diff --git a/assimp-vc141-mt.dll b/assimp-vc141-mt.dll
new file mode 100644
index 0000000..86f52a0
Binary files /dev/null and b/assimp-vc141-mt.dll differ
diff --git a/assimp-vc141-mtd.dll b/assimp-vc141-mtd.dll
new file mode 100644
index 0000000..2e5fd66
Binary files /dev/null and b/assimp-vc141-mtd.dll differ
diff --git a/cw 9/Zadania 9.html b/cw 9/Zadania 9.html
new file mode 100644
index 0000000..50d88b4
--- /dev/null
+++ b/cw 9/Zadania 9.html
@@ -0,0 +1,183 @@
+
+
+
+
+
+
+ Zadania 9
+
+
+
+
+
+
Shadow mapping
+
Cienie są ważnym elementem oświetlenie. Dodają realizmu do sceny i dzięki nim łatwiej jest graczowi zorientować się w przestrzennym rozłożeniu obiektów. Na powyższym obrazku możesz zobaczyć, że dużo łatwiej jest okreśłić położenie kostek, gdy rzucają one cienie. Podstawową techniką generowania cieni jest shadow mapping i wiele bardziej zaawansowanych technik na niej bazuje.
+
Składa się on z dwóch kroków. W pierwszym obliczamy mapy głębokości z perspektywy źródła światła i zapisaniu do tekstury (tą teksturę nazywamy shadowmap), w drugim przy rysowaniu fragmentu porównujemy jego odległość do źródła światła z odległością zapisaną w teksturze.
+
Celem tych zajęć będzie dodanie cieni do początkowej sceny. W obecnym projekcie są 3 źródła światła: światło słoneczne, reflektor samolociku i lampa planetarna. W trakcie zajęć skupimy się na świetle słonecznym.
+
Mapa głębokości - Framebuffers
+
Framebuffer to obiekt, do którego rednerowana jest scena w postaci tekstury. Do tej pory korzystaliśmy z domyślnego Famebuffora, który był wyświetlany na ekranie. Teraz potrzebujemy dodatkowy, który będzie przechwytywał mapę głębokości. Tworzymy go w następujący sposób
+
glGenFramebuffers(1, &depthMapFBO);
+
Zmienna depthMapFBO jest jak to typu unsignet int i powinna być dostępna globalnie. Kolejnym krokiem jest stworzenie tekstury głębokości
Również zmienna depthMap jest jak to typu unsignet int i powinna być dostępna globalnie. Tworzymy teksturę, zaznaczamy, że jest to tekstura głębokości nadając jej format GL_DEPTH_COMPONENT. Parametry SHADOW_WIDTH, SHADOW_HEIGHT są ustalone globalnie i oba wynoszą 1024.
Utwórz funkcję initDepthMap, w której zainicjalizujesz mapę głębokości. Wywołaj ją w funkcji init.
+
Renderowanie mapy głębokości
+
W tej części będziemy uzupełniać funkcję renderShadowapSun. Funkcja ma za zadanie zapisać w FBO mapę głębokości z perspektywy słońca. Pierwsze co musimy mieć, to parę shaderów, która będzie renderować mapę. Ponieważ jedyne co potrzebujemy tylko rozmieścić obiekty w odpowiednich miejscach. Shader wierzchołków ustawia tylko pozycję w oparciu o przesyłanie macierze.
Natomiast shader fragmentów jest pusty, ponieważ nic nie wysyłamy a głębokość zapisywana jest automatycznie.
+
#version 430 core
+
+void main()
+{
+}
+
Zadanie
+
Dodaj utwórz parę shaderów jak powyżej, załaduj je do zmiennej globalnej o nazwie programDepth i aktywuj go w funkcji renderShadowapSun. Utwórz funkcję drawObjectDepth. która będzie przyjmować referencję do RenderContext, macierz viewProjection i macierz modelu oraz przesyłać macierze do GPU i rysować RenderContext
+
Musimy zdefiniować macierz widoku i rzutowania, które mamy przesłać. Implementujemy cienie dla oświetlenia kierunkowego, gdzie dla każdego punktu kierunek światła jest taki sam. W takim wypadku skorzystamy z rzutowania prostopadłego.
+
Do stworzenia macierzy rzutowania perspektywicznego wykorzystamy funkcję:
Tworzy ona macierz rzutowania prostopadłego dla zadanych wymiarów. Musimy tak je dobrać, żeby rzutowanie zawierało całą interesującą scenę. Za małe wartości spowodują artefakty a za duże pogorszą jakość. Przykładowo możesz wziąć glm::ortho(-10.f, 10.f, -10.f, 10.f, 1.0f, 30.0f). Do stworzenia macierzy kamery wykorzystamy funkcję glm::lookAt z poniższymi argumentami argumentami
następnie wywołaj drawObjectDepth dla każdego obiektu, który rysujemy w naszej scenie. wykorzystaj macierz viewProjection zdefiniowaną wyżej, użyj tej samej macierzy modelu co przy właściwym rysowaniu.
+
Zakończ funkcję linią glBindFramebuffer(GL_FRAMEBUFFER, 0);, która przywraca domyślny FBO.
znajdujące się w renderScene. Rysują one prostokąt z mapą głębokości jako teksturą. Jeżeli wszystko zostało wykonane poprawnie, to powinien on zawierać rzutowanie naszego pokoju.
+
Zadanie
+
W tej chwili rzutowanie jest nieoptymalne. popraw je na lepsze. Zmodyfikuj wartości w glm::ortho(-10.f, 10.f, -10.f, 10.f, 1.0f, 30.0f) do takich, żeby pokój wypełniał jak największą część tekstury.
+
Rysowanie cieni
+
Na tym etapie powinniśmy mieć poprawnie stworzoną mapę głębokości. Pozostaje wykorzystać ją w oświetleniu. ### Zadanie #### Przesłanie danych W funkcji drawObjectPBR prześlij teksturę za pomocą instrukcji:
oraz macierz LightVP, która musi być taka sama jak w drawObjectDepth.
+
Następnie w shaderze wierzchołków odbierz LightVP.
+
Shader wierzchołków
+
Oblicz pozycję wierzchołka z perspektywy słońca sunSpacePos=LightVP*modelMatrix*vec4(vertexPosition,1), wynik prześlij do shadera fragmentów #### Shader fragmentów odbierz sunSpacePos i napisz funkcję calculateShadow, która sprawdza czy obiekt jest zacieniony. Aby to zrobić kolejno w funkcji: * ujednorodnij zmienną lightSpacePos dzieląc ją przez współrzędną w, * przeskaluj ją, ma wartości od -1 do 1 a potrzebujemy wartości od 0 do 1 (pomnóż przez 0.5 i dodaj 0.5) wynik zapisz do zmiennej lightSpacePosNormalized, * pobierz głębokość z depthMap próbkuj za pomocą współrzędnych x i y. Pobierz tylko kanał r, zapisz go do zmiennej closestDepth, * porównaj closestDepth ze współrzędną zlightSpacePosNormalized. jeżeli closestDepth jest większa zwróć 1.0, w przeciwnym wypadku zwróć 0.0. * wynik funkcji przemnóż z sunColor w trakcie oblicznia.
Powinniśmy dostać cienie , jednak w niezacienionych strefach pojawiły się paski, które znane są jako shadow acne wynikają one z błędu przybliżenia liczb zmiennoprzecinkowych. Można się go pozbyć na dwa sposoby 1. dodać bias. zamiast sprawdzać closestDepth<lightSpacePosNormalized można wziąć closestDepth+bias>lightSpacePosNormalized, gdzie bias to mała wartość (np 0.01). 2. innym rozwiązaniem jest, żeby przy renderowaniu cieni włączyć front face culling. dzięki temu rysowane będą część modelu, które są dalej niż te, które odpytujemy.
+
Zadanie*
+
Dodaj rysowane cieni również dla latarki doczepionej do statku. pamiętaj, że musisz wykorzystać tutaj macierz rzutowania perspektywicznego
+
+
diff --git a/cw 9/Zadania 9.md b/cw 9/Zadania 9.md
new file mode 100644
index 0000000..d457925
--- /dev/null
+++ b/cw 9/Zadania 9.md
@@ -0,0 +1,158 @@
+# Shadow mapping
+Cienie są ważnym elementem oświetlenie. Dodają realizmu do sceny i dzięki nim łatwiej jest graczowi zorientować się w przestrzennym rozłożeniu obiektów.
+
+Na powyższym obrazku możesz zobaczyć, że dużo łatwiej jest okreśłić położenie kostek, gdy rzucają one cienie. Podstawową techniką generowania cieni jest **shadow mapping** i wiele bardziej zaawansowanych technik na niej bazuje.
+
+Składa się on z dwóch kroków. W pierwszym obliczamy mapy głębokości z perspektywy źródła światła i zapisaniu do tekstury (tą teksturę nazywamy *shadowmap*), w drugim przy rysowaniu fragmentu porównujemy jego odległość do źródła światła z odległością zapisaną w teksturze.
+
+Celem tych zajęć będzie dodanie cieni do początkowej sceny. W obecnym projekcie są 3 źródła światła: światło słoneczne, reflektor samolociku i lampa planetarna. W trakcie zajęć skupimy się na świetle słonecznym.
+
+## Mapa głębokości - Framebuffers
+Framebuffer to obiekt, do którego rednerowana jest scena w postaci tekstury. Do tej pory korzystaliśmy z domyślnego Famebuffora, który był wyświetlany na ekranie. Teraz potrzebujemy dodatkowy, który będzie przechwytywał mapę głębokości. Tworzymy go w następujący sposób
+
+```C++
+glGenFramebuffers(1, &depthMapFBO);
+```
+Zmienna `depthMapFBO` jest jak to typu `unsignet int` i powinna być dostępna globalnie. Kolejnym krokiem jest stworzenie tekstury głębokości
+```C++
+glGenTextures(1, &depthMap);
+glBindTexture(GL_TEXTURE_2D, depthMap);
+glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT,
+ SHADOW_WIDTH, SHADOW_HEIGHT, 0, GL_DEPTH_COMPONENT, GL_FLOAT, NULL);
+glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
+glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
+glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT);
+glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT);
+```
+Również zmienna `depthMap` jest jak to typu `unsignet int` i powinna być dostępna globalnie.
+Tworzymy teksturę, zaznaczamy, że jest to tekstura głębokości nadając jej format `GL_DEPTH_COMPONENT`. Parametry `SHADOW_WIDTH`, `SHADOW_HEIGHT` są ustalone globalnie i oba wynoszą 1024.
+
+I w końcu podpinamy teksturę pod FBO.
+```C++
+
+glBindFramebuffer(GL_FRAMEBUFFER, depthMapFBO);
+glFramebufferTexture2D(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, GL_TEXTURE_2D, depthMap, 0);
+glDrawBuffer(GL_NONE);
+glReadBuffer(GL_NONE);
+glBindFramebuffer(GL_FRAMEBUFFER, 0);
+```
+
+### Zadanie
+Utwórz funkcję `initDepthMap`, w której zainicjalizujesz mapę głębokości. Wywołaj ją w funkcji `init`.
+
+### Renderowanie mapy głębokości
+
+W tej części będziemy uzupełniać funkcję `renderShadowapSun`. Funkcja ma za zadanie zapisać w FBO mapę głębokości z perspektywy słońca.
+Pierwsze co musimy mieć, to parę shaderów, która będzie renderować mapę. Ponieważ jedyne co potrzebujemy tylko rozmieścić obiekty w odpowiednich miejscach.
+Shader wierzchołków ustawia tylko pozycję w oparciu o przesyłanie macierze.
+```C++
+#version 430 core
+
+layout(location = 0) in vec3 vertexPosition;
+layout(location = 1) in vec3 vertexNormal;
+layout(location = 2) in vec2 vertexTexCoord;
+
+uniform mat4 viewProjectionMatrix;
+uniform mat4 modelMatrix;
+
+void main()
+{
+ gl_Position = viewProjectionMatrix * modelMatrix * vec4(vertexPosition, 1.0);
+}
+```
+Natomiast shader fragmentów jest pusty, ponieważ nic nie wysyłamy a głębokość zapisywana jest automatycznie.
+```C++
+#version 430 core
+
+void main()
+{
+}
+```
+### Zadanie
+Dodaj utwórz parę shaderów jak powyżej, załaduj je do zmiennej globalnej o nazwie `programDepth` i aktywuj go w funkcji `renderShadowapSun`. Utwórz funkcję `drawObjectDepth`. która będzie przyjmować referencję do `RenderContext`, macierz `viewProjection` i macierz modelu oraz przesyłać macierze do GPU i rysować `RenderContext`
+
+Musimy zdefiniować macierz widoku i rzutowania, które mamy przesłać. Implementujemy cienie dla oświetlenia kierunkowego, gdzie dla każdego punktu kierunek światła jest taki sam. W takim wypadku skorzystamy z rzutowania prostopadłego.
+
+
+Do stworzenia macierzy rzutowania perspektywicznego wykorzystamy funkcję:
+```C++
+glm::mat4 lightProjection = glm::ortho(
+ float left,
+ float right,
+ float bottom,
+ float top,
+ float zNear,
+ float zFar
+)
+```
+Tworzy ona macierz rzutowania prostopadłego dla zadanych wymiarów. Musimy tak je dobrać, żeby rzutowanie zawierało całą interesującą scenę. Za małe wartości spowodują artefakty a za duże pogorszą jakość. Przykładowo możesz wziąć `glm::ortho(-10.f, 10.f, -10.f, 10.f, 1.0f, 30.0f)`.
+Do stworzenia macierzy kamery wykorzystamy funkcję `glm::lookAt` z poniższymi argumentami argumentami
+```C++
+glm::lookAt(sunPos, sunPos - sunDir, glm::vec3(0, 1, 0))
+```
+### Zadanie
+Uzupełnij funkcję `renderShadowapSun`. Wywołaj w niej instrukcje
+```C++
+ //ustawianie przestrzeni rysowania
+ glViewport(0, 0, SHADOW_WIDTH, SHADOW_HEIGHT);
+ //bindowanie FBO
+ glBindFramebuffer(GL_FRAMEBUFFER, depthMapFBO);
+ //czyszczenie mapy głębokości
+ glClear(GL_DEPTH_BUFFER_BIT);
+ //ustawianie programu
+ glUseProgram(programDepth);
+```
+Stwórz macierz `viewProjection`
+```C++
+ glm::mat4 lightVP = glm::ortho(-10.f, 10.f, -10.f, 10.f, 1.0f, 30.0f) * glm::lookAt(sunPos, sunPos - sunDir, glm::vec3(0, 1, 0));
+```
+następnie wywołaj `drawObjectDepth` dla każdego obiektu, który rysujemy w naszej scenie. wykorzystaj macierz `viewProjection` zdefiniowaną wyżej, użyj tej samej macierzy modelu co przy właściwym rysowaniu.
+
+Zakończ funkcję linią `glBindFramebuffer(GL_FRAMEBUFFER, 0);`, która przywraca domyślny FBO.
+
+### Wizualizacja mapy głębokości
+odkomentuj linie:
+```C++
+ //glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
+ //glUseProgram(programTest);
+ //glActiveTexture(GL_TEXTURE0);
+ //glBindTexture(GL_TEXTURE_2D, depthMap);
+ //Core::DrawContext(models::testContext);
+```
+znajdujące się w `renderScene`. Rysują one prostokąt z mapą głębokości jako teksturą. Jeżeli wszystko zostało wykonane poprawnie, to powinien on zawierać rzutowanie naszego pokoju.
+
+### Zadanie
+W tej chwili rzutowanie jest nieoptymalne. popraw je na lepsze. Zmodyfikuj wartości w `glm::ortho(-10.f, 10.f, -10.f, 10.f, 1.0f, 30.0f)` do takich, żeby pokój wypełniał jak największą część tekstury.
+
+### Rysowanie cieni
+Na tym etapie powinniśmy mieć poprawnie stworzoną mapę głębokości. Pozostaje wykorzystać ją w oświetleniu.
+### Zadanie
+#### Przesłanie danych
+W funkcji `drawObjectPBR` prześlij teksturę za pomocą instrukcji:
+```C++
+ glActiveTexture(GL_TEXTURE0);
+ glBindTexture(GL_TEXTURE_2D, depthMap);
+```
+oraz macierz `LightVP`, która musi być taka sama jak w `drawObjectDepth`.
+
+Następnie w shaderze wierzchołków odbierz `LightVP`.
+
+#### Shader wierzchołków
+Oblicz pozycję wierzchołka z perspektywy słońca `sunSpacePos=LightVP*modelMatrix*vec4(vertexPosition,1)`, wynik prześlij do shadera fragmentów
+#### Shader fragmentów
+odbierz `sunSpacePos` i napisz funkcję `calculateShadow`, która sprawdza czy obiekt jest zacieniony. Aby to zrobić kolejno w funkcji:
+* ujednorodnij zmienną `lightSpacePos` dzieląc ją przez współrzędną **w**,
+* przeskaluj ją, ma wartości od -1 do 1 a potrzebujemy wartości od 0 do 1 (pomnóż przez 0.5 i dodaj 0.5) wynik zapisz do zmiennej `lightSpacePosNormalized`,
+* pobierz głębokość z `depthMap` próbkuj za pomocą współrzędnych **x** i **y**. Pobierz tylko kanał `r`, zapisz go do zmiennej `closestDepth`,
+* porównaj `closestDepth` ze współrzędną **z** `lightSpacePosNormalized`. jeżeli `closestDepth` jest większa zwróć 1.0, w przeciwnym wypadku zwróć 0.0.
+* wynik funkcji przemnóż z `sunColor` w trakcie oblicznia.
+```
+ ilumination=ilumination+PBRLight(sunDir,sunColor,normal,viewDir);
+```
+#### shadow acne
+Powinniśmy dostać cienie , jednak w niezacienionych strefach pojawiły się paski, które znane są jako *shadow acne* wynikają one z błędu przybliżenia liczb zmiennoprzecinkowych. Można się go pozbyć na dwa sposoby
+1. dodać bias. zamiast sprawdzać `closestDepthlightSpacePosNormalized`, gdzie `bias` to mała wartość (np 0.01).
+2. innym rozwiązaniem jest, żeby przy renderowaniu cieni włączyć front face culling. dzięki temu rysowane będą część modelu, które są dalej niż te, które odpytujemy.
+
+### Zadanie*
+Dodaj rysowane cieni również dla latarki doczepionej do statku. pamiętaj, że musisz wykorzystać tutaj macierz rzutowania perspektywicznego
diff --git a/cw 9/assimp-vc141-mt.dll b/cw 9/assimp-vc141-mt.dll
new file mode 100644
index 0000000..86f52a0
Binary files /dev/null and b/cw 9/assimp-vc141-mt.dll differ
diff --git a/cw 9/assimp-vc141-mtd.dll b/cw 9/assimp-vc141-mtd.dll
new file mode 100644
index 0000000..2e5fd66
Binary files /dev/null and b/cw 9/assimp-vc141-mtd.dll differ
diff --git a/cw 9/freeglut.dll b/cw 9/freeglut.dll
new file mode 100644
index 0000000..c768c15
Binary files /dev/null and b/cw 9/freeglut.dll differ
diff --git a/cw 9/glew32.dll b/cw 9/glew32.dll
new file mode 100644
index 0000000..8a49a3b
Binary files /dev/null and b/cw 9/glew32.dll differ
diff --git a/cw 9/grk-cw9.vcxproj b/cw 9/grk-cw9.vcxproj
new file mode 100644
index 0000000..8493ad1
--- /dev/null
+++ b/cw 9/grk-cw9.vcxproj
@@ -0,0 +1,126 @@
+
+
+
+
+ Debug
+ Win32
+
+
+ Release
+ Win32
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ {5BACD057-4B83-4CB6-A367-40A10BCE2149}
+ Win32Proj
+ grk-cw9
+ 10.0.19041.0
+ grk-cw9
+
+
+
+ Application
+ true
+ Unicode
+ v143
+
+
+ Application
+ false
+ true
+ Unicode
+ v142
+
+
+
+
+
+
+
+
+
+
+
+
+ false
+ $(SolutionDir)dependencies\freeglut\lib;$(SolutionDir)dependencies\glew-2.0.0\lib\Release\Win32;$(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019;$(SolutionDir)dependencies\assimp;$(LibraryPath)
+ $(SolutionDir)dependencies\freeglut\include\GL;$(SolutionDir)dependencies\glew-2.0.0\include\GL;$(SolutionDir)dependencies\glm;$(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\include;$(SolutionDir)dependencies\assimp\include;$(IncludePath)
+ $(ExecutablePath)
+
+
+ false
+ $(SolutionDir)dependencies\freeglut\lib;$(SolutionDir)dependencies\glew-2.0.0\lib\Release\Win32;$(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019;$(SolutionDir)dependencies\assimp;$(LibraryPath)
+ $(SolutionDir)dependencies\freeglut\include\GL;$(SolutionDir)dependencies\glew-2.0.0\include\GL;$(SolutionDir)dependencies\glm;$(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\include;$(SolutionDir)dependencies\assimp\include;$(IncludePath)
+
+
+
+
+
+ Level3
+ Disabled
+ WIN32;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)
+
+
+ Console
+ true
+ opengl32.lib;freeglut.lib;glew32.lib;$(SolutionDir)\dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019\glfw3.lib;$(SolutionDir)\dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019\glfw3dll.lib;zlibd.lib;assimp-vc141-mtd.lib;%(AdditionalDependencies)
+ $(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019;%(AdditionalLibraryDirectories)
+
+
+
+
+ Level3
+
+
+ MaxSpeed
+ true
+ true
+ WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)
+
+
+ Console
+ true
+ true
+ true
+ opengl32.lib;freeglut.lib;glew32.lib;$(SolutionDir)\dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019\glfw3.lib;$(SolutionDir)\dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019\glfw3dll.lib;zlibd.lib;assimp-vc141-mtd.lib;%(AdditionalDependencies)
+ $(SolutionDir)dependencies\glfw-3.3.8.bin.WIN32\lib-vc2019;%(AdditionalLibraryDirectories)
+
+
+
+
+
+
\ No newline at end of file
diff --git a/cw 9/grk-cw9.vcxproj.filters b/cw 9/grk-cw9.vcxproj.filters
new file mode 100644
index 0000000..a2b8ba9
--- /dev/null
+++ b/cw 9/grk-cw9.vcxproj.filters
@@ -0,0 +1,113 @@
+
+
+
+
+ {4FC737F1-C7A5-4376-A066-2A32D752A2FF}
+ cpp;c;cc;cxx;def;odl;idl;hpj;bat;asm;asmx
+
+
+ {93995380-89BD-4b04-88EB-625FBE52EBFB}
+ h;hpp;hxx;hm;inl;inc;xsd
+
+
+ {67DA6AB6-F800-4c08-8B7A-83BB121AAD01}
+ rc;ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe;resx;tiff;tif;png;wav;mfcribbon-ms
+
+
+ {0a247bb8-2e8e-4a90-b0ef-17415b0941ba}
+
+
+ {0af44075-33f4-4953-b1d6-1d28d61d758f}
+
+
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files\SOIL
+
+
+ Source Files
+
+
+
+
+ Shader Files
+
+
+ Shader Files
+
+
+ Shader Files
+
+
+ Shader Files
+
+
+ Shader Files
+
+
+ Shader Files
+
+
+
\ No newline at end of file
diff --git a/cw 9/img/2.png b/cw 9/img/2.png
new file mode 100644
index 0000000..e0537ce
Binary files /dev/null and b/cw 9/img/2.png differ
diff --git a/cw 9/img/3.webp b/cw 9/img/3.webp
new file mode 100644
index 0000000..cda42cc
Binary files /dev/null and b/cw 9/img/3.webp differ
diff --git a/cw 9/img/384px-PerlinNoiseDotProducts.png b/cw 9/img/384px-PerlinNoiseDotProducts.png
new file mode 100644
index 0000000..eb610e2
Binary files /dev/null and b/cw 9/img/384px-PerlinNoiseDotProducts.png differ
diff --git a/cw 9/img/384px-PerlinNoiseInterpolated.png b/cw 9/img/384px-PerlinNoiseInterpolated.png
new file mode 100644
index 0000000..97caa53
Binary files /dev/null and b/cw 9/img/384px-PerlinNoiseInterpolated.png differ
diff --git a/cw 9/img/4.png b/cw 9/img/4.png
new file mode 100644
index 0000000..0d74746
Binary files /dev/null and b/cw 9/img/4.png differ
diff --git a/cw 9/img/5.webp b/cw 9/img/5.webp
new file mode 100644
index 0000000..2157419
Binary files /dev/null and b/cw 9/img/5.webp differ
diff --git a/cw 9/img/6.webp b/cw 9/img/6.webp
new file mode 100644
index 0000000..6171ecc
Binary files /dev/null and b/cw 9/img/6.webp differ
diff --git a/cw 9/img/7.webp b/cw 9/img/7.webp
new file mode 100644
index 0000000..c13c85d
Binary files /dev/null and b/cw 9/img/7.webp differ
diff --git a/cw 9/img/8.webp b/cw 9/img/8.webp
new file mode 100644
index 0000000..9c22d8c
Binary files /dev/null and b/cw 9/img/8.webp differ
diff --git a/cw 9/img/Figure_1.eps b/cw 9/img/Figure_1.eps
new file mode 100644
index 0000000..8f26308
--- /dev/null
+++ b/cw 9/img/Figure_1.eps
@@ -0,0 +1,1254 @@
+%!PS-Adobe-3.0 EPSF-3.0
+%%Title: Figure_1.eps
+%%Creator: Matplotlib v3.4.3, https://matplotlib.org/
+%%CreationDate: Thu Nov 3 12:06:06 2022
+%%Orientation: portrait
+%%BoundingBox: 75 223 537 569
+%%HiResBoundingBox: 75.600000 223.200000 536.400000 568.800000
+%%EndComments
+%%BeginProlog
+/mpldict 11 dict def
+mpldict begin
+/_d { bind def } bind def
+/m { moveto } _d
+/l { lineto } _d
+/r { rlineto } _d
+/c { curveto } _d
+/cl { closepath } _d
+/ce { closepath eofill } _d
+/box {
+ m
+ 1 index 0 r
+ 0 exch r
+ neg 0 r
+ cl
+ } _d
+/clipbox {
+ box
+ clip
+ newpath
+ } _d
+/sc { setcachedevice } _d
+%!PS-Adobe-3.0 Resource-Font
+%%Creator: Converted from TrueType to Type 3 by Matplotlib.
