diff --git a/example.cpp b/example.cpp index 6f5a6b0..980f20c 100644 --- a/example.cpp +++ b/example.cpp @@ -8,6 +8,8 @@ static void completion(const char *buf, linenoiseCompletions *lc) { if (buf[0] == 'h') { linenoiseAddCompletion(lc,"hello"); linenoiseAddCompletion(lc,"hello there"); + linenoiseAddCompletion(lc,"hello πŸ˜€"); + linenoiseAddCompletion(lc,"こんにけは"); } } @@ -17,6 +19,11 @@ static const char *hints(const char *buf, int *color, int *bold) { *bold = 0; return " World"; } + if (!strcasecmp(buf,"こんにけは")) { + *color = 35; + *bold = 0; + return " δΈ–η•Œ"; + } return NULL; } @@ -61,7 +68,7 @@ int main(int argc, char **argv) { while(1) { if (!async) { - line = linenoise("hello> "); + line = linenoise("πŸ˜€ \033[32mhello\x1b[0m> "); if (line == NULL) break; } else { /* Asynchronous mode using the multiplexing API: wait for @@ -116,7 +123,7 @@ int main(int argc, char **argv) { } else if (!strncmp(line, "/unmask", 7)) { linenoiseMaskModeDisable(); } else if (line[0] == '/') { - printf("Unreconized command: %s\n", line); + printf("Unrecognized command: %s\n", line); } free((void*) line); } diff --git a/linenoise.cpp b/linenoise.cpp index a68f12a..6e8ec44 100644 --- a/linenoise.cpp +++ b/linenoise.cpp @@ -227,6 +227,445 @@ static void lndebug(const char *, ...) { } #endif +/* ========================== Encoding functions ============================= */ + +/* Get length of previous UTF8 codepoint */ +static size_t prevUtf8CodePointLen(const char* buf, int pos) { + int end = pos--; + while (pos >= 0 && ((unsigned char)buf[pos] & 0xC0) == 0x80) + pos--; + return end - pos; +} + +/* Convert UTF8 to Unicode code point */ +static size_t utf8BytesToCodePoint(const char* buf, size_t len, int* cp) { + if (len) { + unsigned char byte = buf[0]; + if ((byte & 0x80) == 0) { + *cp = byte; + return 1; + } else if ((byte & 0xE0) == 0xC0) { + if (len >= 2) { + *cp = (((unsigned long)(buf[0] & 0x1F)) << 6) | + ((unsigned long)(buf[1] & 0x3F)); + return 2; + } + } else if ((byte & 0xF0) == 0xE0) { + if (len >= 3) { + *cp = (((unsigned long)(buf[0] & 0x0F)) << 12) | + (((unsigned long)(buf[1] & 0x3F)) << 6) | + ((unsigned long)(buf[2] & 0x3F)); + return 3; + } + } else if ((byte & 0xF8) == 0xF0) { + if (len >= 4) { + *cp = (((unsigned long)(buf[0] & 0x07)) << 18) | + (((unsigned long)(buf[1] & 0x3F)) << 12) | + (((unsigned long)(buf[2] & 0x3F)) << 6) | + ((unsigned long)(buf[3] & 0x3F)); + return 4; + } + } + } + return 0; +} + +/* Check if the code is a wide character */ +static const unsigned long wideCharTable[][2] = { + /* BEGIN: WIDE CHAR TABLE */ + { 0x1100, 0x115F }, { 0x231A, 0x231B }, { 0x2329, 0x232A }, { 0x23E9, 0x23EC }, + { 0x23F0, 0x23F0 }, { 0x23F3, 0x23F3 }, { 0x25FD, 0x25FE }, { 0x2614, 0x2615 }, + { 0x2630, 0x2637 }, { 0x2648, 0x2653 }, { 0x267F, 0x267F }, { 0x268A, 0x268F }, + { 0x2693, 0x2693 }, { 0x26A1, 0x26A1 }, { 0x26AA, 0x26AB }, { 0x26BD, 0x26BE }, + { 0x26C4, 0x26C5 }, { 0x26CE, 0x26CE }, { 0x26D4, 0x26D4 }, { 0x26EA, 0x26EA }, + { 0x26F2, 0x26F3 }, { 0x26F5, 0x26F5 }, { 0x26FA, 0x26FA }, { 0x26FD, 0x26FD }, + { 0x2705, 0x2705 }, { 0x270A, 0x270B }, { 0x2728, 0x2728 }, { 0x274C, 0x274C }, + { 0x274E, 0x274E }, { 0x2753, 0x2755 }, { 0x2757, 0x2757 }, { 0x2795, 0x2797 }, + { 0x27B0, 0x27B0 }, { 0x27BF, 0x27BF }, { 0x2B1B, 0x2B1C }, { 0x2B50, 0x2B50 }, + { 0x2B55, 0x2B55 }, { 0x2E80, 0x2E99 }, { 0x2E9B, 0x2EF3 }, { 0x2F00, 0x2FD5 }, + { 0x2FF0, 0x303E }, { 0x3041, 0x3096 }, { 0x3099, 0x30FF }, { 0x3105, 0x312F }, + { 0x3131, 0x318E }, { 0x3190, 0x31E5 }, { 0x31EF, 0x321E }, { 0x3220, 0x3247 }, + { 0x3250, 0xA48C }, { 0xA490, 0xA4C6 }, { 0xA960, 0xA97C }, { 0xAC00, 0xD7A3 }, + { 0xF900, 0xFAFF }, { 0xFE10, 0xFE19 }, { 0xFE30, 0xFE52 }, { 0xFE54, 0xFE66 }, + { 0xFE68, 0xFE6B }, { 0xFF01, 0xFF60 }, { 0xFFE0, 0xFFE6 }, { 0x16FE0, 0x16FE4 }, + { 0x16FF0, 0x16FF1 }, { 0x17000, 0x187F7 }, { 0x18800, 0x18CD5 }, { 0x18CFF, 0x18D08 }, + { 0x1AFF0, 0x1AFF3 }, { 0x1AFF5, 0x1AFFB }, { 0x1AFFD, 0x1AFFE }, { 0x1B000, 0x1B122 }, + { 0x1B132, 0x1B132 }, { 0x1B150, 0x1B152 }, { 0x1B155, 0x1B155 }, { 0x1B164, 0x1B167 }, + { 0x1B170, 0x1B2FB }, { 0x1D300, 0x1D356 }, { 0x1D360, 0x1D376 }, { 0x1F004, 0x1F004 }, + { 0x1F0CF, 0x1F0CF }, { 0x1F18E, 0x1F18E }, { 0x1F191, 0x1F19A }, { 0x1F200, 0x1F202 }, + { 0x1F210, 0x1F23B }, { 0x1F240, 0x1F248 }, { 0x1F250, 0x1F251 }, { 0x1F260, 0x1F265 }, + { 0x1F300, 0x1F320 }, { 0x1F32D, 0x1F335 }, { 0x1F337, 0x1F37C }, { 0x1F37E, 0x1F393 }, + { 0x1F3A0, 0x1F3CA }, { 0x1F3CF, 0x1F3D3 }, { 0x1F3E0, 0x1F3F0 }, { 0x1F3F4, 0x1F3F4 }, + { 0x1F3F8, 0x1F43E }, { 0x1F440, 0x1F440 }, { 0x1F442, 0x1F4FC }, { 0x1F4FF, 0x1F53D }, + { 0x1F54B, 0x1F54E }, { 0x1F550, 0x1F567 }, { 0x1F57A, 0x1F57A }, { 0x1F595, 0x1F596 }, + { 0x1F5A4, 0x1F5A4 }, { 0x1F5FB, 0x1F64F }, { 0x1F680, 0x1F6C5 }, { 0x1F6CC, 0x1F6CC }, + { 0x1F6D0, 0x1F6D2 }, { 0x1F6D5, 0x1F6D7 }, { 0x1F6DC, 0x1F6DF }, { 0x1F6EB, 0x1F6EC }, + { 0x1F6F4, 0x1F6FC }, { 0x1F7E0, 0x1F7EB }, { 0x1F7F0, 0x1F7F0 }, { 0x1F90C, 0x1F93A }, + { 0x1F93C, 0x1F945 }, { 0x1F947, 0x1F9FF }, { 0x1FA70, 0x1FA7C }, { 0x1FA80, 0x1FA89 }, + { 0x1FA8F, 0x1FAC6 }, { 0x1FACE, 0x1FADC }, { 0x1FADF, 0x1FAE9 }, { 0x1FAF0, 0x1FAF8 }, + { 0x20000, 0x2FFFD }, { 0x30000, 0x3FFFD } + /* END: WIDE CHAR TABLE */ +}; + +static const size_t wideCharTableSize = + sizeof(wideCharTable) / sizeof(wideCharTable[0]); + +static int isWideChar(unsigned long cp) { + size_t i; + for (i = 0; i < wideCharTableSize; i++) + if (wideCharTable[i][0] <= cp && cp <= wideCharTable[i][1]) return 1; + return 0; +} + +/* Check if the code is a combining character */ +static const unsigned long combiningCharTable[] = { + /* BEGIN: COMBINING CHAR TABLE */ + 0x0300, 0x0301, 0x0302, 0x0303, 0x0304, 0x0305, 0x0306, 0x0307, + 0x0308, 0x0309, 0x030A, 0x030B, 0x030C, 0x030D, 0x030E, 0x030F, + 0x0310, 0x0311, 0x0312, 0x0313, 0x0314, 0x0315, 0x0316, 0x0317, + 0x0318, 0x0319, 0x031A, 0x031B, 0x031C, 0x031D, 0x031E, 0x031F, + 0x0320, 0x0321, 0x0322, 0x0323, 0x0324, 0x0325, 0x0326, 0x0327, + 0x0328, 0x0329, 0x032A, 0x032B, 0x032C, 0x032D, 0x032E, 0x032F, + 0x0330, 0x0331, 0x0332, 0x0333, 0x0334, 0x0335, 0x0336, 0x0337, + 0x0338, 0x0339, 0x033A, 0x033B, 0x033C, 0x033D, 0x033E, 0x033F, + 0x0340, 0x0341, 0x0342, 0x0343, 0x0344, 0x0345, 0x0346, 0x0347, + 0x0348, 0x0349, 0x034A, 0x034B, 0x034C, 0x034D, 0x034E, 0x034F, + 0x0350, 0x0351, 0x0352, 0x0353, 0x0354, 0x0355, 0x0356, 0x0357, + 0x0358, 0x0359, 0x035A, 0x035B, 0x035C, 0x035D, 0x035E, 0x035F, + 0x0360, 0x0361, 0x0362, 0x0363, 0x0364, 0x0365, 0x0366, 0x0367, + 0x0368, 0x0369, 0x036A, 0x036B, 0x036C, 0x036D, 0x036E, 0x036F, + 0x0483, 0x0484, 0x0485, 0x0486, 0x0487, 0x0591, 0x0592, 0x0593, + 0x0594, 0x0595, 0x0596, 0x0597, 0x0598, 0x0599, 0x059A, 0x059B, + 0x059C, 0x059D, 0x059E, 0x059F, 0x05A0, 0x05A1, 0x05A2, 0x05A3, + 0x05A4, 0x05A5, 0x05A6, 0x05A7, 0x05A8, 0x05A9, 0x05AA, 0x05AB, + 0x05AC, 0x05AD, 0x05AE, 0x05AF, 0x05B0, 0x05B1, 0x05B2, 0x05B3, + 0x05B4, 0x05B5, 0x05B6, 0x05B7, 0x05B8, 0x05B9, 0x05BA, 0x05BB, + 0x05BC, 0x05BD, 0x05BF, 0x05C1, 0x05C2, 0x05C4, 0x05C5, 0x05C7, + 0x0610, 0x0611, 0x0612, 0x0613, 0x0614, 0x0615, 0x0616, 0x0617, + 0x0618, 0x0619, 0x061A, 0x064B, 0x064C, 0x064D, 0x064E, 0x064F, + 0x0650, 0x0651, 0x0652, 0x0653, 0x0654, 0x0655, 0x0656, 0x0657, + 0x0658, 0x0659, 0x065A, 0x065B, 0x065C, 0x065D, 0x065E, 0x065F, + 0x0670, 0x06D6, 0x06D7, 0x06D8, 0x06D9, 0x06DA, 0x06DB, 0x06DC, + 0x06DF, 0x06E0, 0x06E1, 0x06E2, 0x06E3, 0x06E4, 0x06E7, 0x06E8, + 0x06EA, 0x06EB, 0x06EC, 0x06ED, 0x0711, 0x0730, 0x0731, 0x0732, + 0x0733, 0x0734, 0x0735, 0x0736, 0x0737, 0x0738, 0x0739, 0x073A, + 0x073B, 0x073C, 0x073D, 0x073E, 0x073F, 0x0740, 0x0741, 0x0742, + 0x0743, 0x0744, 0x0745, 0x0746, 0x0747, 0x0748, 0x0749, 0x074A, + 0x07A6, 0x07A7, 0x07A8, 0x07A9, 0x07AA, 0x07AB, 0x07AC, 0x07AD, + 0x07AE, 0x07AF, 0x07B0, 0x07EB, 0x07EC, 0x07ED, 0x07EE, 0x07EF, + 0x07F0, 0x07F1, 0x07F2, 0x07F3, 0x07FD, 0x0816, 0x0817, 0x0818, + 0x0819, 0x081B, 0x081C, 0x081D, 0x081E, 0x081F, 0x0820, 0x0821, + 0x0822, 0x0823, 0x0825, 0x0826, 0x0827, 0x0829, 0x082A, 0x082B, + 0x082C, 0x082D, 0x0859, 0x085A, 0x085B, 0x0897, 0x0898, 0x0899, + 0x089A, 0x089B, 0x089C, 0x089D, 0x089E, 0x089F, 0x08CA, 0x08CB, + 0x08CC, 0x08CD, 0x08CE, 0x08CF, 0x08D0, 0x08D1, 0x08D2, 0x08D3, + 0x08D4, 0x08D5, 0x08D6, 0x08D7, 0x08D8, 0x08D9, 0x08DA, 0x08DB, + 0x08DC, 0x08DD, 0x08DE, 0x08DF, 0x08E0, 0x08E1, 0x08E3, 0x08E4, + 0x08E5, 0x08E6, 0x08E7, 0x08E8, 0x08E9, 0x08EA, 0x08EB, 0x08EC, + 0x08ED, 0x08EE, 0x08EF, 0x08F0, 0x08F1, 0x08F2, 0x08F3, 0x08F4, + 0x08F5, 0x08F6, 0x08F7, 0x08F8, 0x08F9, 0x08FA, 0x08FB, 0x08FC, + 0x08FD, 0x08FE, 0x08FF, 0x0900, 0x0901, 0x0902, 0x093A, 0x093C, + 0x0941, 0x0942, 0x0943, 0x0944, 0x0945, 0x0946, 0x0947, 0x0948, + 0x094D, 0x0951, 0x0952, 0x0953, 0x0954, 0x0955, 0x0956, 0x0957, + 0x0962, 0x0963, 0x0981, 0x09BC, 0x09C1, 0x09C2, 0x09C3, 0x09C4, + 0x09CD, 0x09E2, 0x09E3, 0x09FE, 0x0A01, 0x0A02, 0x0A3C, 0x0A41, + 0x0A42, 0x0A47, 0x0A48, 0x0A4B, 0x0A4C, 0x0A4D, 0x0A51, 0x0A70, + 0x0A71, 0x0A75, 0x0A81, 0x0A82, 0x0ABC, 0x0AC1, 0x0AC2, 0x0AC3, + 0x0AC4, 0x0AC5, 0x0AC7, 0x0AC8, 0x0ACD, 0x0AE2, 0x0AE3, 0x0AFA, + 0x0AFB, 0x0AFC, 0x0AFD, 0x0AFE, 0x0AFF, 0x0B01, 0x0B3C, 0x0B3F, + 0x0B41, 0x0B42, 0x0B43, 0x0B44, 0x0B4D, 0x0B55, 0x0B56, 0x0B62, + 0x0B63, 0x0B82, 0x0BC0, 0x0BCD, 0x0C00, 0x0C04, 0x0C3C, 0x0C3E, + 0x0C3F, 0x0C40, 0x0C46, 0x0C47, 0x0C48, 0x0C4A, 0x0C4B, 0x0C4C, + 0x0C4D, 0x0C55, 0x0C56, 0x0C62, 0x0C63, 0x0C81, 0x0CBC, 0x0CBF, + 0x0CC6, 0x0CCC, 0x0CCD, 0x0CE2, 0x0CE3, 0x0D00, 0x0D01, 0x0D3B, + 0x0D3C, 0x0D41, 0x0D42, 0x0D43, 0x0D44, 0x0D4D, 0x0D62, 0x0D63, + 0x0D81, 0x0DCA, 0x0DD2, 0x0DD3, 0x0DD4, 0x0DD6, 0x0E31, 0x0E34, + 0x0E35, 0x0E36, 0x0E37, 0x0E38, 0x0E39, 0x0E3A, 0x0E47, 0x0E48, + 0x0E49, 0x0E4A, 0x0E4B, 0x0E4C, 0x0E4D, 0x0E4E, 0x0EB1, 0x0EB4, + 0x0EB5, 0x0EB6, 0x0EB7, 0x0EB8, 0x0EB9, 0x0EBA, 0x0EBB, 0x0EBC, + 0x0EC8, 0x0EC9, 0x0ECA, 0x0ECB, 0x0ECC, 0x0ECD, 0x0ECE, 0x0F18, + 0x0F19, 0x0F35, 0x0F37, 0x0F39, 0x0F71, 0x0F72, 0x0F73, 0x0F74, + 0x0F75, 0x0F76, 0x0F77, 0x0F78, 0x0F79, 0x0F7A, 0x0F7B, 0x0F7C, + 0x0F7D, 0x0F7E, 0x0F80, 0x0F81, 0x0F82, 0x0F83, 0x0F84, 0x0F86, + 0x0F87, 0x0F8D, 0x0F8E, 0x0F8F, 0x0F90, 0x0F91, 0x0F92, 0x0F93, + 0x0F94, 0x0F95, 0x0F96, 0x0F97, 0x0F99, 0x0F9A, 0x0F9B, 0x0F9C, + 0x0F9D, 0x0F9E, 0x0F9F, 0x0FA0, 0x0FA1, 0x0FA2, 0x0FA3, 0x0FA4, + 0x0FA5, 0x0FA6, 0x0FA7, 0x0FA8, 0x0FA9, 0x0FAA, 0x0FAB, 0x0FAC, + 0x0FAD, 0x0FAE, 0x0FAF, 0x0FB0, 0x0FB1, 0x0FB2, 0x0FB3, 0x0FB4, + 0x0FB5, 0x0FB6, 0x0FB7, 0x0FB8, 0x0FB9, 0x0FBA, 0x0FBB, 0x0FBC, + 0x0FC6, 0x102D, 0x102E, 0x102F, 0x1030, 0x1032, 0x1033, 0x1034, + 0x1035, 0x1036, 