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     if (m_HandlersStack)
  87     {
  88         wxList& tmp = *m_HandlersStack;
  89         wxList::iterator it, en;
  90         for( it = tmp.begin(), en = tmp.end(); it != en; ++it )
  91             delete (wxHashTable*)*it;
  92         tmp.clear();
  93     }
  94     delete m_HandlersStack;
  95     m_HandlersHash.Clear();
  96     WX_CLEAR_LIST(wxList, m_HandlersList);
  97     delete m_entitiesParser;
  98 }
  99
 100 wxObject* wxHtmlParser::Parse(const wxString& source)
 101 {
 102     InitParser(source);
 103     DoParsing();
 104     wxObject *result = GetProduct();
 105     DoneParser();
 106     return result;
 107 }
 108
 109 void wxHtmlParser::InitParser(const wxString& source)
 110 {
 111     SetSource(source);
 112     m_stopParsing = FALSE;
 113 }
 114
 115 void wxHtmlParser::DoneParser()
 116 {
 117     DestroyDOMTree();
 118 }
 119
 120 void wxHtmlParser::SetSource(const wxString& src)
 121 {
 122     DestroyDOMTree();
 123     m_Source = src;
 124     CreateDOMTree();
 125     m_CurTag = NULL;
 126     m_CurTextPiece = 0;
 127 }
 128
 129 void wxHtmlParser::CreateDOMTree()
 130 {
 131     wxHtmlTagsCache cache(m_Source);
 132     m_TextPieces = new wxHtmlTextPieces;
 133     CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache);
 134     m_CurTextPiece = 0;
 135 }
 136
 137 extern bool wxIsCDATAElement(const wxChar *tag);
 138
 139 void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
 140                                     int begin_pos, int end_pos,
 141                                     wxHtmlTagsCache *cache)
 142 {
 143     if (end_pos <= begin_pos) return;
 144
 145     wxChar c;
 146     int i = begin_pos;
 147     int textBeginning = begin_pos;
 148
 149     // If the tag contains CDATA text, we include the text between beginning
 150     // and ending tag verbosely. Setting i=end_pos will skip to the very
 151     // end of this function where text piece is added, bypassing any child
 152     // tags parsing (CDATA element can't have child elements by definition):
 153     if (cur != NULL && wxIsCDATAElement(cur->GetName().c_str()))
 154     {
 155         i = end_pos;
 156     }
 157
 158     while (i < end_pos)
 159     {
 160         c = m_Source.GetChar(i);
 161
 162         if (c == wxT('<'))
 163         {
 164             // add text to m_TextPieces:
 165             if (i - textBeginning > 0)
 166                 m_TextPieces->Add(
 167                     wxHtmlTextPiece(textBeginning, i - textBeginning));
 168
 169             // if it is a comment, skip it:
 170             if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') &&
 171                                  m_Source.GetChar(i+2) == wxT('-') &&
 172                                  m_Source.GetChar(i+3) == wxT('-'))
 173             {
 174                 // Comments begin with "<!--" and end with "--[ \t\r\n]*>"
 175                 // according to HTML 4.0
 176                 int dashes = 0;
 177                 i += 4;
 178                 while (i < end_pos)
 179                 {
 180                     c = m_Source.GetChar(i++);
 181                     if ((c == wxT(' ') || c == wxT('\n') ||
 182                         c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
 183                     else if (c == wxT('>') && dashes >= 2)
 184                     {
 185                         textBeginning = i;
 186                         break;
 187                     }
 188                     else if (c == wxT('-'))
 189                         dashes++;
 190                     else
 191                         dashes = 0;
 192                 }
 193             }
 194
 195             // add another tag to the tree:
 196             else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))
 197                 {
 198                 wxHtmlTag *chd;
 199                 if (cur)
 200                     chd = new wxHtmlTag(cur, m_Source,
