; i++) + { + vec3 n1 = vn[fnindex[i][0]]; + vec3 n2 = vn[fnindex[i][1]]; + vec3 n3 = vn[fnindex[i][2]]; + vec3 fnormal = (n1+n2+n3)/3.0f; + fnormal = normalize(fnormal); + fn.push_back(fnormal); + } + } + else //compute face normal by myself + { + cout< +#include +#include +#include +#include + +using namespace std; +using namespace glm; + + +void OBJreader(vector &vv,vector &fn,vector &vi,string file,vec3 &maxpoint,vec3 &minpoint); \ No newline at end of file diff --git a/src/raytraceKernel.cu b/src/raytraceKernel.cu index 9c7bc7d..9a66d22 100644 --- a/src/raytraceKernel.cu +++ b/src/raytraceKernel.cu @@ -5,10 +5,10 @@ // Peter Kutz and Yining Karl Li's GPU Pathtracer: http://gpupathtracer.blogspot.com/ // Yining Karl Li's TAKUA Render, a massively parallel pathtracing renderer: http://www.yiningkarlli.com +#include #include #include #include - #include "sceneStructs.h" #include "glm/glm.hpp" #include "utilities.h" @@ -16,150 +16,474 @@ #include "intersections.h" #include "interactions.h" +extern bool streamcompact_b; +extern bool texturemap_b; +extern bool bumpmap_b; +extern bool DOF_b; +extern bool MB_b; + void checkCUDAError(const char *msg) { - cudaError_t err = cudaGetLastError(); - if( cudaSuccess != err) { - fprintf(stderr, "Cuda error: %s: %s.\n", msg, cudaGetErrorString( err) ); - exit(EXIT_FAILURE); - } + cudaError_t err = cudaGetLastError(); + if( cudaSuccess != err) { + fprintf(stderr, "Cuda error: %s: %s.\n", msg, cudaGetErrorString( err) ); + exit(EXIT_FAILURE); + } } // LOOK: This function demonstrates how to use thrust for random number generation on the GPU! // Function that generates static. __host__ __device__ glm::vec3 generateRandomNumberFromThread(glm::vec2 resolution, float time, int x, int y){ - int index = x + (y * resolution.x); - - thrust::default_random_engine rng(hash(index*time)); - thrust::uniform_real_distribution u01(0,1); + int index = x + (y * resolution.x); - return glm::vec3((float) u01(rng), (float) u01(rng), (float) u01(rng)); + thrust::default_random_engine rng(hash(index*time)); + thrust::uniform_real_distribution u01(0,1); + + return glm::vec3((float) u01(rng), (float) u01(rng), (float) u01(rng)); } // TODO: IMPLEMENT THIS FUNCTION // Function that does the initial raycast from the camera -__host__ __device__ ray raycastFromCameraKernel(glm::vec2 resolution, float time, int x, int y, glm::vec3 eye, glm::vec3 view, glm::vec3 up, glm::vec2 fov){ - ray r; - r.origin = glm::vec3(0,0,0); - r.direction = glm::vec3(0,0,-1); - return r; +__host__ __device__ ray raycastFromCameraKernel(glm::vec2 resolution, float time, int x, int y, glm::vec3 eye, glm::vec3 view, glm::vec3 up, glm::vec2 fov){ + glm::vec3 A = glm::cross(view,up); + glm::vec3 B = glm::cross(A,view); + glm::vec3 M = eye + view; + float angley = fov.y; + float anglex= fov.x; + glm::vec3 H = glm::normalize(A) * glm::length(view) * tan(glm::radians(anglex)); + glm::vec3 V = glm::normalize(B) * glm::length(view) * tan(glm::radians(angley)); + + float sx = ((float)x)/(resolution.x - 1); + float sy = ((float)y)/(resolution.y - 1); + + glm::vec3 P = M + (2.0f*sx-1)* H + (1-2.0f*sy) * V; //The picture begins + glm::vec3 D = glm::normalize(P - eye); + ray r; + r.origin = eye; + r.direction = D; + return r; } //Kernel that blacks out a given image buffer __global__ void clearImage(glm::vec2 resolution, glm::vec3* image){ - int x = (blockIdx.x * blockDim.x) + threadIdx.x; - int y = (blockIdx.y * blockDim.y) + threadIdx.y; - int index = x + (y * resolution.x); - if(x<=resolution.x && y<=resolution.y){ - image[index] = glm::vec3(0,0,0); - } + int x = (blockIdx.x * blockDim.x) + threadIdx.x; + int y = (blockIdx.y * blockDim.y) + threadIdx.y; + int index = x + (y * resolution.x); + if(x<=resolution.x && y<=resolution.y){ + image[index] = glm::vec3(0,0,0); + } } //Kernel that writes the image to the OpenGL PBO directly. -__global__ void sendImageToPBO(uchar4* PBOpos, glm::vec2 resolution, glm::vec3* image){ - - int x = (blockIdx.x * blockDim.x) + threadIdx.x; - int y = (blockIdx.y * blockDim.y) + threadIdx.y; - int index = x + (y * resolution.x); - - if(x<=resolution.x && y<=resolution.y){ - - glm::vec3 color; - color.x = image[index].x*255.0; - color.y = image[index].y*255.0; - color.z = image[index].z*255.0; - - if(color.x>255){ - color.x = 255; - } - - if(color.y>255){ - color.y = 255; - } - - if(color.z>255){ - color.z = 255; - } - - // Each thread writes one pixel location in the texture (textel) - PBOpos[index].w = 0; - PBOpos[index].x = color.x; - PBOpos[index].y = color.y; - PBOpos[index].z = color.z; - } +__global__ void sendImageToPBO(uchar4* PBOpos, glm::vec2 resolution, glm::vec3* image,float iterations){ + + int x = (blockIdx.x * blockDim.x) + threadIdx.x; + int y = (blockIdx.y * blockDim.y) + threadIdx.y; + int index = x + (y * resolution.x); + + if(x<=resolution.x && y<=resolution.y){ + + glm::vec3 color; + color.x = image[index].x*255.0; + color.y = image[index].y*255.0; + color.z = image[index].z*255.0; + color /= iterations; + + if(color.x>255){ + color.x = 255; + } + + if(color.y>255){ + color.y = 255; + } + + if(color.z>255){ + color.z = 255; + } + + // Each thread writes one pixel location in the texture (textel) + PBOpos[index].w = 0; + PBOpos[index].x = color.x; + PBOpos[index].y = color.y; + PBOpos[index].z = color.z; + } } // TODO: IMPLEMENT THIS FUNCTION // Core raytracer kernel -__global__ void raytraceRay(glm::vec2 resolution, float time, cameraData cam, int rayDepth, glm::vec3* colors, - staticGeom* geoms, int numberOfGeoms){ +__global__ void PathTraceColor(ray* remainrays,int raysnum,int currdepth,int maxdepth, + staticGeom* geoms, int numberOfGeoms, int* lightIndex, + int lightNum,material* materials,float time,uint3* tcolors,int* tnums,bool textureb,uint3* bcolors,int* bnums,bool bumpb) +{ + int index = blockIdx.x * blockDim.x + threadIdx.x; + if(index<=raysnum) + { + ray r = remainrays[index]; + if(!r.exist) //If not exist, its new add color is 0 + { + r.raycolor = glm::vec3(0,0,0); + remainrays[index] = r; + return; + } + + //clear all ray if currdepth == maxdepth + if(currdepth==maxdepth) + { + r.exist = false; + remainrays[index] = r; + return; + } + + bool Intersect = false; + glm::vec3 InterSectP,InterSectN; + int IntersectgeomId = -1; + Intersect = Intersecttest(r,InterSectP,InterSectN,geoms,numberOfGeoms,IntersectgeomId); + if(bumpb) + bumpMap(geoms,IntersectgeomId,InterSectN,InterSectP,bcolors,bnums); + //if the ray intersect with nothing, give it black/backgroundcolor + if(Intersect==false) + { + r.raycolor = glm::vec3(0,0,0); + r.exist = false; + remainrays[index] = r; + return; + } + + + material currMaterial = materials[geoms[IntersectgeomId].materialid]; - int x = (blockIdx.x * blockDim.x) + threadIdx.x; - int y = (blockIdx.y * blockDim.y) + threadIdx.y; - int index = x + (y * resolution.x); + if(textureb) + textureMap(geoms,IntersectgeomId,currMaterial,InterSectN,InterSectP,tcolors,tnums); - if((x<=resolution.x && y<=resolution.y)){ + bool IsLight = false; + for(int i=0;i u01(0, 1); + //ray r = raycastFromCameraKernel(resolution,0.0f, x + float(u01(rng)) -0.5f, y+float(u01(rng))-0.5f,cam.position,cam.view,cam.up,cam.fov); + ray r = raycastFromCameraKernel(resolution,0.0f, x , y,cam.position,cam.view,cam.up,cam.fov); + if(DOFbool) + { + glm::vec3 rand3 = generateRandomNumberFromThread(resolution, time, x, y); + glm::vec2 rand2 = glm::vec2(rand3.x,rand3.y); + glm::vec3 offset = glm::vec3(rand2.x * cos((float)TWO_PI*rand2.y), rand2.x * sin((float)TWO_PI*rand2.y), 0.0f) * cam.blurradius; + glm::vec3 p = r.origin + r.direction * cam.focallength / glm::dot(cam.view, r.direction); + r.origin = r.origin + offset; + r.direction = glm::normalize(p - r.origin); + } + + + + r.exist = true; + r.initindex = index; + r.raycolor = glm::vec3(1,1,1); + r.IOR = 1.0f; + rays[index] = r; + } +} + +struct Is_Exist +{ + __host__ __device__ + bool operator()(const ray x) + { + if(x.exist) return true; + else return false; + } +}; + +struct Is_Not_Exist +{ + __host__ __device__ + bool operator()(const ray x) + { + if(!x.exist) return true; + else return false; + } +}; + +//Stream Compact +void ThrustStreamCompact(thrust::device_ptr origin,int &N) +{ + //Count how many rays still exist + int finallength = thrust::count_if(origin, origin+N,Is_Exist()); + thrust::remove_if(origin, origin+N,Is_Not_Exist()); + N = finallength; + return; +} + + // TODO: FINISH THIS FUNCTION // Wrapper for the __global__ call that sets up the kernel calls and does a ton of memory management -void cudaRaytraceCore(uchar4* PBOpos, camera* renderCam, int frame, int iterations, material* materials, int numberOfMaterials, geom* geoms, int numberOfGeoms){ - - int traceDepth = 1; //determines how many bounces the raytracer traces - - // set up crucial magic - int tileSize = 8; - dim3 threadsPerBlock(tileSize, tileSize); - dim3 fullBlocksPerGrid((int)ceil(float(renderCam->resolution.x)/float(tileSize)), (int)ceil(float(renderCam->resolution.y)/float(tileSize))); - - // send image to GPU - glm::vec3* cudaimage = NULL; - cudaMalloc((void**)&cudaimage, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3)); - cudaMemcpy( cudaimage, renderCam->image, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3), cudaMemcpyHostToDevice); - - // package geometry and materials and sent to GPU - staticGeom* geomList = new staticGeom[numberOfGeoms]; - for(int i=0; iresolution; - cam.position = renderCam->positions[frame]; - cam.view = renderCam->views[frame]; - cam.up = renderCam->ups[frame]; - cam.fov = renderCam->fov; - - // kernel launches - raytraceRay<<>>(renderCam->resolution, (float)iterations, cam, traceDepth, cudaimage, cudageoms, numberOfGeoms); - - sendImageToPBO<<>>(PBOpos, renderCam->resolution, cudaimage); - - // retrieve image from GPU - cudaMemcpy( renderCam->image, cudaimage, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3), cudaMemcpyDeviceToHost); - - // free up stuff, or else we'll leak memory like a madman - cudaFree( cudaimage ); - cudaFree( cudageoms ); - delete geomList; - - // make certain the kernel has completed - cudaThreadSynchronize(); - - checkCUDAError("Kernel failed!"); -} +void cudaRaytraceCore(uchar4* PBOpos, camera* renderCam, int frame, int iterations, material* materials, + int numberOfMaterials, geom* geoms, int numberOfGeoms,std::vector mapcolors,std::vector maplastnums, + std::vector bumcolors,std::vector bumlastnums){ + + int traceDepth = 8; //determines how many bounces the raytracer traces + + // set up crucial magic + int tileSize = 8; + dim3 threadsPerBlock(tileSize, tileSize); + dim3 fullBlocksPerGrid((int)ceil(float(renderCam->resolution.x)/float(tileSize)), (int)ceil(float(renderCam->resolution.y)/float(tileSize))); + + // send image to GPU + glm::vec3* cudaimage = NULL; + cudaMalloc((void**)&cudaimage, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3)); + cudaMemcpy( cudaimage, renderCam->image, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3), cudaMemcpyHostToDevice); + + // package geometry and materials and sent to GPU + staticGeom* geomList = new staticGeom[numberOfGeoms]; + for(int i=0; i=0&&tempit<25) + { + newStaticGeom.translation += (float)tempit * newStaticGeom.MBV; + transform = utilityCore::buildTransformationMatrix(newStaticGeom.translation, newStaticGeom.rotation, newStaticGeom.scale); + } + else if(tempit>24) + { + newStaticGeom.translation -= (float)tempit * newStaticGeom.MBV; + transform = utilityCore::buildTransformationMatrix(newStaticGeom.translation, newStaticGeom.rotation, newStaticGeom.scale); + } + + newStaticGeom.transform = utilityCore::glmMat4ToCudaMat4(transform); + newStaticGeom.inverseTransform = utilityCore::glmMat4ToCudaMat4(glm::inverse(transform)); + } + + newStaticGeom.tri = geoms[i].tri; + newStaticGeom.trinum = geoms[i].trinum; + newStaticGeom.texindex = geoms[i].texindex; + newStaticGeom.theight = geoms[i].theight; + newStaticGeom.twidth = geoms[i].twidth; + newStaticGeom.bumpindex = geoms[i].bumpindex; + newStaticGeom.bheight = geoms[i].bheight; + newStaticGeom.bwidth = geoms[i].bwidth; + geomList[i] = newStaticGeom; + } + + staticGeom* cudageoms = NULL; + cudaMalloc((void**)&cudageoms, numberOfGeoms*sizeof(staticGeom)); + cudaMemcpy( cudageoms, geomList, numberOfGeoms*sizeof(staticGeom), cudaMemcpyHostToDevice); + + //materials + material* materialList = new material[numberOfMaterials]; + for(int i=0; i0) + lcount++; + } + + int *lightIds = new int[lcount]; + lcount = 0; + for(int i=0; i0) + { + lightIds[lcount] = i; + lcount++; + } + } + + int *cudalightIds=NULL; + cudaMalloc((void**)&cudalightIds,lcount * sizeof(int)); + cudaMemcpy( cudalightIds, lightIds,lcount * sizeof(int), cudaMemcpyHostToDevice); + + // package camera + cameraData cam; + cam.resolution = renderCam->resolution; + cam.position = renderCam->positions[frame]; + cam.view = renderCam->views[frame]; + cam.up = renderCam->ups[frame]; + cam.fov = renderCam->fov; + cam.focallength = renderCam->focall; + cam.blurradius = renderCam->blurr; + + //Transfer Texture Map + uint3* cudacolors = NULL; + int* cudalastnums = NULL; + if(texturemap_b) + { + if(iterations==1 && (maplastnums.size()==0||mapcolors.size()==0)) + { + std::cout<<"No Texture Map Set!"<resolution.x*renderCam->resolution.y; + ray* cudarays = NULL; + cudaMalloc((void**)&cudarays, numberOfInitrays*sizeof(ray)); + InitRays<<>>(cudarays,renderCam->resolution,cam,(float)iterations,DOF_b); + + //set path trace dim + int raythreadsPerBlock = (int)(tileSize*tileSize); + int rayblocksPerGrid = ceil((float)numberOfInitrays/(float)raythreadsPerBlock); + + // kernel launches + for(int i=0;i<=traceDepth;i++) + { + if(numberOfInitrays>0) + { + PathTraceColor<<>>(cudarays,numberOfInitrays,i,traceDepth,cudageoms, + numberOfGeoms,cudalightIds,lcount,cudamaterials,(float)iterations,cudacolors,cudalastnums,texturemap_b + ,cudabumcolors,cudabumlastnums,bumpmap_b); + AddColor<<>>(cudaimage, cudarays,numberOfInitrays); + if(streamcompact_b) + { + thrust::device_ptr rayStart(cudarays); + ThrustStreamCompact(rayStart,numberOfInitrays); + rayblocksPerGrid = ceil((float)numberOfInitrays/(float)raythreadsPerBlock); + } + } + } + + + sendImageToPBO<<>>(PBOpos, renderCam->resolution, cudaimage,(float)iterations); + + // retrieve image from GPU + cudaMemcpy( renderCam->image, cudaimage, (int)renderCam->resolution.x*(int)renderCam->resolution.y*sizeof(glm::vec3), cudaMemcpyDeviceToHost); + + // free up stuff, or else we'll leak memory like a madman + cudaFree( cudaimage ); + cudaFree( cudageoms ); + //Added + cudaFree( cudalightIds ); + cudaFree( cudamaterials ); + cudaFree( cudarays ); + cudaFree( cudacolors ); + cudaFree( cudalastnums ); + cudaFree( cudabumcolors ); + cudaFree( cudabumlastnums ); + + delete geomList; + //Added + delete materialList; + delete lightIds; + + + // make certain the kernel has completed + cudaThreadSynchronize(); + + checkCUDAError("Kernel failed!"); +} \ No newline at end of file diff --git a/src/raytraceKernel.h b/src/raytraceKernel.h index 984e89f..08102f6 100755 --- a/src/raytraceKernel.h +++ b/src/raytraceKernel.h @@ -14,6 +14,15 @@ #include #include "sceneStructs.h" -void cudaRaytraceCore(uchar4* pos, camera* renderCam, int frame, int iterations, material* materials, int numberOfMaterials, geom* geoms, int numberOfGeoms); +//Added +#include +#include +#include +#include +#include + +void cudaRaytraceCore(uchar4* pos, camera* renderCam, int frame, int iterations, material* materials, + int numberOfMaterials, geom* geoms, int numberOfGeoms,std::vector allcolors,std::vector alllastnums + ,std::vector bumcolors,std::vector bumlastnums); #endif diff --git a/src/scene.cpp b/src/scene.cpp index 4cbe216..4dbd29b 100644 --- a/src/scene.cpp +++ b/src/scene.cpp @@ -7,8 +7,16 @@ #include #include "scene.h" #include +//Add FreeImage.lib +#pragma comment(lib,"FreeImage.lib") +extern bool texturemap_b; +extern bool bumpmap_b; scene::scene(string filename){ + colors.clear(); + lastnum.clear(); + bump_colors.clear(); + bump_lastnum.clear(); cout << "Reading scene from " << filename << " ..." << endl; cout << " " << endl; char* fname = (char*)filename.c_str(); @@ -34,12 +42,47 @@ scene::scene(string filename){ } } +bool LoadPic(string tfilename,vector &colors,vector &lastnum, int &nHeight,int &nWidth) +{ + FIBITMAP* bitmap; + bitmap =FreeImage_Load(FreeImage_GetFileType(tfilename.c_str(), 0), tfilename.c_str()); + bitmap = FreeImage_ConvertTo32Bits(bitmap); + + nWidth = FreeImage_GetWidth(bitmap); + nHeight = FreeImage_GetHeight(bitmap); + int l = nWidth*nHeight; + if(lastnum.size()>0) + l += lastnum[lastnum.size()-1]; + + lastnum.push_back(l); + for(int i=0;i tris; + glm::vec3 maxp,minp; + //if(id!=objects.size()){ + // cout << "ERROR: OBJECT ID does not match expected number of objects" << endl; + // return -1; + //}else{ cout << "Loading Object " << id << "..." << endl; geom newObject; string line; @@ -63,13 +106,22 @@ int scene::loadObject(string objectid){ if(strcmp(extension.c_str(), "obj")==0){ cout << "Creating new mesh..." << endl; cout << "Reading mesh from " << line << "... " << endl; - newObject.type = MESH; + newObject.type = MESH; + //Add to load obj file + std::vector vv,vi,fn; + + OBJreader(vv,fn,vi,objline,maxp,minp); + for(int i=0;i<(int)vi.size();i++) + { + triangle newtri(vv[vi[i][0]],vv[vi[i][1]],vv[vi[i][2]],fn[i]); + tris.push_back(newtri); + } }else{ cout << "ERROR: " << line << " is not a valid object type!" << endl; return -1; } } - } + //} //link material utilityCore::safeGetline(fp_in,line); @@ -77,7 +129,7 @@ int scene::loadObject(string objectid){ vector tokens = utilityCore::tokenizeString(line); newObject.materialid = atoi(tokens[1].c_str()); cout << "Connecting Object " << objectid << " to Material " << newObject.materialid << "..." << endl; - } + } //load frames int frameCount = 0; @@ -85,6 +137,8 @@ int scene::loadObject(string objectid){ vector translations; vector scales; vector rotations; + vector MBV; + while (!line.empty() && fp_in.good()){ //check frame number @@ -95,7 +149,7 @@ int scene::loadObject(string objectid){ } //load tranformations - for(int i=0; i<3; i++){ + for(int i=0; i<6; i++){ glm::vec3 translation; glm::vec3 rotation; glm::vec3 scale; utilityCore::safeGetline(fp_in,line); tokens = utilityCore::tokenizeString(line); @@ -106,6 +160,59 @@ int scene::loadObject(string objectid){ }else if(strcmp(tokens[0].c_str(), "SCALE")==0){ scales.push_back(glm::vec3(atof(tokens[1].c_str()), atof(tokens[2].c_str()), atof(tokens[3].c_str()))); } + else if(strcmp(tokens[0].c_str(), "MBV")==0){ + MBV.push_back(glm::vec3(atof(tokens[1].c_str()), atof(tokens[2].c_str()), atof(tokens[3].c_str()))); + } + else if(strcmp(tokens[0].c_str(), "MAP")==0){ + if(texturemap_b) + { + string ext = ""; + int h= 0,w=0; + newObject.theight = h; + newObject.twidth = w; + newObject.texindex = -1; + if(tokens.size()>1&&tokens[1].size()>=4) + ext = tokens[1].substr(tokens[1].length()-3,tokens[1].length()); + + if(strcmp(ext.c_str(), "jpg")==0||strcmp(ext.c_str(), "png")==0||strcmp(ext.c_str(), "bmp")==0) + { + if(LoadPic(tokens[1].c_str(),colors,lastnum,h,w)) + { + cout<<"Finish load texture map"<1&&tokens[1].size()>=4) + ext = tokens[1].substr(tokens[1].length()-3,tokens[1].length()); + + if(strcmp(ext.c_str(), "jpg")==0||strcmp(ext.c_str(), "png")==0||strcmp(ext.c_str(), "bmp")==0) + { + if(LoadPic(tokens[1].c_str(),bump_colors,bump_lastnum,h,w)) + { + cout<<"Finish load bump map"< #include +//OBJ +#include "objreader.h" + +//Add bmp read headers +#include "FreeImage.h" + using namespace std; class scene{ @@ -29,6 +35,14 @@ class scene{ vector objects; vector materials; camera renderCam; + + //Data for texture map + vector lastnum; + vector colors; + + //Data for bump map + vector bump_lastnum; + vector bump_colors; }; #endif diff --git a/src/sceneStructs.h b/src/sceneStructs.h index 5e0c853..0a91ae8 100644 --- a/src/sceneStructs.h +++ b/src/sceneStructs.h @@ -10,12 +10,35 @@ #include "cudaMat4.h" #include #include +#include +#include +#include enum GEOMTYPE{ SPHERE, CUBE, MESH }; struct ray { glm::vec3 origin; glm::vec3 direction; + bool exist; + glm::vec3 raycolor; + int initindex; + float IOR; //REFRIOR +}; + +//Mesh struct +struct triangle{ + glm::vec3 p1,p2,p3,normal; + triangle() + { + p1= glm::vec3(0,0,0); p2= glm::vec3(0,0,0); p3= glm::vec3(0,0,0); normal= glm::vec3(0,0,0); + }; + triangle(glm::vec3 t1,glm::vec3 t2,glm::vec3 t3,glm::vec3 n) + { + p1 = t1; + p2 = t2; + p3 = t3; + normal = n; + } }; struct geom { @@ -27,8 +50,29 @@ struct geom { glm::vec3* scales; cudaMat4* transforms; cudaMat4* inverseTransforms; + cudaMat4* transinverseTransforms; + + //data for Motion Blur + glm::vec3* MBV; + + //data for obj + triangle tri; + + //To accelerate + int trinum; + + //Texture map data + int texindex; + int theight; + int twidth; + + //Bump map data + int bumpindex; + int bheight; + int bwidth; }; + struct staticGeom { enum GEOMTYPE type; int materialid; @@ -37,6 +81,25 @@ struct staticGeom { glm::vec3 scale; cudaMat4 transform; cudaMat4 inverseTransform; + cudaMat4 transinverseTransform; + + //data for Motion Blur + glm::vec3 MBV; + + //data for obj + int trinum; + //To accelerate + triangle tri; + + //Texture map data + int texindex; + int theight; + int twidth; + + //Bump map data + int bumpindex; + int bheight; + int bwidth; }; struct cameraData { @@ -45,6 +108,10 @@ struct cameraData { glm::vec3 view; glm::vec3 up; glm::vec2 fov; + + //Added for DOF + float focallength; + float blurradius; }; struct camera { @@ -58,6 +125,10 @@ struct camera { glm::vec3* image; ray* rayList; std::string imageName; + + //Added for DOF + float focall; + float blurr; }; struct material{ diff --git a/src/stb_image.c b/src/stb_image.c new file mode 100644 index 0000000..2c1358b --- /dev/null +++ b/src/stb_image.c @@ -0,0 +1,4674 @@ +// Courtesy of Peter Kutz +/* stbi-1.33 - public domain JPEG/PNG reader - http://nothings.org/stb_image.c + when you control the images you're loading + no warranty implied; use at your own risk + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline (no JPEG progressive) + PNG 8-bit only + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - overridable dequantizing-IDCT, YCbCr-to-RGB conversion (define STBI_SIMD) + + Latest revisions: + 1.33 (2011-07-14) minor fixes suggested by Dave Moore + 1.32 (2011-07-13) info support for all filetypes (SpartanJ) + 1.31 (2011-06-19) a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) added ability to load files via io callbacks (Ben Wenger) + 1.29 (2010-08-16) various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) cast-to-uint8 to fix warnings (Laurent Gomila) + allow trailing 0s at end of image data (Laurent Gomila) + 1.26 (2010-07-24) fix bug in file buffering