0x1037, 0x1039, 0x103A, 0x103D, 0x103E, 0x1058, + 0x1059, 0x105E, 0x105F, 0x1060, 0x1071, 0x1072, 0x1073, 0x1074, + 0x1082, 0x1085, 0x1086, 0x108D, 0x109D, 0x135D, 0x135E, 0x135F, + 0x1712, 0x1713, 0x1714, 0x1732, 0x1733, 0x1752, 0x1753, 0x1772, + 0x1773, 0x17B4, 0x17B5, 0x17B7, 0x17B8, 0x17B9, 0x17BA, 0x17BB, + 0x17BC, 0x17BD, 0x17C6, 0x17C9, 0x17CA, 0x17CB, 0x17CC, 0x17CD, + 0x17CE, 0x17CF, 0x17D0, 0x17D1, 0x17D2, 0x17D3, 0x17DD, 0x180B, + 0x180C, 0x180D, 0x180F, 0x1885, 0x1886, 0x18A9, 0x1920, 0x1921, + 0x1922, 0x1927, 0x1928, 0x1932, 0x1939, 0x193A, 0x193B, 0x1A17, + 0x1A18, 0x1A1B, 0x1A56, 0x1A58, 0x1A59, 0x1A5A, 0x1A5B, 0x1A5C, + 0x1A5D, 0x1A5E, 0x1A60, 0x1A62, 0x1A65, 0x1A66, 0x1A67, 0x1A68, + 0x1A69, 0x1A6A, 0x1A6B, 0x1A6C, 0x1A73, 0x1A74, 0x1A75, 0x1A76, + 0x1A77, 0x1A78, 0x1A79, 0x1A7A, 0x1A7B, 0x1A7C, 0x1A7F, 0x1AB0, + 0x1AB1, 0x1AB2, 0x1AB3, 0x1AB4, 0x1AB5, 0x1AB6, 0x1AB7, 0x1AB8, + 0x1AB9, 0x1ABA, 0x1ABB, 0x1ABC, 0x1ABD, 0x1ABF, 0x1AC0, 0x1AC1, + 0x1AC2, 0x1AC3, 0x1AC4, 0x1AC5, 0x1AC6, 0x1AC7, 0x1AC8, 0x1AC9, + 0x1ACA, 0x1ACB, 0x1ACC, 0x1ACD, 0x1ACE, 0x1B00, 0x1B01, 0x1B02, + 0x1B03, 0x1B34, 0x1B36, 0x1B37, 0x1B38, 0x1B39, 0x1B3A, 0x1B3C, + 0x1B42, 0x1B6B, 0x1B6C, 0x1B6D, 0x1B6E, 0x1B6F, 0x1B70, 0x1B71, + 0x1B72, 0x1B73, 0x1B80, 0x1B81, 0x1BA2, 0x1BA3, 0x1BA4, 0x1BA5, + 0x1BA8, 0x1BA9, 0x1BAB, 0x1BAC, 0x1BAD, 0x1BE6, 0x1BE8, 0x1BE9, + 0x1BED, 0x1BEF, 0x1BF0, 0x1BF1, 0x1C2C, 0x1C2D, 0x1C2E, 0x1C2F, + 0x1C30, 0x1C31, 0x1C32, 0x1C33, 0x1C36, 0x1C37, 0x1CD0, 0x1CD1, + 0x1CD2, 0x1CD4, 0x1CD5, 0x1CD6, 0x1CD7, 0x1CD8, 0x1CD9, 0x1CDA, + 0x1CDB, 0x1CDC, 0x1CDD, 0x1CDE, 0x1CDF, 0x1CE0, 0x1CE2, 0x1CE3, + 0x1CE4, 0x1CE5, 0x1CE6, 0x1CE7, 0x1CE8, 0x1CED, 0x1CF4, 0x1CF8, + 0x1CF9, 0x1DC0, 0x1DC1, 0x1DC2, 0x1DC3, 0x1DC4, 0x1DC5, 0x1DC6, + 0x1DC7, 0x1DC8, 0x1DC9, 0x1DCA, 0x1DCB, 0x1DCC, 0x1DCD, 0x1DCE, + 0x1DCF, 0x1DD0, 0x1DD1, 0x1DD2, 0x1DD3, 0x1DD4, 0x1DD5, 0x1DD6, + 0x1DD7, 0x1DD8, 0x1DD9, 0x1DDA, 0x1DDB, 0x1DDC, 0x1DDD, 0x1DDE, + 0x1DDF, 0x1DE0, 0x1DE1, 0x1DE2, 0x1DE3, 0x1DE4, 0x1DE5, 0x1DE6, + 0x1DE7, 0x1DE8, 0x1DE9, 0x1DEA, 0x1DEB, 0x1DEC, 0x1DED, 0x1DEE, + 0x1DEF, 0x1DF0, 0x1DF1, 0x1DF2, 0x1DF3, 0x1DF4, 0x1DF5, 0x1DF6, + 0x1DF7, 0x1DF8, 0x1DF9, 0x1DFA, 0x1DFB, 0x1DFC, 0x1DFD, 0x1DFE, + 0x1DFF, 0x20D0, 0x20D1, 0x20D2, 0x20D3, 0x20D4, 0x20D5, 0x20D6, + 0x20D7, 0x20D8, 0x20D9, 0x20DA, 0x20DB, 0x20DC, 0x20E1, 0x20E5, + 0x20E6, 0x20E7, 0x20E8, 0x20E9, 0x20EA, 0x20EB, 0x20EC, 0x20ED, + 0x20EE, 0x20EF, 0x20F0, 0x2CEF, 0x2CF0, 0x2CF1, 0x2D7F, 0x2DE0, + 0x2DE1, 0x2DE2, 0x2DE3, 0x2DE4, 0x2DE5, 0x2DE6, 0x2DE7, 0x2DE8, + 0x2DE9, 0x2DEA, 0x2DEB, 0x2DEC, 0x2DED, 0x2DEE, 0x2DEF, 0x2DF0, + 0x2DF1, 0x2DF2, 0x2DF3, 0x2DF4, 0x2DF5, 0x2DF6, 0x2DF7, 0x2DF8, + 0x2DF9, 0x2DFA, 0x2DFB, 0x2DFC, 0x2DFD, 0x2DFE, 0x2DFF, 0x302A, + 0x302B, 0x302C, 0x302D, 0x3099, 0x309A, 0xA66F, 0xA674, 0xA675, + 0xA676, 0xA677, 0xA678, 0xA679, 0xA67A, 0xA67B, 0xA67C, 0xA67D, + 0xA69E, 0xA69F, 0xA6F0, 0xA6F1, 0xA802, 0xA806, 0xA80B, 0xA825, + 0xA826, 0xA82C, 0xA8C4, 0xA8C5, 0xA8E0, 0xA8E1, 0xA8E2, 0xA8E3, + 0xA8E4, 0xA8E5, 0xA8E6, 0xA8E7, 0xA8E8, 0xA8E9, 0xA8EA, 0xA8EB, + 0xA8EC, 0xA8ED, 0xA8EE, 0xA8EF, 0xA8F0, 0xA8F1, 0xA8FF, 0xA926, + 0xA927, 0xA928, 0xA929, 0xA92A, 0xA92B, 0xA92C, 0xA92D, 0xA947, + 0xA948, 0xA949, 0xA94A, 0xA94B, 0xA94C, 0xA94D, 0xA94E, 0xA94F, + 0xA950, 0xA951, 0xA980, 0xA981, 0xA982, 0xA9B3, 0xA9B6, 0xA9B7, + 0xA9B8, 0xA9B9, 0xA9BC, 0xA9BD, 0xA9E5, 0xAA29, 0xAA2A, 0xAA2B, + 0xAA2C, 0xAA2D, 0xAA2E, 0xAA31, 0xAA32, 0xAA35, 0xAA36, 0xAA43, + 0xAA4C, 0xAA7C, 0xAAB0, 0xAAB2, 0xAAB3, 0xAAB4, 0xAAB7, 0xAAB8, + 0xAABE, 0xAABF, 0xAAC1, 0xAAEC, 0xAAED, 0xAAF6, 0xABE5, 0xABE8, + 0xABED, 0xFB1E, 0xFE00, 0xFE01, 0xFE02, 0xFE03, 0xFE04, 0xFE05, + 0xFE06, 0xFE07, 0xFE08, 0xFE09, 0xFE0A, 0xFE0B, 0xFE0C, 0xFE0D, + 0xFE0E, 0xFE0F, 0xFE20, 0xFE21, 0xFE22, 0xFE23, 0xFE24, 0xFE25, + 0xFE26, 0xFE27, 0xFE28, 0xFE29, 0xFE2A, 0xFE2B, 0xFE2C, 0xFE2D, + 0xFE2E, 0xFE2F, 0x101FD, 0x102E0, 0x10376, 0x10377, 0x10378, 0x10379, + 0x1037A, 0x10A01, 0x10A02, 0x10A03, 0x10A05, 0x10A06, 0x10A0C, 0x10A0D, + 0x10A0E, 0x10A0F, 0x10A38, 0x10A39, 0x10A3A, 0x10A3F, 0x10AE5, 0x10AE6, + 0x10D24, 0x10D25, 0x10D26, 0x10D27, 0x10D69, 0x10D6A, 0x10D6B, 0x10D6C, + 0x10D6D, 0x10EAB, 0x10EAC, 0x10EFC, 0x10EFD, 0x10EFE, 0x10EFF, 0x10F46, + 0x10F47, 0x10F48, 0x10F49, 0x10F4A, 0x10F4B, 0x10F4C, 0x10F4D, 0x10F4E, + 0x10F4F, 0x10F50, 0x10F82, 0x10F83, 0x10F84, 0x10F85, 0x11001, 0x11038, + 0x11039, 0x1103A, 0x1103B, 0x1103C, 0x1103D, 0x1103E, 0x1103F, 0x11040, + 0x11041, 0x11042, 0x11043, 0x11044, 0x11045, 0x11046, 0x11070, 0x11073, + 0x11074, 0x1107F, 0x11080, 0x11081, 0x110B3, 0x110B4, 0x110B5, 0x110B6, + 0x110B9, 0x110BA, 0x110C2, 0x11100, 0x11101, 0x11102, 0x11127, 0x11128, + 0x11129, 0x1112A, 0x1112B, 0x1112D, 0x1112E, 0x1112F, 0x11130, 0x11131, + 0x11132, 0x11133, 0x11134, 0x11173, 0x11180, 0x11181, 0x111B6, 0x111B7, + 0x111B8, 0x111B9, 0x111BA, 0x111BB, 0x111BC, 0x111BD, 0x111BE, 0x111C9, + 0x111CA, 0x111CB, 0x111CC, 0x111CF, 0x1122F, 0x11230, 0x11231, 0x11234, + 0x11236, 0x11237, 0x1123E, 0x11241, 0x112DF, 0x112E3, 0x112E4, 0x112E5, + 0x112E6, 0x112E7, 0x112E8, 0x112E9, 0x112EA, 0x11300, 0x11301, 0x1133B, + 0x1133C, 0x11340, 0x11366, 0x11367, 0x11368, 0x11369, 0x1136A, 0x1136B, + 0x1136C, 0x11370, 0x11371, 0x11372, 0x11373, 0x11374, 0x113BB, 0x113BC, + 