1 /* $Header: /src/pub/tcsh/tc.nls.c,v 3.6 2005/02/15 21:09:02 christos Exp $ */ 2 /* 3 * tc.nls.c: NLS handling 4 */ 5 /*- 6 * Copyright (c) 1980, 1991 The Regents of the University of California. 7 * All rights reserved. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 3. Neither the name of the University nor the names of its contributors 18 * may be used to endorse or promote products derived from this software 19 * without specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 */ 33 #include "sh.h" 34 35 RCSID("$Id: tc.nls.c,v 3.6 2005/02/15 21:09:02 christos Exp $") 36 37 #ifdef SHORT_STRINGS 38 int 39 NLSWidth(c) 40 NLSChar c; 41 { 42 # ifdef HAVE_WCWIDTH 43 int l; 44 if (c & NLS_ILLEGAL) 45 return 1; 46 l = wcwidth(c); 47 return l >= 0 ? l : 0; 48 # else 49 return c != 0; 50 # endif 51 } 52 #endif 53 54 #if defined (WIDE_STRINGS) || !defined (SHORT_STRINGS) 55 Char * 56 NLSChangeCase(Char *p, int mode) 57 { 58 Char c, *op = p, *n, c2 = 0; 59 for (; (c = *p) != 0; p++) { 60 if (mode == 0 && Islower(c)) { 61 c2 = Toupper(c); 62 break; 63 } else if (mode && Isupper(c)) { 64 c2 = Tolower(c); 65 break; 66 } 67 } 68 if (!*p) 69 return 0; 70 n = Strsave(op); 71 n[p - op] = c2; 72 return n; 73 } 74 75 int 76 NLSExtend(Char *from, int max, int num) 77 { 78 (void)from; 79 num = abs (num); 80 if (num > max) 81 num = max; 82 return num; 83 } 84 #endif 85 86 #ifdef WIDE_STRINGS 87 88 int 89 NLSStringWidth(Char *s) 90 { 91 int w = 0; 92 while (*s) 93 w += wcwidth(*s++); 94 return w; 95 } 96 97 #elif defined (SHORT_STRINGS) 98 99 int 100 NLSFrom(const Char *p, size_t l, NLSChar *cp) 101 { 102 size_t i; 103 int len; 104 wchar_t c; 105 char b[MB_LEN_MAX]; 106 107 if (l == NLSZEROT) { 108 for (i = 0; i < MB_CUR_MAX && *p; i++) 109 b[i] = p[i] & CHAR; 110 } else { 111 for (i = 0; i < MB_CUR_MAX && i < l; i++) 112 b[i] = p[i] & CHAR; 113 } 114 mbtowc(0, 0, 0); 115 len = rt_mbtowc(&c, b, i); 116 if (len <= 0) { 117 if (cp) 118 *cp = *p ? *p | NLS_ILLEGAL : 0; 119 return 1; 120 } 121 if (cp) 122 *cp = (int)c; 123 return len; 124 } 125 126 int 127 NLSFinished(Char *p, size_t l, eChar extra) 128 { 129 size_t i, r; 130 wchar_t c; 131 char b[MB_LEN_MAX + 1], back[MB_LEN_MAX]; 132 mbstate_t state; 133 for (i = 0; i < MB_CUR_MAX && i < l; i++) 134 b[i] = p[i]; 135 if (extra != CHAR_ERR) 136 b[i++] = extra; 137 memset(&state, 0, sizeof(state)); 138 r = mbrtowc((wchar_t *)&c, b, i, (mbstate_t *)&state); 139 if (r == (size_t)-2) 140 return 0; 141 if (r == (size_t)-1 || (size_t)wctomb(back, c) != r || 142 memcmp(b, back, r) != 0) 143 return -1; 144 return r == i ? 