for PNG reported by SpartanJ + + See end of file for full revision history. + + TODO: + stbi_info support for BMP,PSD,HDR,PIC + + + ============================ Contributors ========================= + + Image formats Optimizations & bugfixes + Sean Barrett (jpeg, png, bmp) Fabian "ryg" Giesen + Nicolas Schulz (hdr, psd) + Jonathan Dummer (tga) Bug fixes & warning fixes + Jean-Marc Lienher (gif) Marc LeBlanc + Tom Seddon (pic) Christpher Lloyd + Thatcher Ulrich (psd) Dave Moore + Won Chun + the Horde3D community + Extensions, features Janez Zemva + Jetro Lauha (stbi_info) Jonathan Blow + James "moose2000" Brown (iPhone PNG) Laurent Gomila + Ben "Disch" Wenger (io callbacks) Aruelien Pocheville + Martin "SpartanJ" Golini Ryamond Barbiero + David Woo + + + If your name should be here but isn't, let Sean know. + +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// To get a header file for this, either cut and paste the header, +// or create stb_image.h, #define STBI_HEADER_FILE_ONLY, and +// then include stb_image.c from it. + +//// begin header file //////////////////////////////////////////////////// +// +// Limitations: +// - no jpeg progressive support +// - non-HDR formats support 8-bit samples only (jpeg, png) +// - no delayed line count (jpeg) -- IJG doesn't support either +// - no 1-bit BMP +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below): +// int x,y,n; +// unsigned char *data = stbi_load(filename, &x, &y, &n, 0); +// // ... process data if not NULL ... +// // ... x = width, y = height, n = # 8-bit components per pixel ... +// // ... replace '0' with '1'..'4' to force that many components per pixel +// // ... but 'n' will always be the number that it would have been if you said 0 +// stbi_image_free(data) +// +// Standard parameters: +// int *x -- outputs image width in pixels +// int *y -- outputs image height in pixels +// int *comp -- outputs # of image components in image file +// int req_comp -- if non-zero, # of image components requested in result +// +// The return value from an image loader is an 'unsigned char *' which points +// to the pixel data. The pixel data consists of *y scanlines of *x pixels, +// with each pixel consisting of N interleaved 8-bit components; the first +// pixel pointed to is top-left-most in the image. There is no padding between +// image scanlines or between pixels, regardless of format. The number of +// components N is 'req_comp' if req_comp is non-zero, or *comp otherwise. +// If req_comp is non-zero, *comp has the number of components that _would_ +// have been output otherwise. E.g. if you set req_comp to 4, you will always +// get RGBA output, but you can check *comp to easily see if it's opaque. +// +// An output image with N components has the following components interleaved +// in this order in each pixel: +// +// N=#comp components +// 1 grey +// 2 grey, alpha +// 3 red, green, blue +// 4 red, green, blue, alpha +// +// If image loading fails for any reason, the return value will be NULL, +// and *x, *y, *comp will be unchanged. The function stbi_failure_reason() +// can be queried for an extremely brief, end-user unfriendly explanation +// of why the load failed. Define STBI_NO_FAILURE_STRINGS to avoid +// compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly +// more user-friendly ones. +// +// Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. +// +// =========================================================================== +// +// iPhone PNG support: +// +// By default we convert iphone-formatted PNGs back to RGB; nominally they +// would silently load as BGR, except the existing code should have just +// failed on such iPhone PNGs. But you can disable this conversion by +// by calling stbi_convert_iphone_png_to_rgb(0), in which case +// you will always just get the native iphone "format" through. +// +// Call stbi_set_unpremultiply_on_load(1) as well to force a divide per +// pixel to remove any premultiplied alpha *only* if the image file explicitly +// says there's premultiplied data (currently only happens in iPhone images, +// and only if iPhone convert-to-rgb processing is on). +// +// =========================================================================== +// +// HDR image support (disable by defining STBI_NO_HDR) +// +// stb_image now supports loading HDR images in general, and currently +// the Radiance .HDR file format, although the support is provided +// generically. You can still load any file through the existing interface; +// if you attempt to load an HDR file, it will be automatically remapped to +// LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; +// both of these constants can be reconfigured through this interface: +// +// stbi_hdr_to_ldr_gamma(2.2f); +// stbi_hdr_to_ldr_scale(1.0f); +// +// (note, do not use _inverse_ constants; stbi_image will invert them +// appropriately). +// +// Additionally, there is a new, parallel interface for loading files as +// (linear) floats to preserve the full dynamic range: +// +// float *data = stbi_loadf(filename, &x, &y, &n, 0); +// +// If you load LDR images through this interface, those images will +// be promoted to floating point values, run through the inverse of +// constants corresponding to the above: +// +// stbi_ldr_to_hdr_scale(1.0f); +// stbi_ldr_to_hdr_gamma(2.2f); +// +// Finally, given a filename (or an open file or memory block--see header +// file for details) containing image data, you can query for the "most +// appropriate" interface to use (that is, whether the image is HDR or +// not), using: +// +// stbi_is_hdr(char *filename); +// +// =========================================================================== +// +// I/O callbacks +// +// I/O callbacks allow you to read from arbitrary sources, like packaged +// files or some other source. Data read from callbacks are processed +// through a small internal buffer (currently 128 bytes) to try to reduce +// overhead. +// +// The three functions you must define are "read" (reads some bytes of data), +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). + + +#ifndef STBI_NO_STDIO + +#if defined(_MSC_VER) // Modified by Peter Kutz. +#define _CRT_SECURE_NO_WARNINGS // suppress bogus warnings about fopen() +#endif + +#include +#endif + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for req_comp + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +typedef unsigned char stbi_uc; + +#ifdef __cplusplus +extern "C" { +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// PRIMARY API - works on images of any type +// + +// +// load image by filename, open file, or memory buffer +// + +extern stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp); + +#ifndef STBI_NO_STDIO +extern stbi_uc *stbi_load (char const *filename, int *x, int *y, int *comp, int req_comp); +extern stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp); +// for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +typedef struct +{ + int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read + void (*skip) (void *user,unsigned n); // skip the next 'n' bytes + int (*eof) (void *user); // returns nonzero if we are at end of file/data +} stbi_io_callbacks; + +extern stbi_uc *stbi_load_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp); + +#ifndef STBI_NO_HDR + extern float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp); + + #ifndef STBI_NO_STDIO + extern float *stbi_loadf (char const *filename, int *x, int *y, int *comp, int req_comp); + extern float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *comp, int req_comp); + #endif + + extern float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp); + + extern void stbi_hdr_to_ldr_gamma(float gamma); + extern void stbi_hdr_to_ldr_scale(float scale); + + extern void stbi_ldr_to_hdr_gamma(float gamma); + extern void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_HDR + +// stbi_is_hdr is always defined +extern int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); +extern int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); +#ifndef STBI_NO_STDIO +extern int stbi_is_hdr (char const *filename); +extern int stbi_is_hdr_from_file(FILE *f); +#endif // STBI_NO_STDIO + + +// get a VERY brief reason for failure +// NOT THREADSAFE +extern const char *stbi_failure_reason (void); + +// free the loaded image -- this is just free() +extern void stbi_image_free (void *retval_from_stbi_load); + +// get image dimensions & components without fully decoding +extern int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); +extern int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); + +#ifndef STBI_NO_STDIO +extern int stbi_info (char const *filename, int *x, int *y, int *comp); +extern int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); + +#endif + + + +// for image formats that explicitly notate that they have premultiplied alpha, +// we just return the colors as stored in the file. set this flag to force +// unpremultiplication. results are undefined if the unpremultiply overflow. +extern void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); + +// indicate whether we should process iphone images back to canonical format, +// or just pass them through "as-is" +extern void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); + + +// ZLIB client - used by PNG, available for other purposes + +extern char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); +extern char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); +extern int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + +extern char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); +extern int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + + +// define faster low-level operations (typically SIMD support) +#ifdef STBI_SIMD +typedef void (*stbi_idct_8x8)(stbi_uc *out, int out_stride, short data[64], unsigned short *dequantize); +// compute an integer IDCT on "input" +// input[x] = data[x] * dequantize[x] +// write results to 'out': 64 samples, each run of 8 spaced by 'out_stride' +// CLAMP results to 0..255 +typedef void (*stbi_YCbCr_to_RGB_run)(stbi_uc *output, stbi_uc const *y, stbi_uc const *cb, stbi_uc const *cr, int count, int step); +// compute a conversion from YCbCr to RGB +// 'count' pixels +// write pixels to 'output'; each pixel is 'step' bytes (either 3 or 4; if 4, write '255' as 4th), order R,G,B +// y: Y input channel +// cb: Cb input channel; scale/biased to be 0..255 +// cr: Cr input channel; scale/biased to be 0..255 + +extern void stbi_install_idct(stbi_idct_8x8 func); +extern void stbi_install_YCbCr_to_RGB(stbi_YCbCr_to_RGB_run func); +#endif // STBI_SIMD + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifndef STBI_HEADER_FILE_ONLY + +#ifndef STBI_NO_HDR +#include // ldexp +#include // strcmp, strtok +#endif + +#ifndef STBI_NO_STDIO +#include +#endif +#include +#include +#include +#include + +#ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif +#else + #define stbi_inline __forceinline +#endif + + +// implementation: +typedef unsigned char uint8; +typedef unsigned short uint16; +typedef signed short int16; +typedef unsigned int uint32; +typedef signed int int32; +typedef unsigned int uint; + +// should produce compiler error if size is wrong +typedef unsigned char validate_uint32[sizeof(uint32)==4 ? 1 : -1]; + +#if defined(STBI_NO_STDIO) && !defined(STBI_NO_WRITE) +#define STBI_NO_WRITE +#endif + +#define STBI_NOTUSED(v) (void)sizeof(v) + +#ifdef _MSC_VER +#define STBI_HAS_LROTL +#endif + +#ifdef STBI_HAS_LROTL + #define stbi_lrot(x,y) _lrotl(x,y) +#else + #define stbi_lrot(x,y) (((x) << (y)) | ((x) >> (32 - (y)))) +#endif + +/////////////////////////////////////////////// +// +// stbi struct and start_xxx functions + +// stbi structure is our basic context used by all images, so it +// contains all the IO context, plus some basic image information +typedef struct +{ + uint32 img_x, img_y; + int img_n, img_out_n; + + stbi_io_callbacks io; + void *io_user_data; + + int read_from_callbacks; + int buflen; + uint8 buffer_start[128]; + + uint8 *img_buffer, *img_buffer_end; + uint8 *img_buffer_original; +} stbi; + + +static void refill_buffer(stbi *s); + +// initialize a memory-decode context +static void start_mem(stbi *s, uint8 const *buffer, int len) +{ + s->io.read = NULL; + s->read_from_callbacks = 0; + s->img_buffer = s->img_buffer_original = (uint8 *) buffer; + s->img_buffer_end = (uint8 *) buffer+len; +} + +// initialize a callback-based context +static void start_callbacks(stbi *s, stbi_io_callbacks *c, void *user) +{ + s->io = *c; + s->io_user_data = user; + s->buflen = sizeof(s->buffer_start); + s->read_from_callbacks = 1; + s->img_buffer_original = s->buffer_start; + refill_buffer(s); +} + +#ifndef STBI_NO_STDIO + +static int stdio_read(void *user, char *data, int size) +{ + return (int) fread(data,1,size,(FILE*) user); +} + +static void stdio_skip(void *user, unsigned n) +{ + fseek((FILE*) user, n, SEEK_CUR); +} + +static int stdio_eof(void *user) +{ + return feof((FILE*) user); +} + +static stbi_io_callbacks stbi_stdio_callbacks = +{ + stdio_read, + stdio_skip, + stdio_eof, +}; + +static void start_file(stbi *s, FILE *f) +{ + start_callbacks(s, &stbi_stdio_callbacks, (void *) f); +} + +//static void stop_file(stbi *s) { } + +#endif // !STBI_NO_STDIO + +static void stbi_rewind(stbi *s) +{ + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 bytes + s->img_buffer = s->img_buffer_original; +} + +static int stbi_jpeg_test(stbi *s); +static stbi_uc *stbi_jpeg_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_jpeg_info(stbi *s, int *x, int *y, int *comp); +static int stbi_png_test(stbi *s); +static stbi_uc *stbi_png_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_png_info(stbi *s, int *x, int *y, int *comp); +static int stbi_bmp_test(stbi *s); +static stbi_uc *stbi_bmp_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_tga_test(stbi *s); +static stbi_uc *stbi_tga_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_tga_info(stbi *s, int *x, int *y, int *comp); +static int stbi_psd_test(stbi *s); +static stbi_uc *stbi_psd_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_hdr_test(stbi *s); +static float *stbi_hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_pic_test(stbi *s); +static stbi_uc *stbi_pic_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_gif_test(stbi *s); +static stbi_uc *stbi_gif_load(stbi *s, int *x, int *y, int *comp, int req_comp); +static int stbi_gif_info(stbi *s, int *x, int *y, int *comp); + + +// this is not threadsafe +static const char *failure_reason; + +const char *stbi_failure_reason(void) +{ + return failure_reason; +} + +static int e(const char *str) +{ + failure_reason = str; + return 0; +} + +// e - error +// epf - error returning pointer to float +// epuc - error returning pointer to unsigned char + +#ifdef STBI_NO_FAILURE_STRINGS + #define e(x,y) 0 +#elif defined(STBI_FAILURE_USERMSG) + #define e(x,y) e(y) +#else + #define e(x,y) e(x) +#endif + +#define epf(x,y) ((float *) (e(x,y)?NULL:NULL)) +#define epuc(x,y) ((unsigned char *) (e(x,y)?NULL:NULL)) + +void stbi_image_free(void *retval_from_stbi_load) +{ + free(retval_from_stbi_load); +} + +#ifndef STBI_NO_HDR +static float *ldr_to_hdr(stbi_uc *data, int x, int y, int comp); +static stbi_uc *hdr_to_ldr(float *data, int x, int y, int comp); +#endif + +static unsigned char *stbi_load_main(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + if (stbi_jpeg_test(s)) return stbi_jpeg_load(s,x,y,comp,req_comp); + if (stbi_png_test(s)) return stbi_png_load(s,x,y,comp,req_comp); + if (stbi_bmp_test(s)) return stbi_bmp_load(s,x,y,comp,req_comp); + if (stbi_gif_test(s)) return stbi_gif_load(s,x,y,comp,req_comp); + if (stbi_psd_test(s)) return stbi_psd_load(s,x,y,comp,req_comp); + if (stbi_pic_test(s)) return stbi_pic_load(s,x,y,comp,req_comp); + + #ifndef STBI_NO_HDR + if (stbi_hdr_test(s)) { + float *hdr = stbi_hdr_load(s, x,y,comp,req_comp); + return hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif + + // test tga last because it's a crappy test! + if (stbi_tga_test(s)) + return stbi_tga_load(s,x,y,comp,req_comp); + return epuc("unknown image type", "Image not of any known type, or corrupt"); +} + +#ifndef STBI_NO_STDIO +unsigned char *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = fopen(filename, "rb"); + unsigned char *result; + if (!f) return epuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +unsigned char *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_file(&s,f); + return stbi_load_main(&s,x,y,comp,req_comp); +} +#endif //!STBI_NO_STDIO + +unsigned char *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_mem(&s,buffer,len); + return stbi_load_main(&s,x,y,comp,req_comp); +} + +unsigned char *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi_load_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_HDR + +float *stbi_loadf_main(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi_hdr_test(s)) + return stbi_hdr_load(s,x,y,comp,req_comp); + #endif + data = stbi_load_main(s, x, y, comp, req_comp); + if (data) + return ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return epf("unknown image type", "Image not of any known type, or corrupt"); +} + +float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_mem(&s,buffer,len); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} + +float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_STDIO +float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = fopen(filename, "rb"); + float *result; + if (!f) return epf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi s; + start_file(&s,f); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} +#endif // !STBI_NO_STDIO + +#endif // !STBI_NO_HDR + +// these is-hdr-or-not is defined independent of whether STBI_NO_HDR is +// defined, for API simplicity; if STBI_NO_HDR is defined, it always +// reports false! + +int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) +{ + #ifndef STBI_NO_HDR + stbi s; + start_mem(&s,buffer,len); + return stbi_hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif +} + +#ifndef STBI_NO_STDIO +extern int stbi_is_hdr (char const *filename) +{ + FILE *f = fopen(filename, "rb"); + int result=0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; +} + +extern int stbi_is_hdr_from_file(FILE *f) +{ + #ifndef STBI_NO_HDR + stbi s; + start_file(&s,f); + return stbi_hdr_test(&s); + #else + return 0; + #endif +} +#endif // !STBI_NO_STDIO + +extern int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) +{ + #ifndef STBI_NO_HDR + stbi s; + start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi_hdr_test(&s); + #else + return 0; + #endif +} + +#ifndef STBI_NO_HDR +static float h2l_gamma_i=1.0f/2.2f, h2l_scale_i=1.0f; +static float l2h_gamma=2.2f, l2h_scale=1.0f; + +void stbi_hdr_to_ldr_gamma(float gamma) { h2l_gamma_i = 1/gamma; } +void stbi_hdr_to_ldr_scale(float scale) { h2l_scale_i = 1/scale; } + +void stbi_ldr_to_hdr_gamma(float gamma) { l2h_gamma = gamma; } +void stbi_ldr_to_hdr_scale(float scale) { l2h_scale = scale; } +#endif + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + SCAN_load=0, + SCAN_type, + SCAN_header +}; + +static void refill_buffer(stbi *s) +{ + int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); + if (n == 0) { + // at end of file, treat same as if from memory + s->read_from_callbacks = 0; + s->img_buffer = s->img_buffer_end-1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } +} + +stbi_inline static int get8(stbi *s) +{ + if (s->img_buffer < s->img_buffer_end) + return *s->img_buffer++; + if (s->read_from_callbacks) { + refill_buffer(s); + return *s->img_buffer++; + } + return 0; +} + +stbi_inline static int at_eof(stbi *s) +{ + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} + +stbi_inline static uint8 get8u(stbi *s) +{ + return (uint8) get8(s); +} + +static void skip(stbi *s, int n) +{ + if (s->io.read) { + int blen = s->img_buffer_end - s->img_buffer; + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} + +static int getn(stbi *s, stbi_uc *buffer, int n) +{ + if (s->io.read) { + int blen = s->img_buffer_end - s->img_buffer; + if (blen < n) { + int res, count; + + memcpy(buffer, s->img_buffer, blen); + + count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen); + res = (count == (n-blen)); + s->img_buffer = s->img_buffer_end; + return res; + } + } + + if (s->img_buffer+n <= s->img_buffer_end) { + memcpy(buffer, s->img_buffer, n); + s->img_buffer += n; + return 1; + } else + return 0; +} + +static int get16(stbi *s) +{ + int z = get8(s); + return (z << 8) + get8(s); +} + +static uint32 get32(stbi *s) +{ + uint32 z = get16(s); + return (z << 16) + get16(s); +} + +static int get16le(stbi *s) +{ + int z = get8(s); + return z + (get8(s) << 8); +} + +static uint32 get32le(stbi *s) +{ + uint32 z = get16le(s); + return z + (get16le(s) << 16); +} + +////////////////////////////////////////////////////////////////////////////// +// +// generic converter from built-in img_n to req_comp +// individual types do this automatically as much as possible (e.g. jpeg +// does all cases internally since it needs to colorspace convert anyway, +// and it never has alpha, so very few cases ). png can automatically +// interleave an alpha=255 channel, but falls back to this for other cases +// +// assume data buffer is malloced, so malloc a new one and free that one +// only failure mode is malloc failing + +static uint8 compute_y(int r, int g, int b) +{ + return (uint8) (((r*77) + (g*150) + (29*b)) >> 8); +} + +static unsigned char *convert_format(unsigned char *data, int img_n, int req_comp, uint x, uint y) +{ + int i,j; + unsigned char *good; + + if (req_comp == img_n) return data; + assert(req_comp >= 1 && req_comp <= 4); + + good = (unsigned char *) malloc(req_comp * x * y); + if (good == NULL) { + free(data); + return epuc("outofmem", "Out of memory"); + } + + for (j=0; j < (int) y; ++j) { + unsigned char *src = data + j * x * img_n ; + unsigned char *dest = good + j * x * req_comp; + + #define COMBO(a,b) ((a)*8+(b)) + #define CASE(a,b) case COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (COMBO(img_n, req_comp)) { + CASE(1,2) dest[0]=src[0], dest[1]=255; break; + CASE(1,3) dest[0]=dest[1]=dest[2]=src[0]; break; + CASE(1,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=255; break; + CASE(2,1) dest[0]=src[0]; break; + CASE(2,3) dest[0]=dest[1]=dest[2]=src[0]; break; + CASE(2,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=src[1]; break; + CASE(3,4) dest[0]=src[0],dest[1]=src[1],dest[2]=src[2],dest[3]=255; break; + CASE(3,1) dest[0]=compute_y(src[0],src[1],src[2]); break; + CASE(3,2) dest[0]=compute_y(src[0],src[1],src[2]), dest[1] = 