1 /* 2 * Copyright (C) 1999-2002 Free Software Foundation, Inc. 3 * This file is part of the GNU LIBICONV Library. 4 * 5 * The GNU LIBICONV Library is free software; you can redistribute it 6 * and/or modify it under the terms of the GNU Library General Public 7 * License as published by the Free Software Foundation; either version 2 8 * of the License, or (at your option) any later version. 9 * 10 * The GNU LIBICONV Library is distributed in the hope that it will be 11 * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of 12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13 * Library General Public License for more details. 14 * 15 * You should have received a copy of the GNU Library General Public 16 * License along with the GNU LIBICONV Library; see the file COPYING.LIB. 17 * If not, write to the Free Software Foundation, Inc., 51 Franklin Street, 18 * Fifth Floor, Boston, MA 02110-1301, USA. 19 */ 20 21 /* 22 * JAVA 23 * This is ISO 8859-1 with \uXXXX escape sequences, denoting Unicode BMP 24 * characters. Consecutive pairs of \uXXXX escape sequences in the surrogate 25 * range, as in UTF-16, denote Unicode characters outside the BMP. 26 */ 27 28 static int 29 java_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n) 30 { 31 unsigned char c; 32 ucs4_t wc, wc2; 33 int i; 34 35 c = s[0]; 36 if (c != '\\') { 37 *pwc = c; 38 return 1; 39 } 40 if (n < 2) 41 return RET_TOOFEW(0); 42 if (s[1] != 'u') 43 goto simply_backslash; 44 wc = 0; 45 for (i = 2; i < 6; i++) { 46 if (n <= i) 47 return RET_TOOFEW(0); 48 c = s[i]; 49 if (c >= '0' && c <= '9') 50 c -= '0'; 51 else if (c >= 'A' && c <= 'Z') 52 c -= 'A'-10; 53 else if (c >= 'a' && c <= 'z') 54 c -= 'a'-10; 55 else 56 goto simply_backslash; 57 wc |= (ucs4_t) c << (4 * (5-i)); 58 } 59 if (!(wc >= 0xd800 && wc < 0xe000)) { 60 *pwc = wc; 61 return 6; 62 } 63 if (wc >= 0xdc00) 64 goto simply_backslash; 65 if (n < 7) 66 return RET_TOOFEW(0); 67 if (s[6] != '\\') 68 goto simply_backslash; 69 if (n < 8) 70 return RET_TOOFEW(0); 71 if (s[7] != 'u') 72 goto simply_backslash; 73 wc2 = 0; 74 for (i = 8; i < 12; i++) { 75 if (n <= i) 76 return RET_TOOFEW(0); 77 c = s[i]; 78 if (c >= '0' && c <= '9') 79 c -= '0'; 80 else if (c >= 'A' && c <= 'Z') 81 c -= 'A'-10; 82 else if (c >= 'a' && c <= 'z') 83 c -= 'a'-10; 84 else 85 goto simply_backslash; 86 wc2 |= (ucs4_t) c << (4 * (11-i)); 87 } 88 if (!(wc2 >= 0xdc00 && wc2 < 0xe000)) 89 goto simply_backslash; 90 *pwc = 0x10000 + ((wc - 0xd800) << 10) + (wc2 - 0xdc00); 91 return 12; 92 simply_backslash: 93 *pwc = '\\'; 94 return 1; 95 } 96 97 static int 98 java_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) 99 { 100 if (wc < 0x80) { 101 *r = wc; 102 return 1; 103 } else if (wc < 0x10000) { 104 if (n >= 6) { 105 unsigned int i; 106 r[0] = '\\'; 107 r[1] = 'u'; 108 i = (wc >> 12) & 0x0f; r[2] = (i < 10 ? '0'+i : 'a'-10+i); 109 i = (wc >> 8) & 0x0f; r[3] = (i < 10 ? '0'+i : 'a'-10+i); 110 i = (wc >> 4) & 0x0f; r[4] = (i < 10 ? '0'+i : 'a'-10+i); 111 i = wc & 0x0f; r[5] = (i < 10 ? '0'+i : 'a'-10+i); 112 return 6; 113 } else 114 return RET_TOOSMALL; 115 } else if (wc < 0x110000) { 116 if (n >= 12) { 117 ucs4_t wc1 = 0xd800 + ((wc - 0x10000) >> 10); 118 ucs4_t wc2 = 0xdc00 + ((wc - 0x10000) & 0x3ff); 119 unsigned int i; 120 r[0] = '\\'; 121 r[1] = 'u'; 122 i = (wc1 >> 12) & 0x0f; r[2] = (i < 10 ? '0'+i : 'a'-10+i); 123 i = (wc1 >> 8) & 0x0f; r[3] = (i < 10 ? '0'+i : 'a'-10+i); 124 i = (wc1 >> 4) & 0x0f; r[4] = (i < 10 ? '0'+i : 'a'-10+i); 125 i = wc1 & 0x0f; r[5] = (i < 10 ? '0'+i : 'a'-10+i); 126 r[6] = '\\'; 127 r[7] = 'u'; 128 i = (wc2 >> 12) & 0x0f; r[8] = (i < 10 ? '0'+i : 'a'-10+i); 129 i = (wc2 >> 8) & 0x0f; r[9] = (i < 10 ? '0'+i : 'a'-10+i); 130 i = (wc2 >> 4) & 0x0f; r[10] = (i < 10 ? '0'+i : 'a'-10+i); 131 i = wc2 & 0x0f; r[11] = (i < 10 ? '0'+i : 'a'-10+i); 132 return 12; 133 } else 134 return RET_TOOSMALL; 135 } 136 return RET_ILUNI; 137 } 138