lib/libc/gen/unvis.c

   1 /*      @(#)unvis.c     8.1 (Berkeley) 6/4/93   */
   2 /*      $NetBSD: unvis.c,v 1.44 2014/09/26 15:43:36 roy Exp $   */
   3
   4 /*-
   5  * Copyright (c) 1989, 1993
   6  *      The Regents of the University of California.  All rights reserved.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 3. Neither the name of the University nor the names of its contributors
  17  *    may be used to endorse or promote products derived from this software
  18  *    without specific prior written permission.
  19  *
  20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  30  * SUCH DAMAGE.
  31  */
  32
  33 #include "namespace.h"
  34 #include <sys/types.h>
  35
  36 #include <assert.h>
  37 #include <ctype.h>
  38 #include <stdint.h>
  39 #include <stdio.h>
  40 #include <errno.h>
  41 #include <vis.h>
  42 #include "un-namespace.h"
  43
  44 /*
  45  * Return the number of elements in a statically-allocated array,
  46  * __x.
  47  */
  48 #define __arraycount(__x)       (sizeof(__x) / sizeof(__x[0]))
  49
  50 #ifdef __weak_alias
  51 __weak_alias(strnunvisx,_strnunvisx)
  52 #endif
  53
  54 #if !HAVE_VIS
  55 /*
  56  * decode driven by state machine
  57  */
  58 #define S_GROUND        0       /* haven't seen escape char */
  59 #define S_START         1       /* start decoding special sequence */
  60 #define S_META          2       /* metachar started (M) */
  61 #define S_META1         3       /* metachar more, regular char (-) */
  62 #define S_CTRL          4       /* control char started (^) */
  63 #define S_OCTAL2        5       /* octal digit 2 */
  64 #define S_OCTAL3        6       /* octal digit 3 */
  65 #define S_HEX           7       /* mandatory hex digit */
  66 #define S_HEX1          8       /* http hex digit */
  67 #define S_HEX2          9       /* http hex digit 2 */
  68 #define S_MIME1         10      /* mime hex digit 1 */
  69 #define S_MIME2         11      /* mime hex digit 2 */
  70 #define S_EATCRNL       12      /* mime eating CRNL */
  71 #define S_AMP           13      /* seen & */
  72 #define S_NUMBER        14      /* collecting number */
  73 #define S_STRING        15      /* collecting string */
  74
  75 #define isoctal(c)      (((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
  76 #define xtod(c)         (isdigit(c) ? (c - '0') : ((tolower(c) - 'a') + 10))
  77 #define XTOD(c)         (isdigit(c) ? (c - '0') : ((c - 'A') + 10))
  78
  79 /*
  80  * RFC 1866
  81  */
  82 static const struct nv {
  83         char name[7];
  84         uint8_t value;
  85 } nv[] = {
  86         { "AElig",      198 }, /* capital AE diphthong (ligature)  */
  87         { "Aacute",     193 }, /* capital A, acute accent  */
  88         { "Acirc",      194 }, /* capital A, circumflex accent  */
  89         { "Agrave",     192 }, /* capital A, grave accent  */
  90         { "Aring",      197 }, /* capital A, ring  */
  91         { "Atilde",     195 }, /* capital A, tilde  */
  92         { "Auml",       196 }, /* capital A, dieresis or umlaut mark  */
  93         { "Ccedil",     199 }, /* capital C, cedilla  */
  94         { "ETH",        208 }, /* capital Eth, Icelandic  */
  95         { "Eacute",     201 }, /* capital E, acute accent  */
  96         { "Ecirc",      202 }, /* capital E, circumflex accent  */
  97         { "Egrave",     200 }, /* capital E, grave accent  */
  98         { "Euml",       203 }, /* capital E, dieresis or umlaut mark  */
  99         { "Iacute",     205 }, /* capital I, acute accent  */
 100         { "Icirc",      206 }, /* capital I, circumflex accent  */
 101         { "Igrave",     204 }, /* capital I, grave accent  */
 102         { "Iuml",       207 }, /* capital I, dieresis or umlaut mark  */
 103         { "Ntilde",     209 }, /* capital N, tilde  */
 104         { "Oacute",     211 }, /* capital O, acute accent  */
 105         { "Ocirc",      212 }, /* capital O, circumflex accent  */
 106         { "Ograve",     210 }, /* capital O, grave accent  */
 107         { "Oslash",     216 }, /* capital O, slash  */
 108         { "Otilde",     213 }, /* capital O, tilde  */
 109         { "Ouml",       214 }, /* capital O, dieresis or umlaut mark  */
 110         { "THORN",      222 }, /* capital THORN, Icelandic  */
 111         { "Uacute",     218 }, /* capital U, acute accent  */
 112         { "Ucirc",      219 }, /* capital U, circumflex accent  */
 113         { "Ugrave",     217 }, /* capital U, grave accent  */
 114         { "Uuml",       220 }, /* capital U, dieresis or umlaut mark  */
 115         { "Yacute",     221 }, /* capital Y, acute accent  */
 116         { "aacute",     225 }, /* small a, acute accent  */