255; break; + CASE(4,1) dest[0]=compute_y(src[0],src[1],src[2]); break; + CASE(4,2) dest[0]=compute_y(src[0],src[1],src[2]), dest[1] = src[3]; break; + CASE(4,3) dest[0]=src[0],dest[1]=src[1],dest[2]=src[2]; break; + default: assert(0); + } + #undef CASE + } + + free(data); + return good; +} + +#ifndef STBI_NO_HDR +static float *ldr_to_hdr(stbi_uc *data, int x, int y, int comp) +{ + int i,k,n; + float *output = (float *) malloc(x * y * comp * sizeof(float)); + if (output == NULL) { free(data); return epf("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + output[i*comp + k] = (float) pow(data[i*comp+k]/255.0f, l2h_gamma) * l2h_scale; + } + if (k < comp) output[i*comp + k] = data[i*comp+k]/255.0f; + } + free(data); + return output; +} + +#define float2int(x) ((int) (x)) +static stbi_uc *hdr_to_ldr(float *data, int x, int y, int comp) +{ + int i,k,n; + stbi_uc *output = (stbi_uc *) malloc(x * y * comp); + if (output == NULL) { free(data); return epuc("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + float z = (float) pow(data[i*comp+k]*h2l_scale_i, h2l_gamma_i) * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (uint8) float2int(z); + } + if (k < comp) { + float z = data[i*comp+k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (uint8) float2int(z); + } + } + free(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder (not actually fully baseline implementation) +// +// simple implementation +// - channel subsampling of at most 2 in each dimension +// - doesn't support delayed output of y-dimension +// - simple interface (only one output format: 8-bit interleaved RGB) +// - doesn't try to recover corrupt jpegs +// - doesn't allow partial loading, loading multiple at once +// - still fast on x86 (copying globals into locals doesn't help x86) +// - allocates lots of intermediate memory (full size of all components) +// - non-interleaved case requires this anyway +// - allows good upsampling (see next) +// high-quality +// - upsampled channels are bilinearly interpolated, even across blocks +// - quality integer IDCT derived from IJG's 'slow' +// performance +// - fast huffman; reasonable integer IDCT +// - uses a lot of intermediate memory, could cache poorly +// - load http://nothings.org/remote/anemones.jpg 3 times on 2.8Ghz P4 +// stb_jpeg: 1.34 seconds (MSVC6, default release build) +// stb_jpeg: 1.06 seconds (MSVC6, processor = Pentium Pro) +// IJL11.dll: 1.08 seconds (compiled by intel) +// IJG 1998: 0.98 seconds (MSVC6, makefile provided by IJG) +// IJG 1998: 0.95 seconds (MSVC6, makefile + proc=PPro) + +// huffman decoding acceleration +#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + +typedef struct +{ + uint8 fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + uint16 code[256]; + uint8 values[256]; + uint8 size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' +} huffman; + +typedef struct +{ + #ifdef STBI_SIMD + unsigned short dequant2[4][64]; + #endif + stbi *s; + huffman huff_dc[4]; + huffman huff_ac[4]; + uint8 dequant[4][64]; + +// sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + +// definition of jpeg image component + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + uint8 *data; + void *raw_data; + uint8 *linebuf; + } img_comp[4]; + + uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop + + int scan_n, order[4]; + int restart_interval, todo; +} jpeg; + +static int build_huffman(huffman *h, int *count) +{ + int i,j,k=0,code; + // build size list for each symbol (from JPEG spec) + for (i=0; i < 16; ++i) + for (j=0; j < count[i]; ++j) + h->size[k++] = (uint8) (i+1); + h->size[k] = 0; + + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (uint16) (code++); + if (code-1 >= (1 << j)) return e("bad code lengths","Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i=0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS-s); + int m = 1 << (FAST_BITS-s); + for (j=0; j < m; ++j) { + h->fast[c+j] = (uint8) i; + } + } + } + return 1; +} + +static void grow_buffer_unsafe(jpeg *j) +{ + do { + int b = j->nomore ? 0 : get8(j->s); + if (b == 0xff) { + int c = get8(j->s); + if (c != 0) { + j->marker = (unsigned char) c; + j->nomore = 1; + return; + } + } + j->code_buffer |= b << (24 - j->code_bits); + j->code_bits += 8; + } while (j->code_bits <= 24); +} + +// (1 << n) - 1 +static uint32 bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +// decode a jpeg huffman value from the bitstream +stbi_inline static int decode(jpeg *j, huffman *h) +{ + unsigned int temp; + int c,k; + + if (j->code_bits < 16) grow_buffer_unsafe(j); + + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) + return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } + + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & bmask[k]) + h->delta[k]; + assert((((j->code_buffer) >> (32 - h->size[c])) & bmask[h->size[c]]) == h->code[c]); + + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +// combined JPEG 'receive' and JPEG 'extend', since baseline +// always extends everything it receives. +stbi_inline static int extend_receive(jpeg *j, int n) +{ + unsigned int m = 1 << (n-1); + unsigned int k; + if (j->code_bits < n) grow_buffer_unsafe(j); + + #if 1 + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~bmask[n]; + k &= bmask[n]; + j->code_bits -= n; + #else + k = (j->code_buffer >> (32 - n)) & bmask[n]; + j->code_bits -= n; + j->code_buffer <<= n; + #endif + // the following test is probably a random branch that won't + // predict well. I tried to table accelerate it but failed. + // maybe it's compiling as a conditional move? + if (k < m) + return (-1 << n) + k + 1; + else + return k; +} + +// given a value that's at position X in the zigzag stream, +// where does it appear in the 8x8 matrix coded as row-major? +static uint8 dezigzag[64+15] = +{ + 0, 1, 8, 16, 9, 2, 3, 10, + 17, 24, 32, 25, 18, 11, 4, 5, + 12, 19, 26, 33, 40, 48, 41, 34, + 27, 20, 13, 6, 7, 14, 21, 28, + 35, 42, 49, 56, 57, 50, 43, 36, + 29, 22, 15, 23, 30, 37, 44, 51, + 58, 59, 52, 45, 38, 31, 39, 46, + 53, 60, 61, 54, 47, 55, 62, 63, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +// decode one 64-entry block-- +static int decode_block(jpeg *j, short data[64], huffman *hdc, huffman *hac, int b) +{ + int diff,dc,k; + int t = decode(j, hdc); + if (t < 0) return e("bad huffman code","Corrupt JPEG"); + + // 0 all the ac values now so we can do it 32-bits at a time + memset(data,0,64*sizeof(data[0])); + + diff = t ? extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short) dc; + + // decode AC components, see JPEG spec + k = 1; + do { + int r,s; + int rs = decode(j, hac); + if (rs < 0) return e("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } else { + k += r; + // decode into unzigzag'd location + data[dezigzag[k++]] = (short) extend_receive(j,s); + } + } while (k < 64); + return 1; +} + +// take a -128..127 value and clamp it and convert to 0..255 +stbi_inline static uint8 clamp(int x) +{ + // trick to use a single test to catch both cases + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (uint8) x; +} + +#define f2f(x) (int) (((x) * 4096 + 0.5)) +#define fsh(x) ((x) << 12) + +// derived from jidctint -- DCT_ISLOW +#define IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7) \ + int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \ + p2 = s2; \ + p3 = s6; \ + p1 = (p2+p3) * f2f(0.5411961f); \ + t2 = p1 + p3*f2f(-1.847759065f); \ + t3 = p1 + p2*f2f( 0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = fsh(p2+p3); \ + t1 = fsh(p2-p3); \ + x0 = t0+t3; \ + x3 = t0-t3; \ + x1 = t1+t2; \ + x2 = t1-t2; \ + t0 = s7; \ + t1 = s5; \ + t2 = s3; \ + t3 = s1; \ + p3 = t0+t2; \ + p4 = t1+t3; \ + p1 = t0+t3; \ + p2 = t1+t2; \ + p5 = (p3+p4)*f2f( 1.175875602f); \ + t0 = t0*f2f( 0.298631336f); \ + t1 = t1*f2f( 2.053119869f); \ + t2 = t2*f2f( 3.072711026f); \ + t3 = t3*f2f( 1.501321110f); \ + p1 = p5 + p1*f2f(-0.899976223f); \ + p2 = p5 + p2*f2f(-2.562915447f); \ + p3 = p3*f2f(-1.961570560f); \ + p4 = p4*f2f(-0.390180644f); \ + t3 += p1+p4; \ + t2 += p2+p3; \ + t1 += p2+p4; \ + t0 += p1+p3; + +#ifdef STBI_SIMD +typedef unsigned short stbi_dequantize_t; +#else +typedef uint8 stbi_dequantize_t; +#endif + +// .344 seconds on 3*anemones.jpg +static void idct_block(uint8 *out, int out_stride, short data[64], stbi_dequantize_t *dequantize) +{ + int i,val[64],*v=val; + stbi_dequantize_t *dq = dequantize; + uint8 *o; + short *d = data; + + // columns + for (i=0; i < 8; ++i,++d,++dq, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0] * dq[0] << 2; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + IDCT_1D(d[ 0]*dq[ 0],d[ 8]*dq[ 8],d[16]*dq[16],d[24]*dq[24], + d[32]*dq[32],d[40]*dq[40],d[48]*dq[48],d[56]*dq[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; x1 += 512; x2 += 512; x3 += 512; + v[ 0] = (x0+t3) >> 10; + v[56] = (x0-t3) >> 10; + v[ 8] = (x1+t2) >> 10; + v[48] = (x1-t2) >> 10; + v[16] = (x2+t1) >> 10; + v[40] = (x2-t1) >> 10; + v[24] = (x3+t0) >> 10; + v[32] = (x3-t0) >> 10; + } + } + + for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { + // no fast case since the first 1D IDCT spread components out + IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = clamp((x0+t3) >> 17); + o[7] = clamp((x0-t3) >> 17); + o[1] = clamp((x1+t2) >> 17); + o[6] = clamp((x1-t2) >> 17); + o[2] = clamp((x2+t1) >> 17); + o[5] = clamp((x2-t1) >> 17); + o[3] = clamp((x3+t0) >> 17); + o[4] = clamp((x3-t0) >> 17); + } +} + +#ifdef STBI_SIMD +static stbi_idct_8x8 stbi_idct_installed = idct_block; + +void stbi_install_idct(stbi_idct_8x8 func) +{ + stbi_idct_installed = func; +} +#endif + +#define MARKER_none 0xff +// if there's a pending marker from the entropy stream, return that +// otherwise, fetch from the stream and get a marker. if there's no +// marker, return 0xff, which is never a valid marker value +static uint8 get_marker(jpeg *j) +{ + uint8 x; + if (j->marker != MARKER_none) { x = j->marker; j->marker = MARKER_none; return x; } + x = get8u(j->s); + if (x != 0xff) return MARKER_none; + while (x == 0xff) + x = get8u(j->s); + return x; +} + +// in each scan, we'll have scan_n components, and the order +// of the components is specified by order[] +#define RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +// after a restart interval, reset the entropy decoder and +// the dc prediction +static void reset(jpeg *j) +{ + j->code_bits = 0; + j->code_buffer = 0; + j->nomore = 0; + j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = 0; + j->marker = MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels +} + +static int parse_entropy_coded_data(jpeg *z) +{ + reset(z); + if (z->scan_n == 1) { + int i,j; + #ifdef STBI_SIMD + __declspec(align(16)) + #endif + short data[64]; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + if (!decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0; + #ifdef STBI_SIMD + stbi_idct_installed(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]); + #else + idct_block(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]); + #endif + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!RESTART(z->marker)) return 1; + reset(z); + } + } + } + } else { // interleaved! + int i,j,k,x,y; + short data[64]; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x)*8; + int y2 = (j*z->img_comp[n].v + y)*8; + if (!decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0; + #ifdef STBI_SIMD + stbi_idct_installed(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]); + #else + idct_block(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]); + #endif + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!RESTART(z->marker)) return 1; + reset(z); + } + } + } + } + return 1; +} + +static int process_marker(jpeg *z, int m) +{ + int L; + switch (m) { + case MARKER_none: // no marker found + return e("expected marker","Corrupt JPEG"); + + case 0xC2: // SOF - progressive + return e("progressive jpeg","JPEG format not supported (progressive)"); + + case 0xDD: // DRI - specify restart interval + if (get16(z->s) != 4) return e("bad DRI len","Corrupt JPEG"); + z->restart_interval = get16(z->s); + return 1; + + case 0xDB: // DQT - define quantization table + L = get16(z->s)-2; + while (L > 0) { + int q = get8(z->s); + int p = q >> 4; + int t = q & 15,i; + if (p != 0) return e("bad DQT type","Corrupt JPEG"); + if (t > 3) return e("bad DQT table","Corrupt JPEG"); + for (i=0; i < 64; ++i) + z->dequant[t][dezigzag[i]] = get8u(z->s); + #ifdef STBI_SIMD + for (i=0; i < 64; ++i) + z->dequant2[t][i] = z->dequant[t][i]; + #endif + L -= 65; + } + return L==0; + + case 0xC4: // DHT - define huffman table + L = get16(z->s)-2; + while (L > 0) { + uint8 *v; + int sizes[16],i,m=0; + int q = get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) return e("bad DHT header","Corrupt JPEG"); + for (i=0; i < 16; ++i) { + sizes[i] = get8(z->s); + m += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!build_huffman(z->huff_dc+th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!build_huffman(z->huff_ac+th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i=0; i < m; ++i) + v[i] = get8u(z->s); + L -= m; + } + return L==0; + } + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + skip(z->s, get16(z->s)-2); + return 1; + } + return 0; +} + +// after we see SOS +static int process_scan_header(jpeg *z) +{ + int i; + int Ls = get16(z->s); + z->scan_n = get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return e("bad SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return e("bad SOS len","Corrupt JPEG"); + for (i=0; i < z->scan_n; ++i) { + int id = get8(z->s), which; + int q = get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return e("bad DC huff","Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return e("bad AC huff","Corrupt JPEG"); + z->order[i] = which; + } + if (get8(z->s) != 0) return e("bad SOS","Corrupt JPEG"); + get8(z->s); // should be 63, but might be 0 + if (get8(z->s) != 0) return e("bad SOS","Corrupt JPEG"); + + return 1; +} + +static int process_frame_header(jpeg *z, int scan) +{ + stbi *s = z->s; + int Lf,p,i,q, h_max=1,v_max=1,c; + Lf = get16(s); if (Lf < 11) return e("bad SOF len","Corrupt JPEG"); // JPEG + p = get8(s); if (p != 8) return e("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = get16(s); if (s->img_y == 0) return e("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG + s->img_x = get16(s); if (s->img_x == 0) return e("0 width","Corrupt JPEG"); // JPEG requires + c = get8(s); + if (c != 3 && c != 1) return e("bad component count","Corrupt JPEG"); // JFIF requires + s->img_n = c; + for (i=0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8+3*s->img_n) return e("bad SOF len","Corrupt JPEG"); + + for (i=0; i < s->img_n; ++i) { + z->img_comp[i].id = get8(s); + if (z->img_comp[i].id != i+1) // JFIF requires + if (z->img_comp[i].id != i) // some version of jpegtran outputs non-JFIF-compliant files! + return e("bad component ID","Corrupt JPEG"); + q = get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return e("bad H","Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return e("bad V","Corrupt JPEG"); + z->img_comp[i].tq = get8(s); if (z->img_comp[i].tq > 3) return e("bad TQ","Corrupt JPEG"); + } + + if (scan != SCAN_load) return 1; + + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return e("too large", "Image too large to decode"); + + for (i=0; i < s->img_n; ++i) { + if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h; + if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; + + for (i=0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].raw_data = malloc(z->img_comp[i].w2 * z->img_comp[i].h2+15); + if (z->img_comp[i].raw_data == NULL) { + for(--i; i >= 0; --i) { + free(z->img_comp[i].raw_data); + z->img_comp[i].data = NULL; + } + return e("outofmem", "Out of memory"); + } + // align blocks for installable-idct using mmx/sse + z->img_comp[i].data = (uint8*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + z->img_comp[i].linebuf = NULL; + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. SOF) +#define DNL(x) ((x) == 0xdc) +#define SOI(x) ((x) == 0xd8) +#define EOI(x) ((x) == 0xd9) +#define SOF(x) ((x) == 0xc0 || (x) == 0xc1) +#define SOS(x) ((x) == 0xda) + +static int decode_jpeg_header(jpeg *z, int scan) +{ + int m; + z->marker = MARKER_none; // initialize cached marker to empty + m = get_marker(z); + if (!SOI(m)) return e("no SOI","Corrupt JPEG"); + if (scan == SCAN_type) return 1; + m = get_marker(z); + while (!SOF(m)) { + if (!process_marker(z,m)) return 0; + m = get_marker(z); + while (m == MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (at_eof(z->s)) return e("no SOF", "Corrupt JPEG"); + m = get_marker(z); + } + } + if (!process_frame_header(z, scan)) return 0; + return 1; +} + +static int decode_jpeg_image(jpeg *j) +{ + int m; + j->restart_interval = 0; + if (!decode_jpeg_header(j, SCAN_load)) return 0; + m = get_marker(j); + while (!EOI(m)) { + if (SOS(m)) { + if (!process_scan_header(j)) return 0; + if (!parse_entropy_coded_data(j)) return 0; + if (j->marker == MARKER_none ) { + // handle 0s at the end of image data from IP Kamera 9060 + while (!at_eof(j->s)) { + int x = get8(j->s); + if (x == 255) { + j->marker = get8u(j->s); + break; + } else if (x != 0) { + return 0; + } + } + // if we reach eof without hitting a marker, get_marker() below will fail and we'll eventually return 0 + } + } else { + if (!process_marker(j, m)) return 0; + } + m = get_marker(j); + } + return 1; +} + +// static jfif-centered resampling (across block boundaries) + +typedef uint8 *(*resample_row_func)(uint8 *out, uint8 *in0, uint8 *in1, + int w, int hs); + +#define div4(x) ((uint8) ((x) >> 2)) + +static uint8 *resample_row_1(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static uint8* resample_row_v_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs) +{ + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static uint8* resample_row_h_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs) +{ + // need to generate two samples horizontally for every one in input + int i; + uint8 *input = in_near; + + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = div4(input[0]*3 + input[1] + 2); + for (i=1; i < w-1; ++i) { + int n = 3*input[i]+2; + out[i*2+0] = div4(n+input[i-1]); + out[i*2+1] = div4(n+input[i+1]); + } + out[i*2+0] = div4(input[w-2]*3 + input[w-1] + 2); + out[i*2+1] = input[w-1]; + + STBI_NOTUSED(in_far); + STBI_NOTUSED(hs); + + return out; +} + +#define div16(x) ((uint8) ((x) >> 4)) + +static uint8 *resample_row_hv_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + out[0] = div4(t1+2); + for (i=1; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = div16(3*t0 + t1 + 8); + out[i*2 ] = div16(3*t1 + t0 + 8); + } + out[w*2-1] = div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} + +static uint8 *resample_row_generic(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs) +{ + // resample with nearest-neighbor + int i,j; + in_far = in_far; + for (i=0; i < w; ++i) + for (j=0; j < hs; ++j) + out[i*hs+j] = in_near[i]; + return out; +} + +#define float2fixed(x) ((int) ((x) * 65536 + 0.5)) + +// 0.38 seconds on 3*anemones.jpg (0.25 with processor = Pro) +// VC6 without processor=Pro is generating multiple LEAs per multiply! +static void YCbCr_to_RGB_row(uint8 *out, const uint8 *y, const uint8 *pcb, const uint8 *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 16) + 32768; // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr*float2fixed(1.40200f); + g = y_fixed - cr*float2fixed(0.71414f) - cb*float2fixed(0.34414f); + b = y_fixed + cb*float2fixed(1.77200f); + r >>= 16; + g >>= 16; + b >>= 16; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (uint8)r; + out[1] = (uint8)g; + out[2] = (uint8)b; + out[3] = 255; + out += step; + } +} + +#ifdef STBI_SIMD +static stbi_YCbCr_to_RGB_run stbi_YCbCr_installed = YCbCr_to_RGB_row; + +void stbi_install_YCbCr_to_RGB(stbi_YCbCr_to_RGB_run func) +{ + stbi_YCbCr_installed = func; +} +#endif + + +// clean up the temporary component buffers +static void cleanup_jpeg(jpeg *j) +{ + int i; + for (i=0; i < j->s->img_n; ++i) { + if (j->img_comp[i].data) { + free(j->img_comp[i].raw_data); + j->img_comp[i].data = NULL; + } + if (j->img_comp[i].linebuf) { + free(j->img_comp[i].linebuf); + j->img_comp[i].linebuf = NULL; + } + } +} + +typedef struct +{ + resample_row_func resample; + uint8 *line0,*line1; + int hs,vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on +} stbi_resample; + +static uint8 *load_jpeg_image(jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) +{ + int n, decode_n; + // validate req_comp + if (req_comp < 0 || req_comp > 4) return epuc("bad req_comp", "Internal error"); + z->s->img_n = 0; + + // load a jpeg image from whichever source + if (!decode_jpeg_image(z)) { cleanup_jpeg(z); return NULL; } + + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n; + + if (z->s->img_n == 3 && n < 3) + decode_n = 1; + else + decode_n = z->s->img_n; + + // resample and color-convert + { + int k; + uint i,j; + uint8 *output; + uint8 *coutput[4]; + + stbi_resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi_resample *r = &res_comp[k]; + + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (uint8 *) malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { cleanup_jpeg(z); return epuc("outofmem", "Out of memory"); } + + r->hs = z->img_h_max / z->img_comp[k].h; + r->vs = z->img_v_max / z->img_comp[k].v; + r->ystep = r->vs >> 1; + r->w_lores = (z->s->img_x + r->hs-1) / r->hs; + r->ypos = 0; + r->line0 = r->line1 = z->img_comp[k].data; + + if (r->hs == 1 && r->vs == 1) r->resample = resample_row_1; + else if (r->hs == 1 && r->vs == 2) r->resample = resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) r->resample = resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) r->resample = resample_row_hv_2; + else r->resample = resample_row_generic; + } + + // can't error after this so, this is safe + output = (uint8 *) malloc(n * z->s->img_x * z->s->img_y + 1); + if (!output) { cleanup_jpeg(z); return epuc("outofmem", "Out of memory"); } + + // now go ahead and resample + for (j=0; j < z->s->img_y; ++j) { + uint8 *out = output + n * z->s->img_x * j; + for (k=0; k < decode_n; ++k) { + stbi_resample *r = &res_comp[k]; + int y_bot = r->ystep >= (r->vs >> 1); + coutput[k] = r->resample(z->img_comp[k].linebuf, + y_bot ? r->line1 : r->line0, + y_bot ? r->line0 : r->line1, + r->w_lores, r->hs); + if (++r->ystep >= r->vs) { + r->ystep = 0; + r->line0 = r->line1; + if (++r->ypos < z->img_comp[k].y) + r->line1 += z->img_comp[k].w2; + } + } + if (n >= 3) { + uint8 *y = coutput[0]; + if (z->s->img_n == 3) { + #ifdef STBI_SIMD + stbi_YCbCr_installed(out, y, coutput[1], coutput[2], z->s.img_x, n); + #else + YCbCr_to_RGB_row(out, y, coutput[1], coutput[2], z->s->img_x, n); + #endif + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + uint8 *y = coutput[0]; + if (n == 1) + for (i=0; i < z->s->img_x; ++i) out[i] = y[i]; + else + for (i=0; i < z->s->img_x; ++i) *out++ = y[i], *out++ = 255; + } + } + cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) *comp = z->s->img_n; // report original components, not output + return output; + } +} + +static unsigned char *stbi_jpeg_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + jpeg j; + j.s = s; + return load_jpeg_image(&j, x,y,comp,req_comp); +} + +static int stbi_jpeg_test(stbi *s) +{ + int r; + jpeg j; + j.s = s; + r = decode_jpeg_header(&j, SCAN_type); + stbi_rewind(s); + return r; +} + +static int stbi_jpeg_info_raw(jpeg *j, int *x, int *y, int *comp) +{ + if (!decode_jpeg_header(j, SCAN_header)) { + stbi_rewind( j->s ); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n; + return 1; +} + +static int stbi_jpeg_info(stbi *s, int *x, int *y, int *comp) +{ + jpeg j; + j.s = s; + return stbi_jpeg_info_raw(&j, x, y, comp); +} + +// public domain zlib decode v0.2 Sean Barrett 2006-11-18 +// simple implementation +// - all input must be provided in an upfront buffer +// - all output is written to a single output buffer (can malloc/realloc) +// performance +// - fast huffman + +// fast-way is faster to check than jpeg huffman, but slow way is slower +#define ZFAST_BITS 9 // accelerate all cases in default tables +#define ZFAST_MASK ((1 << ZFAST_BITS) - 1) + +// zlib-style huffman encoding +// (jpegs packs from left, zlib from right, so can't share code) +typedef struct +{ + uint16 fast[1 << ZFAST_BITS]; + uint16 firstcode[16]; + int maxcode[17]; + uint16 firstsymbol[16]; + uint8 size[288]; + uint16 value[288]; +} zhuffman; + +stbi_inline static int bitreverse16(int n) +{ + n = ((n & 0xAAAA) >> 1) | ((n & 0x5555) << 1); + n = ((n & 0xCCCC) >> 2) | ((n & 0x3333) << 2); + n = ((n & 0xF0F0) >> 4) | ((n & 0x0F0F) << 4); + n = ((n & 0xFF00) >> 8) | ((n & 0x00FF) << 8); + return n; +} + +stbi_inline static int bit_reverse(int v, int bits) +{ + assert(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return bitreverse16(v) >> (16-bits); +} + +static int zbuild_huffman(zhuffman *z, uint8 *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 255, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + assert(sizes[i] <= (1 << i)); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (uint16) code; + z->firstsymbol[i] = (uint16) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return e("bad codelengths","Corrupt JPEG"); + z->maxcode[i] = code << (16-i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i=0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + z->size[c] = (uint8)s; + z->value[c] = (uint16)i; + if (s <= ZFAST_BITS) { + int k = bit_reverse(next_code[s],s); + while (k < (1 << ZFAST_BITS)) { + z->fast[k] = (uint16) c; + k += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +// zlib-from-memory implementation for PNG reading +// because PNG allows splitting the zlib stream arbitrarily, +// and it's annoying structurally to have PNG call ZLIB call PNG, +// we require PNG read all the IDATs and combine them into a single +// memory buffer + +typedef struct +{ + uint8 *zbuffer, *zbuffer_end; + int num_bits; + uint32 code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + zhuffman z_length, z_distance; +} zbuf; + +stbi_inline static int zget8(zbuf *z) +{ + if (z->zbuffer >= z->zbuffer_end) return 0; + return *z->zbuffer++; +} + +static void fill_bits(zbuf *z) +{ + do { + assert(z->code_buffer < (1U << z->num_bits)); + z->code_buffer |= zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); +} + +stbi_inline static unsigned int zreceive(zbuf *z, int n) +{ + unsigned int k; + if (z->num_bits < n) fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; +} + +stbi_inline static int zhuffman_decode(zbuf *a, zhuffman *z) +{ + int b,s,k; + if (a->num_bits < 16) fill_bits(a); + b = z->fast[a->code_buffer & ZFAST_MASK]; + if (b < 0xffff) { + s = z->size[b]; + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; + } + + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = bit_reverse(a->code_buffer, 16); + for (s=ZFAST_BITS+1; ; ++s) + if (k < z->maxcode[s]) + break; + if (s == 16) return -1; // invalid code! + // code size is s, so: + b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; + assert(z->size[b] == s); + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; +} + +static int expand(zbuf *z, int n) // need to make room for n bytes +{ + char *q; + int cur, limit; + if (!z->z_expandable) return e("output buffer limit","Corrupt PNG"); + cur = (int) (z->zout - z->zout_start); + limit = (int) (z->zout_end - z->zout_start); + while (cur + n > limit) + limit *= 2; + q = (char *) realloc(z->zout_start, limit); + if (q == NULL) return e("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; +} + +static int length_base[31] = { + 3,4,5,6,7,8,9,10,11,13, + 15,17,19,23,27,31,35,43,51,59, + 67,83,99,115,131,163,195,227,258,0,0 }; + +static int length_extra[31]= +{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 }; + +static int dist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193, +257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0}; + +static int dist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +static int parse_huffman_block(zbuf *a) +{ + for(;;) { + int z = zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return e("bad huffman code","Corrupt PNG"); // error in huffman codes + if (a->zout >= a->zout_end) if (!expand(a, 1)) return 0; + *a->zout++ = (char) z; + } else { + uint8 *p; + int len,dist; + if (z == 256) return 1; + z -= 257; + len = length_base[z]; + if (length_extra[z]) len += zreceive(a, length_extra[z]); + z = zhuffman_decode(a, &a->z_distance); + if (z < 0) return e("bad huffman code","Corrupt PNG"); + dist = dist_base[z]; + if (dist_extra[z]) dist += zreceive(a, dist_extra[z]); + if (a->zout - a->zout_start < dist) return e("bad dist","Corrupt PNG"); + if (a->zout + len > a->zout_end) if (!expand(a, len)) return 0; + p = (uint8 *) (a->zout - dist); + while (len--) + *a->zout++ = *p++; + } + } +} + +static int compute_huffman_codes(zbuf *a) +{ + static uint8 length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + zhuffman z_codelength; + uint8 lencodes[286+32+137];//padding for maximum single op + uint8 codelength_sizes[19]; + int i,n; + + int hlit = zreceive(a,5) + 257; + int hdist = zreceive(a,5) + 1; + int hclen = zreceive(a,4) + 4; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (uint8) s; + } + if (!zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < hlit + hdist) { + int c = zhuffman_decode(a, &z_codelength); + assert(c >= 0 && c < 19); + if (c < 16) + lencodes[n++] = (uint8) c; + else if (c == 16) { + c = zreceive(a,2)+3; + memset(lencodes+n, lencodes[n-1], c); + n += c; + } else if (c == 17) { + c = zreceive(a,3)+3; + memset(lencodes+n, 0, c); + n += c; + } else { + assert(c == 18); + c = zreceive(a,7)+11; + memset(lencodes+n, 0, c); + n += c; + } + } + if (n != hlit+hdist) return e("bad codelengths","Corrupt PNG"); + if (!zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; + return 1; +} + +static int parse_uncompressed_block(zbuf *a) +{ + uint8 header[4]; + int len,nlen,k; + if (a->num_bits & 7) + zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = (uint8) (a->code_buffer & 255); // wtf this warns? + a->code_buffer >>= 8; + a->num_bits -= 8; + } + assert(a->num_bits == 0); + // now fill header the normal way + while (k < 4) + header[k++] = (uint8) zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return e("zlib corrupt","Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) return e("read past buffer","Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!expand(a, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int parse_zlib_header(zbuf *a) +{ + int cmf = zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = zget8(a); + if ((cmf*256+flg) % 31 != 0) return e("bad zlib header","Corrupt PNG"); // zlib spec + if (flg & 32) return e("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) return e("bad compression","Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; +} + +// @TODO: should statically initialize these for optimal thread safety +static uint8 default_length[288], default_distance[32]; +static void init_defaults(void) +{ + int i; // use <= to match clearly with spec + for (i=0; i <= 143; ++i) default_length[i] = 8; + for ( ; i <= 255; ++i) default_length[i] = 9; + for ( ; i <= 279; ++i) default_length[i] = 7; + for ( ; i <= 287; ++i) default_length[i] = 8; + + for (i=0; i <= 31; ++i) default_distance[i] = 5; +} + +int stbi_png_partial; // a quick hack to only allow decoding some of a PNG... I should implement real streaming support instead +static int parse_zlib(zbuf *a, int parse_header) +{ + int final, type; + if (parse_header) + if (!parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = zreceive(a,1); + type = zreceive(a,2); + if (type == 0) { + if (!parse_uncompressed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + // use fixed code lengths + if (!default_distance[31]) init_defaults(); + if (!zbuild_huffman(&a->z_length , default_length , 288)) return 0; + if (!zbuild_huffman(&a->z_distance, default_distance, 32)) return 0; + } else { + if (!compute_huffman_codes(a)) return 0; + } + if (!parse_huffman_block(a)) return 0; + } + if (stbi_png_partial && a->zout - a->zout_start > 65536) + break; + } while (!final); + return 1; +} + +static int do_zlib(zbuf *a, char *obuf, int olen, int exp, int parse_header) +{ + a->zout_start = obuf; + a->zout = obuf; + a->zout_end = obuf + olen; + a->z_expandable = exp; + + return parse_zlib(a, parse_header); +} + +char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen) +{ + zbuf a; + char *p = (char *) malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (uint8 *) buffer; + a.zbuffer_end = (uint8 *) buffer + len; + if (do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + free(a.zout_start); + return NULL; + } +} + +char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) +{ + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +} + +char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) +{ + zbuf a; + char *p = (char *) malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (uint8 *) buffer; + a.zbuffer_end = (uint8 *) buffer + len; + if (do_zlib(&a, p, initial_size, 1, parse_header)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + free(a.zout_start); + return NULL; + } +} + +int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) +{ + zbuf a; + a.zbuffer = (uint8 *) ibuffer; + a.zbuffer_end = (uint8 *) ibuffer + ilen; + if (do_zlib(&a, obuffer, olen, 0, 1)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) +{ + zbuf a; + char *p = (char *) malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (uint8 *) buffer; + a.zbuffer_end = (uint8 *) buffer+len; + if (do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + free(a.zout_start); + return NULL; + } +} + +int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) +{ + zbuf a; + a.zbuffer = (uint8 *) ibuffer; + a.zbuffer_end = (uint8 *) ibuffer + ilen; + if (do_zlib(&a, obuffer, olen, 0, 0)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +// public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 +// simple implementation +// - only 8-bit samples +// - no CRC checking +// - allocates lots of intermediate memory +// - avoids problem of streaming data between subsystems +// - avoids explicit window management +// performance +// - uses stb_zlib, a PD zlib implementation with fast huffman decoding + + +typedef struct +{ + uint32 length; + uint32 type; +} chunk; + +#define PNG_TYPE(a,b,c,d) (((a) << 24) + ((b) << 16) + ((c) << 8) + (d)) + +static chunk get_chunk_header(stbi *s) +{ + chunk c; + c.length = get32(s); + c.type = get32(s); + return c; +} + +static int check_png_header(stbi *s) +{ + static uint8 png_sig[8] = { 137,80,78,71,13,10,26,10 }; + int i; + for (i=0; i < 8; ++i) + if (get8u(s) != png_sig[i]) return e("bad png sig","Not a PNG"); + return 1; +} + +typedef struct +{ + stbi *s; + uint8 *idata, *expanded, *out; +} png; + + +enum { + F_none=0, F_sub=1, F_up=2, F_avg=3, F_paeth=4, + F_avg_first, F_paeth_first +}; + +static uint8 first_row_filter[5] = +{ + F_none, F_sub, F_none, F_avg_first, F_paeth_first +}; + +static int paeth(int a, int b, int c) +{ + int p = a + b - c; + int pa = abs(p-a); + int pb = abs(p-b); + int pc = abs(p-c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; +} + +// create the png data from post-deflated data +static int create_png_image_raw(png *a, uint8 *raw, uint32 raw_len, int out_n, uint32 x, uint32 y) +{ + stbi *s = a->s; + uint32 i,j,stride = x*out_n; + int k; + int img_n = s->img_n; // copy it into a local for later + assert(out_n == s->img_n || out_n == s->img_n+1); + if (stbi_png_partial) y = 1; + a->out = (uint8 *) malloc(x * y * out_n); + if (!a->out) return e("outofmem", "Out of memory"); + if (!stbi_png_partial) { + if (s->img_x == x && s->img_y == y) { + if (raw_len != (img_n * x + 1) * y) return e("not enough pixels","Corrupt PNG"); + } else { // interlaced: + if (raw_len < (img_n * x + 1) * y) return e("not enough pixels","Corrupt PNG"); + } + } + for (j=0; j < y; ++j) { + uint8 *cur = a->out + stride*j; + uint8 *prior = cur - stride; + int filter = *raw++; + if (filter > 4) return e("invalid filter","Corrupt PNG"); + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + // handle first pixel explicitly + for (k=0; k < img_n; ++k) { + switch (filter) { + case F_none : cur[k] = raw[k]; break; + case F_sub : cur[k] = raw[k]; break; + case F_up : cur[k] = raw[k] + prior[k]; break; + case F_avg : cur[k] = raw[k] + (prior[k]>>1); break; + case F_paeth : cur[k] = (uint8) (raw[k] + paeth(0,prior[k],0)); break; + case F_avg_first : cur[k] = raw[k]; break; + case F_paeth_first: cur[k] = raw[k]; break; + } + } + if (img_n != out_n) cur[img_n] = 255; + raw += img_n; + cur += out_n; + prior += out_n; + // this is a little gross, so that we don't switch per-pixel or per-component + if (img_n == out_n) { + #define CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, raw+=img_n,cur+=img_n,prior+=img_n) \ + for (k=0; k < img_n; ++k) + switch (filter) { + CASE(F_none) cur[k] = raw[k]; break; + CASE(F_sub) cur[k] = raw[k] + cur[k-img_n]; break; + CASE(F_up) cur[k] = raw[k] + prior[k]; break; + CASE(F_avg) cur[k] = raw[k] + ((prior[k] + cur[k-img_n])>>1); break; + CASE(F_paeth) cur[k] = (uint8) (raw[k] + paeth(cur[k-img_n],prior[k],prior[k-img_n])); break; + CASE(F_avg_first) cur[k] = raw[k] + (cur[k-img_n] >> 1); break; + CASE(F_paeth_first) cur[k] = (uint8) (raw[k] + paeth(cur[k-img_n],0,0)); break; + } + #undef CASE + } else { + assert(img_n+1 == out_n); + #define CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, cur[img_n]=255,raw+=img_n,cur+=out_n,prior+=out_n) \ + for (k=0; k < img_n; ++k) + switch (filter) { + CASE(F_none) cur[k] = raw[k]; break; + CASE(F_sub) cur[k] = raw[k] + cur[k-out_n]; break; + CASE(F_up) cur[k] = raw[k] + prior[k]; break; + CASE(F_avg) cur[k] = raw[k] + ((prior[k] + cur[k-out_n])>>1); break; + CASE(F_paeth) cur[k] = (uint8) (raw[k] + paeth(cur[k-out_n],prior[k],prior[k-out_n])); break; + CASE(F_avg_first) cur[k] = raw[k] + (cur[k-out_n] >> 1); break; + CASE(F_paeth_first) cur[k] = (uint8) (raw[k] + paeth(cur[k-out_n],0,0)); break; + } + #undef CASE + } + } + return 1; +} + +static int create_png_image(png *a, uint8 *raw, uint32 raw_len, int out_n, int interlaced) +{ + uint8 *final; + int p; + int save; + if (!interlaced) + return create_png_image_raw(a, raw, raw_len, out_n, a->s->img_x, a->s->img_y); + save = stbi_png_partial; + stbi_png_partial = 0; + + // de-interlacing + final = (uint8 *) malloc(a->s->img_x * a->s->img_y * out_n); + for (p=0; p < 7; ++p) { + int xorig[] = { 0,4,0,2,0,1,0 }; + int yorig[] = { 0,0,4,0,2,0,1 }; + int xspc[] = { 8,8,4,4,2,2,1 }; + int yspc[] = { 8,8,8,4,4,2,2 }; + int i,j,x,y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; + if (x && y) { + if (!create_png_image_raw(a, raw, raw_len, out_n, x, y)) { + free(final); + return 0; + } + for (j=0; j < y; ++j) + for (i=0; i < x; ++i) + memcpy(final + (j*yspc[p]+yorig[p])*a->s->img_x*out_n + (i*xspc[p]+xorig[p])*out_n, + a->out + (j*x+i)*out_n, out_n); + free(a->out); + raw += (x*out_n+1)*y; + raw_len -= (x*out_n+1)*y; + } + } + a->out = final; + + stbi_png_partial = save; + return 1; +} + +static int compute_transparency(png *z, uint8 tc[3], int out_n) +{ + stbi *s = z->s; + uint32 i, pixel_count = s->img_x * s->img_y; + uint8 *p = z->out; + + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + assert(out_n == 2 || out_n == 4); + + if (out_n == 2) { + for (i=0; i < pixel_count; ++i) { + p[1] = (p[0] == tc[0] ? 0 : 255); + p += 2; + } + } else { + for (i=0; i < pixel_count; ++i) { + if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2]) + p[3] = 0; + p += 4; + } + } + return 1; +} + +static int expand_palette(png *a, uint8 *palette, int len, int pal_img_n) +{ + uint32 i, pixel_count = a->s->img_x * a->s->img_y; + uint8 *p, *temp_out, *orig = a->out; + + p = (uint8 *) malloc(pixel_count * pal_img_n); + if (p == NULL) return e("outofmem", "Out of memory"); + + // between here and free(out) below, exitting would leak + temp_out = p; + + if (pal_img_n == 3) { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p += 3; + } + } else { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p[3] = palette[n+3]; + p += 4; + } + } + free(a->out); + a->out = temp_out; + + STBI_NOTUSED(len); + + return 1; +} + +static int stbi_unpremultiply_on_load = 0; +static int stbi_de_iphone_flag = 0; + +void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) +{ + stbi_unpremultiply_on_load = flag_true_if_should_unpremultiply; +} +void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) +{ + stbi_de_iphone_flag = flag_true_if_should_convert; +} + +static void stbi_de_iphone(png *z) +{ + stbi *s = z->s; + uint32 i, pixel_count = s->img_x * s->img_y; + uint8 *p = z->out; + + if (s->img_out_n == 3) { // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + uint8 t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } else { + assert(s->img_out_n == 4); + if (stbi_unpremultiply_on_load) { + // convert bgr to rgb and unpremultiply + for (i=0; i < pixel_count; ++i) { + uint8 a = p[3]; + uint8 t = p[0]; + if (a) { + p[0] = p[2] * 255 / a; + p[1] = p[1] * 255 / a; + p[2] = t * 255 / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; + } + } else { + // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + uint8 t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +static int parse_png_file(png *z, int scan, int req_comp) +{ + uint8 palette[1024], pal_img_n=0; + uint8 has_trans=0, tc[3]; + uint32 ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=0, iphone=0; + stbi *s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!check_png_header(s)) return 0; + + if (scan == SCAN_type) return 1; + + for (;;) { + chunk c = get_chunk_header(s); + switch (c.type) { + case PNG_TYPE('C','g','B','I'): + iphone = stbi_de_iphone_flag; + skip(s, c.length); + break; + case PNG_TYPE('I','H','D','R'): { + int depth,color,comp,filter; + if (!first) return e("multiple IHDR","Corrupt PNG"); + first = 0; + if (c.length != 13) return e("bad IHDR len","Corrupt PNG"); + s->img_x = get32(s); if (s->img_x > (1 << 24)) return e("too large","Very large image (corrupt?)"); + s->img_y = get32(s); if (s->img_y > (1 << 24)) return e("too large","Very large image (corrupt?)"); + depth = get8(s); if (depth != 8) return e("8bit only","PNG not supported: 8-bit only"); + color = get8(s); if (color > 6) return e("bad ctype","Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return e("bad ctype","Corrupt PNG"); + comp = get8(s); if (comp) return e("bad comp method","Corrupt PNG"); + filter= get8(s); if (filter) return e("bad filter method","Corrupt PNG"); + interlace = get8(s); if (interlace>1) return e("bad interlace method","Corrupt PNG"); + if (!s->img_x || !s->img_y) return e("0-pixel image","Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return e("too large", "Image too large to decode"); + if (scan == SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return e("too large","Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } + + case PNG_TYPE('P','L','T','E'): { + if (first) return e("first not IHDR", "Corrupt PNG"); + if (c.length > 256*3) return e("invalid PLTE","Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return e("invalid PLTE","Corrupt PNG"); + for (i=0; i < pal_len; ++i) { + palette[i*4+0] = get8u(s); + palette[i*4+1] = get8u(s); + palette[i*4+2] = get8u(s); + palette[i*4+3] = 255; + } + break; + } + + case PNG_TYPE('t','R','N','S'): { + if (first) return e("first not IHDR", "Corrupt PNG"); + if (z->idata) return e("tRNS after IDAT","Corrupt PNG"); + if (pal_img_n) { + if (scan == SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return e("tRNS before PLTE","Corrupt PNG"); + if (c.length > pal_len) return e("bad tRNS len","Corrupt PNG"); + pal_img_n = 4; + for (i=0; i < c.length; ++i) + palette[i*4+3] = get8u(s); + } else { + if (!