Merge remote-tracking branch 'origin/master'

# Conflicts:
#	README.md
#	libs/raylib/.gitignore
#	libs/raylib/BINDINGS.md
#	libs/raylib/CHANGELOG
#	libs/raylib/CMakeLists.txt
#	libs/raylib/CONTRIBUTING.md
#	libs/raylib/HISTORY.md
#	libs/raylib/LICENSE
#	libs/raylib/README.md
#	libs/raylib/ROADMAP.md
#	libs/raylib/SPONSORS.md
#	libs/raylib/raylib.pc.in
#	libs/raylib/src/CMakeLists.txt
#	libs/raylib/src/Makefile
#	libs/raylib/src/camera.h
#	libs/raylib/src/config.h
#	libs/raylib/src/core.c
#	libs/raylib/src/easings.h
#	libs/raylib/src/external/cgltf.h
#	libs/raylib/src/external/dr_flac.h
#	libs/raylib/src/external/dr_mp3.h
#	libs/raylib/src/external/dr_wav.h
#	libs/raylib/src/external/glfw/CMake/glfw3.pc.in
#	libs/raylib/src/external/glfw/CMake/glfw3Config.cmake.in
#	libs/raylib/src/external/glfw/CMakeLists.txt
#	libs/raylib/src/external/glfw/README.md
#	libs/raylib/src/external/glfw/include/GLFW/glfw3.h
#	libs/raylib/src/external/glfw/src/cocoa_init.m
#	libs/raylib/src/external/glfw/src/cocoa_joystick.h
#	libs/raylib/src/external/glfw/src/cocoa_joystick.m
#	libs/raylib/src/external/glfw/src/cocoa_monitor.m
#	libs/raylib/src/external/glfw/src/cocoa_platform.h
#	libs/raylib/src/external/glfw/src/cocoa_window.m
#	libs/raylib/src/external/glfw/src/egl_context.c
#	libs/raylib/src/external/glfw/src/egl_context.h
#	libs/raylib/src/external/glfw/src/glfw_config.h.in
#	libs/raylib/src/external/glfw/src/init.c
#	libs/raylib/src/external/glfw/src/input.c
#	libs/raylib/src/external/glfw/src/internal.h
#	libs/raylib/src/external/glfw/src/linux_joystick.c
#	libs/raylib/src/external/glfw/src/linux_joystick.h
#	libs/raylib/src/external/glfw/src/mappings.h
#	libs/raylib/src/external/glfw/src/nsgl_context.m
#	libs/raylib/src/external/glfw/src/null_init.c
#	libs/raylib/src/external/glfw/src/null_joystick.c
#	libs/raylib/src/external/glfw/src/null_monitor.c
#	libs/raylib/src/external/glfw/src/null_platform.h
#	libs/raylib/src/external/glfw/src/null_window.c
#	libs/raylib/src/external/glfw/src/osmesa_context.h
#	libs/raylib/src/external/glfw/src/wgl_context.h
#	libs/raylib/src/external/glfw/src/win32_init.c
#	libs/raylib/src/external/glfw/src/win32_joystick.c
#	libs/raylib/src/external/glfw/src/win32_joystick.h
#	libs/raylib/src/external/glfw/src/win32_monitor.c
#	libs/raylib/src/external/glfw/src/win32_platform.h
#	libs/raylib/src/external/glfw/src/win32_window.c
#	libs/raylib/src/external/glfw/src/window.c
#	libs/raylib/src/external/glfw/src/wl_init.c
#	libs/raylib/src/external/glfw/src/wl_monitor.c
#	libs/raylib/src/external/glfw/src/wl_platform.h
#	libs/raylib/src/external/glfw/src/wl_window.c
#	libs/raylib/src/external/glfw/src/x11_init.c
#	libs/raylib/src/external/glfw/src/x11_monitor.c
#	libs/raylib/src/external/glfw/src/x11_platform.h
#	libs/raylib/src/external/glfw/src/x11_window.c
#	libs/raylib/src/external/jar_xm.h
#	libs/raylib/src/external/miniaudio.h
#	libs/raylib/src/external/stb_image.h
#	libs/raylib/src/external/stb_image_write.h
#	libs/raylib/src/external/stb_vorbis.h
#	libs/raylib/src/external/tinyobj_loader_c.h
#	libs/raylib/src/gestures.h
#	libs/raylib/src/models.c
#	libs/raylib/src/physac.h
#	libs/raylib/src/raudio.c
#	libs/raylib/src/raudio.h
#	libs/raylib/src/raylib.dll.rc
#	libs/raylib/src/raylib.dll.rc.data
#	libs/raylib/src/raylib.h
#	libs/raylib/src/raylib.rc
#	libs/raylib/src/raylib.rc.data
#	libs/raylib/src/raymath.h
#	libs/raylib/src/rglfw.c
#	libs/raylib/src/rlgl.h
#	libs/raylib/src/rmem.h
#	libs/raylib/src/rnet.h
#	libs/raylib/src/shapes.c
#	libs/raylib/src/text.c
#	libs/raylib/src/textures.c
#	libs/raylib/src/utils.c
#	libs/raylib/src/utils.h

4 files changed, 1888 insertions, 0 deletions

diff --git a/libs/raylib/src/external/msf_gif.h b/libs/raylib/src/external/msf_gif.h
new file mode 100644
index 0000000..348b1b1
--- /dev/null
+++ b/libs/raylib/src/external/msf_gif.h
@@ -0,0 +1,608 @@
+/*
+HOW TO USE:
+
+    In exactly one translation unit (.c or .cpp file), #define MSF_GIF_IMPL before including the header, like so:
+
+    #define MSF_GIF_IMPL
+    #include "msf_gif.h"
+
+    Everywhere else, just include the header like normal.
+
+
+USAGE EXAMPLE:
+
+    int width = 480, height = 320, centisecondsPerFrame = 5, bitDepth = 16;
+    MsfGifState gifState = {};
+    msf_gif_begin(&gifState, width, height);
+    msf_gif_frame(&gifState, ..., centisecondsPerFrame, bitDepth, width * 4); //frame 1
+    msf_gif_frame(&gifState, ..., centisecondsPerFrame, bitDepth, width * 4); //frame 2
+    msf_gif_frame(&gifState, ..., centisecondsPerFrame, bitDepth, width * 4); //frame 3, etc...
+    MsfGifResult result = msf_gif_end(&gifState);
+    FILE * fp = fopen("MyGif.gif", "wb");
+    fwrite(result.data, result.dataSize, 1, fp);
+    fclose(fp);
+    msf_gif_free(result);
+
+Detailed function documentation can be found in the header section below.
+
+
+REPLACING MALLOC:
+
+    This library uses malloc+realloc+free internally for memory allocation.
+    To facilitate integration with custom memory allocators, these calls go through macros, which can be redefined.
+    The expected function signature equivalents of the macros are as follows:
+
+    void * MSF_GIF_MALLOC(void * context, size_t newSize)
+    void * MSF_GIF_REALLOC(void * context, void * oldMemory, size_t oldSize, size_t newSize)
+    void MSF_GIF_FREE(void * context, void * oldMemory, size_t oldSize)
+
+    If your allocator needs a context pointer, you can set the `customAllocatorContext` field of the MsfGifState struct
+    before calling msf_gif_begin(), and it will be passed to all subsequent allocator macro calls.
+
+See end of file for license information.
+*/
+
+//version 2.1
+
+#ifndef MSF_GIF_H
+#define MSF_GIF_H
+
+#include <stdint.h>
+#include <stddef.h>
+
+typedef struct {
+    void * data;
+    size_t dataSize;
+
+    size_t allocSize; //internal use
+    void * contextPointer; //internal use
+} MsfGifResult;
+
+typedef struct { //internal use
+    uint32_t * pixels;
+    int depth, count, rbits, gbits, bbits;
+} MsfCookedFrame;
+
+typedef struct {
+    MsfCookedFrame previousFrame;
+    uint8_t * listHead;
+    uint8_t * listTail;
+    int width, height;
+    void * customAllocatorContext;
+} MsfGifState;
+
+#ifdef __cplusplus
+extern "C" {
+#endif //__cplusplus
+
+/**
+ * @param width                Image width in pixels.
+ * @param height               Image height in pixels.
+ * @return                     Non-zero on success, 0 on error.
+ */
+int msf_gif_begin(MsfGifState * handle, int width, int height);
+
+/**
+ * @param pixelData            Pointer to raw framebuffer data. Rows must be contiguous in memory, in RGBA8 format.
+ *                             Note: This function does NOT free `pixelData`. You must free it yourself afterwards.
+ * @param centiSecondsPerFrame How many hundredths of a second this frame should be displayed for.
+ *                             Note: This being specified in centiseconds is a limitation of the GIF format.
+ * @param maxBitDepth          Limits how many bits per pixel can be used when quantizing the gif.
+ *                             The actual bit depth chosen for a given frame will be less than or equal to
+ *                             the supplied maximum, depending on the variety of colors used in the frame.
+ *                             `maxBitDepth` will be clamped between 1 and 16. The recommended default is 16.
+ *                             Lowering this value can result in faster exports and smaller gifs,
+ *                             but the quality may suffer.
+ *                             Please experiment with this value to find what works best for your application.
+ * @param pitchInBytes         The number of bytes from the beginning of one row of pixels to the beginning of the next.
+ *                             If you want to flip the image, just pass in a negative pitch.
+ * @return                     Non-zero on success, 0 on error.
+ */
+int msf_gif_frame(MsfGifState * handle, uint8_t * pixelData, int centiSecondsPerFame, int maxBitDepth, int pitchInBytes);
+
+/**
+ * @return                     A block of memory containing the gif file data, or NULL on error.
+ *                             You are responsible for freeing this via `msf_gif_free()`.
+ */
+MsfGifResult msf_gif_end(MsfGifState * handle);
+
+/**
+ * @param result                The MsfGifResult struct, verbatim as it was returned from `msf_gif_end()`.
+ */
+void msf_gif_free(MsfGifResult result);
+
+#ifdef __cplusplus
+}
+#endif //__cplusplus
+
+#endif //MSF_GIF_H
+
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+/// IMPLEMENTATION                                                                                                   ///
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+
+#ifdef MSF_GIF_IMPL
+#ifndef MSF_GIF_ALREADY_IMPLEMENTED_IN_THIS_TRANSLATION_UNIT
+#define MSF_GIF_ALREADY_IMPLEMENTED_IN_THIS_TRANSLATION_UNIT
+
+#ifndef MSF_GIF_BUFFER_INIT_SIZE
+#define MSF_GIF_BUFFER_INIT_SIZE 1024 * 1024 * 4 //4MB by default, you can increase this if you want to realloc less
+#endif
+
+//ensure the library user has either defined all of malloc/realloc/free, or none
+#if defined(MSF_GIF_MALLOC) && defined(MSF_GIF_REALLOC) && defined(MSF_GIF_FREE) //ok
+#elif !defined(MSF_GIF_MALLOC) && !defined(MSF_GIF_REALLOC) && !defined(MSF_GIF_FREE) //ok
+#else
+#error "You must either define all of MSF_GIF_MALLOC, MSF_GIF_REALLOC, and MSF_GIF_FREE, or define none of them"
+#endif
+
+//provide default allocator definitions that redirect to the standard global allocator
+#if !defined(MSF_GIF_MALLOC)
+#include <stdlib.h> //malloc, etc.