+10 dict begin
+/FontName /DejaVuSans def
+/PaintType 0 def
+/FontMatrix [0.00048828125 0 0 0.00048828125 0 0] def
+/FontBBox [-2090 -948 3673 2524] def
+/FontType 3 def
+/Encoding [/aogonek /minus /space /period /zero /one /two /five /seven /a /c /d /e /g /h /i /j /m /n /o /p /r /z] def
+/CharStrings 24 dict dup begin
+/.notdef 0 def
+/aogonek{1255 0 123 -395 1152 1147 sc
+702 563 m
+553 563 450 546 393 512 c
+336 478 307 420 307 338 c
+307 273 328 221 371 182 c
+414 144 473 125 547 125 c
+649 125 731 161 792 233 c
+854 306 885 402 885 522 c
+885 563 l
+702 563 l
+
+1069 639 m
+1069 0 l
+885 0 l
+885 170 l
+843 102 791 52 728 19 c
+665 -13 589 -29 498 -29 c
+383 -29 292 3 224 67 c
+157 132 123 218 123 326 c
+123 452 165 547 249 611 c
+334 675 460 707 627 707 c
+885 707 l
+885 725 l
+885 810 857 875 801 921 c
+746 968 668 991 567 991 c
+503 991 441 983 380 968 c
+319 953 261 930 205 899 c
+205 1069 l
+272 1095 338 1114 401 1127 c
+464 1140 526 1147 586 1147 c
+748 1147 869 1105 949 1021 c
+1029 937 1069 810 1069 639 c
+
+887 0 m
+1006 0 l
+976 -41 954 -76 939 -105 c
+925 -134 918 -159 918 -180 c
+918 -211 927 -234 945 -249 c
+964 -264 991 -272 1027 -272 c
+1048 -272 1069 -269 1090 -264 c
+1111 -259 1131 -252 1152 -242 c
+1152 -375 l
+1127 -382 1103 -387 1080 -390 c
+1057 -393 1036 -395 1016 -395 c
+935 -395 875 -380 836 -351 c
+798 -322 779 -277 779 -215 c
+779 -183 788 -149 805 -114 c
+823 -79 850 -41 887 0 c
+
+ce} _d
+/minus{1716 0 217 557 1499 727 sc
+217 727 m
+1499 727 l
+1499 557 l
+217 557 l
+217 727 l
+
+ce} _d
+/space{651 0 0 0 0 0 sc
+ce} _d
+/period{651 0 219 0 430 254 sc
+219 254 m
+430 254 l
+430 0 l
+219 0 l
+219 254 l
+
+ce} _d
+/zero{1303 0 135 -29 1167 1520 sc
+651 1360 m
+547 1360 469 1309 416 1206 c
+364 1104 338 950 338 745 c
+338 540 364 387 416 284 c
+469 182 547 131 651 131 c
+756 131 834 182 886 284 c
+939 387 965 540 965 745 c
+965 950 939 1104 886 1206 c
+834 1309 756 1360 651 1360 c
+
+651 1520 m
+818 1520 946 1454 1034 1321 c
+1123 1189 1167 997 1167 745 c
+1167 494 1123 302 1034 169 c
+946 37 818 -29 651 -29 c
+484 -29 356 37 267 169 c
+179 302 135 494 135 745 c
+135 997 179 1189 267 1321 c
+356 1454 484 1520 651 1520 c
+
+ce} _d
+/one{1303 0 225 0 1114 1493 sc
+254 170 m
+584 170 l
+584 1309 l
+225 1237 l
+225 1421 l
+582 1493 l
+784 1493 l
+784 170 l
+1114 170 l
+1114 0 l
+254 0 l
+254 170 l
+
+ce} _d
+/two{1303 0 150 0 1098 1520 sc
+393 170 m
+1098 170 l
+1098 0 l
+150 0 l
+150 170 l
+227 249 331 356 463 489 c
+596 623 679 709 713 748 c
+778 821 823 882 848 932 c
+874 983 887 1032 887 1081 c
+887 1160 859 1225 803 1275 c
+748 1325 675 1350 586 1350 c
+523 1350 456 1339 385 1317 c
+315 1295 240 1262 160 1217 c
+160 1421 l
+241 1454 317 1478 388 1495 c
+459 1512 523 1520 582 1520 c
+737 1520 860 1481 952 1404 c
+1044 1327 1090 1223 1090 1094 c
+1090 1033 1078 974 1055 919 c
+1032 864 991 800 930 725 c
+913 706 860 650 771 557 c
+682 465 556 336 393 170 c
+
+ce} _d
+/five{1303 0 158 -29 1124 1493 sc
+221 1493 m
+1014 1493 l
+1014 1323 l
+406 1323 l
+406 957 l
+435 967 465 974 494 979 c
+523 984 553 987 582 987 c
+749 987 881 941 978 850 c
+1075 759 1124 635 1124 479 c
+1124 318 1074 193 974 104 c
+874 15 733 -29 551 -29 c
+488 -29 424 -24 359 -13 c
+294 -2 227 14 158 35 c
+158 238 l
+218 205 280 181 344 165 c
+408 149 476 141 547 141 c
+662 141 754 171 821 232 c
+888 293 922 375 922 479 c
+922 583 888 665 821 726 c
+754 787 662 817 547 817 c
+493 817 439 811 385 799 c
+332 787 277 768 221 743 c
+221 1493 l
+
+ce} _d
+/seven{1303 0 168 0 1128 1493 sc
+168 1493 m
+1128 1493 l
+1128 1407 l
+586 0 l
+375 0 l
+885 1323 l
+168 1323 l
+168 1493 l
+
+ce} _d
+/a{1255 0 123 -29 1069 1147 sc
+702 563 m
+553 563 450 546 393 512 c
+336 478 307 420 307 338 c
+307 273 328 221 371 182 c
+414 144 473 125 547 125 c
+649 125 731 161 792 233 c
+854 306 885 402 885 522 c
+885 563 l
+702 563 l
+
+1069 639 m
+1069 0 l
+885 0 l
+885 170 l
+843 102 791 52 728 19 c
+665 -13 589 -29 498 -29 c
+383 -29 292 3 224 67 c
+157 132 123 218 123 326 c
+123 452 165 547 249 611 c
+334 675 460 707 627 707 c
+885 707 l
+885 725 l
+885 810 857 875 801 921 c
+746 968 668 991 567 991 c
+503 991 441 983 380 968 c
+319 953 261 930 205 899 c
+205 1069 l
+272 1095 338 1114 401 1127 c
+464 1140 526 1147 586 1147 c
+748 1147 869 1105 949 1021 c
+1029 937 1069 810 1069 639 c
+
+ce} _d
+/c{1126 0 113 -29 999 1147 sc
+999 1077 m
+999 905 l
+947 934 895 955 842 969 c
+790 984 737 991 684 991 c
+565 991 472 953 406 877 c
+340 802 307 696 307 559 c
+307 422 340 316 406 240 c
+472 165 565 127 684 127 c
+737 127 790 134 842 148 c
+895 163 947 184 999 213 c
+999 43 l
+948 19 894 1 839 -11 c
+784 -23 726 -29 664 -29 c
+495 -29 361 24 262 130 c
+163 236 113 379 113 559 c
+113 742 163 885 263 990 c
+364 1095 501 1147 676 1147 c
+733 1147 788 1141 842 1129 c
+896 1118 948 1100 999 1077 c
+
+ce} _d
+/d{1300 0 113 -29 1114 1556 sc
+930 950 m
+930 1556 l
+1114 1556 l
+1114 0 l
+930 0 l
+930 168 l
+891 101 842 52 783 19 c
+724 -13 654 -29 571 -29 c
+436 -29 325 25 240 133 c
+155 241 113 383 113 559 c
+113 735 155 877 240 985 c
+325 1093 436 1147 571 1147 c
+654 1147 724 1131 783 1098 c
+842 1066 891 1017 930 950 c
+
+303 559 m
+303 424 331 317 386 240 c
+442 163 519 125 616 125 c
+713 125 790 163 846 240 c
+902 317 930 424 930 559 c
+930 694 902 800 846 877 c
+790 954 713 993 616 993 c
+519 993 442 954 386 877 c
+331 800 303 694 303 559 c
+
+ce} _d
+/e{1260 0 113 -29 1151 1147 sc
+1151 606 m
+1151 516 l
+305 516 l
+313 389 351 293 419 226 c
+488 160 583 127 705 127 c
+776 127 844 136 910 153 c
+977 170 1043 196 1108 231 c
+1108 57 l
+1042 29 974 8 905 -7 c
+836 -22 765 -29 694 -29 c
+515 -29 374 23 269 127 c
+165 231 113 372 113 549 c
+113 732 162 878 261 985 c
+360 1093 494 1147 662 1147 c
+813 1147 932 1098 1019 1001 c
+1107 904 1151 773 1151 606 c
+
+967 660 m
+966 761 937 841 882 901 c
+827 961 755 991 664 991 c
+561 991 479 962 417 904 c
+356 846 320 764 311 659 c
+967 660 l
+
+ce} _d
+/g{1300 0 113 -426 1114 1147 sc
+930 573 m
+930 706 902 810 847 883 c
+792 956 715 993 616 993 c
+517 993 440 956 385 883 c
+330 810 303 706 303 573 c
+303 440 330 337 385 264 c
+440 191 517 154 616 154 c
+715 154 792 191 847 264 c
+902 337 930 440 930 573 c
+
+1114 139 m
+1114 -52 1072 -193 987 -286 c
+902 -379 773 -426 598 -426 c
+533 -426 472 -421 415 -411 c
+358 -402 302 -387 248 -367 c
+248 -188 l
+302 -217 355 -239 408 -253 c
+461 -267 514 -274 569 -274 c
+690 -274 780 -242 840 -179 c
+900 -116 930 -21 930 106 c
+930 197 l
+892 131 843 82 784 49 c
+725 16 654 0 571 0 c
+434 0 323 52 239 157 c
+155 262 113 400 113 573 c
+113 746 155 885 239 990 c
+323 1095 434 1147 571 1147 c
+654 1147 725 1131 784 1098 c
+843 1065 892 1016 930 950 c
+930 1120 l
+1114 1120 l
+1114 139 l
+
+ce} _d
+/h{1298 0 186 0 1124 1556 sc
+1124 676 m
+1124 0 l
+940 0 l
+940 670 l
+940 776 919 855 878 908 c
+837 961 775 987 692 987 c
+593 987 514 955 457 892 c
+400 829 371 742 371 633 c
+371 0 l
+186 0 l
+186 1556 l
+371 1556 l
+371 946 l
+415 1013 467 1064 526 1097 c
+586 1130 655 1147 733 1147 c
+862 1147 959 1107 1025 1027 c
+1091 948 1124 831 1124 676 c
+
+ce} _d
+/i{569 0 193 0 377 1556 sc
+193 1120 m
+377 1120 l
+377 0 l
+193 0 l
+193 1120 l
+
+193 1556 m
+377 1556 l
+377 1323 l
+193 1323 l
+193 1556 l
+
+ce} _d
+/j{569 0 -37 -426 377 1556 sc
+193 1120 m
+377 1120 l
+377 -20 l
+377 -163 350 -266 295 -330 c
+241 -394 154 -426 33 -426 c
+-37 -426 l
+-37 -270 l
+12 -270 l
+82 -270 130 -254 155 -221 c
+180 -189 193 -122 193 -20 c
+193 1120 l
+
+193 1556 m
+377 1556 l
+377 1323 l
+193 1323 l
+193 1556 l
+
+ce} _d
+/m{1995 0 186 0 1821 1147 sc
+1065 905 m
+1111 988 1166 1049 1230 1088 c
+1294 1127 1369 1147 1456 1147 c
+1573 1147 1663 1106 1726 1024 c
+1789 943 1821 827 1821 676 c
+1821 0 l
+1636 0 l
+1636 670 l
+1636 777 1617 857 1579 909 c
+1541 961 1483 987 1405 987 c
+1310 987 1234 955 1179 892 c
+1124 829 1096 742 1096 633 c
+1096 0 l
+911 0 l
+911 670 l
+911 778 892 858 854 909 c
+816 961 757 987 678 987 c
+584 987 509 955 454 891 c
+399 828 371 742 371 633 c
+371 0 l
+186 0 l
+186 1120 l
+371 1120 l
+371 946 l
+413 1015 463 1065 522 1098 c
+581 1131 650 1147 731 1147 c
+812 1147 881 1126 938 1085 c
+995 1044 1038 984 1065 905 c
+
+ce} _d
+/n{1298 0 186 0 1124 1147 sc
+1124 676 m
+1124 0 l
+940 0 l
+940 670 l
+940 776 919 855 878 908 c
+837 961 775 987 692 987 c
+593 987 514 955 457 892 c
+400 829 371 742 371 633 c
+371 0 l
+186 0 l
+186 1120 l
+371 1120 l
+371 946 l
+415 1013 467 1064 526 1097 c
+586 1130 655 1147 733 1147 c
+862 1147 959 1107 1025 1027 c
+1091 948 1124 831 1124 676 c
+
+ce} _d
+/o{1253 0 113 -29 1141 1147 sc
+627 991 m
+528 991 450 952 393 875 c
+336 798 307 693 307 559 c
+307 425 335 319 392 242 c
+449 165 528 127 627 127 c
+725 127 803 166 860 243 c
+917 320 946 426 946 559 c
+946 692 917 797 860 874 c
+803 952 725 991 627 991 c
+
+627 1147 m
+787 1147 913 1095 1004 991 c
+1095 887 1141 743 1141 559 c
+1141 376 1095 232 1004 127 c
+913 23 787 -29 627 -29 c
+466 -29 340 23 249 127 c
+158 232 113 376 113 559 c
+113 743 158 887 249 991 c
+340 1095 466 1147 627 1147 c
+
+ce} _d
+/p{1300 0 186 -426 1188 1147 sc
+371 168 m
+371 -426 l
+186 -426 l
+186 1120 l
+371 1120 l
+371 950 l
+410 1017 458 1066 517 1098 c
+576 1131 647 1147 729 1147 c
+865 1147 975 1093 1060 985 c
+1145 877 1188 735 1188 559 c
+1188 383 1145 241 1060 133 c
+975 25 865 -29 729 -29 c
+647 -29 576 -13 517 19 c
+458 52 410 101 371 168 c
+
+997 559 m
+997 694 969 800 913 877 c
+858 954 781 993 684 993 c
+587 993 510 954 454 877 c
+399 800 371 694 371 559 c
+371 424 399 317 454 240 c
+510 163 587 125 684 125 c
+781 125 858 163 913 240 c
+969 317 997 424 997 559 c
+
+ce} _d
+/r{842 0 186 0 842 1147 sc
+842 948 m
+821 960 799 969 774 974 c
+750 980 723 983 694 983 c
+590 983 510 949 454 881 c
+399 814 371 717 371 590 c
+371 0 l
+186 0 l
+186 1120 l
+371 1120 l
+371 946 l
+410 1014 460 1064 522 1097 c
+584 1130 659 1147 748 1147 c
+761 1147 775 1146 790 1144 c
+805 1143 822 1140 841 1137 c
+842 948 l
+
+ce} _d
+/z{1075 0 88 0 987 1120 sc
+113 1120 m
+987 1120 l
+987 952 l
+295 147 l
+987 147 l
+987 0 l
+88 0 l
+88 168 l
+780 973 l
+113 973 l
+113 1120 l
+
+ce} _d
+end readonly def
+
+/BuildGlyph {
+ exch begin
+ CharStrings exch
+ 2 copy known not {pop /.notdef} if
+ true 3 1 roll get exec
+ end
+} _d
+
+/BuildChar {
+ 1 index /Encoding get exch get
+ 1 index /BuildGlyph get exec
+} _d
+
+FontName currentdict end definefont pop
+end
+%%EndProlog
+mpldict begin
+75.6 223.2 translate
+460.8 345.6 0 0 clipbox
+gsave
+0 0 m
+460.8 0 l
+460.8 345.6 l
+0 345.6 l
+cl
+1.000 setgray
+fill
+grestore
+gsave
+57.6 38.016 m
+414.72 38.016 l
+414.72 304.128 l
+57.6 304.128 l
+cl
+1.000 setgray
+fill
+grestore
+0.800 setlinewidth
+1 setlinejoin
+0 setlinecap
+[] 0 setdash
+0.000 setgray
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+73.8327 38.016 o
+grestore
+/DejaVuSans 10.000 selectfont
+gsave
+
+58.512415 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /two glyphshow
+14.741211 0 m /zero glyphshow
+21.103516 0 m /period glyphshow
+24.282227 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+116.55 38.016 o
+grestore
+gsave
+
+101.230118 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /one glyphshow
+14.741211 0 m /seven glyphshow
+21.103516 0 m /period glyphshow
+24.282227 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+159.268 38.016 o
+grestore
+gsave
+
+143.947821 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /one glyphshow
+14.741211 0 m /five glyphshow
+21.103516 0 m /period glyphshow
+24.282227 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+201.986 38.016 o
+grestore
+gsave
+
+186.665525 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /one glyphshow
+14.741211 0 m /two glyphshow
+21.103516 0 m /period glyphshow
+24.282227 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+244.704 38.016 o
+grestore
+gsave
+
+229.383228 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /one glyphshow
+14.741211 0 m /zero glyphshow
+21.103516 0 m /period glyphshow
+24.282227 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+287.421 38.016 o
+grestore
+gsave
+
+275.280619 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /seven glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+330.139 38.016 o
+grestore
+gsave
+
+317.998322 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /five glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+0 0 m
+0 -3.5 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+372.857 38.016 o
+grestore
+gsave
+
+360.716026 23.422250 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /two glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /five glyphshow
+grestore
+gsave
+
+178.370937 9.750375 translate
+0.000000 rotate
+0.000000 0 m /z glyphshow
+5.249023 0 m /space glyphshow
+8.427734 0 m /p glyphshow
+14.775391 0 m /r glyphshow
+18.886719 0 m /z glyphshow
+24.135742 0 m /e glyphshow
+30.288086 0 m /d glyphshow
+36.635742 0 m /space glyphshow
+39.814453 0 m /h glyphshow
+46.152344 0 m /o glyphshow
+52.270508 0 m /m glyphshow
+62.011719 0 m /o glyphshow
+68.129883 0 m /g glyphshow
+74.477539 0 m /e glyphshow
+80.629883 0 m /n glyphshow
+86.967773 0 m /i glyphshow
+89.746094 0 m /z glyphshow
+94.995117 0 m /a glyphshow
+101.123047 0 m /c glyphshow
+106.621094 0 m /j glyphshow
+109.399414 0 m /aogonek glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 50.112 o
+grestore
+gsave
+
+19.959375 46.315125 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /one glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /zero glyphshow
+24.282227 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 80.352 o
+grestore
+gsave
+
+19.959375 76.555125 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /zero glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /seven glyphshow
+24.282227 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 110.592 o
+grestore
+gsave
+
+19.959375 106.795125 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /zero glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /five glyphshow
+24.282227 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 140.832 o
+grestore
+gsave
+
+19.959375 137.035125 translate
+0.000000 rotate
+0.000000 0 m /minus glyphshow
+8.378906 0 m /zero glyphshow
+14.741211 0 m /period glyphshow
+17.919922 0 m /two glyphshow
+24.282227 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 171.072 o
+grestore
+gsave
+
+28.334375 167.275125 translate
+0.000000 rotate
+0.000000 0 m /zero glyphshow
+6.362305 0 m /period glyphshow
+9.541016 0 m /zero glyphshow
+15.903320 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 201.312 o
+grestore
+gsave
+
+28.334375 197.515125 translate
+0.000000 rotate
+0.000000 0 m /zero glyphshow
+6.362305 0 m /period glyphshow
+9.541016 0 m /two glyphshow
+15.903320 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 231.552 o
+grestore
+gsave
+
+28.334375 227.755125 translate
+0.000000 rotate
+0.000000 0 m /zero glyphshow
+6.362305 0 m /period glyphshow
+9.541016 0 m /five glyphshow
+15.903320 0 m /zero glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 261.792 o
+grestore
+gsave
+
+28.334375 257.995125 translate
+0.000000 rotate
+0.000000 0 m /zero glyphshow
+6.362305 0 m /period glyphshow
+9.541016 0 m /seven glyphshow
+15.903320 0 m /five glyphshow
+grestore
+gsave
+/o {
+gsave
+newpath
+translate
+0.8 setlinewidth
+1 setlinejoin
+
+0 setlinecap
+
+-0 0 m
+-3.5 0 l
+
+gsave
+0.000 setgray
+fill
+grestore
+stroke
+grestore
+} bind def
+57.6 292.032 o
+grestore
+gsave
+
+28.334375 288.235125 translate
+0.000000 rotate
+0.000000 0 m /one glyphshow
+6.362305 0 m /period glyphshow
+9.541016 0 m /zero glyphshow
+15.903320 0 m /zero glyphshow
+grestore
+gsave
+
+13.881250 122.822000 translate
+90.000000 rotate
+0.000000 0 m /z glyphshow
+5.249023 0 m /space glyphshow
+8.427734 0 m /p glyphshow
+14.775391 0 m /o glyphshow
+20.893555 0 m /space glyphshow
+24.072266 0 m /h glyphshow
+30.410156 0 m /o glyphshow
+36.528320 0 m /m glyphshow
+46.269531 0 m /o glyphshow
+52.387695 0 m /g glyphshow
+58.735352 0 m /e glyphshow
+64.887695 0 m /n glyphshow
+71.225586 0 m /i glyphshow
+74.003906 0 m /z glyphshow
+79.252930 0 m /a glyphshow
+85.380859 0 m /c glyphshow
+90.878906 0 m /j glyphshow
+93.657227 0 m /i glyphshow
+grestore
+1.500 setlinewidth
+2 setlinecap
+0.122 0.467 0.706 setrgbcolor
+gsave
+357.12 266.112 57.6 38.016 clipbox
+73.832727 292.032 m
+104.411412 290.780736 l
+131.086435 289.469535 l
+154.508406 288.09731 l
+174.677326 286.701851 l
+192.894414 285.224164 l
+209.159672 283.684395 l
+223.473099 282.113711 l
+236.485305 280.467946 l
+248.196291 278.767974 l
+258.606056 277.043937 l
+268.36521 275.206208 l
+277.473755 273.256657 l
+285.281079 271.368662 l
+292.437792 269.427684 l
+298.943895 267.456451 l
+305.449998 265.252299 l
+311.305491 263.033379 l
+316.510373 260.840794 l
+321.715256 258.405032 l
+326.269528 256.040834 l
+330.8238 253.422545 l
+334.727462 250.943512 l
+338.631124 248.212935 l
+342.534786 245.190481 l
+345.787837 242.413472 l
+349.040889 239.364908 l
+352.293941 236.002909 l
+355.546992 232.276518 l
+358.149433 228.991428 l
+360.751875 225.394448 l
+363.354316 221.438951 l
+365.956757 217.068522 l
+368.559198 212.214258 l
+371.16164 206.791106 l
+373.764081 200.692835 l
+375.715912 195.596536 l
+377.667743 189.975415 l
+379.619574 183.744 l
+381.571405 176.797196 l
+383.523236 169.004308 l
+385.475066 160.200737 l
+387.426897 150.176234 l
+389.378728 138.657882 l
+391.330559 125.284881 l
+392.63178 115.105433 l
+393.933001 103.701873 l
+395.234221 90.839273 l
+396.535442 76.218475 l
+397.836662 59.452541 l
+398.487273 50.112 l
+398.487273 50.112 l
+stroke
+grestore
+0.800 setlinewidth
+0 setlinejoin
+0.000 setgray
+gsave
+57.6 38.016 m
+57.6 304.128 l
+stroke
+grestore
+gsave
+414.72 38.016 m
+414.72 304.128 l
+stroke
+grestore
+gsave
+57.6 38.016 m
+414.72 38.016 l
+stroke
+grestore
+gsave
+57.6 304.128 m
+414.72 304.128 l
+stroke
+grestore
+
+end
+showpage
diff --git a/cw 9/img/PerlinNoiseDotProducts.png b/cw 9/img/PerlinNoiseDotProducts.png
new file mode 100644
index 0000000..8d68c33
Binary files /dev/null and b/cw 9/img/PerlinNoiseDotProducts.png differ
diff --git a/cw 9/img/PerlinNoiseGradientGrid.png b/cw 9/img/PerlinNoiseGradientGrid.png
new file mode 100644
index 0000000..3841843
Binary files /dev/null and b/cw 9/img/PerlinNoiseGradientGrid.png differ
diff --git a/cw 9/img/PerlinNoiseInterpolated.png b/cw 9/img/PerlinNoiseInterpolated.png
new file mode 100644
index 0000000..c535de9
Binary files /dev/null and b/cw 9/img/PerlinNoiseInterpolated.png differ
diff --git a/cw 9/img/blending1.webp b/cw 9/img/blending1.webp
new file mode 100644
index 0000000..4d9e4ec
Binary files /dev/null and b/cw 9/img/blending1.webp differ
diff --git a/cw 9/img/camera.jpg b/cw 9/img/camera.jpg
new file mode 100644
index 0000000..edb497c
Binary files /dev/null and b/cw 9/img/camera.jpg differ
diff --git a/cw 9/img/capture.webm b/cw 9/img/capture.webm
new file mode 100644
index 0000000..34357e5
Binary files /dev/null and b/cw 9/img/capture.webm differ
diff --git a/cw 9/img/coordinate_systems.jpg b/cw 9/img/coordinate_systems.jpg
new file mode 100644
index 0000000..be4c999
Binary files /dev/null and b/cw 9/img/coordinate_systems.jpg differ
diff --git a/cw 9/img/cubemaps_sampling.png b/cw 9/img/cubemaps_sampling.png
new file mode 100644
index 0000000..2e897a0
Binary files /dev/null and b/cw 9/img/cubemaps_sampling.png differ
diff --git a/cw 9/img/gamma_correction_brightness.png b/cw 9/img/gamma_correction_brightness.png
new file mode 100644
index 0000000..52e33b4
Binary files /dev/null and b/cw 9/img/gamma_correction_brightness.png differ
diff --git a/cw 9/img/lighting_result.png b/cw 9/img/lighting_result.png
new file mode 100644
index 0000000..3bbca23
Binary files /dev/null and b/cw 9/img/lighting_result.png differ
diff --git a/cw 9/img/noise_examples.jpg b/cw 9/img/noise_examples.jpg
new file mode 100644
index 0000000..e7c4878
Binary files /dev/null and b/cw 9/img/noise_examples.jpg differ
diff --git a/cw 9/img/shadow_mapping_projection.png b/cw 9/img/shadow_mapping_projection.png
new file mode 100644
index 0000000..b1843e7
Binary files /dev/null and b/cw 9/img/shadow_mapping_projection.png differ
diff --git a/cw 9/img/shadows1.bmp b/cw 9/img/shadows1.bmp
new file mode 100644
index 0000000..2ca100f
Binary files /dev/null and b/cw 9/img/shadows1.bmp differ
diff --git a/cw 9/img/spotlight.png b/cw 9/img/spotlight.png
new file mode 100644
index 0000000..d5a54a8
Binary files /dev/null and b/cw 9/img/spotlight.png differ
diff --git a/cw 9/img/stride_offest.jpg b/cw 9/img/stride_offest.jpg
new file mode 100644
index 0000000..216771f
Binary files /dev/null and b/cw 9/img/stride_offest.jpg differ
diff --git a/cw 9/img/stride_offest_2_2.jpg b/cw 9/img/stride_offest_2_2.jpg
new file mode 100644
index 0000000..d52cd14
Binary files /dev/null and b/cw 9/img/stride_offest_2_2.jpg differ
diff --git a/cw 9/img/struktura.jpg b/cw 9/img/struktura.jpg
new file mode 100644
index 0000000..4f93e04
Binary files /dev/null and b/cw 9/img/struktura.jpg differ
diff --git a/cw 9/img/sun.png b/cw 9/img/sun.png
new file mode 100644
index 0000000..ce2a691
Binary files /dev/null and b/cw 9/img/sun.png differ
diff --git a/cw 9/img/textures.png b/cw 9/img/textures.png
new file mode 100644
index 0000000..d8b23b7
Binary files /dev/null and b/cw 9/img/textures.png differ
diff --git a/cw 9/img/z_depth_graph.jpg b/cw 9/img/z_depth_graph.jpg
new file mode 100644
index 0000000..a5e6181
Binary files /dev/null and b/cw 9/img/z_depth_graph.jpg differ
diff --git a/cw 9/img/z_depth_graph2.jpg b/cw 9/img/z_depth_graph2.jpg
new file mode 100644
index 0000000..449d66b
Binary files /dev/null and b/cw 9/img/z_depth_graph2.jpg differ
diff --git a/cw 9/render.py b/cw 9/render.py
new file mode 100644
index 0000000..c53ac7e
--- /dev/null
+++ b/cw 9/render.py
@@ -0,0 +1,7 @@
+
+import os
+rootdir = './'
+for filename in os.listdir(rootdir):
+ if filename.endswith(".md"):
+ name = filename[:-3]
+ os.system(f'pandoc -s -o "{name}.html" "{name}.md" --mathjax --css style.css')
\ No newline at end of file
diff --git a/cw 9/shaders/shader_5_1_tex.frag b/cw 9/shaders/shader_5_1_tex.frag
new file mode 100644
index 0000000..951c43e
--- /dev/null
+++ b/cw 9/shaders/shader_5_1_tex.frag
@@ -0,0 +1,18 @@
+#version 430 core
+
+float AMBIENT = 0.1;
+
+uniform vec3 color;
+uniform vec3 lightPos;
+
+in vec3 vecNormal;
+in vec3 worldPos;
+
+out vec4 outColor;
+void main()
+{
+ vec3 lightDir = normalize(lightPos-worldPos);
+ vec3 normal = normalize(vecNormal);
+ float diffuse=max(0,dot(normal,lightDir));
+ outColor = vec4(color*min(1,AMBIENT+diffuse), 1.0);
+}
diff --git a/cw 9/shaders/shader_5_1_tex.vert b/cw 9/shaders/shader_5_1_tex.vert
new file mode 100644
index 0000000..1b569f5
--- /dev/null
+++ b/cw 9/shaders/shader_5_1_tex.vert
@@ -0,0 +1,18 @@
+#version 430 core
+
+layout(location = 0) in vec3 vertexPosition;
+layout(location = 1) in vec3 vertexNormal;
+layout(location = 2) in vec2 vertexTexCoord;
+
+uniform mat4 transformation;
+uniform mat4 modelMatrix;
+
+out vec3 vecNormal;
+out vec3 worldPos;
+
+void main()
+{
+ worldPos = (modelMatrix* vec4(vertexPosition,1)).xyz;
+ vecNormal = (modelMatrix* vec4(vertexNormal,0)).xyz;
+ gl_Position = transformation * vec4(vertexPosition, 1.0);
+}
diff --git a/cw 9/shaders/shader_8_sun.frag b/cw 9/shaders/shader_8_sun.frag
new file mode 100644
index 0000000..1dd40b8
--- /dev/null
+++ b/cw 9/shaders/shader_8_sun.frag
@@ -0,0 +1,11 @@
+#version 430 core
+
+uniform vec3 color;
+uniform float exposition;
+
+
+out vec4 outColor;
+void main()
+{
+ outColor = vec4(vec3(1.0) - exp(-color*exposition),1);
+}
diff --git a/cw 9/shaders/shader_8_sun.vert b/cw 9/shaders/shader_8_sun.vert
new file mode 100644
index 0000000..57204e2
--- /dev/null
+++ b/cw 9/shaders/shader_8_sun.vert
@@ -0,0 +1,13 @@
+#version 430 core
+
+layout(location = 0) in vec3 vertexPosition;
+layout(location = 1) in vec3 vertexNormal;
+layout(location = 2) in vec2 vertexTexCoord;
+
+uniform mat4 transformation;
+
+void main()
+{
+ gl_Position = transformation * vec4(vertexPosition, 1.0);
+ //gl_Position = vec4(vertexPosition, 1.0);
+}
diff --git a/cw 9/shaders/shader_9_1.frag b/cw 9/shaders/shader_9_1.frag
new file mode 100644
index 0000000..550f6c1
--- /dev/null
+++ b/cw 9/shaders/shader_9_1.frag
@@ -0,0 +1,135 @@
+#version 430 core
+
+float AMBIENT = 0.03;
+float PI = 3.14;
+
+uniform sampler2D depthMap;
+
+uniform vec3 cameraPos;
+
+uniform vec3 color;
+
+uniform vec3 sunDir;
+uniform vec3 sunColor;
+
+uniform vec3 lightPos;
+uniform vec3 lightColor;
+
+uniform vec3 spotlightPos;
+uniform vec3 spotlightColor;
+uniform vec3 spotlightConeDir;
+uniform vec3 spotlightPhi;
+
+uniform float metallic;
+uniform float roughness;
+
+uniform float exposition;
+
+in vec3 vecNormal;
+in vec3 worldPos;
+
+out vec4 outColor;
+
+
+in vec3 viewDirTS;
+in vec3 lightDirTS;
+in vec3 spotlightDirTS;
+in vec3 sunDirTS;
+
+in vec3 test;
+
+float DistributionGGX(vec3 normal, vec3 H, float roughness){
+ float a = roughness*roughness;
+ float a2 = a*a;
+ float NdotH = max(dot(normal, H), 0.0);
+ float NdotH2 = NdotH*NdotH;
+
+ float num = a2;
+ float denom = (NdotH2 * (a2 - 1.0) + 1.0);
+ denom = PI * denom * denom;
+
+ return num / denom;
+}
+float GeometrySchlickGGX(float NdotV, float roughness){
+ float r = (roughness + 1.0);
+ float k = (r*r) / 8.0;
+
+ float num = NdotV;
+ float denom = NdotV * (1.0 - k) + k;
+
+ return num / denom;
+}
+float GeometrySmith(vec3 normal, vec3 V, vec3 lightDir, float roughness){
+ float NdotV = max(dot(normal, V), 0.0);
+ float NdotL = max(dot(normal, lightDir), 0.0);
+ float ggx2 = GeometrySchlickGGX(NdotV, roughness);
+ float ggx1 = GeometrySchlickGGX(NdotL, roughness);
+
+ return ggx1 * ggx2;
+}
+vec3 fresnelSchlick(float cosTheta, vec3 F0){
+ return F0 + (1.0 - F0) * pow(clamp(1.0 - cosTheta, 0.0, 1.0), 5.0);
+}
+
+vec3 PBRLight(vec3 lightDir, vec3 radiance, vec3 normal, vec3 V){
+ float diffuse=max(0,dot(normal,lightDir));
+
+ //vec3 V = normalize(cameraPos-worldPos);
+ vec3 F0 = vec3(0.04);
+ F0 = mix(F0, color, metallic);
+
+ vec3 H = normalize(V + lightDir);
+
+ // cook-torrance brdf
+ float NDF = DistributionGGX(normal, H, roughness);
+ float G = GeometrySmith(normal, V, lightDir, roughness);
+ vec3 F = fresnelSchlick(max(dot(H, V), 0.0), F0);
+
+ vec3 kS = F;
+ vec3 kD = vec3(1.0) - kS;
+ kD *= 1.0 - metallic;
+
+ vec3 numerator = NDF * G * F;
+ float denominator = 4.0 * max(dot(normal, V), 0.0) * max(dot(normal, lightDir), 0.0) + 0.0001;
+ vec3 specular = numerator / denominator;
+
+ // add to outgoing radiance Lo
+ float NdotL = max(dot(normal, lightDir), 0.0);
+ return (kD * color / PI + specular) * radiance * NdotL;
+}
+
+
+void main()
+{
+ //vec3 normal = vec3(0,0,1);
+ vec3 normal = normalize(vecNormal);
+
+ //vec3 viewDir = normalize(viewDirTS);
+ vec3 viewDir = normalize(cameraPos-worldPos);
+
+ //vec3 lightDir = normalize(lightDirTS);
+ vec3 lightDir = normalize(lightPos-worldPos);
+
+
+ vec3 ambient = AMBIENT*color;
+ vec3 attenuatedlightColor = lightColor/pow(length(lightPos-worldPos),2);
+ vec3 ilumination;
+ ilumination = ambient+PBRLight(lightDir,attenuatedlightColor,normal,viewDir);
+
+ //flashlight
+ //vec3 spotlightDir= normalize(spotlightDirTS);
+ vec3 spotlightDir= normalize(spotlightPos-worldPos);
+
+
+ float angle_atenuation = clamp((dot(-normalize(spotlightPos-worldPos),spotlightConeDir)-0.5)*3,0,1);
+ attenuatedlightColor = angle_atenuation*spotlightColor/pow(length(spotlightPos-worldPos),2);
+ ilumination=ilumination+PBRLight(spotlightDir,attenuatedlightColor,normal,viewDir);
+
+ //sun
+ ilumination=ilumination+PBRLight(sunDir,sunColor,normal,viewDir);
+
+
+ outColor = vec4(vec3(1.0) - exp(-ilumination*exposition),1);
+ //outColor = vec4(roughness,metallic,0,1);
+ //outColor = vec4(test;
+}
diff --git a/cw 9/shaders/shader_9_1.vert b/cw 9/shaders/shader_9_1.vert
new file mode 100644
index 0000000..750196d
--- /dev/null
+++ b/cw 9/shaders/shader_9_1.vert
@@ -0,0 +1,43 @@
+#version 430 core
+
+layout(location = 0) in vec3 vertexPosition;
+layout(location = 1) in vec3 vertexNormal;
+layout(location = 2) in vec2 vertexTexCoord;
+layout(location = 3) in vec3 vertexTangent;
+layout(location = 4) in vec3 vertexBitangent;
+
+uniform mat4 transformation;
+uniform mat4 modelMatrix;
+
+out vec3 vecNormal;
+out vec3 worldPos;
+
+uniform vec3 lightPos;
+uniform vec3 spotlightPos;
+uniform vec3 cameraPos;
+uniform vec3 sunDir;
+
+out vec3 viewDirTS;
+out vec3 lightDirTS;
+out vec3 spotlightDirTS;
+out vec3 sunDirTS;
+
+void main()
+{
+ worldPos = (modelMatrix* vec4(vertexPosition,1)).xyz;
+ vecNormal = (modelMatrix* vec4(vertexNormal,0)).xyz;
+ gl_Position = transformation * vec4(vertexPosition, 1.0);
+
+ vec3 w_tangent = normalize(mat3(modelMatrix)*vertexTangent);
+ vec3 w_bitangent = normalize(mat3(modelMatrix)*vertexBitangent);
+ mat3 TBN = transpose(mat3(w_tangent, w_bitangent, vecNormal));
+
+ vec3 V = normalize(cameraPos-worldPos);
+ viewDirTS = TBN*V;
+ vec3 L = normalize(lightPos-worldPos);
+ lightDirTS = TBN*L;
+ vec3 SL = normalize(spotlightPos-worldPos);
+ spotlightDirTS = TBN*SL;
+ sunDirTS = TBN*sunDir;
+
+}
diff --git a/cw 9/shaders/test.frag b/cw 9/shaders/test.frag
new file mode 100644
index 0000000..404c2c6
--- /dev/null
+++ b/cw 9/shaders/test.frag
@@ -0,0 +1,12 @@
+#version 330 core
+out vec4 FragColor;
+
+in vec2 tc;
+
+uniform sampler2D depthMap;
+
+void main()
+{
+ float depthValue = texture(depthMap, tc).r;
+ FragColor = vec4(vec3(depthValue+0.5), 1.0);
+}
\ No newline at end of file
diff --git a/cw 9/shaders/test.vert b/cw 9/shaders/test.vert
new file mode 100644
index 0000000..f65f59d
--- /dev/null
+++ b/cw 9/shaders/test.vert
@@ -0,0 +1,14 @@
+#version 430 core
+
+layout(location = 0) in vec3 vertexPosition;
+layout(location = 1) in vec3 vertexNormal;
+layout(location = 2) in vec2 vertexTexCoord;
+
+
+out vec2 tc;
+
+void main()
+{
+ tc = vertexTexCoord;
+ gl_Position = vec4(vertexPosition*0.9, 1.0);
+}
diff --git a/cw 9/src/Box.cpp b/cw 9/src/Box.cpp
new file mode 100644
index 0000000..28445d3
--- /dev/null
+++ b/cw 9/src/Box.cpp
@@ -0,0 +1,55 @@
+
+
+// dane 36 wierzcholkow i kolorow opisujace model pudelka
+const float box[] = {
+
+ // points colors
+ 0.25f, 0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+ 0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+ -0.25f, 0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+
+ 0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+ -0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+ -0.25f, 0.25f, 0.75f, 1.0f, 0.0f, 0.0f, 1.0f, 1.0f,
+
+ 0.25f, 0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+ -0.25f, 0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+ -0.25f, 0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+ -0.25f, -0.25f, -0.75f, 1.0f, 0.8f, 0.8f, 0.8f, 1.0f,
+
+ -0.25f, 0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+ -0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+ -0.25f, -0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+
+ -0.25f, 0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+ -0.25f, -0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+ -0.25f, 0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 0.0f, 1.0f,
+
+ 0.25f, 0.25f, 0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+ 0.25f, -0.25f, 0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+
+ 0.25f, 0.25f, 0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+ 0.25f, 0.25f, -0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.5f, 0.5f, 0.0f, 1.0f,
+
+ 0.25f, 0.25f, -0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+ 0.25f, 0.25f, 0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+ -0.25f, 0.25f, 0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+
+ 0.25f, 0.25f, -0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+ -0.25f, 0.25f, 0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+ -0.25f, 0.25f, -0.75f, 1.0f, 1.0f, 0.0f, 0.0f, 1.0f,
+
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+ -0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+ 0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+
+ 0.25f, -0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+ -0.25f, -0.25f, -0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+ -0.25f, -0.25f, 0.75f, 1.0f, 0.0f, 1.0f, 1.0f, 1.0f,
+};
+
diff --git a/cw 9/src/Camera.cpp b/cw 9/src/Camera.cpp
new file mode 100644
index 0000000..90cc40c
--- /dev/null
+++ b/cw 9/src/Camera.cpp
@@ -0,0 +1,31 @@
+#include "Camera.h"
+
+glm::mat4 Core::createPerspectiveMatrix(float zNear, float zFar, float frustumScale)
+{
+ glm::mat4 perspective;
+ perspective[0][0] = 1.f;
+ perspective[1][1] = frustumScale;
+ perspective[2][2] = (zFar + zNear) / (zNear - zFar);
+ perspective[3][2] = (2 * zFar * zNear) / (zNear - zFar);
+ perspective[2][3] = -1;
+ perspective[3][3] = 0;
+
+ return perspective;
+}
+
+glm::mat4 Core::createViewMatrix( glm::vec3 position, glm::vec3 forward, glm::vec3 up )
+{
+ glm::vec3 side = glm::cross(forward, up);
+
+ // Trzeba pamietac o minusie przy ustawianiu osi Z kamery.
