From: Mike D. Lowis Date: Fri, 18 Sep 2015 15:25:48 +0000 (-0400) Subject: Added generated rune type files X-Git-Url: https://git.mdlowis.com/?a=commitdiff_plain;h=4d32ea0a81eaa72b8c05e511aef955deccea3866;p=archive%2Fcarl.git Added generated rune type files --- diff --git a/build.rb b/build.rb index bbd4644..88ffce3 100755 --- a/build.rb +++ b/build.rb @@ -13,6 +13,10 @@ end #------------------------------------------------------------------------------ # Release Build Targets #------------------------------------------------------------------------------ +# Generate the rune mappings +main_env.Command('Unicode', [], + "CMD" => ['./tools/unicode.rb', './tools/UnicodeData-8.0.0.txt', 'source/utf']) + # Build the library main_env.Library('libc.a', FileList['source/**/*.c']) @@ -20,7 +24,6 @@ main_env.Library('libc.a', FileList['source/**/*.c']) # Test Build Targets #------------------------------------------------------------------------------ if Opts[:profile].include? "test" - # Do nothing for now main_env.Program('test_libc', Dir["tests/**/*.c", "modules/atf/source/*.c"] + ['./libc.a']) main_env.Command('Unit Tests', 'test_libc', "CMD" => './test_libc') end diff --git a/source/utf/chartorune.c b/source/utf/chartorune.c new file mode 100644 index 0000000..db0ea93 --- /dev/null +++ b/source/utf/chartorune.c @@ -0,0 +1,9 @@ +#include "libc.h" + +int chartorune(Rune* r, char* s) +{ + (void)r; + (void)s; + return 0; +} + diff --git a/source/utf/controls.c b/source/utf/controls.c new file mode 100644 index 0000000..16e6f4d --- /dev/null +++ b/source/utf/controls.c @@ -0,0 +1,13 @@ +#include + +static Rune controls[4][2] = { + { 0x0, 0x8 }, + { 0xe, 0x1b }, + { 0x7f, 0x84 }, + { 0x86, 0x9f } +}; + +extern int runeinrange(const void* a, const void* b); +bool iscontrolrune(Rune ch) { + return (NULL != bsearch(&ch, controls, 4, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/fullrune.c b/source/utf/fullrune.c new file mode 100644 index 0000000..f3b8063 --- /dev/null +++ b/source/utf/fullrune.c @@ -0,0 +1,9 @@ +#include "libc.h" + +bool fullrune(char* s, int n) +{ + (void)s; + (void)n; + return false; +} + diff --git a/source/utf/lowers.c b/source/utf/lowers.c new file mode 100644 index 0000000..3f8b7da --- /dev/null +++ b/source/utf/lowers.c @@ -0,0 +1,639 @@ +#include + +static Rune lowers[630][2] = { + { 0x61, 0x7a }, + { 0xb5, 0xb5 }, + { 0xdf, 0xf6 }, + { 0xf8, 0xff }, + { 0x101, 0x101 }, + { 0x103, 0x103 }, + { 0x105, 0x105 }, + { 0x107, 0x107 }, + { 0x109, 0x109 }, + { 0x10b, 0x10b }, + { 0x10d, 0x10d }, + { 0x10f, 0x10f }, + { 0x111, 0x111 }, + { 0x113, 0x113 }, + { 0x115, 0x115 }, + { 0x117, 0x117 }, + { 0x119, 0x119 }, + { 0x11b, 0x11b }, + { 0x11d, 0x11d }, + { 0x11f, 0x11f }, + { 0x121, 0x121 }, + { 0x123, 0x123 }, + { 0x125, 0x125 }, + { 0x127, 0x127 }, + { 0x129, 0x129 }, + { 0x12b, 0x12b }, + { 0x12d, 0x12d }, + { 0x12f, 0x12f }, + { 0x131, 0x131 }, + { 0x133, 0x133 }, + { 0x135, 0x135 }, + { 0x137, 0x138 }, + { 0x13a, 0x13a }, + { 0x13c, 0x13c }, + { 0x13e, 0x13e }, + { 0x140, 0x140 }, + { 0x142, 0x142 }, + { 0x144, 0x144 }, + { 0x146, 0x146 }, + { 0x148, 0x149 }, + { 0x14b, 0x14b }, + { 0x14d, 0x14d }, + { 0x14f, 0x14f }, + { 0x151, 0x151 }, + { 0x153, 0x153 }, + { 0x155, 0x155 }, + { 0x157, 0x157 }, + { 0x159, 0x159 }, + { 0x15b, 0x15b }, + { 0x15d, 0x15d }, + { 0x15f, 0x15f }, + { 0x161, 0x161 }, + { 0x163, 0x163 }, + { 0x165, 0x165 }, + { 0x167, 0x167 }, + { 0x169, 0x169 }, + { 0x16b, 0x16b }, + { 0x16d, 0x16d }, + { 0x16f, 0x16f }, + { 0x171, 0x171 }, + { 0x173, 0x173 }, + { 0x175, 0x175 }, + { 0x177, 0x177 }, + { 0x17a, 0x17a }, + { 0x17c, 0x17c }, + { 0x17e, 0x180 }, + { 0x183, 0x183 }, + { 0x185, 0x185 }, + { 0x188, 0x188 }, + { 0x18c, 0x18d }, + { 0x192, 0x192 }, + { 0x195, 0x195 }, + { 0x199, 0x19b }, + { 0x19e, 0x19e }, + { 0x1a1, 0x1a1 }, + { 0x1a3, 0x1a3 }, + { 0x1a5, 0x1a5 }, + { 0x1a8, 0x1a8 }, + { 0x1aa, 0x1ab }, + { 0x1ad, 0x1ad }, + { 0x1b0, 0x1b0 }, + { 0x1b4, 0x1b4 }, + { 0x1b6, 0x1b6 }, + { 0x1b9, 0x1ba }, + { 0x1bd, 0x1bf }, + { 0x1c6, 0x1c6 }, + { 0x1c9, 0x1c9 }, + { 0x1cc, 0x1cc }, + { 0x1ce, 0x1ce }, + { 0x1d0, 0x1d0 }, + { 0x1d2, 0x1d2 }, + { 0x1d4, 0x1d4 }, + { 0x1d6, 0x1d6 }, + { 0x1d8, 0x1d8 }, + { 0x1da, 0x1da }, + { 0x1dc, 0x1dd }, + { 0x1df, 0x1df }, + { 0x1e1, 0x1e1 }, + { 0x1e3, 0x1e3 }, + { 0x1e5, 0x1e5 }, + { 0x1e7, 0x1e7 }, + { 0x1e9, 0x1e9 }, + { 0x1eb, 0x1eb }, + { 0x1ed, 0x1ed }, + { 0x1ef, 0x1f0 }, + { 0x1f3, 0x1f3 }, + { 0x1f5, 0x1f5 }, + { 0x1f9, 0x1f9 }, + { 0x1fb, 0x1fb }, + { 0x1fd, 0x1fd }, + { 0x1ff, 0x1ff }, + { 0x201, 0x201 }, + { 0x203, 0x203 }, + { 0x205, 0x205 }, + { 0x207, 0x207 }, + { 0x209, 0x209 }, + { 0x20b, 0x20b }, + { 0x20d, 0x20d }, + { 0x20f, 0x20f }, + { 0x211, 0x211 }, + { 0x213, 0x213 }, + { 0x215, 0x215 }, + { 0x217, 0x217 }, + { 0x219, 0x219 }, + { 0x21b, 0x21b }, + { 0x21d, 0x21d }, + { 0x21f, 0x21f }, + { 0x221, 0x221 }, + { 0x223, 0x223 }, + { 0x225, 0x225 }, + { 0x227, 0x227 }, + { 0x229, 0x229 }, + { 0x22b, 0x22b }, + { 0x22d, 0x22d }, + { 0x22f, 0x22f }, + { 0x231, 0x231 }, + { 0x233, 0x239 }, + { 0x23c, 0x23c }, + { 0x23f, 0x240 }, + { 0x242, 0x242 }, + { 0x247, 0x247 }, + { 0x249, 0x249 }, + { 0x24b, 0x24b }, + { 0x24d, 0x24d }, + { 0x24f, 0x293 }, + { 0x295, 0x2af }, + { 0x371, 0x371 }, + { 0x373, 0x373 }, + { 0x377, 0x377 }, + { 0x37b, 0x37d }, + { 0x390, 0x390 }, + { 0x3ac, 0x3ce }, + { 0x3d0, 0x3d1 }, + { 0x3d5, 0x3d7 }, + { 0x3d9, 0x3d9 }, + { 0x3db, 0x3db }, + { 0x3dd, 0x3dd }, + { 0x3df, 0x3df }, + { 0x3e1, 0x3e1 }, + { 0x3e3, 0x3e3 }, + { 0x3e5, 0x3e5 }, + { 0x3e7, 0x3e7 }, + { 0x3e9, 0x3e9 }, + { 0x3eb, 0x3eb }, + { 0x3ed, 0x3ed }, + { 0x3ef, 0x3f3 }, + { 0x3f5, 0x3f5 }, + { 0x3f8, 0x3f8 }, + { 0x3fb, 0x3fc }, + { 0x430, 0x45f }, + { 0x461, 0x461 }, + { 0x463, 0x463 }, + { 0x465, 0x465 }, + { 0x467, 0x467 }, + { 0x469, 0x469 }, + { 0x46b, 0x46b }, + { 0x46d, 0x46d }, + { 0x46f, 0x46f }, + { 0x471, 0x471 }, + { 0x473, 0x473 }, + { 0x475, 0x475 }, + { 0x477, 0x477 }, + { 0x479, 0x479 }, + { 0x47b, 0x47b }, + { 0x47d, 0x47d }, + { 0x47f, 0x47f }, + { 0x481, 0x481 }, + { 0x48b, 0x48b }, + { 0x48d, 0x48d }, + { 0x48f, 0x48f }, + { 0x491, 0x491 }, + { 0x493, 0x493 }, + { 0x495, 0x495 }, + { 0x497, 0x497 }, + { 0x499, 0x499 }, + { 0x49b, 0x49b }, + { 0x49d, 0x49d }, + { 0x49f, 0x49f }, + { 0x4a1, 0x4a1 }, + { 0x4a3, 0x4a3 }, + { 0x4a5, 0x4a5 }, + { 0x4a7, 0x4a7 }, + { 0x4a9, 0x4a9 }, + { 0x4ab, 0x4ab }, + { 0x4ad, 0x4ad }, + { 0x4af, 0x4af }, + { 0x4b1, 0x4b1 }, + { 0x4b3, 0x4b3 }, + { 0x4b5, 0x4b5 }, + { 0x4b7, 0x4b7 }, + { 0x4b9, 0x4b9 }, + { 0x4bb, 0x4bb }, + { 0x4bd, 0x4bd }, + { 0x4bf, 0x4bf }, + { 0x4c2, 0x4c2 }, + { 0x4c4, 0x4c4 }, + { 0x4c6, 0x4c6 }, + { 0x4c8, 0x4c8 }, + { 0x4ca, 0x4ca }, + { 0x4cc, 0x4cc }, + { 0x4ce, 0x4cf }, + { 0x4d1, 0x4d1 }, + { 0x4d3, 0x4d3 }, + { 0x4d5, 0x4d5 }, + { 0x4d7, 0x4d7 }, + { 0x4d9, 0x4d9 }, + { 0x4db, 0x4db }, + { 0x4dd, 0x4dd }, + { 0x4df, 0x4df }, + { 0x4e1, 0x4e1 }, + { 0x4e3, 0x4e3 }, + { 0x4e5, 0x4e5 }, + { 0x4e7, 0x4e7 }, + { 0x4e9, 0x4e9 }, + { 0x4eb, 0x4eb }, + { 0x4ed, 0x4ed }, + { 0x4ef, 0x4ef }, + { 0x4f1, 0x4f1 }, + { 0x4f3, 0x4f3 }, + { 0x4f5, 0x4f5 }, + { 0x4f7, 0x4f7 }, + { 0x4f9, 0x4f9 }, + { 0x4fb, 0x4fb }, + { 0x4fd, 0x4fd }, + { 0x4ff, 0x4ff }, + { 0x501, 0x501 }, + { 0x503, 0x503 }, + { 0x505, 0x505 }, + { 0x507, 0x507 }, + { 0x509, 0x509 }, + { 0x50b, 0x50b }, + { 0x50d, 0x50d }, + { 0x50f, 0x50f }, + { 0x511, 0x511 }, + { 0x513, 0x513 }, + { 0x515, 0x515 }, + { 0x517, 0x517 }, + { 0x519, 0x519 }, + { 0x51b, 0x51b }, + { 0x51d, 0x51d }, + { 0x51f, 0x51f }, + { 0x521, 0x521 }, + { 0x523, 0x523 }, + { 0x525, 0x525 }, + { 0x527, 0x527 }, + { 0x529, 0x529 }, + { 0x52b, 0x52b }, + { 0x52d, 0x52d }, + { 0x52f, 0x52f }, + { 0x561, 0x587 }, + { 0x13f8, 0x13fd }, + { 0x1d00, 0x1d2b }, + { 0x1d6b, 0x1d77 }, + { 0x1d79, 0x1d9a }, + { 0x1e01, 0x1e01 }, + { 0x1e03, 0x1e03 }, + { 0x1e05, 0x1e05 }, + { 0x1e07, 0x1e07 }, + { 0x1e09, 