1 : 2; 145 } 146 147 int 148 NLSChars(Char *s) 149 { 150 int l; 151 for (l = 0; *s; l++) 152 s += NLSSize(s, -1); 153 return l; 154 } 155 156 int 157 NLSStringWidth(Char *s) 158 { 159 int w = 0; 160 NLSChar c; 161 while (*s) { 162 s += NLSFrom(s, NLSZEROT, &c); 163 w += NLSWidth(c); 164 } 165 return w; 166 } 167 168 int 169 NLSTo(Char *p, NLSChar c) 170 { 171 char b[MB_LEN_MAX]; 172 int i, j; 173 174 if (c & NLS_ILLEGAL) { 175 if (p) 176 *p = c; 177 return 1; 178 } 179 i = wctomb(b, (wchar_t)c); 180 if (i == -1) 181 return 0; 182 if (p) 183 for (j = 0; j < i; j++) 184 p[j] = b[j]; 185 return i; 186 } 187 188 189 int 190 NLSExtend(Char *from, int max, int num) 191 { 192 int l, n, i; 193 Char *p; 194 195 if (num == 0) 196 return 0; 197 if (num > 0) { 198 n = 0; 199 while (num > 0 && max > 0) { 200 l = NLSSize(from, max); 201 n += l; 202 from += l; 203 max -= l; 204 num--; 205 } 206 return n; 207 } 208 from -= max; 209 p = from; 210 i = max; 211 n = 0; 212 while (i > 0) { 213 l = NLSSize(p, i); 214 p += l; 215 i -= l; 216 n++; 217 } 218 if (n >= -num) 219 n += num; 220 else 221 n = 0; 222 i = max; 223 while (n > 0) { 224 l = NLSSize(from, max); 225 from += l; 226 max -= l; 227 i -= l; 228 n--; 229 } 230 return i; 231 } 232 233 void 234 NLSQuote(Char *cp) 235 { 236 int l; 237 while (*cp) { 238 l = NLSSize(cp, -1); 239 cp++; 240 while (l-- > 1) 241 *cp++ |= QUOTE; 242 } 243 } 244 245 Char * 246 NLSChangeCase(Char *p, int mode) 247 { 248 Char *n, *op = p; 249 NLSChar c, c2 = 0; 250 int l, l2; 251 252 while (*p) { 253 l = NLSFrom(p, NLSZEROT, &c); 254 if (mode == 0 && iswlower((wint_t)c)) { 255 c2 = (int)towupper((wint_t)c); 256 break; 257 } else if (mode && iswupper((wint_t)c)) { 258 c2 = (int)towlower((wint_t)c); 259 break; 260 } 261 p += l; 262 } 263 if (!*p) 264 return 0; 265 l2 = NLSTo((Char *)0, c2); 266 n = (Char *)xmalloc((size_t)((op - p + l2 + Strlen(p + l) + 1) * sizeof(Char))); 267 if (p != op) 268 memcpy(n, op, (p - op) * sizeof(Char)); 269 NLSTo(n + (p - op), c2); 270 memcpy(n + (p - op + l2), p + l, (Strlen(p + l) + 1) * sizeof(Char)); 271 return n; 272 } 273 #endif 274 275 int 276 NLSClassify(c, nocomb) 277 NLSChar c; 278 int nocomb; 279 { 280 int w; 281 if (c & NLS_ILLEGAL) 282 return NLSCLASS_ILLEGAL; 283 w = NLSWidth(c); 284 if (w > 0 || (Iswprint(c) && !nocomb)) 285 return w; 286 if (Iswcntrl(c) && c < 0x100) { 287 if (c == '\n') 288 return NLSCLASS_NL; 289 if (c == '\t') 290 return NLSCLASS_TAB; 291 #ifndef ASCII 292 if (!Isupper(_toebcdic[_toascii[c]|0100]) && !strchr("@[\\]^_", _toebcdic[_toascii[c]|0100])) 293 return NLSCLASS_ILLEGAL; 294 #endif 295 return NLSCLASS_CTRL; 296 } 297 if (c >= 0x1000000) 298 return NLSCLASS_ILLEGAL4; 299 if (c >= 0x10000) 300 return NLSCLASS_ILLEGAL3; 301 if (c >= 0x100) 302 return NLSCLASS_ILLEGAL2; 303 return NLSCLASS_ILLEGAL; 304 } 305