src/html/htmlpars.cpp

   1 /////////////////////////////////////////////////////////////////////////////
   2 // Name:        htmlpars.cpp
   3 // Purpose:     wxHtmlParser class (generic parser)
   4 // Author:      Vaclav Slavik
   5 // RCS-ID:      $Id$
   6 // Copyright:   (c) 1999 Vaclav Slavik
   7 // Licence:     wxWindows Licence
   8 /////////////////////////////////////////////////////////////////////////////
   9
  10
  11 #ifdef __GNUG__
  12 #pragma implementation "htmlpars.h"
  13 #endif
  14
  15 #include "wx/wxprec.h"
  16
  17 #include "wx/defs.h"
  18 #if wxUSE_HTML && wxUSE_STREAMS
  19
  20 #ifdef __BORLANDC__
  21 #pragma hdrstop
  22 #endif
  23
  24 #ifndef WXPRECOMP
  25     #include "wx/log.h"
  26     #include "wx/intl.h"
  27 #endif
  28
  29 #include "wx/tokenzr.h"
  30 #include "wx/wfstream.h"
  31 #include "wx/url.h"
  32 #include "wx/fontmap.h"
  33 #include "wx/html/htmldefs.h"
  34 #include "wx/html/htmlpars.h"
  35 #include "wx/dynarray.h"
  36 #include "wx/arrimpl.cpp"
  37
  38 //-----------------------------------------------------------------------------
  39 // wxHtmlParser helpers
  40 //-----------------------------------------------------------------------------
  41
  42 class wxHtmlTextPiece
  43 {
  44 public:
  45     wxHtmlTextPiece(int pos, int lng) : m_pos(pos), m_lng(lng) {}
  46     int m_pos, m_lng;
  47 };
  48
  49 WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
  50 WX_DEFINE_OBJARRAY(wxHtmlTextPieces);
  51
  52 class wxHtmlParserState
  53 {
  54 public:
  55     wxHtmlTag         *m_curTag;
  56     wxHtmlTag         *m_tags;
  57     wxHtmlTextPieces  *m_textPieces;
  58     int                m_curTextPiece;
  59     wxString           m_source;
  60     wxHtmlParserState *m_nextState;
  61 };
  62
  63 //-----------------------------------------------------------------------------
  64 // wxHtmlParser
  65 //-----------------------------------------------------------------------------
  66
  67 IMPLEMENT_ABSTRACT_CLASS(wxHtmlParser,wxObject)
  68
  69 wxHtmlParser::wxHtmlParser()
  70     : wxObject(), m_HandlersHash(wxKEY_STRING),
  71       m_FS(NULL), m_HandlersStack(NULL)
  72 {
  73     m_entitiesParser = new wxHtmlEntitiesParser;
  74     m_Tags = NULL;
  75     m_CurTag = NULL;
  76     m_TextPieces = NULL;
  77     m_CurTextPiece = 0;
  78     m_SavedStates = NULL;
  79 }
  80
  81 wxHtmlParser::~wxHtmlParser()
  82 {
  83     while (RestoreState()) {}
  84     DestroyDOMTree();
  85
  86     delete m_HandlersStack;
  87     m_HandlersHash.Clear();
  88     m_HandlersList.DeleteContents(TRUE);
  89     m_HandlersList.Clear();
  90     delete m_entitiesParser;
  91 }
  92
  93 wxObject* wxHtmlParser::Parse(const wxString& source)
  94 {
  95     InitParser(source);
  96     DoParsing();
  97     wxObject *result = GetProduct();
  98     DoneParser();
  99     return result;
 100 }
 101
 102 void wxHtmlParser::InitParser(const wxString& source)
 103 {
 104     SetSource(source);
 105     m_stopParsing = FALSE;
 106 }
 107
 108 void wxHtmlParser::DoneParser()
 109 {
 110     DestroyDOMTree();
 111 }
 112
 113 void wxHtmlParser::SetSource(const wxString& src)
 114 {
 115     DestroyDOMTree();
 116     m_Source = src;
 117     CreateDOMTree();
 118     m_CurTag = NULL;
 119     m_CurTextPiece = 0;
 120 }
 121
 122 void wxHtmlParser::CreateDOMTree()
 123 {
 124     wxHtmlTagsCache cache(m_Source);
 125     m_TextPieces = new wxHtmlTextPieces;
 126     CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache);
 127     m_CurTextPiece = 0;
 128 }
 129
 130 extern bool wxIsCDATAElement(const wxChar *tag);
 131
 132 void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
 133                                     int begin_pos, int end_pos,
