/* * Functions for characters and strings for CLISP * Bruno Haible 1990-2008 * Sam Steingold 1998-2009, 2012 * German comments translated into English: Stefan Kain 2002-09-20 */ #include "lispbibl.c" /* character conversion tables: */ #if defined(ENABLE_UNICODE) /* here are the registered bijective case (small<-->CAP) transformations for Unicode. */ #elif defined(ISOLATIN_CHS) /* here are the registered bijective case (small<-->CAP) transformations small 61 ... 7A E0 ... F6 F8 ... FE CAP 41 ... 5A C0 ... D6 D8 ... DE both aA ... zZ àÀ ... öÖ øØ ... þÞ */ #elif defined(HPROMAN8_CHS) /* here are the registered bijective case (small<-->CAP) transformations small 61 ... 7A C4 C5 D5 C6 C7 B2 C0 C1 D1 C2 C3 C8 C9 D9 CA CB CAP 41 ... 5A E0 DC E5 E7 ED B1 A2 A4 A6 DF AE A1 A3 E6 E8 AD which aA ... zZ áÁ éÉ íÍ óÓ úÚ ýÝ â êÊ îÎ ôÔ ûÛ àÀ èÈ ìÌ òÒ ùÙ small CC CD DD CE CF EF E2 B7 EA D4 D7 D6 B5 EC E4 F1 CAP D8 A5 A7 DA DB EE E1 B6 E9 D0 D3 D2 B4 EB E3 F0 which äÄ ëË ïÏ öÖ üÜ ÿŸ ãà ñÑ õÕ åÅ æÆ øØ çÇ šŠ ðÐ þÞ */ #else /* defined(ASCII_CHS) */ /* here are the registered bijective case (small<-->CAP) transformations small 61 ... 7A CAP 41 ... 5A both aA ... zZ */ #endif #ifdef ENABLE_UNICODE /* No-conversion table, used by up_case_table and down_case_table. */ static const uint16 nop_page[256] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }; #endif /* Converts byte ch into an uppercase letter */ modexp chart up_case (chart ch) { #ifdef ENABLE_UNICODE #include "uni_upcase.c" var cint c = as_cint(ch); if (c < sizeof(up_case_table)/sizeof(up_case_table[0]) << 8) return as_chart(c+(sint16)up_case_table[c>>8][c&0xFF]); else return ch; #else /* table for conversion into uppercase letters: */ local const cint up_case_table[char_code_limit] = #if defined(ISOLATIN_CHS) { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 0xD0,0xD1,0xD2,0xD3,0xD4,0xD5,0xD6,0xD7,0xD8,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xDF, 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 0xD0,0xD1,0xD2,0xD3,0xD4,0xD5,0xD6,0xF7,0xD8,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xFF, }; #elif defined(HPROMAN8_CHS) { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 0xB0,0xB1,0xB2,0xB3,0xB4,0xB4,0xB6,0xB6,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xA2,0xA4,0xDF,0xAE,0xE0,0xDC,0xE7,0xB2,0xA1,0xA3,0xE8,0xAD,0xD8,0xA5,0xDA,0xDB, 0xD0,0xA6,0xD2,0xD3,0xD0,0xE5,0xD2,0xD3,0xD8,0xE6,0xDA,0xDB,0xDC,0xA7,0xDE,0xDF, 0xE0,0xE1,0xE1,0xE3,0xE3,0xE5,0xE6,0xE7,0xE8,0xE9,0xE9,0xEB,0xEB,0xED,0xEE,0xEE, 0xF0,0xF0,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF, }; #else /* standard ascii conversion table: only a..z --> A..Z */ { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F, 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 0xD0,0xD1,0xD2,0xD3,0xD4,0xD5,0xD6,0xD7,0xD8,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xDF, 0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7,0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF, 0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF, }; #endif return as_chart(up_case_table[as_cint(ch)]); #endif } /* Converts byte ch into a lowercase letter */ modexp chart down_case (chart ch) { #ifdef ENABLE_UNICODE #include "uni_downcase.c" var cint c = as_cint(ch); if (c < sizeof(down_case_table)/sizeof(down_case_table[0]) << 8) return as_chart(c+(sint16)down_case_table[c>>8][c&0xFF]); else return ch; #else /* table for conversion into lowercase letters: */ local const cint down_case_table[char_code_limit] = #if defined(ISOLATIN_CHS) { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7,0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF, 0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xD7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xDF, 0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7,0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF, 0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF, }; #elif defined(HPROMAN8_CHS) { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xC8,0xC0,0xC9,0xC1,0xCD,0xD1,0xDD,0xA8,0xA9,0xAA,0xAB,0xAC,0xCB,0xC3,0xAF, 0xB0,0xB2,0xB2,0xB3,0xB5,0xB5,0xB7,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 0xD4,0xD1,0xD6,0xD7,0xD4,0xD5,0xD6,0xD7,0xCC,0xD9,0xCE,0xCF,0xC5,0xDD,0xDE,0xC2, 0xC4,0xE2,0xE2,0xE4,0xE4,0xD5,0xD9,0xC6,0xCA,0xEA,0xEA,0xEC,0xEC,0xC7,0xEF,0xEF, 0xF1,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF, }; #else /* standard ascii conversion table: only A..Z --> a..z */ { 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F, 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x1A,0x1B,0x1C,0x1D,0x1E,0x1F, 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,0x29,0x2A,0x2B,0x2C,0x2D,0x2E,0x2F, 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0x3A,0x3B,0x3C,0x3D,0x3E,0x3F, 0x40,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x5B,0x5C,0x5D,0x5E,0x5F, 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F, 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0x7B,0x7C,0x7D,0x7E,0x7F, 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87,0x88,0x89,0x8A,0x8B,0x8C,0x8D,0x8E,0x8F, 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97,0x98,0x99,0x9A,0x9B,0x9C,0x9D,0x9E,0x9F, 0xA0,0xA1,0xA2,0xA3,0xA4,0xA5,0xA6,0xA7,0xA8,0xA9,0xAA,0xAB,0xAC,0xAD,0xAE,0xAF, 0xB0,0xB1,0xB2,0xB3,0xB4,0xB5,0xB6,0xB7,0xB8,0xB9,0xBA,0xBB,0xBC,0xBD,0xBE,0xBF, 0xC0,0xC1,0xC2,0xC3,0xC4,0xC5,0xC6,0xC7,0xC8,0xC9,0xCA,0xCB,0xCC,0xCD,0xCE,0xCF, 0xD0,0xD1,0xD2,0xD3,0xD4,0xD5,0xD6,0xD7,0xD8,0xD9,0xDA,0xDB,0xDC,0xDD,0xDE,0xDF, 0xE0,0xE1,0xE2,0xE3,0xE4,0xE5,0xE6,0xE7,0xE8,0xE9,0xEA,0xEB,0xEC,0xED,0xEE,0xEF, 0xF0,0xF1,0xF2,0xF3,0xF4,0xF5,0xF6,0xF7,0xF8,0xF9,0xFA,0xFB,0xFC,0xFD,0xFE,0xFF, }; #endif return as_chart(down_case_table[as_cint(ch)]); #endif } #ifdef ENABLE_UNICODE /* Table of Unicode character attributes. unicode_attribute(c) > cint c: a character code < result: 0 = non-graphic 1 = graphic, but not alphanumeric 2 = graphic and numeric 3 = graphic and alphabetic */ #include "uni_attribute.c" #define unicode_attribute(c) \ ((c) < sizeof(unicode_attribute_table)/sizeof(unicode_attribute_table[0]) << 10 \ ? (unicode_attribute_table[(c)>>10][((c)>>2)&0xFF] >> (((c)&0x3)*2)) & 0x3 \ : 0) #endif /* UP: Determines, if a character is alphabetic. alphap(ch) > ch: character-code < result: true if alphabetic, otherwise false. Alphabetic characters have a code c, with */ #if defined(ENABLE_UNICODE) /* java.lang.Character.isLetter(c) */ #else /* $41 <= c <= $5A or $61 <= c <= $7A */ #if defined(ISOLATIN_CHS) /* or $C0 <= c except c=$D7,$F7. */ #elif defined(HPROMAN8_CHS) /* or $A1 <= c <= $A7 or $AD <= c <= $AE or $B1 <= c <= $B7 except c=$B3 or $C0 <= c <= $F1. */ #endif #endif /* Therein, all uppercase- and all lowercase- characters are enclosed (see CLTL p. 236 top). */ local bool alphap (chart ch) { var cint c = as_cint(ch); #ifdef ENABLE_UNICODE return (unicode_attribute(c) == 3); #else if (c < 0x41) goto no; if (c <= 0x5A) goto yes; if (c < 0x61) goto no; if (c <= 0x7A) goto yes; #if defined(ISOLATIN_CHS) if (c < 0xC0) goto no; if ((c == 0xD7) || (c == 0xF7)) goto no; else goto yes; #elif defined(HPROMAN8_CHS) if (c < 0xA1) goto no; if (c > 0xF1) goto no; if (c >= 0xC0) goto yes; if (c <= 0xA7) goto yes; if (c < 0xB1) { if (c < 0xAD) goto no; if (c <= 0xAE) goto yes; goto no; } else { if (c > 0xB7) goto no; if (c == 0xB3) goto no; else goto yes; } #endif no: return false; yes: return true; #endif } /* Determines, if a character is numeric. numericp(ch) > ch: character-code < result: true if numeric, otherwise false. */ local bool numericp (chart ch); #ifdef ENABLE_UNICODE #define numericp(ch) (unicode_attribute(as_cint(ch)) == 2) #else #define numericp(ch) (('0' <= as_cint(ch)) && (as_cint(ch) <= '9')) #endif /* Determines, if a character is alphanumeric. alphanumericp(ch) > ch: character-code < result: true if alphanumeric, otherwise false. Alphanumeric characters comprise the alphabetic characters and the digits. */ global bool alphanumericp (chart ch) { #ifdef ENABLE_UNICODE var cint c = as_cint(ch); return (unicode_attribute(c) >= 2); #else return (numericp(ch) || alphap(ch)); #endif } /* Determines, if a character is a Graphic-Character ("printing") . graphic_char_p(ch) > ch: character-code < result: true if printing, otherwise false. Graphic-Characters are those with a Code c, with */ #if defined(ENABLE_UNICODE) /* (java.lang.Character.isDefined(c) || c == 0x20AC) && !(c < 0x0020 || (0x007F <= c <= 0x009F)) */ #elif defined(ISOLATIN_CHS) || defined(HPROMAN8_CHS) /* $20 <= c <= $7E or $A0 <= c < $100. */ #else /* defined(ASCII_CHS) */ /* $20 <= c <= $7E. */ #endif global bool graphic_char_p (chart ch) { /* This would be the same as iswprint(ch), assuming wide characters were Unicode. */ var cint c = as_cint(ch); #ifdef ENABLE_UNICODE return (unicode_attribute(c) == 0 ? false : true); #else #if defined(ISOLATIN_CHS) || defined(HPROMAN8_CHS) if ((('~' >= c) && (c >= ' ')) || (c >= 0xA0)) goto yes; else goto no; #else /* defined(ASCII_CHS) */ if (c >= ' ') goto yes; else goto no; #endif no: return false; yes: return true; #endif } /* Returns the screen display width of a character. char_width(ch) > ch: character code < result: number of output columns occupied by ch */ global uintL char_width (chart ch); #ifdef ENABLE_UNICODE #include "uniwidth.h" /* from gnulib */ global uintL char_width (chart ch) { /* This would be the same as wcwidth(ch), assuming wide characters were Unicode, except that for non-printable characters we return 0, not -1. */ extern const char* locale_encoding; int w = uc_width(as_cint(ch),locale_encoding); return (w >= 0 ? w : 0); } #else global uintL char_width (chart ch) { return (graphic_char_p(ch) ? 1 : 0); } #endif #if !defined(ENABLE_UNICODE) || defined(HAVE_SMALL_SSTRING) /* Copies an array of uint8 to an array of uint8. copy_8bit_8bit(src,dest,len); > uint8* src: source > uint8* dest: destination > uintL len: number of elements to be copied, > 0 */ global void copy_8bit_8bit (const uint8* src, uint8* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint8 to an array of uint16. copy_8bit_16bit(src,dest,len); > uint8* src: source > uint16* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_8bit_16bit (const uint8* src, uint16* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint8 to an array of uint32. copy_8bit_32bit(src,dest,len); > uint8* src: source > uint32* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_8bit_32bit (const uint8* src, uint32* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint16 to an array of uint8. All source elements must fit into uint8. copy_16bit_8bit(src,dest,len); > uint16* src: source > uint8* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_16bit_8bit (const uint16* src, uint8* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint16 to an array of uint16. copy_16bit_16bit(src,dest,len); > uint16* src: source > uint16* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_16bit_16bit (const uint16* src, uint16* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint16 to an array of uint32. copy_16bit_32bit(src,dest,len); > uint16* src: source > uint32* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_16bit_32bit (const uint16* src, uint32* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint32 to an array of uint8. All source elements must fit into uint8. copy_32bit_8bit(src,dest,len); > uint32* src: source > uint8* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_32bit_8bit (const uint32* src, uint8* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(HAVE_SMALL_SSTRING) /* Copies an array of uint32 to an array of uint16. All source elements must fit into uint16. copy_32bit_16bit(src,dest,len); > uint32* src: source > uint16* dest: destination > uintL len: number of elements to be copied, > 0 */ modexp void copy_32bit_16bit (const uint32* src, uint16* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #if defined(ENABLE_UNICODE) /* Copies an array of uint32 to an array of uint32. copy_32bit_32bit(src,dest,len); > uint32* src: source > uint32* dest: destination > uintL len: number of elements to be copied, > 0 */ global void copy_32bit_32bit (const uint32* src, uint32* dest, uintL len) { do { *dest++ = *src++; } while (--len); } #endif #ifdef HAVE_SMALL_SSTRING /* Determines the smallest string element type capable of holding a set of 16-bit characters. smallest_string_flavour16(src,len) > uint16* src: source > uintL len: number of characters at src < result: Sstringtype_8Bit or Sstringtype_16Bit */ global uintBWL smallest_string_flavour16 (const uint16* src, uintL len) { var uintBWL result = Sstringtype_8Bit; if (len > 0) { var uintL count; dotimespL(count,len, { if (!(*src < cint8_limit)) { result = Sstringtype_16Bit; break; } src++; }); } return result; } /* Determines the smallest string element type capable of holding a set of 32-bit characters. smallest_string_flavour32(src,len) > uint32* src: source > uintL len: number of characters at src < result: Sstringtype_8Bit or Sstringtype_16Bit or Sstringtype_32Bit */ global uintBWL smallest_string_flavour32 (const uint32* src, uintL len) { var uintBWL result = Sstringtype_8Bit; if (len > 0) { var uintL count; dotimespL(count,len, { if (!(*src < cint8_limit)) result = Sstringtype_16Bit; if (!