0x1e09 }, + { 0x1e0b, 0x1e0b }, + { 0x1e0d, 0x1e0d }, + { 0x1e0f, 0x1e0f }, + { 0x1e11, 0x1e11 }, + { 0x1e13, 0x1e13 }, + { 0x1e15, 0x1e15 }, + { 0x1e17, 0x1e17 }, + { 0x1e19, 0x1e19 }, + { 0x1e1b, 0x1e1b }, + { 0x1e1d, 0x1e1d }, + { 0x1e1f, 0x1e1f }, + { 0x1e21, 0x1e21 }, + { 0x1e23, 0x1e23 }, + { 0x1e25, 0x1e25 }, + { 0x1e27, 0x1e27 }, + { 0x1e29, 0x1e29 }, + { 0x1e2b, 0x1e2b }, + { 0x1e2d, 0x1e2d }, + { 0x1e2f, 0x1e2f }, + { 0x1e31, 0x1e31 }, + { 0x1e33, 0x1e33 }, + { 0x1e35, 0x1e35 }, + { 0x1e37, 0x1e37 }, + { 0x1e39, 0x1e39 }, + { 0x1e3b, 0x1e3b }, + { 0x1e3d, 0x1e3d }, + { 0x1e3f, 0x1e3f }, + { 0x1e41, 0x1e41 }, + { 0x1e43, 0x1e43 }, + { 0x1e45, 0x1e45 }, + { 0x1e47, 0x1e47 }, + { 0x1e49, 0x1e49 }, + { 0x1e4b, 0x1e4b }, + { 0x1e4d, 0x1e4d }, + { 0x1e4f, 0x1e4f }, + { 0x1e51, 0x1e51 }, + { 0x1e53, 0x1e53 }, + { 0x1e55, 0x1e55 }, + { 0x1e57, 0x1e57 }, + { 0x1e59, 0x1e59 }, + { 0x1e5b, 0x1e5b }, + { 0x1e5d, 0x1e5d }, + { 0x1e5f, 0x1e5f }, + { 0x1e61, 0x1e61 }, + { 0x1e63, 0x1e63 }, + { 0x1e65, 0x1e65 }, + { 0x1e67, 0x1e67 }, + { 0x1e69, 0x1e69 }, + { 0x1e6b, 0x1e6b }, + { 0x1e6d, 0x1e6d }, + { 0x1e6f, 0x1e6f }, + { 0x1e71, 0x1e71 }, + { 0x1e73, 0x1e73 }, + { 0x1e75, 0x1e75 }, + { 0x1e77, 0x1e77 }, + { 0x1e79, 0x1e79 }, + { 0x1e7b, 0x1e7b }, + { 0x1e7d, 0x1e7d }, + { 0x1e7f, 0x1e7f }, + { 0x1e81, 0x1e81 }, + { 0x1e83, 0x1e83 }, + { 0x1e85, 0x1e85 }, + { 0x1e87, 0x1e87 }, + { 0x1e89, 0x1e89 }, + { 0x1e8b, 0x1e8b }, + { 0x1e8d, 0x1e8d }, + { 0x1e8f, 0x1e8f }, + { 0x1e91, 0x1e91 }, + { 0x1e93, 0x1e93 }, + { 0x1e95, 0x1e9d }, + { 0x1e9f, 0x1e9f }, + { 0x1ea1, 0x1ea1 }, + { 0x1ea3, 0x1ea3 }, + { 0x1ea5, 0x1ea5 }, + { 0x1ea7, 0x1ea7 }, + { 0x1ea9, 0x1ea9 }, + { 0x1eab, 0x1eab }, + { 0x1ead, 0x1ead }, + { 0x1eaf, 0x1eaf }, + { 0x1eb1, 0x1eb1 }, + { 0x1eb3, 0x1eb3 }, + { 0x1eb5, 0x1eb5 }, + { 0x1eb7, 0x1eb7 }, + { 0x1eb9, 0x1eb9 }, + { 0x1ebb, 0x1ebb }, + { 0x1ebd, 0x1ebd }, + { 0x1ebf, 0x1ebf }, + { 0x1ec1, 0x1ec1 }, + { 0x1ec3, 0x1ec3 }, + { 0x1ec5, 0x1ec5 }, + { 0x1ec7, 0x1ec7 }, + { 0x1ec9, 0x1ec9 }, + { 0x1ecb, 0x1ecb }, + { 0x1ecd, 0x1ecd }, + { 0x1ecf, 0x1ecf }, + { 0x1ed1, 0x1ed1 }, + { 0x1ed3, 0x1ed3 }, + { 0x1ed5, 0x1ed5 }, + { 0x1ed7, 0x1ed7 }, + { 0x1ed9, 0x1ed9 }, + { 0x1edb, 0x1edb }, + { 0x1edd, 0x1edd }, + { 0x1edf, 0x1edf }, + { 0x1ee1, 0x1ee1 }, + { 0x1ee3, 0x1ee3 }, + { 0x1ee5, 0x1ee5 }, + { 0x1ee7, 0x1ee7 }, + { 0x1ee9, 0x1ee9 }, + { 0x1eeb, 0x1eeb }, + { 0x1eed, 0x1eed }, + { 0x1eef, 0x1eef }, + { 0x1ef1, 0x1ef1 }, + { 0x1ef3, 0x1ef3 }, + { 0x1ef5, 0x1ef5 }, + { 0x1ef7, 0x1ef7 }, + { 0x1ef9, 0x1ef9 }, + { 0x1efb, 0x1efb }, + { 0x1efd, 0x1efd }, + { 0x1eff, 0x1f07 }, + { 0x1f10, 0x1f15 }, + { 0x1f20, 0x1f27 }, + { 0x1f30, 0x1f37 }, + { 0x1f40, 0x1f45 }, + { 0x1f50, 0x1f57 }, + { 0x1f60, 0x1f67 }, + { 0x1f70, 0x1f7d }, + { 0x1f80, 0x1f87 }, + { 0x1f90, 0x1f97 }, + { 0x1fa0, 0x1fa7 }, + { 0x1fb0, 0x1fb4 }, + { 0x1fb6, 0x1fb7 }, + { 0x1fbe, 0x1fbe }, + { 0x1fc2, 0x1fc4 }, + { 0x1fc6, 0x1fc7 }, + { 0x1fd0, 0x1fd3 }, + { 0x1fd6, 0x1fd7 }, + { 0x1fe0, 0x1fe7 }, + { 0x1ff2, 0x1ff4 }, + { 0x1ff6, 0x1ff7 }, + { 0x210a, 0x210a }, + { 0x210e, 0x210f }, + { 0x2113, 0x2113 }, + { 0x212f, 0x212f }, + { 0x2134, 0x2134 }, + { 0x2139, 0x2139 }, + { 0x213c, 0x213d }, + { 0x2146, 0x2149 }, + { 0x214e, 0x214e }, + { 0x2184, 0x2184 }, + { 0x2c30, 0x2c5e }, + { 0x2c61, 0x2c61 }, + { 0x2c65, 0x2c66 }, + { 0x2c68, 0x2c68 }, + { 0x2c6a, 0x2c6a }, + { 0x2c6c, 0x2c6c }, + { 0x2c71, 0x2c71 }, + { 0x2c73, 0x2c74 }, + { 0x2c76, 0x2c7b }, + { 0x2c81, 0x2c81 }, + { 0x2c83, 0x2c83 }, + { 0x2c85, 0x2c85 }, + { 0x2c87, 0x2c87 }, + { 0x2c89, 0x2c89 }, + { 0x2c8b, 0x2c8b }, + { 0x2c8d, 0x2c8d }, + { 0x2c8f, 0x2c8f }, + { 0x2c91, 0x2c91 }, + { 0x2c93, 0x2c93 }, + { 0x2c95, 0x2c95 }, + { 0x2c97, 0x2c97 }, + { 0x2c99, 0x2c99 }, + { 0x2c9b, 0x2c9b }, + { 0x2c9d, 0x2c9d }, + { 0x2c9f, 0x2c9f }, + { 0x2ca1, 0x2ca1 }, + { 0x2ca3, 0x2ca3 }, + { 0x2ca5, 0x2ca5 }, + { 0x2ca7, 0x2ca7 }, + { 0x2ca9, 0x2ca9 }, + { 0x2cab, 0x2cab }, + { 0x2cad, 0x2cad }, + { 0x2caf, 0x2caf }, + { 0x2cb1, 0x2cb1 }, + { 0x2cb3, 0x2cb3 }, + { 0x2cb5, 0x2cb5 }, + { 0x2cb7, 0x2cb7 }, + { 0x2cb9, 0x2cb9 }, + { 0x2cbb, 0x2cbb }, + { 0x2cbd, 0x2cbd }, + { 0x2cbf, 0x2cbf }, + { 0x2cc1, 0x2cc1 }, + { 0x2cc3, 0x2cc3 }, + { 0x2cc5, 0x2cc5 }, + { 0x2cc7, 0x2cc7 }, + { 0x2cc9, 0x2cc9 }, + { 0x2ccb, 0x2ccb }, + { 0x2ccd, 0x2ccd }, + { 0x2ccf, 0x2ccf }, + { 0x2cd1, 0x2cd1 }, + { 0x2cd3, 0x2cd3 }, + { 0x2cd5, 0x2cd5 }, + { 0x2cd7, 0x2cd7 }, + { 0x2cd9, 0x2cd9 }, + { 0x2cdb, 0x2cdb }, + { 0x2cdd, 0x2cdd }, + { 0x2cdf, 0x2cdf }, + { 0x2ce1, 0x2ce1 }, + { 0x2ce3, 0x2ce4 }, + { 0x2cec, 0x2cec }, + { 0x2cee, 0x2cee }, + { 0x2cf3, 0x2cf3 }, + { 0x2d00, 0x2d25 }, + { 0x2d27, 0x2d27 }, + { 0x2d2d, 0x2d2d }, + { 0xa641, 0xa641 }, + { 0xa643, 0xa643 }, + { 0xa645, 0xa645 }, + { 0xa647, 0xa647 }, + { 0xa649, 0xa649 }, + { 0xa64b, 0xa64b }, + { 0xa64d, 0xa64d }, + { 0xa64f, 0xa64f }, + { 0xa651, 0xa651 }, + { 0xa653, 0xa653 }, + { 0xa655, 0xa655 }, + { 0xa657, 0xa657 }, + { 0xa659, 0xa659 }, + { 0xa65b, 0xa65b }, + { 0xa65d, 0xa65d }, + { 0xa65f, 0xa65f }, + { 0xa661, 0xa661 }, + { 0xa663, 0xa663 }, + { 0xa665, 0xa665 }, + { 0xa667, 0xa667 }, + { 0xa669, 0xa669 }, + { 0xa66b, 0xa66b }, + { 0xa66d, 0xa66d }, + { 0xa681, 0xa681 }, + { 0xa683, 0xa683 }, + { 0xa685, 0xa685 }, + { 0xa687, 0xa687 }, + { 0xa689, 0xa689 }, + { 0xa68b, 0xa68b }, + { 0xa68d, 0xa68d }, + { 0xa68f, 0xa68f }, + { 0xa691, 0xa691 }, + { 0xa693, 0xa693 }, + { 0xa695, 0xa695 }, + { 0xa697, 0xa697 }, + { 0xa699, 0xa699 }, + { 0xa69b, 0xa69b }, + { 0xa723, 0xa723 }, + { 0xa725, 0xa725 }, + { 0xa727, 0xa727 }, + { 0xa729, 0xa729 }, + { 0xa72b, 0xa72b }, + { 0xa72d, 0xa72d }, + { 0xa72f, 0xa731 }, + { 0xa733, 0xa733 }, + { 0xa735, 0xa735 }, + { 0xa737, 0xa737 }, + { 0xa739, 0xa739 }, + { 0xa73b, 0xa73b }, + { 0xa73d, 0xa73d }, + { 0xa73f, 0xa73f }, + { 0xa741, 0xa741 }, + { 0xa743, 0xa743 }, + { 0xa745, 0xa745 }, + { 0xa747, 0xa747 }, + { 0xa749, 0xa749 }, + { 0xa74b, 0xa74b }, + { 0xa74d, 0xa74d }, + { 0xa74f, 0xa74f }, + { 0xa751, 0xa751 }, + { 0xa753, 0xa753 }, + { 0xa755, 0xa755 }, + { 0xa757, 0xa757 }, + { 0xa759, 0xa759 }, + { 0xa75b, 0xa75b }, + { 0xa75d, 0xa75d }, + { 0xa75f, 0xa75f }, + { 0xa761, 0xa761 }, + { 0xa763, 0xa763 }, + { 0xa765, 0xa765 }, + { 0xa767, 0xa767 }, + { 0xa769, 0xa769 }, + { 0xa76b, 0xa76b }, + { 0xa76d, 0xa76d }, + { 0xa76f, 0xa76f }, + { 0xa771, 0xa778 }, + { 0xa77a, 0xa77a }, + { 0xa77c, 0xa77c }, + { 0xa77f, 0xa77f }, + { 0xa781, 0xa781 }, + { 0xa783, 0xa783 }, + { 0xa785, 0xa785 }, + { 0xa787, 0xa787 }, + { 0xa78c, 0xa78c }, + { 0xa78e, 0xa78e }, + { 0xa791, 0xa791 }, + { 0xa793, 0xa795 }, + { 0xa797, 0xa797 }, + { 0xa799, 0xa799 }, + { 0xa79b, 0xa79b }, + { 0xa79d, 0xa79d }, + { 0xa79f, 0xa79f }, + { 0xa7a1, 0xa7a1 }, + { 0xa7a3, 0xa7a3 }, + { 0xa7a5, 0xa7a5 }, + { 0xa7a7, 0xa7a7 }, + { 0xa7a9, 0xa7a9 }, + { 0xa7b5, 0xa7b5 }, + { 0xa7b7, 0xa7b7 }, + { 0xa7fa, 0xa7fa }, + { 0xab30, 0xab5a }, + { 0xab60, 0xab65 }, + { 0xab70, 0xabbf }, + { 0xfb00, 0xfb06 }, + { 0xfb13, 0xfb17 }, + { 0xff41, 0xff5a }, + { 0x10428, 0x1044f }, + { 0x10cc0, 0x10cf2 }, + { 0x118c0, 0x118df }, + { 0x1d41a, 0x1d433 }, + { 0x1d44e, 0x1d454 }, + { 0x1d456, 0x1d467 }, + { 0x1d482, 0x1d49b }, + { 0x1d4b6, 0x1d4b9 }, + { 0x1d4bb, 0x1d4bb }, + { 0x1d4bd, 0x1d4c3 }, + { 0x1d4c5, 0x1d4cf }, + { 0x1d4ea, 0x1d503 }, + { 0x1d51e, 0x1d537 }, + { 0x1d552, 0x1d56b }, + { 0x1d586, 0x1d59f }, + { 0x1d5ba, 0x1d5d3 }, + { 0x1d5ee, 0x1d607 }, + { 0x1d622, 0x1d63b }, + { 0x1d656, 0x1d66f }, + { 0x1d68a, 0x1d6a5 }, + { 0x1d6c2, 0x1d6da }, + { 0x1d6dc, 0x1d6e1 }, + { 0x1d6fc, 0x1d714 }, + { 0x1d716, 0x1d71b }, + { 0x1d736, 0x1d74e }, + { 0x1d750, 0x1d755 }, + { 0x1d770, 0x1d788 }, + { 0x1d78a, 0x1d78f }, + { 0x1d7aa, 0x1d7c2 }, + { 0x1d7c4, 0x1d7c9 }, + { 0x1d7cb, 0x1d7cb } +}; + +extern int runeinrange(const void* a, const void* b); +bool islowerrune(Rune ch) { + return (NULL != bsearch(&ch, lowers, 630, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/marks.c b/source/utf/marks.c new file mode 100644 index 0000000..df397e7 --- /dev/null +++ b/source/utf/marks.c @@ -0,0 +1,245 @@ +#include + +static Rune marks[236][2] = { + { 0x300, 0x36f }, + { 0x483, 0x489 }, + { 0x591, 0x5bd }, + { 0x5bf, 0x5bf }, + { 0x5c1, 0x5c2 }, + { 0x5c4, 0x5c5 }, + { 0x5c7, 0x5c7 }, + { 0x610, 0x61a }, + { 0x64b, 0x65f }, + { 0x670, 0x670 }, + { 0x6d6, 0x6dc }, + { 0x6df, 0x6e4 }, + { 0x6e7, 0x6e8 }, + { 0x6ea, 0x6ed }, + { 0x711, 0x711 }, + { 0x730, 0x74a }, + { 0x7a6, 0x7b0 }, + { 0x7eb, 0x7f3 }, + { 0x816, 0x819 }, + { 0x81b, 0x823 }, + { 0x825, 0x827 }, + { 0x829, 0x82d }, + { 0x859, 0x85b }, + { 0x8e3, 0x903 }, + { 0x93a, 0x93c }, + { 0x93e, 0x94f }, + { 0x951, 0x957 }, + { 0x962, 0x963 }, + { 0x981, 0x983 }, + { 0x9bc, 0x9bc }, + { 0x9be, 0x9c4 }, + { 0x9c7, 0x9c8 }, + { 0x9cb, 0x9cd }, + { 0x9d7, 0x9d7 }, + { 0x9e2, 0x9e3 }, + { 0xa01, 0xa03 }, + { 0xa3c, 0xa3c }, + { 0xa3e, 0xa42 }, + { 0xa47, 0xa48 }, + { 0xa4b, 0xa4d }, + { 0xa51, 0xa51 }, + { 0xa70, 0xa71 }, + { 0xa75, 0xa75 }, + { 0xa81, 0xa83 }, + { 0xabc, 0xabc }, + { 0xabe, 0xac5 }, + { 0xac7, 0xac9 }, + { 0xacb, 0xacd }, + { 0xae2, 0xae3 }, + { 0xb01, 0xb03 }, + { 0xb3c, 0xb3c }, + { 0xb3e, 0xb44 }, + { 0xb47, 0xb48 }, + { 0xb4b, 0xb4d }, + { 0xb56, 0xb57 }, + { 0xb62, 0xb63 }, + { 0xb82, 0xb82 }, + { 0xbbe, 0xbc2 }, + { 0xbc6, 0xbc8 }, + { 0xbca, 0xbcd }, + { 0xbd7, 0xbd7 }, + { 0xc00, 0xc03 }, + { 0xc3e, 0xc44 }, + { 0xc46, 0xc48 }, + { 0xc4a, 0xc4d }, + { 0xc55, 0xc56 }, + { 0xc62, 0xc63 }, + { 0xc81, 0xc83 }, + { 0xcbc, 0xcbc }, + { 0xcbe, 0xcc4 }, + { 0xcc6, 0xcc8 }, + { 0xcca, 0xccd }, + { 0xcd5, 0xcd6 }, + { 0xce2, 0xce3 }, + { 0xd01, 0xd03 }, + { 0xd3e, 0xd44 }, + { 0xd46, 0xd48 }, + { 0xd4a, 0xd4d }, + { 0xd57, 0xd57 }, + { 0xd62, 0xd63 }, + { 0xd82, 0xd83 }, + { 0xdca, 0xdca }, + { 0xdcf, 0xdd4 }, + { 0xdd6, 0xdd6 }, + { 0xdd8, 0xddf }, + { 0xdf2, 0xdf3 }, + { 0xe31, 0xe31 }, + { 0xe34, 0xe3a }, + { 0xe47, 0xe4e }, + { 0xeb1, 0xeb1 }, + { 0xeb4, 0xeb9 }, + { 0xebb, 0xebc }, + { 0xec8, 0xecd }, + { 0xf18, 0xf19 }, + { 0xf35, 0xf35 }, + { 0xf37, 0xf37 }, + { 0xf39, 0xf39 }, + { 0xf3e, 0xf3f }, + { 0xf71, 0xf84 }, + { 0xf86, 0xf87 }, + { 0xf8d, 0xf97 }, + { 0xf99, 0xfbc }, + { 0xfc6, 0xfc6 }, + { 0x102b, 0x103e }, + { 0x1056, 0x1059 }, + { 0x105e, 0x1060 }, + { 0x1062, 0x1064 }, + { 0x1067, 0x106d }, + { 0x1071, 0x1074 }, + { 0x1082, 0x108d }, + { 0x108f, 0x108f }, + { 0x109a, 0x109d }, + { 0x135d, 0x135f }, + { 0x1712, 0x1714 }, + { 0x1732, 0x1734 }, + { 0x1752, 0x1753 }, + { 0x1772, 0x1773 }, + { 0x17b4, 0x17d3 }, + { 0x17dd, 0x17dd }, + { 0x180b, 0x180d }, + { 0x18a9, 0x18a9 }, + { 0x1920, 0x192b }, + { 0x1930, 0x193b }, + { 0x1a17, 0x1a1b }, + { 0x1a55, 0x1a5e }, + { 0x1a60, 0x1a7c }, + { 0x1a7f, 0x1a7f }, + { 0x1ab0, 0x1abe }, + { 0x1b00, 0x1b04 }, + { 0x1b34, 0x1b44 }, + { 0x1b6b, 0x1b73 }, + { 0x1b80, 0x1b82 }, + { 0x1ba1, 0x1bad }, + { 0x1be6, 0x1bf3 }, + { 0x1c24, 0x1c37 }, + { 0x1cd0, 0x1cd2 }, + { 0x1cd4, 0x1ce8 }, + { 0x1ced, 0x1ced }, + { 0x1cf2, 0x1cf4 }, + { 0x1cf8, 0x1cf9 }, + { 0x1dc0, 0x1df5 }, + { 0x1dfc, 0x1dff }, + { 0x20d0, 0x20f0 }, + { 0x2cef, 0x2cf1 }, + { 0x2d7f, 0x2d7f }, + { 0x2de0, 0x2dff }, + { 0x302a, 0x302f }, + { 0x3099, 0x309a }, + { 0xa66f, 0xa672 }, + { 0xa674, 0xa67d }, + { 0xa69e, 0xa69f }, + { 0xa6f0, 0xa6f1 }, + { 0xa802, 0xa802 }, + { 0xa806, 0xa806 }, + { 0xa80b, 0xa80b }, + { 0xa823, 0xa827 }, + { 0xa880, 0xa881 }, + { 0xa8b4, 0xa8c4 }, + { 0xa8e0, 0xa8f1 }, + { 0xa926, 0xa92d }, + { 0xa947, 0xa953 }, + { 0xa980, 0xa983 }, + { 0xa9b3, 0xa9c0 }, + { 0xa9e5, 0xa9e5 }, + { 0xaa29, 0xaa36 }, + { 0xaa43, 0xaa43 }, + { 0xaa4c, 0xaa4d }, + { 0xaa7b, 0xaa7d }, + { 0xaab0, 0xaab0 }, + { 0xaab2, 0xaab4 }, + { 0xaab7, 0xaab8 }, + { 0xaabe, 0xaabf }, + { 0xaac1, 0xaac1 }, + { 0xaaeb, 0xaaef }, + { 0xaaf5, 0xaaf6 }, + { 0xabe3, 0xabea }, + { 0xabec, 0xabed }, + { 0xfb1e, 0xfb1e }, + { 0xfe00, 0xfe0f }, + { 0xfe20, 0xfe2f }, + { 0x101fd, 0x101fd }, + { 0x102e0, 0x102e0 }, + { 0x10376, 0x1037a }, + { 0x10a01, 0x10a03 }, + { 0x10a05, 0x10a06 }, + { 0x10a0c, 0x10a0f }, + { 0x10a38, 0x10a3a }, + { 0x10a3f, 0x10a3f }, + { 0x10ae5, 0x10ae6 }, + { 0x11000, 0x11002 }, + { 0x11038, 0x11046 }, + { 0x1107f, 0x11082 }, + { 0x110b0, 0x110ba }, + { 0x11100, 0x11102 }, + { 0x11127, 0x11134 }, + { 0x11173, 0x11173 }, + { 0x11180, 0x11182 }, + { 0x111b3, 0x111c0 }, + { 0x111ca, 0x111cc }, + { 0x1122c, 0x11237 }, + { 0x112df, 0x112ea }, + { 0x11300, 0x11303 }, + { 0x1133c, 0x1133c }, + { 0x1133e, 0x11344 }, + { 0x11347, 0x11348 }, + { 0x1134b, 0x1134d }, + { 0x11357, 0x11357 }, + { 0x11362, 0x11363 }, + { 0x11366, 0x1136c }, + { 0x11370, 0x11374 }, + { 0x114b0, 0x114c3 }, + { 0x115af, 0x115b5 }, + { 0x115b8, 0x115c0 }, + { 0x115dc, 0x115dd }, + { 0x11630, 0x11640 }, + { 0x116ab, 0x116b7 }, + { 0x1171d, 0x1172b }, + { 0x16af0, 0x16af4 }, + { 0x16b30, 0x16b36 }, + { 0x16f51, 0x16f7e }, + { 0x16f8f, 0x16f92 }, + { 0x1bc9d, 0x1bc9e }, + { 0x1d165, 0x1d169 }, + { 0x1d16d, 0x1d172 }, + { 0x1d17b, 0x1d182 }, + { 0x1d185, 0x1d18b }, + { 0x1d1aa, 0x1d1ad }, + { 0x1d242, 0x1d244 }, + { 0x1da00, 0x1da36 }, + { 0x1da3b, 0x1da6c }, + { 0x1da75, 0x1da75 }, + { 0x1da84, 0x1da84 }, + { 0x1da9b, 0x1da9f }, + { 0x1daa1, 0x1daaf }, + { 0x1e8d0, 0x1e8d6 }, + { 0xe0100, 0xe01ef } +}; + +extern int runeinrange(const void* a, const void* b); +bool ismarkrune(Rune ch) { + return (NULL != bsearch(&ch, marks, 236, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/numbers.c b/source/utf/numbers.c new file mode 100644 index 0000000..e637859 --- /dev/null +++ b/source/utf/numbers.c @@ -0,0 +1,120 @@ +#include + +static Rune numbers[111][2] = { + { 0x30, 0x39 }, + { 0xb2, 0xb3 }, + { 0xb9, 0xb9 }, + { 0xbc, 0xbe }, + { 0x660, 0x669 }, + { 0x6f0, 0x6f9 }, + { 0x7c0, 0x7c9 }, + { 0x966, 0x96f }, + { 0x9e6, 0x9ef }, + { 0x9f4, 0x9f9 }, + { 0xa66, 0xa6f }, + { 0xae6, 0xaef }, + { 0xb66, 0xb6f }, + { 0xb72, 0xb77 }, + { 0xbe6, 0xbf2 }, + { 0xc66, 0xc6f }, + { 0xc78, 0xc7e }, + { 0xce6, 0xcef }, + { 0xd66, 0xd75 }, + { 0xde6, 0xdef }, + { 0xe50, 0xe59 }, + { 0xed0, 0xed9 }, + { 0xf20, 0xf33 }, + { 0x1040, 0x1049 }, + { 0x1090, 0x1099 }, + { 0x1369, 0x137c }, + { 0x16ee, 0x16f0 }, + { 0x17e0, 0x17e9 }, + { 0x17f0, 0x17f9 }, + { 0x1810, 0x1819 }, + { 0x1946, 0x194f }, + { 0x19d0, 0x19da }, + { 0x1a80, 0x1a89 }, + { 0x1a90, 0x1a99 }, + { 0x1b50, 0x1b59 }, + { 0x1bb0, 0x1bb9 }, + { 0x1c40, 0x1c49 }, + { 0x1c50, 0x1c59 }, + { 0x2070, 0x2070 }, + { 0x2074, 0x2079 }, + { 0x2080, 0x2089 }, + { 0x2150, 0x2182 }, + { 0x2185, 0x2189 }, + { 0x2460, 0x249b }, + { 0x24ea, 0x24ff }, + { 0x2776, 0x2793 }, + { 0x2cfd, 0x2cfd }, + { 0x3007, 0x3007 }, + { 0x3021, 0x3029 }, + { 0x3038, 0x303a }, + { 0x3192, 0x3195 }, + { 0x3220, 0x3229 }, + { 0x3248, 0x324f }, + { 0x3251, 0x325f }, + { 0x3280, 0x3289 }, + { 0x32b1, 0x32bf }, + { 0xa620, 0xa629 }, + { 0xa6e6, 0xa6ef }, + { 0xa830, 0xa835 }, + { 0xa8d0, 0xa8d9 }, + { 0xa900, 0xa909 }, + { 0xa9d0, 0xa9d9 }, + { 0xa9f0, 0xa9f9 }, + { 0xaa50, 0xaa59 }, + { 0xabf0, 0xabf9 }, + { 0xff10, 0xff19 }, + { 0x10107, 0x10133 }, + { 0x10140, 0x10178 }, + { 0x1018a, 0x1018b }, + { 0x102e1, 0x102fb }, + { 0x10320, 0x10323 }, + { 0x10341, 0x10341 }, + { 0x1034a, 0x1034a }, + { 0x103d1, 0x103d5 }, + { 0x104a0, 0x104a9 }, + { 0x10858, 0x1085f }, + { 0x10879, 0x1087f }, + { 0x108a7, 0x108af }, + { 0x108fb, 0x108ff }, + { 0x10916, 0x1091b }, + { 0x109bc, 0x109bd }, + { 0x109c0, 0x109cf }, + { 0x109d2, 0x109ff }, + { 0x10a40, 0x10a47 }, + { 0x10a7d, 0x10a7e }, + { 0x10a9d, 0x10a9f }, + { 0x10aeb, 0x10aef }, + { 0x10b58, 0x10b5f }, + { 0x10b78, 0x10b7f }, + { 0x10ba9, 0x10baf }, + { 0x10cfa, 0x10cff }, + { 0x10e60, 0x10e7e }, + { 0x11052, 0x1106f }, + { 0x110f0, 0x110f9 }, + { 0x11136, 0x1113f }, + { 0x111d0, 0x111d9 }, + { 0x111e1, 0x111f4 }, + { 0x112f0, 0x112f9 }, + { 0x114d0, 0x114d9 }, + { 0x11650, 0x11659 }, + { 0x116c0, 0x116c9 }, + { 0x11730, 0x1173b }, + { 0x118e0, 0x118f2 }, + { 0x12400, 0x1246e }, + { 0x16a60, 0x16a69 }, + { 0x16b50, 0x16b59 }, + { 0x16b5b, 0x16b61 }, + { 0x1d360, 0x1d371 }, + { 0x1d7ce, 0x1d7ff }, + { 0x1e8c7, 0x1e8cf }, + { 0x1f100, 0x1f10c } +}; + +extern int runeinrange(const void* a, const void* b); +bool isnumberrune(Rune ch) { + return (NULL != bsearch(&ch, numbers, 111, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/otherletters.c b/source/utf/otherletters.c new file mode 100644 index 0000000..00bdee0 --- /dev/null +++ b/source/utf/otherletters.c @@ -0,0 +1,479 @@ +#include + +static Rune otherletters[470][2] = { + { 0xaa, 0xaa }, + { 0xba, 0xba }, + { 0x1bb, 0x1bb }, + { 0x1c0, 0x1c3 }, + { 0x294, 0x294 }, + { 0x2b0, 0x2c1 }, + { 0x2c6, 0x2d1 }, + { 0x2e0, 0x2e4 }, + { 0x2ec, 0x2ec }, + { 0x2ee, 0x2ee }, + { 0x374, 0x374 }, + { 0x37a, 0x37a }, + { 0x559, 0x559 }, + { 0x5d0, 0x5ea }, + { 0x5f0, 0x5f2 }, + { 0x620, 0x64a }, + { 0x66e, 0x66f }, + { 0x671, 0x6d3 }, + { 0x6d5, 0x6d5 }, + { 0x6e5, 0x6e6 }, + { 0x6ee, 0x6ef }, + { 0x6fa, 0x6fc }, + { 0x6ff, 0x6ff }, + { 0x710, 0x710 }, + { 0x712, 0x72f }, + { 0x74d, 0x7a5 }, + { 0x7b1, 0x7b1 }, + { 0x7ca, 0x7ea }, + { 0x7f4, 0x7f5 }, + { 0x7fa, 0x7fa }, + { 0x800, 0x815 }, + { 0x81a, 0x81a }, + { 0x824, 0x824 }, + { 0x828, 0x828 }, + { 0x840, 0x858 }, + { 0x8a0, 0x8b4 }, + { 0x904, 0x939 }, + { 0x93d, 0x93d }, + { 0x950, 0x950 }, + { 0x958, 0x961 }, + { 0x971, 0x980 }, + { 0x985, 0x98c }, + { 0x98f, 0x990 }, + { 0x993, 0x9a8 }, + { 0x9aa, 0x9b0 }, + { 0x9b2, 0x9b2 }, + { 0x9b6, 0x9b9 }, + { 0x9bd, 0x9bd }, + { 0x9ce, 0x9ce }, + { 0x9dc, 0x9dd }, + { 0x9df, 0x9e1 }, + { 0x9f0, 0x9f1 }, + { 0xa05, 0xa0a }, + { 0xa0f, 0xa10 }, + { 0xa13, 0xa28 }, + { 0xa2a, 0xa30 }, + { 0xa32, 0xa33 }, + { 0xa35, 0xa36 }, + { 0xa38, 0xa39 }, + { 0xa59, 0xa5c }, + { 0xa5e, 0xa5e }, + { 0xa72, 0xa74 }, + { 0xa85, 0xa8d }, + { 0xa8f, 0xa91 }, + { 0xa93, 0xaa8 }, + { 0xaaa, 0xab0 }, + { 0xab2, 0xab3 }, + { 0xab5, 0xab9 }, + { 0xabd, 0xabd }, + { 0xad0, 0xad0 }, + { 0xae0, 0xae1 }, + { 0xaf9, 0xaf9 }, + { 0xb05, 0xb0c }, + { 0xb0f, 0xb10 }, + { 0xb13, 0xb28 }, + { 0xb2a, 0xb30 }, + { 0xb32, 0xb33 }, + { 0xb35, 0xb39 }, + { 0xb3d, 0xb3d }, + { 0xb5c, 0xb5d }, + { 0xb5f, 0xb61 }, + { 0xb71, 0xb71 }, + { 0xb83, 0xb83 }, + { 0xb85, 0xb8a }, + { 0xb8e, 0xb90 }, + { 0xb92, 0xb95 }, + { 0xb99, 0xb9a }, + { 0xb9c, 0xb9c }, + { 0xb9e, 0xb9f }, + { 0xba3, 0xba4 }, + { 0xba8, 0xbaa }, + { 0xbae, 0xbb9 }, + { 0xbd0, 0xbd0 }, + { 0xc05, 0xc0c }, + { 0xc0e, 0xc10 }, + { 0xc12, 0xc28 }, + { 0xc2a, 0xc39 }, + { 0xc3d, 0xc3d }, + { 0xc58, 0xc5a }, + { 0xc60, 0xc61 }, + { 0xc85, 0xc8c }, + { 0xc8e, 0xc90 }, + { 0xc92, 0xca8 }, + { 0xcaa, 0xcb3 }, + { 0xcb5, 0xcb9 }, + { 0xcbd, 0xcbd }, + { 0xcde, 0xcde }, + { 0xce0, 0xce1 }, + { 0xcf1, 0xcf2 }, + { 0xd05, 0xd0c }, + { 0xd0e, 0xd10 }, + { 0xd12, 0xd3a }, + { 0xd3d, 0xd3d }, + { 0xd4e, 0xd4e }, + { 0xd5f, 0xd61 }, + { 0xd7a, 0xd7f }, + { 0xd85, 0xd96 }, + { 0xd9a, 0xdb1 }, + { 0xdb3, 0xdbb }, + { 0xdbd, 0xdbd }, + { 0xdc0, 0xdc6 }, + { 0xe01, 0xe30 }, + { 0xe32, 0xe33 }, + { 0xe40, 0xe46 }, + { 0xe81, 0xe82 }, + { 0xe84, 0xe84 }, + { 0xe87, 0xe88 }, + { 0xe8a, 0xe8a }, + { 0xe8d, 0xe8d }, + { 0xe94, 0xe97 }, + { 0xe99, 0xe9f }, + { 0xea1, 0xea3 }, + { 0xea5, 0xea5 }, + { 0xea7, 0xea7 }, + { 0xeaa, 0xeab }, + { 0xead, 0xeb0 }, + { 0xeb2, 0xeb3 }, + { 0xebd, 0xebd }, + { 0xec0, 0xec4 }, + { 0xec6, 0xec6 }, + { 0xedc, 0xedf }, + { 0xf00, 0xf00 }, + { 0xf40, 0xf47 }, + { 0xf49, 0xf6c }, + { 0xf88, 0xf8c }, + { 0x1000, 0x102a }, + { 0x103f, 0x103f }, + { 0x1050, 0x1055 }, + { 0x105a, 0x105d }, + { 0x1061, 0x1061 }, + { 0x1065, 0x1066 }, + { 0x106e, 0x1070 }, + { 0x1075, 0x1081 }, + { 0x108e, 0x108e }, + { 0x10d0, 0x10fa }, + { 0x10fc, 0x1248 }, + { 0x124a, 0x124d }, + { 0x1250, 0x1256 }, + { 0x1258, 0x1258 }, + { 0x125a, 0x125d }, + { 0x1260, 0x1288 }, + { 0x128a, 0x128d }, + { 0x1290, 0x12b0 }, + { 0x12b2, 0x12b5 }, + { 0x12b8, 0x12be }, + { 0x12c0, 0x12c0 }, + { 0x12c2, 0x12c5 }, + { 0x12c8, 0x12d6 }, + { 0x12d8, 0x1310 }, + { 0x1312, 0x1315 }, + { 0x1318, 0x135a }, + { 0x1380, 0x138f }, + { 0x1401, 0x166c }, + { 0x166f, 0x167f }, + { 0x1681, 0x169a }, + { 0x16a0, 0x16ea }, + { 0x16f1, 0x16f8 }, + { 0x1700, 0x170c }, + { 0x170e, 0x1711 }, + { 0x1720, 0x1731 }, + { 0x1740, 0x1751 }, + { 0x1760, 0x176c }, + { 0x176e, 0x1770 }, + { 0x1780, 0x17b3 }, + { 0x17d7, 0x17d7 }, + { 0x17dc, 0x17dc }, + { 0x1820, 0x1877 }, + { 0x1880, 0x18a8 }, + { 0x18aa, 0x18aa }, + { 0x18b0, 0x18f5 }, + { 0x1900, 0x191e }, + { 0x1950, 0x196d }, + { 0x1970, 0x1974 }, + { 0x1980, 0x19ab }, + { 0x19b0, 0x19c9 }, + { 0x1a00, 0x1a16 }, + { 0x1a20, 0x1a54 }, + { 0x1aa7, 0x1aa7 }, + { 0x1b05, 0x1b33 }, + { 0x1b45, 0x1b4b }, + { 0x1b83, 0x1ba0 }, + { 0x1bae, 0x1baf }, + { 0x1bba, 0x1be5 }, + { 0x1c00, 0x1c23 }, + { 0x1c4d, 0x1c4f }, + { 0x1c5a, 0x1c7d }, + { 0x1ce9, 0x1cec }, + { 0x1cee, 0x1cf1 }, + { 0x1cf5, 0x1cf6 }, + { 0x1d2c, 0x1d6a }, + { 0x1d78, 0x1d78 }, + { 0x1d9b, 0x1dbf }, + { 0x2071, 0x2071 }, + { 0x207f, 0x207f }, + { 0x2090, 0x209c }, + { 0x2135, 0x2138 }, + { 0x2c7c, 0x2c7d }, + { 0x2d30, 0x2d67 }, + { 0x2d6f, 0x2d6f }, + { 0x2d80, 0x2d96 }, + { 0x2da0, 0x2da6 }, + { 0x2da8, 0x2dae }, + { 0x2db0, 0x2db6 }, + { 0x2db8, 0x2dbe }, + { 0x2dc0, 0x2dc6 }, + { 0x2dc8, 0x2dce }, + { 0x2dd0, 0x2dd6 }, + { 0x2dd8, 0x2dde }, + { 0x2e2f, 0x2e2f }, + { 0x3005, 0x3006 }, + { 0x3031, 0x3035 }, + { 0x303b, 0x303c }, + { 0x3041, 0x3096 }, + { 0x309d, 0x309f }, + { 0x30a1, 0x30fa }, + { 0x30fc, 0x30ff }, + { 0x3105, 0x312d }, + { 0x3131, 0x318e }, + { 0x31a0, 0x31ba }, + { 0x31f0, 0x31ff }, + { 0x3400, 0x3400 }, + { 0x4db5, 0x4db5 }, + { 0x4e00, 0x4e00 }, + { 0x9fd5, 0x9fd5 }, + { 0xa000, 0xa48c }, + { 0xa4d0, 0xa4fd }, + { 0xa500, 0xa60c }, + { 0xa610, 0xa61f }, + { 0xa62a, 0xa62b }, + { 0xa66e, 0xa66e }, + { 0xa67f, 0xa67f }, + { 0xa69c, 0xa69d }, + { 0xa6a0, 0xa6e5 }, + { 0xa717, 0xa71f }, + { 0xa770, 0xa770 }, + { 0xa788, 0xa788 }, + { 0xa78f, 0xa78f }, + { 0xa7f7, 0xa7f9 }, + { 0xa7fb, 0xa801 }, + { 0xa803, 0xa805 }, + { 0xa807, 0xa80a }, + { 0xa80c, 0xa822 }, + { 0xa840, 0xa873 }, + { 0xa882, 0xa8b3 }, + { 0xa8f2, 0xa8f7 }, + { 0xa8fb, 0xa8fb }, + { 0xa8fd, 0xa8fd }, + { 0xa90a, 0xa925 }, + { 0xa930, 0xa946 }, + { 0xa960, 0xa97c }, + { 0xa984, 0xa9b2 }, + { 0xa9cf, 0xa9cf }, + { 0xa9e0, 0xa9e4 }, + { 0xa9e6, 0xa9ef }, + { 0xa9fa, 0xa9fe }, + { 0xaa00, 0xaa28 }, + { 0xaa40, 0xaa42 }, + { 0xaa44, 0xaa4b }, + { 0xaa60, 0xaa76 }, + { 0xaa7a, 0xaa7a }, + { 0xaa7e, 0xaaaf }, + { 0xaab1, 0xaab1 }, + { 0xaab5, 0xaab6 }, + { 0xaab9, 0xaabd }, + { 0xaac0, 0xaac0 }, + { 0xaac2, 0xaac2 }, + { 0xaadb, 0xaadd }, + { 0xaae0, 0xaaea }, + { 0xaaf2, 0xaaf4 }, + { 0xab01, 0xab06 }, + { 0xab09, 0xab0e }, + { 0xab11, 0xab16 }, + { 0xab20, 0xab26 }, + { 0xab28, 0xab2e }, + { 0xab5c, 0xab5f }, + { 0xabc0, 0xabe2 }, + { 0xac00, 0xac00 }, + { 0xd7a3, 0xd7a3 }, + { 0xd7b0, 0xd7c6 }, + { 0xd7cb, 0xd7fb }, + { 0xf900, 0xfa6d }, + { 0xfa70, 0xfad9 }, + { 0xfb1d, 0xfb1d }, + { 0xfb1f, 0xfb28 }, + { 0xfb2a, 0xfb36 }, + { 0xfb38, 0xfb3c }, + { 0xfb3e, 0xfb3e }, + { 0xfb40, 0xfb41 }, + { 0xfb43, 0xfb44 }, + { 0xfb46, 0xfbb1 }, + { 0xfbd3, 0xfd3d }, + { 0xfd50, 0xfd8f }, + { 0xfd92, 0xfdc7 }, + { 0xfdf0, 0xfdfb }, + { 0xfe70, 0xfe74 }, + { 0xfe76, 0xfefc }, + { 0xff66, 0xffbe }, + { 0xffc2, 0xffc7 }, + { 0xffca, 0xffcf }, + { 0xffd2, 0xffd7 }, + { 0xffda, 0xffdc }, + { 0x10000, 0x1000b }, + { 0x1000d, 0x10026 }, + { 0x10028, 0x1003a }, + { 0x1003c, 0x1003d }, + { 0x1003f, 0x1004d }, + { 0x10050, 0x1005d }, + { 0x10080, 0x100fa }, + { 0x10280, 0x1029c }, + { 0x102a0, 0x102d0 }, + { 0x10300, 0x1031f }, + { 0x10330, 0x10340 }, + { 0x10342, 0x10349 }, + { 0x10350, 0x10375 }, + { 0x10380, 0x1039d }, + { 0x103a0, 0x103c3 }, + { 0x103c8, 0x103cf }, + { 0x10450, 0x1049d }, + { 0x10500, 0x10527 }, + { 0x10530, 0x10563 }, + { 0x10600, 0x10736 }, + { 0x10740, 0x10755 }, + { 0x10760, 