 134                                     wxHtmlTagsCache *cache)
 135 {
 136     if (end_pos <= begin_pos) return;
 137
 138     wxChar c;
 139     int i = begin_pos;
 140     int textBeginning = begin_pos;
 141
 142     // If the tag contains CDATA text, we include the text between beginning
 143     // and ending tag verbosely. Setting i=end_pos will skip to the very
 144     // end of this function where text piece is added, bypassing any child
 145     // tags parsing (CDATA element can't have child elements by definition):
 146     if (cur != NULL && wxIsCDATAElement(cur->GetName().c_str()))
 147     {
 148         i = end_pos;
 149     }
 150
 151     while (i < end_pos)
 152     {
 153         c = m_Source.GetChar(i);
 154
 155         if (c == wxT('<'))
 156         {
 157             // add text to m_TextPieces:
 158             if (i - textBeginning > 0)
 159                 m_TextPieces->Add(
 160                     wxHtmlTextPiece(textBeginning, i - textBeginning));
 161
 162             // if it is a comment, skip it:
 163             if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') &&
 164                                  m_Source.GetChar(i+2) == wxT('-') &&
 165                                  m_Source.GetChar(i+3) == wxT('-'))
 166             {
 167                 // Comments begin with "<!--" and end with "--[ \t\r\n]*>"
 168                 // according to HTML 4.0
 169                 int dashes = 0;
 170                 i += 4;
 171                 while (i < end_pos)
 172                 {
 173                     c = m_Source.GetChar(i++);
 174                     if ((c == wxT(' ') || c == wxT('\n') ||
 175                         c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
 176                     else if (c == wxT('>') && dashes >= 2)
 177                     {
 178                         textBeginning = i;
 179                         break;
 180                     }
 181                     else if (c == wxT('-'))
 182                         dashes++;
 183                     else
 184                         dashes = 0;
 185                 }
 186             }
 187
 188             // add another tag to the tree:
 189             else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))
 190                 {
 191                 wxHtmlTag *chd;
 192                 if (cur)
 193                     chd = new wxHtmlTag(cur, m_Source,
 194                                         i, end_pos, cache, m_entitiesParser);
 195                 else
 196                 {
 197                     chd = new wxHtmlTag(NULL, m_Source,
 198                                         i, end_pos, cache, m_entitiesParser);
 199                     if (!m_Tags)
 200                     {
 201                         // if this is the first tag to be created make the root
 202                         // m_Tags point to it:
 203                         m_Tags = chd;
 204                     }
 205                     else
 206                     {
 207                         // if there is already a root tag add this tag as
 208                         // the last sibling:
 209                         chd->m_Prev = m_Tags->GetLastSibling();
 210                         chd->m_Prev->m_Next = chd;
 211                     }
 212                 }
 213
 214                 if (chd->HasEnding())
 215                 {
 216                     CreateDOMSubTree(chd,
 217                                      chd->GetBeginPos(), chd->GetEndPos1(),
 218                                      cache);
 219                     i = chd->GetEndPos2();
 220                 }
 221                 else
 222                     i = chd->GetBeginPos();