(s->img_n & 1)) return e("tRNS with alpha","Corrupt PNG"); + if (c.length != (uint32) s->img_n*2) return e("bad tRNS len","Corrupt PNG"); + has_trans = 1; + for (k=0; k < s->img_n; ++k) + tc[k] = (uint8) get16(s); // non 8-bit images will be larger + } + break; + } + + case PNG_TYPE('I','D','A','T'): { + if (first) return e("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return e("no PLTE","Corrupt PNG"); + if (scan == SCAN_header) { s->img_n = pal_img_n; return 1; } + if (ioff + c.length > idata_limit) { + uint8 *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + p = (uint8 *) realloc(z->idata, idata_limit); if (p == NULL) return e("outofmem", "Out of memory"); + z->idata = p; + } + if (!getn(s, z->idata+ioff,c.length)) return e("outofdata","Corrupt PNG"); + ioff += c.length; + break; + } + + case PNG_TYPE('I','E','N','D'): { + uint32 raw_len; + if (first) return e("first not IHDR", "Corrupt PNG"); + if (scan != SCAN_load) return 1; + if (z->idata == NULL) return e("no IDAT","Corrupt PNG"); + z->expanded = (uint8 *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, 16384, (int *) &raw_len, !iphone); + if (z->expanded == NULL) return 0; // zlib should set error + free(z->idata); z->idata = NULL; + if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) + s->img_out_n = s->img_n+1; + else + s->img_out_n = s->img_n; + if (!create_png_image(z, z->expanded, raw_len, s->img_out_n, interlace)) return 0; + if (has_trans) + if (!compute_transparency(z, tc, s->img_out_n)) return 0; + if (iphone && s->img_out_n > 2) + stbi_de_iphone(z); + if (pal_img_n) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!expand_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } + free(z->expanded); z->expanded = NULL; + return 1; + } + + default: + // if critical, fail + if (first) return e("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + static char invalid_chunk[] = "XXXX chunk not known"; + invalid_chunk[0] = (uint8) (c.type >> 24); + invalid_chunk[1] = (uint8) (c.type >> 16); + invalid_chunk[2] = (uint8) (c.type >> 8); + invalid_chunk[3] = (uint8) (c.type >> 0); + #endif + return e(invalid_chunk, "PNG not supported: unknown chunk type"); + } + skip(s, c.length); + break; + } + // end of chunk, read and skip CRC + get32(s); + } +} + +static unsigned char *do_png(png *p, int *x, int *y, int *n, int req_comp) +{ + unsigned char *result=NULL; + if (req_comp < 0 || req_comp > 4) return epuc("bad req_comp", "Internal error"); + if (parse_png_file(p, SCAN_load, req_comp)) { + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + result = convert_format(result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n) *n = p->s->img_n; + } + free(p->out); p->out = NULL; + free(p->expanded); p->expanded = NULL; + free(p->idata); p->idata = NULL; + + return result; +} + +static unsigned char *stbi_png_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + png p; + p.s = s; + return do_png(&p, x,y,comp,req_comp); +} + +static int stbi_png_test(stbi *s) +{ + int r; + r = check_png_header(s); + stbi_rewind(s); + return r; +} + +static int stbi_png_info_raw(png *p, int *x, int *y, int *comp) +{ + if (!parse_png_file(p, SCAN_header, 0)) { + stbi_rewind( p->s ); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; +} + +static int stbi_png_info(stbi *s, int *x, int *y, int *comp) +{ + png p; + p.s = s; + return stbi_png_info_raw(&p, x, y, comp); +} + +// Microsoft/Windows BMP image + +static int bmp_test(stbi *s) +{ + int sz; + if (get8(s) != 'B') return 0; + if (get8(s) != 'M') return 0; + get32le(s); // discard filesize + get16le(s); // discard reserved + get16le(s); // discard reserved + get32le(s); // discard data offset + sz = get32le(s); + if (sz == 12 || sz == 40 || sz == 56 || sz == 108) return 1; + return 0; +} + +static int stbi_bmp_test(stbi *s) +{ + int r = bmp_test(s); + stbi_rewind(s); + return r; +} + + +// returns 0..31 for the highest set bit +static int high_bit(unsigned int z) +{ + int n=0; + if (z == 0) return -1; + if (z >= 0x10000) n += 16, z >>= 16; + if (z >= 0x00100) n += 8, z >>= 8; + if (z >= 0x00010) n += 4, z >>= 4; + if (z >= 0x00004) n += 2, z >>= 2; + if (z >= 0x00002) n += 1, z >>= 1; + return n; +} + +static int bitcount(unsigned int a) +{ + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; +} + +static int shiftsigned(int v, int shift, int bits) +{ + int result; + int z=0; + + if (shift < 0) v <<= -shift; + else v >>= shift; + result = v; + + z = bits; + while (z < 8) { + result += v >> z; + z += bits; + } + return result; +} + +static stbi_uc *bmp_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + uint8 *out; + unsigned int mr=0,mg=0,mb=0,ma=0, fake_a=0; + stbi_uc pal[256][4]; + int psize=0,i,j,compress=0,width; + int bpp, flip_vertically, pad, target, offset, hsz; + if (get8(s) != 'B' || get8(s) != 'M') return epuc("not BMP", "Corrupt BMP"); + get32le(s); // discard filesize + get16le(s); // discard reserved + get16le(s); // discard reserved + offset = get32le(s); + hsz = get32le(s); + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108) return epuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = get16le(s); + s->img_y = get16le(s); + } else { + s->img_x = get32le(s); + s->img_y = get32le(s); + } + if (get16le(s) != 1) return epuc("bad BMP", "bad BMP"); + bpp = get16le(s); + if (bpp == 1) return epuc("monochrome", "BMP type not supported: 1-bit"); + flip_vertically = ((int) s->img_y) > 0; + s->img_y = abs((int) s->img_y); + if (hsz == 12) { + if (bpp < 24) + psize = (offset - 14 - 24) / 3; + } else { + compress = get32le(s); + if (compress == 1 || compress == 2) return epuc("BMP RLE", "BMP type not supported: RLE"); + get32le(s); // discard sizeof + get32le(s); // discard hres + get32le(s); // discard vres + get32le(s); // discard colorsused + get32le(s); // discard max important + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + get32le(s); + get32le(s); + get32le(s); + get32le(s); + } + if (bpp == 16 || bpp == 32) { + mr = mg = mb = 0; + if (compress == 0) { + if (bpp == 32) { + mr = 0xffu << 16; + mg = 0xffu << 8; + mb = 0xffu << 0; + ma = 0xffu << 24; + fake_a = 1; // @TODO: check for cases like alpha value is all 0 and switch it to 255 + } else { + mr = 31u << 10; + mg = 31u << 5; + mb = 31u << 0; + } + } else if (compress == 3) { + mr = get32le(s); + mg = get32le(s); + mb = get32le(s); + // not documented, but generated by photoshop and handled by mspaint + if (mr == mg && mg == mb) { + // ?!?!? + return epuc("bad BMP", "bad BMP"); + } + } else + return epuc("bad BMP", "bad BMP"); + } + } else { + assert(hsz == 108); + mr = get32le(s); + mg = get32le(s); + mb = get32le(s); + ma = get32le(s); + get32le(s); // discard color space + for (i=0; i < 12; ++i) + get32le(s); // discard color space parameters + } + if (bpp < 16) + psize = (offset - 14 - hsz) >> 2; + } + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + out = (stbi_uc *) malloc(target * s->img_x * s->img_y); + if (!out) return epuc("outofmem", "Out of memory"); + if (bpp < 16) { + int z=0; + if (psize == 0 || psize > 256) { free(out); return epuc("invalid", "Corrupt BMP"); } + for (i=0; i < psize; ++i) { + pal[i][2] = get8u(s); + pal[i][1] = get8u(s); + pal[i][0] = get8u(s); + if (hsz != 12) get8(s); + pal[i][3] = 255; + } + skip(s, offset - 14 - hsz - psize * (hsz == 12 ? 3 : 4)); + if (bpp == 4) width = (s->img_x + 1) >> 1; + else if (bpp == 8) width = s->img_x; + else { free(out); return epuc("bad bpp", "Corrupt BMP"); } + pad = (-width)&3; + for (j=0; j < (int) s->img_y; ++j) { + for (i=0; i < (int) s->img_x; i += 2) { + int v=get8(s),v2=0; + if (bpp == 4) { + v2 = v & 15; + v >>= 4; + } + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + if (i+1 == (int) s->img_x) break; + v = (bpp == 8) ? get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + skip(s, pad); + } + } else { + int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0; + int z = 0; + int easy=0; + skip(s, offset - 14 - hsz); + if (bpp == 24) width = 3 * s->img_x; + else if (bpp == 16) width = 2*s->img_x; + else /* bpp = 32 and pad = 0 */ width=0; + pad = (-width) & 3; + if (bpp == 24) { + easy = 1; + } else if (bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { free(out); return epuc("bad masks", "Corrupt BMP"); } + // right shift amt to put high bit in position #7 + rshift = high_bit(mr)-7; rcount = bitcount(mr); + gshift = high_bit(mg)-7; gcount = bitcount(mr); + bshift = high_bit(mb)-7; bcount = bitcount(mr); + ashift = high_bit(ma)-7; acount = bitcount(mr); + } + for (j=0; j < (int) s->img_y; ++j) { + if (easy) { + for (i=0; i < (int) s->img_x; ++i) { + int a; + out[z+2] = get8u(s); + out[z+1] = get8u(s); + out[z+0] = get8u(s); + z += 3; + a = (easy == 2 ? get8(s) : 255); + if (target == 4) out[z++] = (uint8) a; + } + } else { + for (i=0; i < (int) s->img_x; ++i) { + uint32 v = (bpp == 16 ? get16le(s) : get32le(s)); + int a; + out[z++] = (uint8) shiftsigned(v & mr, rshift, rcount); + out[z++] = (uint8) shiftsigned(v & mg, gshift, gcount); + out[z++] = (uint8) shiftsigned(v & mb, bshift, bcount); + a = (ma ? shiftsigned(v & ma, ashift, acount) : 255); + if (target == 4) out[z++] = (uint8) a; + } + } + skip(s, pad); + } + } + if (flip_vertically) { + stbi_uc t; + for (j=0; j < (int) s->img_y>>1; ++j) { + stbi_uc *p1 = out + j *s->img_x*target; + stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; + for (i=0; i < (int) s->img_x*target; ++i) { + t = p1[i], p1[i] = p2[i], p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} + +static stbi_uc *stbi_bmp_load(stbi *s,int *x, int *y, int *comp, int req_comp) +{ + return bmp_load(s, x,y,comp,req_comp); +} + + +// Targa Truevision - TGA +// by Jonathan Dummer + +static int tga_info(stbi *s, int *x, int *y, int *comp) +{ + int tga_w, tga_h, tga_comp; + int sz; + get8u(s); // discard Offset + sz = get8u(s); // color type + if( sz > 1 ) { + stbi_rewind(s); + return 0; // only RGB or indexed allowed + } + sz = get8u(s); // image type + // only RGB or grey allowed, +/- RLE + if ((sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11)) return 0; + skip(s,9); + tga_w = get16le(s); + if( tga_w < 1 ) { + stbi_rewind(s); + return 0; // test width + } + tga_h = get16le(s); + if( tga_h < 1 ) { + stbi_rewind(s); + return 0; // test height + } + sz = get8(s); // bits per pixel + // only RGB or RGBA or grey allowed + if ((sz != 8) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi_rewind(s); + return 0; + } + tga_comp = sz; + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp / 8; + return 1; // seems to have passed everything +} + +int stbi_tga_info(stbi *s, int *x, int *y, int *comp) +{ + return tga_info(s, x, y, comp); +} + +static int tga_test(stbi *s) +{ + int sz; + get8u(s); // discard Offset + sz = get8u(s); // color type + if ( sz > 1 ) return 0; // only RGB or indexed allowed + sz = get8u(s); // image type + if ( (sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11) ) return 0; // only RGB or grey allowed, +/- RLE + get16(s); // discard palette start + get16(s); // discard palette length + get8(s); // discard bits per palette color entry + get16(s); // discard x origin + get16(s); // discard y origin + if ( get16(s) < 1 ) return 0; // test width + if ( get16(s) < 1 ) return 0; // test height + sz = get8(s); // bits per pixel + if ( (sz != 8) && (sz != 16) && (sz != 24) && (sz != 32) ) return 0; // only RGB or RGBA or grey allowed + return 1; // seems to have passed everything +} + +static int stbi_tga_test(stbi *s) +{ + int res = tga_test(s); + stbi_rewind(s); + return res; +} + +static stbi_uc *tga_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + // read in the TGA header stuff + int tga_offset = get8u(s); + int tga_indexed = get8u(s); + int tga_image_type = get8u(s); + int tga_is_RLE = 0; + int tga_palette_start = get16le(s); + int tga_palette_len = get16le(s); + int tga_palette_bits = get8u(s); + int tga_x_origin = get16le(s); + int tga_y_origin = get16le(s); + int tga_width = get16le(s); + int tga_height = get16le(s); + int tga_bits_per_pixel = get8u(s); + int tga_inverted = get8u(s); + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4]; + unsigned char trans_data[4]; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + + // do a tiny bit of precessing + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + /* int tga_alpha_bits = tga_inverted & 15; */ + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // error check + if ( //(tga_indexed) || + (tga_width < 1) || (tga_height < 1) || + (tga_image_type < 1) || (tga_image_type > 3) || + ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16) && + (tga_bits_per_pixel != 24) && (tga_bits_per_pixel != 32)) + ) + { + return NULL; // we don't report this as a bad TGA because we don't even know if it's TGA + } + + // If I'm paletted, then I'll use the number of bits from the palette + if ( tga_indexed ) + { + tga_bits_per_pixel = tga_palette_bits; + } + + // tga info + *x = tga_width; + *y = tga_height; + if ( (req_comp < 1) || (req_comp > 4) ) + { + // just use whatever the file was + req_comp = tga_bits_per_pixel / 8; + *comp = req_comp; + } else + { + // force a new number of components + *comp = tga_bits_per_pixel/8; + } + tga_data = (unsigned char*)malloc( tga_width * tga_height * req_comp ); + if (!tga_data) return epuc("outofmem", "Out of memory"); + + // skip to the data's starting position (offset usually = 0) + skip(s, tga_offset ); + // do I need to load a palette? + if ( tga_indexed ) + { + // any data to skip? (offset usually = 0) + skip(s, tga_palette_start ); + // load the palette + tga_palette = (unsigned char*)malloc( tga_palette_len * tga_palette_bits / 8 ); + if (!tga_palette) return epuc("outofmem", "Out of memory"); + if (!getn(s, tga_palette, tga_palette_len * tga_palette_bits / 8 )) { + free(tga_data); + free(tga_palette); + return epuc("bad palette", "Corrupt TGA"); + } + } + // load the data + trans_data[0] = trans_data[1] = trans_data[2] = trans_data[3] = 0; + for (i=0; i < tga_width * tga_height; ++i) + { + // if I'm in RLE mode, do I need to get a RLE chunk? + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + // yep, get the next byte as a RLE command + int RLE_cmd = get8u(s); + RLE_count = 1 + (RLE_cmd & 127); + RLE_repeating = RLE_cmd >> 7; + read_next_pixel = 1; + } else if ( !RLE_repeating ) + { + read_next_pixel = 1; + } + } else + { + read_next_pixel = 1; + } + // OK, if I need to read a pixel, do it now + if ( read_next_pixel ) + { + // load however much data we did have + if ( tga_indexed ) + { + // read in 1 byte, then perform the lookup + int pal_idx = get8u(s); + if ( pal_idx >= tga_palette_len ) + { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_bits_per_pixel / 8; + for (j = 0; j*8 < tga_bits_per_pixel; ++j) + { + raw_data[j] = tga_palette[pal_idx+j]; + } + } else + { + // read in the data raw + for (j = 0; j*8 < tga_bits_per_pixel; ++j) + { + raw_data[j] = get8u(s); + } + } + // convert raw to the intermediate format + switch (tga_bits_per_pixel) + { + case 8: + // Luminous => RGBA + trans_data[0] = raw_data[0]; + trans_data[1] = raw_data[0]; + trans_data[2] = raw_data[0]; + trans_data[3] = 255; + break; + case 16: + // Luminous,Alpha => RGBA + trans_data[0] = raw_data[0]; + trans_data[1] = raw_data[0]; + trans_data[2] = raw_data[0]; + trans_data[3] = raw_data[1]; + break; + case 24: + // BGR => RGBA + trans_data[0] = raw_data[2]; + trans_data[1] = raw_data[1]; + trans_data[2] = raw_data[0]; + trans_data[3] = 255; + break; + case 32: + // BGRA => RGBA + trans_data[0] = raw_data[2]; + trans_data[1] = raw_data[1]; + trans_data[2] = raw_data[0]; + trans_data[3] = raw_data[3]; + break; + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + // convert to final format + switch (req_comp) + { + case 1: + // RGBA => Luminance + tga_data[i*req_comp+0] = compute_y(trans_data[0],trans_data[1],trans_data[2]); + break; + case 2: + // RGBA => Luminance,Alpha + tga_data[i*req_comp+0] = compute_y(trans_data[0],trans_data[1],trans_data[2]); + tga_data[i*req_comp+1] = trans_data[3]; + break; + case 3: + // RGBA => RGB + tga_data[i*req_comp+0] = trans_data[0]; + tga_data[i*req_comp+1] = trans_data[1]; + tga_data[i*req_comp+2] = trans_data[2]; + break; + case 4: + // RGBA => RGBA + tga_data[i*req_comp+0] = trans_data[0]; + tga_data[i*req_comp+1] = trans_data[1]; + tga_data[i*req_comp+2] = trans_data[2]; + tga_data[i*req_comp+3] = trans_data[3]; + break; + } + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * req_comp; + int index2 = (tga_height - 1 - j) * tga_width * req_comp; + for (i = tga_width * req_comp; i > 0; --i) + { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } + } + } + // clear my palette, if I had one + if ( tga_palette != NULL ) + { + free( tga_palette ); + } + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + // OK, done + return tga_data; +} + +static stbi_uc *stbi_tga_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + return tga_load(s,x,y,comp,req_comp); +} + + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +static int psd_test(stbi *s) +{ + if (get32(s) != 0x38425053) return 0; // "8BPS" + else return 1; +} + +static int stbi_psd_test(stbi *s) +{ + int r = psd_test(s); + stbi_rewind(s); + return r; +} + +static stbi_uc *psd_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + int pixelCount; + int channelCount, compression; + int channel, i, count, len; + int w,h; + uint8 *out; + + // Check identifier + if (get32(s) != 0x38425053) // "8BPS" + return epuc("not PSD", "Corrupt PSD image"); + + // Check file type version. + if (get16(s) != 1) + return epuc("wrong version", "Unsupported version of PSD image"); + + // Skip 6 reserved bytes. + skip(s, 6 ); + + // Read the number of channels (R, G, B, A, etc). + channelCount = get16(s); + if (channelCount < 0 || channelCount > 16) + return epuc("wrong channel count", "Unsupported number of channels in PSD image"); + + // Read the rows and columns of the image. + h = get32(s); + w = get32(s); + + // Make sure the depth is 8 bits. + if (get16(s) != 8) + return epuc("unsupported bit depth", "PSD bit depth is not 8 bit"); + + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (get16(s) != 3) + return epuc("wrong color format", "PSD is not in RGB color format"); + + // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) + skip(s,get32(s) ); + + // Skip the image resources. (resolution, pen tool paths, etc) + skip(s, get32(s) ); + + // Skip the reserved data. + skip(s, get32(s) ); + + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = get16(s); + if (compression > 1) + return epuc("bad compression", "PSD has an unknown compression format"); + + // Create the destination image. + out = (stbi_uc *) malloc(4 * w*h); + if (!out) return epuc("outofmem", "Out of memory"); + pixelCount = w*h; + + // Initialize the data to zero. + //memset( out, 0, pixelCount * 4 ); + + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. + // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. + // Else if n is 128, noop. + // Endloop + + // The RLE-compressed data is preceeded by a 2-byte data count for each row in the data, + // which we're going to just skip. + skip(s, h * channelCount * 2 ); + + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + uint8 *p; + + p = out+channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++) *p = (channel == 3 ? 255 : 0), p += 4; + } else { + // Read the RLE data. + count = 0; + while (count < pixelCount) { + len = get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + count += len; + while (len) { + *p = get8u(s); + p += 4; + len--; + } + } else if (len > 128) { + uint8 val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len ^= 0x0FF; + len += 2; + val = get8u(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + } + } + + } else { + // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) + // where each channel consists of an 8-bit value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + uint8 *p; + + p = out + channel; + if (channel > channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++) *p = channel == 3 ? 255 : 0, p += 4; + } else { + // Read the data. + for (i = 0; i < pixelCount; i++) + *p = get8u(s), p += 4; + } + } + } + + if (req_comp && req_comp != 4) { + out = convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // convert_format frees input on failure + } + + if (comp) *comp = channelCount; + *y = h; + *x = w; + + return out; +} + +static stbi_uc *stbi_psd_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + return psd_load(s,x,y,comp,req_comp); +} + +// ************************************************************************************************* +// Softimage PIC loader +// by Tom Seddon +// +// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format +// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ + +static int pic_is4(stbi *s,const char *str) +{ + int i; + for (i=0; i<4; ++i) + if (get8(s) != (stbi_uc)str[i]) + return 0; + + return 1; +} + +static int pic_test(stbi *s) +{ + int i; + + if (!pic_is4(s,"\x53\x80\xF6\x34")) + return 0; + + for(i=0;i<84;++i) + get8(s); + + if (!pic_is4(s,"PICT")) + return 0; + + return 1; +} + +typedef struct +{ + stbi_uc size,type,channel; +} pic_packet_t; + +static stbi_uc *pic_readval(stbi *s, int channel, stbi_uc *dest) +{ + int mask=0x80, i; + + for (i=0; i<4; ++i, mask>>=1) { + if (channel & mask) { + if (at_eof(s)) return epuc("bad file","PIC file too short"); + dest[i]=get8u(s); + } + } + + return dest; +} + +static void pic_copyval(int channel,stbi_uc *dest,const stbi_uc *src) +{ + int mask=0x80,i; + + for (i=0;i<4; ++i, mask>>=1) + if (channel&mask) + dest[i]=src[i]; +} + +static stbi_uc *pic_load2(stbi *s,int width,int height,int *comp, stbi_uc *result) +{ + int act_comp=0,num_packets=0,y,chained; + pic_packet_t packets[10]; + + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + pic_packet_t *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return epuc("bad format","too many packets"); + + packet = &packets[num_packets++]; + + chained = get8(s); + packet->size = get8u(s); + packet->type = get8u(s); + packet->channel = get8u(s); + + act_comp |= packet->channel; + + if (at_eof(s)) return epuc("bad file","file too short (reading packets)"); + if (packet->size != 8) return epuc("bad format","packet isn't 8bpp"); + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + + for(y=0; ytype) { + default: + return epuc("bad format","packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for(x=0;xchannel,dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left=width, i; + + while (left>0) { + stbi_uc count,value[4]; + + count=get8u(s); + if (at_eof(s)) return epuc("bad file","file too short (pure read count)"); + + if (count > left) + count = (uint8) left; + + if (!pic_readval(s,packet->channel,value)) return 0; + + for(i=0; ichannel,dest,value); + left -= count; + } + } + break; + + case 2: {//Mixed RLE + int left=width; + while (left>0) { + int count = get8(s), i; + if (at_eof(s)) return epuc("bad file","file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + int i; + + if (count==128) + count = get16(s); + else + count -= 127; + if (count > left) + return epuc("bad file","scanline overrun"); + + if (!pic_readval(s,packet->channel,value)) + return 0; + + for(i=0;ichannel,dest,value); + } else { // Raw + ++count; + if (count>left) return epuc("bad file","scanline overrun"); + + for(i=0;ichannel,dest)) + return 0; + } + left-=count; + } + break; + } + } + } + } + + return result; +} + +static stbi_uc *pic_load(stbi *s,int *px,int *py,int *comp,int req_comp) +{ + stbi_uc *result; + int i, x,y; + + for (i=0; i<92; ++i) + get8(s); + + x = get16(s); + y = get16(s); + if (at_eof(s)) return epuc("bad file","file too short (pic header)"); + if ((1 << 28) / x < y) return epuc("too large", "Image too large to decode"); + + get32(s); //skip `ratio' + get16(s); //skip `fields' + get16(s); //skip `pad' + + // intermediate buffer is RGBA + result = (stbi_uc *) malloc(x*y*4); + memset(result, 0xff, x*y*4); + + if (!pic_load2(s,x,y,comp, result)) { + free(result); + result=0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result=convert_format(result,4,req_comp,x,y); + + return result; +} + +static int stbi_pic_test(stbi *s) +{ + int r = pic_test(s); + stbi_rewind(s); + return r; +} + +static stbi_uc *stbi_pic_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + return pic_load(s,x,y,comp,req_comp); +} + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb +typedef struct stbi_gif_lzw_struct { + int16 prefix; + uint8 first; + uint8 suffix; +} stbi_gif_lzw; + +typedef struct stbi_gif_struct +{ + int w,h; + stbi_uc *out; // output buffer (always 4 components) + int flags, bgindex, ratio, transparent, eflags; + uint8 pal[256][4]; + uint8 lpal[256][4]; + stbi_gif_lzw codes[4096]; + uint8 *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; +} stbi_gif; + +static int gif_test(stbi *s) +{ + int sz; + if (get8(s) != 'G' || get8(s) != 'I' || get8(s) != 'F' || get8(s) != '8') return 0; + sz = get8(s); + if (sz != '9' && sz != '7') return 0; + if (get8(s) != 'a') return 0; + return 1; +} + +static int stbi_gif_test(stbi *s) +{ + int r = gif_test(s); + stbi_rewind(s); + return r; +} + +static void stbi_gif_parse_colortable(stbi *s, uint8 pal[256][4], int num_entries, int transp) +{ + int i; + for (i=0; i < num_entries; ++i) { + pal[i][2] = get8u(s); + pal[i][1] = get8u(s); + pal[i][0] = get8u(s); + pal[i][3] = transp ? 