 117         { "acirc",      226 }, /* small a, circumflex accent  */
 118         { "acute",      180 }, /* acute accent  */
 119         { "aelig",      230 }, /* small ae diphthong (ligature)  */
 120         { "agrave",     224 }, /* small a, grave accent  */
 121         { "amp",         38 }, /* ampersand  */
 122         { "aring",      229 }, /* small a, ring  */
 123         { "atilde",     227 }, /* small a, tilde  */
 124         { "auml",       228 }, /* small a, dieresis or umlaut mark  */
 125         { "brvbar",     166 }, /* broken (vertical) bar  */
 126         { "ccedil",     231 }, /* small c, cedilla  */
 127         { "cedil",      184 }, /* cedilla  */
 128         { "cent",       162 }, /* cent sign  */
 129         { "copy",       169 }, /* copyright sign  */
 130         { "curren",     164 }, /* general currency sign  */
 131         { "deg",        176 }, /* degree sign  */
 132         { "divide",     247 }, /* divide sign  */
 133         { "eacute",     233 }, /* small e, acute accent  */
 134         { "ecirc",      234 }, /* small e, circumflex accent  */
 135         { "egrave",     232 }, /* small e, grave accent  */
 136         { "eth",        240 }, /* small eth, Icelandic  */
 137         { "euml",       235 }, /* small e, dieresis or umlaut mark  */
 138         { "frac12",     189 }, /* fraction one-half  */
 139         { "frac14",     188 }, /* fraction one-quarter  */
 140         { "frac34",     190 }, /* fraction three-quarters  */
 141         { "gt",          62 }, /* greater than  */
 142         { "iacute",     237 }, /* small i, acute accent  */
 143         { "icirc",      238 }, /* small i, circumflex accent  */
 144         { "iexcl",      161 }, /* inverted exclamation mark  */
 145         { "igrave",     236 }, /* small i, grave accent  */
 146         { "iquest",     191 }, /* inverted question mark  */
 147         { "iuml",       239 }, /* small i, dieresis or umlaut mark  */
 148         { "laquo",      171 }, /* angle quotation mark, left  */
 149         { "lt",          60 }, /* less than  */
 150         { "macr",       175 }, /* macron  */
 151         { "micro",      181 }, /* micro sign  */
 152         { "middot",     183 }, /* middle dot  */
 153         { "nbsp",       160 }, /* no-break space  */
 154         { "not",        172 }, /* not sign  */
 155         { "ntilde",     241 }, /* small n, tilde  */
 156         { "oacute",     243 }, /* small o, acute accent  */
 157         { "ocirc",      244 }, /* small o, circumflex accent  */
 158         { "ograve",     242 }, /* small o, grave accent  */
 159         { "ordf",       170 }, /* ordinal indicator, feminine  */
 160         { "ordm",       186 }, /* ordinal indicator, masculine  */
 161         { "oslash",     248 }, /* small o, slash  */
 162         { "otilde",     245 }, /* small o, tilde  */
 163         { "ouml",       246 }, /* small o, dieresis or umlaut mark  */
 164         { "para",       182 }, /* pilcrow (paragraph sign)  */
 165         { "plusmn",     177 }, /* plus-or-minus sign  */
 166         { "pound",      163 }, /* pound sterling sign  */
 167         { "quot",        34 }, /* double quote  */
 168         { "raquo",      187 }, /* angle quotation mark, right  */
 169         { "reg",        174 }, /* registered sign  */
 170         { "sect",       167 }, /* section sign  */
 171         { "shy",        173 }, /* soft hyphen  */
 172         { "sup1",       185 }, /* superscript one  */
 173         { "sup2",       178 }, /* superscript two  */
 174         { "sup3",       179 }, /* superscript three  */
 175         { "szlig",      223 }, /* small sharp s, German (sz ligature)  */
 176         { "thorn",      254 }, /* small thorn, Icelandic  */
 177         { "times",      215 }, /* multiply sign  */
 178         { "uacute",     250 }, /* small u, acute accent  */
 179         { "ucirc",      251 }, /* small u, circumflex accent  */
 180         { "ugrave",     249 }, /* small u, grave accent  */
 181         { "uml",        168 }, /* umlaut (dieresis)  */
 182         { "uuml",       252 }, /* small u, dieresis or umlaut mark  */
 183         { "yacute",     253 }, /* small y, acute accent  */
 184         { "yen",        165 }, /* yen sign  */
 185         { "yuml",       255 }, /* small y, dieresis or umlaut mark  */
 186 };
 187
 188 /*
 189  * unvis - decode characters previously encoded by vis
 190  */
 191 int
 192 unvis(char *cp, int c, int *astate, int flag)
 193 {
 194         unsigned char uc = (unsigned char)c;
 195         unsigned char st, ia, is, lc;
 196
 197 /*
 198  * Bottom 8 bits of astate hold the state machine state.