(*src < cint16_limit)) { result = Sstringtype_32Bit; break; } src++; }); } return result; } #endif /* UP: unpack a string unpack_string_ro(string,&len,&offset) [for read-only access] > object string: a string < uintL len: the fill-pointer length of the string < uintL offset: offset into the datastorage vector < object result: datastorage vector, a simple-string or NIL */ modexp object unpack_string_ro (object string, uintL* len, uintL* offset) { if (simple_string_p(string)) { sstring_un_realloc(string); *len = Sstring_length(string); *offset = 0; return string; } else { /* string, but not simple-string => follow the displacement determine the length (like vector_length() in array.d): */ var uintL size; { var Iarray addr = TheIarray(string); var uintL offset_fil = offsetofa(iarray_,dims); if (iarray_flags(addr) & bit(arrayflags_dispoffset_bit)) offset_fil += sizeof(uintL); if (iarray_flags(addr) & bit(arrayflags_fillp_bit)) offset_fil += sizeof(uintL); size = *(uintL*)pointerplus(addr,offset_fil); } *len = size; /* follow the displacement: */ *offset = 0; return iarray_displace_check(string,size,offset); } } #if 0 /* not used */ /* UP: unpack a string unpack_string_rw(string,&len,&offset) [for read-write access] > object string: a string < uintL len: the fill-pointer length of the string < uintL offset: offset in the datastorage vector < object result: datastorage vector, a simple-string or NIL */ global object unpack_string_rw (object string, uintL* len, uintL* offset) { var object unpacked = unpack_string_ro(string,len,offset); if (*len > 0) { if (simple_nilarray_p(unpacked)) error_nilarray_access(); check_sstring_mutable(unpacked); } return unpacked; } #endif /* UP: compares two strings for equality string_eq(string1,string2) > string1: string > string2: simple-string < result: /=0, if equal */ global bool string_eq (object string1, object string2) { var uintL len1; var uintL offset1; string1 = unpack_string_ro(string1,&len1,&offset1); sstring_un_realloc(string2); if (len1 != Sstring_length(string2)) return false; /* Now both strings have exactly len1 characters. Compare them. */ if (len1 > 0) return string_eqcomp(string1,offset1,string2,0,len1); return true; } /* UP: compares two strings for equality, case-insensitive string_equal(string1,string2) > string1: string > string2: simple-string < result: /=0, if equal */ modexp bool string_equal (object string1, object string2) { var uintL len1; var uintL offset1; string1 = unpack_string_ro(string1,&len1,&offset1); sstring_un_realloc(string2); if (len1 != Sstring_length(string2)) return false; /* Now both strings have exactly len1 characters. Compare them. */ if (len1 > 0) return string_eqcomp_ci(string1,offset1,string2,0,len1); return true; } /* UP: Stores a character in a string. > string: a mutable string that is or was simple > index: (already checked) index into the string > element: a character < result: the possibly reallocated string can trigger GC */ global maygc object sstring_store (object string, uintL index, chart element) { var object inner_string = string; sstring_un_realloc(inner_string); switch (sstring_eltype(TheSstring(inner_string))) { #ifdef ENABLE_UNICODE #ifdef HAVE_SMALL_SSTRING case Sstringtype_8Bit: /* mutable Simple-String */ if (as_cint(element) < cint8_limit) { TheS8string(inner_string)->data[index] = as_cint(element); break; } ASSERT(eq(string,inner_string)); if (as_cint(element) < cint16_limit) { string = reallocate_small_string(inner_string,Sstringtype_16Bit); inner_string = TheSistring(string)->data; TheS16string(inner_string)->data[index] = as_cint(element); break; } string = reallocate_small_string(inner_string,Sstringtype_32Bit); inner_string = TheSistring(string)->data; TheS32string(inner_string)->data[index] = as_cint(element); break; case Sstringtype_16Bit: /* mutable Simple-String */ if (as_cint(element) < cint16_limit) { TheS16string(inner_string)->data[index] = as_cint(element); break; } pushSTACK(string); inner_string = reallocate_small_string(inner_string,Sstringtype_32Bit); string = popSTACK(); inner_string = TheSistring(inner_string)->data; /*FALLTHROUGH*/ #endif case Sstringtype_32Bit: /* mutable Simple-String */ TheS32string(inner_string)->data[index] = as_cint(element); break; #else case Sstringtype_8Bit: /* mutable Simple-String */ TheS8string(inner_string)->data[index] = as_cint(element); break; #endif default: NOTREACHED; } return string; } /* UP: Stores an array of characters in a string. > string: a mutable string that is or was simple > offset: (already checked) offset into the string > charptr[0..len-1]: a character array, not GC affected < result: the possibly reallocated string can trigger GC */ global maygc object sstring_store_array (object string, uintL offset, const chart *charptr, uintL len) { if (len > 0) { var object inner_string = string; sstring_un_realloc(inner_string); switch (sstring_eltype(TheSstring(inner_string))) { #ifdef ENABLE_UNICODE #ifdef HAVE_SMALL_SSTRING case Sstringtype_8Bit: { /* mutable Simple-String */ var bool fits_in_8bit = true; var bool fits_in_16bit = true; { var uintL n = len; var const chart* p = charptr; do { if (!(as_cint(*p) < cint8_limit)) fits_in_8bit = false; if (!(as_cint(*p) < cint16_limit)) { fits_in_16bit = false; break; } p++; } while (--n); } if (fits_in_8bit) { var const chart* p = charptr; var cint8* q = &TheS8string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); break; } ASSERT(eq(string,inner_string)); if (fits_in_16bit) { string = reallocate_small_string(inner_string,Sstringtype_16Bit); inner_string = TheSistring(string)->data; var const chart* p = charptr; var cint16* q = &TheS16string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); break; } string = reallocate_small_string(inner_string,Sstringtype_32Bit); inner_string = TheSistring(string)->data; var const chart* p = charptr; var cint32* q = &TheS32string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); } break; case Sstringtype_16Bit: { /* mutable Simple-String */ var bool fits_in_16bit = true; { var uintL n = len; var const chart* p = charptr; do { if (!(as_cint(*p) < cint16_limit)) { fits_in_16bit = false; break; } p++; } while (--n); } if (fits_in_16bit) { var const chart* p = charptr; var cint16* q = &TheS16string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); break; } pushSTACK(string); inner_string = reallocate_small_string(inner_string,Sstringtype_32Bit); string = popSTACK(); inner_string = TheSistring(inner_string)->data; } /*FALLTHROUGH*/ #endif case Sstringtype_32Bit: { /* mutable Simple-String */ var const chart* p = charptr; var cint32* q = &TheS32string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); } break; #else case Sstringtype_8Bit: { /* mutable Simple-String */ var const chart* p = charptr; var cint8* q = &TheS8string(inner_string)->data[offset]; do { *q = as_cint(*p); p++; q++; } while (--len); } break; #endif default: NOTREACHED; } } return string; } #ifdef ENABLE_UNICODE /* UP: Creates a Simple-String with given elements. stringof(len) > uintL len: desired vector length > on STACK: len characters, first on top < result: Simple-String with these objects increases STACK changes STACK, can trigger GC */ global maygc object stringof (uintL len) { check_stringsize(len); var object new_string = allocate_string(len); if (len > 0) { var gcv_object_t* topargptr = STACK STACKop len; var gcv_object_t* argptr = topargptr; var chart* ptr = &TheSnstring(new_string)->data[0]; var uintL count; dotimespL(count,len, { *ptr++ = char_code(NEXT(argptr)); }); set_args_end_pointer(topargptr); #ifdef HAVE_SMALL_SSTRING /* We use alloca for small-simple-strings, therefore their length should not be too large, or we risk an SP overflow and core dump. */ if (len < 0x10000) { var uintBWL flavour = smallest_string_flavour(&TheSnstring(new_string)->data[0],len); if (flavour == Sstringtype_8Bit) { pushSTACK(new_string); var object copied_string = allocate_s8string(len); copy_32bit_8bit(&TheS32string(popSTACK())->data[0], &TheS8string(copied_string)->data[0],len); new_string = copied_string; } else if (flavour == Sstringtype_16Bit) { pushSTACK(new_string); var object copied_string = allocate_s16string(len); copy_32bit_16bit(&TheS32string(popSTACK())->data[0], &TheS16string(copied_string)->data[0],len); new_string = copied_string; } } #endif } return new_string; } #endif /* UP: Copies a string and thereby turns it into a Simple-String. copy_string_normal(string) > string: String < result: mutable Normal-Simple-String with the same characters can trigger GC */ global maygc object copy_string_normal (object string) { var uintL len; var uintL offset; string = unpack_string_ro(string,&len,&offset); pushSTACK(string); /* save string */ var object new_string = allocate_string(len); /* new_string = new Normal-Simple-String with given length len */ string = popSTACK(); /* return string */ if (len > 0) { #ifdef ENABLE_UNICODE SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_16bit_32bit(&TheS16string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_32bit_32bit(&TheS32string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { error_nilarray_retrieve(); }); #else SstringCase(string, { NOTREACHED; }, { NOTREACHED; }, { copy_8bit_8bit(&TheS8string(string)->data[offset], &TheS8string(new_string)->data[0],len); }, { error_nilarray_retrieve(); }); #endif } return new_string; } #ifdef HAVE_SMALL_SSTRING /* UP: Copies a string and thereby turns it into a Simple-String. copy_string(string) > string: String < result: mutable Simple-String with the same characters can trigger GC */ global maygc object copy_string (object string) { var uintL len; var uintL offset; string = unpack_string_ro(string,&len,&offset); var uintBWL flavour; /* We use alloca for small-simple-strings, therefore their length should not be too large, or we risk an SP overflow and core dump. */ if (len < 0x10000) { SstringCase(string, { flavour = smallest_string_flavour8(&TheS8string(string)->data[offset],len); }, { flavour = smallest_string_flavour16(&TheS16string(string)->data[offset],len); }, { flavour = smallest_string_flavour32(&TheS32string(string)->data[offset],len); }, { flavour = Sstringtype_8Bit; }); } else flavour = Sstringtype_32Bit; pushSTACK(string); /* save string */ var object new_string = (flavour == Sstringtype_8Bit ? allocate_s8string(len) : flavour == Sstringtype_16Bit ? allocate_s16string(len) : allocate_s32string(len)); /* new_string = new Simple-String with given length len */ string = popSTACK(); /* return string */ if (len > 0) { if (flavour == Sstringtype_8Bit) { SstringCase(string, { copy_8bit_8bit(&TheS8string(string)->data[offset], &TheS8string(new_string)->data[0],len); }, { copy_16bit_8bit(&TheS16string(string)->data[offset], &TheS8string(new_string)->data[0],len); }, { copy_32bit_8bit(&TheS32string(string)->data[offset], &TheS8string(new_string)->data[0],len); }, { error_nilarray_retrieve(); }); } else if (flavour == Sstringtype_16Bit) { SstringCase(string, { copy_8bit_16bit(&TheS8string(string)->data[offset], &TheS16string(new_string)->data[0],len); }, { copy_16bit_16bit(&TheS16string(string)->data[offset], &TheS16string(new_string)->data[0],len); }, { copy_32bit_16bit(&TheS32string(string)->data[offset], &TheS16string(new_string)->data[0],len); }, { NOTREACHED; }); } else { SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_16bit_32bit(&TheS16string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_32bit_32bit(&TheS32string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { NOTREACHED; }); } } return new_string; } #endif /* UP: converts a string into a Simple-String. coerce_ss(obj) > obj: Lisp-object, should be a string. < result: Simple-String with the same characters can trigger GC */ global maygc object coerce_ss (object obj) { start: #ifdef TYPECODES switch (typecode(obj)) #else if (orecordp(obj)) switch (Record_type(obj)) { case_Rectype_Sstring_above; case_Rectype_ostring_above; default: break; } switch (0) #endif { case_sstring: /* Simple-String, returned unchanged */ DBGREALLOC(obj); return obj; case_ostring: /* other string, copy it */ return copy_string(obj); default: break; } obj = check_string(obj); goto start; } /* UP: converts a string into an immutable Simple-String. coerce_imm_ss(obj) > obj: Lisp-object, should be a string. < result: immutable Simple-String with the same characters can trigger GC */ global maygc object coerce_imm_ss (object obj) { start: #ifdef TYPECODES switch (typecode(obj)) #else if (orecordp(obj)) switch (Record_type(obj)) { case_Rectype_Sstring_above; case_Rectype_ostring_above; default: break; } switch (0) #endif { case_sstring: /* Simple-String */ if (sstring_immutable(TheSstring(obj))) /* already immutable, return unchanged */ return obj; /*FALLTHROUGH*/ case_ostring: { /* other string, copy it */ var uintL len; var uintL offset; var object string = unpack_string_ro(obj,&len,&offset); if (simple_nilarray_p(string)) { if (len > 0) error_nilarray_retrieve(); return allocate_imm_string(0); } #ifdef ENABLE_UNICODE #ifdef HAVE_SMALL_SSTRING if (sstring_eltype(TheSstring(string)) == Sstringtype_8Bit) { pushSTACK(string); var object new_string = allocate_imm_s8string(len); string = popSTACK(); if (len > 0) copy_8bit_8bit(&TheS8string(string)->data[offset], &TheS8string(new_string)->data[0],len); return new_string; } if (sstring_eltype(TheSstring(string)) == Sstringtype_16Bit) { /* Check if all characters fit into an 8-bit character string. */ var bool fits_in_8bit = true; if (len > 0) { var const cint16* ptr = &TheS16string(string)->data[offset]; var uintL count = len; do { if (!