0x113BD, 0x113BE, 0x113BF, 0x113C0, 0x113CE, 0x113D0, 0x113D2, 0x113E1, + 0x113E2, 0x11438, 0x11439, 0x1143A, 0x1143B, 0x1143C, 0x1143D, 0x1143E, + 0x1143F, 0x11442, 0x11443, 0x11444, 0x11446, 0x1145E, 0x114B3, 0x114B4, + 0x114B5, 0x114B6, 0x114B7, 0x114B8, 0x114BA, 0x114BF, 0x114C0, 0x114C2, + 0x114C3, 0x115B2, 0x115B3, 0x115B4, 0x115B5, 0x115BC, 0x115BD, 0x115BF, + 0x115C0, 0x115DC, 0x115DD, 0x11633, 0x11634, 0x11635, 0x11636, 0x11637, + 0x11638, 0x11639, 0x1163A, 0x1163D, 0x1163F, 0x11640, 0x116AB, 0x116AD, + 0x116B0, 0x116B1, 0x116B2, 0x116B3, 0x116B4, 0x116B5, 0x116B7, 0x1171D, + 0x1171F, 0x11722, 0x11723, 0x11724, 0x11725, 0x11727, 0x11728, 0x11729, + 0x1172A, 0x1172B, 0x1182F, 0x11830, 0x11831, 0x11832, 0x11833, 0x11834, + 0x11835, 0x11836, 0x11837, 0x11839, 0x1183A, 0x1193B, 0x1193C, 0x1193E, + 0x11943, 0x119D4, 0x119D5, 0x119D6, 0x119D7, 0x119DA, 0x119DB, 0x119E0, + 0x11A01, 0x11A02, 0x11A03, 0x11A04, 0x11A05, 0x11A06, 0x11A07, 0x11A08, + 0x11A09, 0x11A0A, 0x11A33, 0x11A34, 0x11A35, 0x11A36, 0x11A37, 0x11A38, + 0x11A3B, 0x11A3C, 0x11A3D, 0x11A3E, 0x11A47, 0x11A51, 0x11A52, 0x11A53, + 0x11A54, 0x11A55, 0x11A56, 0x11A59, 0x11A5A, 0x11A5B, 0x11A8A, 0x11A8B, + 0x11A8C, 0x11A8D, 0x11A8E, 0x11A8F, 0x11A90, 0x11A91, 0x11A92, 0x11A93, + 0x11A94, 0x11A95, 0x11A96, 0x11A98, 0x11A99, 0x11C30, 0x11C31, 0x11C32, + 0x11C33, 0x11C34, 0x11C35, 0x11C36, 0x11C38, 0x11C39, 0x11C3A, 0x11C3B, + 0x11C3C, 0x11C3D, 0x11C3F, 0x11C92, 0x11C93, 0x11C94, 0x11C95, 0x11C96, + 0x11C97, 0x11C98, 0x11C99, 0x11C9A, 0x11C9B, 0x11C9C, 0x11C9D, 0x11C9E, + 0x11C9F, 0x11CA0, 0x11CA1, 0x11CA2, 0x11CA3, 0x11CA4, 0x11CA5, 0x11CA6, + 0x11CA7, 0x11CAA, 0x11CAB, 0x11CAC, 0x11CAD, 0x11CAE, 0x11CAF, 0x11CB0, + 0x11CB2, 0x11CB3, 0x11CB5, 0x11CB6, 0x11D31, 0x11D32, 0x11D33, 0x11D34, + 0x11D35, 0x11D36, 0x11D3A, 0x11D3C, 0x11D3D, 0x11D3F, 0x11D40, 0x11D41, + 0x11D42, 0x11D43, 0x11D44, 0x11D45, 0x11D47, 0x11D90, 0x11D91, 0x11D95, + 0x11D97, 0x11EF3, 0x11EF4, 0x11F00, 0x11F01, 0x11F36, 0x11F37, 0x11F38, + 0x11F39, 0x11F3A, 0x11F40, 0x11F42, 0x11F5A, 0x13440, 0x13447, 0x13448, + 0x13449, 0x1344A, 0x1344B, 0x1344C, 0x1344D, 0x1344E, 0x1344F, 0x13450, + 0x13451, 0x13452, 0x13453, 0x13454, 0x13455, 0x1611E, 0x1611F, 0x16120, + 0x16121, 0x16122, 0x16123, 0x16124, 0x16125, 0x16126, 0x16127, 0x16128, + 0x16129, 0x1612D, 0x1612E, 0x1612F, 0x16AF0, 0x16AF1, 0x16AF2, 0x16AF3, + 0x16AF4, 0x16B30, 0x16B31, 0x16B32, 0x16B33, 0x16B34, 0x16B35, 0x16B36, + 0x16F4F, 0x16F8F, 0x16F90, 0x16F91, 0x16F92, 0x16FE4, 0x1BC9D, 0x1BC9E, + 0x1CF00, 0x1CF01, 0x1CF02, 0x1CF03, 0x1CF04, 0x1CF05, 0x1CF06, 0x1CF07, + 0x1CF08, 0x1CF09, 0x1CF0A, 0x1CF0B, 0x1CF0C, 0x1CF0D, 0x1CF0E, 0x1CF0F, + 0x1CF10, 0x1CF11, 0x1CF12, 0x1CF13, 0x1CF14, 0x1CF15, 0x1CF16, 0x1CF17, + 0x1CF18, 0x1CF19, 0x1CF1A, 0x1CF1B, 0x1CF1C, 0x1CF1D, 0x1CF1E, 0x1CF1F, + 0x1CF20, 0x1CF21, 0x1CF22, 0x1CF23, 0x1CF24, 0x1CF25, 0x1CF26, 0x1CF27, + 0x1CF28, 0x1CF29, 0x1CF2A, 0x1CF2B, 0x1CF2C, 0x1CF2D, 0x1CF30, 0x1CF31, + 0x1CF32, 0x1CF33, 0x1CF34, 0x1CF35, 0x1CF36, 0x1CF37, 0x1CF38, 0x1CF39, + 0x1CF3A, 0x1CF3B, 0x1CF3C, 0x1CF3D, 0x1CF3E, 0x1CF3F, 0x1CF40, 0x1CF41, + 0x1CF42, 0x1CF43, 0x1CF44, 0x1CF45, 0x1CF46, 0x1D167, 0x1D168, 0x1D169, + 0x1D17B, 0x1D17C, 0x1D17D, 0x1D17E, 0x1D17F, 0x1D180, 0x1D181, 0x1D182, + 0x1D185, 0x1D186, 0x1D187, 0x1D188, 0x1D189, 0x1D18A, 0x1D18B, 0x1D1AA, + 0x1D1AB, 0x1D1AC, 0x1D1AD, 0x1D242, 0x1D243, 0x1D244, 0x1DA00, 0x1DA01, + 0x1DA02, 0x1DA03, 0x1DA04, 0x1DA05, 0x1DA06, 0x1DA07, 0x1DA08, 0x1DA09, + 0x1DA0A, 0x1DA0B, 0x1DA0C, 0x1DA0D, 0x1DA0E, 0x1DA0F, 0x1DA10, 0x1DA11, + 0x1DA12, 0x1DA13, 0x1DA14, 0x1DA15, 0x1DA16, 0x1DA17, 0x1DA18, 0x1DA19, + 0x1DA1A, 0x1DA1B, 0x1DA1C, 0x1DA1D, 0x1DA1E, 0x1DA1F, 0x1DA20, 0x1DA21, + 0x1DA22, 0x1DA23, 0x1DA24, 0x1DA25, 0x1DA26, 0x1DA27, 0x1DA28, 0x1DA29, + 0x1DA2A, 0x1DA2B, 0x1DA2C, 0x1DA2D, 0x1DA2E, 0x1DA2F, 0x1DA30, 0x1DA31, + 0x1DA32, 0x1DA33, 0x1DA34, 0x1DA35, 0x1DA36, 0x1DA3B, 0x1DA3C, 0x1DA3D, + 0x1DA3E, 0x1DA3F, 0x1DA40, 0x1DA41, 0x1DA42, 0x1DA43, 0x1DA44, 0x1DA45, + 0x1DA46, 0x1DA47, 0x1DA48, 0x1DA49, 0x1DA4A, 0x1DA4B, 0x1DA4C, 0x1DA4D, + 0x1DA4E, 0x1DA4F, 0x1DA50, 0x1DA51, 0x1DA52, 0x1DA53, 0x1DA54, 0x1DA55, + 0x1DA56, 0x1DA57, 0x1DA58, 0x1DA59, 0x1DA5A, 0x1DA5B, 0x1DA5C, 0x1DA5D, + 0x1DA5E, 0x1DA5F, 0x1DA60, 0x1DA61, 0x1DA62, 0x1DA63, 0x1DA64, 0x1DA65, + 0x1DA66, 0x1DA67, 0x1DA68, 0x1DA69, 0x1DA6A, 0x1DA6B, 0x1DA6C, 0x1DA75, + 0x1DA84, 0x1DA9B, 0x1DA9C, 0x1DA9D, 0x1DA9E, 0x1DA9F, 0x1DAA1, 0x1DAA2, + 0x1DAA3, 0x1DAA4, 0x1DAA5, 0x1DAA6, 0x1DAA7, 0x1DAA8, 0x1DAA9, 0x1DAAA, + 0x1DAAB, 0x1DAAC, 0x1DAAD, 0x1DAAE, 0x1DAAF, 0x1E000, 0x1E001, 0x1E002, + 0x1E003, 0x1E004, 0x1E005, 0x1E006, 0x1E008, 0x1E009, 0x1E00A, 0x1E00B, + 0x1E00C, 0x1E00D, 0x1E00E, 0x1E00F, 0x1E010, 0x1E011, 0x1E012, 0x1E013, + 0x1E014, 0x1E015, 0x1E016, 0x1E017, 0x1E018, 0x1E01B, 0x1E01C, 0x1E01D, + 0x1E01E, 0x1E01F, 0x1E020, 0x1E021, 0x1E023, 0x1E024, 0x1E026, 0x1E027, + 0x1E028, 0x1E029, 0x1E02A, 0x1E08F, 0x1E130, 0x1E131, 0x1E132, 0x1E133, + 0x1E134, 0x1E135, 0x1E136, 0x1E2AE, 0x1E2EC, 0x1E2ED, 0x1E2EE, 0x1E2EF, + 0x1E4EC, 0x1E4ED, 0x1E4EE, 0x1E4EF, 0x1E5EE, 0x1E5EF, 0x1E8D0, 0x1E8D1, + 0x1E8D2, 0x1E8D3, 0x1E8D4, 0x1E8D5, 0x1E8D6, 0x1E944, 0x1E945, 0x1E946, + 0x1E947, 0x1E948, 0x1E949, 0x1E94A, 0xE0100, 0xE0101, 0xE0102, 0xE0103, + 0xE0104, 0xE0105, 0xE0106, 