 199  * Top 8 bits hold the current character in the http 1866 nv string decoding
 200  */
 201 #define GS(a)           ((a) & 0xff)
 202 #define SS(a, b)        (((uint32_t)(a) << 24) | (b))
 203 #define GI(a)           ((uint32_t)(a) >> 24)
 204
 205         _DIAGASSERT(cp != NULL);
 206         _DIAGASSERT(astate != NULL);
 207         st = GS(*astate);
 208
 209         if (flag & UNVIS_END) {
 210                 switch (st) {
 211                 case S_OCTAL2:
 212                 case S_OCTAL3:
 213                 case S_HEX2:
 214                         *astate = SS(0, S_GROUND);
 215                         return UNVIS_VALID;
 216                 case S_GROUND:
 217                         return UNVIS_NOCHAR;
 218                 default:
 219                         return UNVIS_SYNBAD;
 220                 }
 221         }
 222
 223         switch (st) {
 224
 225         case S_GROUND:
 226                 *cp = 0;
 227                 if ((flag & VIS_NOESCAPE) == 0 && c == '\\') {
 228                         *astate = SS(0, S_START);
 229                         return UNVIS_NOCHAR;
 230                 }
 231                 if ((flag & VIS_HTTP1808) && c == '%') {
 232                         *astate = SS(0, S_HEX1);
 233                         return UNVIS_NOCHAR;
 234                 }
 235                 if ((flag & VIS_HTTP1866) && c == '&') {
 236                         *astate = SS(0, S_AMP);
 237                         return UNVIS_NOCHAR;
 238                 }
 239                 if ((flag & VIS_MIMESTYLE) && c == '=') {
 240                         *astate = SS(0, S_MIME1);
 241                         return UNVIS_NOCHAR;
 242                 }
 243                 *cp = c;
 244                 return UNVIS_VALID;
 245
 246         case S_START:
 247                 switch(c) {
 248                 case '-':
 249                         *cp = 0;
 250                         *astate = SS(0, S_GROUND);
 251                         return UNVIS_NOCHAR;
 252                 case '\\':
 253                         *cp = c;
 254                         *astate = SS(0, S_GROUND);
 255                         return UNVIS_VALID;
 256                 case '0': case '1': case '2': case '3':
 257                 case '4': case '5': case '6': case '7':
 258                         *cp = (c - '0');
 259                         *astate = SS(0, S_OCTAL2);
 260                         return UNVIS_NOCHAR;
 261                 case 'M':
 262                         *cp = (char)0200;
 263                         *astate = SS(0, S_META);
 264                         return UNVIS_NOCHAR;
 265                 case '^':
 266                         *astate = SS(0, S_CTRL);
 267                         return UNVIS_NOCHAR;
 268                 case 'n':
 269                         *cp = '\n';
 270                         *astate = SS(0, S_GROUND);
 271                         return UNVIS_VALID;
 272                 case 'r':
 273                         *cp = '\r';
 274                         *astate = SS(0, S_GROUND);
 275                         return UNVIS_VALID;
 276                 case 'b':
 277                         *cp = '\b';
 278                         *astate = SS(0, S_GROUND);
 279                         return UNVIS_VALID;
 280                 case 'a':
 281                         *cp = '\007';
 282                         *astate = SS(0, S_GROUND);
 283                         return UNVIS_VALID;
 284                 case 'v':
 285                         *cp = '\v';
 286                         *astate = SS(0, S_GROUND);
 287                         return UNVIS_VALID;
 288                 case 't':
 289                         *cp = '\t';
 290                         *astate = SS(0, S_GROUND);
 291                         return UNVIS_VALID;
 292                 case 'f':
 293                         *cp = '\f';
 294                         *astate = SS(0, S_GROUND);
 295                         return UNVIS_VALID;
 296                 case 's':
 297                         *cp = ' ';
 298                         *astate = SS(0, S_GROUND);
 299                         return UNVIS_VALID;
 300                 case 'E':
 301                         *cp = '\033';
 302                         *astate = SS(0, S_GROUND);
 303                         return UNVIS_VALID;
 304                 case 'x':
 305                         *astate = SS(0, S_HEX);
 306                         return UNVIS_NOCHAR;
 307                 case '\n':
