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 #if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
  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 #ifdef __WXWINCE__
  39     #include "wx/msw/wince/missing.h"       // for bsearch()
  40 #endif
  41
  42 // DLL options compatibility check:
  43 #include "wx/app.h"
  44 WX_CHECK_BUILD_OPTIONS("wxHTML")
  45
  46 const wxChar *wxTRACE_HTML_DEBUG = _T("htmldebug");
  47
  48 //-----------------------------------------------------------------------------
  49 // wxHtmlParser helpers
  50 //-----------------------------------------------------------------------------
  51
  52 class wxHtmlTextPiece
  53 {
  54 public:
  55     wxHtmlTextPiece(int pos, int lng) : m_pos(pos), m_lng(lng) {}
  56     int m_pos, m_lng;
  57 };
  58
  59 WX_DECLARE_OBJARRAY(wxHtmlTextPiece, wxHtmlTextPieces);
  60 WX_DEFINE_OBJARRAY(wxHtmlTextPieces);
  61
  62 class wxHtmlParserState
  63 {
  64 public:
  65     wxHtmlTag         *m_curTag;
  66     wxHtmlTag         *m_tags;
  67     wxHtmlTextPieces  *m_textPieces;
  68     int                m_curTextPiece;
  69     wxString           m_source;
  70     wxHtmlParserState *m_nextState;
  71 };
  72
  73 //-----------------------------------------------------------------------------
  74 // wxHtmlParser
  75 //-----------------------------------------------------------------------------
  76
  77 IMPLEMENT_ABSTRACT_CLASS(wxHtmlParser,wxObject)
  78
  79 wxHtmlParser::wxHtmlParser()
  80     : wxObject(), m_HandlersHash(wxKEY_STRING),
  81       m_FS(NULL), m_HandlersStack(NULL)
  82 {
  83     m_entitiesParser = new wxHtmlEntitiesParser;
  84     m_Tags = NULL;
  85     m_CurTag = NULL;
  86     m_TextPieces = NULL;
  87     m_CurTextPiece = 0;
  88     m_SavedStates = NULL;
  89 }
  90
  91 wxHtmlParser::~wxHtmlParser()
  92 {
  93     while (RestoreState()) {}
  94     DestroyDOMTree();
  95
  96     if (m_HandlersStack)
  97     {
  98         wxList& tmp = *m_HandlersStack;
  99         wxList::iterator it, en;
 100         for( it = tmp.begin(), en = tmp.end(); it != en; ++it )
 101             delete (wxHashTable*)*it;
 102         tmp.clear();
 103     }
 104     delete m_HandlersStack;
 105     m_HandlersHash.Clear();
 106     WX_CLEAR_LIST(wxList, m_HandlersList);
 107     delete m_entitiesParser;
 108 }
 109
 110 wxObject* wxHtmlParser::Parse(const wxString& source)
 111 {
 112     InitParser(source);
 113     DoParsing();
 114     wxObject *result = GetProduct();
 115     DoneParser();
 116     return result;
 117 }
 118
 119 void wxHtmlParser::InitParser(const wxString& source)
 120 {
 121     SetSource(source);
 122     m_stopParsing = false;
 123 }
 124
 125 void wxHtmlParser::DoneParser()
 126 {
 127     DestroyDOMTree();
 128 }
 129
 130 void wxHtmlParser::SetSource(const wxString& src)
 131 {
 132     DestroyDOMTree();
 133     m_Source = src;
 134     CreateDOMTree();
 135     m_CurTag = NULL;
 136     m_CurTextPiece = 0;
 137 }
 138
 139 void wxHtmlParser::CreateDOMTree()
 140 {
 141     wxHtmlTagsCache cache(m_Source);
 142     m_TextPieces = new wxHtmlTextPieces;
 143     CreateDOMSubTree(NULL, 0, m_Source.Length(), &cache);
 144     m_CurTextPiece = 0;
 145 }
 146
 147 extern bool wxIsCDATAElement(const wxChar *tag);
 148
 149 void wxHtmlParser::CreateDOMSubTree(wxHtmlTag *cur,
 150                                     int begin_pos, int end_pos,
 151                                     wxHtmlTagsCache *cache)
 152 {
 153     if (end_pos <= begin_pos) return;
 154
 155     wxChar c;
 156     int i = begin_pos;
 157     int textBeginning = begin_pos;
 158
 159     // If the tag contains CDATA text, we include the text between beginning
 160     // and ending tag verbosely. Setting i=end_pos will skip to the very
 161     // end of this function where text piece is added, bypassing any child
 162     // tags parsing (CDATA element can't have child elements by definition):
 163     if (cur != NULL && wxIsCDATAElement(cur->GetName().c_str()))
