entities.c

   1 #include "entities.h"
   2
   3 #include <errno.h>
   4 #include <string.h>
   5 #include <stdlib.h>
   6
   7 #define UNICODE_MAX 0x10FFFFul
   8
   9 static const char *named_entities[][2] =
  10 {
  11         { "AElig;", "Æ" },
  12         { "Aacute;", "Á" },
  13         { "Acirc;", "Â" },
  14         { "Agrave;", "À" },
  15         { "Alpha;", "Α" },
  16         { "Aring;", "Å" },
  17         { "Atilde;", "Ã" },
  18         { "Auml;", "Ä" },
  19         { "Beta;", "Β" },
  20         { "Ccedil;", "Ç" },
  21         { "Chi;", "Χ" },
  22         { "Dagger;", "‡" },
  23         { "Delta;", "Δ" },
  24         { "ETH;", "Ð" },
  25         { "Eacute;", "É" },
  26         { "Ecirc;", "Ê" },
  27         { "Egrave;", "È" },
  28         { "Epsilon;", "Ε" },
  29         { "Eta;", "Η" },
  30         { "Euml;", "Ë" },
  31         { "Gamma;", "Γ" },
  32         { "Iacute;", "Í" },
  33         { "Icirc;", "Î" },
  34         { "Igrave;", "Ì" },
  35         { "Iota;", "Ι" },
  36         { "Iuml;", "Ï" },
  37         { "Kappa;", "Κ" },
  38         { "Lambda;", "Λ" },
  39         { "Mu;", "Μ" },
  40         { "Ntilde;", "Ñ" },
  41         { "Nu;", "Ν" },
  42         { "OElig;", "Œ" },
  43         { "Oacute;", "Ó" },
  44         { "Ocirc;", "Ô" },
  45         { "Ograve;", "Ò" },
  46         { "Omega;", "Ω" },
  47         { "Omicron;", "Ο" },
  48         { "Oslash;", "Ø" },
  49         { "Otilde;", "Õ" },
  50         { "Ouml;", "Ö" },
  51         { "Phi;", "Φ" },
  52         { "Pi;", "Π" },
  53         { "Prime;", "″" },
  54         { "Psi;", "Ψ" },
  55         { "Rho;", "Ρ" },
  56         { "Scaron;", "Š" },
  57         { "Sigma;", "Σ" },
  58         { "THORN;", "Þ" },
  59         { "Tau;", "Τ" },
  60         { "Theta;", "Θ" },
  61         { "Uacute;", "Ú" },
  62         { "Ucirc;", "Û" },
  63         { "Ugrave;", "Ù" },
  64         { "Upsilon;", "Υ" },
  65         { "Uuml;", "Ü" },
  66         { "Xi;", "Ξ" },
  67         { "Yacute;", "Ý" },
  68         { "Yuml;", "Ÿ" },
  69         { "Zeta;", "Ζ" },
  70         { "aacute;", "á" },
  71         { "acirc;", "â" },
  72         { "acute;", "´" },
  73         { "aelig;", "æ" },
  74         { "agrave;", "à" },
  75         { "alefsym;", "ℵ" },
  76         { "alpha;", "α" },
  77         { "amp;", "&" },
  78         { "and;", "∧" },
  79         { "ang;", "∠" },
  80         { "apos;", "'" },
  81         { "aring;", "å" },
  82         { "asymp;", "≈" },
  83         { "atilde;", "ã" },
  84         { "auml;", "ä" },
  85         { "bdquo;", "„" },
  86         { "beta;", "β" },
  87         { "brvbar;", "¦" },
  88         { "bull;", "•" },
  89         { "cap;", "∩" },
  90         { "ccedil;", "ç" },
  91         { "cedil;", "¸" },
  92         { "cent;", "¢" },
  93         { "chi;", "χ" },
  94         { "circ;", "ˆ" },
  95         { "clubs;", "♣" },
  96         { "cong;", "≅" },
  97         { "copy;", "©" },
  98         { "crarr;", "↵" },
  99         { "cup;", "∪" },
 100         { "curren;", "¤" },
 101         { "dArr;", "⇓" },
 102         { "dagger;", "†" },
 103         { "darr;", "↓" },
 104         { "deg;", "°" },
 105         { "delta;", "δ" },
 106         { "diams;", "♦" },
 107         { "divide;", "÷" },
 108         { "eacute;", "é" },
 109         { "ecirc;", "ê" },
 110         { "egrave;", "è" },
 111         { "empty;", "∅" },
 112         { "emsp;", " " },
 113         { "ensp;", " " },
 114         { "epsilon;", "ε" },
 115         { "equiv;", "≡" },
 116         { "eta;", "η" },
 117         { "eth;", "ð" },
 118         { "euml;", "ë" },
 119         { "euro;", "€" },
 120         { "exist;", "∃" },
 121         { "fnof;", "ƒ" },
 122         { "forall;", "∀" },
 123         { "frac12;", "½" },
 124         { "frac14;", "¼" },
 125         { "frac34;", "¾" },
 126         { "frasl;", "⁄" },
 127         { "gamma;", "γ" },
 128         { "ge;", "≥" },
 129         { "gt;", ">" },
 130         { "hArr;", "⇔" },
 131         { "harr;", "↔" },