+ // Wynika to z tego, ze standardowa macierz perspektywiczna zaklada, ze "z przodu" jest ujemna (a nie dodatnia) czesc osi Z.
+ glm::mat4 cameraRotation;
+ cameraRotation[0][0] = side.x; cameraRotation[1][0] = side.y; cameraRotation[2][0] = side.z;
+ cameraRotation[0][1] = up.x; cameraRotation[1][1] = up.y; cameraRotation[2][1] = up.z;
+ cameraRotation[0][2] = -forward.x; cameraRotation[1][2] = -forward.y; cameraRotation[2][2] = -forward.z;
+
+ glm::mat4 cameraTranslation;
+ cameraTranslation[3] = glm::vec4(-position, 1.0f);
+
+ return cameraRotation * cameraTranslation;
+}
diff --git a/cw 9/src/Camera.h b/cw 9/src/Camera.h
new file mode 100644
index 0000000..54c29b3
--- /dev/null
+++ b/cw 9/src/Camera.h
@@ -0,0 +1,14 @@
+#pragma once
+
+#include "glm.hpp"
+
+namespace Core
+{
+ glm::mat4 createPerspectiveMatrix(float zNear = 0.1f, float zFar = 100.0f, float frustumScale = 1.f);
+
+ // position - pozycja kamery
+ // forward - wektor "do przodu" kamery (jednostkowy)
+ // up - wektor "w gore" kamery (jednostkowy)
+ // up i forward musza byc ortogonalne!
+ glm::mat4 createViewMatrix(glm::vec3 position, glm::vec3 forward, glm::vec3 up);
+}
\ No newline at end of file
diff --git a/cw 9/src/Render_Utils.cpp b/cw 9/src/Render_Utils.cpp
new file mode 100644
index 0000000..18f71f9
--- /dev/null
+++ b/cw 9/src/Render_Utils.cpp
@@ -0,0 +1,128 @@
+#include "Render_Utils.h"
+
+#include
+
+#include "glew.h"
+#include "freeglut.h"
+#include
+#include
+#include
+
+
+
+void Core::RenderContext::initFromAssimpMesh(aiMesh* mesh) {
+ vertexArray = 0;
+ vertexBuffer = 0;
+ vertexIndexBuffer = 0;
+
+ std::vector textureCoord;
+ std::vector indices;
+ //tex coord must be converted to 2d vecs
+ for (unsigned int i = 0; i < mesh->mNumVertices; i++)
+ {
+ if (mesh->mTextureCoords[0] != nullptr) {
+ textureCoord.push_back(mesh->mTextureCoords[0][i].x);
+ textureCoord.push_back(mesh->mTextureCoords[0][i].y);
+ }
+ else {
+ textureCoord.push_back(0.0f);
+ textureCoord.push_back(0.0f);
+ }
+ }
+ if (mesh->mTextureCoords[0] == nullptr) {
+ std::cout << "no uv coords\n";
+ }
+ for (unsigned int i = 0; i < mesh->mNumFaces; i++)
+ {
+ aiFace face = mesh->mFaces[i];
+ // retrieve all indices of the face and store them in the indices vector
+ for (unsigned int j = 0; j < face.mNumIndices; j++)
+ indices.push_back(face.mIndices[j]);
+ }
+
+ unsigned int vertexDataBufferSize = sizeof(float) * mesh->mNumVertices * 3;
+ unsigned int vertexNormalBufferSize = sizeof(float) * mesh->mNumVertices * 3;
+ unsigned int vertexTexBufferSize = sizeof(float) * mesh->mNumVertices * 2;
+ unsigned int vertexTangentBufferSize = sizeof(float) * mesh->mNumVertices * 3;
+ unsigned int vertexBiTangentBufferSize = sizeof(float) * mesh->mNumVertices * 3;
+
+ unsigned int vertexElementBufferSize = sizeof(unsigned int) * indices.size();
+ size = indices.size();
+
+ glGenVertexArrays(1, &vertexArray);
+ glBindVertexArray(vertexArray);
+
+
+ glGenBuffers(1, &vertexIndexBuffer);
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, vertexIndexBuffer);
+ glBufferData(GL_ELEMENT_ARRAY_BUFFER, vertexElementBufferSize, &indices[0], GL_STATIC_DRAW);
+
+ glGenBuffers(1, &vertexBuffer);
+ glBindBuffer(GL_ARRAY_BUFFER, vertexBuffer);
+ //std::cout << vertexBuffer;
+ glEnableVertexAttribArray(0);
+ glEnableVertexAttribArray(1);
+ glEnableVertexAttribArray(2);
+ glEnableVertexAttribArray(3);
+ glEnableVertexAttribArray(4);
+
+ glBufferData(GL_ARRAY_BUFFER, vertexDataBufferSize + vertexNormalBufferSize + vertexTexBufferSize + vertexTangentBufferSize + vertexBiTangentBufferSize, NULL, GL_STATIC_DRAW);
+
+ glBufferSubData(GL_ARRAY_BUFFER, 0, vertexDataBufferSize, mesh->mVertices);
+
+ glBufferSubData(GL_ARRAY_BUFFER, vertexDataBufferSize, vertexNormalBufferSize, mesh->mNormals);
+
+ glBufferSubData(GL_ARRAY_BUFFER, vertexDataBufferSize + vertexNormalBufferSize, vertexTexBufferSize, &textureCoord[0]);
+
+ glBufferSubData(GL_ARRAY_BUFFER, vertexDataBufferSize + vertexNormalBufferSize + vertexTexBufferSize, vertexTangentBufferSize, mesh->mTangents);
+
+ glBufferSubData(GL_ARRAY_BUFFER, vertexDataBufferSize + vertexNormalBufferSize + vertexTexBufferSize + vertexTangentBufferSize, vertexBiTangentBufferSize, mesh->mBitangents);
+
+ glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (void*)(0));
+ glVertexAttribPointer(1, 3, GL_FLOAT, GL_FALSE, 0, (void*)(vertexDataBufferSize));
+ glVertexAttribPointer(2, 2, GL_FLOAT, GL_FALSE, 0, (void*)(vertexNormalBufferSize + vertexDataBufferSize));
+ glVertexAttribPointer(3, 3, GL_FLOAT, GL_FALSE, 0, (void*)(vertexDataBufferSize + vertexNormalBufferSize + vertexTexBufferSize));
+ glVertexAttribPointer(4, 3, GL_FLOAT, GL_FALSE, 0, (void*)(vertexDataBufferSize + vertexNormalBufferSize + vertexTexBufferSize + vertexTangentBufferSize));
+
+}
+
+void Core::DrawVertexArray(const float * vertexArray, int numVertices, int elementSize )
+{
+ glVertexAttribPointer(0, elementSize, GL_FLOAT, false, 0, vertexArray);
+ glEnableVertexAttribArray(0);
+
+ glDrawArrays(GL_TRIANGLES, 0, numVertices);
+}
+
+void Core::DrawVertexArrayIndexed( const float * vertexArray, const int * indexArray, int numIndexes, int elementSize )
+{
+ glVertexAttribPointer(0, elementSize, GL_FLOAT, false, 0, vertexArray);
+ glEnableVertexAttribArray(0);
+
+ glDrawElements(GL_TRIANGLES, numIndexes, GL_UNSIGNED_INT, indexArray);
+}
+
+
+void Core::DrawVertexArray( const VertexData & data )
+{
+ int numAttribs = std::min(VertexData::MAX_ATTRIBS, data.NumActiveAttribs);
+ for(int i = 0; i < numAttribs; i++)
+ {
+ glVertexAttribPointer(i, data.Attribs[i].Size, GL_FLOAT, false, 0, data.Attribs[i].Pointer);
+ glEnableVertexAttribArray(i);
+ }
+ glDrawArrays(GL_TRIANGLES, 0, data.NumVertices);
+}
+
+void Core::DrawContext(Core::RenderContext& context)
+{
+
+ glBindVertexArray(context.vertexArray);
+ glDrawElements(
+ GL_TRIANGLES, // mode
+ context.size, // count
+ GL_UNSIGNED_INT, // type
+ (void*)0 // element array buffer offset
+ );
+ glBindVertexArray(0);
+}
diff --git a/cw 9/src/Render_Utils.h b/cw 9/src/Render_Utils.h
new file mode 100644
index 0000000..052008d
--- /dev/null
+++ b/cw 9/src/Render_Utils.h
@@ -0,0 +1,72 @@
+#pragma once
+#include "glm.hpp"
+#include "glew.h"
+#include "objload.h"
+#include
+#include
+#include
+
+#define BUFFER_OFFSET(i) ((char *)NULL + (i))
+
+namespace Core
+{
+ struct RenderContext
+ {
+ GLuint vertexArray;
+ GLuint vertexBuffer;
+ GLuint vertexIndexBuffer;
+ int size = 0;
+
+ void initFromOBJ(obj::Model& model);
+
+ void initFromAssimpMesh(aiMesh* mesh);
+ };
+
+ // vertexArray - jednowymiarowa tablica zawierajaca wartosci opisujace pozycje kolejnych wierzcholkow w jednym ciagu (x1, y1, z1, w1, x2, y2, z2, w2, ...)
+ // numVertices - liczba wierzcholkow do narysowania
+ // elementSize - liczba wartosci opisujacych pojedynczy wierzcholek (np. 3 gdy wierzcholek opisany jest trojka (x, y, z))
+ void DrawVertexArray(const float * vertexArray, int numVertices, int elementSize);
+
+ // indexArray - jednowymiarowa tablica zawierajaca indeksy wierzcholkow kolejnych trojkatow w jednym ciagu (t1_i1, t1_i2, t1_i3, t2_i1, t2_i2, t2_i3, ...)
+ // numIndexes - liczba indeksow w tablicy indexArray
+ void DrawVertexArrayIndexed(const float * vertexArray, const int * indexArray, int numIndexes, int elementSize);
+
+
+ struct VertexAttribute
+ {
+ const void * Pointer;
+ int Size;
+ };
+
+ struct VertexData
+ {
+ static const int MAX_ATTRIBS = 8;
+ VertexAttribute Attribs[MAX_ATTRIBS];
+ int NumActiveAttribs;
+ int NumVertices;
+ };
+
+ // Ta funkcja sluzy do rysowania trojkatow, ktorych wierzcholki moga byc opisane wiecej niz jednym atrybutem.
+ // Funkcja przyjmuje jako parametr strukture, w ktorej nalezy zawrzec wszystkie potrzebne dane.
+ //
+ // Przykladowe wywolanie funkcji - narysowanie trojkata jak na pierwszych zajeciach:
+ /*
+
+ const float vertices[] = {
+ 0.25f, 0.25f, 0.0f, 1.0f,
+ 0.25f, -0.25f, 0.0f, 1.0f,
+ -0.25f, -0.25f, 0.0f, 1.0f
+ };
+
+ Core::VertexData vertexData;
+ vertexData.NumActiveAttribs = 1; // Liczba uzywanych atrybutow wierzcholka
+ vertexData.Attribs[0].Pointer = vertices; // Wskaznik na dane zerowego atrybutu
+ vertexData.Attribs[0].Size = 4; // Wielkosc zerowego atrybutu (ilosc liczb opisujacych ten atrybut w pojedynczym wierzcholku)
+ vertexData.NumVertices = 3; // Liczba wierzcholkow do narysowania
+ Core::DrawVertexArray(vertexData);
+
+ */
+ void DrawVertexArray(const VertexData & data);
+
+ void DrawContext(RenderContext& context);
+}
\ No newline at end of file
diff --git a/cw 9/src/SOIL/SOIL.c b/cw 9/src/SOIL/SOIL.c
new file mode 100644
index 0000000..1ee4daf
--- /dev/null
+++ b/cw 9/src/SOIL/SOIL.c
@@ -0,0 +1,2024 @@
+/*
+ Jonathan Dummer
+ 2007-07-26-10.36
+
+ Simple OpenGL Image Library
+
+ Public Domain
+ using Sean Barret's stb_image as a base
+
+ Thanks to:
+ * Sean Barret - for the awesome stb_image
+ * Dan Venkitachalam - for finding some non-compliant DDS files, and patching some explicit casts
+ * everybody at gamedev.net
+*/
+
+#define SOIL_CHECK_FOR_GL_ERRORS 0
+
+#ifdef WIN32
+ #define WIN32_LEAN_AND_MEAN
+ #include
+ #include
+ #include
+#elif defined(__APPLE__) || defined(__APPLE_CC__)
+ /* I can't test this Apple stuff! */
+ #include
+ #include
+ #define APIENTRY
+#else
+ #include
+ #include
+#endif
+
+#include "SOIL.h"
+#include "stb_image_aug.h"
+#include "image_helper.h"
+#include "image_DXT.h"
+
+#include
+#include
+
+/* error reporting */
+char *result_string_pointer = "SOIL initialized";
+
+/* for loading cube maps */
+enum{
+ SOIL_CAPABILITY_UNKNOWN = -1,
+ SOIL_CAPABILITY_NONE = 0,
+ SOIL_CAPABILITY_PRESENT = 1
+};
+static int has_cubemap_capability = SOIL_CAPABILITY_UNKNOWN;
+int query_cubemap_capability( void );
+#define SOIL_TEXTURE_WRAP_R 0x8072
+#define SOIL_CLAMP_TO_EDGE 0x812F
+#define SOIL_NORMAL_MAP 0x8511
+#define SOIL_REFLECTION_MAP 0x8512
+#define SOIL_TEXTURE_CUBE_MAP 0x8513
+#define SOIL_TEXTURE_BINDING_CUBE_MAP 0x8514
+#define SOIL_TEXTURE_CUBE_MAP_POSITIVE_X 0x8515
+#define SOIL_TEXTURE_CUBE_MAP_NEGATIVE_X 0x8516
+#define SOIL_TEXTURE_CUBE_MAP_POSITIVE_Y 0x8517
+#define SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Y 0x8518
+#define SOIL_TEXTURE_CUBE_MAP_POSITIVE_Z 0x8519
+#define SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Z 0x851A
+#define SOIL_PROXY_TEXTURE_CUBE_MAP 0x851B
+#define SOIL_MAX_CUBE_MAP_TEXTURE_SIZE 0x851C
+/* for non-power-of-two texture */
+static int has_NPOT_capability = SOIL_CAPABILITY_UNKNOWN;
+int query_NPOT_capability( void );
+/* for texture rectangles */
+static int has_tex_rectangle_capability = SOIL_CAPABILITY_UNKNOWN;
+int query_tex_rectangle_capability( void );
+#define SOIL_TEXTURE_RECTANGLE_ARB 0x84F5
+#define SOIL_MAX_RECTANGLE_TEXTURE_SIZE_ARB 0x84F8
+/* for using DXT compression */
+static int has_DXT_capability = SOIL_CAPABILITY_UNKNOWN;
+int query_DXT_capability( void );
+#define SOIL_RGB_S3TC_DXT1 0x83F0
+#define SOIL_RGBA_S3TC_DXT1 0x83F1
+#define SOIL_RGBA_S3TC_DXT3 0x83F2
+#define SOIL_RGBA_S3TC_DXT5 0x83F3
+typedef void (APIENTRY * P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC) (GLenum target, GLint level, GLenum internalformat, GLsizei width, GLsizei height, GLint border, GLsizei imageSize, const GLvoid * data);
+P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC soilGlCompressedTexImage2D = NULL;
+unsigned int SOIL_direct_load_DDS(
+ const char *filename,
+ unsigned int reuse_texture_ID,
+ int flags,
+ int loading_as_cubemap );
+unsigned int SOIL_direct_load_DDS_from_memory(
+ const unsigned char *const buffer,
+ int buffer_length,
+ unsigned int reuse_texture_ID,
+ int flags,
+ int loading_as_cubemap );
+/* other functions */
+unsigned int
+ SOIL_internal_create_OGL_texture
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags,
+ unsigned int opengl_texture_type,
+ unsigned int opengl_texture_target,
+ unsigned int texture_check_size_enum
+ );
+
+/* and the code magic begins here [8^) */
+unsigned int
+ SOIL_load_OGL_texture
+ (
+ const char *filename,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels;
+ unsigned int tex_id;
+ /* does the user want direct uploading of the image as a DDS file? */
+ if( flags & SOIL_FLAG_DDS_LOAD_DIRECT )
+ {
+ /* 1st try direct loading of the image as a DDS file
+ note: direct uploading will only load what is in the
+ DDS file, no MIPmaps will be generated, the image will
+ not be flipped, etc. */
+ tex_id = SOIL_direct_load_DDS( filename, reuse_texture_ID, flags, 0 );
+ if( tex_id )
+ {
+ /* hey, it worked!! */
+ return tex_id;
+ }
+ }
+ /* try to load the image */
+ img = SOIL_load_image( filename, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* OK, make it a texture! */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ reuse_texture_ID, flags,
+ GL_TEXTURE_2D, GL_TEXTURE_2D,
+ GL_MAX_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_HDR_texture
+ (
+ const char *filename,
+ int fake_HDR_format,
+ int rescale_to_max,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels;
+ unsigned int tex_id;
+ /* no direct uploading of the image as a DDS file */
+ /* error check */
+ if( (fake_HDR_format != SOIL_HDR_RGBE) &&
+ (fake_HDR_format != SOIL_HDR_RGBdivA) &&
+ (fake_HDR_format != SOIL_HDR_RGBdivA2) )
+ {
+ result_string_pointer = "Invalid fake HDR format specified";
+ return 0;
+ }
+ /* try to load the image (only the HDR type) */
+ img = stbi_hdr_load_rgbe( filename, &width, &height, &channels, 4 );
+ /* channels holds the original number of channels, which may have been forced */
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* the load worked, do I need to convert it? */
+ if( fake_HDR_format == SOIL_HDR_RGBdivA )
+ {
+ RGBE_to_RGBdivA( img, width, height, rescale_to_max );
+ } else if( fake_HDR_format == SOIL_HDR_RGBdivA2 )
+ {
+ RGBE_to_RGBdivA2( img, width, height, rescale_to_max );
+ }
+ /* OK, make it a texture! */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ reuse_texture_ID, flags,
+ GL_TEXTURE_2D, GL_TEXTURE_2D,
+ GL_MAX_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_texture_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels;
+ unsigned int tex_id;
+ /* does the user want direct uploading of the image as a DDS file? */
+ if( flags & SOIL_FLAG_DDS_LOAD_DIRECT )
+ {
+ /* 1st try direct loading of the image as a DDS file
+ note: direct uploading will only load what is in the
+ DDS file, no MIPmaps will be generated, the image will
+ not be flipped, etc. */
+ tex_id = SOIL_direct_load_DDS_from_memory(
+ buffer, buffer_length,
+ reuse_texture_ID, flags, 0 );
+ if( tex_id )
+ {
+ /* hey, it worked!! */
+ return tex_id;
+ }
+ }
+ /* try to load the image */
+ img = SOIL_load_image_from_memory(
+ buffer, buffer_length,
+ &width, &height, &channels,
+ force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* OK, make it a texture! */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ reuse_texture_ID, flags,
+ GL_TEXTURE_2D, GL_TEXTURE_2D,
+ GL_MAX_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_cubemap
+ (
+ const char *x_pos_file,
+ const char *x_neg_file,
+ const char *y_pos_file,
+ const char *y_neg_file,
+ const char *z_pos_file,
+ const char *z_neg_file,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels;
+ unsigned int tex_id;
+ /* error checking */
+ if( (x_pos_file == NULL) ||
+ (x_neg_file == NULL) ||
+ (y_pos_file == NULL) ||
+ (y_neg_file == NULL) ||
+ (z_pos_file == NULL) ||
+ (z_neg_file == NULL) )
+ {
+ result_string_pointer = "Invalid cube map files list";
+ return 0;
+ }
+ /* capability checking */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ result_string_pointer = "No cube map capability present";
+ return 0;
+ }
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( x_pos_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, and create a texture ID if necessary */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ reuse_texture_ID, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_X,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( x_neg_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_X,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( y_pos_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_Y,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( y_neg_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Y,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( z_pos_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_Z,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image( z_neg_file, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Z,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_cubemap_from_memory
+ (
+ const unsigned char *const x_pos_buffer,
+ int x_pos_buffer_length,
+ const unsigned char *const x_neg_buffer,
+ int x_neg_buffer_length,
+ const unsigned char *const y_pos_buffer,
+ int y_pos_buffer_length,
+ const unsigned char *const y_neg_buffer,
+ int y_neg_buffer_length,
+ const unsigned char *const z_pos_buffer,
+ int z_pos_buffer_length,
+ const unsigned char *const z_neg_buffer,
+ int z_neg_buffer_length,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels;
+ unsigned int tex_id;
+ /* error checking */
+ if( (x_pos_buffer == NULL) ||
+ (x_neg_buffer == NULL) ||
+ (y_pos_buffer == NULL) ||
+ (y_neg_buffer == NULL) ||
+ (z_pos_buffer == NULL) ||
+ (z_neg_buffer == NULL) )
+ {
+ result_string_pointer = "Invalid cube map buffers list";
+ return 0;
+ }
+ /* capability checking */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ result_string_pointer = "No cube map capability present";
+ return 0;
+ }
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ x_pos_buffer, x_pos_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, and create a texture ID if necessary */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ reuse_texture_ID, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_X,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ x_neg_buffer, x_neg_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_X,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ y_pos_buffer, y_pos_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_Y,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ y_neg_buffer, y_neg_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Y,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ z_pos_buffer, z_pos_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_POSITIVE_Z,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* continue? */
+ if( tex_id != 0 )
+ {
+ /* 1st face: try to load the image */
+ img = SOIL_load_image_from_memory(
+ z_neg_buffer, z_neg_buffer_length,
+ &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* upload the texture, but reuse the assigned texture ID */
+ tex_id = SOIL_internal_create_OGL_texture(
+ img, width, height, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP, SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Z,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ /* and nuke the image data */
+ SOIL_free_image_data( img );
+ }
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_single_cubemap
+ (
+ const char *filename,
+ const char face_order[6],
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels, i;
+ unsigned int tex_id = 0;
+ /* error checking */
+ if( filename == NULL )
+ {
+ result_string_pointer = "Invalid single cube map file name";
+ return 0;
+ }
+ /* does the user want direct uploading of the image as a DDS file? */
+ if( flags & SOIL_FLAG_DDS_LOAD_DIRECT )
+ {
+ /* 1st try direct loading of the image as a DDS file
+ note: direct uploading will only load what is in the
+ DDS file, no MIPmaps will be generated, the image will
+ not be flipped, etc. */
+ tex_id = SOIL_direct_load_DDS( filename, reuse_texture_ID, flags, 1 );
+ if( tex_id )
+ {
+ /* hey, it worked!! */
+ return tex_id;
+ }
+ }
+ /* face order checking */
+ for( i = 0; i < 6; ++i )
+ {
+ if( (face_order[i] != 'N') &&
+ (face_order[i] != 'S') &&
+ (face_order[i] != 'W') &&
+ (face_order[i] != 'E') &&
+ (face_order[i] != 'U') &&
+ (face_order[i] != 'D') )
+ {
+ result_string_pointer = "Invalid single cube map face order";
+ return 0;
+ };
+ }
+ /* capability checking */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ result_string_pointer = "No cube map capability present";
+ return 0;
+ }
+ /* 1st off, try to load the full image */
+ img = SOIL_load_image( filename, &width, &height, &channels, force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* now, does this image have the right dimensions? */
+ if( (width != 6*height) &&
+ (6*width != height) )
+ {
+ SOIL_free_image_data( img );
+ result_string_pointer = "Single cubemap image must have a 6:1 ratio";
+ return 0;
+ }
+ /* try the image split and create */
+ tex_id = SOIL_create_OGL_single_cubemap(
+ img, width, height, channels,
+ face_order, reuse_texture_ID, flags
+ );
+ /* nuke the temporary image data and return the texture handle */
+ SOIL_free_image_data( img );
+ return tex_id;
+}
+
+unsigned int
+ SOIL_load_OGL_single_cubemap_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ const char face_order[6],
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* img;
+ int width, height, channels, i;
+ unsigned int tex_id = 0;
+ /* error checking */
+ if( buffer == NULL )
+ {
+ result_string_pointer = "Invalid single cube map buffer";
+ return 0;
+ }
+ /* does the user want direct uploading of the image as a DDS file? */
+ if( flags & SOIL_FLAG_DDS_LOAD_DIRECT )
+ {
+ /* 1st try direct loading of the image as a DDS file
+ note: direct uploading will only load what is in the
+ DDS file, no MIPmaps will be generated, the image will
+ not be flipped, etc. */
+ tex_id = SOIL_direct_load_DDS_from_memory(
+ buffer, buffer_length,
+ reuse_texture_ID, flags, 1 );
+ if( tex_id )
+ {
+ /* hey, it worked!! */
+ return tex_id;
+ }
+ }
+ /* face order checking */
+ for( i = 0; i < 6; ++i )
+ {
+ if( (face_order[i] != 'N') &&
+ (face_order[i] != 'S') &&
+ (face_order[i] != 'W') &&
+ (face_order[i] != 'E') &&
+ (face_order[i] != 'U') &&
+ (face_order[i] != 'D') )
+ {
+ result_string_pointer = "Invalid single cube map face order";
+ return 0;
+ };
+ }
+ /* capability checking */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ result_string_pointer = "No cube map capability present";
+ return 0;
+ }
+ /* 1st off, try to load the full image */
+ img = SOIL_load_image_from_memory(
+ buffer, buffer_length,
+ &width, &height, &channels,
+ force_channels );
+ /* channels holds the original number of channels, which may have been forced */
+ if( (force_channels >= 1) && (force_channels <= 4) )
+ {
+ channels = force_channels;
+ }
+ if( NULL == img )
+ {
+ /* image loading failed */
+ result_string_pointer = stbi_failure_reason();
+ return 0;
+ }
+ /* now, does this image have the right dimensions? */
+ if( (width != 6*height) &&
+ (6*width != height) )
+ {
+ SOIL_free_image_data( img );
+ result_string_pointer = "Single cubemap image must have a 6:1 ratio";
+ return 0;
+ }
+ /* try the image split and create */
+ tex_id = SOIL_create_OGL_single_cubemap(
+ img, width, height, channels,
+ face_order, reuse_texture_ID, flags
+ );
+ /* nuke the temporary image data and return the texture handle */
+ SOIL_free_image_data( img );
+ return tex_id;
+}
+
+unsigned int
+ SOIL_create_OGL_single_cubemap
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ const char face_order[6],
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* variables */
+ unsigned char* sub_img;
+ int dw, dh, sz, i;
+ unsigned int tex_id;
+ /* error checking */
+ if( data == NULL )
+ {
+ result_string_pointer = "Invalid single cube map image data";
+ return 0;
+ }
+ /* face order checking */
+ for( i = 0; i < 6; ++i )
+ {
+ if( (face_order[i] != 'N') &&
+ (face_order[i] != 'S') &&
+ (face_order[i] != 'W') &&
+ (face_order[i] != 'E') &&
+ (face_order[i] != 'U') &&
+ (face_order[i] != 'D') )
+ {
+ result_string_pointer = "Invalid single cube map face order";
+ return 0;
+ };
+ }
+ /* capability checking */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ result_string_pointer = "No cube map capability present";
+ return 0;
+ }
+ /* now, does this image have the right dimensions? */
+ if( (width != 6*height) &&
+ (6*width != height) )
+ {
+ result_string_pointer = "Single cubemap image must have a 6:1 ratio";
+ return 0;
+ }
+ /* which way am I stepping? */
+ if( width > height )
+ {
+ dw = height;
+ dh = 0;
+ } else
+ {
+ dw = 0;
+ dh = width;
+ }
+ sz = dw+dh;
+ sub_img = (unsigned char *)malloc( sz*sz*channels );
+ /* do the splitting and uploading */
+ tex_id = reuse_texture_ID;
+ for( i = 0; i < 6; ++i )
+ {
+ int x, y, idx = 0;
+ unsigned int cubemap_target = 0;
+ /* copy in the sub-image */
+ for( y = i*dh; y < i*dh+sz; ++y )
+ {
+ for( x = i*dw*channels; x < (i*dw+sz)*channels; ++x )
+ {
+ sub_img[idx++] = data[y*width*channels+x];
+ }
+ }
+ /* what is my texture target?
+ remember, this coordinate system is
+ LHS if viewed from inside the cube! */
+ switch( face_order[i] )
+ {
+ case 'N':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_POSITIVE_Z;
+ break;
+ case 'S':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Z;
+ break;
+ case 'W':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_NEGATIVE_X;
+ break;
+ case 'E':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_POSITIVE_X;
+ break;
+ case 'U':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_POSITIVE_Y;
+ break;
+ case 'D':
+ cubemap_target = SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Y;
+ break;
+ }
+ /* upload it as a texture */
+ tex_id = SOIL_internal_create_OGL_texture(
+ sub_img, sz, sz, channels,
+ tex_id, flags,
+ SOIL_TEXTURE_CUBE_MAP,
+ cubemap_target,
+ SOIL_MAX_CUBE_MAP_TEXTURE_SIZE );
+ }
+ /* and nuke the image and sub-image data */
+ SOIL_free_image_data( sub_img );
+ /* and return the handle, such as it is */
+ return tex_id;
+}
+
+unsigned int
+ SOIL_create_OGL_texture
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ )
+{
+ /* wrapper function for 2D textures */
+ return SOIL_internal_create_OGL_texture(
+ data, width, height, channels,
+ reuse_texture_ID, flags,
+ GL_TEXTURE_2D, GL_TEXTURE_2D,
+ GL_MAX_TEXTURE_SIZE );
+}
+
+#if SOIL_CHECK_FOR_GL_ERRORS
+void check_for_GL_errors( const char *calling_location )
+{
+ /* check for errors */
+ GLenum err_code = glGetError();
+ while( GL_NO_ERROR != err_code )
+ {
+ printf( "OpenGL Error @ %s: %i", calling_location, err_code );
+ err_code = glGetError();
+ }
+}
+#else
+void check_for_GL_errors( const char *calling_location )
+{
+ /* no check for errors */
+}
+#endif
+
+unsigned int
+ SOIL_internal_create_OGL_texture
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags,
+ unsigned int opengl_texture_type,
+ unsigned int opengl_texture_target,
+ unsigned int texture_check_size_enum
+ )
+{
+ /* variables */
+ unsigned char* img;
+ unsigned int tex_id;
+ unsigned int internal_texture_format = 0, original_texture_format = 0;
+ int DXT_mode = SOIL_CAPABILITY_UNKNOWN;
+ int max_supported_size;
+ /* If the user wants to use the texture rectangle I kill a few flags */
+ if( flags & SOIL_FLAG_TEXTURE_RECTANGLE )
+ {
+ /* well, the user asked for it, can we do that? */
+ if( query_tex_rectangle_capability() == SOIL_CAPABILITY_PRESENT )
+ {
+ /* only allow this if the user in _NOT_ trying to do a cubemap! */
+ if( opengl_texture_type == GL_TEXTURE_2D )
+ {
+ /* clean out the flags that cannot be used with texture rectangles */
+ flags &= ~(
+ SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS |
+ SOIL_FLAG_TEXTURE_REPEATS
+ );
+ /* and change my target */
+ opengl_texture_target = SOIL_TEXTURE_RECTANGLE_ARB;
+ opengl_texture_type = SOIL_TEXTURE_RECTANGLE_ARB;
+ } else
+ {
+ /* not allowed for any other uses (yes, I'm looking at you, cubemaps!) */
+ flags &= ~SOIL_FLAG_TEXTURE_RECTANGLE;
+ }
+
+ } else
+ {
+ /* can't do it, and that is a breakable offense (uv coords use pixels instead of [0,1]!) */
+ result_string_pointer = "Texture Rectangle extension unsupported";
+ return 0;
+ }
+ }
+ /* create a copy the image data */
+ img = (unsigned char*)malloc( width*height*channels );
+ memcpy( img, data, width*height*channels );
+ /* does the user want me to invert the image? */
+ if( flags & SOIL_FLAG_INVERT_Y )
+ {
+ int i, j;
+ for( j = 0; j*2 < height; ++j )
+ {
+ int index1 = j * width * channels;
+ int index2 = (height - 1 - j) * width * channels;
+ for( i = width * channels; i > 0; --i )
+ {
+ unsigned char temp = img[index1];
+ img[index1] = img[index2];
+ img[index2] = temp;
+ ++index1;
+ ++index2;
+ }
+ }
+ }
+ /* does the user want me to scale the colors into the NTSC safe RGB range? */
+ if( flags & SOIL_FLAG_NTSC_SAFE_RGB )
+ {
+ scale_image_RGB_to_NTSC_safe( img, width, height, channels );
+ }
+ /* does the user want me to convert from straight to pre-multiplied alpha?