 164     {
 165         i = end_pos;
 166     }
 167
 168     while (i < end_pos)
 169     {
 170         c = m_Source.GetChar(i);
 171
 172         if (c == wxT('<'))
 173         {
 174             // add text to m_TextPieces:
 175             if (i - textBeginning > 0)
 176                 m_TextPieces->Add(
 177                     wxHtmlTextPiece(textBeginning, i - textBeginning));
 178
 179             // if it is a comment, skip it:
 180             if (i < end_pos-6 && m_Source.GetChar(i+1) == wxT('!') &&
 181                                  m_Source.GetChar(i+2) == wxT('-') &&
 182                                  m_Source.GetChar(i+3) == wxT('-'))
 183             {
 184                 // Comments begin with "<!--" and end with "--[ \t\r\n]*>"
 185                 // according to HTML 4.0
 186                 int dashes = 0;
 187                 i += 4;
 188                 while (i < end_pos)
 189                 {
 190                     c = m_Source.GetChar(i++);
 191                     if ((c == wxT(' ') || c == wxT('\n') ||
 192                         c == wxT('\r') || c == wxT('\t')) && dashes >= 2) {}
 193                     else if (c == wxT('>') && dashes >= 2)
 194                     {
 195                         textBeginning = i;
 196                         break;
 197                     }
 198                     else if (c == wxT('-'))
 199                         dashes++;
 200                     else
 201                         dashes = 0;
 202                 }
 203             }
 204
 205             // add another tag to the tree:
 206             else if (i < end_pos-1 && m_Source.GetChar(i+1) != wxT('/'))
 207             {
 208                 wxHtmlTag *chd;
 209                 if (cur)
 210                     chd = new wxHtmlTag(cur, m_Source,
 211                                         i, end_pos, cache, m_entitiesParser);
 212                 else
 213                 {
 214                     chd = new wxHtmlTag(NULL, m_Source,
 215                                         i, end_pos, cache, m_entitiesParser);
 216                     if (!m_Tags)
 217                     {
 218                         // if this is the first tag to be created make the root
 219                         // m_Tags point to it:
 220                         m_Tags = chd;
 221                     }
 222                     else
 223                     {
 224                         // if there is already a root tag add this tag as
 225                         // the last sibling:
 226                         chd->m_Prev = m_Tags->GetLastSibling();
 227                         chd->m_Prev->m_Next = chd;
 228                     }
 229                 }
 230
 231                 if (chd->HasEnding())
 232                 {
 233                     CreateDOMSubTree(chd,
 234                                      chd->GetBeginPos(), chd->GetEndPos1(),
 235                                      cache);
 236                     i = chd->GetEndPos2();
 237                 }
 238                 else
 239                     i = chd->GetBeginPos();
 240
 241                 textBeginning = i;
 242             }
 243
 244             // ... or skip ending tag:
 245             else
 246             {
 247                 while (i < end_pos && m_Source.GetChar(i) != wxT('>')) i++;
 248                 textBeginning = i+1;
 249             }
 250         }
 251         else i++;
 252     }
 253
 254     // add remaining text to m_TextPieces:
 255     if (end_pos - textBeginning > 0)
 256         m_TextPieces->Add(
 257             wxHtmlTextPiece(textBeginning, end_pos - textBeginning));
 258 }
 259
 260 void wxHtmlParser::DestroyDOMTree()
 261 {
 262     wxHtmlTag *t1, *t2;
 263     t1 = m_Tags;
 264     while (t1)
 265     {
 266         t2 = t1->GetNextSibling();
 267         delete t1;
 268         t1 = t2;
 269     }
 270     m_Tags = m_CurTag = NULL;
 271
 272     delete m_TextPieces;
 273     m_TextPieces = NULL;
