diff options
author | Nao Pross <np@0hm.ch> | 2024-02-12 14:52:43 +0100 |
---|---|---|
committer | Nao Pross <np@0hm.ch> | 2024-02-12 14:52:43 +0100 |
commit | eda5bc26f44ee9a6f83dcf8c91f17296d7fc509d (patch) | |
tree | bc2efa38ff4e350f9a111ac87065cd7ae9a911c7 /src/imgui/misc/fonts | |
download | fsisotool-eda5bc26f44ee9a6f83dcf8c91f17296d7fc509d.tar.gz fsisotool-eda5bc26f44ee9a6f83dcf8c91f17296d7fc509d.zip |
Move into version control
Diffstat (limited to '')
-rw-r--r-- | src/imgui/misc/fonts/Cousine-Regular.ttf | bin | 0 -> 43912 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/DroidSans.ttf | bin | 0 -> 190044 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/Karla-Regular.ttf | bin | 0 -> 16848 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/ProggyClean.ttf | bin | 0 -> 41208 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/ProggyTiny.ttf | bin | 0 -> 35656 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/Roboto-Medium.ttf | bin | 0 -> 162588 bytes | |||
-rw-r--r-- | src/imgui/misc/fonts/binary_to_compressed_c.cpp | 388 |
7 files changed, 388 insertions, 0 deletions
diff --git a/src/imgui/misc/fonts/Cousine-Regular.ttf b/src/imgui/misc/fonts/Cousine-Regular.ttf Binary files differnew file mode 100644 index 0000000..70a0bf9 --- /dev/null +++ b/src/imgui/misc/fonts/Cousine-Regular.ttf diff --git a/src/imgui/misc/fonts/DroidSans.ttf b/src/imgui/misc/fonts/DroidSans.ttf Binary files differnew file mode 100644 index 0000000..767c63a --- /dev/null +++ b/src/imgui/misc/fonts/DroidSans.ttf diff --git a/src/imgui/misc/fonts/Karla-Regular.ttf b/src/imgui/misc/fonts/Karla-Regular.ttf Binary files differnew file mode 100644 index 0000000..81b3de6 --- /dev/null +++ b/src/imgui/misc/fonts/Karla-Regular.ttf diff --git a/src/imgui/misc/fonts/ProggyClean.ttf b/src/imgui/misc/fonts/ProggyClean.ttf Binary files differnew file mode 100644 index 0000000..0270cdf --- /dev/null +++ b/src/imgui/misc/fonts/ProggyClean.ttf diff --git a/src/imgui/misc/fonts/ProggyTiny.ttf b/src/imgui/misc/fonts/ProggyTiny.ttf Binary files differnew file mode 100644 index 0000000..1c4312c --- /dev/null +++ b/src/imgui/misc/fonts/ProggyTiny.ttf diff --git a/src/imgui/misc/fonts/Roboto-Medium.ttf b/src/imgui/misc/fonts/Roboto-Medium.ttf Binary files differnew file mode 100644 index 0000000..39c63d7 --- /dev/null +++ b/src/imgui/misc/fonts/Roboto-Medium.ttf diff --git a/src/imgui/misc/fonts/binary_to_compressed_c.cpp b/src/imgui/misc/fonts/binary_to_compressed_c.cpp new file mode 100644 index 0000000..f41d20f --- /dev/null +++ b/src/imgui/misc/fonts/binary_to_compressed_c.cpp @@ -0,0 +1,388 @@ +// dear imgui +// (binary_to_compressed_c.cpp) +// Helper tool to turn a file into a C array, if you want to embed font data in your source code. + +// The data is first compressed with stb_compress() to reduce source code size, +// then encoded in Base85 to fit in a string so we can fit roughly 4 bytes of compressed data into 5 bytes of source code (suggested by @mmalex) +// (If we used 32-bit constants it would require take 11 bytes of source code to encode 4 bytes, and be endianness dependent) +// Note that even with compression, the output array is likely to be bigger than the binary file.. +// Load compressed TTF fonts with ImGui::GetIO().Fonts->AddFontFromMemoryCompressedTTF() + +// Build with, e.g: +// # cl.exe binary_to_compressed_c.cpp +// # g++ binary_to_compressed_c.cpp +// # clang++ binary_to_compressed_c.cpp +// You can also find a precompiled Windows binary in the binary/demo package available from https://github.com/ocornut/imgui + +// Usage: +// binary_to_compressed_c.exe [-base85] [-nocompress] [-nostatic] <inputfile> <symbolname> +// Usage example: +// # binary_to_compressed_c.exe myfont.ttf MyFont > myfont.cpp +// # binary_to_compressed_c.exe -base85 myfont.ttf MyFont > myfont.cpp + +#define _CRT_SECURE_NO_WARNINGS +#include <stdio.h> +#include <string.h> +#include <stdlib.h> +#include <assert.h> + +// stb_compress* from stb.h - declaration +typedef unsigned int stb_uint; +typedef unsigned char stb_uchar; +stb_uint stb_compress(stb_uchar* out, stb_uchar* in, stb_uint len); + +static bool binary_to_compressed_c(const char* filename, const char* symbol, bool use_base85_encoding, bool use_compression, bool use_static); + +int main(int argc, char** argv) +{ + if (argc < 3) + { + printf("Syntax: %s [-base85] [-nocompress] [-nostatic] <inputfile> <symbolname>\n", argv[0]); + return 0; + } + + int argn = 1; + bool use_base85_encoding = false; + bool use_compression = true; + bool use_static = true; + while (argn < (argc - 2) && argv[argn][0] == '-') + { + if (strcmp(argv[argn], "-base85") == 0) { use_base85_encoding = true; argn++; } + else if (strcmp(argv[argn], "-nocompress") == 0) { use_compression = false; argn++; } + else if (strcmp(argv[argn], "-nostatic") == 0) { use_static = false; argn++; } + else + { + fprintf(stderr, "Unknown argument: '%s'\n", argv[argn]); + return 1; + } + } + + bool ret = binary_to_compressed_c(argv[argn], argv[argn + 1], use_base85_encoding, use_compression, use_static); + if (!ret) + fprintf(stderr, "Error opening or reading file: '%s'\n", argv[argn]); + return ret ? 0 : 1; +} + +char Encode85Byte(unsigned int x) +{ + x = (x % 85) + 35; + return (char)((x >= '\\') ? x + 1 : x); +} + +bool binary_to_compressed_c(const char* filename, const char* symbol, bool use_base85_encoding, bool use_compression, bool use_static) +{ + // Read file + FILE* f = fopen(filename, "rb"); + if (!f) return false; + int data_sz; + if (fseek(f, 0, SEEK_END) || (data_sz = (int)ftell(f)) == -1 || fseek(f, 0, SEEK_SET)) { fclose(f); return false; } + char* data = new char[data_sz + 4]; + if (fread(data, 1, data_sz, f) != (size_t)data_sz) { fclose(f); delete[] data; return false; } + memset((void*)(((char*)data) + data_sz), 0, 4); + fclose(f); + + // Compress + int maxlen = data_sz + 512 + (data_sz >> 2) + sizeof(int); // total guess + char* compressed = use_compression ? new char[maxlen] : data; + int compressed_sz = use_compression ? stb_compress((stb_uchar*)compressed, (stb_uchar*)data, data_sz) : data_sz; + if (use_compression) + memset(compressed + compressed_sz, 0, maxlen - compressed_sz); + + // Output as Base85 encoded + FILE* out = stdout; + fprintf(out, "// File: '%s' (%d bytes)\n", filename, (int)data_sz); + fprintf(out, "// Exported using binary_to_compressed_c.cpp\n"); + const char* static_str = use_static ? "static " : ""; + const char* compressed_str = use_compression ? "compressed_" : ""; + if (use_base85_encoding) + { + fprintf(out, "%sconst char %s_%sdata_base85[%d+1] =\n \"", static_str, symbol, compressed_str, (int)((compressed_sz + 3) / 4)*5); + char prev_c = 0; + for (int src_i = 0; src_i < compressed_sz; src_i += 4) + { + // This is made a little more complicated by the fact that ??X sequences are interpreted as trigraphs by old C/C++ compilers. So we need to escape pairs of ??. + unsigned int d = *(unsigned int*)(compressed + src_i); + for (unsigned int n5 = 0; n5 < 5; n5++, d /= 85) + { + char c = Encode85Byte(d); + fprintf(out, (c == '?' && prev_c == '?') ? "\\%c" : "%c", c); + prev_c = c; + } + if ((src_i % 112) == 112 - 4) + fprintf(out, "\"\n \""); + } + fprintf(out, "\";\n\n"); + } + else + { + fprintf(out, "%sconst unsigned int %s_%ssize = %d;\n", static_str, symbol, compressed_str, (int)compressed_sz); + fprintf(out, "%sconst unsigned int %s_%sdata[%d/4] =\n{", static_str, symbol, compressed_str, (int)((compressed_sz + 3) / 4)*4); + int column = 0; + for (int i = 0; i < compressed_sz; i += 4) + { + unsigned int d = *(unsigned int*)(compressed + i); + if ((column++ % 12) == 0) + fprintf(out, "\n 0x%08x, ", d); + else + fprintf(out, "0x%08x, ", d); + } + fprintf(out, "\n};\n\n"); + } + + // Cleanup + delete[] data; + if (use_compression) + delete[] compressed; + return true; +} + +// stb_compress* from stb.h - definition + +//////////////////// compressor /////////////////////// + +static stb_uint stb_adler32(stb_uint adler32, stb_uchar *buffer, stb_uint buflen) +{ + const unsigned long ADLER_MOD = 65521; + unsigned long s1 = adler32 & 0xffff, s2 = adler32 >> 16; + unsigned long blocklen, i; + + blocklen = buflen % 5552; + while (buflen) { + for (i=0; i + 7 < blocklen; i += 8) { + s1 += buffer[0], s2 += s1; + s1 += buffer[1], s2 += s1; + s1 += buffer[2], s2 += s1; + s1 += buffer[3], s2 += s1; + s1 += buffer[4], s2 += s1; + s1 += buffer[5], s2 += s1; + s1 += buffer[6], s2 += s1; + s1 += buffer[7], s2 += s1; + + buffer += 8; + } + + for (; i < blocklen; ++i) + s1 += *buffer++, s2 += s1; + + s1 %= ADLER_MOD, s2 %= ADLER_MOD; + buflen -= blocklen; + blocklen = 5552; + } + return (s2 << 16) + s1; +} + +static unsigned int stb_matchlen(stb_uchar *m1, stb_uchar *m2, stb_uint maxlen) +{ + stb_uint i; + for (i=0; i < maxlen; ++i) + if (m1[i] != m2[i]) return i; + return i; +} + +// simple implementation that just takes the source data in a big block + +static stb_uchar *stb__out; +static FILE *stb__outfile; +static stb_uint stb__outbytes; + +static void stb__write(unsigned char v) +{ + fputc(v, stb__outfile); + ++stb__outbytes; +} + +//#define stb_out(v) (stb__out ? *stb__out++ = (stb_uchar) (v) : stb__write((stb_uchar) (v))) +#define stb_out(v) do { if (stb__out) *stb__out++ = (stb_uchar) (v); else stb__write((stb_uchar) (v)); } while (0) + +static void stb_out2(stb_uint v) { stb_out(v >> 8); stb_out(v); } +static void stb_out3(stb_uint v) { stb_out(v >> 16); stb_out(v >> 8); stb_out(v); } +static void stb_out4(stb_uint v) { stb_out(v >> 24); stb_out(v >> 16); stb_out(v >> 8 ); stb_out(v); } + +static void outliterals(stb_uchar *in, int numlit) +{ + while (numlit > 65536) { + outliterals(in,65536); + in += 65536; + numlit -= 65536; + } + + if (numlit == 0) ; + else if (numlit <= 32) stb_out (0x000020 + numlit-1); + else if (numlit <= 2048) stb_out2(0x000800 + numlit-1); + else /* numlit <= 65536) */ stb_out3(0x070000 + numlit-1); + + if (stb__out) { + memcpy(stb__out,in,numlit); + stb__out += numlit; + } else + fwrite(in, 1, numlit, stb__outfile); +} + +static int stb__window = 0x40000; // 256K + +static int stb_not_crap(int best, int dist) +{ + return ((best > 2 && dist <= 0x00100) + || (best > 5 && dist <= 0x04000) + || (best > 7 && dist <= 0x80000)); +} + +static stb_uint stb__hashsize = 32768; + +// note that you can play with the hashing functions all you +// want without needing to change the decompressor +#define stb__hc(q,h,c) (((h) << 7) + ((h) >> 25) + q[c]) +#define stb__hc2(q,h,c,d) (((h) << 14) + ((h) >> 18) + (q[c] << 7) + q[d]) +#define stb__hc3(q,c,d,e) ((q[c] << 14) + (q[d] << 7) + q[e]) + +static unsigned int stb__running_adler; + +static int stb_compress_chunk(stb_uchar *history, + stb_uchar *start, + stb_uchar *end, + int length, + int *pending_literals, + stb_uchar **chash, + stb_uint mask) +{ + (void)history; + int window = stb__window; + stb_uint match_max; + stb_uchar *lit_start = start - *pending_literals; + stb_uchar *q = start; + +#define STB__SCRAMBLE(h) (((h) + ((h) >> 16)) & mask) + + // stop short of the end so we don't scan off the end doing + // the hashing; this means we won't compress the last few bytes + // unless they were part of something longer + while (q < start+length && q+12 < end) { + int m; + stb_uint h1,h2,h3,h4, h; + stb_uchar *t; + int best = 2, dist=0; + + if (q+65536 > end) + match_max = (stb_uint)(end-q); + else + match_max = 65536; + +#define stb__nc(b,d) ((d) <= window && ((b) > 9 || stb_not_crap((int)(b),(int)(d)))) + +#define STB__TRY(t,p) /* avoid retrying a match we already tried */ \ + if (p ? dist != (int)(q-t) : 1) \ + if ((m = stb_matchlen(t, q, match_max)) > best) \ + if (stb__nc(m,q-(t))) \ + best = m, dist = (int)(q - (t)) + + // rather than search for all matches, only try 4 candidate locations, + // chosen based on 4 different hash functions of different lengths. + // this strategy is inspired by LZO; hashing is unrolled here using the + // 'hc' macro + h = stb__hc3(q,0, 1, 2); h1 = STB__SCRAMBLE(h); + t = chash[h1]; if (t) STB__TRY(t,0); + h = stb__hc2(q,h, 3, 4); h2 = STB__SCRAMBLE(h); + h = stb__hc2(q,h, 5, 6); t = chash[h2]; if (t) STB__TRY(t,1); + h = stb__hc2(q,h, 7, 8); h3 = STB__SCRAMBLE(h); + h = stb__hc2(q,h, 9,10); t = chash[h3]; if (t) STB__TRY(t,1); + h = stb__hc2(q,h,11,12); h4 = STB__SCRAMBLE(h); + t = chash[h4]; if (t) STB__TRY(t,1); + + // because we use a shared hash table, can only update it + // _after_ we've probed all of them + chash[h1] = chash[h2] = chash[h3] = chash[h4] = q; + + if (best > 2) + assert(dist > 0); + + // see if our best match qualifies + if (best < 3) { // fast path literals + ++q; + } else if (best > 2 && best <= 0x80 && dist <= 0x100) { + outliterals(lit_start, (int)(q-lit_start)); lit_start = (q += best); + stb_out(0x80 + best-1); + stb_out(dist-1); + } else if (best > 5 && best <= 0x100 && dist <= 0x4000) { + outliterals(lit_start, (int)(q-lit_start)); lit_start = (q += best); + stb_out2(0x4000 + dist-1); + stb_out(best-1); + } else if (best > 7 && best <= 0x100 && dist <= 0x80000) { + outliterals(lit_start, (int)(q-lit_start)); lit_start = (q += best); + stb_out3(0x180000 + dist-1); + stb_out(best-1); + } else if (best > 8 && best <= 0x10000 && dist <= 0x80000) { + outliterals(lit_start, (int)(q-lit_start)); lit_start = (q += best); + stb_out3(0x100000 + dist-1); + stb_out2(best-1); + } else if (best > 9 && dist <= 0x1000000) { + if (best > 65536) best = 65536; + outliterals(lit_start, (int)(q-lit_start)); lit_start = (q += best); + if (best <= 0x100) { + stb_out(0x06); + stb_out3(dist-1); + stb_out(best-1); + } else { + stb_out(0x04); + stb_out3(dist-1); + stb_out2(best-1); + } + } else { // fallback literals if no match was a balanced tradeoff + ++q; + } + } + + // if we didn't get all the way, add the rest to literals + if (q-start < length) + q = start+length; + + // the literals are everything from lit_start to q + *pending_literals = (int)(q - lit_start); + + stb__running_adler = stb_adler32(stb__running_adler, start, (stb_uint)(q - start)); + return (int)(q - start); +} + +static int stb_compress_inner(stb_uchar *input, stb_uint length) +{ + int literals = 0; + stb_uint len,i; + + stb_uchar **chash; + chash = (stb_uchar**) malloc(stb__hashsize * sizeof(stb_uchar*)); + if (chash == nullptr) return 0; // failure + for (i=0; i < stb__hashsize; ++i) + chash[i] = nullptr; + + // stream signature + stb_out(0x57); stb_out(0xbc); + stb_out2(0); + + stb_out4(0); // 64-bit length requires 32-bit leading 0 + stb_out4(length); + stb_out4(stb__window); + + stb__running_adler = 1; + + len = stb_compress_chunk(input, input, input+length, length, &literals, chash, stb__hashsize-1); + assert(len == length); + + outliterals(input+length - literals, literals); + + free(chash); + + stb_out2(0x05fa); // end opcode + + stb_out4(stb__running_adler); + + return 1; // success +} + +stb_uint stb_compress(stb_uchar *out, stb_uchar *input, stb_uint length) +{ + stb__out = out; + stb__outfile = nullptr; + + stb_compress_inner(input, length); + + return (stb_uint)(stb__out - out); +} |