+ (and do we even _have_ alpha?) */
+ if( flags & SOIL_FLAG_MULTIPLY_ALPHA )
+ {
+ int i;
+ switch( channels )
+ {
+ case 2:
+ for( i = 0; i < 2*width*height; i += 2 )
+ {
+ img[i] = (img[i] * img[i+1] + 128) >> 8;
+ }
+ break;
+ case 4:
+ for( i = 0; i < 4*width*height; i += 4 )
+ {
+ img[i+0] = (img[i+0] * img[i+3] + 128) >> 8;
+ img[i+1] = (img[i+1] * img[i+3] + 128) >> 8;
+ img[i+2] = (img[i+2] * img[i+3] + 128) >> 8;
+ }
+ break;
+ default:
+ /* no other number of channels contains alpha data */
+ break;
+ }
+ }
+ /* if the user can't support NPOT textures, make sure we force the POT option */
+ if( (query_NPOT_capability() == SOIL_CAPABILITY_NONE) &&
+ !(flags & SOIL_FLAG_TEXTURE_RECTANGLE) )
+ {
+ /* add in the POT flag */
+ flags |= SOIL_FLAG_POWER_OF_TWO;
+ }
+ /* how large of a texture can this OpenGL implementation handle? */
+ /* texture_check_size_enum will be GL_MAX_TEXTURE_SIZE or SOIL_MAX_CUBE_MAP_TEXTURE_SIZE */
+ glGetIntegerv( texture_check_size_enum, &max_supported_size );
+ /* do I need to make it a power of 2? */
+ if(
+ (flags & SOIL_FLAG_POWER_OF_TWO) || /* user asked for it */
+ (flags & SOIL_FLAG_MIPMAPS) || /* need it for the MIP-maps */
+ (width > max_supported_size) || /* it's too big, (make sure it's */
+ (height > max_supported_size) ) /* 2^n for later down-sampling) */
+ {
+ int new_width = 1;
+ int new_height = 1;
+ while( new_width < width )
+ {
+ new_width *= 2;
+ }
+ while( new_height < height )
+ {
+ new_height *= 2;
+ }
+ /* still? */
+ if( (new_width != width) || (new_height != height) )
+ {
+ /* yep, resize */
+ unsigned char *resampled = (unsigned char*)malloc( channels*new_width*new_height );
+ up_scale_image(
+ img, width, height, channels,
+ resampled, new_width, new_height );
+ /* OJO this is for debug only! */
+ /*
+ SOIL_save_image( "\\showme.bmp", SOIL_SAVE_TYPE_BMP,
+ new_width, new_height, channels,
+ resampled );
+ */
+ /* nuke the old guy, then point it at the new guy */
+ SOIL_free_image_data( img );
+ img = resampled;
+ width = new_width;
+ height = new_height;
+ }
+ }
+ /* now, if it is too large... */
+ if( (width > max_supported_size) || (height > max_supported_size) )
+ {
+ /* I've already made it a power of two, so simply use the MIPmapping
+ code to reduce its size to the allowable maximum. */
+ unsigned char *resampled;
+ int reduce_block_x = 1, reduce_block_y = 1;
+ int new_width, new_height;
+ if( width > max_supported_size )
+ {
+ reduce_block_x = width / max_supported_size;
+ }
+ if( height > max_supported_size )
+ {
+ reduce_block_y = height / max_supported_size;
+ }
+ new_width = width / reduce_block_x;
+ new_height = height / reduce_block_y;
+ resampled = (unsigned char*)malloc( channels*new_width*new_height );
+ /* perform the actual reduction */
+ mipmap_image( img, width, height, channels,
+ resampled, reduce_block_x, reduce_block_y );
+ /* nuke the old guy, then point it at the new guy */
+ SOIL_free_image_data( img );
+ img = resampled;
+ width = new_width;
+ height = new_height;
+ }
+ /* does the user want us to use YCoCg color space? */
+ if( flags & SOIL_FLAG_CoCg_Y )
+ {
+ /* this will only work with RGB and RGBA images */
+ convert_RGB_to_YCoCg( img, width, height, channels );
+ /*
+ save_image_as_DDS( "CoCg_Y.dds", width, height, channels, img );
+ */
+ }
+ /* create the OpenGL texture ID handle
+ (note: allowing a forced texture ID lets me reload a texture) */
+ tex_id = reuse_texture_ID;
+ if( tex_id == 0 )
+ {
+ glGenTextures( 1, &tex_id );
+ }
+ check_for_GL_errors( "glGenTextures" );
+ /* Note: sometimes glGenTextures fails (usually no OpenGL context) */
+ if( tex_id )
+ {
+ /* and what type am I using as the internal texture format? */
+ switch( channels )
+ {
+ case 1:
+ original_texture_format = GL_LUMINANCE;
+ break;
+ case 2:
+ original_texture_format = GL_LUMINANCE_ALPHA;
+ break;
+ case 3:
+ original_texture_format = GL_RGB;
+ break;
+ case 4:
+ original_texture_format = GL_RGBA;
+ break;
+ }
+ internal_texture_format = original_texture_format;
+ /* does the user want me to, and can I, save as DXT? */
+ if( flags & SOIL_FLAG_COMPRESS_TO_DXT )
+ {
+ DXT_mode = query_DXT_capability();
+ if( DXT_mode == SOIL_CAPABILITY_PRESENT )
+ {
+ /* I can use DXT, whether I compress it or OpenGL does */
+ if( (channels & 1) == 1 )
+ {
+ /* 1 or 3 channels = DXT1 */
+ internal_texture_format = SOIL_RGB_S3TC_DXT1;
+ } else
+ {
+ /* 2 or 4 channels = DXT5 */
+ internal_texture_format = SOIL_RGBA_S3TC_DXT5;
+ }
+ }
+ }
+ /* bind an OpenGL texture ID */
+ glBindTexture( opengl_texture_type, tex_id );
+ check_for_GL_errors( "glBindTexture" );
+ /* upload the main image */
+ if( DXT_mode == SOIL_CAPABILITY_PRESENT )
+ {
+ /* user wants me to do the DXT conversion! */
+ int DDS_size;
+ unsigned char *DDS_data = NULL;
+ if( (channels & 1) == 1 )
+ {
+ /* RGB, use DXT1 */
+ DDS_data = convert_image_to_DXT1( img, width, height, channels, &DDS_size );
+ } else
+ {
+ /* RGBA, use DXT5 */
+ DDS_data = convert_image_to_DXT5( img, width, height, channels, &DDS_size );
+ }
+ if( DDS_data )
+ {
+ soilGlCompressedTexImage2D(
+ opengl_texture_target, 0,
+ internal_texture_format, width, height, 0,
+ DDS_size, DDS_data );
+ check_for_GL_errors( "glCompressedTexImage2D" );
+ SOIL_free_image_data( DDS_data );
+ /* printf( "Internal DXT compressor\n" ); */
+ } else
+ {
+ /* my compression failed, try the OpenGL driver's version */
+ glTexImage2D(
+ opengl_texture_target, 0,
+ internal_texture_format, width, height, 0,
+ original_texture_format, GL_UNSIGNED_BYTE, img );
+ check_for_GL_errors( "glTexImage2D" );
+ /* printf( "OpenGL DXT compressor\n" ); */
+ }
+ } else
+ {
+ /* user want OpenGL to do all the work! */
+ glTexImage2D(
+ opengl_texture_target, 0,
+ internal_texture_format, width, height, 0,
+ original_texture_format, GL_UNSIGNED_BYTE, img );
+ check_for_GL_errors( "glTexImage2D" );
+ /*printf( "OpenGL DXT compressor\n" ); */
+ }
+ /* are any MIPmaps desired? */
+ if( flags & SOIL_FLAG_MIPMAPS )
+ {
+ int MIPlevel = 1;
+ int MIPwidth = (width+1) / 2;
+ int MIPheight = (height+1) / 2;
+ unsigned char *resampled = (unsigned char*)malloc( channels*MIPwidth*MIPheight );
+ while( ((1< 0; --i )
+ {
+ unsigned char temp = pixel_data[index1];
+ pixel_data[index1] = pixel_data[index2];
+ pixel_data[index2] = temp;
+ ++index1;
+ ++index2;
+ }
+ }
+
+ /* save the image */
+ save_result = SOIL_save_image( filename, image_type, width, height, 3, pixel_data);
+
+ /* And free the memory */
+ SOIL_free_image_data( pixel_data );
+ return save_result;
+}
+
+unsigned char*
+ SOIL_load_image
+ (
+ const char *filename,
+ int *width, int *height, int *channels,
+ int force_channels
+ )
+{
+ unsigned char *result = stbi_load( filename,
+ width, height, channels, force_channels );
+ if( result == NULL )
+ {
+ result_string_pointer = stbi_failure_reason();
+ } else
+ {
+ result_string_pointer = "Image loaded";
+ }
+ return result;
+}
+
+unsigned char*
+ SOIL_load_image_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ int *width, int *height, int *channels,
+ int force_channels
+ )
+{
+ unsigned char *result = stbi_load_from_memory(
+ buffer, buffer_length,
+ width, height, channels,
+ force_channels );
+ if( result == NULL )
+ {
+ result_string_pointer = stbi_failure_reason();
+ } else
+ {
+ result_string_pointer = "Image loaded from memory";
+ }
+ return result;
+}
+
+int
+ SOIL_save_image
+ (
+ const char *filename,
+ int image_type,
+ int width, int height, int channels,
+ const unsigned char *const data
+ )
+{
+ int save_result;
+
+ /* error check */
+ if( (width < 1) || (height < 1) ||
+ (channels < 1) || (channels > 4) ||
+ (data == NULL) ||
+ (filename == NULL) )
+ {
+ return 0;
+ }
+ if( image_type == SOIL_SAVE_TYPE_BMP )
+ {
+ save_result = stbi_write_bmp( filename,
+ width, height, channels, (void*)data );
+ } else
+ if( image_type == SOIL_SAVE_TYPE_TGA )
+ {
+ save_result = stbi_write_tga( filename,
+ width, height, channels, (void*)data );
+ } else
+ if( image_type == SOIL_SAVE_TYPE_DDS )
+ {
+ save_result = save_image_as_DDS( filename,
+ width, height, channels, (const unsigned char *const)data );
+ } else
+ {
+ save_result = 0;
+ }
+ if( save_result == 0 )
+ {
+ result_string_pointer = "Saving the image failed";
+ } else
+ {
+ result_string_pointer = "Image saved";
+ }
+ return save_result;
+}
+
+void
+ SOIL_free_image_data
+ (
+ unsigned char *img_data
+ )
+{
+ free( (void*)img_data );
+}
+
+const char*
+ SOIL_last_result
+ (
+ void
+ )
+{
+ return result_string_pointer;
+}
+
+unsigned int SOIL_direct_load_DDS_from_memory(
+ const unsigned char *const buffer,
+ int buffer_length,
+ unsigned int reuse_texture_ID,
+ int flags,
+ int loading_as_cubemap )
+{
+ /* variables */
+ DDS_header header;
+ unsigned int buffer_index = 0;
+ unsigned int tex_ID = 0;
+ /* file reading variables */
+ unsigned int S3TC_type = 0;
+ unsigned char *DDS_data;
+ unsigned int DDS_main_size;
+ unsigned int DDS_full_size;
+ unsigned int width, height;
+ int mipmaps, cubemap, uncompressed, block_size = 16;
+ unsigned int flag;
+ unsigned int cf_target, ogl_target_start, ogl_target_end;
+ unsigned int opengl_texture_type;
+ int i;
+ /* 1st off, does the filename even exist? */
+ if( NULL == buffer )
+ {
+ /* we can't do it! */
+ result_string_pointer = "NULL buffer";
+ return 0;
+ }
+ if( buffer_length < sizeof( DDS_header ) )
+ {
+ /* we can't do it! */
+ result_string_pointer = "DDS file was too small to contain the DDS header";
+ return 0;
+ }
+ /* try reading in the header */
+ memcpy ( (void*)(&header), (const void *)buffer, sizeof( DDS_header ) );
+ buffer_index = sizeof( DDS_header );
+ /* guilty until proven innocent */
+ result_string_pointer = "Failed to read a known DDS header";
+ /* validate the header (warning, "goto"'s ahead, shield your eyes!!) */
+ flag = ('D'<<0)|('D'<<8)|('S'<<16)|(' '<<24);
+ if( header.dwMagic != flag ) {goto quick_exit;}
+ if( header.dwSize != 124 ) {goto quick_exit;}
+ /* I need all of these */
+ flag = DDSD_CAPS | DDSD_HEIGHT | DDSD_WIDTH | DDSD_PIXELFORMAT;
+ if( (header.dwFlags & flag) != flag ) {goto quick_exit;}
+ /* According to the MSDN spec, the dwFlags should contain
+ DDSD_LINEARSIZE if it's compressed, or DDSD_PITCH if
+ uncompressed. Some DDS writers do not conform to the
+ spec, so I need to make my reader more tolerant */
+ /* I need one of these */
+ flag = DDPF_FOURCC | DDPF_RGB;
+ if( (header.sPixelFormat.dwFlags & flag) == 0 ) {goto quick_exit;}
+ if( header.sPixelFormat.dwSize != 32 ) {goto quick_exit;}
+ if( (header.sCaps.dwCaps1 & DDSCAPS_TEXTURE) == 0 ) {goto quick_exit;}
+ /* make sure it is a type we can upload */
+ if( (header.sPixelFormat.dwFlags & DDPF_FOURCC) &&
+ !(
+ (header.sPixelFormat.dwFourCC == (('D'<<0)|('X'<<8)|('T'<<16)|('1'<<24))) ||
+ (header.sPixelFormat.dwFourCC == (('D'<<0)|('X'<<8)|('T'<<16)|('3'<<24))) ||
+ (header.sPixelFormat.dwFourCC == (('D'<<0)|('X'<<8)|('T'<<16)|('5'<<24)))
+ ) )
+ {
+ goto quick_exit;
+ }
+ /* OK, validated the header, let's load the image data */
+ result_string_pointer = "DDS header loaded and validated";
+ width = header.dwWidth;
+ height = header.dwHeight;
+ uncompressed = 1 - (header.sPixelFormat.dwFlags & DDPF_FOURCC) / DDPF_FOURCC;
+ cubemap = (header.sCaps.dwCaps2 & DDSCAPS2_CUBEMAP) / DDSCAPS2_CUBEMAP;
+ if( uncompressed )
+ {
+ S3TC_type = GL_RGB;
+ block_size = 3;
+ if( header.sPixelFormat.dwFlags & DDPF_ALPHAPIXELS )
+ {
+ S3TC_type = GL_RGBA;
+ block_size = 4;
+ }
+ DDS_main_size = width * height * block_size;
+ } else
+ {
+ /* can we even handle direct uploading to OpenGL DXT compressed images? */
+ if( query_DXT_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ /* we can't do it! */
+ result_string_pointer = "Direct upload of S3TC images not supported by the OpenGL driver";
+ return 0;
+ }
+ /* well, we know it is DXT1/3/5, because we checked above */
+ switch( (header.sPixelFormat.dwFourCC >> 24) - '0' )
+ {
+ case 1:
+ S3TC_type = SOIL_RGBA_S3TC_DXT1;
+ block_size = 8;
+ break;
+ case 3:
+ S3TC_type = SOIL_RGBA_S3TC_DXT3;
+ block_size = 16;
+ break;
+ case 5:
+ S3TC_type = SOIL_RGBA_S3TC_DXT5;
+ block_size = 16;
+ break;
+ }
+ DDS_main_size = ((width+3)>>2)*((height+3)>>2)*block_size;
+ }
+ if( cubemap )
+ {
+ /* does the user want a cubemap? */
+ if( !loading_as_cubemap )
+ {
+ /* we can't do it! */
+ result_string_pointer = "DDS image was a cubemap";
+ return 0;
+ }
+ /* can we even handle cubemaps with the OpenGL driver? */
+ if( query_cubemap_capability() != SOIL_CAPABILITY_PRESENT )
+ {
+ /* we can't do it! */
+ result_string_pointer = "Direct upload of cubemap images not supported by the OpenGL driver";
+ return 0;
+ }
+ ogl_target_start = SOIL_TEXTURE_CUBE_MAP_POSITIVE_X;
+ ogl_target_end = SOIL_TEXTURE_CUBE_MAP_NEGATIVE_Z;
+ opengl_texture_type = SOIL_TEXTURE_CUBE_MAP;
+ } else
+ {
+ /* does the user want a non-cubemap? */
+ if( loading_as_cubemap )
+ {
+ /* we can't do it! */
+ result_string_pointer = "DDS image was not a cubemap";
+ return 0;
+ }
+ ogl_target_start = GL_TEXTURE_2D;
+ ogl_target_end = GL_TEXTURE_2D;
+ opengl_texture_type = GL_TEXTURE_2D;
+ }
+ if( (header.sCaps.dwCaps1 & DDSCAPS_MIPMAP) && (header.dwMipMapCount > 1) )
+ {
+ int shift_offset;
+ mipmaps = header.dwMipMapCount - 1;
+ DDS_full_size = DDS_main_size;
+ if( uncompressed )
+ {
+ /* uncompressed DDS, simple MIPmap size calculation */
+ shift_offset = 0;
+ } else
+ {
+ /* compressed DDS, MIPmap size calculation is block based */
+ shift_offset = 2;
+ }
+ for( i = 1; i <= mipmaps; ++ i )
+ {
+ int w, h;
+ w = width >> (shift_offset + i);
+ h = height >> (shift_offset + i);
+ if( w < 1 )
+ {
+ w = 1;
+ }
+ if( h < 1 )
+ {
+ h = 1;
+ }
+ DDS_full_size += w*h*block_size;
+ }
+ } else
+ {
+ mipmaps = 0;
+ DDS_full_size = DDS_main_size;
+ }
+ DDS_data = (unsigned char*)malloc( DDS_full_size );
+ /* got the image data RAM, create or use an existing OpenGL texture handle */
+ tex_ID = reuse_texture_ID;
+ if( tex_ID == 0 )
+ {
+ glGenTextures( 1, &tex_ID );
+ }
+ /* bind an OpenGL texture ID */
+ glBindTexture( opengl_texture_type, tex_ID );
+ /* do this for each face of the cubemap! */
+ for( cf_target = ogl_target_start; cf_target <= ogl_target_end; ++cf_target )
+ {
+ if( buffer_index + DDS_full_size <= buffer_length )
+ {
+ unsigned int byte_offset = DDS_main_size;
+ memcpy( (void*)DDS_data, (const void*)(&buffer[buffer_index]), DDS_full_size );
+ buffer_index += DDS_full_size;
+ /* upload the main chunk */
+ if( uncompressed )
+ {
+ /* and remember, DXT uncompressed uses BGR(A),
+ so swap to RGB(A) for ALL MIPmap levels */
+ for( i = 0; i < DDS_full_size; i += block_size )
+ {
+ unsigned char temp = DDS_data[i];
+ DDS_data[i] = DDS_data[i+2];
+ DDS_data[i+2] = temp;
+ }
+ glTexImage2D(
+ cf_target, 0,
+ S3TC_type, width, height, 0,
+ S3TC_type, GL_UNSIGNED_BYTE, DDS_data );
+ } else
+ {
+ soilGlCompressedTexImage2D(
+ cf_target, 0,
+ S3TC_type, width, height, 0,
+ DDS_main_size, DDS_data );
+ }
+ /* upload the mipmaps, if we have them */
+ for( i = 1; i <= mipmaps; ++i )
+ {
+ int w, h, mip_size;
+ w = width >> i;
+ h = height >> i;
+ if( w < 1 )
+ {
+ w = 1;
+ }
+ if( h < 1 )
+ {
+ h = 1;
+ }
+ /* upload this mipmap */
+ if( uncompressed )
+ {
+ mip_size = w*h*block_size;
+ glTexImage2D(
+ cf_target, i,
+ S3TC_type, w, h, 0,
+ S3TC_type, GL_UNSIGNED_BYTE, &DDS_data[byte_offset] );
+ } else
+ {
+ mip_size = ((w+3)/4)*((h+3)/4)*block_size;
+ soilGlCompressedTexImage2D(
+ cf_target, i,
+ S3TC_type, w, h, 0,
+ mip_size, &DDS_data[byte_offset] );
+ }
+ /* and move to the next mipmap */
+ byte_offset += mip_size;
+ }
+ /* it worked! */
+ result_string_pointer = "DDS file loaded";
+ } else
+ {
+ glDeleteTextures( 1, & tex_ID );
+ tex_ID = 0;
+ cf_target = ogl_target_end + 1;
+ result_string_pointer = "DDS file was too small for expected image data";
+ }
+ }/* end reading each face */
+ SOIL_free_image_data( DDS_data );
+ if( tex_ID )
+ {
+ /* did I have MIPmaps? */
+ if( mipmaps > 0 )
+ {
+ /* instruct OpenGL to use the MIPmaps */
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_MAG_FILTER, GL_LINEAR );
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_MIN_FILTER, GL_LINEAR_MIPMAP_LINEAR );
+ } else
+ {
+ /* instruct OpenGL _NOT_ to use the MIPmaps */
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_MAG_FILTER, GL_LINEAR );
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_MIN_FILTER, GL_LINEAR );
+ }
+ /* does the user want clamping, or wrapping? */
+ if( flags & SOIL_FLAG_TEXTURE_REPEATS )
+ {
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_WRAP_S, GL_REPEAT );
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_WRAP_T, GL_REPEAT );
+ glTexParameteri( opengl_texture_type, SOIL_TEXTURE_WRAP_R, GL_REPEAT );
+ } else
+ {
+ /* unsigned int clamp_mode = SOIL_CLAMP_TO_EDGE; */
+ unsigned int clamp_mode = GL_CLAMP;
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_WRAP_S, clamp_mode );
+ glTexParameteri( opengl_texture_type, GL_TEXTURE_WRAP_T, clamp_mode );
+ glTexParameteri( opengl_texture_type, SOIL_TEXTURE_WRAP_R, clamp_mode );
+ }
+ }
+
+quick_exit:
+ /* report success or failure */
+ return tex_ID;
+}
+
+unsigned int SOIL_direct_load_DDS(
+ const char *filename,
+ unsigned int reuse_texture_ID,
+ int flags,
+ int loading_as_cubemap )
+{
+ FILE *f;
+ unsigned char *buffer;
+ size_t buffer_length, bytes_read;
+ unsigned int tex_ID = 0;
+ /* error checks */
+ if( NULL == filename )
+ {
+ result_string_pointer = "NULL filename";
+ return 0;
+ }
+ f = fopen( filename, "rb" );
+ if( NULL == f )
+ {
+ /* the file doesn't seem to exist (or be open-able) */
+ result_string_pointer = "Can not find DDS file";
+ return 0;
+ }
+ fseek( f, 0, SEEK_END );
+ buffer_length = ftell( f );
+ fseek( f, 0, SEEK_SET );
+ buffer = (unsigned char *) malloc( buffer_length );
+ if( NULL == buffer )
+ {
+ result_string_pointer = "malloc failed";
+ fclose( f );
+ return 0;
+ }
+ bytes_read = fread( (void*)buffer, 1, buffer_length, f );
+ fclose( f );
+ if( bytes_read < buffer_length )
+ {
+ /* huh? */
+ buffer_length = bytes_read;
+ }
+ /* now try to do the loading */
+ tex_ID = SOIL_direct_load_DDS_from_memory(
+ (const unsigned char *const)buffer, buffer_length,
+ reuse_texture_ID, flags, loading_as_cubemap );
+ SOIL_free_image_data( buffer );
+ return tex_ID;
+}
+
+int query_NPOT_capability( void )
+{
+ /* check for the capability */
+ if( has_NPOT_capability == SOIL_CAPABILITY_UNKNOWN )
+ {
+ /* we haven't yet checked for the capability, do so */
+ if(
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_ARB_texture_non_power_of_two" ) )
+ )
+ {
+ /* not there, flag the failure */
+ has_NPOT_capability = SOIL_CAPABILITY_NONE;
+ } else
+ {
+ /* it's there! */
+ has_NPOT_capability = SOIL_CAPABILITY_PRESENT;
+ }
+ }
+ /* let the user know if we can do non-power-of-two textures or not */
+ return has_NPOT_capability;
+}
+
+int query_tex_rectangle_capability( void )
+{
+ /* check for the capability */
+ if( has_tex_rectangle_capability == SOIL_CAPABILITY_UNKNOWN )
+ {
+ /* we haven't yet checked for the capability, do so */
+ if(
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_ARB_texture_rectangle" ) )
+ &&
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_EXT_texture_rectangle" ) )
+ &&
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_NV_texture_rectangle" ) )
+ )
+ {
+ /* not there, flag the failure */
+ has_tex_rectangle_capability = SOIL_CAPABILITY_NONE;
+ } else
+ {
+ /* it's there! */
+ has_tex_rectangle_capability = SOIL_CAPABILITY_PRESENT;
+ }
+ }
+ /* let the user know if we can do texture rectangles or not */
+ return has_tex_rectangle_capability;
+}
+
+int query_cubemap_capability( void )
+{
+ /* check for the capability */
+ if( has_cubemap_capability == SOIL_CAPABILITY_UNKNOWN )
+ {
+ /* we haven't yet checked for the capability, do so */
+ if(
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_ARB_texture_cube_map" ) )
+ &&
+ (NULL == strstr( (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_EXT_texture_cube_map" ) )
+ )
+ {
+ /* not there, flag the failure */
+ has_cubemap_capability = SOIL_CAPABILITY_NONE;
+ } else
+ {
+ /* it's there! */
+ has_cubemap_capability = SOIL_CAPABILITY_PRESENT;
+ }
+ }
+ /* let the user know if we can do cubemaps or not */
+ return has_cubemap_capability;
+}
+
+int query_DXT_capability( void )
+{
+ /* check for the capability */
+ if( has_DXT_capability == SOIL_CAPABILITY_UNKNOWN )
+ {
+ /* we haven't yet checked for the capability, do so */
+ if( NULL == strstr(
+ (char const*)glGetString( GL_EXTENSIONS ),
+ "GL_EXT_texture_compression_s3tc" ) )
+ {
+ /* not there, flag the failure */
+ has_DXT_capability = SOIL_CAPABILITY_NONE;
+ } else
+ {
+ /* and find the address of the extension function */
+ P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC ext_addr = NULL;
+ #ifdef WIN32
+ ext_addr = (P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC)
+ wglGetProcAddress
+ (
+ "glCompressedTexImage2DARB"
+ );
+ #elif defined(__APPLE__) || defined(__APPLE_CC__)
+ /* I can't test this Apple stuff! */
+ CFBundleRef bundle;
+ CFURLRef bundleURL =
+ CFURLCreateWithFileSystemPath(
+ kCFAllocatorDefault,
+ CFSTR("/System/Library/Frameworks/OpenGL.framework"),
+ kCFURLPOSIXPathStyle,
+ true );
+ CFStringRef extensionName =
+ CFStringCreateWithCString(
+ kCFAllocatorDefault,
+ "glCompressedTexImage2DARB",
+ kCFStringEncodingASCII );
+ bundle = CFBundleCreate( kCFAllocatorDefault, bundleURL );
+ assert( bundle != NULL );
+ ext_addr = (P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC)
+ CFBundleGetFunctionPointerForName
+ (
+ bundle, extensionName
+ );
+ CFRelease( bundleURL );
+ CFRelease( extensionName );
+ CFRelease( bundle );
+ #else
+ ext_addr = (P_SOIL_GLCOMPRESSEDTEXIMAGE2DPROC)
+ glXGetProcAddressARB
+ (
+ (const GLubyte *)"glCompressedTexImage2DARB"
+ );
+ #endif
+ /* Flag it so no checks needed later */
+ if( NULL == ext_addr )
+ {
+ /* hmm, not good!! This should not happen, but does on my
+ laptop's VIA chipset. The GL_EXT_texture_compression_s3tc
+ spec requires that ARB_texture_compression be present too.
+ this means I can upload and have the OpenGL drive do the
+ conversion, but I can't use my own routines or load DDS files
+ from disk and upload them directly [8^( */
+ has_DXT_capability = SOIL_CAPABILITY_NONE;
+ } else
+ {
+ /* all's well! */
+ soilGlCompressedTexImage2D = ext_addr;
+ has_DXT_capability = SOIL_CAPABILITY_PRESENT;
+ }
+ }
+ }
+ /* let the user know if we can do DXT or not */
+ return has_DXT_capability;
+}
diff --git a/cw 9/src/SOIL/SOIL.h b/cw 9/src/SOIL/SOIL.h
new file mode 100644
index 0000000..43f634f
--- /dev/null
+++ b/cw 9/src/SOIL/SOIL.h
@@ -0,0 +1,433 @@
+/**
+ @mainpage SOIL
+
+ Jonathan Dummer
+ 2007-07-26-10.36
+
+ Simple OpenGL Image Library
+
+ A tiny c library for uploading images as
+ textures into OpenGL. Also saving and
+ loading of images is supported.
+
+ I'm using Sean's Tool Box image loader as a base:
+ http://www.nothings.org/
+
+ I'm upgrading it to load TGA and DDS files, and a direct
+ path for loading DDS files straight into OpenGL textures,
+ when applicable.
+
+ Image Formats:
+ - BMP load & save
+ - TGA load & save
+ - DDS load & save
+ - PNG load
+ - JPG load
+
+ OpenGL Texture Features:
+ - resample to power-of-two sizes
+ - MIPmap generation
+ - compressed texture S3TC formats (if supported)
+ - can pre-multiply alpha for you, for better compositing
+ - can flip image about the y-axis (except pre-compressed DDS files)
+
+ Thanks to:
+ * Sean Barret - for the awesome stb_image
+ * Dan Venkitachalam - for finding some non-compliant DDS files, and patching some explicit casts
+ * everybody at gamedev.net
+**/
+
+#ifndef HEADER_SIMPLE_OPENGL_IMAGE_LIBRARY
+#define HEADER_SIMPLE_OPENGL_IMAGE_LIBRARY
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ The format of images that may be loaded (force_channels).
+ SOIL_LOAD_AUTO leaves the image in whatever format it was found.
+ SOIL_LOAD_L forces the image to load as Luminous (greyscale)
+ SOIL_LOAD_LA forces the image to load as Luminous with Alpha
+ SOIL_LOAD_RGB forces the image to load as Red Green Blue
+ SOIL_LOAD_RGBA forces the image to load as Red Green Blue Alpha
+**/
+enum
+{
+ SOIL_LOAD_AUTO = 0,
+ SOIL_LOAD_L = 1,
+ SOIL_LOAD_LA = 2,
+ SOIL_LOAD_RGB = 3,
+ SOIL_LOAD_RGBA = 4
+};
+
+/**
+ Passed in as reuse_texture_ID, will cause SOIL to
+ register a new texture ID using glGenTextures().
+ If the value passed into reuse_texture_ID > 0 then
+ SOIL will just re-use that texture ID (great for
+ reloading image assets in-game!)
+**/
+enum
+{
+ SOIL_CREATE_NEW_ID = 0
+};
+
+/**
+ flags you can pass into SOIL_load_OGL_texture()
+ and SOIL_create_OGL_texture().
+ (note that if SOIL_FLAG_DDS_LOAD_DIRECT is used
+ the rest of the flags with the exception of
+ SOIL_FLAG_TEXTURE_REPEATS will be ignored while
+ loading already-compressed DDS files.)
+
+ SOIL_FLAG_POWER_OF_TWO: force the image to be POT
+ SOIL_FLAG_MIPMAPS: generate mipmaps for the texture
+ SOIL_FLAG_TEXTURE_REPEATS: otherwise will clamp
+ SOIL_FLAG_MULTIPLY_ALPHA: for using (GL_ONE,GL_ONE_MINUS_SRC_ALPHA) blending
+ SOIL_FLAG_INVERT_Y: flip the image vertically
+ SOIL_FLAG_COMPRESS_TO_DXT: if the card can display them, will convert RGB to DXT1, RGBA to DXT5
+ SOIL_FLAG_DDS_LOAD_DIRECT: will load DDS files directly without _ANY_ additional processing
+ SOIL_FLAG_NTSC_SAFE_RGB: clamps RGB components to the range [16,235]
+ SOIL_FLAG_CoCg_Y: Google YCoCg; RGB=>CoYCg, RGBA=>CoCgAY
+ SOIL_FLAG_TEXTURE_RECTANGE: uses ARB_texture_rectangle ; pixel indexed & no repeat or MIPmaps or cubemaps
+**/
+enum
+{
+ SOIL_FLAG_POWER_OF_TWO = 1,
+ SOIL_FLAG_MIPMAPS = 2,
+ SOIL_FLAG_TEXTURE_REPEATS = 4,
+ SOIL_FLAG_MULTIPLY_ALPHA = 8,
+ SOIL_FLAG_INVERT_Y = 16,
+ SOIL_FLAG_COMPRESS_TO_DXT = 32,
+ SOIL_FLAG_DDS_LOAD_DIRECT = 64,
+ SOIL_FLAG_NTSC_SAFE_RGB = 128,
+ SOIL_FLAG_CoCg_Y = 256,
+ SOIL_FLAG_TEXTURE_RECTANGLE = 512
+};
+
+/**
+ The types of images that may be saved.