0x10767 }, + { 0x10800, 0x10805 }, + { 0x10808, 0x10808 }, + { 0x1080a, 0x10835 }, + { 0x10837, 0x10838 }, + { 0x1083c, 0x1083c }, + { 0x1083f, 0x10855 }, + { 0x10860, 0x10876 }, + { 0x10880, 0x1089e }, + { 0x108e0, 0x108f2 }, + { 0x108f4, 0x108f5 }, + { 0x10900, 0x10915 }, + { 0x10920, 0x10939 }, + { 0x10980, 0x109b7 }, + { 0x109be, 0x109bf }, + { 0x10a00, 0x10a00 }, + { 0x10a10, 0x10a13 }, + { 0x10a15, 0x10a17 }, + { 0x10a19, 0x10a33 }, + { 0x10a60, 0x10a7c }, + { 0x10a80, 0x10a9c }, + { 0x10ac0, 0x10ac7 }, + { 0x10ac9, 0x10ae4 }, + { 0x10b00, 0x10b35 }, + { 0x10b40, 0x10b55 }, + { 0x10b60, 0x10b72 }, + { 0x10b80, 0x10b91 }, + { 0x10c00, 0x10c48 }, + { 0x11003, 0x11037 }, + { 0x11083, 0x110af }, + { 0x110d0, 0x110e8 }, + { 0x11103, 0x11126 }, + { 0x11150, 0x11172 }, + { 0x11176, 0x11176 }, + { 0x11183, 0x111b2 }, + { 0x111c1, 0x111c4 }, + { 0x111da, 0x111da }, + { 0x111dc, 0x111dc }, + { 0x11200, 0x11211 }, + { 0x11213, 0x1122b }, + { 0x11280, 0x11286 }, + { 0x11288, 0x11288 }, + { 0x1128a, 0x1128d }, + { 0x1128f, 0x1129d }, + { 0x1129f, 0x112a8 }, + { 0x112b0, 0x112de }, + { 0x11305, 0x1130c }, + { 0x1130f, 0x11310 }, + { 0x11313, 0x11328 }, + { 0x1132a, 0x11330 }, + { 0x11332, 0x11333 }, + { 0x11335, 0x11339 }, + { 0x1133d, 0x1133d }, + { 0x11350, 0x11350 }, + { 0x1135d, 0x11361 }, + { 0x11480, 0x114af }, + { 0x114c4, 0x114c5 }, + { 0x114c7, 0x114c7 }, + { 0x11580, 0x115ae }, + { 0x115d8, 0x115db }, + { 0x11600, 0x1162f }, + { 0x11644, 0x11644 }, + { 0x11680, 0x116aa }, + { 0x11700, 0x11719 }, + { 0x118ff, 0x118ff }, + { 0x11ac0, 0x11af8 }, + { 0x12000, 0x12399 }, + { 0x12480, 0x12543 }, + { 0x13000, 0x1342e }, + { 0x14400, 0x14646 }, + { 0x16800, 0x16a38 }, + { 0x16a40, 0x16a5e }, + { 0x16ad0, 0x16aed }, + { 0x16b00, 0x16b2f }, + { 0x16b40, 0x16b43 }, + { 0x16b63, 0x16b77 }, + { 0x16b7d, 0x16b8f }, + { 0x16f00, 0x16f44 }, + { 0x16f50, 0x16f50 }, + { 0x16f93, 0x16f9f }, + { 0x1b000, 0x1b001 }, + { 0x1bc00, 0x1bc6a }, + { 0x1bc70, 0x1bc7c }, + { 0x1bc80, 0x1bc88 }, + { 0x1bc90, 0x1bc99 }, + { 0x1e800, 0x1e8c4 }, + { 0x1ee00, 0x1ee03 }, + { 0x1ee05, 0x1ee1f }, + { 0x1ee21, 0x1ee22 }, + { 0x1ee24, 0x1ee24 }, + { 0x1ee27, 0x1ee27 }, + { 0x1ee29, 0x1ee32 }, + { 0x1ee34, 0x1ee37 }, + { 0x1ee39, 0x1ee39 }, + { 0x1ee3b, 0x1ee3b }, + { 0x1ee42, 0x1ee42 }, + { 0x1ee47, 0x1ee47 }, + { 0x1ee49, 0x1ee49 }, + { 0x1ee4b, 0x1ee4b }, + { 0x1ee4d, 0x1ee4f }, + { 0x1ee51, 0x1ee52 }, + { 0x1ee54, 0x1ee54 }, + { 0x1ee57, 0x1ee57 }, + { 0x1ee59, 0x1ee59 }, + { 0x1ee5b, 0x1ee5b }, + { 0x1ee5d, 0x1ee5d }, + { 0x1ee5f, 0x1ee5f }, + { 0x1ee61, 0x1ee62 }, + { 0x1ee64, 0x1ee64 }, + { 0x1ee67, 0x1ee6a }, + { 0x1ee6c, 0x1ee72 }, + { 0x1ee74, 0x1ee77 }, + { 0x1ee79, 0x1ee7c }, + { 0x1ee7e, 0x1ee7e }, + { 0x1ee80, 0x1ee89 }, + { 0x1ee8b, 0x1ee9b }, + { 0x1eea1, 0x1eea3 }, + { 0x1eea5, 0x1eea9 }, + { 0x1eeab, 0x1eebb }, + { 0x20000, 0x20000 }, + { 0x2a6d6, 0x2a6d6 }, + { 0x2a700, 0x2a700 }, + { 0x2b734, 0x2b734 }, + { 0x2b740, 0x2b740 }, + { 0x2b81d, 0x2b81d }, + { 0x2b820, 0x2b820 }, + { 0x2cea1, 0x2cea1 }, + { 0x2f800, 0x2fa1d } +}; + +extern int runeinrange(const void* a, const void* b); +bool isotherletterrune(Rune ch) { + return (NULL != bsearch(&ch, otherletters, 470, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/punctuation.c b/source/utf/punctuation.c new file mode 100644 index 0000000..f0da153 --- /dev/null +++ b/source/utf/punctuation.c @@ -0,0 +1,170 @@ +#include + +static Rune punctuation[161][2] = { + { 0x21, 0x23 }, + { 0x25, 0x2a }, + { 0x2c, 0x2f }, + { 0x3a, 0x3b }, + { 0x3f, 0x40 }, + { 0x5b, 0x5d }, + { 0x5f, 0x5f }, + { 0x7b, 0x7b }, + { 0x7d, 0x7d }, + { 0xa1, 0xa1 }, + { 0xa7, 0xa7 }, + { 0xab, 0xab }, + { 0xb6, 0xb7 }, + { 0xbb, 0xbb }, + { 0xbf, 0xbf }, + { 0x37e, 0x37e }, + { 0x387, 0x387 }, + { 0x55a, 0x55f }, + { 0x589, 0x58a }, + { 0x5be, 0x5be }, + { 0x5c0, 0x5c0 }, + { 0x5c3, 0x5c3 }, + { 0x5c6, 0x5c6 }, + { 0x5f3, 0x5f4 }, + { 0x609, 0x60a }, + { 0x60c, 0x60d }, + { 0x61b, 0x61b }, + { 0x61e, 0x61f }, + { 0x66a, 0x66d }, + { 0x6d4, 0x6d4 }, + { 0x700, 0x70d }, + { 0x7f7, 0x7f9 }, + { 0x830, 0x83e }, + { 0x85e, 0x85e }, + { 0x964, 0x965 }, + { 0x970, 0x970 }, + { 0xaf0, 0xaf0 }, + { 0xdf4, 0xdf4 }, + { 0xe4f, 0xe4f }, + { 0xe5a, 0xe5b }, + { 0xf04, 0xf12 }, + { 0xf14, 0xf14 }, + { 0xf3a, 0xf3d }, + { 0xf85, 0xf85 }, + { 0xfd0, 0xfd4 }, + { 0xfd9, 0xfda }, + { 0x104a, 0x104f }, + { 0x10fb, 0x10fb }, + { 0x1360, 0x1368 }, + { 0x1400, 0x1400 }, + { 0x166d, 0x166e }, + { 0x169b, 0x169c }, + { 0x16eb, 0x16ed }, + { 0x1735, 0x1736 }, + { 0x17d4, 0x17d6 }, + { 0x17d8, 0x17da }, + { 0x1800, 0x180a }, + { 0x1944, 0x1945 }, + { 0x1a1e, 0x1a1f }, + { 0x1aa0, 0x1aa6 }, + { 0x1aa8, 0x1aad }, + { 0x1b5a, 0x1b60 }, + { 0x1bfc, 0x1bff }, + { 0x1c3b, 0x1c3f }, + { 0x1c7e, 0x1c7f }, + { 0x1cc0, 0x1cc7 }, + { 0x1cd3, 0x1cd3 }, + { 0x2010, 0x2027 }, + { 0x2030, 0x2043 }, + { 0x2045, 0x2051 }, + { 0x2053, 0x205e }, + { 0x207d, 0x207e }, + { 0x208d, 0x208e }, + { 0x2308, 0x230b }, + { 0x2329, 0x232a }, + { 0x2768, 0x2775 }, + { 0x27c5, 0x27c6 }, + { 0x27e6, 0x27ef }, + { 0x2983, 0x2998 }, + { 0x29d8, 0x29db }, + { 0x29fc, 0x29fd }, + { 0x2cf9, 0x2cfc }, + { 0x2cfe, 0x2cff }, + { 0x2d70, 0x2d70 }, + { 0x2e00, 0x2e2e }, + { 0x2e30, 0x2e42 }, + { 0x3001, 0x3003 }, + { 0x3008, 0x3011 }, + { 0x3014, 0x301f }, + { 0x3030, 0x3030 }, + { 0x303d, 0x303d }, + { 0x30a0, 0x30a0 }, + { 0x30fb, 0x30fb }, + { 0xa4fe, 0xa4ff }, + { 0xa60d, 0xa60f }, + { 0xa673, 0xa673 }, + { 0xa67e, 0xa67e }, + { 0xa6f2, 0xa6f7 }, + { 0xa874, 0xa877 }, + { 0xa8ce, 0xa8cf }, + { 0xa8f8, 0xa8fa }, + { 0xa8fc, 0xa8fc }, + { 0xa92e, 0xa92f }, + { 0xa95f, 0xa95f }, + { 0xa9c1, 0xa9cd }, + { 0xa9de, 0xa9df }, + { 0xaa5c, 0xaa5f }, + { 0xaade, 0xaadf }, + { 0xaaf0, 0xaaf1 }, + { 0xabeb, 0xabeb }, + { 0xfd3e, 0xfd3f }, + { 0xfe10, 0xfe19 }, + { 0xfe30, 0xfe52 }, + { 0xfe54, 0xfe61 }, + { 0xfe63, 0xfe63 }, + { 0xfe68, 0xfe68 }, + { 0xfe6a, 0xfe6b }, + { 0xff01, 0xff03 }, + { 0xff05, 0xff0a }, + { 0xff0c, 0xff0f }, + { 0xff1a, 0xff1b }, + { 0xff1f, 0xff20 }, + { 0xff3b, 0xff3d }, + { 0xff3f, 0xff3f }, + { 0xff5b, 0xff5b }, + { 0xff5d, 0xff5d }, + { 0xff5f, 0xff65 }, + { 0x10100, 0x10102 }, + { 0x1039f, 0x1039f }, + { 0x103d0, 0x103d0 }, + { 0x1056f, 0x1056f }, + { 0x10857, 0x10857 }, + { 0x1091f, 0x1091f }, + { 0x1093f, 0x1093f }, + { 0x10a50, 0x10a58 }, + { 0x10a7f, 0x10a7f }, + { 0x10af0, 0x10af6 }, + { 0x10b39, 0x10b3f }, + { 0x10b99, 0x10b9c }, + { 0x11047, 0x1104d }, + { 0x110bb, 0x110bc }, + { 0x110be, 0x110c1 }, + { 0x11140, 0x11143 }, + { 0x11174, 0x11175 }, + { 0x111c5, 0x111c9 }, + { 0x111cd, 0x111cd }, + { 0x111db, 0x111db }, + { 0x111dd, 0x111df }, + { 0x11238, 0x1123d }, + { 0x112a9, 0x112a9 }, + { 0x114c6, 0x114c6 }, + { 0x115c1, 0x115d7 }, + { 0x11641, 0x11643 }, + { 0x1173c, 0x1173e }, + { 0x12470, 0x12474 }, + { 0x16a6e, 0x16a6f }, + { 0x16af5, 0x16af5 }, + { 0x16b37, 0x16b3b }, + { 0x16b44, 0x16b44 }, + { 0x1bc9f, 0x1bc9f }, + { 0x1da87, 0x1da8b } +}; + +extern int runeinrange(const void* a, const void* b); +bool ispunctuationrune(Rune ch) { + return (NULL != bsearch(&ch, punctuation, 161, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/runeinrange.c b/source/utf/runeinrange.c new file mode 100644 index 0000000..7309a3c --- /dev/null +++ b/source/utf/runeinrange.c @@ -0,0 +1,14 @@ +#include "libc.h" + +/* Used by rune type checking functions to find the rune in the type tables */ +int runeinrange(const void* a, const void* b) { + Rune* item = (Rune*)a; + Rune* range = (Rune*)b; + if (*item < range[0]) + return -1; + else if (*item > range[1]) + return 1; + else + return 0; +} + diff --git a/source/utf/rune.c b/source/utf/runelen.c similarity index 56% rename from source/utf/rune.c rename to source/utf/runelen.c index 5c9df42..1e0c102 100644 --- a/source/utf/rune.c +++ b/source/utf/runelen.c @@ -1,19 +1,5 @@ #include "libc.h" -int runetochar(char* s, Rune* r) -{ - (void)s; - (void)r; - return 0; -} - -int