0 : 255; + } +} + +static int stbi_gif_header(stbi *s, stbi_gif *g, int *comp, int is_info) +{ + uint8 version; + if (get8(s) != 'G' || get8(s) != 'I' || get8(s) != 'F' || get8(s) != '8') + return e("not GIF", "Corrupt GIF"); + + version = get8u(s); + if (version != '7' && version != '9') return e("not GIF", "Corrupt GIF"); + if (get8(s) != 'a') return e("not GIF", "Corrupt GIF"); + + failure_reason = ""; + g->w = get16le(s); + g->h = get16le(s); + g->flags = get8(s); + g->bgindex = get8(s); + g->ratio = get8(s); + g->transparent = -1; + + if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + + if (is_info) return 1; + + if (g->flags & 0x80) + stbi_gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + + return 1; +} + +static int stbi_gif_info_raw(stbi *s, int *x, int *y, int *comp) +{ + stbi_gif g; + if (!stbi_gif_header(s, &g, comp, 1)) { + stbi_rewind( s ); + return 0; + } + if (x) *x = g.w; + if (y) *y = g.h; + return 1; +} + +static void stbi_out_gif_code(stbi_gif *g, uint16 code) +{ + uint8 *p, *c; + + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) + stbi_out_gif_code(g, g->codes[code].prefix); + + if (g->cur_y >= g->max_y) return; + + p = &g->out[g->cur_x + g->cur_y]; + c = &g->color_table[g->codes[code].suffix * 4]; + + if (c[3] >= 128) { + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; + + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; + + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static uint8 *stbi_process_gif_raster(stbi *s, stbi_gif *g) +{ + uint8 lzw_cs; + int32 len, code; + uint32 first; + int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi_gif_lzw *p; + + lzw_cs = get8u(s); + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (code = 0; code < clear; code++) { + g->codes[code].prefix = -1; + g->codes[code].first = (uint8) code; + g->codes[code].suffix = (uint8) code; + } + + // support no starting clear code + avail = clear+2; + oldcode = -1; + + len = 0; + for(;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = get8(s); // start new block + if (len == 0) + return g->out; + } + --len; + bits |= (int32) get8(s) << valid_bits; + valid_bits += 8; + } else { + int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + skip(s, len); + while ((len = get8(s)) > 0) + skip(s,len); + return g->out; + } else if (code <= avail) { + if (first) return epuc("no clear code", "Corrupt GIF"); + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 4096) return epuc("too many codes", "Corrupt GIF"); + p->prefix = (int16) oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return epuc("illegal code in raster", "Corrupt GIF"); + + stbi_out_gif_code(g, (uint16) code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return epuc("illegal code in raster", "Corrupt GIF"); + } + } + } +} + +static void stbi_fill_gif_background(stbi_gif *g) +{ + int i; + uint8 *c = g->pal[g->bgindex]; + // @OPTIMIZE: write a dword at a time + for (i = 0; i < g->w * g->h * 4; i += 4) { + uint8 *p = &g->out[i]; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +static uint8 *stbi_gif_load_next(stbi *s, stbi_gif *g, int *comp, int req_comp) +{ + int i; + uint8 *old_out = 0; + + if (g->out == 0) { + if (!stbi_gif_header(s, g, comp,0)) return 0; // failure_reason set by stbi_gif_header + g->out = (uint8 *) malloc(4 * g->w * g->h); + if (g->out == 0) return epuc("outofmem", "Out of memory"); + stbi_fill_gif_background(g); + } else { + // animated-gif-only path + if (((g->eflags & 0x1C) >> 2) == 3) { + old_out = g->out; + g->out = (uint8 *) malloc(4 * g->w * g->h); + if (g->out == 0) return epuc("outofmem", "Out of memory"); + memcpy(g->out, old_out, g->w*g->h*4); + } + } + + for (;;) { + switch (get8(s)) { + case 0x2C: /* Image Descriptor */ + { + int32 x, y, w, h; + uint8 *o; + + x = get16le(s); + y = get16le(s); + w = get16le(s); + h = get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return epuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + g->lflags = get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi_gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (uint8 *) g->lpal; + } else if (g->flags & 0x80) { + for (i=0; i < 256; ++i) // @OPTIMIZE: reset only the previous transparent + g->pal[i][3] = 255; + if (g->transparent >= 0 && (g->eflags & 0x01)) + g->pal[g->transparent][3] = 0; + g->color_table = (uint8 *) g->pal; + } else + return epuc("missing color table", "Corrupt GIF"); + + o = stbi_process_gif_raster(s, g); + if (o == NULL) return NULL; + + if (req_comp && req_comp != 4) + o = convert_format(o, 4, req_comp, g->w, g->h); + return o; + } + + case 0x21: // Comment Extension. + { + int len; + if (get8(s) == 0xF9) { // Graphic Control Extension. + len = get8(s); + if (len == 4) { + g->eflags = get8(s); + get16le(s); // delay + g->transparent = get8(s); + } else { + skip(s, len); + break; + } + } + while ((len = get8(s)) != 0) + skip(s, len); + break; + } + + case 0x3B: // gif stream termination code + return (uint8 *) 1; + + default: + return epuc("unknown code", "Corrupt GIF"); + } + } +} + +static stbi_uc *stbi_gif_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + uint8 *u = 0; + stbi_gif g={0}; + + u = stbi_gif_load_next(s, &g, comp, req_comp); + if (u == (void *) 1) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + } + + return u; +} + +static int stbi_gif_info(stbi *s, int *x, int *y, int *comp) +{ + return stbi_gif_info_raw(s,x,y,comp); +} + + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int hdr_test(stbi *s) +{ + const char *signature = "#?RADIANCE\n"; + int i; + for (i=0; signature[i]; ++i) + if (get8(s) != signature[i]) + return 0; + return 1; +} + +static int stbi_hdr_test(stbi* s) +{ + int r = hdr_test(s); + stbi_rewind(s); + return r; +} + +#define HDR_BUFLEN 1024 +static char *hdr_gettoken(stbi *z, char *buffer) +{ + int len=0; + char c = '\0'; + + c = (char) get8(z); + + while (!at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == HDR_BUFLEN-1) { + // flush to end of line + while (!at_eof(z) && get8(z) != '\n') + ; + break; + } + c = (char) get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void hdr_convert(float *output, stbi_uc *input, int req_comp) +{ + if ( input[3] != 0 ) { + float f1; + // Exponent + f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: output[3] = 1; /* fallthrough */ + case 3: output[0] = output[1] = output[2] = 0; + break; + case 2: output[1] = 1; /* fallthrough */ + case 1: output[0] = 0; + break; + } + } +} + +static float *hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + char buffer[HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1,c2, z; + + + // Check identifier + if (strcmp(hdr_gettoken(s,buffer), "#?RADIANCE") != 0) + return epf("not HDR", "Corrupt HDR image"); + + // Parse header + for(;;) { + token = hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) return epf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) return epf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) return epf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = strtol(token, NULL, 10); + + *x = width; + *y = height; + + *comp = 3; + if (req_comp == 0) req_comp = 3; + + // Read data + hdr_data = (float *) malloc(height * width * req_comp * sizeof(float)); + + // Load image data + // image data is stored as some number of sca + if ( width < 8 || width >= 32768) { + // Read flat data + for (j=0; j < height; ++j) { + for (i=0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + getn(s, rgbe, 4); + hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = get8(s); + c2 = get8(s); + len = get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a decoded + // pixel (note this can't be a valid pixel--one of RGB must be >= 128) + uint8 rgbe[4]; + rgbe[0] = (uint8) c1; + rgbe[1] = (uint8) c2; + rgbe[2] = (uint8) len; + rgbe[3] = (uint8) get8u(s); + hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + free(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= get8(s); + if (len != width) { free(hdr_data); free(scanline); return epf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) scanline = (stbi_uc *) malloc(width * 4); + + for (k = 0; k < 4; ++k) { + i = 0; + while (i < width) { + count = get8u(s); + if (count > 128) { + // Run + value = get8u(s); + count -= 128; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = get8u(s); + } + } + } + for (i=0; i < width; ++i) + hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); + } + free(scanline); + } + + return hdr_data; +} + +static float *stbi_hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp) +{ + return hdr_load(s,x,y,comp,req_comp); +} + +static int stbi_hdr_info(stbi *s, int *x, int *y, int *comp) +{ + char buffer[HDR_BUFLEN]; + char *token; + int valid = 0; + + if (strcmp(hdr_gettoken(s,buffer), "#?RADIANCE") != 0) { + stbi_rewind( s ); + return 0; + } + + for(;;) { + token = hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi_rewind( s ); + return 0; + } + token = hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) { + stbi_rewind( s ); + return 0; + } + token += 3; + *y = strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) { + stbi_rewind( s ); + return 0; + } + token += 3; + *x = strtol(token, NULL, 10); + *comp = 3; + return 1; +} +#endif // STBI_NO_HDR + +static int stbi_bmp_info(stbi *s, int *x, int *y, int *comp) +{ + int hsz; + if (get8(s) != 'B' || get8(s) != 'M') { + stbi_rewind( s ); + return 0; + } + skip(s,12); + hsz = get32le(s); + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108) { + stbi_rewind( s ); + return 0; + } + if (hsz == 12) { + *x = get16le(s); + *y = get16le(s); + } else { + *x = get32le(s); + *y = get32le(s); + } + if (get16le(s) != 1) { + stbi_rewind( s ); + return 0; + } + *comp = get16le(s) / 8; + return 1; +} + +static int stbi_psd_info(stbi *s, int *x, int *y, int *comp) +{ + int channelCount; + if (get32(s) != 0x38425053) { + stbi_rewind( s ); + return 0; + } + if (get16(s) != 1) { + stbi_rewind( s ); + return 0; + } + skip(s, 6); + channelCount = get16(s); + if (channelCount < 0 || channelCount > 16) { + stbi_rewind( s ); + return 0; + } + *y = get32(s); + *x = get32(s); + if (get16(s) != 8) { + stbi_rewind( s ); + return 0; + } + if (get16(s) != 3) { + stbi_rewind( s ); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi_pic_info(stbi *s, int *x, int *y, int *comp) +{ + int act_comp=0,num_packets=0,chained; + pic_packet_t packets[10]; + + skip(s, 92); + + *x = get16(s); + *y = get16(s); + if (at_eof(s)) return 0; + if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi_rewind( s ); + return 0; + } + + skip(s, 8); + + do { + pic_packet_t *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return 0; + + packet = &packets[num_packets++]; + chained = get8(s); + packet->size = get8u(s); + packet->type = get8u(s); + packet->channel = get8u(s); + act_comp |= packet->channel; + + if (at_eof(s)) { + stbi_rewind( s ); + return 0; + } + if (packet->size != 8) { + stbi_rewind( s ); + return 0; + } + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); + + return 1; +} + +static int stbi_info_main(stbi *s, int *x, int *y, int *comp) +{ + if (stbi_jpeg_info(s, x, y, comp)) + return 1; + if (stbi_png_info(s, x, y, comp)) + return 1; + if (stbi_gif_info(s, x, y, comp)) + return 1; + if (stbi_bmp_info(s, x, y, comp)) + return 1; + if (stbi_psd_info(s, x, y, comp)) + return 1; + if (stbi_pic_info(s, x, y, comp)) + return 1; + #ifndef STBI_NO_HDR + if (stbi_hdr_info(s, x, y, comp)) + return 1; + #endif + // test tga last because it's a crappy test! + if (stbi_tga_info(s, x, y, comp)) + return 1; + return e("unknown image type", "Image not of any known type, or corrupt"); +} + +#ifndef STBI_NO_STDIO +int stbi_info(char const *filename, int *x, int *y, int *comp) +{ + FILE *f = fopen(filename, "rb"); + int result; + if (!f) return e("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; +} + +int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) +{ + int r; + stbi s; + long pos = ftell(f); + start_file(&s, f); + r = stbi_info_main(&s,x,y,comp); + fseek(f,pos,SEEK_SET); + return r; +} +#endif // !STBI_NO_STDIO + +int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) +{ + stbi s; + start_mem(&s,buffer,len); + return stbi_info_main(&s,x,y,comp); +} + +int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) +{ + stbi s; + start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi_info_main(&s,x,y,comp); +} + +#endif // STBI_HEADER_FILE_ONLY + +/* + revision history: + 1.33 (2011-07-14) + make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements + 1.32 (2011-07-13) + support for "info" function for all supported filetypes (SpartanJ) + 1.31 (2011-06-20) + a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) + added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + removed deprecated format-specific test/load functions + removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway + error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) + fix inefficiency in decoding 32-bit BMP (David Woo) + 1.29 (2010-08-16) + various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) + fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) + cast-to-uint8 to fix warnings + 1.26 (2010-07-24) + fix bug in file buffering for PNG reported by SpartanJ + 1.25 (2010-07-17) + refix trans_data warning (Won Chun) + 1.24 (2010-07-12) + perf improvements reading from files on platforms with lock-heavy fgetc() + minor perf improvements for jpeg + deprecated type-specific functions so we'll get feedback if they're needed + attempt to fix trans_data warning (Won Chun) + 1.23 fixed bug in iPhone support + 1.22 (2010-07-10) + removed image *writing* support + stbi_info support from Jetro Lauha + GIF support from Jean-Marc Lienher + iPhone PNG-extensions from James Brown + warning-fixes from Nicolas Schulz and Janez Zemva (i.e. Janez (U+017D)emva) + 1.21 fix use of 'uint8' in header (reported by jon blow) + 1.20 added support for Softimage PIC, by Tom Seddon + 1.19 bug in interlaced PNG corruption check (found by ryg) + 1.18 2008-08-02 + fix a threading bug (local mutable static) + 1.17 support interlaced PNG + 1.16 major bugfix - convert_format converted one too many pixels + 1.15 initialize some fields for thread safety + 1.14 fix threadsafe conversion bug + header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + 1.13 threadsafe + 1.12 const qualifiers in the API + 1.11 Support installable IDCT, colorspace conversion routines + 1.10 Fixes for 64-bit (don't use "unsigned long") + optimized upsampling by Fabian "ryg" Giesen + 1.09 Fix format-conversion for PSD code (bad global variables!) + 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz + 1.07 attempt to fix C++ warning/errors again + 1.06 attempt to fix C++ warning/errors again + 1.05 fix TGA loading to return correct *comp and use good luminance calc + 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free + 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR + 1.02 support for (subset of) HDR files, float interface for preferred access to them + 1.01 fix bug: possible bug in handling right-side up bmps... not sure + fix bug: the stbi_bmp_load() and stbi_tga_load() functions didn't work at all + 1.00 interface to zlib that skips zlib header + 0.99 correct handling of alpha in palette + 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 0.97 jpeg errors on too large a file; also catch another malloc failure + 0.96 fix detection of invalid v value - particleman@mollyrocket forum + 0.95 during header scan, seek to markers in case of padding + 0.94 STBI_NO_STDIO to disable stdio usage; rename all #defines the same + 0.93 handle jpegtran output; verbose errors + 0.92 read 4,8,16,24,32-bit BMP files of several formats + 0.91 output 24-bit Windows 3.0 BMP files + 0.90 fix a few more warnings; bump version number to approach 1.0 + 0.61 bugfixes due to Marc LeBlanc, Christopher Lloyd + 0.60 fix compiling as c++ + 0.59 fix warnings: merge Dave Moore's -Wall fixes + 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available + 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.55 fix bug: restart_interval not initialized to 0 + 0.54 allow NULL for 'int *comp' + 0.53 fix bug in png 3->4; speedup png decoding + 0.52 png handles req_comp=3,4 directly; minor cleanup; jpeg comments + 0.51 obey req_comp requests, 1-component jpegs return as 1-component, + on 'test' only check type, not whether we support this variant + 0.50 first released version +*/ diff --git a/src/stb_image.h b/src/stb_image.h new file mode 100644 index 0000000..bcdbf3a --- /dev/null +++ b/src/stb_image.h @@ -0,0 +1,8 @@ +// Courtesy of Peter Kutz +#ifndef STB_IMAGE_H +#define STB_IMAGE_H + +#define STBI_HEADER_FILE_ONLY +#include "stb_image.c" + +#endif // STB_IMAGE_H \ No newline at end of file diff --git a/src/stb_image_write.c b/src/stb_image_write.c new file mode 100644 index 0000000..474291c --- /dev/null +++ b/src/stb_image_write.c @@ -0,0 +1,3 @@ +// Courtesy of Peter Kutz +#define STB_IMAGE_WRITE_IMPLEMENTATION +#include "stb_image_write.h" diff --git a/src/stb_image_write.h b/src/stb_image_write.h new file mode 100644 index 0000000..54a3cf7 --- /dev/null +++ b/src/stb_image_write.h @@ -0,0 +1,517 @@ +// Courtesy of Peter Kutz +/* stbiw-0.92 - public domain - http://nothings.org/stb/stb_image_write.h + writes out PNG/BMP/TGA images to C stdio - Sean Barrett 2010 + no warranty implied; use at your own risk + + +Before including, + + #define STB_IMAGE_WRITE_IMPLEMENTATION + +in the file that you want to have the implementation. + + +ABOUT: + + This header file is a library for writing images to C stdio. It could be + adapted to write to memory or a general streaming interface; let me know. + + The PNG output is not optimal; it is 20-50% larger than the file + written by a decent optimizing implementation. This library is designed + for source code compactness and simplicitly, not optimal image file size + or run-time performance. + +USAGE: + + There are three functions, one for each image file format: + + int stbi_write_png(char const *filename, int w, int h, int comp, const void *data, int stride_in_bytes); + int stbi_write_bmp(char const *filename, int w, int h, int comp, const void *data); + int stbi_write_tga(char const *filename, int w, int h, int comp, const void *data); + + Each function returns 0 on failure and non-0 on success. + + The functions create an image file defined by the parameters. The image + is a rectangle of pixels stored from left-to-right, top-to-bottom. + Each pixel contains 'comp' channels of data stored interleaved with 8-bits + per channel, in the following order: 1=Y, 2=YA, 3=RGB, 4=RGBA. (Y is + monochrome color.) The rectangle is 'w' pixels wide and 'h' pixels tall. + The *data pointer points to the first byte of the top-left-most pixel. + For PNG, "stride_in_bytes" is the distance in bytes from the first byte of + a row of pixels to the first byte of the next row of pixels. + + PNG creates output files with the same number of components as the input. + The BMP and TGA formats expand Y to RGB in the file format. BMP does not + output alpha. + + PNG supports writing rectangles of data even when the bytes storing rows of + data are not consecutive in memory (e.g. sub-rectangles of a larger image), + by supplying the stride between the beginning of adjacent rows. The other + formats do not. (Thus you cannot write a native-format BMP through the BMP + writer, both because it is in BGR order and because it may have padding + at the end of the line.) +*/ + +#ifndef INCLUDE_STB_IMAGE_WRITE_H +#define INCLUDE_STB_IMAGE_WRITE_H + +// Added by Peter Kutz: +#if defined(_MSC_VER) +#define _CRT_SECURE_NO_WARNINGS // suppress bogus warnings about fopen() +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +extern int stbi_write_png(char const *filename, int w, int h, int comp, const void *data, int stride_in_bytes); +extern int stbi_write_bmp(char const *filename, int w, int h, int comp, const void *data); +extern int stbi_write_tga(char const *filename, int w, int h, int comp, const void *data); + +#ifdef __cplusplus +} +#endif + +#endif//INCLUDE_STB_IMAGE_WRITE_H + +#ifdef STB_IMAGE_WRITE_IMPLEMENTATION + +#include +#include +#include +#include +#include + +typedef unsigned int stbiw_uint32; +typedef int stb_image_write_test[sizeof(stbiw_uint32)==4 ? 1 : -1]; + +static void writefv(FILE *f, const char *fmt, va_list v) +{ + while (*fmt) { + switch (*fmt++) { + case ' ': break; + case '1': { unsigned char x = (unsigned char) va_arg(v, int); fputc(x,f); break; } + case '2': { int x = va_arg(v,int); unsigned char b[2]; + b[0] = (unsigned char) x; b[1] = (unsigned char) (x>>8); + fwrite(b,2,1,f); break; } + case '4': { stbiw_uint32 x = va_arg(v,int); unsigned char b[4]; + b[0]=(unsigned char)x; b[1]=(unsigned char)(x>>8); + b[2]=(unsigned char)(x>>16); b[3]=(unsigned char)(x>>24); + fwrite(b,4,1,f); break; } + default: + assert(0); + return; + } + } +} + +static void write3(FILE *f, unsigned char a, unsigned char b, unsigned char c) +{ + unsigned char arr[3]; + arr[0] = a, arr[1] = b, arr[2] = c; + fwrite(arr, 3, 1, f); +} + +static void write_pixels(FILE *f, int rgb_dir, int vdir, int x, int y, int comp, void *data, int write_alpha, int scanline_pad) +{ + unsigned char bg[3] = { 255, 0, 255}, px[3]; + stbiw_uint32 zero = 0; + int i,j,k, j_end; + + if (y <= 0) + return; + + if (vdir < 0) + j_end = -1, j = y-1; + else + j_end = y, j = 0; + + for (; j != j_end; j += vdir) { + for (i=0; i < x; ++i) { + unsigned char *d = (unsigned char *) data + (j*x+i)*comp; + if (write_alpha < 0) + fwrite(&d[comp-1], 1, 1, f); + switch (comp) { + case 1: + case 2: write3(f, d[0],d[0],d[0]); + break; + case 4: + if (!write_alpha) { + // composite against pink background + for (k=0; k < 3; ++k) + px[k] = bg[k] + ((d[k] - bg[k]) * d[3])/255; + write3(f, px[1-rgb_dir],px[1],px[1+rgb_dir]); + break; + } + /* FALLTHROUGH */ + case 3: + write3(f, d[1-rgb_dir],d[1],d[1+rgb_dir]); + break; + } + if (write_alpha > 0) + fwrite(&d[comp-1], 1, 1, f); + } + fwrite(&zero,scanline_pad,1,f); + } +} + +static int outfile(char const *filename, int rgb_dir, int vdir, int x, int y, int comp, void *data, int alpha, int pad, const char *fmt, ...) +{ + FILE *f; + if (y < 0 || x < 0) return 0; + f = fopen(filename, "wb"); + if (f) { + va_list v; + va_start(v, fmt); + writefv(f, fmt, v); + va_end(v); + write_pixels(f,rgb_dir,vdir,x,y,comp,data,alpha,pad); + fclose(f); + } + return f != NULL; +} + +int stbi_write_bmp(char const *filename, int x, int y, int comp, const void *data) +{ + int pad = (-x*3) & 3; + return outfile(filename,-1,-1,x,y,comp,(void *) data,0,pad, + "11 4 22 4" "4 44 22 444444", + 'B', 'M', 14+40+(x*3+pad)*y, 0,0, 14+40, // file header + 40, x,y, 1,24, 0,0,0,0,0,0); // bitmap header +} + +int stbi_write_tga(char const *filename, int x, int y, int comp, const void *data) +{ + int has_alpha = !