 223
 224                 textBeginning = i;
 225             }
 226
 227             // ... or skip ending tag:
 228             else
 229             {
 230                 while (i < end_pos && m_Source.GetChar(i) != wxT('>')) i++;
 231                 textBeginning = i+1;
 232             }
 233         }
 234         else i++;
 235     }
 236
 237     // add remaining text to m_TextPieces:
 238     if (end_pos - textBeginning > 0)
 239         m_TextPieces->Add(
 240             wxHtmlTextPiece(textBeginning, end_pos - textBeginning));
 241 }
 242
 243 void wxHtmlParser::DestroyDOMTree()
 244 {
 245     wxHtmlTag *t1, *t2;
 246     t1 = m_Tags;
 247     while (t1)
 248     {
 249         t2 = t1->GetNextSibling();
 250         delete t1;
 251         t1 = t2;
 252     }
 253     m_Tags = m_CurTag = NULL;
 254
 255     delete m_TextPieces;
 256     m_TextPieces = NULL;
 257 }
 258
 259 void wxHtmlParser::DoParsing()
 260 {
 261     m_CurTag = m_Tags;
 262     m_CurTextPiece = 0;
 263     DoParsing(0, m_Source.Length());
 264 }
 265
 266 void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
 267 {
 268     if (end_pos <= begin_pos) return;
 269
 270     wxHtmlTextPieces& pieces = *m_TextPieces;
 271     size_t piecesCnt = pieces.GetCount();
 272
 273     while (begin_pos < end_pos)
 274     {
 275         while (m_CurTag && m_CurTag->GetBeginPos() < begin_pos)
 276             m_CurTag = m_CurTag->GetNextTag();
 277         while (m_CurTextPiece < piecesCnt &&
 278                pieces[m_CurTextPiece].m_pos < begin_pos)
 279             m_CurTextPiece++;
 280
 281         if (m_CurTextPiece < piecesCnt &&
 282             (!m_CurTag ||
 283              pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
 284         {
 285             // Add text:
 286             AddText(GetEntitiesParser()->Parse(
 287                        m_Source.Mid(pieces[m_CurTextPiece].m_pos,
 288                                     pieces[m_CurTextPiece].m_lng)));
 289             begin_pos = pieces[m_CurTextPiece].m_pos +
 290                         pieces[m_CurTextPiece].m_lng;
 291             m_CurTextPiece++;
 292         }
 293         else if (m_CurTag)
 294         {
 295             // Add tag:
 296             if (m_CurTag)
 297             {
 298                 if (m_CurTag->HasEnding())
 299                     begin_pos = m_CurTag->GetEndPos2();
 300                 else
 301                     begin_pos = m_CurTag->GetBeginPos();
 302             }
 303             wxHtmlTag *t = m_CurTag;
 304             m_CurTag = m_CurTag->GetNextTag();
 305             AddTag(*t);
 306             if (m_stopParsing)
 307                 return;
 308         }
 309         else break;
 310     }
 311 }
 312
 313 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 314 {
 315     wxHtmlTagHandler *h;
 316     bool inner = FALSE;
 317
 318     h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
 319     if (h)
 320     {
 321         inner = h->HandleTag(tag);
 322         if (m_stopParsing)
 323             return;
 324     }
 325     if (!inner)
 326     {
 327         if (tag.HasEnding())
 328             DoParsing(tag.GetBeginPos(), tag.GetEndPos1());
 329     }
 330 }
 331
 332 void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 333 {
 334     wxString s(handler->GetSupportedTags());
 335     wxStringTokenizer tokenizer(s, wxT(", "));
 336
 337     while (tokenizer.HasMoreTokens())
 338         m_HandlersHash.Put(tokenizer.GetNextToken(), handler);
 339
 340     if (m_HandlersList.IndexOf(handler) == wxNOT_FOUND)
 341         m_HandlersList.Append(handler);
 342
 343     handler->SetParser(this);
 344 }
 345
 346 void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 347 {