 201                                         i, end_pos, cache, m_entitiesParser);
 202                 else
 203                 {
 204                     chd = new wxHtmlTag(NULL, m_Source,
 205                                         i, end_pos, cache, m_entitiesParser);
 206                     if (!m_Tags)
 207                     {
 208                         // if this is the first tag to be created make the root
 209                         // m_Tags point to it:
 210                         m_Tags = chd;
 211                     }
 212                     else
 213                     {
 214                         // if there is already a root tag add this tag as
 215                         // the last sibling:
 216                         chd->m_Prev = m_Tags->GetLastSibling();
 217                         chd->m_Prev->m_Next = chd;
 218                     }
 219                 }
 220
 221                 if (chd->HasEnding())
 222                 {
 223                     CreateDOMSubTree(chd,
 224                                      chd->GetBeginPos(), chd->GetEndPos1(),
 225                                      cache);
 226                     i = chd->GetEndPos2();
 227                 }
 228                 else
 229                     i = chd->GetBeginPos();
 230
 231                 textBeginning = i;
 232             }
 233
 234             // ... or skip ending tag:
 235             else
 236             {
 237                 while (i < end_pos && m_Source.GetChar(i) != wxT('>')) i++;
 238                 textBeginning = i+1;
 239             }
 240         }
 241         else i++;
 242     }
 243
 244     // add remaining text to m_TextPieces:
 245     if (end_pos - textBeginning > 0)
 246         m_TextPieces->Add(
 247             wxHtmlTextPiece(textBeginning, end_pos - textBeginning));
 248 }
 249
 250 void wxHtmlParser::DestroyDOMTree()
 251 {
 252     wxHtmlTag *t1, *t2;
 253     t1 = m_Tags;
 254     while (t1)
 255     {
 256         t2 = t1->GetNextSibling();
 257         delete t1;
 258         t1 = t2;
 259     }
 260     m_Tags = m_CurTag = NULL;
 261
 262     delete m_TextPieces;
 263     m_TextPieces = NULL;
 264 }
 265
 266 void wxHtmlParser::DoParsing()
 267 {
 268     m_CurTag = m_Tags;
 269     m_CurTextPiece = 0;
 270     DoParsing(0, m_Source.Length());
 271 }
 272
 273 void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
 274 {
 275     if (end_pos <= begin_pos) return;
 276
 277     wxHtmlTextPieces& pieces = *m_TextPieces;
 278     size_t piecesCnt = pieces.GetCount();
 279
 280     while (begin_pos < end_pos)
 281     {
 282         while (m_CurTag && m_CurTag->GetBeginPos() < begin_pos)
 283             m_CurTag = m_CurTag->GetNextTag();
 284         while (m_CurTextPiece < piecesCnt &&
 285                pieces[m_CurTextPiece].m_pos < begin_pos)
 286             m_CurTextPiece++;
 287
 288         if (m_CurTextPiece < piecesCnt &&
 289             (!m_CurTag ||
 290              pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
 291         {
 292             // Add text:
 293             AddText(GetEntitiesParser()->Parse(
 294                        m_Source.Mid(pieces[m_CurTextPiece].m_pos,
 295                                     pieces[m_CurTextPiece].m_lng)));
 296             begin_pos = pieces[m_CurTextPiece].m_pos +
 297                         pieces[m_CurTextPiece].m_lng;
 298             m_CurTextPiece++;
 299         }
 300         else if (m_CurTag)
 301         {
 302             // Add tag:
 303             if (m_CurTag)
 304             {
 305                 if (m_CurTag->HasEnding())
 306                     begin_pos = m_CurTag->GetEndPos2();
 307                 else
 308                     begin_pos = m_CurTag->GetBeginPos();
 309             }
 310             wxHtmlTag *t = m_CurTag;
 311             m_CurTag = m_CurTag->GetNextTag();
 312             AddTag(*t);
 313             if (m_stopParsing)
 314                 return;
 315         }
 316         else break;