(comp & 1); + return outfile(filename, -1,-1, x, y, comp, (void *) data, has_alpha, 0, + "111 221 2222 11", 0,0,2, 0,0,0, 0,0,x,y, 24+8*has_alpha, 8*has_alpha); +} + +// stretchy buffer; stbi__sbpush() == vector<>::push_back() -- stbi__sbcount() == vector<>::size() +#define stbi__sbraw(a) ((int *) (a) - 2) +#define stbi__sbm(a) stbi__sbraw(a)[0] +#define stbi__sbn(a) stbi__sbraw(a)[1] + +#define stbi__sbneedgrow(a,n) ((a)==0 || stbi__sbn(a)+n >= stbi__sbm(a)) +#define stbi__sbmaybegrow(a,n) (stbi__sbneedgrow(a,(n)) ? stbi__sbgrow(a,n) : 0) +#define stbi__sbgrow(a,n) stbi__sbgrowf((void **) &(a), (n), sizeof(*(a))) + +#define stbi__sbpush(a, v) (stbi__sbmaybegrow(a,1), (a)[stbi__sbn(a)++] = (v)) +#define stbi__sbcount(a) ((a) ? stbi__sbn(a) : 0) +#define stbi__sbfree(a) ((a) ? free(stbi__sbraw(a)),0 : 0) + +static void *stbi__sbgrowf(void **arr, int increment, int itemsize) +{ + int m = *arr ? 2*stbi__sbm(*arr)+increment : increment+1; + void *p = realloc(*arr ? stbi__sbraw(*arr) : 0, itemsize * m + sizeof(int)*2); + assert(p); + if (p) { + if (!*arr) ((int *) p)[1] = 0; + *arr = (void *) ((int *) p + 2); + stbi__sbm(*arr) = m; + } + return *arr; +} + +static unsigned char *stbi__zlib_flushf(unsigned char *data, unsigned int *bitbuffer, int *bitcount) +{ + while (*bitcount >= 8) { + stbi__sbpush(data, (unsigned char) *bitbuffer); + *bitbuffer >>= 8; + *bitcount -= 8; + } + return data; +} + +static int stbi__zlib_bitrev(int code, int codebits) +{ + int res=0; + while (codebits--) { + res = (res << 1) | (code & 1); + code >>= 1; + } + return res; +} + +static unsigned int stbi__zlib_countm(unsigned char *a, unsigned char *b, int limit) +{ + int i; + for (i=0; i < limit && i < 258; ++i) + if (a[i] != b[i]) break; + return i; +} + +static unsigned int stbi__zhash(unsigned char *data) +{ + stbiw_uint32 hash = data[0] + (data[1] << 8) + (data[2] << 16); + hash ^= hash << 3; + hash += hash >> 5; + hash ^= hash << 4; + hash += hash >> 17; + hash ^= hash << 25; + hash += hash >> 6; + return hash; +} + +#define stbi__zlib_flush() (out = stbi__zlib_flushf(out, &bitbuf, &bitcount)) +#define stbi__zlib_add(code,codebits) \ + (bitbuf |= (code) << bitcount, bitcount += (codebits), stbi__zlib_flush()) +#define stbi__zlib_huffa(b,c) stbi__zlib_add(stbi__zlib_bitrev(b,c),c) +// default huffman tables +#define stbi__zlib_huff1(n) stbi__zlib_huffa(0x30 + (n), 8) +#define stbi__zlib_huff2(n) stbi__zlib_huffa(0x190 + (n)-144, 9) +#define stbi__zlib_huff3(n) stbi__zlib_huffa(0 + (n)-256,7) +#define stbi__zlib_huff4(n) stbi__zlib_huffa(0xc0 + (n)-280,8) +#define stbi__zlib_huff(n) ((n) <= 143 ? stbi__zlib_huff1(n) : (n) <= 255 ? stbi__zlib_huff2(n) : (n) <= 279 ? stbi__zlib_huff3(n) : stbi__zlib_huff4(n)) +#define stbi__zlib_huffb(n) ((n) <= 143 ? stbi__zlib_huff1(n) : stbi__zlib_huff2(n)) + +#define stbi__ZHASH 16384 + +unsigned char * stbi_zlib_compress(unsigned char *data, int data_len, int *out_len, int quality) +{ + static unsigned short lengthc[] = { 3,4,5,6,7,8,9,10,11,13,15,17,19,23,27,31,35,43,51,59,67,83,99,115,131,163,195,227,258, 259 }; + static unsigned char lengtheb[]= { 0,0,0,0,0,0,0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0 }; + static unsigned short distc[] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193,257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577, 32768 }; + static unsigned char disteb[] = { 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13 }; + unsigned int bitbuf=0; + int i,j, bitcount=0; + unsigned char *out = NULL; + unsigned char **hash_table[stbi__ZHASH]; // 64KB on the stack! + if (quality < 5) quality = 5; + + stbi__sbpush(out, 0x78); // DEFLATE 32K window + stbi__sbpush(out, 0x5e); // FLEVEL = 1 + stbi__zlib_add(1,1); // BFINAL = 1 + stbi__zlib_add(1,2); // BTYPE = 1 -- fixed huffman + + for (i=0; i < stbi__ZHASH; ++i) + hash_table[i] = NULL; + + i=0; + while (i < data_len-3) { + // hash next 3 bytes of data to be compressed + int h = stbi__zhash(data+i)&(stbi__ZHASH-1), best=3; + unsigned char *bestloc = 0; + unsigned char **hlist = hash_table[h]; + int n = stbi__sbcount(hlist); + for (j=0; j < n; ++j) { + if (hlist[j]-data > i-32768) { // if entry lies within window + int d = stbi__zlib_countm(hlist[j], data+i, data_len-i); + if (d >= best) best=d,bestloc=hlist[j]; + } + } + // when hash table entry is too long, delete half the entries + if (hash_table[h] && stbi__sbn(hash_table[h]) == 2*quality) { + memcpy(hash_table[h], hash_table[h]+quality, sizeof(hash_table[h][0])*quality); + stbi__sbn(hash_table[h]) = quality; + } + stbi__sbpush(hash_table[h],data+i); + + if (bestloc) { + // "lazy matching" - check match at *next* byte, and if it's better, do cur byte as literal + h = stbi__zhash(data+i+1)&(stbi__ZHASH-1); + hlist = hash_table[h]; + n = stbi__sbcount(hlist); + for (j=0; j < n; ++j) { + if (hlist[j]-data > i-32767) { + int e = stbi__zlib_countm(hlist[j], data+i+1, data_len-i-1); + if (e > best) { // if next match is better, bail on current match + bestloc = NULL; + break; + } + } + } + } + + if (bestloc) { + int d = data+i - bestloc; // distance back + assert(d <= 32767 && best <= 258); + for (j=0; best > lengthc[j+1]-1; ++j); + stbi__zlib_huff(j+257); + if (lengtheb[j]) stbi__zlib_add(best - lengthc[j], lengtheb[j]); + for (j=0; d > distc[j+1]-1; ++j); + stbi__zlib_add(stbi__zlib_bitrev(j,5),5); + if (disteb[j]) stbi__zlib_add(d - distc[j], disteb[j]); + i += best; + } else { + stbi__zlib_huffb(data[i]); + ++i; + } + } + // write out final bytes + for (;i < data_len; ++i) + stbi__zlib_huffb(data[i]); + stbi__zlib_huff(256); // end of block + // pad with 0 bits to byte boundary + while (bitcount) + stbi__zlib_add(0,1); + + for (i=0; i < stbi__ZHASH; ++i) + (void) stbi__sbfree(hash_table[i]); + + { + // compute adler32 on input + unsigned int i=0, s1=1, s2=0, blocklen = data_len % 5552; + int j=0; + while (j < data_len) { + for (i=0; i < blocklen; ++i) s1 += data[j+i], s2 += s1; + s1 %= 65521, s2 %= 65521; + j += blocklen; + blocklen = 5552; + } + stbi__sbpush(out, (unsigned char) (s2 >> 8)); + stbi__sbpush(out, (unsigned char) s2); + stbi__sbpush(out, (unsigned char) (s1 >> 8)); + stbi__sbpush(out, (unsigned char) s1); + } + *out_len = stbi__sbn(out); + // make returned pointer freeable + memmove(stbi__sbraw(out), out, *out_len); + return (unsigned char *) stbi__sbraw(out); +} + +unsigned int stbi__crc32(unsigned char *buffer, int len) +{ + static unsigned int crc_table[256]; + unsigned int crc = ~0u; + int i,j; + if (crc_table[1] == 0) + for(i=0; i < 256; i++) + for (crc_table[i]=i, j=0; j < 8; ++j) + crc_table[i] = (crc_table[i] >> 1) ^ (crc_table[i] & 1 ? 0xedb88320 : 0); + for (i=0; i < len; ++i) + crc = (crc >> 8) ^ crc_table[buffer[i] ^ (crc & 0xff)]; + return ~crc; +} + +#define stbi__wpng4(o,a,b,c,d) ((o)[0]=(unsigned char)(a),(o)[1]=(unsigned char)(b),(o)[2]=(unsigned char)(c),(o)[3]=(unsigned char)(d),(o)+=4) +#define stbi__wp32(data,v) stbi__wpng4(data, (v)>>24,(v)>>16,(v)>>8,(v)); +#define stbi__wptag(data,s) stbi__wpng4(data, s[0],s[1],s[2],s[3]) + +static void stbi__wpcrc(unsigned char **data, int len) +{ + unsigned int crc = stbi__crc32(*data - len - 4, len+4); + stbi__wp32(*data, crc); +} + +static unsigned char stbi__paeth(int a, int b, int c) +{ + int p = a + b - c, pa = abs(p-a), pb = abs(p-b), pc = abs(p-c); + if (pa <= pb && pa <= pc) return (unsigned char) a; + if (pb <= pc) return (unsigned char) b; + return (unsigned char) c; +} + +unsigned char *stbi_write_png_to_mem(unsigned char *pixels, int stride_bytes, int x, int y, int n, int *out_len) +{ + int ctype[5] = { -1, 0, 4, 2, 6 }; + unsigned char sig[8] = { 137,80,78,71,13,10,26,10 }; + unsigned char *out,*o, *filt, *zlib; + signed char *line_buffer; + int i,j,k,p,zlen; + + if (stride_bytes == 0) + stride_bytes = x * n; + + filt = (unsigned char *) malloc((x*n+1) * y); if (!filt) return 0; + line_buffer = (signed char *) malloc(x * n); if (!line_buffer) { free(filt); return 0; } + for (j=0; j < y; ++j) { + static int mapping[] = { 0,1,2,3,4 }; + static int firstmap[] = { 0,1,0,5,6 }; + int *mymap = j ? mapping : firstmap; + int best = 0, bestval = 0x7fffffff; + for (p=0; p < 2; ++p) { + for (k= p?best:0; k < 5; ++k) { + int type = mymap[k],est=0; + unsigned char *z = pixels + stride_bytes*j; + for (i=0; i < n; ++i) + switch (type) { + case 0: line_buffer[i] = z[i]; break; + case 1: line_buffer[i] = z[i]; break; + case 2: line_buffer[i] = z[i] - z[i-stride_bytes]; break; + case 3: line_buffer[i] = z[i] - (z[i-stride_bytes]>>1); break; + case 4: line_buffer[i] = (signed char) (z[i] - stbi__paeth(0,z[i-stride_bytes],0)); break; + case 5: line_buffer[i] = z[i]; break; + case 6: line_buffer[i] = z[i]; break; + } + for (i=n; i < x*n; ++i) { + switch (type) { + case 0: line_buffer[i] = z[i]; break; + case 1: line_buffer[i] = z[i] - z[i-n]; break; + case 2: line_buffer[i] = z[i] - z[i-stride_bytes]; break; + case 3: line_buffer[i] = z[i] - ((z[i-n] + z[i-stride_bytes])>>1); break; + case 4: line_buffer[i] = z[i] - stbi__paeth(z[i-n], z[i-stride_bytes], z[i-stride_bytes-n]); break; + case 5: line_buffer[i] = z[i] - (z[i-n]>>1); break; + case 6: line_buffer[i] = z[i] - stbi__paeth(z[i-n], 0,0); break; + } + } + if (p) break; + for (i=0; i < x*n; ++i) + est += abs((signed char) line_buffer[i]); + if (est < bestval) { bestval = est; best = k; } + } + } + // when we get here, best contains the filter type, and line_buffer contains the data + filt[j*(x*n+1)] = (unsigned char) best; + memcpy(filt+j*(x*n+1)+1, line_buffer, x*n); + } + free(line_buffer); + zlib = stbi_zlib_compress(filt, y*( x*n+1), &zlen, 8); // increase 8 to get smaller but use more memory + free(filt); + if (!zlib) return 0; + + // each tag requires 12 bytes of overhead + out = (unsigned char *) malloc(8 + 12+13 + 12+zlen + 12); + if (!out) return 0; + *out_len = 8 + 12+13 + 12+zlen + 12; + + o=out; + memcpy(o,sig,8); o+= 8; + stbi__wp32(o, 13); // header length + stbi__wptag(o, "IHDR"); + stbi__wp32(o, x); + stbi__wp32(o, y); + *o++ = 8; + *o++ = (unsigned char) ctype[n]; + *o++ = 0; + *o++ = 0; + *o++ = 0; + stbi__wpcrc(&o,13); + + stbi__wp32(o, zlen); + stbi__wptag(o, "IDAT"); + memcpy(o, zlib, zlen); o += zlen; free(zlib); + stbi__wpcrc(&o, zlen); + + stbi__wp32(o,0); + stbi__wptag(o, "IEND"); + stbi__wpcrc(&o,0); + + assert(o == out + *out_len); + + return out; +} + +int stbi_write_png(char const *filename, int x, int y, int comp, const void *data, int stride_bytes) +{ + FILE *f; + int len; + unsigned char *png = stbi_write_png_to_mem((unsigned char *) data, stride_bytes, x, y, comp, &len); + if (!png) return 0; + f = fopen(filename, "wb"); + if (!f) { free(png); return 0; } + fwrite(png, 1, len, f); + fclose(f); + free(png); + return 1; +} +#endif // STB_IMAGE_WRITE_IMPLEMENTATION + +/* Revision history + + 0.92 (2010-08-01) + casts to unsigned char to fix warnings + 0.91 (2010-07-17) + first public release + 0.90 first internal release +*/ diff --git a/src/timer.h b/src/timer.h new file mode 100644 index 0000000..e0f4b68 --- /dev/null +++ b/src/timer.h @@ -0,0 +1,148 @@ +#ifndef __MMC_TIMER_H__ +#define __MMC_TIMER_H__ + +#ifdef WIN32 + +/*! win32 implementation -- use OS-provided high-precision timer */ +#ifndef WIN32_LEAN_AND_MEAN +#define WIN32_LEAN_AND_MEAN +#endif // WIN32_LEAN_AND_MEAN +#include + +typedef _int64 MMC_TIME_TYPE; + +#else +#ifdef LINUX + +#include +#include +typedef double MMC_TIME_TYPE; + +#else + +/*! not win32 -- use C-provided clock timing methods; less precise, but + but provide same basic functionality; + \todo Other OS's may provide similar OS-specific high-precision + timer routines, which are preferable to the clock() method. */ +#include +typedef clock_t MMC_TIME_TYPE; + +#endif // LINUX +#endif // win32 + +namespace mmc +{ + + /*! \brief Class for timing application events. + Only one of these is needed per application. */ + class Timer + { + public: + + /*! Get a new timer, initially stopped. */ + inline Timer (); + + /*! Start the timer. */ + inline void start (); + + /*! Call this once per frame to advance internal timer state. */ + inline void inc (); + + /*! Returns number of ms elapsed since start() was called. Use + this to determine how long the timer has been running. */ + inline MMC_TIME_TYPE queryElapsed () const; + + /*! Returns number of ms elapsed between the two previous calls to + \p inc(). Use this to determine the time between frames. */ + inline MMC_TIME_TYPE queryInc () const; + + /*! Returns 1 / frequency of the internal timing clock actually + being used. */ + inline double getInvFreq () const; + + private: + MMC_TIME_TYPE freq_; + MMC_TIME_TYPE start_; + MMC_TIME_TYPE now_; + MMC_TIME_TYPE last_; + MMC_TIME_TYPE elapsed_; + double invFreq_; + }; + + ////////////////////////////////////////////////////////////////////// + // Implementation. + // + + inline + Timer::Timer () + : start_(0), last_(0), now_(0), elapsed_(0) + { +#ifdef WIN32 + QueryPerformanceFrequency((LARGE_INTEGER *) &freq_); +#else +#ifdef LINUX + freq_ = 1.0; +#else + freq_ = CLOCKS_PER_SEC; +#endif +#endif + invFreq_ = 1.0 / (double) freq_; + } + + inline void + Timer::start () + { +#ifdef WIN32 + QueryPerformanceCounter((LARGE_INTEGER *) &start_); +#else +#ifdef LINUX + struct timeval tv; + struct timezone tz; + gettimeofday(&tv, &tz); + start_ = (double) (tv.tv_sec + 1e-6 * tv.tv_usec); +#else + start_ = clock(); +#endif +#endif + last_ = now_ = start_; + } + + inline MMC_TIME_TYPE + Timer::queryElapsed () const + { + return now_ - start_; + } + + inline void + Timer::inc () + { + last_ = now_; +#ifdef WIN32 + QueryPerformanceCounter((LARGE_INTEGER *) &now_); +#else +#ifdef LINUX + struct timeval tv; + struct timezone tz; + gettimeofday(&tv, &tz); + now_ = (double) (tv.tv_sec + 1e-6 * tv.tv_usec); +#else + now_ = clock(); +#endif +#endif + } + + inline MMC_TIME_TYPE + Timer::queryInc () const + { + return now_ - last_; + } + + inline double + Timer::getInvFreq () const + { + return invFreq_; + } + +} // namespace mmc + +#endif // __MMC_TIMER_H__ \ No newline at end of file diff --git a/src/utilities.cpp b/src/utilities.cpp index a8e5d90..6f0ae4d 100755 --- a/src/utilities.cpp +++ b/src/utilities.cpp @@ -4,7 +4,7 @@ // File: utilities.cpp // A collection/kitchen sink of generally useful functions -#define GLM_FORCE_RADIANS +//#define GLM_FORCE_RADIANS #include #include diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithNoSSAA.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithNoSSAA.bmp new file mode 100644 index 0000000..fddff53 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithNoSSAA.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithSSAA.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithSSAA.bmp new file mode 100644 index 0000000..d1749fe Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/DetailWithSSAA.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_AllEffects.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_AllEffects.png new file mode 100644 index 0000000..d3a26b4 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_AllEffects.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_DepthOfField.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_DepthOfField.png new file mode 100644 index 0000000..3ad3ae4 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_DepthOfField.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_MotionBlur.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_MotionBlur.png new file mode 100644 index 0000000..e66c8eb Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_MotionBlur.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_NormalMap.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_NormalMap.png new file mode 100644 index 0000000..318b481 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_NormalMap.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_TextureMap.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_TextureMap.png new file mode 100644 index 0000000..92ad7bc Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_TextureMap.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithNoSSAA.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithNoSSAA.png new file mode 100644 index 0000000..df17d4e Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithNoSSAA.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithOBJ.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithOBJ.png new file mode 100644 index 0000000..b31d6ae Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithOBJ.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithRefraction.png b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithRefraction.png new file mode 100644 index 0000000..5b038a3 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/FinalScene_WithRefraction.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/white line bug.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/white line bug.bmp new file mode 100644 index 0000000..4018161 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/Final Images/white line bug.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj b/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj index c45dd79..cd338a8 100644 --- a/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj @@ -46,10 +46,13 @@ true - $(SolutionDir)/../../external/lib/win/GLFW/;$(SolutionDir)/../../external/lib/win/GL/;%(AdditionalLibraryDirectories) + $(SolutionDir)/../../external/lib/win/FreeImage/;$(SolutionDir)/../../external/lib/win/GLFW/;$(SolutionDir)/../../external/lib/win/GL/;%(AdditionalLibraryDirectories) cudart.lib;glew32s.lib;glfw3.lib;opengl32.lib;%(AdditionalDependencies) Console + + compute_20,sm_20 + @@ -63,7 +66,7 @@ true true true - $(SolutionDir)/../../external/lib/win/GLFW/;$(SolutionDir)/../../external/lib/win/GL/;%(AdditionalLibraryDirectories) + $(SolutionDir)/../../external/lib/win/FreeImage/;$(SolutionDir)/../../external/lib/win/GLFW/;$(SolutionDir)/../../external/lib/win/GL/;%(AdditionalLibraryDirectories) cudart.lib;glew32s.lib;glfw3.lib;opengl32.lib;%(AdditionalDependencies) Console @@ -74,10 +77,16 @@ + + + + + + @@ -85,8 +94,10 @@ + + diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj.filters b/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj.filters index 584fd19..4a20021 100644 --- a/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj.filters +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/Project3-Pathtracer.vcxproj.filters @@ -48,6 +48,24 @@ Header Files + + Header Files + + + Header Files + + + Header Files + + + Header Files + + + Header Files + + + Header Files + @@ -71,6 +89,12 @@ Source Files\stb_image + + Source Files + + + Source Files + diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/clock.h b/windows/Project3-Pathtracer/Project3-Pathtracer/clock.h new file mode 100644 index 0000000..ca13984 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/clock.h @@ -0,0 +1,113 @@ + +#ifndef __MMC_CLOCK_H__ +#define __MMC_CLOCK_H__ + +#include "timer.h" + +namespace mmc +{ + + /*! \brief High precision clock for timing events. + + This class is implemented on top of the Timer class. Usually it is + not necessary to use the Timer class directly; instead, use this + class. */ + class Clock + { + public: + /*! Construct a clock. */ + inline Clock (); + + /*! Reset the clock (0 time elapsed). */ + inline void reset (); + + /*! Call once per frame to update the internal clock state. */ + inline void inc (); + + /*! Returns the amount of time (in ms) elapsed between last + two calls to \p inc(). */ + inline long queryInc () const; + + /*! Returns the amount of time (in ms) elapsed since clock creation or + reset() was called. */ + inline long queryTime () const; + + /*! Pause the clock. */ + inline void pauseToggle (); + + private: + Timer *timer_; + MMC_TIME_TYPE inc_; + MMC_TIME_TYPE curTime_; + MMC_TIME_TYPE start_; + bool paused_; + double invFreq_; + }; + + ////////////////////////////////////////////////////////////////////// + // Implementation. + // + + inline + Clock::Clock () + : paused_(false) + { + timer_ = new Timer; + invFreq_ = timer_->getInvFreq(); + timer_->start(); + reset(); + } + + void + Clock::reset () + { + start_ = timer_->queryElapsed(); + inc_ = curTime_ = 0; + } + + void + Clock::inc () + { + timer_->inc(); + if (!paused_) + { + inc_ = timer_->queryInc(); + curTime_ += inc_; + } + else + { + inc_ = 0; + } + } + + // Converts internal clock increment to long in milliseconds. + long + Clock::queryInc () const + { + return (long) (1000.0 * inc_ * invFreq_); + } + + // Converts internal clock time to long in milliseconds. + long + Clock::queryTime () const + { + return (long) (1000.0 * curTime_ * invFreq_); + } + + void + Clock::pauseToggle () + { + if (paused_) + { + paused_ = false; + } + else + { + paused_ = true; + inc_ = 0; + } + } + +} // namespace mmc + +#endif // __MMC_CLOCK_H__ \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/fps.cpp b/windows/Project3-Pathtracer/Project3-Pathtracer/fps.cpp new file mode 100644 index 0000000..b1d29ba --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/fps.cpp @@ -0,0 +1,57 @@ +#include "fps.h" +#include "clock.h" + +using namespace mmc; + +FpsTracker::FpsTracker (int smoothSteps) + : steps_(smoothSteps), nSnaps_(0) +{ + clock_ = new Clock; + snaps_ = new long[steps_]; + memset(snaps_, 0, sizeof(long) * steps_); +} + +FpsTracker::~FpsTracker () +{ + delete snaps_; + delete clock_; +} + +void + FpsTracker::timestamp () +{ + clock_->inc(); + snaps_[nSnaps_ % steps_] = clock_->queryInc(); + nSnaps_++; +} + +float + FpsTracker::fpsAverage () const +{ + int count = (nSnaps_ < steps_) ? nSnaps_ : steps_; + long sum = 0L; + for (int i = 0; i < count; ++i) + { + sum += snaps_[i]; + } + if (!sum) sum = 1L; // prevent div-by-zero + return 1000.0f * count / (float) sum; +} + +float + FpsTracker::fpsInstant () const +{ + long inc = clock_->queryInc(); + if (!inc) inc = 1L; // prevent div-by-zero + return 1000.0f / (float) inc; +} + +void + FpsTracker::setNumSteps (int smoothSteps) +{ + steps_ = smoothSteps; + delete snaps_; + snaps_ = new long[steps_]; + memset(snaps_, 0, sizeof(long) * steps_); + nSnaps_ = 0; +} \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/fps.h b/windows/Project3-Pathtracer/Project3-Pathtracer/fps.h new file mode 100644 index 0000000..4088c6a --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/fps.h @@ -0,0 +1,54 @@ + +/*! \file fps.h + \brief Defines a class for keeping track of frames/sec. + \author Eric Chan + \date July 2003 +*/ + +#ifndef __MMC_FPS_H__ +#define __MMC_FPS_H__ + +#include + +namespace mmc +{ + class Clock; + + /*! \brief Utility class for measuring framerate. */ + class FpsTracker + { + public: + /*! Create a FpsTracker. \p smoothSteps is the window size + over which to average the time measurements. */ + FpsTracker (int smoothSteps = 4); + + /*! Deallocate a FpsTracker. */ + ~FpsTracker (); + + /*! Specify the window size \p smoothSteps over which the time + measurements will be averaged. */ + void setNumSteps (int smoothSteps); + + /*! Makes a timestamp (i.e. takes a snapshot); measures + time intervals between successful calls. Usually, you want + to call this function once per rendering loop. */ + void timestamp (); + + /*! Get the average FPS (averaged over \p smoothSteps + interval). */ + float fpsAverage () const; + + /*! Get the instantaneous FPS (estimated from the last frame + only). */ + float fpsInstant () const; + + private: + Clock *clock_; + int steps_; + int nSnaps_; + long *snaps_; + }; + +} // namespace mmc + +#endif // __MMC_FPS_H__ \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/CompareScene_4999.