+#define MSF_GIF_MALLOC(contextPointer, newSize) malloc(newSize)
+#define MSF_GIF_REALLOC(contextPointer, oldMemory, oldSize, newSize) realloc(oldMemory, newSize)
+#define MSF_GIF_FREE(contextPointer, oldMemory, oldSize) free(oldMemory)
+#endif
+
+//instrumentation for capturing profiling traces (useless for the library user, but useful for the library author)
+#ifdef MSF_GIF_ENABLE_TRACING
+#define MsfTimeFunc TimeFunc
+#define MsfTimeLoop TimeLoop
+#define msf_init_profiling_thread init_profiling_thread
+#else
+#define MsfTimeFunc
+#define MsfTimeLoop(name)
+#define msf_init_profiling_thread()
+#endif //MSF_GIF_ENABLE_TRACING
+
+#include <string.h> //memcpy
+
+//TODO: use compiler-specific notation to force-inline functions currently marked inline
+#if defined(__GNUC__) //gcc, clang
+static inline int msf_bit_log(int i) { return 32 - __builtin_clz(i); }
+#elif defined(_MSC_VER) //msvc
+#include <intrin.h>
+static inline int msf_bit_log(int i) { unsigned long idx; _BitScanReverse(&idx, i); return idx + 1; }
+#else //fallback implementation for other compilers
+//from https://stackoverflow.com/a/31718095/3064745 - thanks!
+static inline int msf_bit_log(int i) {
+    static const int MultiplyDeBruijnBitPosition[32] = {
+        0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30,
+        8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31,
+    };
+    i |= i >> 1;
+    i |= i >> 2;
+    i |= i >> 4;
+    i |= i >> 8;
+    i |= i >> 16;
+    return MultiplyDeBruijnBitPosition[(uint32_t)(i * 0x07C4ACDDU) >> 27] + 1;
+}
+#endif
+static inline int msf_imin(int a, int b) { return a < b? a : b; }
+static inline int msf_imax(int a, int b) { return b < a? a : b; }
+
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+/// Frame Cooking                                                                                                    ///
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+
+#if (defined (__SSE2__) || defined (_M_X64) || _M_IX86_FP == 2) && !defined(MSF_GIF_NO_SSE2)
+#include <emmintrin.h>
+#endif
+
+static MsfCookedFrame msf_cook_frame(void * allocContext, uint8_t * raw, uint8_t * used,
+                                     int width, int height, int pitch, int depth)
+{ MsfTimeFunc
+    //bit depth for each channel
+    const static int rdepths[17] = { 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5 };
+    const static int gdepths[17] = { 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6 };
+    const static int bdepths[17] = { 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5 };
+
+    const static int ditherKernel[16] = {
+         0 << 12,  8 << 12,  2 << 12, 10 << 12,
+        12 << 12,  4 << 12, 14 << 12,  6 << 12,
+         3 << 12, 11 << 12,  1 << 12,  9 << 12,
+        15 << 12,  7 << 12, 13 << 12,  5 << 12,
+    };
+
+    uint32_t * cooked = (uint32_t *) MSF_GIF_MALLOC(allocContext, width * height * sizeof(uint32_t));
+    if (!cooked) { MsfCookedFrame blank = {0}; return blank; }
+
+    int count = 0;
+    MsfTimeLoop("do") do {
+        int rbits = rdepths[depth], gbits = gdepths[depth], bbits = bdepths[depth];
+        int paletteSize = 1 << (rbits + gbits + bbits);
+        memset(used, 0, paletteSize * sizeof(uint8_t));
+
+        //TODO: document what this math does and why it's correct
+        int rdiff = (1 << (8 - rbits)) - 1;
+        int gdiff = (1 << (8 - gbits)) - 1;
+        int bdiff = (1 << (8 - bbits)) - 1;
+        short rmul = (short) ((255.0f - rdiff) / 255.0f * 257);
+        short gmul = (short) ((255.0f - gdiff) / 255.0f * 257);
+        short bmul = (short) ((255.0f - bdiff) / 255.0f * 257);
+
+        int gmask = ((1 << gbits) - 1) << rbits;
+        int bmask = ((1 << bbits) - 1) << rbits << gbits;
+
+        MsfTimeLoop("cook") for (int y = 0; y < height; ++y) {
+            int x = 0;
+
+            #if (defined (__SSE2__) || defined (_M_X64) || _M_IX86_FP == 2) && !defined(MSF_GIF_NO_SSE2)
+                __m128i k = _mm_loadu_si128((__m128i *) &ditherKernel[(y & 3) * 4]);
+                __m128i k2 = _mm_or_si128(_mm_srli_epi32(k, rbits), _mm_slli_epi32(_mm_srli_epi32(k, bbits), 16));
+                // MsfTimeLoop("SIMD")
+                for (; x < width - 3; x += 4) {
+                    uint8_t * pixels = &raw[y * pitch + x * 4];
+                    __m128i p = _mm_loadu_si128((__m128i *) pixels);
+
+                    __m128i rb = _mm_and_si128(p, _mm_set1_epi32(0x00FF00FF));
+                    __m128i rb1 = _mm_mullo_epi16(rb, _mm_set_epi16(bmul, rmul, bmul, rmul, bmul, rmul, bmul, rmul));
+                    __m128i rb2 = _mm_adds_epu16(rb1, k2);
+                    __m128i r3 = _mm_srli_epi32(_mm_and_si128(rb2, _mm_set1_epi32(0x0000FFFF)), 16 - rbits);
+                    __m128i b3 = _mm_and_si128(_mm_srli_epi32(rb2, 32 - rbits - gbits - bbits), _mm_set1_epi32(bmask));
+
+                    __m128i g = _mm_and_si128(_mm_srli_epi32(p, 8), _mm_set1_epi32(0x000000FF));
+                    __m128i g1 = _mm_mullo_epi16(g, _mm_set1_epi32(gmul));
+                    __m128i g2 = _mm_adds_epu16(g1, _mm_srli_epi32(k, gbits));
+                    __m128i g3 = _mm_and_si128(_mm_srli_epi32(g2, 16 - rbits - gbits), _mm_set1_epi32(gmask));
+
+                    //TODO: does storing this as a __m128i then reading it back as a uint32_t violate strict aliasing?
+                    uint32_t * c = &cooked[y * width + x];
+                    __m128i out = _mm_or_si128(_mm_or_si128(r3, g3), b3);
+                    _mm_storeu_si128((__m128i *) c, out);
+                }
+            #endif
+
+            //scalar cleanup loop
+            // MsfTimeLoop("scalar")
+            for (; x < width; ++x) {
+                uint8_t * p = &raw[y * pitch + x * 4];
+                int dx = x & 3, dy = y & 3;
+                int k = ditherKernel[dy * 4 + dx];
+                cooked[y * width + x] =
+                    (msf_imin(65535, p[2] * bmul + (k >> bbits)) >> (16 - rbits - gbits - bbits) & bmask) |
+                    (msf_imin(65535, p[1] * gmul + (k >> gbits)) >> (16 - rbits - gbits        ) & gmask) |
+                     msf_imin(65535, p[0] * rmul + (k >> rbits)) >> (16 - rbits                );
+            }
+        }
+
+        count = 0;
+        MsfTimeLoop("mark and count") for (int i = 0; i < width * height; ++i) {
+            used[cooked[i]] = 1;
+        }
+
+        //count used colors
+        MsfTimeLoop("count") for (int j = 0; j < paletteSize; ++j) {
+            count += used[j];
+        }
+    } while (count >= 256 && --depth);
+
+    MsfCookedFrame ret = { cooked, depth, count, rdepths[depth], gdepths[depth], bdepths[depth] };
+    return ret;
+}
+
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+/// Frame Compression                                                                                                ///
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+
+typedef struct {
+    uint8_t * next;
+    size_t size;
+} MsfBufferHeader;
+
+static inline int msf_put_code(uint8_t * * writeHead, uint32_t * blockBits, int len, uint32_t code) {
+    //insert new code into block buffer
+    int idx = *blockBits / 8;
+    int bit = *blockBits % 8;
+    (*writeHead)[idx + 0] |= code <<       bit ;
+    (*writeHead)[idx + 1] |= code >> ( 8 - bit);
+    (*writeHead)[idx + 2] |= code >> (16 - bit);
+    *blockBits += len;
+
+    //prep the next block buffer if the current one is full
+    if (*blockBits >= 256 * 8) {
+        *blockBits -= 255 * 8;
+        (*writeHead) += 256;
+        (*writeHead)[2] = (*writeHead)[1];
+        (*writeHead)[1] = (*writeHead)[0];
+        (*writeHead)[0] = 255;
+        memset((*writeHead) + 4, 0, 256);
+    }
+
+    return 1;
+}
+
+typedef struct {
+    int16_t * data;
+    int len;
+    int stride;
+} MsfStridedList;
+
+static inline void msf_lzw_reset(MsfStridedList * lzw, int tableSize, int stride) { MsfTimeFunc
+    memset(lzw->data, 0xFF, 4096 * stride * sizeof(int16_t));
+    lzw->len = tableSize + 2;
+    lzw->stride = stride;
+}
+
+static uint8_t * msf_compress_frame(void * allocContext, int width, int height, int centiSeconds,
+                                    MsfCookedFrame frame, MsfCookedFrame previous, uint8_t * used)
+{ MsfTimeFunc
+    //NOTE: we reserve enough memory for theoretical the worst case upfront because it's a reasonable amount,
+    //      and prevents us from ever having to check size or realloc during compression
+    int maxBufSize = sizeof(MsfBufferHeader) + 32 + 256 * 3 + width * height * 3 / 2; //headers + color table + data
+    uint8_t * allocation = (uint8_t *) MSF_GIF_MALLOC(allocContext, maxBufSize);
+    if (!allocation) { return NULL; }
+    uint8_t * writeBase = allocation + sizeof(MsfBufferHeader);
+    uint8_t * writeHead = writeBase;
+    int lzwAllocSize = 4096 * (frame.count + 1) * sizeof(int16_t);
+    MsfStridedList lzw = { (int16_t *) MSF_GIF_MALLOC(allocContext, lzwAllocSize) };
+    if (!lzw.data) { MSF_GIF_FREE(allocContext, allocation, maxBufSize); return NULL; }
+
+    //allocate tlb
+    int totalBits = frame.rbits + frame.gbits + frame.bbits;
+    int tlbSize = 1 << totalBits;
+    uint8_t tlb[1 << 16]; //only 64k, so stack allocating is fine
+
+    //generate palette
+    typedef struct { uint8_t r, g, b; } Color3;
+    Color3 table[256] = { {0} };
+    int tableIdx = 1; //we start counting at 1 because 0 is the transparent color
+    MsfTimeLoop("table") for (int i = 0; i < tlbSize; ++i) {
+        if (used[i]) {
+            tlb[i] = tableIdx;
+            int rmask = (1 << frame.rbits) - 1;
+            int gmask = (1 << frame.gbits) - 1;
+            //isolate components
+            int r = i & rmask;
+            int g = i >> frame.rbits & gmask;
+            int b = i >> (frame.rbits + frame.gbits);
+            //shift into highest bits
+            r <<= 8 - frame.rbits;
+            g <<= 8 - frame.gbits;
+            b <<= 8 - frame.bbits;
+            table[tableIdx].r = r | r >> frame.rbits | r >> (frame.rbits * 2) | r >> (frame.rbits * 3);
+            table[tableIdx].g = g | g >> frame.gbits | g >> (frame.gbits * 2) | g >> (frame.gbits * 3);
+            table[tableIdx].b = b | b >> frame.bbits | b >> (frame.bbits * 2) | b >> (frame.bbits * 3);
+            ++tableIdx;
+        }
+    }
+
+    //SPEC: "Because of some algorithmic constraints however, black & white images which have one color bit
+    //       must be indicated as having a code size of 2."