 308                         /*
 309                          * hidden newline
 310                          */
 311                         *astate = SS(0, S_GROUND);
 312                         return UNVIS_NOCHAR;
 313                 case '$':
 314                         /*
 315                          * hidden marker
 316                          */
 317                         *astate = SS(0, S_GROUND);
 318                         return UNVIS_NOCHAR;
 319                 default:
 320                         if (isgraph(c)) {
 321                                 *cp = c;
 322                                 *astate = SS(0, S_GROUND);
 323                                 return UNVIS_VALID;
 324                         }
 325                 }
 326                 goto bad;
 327
 328         case S_META:
 329                 if (c == '-')
 330                         *astate = SS(0, S_META1);
 331                 else if (c == '^')
 332                         *astate = SS(0, S_CTRL);
 333                 else
 334                         goto bad;
 335                 return UNVIS_NOCHAR;
 336
 337         case S_META1:
 338                 *astate = SS(0, S_GROUND);
 339                 *cp |= c;
 340                 return UNVIS_VALID;
 341
 342         case S_CTRL:
 343                 if (c == '?')
 344                         *cp |= 0177;
 345                 else
 346                         *cp |= c & 037;
 347                 *astate = SS(0, S_GROUND);
 348                 return UNVIS_VALID;
 349
 350         case S_OCTAL2:  /* second possible octal digit */
 351                 if (isoctal(uc)) {
 352                         /*
 353                          * yes - and maybe a third
 354                          */
 355                         *cp = (*cp << 3) + (c - '0');
 356                         *astate = SS(0, S_OCTAL3);
 357                         return UNVIS_NOCHAR;
 358                 }
 359                 /*
 360                  * no - done with current sequence, push back passed char
 361                  */
 362                 *astate = SS(0, S_GROUND);
 363                 return UNVIS_VALIDPUSH;
 364
 365         case S_OCTAL3:  /* third possible octal digit */
 366                 *astate = SS(0, S_GROUND);
 367                 if (isoctal(uc)) {
 368                         *cp = (*cp << 3) + (c - '0');
 369                         return UNVIS_VALID;
 370                 }
 371                 /*
 372                  * we were done, push back passed char
 373                  */
 374                 return UNVIS_VALIDPUSH;
 375
 376         case S_HEX:
 377                 if (!isxdigit(uc))
 378                         goto bad;
 379                 /*FALLTHROUGH*/
 380         case S_HEX1:
 381                 if (isxdigit(uc)) {
 382                         *cp = xtod(uc);
 383                         *astate = SS(0, S_HEX2);
 384                         return UNVIS_NOCHAR;
 385                 }
 386                 /*
 387                  * no - done with current sequence, push back passed char
 388                  */
 389                 *astate = SS(0, S_GROUND);
 390                 return UNVIS_VALIDPUSH;
 391
 392         case S_HEX2:
 393                 *astate = S_GROUND;
 394                 if (isxdigit(uc)) {
 395                         *cp = xtod(uc) | (*cp << 4);
 396                         return UNVIS_VALID;
 397                 }
 398                 return UNVIS_VALIDPUSH;
 399
 400         case S_MIME1:
 401                 if (uc == '\n' || uc == '\r') {
 402                         *astate = SS(0, S_EATCRNL);
 403                         return UNVIS_NOCHAR;
 404                 }
 405                 if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
 406                         *cp = XTOD(uc);
 407                         *astate = SS(0, S_MIME2);
 408                         return UNVIS_NOCHAR;
 409                 }
 410                 goto bad;
 411
 412         case S_MIME2:
 413                 if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
 414                         *astate = SS(0, S_GROUND);
 415                         *cp = XTOD(uc) | (*cp << 4);
 416                         return UNVIS_VALID;
 417                 }
 418                 goto bad;
 419
 420         case S_EATCRNL:
 421                 switch (uc) {
 422                 case '\r':
 423                 case '\n':
 424                         return UNVIS_NOCHAR;