 348     wxStringTokenizer tokenizer(tags, wxT(", "));
 349     wxString key;
 350
 351     if (m_HandlersStack == NULL)
 352     {
 353         m_HandlersStack = new wxList;
 354         m_HandlersStack->DeleteContents(TRUE);
 355     }
 356
 357     m_HandlersStack->Insert(new wxHashTable(m_HandlersHash));
 358
 359     while (tokenizer.HasMoreTokens())
 360     {
 361         key = tokenizer.GetNextToken();
 362         m_HandlersHash.Delete(key);
 363         m_HandlersHash.Put(key, handler);
 364     }
 365 }
 366
 367 void wxHtmlParser::PopTagHandler()
 368 {
 369     wxNode *first;
 370
 371     if (m_HandlersStack == NULL ||
 372         (first = m_HandlersStack->GetFirst()) == NULL)
 373     {
 374         wxLogWarning(_("Warning: attempt to remove HTML tag handler from empty stack."));
 375         return;
 376     }
 377     m_HandlersHash = *((wxHashTable*) first->GetData());
 378     m_HandlersStack->DeleteNode(first);
 379 }
 380
 381 void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
 382 {
 383     wxHtmlParserState *s = new wxHtmlParserState;
 384
 385     s->m_curTag = m_CurTag;
 386     s->m_tags = m_Tags;
 387     s->m_textPieces = m_TextPieces;
 388     s->m_curTextPiece = m_CurTextPiece;
 389     s->m_source = m_Source;
 390
 391     s->m_nextState = m_SavedStates;
 392     m_SavedStates = s;
 393
 394     m_CurTag = NULL;
 395     m_Tags = NULL;
 396     m_TextPieces = NULL;
 397     m_CurTextPiece = 0;
 398     m_Source = wxEmptyString;
 399
 400     SetSource(src);
 401 }
 402
 403 bool wxHtmlParser::RestoreState()
 404 {
 405     if (!m_SavedStates) return FALSE;
 406
 407     DestroyDOMTree();
 408
 409     wxHtmlParserState *s = m_SavedStates;
 410     m_SavedStates = s->m_nextState;
 411
 412     m_CurTag = s->m_curTag;
 413     m_Tags = s->m_tags;
 414     m_TextPieces = s->m_textPieces;
 415     m_CurTextPiece = s->m_curTextPiece;
 416     m_Source = s->m_source;
 417
 418     delete s;
 419     return TRUE;
 420 }
 421
 422 //-----------------------------------------------------------------------------
 423 // wxHtmlTagHandler
 424 //-----------------------------------------------------------------------------
 425
 426 IMPLEMENT_ABSTRACT_CLASS(wxHtmlTagHandler,wxObject)
 427
 428
 429 //-----------------------------------------------------------------------------
 430 // wxHtmlEntitiesParser
 431 //-----------------------------------------------------------------------------
 432
 433 IMPLEMENT_DYNAMIC_CLASS(wxHtmlEntitiesParser,wxObject)
 434
 435 wxHtmlEntitiesParser::wxHtmlEntitiesParser()
 436 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 437     : m_conv(NULL), m_encoding(wxFONTENCODING_SYSTEM)
 438 #endif
 439 {
 440 }
 441
 442 wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 443 {
 444 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 445     delete m_conv;
 446 #endif
 447 }
 448
 449 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 450 {
 451 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 452     if (encoding == m_encoding)
 453         return;
 454
 455     delete m_conv;
 456
 457     m_encoding = encoding;
 458     if (m_encoding == wxFONTENCODING_SYSTEM)
 459         m_conv = NULL;
 460     else
 461         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
 462 #else
 463     (void) encoding;
 464 #endif
 465 }
 466
 467 wxString wxHtmlEntitiesParser::Parse(const wxString& input)
 468 {
 469     const wxChar *c, *last;
 470     const wxChar *in_str = input.c_str();
 471     wxString output;
 472
 473     output.reserve(input.length());
 474
 475     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
 476     {
 477         if (*c == wxT('&'))
 478         {