+    int tableBits = msf_imax(2, msf_bit_log(tableIdx - 1));
+    int tableSize = 1 << tableBits;
+    //NOTE: we don't just compare `depth` field here because it will be wrong for the first frame and we will segfault
+    int hasSamePal = frame.rbits == previous.rbits && frame.gbits == previous.gbits && frame.bbits == previous.bbits;
+
+    //NOTE: because __attribute__((__packed__)) is annoyingly compiler-specific, we do this unreadable weirdness
+    char headerBytes[19] = "\x21\xF9\x04\x05\0\0\0\0" "\x2C\0\0\0\0\0\0\0\0\x80";
+    memcpy(&headerBytes[4], &centiSeconds, 2);
+    memcpy(&headerBytes[13], &width, 2);
+    memcpy(&headerBytes[15], &height, 2);
+    headerBytes[17] |= tableBits - 1;
+    memcpy(writeHead, headerBytes, 18);
+    writeHead += 18;
+
+    //local color table
+    memcpy(writeHead, table, tableSize * sizeof(Color3));
+    writeHead += tableSize * sizeof(Color3);
+    *writeHead++ = tableBits;
+
+    //prep block
+    memset(writeHead, 0, 260);
+    writeHead[0] = 255;
+    uint32_t blockBits = 8; //relative to block.head
+
+    //SPEC: "Encoders should output a Clear code as the first code of each image data stream."
+    msf_lzw_reset(&lzw, tableSize, tableIdx);
+    msf_put_code(&writeHead, &blockBits, msf_bit_log(lzw.len - 1), tableSize);
+
+    int lastCode = hasSamePal && frame.pixels[0] == previous.pixels[0]? 0 : tlb[frame.pixels[0]];
+    MsfTimeLoop("compress") for (int i = 1; i < width * height; ++i) {
+        //PERF: branching vs. branchless version of this line is observed to have no discernable impact on speed
+        int color = hasSamePal && frame.pixels[i] == previous.pixels[i]? 0 : tlb[frame.pixels[i]];
+        //PERF: branchless version must use && otherwise it will segfault on frame 1, but it's well-predicted so OK
+        // int color = (!(hasSamePal && frame.pixels[i] == previous.pixels[i])) * tlb[frame.pixels[i]];
+        int code = (&lzw.data[lastCode * lzw.stride])[color];
+        if (code < 0) {
+            //write to code stream
+            int codeBits = msf_bit_log(lzw.len - 1);
+            msf_put_code(&writeHead, &blockBits, codeBits, lastCode);
+
+            if (lzw.len > 4095) {
+                //reset buffer code table
+                msf_put_code(&writeHead, &blockBits, codeBits, tableSize);
+                msf_lzw_reset(&lzw, tableSize, tableIdx);
+            } else {
+                (&lzw.data[lastCode * lzw.stride])[color] = lzw.len;
+                ++lzw.len;
+            }
+
+            lastCode = color;
+        } else {
+            lastCode = code;
+        }
+    }
+
+    MSF_GIF_FREE(allocContext, lzw.data, lzwAllocSize);
+    MSF_GIF_FREE(allocContext, previous.pixels, width * height * sizeof(uint32_t));
+
+    //write code for leftover index buffer contents, then the end code
+    msf_put_code(&writeHead, &blockBits, msf_imin(12, msf_bit_log(lzw.len - 1)), lastCode);
+    msf_put_code(&writeHead, &blockBits, msf_imin(12, msf_bit_log(lzw.len)), tableSize + 1);
+
+    //flush remaining data
+    if (blockBits > 8) {
+        int bytes = (blockBits + 7) / 8; //round up
+        writeHead[0] = bytes - 1;
+        writeHead += bytes;
+    }
+    *writeHead++ = 0; //terminating block
+
+    //filling in buffer header and shrink buffer to fit data
+    MsfBufferHeader * header = (MsfBufferHeader *) allocation;
+    header->next = NULL;
+    header->size = writeHead - writeBase;
+    uint8_t * moved = (uint8_t *) MSF_GIF_REALLOC(allocContext, allocation, maxBufSize, writeHead - allocation);
+    if (!moved) { MSF_GIF_FREE(allocContext, allocation, maxBufSize); return NULL; }
+    return moved;
+}
+
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+/// Incremental API                                                                                                  ///
+////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+
+int msf_gif_begin(MsfGifState * handle, int width, int height) { MsfTimeFunc
+    MsfCookedFrame empty = {0}; //god I hate MSVC...
+    handle->previousFrame = empty;
+    handle->width = width;
+    handle->height = height;
+
+    //setup header buffer header (lol)
+    handle->listHead = (uint8_t *) MSF_GIF_MALLOC(handle->customAllocatorContext, sizeof(MsfBufferHeader) + 32);
+    if (!handle->listHead) { return 0; }
+    handle->listTail = handle->listHead;
+    MsfBufferHeader * header = (MsfBufferHeader *) handle->listHead;
+    header->next = NULL;
+    header->size = 32;
+
+    //NOTE: because __attribute__((__packed__)) is annoyingly compiler-specific, we do this unreadable weirdness
+    char headerBytes[33] = "GIF89a\0\0\0\0\x10\0\0" "\x21\xFF\x0BNETSCAPE2.0\x03\x01\0\0\0";
+    memcpy(&headerBytes[6], &width, 2);
+    memcpy(&headerBytes[8], &height, 2);
+    memcpy(handle->listHead + sizeof(MsfBufferHeader), headerBytes, 32);
+    return 1;
+}
+
+int msf_gif_frame(MsfGifState * handle, uint8_t * pixelData, int centiSecondsPerFame, int maxBitDepth, int pitchInBytes)
+{ MsfTimeFunc
+    if (!handle->listHead) { return 0; }
+
+    maxBitDepth = msf_imax(1, msf_imin(16, maxBitDepth));
+    if (pitchInBytes == 0) pitchInBytes = handle->width * 4;
+    if (pitchInBytes < 0) pixelData -= pitchInBytes * (handle->height - 1);
+
+    uint8_t used[1 << 16]; //only 64k, so stack allocating is fine
+    MsfCookedFrame frame =
+        msf_cook_frame(handle->customAllocatorContext, pixelData, used, handle->width, handle->height, pitchInBytes,
+            msf_imin(maxBitDepth, handle->previousFrame.depth + 160 / msf_imax(1, handle->previousFrame.count)));
+    //TODO: de-duplicate cleanup code
+    if (!frame.pixels) {
+        MSF_GIF_FREE(handle->customAllocatorContext,
+                     handle->previousFrame.pixels, handle->width * handle->height * sizeof(uint32_t));
+        for (uint8_t * node = handle->listHead; node;) {
+            MsfBufferHeader * header = (MsfBufferHeader *) node;
+            node = header->next;
+            MSF_GIF_FREE(handle->customAllocatorContext, header, sizeof(MsfBufferHeader) + header->size);
+        }
+        handle->listHead = handle->listTail = NULL;
+        return 0;
+    }
+
+    uint8_t * buffer = msf_compress_frame(handle->customAllocatorContext,
+        handle->width, handle->height, centiSecondsPerFame, frame, handle->previousFrame, used);
+    ((MsfBufferHeader *) handle->listTail)->next = buffer;
+    handle->listTail = buffer;
+    if (!buffer) {
+        MSF_GIF_FREE(handle->customAllocatorContext, frame.pixels, handle->width * handle->height * sizeof(uint32_t));
+        MSF_GIF_FREE(handle->customAllocatorContext,
+                     handle->previousFrame.pixels, handle->width * handle->height * sizeof(uint32_t));
+        for (uint8_t * node = handle->listHead; node;) {
+            MsfBufferHeader * header = (MsfBufferHeader *) node;
+            node = header->next;
+            MSF_GIF_FREE(handle->customAllocatorContext, header, sizeof(MsfBufferHeader) + header->size);
+        }
+        handle->listHead = handle->listTail = NULL;
+        return 0;
+    }
+
+    handle->previousFrame = frame;
+    return 1;
+}
+
+MsfGifResult msf_gif_end(MsfGifState * handle) { MsfTimeFunc
+    if (!handle->listHead) { MsfGifResult empty = {0}; return empty; }
+
+    MSF_GIF_FREE(handle->customAllocatorContext,
+                 handle->previousFrame.pixels, handle->width * handle->height * sizeof(uint32_t));
+
+    //first pass: determine total size
+    size_t total = 1; //1 byte for trailing marker
+    for (uint8_t * node = handle->listHead; node;) {
+        MsfBufferHeader * header = (MsfBufferHeader *) node;
+        node = header->next;
+        total += header->size;
+    }
+
+    //second pass: write data
+    uint8_t * buffer = (uint8_t *) MSF_GIF_MALLOC(handle->customAllocatorContext, total);
+    if (buffer) {
+        uint8_t * writeHead = buffer;
+        for (uint8_t * node = handle->listHead; node;) {
+            MsfBufferHeader * header = (MsfBufferHeader *) node;
+            memcpy(writeHead, node + sizeof(MsfBufferHeader), header->size);
+            writeHead += header->size;
+            node = header->next;
+        }
+        *writeHead++ = 0x3B;
+    }
+
+    //third pass: free buffers
+    for (uint8_t * node = handle->listHead; node;) {
+        MsfBufferHeader * header = (MsfBufferHeader *) node;
+        node = header->next;
+        MSF_GIF_FREE(handle->customAllocatorContext, header, sizeof(MsfBufferHeader) + header->size);
+    }
+
+    MsfGifResult ret = { buffer, total, total, handle->customAllocatorContext };
+    return ret;
+}
+
+void msf_gif_free(MsfGifResult result) {
+    if (result.data) { MSF_GIF_FREE(result.contextPointer, result.data, result.allocSize); }
+}
+
+#endif //MSF_GIF_ALREADY_IMPLEMENTED_IN_THIS_TRANSLATION_UNIT
+#endif //MSF_GIF_IMPL
+
+/*
+------------------------------------------------------------------------------
+This software is available under 2 licenses -- choose whichever you prefer.
+------------------------------------------------------------------------------
+ALTERNATIVE A - MIT License
+Copyright (c) 2020 Miles Fogle
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+------------------------------------------------------------------------------
+ALTERNATIVE B - Public Domain (www.unlicense.org)
+This is free and unencumbered software released into the public domain.
+Anyone is free to copy, modify, publish, use, compile, sell, or distribute this
+software, either in source code form or as a compiled binary, for any purpose,
+commercial or non-commercial, and by any means.