 425                 case '=':
 426                         *astate = SS(0, S_MIME1);
 427                         return UNVIS_NOCHAR;
 428                 default:
 429                         *cp = uc;
 430                         *astate = SS(0, S_GROUND);
 431                         return UNVIS_VALID;
 432                 }
 433
 434         case S_AMP:
 435                 *cp = 0;
 436                 if (uc == '#') {
 437                         *astate = SS(0, S_NUMBER);
 438                         return UNVIS_NOCHAR;
 439                 }
 440                 *astate = SS(0, S_STRING);
 441                 /*FALLTHROUGH*/
 442
 443         case S_STRING:
 444                 ia = *cp;               /* index in the array */
 445                 is = GI(*astate);       /* index in the string */
 446                 lc = is == 0 ? 0 : nv[ia].name[is - 1]; /* last character */
 447
 448                 if (uc == ';')
 449                         uc = '\0';
 450
 451                 for (; ia < __arraycount(nv); ia++) {
 452                         if (is != 0 && nv[ia].name[is - 1] != lc)
 453                                 goto bad;
 454                         if (nv[ia].name[is] == uc)
 455                                 break;
 456                 }
 457
 458                 if (ia == __arraycount(nv))
 459                         goto bad;
 460
 461                 if (uc != 0) {
 462                         *cp = ia;
 463                         *astate = SS(is + 1, S_STRING);
 464                         return UNVIS_NOCHAR;
 465                 }
 466
 467                 *cp = nv[ia].value;
 468                 *astate = SS(0, S_GROUND);
 469                 return UNVIS_VALID;
 470
 471         case S_NUMBER:
 472                 if (uc == ';')
 473                         return UNVIS_VALID;
 474                 if (!isdigit(uc))
 475                         goto bad;
 476                 *cp += (*cp * 10) + uc - '0';
 477                 return UNVIS_NOCHAR;
 478
 479         default:
 480         bad:
 481                 /*
 482                  * decoder in unknown state - (probably uninitialized)
 483                  */
 484                 *astate = SS(0, S_GROUND);
 485                 return UNVIS_SYNBAD;
 486         }
 487 }
 488
 489 /*
 490  * strnunvisx - decode src into dst
 491  *
 492  *      Number of chars decoded into dst is returned, -1 on error.
 493  *      Dst is null terminated.
 494  */
 495
 496 int
 497 strnunvisx(char *dst, size_t dlen, const char *src, int flag)
 498 {
 499         char c;
 500         char t = '\0', *start = dst;
 501         int state = 0;
 502
 503         _DIAGASSERT(src != NULL);
 504         _DIAGASSERT(dst != NULL);
 505 #define CHECKSPACE() \
 506         do { \
 507                 if (dlen-- == 0) { \
 508                         errno = ENOSPC; \
 509                         return -1; \
 510                 } \
 511         } while (/*CONSTCOND*/0)
 512
 513         while ((c = *src++) != '\0') {
 514  again:
 515                 switch (unvis(&t, c, &state, flag)) {
 516                 case UNVIS_VALID:
 517                         CHECKSPACE();
 518                         *dst++ = t;
 519                         break;
 520                 case UNVIS_VALIDPUSH:
 521                         CHECKSPACE();
 522                         *dst++ = t;
 523                         goto again;
 524                 case 0:
 525                 case UNVIS_NOCHAR:
 526                         break;
 527                 case UNVIS_SYNBAD:
 528                         errno = EINVAL;
 529                         return -1;
 530                 default:
 531                         _DIAGASSERT(/*CONSTCOND*/0);
 532                         errno = EINVAL;
 533                         return -1;
 534                 }
 535         }
 536         if (unvis(&t, c, &state, UNVIS_END) == UNVIS_VALID) {
 537                 CHECKSPACE();
 538                 *dst++ = t;
 539         }
 540         CHECKSPACE();
 541         *dst = '\0';
 542         return (int)(dst - start);
 543 }
 544
 545 int
 546 strunvisx(char *dst, const char *src, int flag)
 547 {
 548         return strnunvisx(dst, (size_t)~0, src, flag);
 549 }
 550
 551 int
 552 strunvis(char *dst, const char *src)
 553 {
 554         return strnunvisx(dst, (size_t)~0, src, 0);
 555 }
 556
 557 int
 558 strnunvis(char *dst, size_t dlen, const char *src)
 559 {
 560         return strnunvisx(dst, dlen, src, 0);
 561 }
 562 #endif