chartorune(Rune* r, char* s) -{ - (void)r; - (void)s; - return 0; -} - int runelen(long r) { if(r <= 0x7F) @@ -34,16 +20,3 @@ int runelen(long r) return 0; /* rune too large */ } -int runenlen(Rune* r, int num) -{ - size_t i, n = 0; - for(i = 0; i < num; i++) - n += runelen(r[i]); - return n; -} - -bool fullrune(char* s, int n) -{ - return false; -} - diff --git a/source/utf/runenlen.c b/source/utf/runenlen.c new file mode 100644 index 0000000..326d329 --- /dev/null +++ b/source/utf/runenlen.c @@ -0,0 +1,9 @@ +#include "libc.h" + +int runenlen(Rune* r, int num) +{ + int i = 0, n = 0; + for(i = 0; i < num; i++) + n += runelen(r[i]); + return n; +} diff --git a/source/utf/runetochar.c b/source/utf/runetochar.c new file mode 100644 index 0000000..faca3f2 --- /dev/null +++ b/source/utf/runetochar.c @@ -0,0 +1,9 @@ +#include "libc.h" + +int runetochar(char* s, Rune* r) +{ + (void)s; + (void)r; + return 0; +} + diff --git a/source/utf/runetype.c b/source/utf/runetype.c index c80c109..701d830 100644 --- a/source/utf/runetype.c +++ b/source/utf/runetype.c @@ -24,43 +24,21 @@ Rune tobaserune(Rune ch) return 0; } -bool isalpharune(Rune ch) -{ - (void)ch; - return false; -} -bool isbaserune(Rune ch) -{ - (void)ch; - return false; -} -bool isdigitrune(Rune ch) -{ - (void)ch; - return false; -} - -bool islowerrune(Rune ch) -{ - (void)ch; - return false; -} -bool isspacerune(Rune ch) +bool isalpharune(Rune ch) { - (void)ch; - return false; + return (isupperrune(ch) || islowerrune(ch)); } -bool istitlerune(Rune ch) +bool isbaserune(Rune ch) { (void)ch; return false; } -bool isupperrune(Rune ch) +bool isdigitrune(Rune ch) { (void)ch; return false; diff --git a/source/utf/spaces.c b/source/utf/spaces.c new file mode 100644 index 0000000..26973d9 --- /dev/null +++ b/source/utf/spaces.c @@ -0,0 +1,19 @@ +#include + +static Rune spaces[10][2] = { + { 0x9, 0xd }, + { 0x1c, 0x20 }, + { 0x85, 0x85 }, + { 0xa0, 0xa0 }, + { 0x1680, 0x1680 }, + { 0x2000, 0x200a }, + { 0x2028, 0x2029 }, + { 0x202f, 0x202f }, + { 0x205f, 0x205f }, + { 0x3000, 0x3000 } +}; + +extern int runeinrange(const void* a, const void* b); +bool isspacerune(Rune ch) { + return (NULL != bsearch(&ch, spaces, 10, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/symbols.c b/source/utf/symbols.c new file mode 100644 index 0000000..f83e6a4 --- /dev/null +++ b/source/utf/symbols.c @@ -0,0 +1,223 @@ +#include + +static Rune symbols[214][2] = { + { 0x24, 0x24 }, + { 0x2b, 0x2b }, + { 0x3c, 0x3e }, + { 0x5e, 0x5e }, + { 0x60, 0x60 }, + { 0x7c, 0x7c }, + { 0x7e, 0x7e }, + { 0xa2, 0xa6 }, + { 0xa8, 0xa9 }, + { 0xac, 0xac }, + { 0xae, 0xb1 }, + { 0xb4, 0xb4 }, + { 0xb8, 0xb8 }, + { 0xd7, 0xd7 }, + { 0xf7, 0xf7 }, + { 0x2c2, 0x2c5 }, + { 0x2d2, 0x2df }, + { 0x2e5, 0x2eb }, + { 0x2ed, 0x2ed }, + { 0x2ef, 0x2ff }, + { 0x375, 0x375 }, + { 0x384, 0x385 }, + { 0x3f6, 0x3f6 }, + { 0x482, 0x482 }, + { 0x58d, 0x58f }, + { 0x606, 0x608 }, + { 0x60b, 0x60b }, + { 0x60e, 0x60f }, + { 0x6de, 0x6de }, + { 0x6e9, 0x6e9 }, + { 0x6fd, 0x6fe }, + { 0x7f6, 0x7f6 }, + { 0x9f2, 0x9f3 }, + { 0x9fa, 0x9fb }, + { 0xaf1, 0xaf1 }, + { 0xb70, 0xb70 }, + { 0xbf3, 0xbfa }, + { 0xc7f, 0xc7f }, + { 0xd79, 0xd79 }, + { 0xe3f, 0xe3f }, + { 0xf01, 0xf03 }, + { 0xf13, 0xf13 }, + { 0xf15, 0xf17 }, + { 0xf1a, 0xf1f }, + { 0xf34, 0xf34 }, + { 0xf36, 0xf36 }, + { 0xf38, 0xf38 }, + { 0xfbe, 0xfc5 }, + { 0xfc7, 0xfcc }, + { 0xfce, 0xfcf }, + { 0xfd5, 0xfd8 }, + { 0x109e, 0x109f }, + { 0x1390, 0x1399 }, + { 0x17db, 0x17db }, + { 0x1940, 0x1940 }, + { 0x19de, 0x19ff }, + { 0x1b61, 0x1b6a }, + { 0x1b74, 0x1b7c }, + { 0x1fbd, 0x1fbd }, + { 0x1fbf, 0x1fc1 }, + { 0x1fcd, 0x1fcf }, + { 0x1fdd, 0x1fdf }, + { 0x1fed, 0x1fef }, + { 0x1ffd, 0x1ffe }, + { 0x2044, 0x2044 }, + { 0x2052, 0x2052 }, + { 0x207a, 0x207c }, + { 0x208a, 0x208c }, + { 0x20a0, 0x20be }, + { 0x2100, 0x2101 }, + { 0x2103, 0x2106 }, + { 0x2108, 0x2109 }, + { 0x2114, 0x2114 }, + { 0x2116, 0x2118 }, + { 0x211e, 0x2123 }, + { 0x2125, 0x2125 }, + { 0x2127, 0x2127 }, + { 0x2129, 0x2129 }, + { 0x212e, 0x212e }, + { 0x213a, 0x213b }, + { 0x2140, 0x2144 }, + { 0x214a, 0x214d }, + { 0x214f, 0x214f }, + { 0x218a, 0x218b }, + { 0x2190, 0x2307 }, + { 0x230c, 0x2328 }, + { 0x232b, 0x23fa }, + { 0x2400, 0x2426 }, + { 0x2440, 0x244a }, + { 0x249c, 0x24e9 }, + { 0x2500, 0x2767 }, + { 0x2794, 0x27c4 }, + { 0x27c7, 0x27e5 }, + { 0x27f0, 0x2982 }, + { 0x2999, 0x29d7 }, + { 0x29dc, 0x29fb }, + { 0x29fe, 0x2b73 }, + { 0x2b76, 0x2b95 }, + { 0x2b98, 0x2bb9 }, + { 0x2bbd, 0x2bc8 }, + { 0x2bca, 0x2bd1 }, + { 0x2bec, 0x2bef }, + { 0x2ce5, 0x2cea }, + { 0x2e80, 0x2e99 }, + { 0x2e9b, 0x2ef3 }, + { 0x2f00, 0x2fd5 }, + { 0x2ff0, 0x2ffb }, + { 0x3004, 0x3004 }, + { 0x3012, 0x3013 }, + { 0x3020, 0x3020 }, + { 0x3036, 0x3037 }, + { 0x303e, 0x303f }, + { 0x309b, 0x309c }, + { 0x3190, 0x3191 }, + { 0x3196, 0x319f }, + { 0x31c0, 0x31e3 }, + { 0x3200, 0x321e }, + { 0x322a, 0x3247 }, + { 0x3250, 0x3250 }, + { 0x3260, 0x327f }, + { 0x328a, 0x32b0 }, + { 0x32c0, 0x32fe }, + { 0x3300, 0x33ff }, + { 0x4dc0, 0x4dff }, + { 0xa490, 0xa4c6 }, + { 0xa700, 0xa716 }, + { 0xa720, 0xa721 }, + { 0xa789, 0xa78a }, + { 0xa828, 0xa82b }, + { 0xa836, 0xa839 }, + { 0xaa77, 0xaa79 }, + { 0xab5b, 0xab5b }, + { 0xfb29, 0xfb29 }, + { 0xfbb2, 0xfbc1 }, + { 0xfdfc, 0xfdfd }, + { 0xfe62, 0xfe62 }, + { 0xfe64, 0xfe66 }, + { 0xfe69, 0xfe69 }, + { 0xff04, 0xff04 }, + { 0xff0b, 0xff0b }, + { 0xff1c, 0xff1e }, + { 0xff3e, 0xff3e }, + { 0xff40, 0xff40 }, + { 0xff5c, 0xff5c }, + { 0xff5e, 0xff5e }, + { 0xffe0, 0xffe6 }, + { 0xffe8, 0xffee }, + { 0xfffc, 0xfffd }, + { 0x10137, 0x1013f }, + { 0x10179, 0x10189 }, + { 0x1018c, 0x1018c }, + { 0x10190, 0x1019b }, + { 0x101a0, 0x101a0 }, + { 0x101d0, 0x101fc }, + { 0x10877, 0x10878 }, + { 0x10ac8, 0x10ac8 }, + { 0x1173f, 0x1173f }, + { 0x16b3c, 0x16b3f }, + { 0x16b45, 0x16b45 }, + { 0x1bc9c, 0x1bc9c }, + { 0x1d000, 0x1d0f5 }, + { 0x1d100, 0x1d126 }, + { 0x1d129, 0x1d164 }, + { 0x1d16a, 0x1d16c }, + { 0x1d183, 0x1d184 }, + { 0x1d18c, 0x1d1a9 }, + { 0x1d1ae, 0x1d1e8 }, + { 0x1d200, 0x1d241 }, + { 0x1d245, 0x1d245 }, + { 0x1d300, 0x1d356 }, + { 0x1d6c1, 0x1d6c1 }, + { 0x1d6db, 0x1d6db }, + { 0x1d6fb, 0x1d6fb }, + { 0x1d715, 0x1d715 }, + { 0x1d735, 0x1d735 }, + { 0x1d74f, 0x1d74f }, + { 0x1d76f, 0x1d76f }, + { 0x1d789, 0x1d789 }, + { 0x1d7a9, 0x1d7a9 }, + { 0x1d7c3, 0x1d7c3 }, + { 0x1d800, 0x1d9ff }, + { 0x1da37, 0x1da3a }, + { 0x1da6d, 0x1da74 }, + { 0x1da76, 0x1da83 }, + { 0x1da85, 0x1da86 }, + { 0x1eef0, 0x1eef1 }, + { 0x1f000, 0x1f02b }, + { 0x1f030, 0x1f093 }, + { 0x1f0a0, 0x1f0ae }, + { 0x1f0b1, 0x1f0bf }, + { 0x1f0c1, 0x1f0cf }, + { 0x1f0d1, 0x1f0f5 }, + { 0x1f110, 0x1f12e }, + { 0x1f130, 0x1f16b }, + { 0x1f170, 0x1f19a }, + { 0x1f1e6, 0x1f202 }, + { 0x1f210, 0x1f23a }, + { 0x1f240, 0x1f248 }, + { 0x1f250, 0x1f251 }, + { 0x1f300, 0x1f579 }, + { 0x1f57b, 0x1f5a3 }, + { 0x1f5a5, 0x1f6d0 }, + { 0x1f6e0, 0x1f6ec }, + { 0x1f6f0, 0x1f6f3 }, + { 0x1f700, 0x1f773 }, + { 0x1f780, 0x1f7d4 }, + { 0x1f800, 0x1f80b }, + { 0x1f810, 0x1f847 }, + { 0x1f850, 0x1f859 }, + { 0x1f860, 0x1f887 }, + { 0x1f890, 0x1f8ad }, + { 0x1f910, 0x1f918 }, + { 0x1f980, 0x1f984 }, + { 0x1f9c0, 0x1f9c0 } +}; + +extern int runeinrange(const void* a, const void* b); +bool issymbolrune(Rune ch) { + return (NULL != bsearch(&ch, symbols, 214, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/titles.c b/source/utf/titles.c new file mode 100644 index 0000000..5401116 --- /dev/null +++ b/source/utf/titles.c @@ -0,0 +1,19 @@ +#include + +static Rune titles[10][2] = { + { 0x1c5, 0x1c5 }, + { 0x1c8, 0x1c8 }, + { 0x1cb, 0x1cb }, + { 0x1f2, 0x1f2 }, + { 0x1f88, 0x1f8f }, + { 0x1f98, 0x1f9f }, + { 0x1fa8, 0x1faf }, + { 0x1fbc, 0x1fbc }, + { 0x1fcc, 0x1fcc }, + { 0x1ffc, 0x1ffc } +}; + +extern int runeinrange(const void* a, const void* b); +bool istitlerune(Rune ch) { + return (NULL != bsearch(&ch, titles, 10, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/source/utf/uppers.c b/source/utf/uppers.c