+In jurisdictions that recognize copyright laws, the author or authors of this
+software dedicate any and all copyright interest in the software to the public
+domain. We make this dedication for the benefit of the public at large and to
+the detriment of our heirs and successors. We intend this dedication to be an
+overt act of relinquishment in perpetuity of all present and future rights to
+this software under copyright law.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+------------------------------------------------------------------------------
+*/
diff --git a/libs/raylib/src/external/sdefl.h b/libs/raylib/src/external/sdefl.h
new file mode 100644
index 0000000..218e754
--- /dev/null
+++ b/libs/raylib/src/external/sdefl.h
@@ -0,0 +1,697 @@
+/*
+# Small Deflate
+`sdefl` is a small bare bone lossless compression library in ANSI C (ISO C90)
+which implements the Deflate (RFC 1951) compressed data format specification standard.
+It is mainly tuned to get as much speed and compression ratio from as little code
+as needed to keep the implementation as concise as possible.
+
+## Features
+- Portable single header and source file duo written in ANSI C (ISO C90)
+- Dual license with either MIT or public domain
+- Small implementation
+    - Deflate: 525 LoC
+    - Inflate: 320 LoC
+- Webassembly:
+    - Deflate ~3.7 KB (~2.2KB compressed)
+    - Inflate ~3.6 KB (~2.2KB compressed)
+
+## Usage:
+This file behaves differently depending on what symbols you define
+before including it.
+
+Header-File mode:
+If you do not define `SDEFL_IMPLEMENTATION` before including this file, it
+will operate in header only mode. In this mode it declares all used structs
+and the API of the library without including the implementation of the library.
+
+Implementation mode:
+If you define `SDEFL_IMPLEMENTATION` before including this file, it will
+compile the implementation . Make sure that you only include
+this file implementation in *one* C or C++ file to prevent collisions.
+
+### Benchmark
+
+| Compressor name         | Compression| Decompress.| Compr. size | Ratio |
+| ------------------------| -----------| -----------| ----------- | ----- |
+| sdefl 1.0 -0            |   127 MB/s |   233 MB/s |    40004116 | 39.88 |
+| sdefl 1.0 -1            |   111 MB/s |   259 MB/s |    38940674 | 38.82 |
+| sdefl 1.0 -5            |    45 MB/s |   275 MB/s |    36577183 | 36.46 |
+| sdefl 1.0 -7            |    38 MB/s |   276 MB/s |    36523781 | 36.41 |
+| zlib 1.2.11 -1          |    72 MB/s |   307 MB/s |    42298774 | 42.30 |
+| zlib 1.2.11 -6          |    24 MB/s |   313 MB/s |    36548921 | 36.55 |
+| zlib 1.2.11 -9          |    20 MB/s |   314 MB/s |    36475792 | 36.48 |
+| miniz 1.0 -1            |   122 MB/s |   208 MB/s |    48510028 | 48.51 |
+| miniz 1.0 -6            |    27 MB/s |   260 MB/s |    36513697 | 36.51 |
+| miniz 1.0 -9            |    23 MB/s |   261 MB/s |    36460101 | 36.46 |
+| libdeflate 1.3 -1       |   147 MB/s |   667 MB/s |    39597378 | 39.60 |
+| libdeflate 1.3 -6       |    69 MB/s |   689 MB/s |    36648318 | 36.65 |
+| libdeflate 1.3 -9       |    13 MB/s |   672 MB/s |    35197141 | 35.20 |
+| libdeflate 1.3 -12      |  8.13 MB/s |   670 MB/s |    35100568 | 35.10 |
+
+### Compression
+Results on the [Silesia compression corpus](http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia):
+
+| File    |   Original | `sdefl 0`  	| `sdefl 5` 	| `sdefl 7` |
+| :------ | ---------: | -----------------: | ---------: | ----------: |
+| dickens | 10.192.446 |  4,260,187|  3,845,261|   3,833,657 |
+| mozilla | 51.220.480 | 20,774,706 | 19,607,009 |  19,565,867 |
+| mr      |  9.970.564 | 3,860,531 |  3,673,460 |   3,665,627 |
+| nci     | 33.553.445 | 4,030,283 |  3,094,526 |   3,006,075 |
+| ooffice |  6.152.192 | 3,320,063 |  3,186,373 |   3,183,815 |
+| osdb    | 10.085.684 | 3,919,646 |  3,649,510 |   3,649,477 |
+| reymont |  6.627.202 | 2,263,378 |  1,857,588 |   1,827,237 |
+| samba   | 21.606.400 | 6,121,797 |  5,462,670 |   5,450,762 |
+| sao     |  7.251.944 | 5,612,421 |  5,485,380 |   5,481,765 |
+| webster | 41.458.703 | 13,972,648 | 12,059,432 |  11,991,421 |
+| xml     |  5.345.280 | 886,620|    674,009 |     662,141 |
+| x-ray   |  8.474.240 | 6,304,655 |  6,244,779 |   6,244,779 |
+
+## License
+```
+------------------------------------------------------------------------------
+This software is available under 2 licenses -- choose whichever you prefer.
+------------------------------------------------------------------------------
+ALTERNATIVE A - MIT License
+Copyright (c) 2020 Micha Mettke
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+------------------------------------------------------------------------------
+ALTERNATIVE B - Public Domain (www.unlicense.org)
+This is free and unencumbered software released into the public domain.
+Anyone is free to copy, modify, publish, use, compile, sell, or distribute this
+software, either in source code form or as a compiled binary, for any purpose,
+commercial or non-commercial, and by any means.
+In jurisdictions that recognize copyright laws, the author or authors of this
+software dedicate any and all copyright interest in the software to the public
+domain. We make this dedication for the benefit of the public at large and to
+the detriment of our heirs and successors. We intend this dedication to be an
+overt act of relinquishment in perpetuity of all present and future rights to
+this software under copyright law.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+------------------------------------------------------------------------------
+```
+*/
+#ifndef SDEFL_H_INCLUDED
+#define SDEFL_H_INCLUDED
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define SDEFL_MAX_OFF   (1 << 15)
+#define SDEFL_WIN_SIZ   SDEFL_MAX_OFF
+#define SDEFL_WIN_MSK   (SDEFL_WIN_SIZ-1)
+
+#define SDEFL_HASH_BITS 15
+#define SDEFL_HASH_SIZ  (1 << SDEFL_HASH_BITS)
+#define SDEFL_HASH_MSK  (SDEFL_HASH_SIZ-1)
+
+#define SDEFL_MIN_MATCH 4
+#define SDEFL_BLK_MAX   (256*1024)
+#define SDEFL_SEQ_SIZ   ((SDEFL_BLK_MAX + SDEFL_MIN_MATCH)/SDEFL_MIN_MATCH)
+
+#define SDEFL_SYM_MAX   (288)
+#define SDEFL_OFF_MAX   (32)
+#define SDEFL_PRE_MAX   (19)
+
+#define SDEFL_LVL_MIN   0
+#define SDEFL_LVL_DEF   5
+#define SDEFL_LVL_MAX   8
+
+struct sdefl_freq {
+  unsigned lit[SDEFL_SYM_MAX];
+  unsigned off[SDEFL_OFF_MAX];
+};
+struct sdefl_code_words {
+  unsigned lit[SDEFL_SYM_MAX];
+  unsigned off[SDEFL_OFF_MAX];
+};
+struct sdefl_lens {
+  unsigned char lit[SDEFL_SYM_MAX];
+  unsigned char off[SDEFL_OFF_MAX];
+};
+struct sdefl_codes {
+  struct sdefl_code_words word;
+  struct sdefl_lens len;
+};
+struct sdefl_seqt {
+  int off, len;
+};
+struct sdefl {
+  int bits, bitcnt;
+  int tbl[SDEFL_HASH_SIZ];
+  int prv[SDEFL_WIN_SIZ];
+
+  int seq_cnt;
+  struct sdefl_seqt seq[SDEFL_SEQ_SIZ];
+  struct sdefl_freq freq;
+  struct sdefl_codes cod;
+};
+extern int sdefl_bound(int in_len);
+extern int sdeflate(struct sdefl *s, void *o, const void *i, int n, int lvl);
+extern int zsdeflate(struct sdefl *s, void *o, const void *i, int n, int lvl);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* SDEFL_H_INCLUDED */
+
+#ifdef SDEFL_IMPLEMENTATION
+
+#include <assert.h> /* assert */
+#include <string.h> /* memcpy */
+#include <limits.h> /* CHAR_BIT */
+
+#define SDEFL_NIL               (-1)
+#define SDEFL_MAX_MATCH         258
+#define SDEFL_MAX_CODE_LEN      (15)
+#define SDEFL_SYM_BITS          (10u)
+#define SDEFL_SYM_MSK           ((1u << SDEFL_SYM_BITS)-1u)
+#define SDEFL_LIT_LEN_CODES     (14)
+#define SDEFL_OFF_CODES         (15)
+#define SDEFL_PRE_CODES         (7)
+#define SDEFL_CNT_NUM(n)        ((((n)+3u/4u)+3u)&~3u)
+#define SDEFL_EOB               (256)
+
+#define sdefl_npow2(n) (1 << (sdefl_ilog2((n)-1) + 1))
+
+static int
+sdefl_ilog2(int n) {
+  if (!n) return 0;
+#ifdef _MSC_VER
+  unsigned long msbp = 0;
+  _BitScanReverse(&msbp, (unsigned long)n);
+  return (int)msbp;
+#elif defined(__GNUC__) || defined(__clang__)
+  return (int)sizeof(unsigned long) * CHAR_BIT - 1 - __builtin_clzl((unsigned long)n);
+#else
+  #define lt(n) n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n
+  static const char tbl[256] = {
+    0,0,1,1,2,2,2,2,3,3,3,3,3,3,3,3,lt(4), lt(5), lt(5), lt(6), lt(6), lt(6), lt(6),
+    lt(7), lt(7), lt(7), lt(7), lt(7), lt(7), lt(7), lt(7)};
+  int tt, t;
+  if ((tt = (n >> 16))) {
+    return (t = (tt >> 8)) ? 24 + tbl[t] : 16 + tbl[tt];
+  } else {
+    return (t = (n >> 8)) ? 8 + tbl[t] : tbl[n];
+  }
+  #undef lt
+#endif
+}
+static unsigned
+sdefl_uload32(const void *p) {
+  /* hopefully will be optimized to an unaligned read */
+  unsigned n = 0;
+  memcpy(&n, p, sizeof(n));