0xE0107, 0xE0108, 0xE0109, 0xE010A, 0xE010B, + 0xE010C, 0xE010D, 0xE010E, 0xE010F, 0xE0110, 0xE0111, 0xE0112, 0xE0113, + 0xE0114, 0xE0115, 0xE0116, 0xE0117, 0xE0118, 0xE0119, 0xE011A, 0xE011B, + 0xE011C, 0xE011D, 0xE011E, 0xE011F, 0xE0120, 0xE0121, 0xE0122, 0xE0123, + 0xE0124, 0xE0125, 0xE0126, 0xE0127, 0xE0128, 0xE0129, 0xE012A, 0xE012B, + 0xE012C, 0xE012D, 0xE012E, 0xE012F, 0xE0130, 0xE0131, 0xE0132, 0xE0133, + 0xE0134, 0xE0135, 0xE0136, 0xE0137, 0xE0138, 0xE0139, 0xE013A, 0xE013B, + 0xE013C, 0xE013D, 0xE013E, 0xE013F, 0xE0140, 0xE0141, 0xE0142, 0xE0143, + 0xE0144, 0xE0145, 0xE0146, 0xE0147, 0xE0148, 0xE0149, 0xE014A, 0xE014B, + 0xE014C, 0xE014D, 0xE014E, 0xE014F, 0xE0150, 0xE0151, 0xE0152, 0xE0153, + 0xE0154, 0xE0155, 0xE0156, 0xE0157, 0xE0158, 0xE0159, 0xE015A, 0xE015B, + 0xE015C, 0xE015D, 0xE015E, 0xE015F, 0xE0160, 0xE0161, 0xE0162, 0xE0163, + 0xE0164, 0xE0165, 0xE0166, 0xE0167, 0xE0168, 0xE0169, 0xE016A, 0xE016B, + 0xE016C, 0xE016D, 0xE016E, 0xE016F, 0xE0170, 0xE0171, 0xE0172, 0xE0173, + 0xE0174, 0xE0175, 0xE0176, 0xE0177, 0xE0178, 0xE0179, 0xE017A, 0xE017B, + 0xE017C, 0xE017D, 0xE017E, 0xE017F, 0xE0180, 0xE0181, 0xE0182, 0xE0183, + 0xE0184, 0xE0185, 0xE0186, 0xE0187, 0xE0188, 0xE0189, 0xE018A, 0xE018B, + 0xE018C, 0xE018D, 0xE018E, 0xE018F, 0xE0190, 0xE0191, 0xE0192, 0xE0193, + 0xE0194, 0xE0195, 0xE0196, 0xE0197, 0xE0198, 0xE0199, 0xE019A, 0xE019B, + 0xE019C, 0xE019D, 0xE019E, 0xE019F, 0xE01A0, 0xE01A1, 0xE01A2, 0xE01A3, + 0xE01A4, 0xE01A5, 0xE01A6, 0xE01A7, 0xE01A8, 0xE01A9, 0xE01AA, 0xE01AB, + 0xE01AC, 0xE01AD, 0xE01AE, 0xE01AF, 0xE01B0, 0xE01B1, 0xE01B2, 0xE01B3, + 0xE01B4, 0xE01B5, 0xE01B6, 0xE01B7, 0xE01B8, 0xE01B9, 0xE01BA, 0xE01BB, + 0xE01BC, 0xE01BD, 0xE01BE, 0xE01BF, 0xE01C0, 0xE01C1, 0xE01C2, 0xE01C3, + 0xE01C4, 0xE01C5, 0xE01C6, 0xE01C7, 0xE01C8, 0xE01C9, 0xE01CA, 0xE01CB, + 0xE01CC, 0xE01CD, 0xE01CE, 0xE01CF, 0xE01D0, 0xE01D1, 0xE01D2, 0xE01D3, + 0xE01D4, 0xE01D5, 0xE01D6, 0xE01D7, 0xE01D8, 0xE01D9, 0xE01DA, 0xE01DB, + 0xE01DC, 0xE01DD, 0xE01DE, 0xE01DF, 0xE01E0, 0xE01E1, 0xE01E2, 0xE01E3, + 0xE01E4, 0xE01E5, 0xE01E6, 0xE01E7, 0xE01E8, 0xE01E9, 0xE01EA, 0xE01EB, + 0xE01EC, 0xE01ED, 0xE01EE, 0xE01EF + /* END: COMBINING CHAR TABLE */ +}; + +static const unsigned long combiningCharTableSize = + sizeof(combiningCharTable) / sizeof(combiningCharTable[0]); + +static int isCombiningChar(unsigned long cp) { + size_t i; + for (i = 0; i < combiningCharTableSize; i++) + if (combiningCharTable[i] == cp) return 1; + return 0; +} + +/* Get length of previous grapheme */ +static size_t defaultPrevCharLen(const char* buf, size_t /*buf_len*/, size_t pos, size_t *col_len) { + size_t end = pos; + while (pos > 0) { + size_t len = prevUtf8CodePointLen(buf, pos); + pos -= len; + int cp; + utf8BytesToCodePoint(buf + pos, len, &cp); + if (!isCombiningChar(cp)) { + if (col_len != NULL) *col_len = isWideChar(cp) ? 2 : 1; + return end - pos; + } + } + /* NOTREACHED */ + return 0; +} + +/* Get length of next grapheme */ +static size_t defaultNextCharLen(const char* buf, size_t buf_len, size_t pos, size_t *col_len) { + size_t beg = pos; + int cp; + size_t len = utf8BytesToCodePoint(buf + pos, buf_len - pos, &cp); + if (isCombiningChar(cp)) { + /* NOTREACHED */ + return 0; + } + if (col_len != NULL) *col_len = isWideChar(cp) ? 2 : 1; + pos += len; + while (pos < buf_len) { + int cp; + len = utf8BytesToCodePoint(buf + pos, buf_len - pos, &cp); + if (!isCombiningChar(cp)) return pos - beg; + pos += len; + } + return pos - beg; +} + +/* Read a Unicode from file. */ +static size_t defaultReadCode(int fd, char* buf, size_t buf_len, int* cp) { + if (buf_len < 1) return -1; + size_t nread = read(fd,&buf[0],1); + if (nread <= 0) return nread; + + unsigned char byte = buf[0]; + if ((byte & 0x80) == 0) { + ; + } else if ((byte & 0xE0) == 0xC0) { + if (buf_len < 2) return -1; + nread = read(fd,&buf[1],1); + if (nread <= 0) return nread; + } else if ((byte & 0xF0) == 0xE0) { + if (buf_len < 3) return -1; + nread = read(fd,&buf[1],2); + if (nread <= 0) return nread; + } else if ((byte & 0xF8) == 0xF0) { + if (buf_len < 3) return -1; + nread = read(fd,&buf[1],3); + if (nread <= 0) return nread; + } else { + return -1; + } + + return utf8BytesToCodePoint(buf, buf_len, cp); +} + +/* Set default encoding functions */ +static linenoisePrevCharLen *prevCharLen = defaultPrevCharLen; +static linenoiseNextCharLen *nextCharLen = defaultNextCharLen; +static linenoiseReadCode *readCode = defaultReadCode; + +/* Set used defined encoding functions */ +void linenoiseSetEncodingFunctions( + linenoisePrevCharLen *prevCharLenFunc, + linenoiseNextCharLen *nextCharLenFunc, + linenoiseReadCode *readCodeFunc) { + prevCharLen = prevCharLenFunc; + nextCharLen = nextCharLenFunc; + readCode = readCodeFunc; +} + /* ======================= Low level terminal handling ====================== */ /* Enable "mask mode". When it is enabled, instead of the input that @@ -508,16 +947,30 @@ void linenoiseAddCompletion(linenoiseCompletions *lc, const char *str) { lc->cvec[lc->len++] = copy.release(); } +/* Get column length from begining of buffer to current byte position */ +static size_t columnPos(const char *buf, size_t buf_len, size_t pos) { + size_t ret = 0; + size_t off = 0; + while (off < pos) { + size_t col_len; + size_t len = nextCharLen(buf,buf_len,off,&col_len); + off += len; + ret += col_len; + } + return ret; +} + /* Helper of refreshSingleLine() and refreshMultiLine() to show hints * to the right of the prompt. */ -static void refreshShowHints(std::string & ab, struct linenoiseState * l, int plen) { +static void refreshShowHints(std::string & ab, struct linenoiseState * l, int pcollen) { char seq[64]; - if (hintsCallback && plen+l->len < l->cols) { + size_t collen = pcollen+columnPos(l->buf,l->len,l->len); + if (hintsCallback && collen < l->cols) { int color = -1, bold = 0; const char *hint = hintsCallback(l->buf,&color,&bold); if (hint) { int hintlen = strlen(hint); - int hintmaxlen = l->cols-(plen+l->len); + int hintmaxlen = l->cols-collen; if (hintlen > hintmaxlen) hintlen = hintmaxlen; if (bold == 1 && color == -1) color = 37; if (color != -1 || bold != 0) @@ -535,6 +988,39 @@ static void refreshShowHints(std::string & ab, struct linenoiseState * l, int pl } } +/* Check if text is an ANSI escape sequence */ +static int isAnsiEscape(const char *buf, size_t buf_len, size_t* len) { + if (buf_len > 2 && !memcmp("\033[", buf, 2)) { + size_t off = 2; + while (off < buf_len) { + switch (buf[off++]) { + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': case 'G': case 'H': case 'J': case 'K': + case 'S': case 'T': case 'f': case 'm': + *len = off; + return 1; + } + } + } + return 0; +} + +/* Get column length of prompt text */ +static size_t promptTextColumnLen(const char *prompt, size_t plen) { + char buf[LINENOISE_MAX_LINE]; + size_t buf_len = 0; + size_t off = 0; + while (off < plen) { + size_t len; + if (isAnsiEscape(prompt + off, plen - off, &len)) { + off += len; + continue; + } + buf[buf_len++] = prompt[off++]; + } + return columnPos(buf,buf_len,buf_len); +} + /* Single line low level line refresh. * * Rewrite the currently edited line accordingly to the buffer content, @@ -544,19 +1030,21 @@ static void refreshShowHints(std::string & ab, struct linenoiseState * l, int pl * prompt, just write it, or both. */ static void refreshSingleLine(struct linenoiseState *l, int flags) { char seq[64]; - size_t plen = strlen(l->prompt); + size_t pcollen = promptTextColumnLen(l->prompt,strlen(l->prompt)); int fd = l->ofd; char *buf = l->buf; size_t len = l->len; size_t pos = l->pos; std::string ab; - while((plen+pos) >= l->cols) { - buf++; - len--; - pos--; + + while((pcollen+columnPos(buf,len,pos)) >= l->cols) { + int chlen = nextCharLen(buf,len,0,NULL); + buf += chlen; + len -= chlen; + pos -= chlen; } - while (plen+len > l->cols) { - len--; + while (pcollen+columnPos(buf,len,len) > l->cols) { + len -= prevCharLen(buf,len,len,NULL); } /* Cursor to left edge */ @@ -574,7 +1062,7 @@ static void refreshSingleLine(struct linenoiseState *l, int flags) { ab.append(buf, len); } /* Show hits if any. */ - refreshShowHints(ab, l, plen); + refreshShowHints(ab,l,pcollen); } /* Erase to right */ @@ -582,13 +1070,41 @@ static void refreshSingleLine(struct linenoiseState *l, int flags) { ab.append(seq); if (flags & REFRESH_WRITE) { /* Move cursor to original position. */ - snprintf(seq,sizeof(seq),"\r\x1b[%dC", (int)(pos+plen)); + snprintf(seq,sizeof(seq),"\r\x1b[%dC", (int)(columnPos(buf,len,pos)+pcollen)); ab.append(seq); } (void) !write(fd, ab.c_str(), ab.size()); /* Can't recover from write error. */ } +/* Get column length from begining of buffer to current byte position for multiline mode*/ +static size_t columnPosForMultiLine(const char *buf, size_t buf_len, size_t pos, size_t cols, size_t ini_pos) { + size_t ret = 0; + size_t colwid = ini_pos; + + size_t off = 0; + while (off < buf_len) { + size_t col_len; + size_t len = nextCharLen(buf,buf_len,off,&col_len); + + int dif = (int)(colwid + col_len) - (int)cols; + if (dif > 0) { + ret += dif; + colwid = col_len; + } else if (dif == 0) { + colwid = 0; + } else { + colwid += col_len; + } + + if (off >= pos) break; + off += len; + ret += col_len; + } + + return ret; +} + /* Multi line low level line refresh. * * Rewrite the currently edited line accordingly to the buffer content, @@ -598,11 +1114,13 @@ static void refreshSingleLine(struct linenoiseState *l, int flags) { * prompt, just write it, or both. */ static void refreshMultiLine(struct linenoiseState *l, int flags) { char seq[64]; - int plen = strlen(l->prompt); - int rows = (plen+l->len+l->cols-1)/l->cols; /* rows used by current buf. */ - int rpos = (plen+l->oldpos+l->cols)/l->cols; /* cursor relative row. */ + size_t pcollen = promptTextColumnLen(l->prompt,strlen(l->prompt)); + int colpos = columnPosForMultiLine(l->buf, l->len, l->len, l->cols, pcollen); + int colpos2; /* cursor column position. */ + int rows = (pcollen+colpos+l->cols-1)/l->cols; /* rows used by current buf. */ + int rpos = (pcollen+l->oldcolpos+l->cols)/l->cols; /* cursor relative row. */ int rpos2; /* rpos after refresh. */ - int col; /* colum position, zero-based. */ + int col; /* column position, zero-based. */ int old_rows = l->oldrows; int fd = l->ofd, j; std::string ab; @@ -632,6 +1150,9 @@ static void refreshMultiLine(struct linenoiseState *l, int flags) { ab.append(seq); } + /* Get column length to cursor position */ + colpos2 = columnPosForMultiLine(l->buf,l->len,l->pos,l->cols,pcollen); + if (flags & REFRESH_WRITE) { /* Write the prompt and the