(*ptr < cint8_limit)) { fits_in_8bit = false; break; } ptr++; } while (--count); } pushSTACK(string); var object new_string = (fits_in_8bit ? allocate_imm_s8string(len) : allocate_imm_s16string(len)); string = popSTACK(); if (len > 0) { if (fits_in_8bit) copy_16bit_8bit(&TheS16string(string)->data[offset], &TheS8string(new_string)->data[0],len); else copy_16bit_16bit(&TheS16string(string)->data[offset], &TheS16string(new_string)->data[0],len); } return new_string; } ASSERT(sstring_eltype(TheSstring(string)) == Sstringtype_32Bit); /* We use alloca for small-simple-strings, therefore their length should not be too large, or we risk an SP overflow and core dump. */ if (len < 0x10000) { /* Check if all characters fit into an 8-bit or 16-bit character simple string: */ var bool fits_in_8bit = true; var bool fits_in_16bit = true; if (len > 0) { var const cint32* ptr = &TheS32string(string)->data[offset]; var uintL count = len; do { if (!(*ptr < cint8_limit)) fits_in_8bit = false; if (!(*ptr < cint16_limit)) { fits_in_16bit = false; break; } ptr++; } while (--count); } if (fits_in_16bit) { pushSTACK(string); var object new_string = (fits_in_8bit ? allocate_imm_s8string(len) : allocate_imm_s16string(len)); string = popSTACK(); if (len > 0) { if (fits_in_8bit) copy_32bit_8bit(&TheS32string(string)->data[offset], &TheS8string(new_string)->data[0],len); else copy_32bit_16bit(&TheS32string(string)->data[offset], &TheS16string(new_string)->data[0],len); } return new_string; } } #endif pushSTACK(string); var object new_string = allocate_imm_s32string(len); string = popSTACK(); if (len > 0) copy_32bit_32bit(&TheS32string(string)->data[offset], &TheS32string(new_string)->data[0],len); return new_string; #else pushSTACK(string); var object new_string = allocate_imm_string(len); string = popSTACK(); if (len > 0) copy_8bit_8bit(&TheS8string(string)->data[offset], &TheS8string(new_string)->data[0],len); return new_string; #endif } default: break; } obj = check_string(obj); goto start; } #ifdef HAVE_SMALL_SSTRING /* UP: converts a string into a Normal-Simple-String. coerce_normal_ss(obj) > obj: Lisp-object, should be a string. < result: Normal-Simple-String with the same characters can trigger GC */ global maygc object coerce_normal_ss (object obj) { start: #ifdef TYPECODES switch (typecode(obj)) #else if (orecordp(obj)) switch (Record_type(obj)) { case_Rectype_Sstring_above; case_Rectype_ostring_above; default: break; } switch (0) #endif { case_sstring: sstring_un_realloc(obj); if (sstring_eltype(TheSstring(obj)) == Sstringtype_32Bit) /* already a Normal-Simple-String, return unchanged */ return obj; /*FALLTHROUGH*/ case_ostring: /* other string, copy it */ return copy_string_normal(obj); default: break; } obj = check_string(obj); goto start; } #endif #if 0 /* unused */ #ifdef HAVE_SMALL_SSTRING /* UP: converts a string into an immutable Normal-Simple-String. coerce_imm_normal_ss(obj) > obj: Lisp-object, should be a string. < result: immutable Normal-Simple-String with the same characters can trigger GC */ global maygc object coerce_imm_normal_ss (object obj) { start: #ifdef TYPECODES switch (typecode(obj)) #else if (orecordp(obj)) switch (Record_type(obj)) { case_Rectype_Sstring_above; case_Rectype_ostring_above; default: break; } switch (0) #endif { case_sstring: if (sstring_immutable(TheSstring(obj)) && sstring_eltype(TheSstring(obj)) == Sstringtype_32Bit) /* immutable Normal-Simple-String, return unchanged */ return obj; /*FALLTHROUGH*/ case_ostring: { /* other string, copy it */ var uintL len; var uintL offset; var object string = unpack_string_ro(obj,&len,&offset); pushSTACK(string); var object new_string = allocate_imm_string(len); string = popSTACK(); if (len > 0) { SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_16bit_32bit(&TheS16string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { copy_32bit_32bit(&TheS32string(string)->data[offset], &TheS32string(new_string)->data[0],len); }, { error_nilarray_retrieve(); }); } return new_string; } default: break; } obj = check_string(obj); goto start; } #endif #endif LISPFUNNR(string_info,1) { /* (SYS::STRING-INFO str) => char-len(8/16/32); immutable-p; realloc-p */ var object str = popSTACK(); if (stringp(str)) { if (!simple_string_p(str)) { if ((Iarray_flags(str) & arrayflags_atype_mask) == Atype_NIL) goto other; do { str = TheIarray(str)->data; } while (!simple_string_p(str)); } value3 = NIL; while (sstring_reallocatedp(TheSstring(str))) { value3 = T; str = TheSistring(str)->data; } value2 = (sstring_immutable(TheSstring(str)) ? T : NIL); value1 = fixnum(8 << sstring_eltype(TheSstring(str))); } else other: value1 = value2 = value3 = NIL; mv_count = 3; } /* UP: conversion of an object into a character coerce_char(obj) > obj: Lisp-object < result: Character or NIL */ global object coerce_char (object obj) { if (charp(obj)) { return obj; /* return character unchanged */ } else if (symbolp(obj)) { /* obj is a symbol */ obj = TheSymbol(obj)->pname; goto string; } else if (stringp(obj)) string: { /* obj is a string */ var uintL len; var uintL offset; var object string = unpack_string_ro(obj,&len,&offset); /* at ptr are len characters */ if (len==1) return code_char(schar(string,offset)); } else if (nullpSv(coerce_fixnum_char_ansi) && posfixnump(obj)) { var uintV code = posfixnum_to_V(obj); if (code < char_code_limit) /* obj is a fixnum >=0, < char_code_limit */ return code_char(as_chart(code)); } #if defined(KEYBOARD) else if (typep_classname(obj,S(input_character))) { /* obj is an INPUT-CHARACTER. Call (SYS::INPUT-CHARACTER-CHAR obj): */ pushSTACK(obj); funcall(S(input_character_char),1); return charp(value1) ? value1 : NIL; } #endif /* was none of it -> can not be converted into a character */ return NIL; /* NIL as result */ } /* character-names: Only the characters with font 0 and bits 0 have names. Among these are all non-graphic characters and the space. The reader also accepts - the syntax #\A for the character A (etc. for all characters), - the syntax #\^A for the character 'A'-64 (etc. for all control-characters of the ASCII-charset) and - the syntax #\Code231 for the character with the code 231 (decimal) for all characters out of font 0. */ /* table of character-names: defined in CONSTOBJ.D, */ #ifdef WIN32_CHARNAMES #define charname_table_length 14 /* length of the table */ #define charname_table ((gcv_object_t*)(&object_tab.charname_0)) /* table starts with charname_0 */ #endif #ifdef UNIX_CHARNAMES #define charname_table_length 48 /* length of the table */ #define charname_table ((gcv_object_t*)(&object_tab.charname_0_1)) /* table starts with charname_0_1 */ #endif /* table of codes for this name: */ local const uintB charname_table_codes [charname_table_length] #ifdef WIN32_CHARNAMES = { 0,BEL,BS,TAB,NL,11,PG,CR,26,ESC,' ',RUBOUT,LF,27, }; #endif #ifdef UNIX_CHARNAMES = { 0,7,BS,TAB,NL,LF,LF,PG,PG,CR,27,32,RUBOUT,127, 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19, 20,21,22,23,24,25,26,27,28,29,30,31,32,127, }; #endif /* the code charname_table_codes[i] belongs to the name charname_table[i] (for 0 <= i < charname_table_length). */ #ifdef ENABLE_UNICODE #include "uniname.h" /* from gnulib */ #endif /* UP: return the name of a character char_name(code) > chart code: character code < result: simple-string (the name of the character) or NIL Note that the resulting string is ready to be output unmodified; no prior STRING-CAPITALIZE or so is needed. can trigger GC */ global maygc object char_name (chart code) { var cint c = as_cint(code); { var const uintB* codes_ptr = &charname_table_codes[0]; var const gcv_object_t* strings_ptr = &charname_table[0]; var uintC count = charname_table_length; while (count--) { if (c == *codes_ptr++) /* compare code with charname_table_codes[i] */ return *strings_ptr; /* return string charname_table[i] from the table */ strings_ptr++; } } /* not found */ #ifdef ENABLE_UNICODE /* Try to find the long name, from UnicodeDataFull.txt. It is the second semicolon separated field from (sys::unicode-attributes-line c). */ #ifdef AWFULLY_SLOW { pushSTACK(fixnum(c)); funcall(S(unicode_attributes_line),1); var object line = value1; if (!nullp(line)) { var uintL len = Sstring_length(line); var uintL i1; var uintL i2; for (i1 = 0; i1 < len; i1++) if (chareq(TheSstring(line)->data[i1],ascii(';'))) { i1++; for (i2 = i1; i2 < len; i2++) if (chareq(TheSstring(line)->data[i2],ascii(';'))) { if (!chareq(TheSstring(line)->data[i1],ascii('<'))) { var object name = subsstring(line,i1,i2); /* Replace ' ' with '_': */ var uintL count = i2-i1; if (count > 0) { var chart* ptr = &TheSstring(name)->data[0]; do { if (chareq(*ptr,ascii(' '))) *ptr = ascii('_'); ptr++; } while (--count); } return name; } break; } break; } } } #else /* Here is a much faster implementation. */ { var char buf[UNINAME_MAX]; if (unicode_character_name(c,buf)) { var char* ptr = buf; /* Turn the word separators into underscores. */ while (*ptr != '\0') { if (*ptr == ' ') *ptr = '_'; ptr++; } return n_char_to_string(buf,ptr-buf,Symbol_value(S(ascii))); } } #endif /* AWFULLY_SLOW */ /* CLHS (glossary "name" 5) specifies that all non-graphic characters have a name. Let's give a name to all of them, it's more uniform (and avoids printer errors). */ /* if (!graphic_char_p(code)) */ { local const char hex_table[] = "0123456789ABCDEF"; if (c < 0x10000) { var object name = allocate_s8string(5); TheS8string(name)->data[0] = as_cint(ascii('U')); TheS8string(name)->data[1] = as_cint(ascii(hex_table[(c>>12)&0x0F])); TheS8string(name)->data[2] = as_cint(ascii(hex_table[(c>>8)&0x0F])); TheS8string(name)->data[3] = as_cint(ascii(hex_table[(c>>4)&0x0F])); TheS8string(name)->data[4] = as_cint(ascii(hex_table[c&0x0F])); return name; } else { var object name = allocate_s8string(9); TheS8string(name)->data[0] = as_cint(ascii('U')); TheS8string(name)->data[1] = as_cint(ascii('0')); TheS8string(name)->data[2] = as_cint(ascii('0')); TheS8string(name)->data[3] = as_cint(ascii(hex_table[(c>>20)&0x0F])); TheS8string(name)->data[4] = as_cint(ascii(hex_table[(c>>16)&0x0F])); TheS8string(name)->data[5] = as_cint(ascii(hex_table[(c>>12)&0x0F])); TheS8string(name)->data[6] = as_cint(ascii(hex_table[(c>>8)&0x0F])); TheS8string(name)->data[7] = as_cint(ascii(hex_table[(c>>4)&0x0F])); TheS8string(name)->data[8] = as_cint(ascii(hex_table[c&0x0F])); return name; } } #else /* no ENABLE_UNICODE */ { var object name = allocate_string(1); TheSnstring(name)->data[0] = ascii(c); return name; } #endif return NIL; } /* UP: find the character with the given name name_char(string) > string: String < result: character with the name, or NIL if does not exist */ global object name_char (object string) { { var const uintB* codes_ptr = &charname_table_codes[0]; var const gcv_object_t* strings_ptr = &charname_table[0]; var uintC count = charname_table_length; while (count--) { if (string_equal(string,*strings_ptr++)) /* compare string with charname_table[i] */ return code_char(as_chart(*codes_ptr)); /* return Code charname_table_codes[i] from the table */ codes_ptr++; } } /* no character with the name name found */ #ifdef ENABLE_UNICODE { var uintL len; var uintL offset; string = unpack_string_ro(string,&len,&offset); if (len > 1 && len < UNINAME_MAX) { var const chart* charptr; unpack_sstring_alloca(string,len,offset, charptr=); /* Test for Uxxxx or Uxxxxxxxx syntax. */ if ((len == 5 || len == 9) && (chareq(charptr[0],ascii('U')) || chareq(charptr[0],ascii('u')))) { /* decode a hexadecimal number: */ var uintL code = 0; var uintL index = 1; var const chart* tmpcharptr = charptr+1; while (1) { var cint c = as_cint(*tmpcharptr++); /* next character */ /* should be a hexadecimal digit: */ if (c > 'f') break; if (c >= 'a') { c -= 'a'-'A'; } if (c < '0') break; if (c <= '9') { c = c - '0'; } else if ((c >= 'A') && (c <= 'F')) { c = c - 'A' + 10; } else break; code = 16*code + c; /* put in the digit */ /* code should be < char_code_limit: */ if (code >= char_code_limit) break; /* should not occur */ index++; if (index == len) { /* Character name was "Uxxxx" with code = xxxx < char_code_limit. Its length should be 5 or 9, depending on xxxx < 0x10000. */ if (!(len == (code < 0x10000 ? 5 : 9))) break; /* Don't test for graphic_char_p - see comment in char_name(). This also avoids special-casing the #\Uxxxx syntax in io.d. */ /* if (!graphic_char_p(as_chart(code))) */ return code_char(as_chart(code)); } } } { /* Test for word1_word2_... syntax. */ var char buf[UNINAME_MAX]; var char* ptr = buf; while (1) { var cint c = as_cint(*charptr++); if (!