 317     }
 318 }
 319
 320 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 321 {
 322     wxHtmlTagHandler *h;
 323     bool inner = FALSE;
 324
 325     h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
 326     if (h)
 327     {
 328         inner = h->HandleTag(tag);
 329         if (m_stopParsing)
 330             return;
 331     }
 332     if (!inner)
 333     {
 334         if (tag.HasEnding())
 335             DoParsing(tag.GetBeginPos(), tag.GetEndPos1());
 336     }
 337 }
 338
 339 void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 340 {
 341     wxString s(handler->GetSupportedTags());
 342     wxStringTokenizer tokenizer(s, wxT(", "));
 343
 344     while (tokenizer.HasMoreTokens())
 345         m_HandlersHash.Put(tokenizer.GetNextToken(), handler);
 346
 347     if (m_HandlersList.IndexOf(handler) == wxNOT_FOUND)
 348         m_HandlersList.Append(handler);
 349
 350     handler->SetParser(this);
 351 }
 352
 353 void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 354 {
 355     wxStringTokenizer tokenizer(tags, wxT(", "));
 356     wxString key;
 357
 358     if (m_HandlersStack == NULL)
 359     {
 360         m_HandlersStack = new wxList;
 361     }
 362
 363     m_HandlersStack->Insert((wxObject*)new wxHashTable(m_HandlersHash));
 364
 365     while (tokenizer.HasMoreTokens())
 366     {
 367         key = tokenizer.GetNextToken();
 368         m_HandlersHash.Delete(key);
 369         m_HandlersHash.Put(key, handler);
 370     }
 371 }
 372
 373 void wxHtmlParser::PopTagHandler()
 374 {
 375     wxList::compatibility_iterator first;
 376
 377     if ( !m_HandlersStack ||
 378 #if wxUSE_STL
 379          !(first = m_HandlersStack->GetFirst())
 380 #else // !wxUSE_STL
 381          ((first = m_HandlersStack->GetFirst()) == NULL)
 382 #endif // wxUSE_STL/!wxUSE_STL
 383         )
 384     {
 385         wxLogWarning(_("Warning: attempt to remove HTML tag handler from empty stack."));
 386         return;
 387     }
 388     m_HandlersHash = *((wxHashTable*) first->GetData());
 389     delete (wxHashTable*) first->GetData();
 390     m_HandlersStack->Erase(first);
 391 }
 392
 393 void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
 394 {
 395     wxHtmlParserState *s = new wxHtmlParserState;
 396
 397     s->m_curTag = m_CurTag;
 398     s->m_tags = m_Tags;
 399     s->m_textPieces = m_TextPieces;
 400     s->m_curTextPiece = m_CurTextPiece;
 401     s->m_source = m_Source;
 402
 403     s->m_nextState = m_SavedStates;
 404     m_SavedStates = s;
 405
 406     m_CurTag = NULL;
 407     m_Tags = NULL;
 408     m_TextPieces = NULL;
 409     m_CurTextPiece = 0;
 410     m_Source = wxEmptyString;
 411
 412     SetSource(src);
 413 }
 414
 415 bool wxHtmlParser::RestoreState()
 416 {
 417     if (!m_SavedStates) return FALSE;
 418
 419     DestroyDOMTree();
 420
 421     wxHtmlParserState *s = m_SavedStates;
 422     m_SavedStates = s->m_nextState;
 423
 424     m_CurTag = s->m_curTag;
 425     m_Tags = s->m_tags;
 426     m_TextPieces = s->m_textPieces;
 427     m_CurTextPiece = s->m_curTextPiece;
 428     m_Source = s->m_source;
 429
 430     delete s;
 431     return TRUE;
 432 }
 433
 434 //-----------------------------------------------------------------------------
 435 // wxHtmlTagHandler
 436 //-----------------------------------------------------------------------------
 437
 438 IMPLEMENT_ABSTRACT_CLASS(wxHtmlTagHandler,wxObject)
 439
 440
 441 //-----------------------------------------------------------------------------
 442 // wxHtmlEntitiesParser
 443 //-----------------------------------------------------------------------------
 444
 445 IMPLEMENT_DYNAMIC_CLASS(wxHtmlEntitiesParser,wxObject)
 446
 447 wxHtmlEntitiesParser::wxHtmlEntitiesParser()
 448 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 449     : m_conv(NULL), m_encoding(wxFONTENCODING_SYSTEM)
 450 #endif
 451 {
 452 }
 453
 454 wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 455 {