+ (TGA supports uncompressed RGB / RGBA)
+ (BMP supports uncompressed RGB)
+ (DDS supports DXT1 and DXT5)
+**/
+enum
+{
+ SOIL_SAVE_TYPE_TGA = 0,
+ SOIL_SAVE_TYPE_BMP = 1,
+ SOIL_SAVE_TYPE_DDS = 2
+};
+
+/**
+ Defines the order of faces in a DDS cubemap.
+ I recommend that you use the same order in single
+ image cubemap files, so they will be interchangeable
+ with DDS cubemaps when using SOIL.
+**/
+#define SOIL_DDS_CUBEMAP_FACE_ORDER "EWUDNS"
+
+/**
+ The types of internal fake HDR representations
+
+ SOIL_HDR_RGBE: RGB * pow( 2.0, A - 128.0 )
+ SOIL_HDR_RGBdivA: RGB / A
+ SOIL_HDR_RGBdivA2: RGB / (A*A)
+**/
+enum
+{
+ SOIL_HDR_RGBE = 0,
+ SOIL_HDR_RGBdivA = 1,
+ SOIL_HDR_RGBdivA2 = 2
+};
+
+/**
+ Loads an image from disk into an OpenGL texture.
+ \param filename the name of the file to upload as a texture
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_texture
+ (
+ const char *filename,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads 6 images from disk into an OpenGL cubemap texture.
+ \param x_pos_file the name of the file to upload as the +x cube face
+ \param x_neg_file the name of the file to upload as the -x cube face
+ \param y_pos_file the name of the file to upload as the +y cube face
+ \param y_neg_file the name of the file to upload as the -y cube face
+ \param z_pos_file the name of the file to upload as the +z cube face
+ \param z_neg_file the name of the file to upload as the -z cube face
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_cubemap
+ (
+ const char *x_pos_file,
+ const char *x_neg_file,
+ const char *y_pos_file,
+ const char *y_neg_file,
+ const char *z_pos_file,
+ const char *z_neg_file,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads 1 image from disk and splits it into an OpenGL cubemap texture.
+ \param filename the name of the file to upload as a texture
+ \param face_order the order of the faces in the file, any combination of NSWEUD, for North, South, Up, etc.
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_single_cubemap
+ (
+ const char *filename,
+ const char face_order[6],
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads an HDR image from disk into an OpenGL texture.
+ \param filename the name of the file to upload as a texture
+ \param fake_HDR_format SOIL_HDR_RGBE, SOIL_HDR_RGBdivA, SOIL_HDR_RGBdivA2
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_HDR_texture
+ (
+ const char *filename,
+ int fake_HDR_format,
+ int rescale_to_max,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads an image from RAM into an OpenGL texture.
+ \param buffer the image data in RAM just as if it were still in a file
+ \param buffer_length the size of the buffer in bytes
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_texture_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads 6 images from memory into an OpenGL cubemap texture.
+ \param x_pos_buffer the image data in RAM to upload as the +x cube face
+ \param x_pos_buffer_length the size of the above buffer
+ \param x_neg_buffer the image data in RAM to upload as the +x cube face
+ \param x_neg_buffer_length the size of the above buffer
+ \param y_pos_buffer the image data in RAM to upload as the +x cube face
+ \param y_pos_buffer_length the size of the above buffer
+ \param y_neg_buffer the image data in RAM to upload as the +x cube face
+ \param y_neg_buffer_length the size of the above buffer
+ \param z_pos_buffer the image data in RAM to upload as the +x cube face
+ \param z_pos_buffer_length the size of the above buffer
+ \param z_neg_buffer the image data in RAM to upload as the +x cube face
+ \param z_neg_buffer_length the size of the above buffer
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_cubemap_from_memory
+ (
+ const unsigned char *const x_pos_buffer,
+ int x_pos_buffer_length,
+ const unsigned char *const x_neg_buffer,
+ int x_neg_buffer_length,
+ const unsigned char *const y_pos_buffer,
+ int y_pos_buffer_length,
+ const unsigned char *const y_neg_buffer,
+ int y_neg_buffer_length,
+ const unsigned char *const z_pos_buffer,
+ int z_pos_buffer_length,
+ const unsigned char *const z_neg_buffer,
+ int z_neg_buffer_length,
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Loads 1 image from RAM and splits it into an OpenGL cubemap texture.
+ \param buffer the image data in RAM just as if it were still in a file
+ \param buffer_length the size of the buffer in bytes
+ \param face_order the order of the faces in the file, any combination of NSWEUD, for North, South, Up, etc.
+ \param force_channels 0-image format, 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_load_OGL_single_cubemap_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ const char face_order[6],
+ int force_channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Creates a 2D OpenGL texture from raw image data. Note that the raw data is
+ _NOT_ freed after the upload (so the user can load various versions).
+ \param data the raw data to be uploaded as an OpenGL texture
+ \param width the width of the image in pixels
+ \param height the height of the image in pixels
+ \param channels the number of channels: 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_create_OGL_texture
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Creates an OpenGL cubemap texture by splitting up 1 image into 6 parts.
+ \param data the raw data to be uploaded as an OpenGL texture
+ \param width the width of the image in pixels
+ \param height the height of the image in pixels
+ \param channels the number of channels: 1-luminous, 2-luminous/alpha, 3-RGB, 4-RGBA
+ \param face_order the order of the faces in the file, and combination of NSWEUD, for North, South, Up, etc.
+ \param reuse_texture_ID 0-generate a new texture ID, otherwise reuse the texture ID (overwriting the old texture)
+ \param flags can be any of SOIL_FLAG_POWER_OF_TWO | SOIL_FLAG_MIPMAPS | SOIL_FLAG_TEXTURE_REPEATS | SOIL_FLAG_MULTIPLY_ALPHA | SOIL_FLAG_INVERT_Y | SOIL_FLAG_COMPRESS_TO_DXT | SOIL_FLAG_DDS_LOAD_DIRECT
+ \return 0-failed, otherwise returns the OpenGL texture handle
+**/
+unsigned int
+ SOIL_create_OGL_single_cubemap
+ (
+ const unsigned char *const data,
+ int width, int height, int channels,
+ const char face_order[6],
+ unsigned int reuse_texture_ID,
+ unsigned int flags
+ );
+
+/**
+ Captures the OpenGL window (RGB) and saves it to disk
+ \return 0 if it failed, otherwise returns 1
+**/
+int
+ SOIL_save_screenshot
+ (
+ const char *filename,
+ int image_type,
+ int x, int y,
+ int width, int height
+ );
+
+/**
+ Loads an image from disk into an array of unsigned chars.
+ Note that *channels return the original channel count of the
+ image. If force_channels was other than SOIL_LOAD_AUTO,
+ the resulting image has force_channels, but *channels may be
+ different (if the original image had a different channel
+ count).
+ \return 0 if failed, otherwise returns 1
+**/
+unsigned char*
+ SOIL_load_image
+ (
+ const char *filename,
+ int *width, int *height, int *channels,
+ int force_channels
+ );
+
+/**
+ Loads an image from memory into an array of unsigned chars.
+ Note that *channels return the original channel count of the
+ image. If force_channels was other than SOIL_LOAD_AUTO,
+ the resulting image has force_channels, but *channels may be
+ different (if the original image had a different channel
+ count).
+ \return 0 if failed, otherwise returns 1
+**/
+unsigned char*
+ SOIL_load_image_from_memory
+ (
+ const unsigned char *const buffer,
+ int buffer_length,
+ int *width, int *height, int *channels,
+ int force_channels
+ );
+
+/**
+ Saves an image from an array of unsigned chars (RGBA) to disk
+ \return 0 if failed, otherwise returns 1
+**/
+int
+ SOIL_save_image
+ (
+ const char *filename,
+ int image_type,
+ int width, int height, int channels,
+ const unsigned char *const data
+ );
+
+/**
+ Frees the image data (note, this is just C's "free()"...this function is
+ present mostly so C++ programmers don't forget to use "free()" and call
+ "delete []" instead [8^)
+**/
+void
+ SOIL_free_image_data
+ (
+ unsigned char *img_data
+ );
+
+/**
+ This function resturn a pointer to a string describing the last thing
+ that happened inside SOIL. It can be used to determine why an image
+ failed to load.
+**/
+const char*
+ SOIL_last_result
+ (
+ void
+ );
+
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* HEADER_SIMPLE_OPENGL_IMAGE_LIBRARY */
diff --git a/cw 9/src/SOIL/image_DXT.c b/cw 9/src/SOIL/image_DXT.c
new file mode 100644
index 0000000..4206a1b
--- /dev/null
+++ b/cw 9/src/SOIL/image_DXT.c
@@ -0,0 +1,632 @@
+/*
+ Jonathan Dummer
+ 2007-07-31-10.32
+
+ simple DXT compression / decompression code
+
+ public domain
+*/
+
+#include "image_DXT.h"
+#include
+#include
+#include
+#include
+
+/* set this =1 if you want to use the covarince matrix method...
+ which is better than my method of using standard deviations
+ overall, except on the infintesimal chance that the power
+ method fails for finding the largest eigenvector */
+#define USE_COV_MAT 1
+
+/********* Function Prototypes *********/
+/*
+ Takes a 4x4 block of pixels and compresses it into 8 bytes
+ in DXT1 format (color only, no alpha). Speed is valued
+ over prettyness, at least for now.
+*/
+void compress_DDS_color_block(
+ int channels,
+ const unsigned char *const uncompressed,
+ unsigned char compressed[8] );
+/*
+ Takes a 4x4 block of pixels and compresses the alpha
+ component it into 8 bytes for use in DXT5 DDS files.
+ Speed is valued over prettyness, at least for now.
+*/
+void compress_DDS_alpha_block(
+ const unsigned char *const uncompressed,
+ unsigned char compressed[8] );
+
+/********* Actual Exposed Functions *********/
+int
+ save_image_as_DDS
+ (
+ const char *filename,
+ int width, int height, int channels,
+ const unsigned char *const data
+ )
+{
+ /* variables */
+ FILE *fout;
+ unsigned char *DDS_data;
+ DDS_header header;
+ int DDS_size;
+ /* error check */
+ if( (NULL == filename) ||
+ (width < 1) || (height < 1) ||
+ (channels < 1) || (channels > 4) ||
+ (data == NULL ) )
+ {
+ return 0;
+ }
+ /* Convert the image */
+ if( (channels & 1) == 1 )
+ {
+ /* no alpha, just use DXT1 */
+ DDS_data = convert_image_to_DXT1( data, width, height, channels, &DDS_size );
+ } else
+ {
+ /* has alpha, so use DXT5 */
+ DDS_data = convert_image_to_DXT5( data, width, height, channels, &DDS_size );
+ }
+ /* save it */
+ memset( &header, 0, sizeof( DDS_header ) );
+ header.dwMagic = ('D' << 0) | ('D' << 8) | ('S' << 16) | (' ' << 24);
+ header.dwSize = 124;
+ header.dwFlags = DDSD_CAPS | DDSD_HEIGHT | DDSD_WIDTH | DDSD_PIXELFORMAT | DDSD_LINEARSIZE;
+ header.dwWidth = width;
+ header.dwHeight = height;
+ header.dwPitchOrLinearSize = DDS_size;
+ header.sPixelFormat.dwSize = 32;
+ header.sPixelFormat.dwFlags = DDPF_FOURCC;
+ if( (channels & 1) == 1 )
+ {
+ header.sPixelFormat.dwFourCC = ('D' << 0) | ('X' << 8) | ('T' << 16) | ('1' << 24);
+ } else
+ {
+ header.sPixelFormat.dwFourCC = ('D' << 0) | ('X' << 8) | ('T' << 16) | ('5' << 24);
+ }
+ header.sCaps.dwCaps1 = DDSCAPS_TEXTURE;
+ /* write it out */
+ fout = fopen( filename, "wb");
+ fwrite( &header, sizeof( DDS_header ), 1, fout );
+ fwrite( DDS_data, 1, DDS_size, fout );
+ fclose( fout );
+ /* done */
+ free( DDS_data );
+ return 1;
+}
+
+unsigned char* convert_image_to_DXT1(
+ const unsigned char *const uncompressed,
+ int width, int height, int channels,
+ int *out_size )
+{
+ unsigned char *compressed;
+ int i, j, x, y;
+ unsigned char ublock[16*3];
+ unsigned char cblock[8];
+ int index = 0, chan_step = 1;
+ int block_count = 0;
+ /* error check */
+ *out_size = 0;
+ if( (width < 1) || (height < 1) ||
+ (NULL == uncompressed) ||
+ (channels < 1) || (channels > 4) )
+ {
+ return NULL;
+ }
+ /* for channels == 1 or 2, I do not step forward for R,G,B values */
+ if( channels < 3 )
+ {
+ chan_step = 0;
+ }
+ /* get the RAM for the compressed image
+ (8 bytes per 4x4 pixel block) */
+ *out_size = ((width+3) >> 2) * ((height+3) >> 2) * 8;
+ compressed = (unsigned char*)malloc( *out_size );
+ /* go through each block */
+ for( j = 0; j < height; j += 4 )
+ {
+ for( i = 0; i < width; i += 4 )
+ {
+ /* copy this block into a new one */
+ int idx = 0;
+ int mx = 4, my = 4;
+ if( j+4 >= height )
+ {
+ my = height - j;
+ }
+ if( i+4 >= width )
+ {
+ mx = width - i;
+ }
+ for( y = 0; y < my; ++y )
+ {
+ for( x = 0; x < mx; ++x )
+ {
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels];
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels+chan_step];
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels+chan_step+chan_step];
+ }
+ for( x = mx; x < 4; ++x )
+ {
+ ublock[idx++] = ublock[0];
+ ublock[idx++] = ublock[1];
+ ublock[idx++] = ublock[2];
+ }
+ }
+ for( y = my; y < 4; ++y )
+ {
+ for( x = 0; x < 4; ++x )
+ {
+ ublock[idx++] = ublock[0];
+ ublock[idx++] = ublock[1];
+ ublock[idx++] = ublock[2];
+ }
+ }
+ /* compress the block */
+ ++block_count;
+ compress_DDS_color_block( 3, ublock, cblock );
+ /* copy the data from the block into the main block */
+ for( x = 0; x < 8; ++x )
+ {
+ compressed[index++] = cblock[x];
+ }
+ }
+ }
+ return compressed;
+}
+
+unsigned char* convert_image_to_DXT5(
+ const unsigned char *const uncompressed,
+ int width, int height, int channels,
+ int *out_size )
+{
+ unsigned char *compressed;
+ int i, j, x, y;
+ unsigned char ublock[16*4];
+ unsigned char cblock[8];
+ int index = 0, chan_step = 1;
+ int block_count = 0, has_alpha;
+ /* error check */
+ *out_size = 0;
+ if( (width < 1) || (height < 1) ||
+ (NULL == uncompressed) ||
+ (channels < 1) || ( channels > 4) )
+ {
+ return NULL;
+ }
+ /* for channels == 1 or 2, I do not step forward for R,G,B vales */
+ if( channels < 3 )
+ {
+ chan_step = 0;
+ }
+ /* # channels = 1 or 3 have no alpha, 2 & 4 do have alpha */
+ has_alpha = 1 - (channels & 1);
+ /* get the RAM for the compressed image
+ (16 bytes per 4x4 pixel block) */
+ *out_size = ((width+3) >> 2) * ((height+3) >> 2) * 16;
+ compressed = (unsigned char*)malloc( *out_size );
+ /* go through each block */
+ for( j = 0; j < height; j += 4 )
+ {
+ for( i = 0; i < width; i += 4 )
+ {
+ /* local variables, and my block counter */
+ int idx = 0;
+ int mx = 4, my = 4;
+ if( j+4 >= height )
+ {
+ my = height - j;
+ }
+ if( i+4 >= width )
+ {
+ mx = width - i;
+ }
+ for( y = 0; y < my; ++y )
+ {
+ for( x = 0; x < mx; ++x )
+ {
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels];
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels+chan_step];
+ ublock[idx++] = uncompressed[(j+y)*width*channels+(i+x)*channels+chan_step+chan_step];
+ ublock[idx++] =
+ has_alpha * uncompressed[(j+y)*width*channels+(i+x)*channels+channels-1]
+ + (1-has_alpha)*255;
+ }
+ for( x = mx; x < 4; ++x )
+ {
+ ublock[idx++] = ublock[0];
+ ublock[idx++] = ublock[1];
+ ublock[idx++] = ublock[2];
+ ublock[idx++] = ublock[3];
+ }
+ }
+ for( y = my; y < 4; ++y )
+ {
+ for( x = 0; x < 4; ++x )
+ {
+ ublock[idx++] = ublock[0];
+ ublock[idx++] = ublock[1];
+ ublock[idx++] = ublock[2];
+ ublock[idx++] = ublock[3];
+ }
+ }
+ /* now compress the alpha block */
+ compress_DDS_alpha_block( ublock, cblock );
+ /* copy the data from the compressed alpha block into the main buffer */
+ for( x = 0; x < 8; ++x )
+ {
+ compressed[index++] = cblock[x];
+ }
+ /* then compress the color block */
+ ++block_count;
+ compress_DDS_color_block( 4, ublock, cblock );
+ /* copy the data from the compressed color block into the main buffer */
+ for( x = 0; x < 8; ++x )
+ {
+ compressed[index++] = cblock[x];
+ }
+ }
+ }
+ return compressed;
+}
+
+/********* Helper Functions *********/
+int convert_bit_range( int c, int from_bits, int to_bits )
+{
+ int b = (1 << (from_bits - 1)) + c * ((1 << to_bits) - 1);
+ return (b + (b >> from_bits)) >> from_bits;
+}
+
+int rgb_to_565( int r, int g, int b )
+{
+ return
+ (convert_bit_range( r, 8, 5 ) << 11) |
+ (convert_bit_range( g, 8, 6 ) << 05) |
+ (convert_bit_range( b, 8, 5 ) << 00);
+}
+
+void rgb_888_from_565( unsigned int c, int *r, int *g, int *b )
+{
+ *r = convert_bit_range( (c >> 11) & 31, 5, 8 );
+ *g = convert_bit_range( (c >> 05) & 63, 6, 8 );
+ *b = convert_bit_range( (c >> 00) & 31, 5, 8 );
+}
+
+void compute_color_line_STDEV(
+ const unsigned char *const uncompressed,
+ int channels,
+ float point[3], float direction[3] )
+{
+ const float inv_16 = 1.0f / 16.0f;
+ int i;
+ float sum_r = 0.0f, sum_g = 0.0f, sum_b = 0.0f;
+ float sum_rr = 0.0f, sum_gg = 0.0f, sum_bb = 0.0f;
+ float sum_rg = 0.0f, sum_rb = 0.0f, sum_gb = 0.0f;
+ /* calculate all data needed for the covariance matrix
+ ( to compare with _rygdxt code) */
+ for( i = 0; i < 16*channels; i += channels )
+ {
+ sum_r += uncompressed[i+0];
+ sum_rr += uncompressed[i+0] * uncompressed[i+0];
+ sum_g += uncompressed[i+1];
+ sum_gg += uncompressed[i+1] * uncompressed[i+1];
+ sum_b += uncompressed[i+2];
+ sum_bb += uncompressed[i+2] * uncompressed[i+2];
+ sum_rg += uncompressed[i+0] * uncompressed[i+1];
+ sum_rb += uncompressed[i+0] * uncompressed[i+2];
+ sum_gb += uncompressed[i+1] * uncompressed[i+2];
+ }
+ /* convert the sums to averages */
+ sum_r *= inv_16;
+ sum_g *= inv_16;
+ sum_b *= inv_16;
+ /* and convert the squares to the squares of the value - avg_value */
+ sum_rr -= 16.0f * sum_r * sum_r;
+ sum_gg -= 16.0f * sum_g * sum_g;
+ sum_bb -= 16.0f * sum_b * sum_b;
+ sum_rg -= 16.0f * sum_r * sum_g;
+ sum_rb -= 16.0f * sum_r * sum_b;
+ sum_gb -= 16.0f * sum_g * sum_b;
+ /* the point on the color line is the average */
+ point[0] = sum_r;
+ point[1] = sum_g;
+ point[2] = sum_b;
+ #if USE_COV_MAT
+ /*
+ The following idea was from ryg.
+ (https://mollyrocket.com/forums/viewtopic.php?t=392)
+ The method worked great (less RMSE than mine) most of
+ the time, but had some issues handling some simple
+ boundary cases, like full green next to full red,
+ which would generate a covariance matrix like this:
+
+ | 1 -1 0 |
+ | -1 1 0 |
+ | 0 0 0 |
+
+ For a given starting vector, the power method can
+ generate all zeros! So no starting with {1,1,1}
+ as I was doing! This kind of error is still a
+ slight posibillity, but will be very rare.
+ */
+ /* use the covariance matrix directly
+ (1st iteration, don't use all 1.0 values!) */
+ sum_r = 1.0f;
+ sum_g = 2.718281828f;
+ sum_b = 3.141592654f;
+ direction[0] = sum_r*sum_rr + sum_g*sum_rg + sum_b*sum_rb;
+ direction[1] = sum_r*sum_rg + sum_g*sum_gg + sum_b*sum_gb;
+ direction[2] = sum_r*sum_rb + sum_g*sum_gb + sum_b*sum_bb;
+ /* 2nd iteration, use results from the 1st guy */
+ sum_r = direction[0];
+ sum_g = direction[1];
+ sum_b = direction[2];
+ direction[0] = sum_r*sum_rr + sum_g*sum_rg + sum_b*sum_rb;
+ direction[1] = sum_r*sum_rg + sum_g*sum_gg + sum_b*sum_gb;
+ direction[2] = sum_r*sum_rb + sum_g*sum_gb + sum_b*sum_bb;
+ /* 3rd iteration, use results from the 2nd guy */
+ sum_r = direction[0];
+ sum_g = direction[1];
+ sum_b = direction[2];
+ direction[0] = sum_r*sum_rr + sum_g*sum_rg + sum_b*sum_rb;
+ direction[1] = sum_r*sum_rg + sum_g*sum_gg + sum_b*sum_gb;
+ direction[2] = sum_r*sum_rb + sum_g*sum_gb + sum_b*sum_bb;
+ #else
+ /* use my standard deviation method
+ (very robust, a tiny bit slower and less accurate) */
+ direction[0] = sqrt( sum_rr );
+ direction[1] = sqrt( sum_gg );
+ direction[2] = sqrt( sum_bb );
+ /* which has a greater component */
+ if( sum_gg > sum_rr )
+ {
+ /* green has greater component, so base the other signs off of green */
+ if( sum_rg < 0.0f )
+ {
+ direction[0] = -direction[0];
+ }
+ if( sum_gb < 0.0f )
+ {
+ direction[2] = -direction[2];
+ }
+ } else
+ {
+ /* red has a greater component */
+ if( sum_rg < 0.0f )
+ {
+ direction[1] = -direction[1];
+ }
+ if( sum_rb < 0.0f )
+ {
+ direction[2] = -direction[2];
+ }
+ }
+ #endif
+}
+
+void LSE_master_colors_max_min(
+ int *cmax, int *cmin,
+ int channels,
+ const unsigned char *const uncompressed )
+{
+ int i, j;
+ /* the master colors */
+ int c0[3], c1[3];
+ /* used for fitting the line */
+ float sum_x[] = { 0.0f, 0.0f, 0.0f };
+ float sum_x2[] = { 0.0f, 0.0f, 0.0f };
+ float dot_max = 1.0f, dot_min = -1.0f;
+ float vec_len2 = 0.0f;
+ float dot;
+ /* error check */
+ if( (channels < 3) || (channels > 4) )
+ {
+ return;
+ }
+ compute_color_line_STDEV( uncompressed, channels, sum_x, sum_x2 );
+ vec_len2 = 1.0f / ( 0.00001f +
+ sum_x2[0]*sum_x2[0] + sum_x2[1]*sum_x2[1] + sum_x2[2]*sum_x2[2] );
+ /* finding the max and min vector values */
+ dot_max =
+ (
+ sum_x2[0] * uncompressed[0] +
+ sum_x2[1] * uncompressed[1] +
+ sum_x2[2] * uncompressed[2]
+ );
+ dot_min = dot_max;
+ for( i = 1; i < 16; ++i )
+ {
+ dot =
+ (
+ sum_x2[0] * uncompressed[i*channels+0] +
+ sum_x2[1] * uncompressed[i*channels+1] +
+ sum_x2[2] * uncompressed[i*channels+2]
+ );
+ if( dot < dot_min )
+ {
+ dot_min = dot;
+ } else if( dot > dot_max )
+ {
+ dot_max = dot;
+ }
+ }
+ /* and the offset (from the average location) */
+ dot = sum_x2[0]*sum_x[0] + sum_x2[1]*sum_x[1] + sum_x2[2]*sum_x[2];
+ dot_min -= dot;
+ dot_max -= dot;
+ /* post multiply by the scaling factor */
+ dot_min *= vec_len2;
+ dot_max *= vec_len2;
+ /* OK, build the master colors */
+ for( i = 0; i < 3; ++i )
+ {
+ /* color 0 */
+ c0[i] = (int)(0.5f + sum_x[i] + dot_max * sum_x2[i]);
+ if( c0[i] < 0 )
+ {
+ c0[i] = 0;
+ } else if( c0[i] > 255 )
+ {
+ c0[i] = 255;
+ }
+ /* color 1 */
+ c1[i] = (int)(0.5f + sum_x[i] + dot_min * sum_x2[i]);
+ if( c1[i] < 0 )
+ {
+ c1[i] = 0;
+ } else if( c1[i] > 255 )
+ {
+ c1[i] = 255;
+ }
+ }
+ /* down_sample (with rounding?) */
+ i = rgb_to_565( c0[0], c0[1], c0[2] );
+ j = rgb_to_565( c1[0], c1[1], c1[2] );
+ if( i > j )
+ {
+ *cmax = i;
+ *cmin = j;
+ } else
+ {
+ *cmax = j;
+ *cmin = i;
+ }
+}
+
+void
+ compress_DDS_color_block
+ (
+ int channels,
+ const unsigned char *const uncompressed,
+ unsigned char compressed[8]
+ )
+{
+ /* variables */
+ int i;
+ int next_bit;
+ int enc_c0, enc_c1;
+ int c0[4], c1[4];
+ float color_line[] = { 0.0f, 0.0f, 0.0f, 0.0f };
+ float vec_len2 = 0.0f, dot_offset = 0.0f;
+ /* stupid order */
+ int swizzle4[] = { 0, 2, 3, 1 };
+ /* get the master colors */
+ LSE_master_colors_max_min( &enc_c0, &enc_c1, channels, uncompressed );
+ /* store the 565 color 0 and color 1 */
+ compressed[0] = (enc_c0 >> 0) & 255;
+ compressed[1] = (enc_c0 >> 8) & 255;
+ compressed[2] = (enc_c1 >> 0) & 255;
+ compressed[3] = (enc_c1 >> 8) & 255;
+ /* zero out the compressed data */
+ compressed[4] = 0;
+ compressed[5] = 0;
+ compressed[6] = 0;
+ compressed[7] = 0;
+ /* reconstitute the master color vectors */
+ rgb_888_from_565( enc_c0, &c0[0], &c0[1], &c0[2] );
+ rgb_888_from_565( enc_c1, &c1[0], &c1[1], &c1[2] );
+ /* the new vector */
+ vec_len2 = 0.0f;
+ for( i = 0; i < 3; ++i )
+ {
+ color_line[i] = (float)(c1[i] - c0[i]);
+ vec_len2 += color_line[i] * color_line[i];
+ }
+ if( vec_len2 > 0.0f )
+ {
+ vec_len2 = 1.0f / vec_len2;
+ }
+ /* pre-proform the scaling */
+ color_line[0] *= vec_len2;
+ color_line[1] *= vec_len2;
+ color_line[2] *= vec_len2;
+ /* compute the offset (constant) portion of the dot product */
+ dot_offset = color_line[0]*c0[0] + color_line[1]*c0[1] + color_line[2]*c0[2];
+ /* store the rest of the bits */
+ next_bit = 8*4;
+ for( i = 0; i < 16; ++i )
+ {
+ /* find the dot product of this color, to place it on the line
+ (should be [-1,1]) */
+ int next_value = 0;
+ float dot_product =
+ color_line[0] * uncompressed[i*channels+0] +
+ color_line[1] * uncompressed[i*channels+1] +
+ color_line[2] * uncompressed[i*channels+2] -
+ dot_offset;
+ /* map to [0,3] */
+ next_value = (int)( dot_product * 3.0f + 0.5f );
+ if( next_value > 3 )
+ {
+ next_value = 3;
+ } else if( next_value < 0 )
+ {
+ next_value = 0;
+ }
+ /* OK, store this value */
+ compressed[next_bit >> 3] |= swizzle4[ next_value ] << (next_bit & 7);
+ next_bit += 2;
+ }
+ /* done compressing to DXT1 */
+}
+
+void
+ compress_DDS_alpha_block
+ (
+ const unsigned char *const uncompressed,
+ unsigned char compressed[8]
+ )
+{
+ /* variables */
+ int i;
+ int next_bit;
+ int a0, a1;
+ float scale_me;
+ /* stupid order */
+ int swizzle8[] = { 1, 7, 6, 5, 4, 3, 2, 0 };
+ /* get the alpha limits (a0 > a1) */
+ a0 = a1 = uncompressed[3];
+ for( i = 4+3; i < 16*4; i += 4 )
+ {
+ if( uncompressed[i] > a0 )
+ {
+ a0 = uncompressed[i];
+ } else if( uncompressed[i] < a1 )
+ {
+ a1 = uncompressed[i];
+ }
+ }
+ /* store those limits, and zero the rest of the compressed dataset */
+ compressed[0] = a0;
+ compressed[1] = a1;
+ /* zero out the compressed data */
+ compressed[2] = 0;
+ compressed[3] = 0;
+ compressed[4] = 0;
+ compressed[5] = 0;
+ compressed[6] = 0;
+ compressed[7] = 0;
+ /* store the all of the alpha values */
+ next_bit = 8*2;
+ scale_me = 7.9999f / (a0 - a1);
+ for( i = 3; i < 16*4; i += 4 )
+ {
+ /* convert this alpha value to a 3 bit number */
+ int svalue;
+ int value = (int)((uncompressed[i] - a1) * scale_me);
+ svalue = swizzle8[ value&7 ];
+ /* OK, store this value, start with the 1st byte */
+ compressed[next_bit >> 3] |= svalue << (next_bit & 7);
+ if( (next_bit & 7) > 5 )
+ {
+ /* spans 2 bytes, fill in the start of the 2nd byte */
+ compressed[1 + (next_bit >> 3)] |= svalue >> (8 - (next_bit & 7) );
+ }
+ next_bit += 3;
+ }
+ /* done compressing to DXT1 */
+}
diff --git a/cw 9/src/SOIL/image_DXT.h b/cw 9/src/SOIL/image_DXT.h
new file mode 100644
index 0000000..75f604f
--- /dev/null
+++ b/cw 9/src/SOIL/image_DXT.h
@@ -0,0 +1,123 @@
+/*
+ Jonathan Dummer
+ 2007-07-31-10.32
+
+ simple DXT compression / decompression code
+
+ public domain
+*/
+
+#ifndef HEADER_IMAGE_DXT
+#define HEADER_IMAGE_DXT
+
+/**
+ Converts an image from an array of unsigned chars (RGB or RGBA) to
+ DXT1 or DXT5, then saves the converted image to disk.