new file mode 100644 index 0000000..fd8499f --- /dev/null +++ b/source/utf/uppers.c @@ -0,0 +1,634 @@ +#include + +static Rune uppers[625][2] = { + { 0x41, 0x5a }, + { 0xc0, 0xd6 }, + { 0xd8, 0xde }, + { 0x100, 0x100 }, + { 0x102, 0x102 }, + { 0x104, 0x104 }, + { 0x106, 0x106 }, + { 0x108, 0x108 }, + { 0x10a, 0x10a }, + { 0x10c, 0x10c }, + { 0x10e, 0x10e }, + { 0x110, 0x110 }, + { 0x112, 0x112 }, + { 0x114, 0x114 }, + { 0x116, 0x116 }, + { 0x118, 0x118 }, + { 0x11a, 0x11a }, + { 0x11c, 0x11c }, + { 0x11e, 0x11e }, + { 0x120, 0x120 }, + { 0x122, 0x122 }, + { 0x124, 0x124 }, + { 0x126, 0x126 }, + { 0x128, 0x128 }, + { 0x12a, 0x12a }, + { 0x12c, 0x12c }, + { 0x12e, 0x12e }, + { 0x130, 0x130 }, + { 0x132, 0x132 }, + { 0x134, 0x134 }, + { 0x136, 0x136 }, + { 0x139, 0x139 }, + { 0x13b, 0x13b }, + { 0x13d, 0x13d }, + { 0x13f, 0x13f }, + { 0x141, 0x141 }, + { 0x143, 0x143 }, + { 0x145, 0x145 }, + { 0x147, 0x147 }, + { 0x14a, 0x14a }, + { 0x14c, 0x14c }, + { 0x14e, 0x14e }, + { 0x150, 0x150 }, + { 0x152, 0x152 }, + { 0x154, 0x154 }, + { 0x156, 0x156 }, + { 0x158, 0x158 }, + { 0x15a, 0x15a }, + { 0x15c, 0x15c }, + { 0x15e, 0x15e }, + { 0x160, 0x160 }, + { 0x162, 0x162 }, + { 0x164, 0x164 }, + { 0x166, 0x166 }, + { 0x168, 0x168 }, + { 0x16a, 0x16a }, + { 0x16c, 0x16c }, + { 0x16e, 0x16e }, + { 0x170, 0x170 }, + { 0x172, 0x172 }, + { 0x174, 0x174 }, + { 0x176, 0x176 }, + { 0x178, 0x179 }, + { 0x17b, 0x17b }, + { 0x17d, 0x17d }, + { 0x181, 0x182 }, + { 0x184, 0x184 }, + { 0x186, 0x187 }, + { 0x189, 0x18b }, + { 0x18e, 0x191 }, + { 0x193, 0x194 }, + { 0x196, 0x198 }, + { 0x19c, 0x19d }, + { 0x19f, 0x1a0 }, + { 0x1a2, 0x1a2 }, + { 0x1a4, 0x1a4 }, + { 0x1a6, 0x1a7 }, + { 0x1a9, 0x1a9 }, + { 0x1ac, 0x1ac }, + { 0x1ae, 0x1af }, + { 0x1b1, 0x1b3 }, + { 0x1b5, 0x1b5 }, + { 0x1b7, 0x1b8 }, + { 0x1bc, 0x1bc }, + { 0x1c4, 0x1c4 }, + { 0x1c7, 0x1c7 }, + { 0x1ca, 0x1ca }, + { 0x1cd, 0x1cd }, + { 0x1cf, 0x1cf }, + { 0x1d1, 0x1d1 }, + { 0x1d3, 0x1d3 }, + { 0x1d5, 0x1d5 }, + { 0x1d7, 0x1d7 }, + { 0x1d9, 0x1d9 }, + { 0x1db, 0x1db }, + { 0x1de, 0x1de }, + { 0x1e0, 0x1e0 }, + { 0x1e2, 0x1e2 }, + { 0x1e4, 0x1e4 }, + { 0x1e6, 0x1e6 }, + { 0x1e8, 0x1e8 }, + { 0x1ea, 0x1ea }, + { 0x1ec, 0x1ec }, + { 0x1ee, 0x1ee }, + { 0x1f1, 0x1f1 }, + { 0x1f4, 0x1f4 }, + { 0x1f6, 0x1f8 }, + { 0x1fa, 0x1fa }, + { 0x1fc, 0x1fc }, + { 0x1fe, 0x1fe }, + { 0x200, 0x200 }, + { 0x202, 0x202 }, + { 0x204, 0x204 }, + { 0x206, 0x206 }, + { 0x208, 0x208 }, + { 0x20a, 0x20a }, + { 0x20c, 0x20c }, + { 0x20e, 0x20e }, + { 0x210, 0x210 }, + { 0x212, 0x212 }, + { 0x214, 0x214 }, + { 0x216, 0x216 }, + { 0x218, 0x218 }, + { 0x21a, 0x21a }, + { 0x21c, 0x21c }, + { 0x21e, 0x21e }, + { 0x220, 0x220 }, + { 0x222, 0x222 }, + { 0x224, 0x224 }, + { 0x226, 0x226 }, + { 0x228, 0x228 }, + { 0x22a, 0x22a }, + { 0x22c, 0x22c }, + { 0x22e, 0x22e }, + { 0x230, 0x230 }, + { 0x232, 0x232 }, + { 0x23a, 0x23b }, + { 0x23d, 0x23e }, + { 0x241, 0x241 }, + { 0x243, 0x246 }, + { 0x248, 0x248 }, + { 0x24a, 0x24a }, + { 0x24c, 0x24c }, + { 0x24e, 0x24e }, + { 0x370, 0x370 }, + { 0x372, 0x372 }, + { 0x376, 0x376 }, + { 0x37f, 0x37f }, + { 0x386, 0x386 }, + { 0x388, 0x38a }, + { 0x38c, 0x38c }, + { 0x38e, 0x38f }, + { 0x391, 0x3a1 }, + { 0x3a3, 0x3ab }, + { 0x3cf, 0x3cf }, + { 0x3d2, 0x3d4 }, + { 0x3d8, 0x3d8 }, + { 0x3da, 0x3da }, + { 0x3dc, 0x3dc }, + { 0x3de, 0x3de }, + { 0x3e0, 0x3e0 }, + { 0x3e2, 0x3e2 }, + { 0x3e4, 0x3e4 }, + { 0x3e6, 0x3e6 }, + { 0x3e8, 0x3e8 }, + { 0x3ea, 0x3ea }, + { 0x3ec, 0x3ec }, + { 0x3ee, 0x3ee }, + { 0x3f4, 0x3f4 }, + { 0x3f7, 0x3f7 }, + { 0x3f9, 0x3fa }, + { 0x3fd, 0x42f }, + { 0x460, 0x460 }, + { 0x462, 0x462 }, + { 0x464, 0x464 }, + { 0x466, 0x466 }, + { 0x468, 0x468 }, + { 0x46a, 0x46a }, + { 0x46c, 0x46c }, + { 0x46e, 0x46e }, + { 0x470, 0x470 }, + { 0x472, 0x472 }, + { 0x474, 0x474 }, + { 0x476, 0x476 }, + { 0x478, 0x478 }, + { 0x47a, 0x47a }, + { 0x47c, 0x47c }, + { 0x47e, 0x47e }, + { 0x480, 0x480 }, + { 0x48a, 0x48a }, + { 0x48c, 0x48c }, + { 0x48e, 0x48e }, + { 0x490, 0x490 }, + { 0x492, 0x492 }, + { 0x494, 0x494 }, + { 0x496, 0x496 }, + { 0x498, 0x498 }, + { 0x49a, 0x49a }, + { 0x49c, 0x49c }, + { 0x49e, 0x49e }, + { 0x4a0, 0x4a0 }, + { 0x4a2, 0x4a2 }, + { 0x4a4, 0x4a4 }, + { 0x4a6, 0x4a6 }, + { 0x4a8, 0x4a8 }, + { 0x4aa, 0x4aa }, + { 0x4ac, 0x4ac }, + { 0x4ae, 0x4ae }, + { 0x4b0, 0x4b0 }, + { 0x4b2, 0x4b2 }, + { 0x4b4, 0x4b4 }, + { 0x4b6, 0x4b6 }, + { 0x4b8, 0x4b8 }, + { 0x4ba, 0x4ba }, + { 0x4bc, 0x4bc }, + { 0x4be, 0x4be }, + { 0x4c0, 0x4c1 }, + { 0x4c3, 0x4c3 }, + { 0x4c5, 0x4c5 }, + { 0x4c7, 0x4c7 }, + { 0x4c9, 0x4c9 }, + { 0x4cb, 0x4cb }, + { 0x4cd, 0x4cd }, + { 0x4d0, 0x4d0 }, + { 0x4d2, 0x4d2 }, + { 0x4d4, 0x4d4 }, + { 0x4d6, 0x4d6 }, + { 0x4d8, 0x4d8 }, + { 0x4da, 0x4da }, + { 0x4dc, 0x4dc }, + { 0x4de, 0x4de }, + { 0x4e0, 0x4e0 }, + { 0x4e2, 0x4e2 }, + { 0x4e4, 0x4e4 }, + { 0x4e6, 0x4e6 }, + { 0x4e8, 0x4e8 }, + { 0x4ea, 0x4ea }, + { 0x4ec, 0x4ec }, + { 0x4ee, 0x4ee }, + { 0x4f0, 0x4f0 }, + { 0x4f2, 0x4f2 }, + { 0x4f4, 0x4f4 }, + { 0x4f6, 0x4f6 }, + { 0x4f8, 0x4f8 }, + { 0x4fa, 0x4fa }, + { 0x4fc, 0x4fc }, + { 0x4fe, 0x4fe }, + { 0x500, 0x500 }, + { 0x502, 0x502 }, + { 0x504, 0x504 }, + { 0x506, 0x506 }, + { 0x508, 0x508 }, + { 0x50a, 0x50a }, + { 0x50c, 0x50c }, + { 0x50e, 0x50e }, + { 0x510, 0x510 }, + { 0x512, 0x512 }, + { 0x514, 0x514 }, + { 0x516, 0x516 }, + { 0x518, 0x518 }, + { 0x51a, 0x51a }, + { 0x51c, 0x51c }, + { 0x51e, 0x51e }, + { 0x520, 0x520 }, + { 0x522, 0x522 }, + { 0x524, 0x524 }, + { 0x526, 0x526 }, + { 0x528, 0x528 }, + { 0x52a, 0x52a }, + { 0x52c, 0x52c }, + { 0x52e, 0x52e }, + { 0x531, 0x556 }, + { 0x10a0, 0x10c5 }, + { 0x10c7, 0x10c7 }, + { 0x10cd, 0x10cd }, + { 0x13a0, 0x13f5 }, + { 0x1e00, 0x1e00 }, + { 0x1e02, 0x1e02 }, + { 0x1e04, 0x1e04 }, + { 0x1e06, 0x1e06 }, + { 0x1e08, 0x1e08 }, + { 0x1e0a, 0x1e0a }, + { 0x1e0c, 0x1e0c }, + { 0x1e0e, 0x1e0e }, + { 0x1e10, 0x1e10 }, + { 0x1e12, 0x1e12 }, + { 0x1e14, 0x1e14 }, + { 0x1e16, 0x1e16 }, + { 0x1e18, 0x1e18 }, + { 0x1e1a, 0x1e1a }, + { 0x1e1c, 0x1e1c }, + { 0x1e1e, 0x1e1e }, + { 0x1e20, 0x1e20 }, + { 0x1e22, 0x1e22 }, + { 0x1e24, 0x1e24 }, + { 0x1e26, 0x1e26 }, + { 0x1e28, 0x1e28 }, + { 0x1e2a, 0x1e2a }, + { 0x1e2c, 0x1e2c }, + { 0x1e2e, 0x1e2e }, + { 0x1e30, 0x1e30 }, + { 0x1e32, 0x1e32 }, + { 0x1e34, 0x1e34 }, + { 0x1e36, 0x1e36 }, + { 0x1e38, 0x1e38 }, + { 0x1e3a, 0x1e3a }, + { 0x1e3c, 0x1e3c }, + { 0x1e3e, 0x1e3e }, + { 0x1e40, 0x1e40 }, + { 0x1e42, 0x1e42 }, + { 0x1e44, 0x1e44 }, + { 0x1e46, 0x1e46 }, + { 0x1e48, 0x1e48 }, + { 0x1e4a, 0x1e4a }, + { 0x1e4c, 0x1e4c }, + { 0x1e4e, 0x1e4e }, + { 0x1e50, 0x1e50 }, + { 0x1e52, 0x1e52 }, + { 0x1e54, 0x1e54 }, + { 0x1e56, 0x1e56 }, + { 0x1e58, 0x1e58 }, + { 0x1e5a, 0x1e5a }, + { 0x1e5c, 0x1e5c }, + { 0x1e5e, 0x1e5e }, + { 0x1e60, 0x1e60 }, + { 0x1e62, 0x1e62 }, + { 0x1e64, 0x1e64 }, + { 0x1e66, 0x1e66 }, + { 0x1e68, 0x1e68 }, + { 0x1e6a, 0x1e6a }, + { 0x1e6c, 0x1e6c }, + { 0x1e6e, 0x1e6e }, + { 0x1e70, 0x1e70 }, + { 0x1e72, 0x1e72 }, + { 0x1e74, 0x1e74 }, + { 0x1e76, 0x1e76 }, + { 0x1e78, 0x1e78 }, + { 0x1e7a, 0x1e7a }, + { 0x1e7c, 0x1e7c }, + { 0x1e7e, 0x1e7e }, + { 0x1e80, 0x1e80 }, + { 0x1e82, 0x1e82 }, + { 0x1e84, 0x1e84 }, + { 0x1e86, 0x1e86 }, + { 0x1e88, 0x1e88 }, + { 0x1e8a, 0x1e8a }, + { 0x1e8c, 0x1e8c }, + { 0x1e8e, 0x1e8e }, + { 0x1e90, 0x1e90 }, + { 0x1e92, 0x1e92 }, + { 0x1e94, 0x1e94 }, + { 0x1e9e, 0x1e9e }, + { 0x1ea0, 0x1ea0 }, + { 0x1ea2, 0x1ea2 }, + { 0x1ea4, 0x1ea4 }, + { 0x1ea6, 0x1ea6 }, + { 0x1ea8, 0x1ea8 }, + { 0x1eaa, 0x1eaa }, + { 0x1eac, 0x1eac }, + { 0x1eae, 0x1eae }, + { 0x1eb0, 0x1eb0 }, + { 0x1eb2, 0x1eb2 }, + { 0x1eb4, 0x1eb4 }, + { 0x1eb6, 0x1eb6 }, + { 0x1eb8, 0x1eb8 }, + { 0x1eba, 0x1eba }, + { 0x1ebc, 0x1ebc }, + { 0x1ebe, 0x1ebe }, + { 0x1ec0, 0x1ec0 }, + { 0x1ec2, 0x1ec2 }, + { 0x1ec4, 0x1ec4 }, + { 0x1ec6, 0x1ec6 }, + { 0x1ec8, 0x1ec8 }, + { 0x1eca, 0x1eca }, + { 0x1ecc, 0x1ecc }, + { 0x1ece, 0x1ece }, + { 0x1ed0, 0x1ed0 }, + { 0x1ed2, 0x1ed2 }, + { 0x1ed4, 0x1ed4 }, + { 0x1ed6, 0x1ed6 }, + { 0x1ed8, 0x1ed8 }, + { 0x1eda, 0x1eda }, + { 0x1edc, 0x1edc }, + { 0x1ede, 0x1ede }, + { 0x1ee0, 0x1ee0 }, + { 0x1ee2, 0x1ee2 }, + { 0x1ee4, 0x1ee4 }, + { 0x1ee6, 0x1ee6 }, + { 0x1ee8, 0x1ee8 }, + { 0x1eea, 0x1eea }, + { 0x1eec, 0x1eec }, + { 0x1eee, 0x1eee }, + { 0x1ef0, 0x1ef0 }, + { 0x1ef2, 0x1ef2 }, + { 0x1ef4, 0x1ef4 }, + { 0x1ef6, 0x1ef6 }, + { 0x1ef8, 0x1ef8 }, + { 0x1efa, 0x1efa }, + { 0x1efc, 0x1efc }, + { 0x1efe, 0x1efe }, + { 0x1f08, 0x1f0f }, + { 0x1f18, 0x1f1d }, + { 0x1f28, 0x1f2f }, + { 0x1f38, 0x1f3f }, + { 0x1f48, 0x1f4d }, + { 0x1f59, 0x1f59 }, + { 0x1f5b, 0x1f5b }, + { 0x1f5d, 0x1f5d }, + { 0x1f5f, 0x1f5f }, + { 0x1f68, 0x1f6f }, + { 0x1fb8, 0x1fbb }, + { 0x1fc8, 0x1fcb }, + { 0x1fd8, 0x1fdb }, + { 0x1fe8, 0x1fec }, + { 0x1ff8, 0x1ffb }, + { 0x2102, 0x2102 }, + { 0x2107, 0x2107 }, + { 0x210b, 0x210d }, + { 0x2110, 0x2112 }, + { 0x2115, 0x2115 }, + { 0x2119, 0x211d }, + { 0x2124, 0x2124 }, + { 0x2126, 0x2126 }, + { 0x2128, 0x2128 }, + { 0x212a, 0x212d }, + { 0x2130, 0x2133 }, + { 0x213e, 0x213f }, + { 0x2145, 0x2145 }, + { 0x2183, 0x2183 }, + { 0x2c00, 0x2c2e }, + { 0x2c60, 0x2c60 }, + { 0x2c62, 0x2c64 }, + { 0x2c67, 0x2c67 }, + { 0x2c69, 0x2c69 }, + { 0x2c6b, 0x2c6b }, + { 0x2c6d, 0x2c70 }, + { 0x2c72, 0x2c72 }, + { 0x2c75, 0x2c75 }, + { 0x2c7e, 0x2c80 }, + { 0x2c82, 0x2c82 }, + { 0x2c84, 0x2c84 }, + { 0x2c86, 0x2c86 }, + { 0x2c88, 0x2c88 }, + { 0x2c8a, 0x2c8a }, + { 0x2c8c, 0x2c8c }, + { 0x2c8e, 0x2c8e }, + { 0x2c90, 0x2c90 }, + { 0x2c92, 0x2c92 }, + { 0x2c94, 0x2c94 }, + { 0x2c96, 0x2c96 }, + { 0x2c98, 0x2c98 }, + { 0x2c9a, 0x2c9a }, + { 0x2c9c, 0x2c9c }, + { 0x2c9e, 0x2c9e }, + { 0x2ca0, 0x2ca0 }, + { 0x2ca2, 0x2ca2 }, + { 0x2ca4, 0x2ca4 }, + { 0x2ca6, 0x2ca6 }, + { 0x2ca8, 0x2ca8 }, + { 0x2caa, 0x2caa }, + { 0x2cac, 0x2cac }, + { 0x2cae, 0x2cae }, + { 0x2cb0, 0x2cb0 }, + { 0x2cb2, 0x2cb2 }, + { 0x2cb4, 0x2cb4 }, + { 0x2cb6, 0x2cb6 }, + { 0x2cb8, 0x2cb8 }, + { 0x2cba, 0x2cba }, + { 0x2cbc, 0x2cbc }, + { 0x2cbe, 0x2cbe }, + { 0x2cc0, 0x2cc0 }, + { 0x2cc2, 0x2cc2 }, + { 0x2cc4, 0x2cc4 }, + { 0x2cc6, 0x2cc6 }, + { 0x2cc8, 0x2cc8 }, + { 0x2cca, 0x2cca }, + { 0x2ccc, 0x2ccc }, + { 0x2cce, 0x2cce }, + { 0x2cd0, 0x2cd0 }, + { 0x2cd2, 0x2cd2 }, + { 0x2cd4, 0x2cd4 }, + { 0x2cd6, 0x2cd6 }, + { 0x2cd8, 0x2cd8 }, + { 0x2cda, 0x2cda }, + { 0x2cdc, 0x2cdc }, + { 0x2cde, 0x2cde }, + { 0x2ce0, 0x2ce0 }, + { 0x2ce2, 0x2ce2 }, + { 0x2ceb, 0x2ceb }, + { 0x2ced, 0x2ced }, + { 0x2cf2, 0x2cf2 }, + { 0xa640, 0xa640 }, + { 0xa642, 0xa642 }, + { 0xa644, 0xa644 }, + { 0xa646, 0xa646 }, + { 0xa648, 0xa648 }, + { 0xa64a, 0xa64a }, + { 0xa64c, 0xa64c }, + { 0xa64e, 0xa64e }, + { 0xa650, 0xa650 }, + { 0xa652, 0xa652 }, + { 0xa654, 0xa654 }, + { 0xa656, 0xa656 }, + { 0xa658, 0xa658 }, + { 0xa65a, 0xa65a }, + { 0xa65c, 0xa65c }, + { 0xa65e, 0xa65e }, + { 0xa660, 0xa660 }, + { 0xa662, 0xa662 }, + { 0xa664, 0xa664 }, + { 0xa666, 0xa666 }, + { 0xa668, 0xa668 }, + { 0xa66a, 0xa66a }, + { 0xa66c, 0xa66c }, + { 0xa680, 0xa680 }, + { 0xa682, 0xa682 }, + { 0xa684, 0xa684 }, + { 0xa686, 0xa686 }, + { 0xa688, 0xa688 }, + { 0xa68a, 0xa68a }, + { 0xa68c, 0xa68c }, + { 0xa68e, 0xa68e }, + { 0xa690, 0xa690 }, + { 0xa692, 0xa692 }, + { 0xa694, 0xa694 }, + { 0xa696, 0xa696 }, + { 0xa698, 0xa698 }, + { 0xa69a, 0xa69a }, + { 0xa722, 0xa722 }, + { 0xa724, 0xa724 }, + { 0xa726, 0xa726 }, + { 0xa728, 0xa728 }, + { 0xa72a, 0xa72a }, + { 0xa72c, 0xa72c }, + { 0xa72e, 0xa72e }, + { 0xa732, 0xa732 }, + { 0xa734, 0xa734 }, + { 0xa736, 0xa736 }, + { 0xa738, 0xa738 }, + { 0xa73a, 0xa73a }, + { 0xa73c, 0xa73c }, + { 0xa73e, 0xa73e }, + { 0xa740, 0xa740 }, + { 0xa742, 0xa742 }, + { 0xa744, 0xa744 }, + { 0xa746, 0xa746 }, + { 0xa748, 0xa748 }, + { 0xa74a, 0xa74a }, + { 0xa74c, 0xa74c }, + { 0xa74e, 0xa74e }, + { 0xa750, 0xa750 }, + { 0xa752, 0xa752 }, + { 0xa754, 0xa754 }, + { 0xa756, 0xa756 }, + { 0xa758, 0xa758 }, + { 0xa75a, 0xa75a }, + { 0xa75c, 0xa75c }, + { 0xa75e, 0xa75e }, + { 0xa760, 0xa760 }, + { 0xa762, 0xa762 }, + { 0xa764, 0xa764 }, + { 0xa766, 0xa766 }, + { 0xa768, 0xa768 }, + { 0xa76a, 0xa76a }, + { 0xa76c, 0xa76c }, + { 0xa76e, 0xa76e }, + { 0xa779, 0xa779 }, + { 0xa77b, 0xa77b }, + { 0xa77d, 0xa77e }, + { 0xa780, 0xa780 }, + { 0xa782, 0xa782 }, + { 0xa784, 0xa784 }, + { 0xa786, 0xa786 }, + { 0xa78b, 0xa78b }, + { 0xa78d, 0xa78d }, + { 0xa790, 0xa790 }, + { 0xa792, 0xa792 }, + { 0xa796, 0xa796 }, + { 0xa798, 0xa798 }, + { 0xa79a, 0xa79a }, + { 0xa79c, 0xa79c }, + { 0xa79e, 0xa79e }, + { 0xa7a0, 0xa7a0 }, + { 0xa7a2, 0xa7a2 }, + { 0xa7a4, 0xa7a4 }, + { 0xa7a6, 0xa7a6 }, + { 0xa7a8, 0xa7a8 }, + { 0xa7aa, 0xa7ad }, + { 0xa7b0, 0xa7b4 }, + { 0xa7b6, 0xa7b6 }, + { 0xff21, 0xff3a }, + { 0x10400, 0x10427 }, + { 0x10c80, 0x10cb2 }, + { 0x118a0, 0x118bf }, + { 0x1d400, 0x1d419 }, + { 0x1d434, 0x1d44d }, + { 0x1d468, 0x1d481 }, + { 0x1d49c, 0x1d49c }, + { 0x1d49e, 0x1d49f }, + { 0x1d4a2, 0x1d4a2 }, + { 0x1d4a5, 0x1d4a6 }, + { 0x1d4a9, 0x1d4ac }, + { 0x1d4ae, 0x1d4b5 }, + { 0x1d4d0, 0x1d4e9 }, + { 0x1d504, 0x1d505 }, + { 0x1d507, 0x1d50a }, + { 0x1d50d, 0x1d514 }, + { 0x1d516, 0x1d51c }, + { 0x1d538, 0x1d539 }, + { 0x1d53b, 0x1d53e }, + { 0x1d540, 0x1d544 }, + { 0x1d546, 0x1d546 }, + { 0x1d54a, 0x1d550 }, + { 0x1d56c, 0x1d585 }, + { 0x1d5a0, 0x1d5b9 }, + { 0x1d5d4, 0x1d5ed }, + { 0x1d608, 0x1d621 }, + { 0x1d63c, 0x1d655 }, + { 0x1d670, 0x1d689 }, + { 0x1d6a8, 0x1d6c0 }, + { 0x1d6e2, 0x1d6fa }, + { 0x1d71c, 0x1d734 }, + { 0x1d756, 0x1d76e }, + { 0x1d790, 0x1d7a8 }, + { 0x1d7ca, 0x1d7ca } +}; + +extern int runeinrange(const void* a, const void* b); +bool isupperrune(Rune ch) { + return (NULL != bsearch(&ch, uppers, 625, 2 * sizeof(Rune), &runeinrange)); +} diff --git a/tools/unicode.rb b/tools/unicode.rb index 80bca9d..cdbecf1 100755 --- a/tools/unicode.rb +++ b/tools/unicode.rb @@ -1,5 +1,7 @@ #!/usr/bin/env ruby +require 'fileutils' + $types = {} # Map of of all rune types to lookup table name $typemap = { @@ -7,9 +9,9 @@ $typemap = { "Lu" => :uppers, # Upper Case "Ll" => :lowers, # Lower Case "Lt" => :titles, # Title Case - "LC" => :other_letters, # Cased Letter - "Lm" => :other_letters, # Modifier Letter - "Lo" => :other_letters, # Other Letter + "LC" => :otherletters, # Cased Letter + "Lm" => :otherletters, # Modifier Letter + "Lo" => :otherletters, # Other Letter # Combining Marks "Mn" => :marks, # Non-Spacing Mark "Mc" => :marks, # Spacing Mark @@ -48,6 +50,12 @@ def register_codepoint(type, val) $types[type] << val end +if ARGV.length != 2 then + puts "Usage: unicode.rb DATABASE OUTDIR" + puts "\nError: Incorrect number of arguments" + exit 1 +end + unicode_data = File.open(ARGV[0],"r") unicode_data.each_line do |data| fields = data.split(';') @@ -81,21 +89,31 @@ def get_ranges(table) end end +def generate_typecheck_func(type, count) + "extern int runeinrange(const void* a, const void* b);\n" + + "bool is#{type.to_s.gsub(/s$/,'')}rune(Rune ch) {\n" + + " return (NULL != bsearch(&ch, #{type}, #{count}, 2 * sizeof(Rune), &runeinrange));\n" + + "}\n" +end + def generate_type_table(type, altcase = []) ranges = get_ranges($types[type]) - pairs = ranges.map{|r| "0x#{r.first.to_s(16)}, 0x#{r.last.to_s(16)}" }.join(",\n ") - File.open("#{type.to_s}.c", "w") do |f| - f.puts("Rune #{type.to_s}[#{ranges.length*2}] = {") + pairs = ranges.map{|r| "{ 0x#{r.first.to_s(16)}, 0x#{r.last.to_s(16)} }" }.join(",\n ") + File.open("#{ARGV[1]}/#{type.to_s}.c", "w") do |f| + f.puts("#include \n\n") + f.puts("static Rune #{type.to_s}[#{ranges.length}][2] = {") f.print(' ') f.puts(pairs) - f.puts("};"); + f.print("};\n\n"); + f.print(generate_typecheck_func(type, ranges.length)) end end +FileUtils.mkdir_p ARGV[1] generate_type_table(:uppers, :tolowers) generate_type_table(:lowers, :touppers) generate_type_table(:titles) -generate_type_table(:other_letters) +generate_type_table(:otherletters) generate_type_table(:marks) generate_type_table(:numbers) generate_type_table(:punctuation)