(c >= ' ' && c <= '~')) break; *ptr++ = (char)(c == '_' ? ' ' : c); if (--len == 0) goto filled_buf; } if (false) { filled_buf: *ptr = '\0'; var cint32 code = unicode_name_character(buf); if (code != UNINAME_INVALID) return code_char(as_chart(code)); } } } } #else /* no ENABLE_UNICODE */ return coerce_char(string); #endif return NIL; } LISPFUNNF(standard_char_p,1) { /* (STANDARD-CHAR-P char), CLTL p. 234 (standard-char-p char) == (or (char= char #\Newline) (char<= #\Space char #\~)) Standard-Chars have a code c, with $20 <= c <= $7E or c = NL. */ var object arg = check_char(popSTACK()); var chart ch = char_code(arg); var cint c = as_cint(ch); VALUES_IF(standard_cint_p(c)); } LISPFUNNF(graphic_char_p,1) { /* (GRAPHIC-CHAR-P char), CLTL p. 234 */ var object arg = check_char(popSTACK()); VALUES_IF(graphic_char_p(char_code(arg))); } LISPFUNN(char_width,1) /* (CHAR-WIDTH char) */ { var object arg = check_char(popSTACK()); VALUES1(fixnum(char_width(char_code(arg)))); } LISPFUNNF(string_char_p,1) { /* (STRING-CHAR-P char), CLTL p. 235 - all characters are string-chars. */ var object arg = check_char(popSTACK()); VALUES1(T); } #if (base_char_code_limit < char_code_limit) LISPFUNN(base_char_p,1) /* (SYSTEM::BASE-CHAR-P char) */ { var object arg = check_char(popSTACK()); VALUES_IF(as_cint(char_code(arg)) < base_char_code_limit); } #endif LISPFUNNF(alpha_char_p,1) { /* (ALPHA-CHAR-P char), CLTL p. 235 - test with ALPHAP. */ var object arg = check_char(popSTACK()); VALUES_IF(alphap(char_code(arg))); } LISPFUNNF(upper_case_p,1) { /* (UPPER-CASE-P char), CLTL p. 235: upper-case-characters are those with a code c with 0 <= c < $100, that are different from (downcase char) . */ var object arg = check_char(popSTACK()); var chart ch = char_code(arg); VALUES_IF(!chareq(down_case(ch),ch)); } LISPFUNNF(lower_case_p,1) { /* (LOWER-CASE-P char), CLTL p. 235: lower-case-characters are those with a code c with 0 <= c < $100, that are different from (upcase char) . */ var object arg = check_char(popSTACK()); var chart ch = char_code(arg); VALUES_IF(!chareq(up_case(ch),ch)); } LISPFUNNF(both_case_p,1) { /* (BOTH-CASE-P char), CLTL p. 235 (both-case-p char) == (or (upper-case-p char) (lower-case-p char)) both-case-characters are those with a code c with 0 <= c < $100. For them (downcase char) and (upcase char) are different. */ var object arg = check_char(popSTACK()); var chart ch = char_code(arg); VALUES_IF(!chareq(down_case(ch),up_case(ch))); } /* UP: Checks an optional radix-argument test_radix_arg() > STACK_0: argument, default is 10 < result: radix, an integer >=2, <=36 removes one element from STACK can trigger GC */ local maygc uintWL test_radix_arg (void) { var object arg = popSTACK(); /* argument */ restart_radix_check: if (!boundp(arg)) return 10; if (posfixnump(arg)) { var uintV radix = posfixnum_to_V(arg); if ((2 <= radix) && (radix <= 36)) return radix; } pushSTACK(NIL); /* no PLACE */ pushSTACK(arg); /* TYPE-ERROR slot DATUM */ pushSTACK(O(type_radix)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(arg); pushSTACK(TheSubr(subr_self)->name); check_value(type_error,GETTEXT("~S: the radix must be an integer between 2 and 36, not ~S")); arg = value1; goto restart_radix_check; } LISPFUN(digit_char_p,seclass_foldable,1,1,norest,nokey,0,NIL) { /* (DIGIT-CHAR-P char [radix]), CLTL p. 236 method: test, if radix is an integer >=2 and <=36 . char must be a character <= 'z' , otherwise return NIL as result. if radix<=10: c must be >= '0' and < '0'+radix , else NIL. if radix>=10: c must be >= '0' and <= '9' or (upcase c) must be >= 'A' and < 'A'-10+radix , else NIL. */ var uintWL radix = test_radix_arg(); /* basis >=2, <=36 */ var object arg = check_char(popSTACK()); var chart ch = char_code(arg); var cint c = as_cint(ch); #ifdef ENABLE_UNICODE switch (c >> 8) { case 0x00: /* ASCII */ if ((c >= 0x0030) && (c <= 0x0039)) { c -= 0x0030; break; } if ((c >= 0x0041) && (c <= 0x005a)) { c -= 0x0037; break; } if ((c >= 0x0061) && (c <= 0x007a)) { c -= 0x0057; break; } goto no; case 0x06: /* (EXTENDED)?_ARABIC-INDIC_DIGIT_* */ if ((c >= 0x0660) && (c <= 0x0669)) { c -= 0x0660; break; } if ((c >= 0x06f0) && (c <= 0x06f9)) { c -= 0x06f0; break; } goto no; case 0x09: /* DEVANAGARI_DIGIT_*, BENGALI_DIGIT_* */ if ((c >= 0x0966) && (c <= 0x096f)) { c -= 0x0966; break; } if ((c >= 0x09e6) && (c <= 0x09ef)) { c -= 0x09e6; break; } goto no; case 0x0A: /* GURMUKHI_DIGIT_*, GUJARATI_DIGIT_* */ if ((c >= 0x0a66) && (c <= 0x0a6f)) { c -= 0x0a66; break; } if ((c >= 0x0ae6) && (c <= 0x0aef)) { c -= 0x0ae6; break; } goto no; case 0x0B: /* ORIYA_DIGIT_*, TAMIL_DIGIT_* */ if ((c >= 0x0b66) && (c <= 0x0b6f)) { c -= 0x0b66; break; } if ((c >= 0x0be7) && (c <= 0x0bef)) { c -= 0x0be6; break; } goto no; case 0x0C: /* TELUGU_DIGIT_*, KANNADA_DIGIT_* */ if ((c >= 0x0c66) && (c <= 0x0c6f)) { c -= 0x0c66; break; } if ((c >= 0x0ce6) && (c <= 0x0cef)) { c -= 0x0ce6; break; } goto no; case 0x0D: /* MALAYALAM_DIGIT_* */ if ((c >= 0x0d66) && (c <= 0x0d6f)) { c -= 0x0d66; break; } goto no; case 0x0E: /* THAI_DIGIT_*, LAO_DIGIT_* */ if ((c >= 0x0e50) && (c <= 0x0e59)) { c -= 0x0e50; break; } if ((c >= 0x0ed0) && (c <= 0x0ed9)) { c -= 0x0ed0; break; } goto no; case 0x0F: /* TIBETAN_DIGIT_* */ if ((c >= 0x0f20) && (c <= 0x0f29)) { c -= 0x0f20; break; } goto no; case 0x10: /* MYANMAR_DIGIT_* */ if ((c >= 0x1040) && (c <= 0x1049)) { c -= 0x1040; break; } goto no; case 0x13: /* ETHIOPIC_DIGIT_* */ if ((c >= 0x1369) && (c <= 0x1371)) { c -= 0x1369; break; } goto no; case 0x17: /* KHMER_DIGIT_* */ if ((c >= 0x17E0) && (c <= 0x17E9)) { c -= 0x17E0; break; } goto no; case 0x18: /* MONGOLIAN_DIGIT_* */ if ((c >= 0x1810) && (c <= 0x1819)) { c -= 0x1810; break; } goto no; case 0xFF: /* FULLWIDTH_DIGIT_* */ if ((c >= 0xff10) && (c <= 0xff19)) { c -= 0xff10; break; } goto no; case 0x1d7: /* MATHEMATICAL_* SANS-SERIF/BOLD DOUBLE-STRUCK MONOSPACE */ if ((c >= 0x1d7ce) && (c <= 0x1d7ff)) { c -= 0x1d7ce; c %= 10; break; } goto no; default: goto no; } #else if (c > 'z') goto no; /* too big -> no */ if (c >= 'a') { c -= 'a'-'A'; } /* convert character >='a',<='z' into uppercase letter */ /* now: $00 <= ch <= $60. */ if (c < '0') goto no; /* convert $30 <= c <= $60 into number value: */ if (c <= '9') { c = c - '0'; } else if (c >= 'A') { c = c - 'A' + 10; } else goto no; #endif /* now, c is the number value of the digit, >=0, <=41. */ if (c >= radix) goto no; /* only valid, if 0 <= c < radix. */ /* return value as fixnum: */ VALUES1(fixnum(c)); return; no: VALUES1(NIL); return; } LISPFUNNF(alphanumericp,1) { /* (ALPHANUMERICP char), CLTL p. 236 alphanumeric characters are the digits '0',...,'9' and the alphabetic characters. */ var object arg = check_char(popSTACK()); VALUES_IF(alphanumericp(char_code(arg))); } /* comparison functions for characters: The comparisons CHAR=,... compare the entire oint (or equivalent, only the cint). The comparisons CHAR-EQUAL,... convert the codes into uppercase letters and compare those. */ /* UP: tests, if all argcount+1 arguments below args_pointer are characters. if not, Error. > argcount: number of arguments - 1 > args_pointer: pointer to the arguments can trigger GC */ local maygc void test_char_args (uintC argcount, gcv_object_t* args_pointer) { do { var gcv_object_t* argptr = &NEXT(args_pointer); var object arg = *argptr; /* next argument */ if (!charp(arg)) /* must be a character */ *argptr = check_char(arg); } while (argcount--); /* sic: not --argcount! */ } /* UP: tests, if all argcount+1 arguments below args_pointer are characters. If not, error. Discards bits and font and transforms them into uppercase letters. > argcount: number of arguments - 1 > args_pointer: pointer to the arguments can trigger GC */ local maygc void test_char_args_upcase (uintC argcount, gcv_object_t* args_pointer) { do { var gcv_object_t* argptr = &NEXT(args_pointer); var object arg = *argptr; /* next argument */ if (!charp(arg)) /* must be a character */ arg = check_char(arg); /* replace by uppercase letters: */ *argptr = code_char(up_case(char_code(arg))); } while (argcount--); /* sic: not --argcount! */ } /* UP: (CHAR= char {char}) for checked arguments */ local Values char_eq (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 arguments Arg[0..n]. x:=Arg[n]. for i:=n-1 to 0 step -1 do ( if Arg[i]/=x then return(NIL) ), return(T). */ var object x = popSTACK(); /* take last argument */ while (argcount--) { if (!eq(popSTACK(),x)) goto no; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } /* UP: (CHAR/= char {char}) for checked arguments */ local Values char_noteq (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 arguments Arg[0..n]. for j:=n-1 to 0 step -1 do x:=Arg[j+1], for i:=j to 0 step -1 do if Arg[i]=x then return(NIL), return(T). */ var gcv_object_t* arg_j_ptr = args_end_pointer; var uintC j = argcount; while (j!=0) { var object x = BEFORE(arg_j_ptr); /* second last argument */ /* compare with all previous arguments: */ var gcv_object_t* arg_i_ptr = arg_j_ptr; var uintC i = j; do { if (eq(BEFORE(arg_i_ptr),x)) goto no; } while (--i); j--; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } /* UP: (CHAR< char {char}) for checked arguments */ local Values char_less (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 Arguments Arg[0..n]. for i:=n to 1 step -1 do x:=Arg[i], if x char<= Arg[i-1] then return(NIL), return(T). */ while (argcount--) { var object x = popSTACK(); if (as_oint(x) <= as_oint(STACK_0)) goto no; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } /* UP: (CHAR> char {char}) for checked arguments */ local Values char_greater (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 arguments Arg[0..n]. for i:=n to 1 step -1 do x:=Arg[i], if x char>= Arg[i-1] then return(NIL), return(T). */ while (argcount--) { var object x = popSTACK(); if (as_oint(x) >= as_oint(STACK_0)) goto no; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } /* UP: (CHAR<= char {char}) for checked arguments */ local Values char_ltequal (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 arguments Arg[0..n]. for i:=n to 1 step -1 do x:=Arg[i], if x char< Arg[i-1] then return(NIL), return(T). */ while (argcount--) { var object x = popSTACK(); if (as_oint(x) < as_oint(STACK_0)) goto no; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } /* UP: (CHAR>= char {char}) for checked arguments */ local Values char_gtequal (uintC argcount, gcv_object_t* args_pointer) { /* method: n+1 arguments Arg[0..n]. for i:=n to 1 step -1 do x:=Arg[i], if x char> Arg[i-1] then return(NIL), return(T). */ while (argcount--) { var object x = popSTACK(); if (as_oint(x) > as_oint(STACK_0)) goto no; } yes: VALUES1(T); goto ok; no: VALUES1(NIL); goto ok; ok: set_args_end_pointer(args_pointer); } LISPFUN(char_eq,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR= char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_eq(argcount,args_pointer); } LISPFUN(char_noteq,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR/= char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_noteq(argcount,args_pointer); } LISPFUN(char_less,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR< char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_less(argcount,args_pointer); } LISPFUN(char_greater,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR> char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_greater(argcount,args_pointer); } LISPFUN(char_ltequal,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR<= char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_ltequal(argcount,args_pointer); } LISPFUN(char_gtequal,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR>= char {char}), CLTL p. 237 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args(argcount,args_pointer); return_Values char_gtequal(argcount,args_pointer); } LISPFUN(char_equal,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-EQUAL char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_eq(argcount,args_pointer); } LISPFUN(char_not_equal,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-NOT-EQUAL char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_noteq(argcount,args_pointer); } LISPFUN(char_lessp,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-LESSP char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_less(argcount,args_pointer); } LISPFUN(char_greaterp,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-GREATERP char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_greater(argcount,args_pointer); } LISPFUN(char_not_greaterp,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-NOT-GREATERP char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_ltequal(argcount,args_pointer); } LISPFUN(char_not_lessp,seclass_foldable,1,0,rest,nokey,0,NIL) { /* (CHAR-NOT-LESSP char {char}), CLTL p. 239 */ var gcv_object_t* args_pointer = rest_args_pointer STACKop 1; test_char_args_upcase(argcount,args_pointer); return_Values char_gtequal(argcount,args_pointer); } LISPFUNNF(char_code,1) { /* (CHAR-CODE char), CLTL p. 239 */ var object arg = check_char(popSTACK()); VALUES1(fixnum(as_cint(char_code(arg)))); /* ascii-code as fixnum */ } LISPFUNNF(code_char,1) { /* (CODE-CHAR code) */ var object codeobj = popSTACK(); /* code-argument */ if (!integerp(codeobj)) { /* code-argument is not an integer. */ pushSTACK(codeobj); /* TYPE-ERROR slot DATUM */ pushSTACK(S(integer)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(codeobj); pushSTACK(TheSubr(subr_self)->name); error(type_error, GETTEXT("~S: the code argument should be an integer, not ~S")); } /* codeobj is now an integer. */ var uintV code; /* test, if 0 <= code < char_code_limit : */ if (posfixnump(codeobj) && ((code = posfixnum_to_V(codeobj)) < char_code_limit)) { VALUES1(code_char(as_chart(code))); /* handicraft character */ } else { VALUES1(NIL); /* else value NIL */ } } LISPFUNNR(character,1) { /* (CHARACTER object), CLTL p. 241 */ var object trial = coerce_char(STACK_0); /* convert argument into character */ if (nullp(trial)) { /* unsuccessfully? */ /* Argument still in STACK_0, TYPE-ERROR slot DATUM */ pushSTACK(O(type_designator_character)); /* TYPE-ERROR slot EXPECTED-TYPE*/ pushSTACK(STACK_1); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: cannot coerce ~S to a character")); } else { VALUES1(trial); skipSTACK(1); } } LISPFUNNF(char_upcase,1) { /* (CHAR-UPCASE char), CLTL p. 241 */ var object arg = check_char(popSTACK()); VALUES1(code_char(up_case(char_code(arg)))); /* convert into uppercase letters */ } LISPFUNNF(char_downcase,1) { /* (CHAR-DOWNCASE char), CLTL p. 241 */ var object arg = check_char(popSTACK()); VALUES1(code_char(down_case(char_code(arg)))); /* convert into lowercase letters */ } LISPFUN(digit_char,seclass_foldable,1,1,norest,nokey,0,NIL) { /* (DIGIT-CHAR weight [radix]), CLTL2 p. 384 method: all arguments have to be integers, radix between 2 and 36. if 0 <= weight < radix, construct a character from '0',...,'9','A',...,'Z' with value weight. else value NIL. */ var uintWL radix = test_radix_arg(); /* radix-argument, >=2, <=36 */ var object weightobj = popSTACK(); /* weight-argument */ if (!integerp(weightobj)) { /* weight-Argument is not an integer. */ pushSTACK(weightobj); /* TYPE-ERROR slot DATUM */ pushSTACK(S(integer)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(weightobj); pushSTACK(TheSubr(subr_self)->name); error(type_error, GETTEXT("~S: the weight argument should be an integer, not ~S")); } /* weightobj is now an integer. */ /* test, if 0<=weight '9') weight += 'A'-'0'-10; /* or turn it into a letter */ VALUES1(ascii_char(weight)); /* handicraft character */ } else VALUES1(NIL); } LISPFUNNF(char_int,1) { /* (CHAR-INT char), CLTL p. 242 */ var object arg = check_char(popSTACK()); VALUES1(fixnum(as_cint(char_code(arg)))); } LISPFUNNF(int_char,1) { /* (INT-CHAR integer), CLTL p. 242 */ var object arg = popSTACK(); /* integer-Argument */ if (integerp(arg)) { /* turn into a character if 0 <= arg < char_code_limit, else NIL */ var uintV i; if ((posfixnump(arg)) && ((i = posfixnum_to_V(arg)) < char_code_limit)) { VALUES1(code_char(as_chart(i))); } else { VALUES1(NIL); } } else { /* arg not an integer -> error: */ pushSTACK(arg); /* TYPE-ERROR slot DATUM */ pushSTACK(S(integer)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(arg); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: argument should be an integer, not ~S")); } } LISPFUNNF(char_name,1) { /* (CHAR-NAME char), CLTL p. 242 */ var object arg = check_char(popSTACK()); VALUES1(char_name(char_code(arg))); } /* Support for case-inverted packages. */ /* Converts a character to opposite case. invert_case(ch) > ch: a character < result: a character, either ch or up_case(ch) or down_case(ch) Note that always invert_case(invert_case(ch)) == ch. */ global chart invert_case (chart ch) { var chart up = up_case(ch); if (!chareq(ch,up)) return up; var chart down = down_case(ch); if (!chareq(ch,down)) return down; return ch; } LISPFUNNF(char_invertcase,1) { /* (EXT:CHAR-INVERTCASE char) */ var object arg = check_char(popSTACK()); VALUES1(code_char(invert_case(char_code(arg)))); } /* UP: compares two strings of equal length for equality modulo case-invert > string1,offset1: here are the addressed characters in string1 > string2,offset2: here are the addressed characters in string2 > len: number of addressed characters in String1 and in String2, > 0 < result: true if equal, else false. */ local bool string_eqcomp_inverted (object string1, uintL offset1, object string2, uintL offset2, uintL len) { SstringDispatch(string1,X1, { var const cintX1* charptr1 = &((SstringX1)TheVarobject(string1))->data[offset1]; SstringDispatch(string2,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(string2))->data[offset2]; do { if (!chareq(invert_case(as_chart(*charptr1++)),as_chart(*charptr2++))) goto no; } while (--len); }); }); return true; no: return false; } /* UP: compares two strings for equality modulo case-invert string_eq_inverted(string1,string2) > string1: string > string2: simple-string < result: /=0, if equal modulo case-invert */ global bool string_eq_inverted (object string1, object string2) { var uintL len1; var uintL offset1; string1 = unpack_string_ro(string1,&len1,&offset1); sstring_un_realloc(string2); if (len1 != Sstring_length(string2)) return false; /* Now both strings have exactly len1 characters. Compare them. */ if (len1 > 0) return string_eqcomp_inverted(string1,offset1,string2,0,len1); return true; } /* UP: converts a string piece to opposite case, uppercase characters to lowercase and lowercase characters to uppercase. nstring_invertcase(dv,offset,len); > object dv: the character storage vector > uintL offset: index of first affected character > uintL len: number of affected characters can trigger GC */ local maygc void nstring_invertcase (object dv, uintL offset, uintL len) { restart_it: if (len > 0) SstringCase(dv,{ do { var chart ch = invert_case(as_chart(TheS8string(dv)->data[offset])); if (as_cint(ch) < cint8_limit) { TheS8string(dv)->data[offset] = as_cint(ch); offset++; len--; } else { dv = sstring_store(dv,offset,ch); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } } while (len > 0); },{ do { var chart ch = invert_case(as_chart(TheS16string(dv)->data[offset])); if (as_cint(ch) < cint16_limit) { TheS16string(dv)->data[offset] = as_cint(ch); offset++; len--; } else { dv = sstring_store(dv,offset,ch); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } } while (len > 0); },{ var cint32* charptr = &TheS32string(dv)->data[offset]; do { *charptr = as_cint(invert_case(as_chart(*charptr))); charptr++; } while (--len); },{ error_nilarray_retrieve(); }); } /* UP: converts a string to opposite case string_invertcase(string) > string: string < result: new normal-simple-string can trigger GC */ global maygc object string_invertcase (object string) { string = copy_string_normal(string); /* copy and turn into a normal-simple-string */ pushSTACK(string); nstring_invertcase(string,0,Sstring_length(string)); /* convert */ string = popSTACK(); DBGREALLOC(string); return string; } /* error, if index-argument is not an integer. */ local _Noreturn void error_int (object kw, object obj) { pushSTACK(obj); /* TYPE-ERROR slot DATUM */ pushSTACK(S(integer)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(obj); if (eq(kw,nullobj)) { pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: index should be an integer, not ~S")); } else { pushSTACK(kw); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: ~S-index should be an integer, not ~S")); } } /* error, if index-argument is not an integer or NIL. */ local _Noreturn void error_int_null (object kw, object obj) { pushSTACK(obj); /* TYPE-ERROR slot DATUM */ pushSTACK(O(type_end_index)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(obj); if (eq(kw,nullobj)) { pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: index should be NIL or an integer, not ~S")); } else { pushSTACK(kw); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: ~S-index should be NIL or an integer, not ~S")); } } /* error, if index-argument is not <= limit. */ local _Noreturn void error_cmp_inclusive (object kw, object obj, uintL grenze) { pushSTACK(obj); /* TYPE-ERROR slot DATUM */ pushSTACK(NIL); pushSTACK(obj); { var object tmp; pushSTACK(S(integer)); pushSTACK(Fixnum_0); pushSTACK(UL_to_I(grenze)); tmp = listof(3); STACK_1 = tmp; /* TYPE-ERROR slot EXPECTED-TYPE */ } if (eq(kw,nullobj)) { pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: index ~S should not be greater than the length of the string")); } else { pushSTACK(kw); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: ~S-index ~S should not be greater than the length of the string")); } } /* error, if index-argument is not < limit. */ local _Noreturn void error_cmp_exclusive (object kw, object obj, uintL grenze) { pushSTACK(obj); /* TYPE-ERROR slot DATUM */ pushSTACK(NIL); pushSTACK(obj); { var object tmp; pushSTACK(S(integer)); pushSTACK(Fixnum_0); pushSTACK(UL_to_I(grenze)); tmp = listof(1); pushSTACK(tmp); tmp = listof(3); STACK_1 = tmp; /* TYPE-ERROR slot EXPECTED-TYPE */ } if (eq(kw,nullobj)) { pushSTACK(TheSubr(subr_self)->name); error(type_error, GETTEXT("~S: index ~S should be less than the length of the string")); } else { pushSTACK(kw); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: ~S-index ~S should be less than the length of the string")); } } /* Macro: checks an index-argument test_index(from,to_setter,def,default,uplimit_cmp,upper_limit,ucname,lcname) from : expression, where the index (as object) comes from. to_setter : assigns the result (as uintV) . def : 0 if we do not have to test for default values, 1 if the default is set in on unbound, 2 if the default is set in on unbound or NIL. default : expression, that serves as default value in this case. upper_limit : upper limit uplimit_cmp : comparison with upper limit kw : keyword, that identifies the index, or nullobj */ #define test_index(from,to_setter,def,default,uplimit_cmp,upper_limit,kw) \ { var object index = from; /* index-argument */ \ if (def && (!boundp(index) || (def == 2 && nullp(index)))) \ { to_setter default; } \ else { /* must be an integer: */ \ if (!integerp(index)) \ { if (def==2) error_int_null(kw,index); else error_int(kw,index); } \ /* index is an integer. */ \ if (!(positivep(index))) \ { error_pos_integer(kw,index); } \ /* index is >=0. */ \ if (!((posfixnump(index)) && \ ((to_setter posfixnum_to_V(index)) uplimit_cmp upper_limit))) { \ if (0 uplimit_cmp 0) \ /* "<= upper_limit" - comparison not satisfied (upper_limit == limit) */ \ { error_cmp_inclusive(kw,index,upper_limit); } \ else \ /* "< upper_limit" - comparison not satisfied (upper_limit == limit) */ \ { error_cmp_exclusive(kw,index,upper_limit); } \ } \ }} /* UP: check the index argument for string functions > STACK_0: Argument > len: length of the strings (< array-total-size-limit) < return: index in the string */ local uintL test_index_arg (uintL len) { var uintV i; /* i := Index STACK_0, no default value, must be totalsize; string = iarray_displace_check(string,len,&offset); } var uintL index = test_index_arg(len); VALUES1(code_char(schar(string,offset+index))); skipSTACK(2); } LISPFUNNR(schar,2) { /* (SCHAR string integer), CLTL p. 300 */ var object string = STACK_1; if (!simple_string_p(string)) { /* must be a simple-string */ if (stringp(string) && (Iarray_flags(string) & arrayflags_atype_mask) == Atype_NIL) error_nilarray_store(); else error_sstring(string); } sstring_un_realloc(string); var uintL index = test_index_arg(Sstring_length(string)); VALUES1(code_char(schar(string,index))); skipSTACK(2); } LISPFUNN(store_char,3) { /* (SYSTEM::STORE-CHAR string index newchar) = (SETF (CHAR string index) newchar), CLTL p. 300 */ STACK_2 = check_string(STACK_2); /* string-argument */ var object newchar = check_char(popSTACK()); /* newchar-Argument */ var object string = STACK_1; /* string-argument */ var uintL len; var uintL offset = 0; /* Don't use unpack_string_rw() -- we need (array-dimension string 0), not (length string). */ if (simple_string_p(string)) { sstring_un_realloc(string); len = Sstring_length(string); } else { len = TheIarray(string)->totalsize; string = iarray_displace_check(string,len,&offset); if (simple_nilarray_p(string)) error_nilarray_store(); } check_sstring_mutable(string); offset += test_index_arg(len); /* go to the element addressed by index */ sstring_store(string,offset,char_code(newchar)); /* put in the character */ VALUES1(newchar); skipSTACK(2); } LISPFUNN(store_schar,3) { /* (SYSTEM::STORE-SCHAR simple-string index newchar) = (SETF (SCHAR simple-string index) newchar), CLTL p. 300 */ var object newchar = check_char(popSTACK()); /* newchar-argument */ var object string = STACK_1; /* string-argument */ if (!simple_string_p(string)) { /* must be a simple-string */ if (stringp(string) && (Iarray_flags(string) & arrayflags_atype_mask) == Atype_NIL) error_nilarray_store(); else error_sstring(string); } sstring_un_realloc(string); check_sstring_mutable(string); var uintL offset = test_index_arg(Sstring_length(string)); /* go to the element addressed by index */ sstring_store(string,offset,char_code(newchar)); /* put in the character */ VALUES1(newchar); skipSTACK(2); } /* UP: checks :START and :END limits for a vector argument > STACK_1: optional :start-argument > STACK_0: optional :end-argument > stringarg arg: arg.string its data vector, [arg.offset .. arg.offset+arg.len-1] the range within the data vector corresponding to the entire vector-argument < stringarg arg: arg.string and arg.offset unchanged, [arg.offset+arg.index .. arg.offset+arg.index+arg.len-1] the range within the data vector corresponding to the selected vector slice removes 2 elements from STACK */ modexp void test_vector_limits (stringarg* arg) { if (arg->len > 0 && simple_nilarray_p(arg->string)) error_nilarray_retrieve(); var uintV start; var uintV end; /* arg->len is the length (<2^oint_data_len). check :START-argument: start := Index STACK_1, default value 0, must be <=len : */ test_index(STACK_1,start=,1,0,<=,arg->len,S(Kstart)); /* start is now the value of the :START-argument. check :END-argument: end := Index STACK_0, default value len, musst be <=len : */ test_index(STACK_0,end=,2,arg->len,<=,arg->len,S(Kend)); /* end is now the value of the :END-argument. compare :START and :END arguments: */ if (start > end) { pushSTACK(STACK_0); /* :END index */ pushSTACK(STACK_(1+1)); /* :START index */ pushSTACK(TheSubr(subr_self)->name); error(error_condition,GETTEXT("~S: :START-index ~S must not be greater than :END-index ~S")); } skipSTACK(2); /* fill