+ \return 0 if failed, otherwise returns 1
+**/
+int
+save_image_as_DDS
+(
+ const char *filename,
+ int width, int height, int channels,
+ const unsigned char *const data
+);
+
+/**
+ take an image and convert it to DXT1 (no alpha)
+**/
+unsigned char*
+convert_image_to_DXT1
+(
+ const unsigned char *const uncompressed,
+ int width, int height, int channels,
+ int *out_size
+);
+
+/**
+ take an image and convert it to DXT5 (with alpha)
+**/
+unsigned char*
+convert_image_to_DXT5
+(
+ const unsigned char *const uncompressed,
+ int width, int height, int channels,
+ int *out_size
+);
+
+/** A bunch of DirectDraw Surface structures and flags **/
+typedef struct
+{
+ unsigned int dwMagic;
+ unsigned int dwSize;
+ unsigned int dwFlags;
+ unsigned int dwHeight;
+ unsigned int dwWidth;
+ unsigned int dwPitchOrLinearSize;
+ unsigned int dwDepth;
+ unsigned int dwMipMapCount;
+ unsigned int dwReserved1[ 11 ];
+
+ /* DDPIXELFORMAT */
+ struct
+ {
+ unsigned int dwSize;
+ unsigned int dwFlags;
+ unsigned int dwFourCC;
+ unsigned int dwRGBBitCount;
+ unsigned int dwRBitMask;
+ unsigned int dwGBitMask;
+ unsigned int dwBBitMask;
+ unsigned int dwAlphaBitMask;
+ }
+ sPixelFormat;
+
+ /* DDCAPS2 */
+ struct
+ {
+ unsigned int dwCaps1;
+ unsigned int dwCaps2;
+ unsigned int dwDDSX;
+ unsigned int dwReserved;
+ }
+ sCaps;
+ unsigned int dwReserved2;
+}
+DDS_header ;
+
+/* the following constants were copied directly off the MSDN website */
+
+/* The dwFlags member of the original DDSURFACEDESC2 structure
+ can be set to one or more of the following values. */
+#define DDSD_CAPS 0x00000001
+#define DDSD_HEIGHT 0x00000002
+#define DDSD_WIDTH 0x00000004
+#define DDSD_PITCH 0x00000008
+#define DDSD_PIXELFORMAT 0x00001000
+#define DDSD_MIPMAPCOUNT 0x00020000
+#define DDSD_LINEARSIZE 0x00080000
+#define DDSD_DEPTH 0x00800000
+
+/* DirectDraw Pixel Format */
+#define DDPF_ALPHAPIXELS 0x00000001
+#define DDPF_FOURCC 0x00000004
+#define DDPF_RGB 0x00000040
+
+/* The dwCaps1 member of the DDSCAPS2 structure can be
+ set to one or more of the following values. */
+#define DDSCAPS_COMPLEX 0x00000008
+#define DDSCAPS_TEXTURE 0x00001000
+#define DDSCAPS_MIPMAP 0x00400000
+
+/* The dwCaps2 member of the DDSCAPS2 structure can be
+ set to one or more of the following values. */
+#define DDSCAPS2_CUBEMAP 0x00000200
+#define DDSCAPS2_CUBEMAP_POSITIVEX 0x00000400
+#define DDSCAPS2_CUBEMAP_NEGATIVEX 0x00000800
+#define DDSCAPS2_CUBEMAP_POSITIVEY 0x00001000
+#define DDSCAPS2_CUBEMAP_NEGATIVEY 0x00002000
+#define DDSCAPS2_CUBEMAP_POSITIVEZ 0x00004000
+#define DDSCAPS2_CUBEMAP_NEGATIVEZ 0x00008000
+#define DDSCAPS2_VOLUME 0x00200000
+
+#endif /* HEADER_IMAGE_DXT */
diff --git a/cw 9/src/SOIL/image_helper.c b/cw 9/src/SOIL/image_helper.c
new file mode 100644
index 0000000..d22340f
--- /dev/null
+++ b/cw 9/src/SOIL/image_helper.c
@@ -0,0 +1,435 @@
+/*
+ Jonathan Dummer
+
+ image helper functions
+
+ MIT license
+*/
+
+#include "image_helper.h"
+#include
+#include
+
+/* Upscaling the image uses simple bilinear interpolation */
+int
+ up_scale_image
+ (
+ const unsigned char* const orig,
+ int width, int height, int channels,
+ unsigned char* resampled,
+ int resampled_width, int resampled_height
+ )
+{
+ float dx, dy;
+ int x, y, c;
+
+ /* error(s) check */
+ if ( (width < 1) || (height < 1) ||
+ (resampled_width < 2) || (resampled_height < 2) ||
+ (channels < 1) ||
+ (NULL == orig) || (NULL == resampled) )
+ {
+ /* signify badness */
+ return 0;
+ }
+ /*
+ for each given pixel in the new map, find the exact location
+ from the original map which would contribute to this guy
+ */
+ dx = (width - 1.0f) / (resampled_width - 1.0f);
+ dy = (height - 1.0f) / (resampled_height - 1.0f);
+ for ( y = 0; y < resampled_height; ++y )
+ {
+ /* find the base y index and fractional offset from that */
+ float sampley = y * dy;
+ int inty = (int)sampley;
+ /* if( inty < 0 ) { inty = 0; } else */
+ if( inty > height - 2 ) { inty = height - 2; }
+ sampley -= inty;
+ for ( x = 0; x < resampled_width; ++x )
+ {
+ float samplex = x * dx;
+ int intx = (int)samplex;
+ int base_index;
+ /* find the base x index and fractional offset from that */
+ /* if( intx < 0 ) { intx = 0; } else */
+ if( intx > width - 2 ) { intx = width - 2; }
+ samplex -= intx;
+ /* base index into the original image */
+ base_index = (inty * width + intx) * channels;
+ for ( c = 0; c < channels; ++c )
+ {
+ /* do the sampling */
+ float value = 0.5f;
+ value += orig[base_index]
+ *(1.0f-samplex)*(1.0f-sampley);
+ value += orig[base_index+channels]
+ *(samplex)*(1.0f-sampley);
+ value += orig[base_index+width*channels]
+ *(1.0f-samplex)*(sampley);
+ value += orig[base_index+width*channels+channels]
+ *(samplex)*(sampley);
+ /* move to the next channel */
+ ++base_index;
+ /* save the new value */
+ resampled[y*resampled_width*channels+x*channels+c] =
+ (unsigned char)(value);
+ }
+ }
+ }
+ /* done */
+ return 1;
+}
+
+int
+ mipmap_image
+ (
+ const unsigned char* const orig,
+ int width, int height, int channels,
+ unsigned char* resampled,
+ int block_size_x, int block_size_y
+ )
+{
+ int mip_width, mip_height;
+ int i, j, c;
+
+ /* error check */
+ if( (width < 1) || (height < 1) ||
+ (channels < 1) || (orig == NULL) ||
+ (resampled == NULL) ||
+ (block_size_x < 1) || (block_size_y < 1) )
+ {
+ /* nothing to do */
+ return 0;
+ }
+ mip_width = width / block_size_x;
+ mip_height = height / block_size_y;
+ if( mip_width < 1 )
+ {
+ mip_width = 1;
+ }
+ if( mip_height < 1 )
+ {
+ mip_height = 1;
+ }
+ for( j = 0; j < mip_height; ++j )
+ {
+ for( i = 0; i < mip_width; ++i )
+ {
+ for( c = 0; c < channels; ++c )
+ {
+ const int index = (j*block_size_y)*width*channels + (i*block_size_x)*channels + c;
+ int sum_value;
+ int u,v;
+ int u_block = block_size_x;
+ int v_block = block_size_y;
+ int block_area;
+ /* do a bit of checking so we don't over-run the boundaries
+ (necessary for non-square textures!) */
+ if( block_size_x * (i+1) > width )
+ {
+ u_block = width - i*block_size_y;
+ }
+ if( block_size_y * (j+1) > height )
+ {
+ v_block = height - j*block_size_y;
+ }
+ block_area = u_block*v_block;
+ /* for this pixel, see what the average
+ of all the values in the block are.
+ note: start the sum at the rounding value, not at 0 */
+ sum_value = block_area >> 1;
+ for( v = 0; v < v_block; ++v )
+ for( u = 0; u < u_block; ++u )
+ {
+ sum_value += orig[index + v*width*channels + u*channels];
+ }
+ resampled[j*mip_width*channels + i*channels + c] = sum_value / block_area;
+ }
+ }
+ }
+ return 1;
+}
+
+int
+ scale_image_RGB_to_NTSC_safe
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ )
+{
+ const float scale_lo = 16.0f - 0.499f;
+ const float scale_hi = 235.0f + 0.499f;
+ int i, j;
+ int nc = channels;
+ unsigned char scale_LUT[256];
+ /* error check */
+ if( (width < 1) || (height < 1) ||
+ (channels < 1) || (orig == NULL) )
+ {
+ /* nothing to do */
+ return 0;
+ }
+ /* set up the scaling Look Up Table */
+ for( i = 0; i < 256; ++i )
+ {
+ scale_LUT[i] = (unsigned char)((scale_hi - scale_lo) * i / 255.0f + scale_lo);
+ }
+ /* for channels = 2 or 4, ignore the alpha component */
+ nc -= 1 - (channels & 1);
+ /* OK, go through the image and scale any non-alpha components */
+ for( i = 0; i < width*height*channels; i += channels )
+ {
+ for( j = 0; j < nc; ++j )
+ {
+ orig[i+j] = scale_LUT[orig[i+j]];
+ }
+ }
+ return 1;
+}
+
+unsigned char clamp_byte( int x ) { return ( (x) < 0 ? (0) : ( (x) > 255 ? 255 : (x) ) ); }
+
+/*
+ This function takes the RGB components of the image
+ and converts them into YCoCg. 3 components will be
+ re-ordered to CoYCg (for optimum DXT1 compression),
+ while 4 components will be ordered CoCgAY (for DXT5
+ compression).
+*/
+int
+ convert_RGB_to_YCoCg
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ )
+{
+ int i;
+ /* error check */
+ if( (width < 1) || (height < 1) ||
+ (channels < 3) || (channels > 4) ||
+ (orig == NULL) )
+ {
+ /* nothing to do */
+ return -1;
+ }
+ /* do the conversion */
+ if( channels == 3 )
+ {
+ for( i = 0; i < width*height*3; i += 3 )
+ {
+ int r = orig[i+0];
+ int g = (orig[i+1] + 1) >> 1;
+ int b = orig[i+2];
+ int tmp = (2 + r + b) >> 2;
+ /* Co */
+ orig[i+0] = clamp_byte( 128 + ((r - b + 1) >> 1) );
+ /* Y */
+ orig[i+1] = clamp_byte( g + tmp );
+ /* Cg */
+ orig[i+2] = clamp_byte( 128 + g - tmp );
+ }
+ } else
+ {
+ for( i = 0; i < width*height*4; i += 4 )
+ {
+ int r = orig[i+0];
+ int g = (orig[i+1] + 1) >> 1;
+ int b = orig[i+2];
+ unsigned char a = orig[i+3];
+ int tmp = (2 + r + b) >> 2;
+ /* Co */
+ orig[i+0] = clamp_byte( 128 + ((r - b + 1) >> 1) );
+ /* Cg */
+ orig[i+1] = clamp_byte( 128 + g - tmp );
+ /* Alpha */
+ orig[i+2] = a;
+ /* Y */
+ orig[i+3] = clamp_byte( g + tmp );
+ }
+ }
+ /* done */
+ return 0;
+}
+
+/*
+ This function takes the YCoCg components of the image
+ and converts them into RGB. See above.
+*/
+int
+ convert_YCoCg_to_RGB
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ )
+{
+ int i;
+ /* error check */
+ if( (width < 1) || (height < 1) ||
+ (channels < 3) || (channels > 4) ||
+ (orig == NULL) )
+ {
+ /* nothing to do */
+ return -1;
+ }
+ /* do the conversion */
+ if( channels == 3 )
+ {
+ for( i = 0; i < width*height*3; i += 3 )
+ {
+ int co = orig[i+0] - 128;
+ int y = orig[i+1];
+ int cg = orig[i+2] - 128;
+ /* R */
+ orig[i+0] = clamp_byte( y + co - cg );
+ /* G */
+ orig[i+1] = clamp_byte( y + cg );
+ /* B */
+ orig[i+2] = clamp_byte( y - co - cg );
+ }
+ } else
+ {
+ for( i = 0; i < width*height*4; i += 4 )
+ {
+ int co = orig[i+0] - 128;
+ int cg = orig[i+1] - 128;
+ unsigned char a = orig[i+2];
+ int y = orig[i+3];
+ /* R */
+ orig[i+0] = clamp_byte( y + co - cg );
+ /* G */
+ orig[i+1] = clamp_byte( y + cg );
+ /* B */
+ orig[i+2] = clamp_byte( y - co - cg );
+ /* A */
+ orig[i+3] = a;
+ }
+ }
+ /* done */
+ return 0;
+}
+
+float
+find_max_RGBE
+(
+ unsigned char *image,
+ int width, int height
+)
+{
+ float max_val = 0.0f;
+ unsigned char *img = image;
+ int i, j;
+ for( i = width * height; i > 0; --i )
+ {
+ /* float scale = powf( 2.0f, img[3] - 128.0f ) / 255.0f; */
+ float scale = ldexp( 1.0f / 255.0f, (int)(img[3]) - 128 );
+ for( j = 0; j < 3; ++j )
+ {
+ if( img[j] * scale > max_val )
+ {
+ max_val = img[j] * scale;
+ }
+ }
+ /* next pixel */
+ img += 4;
+ }
+ return max_val;
+}
+
+int
+RGBE_to_RGBdivA
+(
+ unsigned char *image,
+ int width, int height,
+ int rescale_to_max
+)
+{
+ /* local variables */
+ int i, iv;
+ unsigned char *img = image;
+ float scale = 1.0f;
+ /* error check */
+ if( (!image) || (width < 1) || (height < 1) )
+ {
+ return 0;
+ }
+ /* convert (note: no negative numbers, but 0.0 is possible) */
+ if( rescale_to_max )
+ {
+ scale = 255.0f / find_max_RGBE( image, width, height );
+ }
+ for( i = width * height; i > 0; --i )
+ {
+ /* decode this pixel, and find the max */
+ float r,g,b,e, m;
+ /* e = scale * powf( 2.0f, img[3] - 128.0f ) / 255.0f; */
+ e = scale * ldexp( 1.0f / 255.0f, (int)(img[3]) - 128 );
+ r = e * img[0];
+ g = e * img[1];
+ b = e * img[2];
+ m = (r > g) ? r : g;
+ m = (b > m) ? b : m;
+ /* and encode it into RGBdivA */
+ iv = (m != 0.0f) ? (int)(255.0f / m) : 1.0f;
+ iv = (iv < 1) ? 1 : iv;
+ img[3] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * r + 0.5f);
+ img[0] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * g + 0.5f);
+ img[1] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * b + 0.5f);
+ img[2] = (iv > 255) ? 255 : iv;
+ /* and on to the next pixel */
+ img += 4;
+ }
+ return 1;
+}
+
+int
+RGBE_to_RGBdivA2
+(
+ unsigned char *image,
+ int width, int height,
+ int rescale_to_max
+)
+{
+ /* local variables */
+ int i, iv;
+ unsigned char *img = image;
+ float scale = 1.0f;
+ /* error check */
+ if( (!image) || (width < 1) || (height < 1) )
+ {
+ return 0;
+ }
+ /* convert (note: no negative numbers, but 0.0 is possible) */
+ if( rescale_to_max )
+ {
+ scale = 255.0f * 255.0f / find_max_RGBE( image, width, height );
+ }
+ for( i = width * height; i > 0; --i )
+ {
+ /* decode this pixel, and find the max */
+ float r,g,b,e, m;
+ /* e = scale * powf( 2.0f, img[3] - 128.0f ) / 255.0f; */
+ e = scale * ldexp( 1.0f / 255.0f, (int)(img[3]) - 128 );
+ r = e * img[0];
+ g = e * img[1];
+ b = e * img[2];
+ m = (r > g) ? r : g;
+ m = (b > m) ? b : m;
+ /* and encode it into RGBdivA */
+ iv = (m != 0.0f) ? (int)sqrtf( 255.0f * 255.0f / m ) : 1.0f;
+ iv = (iv < 1) ? 1 : iv;
+ img[3] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * img[3] * r / 255.0f + 0.5f);
+ img[0] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * img[3] * g / 255.0f + 0.5f);
+ img[1] = (iv > 255) ? 255 : iv;
+ iv = (int)(img[3] * img[3] * b / 255.0f + 0.5f);
+ img[2] = (iv > 255) ? 255 : iv;
+ /* and on to the next pixel */
+ img += 4;
+ }
+ return 1;
+}
diff --git a/cw 9/src/SOIL/image_helper.h b/cw 9/src/SOIL/image_helper.h
new file mode 100644
index 0000000..3fa2662
--- /dev/null
+++ b/cw 9/src/SOIL/image_helper.h
@@ -0,0 +1,115 @@
+/*
+ Jonathan Dummer
+
+ Image helper functions
+
+ MIT license
+*/
+
+#ifndef HEADER_IMAGE_HELPER
+#define HEADER_IMAGE_HELPER
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ This function upscales an image.
+ Not to be used to create MIPmaps,
+ but to make it square,
+ or to make it a power-of-two sized.
+**/
+int
+ up_scale_image
+ (
+ const unsigned char* const orig,
+ int width, int height, int channels,
+ unsigned char* resampled,
+ int resampled_width, int resampled_height
+ );
+
+/**
+ This function downscales an image.
+ Used for creating MIPmaps,
+ the incoming image should be a
+ power-of-two sized.
+**/
+int
+ mipmap_image
+ (
+ const unsigned char* const orig,
+ int width, int height, int channels,
+ unsigned char* resampled,
+ int block_size_x, int block_size_y
+ );
+
+/**
+ This function takes the RGB components of the image
+ and scales each channel from [0,255] to [16,235].
+ This makes the colors "Safe" for display on NTSC
+ displays. Note that this is _NOT_ a good idea for
+ loading images like normal- or height-maps!
+**/
+int
+ scale_image_RGB_to_NTSC_safe
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ );
+
+/**
+ This function takes the RGB components of the image
+ and converts them into YCoCg. 3 components will be
+ re-ordered to CoYCg (for optimum DXT1 compression),
+ while 4 components will be ordered CoCgAY (for DXT5
+ compression).
+**/
+int
+ convert_RGB_to_YCoCg
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ );
+
+/**
+ This function takes the YCoCg components of the image
+ and converts them into RGB. See above.
+**/
+int
+ convert_YCoCg_to_RGB
+ (
+ unsigned char* orig,
+ int width, int height, int channels
+ );
+
+/**
+ Converts an HDR image from an array
+ of unsigned chars (RGBE) to RGBdivA
+ \return 0 if failed, otherwise returns 1
+**/
+int
+ RGBE_to_RGBdivA
+ (
+ unsigned char *image,
+ int width, int height,
+ int rescale_to_max
+ );
+
+/**
+ Converts an HDR image from an array
+ of unsigned chars (RGBE) to RGBdivA2
+ \return 0 if failed, otherwise returns 1
+**/
+int
+ RGBE_to_RGBdivA2
+ (
+ unsigned char *image,
+ int width, int height,
+ int rescale_to_max
+ );
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* HEADER_IMAGE_HELPER */
diff --git a/cw 9/src/SOIL/stb_image_aug.c b/cw 9/src/SOIL/stb_image_aug.c
new file mode 100644
index 0000000..bb088fc
--- /dev/null
+++ b/cw 9/src/SOIL/stb_image_aug.c
@@ -0,0 +1,3682 @@
+/* stbi-1.16 - public domain JPEG/PNG reader - http://nothings.org/stb_image.c
+ when you control the images you're loading
+
+ 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, no oddball channel decimations)
+ PNG non-interlaced
+ BMP non-1bpp, non-RLE
+ TGA (not sure what subset, if a subset)
+ PSD (composited view only, no extra channels)
+ HDR (radiance rgbE format)
+ writes BMP,TGA (define STBI_NO_WRITE to remove code)
+ decoded from memory or through stdio FILE (define STBI_NO_STDIO to remove code)
+ supports installable dequantizing-IDCT, YCbCr-to-RGB conversion (define STBI_SIMD)
+
+ TODO:
+ stbi_info_*
+
+ history:
+ 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
+*/
+
+#include "stb_image_aug.h"
+
+#ifndef STBI_NO_HDR
+#include // ldexp
+#include // strcmp
+#endif
+
+#ifndef STBI_NO_STDIO
+#include
+#endif
+#include
+#include
+#include
+#include
+
+#ifndef _MSC_VER
+ #ifdef __cplusplus
+ #define __forceinline inline
+ #else
+ #define __forceinline
+ #endif
+#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];
+
+#if defined(STBI_NO_STDIO) && !defined(STBI_NO_WRITE)
+#define STBI_NO_WRITE
+#endif
+
+#ifndef STBI_NO_DDS
+#include "stbi_DDS_aug.h"
+#endif
+
+// I (JLD) want full messages for SOIL
+#define STBI_FAILURE_USERMSG 1
+
+//////////////////////////////////////////////////////////////////////////////
+//
+// Generic API that works on all image types
+//
+
+// this is not threadsafe
+static char *failure_reason;
+
+char *stbi_failure_reason(void)
+{
+ return failure_reason;
+}
+
+static int e(char *str)
+{
+ failure_reason = str;
+ return 0;
+}
+
+#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);
+}
+
+#define MAX_LOADERS 32
+stbi_loader *loaders[MAX_LOADERS];
+static int max_loaders = 0;
+
+int stbi_register_loader(stbi_loader *loader)
+{
+ int i;
+ for (i=0; i < MAX_LOADERS; ++i) {
+ // already present?
+ if (loaders[i] == loader)
+ return 1;
+ // end of the list?
+ if (loaders[i] == NULL) {
+ loaders[i] = loader;
+ max_loaders = i+1;
+ return 1;
+ }
+ }
+ // no room for it
+ return 0;
+}
+
+#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
+
+#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)
+{
+ int i;
+ if (stbi_jpeg_test_file(f))
+ return stbi_jpeg_load_from_file(f,x,y,comp,req_comp);
+ if (stbi_png_test_file(f))
+ return stbi_png_load_from_file(f,x,y,comp,req_comp);
+ if (stbi_bmp_test_file(f))
+ return stbi_bmp_load_from_file(f,x,y,comp,req_comp);
+ if (stbi_psd_test_file(f))
+ return stbi_psd_load_from_file(f,x,y,comp,req_comp);
+ #ifndef STBI_NO_DDS
+ if (stbi_dds_test_file(f))
+ return stbi_dds_load_from_file(f,x,y,comp,req_comp);
+ #endif
+ #ifndef STBI_NO_HDR
+ if (stbi_hdr_test_file(f)) {
+ float *hdr = stbi_hdr_load_from_file(f, x,y,comp,req_comp);
+ return hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp);
+ }
+ #endif
+ for (i=0; i < max_loaders; ++i)
+ if (loaders[i]->test_file(f))
+ return loaders[i]->load_from_file(f,x,y,comp,req_comp);
+ // test tga last because it's a crappy test!
+ if (stbi_tga_test_file(f))
+ return stbi_tga_load_from_file(f,x,y,comp,req_comp);
+ return epuc("unknown image type", "Image not of any known type, or corrupt");
+}
+#endif
+
+unsigned char *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
+{
+ int i;
+ if (stbi_jpeg_test_memory(buffer,len))
+ return stbi_jpeg_load_from_memory(buffer,len,x,y,comp,req_comp);
+ if (stbi_png_test_memory(buffer,len))
+ return stbi_png_load_from_memory(buffer,len,x,y,comp,req_comp);
+ if (stbi_bmp_test_memory(buffer,len))
+ return stbi_bmp_load_from_memory(buffer,len,x,y,comp,req_comp);
+ if (stbi_psd_test_memory(buffer,len))
+ return stbi_psd_load_from_memory(buffer,len,x,y,comp,req_comp);
+ #ifndef STBI_NO_DDS
+ if (stbi_dds_test_memory(buffer,len))
+ return stbi_dds_load_from_memory(buffer,len,x,y,comp,req_comp);
+ #endif
+ #ifndef STBI_NO_HDR
+ if (stbi_hdr_test_memory(buffer, len)) {
+ float *hdr = stbi_hdr_load_from_memory(buffer, len,x,y,comp,req_comp);
+ return hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp);
+ }
+ #endif
+ for (i=0; i < max_loaders; ++i)
+ if (loaders[i]->test_memory(buffer,len))
+ return loaders[i]->load_from_memory(buffer,len,x,y,comp,req_comp);
+ // test tga last because it's a crappy test!
+ if (stbi_tga_test_memory(buffer,len))
+ return stbi_tga_load_from_memory(buffer,len,x,y,comp,req_comp);
+ return epuc("unknown image type", "Image not of any known type, or corrupt");
+}
+
+#ifndef STBI_NO_HDR
+
+#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)
+{
+ unsigned char *data;
+ #ifndef STBI_NO_HDR
+ if (stbi_hdr_test_file(f))
+ return stbi_hdr_load_from_file(f,x,y,comp,req_comp);
+ #endif
+ data = stbi_load_from_file(f, 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");
+}
+#endif
+
+float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
+{
+ stbi_uc *data;
+ #ifndef STBI_NO_HDR
+ if (stbi_hdr_test_memory(buffer, len))
+ return stbi_hdr_load_from_memory(buffer, len,x,y,comp,req_comp);
+ #endif
+ data = stbi_load_from_memory(buffer, len, 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");
+}
+#endif
+
+// 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
+ return stbi_hdr_test_memory(buffer, len);
+ #else
+ 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
+ return stbi_hdr_test_file(f);
+ #else
+ return 0;
+ #endif
+}
+
+#endif
+
+// @TODO: get image dimensions & components without fully decoding
+#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
+extern int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp);
+
+#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,
+};
+
+typedef struct
+{
+ uint32 img_x, img_y;
+ int img_n, img_out_n;
+
+ #ifndef STBI_NO_STDIO
+ FILE *img_file;
+ #endif
+ uint8 *img_buffer, *img_buffer_end;
+} stbi;
+
+#ifndef STBI_NO_STDIO
+static void start_file(stbi *s, FILE *f)
+{
+ s->img_file = f;
+}
+#endif
+
+static void start_mem(stbi *s, uint8 const *buffer, int len)
+{
+#ifndef STBI_NO_STDIO
+ s->img_file = NULL;
+#endif
+ s->img_buffer = (uint8 *) buffer;
+ s->img_buffer_end = (uint8 *) buffer+len;
+}
+
+__forceinline static int get8(stbi *s)
+{
+#ifndef STBI_NO_STDIO
+ if (s->img_file) {
+ int c = fgetc(s->img_file);
+ return c == EOF ? 0 : c;
+ }
+#endif
+ if (s->img_buffer < s->img_buffer_end)
+ return *s->img_buffer++;
+ return 0;
+}
+
+__forceinline static int at_eof(stbi *s)
+{
+#ifndef STBI_NO_STDIO
+ if (s->img_file)
+ return feof(s->img_file);
+#endif
+ return s->img_buffer >= s->img_buffer_end;
+}
+
+__forceinline static uint8 get8u(stbi *s)
+{
+ return (uint8) get8(s);
+}
+
+static void skip(stbi *s, int n)
+{
+#ifndef STBI_NO_STDIO
+ if (s->img_file)
+ fseek(s->img_file, n, SEEK_CUR);
+ else
+#endif
+ s->img_buffer += n;
+}
+
+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);
+}
+
+static void getn(stbi *s, stbi_uc *buffer, int n)
+{
+#ifndef STBI_NO_STDIO
+ if (s->img_file) {
+ fread(buffer, 1, n, s->img_file);
+ return;
+ }
+#endif
+ memcpy(buffer, s->img_buffer, n);
+ s->img_buffer += n;
+}
+
+//////////////////////////////////////////////////////////////////////////////
+//
+// 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] = 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] = 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
+{
+ #if 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 = (j->code_buffer << 8) | b;
+ 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
+__forceinline 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 >> (j->code_bits - FAST_BITS)) & ((1 << FAST_BITS)-1);
+ k = h->fast[c];
+ if (k < 255) {
+ if (h->size[k] > j->code_bits)
+ return -1;
+ j->code_bits -= h->size[k];
+ 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.
+ if (j->code_bits < 16)
+ temp = (j->code_buffer << (16 - j->code_bits)) & 0xffff;
+ else
+ temp = (j->code_buffer >> (j->code_bits - 16)) & 0xffff;
+ 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 >> (j->code_bits - k)) & bmask[k]) + h->delta[k];
+ assert((((j->code_buffer) >> (j->code_bits - h->size[c])) & bmask[h->size[c]]) == h->code[c]);
+
+ // convert the id to a symbol
+ j->code_bits -= k;
+ return h->values[c];
+}
+
+// combined JPEG 'receive' and JPEG 'extend', since baseline
+// always extends everything it receives.
+__forceinline 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);
+ k = (j->code_buffer >> (j->code_bits - n)) & bmask[n];
+ j->code_bits -= n;
+ // 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
+__forceinline static uint8 clamp(int x)
+{
+ x += 128;
+ // 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;
+
+#if !STBI_SIMD
+// .344 seconds on 3*anemones.jpg
+static void idct_block(uint8 *out, int out_stride, short data[64], uint8 *dequantize)
+{
+ int i,val[64],*v=val;
+ uint8 *o,*dq = dequantize;
+ 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.
+ x0 += 65536; x1 += 65536; x2 += 65536; x3 += 65536;
+ 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);
+ }
+}
+#else
+static void idct_block(uint8 *out, int out_stride, short data[64], unsigned short *dequantize)
+{
+ int i,val[64],*v=val;
+ uint8 *o;
+ unsigned short *dq = dequantize;
+ 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.
+ x0 += 65536; x1 += 65536; x2 += 65536; x3 += 65536;
+ 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);
+ }
+}
+static stbi_idct_8x8 stbi_idct_installed = idct_block;
+
+extern 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;
+ #if 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;
+ #if 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;
+ #if 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);
+ #if STBI_SIMD
+ for (i=0; i < 64; ++i)
+ z->dequant2[t][i] = 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;
+ } 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)
+{
+ 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;
+ 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];
+ 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);
+ 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;
+ 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, uint8 *y, uint8 *pcb, 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;
+ }
+}
+
+#if 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) {
+ #if 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;
+ }
+}
+
+#ifndef STBI_NO_STDIO
+unsigned char *stbi_jpeg_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ jpeg j;
+ start_file(&j.s, f);
+ return load_jpeg_image(&j, x,y,comp,req_comp);
+}
+
+unsigned char *stbi_jpeg_load(char const *filename, int *x, int *y, int *comp, int req_comp)
+{
+ unsigned char *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_jpeg_load_from_file(f,x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+#endif
+
+unsigned char *stbi_jpeg_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
+{
+ jpeg j;
+ start_mem(&j.s, buffer,len);
+ return load_jpeg_image(&j, x,y,comp,req_comp);
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_jpeg_test_file(FILE *f)
+{
+ int n,r;
+ jpeg j;
+ n = ftell(f);
+ start_file(&j.s, f);
+ r = decode_jpeg_header(&j, SCAN_type);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_jpeg_test_memory(stbi_uc const *buffer, int len)
+{
+ jpeg j;
+ start_mem(&j.s, buffer,len);
+ return decode_jpeg_header(&j, SCAN_type);
+}
+
+// @TODO:
+#ifndef STBI_NO_STDIO
+extern int stbi_jpeg_info (char const *filename, int *x, int *y, int *comp);
+extern int stbi_jpeg_info_from_file (FILE *f, int *x, int *y, int *comp);
+#endif
+extern int stbi_jpeg_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *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;
+
+__forceinline 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;
+}
+
+__forceinline 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;
+
+__forceinline 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);
+}
+
+__forceinline 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;
+}
+
+__forceinline 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 };
+ static zhuffman z_codelength; // static just to save stack space
+ 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;
+}
+
+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;
+ }
+ } 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);
+}
+
+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 (get8(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(png *a, uint8 *raw, uint32 raw_len, int out_n)
+{
+ stbi *s = &a->s;
+ uint32 i,j,stride = s->img_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);
+ a->out = (uint8 *) malloc(s->img_x * s->img_y * out_n);
+ if (!a->out) return e("outofmem", "Out of memory");
+ if (raw_len != (img_n * s->img_x + 1) * s->img_y) return e("not enough pixels","Corrupt PNG");
+ for (j=0; j < s->img_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=s->img_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=s->img_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 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;
+ return 1;
+}
+
+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;
+ stbi *s = &z->s;
+
+ if (!check_png_header(s)) return 0;
+
+ if (scan == SCAN_type) return 1;
+
+ for(;;first=0) {
+ chunk c = get_chunk_header(s);
+ if (first && c.type != PNG_TYPE('I','H','D','R'))
+ return e("first not IHDR","Corrupt PNG");
+ switch (c.type) {
+ case PNG_TYPE('I','H','D','R'): {
+ int depth,color,interlace,comp,filter;
+ if (!first) return e("multiple IHDR","Corrupt PNG");
+ 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) return e("interlaced","PNG not supported: interlaced mode");
+ 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 (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 (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 (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;
+ }
+ #ifndef STBI_NO_STDIO
+ if (s->img_file)
+ {
+ if (fread(z->idata+ioff,1,c.length,s->img_file) != c.length) return e("outofdata","Corrupt PNG");
+ }
+ else
+ #endif
+ {
+ memcpy(z->idata+ioff, s->img_buffer, c.length);
+ s->img_buffer += c.length;
+ }
+ ioff += c.length;
+ break;
+ }
+
+ case PNG_TYPE('I','E','N','D'): {
+ uint32 raw_len;
+ if (scan != SCAN_load) return 1;
+ if (z->idata == NULL) return e("no IDAT","Corrupt PNG");
+ z->expanded = (uint8 *) stbi_zlib_decode_malloc((char *) z->idata, ioff, (int *) &raw_len);
+ 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)) return 0;
+ if (has_trans)
+ if (!compute_transparency(z, tc, s->img_out_n)) return 0;
+ 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 ((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;
+ p->expanded = NULL;
+ p->idata = NULL;
+ p->out = 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;
+}
+
+#ifndef STBI_NO_STDIO
+unsigned char *stbi_png_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ png p;
+ start_file(&p.s, f);
+ return do_png(&p, x,y,comp,req_comp);
+}
+
+unsigned char *stbi_png_load(char const *filename, int *x, int *y, int *comp, int req_comp)
+{
+ unsigned char *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_png_load_from_file(f,x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+#endif
+
+unsigned char *stbi_png_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
+{
+ png p;
+ start_mem(&p.s, buffer,len);
+ return do_png(&p, x,y,comp,req_comp);
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_png_test_file(FILE *f)
+{
+ png p;
+ int n,r;
+ n = ftell(f);
+ start_file(&p.s, f);
+ r = parse_png_file(&p, SCAN_type,STBI_default);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_png_test_memory(stbi_uc const *buffer, int len)
+{
+ png p;
+ start_mem(&p.s, buffer, len);
+ return parse_png_file(&p, SCAN_type,STBI_default);
+}
+
+// TODO: load header from png
+#ifndef STBI_NO_STDIO
+extern int stbi_png_info (char const *filename, int *x, int *y, int *comp);
+extern int stbi_png_info_from_file (FILE *f, int *x, int *y, int *comp);
+#endif
+extern int stbi_png_info_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *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;
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_bmp_test_file (FILE *f)
+{
+ stbi s;
+ int r,n = ftell(f);
+ start_file(&s,f);
+ r = bmp_test(&s);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_bmp_test_memory (stbi_uc const *buffer, int len)
+{
+ stbi s;
+ start_mem(&s, buffer, len);
+ return bmp_test(&s);
+}
+
+// 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;
+ 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");
+ failure_reason = "bad BMP";
+ 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 0;
+ 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 = 0xff << 16;
+ mg = 0xff << 8;
+ mb = 0xff << 0;
+ } else {
+ mr = 31 << 10;
+ mg = 31 << 5;
+ mb = 31 << 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 NULL;
+ }
+ } else
+ return NULL;
+ }
+ } 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] = get8(s);
+ pal[i][1] = get8(s);
+ pal[i][0] = get8(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 == 0xff000000 && ma == 0xff000000)
+ easy = 2;
+ }
+ if (!easy) {
+ if (!mr || !mg || !mb) 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] = get8(s);
+ out[z+1] = get8(s);
+ out[z+0] = get8(s);
+ z += 3;
+ a = (easy == 2 ? get8(s) : 255);
+ if (target == 4) out[z++] = a;
+ }
+ } else {
+ for (i=0; i < (int) s->img_x; ++i) {
+ uint32 v = (bpp == 16 ? get16le(s) : get32le(s));
+ int a;
+ out[z++] = shiftsigned(v & mr, rshift, rcount);
+ out[z++] = shiftsigned(v & mg, gshift, gcount);
+ out[z++] = shiftsigned(v & mb, bshift, bcount);
+ a = (ma ? shiftsigned(v & ma, ashift, acount) : 255);
+ if (target == 4) out[z++] = 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 = target;
+ return out;
+}
+
+#ifndef STBI_NO_STDIO
+stbi_uc *stbi_bmp_load (char const *filename, int *x, int *y, int *comp, int req_comp)
+{
+ stbi_uc *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_bmp_load_from_file(f, x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+
+stbi_uc *stbi_bmp_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s, f);
+ return bmp_load(&s, x,y,comp,req_comp);
+}
+#endif
+
+stbi_uc *stbi_bmp_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 bmp_load(&s, x,y,comp,req_comp);
+}
+
+// Targa Truevision - TGA
+// by Jonathan Dummer
+
+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
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_tga_test_file (FILE *f)
+{
+ stbi s;
+ int r,n = ftell(f);
+ start_file(&s, f);
+ r = tga_test(&s);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_tga_test_memory (stbi_uc const *buffer, int len)
+{
+ stbi s;
+ start_mem(&s, buffer, len);
+ return tga_test(&s);
+}
+
+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[] = { 0,0,0,0 };
+ 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;
+ }
+
+ // 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 );
+
+ // 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 );
+ getn(s, tga_palette, tga_palette_len * tga_palette_bits / 8 );
+ }
+ // load the data
+ 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;
+}
+
+#ifndef STBI_NO_STDIO
+stbi_uc *stbi_tga_load (char const *filename, int *x, int *y, int *comp, int req_comp)
+{
+ stbi_uc *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_tga_load_from_file(f, x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+
+stbi_uc *stbi_tga_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s, f);
+ return tga_load(&s, x,y,comp,req_comp);
+}
+#endif
+
+stbi_uc *stbi_tga_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 tga_load(&s, x,y,comp,req_comp);
+}
+
+
+// *************************************************************************************************
+// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicholas Schulz, tweaked by STB
+
+static int psd_test(stbi *s)
+{
+ if (get32(s) != 0x38425053) return 0; // "8BPS"
+ else return 1;
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_psd_test_file(FILE *f)
+{
+ stbi s;
+ int r,n = ftell(f);
+ start_file(&s, f);
+ r = psd_test(&s);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_psd_test_memory(stbi_uc const *buffer, int len)
+{
+ stbi s;
+ start_mem(&s, buffer, len);
+ return psd_test(&s);
+}
+
+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 = get8(s);
+ p += 4;
+ len--;
+ }
+ } else if (len > 128) {
+ uint32 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 = get8(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.