current buffer content */ ab.append(l->prompt); @@ -644,13 +1165,13 @@ static void refreshMultiLine(struct linenoiseState *l, int flags) { } /* Show hits if any. */ - refreshShowHints(ab, l, plen); + refreshShowHints(ab,l,pcollen); /* If we are at the very end of the screen with our prompt, we need to * emit a newline and move the prompt to the first column. */ if (l->pos && l->pos == l->len && - (l->pos+plen) % l->cols == 0) + (colpos2+pcollen) % l->cols == 0) { lndebug(""); ab.append("\n"); @@ -661,10 +1182,10 @@ static void refreshMultiLine(struct linenoiseState *l, int flags) { } /* Move cursor to right position. */ - rpos2 = (plen+l->pos+l->cols)/l->cols; /* Current cursor relative row */ + rpos2 = (pcollen+colpos2+l->cols)/l->cols; /* Current cursor relative row */ lndebug("rpos2 %d", rpos2); - /* Go up till we reach the expected positon. */ + /* Go up till we reach the expected position. */ if (rows-rpos2 > 0) { lndebug("go-up %d", rows-rpos2); snprintf(seq,64,"\x1b[%dA", rows-rpos2); @@ -672,7 +1193,7 @@ static void refreshMultiLine(struct linenoiseState *l, int flags) { } /* Set column. */ - col = (plen+(int)l->pos) % (int)l->cols; + col = (pcollen+colpos2) % l->cols; lndebug("set col %d", 1+col); if (col) snprintf(seq,64,"\r\x1b[%dC", col); @@ -682,7 +1203,8 @@ static void refreshMultiLine(struct linenoiseState *l, int flags) { } lndebug("\n"); - l->oldpos = l->pos; + l->oldcolpos = colpos2; + (void) !write(fd, ab.c_str(), ab.size()); /* Can't recover from write error. */ } @@ -720,26 +1242,30 @@ void linenoiseShow(struct linenoiseState *l) { /* Insert the character 'c' at cursor current position. * * On error writing to the terminal -1 is returned, otherwise 0. */ -static int linenoiseEditInsert(struct linenoiseState * l, char c) { - if (l->len < l->buflen) { +static int linenoiseEditInsert(struct linenoiseState *l, const char *cbuf, int clen) { + if (l->len+clen <= l->buflen) { if (l->len == l->pos) { - l->buf[l->pos] = c; - l->pos++; - l->len++; + memcpy(&l->buf[l->pos],cbuf,clen); + l->pos+=clen; + l->len+=clen;; l->buf[l->len] = '\0'; - if ((!mlmode && l->plen+l->len < l->cols && !hintsCallback)) { + if ((!mlmode && promptTextColumnLen(l->prompt,l->plen)+columnPos(l->buf,l->len,l->len) < l->cols && !hintsCallback)) { /* Avoid a full update of the line in the * trivial case. */ - char d = (maskmode==1) ? '*' : c; - if (write(l->ofd,&d,1) == -1) return -1; + if (maskmode == 1) { + static const char d = '*'; + if (write(l->ofd,&d,1) == -1) return -1; + } else { + if (write(l->ofd,cbuf,clen) == -1) return -1; + } } else { refreshLine(l); } } else { - memmove(l->buf+l->pos+1,l->buf+l->pos,l->len-l->pos); - l->buf[l->pos] = c; - l->len++; - l->pos++; + memmove(l->buf+l->pos+clen,l->buf+l->pos,l->len-l->pos); + memcpy(&l->buf[l->pos],cbuf,clen); + l->pos+=clen; + l->len+=clen; l->buf[l->len] = '\0'; refreshLine(l); } @@ -750,7 +1276,7 @@ static int linenoiseEditInsert(struct linenoiseState * l, char c) { /* Move cursor on the left. */ static void linenoiseEditMoveLeft(struct linenoiseState * l) { if (l->pos > 0) { - l->pos--; + l->pos -= prevCharLen(l->buf,l->len,l->pos,NULL); refreshLine(l); } } @@ -758,7 +1284,7 @@ static void linenoiseEditMoveLeft(struct linenoiseState * l) { /* Move cursor on the right. */ static void linenoiseEditMoveRight(struct linenoiseState * l) { if (l->pos != l->len) { - l->pos++; + l->pos += nextCharLen(l->buf,l->len,l->pos,NULL); refreshLine(l); } } @@ -810,8 +1336,9 @@ static void linenoiseEditHistoryNext(struct linenoiseState * l, int dir) { * position. Basically this is what happens with the "Delete" keyboard key. */ static void linenoiseEditDelete(struct linenoiseState * l) { if (l->len > 0 && l->pos < l->len) { - memmove(l->buf+l->pos,l->buf+l->pos+1,l->len-l->pos-1); - l->len--; + int chlen = nextCharLen(l->buf,l->len,l->pos,NULL); + memmove(l->buf+l->pos,l->buf+l->pos+chlen,l->len-l->pos-chlen); + l->len-=chlen; l->buf[l->len] = '\0'; refreshLine(l); } @@ -820,15 +1347,16 @@ static void linenoiseEditDelete(struct linenoiseState * l) { /* Backspace implementation. */ static void linenoiseEditBackspace(struct linenoiseState * l) { if (l->pos > 0 && l->len > 0) { - memmove(l->buf+l->pos-1,l->buf+l->pos,l->len-l->pos); - l->pos--; - l->len--; + int chlen = prevCharLen(l->buf,l->len,l->pos,NULL); + memmove(l->buf+l->pos-chlen,l->buf+l->pos,l->len-l->pos); + l->pos-=chlen; + l->len-=chlen; l->buf[l->len] = '\0'; refreshLine(l); } } -/* Delete the previosu word, maintaining the cursor at the start of the +/* Delete the previous word, maintaining the cursor at the start of the * current word. */ static void linenoiseEditDeletePrevWord(struct linenoiseState * l) { size_t old_pos = l->pos; @@ -878,7 +1406,7 @@ int linenoiseEditStart(struct linenoiseState *l, int stdin_fd, int stdout_fd, ch l->buflen = buflen; l->prompt = prompt; l->plen = strlen(prompt); - l->oldpos = l->pos = 0; + l->oldcolpos = l->pos = 0; l->len = 0; /* Enter raw mode. */ @@ -930,11 +1458,12 @@ const char *linenoiseEditFeed(struct linenoiseState *l) { * count limits. */ if (!isatty(l->ifd)) return linenoiseNoTTY(); - char c; + int c; int nread; + char cbuf[32]; // large enough for any encoding? char seq[3]; - nread = read(l->ifd,&c,1); + nread = readCode(l->ifd,cbuf,sizeof(cbuf),&c); if (nread <= 0) return NULL; /* Only autocomplete when the callback is set. It returns < 0 when @@ -1055,7 +1584,7 @@ const char *linenoiseEditFeed(struct linenoiseState *l) { } break; default: - if (linenoiseEditInsert(l,c)) return NULL; + if (linenoiseEditInsert(l,cbuf,nread)) return NULL; break; case CTRL_U: /* Ctrl+u, delete the whole line. */ l->buf[0] = '\0'; @@ -1136,7 +1665,7 @@ void linenoisePrintKeyCodes(void) { if (memcmp(quit,"quit",sizeof(quit)) == 0) break; printf("'%c' %02x (%d) (type quit to exit)\n", - isprint(c) ? c : '?', (int)c, (int)c); + isprint((int)c) ? c : '?', (int)c, (int)c); printf("\r"); /* Go left edge manually, we are in raw mode. */ fflush(stdout); } diff --git a/linenoise.h b/linenoise.h index a14ec6c..26db591 100644 --- a/linenoise.h +++ b/linenoise.h @@ -63,7 +63,7 @@ struct linenoiseState { const char *prompt; /* Prompt to display. */ size_t plen; /* Prompt length. */ size_t pos; /* Current cursor position. */ - size_t oldpos; /* Previous refresh cursor position. */ + size_t oldcolpos; /* Previous refresh cursor column position. */ size_t len; /* Current edited line length. */ size_t cols; /* Number of columns in terminal. */ size_t oldrows; /* Rows used by last refrehsed line (multiline mode) */ @@ -121,6 +121,16 @@ void linenoisePrintKeyCodes(void); void linenoiseMaskModeEnable(void); void linenoiseMaskModeDisable(void); +/* Encoding functions. */ +typedef size_t (linenoisePrevCharLen)(const char *buf, size_t buf_len, size_t pos, size_t *col_len); +typedef size_t (linenoiseNextCharLen)(const char *buf, size_t buf_len, size_t pos, size_t *col_len); +typedef size_t (linenoiseReadCode)(int fd, char *buf, size_t buf_len, int* c); + +void linenoiseSetEncodingFunctions( + linenoisePrevCharLen *prevCharLenFunc, + linenoiseNextCharLen *nextCharLenFunc, + linenoiseReadCode *readCodeFunc); + #ifdef __cplusplus } #endif diff --git a/scripts/generate_unicode_data_tables.py b/scripts/generate_unicode_data_tables.py new file mode 100755 index 0000000..0067b7c --- /dev/null +++ b/scripts/generate_unicode_data_tables.py @@ -0,0 +1,130 @@ +#!/usr/bin/env python3 +""" Generate a list of combining characters from the latest UnicodeData.txt """ +import sys +import re +import os +import http.client + + +def fetch_unicode_data(host, path): + """ Fetch the Unicode data from the given host and path """ + conn = http.client.HTTPSConnection(host) + conn.request("GET", path) + response = conn.getresponse() + + if response.status != 200: + print(f"Failed to retrieve data: {response.status} {response.reason}") + sys.exit(1) + + return response.read().decode("utf-8").splitlines() + + +def replace_region(lines, text, begin_marker, end_marker): + """ Replace the region between the begin and end markers with the given text """ + results = [] + in_table = False + for line in lines: + match = re.match(f".*{begin_marker}.*", line) + if match: + in_table = True + results.append(line) + results.append(text) + elif in_table: + match = re.match(f".*{end_marker}.*", line) + if match: + in_table = False + results.append(line) + else: + pass + else: + results.append(line) + return results + + +def create_wide_char_table(lines): + """ Create a table of wide characters from the given lines """ + ranges = [] + for line in lines: + match = re.match(r"^(.*?)(?:\.\.(.*?))?\s+;\s+[FW]\s+# .*$", line) + if match: + first = int(match.group(1), 16) + last = int(match.group(2), 16) if match.group(2) else first + if ranges and ranges[-1]["l"] + 1 == first: + ranges[-1]["l"] = last + else: + ranges.append({"f": first, "l": last}) + + pairs = [f"{{ 0x{r['f']:X}, 0x{r['l']:X} }}" for r in ranges] + columns = 4 + return ( + ",\n".join( + [ + " " + ", ".join(pairs[i : i + columns]) + for i in range(0, len(pairs), columns) + ] + ) + + "\n" + ) + + +def create_combining_char_table(lines): + """ Create a table of combining characters from the given lines """ + chars = [] + for line in lines: + match = re.match(r"^(.*);.*;Mn;", line) + if match: + chars.append(f"0x{match.group(1)}") + + columns = 8 + return ( + ",\n".join( + [ + " " + ", ".join(chars[i : i + columns]) + for i in range(0, len(chars), columns) + ] + ) + + "\n" + ) + + +def generate_wide_char_table(lines): + """ Generate the wide character table from the given lines """ + table = create_wide_char_table( + fetch_unicode_data( + "www.unicode.org", "/Public/UCD/latest/ucd/EastAsianWidth.txt" + ) + ) + + return replace_region( + lines, table, "BEGIN: WIDE CHAR TABLE", "END: WIDE CHAR TABLE" + ) + + +def generate_combining_char_table(lines): + """ Generate the combining character table from the given lines """ + table = create_combining_char_table( + fetch_unicode_data("www.unicode.org", "/Public/UCD/latest/ucd/UnicodeData.txt") + ) + + return replace_region( + lines, table, "BEGIN: COMBINING CHAR TABLE", "END: COMBINING CHAR TABLE" + ) + + +def main(): + """ Main function """ + path = f"{os.path.dirname(os.path.abspath(__file__))}/../linenoise.cpp" + + with open(path, "r", encoding="utf-8") as file: + lines = file.readlines() + + lines = generate_wide_char_table(lines) + lines = generate_combining_char_table(lines) + + with open(path, "w", encoding="utf-8") as file: + for line in lines: + file.write(line) + + +if __name__ == "__main__": + main()