+  return n;
+}
+static unsigned
+sdefl_hash32(const void *p) {
+  unsigned n = sdefl_uload32(p);
+  return (n * 0x9E377989) >> (32 - SDEFL_HASH_BITS);
+}
+static void
+sdefl_put(unsigned char **dst, struct sdefl *s, int code, int bitcnt) {
+  s->bits |= (code << s->bitcnt);
+  s->bitcnt += bitcnt;
+  while (s->bitcnt >= 8) {
+    unsigned char *tar = *dst;
+    *tar = (unsigned char)(s->bits & 0xFF);
+    s->bits >>= 8;
+    s->bitcnt -= 8;
+    *dst = *dst + 1;
+  }
+}
+static void
+sdefl_heap_sub(unsigned A[], unsigned len, unsigned sub) {
+  unsigned c, p = sub;
+  unsigned v = A[sub];
+  while ((c = p << 1) <= len) {
+    if (c < len && A[c + 1] > A[c]) c++;
+    if (v >= A[c]) break;
+    A[p] = A[c], p = c;
+  }
+  A[p] = v;
+}
+static void
+sdefl_heap_array(unsigned *A, unsigned len) {
+  unsigned sub;
+  for (sub = len >> 1; sub >= 1; sub--)
+    sdefl_heap_sub(A, len, sub);
+}
+static void
+sdefl_heap_sort(unsigned *A, unsigned n) {
+  A--;
+  sdefl_heap_array(A, n);
+  while (n >= 2) {
+    unsigned tmp = A[n];
+    A[n--] = A[1];
+    A[1] = tmp;
+    sdefl_heap_sub(A, n, 1);
+  }
+}
+static unsigned
+sdefl_sort_sym(unsigned sym_cnt, unsigned *freqs,
+               unsigned char *lens, unsigned *sym_out) {
+  unsigned cnts[SDEFL_CNT_NUM(SDEFL_SYM_MAX)] = {0};
+  unsigned cnt_num = SDEFL_CNT_NUM(sym_cnt);
+  unsigned used_sym = 0;
+  unsigned sym, i;
+  for (sym = 0; sym < sym_cnt; sym++)
+    cnts[freqs[sym] < cnt_num-1 ? freqs[sym]: cnt_num-1]++;
+  for (i = 1; i < cnt_num; i++) {
+    unsigned cnt = cnts[i];
+    cnts[i] = used_sym;
+    used_sym += cnt;
+  }
+  for (sym = 0; sym < sym_cnt; sym++) {
+    unsigned freq = freqs[sym];
+    if (freq) {
+        unsigned idx = freq < cnt_num-1 ? freq : cnt_num-1;
+        sym_out[cnts[idx]++] = sym | (freq << SDEFL_SYM_BITS);
+    } else lens[sym] = 0;
+  }
+  sdefl_heap_sort(sym_out + cnts[cnt_num-2], cnts[cnt_num-1] - cnts[cnt_num-2]);
+  return used_sym;
+}
+static void
+sdefl_build_tree(unsigned *A, unsigned sym_cnt) {
+  unsigned i = 0, b = 0, e = 0;
+  do {
+    unsigned m, n, freq_shift;
+    if (i != sym_cnt && (b == e || (A[i] >> SDEFL_SYM_BITS) <= (A[b] >> SDEFL_SYM_BITS)))
+      m = i++;
+    else m = b++;
+    if (i != sym_cnt && (b == e || (A[i] >> SDEFL_SYM_BITS) <= (A[b] >> SDEFL_SYM_BITS)))
+      n = i++;
+    else n = b++;
+
+    freq_shift = (A[m] & ~SDEFL_SYM_MSK) + (A[n] & ~SDEFL_SYM_MSK);
+    A[m] = (A[m] & SDEFL_SYM_MSK) | (e << SDEFL_SYM_BITS);
+    A[n] = (A[n] & SDEFL_SYM_MSK) | (e << SDEFL_SYM_BITS);
+    A[e] = (A[e] & SDEFL_SYM_MSK) | freq_shift;
+  } while (sym_cnt - ++e > 1);
+}
+static void
+sdefl_gen_len_cnt(unsigned *A, unsigned root, unsigned *len_cnt,
+                  unsigned max_code_len) {
+  int n;
+  unsigned i;
+  for (i = 0; i <= max_code_len; i++)
+    len_cnt[i] = 0;
+  len_cnt[1] = 2;
+
+  A[root] &= SDEFL_SYM_MSK;
+  for (n = (int)root - 1; n >= 0; n--) {
+    unsigned p = A[n] >> SDEFL_SYM_BITS;
+    unsigned pdepth = A[p] >> SDEFL_SYM_BITS;
+    unsigned depth = pdepth + 1;
+    unsigned len = depth;
+
+    A[n] = (A[n] & SDEFL_SYM_MSK) | (depth << SDEFL_SYM_BITS);
+    if (len >= max_code_len) {
+      len = max_code_len;
+      do len--; while (!len_cnt[len]);
+    }
+    len_cnt[len]--;
+    len_cnt[len+1] += 2;
+  }
+}
+static void
+sdefl_gen_codes(unsigned *A, unsigned char *lens, const unsigned *len_cnt,
+                unsigned max_code_word_len, unsigned sym_cnt) {
+  unsigned i, sym, len, nxt[SDEFL_MAX_CODE_LEN + 1];
+  for (i = 0, len = max_code_word_len; len >= 1; len--) {
+    unsigned cnt = len_cnt[len];
+    while (cnt--) lens[A[i++] & SDEFL_SYM_MSK] = (unsigned char)len;
+  }
+  nxt[0] = nxt[1] = 0;
+  for (len = 2; len <= max_code_word_len; len++)
+    nxt[len] = (nxt[len-1] + len_cnt[len-1]) << 1;
+  for (sym = 0; sym < sym_cnt; sym++)
+    A[sym] = nxt[lens[sym]]++;
+}
+static unsigned
+sdefl_rev(unsigned c, unsigned char n) {
+  c = ((c & 0x5555) << 1) | ((c & 0xAAAA) >> 1);
+  c = ((c & 0x3333) << 2) | ((c & 0xCCCC) >> 2);
+  c = ((c & 0x0F0F) << 4) | ((c & 0xF0F0) >> 4);
+  c = ((c & 0x00FF) << 8) | ((c & 0xFF00) >> 8);
+  return c >> (16-n);
+}
+static void
+sdefl_huff(unsigned char *lens, unsigned *codes, unsigned *freqs,
+           unsigned num_syms, unsigned max_code_len) {
+  unsigned c, *A = codes;
+  unsigned len_cnt[SDEFL_MAX_CODE_LEN + 1];
+  unsigned used_syms = sdefl_sort_sym(num_syms, freqs, lens, A);
+  if (!used_syms) return;
+  if (used_syms == 1) {
+    unsigned s = A[0] & SDEFL_SYM_MSK;
+    unsigned i = s ? s : 1;
+    codes[0] = 0, lens[0] = 1;
+    codes[i] = 1, lens[i] = 1;
+    return;
+  }
+  sdefl_build_tree(A, used_syms);
+  sdefl_gen_len_cnt(A, used_syms-2, len_cnt, max_code_len);
+  sdefl_gen_codes(A, lens, len_cnt, max_code_len, num_syms);
+  for (c = 0; c < num_syms; c++) {
+    codes[c] = sdefl_rev(codes[c], lens[c]);
+  }
+}
+struct sdefl_symcnt {
+  int items;
+  int lit;
+  int off;
+};
+static void
+sdefl_precode(struct sdefl_symcnt *cnt, unsigned *freqs, unsigned *items,
+              const unsigned char *litlen, const unsigned char *offlen) {
+  unsigned *at = items;
+  unsigned run_start = 0;
+
+  unsigned total = 0;
+  unsigned char lens[SDEFL_SYM_MAX + SDEFL_OFF_MAX];
+  for (cnt->lit = SDEFL_SYM_MAX; cnt->lit > 257; cnt->lit--)
+    if (litlen[cnt->lit - 1]) break;
+  for (cnt->off = SDEFL_OFF_MAX; cnt->off > 1; cnt->off--)
+    if (offlen[cnt->off - 1]) break;
+
+  total = (unsigned)(cnt->lit + cnt->off);
+  memcpy(lens, litlen, sizeof(unsigned char) * cnt->lit);
+  memcpy(lens + cnt->lit, offlen, sizeof(unsigned char) * cnt->off);
+  do {
+    unsigned len = lens[run_start];
+    unsigned run_end = run_start;
+    do run_end++; while (run_end != total && len == lens[run_end]);
+    if (!len) {
+      while ((run_end - run_start) >= 11) {
+        unsigned n = (run_end - run_start) - 11;
+        unsigned xbits =  n < 0x7f ? n : 0x7f;
+        freqs[18]++;
+        *at++ = 18u | (xbits << 5u);
+        run_start += 11 + xbits;
+      }
+      if ((run_end - run_start) >= 3) {
+        unsigned n = (run_end - run_start) - 3;
+        unsigned xbits =  n < 0x7 ? n : 0x7;
+        freqs[17]++;
+        *at++ = 17u | (xbits << 5u);
+        run_start += 3 + xbits;
+      }
+    } else if ((run_end - run_start) >= 4) {
+      freqs[len]++;
+      *at++ = len;
+      run_start++;
+      do {
+        unsigned xbits = (run_end - run_start) - 3;
+        xbits = xbits < 0x03 ? xbits : 0x03;
+        *at++ = 16 | (xbits << 5);
+        run_start += 3 + xbits;
+        freqs[16]++;
+      } while ((run_end - run_start) >= 3);
+    }
+    while (run_start != run_end) {
+      freqs[len]++;
+      *at++ = len;
+      run_start++;
+    }
+  } while (run_start != total);
+  cnt->items = (int)(at - items);
+}
+struct sdefl_match_codes {
+  int ls, lc;
+  int dc, dx;
+};
+static void
+sdefl_match_codes(struct sdefl_match_codes *cod, int dist, int len) {
+  static const short dxmax[] = {0,6,12,24,48,96,192,384,768,1536,3072,6144,12288,24576};
+  static const unsigned char lslot[258+1] = {
+    0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 12,
+    12, 13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 16, 16,
+    16, 16, 16, 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18,
+    18, 19, 19, 19, 19, 19, 19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20,
+    20, 20, 20, 20, 20, 20, 20, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
+    21, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,
+    22, 22, 22, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,
+    23, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,
+    24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 25, 25, 25,
+    25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,
+    25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 26, 26, 26, 26, 26, 26, 26,
+    26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26,
+    26, 26, 26, 26, 26, 26, 26, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27,
+    27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27,
+    27, 27, 28
+  };
+  cod->ls = lslot[len];
+  cod->lc = 257 + cod->ls;
+  cod->dx = sdefl_ilog2(sdefl_npow2(dist) >> 2);
+  cod->dc = cod->dx ? ((cod->dx + 1) << 1) + (dist > dxmax[cod->dx]) : dist-1;
+}
+static void
+sdefl_match(unsigned char **dst, struct sdefl *s, int dist, int len) {
+  static const char lxn[] = {0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0};
+  static const short lmin[] = {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};
+  static const short dmin[] = {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};
+
+  struct sdefl_match_codes cod;