+ count = 0;
+ for (i = 0; i < pixelCount; i++)
+ *p = get8(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;
+}
+
+#ifndef STBI_NO_STDIO
+stbi_uc *stbi_psd_load(char const *filename, int *x, int *y, int *comp, int req_comp)
+{
+ stbi_uc *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_psd_load_from_file(f, x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+
+stbi_uc *stbi_psd_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s, f);
+ return psd_load(&s, x,y,comp,req_comp);
+}
+#endif
+
+stbi_uc *stbi_psd_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 psd_load(&s, x,y,comp,req_comp);
+}
+
+
+// *************************************************************************************************
+// Radiance RGBE HDR loader
+// originally by Nicolas Schulz
+#ifndef STBI_NO_HDR
+static int hdr_test(stbi *s)
+{
+ char *signature = "#?RADIANCE\n";
+ int i;
+ for (i=0; signature[i]; ++i)
+ if (get8(s) != signature[i])
+ return 0;
+ return 1;
+}
+
+int stbi_hdr_test_memory(stbi_uc const *buffer, int len)
+{
+ stbi s;
+ start_mem(&s, buffer, len);
+ return hdr_test(&s);
+}
+
+#ifndef STBI_NO_STDIO
+int stbi_hdr_test_file(FILE *f)
+{
+ stbi s;
+ int r,n = ftell(f);
+ start_file(&s, f);
+ r = hdr_test(&s);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+#define HDR_BUFLEN 1024
+static char *hdr_gettoken(stbi *z, char *buffer)
+{
+ int len=0;
+ //char *s = buffer,
+ char c = '\0';
+
+ c = 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 = 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
+ while(1) {
+ 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)
+ stbi_uc rgbe[4] = { c1,c2,len, get8(s) };
+ hdr_convert(hdr_data, rgbe, req_comp);
+ i = 1;
+ j = 0;
+ free(scanline);
+ goto main_decode_loop; // yes, this is fucking insane; blame the fucking insane format
+ }
+ 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 = get8(s);
+ if (count > 128) {
+ // Run
+ value = get8(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] = get8(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 stbi_uc *hdr_load_rgbe(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;
+ stbi_uc *rgbe_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 epuc("not HDR", "Corrupt HDR image");
+
+ // Parse header
+ while(1) {
+ token = hdr_gettoken(s,buffer);
+ if (token[0] == 0) break;
+ if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1;
+ }
+
+ if (!valid) return epuc("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 epuc("unsupported data layout", "Unsupported HDR format");
+ token += 3;
+ height = strtol(token, &token, 10);
+ while (*token == ' ') ++token;
+ if (strncmp(token, "+X ", 3)) return epuc("unsupported data layout", "Unsupported HDR format");
+ token += 3;
+ width = strtol(token, NULL, 10);
+
+ *x = width;
+ *y = height;
+
+ // RGBE _MUST_ come out as 4 components
+ *comp = 4;
+ req_comp = 4;
+
+ // Read data
+ rgbe_data = (stbi_uc *) malloc(height * width * req_comp * sizeof(stbi_uc));
+ // point to the beginning
+ scanline = rgbe_data;
+
+ // Load image data
+ // image data is stored as some number of scan lines
+ if( width < 8 || width >= 32768) {
+ // Read flat data
+ for (j=0; j < height; ++j) {
+ for (i=0; i < width; ++i) {
+ main_decode_loop:
+ //getn(rgbe, 4);
+ getn(s,scanline, 4);
+ scanline += 4;
+ }
+ }
+ } else {
+ // Read RLE-encoded data
+ 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)
+ scanline[0] = c1;
+ scanline[1] = c2;
+ scanline[2] = len;
+ scanline[3] = get8(s);
+ scanline += 4;
+ i = 1;
+ j = 0;
+ goto main_decode_loop; // yes, this is insane; blame the insane format
+ }
+ len <<= 8;
+ len |= get8(s);
+ if (len != width) { free(rgbe_data); return epuc("invalid decoded scanline length", "corrupt HDR"); }
+ for (k = 0; k < 4; ++k) {
+ i = 0;
+ while (i < width) {
+ count = get8(s);
+ if (count > 128) {
+ // Run
+ value = get8(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] = get8(s);
+ }
+ }
+ }
+ // move the scanline on
+ scanline += 4 * width;
+ }
+ }
+
+ return rgbe_data;
+}
+
+#ifndef STBI_NO_STDIO
+float *stbi_hdr_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s,f);
+ return hdr_load(&s,x,y,comp,req_comp);
+}
+
+stbi_uc *stbi_hdr_load_rgbe_file(FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s,f);
+ return hdr_load_rgbe(&s,x,y,comp,req_comp);
+}
+
+stbi_uc *stbi_hdr_load_rgbe (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_hdr_load_rgbe_file(f,x,y,comp,req_comp);
+ fclose(f);
+ return result;
+}
+#endif
+
+float *stbi_hdr_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 hdr_load(&s,x,y,comp,req_comp);
+}
+
+stbi_uc *stbi_hdr_load_rgbe_memory(stbi_uc *buffer, int len, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_mem(&s,buffer, len);
+ return hdr_load_rgbe(&s,x,y,comp,req_comp);
+}
+
+#endif // STBI_NO_HDR
+
+/////////////////////// write image ///////////////////////
+
+#ifndef STBI_NO_WRITE
+
+static void write8(FILE *f, int x) { uint8 z = (uint8) x; fwrite(&z,1,1,f); }
+
+static void writefv(FILE *f, char *fmt, va_list v)
+{
+ while (*fmt) {
+ switch (*fmt++) {
+ case ' ': break;
+ case '1': { uint8 x = va_arg(v, int); write8(f,x); break; }
+ case '2': { int16 x = va_arg(v, int); write8(f,x); write8(f,x>>8); break; }
+ case '4': { int32 x = va_arg(v, int); write8(f,x); write8(f,x>>8); write8(f,x>>16); write8(f,x>>24); break; }
+ default:
+ assert(0);
+ va_end(v);
+ return;
+ }
+ }
+}
+
+static void writef(FILE *f, char *fmt, ...)
+{
+ va_list v;
+ va_start(v, fmt);
+ writefv(f,fmt,v);
+ va_end(v);
+}
+
+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)
+{
+ uint8 bg[3] = { 255, 0, 255}, px[3];
+ uint32 zero = 0;
+ int i,j,k, j_end;
+
+ 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) {
+ uint8 *d = (uint8 *) data + (j*x+i)*comp;
+ if (write_alpha < 0)
+ fwrite(&d[comp-1], 1, 1, f);
+ switch (comp) {
+ case 1:
+ case 2: writef(f, "111", d[0],d[0],d[0]);
+ break;
+ case 4:
+ if (!write_alpha) {
+ for (k=0; k < 3; ++k)
+ px[k] = bg[k] + ((d[k] - bg[k]) * d[3])/255;
+ writef(f, "111", px[1-rgb_dir],px[1],px[1+rgb_dir]);
+ break;
+ }
+ /* FALLTHROUGH */
+ case 3:
+ writef(f, "111", 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, char *fmt, ...)
+{
+ FILE *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, void *data)
+{
+ int pad = (-x*3) & 3;
+ return outfile(filename,-1,-1,x,y,comp,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, void *data)
+{
+ int has_alpha = !(comp & 1);
+ return outfile(filename, -1,-1, x, y, comp, 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);
+}
+
+// any other image formats that do interleaved rgb data?
+// PNG: requires adler32,crc32 -- significant amount of code
+// PSD: no, channels output separately
+// TIFF: no, stripwise-interleaved... i think
+
+#endif // STBI_NO_WRITE
+
+// add in my DDS loading support
+#ifndef STBI_NO_DDS
+#include "stbi_DDS_aug_c.h"
+#endif
diff --git a/cw 9/src/SOIL/stb_image_aug.h b/cw 9/src/SOIL/stb_image_aug.h
new file mode 100644
index 0000000..e59f2eb
--- /dev/null
+++ b/cw 9/src/SOIL/stb_image_aug.h
@@ -0,0 +1,354 @@
+/* stbi-1.16 - public domain JPEG/PNG reader - http://nothings.org/stb_image.c
+ when you control the images you're loading
+
+ 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, no oddball channel decimations)
+ PNG non-interlaced
+ BMP non-1bpp, non-RLE
+ TGA (not sure what subset, if a subset)
+ PSD (composited view only, no extra channels)
+ HDR (radiance rgbE format)
+ writes BMP,TGA (define STBI_NO_WRITE to remove code)
+ decoded from memory or through stdio FILE (define STBI_NO_STDIO to remove code)
+ supports installable dequantizing-IDCT, YCbCr-to-RGB conversion (define STBI_SIMD)
+
+ TODO:
+ stbi_info_*
+
+ history:
+ 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
+*/
+
+#ifndef HEADER_STB_IMAGE_AUGMENTED
+#define HEADER_STB_IMAGE_AUGMENTED
+
+//// begin header file ////////////////////////////////////////////////////
+//
+// Limitations:
+// - no progressive/interlaced support (jpeg, png)
+// - 8-bit samples only (jpeg, png)
+// - not threadsafe
+// - channel subsampling of at most 2 in each dimension (jpeg)
+// - no delayed line count (jpeg) -- IJG doesn't support either
+//
+// 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
+// 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 and BMP images are automatically depalettized.
+//
+//
+// ===========================================================================
+//
+// 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);
+
+#ifndef STBI_NO_STDIO
+#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
+
+// WRITING API
+
+#if !defined(STBI_NO_WRITE) && !defined(STBI_NO_STDIO)
+// write a BMP/TGA file given tightly packed 'comp' channels (no padding, nor bmp-stride-padding)
+// (you must include the appropriate extension in the filename).
+// returns TRUE on success, FALSE if couldn't open file, error writing file
+extern int stbi_write_bmp (char const *filename, int x, int y, int comp, void *data);
+extern int stbi_write_tga (char const *filename, int x, int y, int comp, void *data);
+#endif
+
+// PRIMARY API - works on images of any type
+
+// load image by filename, open file, or memory buffer
+#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);
+extern int stbi_info_from_file (FILE *f, int *x, int *y, int *comp);
+#endif
+extern stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+// for stbi_load_from_file, file pointer is left pointing immediately after image
+
+#ifndef STBI_NO_HDR
+#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_memory(stbi_uc const *buffer, int len, 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
+
+// get a VERY brief reason for failure
+// NOT THREADSAFE
+extern 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_is_hdr_from_memory(stbi_uc const *buffer, int len);
+#ifndef STBI_NO_STDIO
+extern int stbi_info (char const *filename, int *x, int *y, int *comp);
+extern int stbi_is_hdr (char const *filename);
+extern int stbi_is_hdr_from_file(FILE *f);
+#endif
+
+// 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);
+
+// TYPE-SPECIFIC ACCESS
+
+// is it a jpeg?
+extern int stbi_jpeg_test_memory (stbi_uc const *buffer, int len);
+extern stbi_uc *stbi_jpeg_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+extern int stbi_jpeg_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp);
+
+#ifndef STBI_NO_STDIO
+extern stbi_uc *stbi_jpeg_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern int stbi_jpeg_test_file (FILE *f);
+extern stbi_uc *stbi_jpeg_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+
+extern int stbi_jpeg_info (char const *filename, int *x, int *y, int *comp);
+extern int stbi_jpeg_info_from_file (FILE *f, int *x, int *y, int *comp);
+#endif
+
+// is it a png?
+extern int stbi_png_test_memory (stbi_uc const *buffer, int len);
+extern stbi_uc *stbi_png_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+extern int stbi_png_info_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp);
+
+#ifndef STBI_NO_STDIO
+extern stbi_uc *stbi_png_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern int stbi_png_info (char const *filename, int *x, int *y, int *comp);
+extern int stbi_png_test_file (FILE *f);
+extern stbi_uc *stbi_png_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+extern int stbi_png_info_from_file (FILE *f, int *x, int *y, int *comp);
+#endif
+
+// is it a bmp?
+extern int stbi_bmp_test_memory (stbi_uc const *buffer, int len);
+
+extern stbi_uc *stbi_bmp_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_bmp_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+#ifndef STBI_NO_STDIO
+extern int stbi_bmp_test_file (FILE *f);
+extern stbi_uc *stbi_bmp_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+#endif
+
+// is it a tga?
+extern int stbi_tga_test_memory (stbi_uc const *buffer, int len);
+
+extern stbi_uc *stbi_tga_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_tga_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+#ifndef STBI_NO_STDIO
+extern int stbi_tga_test_file (FILE *f);
+extern stbi_uc *stbi_tga_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+#endif
+
+// is it a psd?
+extern int stbi_psd_test_memory (stbi_uc const *buffer, int len);
+
+extern stbi_uc *stbi_psd_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_psd_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+#ifndef STBI_NO_STDIO
+extern int stbi_psd_test_file (FILE *f);
+extern stbi_uc *stbi_psd_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+#endif
+
+// is it an hdr?
+extern int stbi_hdr_test_memory (stbi_uc const *buffer, int len);
+
+extern float * stbi_hdr_load (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern float * stbi_hdr_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_hdr_load_rgbe (char const *filename, int *x, int *y, int *comp, int req_comp);
+extern float * stbi_hdr_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+#ifndef STBI_NO_STDIO
+extern int stbi_hdr_test_file (FILE *f);
+extern float * stbi_hdr_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_hdr_load_rgbe_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+#endif
+
+// define new loaders
+typedef struct
+{
+ int (*test_memory)(stbi_uc const *buffer, int len);
+ stbi_uc * (*load_from_memory)(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+ #ifndef STBI_NO_STDIO
+ int (*test_file)(FILE *f);
+ stbi_uc * (*load_from_file)(FILE *f, int *x, int *y, int *comp, int req_comp);
+ #endif
+} stbi_loader;
+
+// register a loader by filling out the above structure (you must defined ALL functions)
+// returns 1 if added or already added, 0 if not added (too many loaders)
+// NOT THREADSAFE
+extern int stbi_register_loader(stbi_loader *loader);
+
+// define faster low-level operations (typically SIMD support)
+#if STBI_SIMD
+typedef void (*stbi_idct_8x8)(uint8 *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)(uint8 *output, uint8 const *y, uint8 const *cb, uint8 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
diff --git a/cw 9/src/SOIL/stbi_DDS_aug.h b/cw 9/src/SOIL/stbi_DDS_aug.h
new file mode 100644
index 0000000..c7da9f7
--- /dev/null
+++ b/cw 9/src/SOIL/stbi_DDS_aug.h
@@ -0,0 +1,21 @@
+/*
+ adding DDS loading support to stbi
+*/
+
+#ifndef HEADER_STB_IMAGE_DDS_AUGMENTATION
+#define HEADER_STB_IMAGE_DDS_AUGMENTATION
+
+// is it a DDS file?
+extern int stbi_dds_test_memory (stbi_uc const *buffer, int len);
+
+extern stbi_uc *stbi_dds_load (char *filename, int *x, int *y, int *comp, int req_comp);
+extern stbi_uc *stbi_dds_load_from_memory (stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp);
+#ifndef STBI_NO_STDIO
+extern int stbi_dds_test_file (FILE *f);
+extern stbi_uc *stbi_dds_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp);
+#endif
+
+//
+//
+//// end header file /////////////////////////////////////////////////////
+#endif // HEADER_STB_IMAGE_DDS_AUGMENTATION
diff --git a/cw 9/src/SOIL/stbi_DDS_aug_c.h b/cw 9/src/SOIL/stbi_DDS_aug_c.h
new file mode 100644
index 0000000..f49407a
--- /dev/null
+++ b/cw 9/src/SOIL/stbi_DDS_aug_c.h
@@ -0,0 +1,511 @@
+
+/// DDS file support, does decoding, _not_ direct uploading
+/// (use SOIL for that ;-)
+
+/// A bunch of DirectDraw Surface structures and flags
+typedef struct {
+ unsigned int dwMagic;
+ unsigned int dwSize;
+ unsigned int dwFlags;
+ unsigned int dwHeight;
+ unsigned int dwWidth;
+ unsigned int dwPitchOrLinearSize;
+ unsigned int dwDepth;
+ unsigned int dwMipMapCount;
+ unsigned int dwReserved1[ 11 ];
+
+ // DDPIXELFORMAT
+ struct {
+ unsigned int dwSize;
+ unsigned int dwFlags;
+ unsigned int dwFourCC;
+ unsigned int dwRGBBitCount;
+ unsigned int dwRBitMask;
+ unsigned int dwGBitMask;
+ unsigned int dwBBitMask;
+ unsigned int dwAlphaBitMask;
+ } sPixelFormat;
+
+ // DDCAPS2
+ struct {
+ unsigned int dwCaps1;
+ unsigned int dwCaps2;
+ unsigned int dwDDSX;
+ unsigned int dwReserved;
+ } sCaps;
+ unsigned int dwReserved2;
+} DDS_header ;
+
+// the following constants were copied directly off the MSDN website
+
+// The dwFlags member of the original DDSURFACEDESC2 structure
+// can be set to one or more of the following values.
+#define DDSD_CAPS 0x00000001
+#define DDSD_HEIGHT 0x00000002
+#define DDSD_WIDTH 0x00000004
+#define DDSD_PITCH 0x00000008
+#define DDSD_PIXELFORMAT 0x00001000
+#define DDSD_MIPMAPCOUNT 0x00020000
+#define DDSD_LINEARSIZE 0x00080000
+#define DDSD_DEPTH 0x00800000
+
+// DirectDraw Pixel Format
+#define DDPF_ALPHAPIXELS 0x00000001
+#define DDPF_FOURCC 0x00000004
+#define DDPF_RGB 0x00000040
+
+// The dwCaps1 member of the DDSCAPS2 structure can be
+// set to one or more of the following values.
+#define DDSCAPS_COMPLEX 0x00000008
+#define DDSCAPS_TEXTURE 0x00001000
+#define DDSCAPS_MIPMAP 0x00400000
+
+// The dwCaps2 member of the DDSCAPS2 structure can be
+// set to one or more of the following values.
+#define DDSCAPS2_CUBEMAP 0x00000200
+#define DDSCAPS2_CUBEMAP_POSITIVEX 0x00000400
+#define DDSCAPS2_CUBEMAP_NEGATIVEX 0x00000800
+#define DDSCAPS2_CUBEMAP_POSITIVEY 0x00001000
+#define DDSCAPS2_CUBEMAP_NEGATIVEY 0x00002000
+#define DDSCAPS2_CUBEMAP_POSITIVEZ 0x00004000
+#define DDSCAPS2_CUBEMAP_NEGATIVEZ 0x00008000
+#define DDSCAPS2_VOLUME 0x00200000
+
+static int dds_test(stbi *s)
+{
+ // check the magic number
+ if (get8(s) != 'D') return 0;
+ if (get8(s) != 'D') return 0;
+ if (get8(s) != 'S') return 0;
+ if (get8(s) != ' ') return 0;
+ // check header size
+ if (get32le(s) != 124) return 0;
+ return 1;
+}
+#ifndef STBI_NO_STDIO
+int stbi_dds_test_file (FILE *f)
+{
+ stbi s;
+ int r,n = ftell(f);
+ start_file(&s,f);
+ r = dds_test(&s);
+ fseek(f,n,SEEK_SET);
+ return r;
+}
+#endif
+
+int stbi_dds_test_memory (stbi_uc const *buffer, int len)
+{
+ stbi s;
+ start_mem(&s,buffer, len);
+ return dds_test(&s);
+}
+
+// helper functions
+int stbi_convert_bit_range( int c, int from_bits, int to_bits )
+{
+ int b = (1 << (from_bits - 1)) + c * ((1 << to_bits) - 1);
+ return (b + (b >> from_bits)) >> from_bits;
+}
+void stbi_rgb_888_from_565( unsigned int c, int *r, int *g, int *b )
+{
+ *r = stbi_convert_bit_range( (c >> 11) & 31, 5, 8 );
+ *g = stbi_convert_bit_range( (c >> 05) & 63, 6, 8 );
+ *b = stbi_convert_bit_range( (c >> 00) & 31, 5, 8 );
+}
+void stbi_decode_DXT1_block(
+ unsigned char uncompressed[16*4],
+ unsigned char compressed[8] )
+{
+ int next_bit = 4*8;
+ int i, r, g, b;
+ int c0, c1;
+ unsigned char decode_colors[4*4];
+ // find the 2 primary colors
+ c0 = compressed[0] + (compressed[1] << 8);
+ c1 = compressed[2] + (compressed[3] << 8);
+ stbi_rgb_888_from_565( c0, &r, &g, &b );
+ decode_colors[0] = r;
+ decode_colors[1] = g;
+ decode_colors[2] = b;
+ decode_colors[3] = 255;
+ stbi_rgb_888_from_565( c1, &r, &g, &b );
+ decode_colors[4] = r;
+ decode_colors[5] = g;
+ decode_colors[6] = b;
+ decode_colors[7] = 255;
+ if( c0 > c1 )
+ {
+ // no alpha, 2 interpolated colors
+ decode_colors[8] = (2*decode_colors[0] + decode_colors[4]) / 3;
+ decode_colors[9] = (2*decode_colors[1] + decode_colors[5]) / 3;
+ decode_colors[10] = (2*decode_colors[2] + decode_colors[6]) / 3;
+ decode_colors[11] = 255;
+ decode_colors[12] = (decode_colors[0] + 2*decode_colors[4]) / 3;
+ decode_colors[13] = (decode_colors[1] + 2*decode_colors[5]) / 3;
+ decode_colors[14] = (decode_colors[2] + 2*decode_colors[6]) / 3;
+ decode_colors[15] = 255;
+ } else
+ {
+ // 1 interpolated color, alpha
+ decode_colors[8] = (decode_colors[0] + decode_colors[4]) / 2;
+ decode_colors[9] = (decode_colors[1] + decode_colors[5]) / 2;
+ decode_colors[10] = (decode_colors[2] + decode_colors[6]) / 2;
+ decode_colors[11] = 255;
+ decode_colors[12] = 0;
+ decode_colors[13] = 0;
+ decode_colors[14] = 0;
+ decode_colors[15] = 0;
+ }
+ // decode the block
+ for( i = 0; i < 16*4; i += 4 )
+ {
+ int idx = ((compressed[next_bit>>3] >> (next_bit & 7)) & 3) * 4;
+ next_bit += 2;
+ uncompressed[i+0] = decode_colors[idx+0];
+ uncompressed[i+1] = decode_colors[idx+1];
+ uncompressed[i+2] = decode_colors[idx+2];
+ uncompressed[i+3] = decode_colors[idx+3];
+ }
+ // done
+}
+void stbi_decode_DXT23_alpha_block(
+ unsigned char uncompressed[16*4],
+ unsigned char compressed[8] )
+{
+ int i, next_bit = 0;
+ // each alpha value gets 4 bits
+ for( i = 3; i < 16*4; i += 4 )
+ {
+ uncompressed[i] = stbi_convert_bit_range(
+ (compressed[next_bit>>3] >> (next_bit&7)) & 15,
+ 4, 8 );
+ next_bit += 4;
+ }
+}
+void stbi_decode_DXT45_alpha_block(
+ unsigned char uncompressed[16*4],
+ unsigned char compressed[8] )
+{
+ int i, next_bit = 8*2;
+ unsigned char decode_alpha[8];
+ // each alpha value gets 3 bits, and the 1st 2 bytes are the range
+ decode_alpha[0] = compressed[0];
+ decode_alpha[1] = compressed[1];
+ if( decode_alpha[0] > decode_alpha[1] )
+ {
+ // 6 step intermediate
+ decode_alpha[2] = (6*decode_alpha[0] + 1*decode_alpha[1]) / 7;
+ decode_alpha[3] = (5*decode_alpha[0] + 2*decode_alpha[1]) / 7;
+ decode_alpha[4] = (4*decode_alpha[0] + 3*decode_alpha[1]) / 7;
+ decode_alpha[5] = (3*decode_alpha[0] + 4*decode_alpha[1]) / 7;
+ decode_alpha[6] = (2*decode_alpha[0] + 5*decode_alpha[1]) / 7;
+ decode_alpha[7] = (1*decode_alpha[0] + 6*decode_alpha[1]) / 7;
+ } else
+ {
+ // 4 step intermediate, pluss full and none
+ decode_alpha[2] = (4*decode_alpha[0] + 1*decode_alpha[1]) / 5;
+ decode_alpha[3] = (3*decode_alpha[0] + 2*decode_alpha[1]) / 5;
+ decode_alpha[4] = (2*decode_alpha[0] + 3*decode_alpha[1]) / 5;
+ decode_alpha[5] = (1*decode_alpha[0] + 4*decode_alpha[1]) / 5;
+ decode_alpha[6] = 0;
+ decode_alpha[7] = 255;
+ }
+ for( i = 3; i < 16*4; i += 4 )
+ {
+ int idx = 0, bit;
+ bit = (compressed[next_bit>>3] >> (next_bit&7)) & 1;
+ idx += bit << 0;
+ ++next_bit;
+ bit = (compressed[next_bit>>3] >> (next_bit&7)) & 1;
+ idx += bit << 1;
+ ++next_bit;
+ bit = (compressed[next_bit>>3] >> (next_bit&7)) & 1;
+ idx += bit << 2;
+ ++next_bit;
+ uncompressed[i] = decode_alpha[idx & 7];
+ }
+ // done
+}
+void stbi_decode_DXT_color_block(
+ unsigned char uncompressed[16*4],
+ unsigned char compressed[8] )
+{
+ int next_bit = 4*8;
+ int i, r, g, b;
+ int c0, c1;
+ unsigned char decode_colors[4*3];
+ // find the 2 primary colors
+ c0 = compressed[0] + (compressed[1] << 8);
+ c1 = compressed[2] + (compressed[3] << 8);
+ stbi_rgb_888_from_565( c0, &r, &g, &b );
+ decode_colors[0] = r;
+ decode_colors[1] = g;
+ decode_colors[2] = b;
+ stbi_rgb_888_from_565( c1, &r, &g, &b );
+ decode_colors[3] = r;
+ decode_colors[4] = g;
+ decode_colors[5] = b;
+ // Like DXT1, but no choicees:
+ // no alpha, 2 interpolated colors
+ decode_colors[6] = (2*decode_colors[0] + decode_colors[3]) / 3;
+ decode_colors[7] = (2*decode_colors[1] + decode_colors[4]) / 3;
+ decode_colors[8] = (2*decode_colors[2] + decode_colors[5]) / 3;
+ decode_colors[9] = (decode_colors[0] + 2*decode_colors[3]) / 3;
+ decode_colors[10] = (decode_colors[1] + 2*decode_colors[4]) / 3;
+ decode_colors[11] = (decode_colors[2] + 2*decode_colors[5]) / 3;
+ // decode the block
+ for( i = 0; i < 16*4; i += 4 )
+ {
+ int idx = ((compressed[next_bit>>3] >> (next_bit & 7)) & 3) * 3;
+ next_bit += 2;
+ uncompressed[i+0] = decode_colors[idx+0];
+ uncompressed[i+1] = decode_colors[idx+1];
+ uncompressed[i+2] = decode_colors[idx+2];
+ }
+ // done
+}
+static stbi_uc *dds_load(stbi *s, int *x, int *y, int *comp, int req_comp)
+{
+ // all variables go up front
+ stbi_uc *dds_data = NULL;
+ stbi_uc block[16*4];
+ stbi_uc compressed[8];
+ int flags, DXT_family;
+ int has_alpha, has_mipmap;
+ int is_compressed, cubemap_faces;
+ int block_pitch, num_blocks;
+ DDS_header header;
+ int i, sz, cf;
+ // load the header
+ if( sizeof( DDS_header ) != 128 )
+ {
+ return NULL;
+ }
+ getn( s, (stbi_uc*)(&header), 128 );
+ // and do some checking
+ if( header.dwMagic != (('D' << 0) | ('D' << 8) | ('S' << 16) | (' ' << 24)) ) return NULL;
+ if( header.dwSize != 124 ) return NULL;
+ flags = DDSD_CAPS | DDSD_HEIGHT | DDSD_WIDTH | DDSD_PIXELFORMAT;
+ if( (header.dwFlags & flags) != flags ) return NULL;
+ /* According to the MSDN spec, the dwFlags should contain
+ DDSD_LINEARSIZE if it's compressed, or DDSD_PITCH if
+ uncompressed. Some DDS writers do not conform to the
+ spec, so I need to make my reader more tolerant */
+ if( header.sPixelFormat.dwSize != 32 ) return NULL;
+ flags = DDPF_FOURCC | DDPF_RGB;
+ if( (header.sPixelFormat.dwFlags & flags) == 0 ) return NULL;
+ if( (header.sCaps.dwCaps1 & DDSCAPS_TEXTURE) == 0 ) return NULL;
+ // get the image data
+ s->img_x = header.dwWidth;
+ s->img_y = header.dwHeight;
+ s->img_n = 4;
+ is_compressed = (header.sPixelFormat.dwFlags & DDPF_FOURCC) / DDPF_FOURCC;
+ has_alpha = (header.sPixelFormat.dwFlags & DDPF_ALPHAPIXELS) / DDPF_ALPHAPIXELS;
+ has_mipmap = (header.sCaps.dwCaps1 & DDSCAPS_MIPMAP) && (header.dwMipMapCount > 1);
+ cubemap_faces = (header.sCaps.dwCaps2 & DDSCAPS2_CUBEMAP) / DDSCAPS2_CUBEMAP;
+ /* I need cubemaps to have square faces */
+ cubemap_faces &= (s->img_x == s->img_y);
+ cubemap_faces *= 5;
+ cubemap_faces += 1;
+ block_pitch = (s->img_x+3) >> 2;
+ num_blocks = block_pitch * ((s->img_y+3) >> 2);
+ /* let the user know what's going on */
+ *x = s->img_x;
+ *y = s->img_y;
+ *comp = s->img_n;
+ /* is this uncompressed? */
+ if( is_compressed )
+ {
+ /* compressed */
+ // note: header.sPixelFormat.dwFourCC is something like (('D'<<0)|('X'<<8)|('T'<<16)|('1'<<24))
+ DXT_family = 1 + (header.sPixelFormat.dwFourCC >> 24) - '1';
+ if( (DXT_family < 1) || (DXT_family > 5) ) return NULL;
+ /* check the expected size...oops, nevermind...