 274 }
 275
 276 void wxHtmlParser::DoParsing()
 277 {
 278     m_CurTag = m_Tags;
 279     m_CurTextPiece = 0;
 280     DoParsing(0, m_Source.Length());
 281 }
 282
 283 void wxHtmlParser::DoParsing(int begin_pos, int end_pos)
 284 {
 285     if (end_pos <= begin_pos) return;
 286
 287     wxHtmlTextPieces& pieces = *m_TextPieces;
 288     size_t piecesCnt = pieces.GetCount();
 289
 290     while (begin_pos < end_pos)
 291     {
 292         while (m_CurTag && m_CurTag->GetBeginPos() < begin_pos)
 293             m_CurTag = m_CurTag->GetNextTag();
 294         while (m_CurTextPiece < piecesCnt &&
 295                pieces[m_CurTextPiece].m_pos < begin_pos)
 296             m_CurTextPiece++;
 297
 298         if (m_CurTextPiece < piecesCnt &&
 299             (!m_CurTag ||
 300              pieces[m_CurTextPiece].m_pos < m_CurTag->GetBeginPos()))
 301         {
 302             // Add text:
 303             AddText(GetEntitiesParser()->Parse(
 304                        m_Source.Mid(pieces[m_CurTextPiece].m_pos,
 305                                     pieces[m_CurTextPiece].m_lng)));
 306             begin_pos = pieces[m_CurTextPiece].m_pos +
 307                         pieces[m_CurTextPiece].m_lng;
 308             m_CurTextPiece++;
 309         }
 310         else if (m_CurTag)
 311         {
 312             // Add tag:
 313             if (m_CurTag)
 314             {
 315                 if (m_CurTag->HasEnding())
 316                     begin_pos = m_CurTag->GetEndPos2();
 317                 else
 318                     begin_pos = m_CurTag->GetBeginPos();
 319             }
 320             wxHtmlTag *t = m_CurTag;
 321             m_CurTag = m_CurTag->GetNextTag();
 322             AddTag(*t);
 323             if (m_stopParsing)
 324                 return;
 325         }
 326         else break;
 327     }
 328 }
 329
 330 void wxHtmlParser::AddTag(const wxHtmlTag& tag)
 331 {
 332     wxHtmlTagHandler *h;
 333     bool inner = false;
 334
 335     h = (wxHtmlTagHandler*) m_HandlersHash.Get(tag.GetName());
 336     if (h)
 337     {
 338         inner = h->HandleTag(tag);
 339         if (m_stopParsing)
 340             return;
 341     }
 342     if (!inner)
 343     {
 344         if (tag.HasEnding())
 345             DoParsing(tag.GetBeginPos(), tag.GetEndPos1());
 346     }
 347 }
 348
 349 void wxHtmlParser::AddTagHandler(wxHtmlTagHandler *handler)
 350 {
 351     wxString s(handler->GetSupportedTags());
 352     wxStringTokenizer tokenizer(s, wxT(", "));
 353
 354     while (tokenizer.HasMoreTokens())
 355         m_HandlersHash.Put(tokenizer.GetNextToken(), handler);
 356
 357     if (m_HandlersList.IndexOf(handler) == wxNOT_FOUND)
 358         m_HandlersList.Append(handler);
 359
 360     handler->SetParser(this);
 361 }
 362
 363 void wxHtmlParser::PushTagHandler(wxHtmlTagHandler *handler, wxString tags)
 364 {
 365     wxStringTokenizer tokenizer(tags, wxT(", "));
 366     wxString key;
 367
 368     if (m_HandlersStack == NULL)
 369     {
 370         m_HandlersStack = new wxList;
 371     }
 372
 373     m_HandlersStack->Insert((wxObject*)new wxHashTable(m_HandlersHash));
 374
 375     while (tokenizer.HasMoreTokens())
 376     {
 377         key = tokenizer.GetNextToken();
 378         m_HandlersHash.Delete(key);
 379         m_HandlersHash.Put(key, handler);
 380     }
 381 }
 382
 383 void wxHtmlParser::PopTagHandler()
 384 {
 385     wxList::compatibility_iterator first;
 386
 387     if ( !m_HandlersStack ||
 388 #if wxUSE_STL
 389          !(first = m_HandlersStack->GetFirst())
 390 #else // !wxUSE_STL
 391          ((first = m_HandlersStack->GetFirst()) == NULL)
 392 #endif // wxUSE_STL/!wxUSE_STL
 393         )
 394     {
 395         wxLogWarning(_("Warning: attempt to remove HTML tag handler from empty stack."));
 396         return;
 397     }
 398     m_HandlersHash = *((wxHashTable*) first->GetData());
 399     delete (wxHashTable*) first->GetData();
 400     m_HandlersStack->Erase(first);