 456 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 457     delete m_conv;
 458 #endif
 459 }
 460
 461 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 462 {
 463 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 464     if (encoding == m_encoding)
 465         return;
 466
 467     delete m_conv;
 468
 469     m_encoding = encoding;
 470     if (m_encoding == wxFONTENCODING_SYSTEM)
 471         m_conv = NULL;
 472     else
 473         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
 474 #else
 475     (void) encoding;
 476 #endif
 477 }
 478
 479 wxString wxHtmlEntitiesParser::Parse(const wxString& input)
 480 {
 481     const wxChar *c, *last;
 482     const wxChar *in_str = input.c_str();
 483     wxString output;
 484
 485     output.reserve(input.length());
 486
 487     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
 488     {
 489         if (*c == wxT('&'))
 490         {
 491             if (c - last > 0)
 492                 output.append(last, c - last);
 493             if (++c == wxT('\0')) break;
 494
 495             wxString entity;
 496             const wxChar *ent_s = c;
 497             wxChar entity_char;
 498
 499             for (; (*c >= wxT('a') && *c <= wxT('z')) ||
 500                    (*c >= wxT('A') && *c <= wxT('Z')) ||
 501                    (*c >= wxT('0') && *c <= wxT('9')) ||
 502                    *c == wxT('_') || *c == wxT('#'); c++) {}
 503             entity.append(ent_s, c - ent_s);
 504             if (*c != wxT(';')) c--;
 505             last = c+1;
 506             entity_char = GetEntityChar(entity);
 507             if (entity_char)
 508                 output << entity_char;
 509             else
 510             {
 511                 output.append(ent_s-1, c-ent_s+2);
 512                 wxLogDebug(wxT("Unrecognized HTML entity: '%s'"), entity.c_str());
 513             }
 514         }
 515     }
 516     if (*last != wxT('\0'))
 517         output.append(last);
 518     return output;
 519 }
 520
 521 struct wxHtmlEntityInfo
 522 {
 523     const wxChar *name;
 524     unsigned code;
 525 };
 526
 527 extern "C" int LINKAGEMODE wxHtmlEntityCompare(const void *key, const void *item)
 528 {
 529     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 530 }
 531
 532 #if !wxUSE_UNICODE
 533 wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
 534 {
 535 #if wxUSE_WCHAR_T
 536     char buf[2];
 537     wchar_t wbuf[2];
 538     wbuf[0] = (wchar_t)code;
 539     wbuf[1] = 0;
 540     wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;
 541     if (conv->WC2MB(buf, wbuf, 2) == (size_t)-1)
 542         return '?';
 543     return buf[0];
 544 #else
 545     return (code < 256) ? (wxChar)code : '?';
 546 #endif
 547 }
 548 #endif
 549
 550 wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
 551 {
 552     unsigned code = 0;
 553
 554     if (entity[0] == wxT('#'))
 555     {
 556         const wxChar *ent_s = entity.c_str();
 557         const wxChar *format;
 558
 559         if (ent_s[1] == wxT('x') || ent_s[1] == wxT('X'))
 560         {
 561             format = wxT("%x");
 562             ent_s++;
 563         }
 564         else
 565             format = wxT("%u");
 566         ent_s++;
 567
 568         if (wxSscanf(ent_s, format, &code) != 1)
 569             code = 0;
 570     }
 571     else
 572     {
 573         static wxHtmlEntityInfo substitutions[] = {
 574             { wxT("AElig"),198 },
 575             { wxT("Aacute"),193 },
 576             { wxT("Acirc"),194 },
 577             { wxT("Agrave"),192 },
 578             { wxT("Alpha"),913 },
 579             { wxT("Aring"),197 },
 580             { wxT("Atilde"),195 },
 581             { wxT("Auml"),196 },
 582             { wxT("Beta"),914 },
 583             { wxT("Ccedil"),199 },
 584             { wxT("Chi"),935 },
 585             { wxT("Dagger"),8225 },
 586             { wxT("Delta"),916 },
 587             { wxT("ETH"),208 },
 588             { wxT("Eacute"),201 },
 589             { wxT("Ecirc"),202 },