 132         { "hearts;", "♥" },
 133         { "hellip;", "…" },
 134         { "iacute;", "í" },
 135         { "icirc;", "î" },
 136         { "iexcl;", "¡" },
 137         { "igrave;", "ì" },
 138         { "image;", "ℑ" },
 139         { "infin;", "∞" },
 140         { "int;", "∫" },
 141         { "iota;", "ι" },
 142         { "iquest;", "¿" },
 143         { "isin;", "∈" },
 144         { "iuml;", "ï" },
 145         { "kappa;", "κ" },
 146         { "lArr;", "⇐" },
 147         { "lambda;", "λ" },
 148         { "lang;", "〈" },
 149         { "laquo;", "«" },
 150         { "larr;", "←" },
 151         { "lceil;", "⌈" },
 152         { "ldquo;", "“" },
 153         { "le;", "≤" },
 154         { "lfloor;", "⌊" },
 155         { "lowast;", "∗" },
 156         { "loz;", "◊" },
 157         { "lrm;", "\xE2\x80\x8E" },
 158         { "lsaquo;", "‹" },
 159         { "lsquo;", "‘" },
 160         { "lt;", "<" },
 161         { "macr;", "¯" },
 162         { "mdash;", "—" },
 163         { "micro;", "µ" },
 164         { "middot;", "·" },
 165         { "minus;", "−" },
 166         { "mu;", "μ" },
 167         { "nabla;", "∇" },
 168         { "nbsp;", " " },
 169         { "ndash;", "–" },
 170         { "ne;", "≠" },
 171         { "ni;", "∋" },
 172         { "not;", "¬" },
 173         { "notin;", "∉" },
 174         { "nsub;", "⊄" },
 175         { "ntilde;", "ñ" },
 176         { "nu;", "ν" },
 177         { "oacute;", "ó" },
 178         { "ocirc;", "ô" },
 179         { "oelig;", "œ" },
 180         { "ograve;", "ò" },
 181         { "oline;", "‾" },
 182         { "omega;", "ω" },
 183         { "omicron;", "ο" },
 184         { "oplus;", "⊕" },
 185         { "or;", "∨" },
 186         { "ordf;", "ª" },
 187         { "ordm;", "º" },
 188         { "oslash;", "ø" },
 189         { "otilde;", "õ" },
 190         { "otimes;", "⊗" },
 191         { "ouml;", "ö" },
 192         { "para;", "¶" },
 193         { "part;", "∂" },
 194         { "permil;", "‰" },
 195         { "perp;", "⊥" },
 196         { "phi;", "φ" },
 197         { "pi;", "π" },
 198         { "piv;", "ϖ" },
 199         { "plusmn;", "±" },
 200         { "pound;", "£" },
 201         { "prime;", "′" },
 202         { "prod;", "∏" },
 203         { "prop;", "∝" },
 204         { "psi;", "ψ" },
 205         { "quot;", "\"" },
 206         { "rArr;", "⇒" },
 207         { "radic;", "√" },
 208         { "rang;", "〉" },
 209         { "raquo;", "»" },
 210         { "rarr;", "→" },
 211         { "rceil;", "⌉" },
 212         { "rdquo;", "”" },
 213         { "real;", "ℜ" },
 214         { "reg;", "®" },
 215         { "rfloor;", "⌋" },
 216         { "rho;", "ρ" },
 217         { "rlm;", "\xE2\x80\x8F" },
 218         { "rsaquo;", "›" },
 219         { "rsquo;", "’" },
 220         { "sbquo;", "‚" },
 221         { "scaron;", "š" },
 222         { "sdot;", "⋅" },
 223         { "sect;", "§" },
 224         { "shy;", "\xC2\xAD" },
 225         { "sigma;", "σ" },
 226         { "sigmaf;", "ς" },
 227         { "sim;", "∼" },
 228         { "spades;", "♠" },
 229         { "sub;", "⊂" },
 230         { "sube;", "⊆" },
 231         { "sum;", "∑" },
 232         { "sup;", "⊃" },
 233         { "sup1;", "¹" },
 234         { "sup2;", "²" },
 235         { "sup3;", "³" },
 236         { "supe;", "⊇" },
 237         { "szlig;", "ß" },
 238         { "tau;", "τ" },
 239         { "there4;", "∴" },
 240         { "theta;", "θ" },
 241         { "thetasym;", "ϑ" },
 242         { "thinsp;", " " },
 243         { "thorn;", "þ" },
 244         { "tilde;", "˜" },
 245         { "times;", "×" },
 246         { "trade;", "™" },
 247         { "uArr;", "⇑" },
 248         { "uacute;", "ú" },
 249         { "uarr;", "↑" },
 250         { "ucirc;", "û" },
 251         { "ugrave;", "ù" },
 252         { "uml;", "¨" },
 253         { "upsih;", "ϒ" },
 254         { "upsilon;", "υ" },
 255         { "uuml;", "ü" },
 256         { "weierp;", "℘" },
 257         { "xi;", "ξ" },
 258         { "yacute;", "ý" },
 259         { "yen;", "¥" },
 260         { "yuml;", "ÿ" },