+  sdefl_match_codes(&cod, dist, len);
+  sdefl_put(dst, s, (int)s->cod.word.lit[cod.lc], s->cod.len.lit[cod.lc]);
+  sdefl_put(dst, s, len - lmin[cod.ls], lxn[cod.ls]);
+  sdefl_put(dst, s, (int)s->cod.word.off[cod.dc], s->cod.len.off[cod.dc]);
+  sdefl_put(dst, s, dist - dmin[cod.dc], cod.dx);
+}
+static void
+sdefl_flush(unsigned char **dst, struct sdefl *s, int is_last,
+            const unsigned char *in) {
+  int j, i = 0, item_cnt = 0;
+  struct sdefl_symcnt symcnt = {0};
+  unsigned codes[SDEFL_PRE_MAX];
+  unsigned char lens[SDEFL_PRE_MAX];
+  unsigned freqs[SDEFL_PRE_MAX] = {0};
+  unsigned items[SDEFL_SYM_MAX + SDEFL_OFF_MAX];
+  static const unsigned char perm[SDEFL_PRE_MAX] = {16,17,18,0,8,7,9,6,10,5,11,
+      4,12,3,13,2,14,1,15};
+
+  /* huffman codes */
+  s->freq.lit[SDEFL_EOB]++;
+  sdefl_huff(s->cod.len.lit, s->cod.word.lit, s->freq.lit, SDEFL_SYM_MAX, SDEFL_LIT_LEN_CODES);
+  sdefl_huff(s->cod.len.off, s->cod.word.off, s->freq.off, SDEFL_OFF_MAX, SDEFL_OFF_CODES);
+  sdefl_precode(&symcnt, freqs, items, s->cod.len.lit, s->cod.len.off);
+  sdefl_huff(lens, codes, freqs, SDEFL_PRE_MAX, SDEFL_PRE_CODES);
+  for (item_cnt = SDEFL_PRE_MAX; item_cnt > 4; item_cnt--) {
+    if (lens[perm[item_cnt - 1]]) break;
+  }
+  /* block header */
+  sdefl_put(dst, s, is_last ? 0x01 : 0x00, 1); /* block */
+  sdefl_put(dst, s, 0x02, 2); /* dynamic huffman */
+  sdefl_put(dst, s, symcnt.lit - 257, 5);
+  sdefl_put(dst, s, symcnt.off - 1, 5);
+  sdefl_put(dst, s, item_cnt - 4, 4);
+  for (i = 0; i < item_cnt; ++i)
+    sdefl_put(dst, s, lens[perm[i]], 3);
+  for (i = 0; i < symcnt.items; ++i) {
+    unsigned sym = items[i] & 0x1F;
+    sdefl_put(dst, s, (int)codes[sym], lens[sym]);
+    if (sym < 16) continue;
+    if (sym == 16) sdefl_put(dst, s, items[i] >> 5, 2);
+    else if(sym == 17) sdefl_put(dst, s, items[i] >> 5, 3);
+    else sdefl_put(dst, s, items[i] >> 5, 7);
+  }
+  /* block sequences */
+  for (i = 0; i < s->seq_cnt; ++i) {
+    if (s->seq[i].off >= 0)
+      for (j = 0; j < s->seq[i].len; ++j) {
+        int c = in[s->seq[i].off + j];
+        sdefl_put(dst, s, (int)s->cod.word.lit[c], s->cod.len.lit[c]);
+      }
+    else sdefl_match(dst, s, -s->seq[i].off, s->seq[i].len);
+  }
+  sdefl_put(dst, s, (int)(s)->cod.word.lit[SDEFL_EOB], (s)->cod.len.lit[SDEFL_EOB]);
+  memset(&s->freq, 0, sizeof(s->freq));
+  s->seq_cnt = 0;
+}
+static void
+sdefl_seq(struct sdefl *s, int off, int len) {
+  assert(s->seq_cnt + 2 < SDEFL_SEQ_SIZ);
+  s->seq[s->seq_cnt].off = off;
+  s->seq[s->seq_cnt].len = len;
+  s->seq_cnt++;
+}
+static void
+sdefl_reg_match(struct sdefl *s, int off, int len) {
+  struct sdefl_match_codes cod;
+  sdefl_match_codes(&cod, off, len);
+  s->freq.lit[cod.lc]++;
+  s->freq.off[cod.dc]++;
+}
+struct sdefl_match {
+  int off;
+  int len;
+};
+static void
+sdefl_fnd(struct sdefl_match *m, const struct sdefl *s,
+          int chain_len, int max_match, const unsigned char *in, int p) {
+  int i = s->tbl[sdefl_hash32(&in[p])];
+  int limit = ((p-SDEFL_WIN_SIZ)<SDEFL_NIL)?SDEFL_NIL:(p-SDEFL_WIN_SIZ);
+  while (i > limit) {
+    if (in[i+m->len] == in[p+m->len] &&
+        (sdefl_uload32(&in[i]) == sdefl_uload32(&in[p]))){
+      int n = SDEFL_MIN_MATCH;
+      while (n < max_match && in[i+n] == in[p+n]) n++;
+      if (n > m->len) {
+        m->len = n, m->off = p - i;
+        if (n == max_match) break;
+      }
+    }
+    if (!(--chain_len)) break;
+    i = s->prv[i&SDEFL_WIN_MSK];
+  }
+}
+static int
+sdefl_compr(struct sdefl *s, unsigned char *out, const unsigned char *in,
+            int in_len, int lvl) {
+  unsigned char *q = out;
+  static const unsigned char pref[] = {8,10,14,24,30,48,65,96,130};
+  int max_chain = (lvl < 8) ? (1 << (lvl + 1)): (1 << 13);
+  int n, i = 0, litlen = 0;
+  for (n = 0; n < SDEFL_HASH_SIZ; ++n) {
+    s->tbl[n] = SDEFL_NIL;
+  }
+  do {int blk_end = i + SDEFL_BLK_MAX < in_len ? i + SDEFL_BLK_MAX : in_len;
+    while (i < blk_end) {
+      struct sdefl_match m = {0};
+      int max_match = ((in_len-i)>SDEFL_MAX_MATCH) ? SDEFL_MAX_MATCH:(in_len-i);
+      int nice_match = pref[lvl] < max_match ? pref[lvl] : max_match;
+      int run = 1, inc = 1, run_inc;
+      if (max_match > SDEFL_MIN_MATCH) {
+        sdefl_fnd(&m, s, max_chain, max_match, in, i);
+      }
+      if (lvl >= 5 && m.len >= SDEFL_MIN_MATCH && m.len < nice_match){
+        struct sdefl_match m2 = {0};
+        sdefl_fnd(&m2, s, max_chain, m.len+1, in, i+1);
+        m.len = (m2.len > m.len) ? 0 : m.len;
+      }
+      if (m.len >= SDEFL_MIN_MATCH) {
+        if (litlen) {
+          sdefl_seq(s, i - litlen, litlen);
+          litlen = 0;
+        }
+        sdefl_seq(s, -m.off, m.len);
+        sdefl_reg_match(s, m.off, m.len);
+        if (lvl < 2 && m.len >= nice_match) {
+          inc = m.len;
+        } else {
+          run = m.len;
+        }
+      } else {
+        s->freq.lit[in[i]]++;
+        litlen++;
+      }
+      run_inc = run * inc;
+      if (in_len - (i + run_inc) > SDEFL_MIN_MATCH) {
+        while (run-- > 0) {
+          unsigned h = sdefl_hash32(&in[i]);
+          s->prv[i&SDEFL_WIN_MSK] = s->tbl[h];
+          s->tbl[h] = i, i += inc;
+        }
+      } else {
+        i += run_inc;
+      }
+    }
+    if (litlen) {
+      sdefl_seq(s, i - litlen, litlen);
+      litlen = 0;
+    }
+    sdefl_flush(&q, s, blk_end == in_len, in);
+  } while (i < in_len);
+
+  if (s->bitcnt)
+    sdefl_put(&q, s, 0x00, 8 - s->bitcnt);
+  return (int)(q - out);
+}
+extern int
+sdeflate(struct sdefl *s, void *out, const void *in, int n, int lvl) {
+  s->bits = s->bitcnt = 0;
+  return sdefl_compr(s, (unsigned char*)out, (const unsigned char*)in, n, lvl);
+}
+static unsigned
+sdefl_adler32(unsigned adler32, const unsigned char *in, int in_len) {
+  #define SDEFL_ADLER_INIT (1)
+  const unsigned ADLER_MOD = 65521;
+  unsigned s1 = adler32 & 0xffff;
+  unsigned s2 = adler32 >> 16;
+  unsigned blk_len, i;
+
+  blk_len = in_len % 5552;
+  while (in_len) {
+    for (i = 0; i + 7 < blk_len; i += 8) {
+      s1 += in[0]; s2 += s1;
+      s1 += in[1]; s2 += s1;
+      s1 += in[2]; s2 += s1;
+      s1 += in[3]; s2 += s1;
+      s1 += in[4]; s2 += s1;
+      s1 += in[5]; s2 += s1;
+      s1 += in[6]; s2 += s1;
+      s1 += in[7]; s2 += s1;
+      in += 8;
+    }
+    for (; i < blk_len; ++i) {
+      s1 += *in++, s2 += s1;
+    }
+    s1 %= ADLER_MOD;
+    s2 %= ADLER_MOD;
+    in_len -= blk_len;
+    blk_len = 5552;
+  }
+  return (unsigned)(s2 << 16) + (unsigned)s1;
+}
+extern int
+zsdeflate(struct sdefl *s, void *out, const void *in, int n, int lvl) {
+  int p = 0;
+  unsigned a = 0;
+  unsigned char *q = (unsigned char*)out;
+
+  s->bits = s->bitcnt = 0;
+  sdefl_put(&q, s, 0x78, 8); /* deflate, 32k window */
+  sdefl_put(&q, s, 0x01, 8); /* fast compression */
+  q += sdefl_compr(s, q, (const unsigned char*)in, n, lvl);
+
+  /* append adler checksum */
+  a = sdefl_adler32(SDEFL_ADLER_INIT, (const unsigned char*)in, n);
+  for (p = 0; p < 4; ++p) {
+    sdefl_put(&q, s, (a >> 24) & 0xFF, 8);
+    a <<= 8;
+  }
+  return (int)(q - (unsigned char*)out);
+}
+extern int
+sdefl_bound(int len) {
+  int a = 128 + (len * 110) / 100;
+  int b = 128 + len + ((len / (31 * 1024)) + 1) * 5;
+  return (a > b) ? a : b;
+}
+#endif /* SDEFL_IMPLEMENTATION */
+
diff --git a/libs/raylib/src/external/sinfl.h b/libs/raylib/src/external/sinfl.h
new file mode 100644
index 0000000..37c1aae
--- /dev/null
+++ b/libs/raylib/src/external/sinfl.h
@@ -0,0 +1,464 @@
+/*
+# Small Deflate
+`sdefl` is a small bare bone lossless compression library in ANSI C (ISO C90)
+which implements the Deflate (RFC 1951) compressed data format specification standard.
+It is mainly tuned to get as much speed and compression ratio from as little code
+as needed to keep the implementation as concise as possible.
+
+## Features
+- Portable single header and source file duo written in ANSI C (ISO C90)
+- Dual license with either MIT or public domain
+- Small implementation
+    - Deflate: 525 LoC
+    - Inflate: 320 LoC
+- Webassembly:
+    - Deflate ~3.7 KB (~2.2KB compressed)
+    - Inflate ~3.6 KB (~2.2KB compressed)
+
+## Usage:
+This file behaves differently depending on what symbols you define
+before including it.
+
+Header-File mode:
+If you do not define `SINFL_IMPLEMENTATION` before including this file, it
+will operate in header only mode. In this mode it declares all used structs
+and the API of the library without including the implementation of the library.
+
+Implementation mode:
+If you define `SINFL_IMPLEMENTATION` before including this file, it will
+compile the implementation. Make sure that you only include
+this file implementation in *one* C or C++ file to prevent collisions.