results: */ arg->index = start; arg->len = end-start; } /* UP: checks the limits for a string-argument test_string_limits_ro(&arg) [for read-only access] > STACK_2: string-argument > STACK_1: optional :start-argument > STACK_0: optional :end-argument < stringarg arg: description of the argument < result: string-argument increases STACK by 3 can trigger GC */ global maygc object test_string_limits_ro (stringarg* arg) { /* check string-argument: */ STACK_2 = check_string(STACK_2); arg->string = unpack_string_ro(STACK_2,&arg->len,&arg->offset); test_vector_limits(arg); return popSTACK(); } /* UP: checks the limits for a string-argument test_string_limits_rw(&arg) [for read-write access] > STACK_2: string-argument > STACK_1: optional :start-argument > STACK_0: optional :end-argument < stringarg arg: description of the argument < result: string-argument increases STACK by 3 can trigger GC */ local maygc object test_string_limits_rw (stringarg* arg) { var object string = test_string_limits_ro(arg); if (arg->len > 0) { if (simple_nilarray_p(arg->string)) error_nilarray_access(); check_sstring_mutable(arg->string); } return string; } /* UP: checks a string/symbol/character-argument test_stringsymchar_arg(obj,invert) > obj: argument > invert: whether to implicitly case-invert a symbol's printname < result: argument as string can trigger GC */ global maygc object test_stringsymchar_arg (object obj, bool invert) { restart_stringsymchar: if (stringp(obj)) /* string: return unchanged */ return obj; if (symbolp(obj)) { /* symbol: use print name */ obj = TheSymbol(obj)->pname; return (invert ? string_invertcase(obj) : obj); } if (charp(obj)) { /* character: turn it into a one-element string: */ var object new_string = allocate_string(1); TheSnstring(new_string)->data[0] = char_code(obj); return new_string; } pushSTACK(NIL); /* no PLACE */ pushSTACK(obj); /* TYPE-ERROR slot DATUM */ pushSTACK(O(type_stringsymchar)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(obj); pushSTACK(TheSubr(subr_self)->name); check_value(type_error,GETTEXT("~S: argument ~S should be a string, a symbol or a character")); obj = value1; goto restart_stringsymchar; } /* UP: checks the limits for 1 string/symbol-argument and copies it test_1_stringsym_limits(invert,&string,&len) > STACK_2: string/symbol-argument > STACK_1: optional :start-argument > STACK_0: optional :end-argument > invert: whether to implicitly case-invert a symbol's printname < object string: copy of the string < uintL offset: index of first affected character < uintL len: number of affected characters increases STACK by 3 can trigger GC */ local maygc void test_1_stringsym_limits (bool invert, object* string_, uintL* offset_, uintL* len_) { var object string; var uintL len; var uintV start; var uintV end; /* check string/symbol-argument: */ string = test_stringsymchar_arg(STACK_2,invert); len = vector_length(string); /* now, len is the length (<2^oint_data_len). check :START-argument: start := Index STACK_1, default value 0, must be <=len : */ test_index(STACK_1,start=,1,0,<=,len,S(Kstart)); /* start is now the value of the :START-argument. check :END-argument: end := Index STACK_0, default value len, must be <=len : */ test_index(STACK_0,end=,2,len,<=,len,S(Kend)); /* end is now the value of the :END-argument. compare :START and :END arguments: */ if (!(start <= end)) { pushSTACK(STACK_0); /* :END-Index */ pushSTACK(STACK_2); /* :START-Index */ pushSTACK(TheSubr(subr_self)->name); error(error_condition,GETTEXT("~S: :START-index ~S must not be greater than :END-index ~S")); } skipSTACK(3); /* copy string and issue results: */ *string_ = copy_string(string); /* copy string */ *offset_ = start; *len_ = end-start; } /* UP: checks the limits for 2 string/symbol-arguments test_2_stringsym_limits(invert,&arg1,&arg2) > STACK_5: string/symbol-argument1 > STACK_4: string/symbol-argument2 > STACK_3: optional :start1-argument > STACK_2: optional :end1-argument > STACK_1: optional :start2-argument > STACK_0: optional :end2-argument > invert: whether to implicitly case-invert a symbol's printname < stringarg arg1: description of argument1 < stringarg arg2: description of argument2 increases STACK by 6 */ local void test_2_stringsym_limits (bool invert, stringarg* arg1, stringarg* arg2) { var uintL len1; var uintL len2; { /* check string/symbol-argument1: */ var object string1 = test_stringsymchar_arg(STACK_5,invert); pushSTACK(string1); /* save string1 */ /* check string/symbol-argument2: */ var object string2 = test_stringsymchar_arg(STACK_(4+1),invert); arg2->string = unpack_string_ro(string2,&len2,&arg2->offset); /* now, len2 is the length (<2^oint_data_len) of string2. */ string1 = popSTACK(); /* restore string1 */ arg1->string = unpack_string_ro(string1,&len1,&arg1->offset); /* now, len1 is the length (<2^oint_data_len) of string1. */ } { /* check :START1 and :END1: */ var uintV start1; var uintV end1; /* check :START1-argument: start1 := Index STACK_3, default value 0, must be <=len1: */ test_index(STACK_3,start1=,1,0,<=,len1,S(Kstart1)); /* start1 is now the value of the :START1-argument. check :END1-argument: end1 := Index STACK_2, default value len1, must be <=len1: */ test_index(STACK_2,end1=,2,len1,<=,len1,S(Kend1)); /* end1 is now the value of the :END1-argument. compare :START1 and :END1 arguments: */ if (!(start1 <= end1)) { pushSTACK(STACK_2); /* :END1-Index */ pushSTACK(STACK_4); /* :START1-Index */ pushSTACK(TheSubr(subr_self)->name); error(error_condition,GETTEXT("~S: :START1-index ~S must not be greater than :END1-index ~S")); } /* issue the results for string1: */ arg1->index = start1; arg1->len = end1-start1; if (arg1->len > 0 && simple_nilarray_p(arg1->string)) error_nilarray_retrieve(); } { /* check :START2 and :END2: */ var uintV start2; var uintV end2; /* check :START2-argument: start2 := Index STACK_1, default value 0, must be <=len2: */ test_index(STACK_1,start2=,1,0,<=,len2,S(Kstart2)); /* start2 is now the value of the :START2-argument. check :END2-argument: end2 := Index STACK_0, default value len2, must be <=len2: */ test_index(STACK_0,end2=,2,len2,<=,len2,S(Kend2)); /* end2 is now the value of the :END2-argument. compare :START2 and :END2 arguments: */ if (!(start2 <= end2)) { pushSTACK(STACK_0); /* :END2-Index */ pushSTACK(STACK_2); /* :START2-Index */ pushSTACK(TheSubr(subr_self)->name); error(error_condition,GETTEXT("~S: :START2-index ~S must not be greater than :END2-index ~S")); } /* issue the results for string2: */ arg2->index = start2; arg2->len = end2-start2; if (arg2->len > 0 && simple_nilarray_p(arg2->string)) error_nilarray_retrieve(); } /* done. */ skipSTACK(6); } /* UP: compares two strings of equal length for equality > string1,offset1: here are the addressed characters in string1 > string2,offset2: here are the addressed characters in string2 > len: number of addressed characters in String1 and in String2, > 0 < result: true if equal, else false. */ global bool string_eqcomp (object string1, uintL offset1, object string2, uintL offset2, uintL len) { SstringDispatch(string1,X1, { var const cintX1* charptr1 = &((SstringX1)TheVarobject(string1))->data[offset1]; SstringDispatch(string2,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(string2))->data[offset2]; do { if (!chareq(as_chart(*charptr1++),as_chart(*charptr2++))) goto no; } while (--len); }); }); return true; no: return false; } /* UP: compares two strings > arg1: here are the addressed characters in string1 > arg2: here are the addressed characters in string2 < arg1.index: location of the first difference i string1 < result: 0 if equal, -1 if string1 is genuinely lesser than string2, +1 if string1 is genuinely bigger than string2. */ local signean string_comp (stringarg* arg1, const stringarg* arg2) { var uintL len1 = arg1->len; var uintL len2 = arg2->len; SstringCase(arg1->string, { var const cint8* charptr1_0 = &TheS8string(arg1->string)->data[arg1->offset]; var const cint8* charptr1 = &charptr1_0[arg1->index]; /* one of the strings empty ? */ if (len1==0) goto A_string1_end; if (len2==0) goto A_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(as_chart(*charptr1++),as_chart(*charptr2++))) break; /* decrease both counters: */ len1--; len2--; /* one of the strings finished ? */ if (len1==0) goto A_string1_end; if (len2==0) goto A_string2_end; } /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(as_chart(*charptr1),as_chart(*--charptr2))) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ }); A_string1_end: /* string1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else /* String1 is a genuine starting piece of string2 */ return signean_minus; A_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }, { var const cint16* charptr1_0 = &TheS16string(arg1->string)->data[arg1->offset]; var const cint16* charptr1 = &charptr1_0[arg1->index]; /* one of the strings empty ? */ if (len1==0) goto B_string1_end; if (len2==0) goto B_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(as_chart(*charptr1++),as_chart(*charptr2++))) break; /* decrease both counters: */ len1--; len2--; /* one of the strings finished ? */ if (len1==0) goto B_string1_end; if (len2==0) goto B_string2_end; } /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(as_chart(*charptr1),as_chart(*--charptr2))) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ }); B_string1_end: /* string1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else /* String1 is a genuine starting piece of string2 */ return signean_minus; B_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }, { var const cint32* charptr1_0 = &TheS32string(arg1->string)->data[arg1->offset]; var const cint32* charptr1 = &charptr1_0[arg1->index]; /* one of the strings empty ? */ if (len1==0) goto C_string1_end; if (len2==0) goto C_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(as_chart(*charptr1++),as_chart(*charptr2++))) break; /* decrease both counters: */ len1--; len2--; /* one of the strings finished ? */ if (len1==0) goto C_string1_end; if (len2==0) goto C_string2_end; } /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(as_chart(*charptr1),as_chart(*--charptr2))) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ }); C_string1_end: /* string1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else /* string1 is a genuine starting piece of string2 */ return signean_minus; C_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }, { /* one of the strings empty ? */ if (len1==0) goto D_string1_end; if (len2==0) goto D_string2_end; error_nilarray_retrieve(); D_string1_end: /* string1 finished */ arg1->index = 0; if (len2==0) return signean_null; /* string1 = string2 */ else /* string1 is a genuine starting piece of string2 */ return signean_minus; D_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = 0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }); } /* (STRING= string1 string2 :start1 :end1 :start2 :end2), CLTL p. 300 */ LISPFUN(string_eq,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES_IF((arg1.len==arg2.len) && ((arg1.len==0) || string_eqcomp(arg1.string,arg1.offset+arg1.index, arg2.string,arg2.offset+arg2.index, arg1.len))); } /* (CS-COMMON-LISP:STRING= string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_eq,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES_IF((arg1.len==arg2.len) && ((arg1.len==0) || string_eqcomp(arg1.string,arg1.offset+arg1.index, arg2.string,arg2.offset+arg2.index, arg1.len))); } /* (STRING/= string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_noteq,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)==0 ? NIL : fixnum(arg1.index)); } /* (CS-COMMON-LISP:STRING/= string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_noteq,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)==0 ? NIL : fixnum(arg1.index)); } /* (STRING< string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_less,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)<0 ? fixnum(arg1.index) : NIL); } /* (CS-COMMON-LISP:STRING< string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_less,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)<0 ? fixnum(arg1.index) : NIL); } /* (STRING> string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_greater,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)>0 ? fixnum(arg1.index) : NIL); } /* (CS-COMMON-LISP:STRING> string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_greater,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)>0 ? fixnum(arg1.index) : NIL); } /* (STRING<= string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_ltequal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)<=0 ? fixnum(arg1.index) : NIL); } /* (CS-COMMON-LISP:STRING<= string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_ltequal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)<=0 ? fixnum(arg1.index) : NIL); } /* (STRING>= string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_gtequal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)>=0 ? fixnum(arg1.index) : NIL); } /* (CS-COMMON-LISP:STRING>= string1 string2 :start1 :end1 :start2 :end2) */ LISPFUN(cs_string_gtequal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(true,&arg1,&arg2); /* compare: */ VALUES1(string_comp(&arg1,&arg2)>=0 ? fixnum(arg1.index) : NIL); } /* UP: compares two strings of equal length for equality, case-insensitive > string1,offset1: here are the addressed characters in string1 > string2,offset2: here are the addressed characters in string2 > len: number of addressed characters in String1 and in String2, > 0 < result: true if equal, else false. */ global bool string_eqcomp_ci (object string1, uintL offset1, object string2, uintL offset2, uintL len) { SstringDispatch(string1,X1, { var const cintX1* charptr1 = &((SstringX1)TheVarobject(string1))->data[offset1]; SstringDispatch(string2,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(string2))->data[offset2]; do { if (!chareq(up_case(as_chart(*charptr1++)),up_case(as_chart(*charptr2++)))) goto no; } while (--len); }); }); return true; no: return false; } /* UP: compares two strings, case-insensitive > arg1: here are the addressed characters in string1 > arg2: here are the addressed characters in string2 < arg1.index: location of the first difference i string1 < result: 0 if equal, -1 if string1 is genuinely lesser than string2, +1 if string1 is genuinely bigger than string2. */ local signean string_comp_ci (stringarg* arg1, const stringarg* arg2) { var uintL len1 = arg1->len; var uintL len2 = arg2->len; SstringCase(arg1->string,{ var const cint8* charptr1_0 = &TheS8string(arg1->string)->data[arg1->offset]; var const cint8* charptr1 = &charptr1_0[arg1->index]; var chart ch1; var chart ch2; /* one of the strings empty ? */ if (len1==0) goto A_string1_end; if (len2==0) goto A_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(ch1 = up_case(as_chart(*charptr1++)), ch2 = up_case(as_chart(*charptr2++)))) break; /* decrease both counters: */ len1--; len2--; /* is one of the strings finished ? */ if (len1==0) goto A_string1_end; if (len2==0) goto A_string2_end; } }); /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(ch1,ch2)) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ A_string1_end: /* string1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else return signean_minus; /* string1 is a genuine starting piece of string2 */ A_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ },{ var const cint16* charptr1_0 = &TheS16string(arg1->string)->data[arg1->offset]; var const cint16* charptr1 = &charptr1_0[arg1->index]; var chart ch1; var chart ch2; /* one of the strings empty ? */ if (len1==0) goto B_string1_end; if (len2==0) goto B_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(ch1 = up_case(as_chart(*charptr1++)), ch2 = up_case(as_chart(*charptr2++)))) break; /* decrease both counters: */ len1--; len2--; /* is one of the strings finished ? */ if (len1==0) goto B_string1_end; if (len2==0) goto B_string2_end; } }); /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(ch1,ch2)) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ B_string1_end: /* string1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else return signean_minus; /* string1 is a genuine starting piece of string2 */ B_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ },{ var const cint32* charptr1_0 = &TheS32string(arg1->string)->data[arg1->offset]; var const cint32* charptr1 = &charptr1_0[arg1->index]; var chart ch1; var chart ch2; /* one of the strings empty ? */ if (len1==0) goto C_string1_end; if (len2==0) goto C_string2_end; SstringDispatch(arg2->string,X2, { var const cintX2* charptr2 = &((SstringX2)TheVarobject(arg2->string))->data[arg2->offset+arg2->index]; while (1) { /* compare next characters: */ if (!chareq(ch1 = up_case(as_chart(*charptr1++)), ch2 = up_case(as_chart(*charptr2++)))) break; /* decrease both counters: */ len1--; len2--; /* is one of the strings finished ? */ if (len1==0) goto C_string1_end; if (len2==0) goto C_string2_end; } }); /* two different characters are found */ arg1->index = --charptr1 - charptr1_0; if (charlt(ch1,ch2)) return signean_minus; /* string1 < string2 */ else return signean_plus; /* string1 > string2 */ C_string1_end: /* String1 finished */ arg1->index = charptr1 - charptr1_0; if (len2==0) return signean_null; /* string1 = string2 */ else return signean_minus; /* string1 is a genuine starting piece of string2 */ C_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = charptr1 - charptr1_0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }, { /* one of the strings empty ? */ if (len1==0) goto D_string1_end; if (len2==0) goto D_string2_end; error_nilarray_retrieve(); D_string1_end: /* string1 finished */ arg1->index = 0; if (len2==0) return signean_null; /* string1 = string2 */ else /* string1 is a genuine starting piece of string2 */ return signean_minus; D_string2_end: /* string2 is finished, string1 is not yet finished */ arg1->index = 0; return signean_plus; /* string2 is a genuine starting piece of string1 */ }); } /* (STRING-EQUAL string1 string2 :start1 :end1 :start2 :end2), CLTL p. 301 */ LISPFUN(string_equal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES_IF((arg1.len==arg2.len) && ((arg1.len==0) || string_eqcomp_ci(arg1.string,arg1.offset+arg1.index, arg2.string,arg2.offset+arg2.index, arg1.len))); } /* (STRING-NOT-EQUAL string1 string2 :start1 :end1 :start2 :end2), CLTL p. 302 */ LISPFUN(string_not_equal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp_ci(&arg1,&arg2)==0 ? NIL : fixnum(arg1.index)); } /* (STRING-LESSP string1 string2 :start1 :end1 :start2 :end2), CLTL p. 302 */ LISPFUN(string_lessp,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp_ci(&arg1,&arg2)<0 ? fixnum(arg1.index) : NIL); } /* (STRING-GREATERP string1 string2 :start1 :end1 :start2 :end2), CLTL p. 302 */ LISPFUN(string_greaterp,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp_ci(&arg1,&arg2)>0 ? fixnum(arg1.index) : NIL); } /* (STRING-NOT-GREATERP string1 string2 :start1 :end1 :start2 :end2), CLTL p. 302 */ LISPFUN(string_not_greaterp,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp_ci(&arg1,&arg2)<=0 ? fixnum(arg1.index) : NIL); } /* (STRING-NOT-LESSP string1 string2 :start1 :end1 :start2 :end2), CLTL p. 302 */ LISPFUN(string_not_lessp,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* compare: */ VALUES1(string_comp_ci(&arg1,&arg2)>=0 ? fixnum(arg1.index) : NIL); } /* UP: searches a string string1 in another string string2 > arg1: here are the addressed characters in string1 > arg2: here are the addressed characters in string2 > eqcomp: comparison function, &string_eqcomp or &string_eqcomp_ci < result: NIL if not found, position in string2 (as fixnum) if found. let eqcomp_fun_t be the type of such a comparison function: */ typedef bool (*eqcomp_fun_t) (object string1, uintL offset1, object string2, uintL offset2, uintL len); local object string_search (const stringarg* arg1, const stringarg* arg2, eqcomp_fun_t eqcomp) { var uintL len1 = arg1->len; var uintL len2 = arg2->len; if (len1 > len2) goto notfound; /* Only if len1<=len2, can string1 occur in string2. */ /* loop: for i=0..len2-len1: compare string1 with the len1 characters at charptr2[i]. Thereto, pass through loop len2-len1+1 times, growing charptr2 and start2. */ { var object string1 = arg1->string; var uintL offset1 = arg1->offset + arg1->index; var object string2 = arg2->string; var uintL offset2 = arg2->offset + arg2->index; var uintL count = len2-len1+1; if (len1==0) goto found; do { if ((*eqcomp)(string1,offset1,string2,offset2,len1)) /* compare */ goto found; offset2++; } while (--count); goto notfound; found: return fixnum(offset2 - arg2->offset); } notfound: return NIL; } LISPFUN(search_string_eq,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { /* (SYS::SEARCH-STRING= string1 string2 [:start1] [:end1] [:start2] [:end2]) = (search string1 string2 :test #'char= [:start1] [:end1] [:start2] [:end2]) */ var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* search string1 in string2: */ VALUES1(string_search(&arg1,&arg2,&string_eqcomp)); } LISPFUN(search_string_equal,seclass_read,2,0,norest,key,4, (kw(start1),kw(end1),kw(start2),kw(end2)) ) { /* (SYS::SEARCH-STRING-EQUAL string1 string2 [:start1] [:end1] [:start2] [:end2]) = (search string1 string2 :test #'char-equal [:start1] [:end1] [:start2] [:end2]) */ var stringarg arg1; var stringarg arg2; /* check arguments: */ test_2_stringsym_limits(false,&arg1,&arg2); /* search string1 in string2: */ VALUES1(string_search(&arg1,&arg2,&string_eqcomp_ci)); } LISPFUN(make_string,seclass_no_se,1,0,norest,key,2, (kw(initial_element),kw(element_type)) ) { /* (MAKE-STRING size :initial-element :element-type) */ var uintV size; /* check size: */ if (!posfixnump(STACK_2)) { /* size must be fixnum >= 0 */ pushSTACK(STACK_2); /* TYPE-ERROR slot DATUM */ pushSTACK(O(type_posfixnum)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(STACK_(2+2)); pushSTACK(TheSubr(subr_self)->name); error(type_error, GETTEXT("~S: the string length ~S should be nonnegative fixnum")); } size = posfixnum_to_V(STACK_2); check_stringsize(size); /* check element-type: */ if (boundp(STACK_0)) { var object eltype = STACK_0; if (!eq(eltype,S(character))) { /* Verify (SUBTYPEP eltype 'CHARACTER): */ pushSTACK(eltype); pushSTACK(S(character)); funcall(S(subtypep),2); if (nullp(value1)) { pushSTACK(STACK_0); /* eltype */ pushSTACK(S(character)); /* CHARACTER */ pushSTACK(S(Kelement_type)); /* :ELEMENT-TYPE */ pushSTACK(S(make_string)); error(error_condition,GETTEXT("~S: ~S argument must be a subtype of ~S, not ~S")); } } } var object new_string; /* maybe fill with initial-element: */ var object initial_element = STACK_1; if (!boundp(initial_element)) { /* Allocate a small-sstring, to save memory in the most frequent case. It will become wider as needed automatically. */ new_string = allocate_s8string(size); } else { if (!charp(initial_element)) { /* must be a character */ pushSTACK(initial_element); /* TYPE-ERROR slot DATUM */ pushSTACK(S(character)); /* TYPE-ERROR slot EXPECTED-TYPE */ pushSTACK(S(character)); pushSTACK(initial_element); pushSTACK(S(Kinitial_element)); pushSTACK(TheSubr(subr_self)->name); error(type_error,GETTEXT("~S: ~S argument ~S should be of type ~S")); } else { var chart ch = char_code(initial_element); #ifdef HAVE_SMALL_SSTRING var cint c = as_cint(ch); if (c < cint8_limit) { new_string = allocate_s8string(size); if (size !=0) { var cint8* pdata = TheS8string(new_string)->data; do { *pdata++ = c; } while (--size); } } else if (c < cint16_limit) { new_string = allocate_s16string(size); if (size !=0) { var cint16* pdata = TheS16string(new_string)->data; do { *pdata++ = c; } while (--size); } } else { new_string = allocate_s32string(size); if (size !=0) { var cint32* pdata = TheS32string(new_string)->data; do { *pdata++ = c; } while (--size); } } #else new_string = allocate_string(size); if (size!=0) { var chart* charptr = &TheSnstring(new_string)->data[0]; do { *charptr++ = ch; } while (--size); } #endif } } DBGREALLOC(new_string); VALUES1(new_string); skipSTACK(3); } LISPFUNNR(string_both_trim,4) /* (SYS::STRING-BOTH-TRIM character-bag-left character-bag-right string invertp) basic function for STRING-TRIM, STRING-LEFT-TRIM, STRING-RIGHT-TRIM, CLTL p. 302 method: (let ((l (length string))) (do ((i 0 (1+ i))) (nil) (when (or (= i l) (not (find (char string i) character-bag-left))) (do ((j l (1- j))) (nil) (when (or (= i j) (not (find (char string (1- j)) character-bag-right))) (return (if (and (= i 0) (= j l)) string (substring string i j)))))))) */ { var object invertp = popSTACK(); var object string = test_stringsymchar_arg(popSTACK(),!nullp(invertp)); /* convert argument into string */ pushSTACK(string); /* and back into stack again */ pushSTACK(fixnum(vector_length(string))); /* length as fixnum into the stack */ pushSTACK(Fixnum_0); /* i := 0 */ /* stack layout: bag-left, bag-right, string, l, i */ while (!eq(STACK_0,STACK_1)) { /* for i = l (both fixnums): loop done */ /* determine (char string i) : */ pushSTACK(STACK_2); pushSTACK(STACK_1); funcall(L(char),2); /* determine (find (char ...) character-bag-left) : */ pushSTACK(value1); pushSTACK(STACK_5); funcall(L(find),2); if (nullp(value1)) break; /* char not in character-bag-left -> loop done */ STACK_0 = fixnum_inc(STACK_0,1); /* i := (1+ i) */ } pushSTACK(STACK_1); /* j := l */ /* stack layout: bag-left, bag-right, string, l, i, j */ while (!eq(STACK_0,STACK_1)) { /* for j = i (both fixnums): loop done */ /* determine (char string (1- j)) : */ pushSTACK(STACK_3); pushSTACK(fixnum_inc(STACK_1,-1)); funcall(L(char),2); /* determine (find (char ...) character-bag-right) : */ pushSTACK(value1); pushSTACK(STACK_5); funcall(L(find),2); if (nullp(value1)) break; /* char not in character-bag-right -> loop done */ STACK_0 = fixnum_inc(STACK_0,-1); /* j := (1- j) */ } /* stack layout: bag-left, bag-right, string, l, i, j throw away the characters of the string with index =j : */ var object j = popSTACK(); var object i = popSTACK(); var object l = popSTACK(); string = popSTACK(); skipSTACK(2); if (eq(i,Fixnum_0) && eq(j,l)) { value1 = string; /* for i=0 and j=l, nothing to do, string as value */ } else { /* copy sub part of the indices >=i, 0) { SstringDispatch(arg.string,X, { var const cintX* charptr = &((SstringX)TheVarobject(arg.string))->data[arg.offset+arg.index]; do { width += char_width(as_chart(*charptr)); charptr++; } while (--len); }); } /* width <= 2*arg.len. */ VALUES1(UL_to_I(width)); } /* UP: converts the characters of a string piece into uppercase letters nstring_upcase(dv,offset,len); > object dv: the character storage vector > uintL offset: index of first affected character > uintL len: number of affected characters can trigger GC */ global maygc void