 401 }
 402
 403 void wxHtmlParser::SetSourceAndSaveState(const wxString& src)
 404 {
 405     wxHtmlParserState *s = new wxHtmlParserState;
 406
 407     s->m_curTag = m_CurTag;
 408     s->m_tags = m_Tags;
 409     s->m_textPieces = m_TextPieces;
 410     s->m_curTextPiece = m_CurTextPiece;
 411     s->m_source = m_Source;
 412
 413     s->m_nextState = m_SavedStates;
 414     m_SavedStates = s;
 415
 416     m_CurTag = NULL;
 417     m_Tags = NULL;
 418     m_TextPieces = NULL;
 419     m_CurTextPiece = 0;
 420     m_Source = wxEmptyString;
 421
 422     SetSource(src);
 423 }
 424
 425 bool wxHtmlParser::RestoreState()
 426 {
 427     if (!m_SavedStates) return false;
 428
 429     DestroyDOMTree();
 430
 431     wxHtmlParserState *s = m_SavedStates;
 432     m_SavedStates = s->m_nextState;
 433
 434     m_CurTag = s->m_curTag;
 435     m_Tags = s->m_tags;
 436     m_TextPieces = s->m_textPieces;
 437     m_CurTextPiece = s->m_curTextPiece;
 438     m_Source = s->m_source;
 439
 440     delete s;
 441     return true;
 442 }
 443
 444 //-----------------------------------------------------------------------------
 445 // wxHtmlTagHandler
 446 //-----------------------------------------------------------------------------
 447
 448 IMPLEMENT_ABSTRACT_CLASS(wxHtmlTagHandler,wxObject)
 449
 450
 451 //-----------------------------------------------------------------------------
 452 // wxHtmlEntitiesParser
 453 //-----------------------------------------------------------------------------
 454
 455 IMPLEMENT_DYNAMIC_CLASS(wxHtmlEntitiesParser,wxObject)
 456
 457 wxHtmlEntitiesParser::wxHtmlEntitiesParser()
 458 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 459     : m_conv(NULL), m_encoding(wxFONTENCODING_SYSTEM)
 460 #endif
 461 {
 462 }
 463
 464 wxHtmlEntitiesParser::~wxHtmlEntitiesParser()
 465 {
 466 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 467     delete m_conv;
 468 #endif
 469 }
 470
 471 void wxHtmlEntitiesParser::SetEncoding(wxFontEncoding encoding)
 472 {
 473 #if wxUSE_WCHAR_T && !wxUSE_UNICODE
 474     if (encoding == m_encoding)
 475         return;
 476
 477     delete m_conv;
 478
 479     m_encoding = encoding;
 480     if (m_encoding == wxFONTENCODING_SYSTEM)
 481         m_conv = NULL;
 482     else
 483         m_conv = new wxCSConv(wxFontMapper::GetEncodingName(m_encoding));
 484 #else
 485     (void) encoding;
 486 #endif
 487 }
 488
 489 wxString wxHtmlEntitiesParser::Parse(const wxString& input)
 490 {
 491     const wxChar *c, *last;
 492     const wxChar *in_str = input.c_str();
 493     wxString output;
 494
 495     output.reserve(input.length());
 496
 497     for (c = in_str, last = in_str; *c != wxT('\0'); c++)
 498     {
 499         if (*c == wxT('&'))
 500         {
 501             if (c - last > 0)
 502                 output.append(last, c - last);
 503             if (++c == wxT('\0')) break;
 504
 505             wxString entity;
 506             const wxChar *ent_s = c;
 507             wxChar entity_char;
 508
 509             for (; (*c >= wxT('a') && *c <= wxT('z')) ||
 510                    (*c >= wxT('A') && *c <= wxT('Z')) ||
 511                    (*c >= wxT('0') && *c <= wxT('9')) ||
 512                    *c == wxT('_') || *c == wxT('#'); c++) {}
 513             entity.append(ent_s, c - ent_s);
 514             if (*c != wxT(';')) c--;
 515             last = c+1;
 516             entity_char = GetEntityChar(entity);
 517             if (entity_char)
 518                 output << entity_char;
 519             else
 520             {
 521                 output.append(ent_s-1, c-ent_s+2);
 522                 wxLogTrace(wxTRACE_HTML_DEBUG,
 523                            wxT("Unrecognized HTML entity: '%s'"),
 524                            entity.c_str());
 525             }
 526         }
 527     }
 528     if (*last != wxT('\0'))
 529         output.append(last);
 530     return output;
 531 }
 532
 533 struct wxHtmlEntityInfo
 534 {
 535     const wxChar *name;