+
+### Benchmark
+
+| Compressor name         | Compression| Decompress.| Compr. size | Ratio |
+| ------------------------| -----------| -----------| ----------- | ----- |
+| sdefl 1.0 -0            |   127 MB/s |   233 MB/s |    40004116 | 39.88 |
+| sdefl 1.0 -1            |   111 MB/s |   259 MB/s |    38940674 | 38.82 |
+| sdefl 1.0 -5            |    45 MB/s |   275 MB/s |    36577183 | 36.46 |
+| sdefl 1.0 -7            |    38 MB/s |   276 MB/s |    36523781 | 36.41 |
+| zlib 1.2.11 -1          |    72 MB/s |   307 MB/s |    42298774 | 42.30 |
+| zlib 1.2.11 -6          |    24 MB/s |   313 MB/s |    36548921 | 36.55 |
+| zlib 1.2.11 -9          |    20 MB/s |   314 MB/s |    36475792 | 36.48 |
+| miniz 1.0 -1            |   122 MB/s |   208 MB/s |    48510028 | 48.51 |
+| miniz 1.0 -6            |    27 MB/s |   260 MB/s |    36513697 | 36.51 |
+| miniz 1.0 -9            |    23 MB/s |   261 MB/s |    36460101 | 36.46 |
+| libdeflate 1.3 -1       |   147 MB/s |   667 MB/s |    39597378 | 39.60 |
+| libdeflate 1.3 -6       |    69 MB/s |   689 MB/s |    36648318 | 36.65 |
+| libdeflate 1.3 -9       |    13 MB/s |   672 MB/s |    35197141 | 35.20 |
+| libdeflate 1.3 -12      |  8.13 MB/s |   670 MB/s |    35100568 | 35.10 |
+
+### Compression
+Results on the [Silesia compression corpus](http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia):
+
+| File    |   Original | `sdefl 0`  	| `sdefl 5` 	| `sdefl 7` |
+| :------ | ---------: | -----------------: | ---------: | ----------: |
+| dickens | 10.192.446 |  4,260,187|  3,845,261|   3,833,657 |
+| mozilla | 51.220.480 | 20,774,706 | 19,607,009 |  19,565,867 |
+| mr      |  9.970.564 | 3,860,531 |  3,673,460 |   3,665,627 |
+| nci     | 33.553.445 | 4,030,283 |  3,094,526 |   3,006,075 |
+| ooffice |  6.152.192 | 3,320,063 |  3,186,373 |   3,183,815 |
+| osdb    | 10.085.684 | 3,919,646 |  3,649,510 |   3,649,477 |
+| reymont |  6.627.202 | 2,263,378 |  1,857,588 |   1,827,237 |
+| samba   | 21.606.400 | 6,121,797 |  5,462,670 |   5,450,762 |
+| sao     |  7.251.944 | 5,612,421 |  5,485,380 |   5,481,765 |
+| webster | 41.458.703 | 13,972,648 | 12,059,432 |  11,991,421 |
+| xml     |  5.345.280 | 886,620|    674,009 |     662,141 |
+| x-ray   |  8.474.240 | 6,304,655 |  6,244,779 |   6,244,779 |
+
+## License
+```
+------------------------------------------------------------------------------
+This software is available under 2 licenses -- choose whichever you prefer.
+------------------------------------------------------------------------------
+ALTERNATIVE A - MIT License
+Copyright (c) 2020 Micha Mettke
+Permission is hereby granted, free of charge, to any person obtaining a copy of
+this software and associated documentation files (the "Software"), to deal in
+the Software without restriction, including without limitation the rights to
+use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+of the Software, and to permit persons to whom the Software is furnished to do
+so, subject to the following conditions:
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+------------------------------------------------------------------------------
+ALTERNATIVE B - Public Domain (www.unlicense.org)
+This is free and unencumbered software released into the public domain.
+Anyone is free to copy, modify, publish, use, compile, sell, or distribute this
+software, either in source code form or as a compiled binary, for any purpose,
+commercial or non-commercial, and by any means.
+In jurisdictions that recognize copyright laws, the author or authors of this
+software dedicate any and all copyright interest in the software to the public
+domain. We make this dedication for the benefit of the public at large and to
+the detriment of our heirs and successors. We intend this dedication to be an
+overt act of relinquishment in perpetuity of all present and future rights to
+this software under copyright law.
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+------------------------------------------------------------------------------
+```
+*/
+#ifndef SINFL_H_INCLUDED
+#define SINFL_H_INCLUDED
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define SINFL_PRE_TBL_SIZE 128
+#define SINFL_LIT_TBL_SIZE 1334
+#define SINFL_OFF_TBL_SIZE 402
+
+struct sinfl {
+  int bits, bitcnt;
+  unsigned lits[SINFL_LIT_TBL_SIZE];
+  unsigned dsts[SINFL_OFF_TBL_SIZE];
+};
+extern int sinflate(void *out, const void *in, int size);
+extern int zsinflate(void *out, const void *in, int size);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* SINFL_H_INCLUDED */
+
+#ifdef SINFL_IMPLEMENTATION
+
+#include <string.h> /* memcpy, memset */
+
+static int
+sinfl_bsr(unsigned n) {
+#ifdef _MSC_VER
+  _BitScanReverse(&n, n);
+  return n;
+#elif defined(__GNUC__) || defined(__clang__)
+  return 31 - __builtin_clz(n);
+#endif
+}
+static int
+sinfl_get(const unsigned char **src, const unsigned char *end, struct sinfl *s,
+          int n) {
+  const unsigned char *in = *src;
+  int v = s->bits & ((1 << n)-1);
+  s->bits >>= n;
+  s->bitcnt = s->bitcnt - n;
+  s->bitcnt = s->bitcnt < 0 ? 0 : s->bitcnt;
+  while (s->bitcnt < 16 && in < end) {
+    s->bits |= (*in++) << s->bitcnt;
+    s->bitcnt += 8;
+  }
+  *src = in;
+  return v;
+}
+struct sinfl_gen {
+  int len;
+  int cnt;
+  int word;
+  short* sorted;
+};
+static int
+sinfl_build_tbl(struct sinfl_gen *gen, unsigned *tbl, int tbl_bits,
+                const int *cnt) {
+  int tbl_end = 0;
+  while (!(gen->cnt = cnt[gen->len])) {
+    ++gen->len;
+  }
+  tbl_end = 1 << gen->len;
+  while (gen->len <= tbl_bits) {
+    do {unsigned bit = 0;
+      tbl[gen->word] = (*gen->sorted++ << 16) | gen->len;
+      if (gen->word == tbl_end - 1) {
+        for (; gen->len < tbl_bits; gen->len++) {
+          memcpy(&tbl[tbl_end], tbl, (size_t)tbl_end * sizeof(tbl[0]));
+          tbl_end <<= 1;
+        }
+        return 1;
+      }
+      bit = 1 << sinfl_bsr((unsigned)(gen->word ^ (tbl_end - 1)));
+      gen->word &= bit - 1;
+      gen->word |= bit;
+    } while (--gen->cnt);
+    do {
+      if (++gen->len <= tbl_bits) {
+        memcpy(&tbl[tbl_end], tbl, (size_t)tbl_end * sizeof(tbl[0]));
+        tbl_end <<= 1;
+      }
+    } while (!(gen->cnt = cnt[gen->len]));
+  }
+  return 0;
+}
+static void
+sinfl_build_subtbl(struct sinfl_gen *gen, unsigned *tbl, int tbl_bits,
+                   const int *cnt) {
+  int sub_bits = 0;
+  int sub_start = 0;
+  int sub_prefix = -1;
+  int tbl_end = 1 << tbl_bits;
+  while (1) {
+    unsigned entry;
+    int bit, stride, i;
+    /* start new subtable */
+    if ((gen->word & ((1 << tbl_bits)-1)) != sub_prefix) {
+      int used = 0;
+      sub_prefix = gen->word & ((1 << tbl_bits)-1);
+      sub_start = tbl_end;
+      sub_bits = gen->len - tbl_bits;
+      used = gen->cnt;
+      while (used < (1 << sub_bits)) {
+        sub_bits++;
+        used = (used << 1) + cnt[tbl_bits + sub_bits];
+      }
+      tbl_end = sub_start + (1 << sub_bits);
+      tbl[sub_prefix] = (sub_start << 16) | 0x10 | (sub_bits & 0xf);
+    }
+    /* fill subtable */
+    entry = (*gen->sorted << 16) | ((gen->len - tbl_bits) & 0xf);
+    gen->sorted++;
+    i = sub_start + (gen->word >> tbl_bits);
+    stride = 1 << (gen->len - tbl_bits);
+    do {
+      tbl[i] = entry;
+      i += stride;
+    } while (i < tbl_end);
+    if (gen->word == (1 << gen->len)-1) {
+      return;
+    }
+    bit = 1 << sinfl_bsr(gen->word ^ ((1 << gen->len) - 1));
+    gen->word &= bit - 1;
+    gen->word |= bit;
+    gen->cnt--;
+    while (!gen->cnt) {
+      gen->cnt = cnt[++gen->len];
+    }
+  }
+}
+static void
+sinfl_build(unsigned *tbl, unsigned char *lens, int tbl_bits, int maxlen,
+            int symcnt) {
+  int i, used = 0;
+  short sort[288];
+  int cnt[16] = {0}, off[16]= {0};
+  struct sinfl_gen gen = {0};
+  gen.sorted = sort;
+  gen.len = 1;
+
+  for (i = 0; i < symcnt; ++i)
+    cnt[lens[i]]++;
+  off[1] = cnt[0];
+  for (i = 1; i < maxlen; ++i) {
+    off[i + 1] = off[i] + cnt[i];
+    used = (used << 1) + cnt[i];
+  }
+  used = (used << 1) + cnt[i];
+  for (i = 0; i < symcnt; ++i)
+    gen.sorted[off[lens[i]]++] = (short)i;
+  gen.sorted += off[0];
+
+  if (used < (1 << maxlen)){
+    for (i = 0; i < 1 << tbl_bits; ++i)
+      tbl[i] = (0 << 16u) | 1;
+    return;
+  }
+  if (!sinfl_build_tbl(&gen, tbl, tbl_bits, cnt)){
+    sinfl_build_subtbl(&gen, tbl, tbl_bits, cnt);
+  }
+}
+static int
+sinfl_decode(const unsigned char **in, const unsigned char *end,
+             struct sinfl *s, const unsigned *tbl, int bit_len) {
+  int idx = s->bits & ((1 << bit_len) - 1);
+  unsigned key = tbl[idx];
+  if (key & 0x10) {
+    /* sub-table lookup */
+    int len = key & 0x0f;
+    sinfl_get(in, end, s, bit_len);
+    idx = s->bits & ((1 << len)-1);
+    key = tbl[((key >> 16) & 0xffff) + (unsigned)idx];
+  }
+  sinfl_get(in, end, s, key & 0x0f);
+  return (key >> 16) & 0x0fff;
+}
+static int
+sinfl_decompress(unsigned char *out, const unsigned char *in, int size) {
+  static const unsigned char order[] = {16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15};
+  static const short dbase[30+2] = {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};
+  static const unsigned char dbits[30+2] = {0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,
+      10,10,11,11,12,12,13,13,0,0};
+  static const short lbase[29+2] = {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 const unsigned char lbits[29+2] = {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};
+
+  const unsigned char *e = in + size, *o = out;
+  enum sinfl_states {hdr,stored,fixed,dyn,blk};
+  enum sinfl_states state = hdr;