 479             if (c - last > 0)
 480                 output.append(last, c - last);
 481             if (++c == wxT('\0')) break;
 482
 483             wxString entity;
 484             const wxChar *ent_s = c;
 485             wxChar entity_char;
 486
 487             for (; (*c >= wxT('a') && *c <= wxT('z')) ||
 488                    (*c >= wxT('A') && *c <= wxT('Z')) ||
 489                    (*c >= wxT('0') && *c <= wxT('9')) ||
 490                    *c == wxT('_') || *c == wxT('#'); c++) {}
 491             entity.append(ent_s, c - ent_s);
 492             if (*c != wxT(';')) c--;
 493             last = c+1;
 494             entity_char = GetEntityChar(entity);
 495             if (entity_char)
 496                 output << entity_char;
 497             else
 498             {
 499                 output.append(ent_s-1, c-ent_s+2);
 500                 wxLogDebug(wxT("Unrecognized HTML entity: '%s'"), entity.c_str());
 501             }
 502         }
 503     }
 504     if (*last != wxT('\0'))
 505         output.append(last);
 506     return output;
 507 }
 508
 509 struct wxHtmlEntityInfo
 510 {
 511     const wxChar *name;
 512     unsigned code;
 513 };
 514
 515 extern "C" int LINKAGEMODE wxHtmlEntityCompare(const void *key, const void *item)
 516 {
 517     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 518 }
 519
 520 #if !wxUSE_UNICODE
 521 wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
 522 {
 523 #if wxUSE_WCHAR_T
 524     char buf[2];
 525     wchar_t wbuf[2];
 526     wbuf[0] = (wchar_t)code;
 527     wbuf[1] = 0;
 528     wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;
 529     if (conv->WC2MB(buf, wbuf, 2) == (size_t)-1)
 530         return '?';
 531     return buf[0];
 532 #else
 533     return (code < 256) ? (wxChar)code : '?';
 534 #endif
 535 }
 536 #endif
 537
 538 wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
 539 {
 540     unsigned code = 0;
 541
 542     if (entity[0] == wxT('#'))
 543     {
 544         const wxChar *ent_s = entity.c_str();
 545         const wxChar *format;
 546
 547         if (ent_s[1] == wxT('x') || ent_s[1] == wxT('X'))
 548         {
 549             format = wxT("%x");
 550             ent_s++;
 551         }
 552         else
 553             format = wxT("%u");
 554         ent_s++;
 555
 556         if (wxSscanf(ent_s, format, &code) != 1)
 557             code = 0;
 558     }
 559     else
 560     {
 561         static wxHtmlEntityInfo substitutions[] = {
 562             { wxT("AElig"),198 },
 563             { wxT("Aacute"),193 },
 564             { wxT("Acirc"),194 },
 565             { wxT("Agrave"),192 },
 566             { wxT("Alpha"),913 },
 567             { wxT("Aring"),197 },
 568             { wxT("Atilde"),195 },
 569             { wxT("Auml"),196 },
 570             { wxT("Beta"),914 },
 571             { wxT("Ccedil"),199 },
 572             { wxT("Chi"),935 },
 573             { wxT("Dagger"),8225 },
 574             { wxT("Delta"),916 },
 575             { wxT("ETH"),208 },
 576             { wxT("Eacute"),201 },
 577             { wxT("Ecirc"),202 },
 578             { wxT("Egrave"),200 },
 579             { wxT("Epsilon"),917 },
 580             { wxT("Eta"),919 },
 581             { wxT("Euml"),203 },
 582             { wxT("Gamma"),915 },
 583             { wxT("Iacute"),205 },
 584             { wxT("Icirc"),206 },
 585             { wxT("Igrave"),204 },
 586             { wxT("Iota"),921 },
 587             { wxT("Iuml"),207 },
 588             { wxT("Kappa"),922 },
 589             { wxT("Lambda"),923 },
 590             { wxT("Mu"),924 },
 591             { wxT("Ntilde"),209 },
 592             { wxT("Nu"),925 },
 593             { wxT("OElig"),338 },