 536     unsigned code;
 537 };
 538
 539 extern "C" int LINKAGEMODE wxHtmlEntityCompare(const void *key, const void *item)
 540 {
 541     return wxStrcmp((wxChar*)key, ((wxHtmlEntityInfo*)item)->name);
 542 }
 543
 544 #if !wxUSE_UNICODE
 545 wxChar wxHtmlEntitiesParser::GetCharForCode(unsigned code)
 546 {
 547 #if wxUSE_WCHAR_T
 548     char buf[2];
 549     wchar_t wbuf[2];
 550     wbuf[0] = (wchar_t)code;
 551     wbuf[1] = 0;
 552     wxMBConv *conv = m_conv ? m_conv : &wxConvLocal;
 553     if (conv->WC2MB(buf, wbuf, 2) == (size_t)-1)
 554         return '?';
 555     return buf[0];
 556 #else
 557     return (code < 256) ? (wxChar)code : '?';
 558 #endif
 559 }
 560 #endif
 561
 562 wxChar wxHtmlEntitiesParser::GetEntityChar(const wxString& entity)
 563 {
 564     unsigned code = 0;
 565
 566     if (entity[0] == wxT('#'))
 567     {
 568         const wxChar *ent_s = entity.c_str();
 569         const wxChar *format;
 570
 571         if (ent_s[1] == wxT('x') || ent_s[1] == wxT('X'))
 572         {
 573             format = wxT("%x");
 574             ent_s++;
 575         }
 576         else
 577             format = wxT("%u");
 578         ent_s++;
 579
 580         if (wxSscanf(ent_s, format, &code) != 1)
 581             code = 0;
 582     }
 583     else
 584     {
 585         static wxHtmlEntityInfo substitutions[] = {
 586             { wxT("AElig"),198 },
 587             { wxT("Aacute"),193 },
 588             { wxT("Acirc"),194 },
 589             { wxT("Agrave"),192 },
 590             { wxT("Alpha"),913 },
 591             { wxT("Aring"),197 },
 592             { wxT("Atilde"),195 },
 593             { wxT("Auml"),196 },
 594             { wxT("Beta"),914 },
 595             { wxT("Ccedil"),199 },
 596             { wxT("Chi"),935 },
 597             { wxT("Dagger"),8225 },
 598             { wxT("Delta"),916 },
 599             { wxT("ETH"),208 },
 600             { wxT("Eacute"),201 },
 601             { wxT("Ecirc"),202 },
 602             { wxT("Egrave"),200 },
 603             { wxT("Epsilon"),917 },
 604             { wxT("Eta"),919 },
 605             { wxT("Euml"),203 },
 606             { wxT("Gamma"),915 },
 607             { wxT("Iacute"),205 },
 608             { wxT("Icirc"),206 },
 609             { wxT("Igrave"),204 },
 610             { wxT("Iota"),921 },
 611             { wxT("Iuml"),207 },
 612             { wxT("Kappa"),922 },
 613             { wxT("Lambda"),923 },
 614             { wxT("Mu"),924 },
 615             { wxT("Ntilde"),209 },
 616             { wxT("Nu"),925 },
 617             { wxT("OElig"),338 },
 618             { wxT("Oacute"),211 },
 619             { wxT("Ocirc"),212 },
 620             { wxT("Ograve"),210 },
 621             { wxT("Omega"),937 },
 622             { wxT("Omicron"),927 },
 623             { wxT("Oslash"),216 },
 624             { wxT("Otilde"),213 },
 625             { wxT("Ouml"),214 },
 626             { wxT("Phi"),934 },
 627             { wxT("Pi"),928 },
 628             { wxT("Prime"),8243 },
 629             { wxT("Psi"),936 },
 630             { wxT("Rho"),929 },
 631             { wxT("Scaron"),352 },
 632             { wxT("Sigma"),931 },
 633             { wxT("THORN"),222 },
 634             { wxT("Tau"),932 },
 635             { wxT("Theta"),920 },
 636             { wxT("Uacute"),218 },
 637             { wxT("Ucirc"),219 },
 638             { wxT("Ugrave"),217 },
 639             { wxT("Upsilon"),933 },
 640             { wxT("Uuml"),220 },
 641             { wxT("Xi"),926 },
 642             { wxT("Yacute"),221 },
 643             { wxT("Yuml"),376 },
 644             { wxT("Zeta"),918 },
 645             { wxT("aacute"),225 },
 646             { wxT("acirc"),226 },
 647             { wxT("acute"),180 },
 648             { wxT("aelig"),230 },
 649             { wxT("agrave"),224 },
 650             { wxT("alefsym"),8501 },
 651             { wxT("alpha"),945 },
 652             { wxT("amp"),38 },
 653             { wxT("and"),8743 },