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/CompareScene_4999.png new file mode 100644 index 0000000..71745ab Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/CompareScene_4999.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1164.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1164.png new file mode 100644 index 0000000..5741297 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1164.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1165.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1165.png new file mode 100644 index 0000000..f5ac81b Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1165.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1878.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1878.png new file mode 100644 index 0000000..0564a94 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_1878.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3429.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3429.png new file mode 100644 index 0000000..29d4381 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3429.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3430.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3430.png new file mode 100644 index 0000000..eca0f89 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3430.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3431.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3431.png new file mode 100644 index 0000000..df17d4e Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3431.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3432.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3432.png new file mode 100644 index 0000000..6ed66ef Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3432.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3433.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3433.png new file mode 100644 index 0000000..254936f Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene2_3433.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0012.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0012.png new file mode 100644 index 0000000..b2da276 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0012.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0013.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0013.png new file mode 100644 index 0000000..3c1143f Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_0013.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4929.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4929.png new file mode 100644 index 0000000..2b7986b Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4929.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4999.png b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4999.png new file mode 100644 index 0000000..55239f3 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/output/FinalScene_4999.png differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/CompareScene.txt b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/CompareScene.txt new file mode 100644 index 0000000..1fea7e2 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/CompareScene.txt @@ -0,0 +1,313 @@ +MATERIAL 0 //white diffuse +RGB 0.75, 0.75, 0.75 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 1 //red diffuse +RGB 0.75 0.25 0.25 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 2 //blue diffuse +RGB 0.25 0.25 0.75 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 3 //red glossy +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 4 //white glossy +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 5 //glass +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 1 +REFRIOR 1.5 +SCATTER 0 +ABSCOEFF .02 5.1 5.7 +RSCTCOEFF 13 +EMITTANCE 0 + +MATERIAL 6 //green glossy +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2.6 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 7 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 1 + +MATERIAL 8 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 10 + +MATERIAL 9 //yellow diffuse +RGB 1 1 0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 10 //red diffuse 2 +RGB 1.0 0.0 0.0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 11 //brown diffuse +RGB 0.9 0.4 0.0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + + +CAMERA +RES 800 800 +FOVY 25 +ITERATIONS 5000 +FILE test.bmp +frame 0 +EYE 0 4.5 15.5 +VIEW 0 0 -1 +UP 0 1 0 +DOFL 14.0 +DOFR 0.7 + +OBJECT 0 +cube +material 0 +frame 0 +TRANS 0 0 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/wood.jpg +BUMP texture/bumpmap.jpg + +OBJECT 1 +cube +material 0 +frame 0 +TRANS 0 5 -5 +ROTAT 0 90 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 2 +cube +material 0 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP +BUMP + +OBJECT 3 +cube +material 1 +frame 0 +TRANS -5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 4 +cube +material 2 +frame 0 +TRANS 5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 5 +sphere +material 5 +frame 0 +TRANS -2.0 5.0 0.0 +ROTAT 0 180 0 +SCALE 1.8 1.8 1.8 +MBV 0 0 0 +MAP +BUMP + +OBJECT 6 +sphere +material 4 +frame 0 +TRANS 0.0 5 0 +ROTAT 0 180 0 +SCALE 1.8 1.8 1.8 +MBV 0 0 0 +MAP +BUMP + +OBJECT 7 +sphere +material 10 +frame 0 +TRANS 2 5 0 +ROTAT 0 180 0 +SCALE 1.8 1.8 1.8 +MBV 0.01 0 0 +MAP texture/worldmap.jpg +BUMP + + +OBJECT 8 +cube +material 8 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .1 3.8 3.8 +MBV 0 0 0 +MAP +BUMP + + +OBJECT 9 +cube +material 0 +frame 0 +TRANS 0 3.5 0 +ROTAT 0 0 0 +SCALE 6.8 0.2 4.8 +MBV 0 0 0 +MAP +BUMP + +OBJECT 10 +cube +material 11 +frame 0 +TRANS 3.0 1.5 1.8 +ROTAT 0 0 0 +SCALE 0.3 4.0 0.3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 11 +cube +material 11 +frame 0 +TRANS 3.0 1.5 -1.8 +ROTAT 0 0 0 +SCALE 0.3 4.0 0.3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 12 +cube +material 11 +frame 0 +TRANS -3.0 1.5 -1.8 +ROTAT 0 0 0 +SCALE 0.3 4.0 0.3 +MBV 0 0 0 +MAP +BUMP + + +OBJECT 13 +cube +material 11 +frame 0 +TRANS -3.0 1.5 1.8 +ROTAT 0 0 0 +SCALE 0.3 4.0 0.3 +MBV 0 0 0 +MAP +BUMP \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene.txt b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene.txt new file mode 100644 index 0000000..82b5165 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene.txt @@ -0,0 +1,231 @@ +MATERIAL 0 //white diffuse +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 1 //red diffuse +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 2 //green diffuse +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 3 //red glossy +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 4 //white glossy +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 5 //glass +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 1 +REFRIOR 2.2 +SCATTER 0 +ABSCOEFF .02 5.1 5.7 +RSCTCOEFF 13 +EMITTANCE 0 + +MATERIAL 6 //green glossy +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2.6 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 7 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 1 + +MATERIAL 8 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 10 + +MATERIAL 9 //yellow diffuse +RGB 1 1 0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +CAMERA +RES 800 800 +FOVY 25 +ITERATIONS 5000 +FILE test.bmp +frame 0 +EYE 0 4.5 14 +VIEW 0 0 -1 +UP 0 1 0 +DOFL 14.0 +DOFR 0.7 + +OBJECT 0 +cube +material 0 +frame 0 +TRANS 0 0 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/wood.jpg +BUMP texture/bumpmap.jpg + +OBJECT 1 +cube +material 9 +frame 0 +TRANS 0 5 -5 +ROTAT 0 90 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 2 +cube +material 0 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP +BUMP + +OBJECT 3 +cube +material 1 +frame 0 +TRANS -5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 4 +cube +material 2 +frame 0 +TRANS 5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 5 +obj/tetra.obj +material 4 +frame 0 +TRANS 0 3 -2.5 +ROTAT 0 180 0 +SCALE 4.5 5 4.5 +MBV 0 0 0 +MAP +BUMP + +OBJECT 6 +sphere +material 5 +frame 0 +TRANS 3 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 7 +sphere +material 2 +frame 0 +TRANS -2 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0.01 0 0 +MAP texture/worldmap.jpg +BUMP + + +OBJECT 8 +cube +material 8 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .3 3 3 +MBV 0 0 0 +MAP +BUMP \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene2.txt b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene2.txt new file mode 100644 index 0000000..628ae52 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene2.txt @@ -0,0 +1,231 @@ +MATERIAL 0 //white diffuse +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 1 //red diffuse +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 2 //green diffuse +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 3 //red glossy +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 4 //white glossy +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 5 //glass +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 1 +REFRIOR 2.2 +SCATTER 0 +ABSCOEFF .02 5.1 5.7 +RSCTCOEFF 13 +EMITTANCE 0 + +MATERIAL 6 //green glossy +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2.6 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 7 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 1 + +MATERIAL 8 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 10 + +MATERIAL 9 //yellow diffuse +RGB 1 1 0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +CAMERA +RES 800 800 +FOVY 25 +ITERATIONS 5000 +FILE test.bmp +frame 0 +EYE 0 4.5 14 +VIEW 0 0 -1 +UP 0 1 0 +DOFL 14.0 +DOFR 0.7 + +OBJECT 0 +cube +material 0 +frame 0 +TRANS 0 0 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/wood.jpg +BUMP texture/bumpmap.jpg + +OBJECT 1 +cube +material 9 +frame 0 +TRANS 0 5 -5 +ROTAT 0 90 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 2 +cube +material 0 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP +BUMP + +OBJECT 3 +cube +material 1 +frame 0 +TRANS -5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 4 +cube +material 2 +frame 0 +TRANS 5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 5 +sphere +material 4 +frame 0 +TRANS 0.5 3 2 +ROTAT 40 0 90 +SCALE 2 2 2 +MBV 0 0 0 +MAP +BUMP + +OBJECT 6 +sphere +material 5 +frame 0 +TRANS 3 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 7 +sphere +material 2 +frame 0 +TRANS -2 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0.01 0 0 +MAP texture/worldmap.jpg +BUMP + + +OBJECT 8 +cube +material 8 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .3 3 3 +MBV 0 0 0 +MAP +BUMP \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene3.txt b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene3.txt new file mode 100644 index 0000000..0b44b70 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/scenes/FinalScene3.txt @@ -0,0 +1,231 @@ +MATERIAL 0 //white diffuse +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 1 //red diffuse +RGB .63 .06 .04 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 2 //green diffuse +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 3 //pink glossy +RGB .9 .7 .7 +SPECEX 0 +SPECRGB 0 1 0 +REFL 1 +REFR 0 +REFRIOR 2 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 4 //white glossy +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 0 0 +REFL 1 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 5 //glass +RGB 1 1 1 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 1 +REFRIOR 2.2 +SCATTER 0 +ABSCOEFF .02 5.1 5.7 +RSCTCOEFF 13 +EMITTANCE 0 + +MATERIAL 6 //green glossy +RGB .15 .48 .09 +SPECEX 0 +SPECRGB 1 1 1 +REFL 1 +REFR 0 +REFRIOR 2.6 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +MATERIAL 7 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 1 + +MATERIAL 8 //light +RGB 1 1 1 +SPECEX 0 +SPECRGB 0 0 0 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 10 + +MATERIAL 9 //yellow diffuse +RGB 1 1 0 +SPECEX 0 +SPECRGB 1 1 1 +REFL 0 +REFR 0 +REFRIOR 0 +SCATTER 0 +ABSCOEFF 0 0 0 +RSCTCOEFF 0 +EMITTANCE 0 + +CAMERA +RES 800 800 +FOVY 25 +ITERATIONS 5000 +FILE test.bmp +frame 0 +EYE 0 4.5 14 +VIEW 0 0 -1 +UP 0 1 0 +DOFL 14.0 +DOFR 0.7 + +OBJECT 0 +cube +material 0 +frame 0 +TRANS 0 0 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/brick.jpg +BUMP texture/brickbump.jpg + +OBJECT 1 +cube +material 9 +frame 0 +TRANS 0 5 -5 +ROTAT 0 90 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 2 +cube +material 0 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .01 10 10 +MBV 0 0 0 +MAP +BUMP + +OBJECT 3 +cube +material 1 +frame 0 +TRANS -5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 4 +cube +material 2 +frame 0 +TRANS 5 5 0 +ROTAT 0 0 0 +SCALE .01 10 10 +MBV 0 0 0 +MAP texture/bricks.bmp +BUMP + +OBJECT 5 +sphere +material 3 +frame 0 +TRANS 0 3 -2.5 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 6 +sphere +material 5 +frame 0 +TRANS 3 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0 0 0 +MAP +BUMP + +OBJECT 7 +sphere +material 2 +frame 0 +TRANS -2 5 0 +ROTAT 0 180 0 +SCALE 3 3 3 +MBV 0.01 0 0 +MAP texture/worldmap.jpg +BUMP + + +OBJECT 8 +cube +material 8 +frame 0 +TRANS 0 10 0 +ROTAT 0 0 90 +SCALE .3 3 3 +MBV 0 0 0 +MAP +BUMP \ No newline at end of file diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/test.0.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/test.0.bmp new file mode 100644 index 0000000..0bc241a Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/test.0.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/WorldBump.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/WorldBump.jpg new file mode 100644 index 0000000..6658823 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/WorldBump.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/apple.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/apple.jpg new file mode 100644 index 0000000..d91c15d Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/apple.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brick.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brick.jpg new file mode 100644 index 0000000..f3882cf Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brick.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brickbump.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brickbump.jpg new file mode 100644 index 0000000..0b27bef Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/brickbump.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bricks.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bricks.bmp new file mode 100644 index 0000000..35ed422 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bricks.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap.jpg new file mode 100644 index 0000000..935508e Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap2.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap2.jpg new file mode 100644 index 0000000..6216bab Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/bumpmap2.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/glass.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/glass.jpg new file mode 100644 index 0000000..b42d740 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/glass.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/lamp.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/lamp.jpg new file mode 100644 index 0000000..5367fee Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/lamp.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/mat.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/mat.jpg new file mode 100644 index 0000000..6f89038 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/mat.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/vase.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/vase.jpg new file mode 100644 index 0000000..e614425 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/vase.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.bmp b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.bmp new file mode 100644 index 0000000..24a1903 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.bmp differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.jpg new file mode 100644 index 0000000..05bd607 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood2.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood2.jpg new file mode 100644 index 0000000..e24507d Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood2.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_bigtable.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_bigtable.jpg new file mode 100644 index 0000000..6242235 Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_bigtable.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_cabinet.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_cabinet.jpg new file mode 100644 index 0000000..2a96c6d Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_cabinet.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_chair.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_chair.jpg new file mode 100644 index 0000000..8d6337d Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_chair.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_table.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_table.jpg new file mode 100644 index 0000000..45b834d Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/wood_table.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/texture/worldmap.jpg b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/worldmap.jpg new file mode 100644 index 0000000..4af20ae Binary files /dev/null and b/windows/Project3-Pathtracer/Project3-Pathtracer/texture/worldmap.jpg differ diff --git a/windows/Project3-Pathtracer/Project3-Pathtracer/timer.h b/windows/Project3-Pathtracer/Project3-Pathtracer/timer.h new file mode 100644 index 0000000..e0f4b68 --- /dev/null +++ b/windows/Project3-Pathtracer/Project3-Pathtracer/timer.h @@ -0,0 +1,148 @@ +#ifndef __MMC_TIMER_H__ +#define __MMC_TIMER_H__ + +#ifdef WIN32 + +/*! win32 implementation -- use OS-provided high-precision timer */ +#ifndef WIN32_LEAN_AND_MEAN +#define WIN32_LEAN_AND_MEAN +#endif // WIN32_LEAN_AND_MEAN +#include + +typedef _int64 MMC_TIME_TYPE; + +#else +#ifdef LINUX + +#include +#include +typedef double MMC_TIME_TYPE; + +#else + +/*! not win32 -- use C-provided clock timing methods; less precise, but + but provide same basic functionality; + \todo Other OS's may provide similar OS-specific high-precision + timer routines, which are preferable to the clock() method. */ +#include +typedef clock_t MMC_TIME_TYPE; + +#endif // LINUX +#endif // win32 + +namespace mmc +{ + + /*! \brief Class for timing application events. + Only one of these is needed per application. */ + class Timer + { + public: + + /*! Get a new timer, initially stopped. */ + inline Timer (); + + /*! Start the timer. */ + inline void start (); + + /*! Call this once per frame to advance internal timer state. */ + inline void inc (); + + /*! Returns number of ms elapsed since start() was called. Use + this to determine how long the timer has been running. */ + inline MMC_TIME_TYPE queryElapsed () const; + + /*! Returns number of ms elapsed between the two previous calls to + \p inc(). Use this to determine the time between frames. */ + inline MMC_TIME_TYPE queryInc () const; + + /*! Returns 1 / frequency of the internal timing clock actually + being used. */ + inline double getInvFreq () const; + + private: + MMC_TIME_TYPE freq_; + MMC_TIME_TYPE start_; + MMC_TIME_TYPE now_; + MMC_TIME_TYPE last_; + MMC_TIME_TYPE elapsed_; + double invFreq_; + }; + + ////////////////////////////////////////////////////////////////////// + // Implementation. + // + + inline + Timer::Timer () + : start_(0), last_(0), now_(0), elapsed_(0) + { +#ifdef WIN32 + QueryPerformanceFrequency((LARGE_INTEGER *) &freq_); +#else +#ifdef LINUX + freq_ = 1.0; +#else + freq_ = CLOCKS_PER_SEC; +#endif +#endif + invFreq_ = 1.0 / (double) freq_; + } + + inline void + Timer::start () + { +#ifdef WIN32 + QueryPerformanceCounter((LARGE_INTEGER *) &start_); +#else +#ifdef LINUX + struct timeval tv; + struct timezone tz; + gettimeofday(&tv, &tz); + start_ = (double) (tv.tv_sec + 1e-6 * tv.tv_usec); +#else + start_ = clock(); +#endif +#endif + last_ = now_ = start_; + } + + inline MMC_TIME_TYPE + Timer::queryElapsed () const + { + return now_ - start_; + } + + inline void + Timer::inc () + { + last_ = now_; +#ifdef WIN32 + QueryPerformanceCounter((LARGE_INTEGER *) &now_); +#else +#ifdef LINUX + struct timeval tv; + struct timezone tz; + gettimeofday(&tv, &tz); + now_ = (double) (tv.tv_sec + 1e-6 * tv.tv_usec); +#else + now_ = clock(); +#endif +#endif + } + + inline MMC_TIME_TYPE + Timer::queryInc () const + { + return now_ - last_; + } + + inline double + Timer::getInvFreq () const + { + return invFreq_; + } + +} // namespace mmc + +#endif // __MMC_TIMER_H__ \ No newline at end of file