+ those non-compliant writers leave
+ dwPitchOrLinearSize == 0 */
+ // passed all the tests, get the RAM for decoding
+ sz = (s->img_x)*(s->img_y)*4*cubemap_faces;
+ dds_data = (unsigned char*)malloc( sz );
+ /* do this once for each face */
+ for( cf = 0; cf < cubemap_faces; ++ cf )
+ {
+ // now read and decode all the blocks
+ for( i = 0; i < num_blocks; ++i )
+ {
+ // where are we?
+ int bx, by, bw=4, bh=4;
+ int ref_x = 4 * (i % block_pitch);
+ int ref_y = 4 * (i / block_pitch);
+ // get the next block's worth of compressed data, and decompress it
+ if( DXT_family == 1 )
+ {
+ // DXT1
+ getn( s, compressed, 8 );
+ stbi_decode_DXT1_block( block, compressed );
+ } else if( DXT_family < 4 )
+ {
+ // DXT2/3
+ getn( s, compressed, 8 );
+ stbi_decode_DXT23_alpha_block ( block, compressed );
+ getn( s, compressed, 8 );
+ stbi_decode_DXT_color_block ( block, compressed );
+ } else
+ {
+ // DXT4/5
+ getn( s, compressed, 8 );
+ stbi_decode_DXT45_alpha_block ( block, compressed );
+ getn( s, compressed, 8 );
+ stbi_decode_DXT_color_block ( block, compressed );
+ }
+ // is this a partial block?
+ if( ref_x + 4 > s->img_x )
+ {
+ bw = s->img_x - ref_x;
+ }
+ if( ref_y + 4 > s->img_y )
+ {
+ bh = s->img_y - ref_y;
+ }
+ // now drop our decompressed data into the buffer
+ for( by = 0; by < bh; ++by )
+ {
+ int idx = 4*((ref_y+by+cf*s->img_x)*s->img_x + ref_x);
+ for( bx = 0; bx < bw*4; ++bx )
+ {
+
+ dds_data[idx+bx] = block[by*16+bx];
+ }
+ }
+ }
+ /* done reading and decoding the main image...
+ skip MIPmaps if present */
+ if( has_mipmap )
+ {
+ int block_size = 16;
+ if( DXT_family == 1 )
+ {
+ block_size = 8;
+ }
+ for( i = 1; i < header.dwMipMapCount; ++i )
+ {
+ int mx = s->img_x >> (i + 2);
+ int my = s->img_y >> (i + 2);
+ if( mx < 1 )
+ {
+ mx = 1;
+ }
+ if( my < 1 )
+ {
+ my = 1;
+ }
+ skip( s, mx*my*block_size );
+ }
+ }
+ }/* per cubemap face */
+ } else
+ {
+ /* uncompressed */
+ DXT_family = 0;
+ s->img_n = 3;
+ if( has_alpha )
+ {
+ s->img_n = 4;
+ }
+ *comp = s->img_n;
+ sz = s->img_x*s->img_y*s->img_n*cubemap_faces;
+ dds_data = (unsigned char*)malloc( sz );
+ /* do this once for each face */
+ for( cf = 0; cf < cubemap_faces; ++ cf )
+ {
+ /* read the main image for this face */
+ getn( s, &dds_data[cf*s->img_x*s->img_y*s->img_n], s->img_x*s->img_y*s->img_n );
+ /* done reading and decoding the main image...
+ skip MIPmaps if present */
+ if( has_mipmap )
+ {
+ for( i = 1; i < header.dwMipMapCount; ++i )
+ {
+ int mx = s->img_x >> i;
+ int my = s->img_y >> i;
+ if( mx < 1 )
+ {
+ mx = 1;
+ }
+ if( my < 1 )
+ {
+ my = 1;
+ }
+ skip( s, mx*my*s->img_n );
+ }
+ }
+ }
+ /* data was BGR, I need it RGB */
+ for( i = 0; i < sz; i += s->img_n )
+ {
+ unsigned char temp = dds_data[i];
+ dds_data[i] = dds_data[i+2];
+ dds_data[i+2] = temp;
+ }
+ }
+ /* finished decompressing into RGBA,
+ adjust the y size if we have a cubemap
+ note: sz is already up to date */
+ s->img_y *= cubemap_faces;
+ *y = s->img_y;
+ // did the user want something else, or
+ // see if all the alpha values are 255 (i.e. no transparency)
+ has_alpha = 0;
+ if( s->img_n == 4)
+ {
+ for( i = 3; (i < sz) && (has_alpha == 0); i += 4 )
+ {
+ has_alpha |= (dds_data[i] < 255);
+ }
+ }
+ if( (req_comp <= 4) && (req_comp >= 1) )
+ {
+ // user has some requirements, meet them
+ if( req_comp != s->img_n )
+ {
+ dds_data = convert_format( dds_data, s->img_n, req_comp, s->img_x, s->img_y );
+ *comp = s->img_n;
+ }
+ } else
+ {
+ // user had no requirements, only drop to RGB is no alpha
+ if( (has_alpha == 0) && (s->img_n == 4) )
+ {
+ dds_data = convert_format( dds_data, 4, 3, s->img_x, s->img_y );
+ *comp = 3;
+ }
+ }
+ // OK, done
+ return dds_data;
+}
+
+#ifndef STBI_NO_STDIO
+stbi_uc *stbi_dds_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp)
+{
+ stbi s;
+ start_file(&s,f);
+ return dds_load(&s,x,y,comp,req_comp);
+}
+
+stbi_uc *stbi_dds_load (char *filename, int *x, int *y, int *comp, int req_comp)
+{
+ stbi_uc *data;
+ FILE *f = fopen(filename, "rb");
+ if (!f) return NULL;
+ data = stbi_dds_load_from_file(f,x,y,comp,req_comp);
+ fclose(f);
+ return data;
+}
+#endif
+
+stbi_uc *stbi_dds_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 dds_load(&s,x,y,comp,req_comp);
+}
diff --git a/cw 9/src/Shader_Loader.cpp b/cw 9/src/Shader_Loader.cpp
new file mode 100644
index 0000000..6e118e9
--- /dev/null
+++ b/cw 9/src/Shader_Loader.cpp
@@ -0,0 +1,102 @@
+#include "Shader_Loader.h"
+#include
+#include
+#include
+
+using namespace Core;
+
+Shader_Loader::Shader_Loader(void){}
+Shader_Loader::~Shader_Loader(void){}
+
+std::string Shader_Loader::ReadShader(char *filename)
+{
+
+ std::string shaderCode;
+ std::ifstream file(filename, std::ios::in);
+
+ if (!file.good())
+ {
+ std::cout << "Can't read file " << filename << std::endl;
+ std::terminate();
+ }
+
+ file.seekg(0, std::ios::end);
+ shaderCode.resize((unsigned int)file.tellg());
+ file.seekg(0, std::ios::beg);
+ file.read(&shaderCode[0], shaderCode.size());
+ file.close();
+ return shaderCode;
+}
+
+GLuint Shader_Loader::CreateShader(GLenum shaderType, std::string
+ source, char* shaderName)
+{
+
+ int compile_result = 0;
+
+ GLuint shader = glCreateShader(shaderType);
+ const char *shader_code_ptr = source.c_str();
+ const int shader_code_size = source.size();
+
+ glShaderSource(shader, 1, &shader_code_ptr, &shader_code_size);
+ glCompileShader(shader);
+ glGetShaderiv(shader, GL_COMPILE_STATUS, &compile_result);
+
+ //sprawdz bledy
+ if (compile_result == GL_FALSE)
+ {
+
+ int info_log_length = 0;
+ glGetShaderiv(shader, GL_INFO_LOG_LENGTH, &info_log_length);
+ std::vector shader_log(info_log_length);
+ glGetShaderInfoLog(shader, info_log_length, NULL, &shader_log[0]);
+ std::cout << "ERROR compiling shader: " << shaderName << std::endl << &shader_log[0] << std::endl;
+ return 0;
+ }
+
+ return shader;
+}
+
+GLuint Shader_Loader::CreateProgram(char* vertexShaderFilename,
+ char* fragmentShaderFilename)
+{
+
+ //wczytaj shadery
+ std::string vertex_shader_code = ReadShader(vertexShaderFilename);
+ std::string fragment_shader_code = ReadShader(fragmentShaderFilename);
+
+ GLuint vertex_shader = CreateShader(GL_VERTEX_SHADER, vertex_shader_code, "vertex shader");
+ GLuint fragment_shader = CreateShader(GL_FRAGMENT_SHADER, fragment_shader_code, "fragment shader");
+
+ int link_result = 0;
+ //stworz shader
+ GLuint program = glCreateProgram();
+ glAttachShader(program, vertex_shader);
+ glAttachShader(program, fragment_shader);
+
+ glLinkProgram(program);
+ glGetProgramiv(program, GL_LINK_STATUS, &link_result);
+ //sprawdz bledy w linkerze
+ if (link_result == GL_FALSE)
+ {
+
+ int info_log_length = 0;
+ glGetProgramiv(program, GL_INFO_LOG_LENGTH, &info_log_length);
+ std::vector program_log(info_log_length);
+ glGetProgramInfoLog(program, info_log_length, NULL, &program_log[0]);
+ std::cout << "Shader Loader : LINK ERROR" << std::endl << &program_log[0] << std::endl;
+ return 0;
+ }
+
+ glDetachShader(program, vertex_shader);
+ glDetachShader(program, fragment_shader);
+ glDeleteShader(vertex_shader);
+ glDeleteShader(fragment_shader);
+
+ return program;
+}
+
+void Shader_Loader::DeleteProgram( GLuint program )
+{
+ glDeleteProgram(program);
+}
diff --git a/cw 9/src/Shader_Loader.h b/cw 9/src/Shader_Loader.h
new file mode 100644
index 0000000..1fb9757
--- /dev/null
+++ b/cw 9/src/Shader_Loader.h
@@ -0,0 +1,29 @@
+#pragma once
+
+#include "glew.h"
+#include "freeglut.h"
+#include
+
+namespace Core
+{
+
+ class Shader_Loader
+ {
+ private:
+
+ std::string ReadShader(char *filename);
+ GLuint CreateShader(GLenum shaderType,
+ std::string source,
+ char* shaderName);
+
+ public:
+
+ Shader_Loader(void);
+ ~Shader_Loader(void);
+ GLuint CreateProgram(char* VertexShaderFilename,
+ char* FragmentShaderFilename);
+
+ void DeleteProgram(GLuint program);
+
+ };
+}
\ No newline at end of file
diff --git a/cw 9/src/Texture.cpp b/cw 9/src/Texture.cpp
new file mode 100644
index 0000000..2548b13
--- /dev/null
+++ b/cw 9/src/Texture.cpp
@@ -0,0 +1,37 @@
+#include "Texture.h"
+
+#include
+#include
+#include
+#include
+#include "SOIL/SOIL.h"
+
+typedef unsigned char byte;
+
+GLuint Core::LoadTexture( const char * filepath )
+{
+ GLuint id;
+ glGenTextures(1, &id);
+ glBindTexture(GL_TEXTURE_2D, id);
+ glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR_MIPMAP_LINEAR);
+ glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR_MIPMAP_LINEAR);
+ glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT);
+ glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT);
+
+ int w, h;
+ unsigned char* image = SOIL_load_image(filepath, &w, &h, 0, SOIL_LOAD_RGBA);
+
+
+ glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, w, h, 0, GL_RGBA, GL_UNSIGNED_BYTE, image);
+ glGenerateMipmap(GL_TEXTURE_2D);
+ SOIL_free_image_data(image);
+
+ return id;
+}
+
+void Core::SetActiveTexture(GLuint textureID, const char * shaderVariableName, GLuint programID, int textureUnit)
+{
+ glUniform1i(glGetUniformLocation(programID, shaderVariableName), textureUnit);
+ glActiveTexture(GL_TEXTURE0 + textureUnit);
+ glBindTexture(GL_TEXTURE_2D, textureID);
+}
diff --git a/cw 9/src/Texture.h b/cw 9/src/Texture.h
new file mode 100644
index 0000000..910228e
--- /dev/null
+++ b/cw 9/src/Texture.h
@@ -0,0 +1,15 @@
+#pragma once
+
+#include "glew.h"
+#include "freeglut.h"
+
+namespace Core
+{
+ GLuint LoadTexture(const char * filepath);
+
+ // textureID - identyfikator tekstury otrzymany z funkcji LoadTexture
+ // shaderVariableName - nazwa zmiennej typu 'sampler2D' w shaderze, z ktora ma zostac powiazana tekstura
+ // programID - identyfikator aktualnego programu karty graficznej
+ // textureUnit - indeks jednostki teksturujacej - liczba od 0 do 7. Jezeli uzywa sie wielu tekstur w jednym shaderze, to kazda z nich nalezy powiazac z inna jednostka.
+ void SetActiveTexture(GLuint textureID, const char * shaderVariableName, GLuint programID, int textureUnit);
+}
\ No newline at end of file
diff --git a/cw 9/src/ex_9_1.hpp b/cw 9/src/ex_9_1.hpp
new file mode 100644
index 0000000..36f1cdc
--- /dev/null
+++ b/cw 9/src/ex_9_1.hpp
@@ -0,0 +1,353 @@
+#include "glew.h"
+#include
+#include "glm.hpp"
+#include "ext.hpp"
+#include
+#include
+
+#include "Shader_Loader.h"
+#include "Render_Utils.h"
+//#include "Texture.h"
+
+#include "Box.cpp"
+#include
+#include
+#include
+#include
+
+const unsigned int SHADOW_WIDTH = 1024, SHADOW_HEIGHT = 1024;
+
+int WIDTH = 500, HEIGHT = 500;
+
+namespace models {
+ Core::RenderContext bedContext;
+ Core::RenderContext chairContext;
+ Core::RenderContext deskContext;
+ Core::RenderContext doorContext;
+ Core::RenderContext drawerContext;
+ Core::RenderContext marbleBustContext;
+ Core::RenderContext materaceContext;
+ Core::RenderContext pencilsContext;
+ Core::RenderContext planeContext;
+ Core::RenderContext roomContext;
+ Core::RenderContext spaceshipContext;
+ Core::RenderContext sphereContext;
+ Core::RenderContext windowContext;
+ Core::RenderContext testContext;
+}
+
+GLuint depthMapFBO;
+GLuint depthMap;
+
+GLuint program;
+GLuint programSun;
+GLuint programTest;
+GLuint programTex;
+
+Core::Shader_Loader shaderLoader;
+
+Core::RenderContext shipContext;
+Core::RenderContext sphereContext;
+
+glm::vec3 sunPos = glm::vec3(-4.740971f, 2.149999f, 0.369280f);
+glm::vec3 sunDir = glm::vec3(-0.93633f, 0.351106, 0.003226f);
+glm::vec3 sunColor = glm::vec3(0.9f, 0.9f, 0.7f)*5;
+
+glm::vec3 cameraPos = glm::vec3(0.479490f, 1.250000f, -2.124680f);
+glm::vec3 cameraDir = glm::vec3(-0.354510f, 0.000000f, 0.935054f);
+
+
+glm::vec3 spaceshipPos = glm::vec3(0.065808f, 1.250000f, -2.189549f);
+glm::vec3 spaceshipDir = glm::vec3(-0.490263f, 0.000000f, 0.871578f);
+GLuint VAO,VBO;
+
+float aspectRatio = 1.f;
+
+float exposition = 1.f;
+
+glm::vec3 pointlightPos = glm::vec3(0, 2, 0);
+glm::vec3 pointlightColor = glm::vec3(0.9, 0.6, 0.6);
+
+glm::vec3 spotlightPos = glm::vec3(0, 0, 0);
+glm::vec3 spotlightConeDir = glm::vec3(0, 0, 0);
+glm::vec3 spotlightColor = glm::vec3(0.4, 0.4, 0.9)*3;
+float spotlightPhi = 3.14 / 4;
+
+
+
+float lastTime = -1.f;
+float deltaTime = 0.f;
+
+void updateDeltaTime(float time) {
+ if (lastTime < 0) {
+ lastTime = time;
+ return;
+ }
+
+ deltaTime = time - lastTime;
+ if (deltaTime > 0.1) deltaTime = 0.1;
+ lastTime = time;
+}
+glm::mat4 createCameraMatrix()
+{
+ glm::vec3 cameraSide = glm::normalize(glm::cross(cameraDir,glm::vec3(0.f,1.f,0.f)));
+ glm::vec3 cameraUp = glm::normalize(glm::cross(cameraSide,cameraDir));
+ glm::mat4 cameraRotrationMatrix = glm::mat4({
+ cameraSide.x,cameraSide.y,cameraSide.z,0,
+ cameraUp.x,cameraUp.y,cameraUp.z ,0,
+ -cameraDir.x,-cameraDir.y,-cameraDir.z,0,
+ 0.,0.,0.,1.,
+ });
+ cameraRotrationMatrix = glm::transpose(cameraRotrationMatrix);
+ glm::mat4 cameraMatrix = cameraRotrationMatrix * glm::translate(-cameraPos);
+
+ return cameraMatrix;
+}
+
+glm::mat4 createPerspectiveMatrix()
+{
+
+ glm::mat4 perspectiveMatrix;
+ float n = 0.05;
+ float f = 20.;
+ float a1 = glm::min(aspectRatio, 1.f);
+ float a2 = glm::min(1 / aspectRatio, 1.f);
+ perspectiveMatrix = glm::mat4({
+ 1,0.,0.,0.,
+ 0.,aspectRatio,0.,0.,
+ 0.,0.,(f+n) / (n - f),2*f * n / (n - f),
+ 0.,0.,-1.,0.,
+ });
+
+
+ perspectiveMatrix=glm::transpose(perspectiveMatrix);
+
+ return perspectiveMatrix;
+}
+
+void drawObjectPBR(Core::RenderContext& context, glm::mat4 modelMatrix, glm::vec3 color, float roughness, float metallic) {
+
+ glm::mat4 viewProjectionMatrix = createPerspectiveMatrix() * createCameraMatrix();
+ glm::mat4 transformation = viewProjectionMatrix * modelMatrix;
+ glUniformMatrix4fv(glGetUniformLocation(program, "transformation"), 1, GL_FALSE, (float*)&transformation);
+ glUniformMatrix4fv(glGetUniformLocation(program, "modelMatrix"), 1, GL_FALSE, (float*)&modelMatrix);
+
+ glUniform1f(glGetUniformLocation(program, "exposition"), exposition);
+
+ glUniform1f(glGetUniformLocation(program, "roughness"), roughness);
+ glUniform1f(glGetUniformLocation(program, "metallic"), metallic);
+
+ glUniform3f(glGetUniformLocation(program, "color"), color.x, color.y, color.z);
+
+ glUniform3f(glGetUniformLocation(program, "cameraPos"), cameraPos.x, cameraPos.y, cameraPos.z);
+
+ glUniform3f(glGetUniformLocation(program, "sunDir"), sunDir.x, sunDir.y, sunDir.z);
+ glUniform3f(glGetUniformLocation(program, "sunColor"), sunColor.x, sunColor.y, sunColor.z);
+
+ glUniform3f(glGetUniformLocation(program, "lightPos"), pointlightPos.x, pointlightPos.y, pointlightPos.z);
+ glUniform3f(glGetUniformLocation(program, "lightColor"), pointlightColor.x, pointlightColor.y, pointlightColor.z);
+
+ glUniform3f(glGetUniformLocation(program, "spotlightConeDir"), spotlightConeDir.x, spotlightConeDir.y, spotlightConeDir.z);
+ glUniform3f(glGetUniformLocation(program, "spotlightPos"), spotlightPos.x, spotlightPos.y, spotlightPos.z);
+ glUniform3f(glGetUniformLocation(program, "spotlightColor"), spotlightColor.x, spotlightColor.y, spotlightColor.z);
+ glUniform1f(glGetUniformLocation(program, "spotlightPhi"), spotlightPhi);
+ Core::DrawContext(context);
+
+}
+
+void renderShadowapSun() {
+ float time = glfwGetTime();
+ glViewport(0, 0, SHADOW_WIDTH, SHADOW_HEIGHT);
+ //uzupelnij o renderowanie glebokosci do tekstury
+
+
+
+
+
+
+ glBindFramebuffer(GL_FRAMEBUFFER, 0);
+ glViewport(0, 0, WIDTH, HEIGHT);
+}
+
+void renderScene(GLFWwindow* window)
+{
+ glClearColor(0.4f, 0.4f, 0.8f, 1.0f);
+ glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
+ float time = glfwGetTime();
+ updateDeltaTime(time);
+ renderShadowapSun();
+
+ //space lamp
+ glUseProgram(programSun);
+ glm::mat4 viewProjectionMatrix = createPerspectiveMatrix() * createCameraMatrix();
+ glm::mat4 transformation = viewProjectionMatrix * glm::translate(pointlightPos) * glm::scale(glm::vec3(0.1));
+ glUniformMatrix4fv(glGetUniformLocation(programSun, "transformation"), 1, GL_FALSE, (float*)&transformation);
+ glUniform3f(glGetUniformLocation(programSun, "color"), sunColor.x / 2, sunColor.y / 2, sunColor.z / 2);
+ glUniform1f(glGetUniformLocation(programSun, "exposition"), exposition);
+ Core::DrawContext(sphereContext);
+
+ glUseProgram(program);
+
+ drawObjectPBR(sphereContext, glm::translate(pointlightPos) * glm::scale(glm::vec3(0.1)) * glm::eulerAngleY(time / 3) * glm::translate(glm::vec3(4.f, 0, 0)) * glm::scale(glm::vec3(0.3f)), glm::vec3(0.2, 0.7, 0.3), 0.3, 0.0);
+
+ drawObjectPBR(sphereContext,
+ glm::translate(pointlightPos) * glm::scale(glm::vec3(0.1)) * glm::eulerAngleY(time / 3) * glm::translate(glm::vec3(4.f, 0, 0)) * glm::eulerAngleY(time) * glm::translate(glm::vec3(1.f, 0, 0)) * glm::scale(glm::vec3(0.1f)),
+ glm::vec3(0.5, 0.5, 0.5), 0.7, 0.0);
+
+ drawObjectPBR(models::bedContext, glm::mat4(), glm::vec3(0.03f, 0.03f, 0.03f), 0.2f, 0.0f);
+ drawObjectPBR(models::chairContext, glm::mat4(), glm::vec3(0.195239f, 0.37728f, 0.8f), 0.4f, 0.0f);
+ drawObjectPBR(models::deskContext, glm::mat4(), glm::vec3(0.428691f, 0.08022f, 0.036889f), 0.2f, 0.0f);
+ drawObjectPBR(models::doorContext, glm::mat4(), glm::vec3(0.402978f, 0.120509f, 0.057729f), 0.2f, 0.0f);
+ drawObjectPBR(models::drawerContext, glm::mat4(), glm::vec3(0.428691f, 0.08022f, 0.036889f), 0.2f, 0.0f);
+ drawObjectPBR(models::marbleBustContext, glm::mat4(), glm::vec3(1.f, 1.f, 1.f), 0.5f, 1.0f);
+ drawObjectPBR(models::materaceContext, glm::mat4(), glm::vec3(0.9f, 0.9f, 0.9f), 0.8f, 0.0f);
+ drawObjectPBR(models::pencilsContext, glm::mat4(), glm::vec3(0.10039f, 0.018356f, 0.001935f), 0.1f, 0.0f);
+ drawObjectPBR(models::planeContext, glm::mat4(), glm::vec3(0.402978f, 0.120509f, 0.057729f), 0.2f, 0.0f);
+ drawObjectPBR(models::roomContext, glm::mat4(), glm::vec3(0.9f, 0.9f, 0.9f), 0.8f, 0.0f);
+ drawObjectPBR(models::windowContext, glm::mat4(), glm::vec3(0.402978f, 0.120509f, 0.057729f), 0.2f, 0.0f);
+
+ glm::vec3 spaceshipSide = glm::normalize(glm::cross(spaceshipDir, glm::vec3(0.f, 1.f, 0.f)));
+ glm::vec3 spaceshipUp = glm::normalize(glm::cross(spaceshipSide, spaceshipDir));
+ glm::mat4 specshipCameraRotrationMatrix = glm::mat4({
+ spaceshipSide.x,spaceshipSide.y,spaceshipSide.z,0,
+ spaceshipUp.x,spaceshipUp.y,spaceshipUp.z ,0,
+ -spaceshipDir.x,-spaceshipDir.y,-spaceshipDir.z,0,
+ 0.,0.,0.,1.,
+ });
+
+
+ //drawObjectColor(shipContext,
+ // glm::translate(cameraPos + 1.5 * cameraDir + cameraUp * -0.5f) * inveseCameraRotrationMatrix * glm::eulerAngleY(glm::pi()),
+ // glm::vec3(0.3, 0.3, 0.5)
+ // );
+ drawObjectPBR(shipContext,
+ glm::translate(spaceshipPos) * specshipCameraRotrationMatrix * glm::eulerAngleY(glm::pi()) * glm::scale(glm::vec3(0.03f)),
+ glm::vec3(0.3, 0.3, 0.5),
+ 0.2,1.0
+ );
+
+ spotlightPos = spaceshipPos + 0.2 * spaceshipDir;
+ spotlightConeDir = spaceshipDir;
+
+
+
+ //test depth buffer
+ //glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
+ //glUseProgram(programTest);
+ //glActiveTexture(GL_TEXTURE0);
+ //glBindTexture(GL_TEXTURE_2D, depthMap);
+ //Core::DrawContext(models::testContext);
+
+ glUseProgram(0);
+ glfwSwapBuffers(window);
+}
+void framebuffer_size_callback(GLFWwindow* window, int width, int height)
+{
+ aspectRatio = width / float(height);
+ glViewport(0, 0, width, height);
+ WIDTH = width;
+ HEIGHT = height;
+}
+void loadModelToContext(std::string path, Core::RenderContext& context)
+{
+ Assimp::Importer import;
+ const aiScene* scene = import.ReadFile(path, aiProcess_Triangulate | aiProcess_CalcTangentSpace);
+
+ if (!scene || scene->mFlags & AI_SCENE_FLAGS_INCOMPLETE || !scene->mRootNode)
+ {
+ std::cout << "ERROR::ASSIMP::" << import.GetErrorString() << std::endl;
+ return;
+ }
+ context.initFromAssimpMesh(scene->mMeshes[0]);
+}
+
+void init(GLFWwindow* window)
+{
+ glfwSetFramebufferSizeCallback(window, framebuffer_size_callback);
+
+ glEnable(GL_DEPTH_TEST);
+ program = shaderLoader.CreateProgram("shaders/shader_9_1.vert", "shaders/shader_9_1.frag");
+ programTest = shaderLoader.CreateProgram("shaders/test.vert", "shaders/test.frag");
+ programSun = shaderLoader.CreateProgram("shaders/shader_8_sun.vert", "shaders/shader_8_sun.frag");
+
+ loadModelToContext("./models/sphere.obj", sphereContext);
+ loadModelToContext("./models/spaceship.obj", shipContext);
+
+
+ loadModelToContext("./models/bed.obj", models::bedContext);
+ loadModelToContext("./models/chair.obj", models::chairContext);
+ loadModelToContext("./models/desk.obj", models::deskContext);
+ loadModelToContext("./models/door.obj", models::doorContext);
+ loadModelToContext("./models/drawer.obj", models::drawerContext);
+ loadModelToContext("./models/marbleBust.obj", models::marbleBustContext);
+ loadModelToContext("./models/materace.obj", models::materaceContext);
+ loadModelToContext("./models/pencils.obj", models::pencilsContext);
+ loadModelToContext("./models/plane.obj", models::planeContext);
+ loadModelToContext("./models/room.obj", models::roomContext);
+ loadModelToContext("./models/spaceship.obj", models::spaceshipContext);
+ loadModelToContext("./models/sphere.obj", models::sphereContext);
+ loadModelToContext("./models/window.obj", models::windowContext);
+ loadModelToContext("./models/test.obj", models::testContext);
+
+}
+
+void shutdown(GLFWwindow* window)
+{
+ shaderLoader.DeleteProgram(program);
+}
+
+//obsluga wejscia
+void processInput(GLFWwindow* window)
+{
+ glm::vec3 spaceshipSide = glm::normalize(glm::cross(spaceshipDir, glm::vec3(0.f,1.f,0.f)));
+ glm::vec3 spaceshipUp = glm::vec3(0.f, 1.f, 0.f);
+ float angleSpeed = 0.05f * deltaTime * 60;
+ float moveSpeed = 0.05f * deltaTime * 60;
+ if (glfwGetKey(window, GLFW_KEY_ESCAPE) == GLFW_PRESS) {
+ glfwSetWindowShouldClose(window, true);
+ }
+ if (glfwGetKey(window, GLFW_KEY_W) == GLFW_PRESS)
+ spaceshipPos += spaceshipDir * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_S) == GLFW_PRESS)
+ spaceshipPos -= spaceshipDir * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_X) == GLFW_PRESS)
+ spaceshipPos += spaceshipSide * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_Z) == GLFW_PRESS)
+ spaceshipPos -= spaceshipSide * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_Q) == GLFW_PRESS)
+ spaceshipPos += spaceshipUp * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_E) == GLFW_PRESS)
+ spaceshipPos -= spaceshipUp * moveSpeed;
+ if (glfwGetKey(window, GLFW_KEY_A) == GLFW_PRESS)
+ spaceshipDir = glm::vec3(glm::eulerAngleY(angleSpeed) * glm::vec4(spaceshipDir, 0));
+ if (glfwGetKey(window, GLFW_KEY_D) == GLFW_PRESS)
+ spaceshipDir = glm::vec3(glm::eulerAngleY(-angleSpeed) * glm::vec4(spaceshipDir, 0));
+
+ cameraPos = spaceshipPos - 0.5 * spaceshipDir + glm::vec3(0, 1, 0) * 0.2f;
+ cameraDir = spaceshipDir;
+
+ if (glfwGetKey(window, GLFW_KEY_1) == GLFW_PRESS)
+ exposition -= 0.05;
+ if (glfwGetKey(window, GLFW_KEY_2) == GLFW_PRESS)
+ exposition += 0.05;
+
+ if (glfwGetKey(window, GLFW_KEY_3) == GLFW_PRESS) {
+ printf("spaceshipPos = glm::vec3(%ff, %ff, %ff);\n", spaceshipPos.x, spaceshipPos.y, spaceshipPos.z);
+ printf("spaceshipDir = glm::vec3(%ff, %ff, %ff);\n", spaceshipDir.x, spaceshipDir.y, spaceshipDir.z);
+ }
+
+ //cameraDir = glm::normalize(-cameraPos);
+
+}
+
+// funkcja jest glowna petla
+void renderLoop(GLFWwindow* window) {
+ while (!glfwWindowShouldClose(window))
+ {
+ processInput(window);
+
+ renderScene(window);
+ glfwPollEvents();
+ }
+}
+//}
\ No newline at end of file
diff --git a/cw 9/src/main.cpp b/cw 9/src/main.cpp
new file mode 100644
index 0000000..4e939cd
--- /dev/null
+++ b/cw 9/src/main.cpp
@@ -0,0 +1,47 @@
+#include "glew.h"
+
+#include
+#include "glm.hpp"
+#include "ext.hpp"
+#include
+#include
+
+#include "ex_9_1.hpp"
+
+
+
+int main(int argc, char** argv)
+{
+ // inicjalizacja glfw
+ glfwInit();
+ glfwWindowHint(GLFW_CONTEXT_VERSION_MAJOR, 3);
+ glfwWindowHint(GLFW_CONTEXT_VERSION_MINOR, 3);
+ glfwWindowHint(GLFW_OPENGL_PROFILE, GLFW_OPENGL_CORE_PROFILE);
+
+#ifdef __APPLE__
+ glfwWindowHint(GLFW_OPENGL_FORWARD_COMPAT, GL_TRUE);
+#endif
+
+ // tworzenie okna za pomoca glfw
+ GLFWwindow* window = glfwCreateWindow(500, 500, "FirstWindow", NULL, NULL);
+ if (window == NULL)
+ {
+ std::cout << "Failed to create GLFW window" << std::endl;
+ glfwTerminate();
+ return -1;
+ }
+ glfwMakeContextCurrent(window);
+
+ // ladowanie OpenGL za pomoca glew
+ glewInit();
+ glViewport(0, 0, 500, 500);
+
+ init(window);
+
+ // uruchomienie glownej petli
+ renderLoop(window);
+
+ shutdown(window);
+ glfwTerminate();
+ return 0;
+}
diff --git a/cw 9/src/objload.h b/cw 9/src/objload.h
new file mode 100644
index 0000000..004ab4e
--- /dev/null
+++ b/cw 9/src/objload.h
@@ -0,0 +1,291 @@
+/* Copyright (c) 2012, Gerhard Reitmayr
+ All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
+ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */
+
+#ifndef OBJLOAD_H_
+#define OBJLOAD_H_
+
+#include
+#include
+#include
+#include
+#include
+#include
Na powyższym obrazku możesz zobaczyć, że dużo łatwiej jest okreśłić położenie kostek, gdy rzucają one cienie. Podstawową techniką generowania cieni jest shadow mapping i wiele bardziej zaawansowanych technik na niej bazuje.