 594             { wxT("Oacute"),211 },
 595             { wxT("Ocirc"),212 },
 596             { wxT("Ograve"),210 },
 597             { wxT("Omega"),937 },
 598             { wxT("Omicron"),927 },
 599             { wxT("Oslash"),216 },
 600             { wxT("Otilde"),213 },
 601             { wxT("Ouml"),214 },
 602             { wxT("Phi"),934 },
 603             { wxT("Pi"),928 },
 604             { wxT("Prime"),8243 },
 605             { wxT("Psi"),936 },
 606             { wxT("Rho"),929 },
 607             { wxT("Scaron"),352 },
 608             { wxT("Sigma"),931 },
 609             { wxT("THORN"),222 },
 610             { wxT("Tau"),932 },
 611             { wxT("Theta"),920 },
 612             { wxT("Uacute"),218 },
 613             { wxT("Ucirc"),219 },
 614             { wxT("Ugrave"),217 },
 615             { wxT("Upsilon"),933 },
 616             { wxT("Uuml"),220 },
 617             { wxT("Xi"),926 },
 618             { wxT("Yacute"),221 },
 619             { wxT("Yuml"),376 },
 620             { wxT("Zeta"),918 },
 621             { wxT("aacute"),225 },
 622             { wxT("acirc"),226 },
 623             { wxT("acute"),180 },
 624             { wxT("aelig"),230 },
 625             { wxT("agrave"),224 },
 626             { wxT("alefsym"),8501 },
 627             { wxT("alpha"),945 },
 628             { wxT("amp"),38 },
 629             { wxT("and"),8743 },
 630             { wxT("ang"),8736 },
 631             { wxT("aring"),229 },
 632             { wxT("asymp"),8776 },
 633             { wxT("atilde"),227 },
 634             { wxT("auml"),228 },
 635             { wxT("bdquo"),8222 },
 636             { wxT("beta"),946 },
 637             { wxT("brvbar"),166 },
 638             { wxT("bull"),8226 },
 639             { wxT("cap"),8745 },
 640             { wxT("ccedil"),231 },
 641             { wxT("cedil"),184 },
 642             { wxT("cent"),162 },
 643             { wxT("chi"),967 },
 644             { wxT("circ"),710 },
 645             { wxT("clubs"),9827 },
 646             { wxT("cong"),8773 },
 647             { wxT("copy"),169 },
 648             { wxT("crarr"),8629 },
 649             { wxT("cup"),8746 },
 650             { wxT("curren"),164 },
 651             { wxT("dArr"),8659 },
 652             { wxT("dagger"),8224 },
 653             { wxT("darr"),8595 },
 654             { wxT("deg"),176 },
 655             { wxT("delta"),948 },
 656             { wxT("diams"),9830 },
 657             { wxT("divide"),247 },
 658             { wxT("eacute"),233 },
 659             { wxT("ecirc"),234 },
 660             { wxT("egrave"),232 },
 661             { wxT("empty"),8709 },
 662             { wxT("emsp"),8195 },
 663             { wxT("ensp"),8194 },
 664             { wxT("epsilon"),949 },
 665             { wxT("equiv"),8801 },
 666             { wxT("eta"),951 },
 667             { wxT("eth"),240 },
 668             { wxT("euml"),235 },
 669             { wxT("euro"),8364 },
 670             { wxT("exist"),8707 },
 671             { wxT("fnof"),402 },
 672             { wxT("forall"),8704 },
 673             { wxT("frac12"),189 },
 674             { wxT("frac14"),188 },
 675             { wxT("frac34"),190 },
 676             { wxT("frasl"),8260 },
 677             { wxT("gamma"),947 },
 678             { wxT("ge"),8805 },
 679             { wxT("gt"),62 },
 680             { wxT("hArr"),8660 },
 681             { wxT("harr"),8596 },
 682             { wxT("hearts"),9829 },
 683             { wxT("hellip"),8230 },
 684             { wxT("iacute"),237 },
 685             { wxT("icirc"),238 },
 686             { wxT("iexcl"),161 },
 687             { wxT("igrave"),236 },
 688             { wxT("image"),8465 },
 689             { wxT("infin"),8734 },