 590             { wxT("Egrave"),200 },
 591             { wxT("Epsilon"),917 },
 592             { wxT("Eta"),919 },
 593             { wxT("Euml"),203 },
 594             { wxT("Gamma"),915 },
 595             { wxT("Iacute"),205 },
 596             { wxT("Icirc"),206 },
 597             { wxT("Igrave"),204 },
 598             { wxT("Iota"),921 },
 599             { wxT("Iuml"),207 },
 600             { wxT("Kappa"),922 },
 601             { wxT("Lambda"),923 },
 602             { wxT("Mu"),924 },
 603             { wxT("Ntilde"),209 },
 604             { wxT("Nu"),925 },
 605             { wxT("OElig"),338 },
 606             { wxT("Oacute"),211 },
 607             { wxT("Ocirc"),212 },
 608             { wxT("Ograve"),210 },
 609             { wxT("Omega"),937 },
 610             { wxT("Omicron"),927 },
 611             { wxT("Oslash"),216 },
 612             { wxT("Otilde"),213 },
 613             { wxT("Ouml"),214 },
 614             { wxT("Phi"),934 },
 615             { wxT("Pi"),928 },
 616             { wxT("Prime"),8243 },
 617             { wxT("Psi"),936 },
 618             { wxT("Rho"),929 },
 619             { wxT("Scaron"),352 },
 620             { wxT("Sigma"),931 },
 621             { wxT("THORN"),222 },
 622             { wxT("Tau"),932 },
 623             { wxT("Theta"),920 },
 624             { wxT("Uacute"),218 },
 625             { wxT("Ucirc"),219 },
 626             { wxT("Ugrave"),217 },
 627             { wxT("Upsilon"),933 },
 628             { wxT("Uuml"),220 },
 629             { wxT("Xi"),926 },
 630             { wxT("Yacute"),221 },
 631             { wxT("Yuml"),376 },
 632             { wxT("Zeta"),918 },
 633             { wxT("aacute"),225 },
 634             { wxT("acirc"),226 },
 635             { wxT("acute"),180 },
 636             { wxT("aelig"),230 },
 637             { wxT("agrave"),224 },
 638             { wxT("alefsym"),8501 },
 639             { wxT("alpha"),945 },
 640             { wxT("amp"),38 },
 641             { wxT("and"),8743 },
 642             { wxT("ang"),8736 },
 643             { wxT("aring"),229 },
 644             { wxT("asymp"),8776 },
 645             { wxT("atilde"),227 },
 646             { wxT("auml"),228 },
 647             { wxT("bdquo"),8222 },
 648             { wxT("beta"),946 },
 649             { wxT("brvbar"),166 },
 650             { wxT("bull"),8226 },
 651             { wxT("cap"),8745 },
 652             { wxT("ccedil"),231 },
 653             { wxT("cedil"),184 },
 654             { wxT("cent"),162 },
 655             { wxT("chi"),967 },
 656             { wxT("circ"),710 },
 657             { wxT("clubs"),9827 },
 658             { wxT("cong"),8773 },
 659             { wxT("copy"),169 },
 660             { wxT("crarr"),8629 },
 661             { wxT("cup"),8746 },
 662             { wxT("curren"),164 },
 663             { wxT("dArr"),8659 },
 664             { wxT("dagger"),8224 },
 665             { wxT("darr"),8595 },
 666             { wxT("deg"),176 },
 667             { wxT("delta"),948 },
 668             { wxT("diams"),9830 },
 669             { wxT("divide"),247 },
 670             { wxT("eacute"),233 },
 671             { wxT("ecirc"),234 },
 672             { wxT("egrave"),232 },
 673             { wxT("empty"),8709 },
 674             { wxT("emsp"),8195 },
 675             { wxT("ensp"),8194 },
 676             { wxT("epsilon"),949 },
 677             { wxT("equiv"),8801 },
 678             { wxT("eta"),951 },
 679             { wxT("eth"),240 },
 680             { wxT("euml"),235 },
 681             { wxT("euro"),8364 },
 682             { wxT("exist"),8707 },
 683             { wxT("fnof"),402 },
 684             { wxT("forall"),8704 },
 685             { wxT("frac12"),189 },
 686             { wxT("frac14"),188 },
 687             { wxT("frac34"),190 },
 688             { wxT("frasl"),8260 },
 689             { wxT("gamma"),947 },
 690             { wxT("ge"),8805 },
 691             { wxT("gt"),62 },
 692             { wxT("hArr"),8660 },