 261         { "zeta;", "ζ" },
 262         { "zwj;", "\xE2\x80\x8D" },
 263         { "zwnj;", "\xE2\x80\x8C" }
 264 };
 265
 266 static int cmp(const void *key, const void *element)
 267 {
 268         return strncmp((const char *)key, *(const char **)element,
 269                 strlen(*(const char **)element));
 270 }
 271
 272 static const char *get_named_entity(const char *name)
 273 {
 274         const char **entity = bsearch(name, named_entities,
 275                 sizeof(named_entities) / sizeof(*named_entities),
 276                 sizeof(*named_entities), cmp);
 277
 278         return entity ? entity[1] : NULL;
 279 }
 280
 281 static size_t putc_utf8(unsigned long cp, char *buffer)
 282 {
 283         unsigned char *bytes = (unsigned char *)buffer;
 284
 285         if(cp <= 0x007Ful)
 286         {
 287                 bytes[0] = (unsigned char)cp;
 288                 return 1;
 289         }
 290
 291         if(cp <= 0x07FFul)
 292         {
 293                 bytes[1] = (unsigned char)((2u << 6) | (cp & 0x3Fu));
 294                 bytes[0] = (unsigned char)((6u << 5) | (cp >> 6));
 295                 return 2;
 296         }
 297
 298         if(cp <= 0xFFFFul)
 299         {
 300                 bytes[2] = (unsigned char)(( 2u << 6) | ( cp       & 0x3Fu));
 301                 bytes[1] = (unsigned char)(( 2u << 6) | ((cp >> 6) & 0x3Fu));
 302                 bytes[0] = (unsigned char)((14u << 4) |  (cp >> 12));
 303                 return 3;
 304         }
 305
 306         if(cp <= 0x10FFFFul)
 307         {
 308                 bytes[3] = (unsigned char)(( 2u << 6) | ( cp        & 0x3Fu));
 309                 bytes[2] = (unsigned char)(( 2u << 6) | ((cp >>  6) & 0x3Fu));
 310                 bytes[1] = (unsigned char)(( 2u << 6) | ((cp >> 12) & 0x3Fu));
 311                 bytes[0] = (unsigned char)((30u << 3) |  (cp >> 18));
 312                 return 4;
 313         }
 314
 315         return 0;
 316 }
 317
 318 static _Bool parse_entity(const char *current, char **to,
 319         const char **from)
 320 {
 321         const char *end = strchr(current, ';');
 322         if(!end) return 0;
 323
 324         if(current[1] == '#')
 325         {
 326                 char *tail = NULL;
 327                 errno = 0;
 328
 329                 _Bool hex = current[2] == 'x' || current[2] == 'X';
 330
 331                 unsigned long cp = strtoul(
 332                         current + (hex ? 3 : 2), &tail, hex ? 16 : 10);
 333
 334                 if(tail == end && !errno && cp <= UNICODE_MAX)
 335                 {
 336                         *to += putc_utf8(cp, *to);
 337                         *from = end + 1;
 338
 339                         return 1;
 340                 }
 341         }
 342         else
 343         {
 344                 const char *entity = get_named_entity(&current[1]);
 345                 if(entity)
 346                 {
 347                         size_t len = strlen(entity);
 348                         memcpy(*to, entity, len);
 349
 350                         *to += len;
 351                         *from = end + 1;
 352
 353                         return 1;
 354                 }
 355         }
 356
 357         return 0;
 358 }
 359
 360 size_t decode_html_entities_utf8(char *dest, const char *src)
 361 {
 362         if(!src) src = dest;
 363
 364         char *to = dest;
 365         const char *from = src;
 366
 367         const char *current;
 368         while((current = strchr(from, '&')))
 369         {
 370                 memcpy(to, from, (size_t)(current - from));
 371                 to += current - from;
 372
 373                 if(parse_entity(current, &to, &from))
 374                         continue;
 375
 376                 from = current;
 377                 *to++ = *from++;
 378         }
 379
 380         size_t remaining = strlen(from);
 381
 382         memcpy(to, from, remaining);
 383         to += remaining;
 384
 385         *to = 0;
 386         return (size_t)(to - dest);
 387 }