 690             { wxT("int"),8747 },
 691             { wxT("iota"),953 },
 692             { wxT("iquest"),191 },
 693             { wxT("isin"),8712 },
 694             { wxT("iuml"),239 },
 695             { wxT("kappa"),954 },
 696             { wxT("lArr"),8656 },
 697             { wxT("lambda"),955 },
 698             { wxT("lang"),9001 },
 699             { wxT("laquo"),171 },
 700             { wxT("larr"),8592 },
 701             { wxT("lceil"),8968 },
 702             { wxT("ldquo"),8220 },
 703             { wxT("le"),8804 },
 704             { wxT("lfloor"),8970 },
 705             { wxT("lowast"),8727 },
 706             { wxT("loz"),9674 },
 707             { wxT("lrm"),8206 },
 708             { wxT("lsaquo"),8249 },
 709             { wxT("lsquo"),8216 },
 710             { wxT("lt"),60 },
 711             { wxT("macr"),175 },
 712             { wxT("mdash"),8212 },
 713             { wxT("micro"),181 },
 714             { wxT("middot"),183 },
 715             { wxT("minus"),8722 },
 716             { wxT("mu"),956 },
 717             { wxT("nabla"),8711 },
 718             { wxT("nbsp"),160 },
 719             { wxT("ndash"),8211 },
 720             { wxT("ne"),8800 },
 721             { wxT("ni"),8715 },
 722             { wxT("not"),172 },
 723             { wxT("notin"),8713 },
 724             { wxT("nsub"),8836 },
 725             { wxT("ntilde"),241 },
 726             { wxT("nu"),957 },
 727             { wxT("oacute"),243 },
 728             { wxT("ocirc"),244 },
 729             { wxT("oelig"),339 },
 730             { wxT("ograve"),242 },
 731             { wxT("oline"),8254 },
 732             { wxT("omega"),969 },
 733             { wxT("omicron"),959 },
 734             { wxT("oplus"),8853 },
 735             { wxT("or"),8744 },
 736             { wxT("ordf"),170 },
 737             { wxT("ordm"),186 },
 738             { wxT("oslash"),248 },
 739             { wxT("otilde"),245 },
 740             { wxT("otimes"),8855 },
 741             { wxT("ouml"),246 },
 742             { wxT("para"),182 },
 743             { wxT("part"),8706 },
 744             { wxT("permil"),8240 },
 745             { wxT("perp"),8869 },
 746             { wxT("phi"),966 },
 747             { wxT("pi"),960 },
 748             { wxT("piv"),982 },
 749             { wxT("plusmn"),177 },
 750             { wxT("pound"),163 },
 751             { wxT("prime"),8242 },
 752             { wxT("prod"),8719 },
 753             { wxT("prop"),8733 },
 754             { wxT("psi"),968 },
 755             { wxT("quot"),34 },
 756             { wxT("rArr"),8658 },
 757             { wxT("radic"),8730 },
 758             { wxT("rang"),9002 },
 759             { wxT("raquo"),187 },
 760             { wxT("rarr"),8594 },
 761             { wxT("rceil"),8969 },
 762             { wxT("rdquo"),8221 },
 763             { wxT("real"),8476 },
 764             { wxT("reg"),174 },
 765             { wxT("rfloor"),8971 },
 766             { wxT("rho"),961 },
 767             { wxT("rlm"),8207 },
 768             { wxT("rsaquo"),8250 },
 769             { wxT("rsquo"),8217 },
 770             { wxT("sbquo"),8218 },
 771             { wxT("scaron"),353 },
 772             { wxT("sdot"),8901 },
 773             { wxT("sect"),167 },
 774             { wxT("shy"),173 },
 775             { wxT("sigma"),963 },
 776             { wxT("sigmaf"),962 },
 777             { wxT("sim"),8764 },
 778             { wxT("spades"),9824 },
 779             { wxT("sub"),8834 },
 780             { wxT("sube"),8838 },
 781             { wxT("sum"),8721 },
 782             { wxT("sup"),8835 },
 783             { wxT("sup1"),185 },
 784             { wxT("sup2"),178 },
 785             { wxT("sup3"),179 },
 786             { wxT("supe"),8839 },