nstring_upcase (object dv, uintL offset, uintL len) { restart_it: if (len > 0) SstringCase(dv,{ do { dv = sstring_store(dv,offset,up_case(as_chart(TheS8string(dv)->data[offset]))); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } while (len > 0); },{ do { dv = sstring_store(dv,offset,up_case(as_chart(TheS16string(dv)->data[offset]))); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } while (len > 0); },{ var cint32* charptr = &TheS32string(dv)->data[offset]; do { *charptr = as_cint(up_case(as_chart(*charptr))); charptr++; } while (--len); },{ error_nilarray_retrieve(); }); } /* UP: converts a string into uppercase letters string_upcase(string) > string: string < result: new normal-simple-string, in uppercase letters can trigger GC */ global maygc object string_upcase (object string) { string = copy_string_normal(string); /* copy and turn into a normal-simple-string */ pushSTACK(string); nstring_upcase(string,0,Sstring_length(string)); /* convert */ string = popSTACK(); DBGREALLOC(string); return string; } LISPFUN(nstring_upcase,seclass_default,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (NSTRING-UPCASE string :start :end), CLTL p. 304 */ var stringarg arg; var object string = test_string_limits_rw(&arg); pushSTACK(string); nstring_upcase(arg.string,arg.offset+arg.index,arg.len); VALUES1(popSTACK()); } LISPFUN(string_upcase,seclass_read,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (STRING-UPCASE string :start :end), CLTL p. 303 */ var object string; var uintL offset; var uintL len; test_1_stringsym_limits(false,&string,&offset,&len); pushSTACK(string); nstring_upcase(string,offset,len); string = popSTACK(); sstring_un_realloc(string); VALUES1(string); } /* UP: converts the characters of a string piece into lowercase letters nstring_downcase(dv,offset,len); > object dv: the character storage vector > uintL offset: index of first affected character > uintL len: number of affected characters can trigger GC */ global maygc void nstring_downcase (object dv, uintL offset, uintL len) { restart_it: if (len > 0) SstringCase(dv,{ do { dv = sstring_store(dv,offset,down_case(as_chart(TheS8string(dv)->data[offset]))); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } while (len > 0); },{ do { dv = sstring_store(dv,offset,down_case(as_chart(TheS16string(dv)->data[offset]))); offset++; len--; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; goto restart_it; } } while (len > 0); },{ var cint32* charptr = &TheS32string(dv)->data[offset]; do { *charptr = as_cint(down_case(as_chart(*charptr))); charptr++; } while (--len); },{ error_nilarray_retrieve(); }); } /* UP: converts a string into lowercase letters string_downcase(string) > string: string < result: new normal-simple-string, in lowercase letters can trigger GC */ global maygc object string_downcase (object string) { string = copy_string_normal(string); /* copy and turn into a normal-simple-string */ pushSTACK(string); nstring_downcase(string,0,Sstring_length(string)); /* convert */ string = popSTACK(); DBGREALLOC(string); return string; } LISPFUN(nstring_downcase,seclass_default,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (NSTRING-DOWNCASE string :start :end), CLTL p. 304 */ var stringarg arg; var object string = test_string_limits_rw(&arg); pushSTACK(string); nstring_downcase(arg.string,arg.offset+arg.index,arg.len); VALUES1(popSTACK()); } LISPFUN(string_downcase,seclass_read,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (STRING-DOWNCASE string :start :end), CLTL p. 303 */ var object string; var uintL offset; var uintL len; test_1_stringsym_limits(false,&string,&offset,&len); pushSTACK(string); nstring_downcase(string,offset,len); string = popSTACK(); sstring_un_realloc(string); VALUES1(string); } /* UP: converts the words of a string piece into words, that that start with a capital and continue with lowercase letters. nstring_capitalize(dv,offset,len); > object dv: the character storage vector > uintL offset: index of first affected character > uintL len: number of affected characters method: alternately, seach for beginning of a word (and do not convert) resp. search for end of word (and do convert). can trigger GC */ global maygc void nstring_capitalize (object dv, uintL offset, uintL len) { if (len > 0) { var chart ch; SstringCase(dv,{ /* Search the start of a word. */ search_wordstart_8: while (len!=0) { ch = as_chart(TheS8string(dv)->data[offset]); if (alphanumericp(ch)) goto wordstart_8; offset++; len--; } return; /* len = 0 -> string terminated */ /* Found the start of a word. */ wordstart_8: dv = sstring_store(dv,offset,up_case(ch)); while (1) { offset++; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; SstringCase(dv, NOTREACHED;, goto in_word_16;, goto in_word_32;, NOTREACHED; ); } in_word_8: if (--len==0) break; ch = as_chart(TheS8string(dv)->data[offset]); if (!alphanumericp(ch)) goto search_wordstart_8; dv = sstring_store(dv,offset,down_case(ch)); } return; /* len = 0 -> string terminated */ },{ /* Search the start of a word. */ search_wordstart_16: while (len!=0) { ch = as_chart(TheS16string(dv)->data[offset]); if (alphanumericp(ch)) goto wordstart_16; offset++; len--; } return; /* len = 0 -> string terminated */ /* Found the start of a word. */ wordstart_16: dv = sstring_store(dv,offset,up_case(ch)); while (1) { offset++; if (sstring_reallocatedp(TheSstring(dv))) { /* has it been reallocated? */ dv = TheSistring(dv)->data; SstringCase(dv, NOTREACHED;, NOTREACHED;, goto in_word_32;, NOTREACHED; ); } in_word_16: if (--len==0) break; ch = as_chart(TheS16string(dv)->data[offset]); if (!alphanumericp(ch)) goto search_wordstart_16; dv = sstring_store(dv,offset,down_case(ch)); } return; /* len = 0 -> string terminated */ },{ /* Search the start of a word. */ search_wordstart_32: while (len!=0) { ch = as_chart(TheS32string(dv)->data[offset]); if (alphanumericp(ch)) goto wordstart_32; offset++; len--; } return; /* len = 0 -> string terminated */ /* Found the start of a word. */ wordstart_32: TheS32string(dv)->data[offset] = as_cint(up_case(ch)); while (1) { offset++; in_word_32: if (--len==0) break; ch = as_chart(TheS32string(dv)->data[offset]); if (!alphanumericp(ch)) goto search_wordstart_32; TheS32string(dv)->data[offset] = as_cint(down_case(ch)); } return; /* len = 0 -> string terminated */ },{ error_nilarray_retrieve(); }); } } LISPFUN(nstring_capitalize,seclass_default,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (NSTRING-CAPITALIZE string :start :end), CLTL p. 304 */ var stringarg arg; var object string = test_string_limits_rw(&arg); pushSTACK(string); nstring_capitalize(arg.string,arg.offset+arg.index,arg.len); VALUES1(popSTACK()); } LISPFUN(string_capitalize,seclass_read,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (STRING-CAPITALIZE string :start :end), CLTL p. 303 */ var object string; var uintL offset; var uintL len; test_1_stringsym_limits(false,&string,&offset,&len); pushSTACK(string); nstring_capitalize(string,offset,len); string = popSTACK(); sstring_un_realloc(string); VALUES1(string); } LISPFUN(nstring_invertcase,seclass_default,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (EXT:NSTRING-INVERTCASE string :start :end) */ var stringarg arg; var object string = test_string_limits_rw(&arg); pushSTACK(string); nstring_invertcase(arg.string,arg.offset+arg.index,arg.len); VALUES1(popSTACK()); } LISPFUN(string_invertcase,seclass_read,1,0,norest,key,2, (kw(start),kw(end)) ) { /* (EXT:STRING-INVERTCASE string :start :end) */ var object string; var uintL offset; var uintL len; test_1_stringsym_limits(false,&string,&offset,&len); pushSTACK(string); nstring_invertcase(string,offset,len); string = popSTACK(); sstring_un_realloc(string); VALUES1(string); } /* (STRING object), CLTL p. 304 */ LISPFUNNR(string,1) { VALUES1(test_stringsymchar_arg(popSTACK(),false)); } /* (CS-COMMON-LISP:STRING object) */ LISPFUNNR(cs_string,1) { VALUES1(test_stringsymchar_arg(popSTACK(),true)); } /* (NAME-CHAR name), CLTL p. 243 */ LISPFUNNR(name_char,1) { /* Convert argument into a string. (Case is not significant here.) Then search a character with this name: */ VALUES1(name_char(test_stringsymchar_arg(popSTACK(),false))); } /* UP: Returns a substring of a simple-string. subsstring(string,start,end) > object string: a simple-string > uintL start: start index > uintL end: end index with 0 <= start <= end <= Sstring_length(string) < object result: (subseq string start end), a freshly created normal-simple-string can trigger GC */ global maygc object subsstring (object string, uintL start, uintL end) { var uintL count = end - start; pushSTACK(string); var object new_string = allocate_string(count); string = popSTACK(); if (count > 0) { #ifdef ENABLE_UNICODE SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[start], &TheS32string(new_string)->data[0],count); }, { copy_16bit_32bit(&TheS16string(string)->data[start], &TheS32string(new_string)->data[0],count); }, { copy_32bit_32bit(&TheS32string(string)->data[start], &TheS32string(new_string)->data[0],count); }, { error_nilarray_retrieve(); }); #else SstringCase(string, { NOTREACHED; }, { NOTREACHED; }, { copy_8bit_8bit(&TheS8string(string)->data[start], &TheS8string(new_string)->data[0],count); }, { error_nilarray_retrieve(); }); #endif } DBGREALLOC(new_string); return new_string; } LISPFUN(substring,seclass_read,2,1,norest,nokey,0,NIL) { /* (SUBSTRING string start [end]) like SUBSEQ, but only for strings */ var object string; var uintL len; var uintV start; var uintV end; /* check string/symbol-argument: */ /* FIXME: This does the wrong thing in a case-sensitive package. */ string = test_stringsymchar_arg(STACK_2,false); len = vector_length(string); /* now, len is the length (<2^oint_data_len). check :START-argument: start := Index STACK_1, default value 0, must be <=len: */ test_index(STACK_1,start=,1,0,<=,len,S(Kstart)); /* start is now the value of the :START-argument. check :end-argument: end := Index STACK_0, default value len, must be <=len: */ test_index(STACK_0,end=,2,len,<=,len,S(Kend)); /* end is now the value of the :END-argument. compare :START and :END arguments: */ if (start > end) { pushSTACK(STACK_0); /* :END-Index */ pushSTACK(STACK_2); /* :START-Index */ pushSTACK(TheSubr(subr_self)->name); error(error_condition,GETTEXT("~S: :START-index ~S must not be greater than :END-index ~S")); } skipSTACK(3); /* extract substring: */ pushSTACK(string); /* save old string */ var uintL count = end-start; /* number of characters to be copied */ var object new_string = allocate_string(count); /* new string */ string = popSTACK(); /* old string */ if (count > 0) { var uintL len; /* again the length of the old string */ var uintL offset; string = unpack_string_ro(string,&len,&offset); #ifdef ENABLE_UNICODE SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[offset+start], &TheS32string(new_string)->data[0],count); }, { copy_16bit_32bit(&TheS16string(string)->data[offset+start], &TheS32string(new_string)->data[0],count); }, { copy_32bit_32bit(&TheS32string(string)->data[offset+start], &TheS32string(new_string)->data[0],count); }, { error_nilarray_retrieve(); }); #else SstringCase(string, { NOTREACHED; }, { NOTREACHED; }, { copy_8bit_8bit(&TheS8string(string)->data[offset+start], &TheS8string(new_string)->data[0],count); }, { error_nilarray_retrieve(); }); #endif } DBGREALLOC(new_string); VALUES1(new_string); } /* UP: concatenates several strings. string_concat(argcount) > uintC argcount: number of arguments > on the STACK: the arguments (should be strings) < result: total string, freshly created < STACK: cleaned up can trigger GC */ modexp maygc object string_concat (uintC argcount) { var gcv_object_t* args_pointer = (args_end_pointer STACKop argcount); /* args_pointer = pointer to the arguments check, if they are all strings, and add the lengths: */ var uintL total_length = 0; if (argcount > 0) { var gcv_object_t* argptr = args_pointer; var uintC count = argcount; do { var gcv_object_t *arg = &(NEXT(argptr)); /* next argument */ if (!stringp(*arg)) *arg = check_string(*arg); total_length += vector_length(*arg); } while (--count); } /* total_length is now the total length. */ check_stringsize(total_length); var object new_string = allocate_string(total_length); /* new string */ if (argcount > 0) { var cint32* charptr2 = &TheS32string(new_string)->data[0]; var gcv_object_t* argptr = args_pointer; do { var object arg = NEXT(argptr); /* next argument-string */ var uintL len; /* its length */ var uintL offset; var object string = unpack_string_ro(arg,&len,&offset); if (len > 0) { /* copy len characters from string to charptr2: */ #ifdef ENABLE_UNICODE SstringCase(string, { copy_8bit_32bit(&TheS8string(string)->data[offset], charptr2,len); }, { copy_16bit_32bit(&TheS16string(string)->data[offset], charptr2,len); }, { copy_32bit_32bit(&TheS32string(string)->data[offset], charptr2,len); }, { error_nilarray_retrieve(); }); #else SstringCase(string, { NOTREACHED; }, { NOTREACHED; }, { copy_8bit_8bit(&TheS8string(string)->data[offset],charptr2,len); }, { error_nilarray_retrieve(); }); #endif charptr2 += len; } } while (--argcount > 0); } set_args_end_pointer(args_pointer); /* clean up STACK */ DBGREALLOC(new_string); return new_string; } LISPFUN(string_concat,seclass_read,0,0,rest,nokey,0,NIL) { /* (STRING-CONCAT {string}) creates a string by concatenating the arguments */ VALUES1(string_concat(argcount)); }