 693             { wxT("harr"),8596 },
 694             { wxT("hearts"),9829 },
 695             { wxT("hellip"),8230 },
 696             { wxT("iacute"),237 },
 697             { wxT("icirc"),238 },
 698             { wxT("iexcl"),161 },
 699             { wxT("igrave"),236 },
 700             { wxT("image"),8465 },
 701             { wxT("infin"),8734 },
 702             { wxT("int"),8747 },
 703             { wxT("iota"),953 },
 704             { wxT("iquest"),191 },
 705             { wxT("isin"),8712 },
 706             { wxT("iuml"),239 },
 707             { wxT("kappa"),954 },
 708             { wxT("lArr"),8656 },
 709             { wxT("lambda"),955 },
 710             { wxT("lang"),9001 },
 711             { wxT("laquo"),171 },
 712             { wxT("larr"),8592 },
 713             { wxT("lceil"),8968 },
 714             { wxT("ldquo"),8220 },
 715             { wxT("le"),8804 },
 716             { wxT("lfloor"),8970 },
 717             { wxT("lowast"),8727 },
 718             { wxT("loz"),9674 },
 719             { wxT("lrm"),8206 },
 720             { wxT("lsaquo"),8249 },
 721             { wxT("lsquo"),8216 },
 722             { wxT("lt"),60 },
 723             { wxT("macr"),175 },
 724             { wxT("mdash"),8212 },
 725             { wxT("micro"),181 },
 726             { wxT("middot"),183 },
 727             { wxT("minus"),8722 },
 728             { wxT("mu"),956 },
 729             { wxT("nabla"),8711 },
 730             { wxT("nbsp"),160 },
 731             { wxT("ndash"),8211 },
 732             { wxT("ne"),8800 },
 733             { wxT("ni"),8715 },
 734             { wxT("not"),172 },
 735             { wxT("notin"),8713 },
 736             { wxT("nsub"),8836 },
 737             { wxT("ntilde"),241 },
 738             { wxT("nu"),957 },
 739             { wxT("oacute"),243 },
 740             { wxT("ocirc"),244 },
 741             { wxT("oelig"),339 },
 742             { wxT("ograve"),242 },
 743             { wxT("oline"),8254 },
 744             { wxT("omega"),969 },
 745             { wxT("omicron"),959 },
 746             { wxT("oplus"),8853 },
 747             { wxT("or"),8744 },
 748             { wxT("ordf"),170 },
 749             { wxT("ordm"),186 },
 750             { wxT("oslash"),248 },
 751             { wxT("otilde"),245 },
 752             { wxT("otimes"),8855 },
 753             { wxT("ouml"),246 },
 754             { wxT("para"),182 },
 755             { wxT("part"),8706 },
 756             { wxT("permil"),8240 },
 757             { wxT("perp"),8869 },
 758             { wxT("phi"),966 },
 759             { wxT("pi"),960 },
 760             { wxT("piv"),982 },
 761             { wxT("plusmn"),177 },
 762             { wxT("pound"),163 },
 763             { wxT("prime"),8242 },
 764             { wxT("prod"),8719 },
 765             { wxT("prop"),8733 },
 766             { wxT("psi"),968 },
 767             { wxT("quot"),34 },
 768             { wxT("rArr"),8658 },
 769             { wxT("radic"),8730 },
 770             { wxT("rang"),9002 },
 771             { wxT("raquo"),187 },
 772             { wxT("rarr"),8594 },
 773             { wxT("rceil"),8969 },
 774             { wxT("rdquo"),8221 },
 775             { wxT("real"),8476 },
 776             { wxT("reg"),174 },
 777             { wxT("rfloor"),8971 },
 778             { wxT("rho"),961 },
 779             { wxT("rlm"),8207 },
 780             { wxT("rsaquo"),8250 },
 781             { wxT("rsquo"),8217 },
 782             { wxT("sbquo"),8218 },
 783             { wxT("scaron"),353 },
 784             { wxT("sdot"),8901 },
 785             { wxT("sect"),167 },
 786             { wxT("shy"),173 },
 787             { wxT("sigma"),963 },
 788             { wxT("sigmaf"),962 },
 789             { wxT("sim"),8764 },
 790             { wxT("spades"),9824 },
 791             { wxT("sub"),8834 },
 792             { wxT("sube"),8838 },
 793             { wxT("sum"),8721 },
 794             { wxT("sup"),8835 },
 795             { wxT("sup1"),185 },