 787             { wxT("szlig"),223 },
 788             { wxT("tau"),964 },
 789             { wxT("there4"),8756 },
 790             { wxT("theta"),952 },
 791             { wxT("thetasym"),977 },
 792             { wxT("thinsp"),8201 },
 793             { wxT("thorn"),254 },
 794             { wxT("tilde"),732 },
 795             { wxT("times"),215 },
 796             { wxT("trade"),8482 },
 797             { wxT("uArr"),8657 },
 798             { wxT("uacute"),250 },
 799             { wxT("uarr"),8593 },
 800             { wxT("ucirc"),251 },
 801             { wxT("ugrave"),249 },
 802             { wxT("uml"),168 },
 803             { wxT("upsih"),978 },
 804             { wxT("upsilon"),965 },
 805             { wxT("uuml"),252 },
 806             { wxT("weierp"),8472 },
 807             { wxT("xi"),958 },
 808             { wxT("yacute"),253 },
 809             { wxT("yen"),165 },
 810             { wxT("yuml"),255 },
 811             { wxT("zeta"),950 },
 812             { wxT("zwj"),8205 },
 813             { wxT("zwnj"),8204 },
 814             {NULL, 0}};
 815         static size_t substitutions_cnt = 0;
 816
 817         if (substitutions_cnt == 0)
 818             while (substitutions[substitutions_cnt].code != 0)
 819                 substitutions_cnt++;
 820
 821         wxHtmlEntityInfo *info;
 822         info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
 823                                            substitutions_cnt,
 824                                            sizeof(wxHtmlEntityInfo),
 825                                            wxHtmlEntityCompare);
 826         if (info)
 827             code = info->code;
 828     }
 829
 830     if (code == 0)
 831         return 0;
 832     else
 833         return GetCharForCode(code);
 834 }
 835
 836 wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
 837                                 const wxString& url) const
 838 {
 839     return GetFS()->OpenFile(url);
 840 }
 841
 842
 843 //-----------------------------------------------------------------------------
 844 // wxHtmlParser::ExtractCharsetInformation
 845 //-----------------------------------------------------------------------------
 846
 847 class wxMetaTagParser : public wxHtmlParser
 848 {
 849 public:
 850     wxObject* GetProduct() { return NULL; }
 851 protected:
 852     virtual void AddText(const wxChar* WXUNUSED(txt)) {}
 853 };
 854
 855 class wxMetaTagHandler : public wxHtmlTagHandler
 856 {
 857 public:
 858     wxMetaTagHandler(wxString *retval) : wxHtmlTagHandler(), m_retval(retval) {}
 859     wxString GetSupportedTags() { return wxT("META,BODY"); }
 860     bool HandleTag(const wxHtmlTag& tag);
 861
 862 private:
 863     wxString *m_retval;
 864 };
 865
 866 bool wxMetaTagHandler::HandleTag(const wxHtmlTag& tag)
 867 {
 868     if (tag.GetName() == _T("BODY"))
 869     {
 870         m_Parser->StopParsing();
 871         return FALSE;
 872     }
 873
 874     if (tag.HasParam(_T("HTTP-EQUIV")) &&
 875         tag.GetParam(_T("HTTP-EQUIV")).IsSameAs(_T("Content-Type"), false) &&
 876         tag.HasParam(_T("CONTENT")))
 877     {
 878         wxString content = tag.GetParam(_T("CONTENT"));
 879         if (content.Left(19) == _T("text/html; charset="))
 880         {
 881             *m_retval = content.Mid(19);
 882             m_Parser->StopParsing();
 883         }
 884     }
 885     return FALSE;
 886 }
 887
 888
 889 /*static*/
 890 wxString wxHtmlParser::ExtractCharsetInformation(const wxString& markup)
 891 {
 892     wxString charset;
 893     wxMetaTagParser parser;
 894     parser.AddTagHandler(new wxMetaTagHandler(&charset));
 895     parser.Parse(markup);
 896     return charset;
 897 }
 898
 899
 900 #endif