+  struct sinfl s = {0};
+  int last = 0;
+
+  sinfl_get(&in,e,&s,0); /* buffer input */
+  while (in < e || s.bitcnt) {
+    switch (state) {
+    case hdr: {
+      int type = 0; /* block header */
+      last = sinfl_get(&in,e,&s,1);
+      type = sinfl_get(&in,e,&s,2);
+
+      switch (type) {default: return (int)(out-o);
+      case 0x00: state = stored; break;
+      case 0x01: state = fixed; break;
+      case 0x02: state = dyn; break;}
+    } break;
+    case stored: {
+      int len; /* uncompressed block */
+      sinfl_get(&in,e,&s,s.bitcnt & 7);
+      len = sinfl_get(&in,e,&s,16);
+      //int nlen = sinfl_get(&in,e,&s,16);
+      in -= 2; s.bitcnt = 0;
+
+      if (len > (e-in) || !len)
+        return (int)(out-o);
+      memcpy(out, in, (size_t)len);
+      in += len, out += len;
+      state = hdr;
+    } break;
+    case fixed: {
+      /* fixed huffman codes */
+      int n; unsigned char lens[288+32];
+      for (n = 0; n <= 143; n++) lens[n] = 8;
+      for (n = 144; n <= 255; n++) lens[n] = 9;
+      for (n = 256; n <= 279; n++) lens[n] = 7;
+      for (n = 280; n <= 287; n++) lens[n] = 8;
+      for (n = 0; n < 32; n++) lens[288+n] = 5;
+
+      /* build lit/dist tables */
+      sinfl_build(s.lits, lens, 10, 15, 288);
+      sinfl_build(s.dsts, lens + 288, 8, 15, 32);
+      state = blk;
+    } break;
+    case dyn: {
+        /* dynamic huffman codes */
+        int n, i;
+        unsigned hlens[SINFL_PRE_TBL_SIZE];
+        unsigned char nlens[19] = {0}, lens[288+32];
+        int nlit = 257 + sinfl_get(&in,e,&s,5);
+        int ndist = 1 + sinfl_get(&in,e,&s,5);
+        int nlen = 4 + sinfl_get(&in,e,&s,4);
+        for (n = 0; n < nlen; n++)
+          nlens[order[n]] = (unsigned char)sinfl_get(&in,e,&s,3);
+        sinfl_build(hlens, nlens, 7, 7, 19);
+
+        /* decode code lengths */
+        for (n = 0; n < nlit + ndist;) {
+          int sym = sinfl_decode(&in, e, &s, hlens, 7);
+          switch (sym) {default: lens[n++] = (unsigned char)sym; break;
+          case 16: for (i=3+sinfl_get(&in,e,&s,2);i;i--,n++) lens[n]=lens[n-1]; break;
+          case 17: for (i=3+sinfl_get(&in,e,&s,3);i;i--,n++) lens[n]=0; break;
+          case 18: for (i=11+sinfl_get(&in,e,&s,7);i;i--,n++) lens[n]=0; break;}
+        }
+        /* build lit/dist tables */
+        sinfl_build(s.lits, lens, 10, 15, nlit);
+        sinfl_build(s.dsts, lens + nlit, 8, 15, ndist);
+        state = blk;
+    } break;
+    case blk: {
+      /* decompress block */
+      int i, sym = sinfl_decode(&in, e, &s, s.lits, 10);
+      if (sym > 256) {sym -= 257; /* match symbol */
+        {int len = sinfl_get(&in, e, &s, lbits[sym]) + lbase[sym];
+        int dsym = sinfl_decode(&in, e, &s, s.dsts, 8);
+        int offs = sinfl_get(&in, e, &s, dbits[dsym]) + dbase[dsym];
+        if (offs > (int)(out-o)) {
+          return (int)(out-o);
+        } else if (offs == 1) {
+          /* rle match copying */
+          unsigned char c = *(out - offs);
+          unsigned long w = (c << 24) | (c << 16) | (c << 8) | c;
+          for (i = 0; i < len >> 2; ++i) {
+            memcpy(out, &w, 4);
+            out += 4;
+          }
+          len = len & 3;
+        } else if (offs >= 4) {
+          /* copy match */
+          int wcnt = len >> 2;
+          for (i = 0; i < wcnt; ++i) {
+            unsigned long w = 0;
+            memcpy(&w, out - offs, 4);
+            memcpy(out, &w, 4);
+            out += 4;
+          }
+          len = len & 3;
+        }
+        for (i = 0; i < len; ++i)
+          {*out = *(out-offs), out++;}
+        }
+      } else if (sym == 256) {
+        /* end of block */
+        if (last) return (int)(out-o);
+        state = hdr;
+        break;
+        /* literal */
+      } else *out++ = (unsigned char)sym;
+    } break;}
+  }
+  return (int)(out-o);
+}
+extern int
+sinflate(void *out, const void *in, int size) {
+  return sinfl_decompress((unsigned char*)out, (const unsigned char*)in, size);
+}
+static unsigned
+sinfl_adler32(unsigned adler32, const unsigned char *in, int in_len) {
+  const unsigned ADLER_MOD = 65521;
+  unsigned s1 = adler32 & 0xffff;
+  unsigned s2 = adler32 >> 16;
+  unsigned blk_len, i;
+
+  blk_len = in_len % 5552;
+  while (in_len) {
+    for (i=0; i + 7 < blk_len; i += 8) {
+      s1 += in[0]; s2 += s1;
+      s1 += in[1]; s2 += s1;
+      s1 += in[2]; s2 += s1;
+      s1 += in[3]; s2 += s1;
+      s1 += in[4]; s2 += s1;
+      s1 += in[5]; s2 += s1;
+      s1 += in[6]; s2 += s1;
+      s1 += in[7]; s2 += s1;
+      in += 8;
+    }
+    for (; i < blk_len; ++i)
+      s1 += *in++, s2 += s1;
+    s1 %= ADLER_MOD; s2 %= ADLER_MOD;
+    in_len -= blk_len;
+    blk_len = 5552;
+  } return (unsigned)(s2 << 16) + (unsigned)s1;
+}
+extern int
+zsinflate(void *out, const void *mem, int size) {
+  const unsigned char *in = (const unsigned char*)mem;
+  if (size >= 6) {
+    const unsigned char *eob = in + size - 4;
+    int n = sinfl_decompress((unsigned char*)out, in + 2u, size);
+    unsigned a = sinfl_adler32(1u, (unsigned char*)out, n);
+    unsigned h = eob[0] << 24 | eob[1] << 16 | eob[2] << 8 | eob[3] << 0;
+    return a == h ? n : -1;
+  } else {
+    return -1;
+  }
+}
+#endif
+
diff --git a/libs/raylib/src/uwp_events.h b/libs/raylib/src/uwp_events.h
new file mode 100644
index 0000000..2c403fd
--- /dev/null
+++ b/libs/raylib/src/uwp_events.h
@@ -0,0 +1,119 @@
+/**********************************************************************************************
+*
+*   raylib.uwp_events - Functions for bootstrapping UWP functionality within raylib's core.
+*
+*
+*   LICENSE: zlib/libpng
+*
+*   Copyright (c) 2020-2020 Reece Mackie (@Rover656)
+*
+*   This software is provided "as-is", without any express or implied warranty. In no event
+*   will the authors be held liable for any damages arising from the use of this software.
+*
+*   Permission is granted to anyone to use this software for any purpose, including commercial
+*   applications, and to alter it and redistribute it freely, subject to the following restrictions:
+*
+*     1. The origin of this software must not be misrepresented; you must not claim that you
+*     wrote the original software. If you use this software in a product, an acknowledgment
+*     in the product documentation would be appreciated but is not required.
+*
+*     2. Altered source versions must be plainly marked as such, and must not be misrepresented
+*     as being the original software.
+*
+*     3. This notice may not be removed or altered from any source distribution.
+*
+**********************************************************************************************/
+
+#ifndef UWP_EVENTS_H
+#define UWP_EVENTS_H
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+#if defined(PLATFORM_UWP)
+
+// Determine if UWP functions are set and ready for raylib's use.
+bool UWPIsConfigured();
+
+// Call this to set the UWP data path you wish for saving and loading.
+void UWPSetDataPath(const char* path);
+
+// Function for getting program time.
+typedef double(*UWPQueryTimeFunc)();
+UWPQueryTimeFunc UWPGetQueryTimeFunc(void);
+void UWPSetQueryTimeFunc(UWPQueryTimeFunc func);
+
+// Function for sleeping the current thread
+typedef void (*UWPSleepFunc)(double sleepUntil);
+UWPSleepFunc UWPGetSleepFunc(void);
+void UWPSetSleepFunc(UWPSleepFunc func);
+
+// Function for querying the display size
+typedef void(*UWPDisplaySizeFunc)(int* width, int* height);
+UWPDisplaySizeFunc UWPGetDisplaySizeFunc(void);
+void UWPSetDisplaySizeFunc(UWPDisplaySizeFunc func);
+
+// Functions for mouse cursor control
+typedef void(*UWPMouseFunc)(void);
+UWPMouseFunc UWPGetMouseLockFunc();
+void UWPSetMouseLockFunc(UWPMouseFunc func);
+UWPMouseFunc UWPGetMouseUnlockFunc();
+void UWPSetMouseUnlockFunc(UWPMouseFunc func);
+UWPMouseFunc UWPGetMouseShowFunc();
+void UWPSetMouseShowFunc(UWPMouseFunc func);
+UWPMouseFunc UWPGetMouseHideFunc();
+void UWPSetMouseHideFunc(UWPMouseFunc func);
+
+// Function for setting mouse cursor position.
+typedef void (*UWPMouseSetPosFunc)(int x, int y);
+UWPMouseSetPosFunc UWPGetMouseSetPosFunc();
+void UWPSetMouseSetPosFunc(UWPMouseSetPosFunc func);
+
+// The below functions are implemented in core.c but are placed here so they can be called by user code.
+// This choice is made as platform-specific code is preferred to be kept away from raylib.h
+
+// Call this when a Key is pressed or released.
+void UWPKeyDownEvent(int key, bool down, bool controlKey);
+
+// Call this on the CoreWindow::CharacterRecieved event
+void UWPKeyCharEvent(int key);
+
+// Call when a mouse button state changes
+void UWPMouseButtonEvent(int button, bool down);
+
+// Call when the mouse cursor moves
+void UWPMousePosEvent(double x, double y);
+
+// Call when the mouse wheel moves
+void UWPMouseWheelEvent(int deltaY);
+
+// Call when the window resizes
+void UWPResizeEvent(int width, int height);
+
+// Call when a gamepad is made active
+void UWPActivateGamepadEvent(int gamepad, bool active);
+
+// Call when a gamepad button state changes
+void UWPRegisterGamepadButton(int gamepad, int button, bool down);
+
+// Call when a gamepad axis state changes
+void UWPRegisterGamepadAxis(int gamepad, int axis, float value);
+
+// Call when the touch point moves
+void UWPGestureMove(int pointer, float x, float y);
+
+// Call when there is a touch down or up
+void UWPGestureTouch(int pointer, float x, float y, bool touch);
+
+// Set the core window pointer so that we can pass it to EGL.
+void* UWPGetCoreWindowPtr();
+void UWPSetCoreWindowPtr(void* ptr);
+
+#if defined(__cplusplus)
+}
+#endif
+
+#endif // PLATFORM_UWP
+
+#endif // UWP_EVENTS_H