 796             { wxT("sup2"),178 },
 797             { wxT("sup3"),179 },
 798             { wxT("supe"),8839 },
 799             { wxT("szlig"),223 },
 800             { wxT("tau"),964 },
 801             { wxT("there4"),8756 },
 802             { wxT("theta"),952 },
 803             { wxT("thetasym"),977 },
 804             { wxT("thinsp"),8201 },
 805             { wxT("thorn"),254 },
 806             { wxT("tilde"),732 },
 807             { wxT("times"),215 },
 808             { wxT("trade"),8482 },
 809             { wxT("uArr"),8657 },
 810             { wxT("uacute"),250 },
 811             { wxT("uarr"),8593 },
 812             { wxT("ucirc"),251 },
 813             { wxT("ugrave"),249 },
 814             { wxT("uml"),168 },
 815             { wxT("upsih"),978 },
 816             { wxT("upsilon"),965 },
 817             { wxT("uuml"),252 },
 818             { wxT("weierp"),8472 },
 819             { wxT("xi"),958 },
 820             { wxT("yacute"),253 },
 821             { wxT("yen"),165 },
 822             { wxT("yuml"),255 },
 823             { wxT("zeta"),950 },
 824             { wxT("zwj"),8205 },
 825             { wxT("zwnj"),8204 },
 826             {NULL, 0}};
 827         static size_t substitutions_cnt = 0;
 828
 829         if (substitutions_cnt == 0)
 830             while (substitutions[substitutions_cnt].code != 0)
 831                 substitutions_cnt++;
 832
 833         wxHtmlEntityInfo *info;
 834         info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
 835                                            substitutions_cnt,
 836                                            sizeof(wxHtmlEntityInfo),
 837                                            wxHtmlEntityCompare);
 838         if (info)
 839             code = info->code;
 840     }
 841
 842     if (code == 0)
 843         return 0;
 844     else
 845         return GetCharForCode(code);
 846 }
 847
 848 wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
 849                                 const wxString& url) const
 850 {
 851     return GetFS()->OpenFile(url);
 852 }
 853
 854
 855 //-----------------------------------------------------------------------------
 856 // wxHtmlParser::ExtractCharsetInformation
 857 //-----------------------------------------------------------------------------
 858
 859 class wxMetaTagParser : public wxHtmlParser
 860 {
 861 public:
 862     wxMetaTagParser() { }
 863
 864     wxObject* GetProduct() { return NULL; }
 865
 866 protected:
 867     virtual void AddText(const wxChar* WXUNUSED(txt)) {}
 868
 869     DECLARE_NO_COPY_CLASS(wxMetaTagParser)
 870 };
 871
 872 class wxMetaTagHandler : public wxHtmlTagHandler
 873 {
 874 public:
 875     wxMetaTagHandler(wxString *retval) : wxHtmlTagHandler(), m_retval(retval) {}
 876     wxString GetSupportedTags() { return wxT("META,BODY"); }
 877     bool HandleTag(const wxHtmlTag& tag);
 878
 879 private:
 880     wxString *m_retval;
 881
 882     DECLARE_NO_COPY_CLASS(wxMetaTagHandler)
 883 };
 884
 885 bool wxMetaTagHandler::HandleTag(const wxHtmlTag& tag)
 886 {
 887     if (tag.GetName() == _T("BODY"))
 888     {
 889         m_Parser->StopParsing();
 890         return FALSE;
 891     }
 892
 893     if (tag.HasParam(_T("HTTP-EQUIV")) &&
 894         tag.GetParam(_T("HTTP-EQUIV")).IsSameAs(_T("Content-Type"), false) &&
 895         tag.HasParam(_T("CONTENT")))
 896     {
 897         wxString content = tag.GetParam(_T("CONTENT"));
 898         if (content.Left(19) == _T("text/html; charset="))
 899         {
 900             *m_retval = content.Mid(19);
 901             m_Parser->StopParsing();
 902         }
 903     }
 904     return FALSE;
 905 }
 906
 907
 908 /*static*/
 909 wxString wxHtmlParser::ExtractCharsetInformation(const wxString& markup)
 910 {
 911     wxString charset;
 912     wxMetaTagParser parser;
 913     parser.AddTagHandler(new wxMetaTagHandler(&charset));
 914     parser.Parse(markup);
 915     return charset;
 916 }
 917
 918
 919 #endif