 654             { wxT("ang"),8736 },
 655             { wxT("aring"),229 },
 656             { wxT("asymp"),8776 },
 657             { wxT("atilde"),227 },
 658             { wxT("auml"),228 },
 659             { wxT("bdquo"),8222 },
 660             { wxT("beta"),946 },
 661             { wxT("brvbar"),166 },
 662             { wxT("bull"),8226 },
 663             { wxT("cap"),8745 },
 664             { wxT("ccedil"),231 },
 665             { wxT("cedil"),184 },
 666             { wxT("cent"),162 },
 667             { wxT("chi"),967 },
 668             { wxT("circ"),710 },
 669             { wxT("clubs"),9827 },
 670             { wxT("cong"),8773 },
 671             { wxT("copy"),169 },
 672             { wxT("crarr"),8629 },
 673             { wxT("cup"),8746 },
 674             { wxT("curren"),164 },
 675             { wxT("dArr"),8659 },
 676             { wxT("dagger"),8224 },
 677             { wxT("darr"),8595 },
 678             { wxT("deg"),176 },
 679             { wxT("delta"),948 },
 680             { wxT("diams"),9830 },
 681             { wxT("divide"),247 },
 682             { wxT("eacute"),233 },
 683             { wxT("ecirc"),234 },
 684             { wxT("egrave"),232 },
 685             { wxT("empty"),8709 },
 686             { wxT("emsp"),8195 },
 687             { wxT("ensp"),8194 },
 688             { wxT("epsilon"),949 },
 689             { wxT("equiv"),8801 },
 690             { wxT("eta"),951 },
 691             { wxT("eth"),240 },
 692             { wxT("euml"),235 },
 693             { wxT("euro"),8364 },
 694             { wxT("exist"),8707 },
 695             { wxT("fnof"),402 },
 696             { wxT("forall"),8704 },
 697             { wxT("frac12"),189 },
 698             { wxT("frac14"),188 },
 699             { wxT("frac34"),190 },
 700             { wxT("frasl"),8260 },
 701             { wxT("gamma"),947 },
 702             { wxT("ge"),8805 },
 703             { wxT("gt"),62 },
 704             { wxT("hArr"),8660 },
 705             { wxT("harr"),8596 },
 706             { wxT("hearts"),9829 },
 707             { wxT("hellip"),8230 },
 708             { wxT("iacute"),237 },
 709             { wxT("icirc"),238 },
 710             { wxT("iexcl"),161 },
 711             { wxT("igrave"),236 },
 712             { wxT("image"),8465 },
 713             { wxT("infin"),8734 },
 714             { wxT("int"),8747 },
 715             { wxT("iota"),953 },
 716             { wxT("iquest"),191 },
 717             { wxT("isin"),8712 },
 718             { wxT("iuml"),239 },
 719             { wxT("kappa"),954 },
 720             { wxT("lArr"),8656 },
 721             { wxT("lambda"),955 },
 722             { wxT("lang"),9001 },
 723             { wxT("laquo"),171 },
 724             { wxT("larr"),8592 },
 725             { wxT("lceil"),8968 },
 726             { wxT("ldquo"),8220 },
 727             { wxT("le"),8804 },
 728             { wxT("lfloor"),8970 },
 729             { wxT("lowast"),8727 },
 730             { wxT("loz"),9674 },
 731             { wxT("lrm"),8206 },
 732             { wxT("lsaquo"),8249 },
 733             { wxT("lsquo"),8216 },
 734             { wxT("lt"),60 },
 735             { wxT("macr"),175 },
 736             { wxT("mdash"),8212 },
 737             { wxT("micro"),181 },
 738             { wxT("middot"),183 },
 739             { wxT("minus"),8722 },
 740             { wxT("mu"),956 },
 741             { wxT("nabla"),8711 },
 742             { wxT("nbsp"),160 },
 743             { wxT("ndash"),8211 },
 744             { wxT("ne"),8800 },
 745             { wxT("ni"),8715 },
 746             { wxT("not"),172 },
 747             { wxT("notin"),8713 },
 748             { wxT("nsub"),8836 },
 749             { wxT("ntilde"),241 },
 750             { wxT("nu"),957 },
 751             { wxT("oacute"),243 },
 752             { wxT("ocirc"),244 },
 753             { wxT("oelig"),339 },
 754             { wxT("ograve"),242 },
 755             { wxT("oline"),8254 },
 756             { wxT("omega"),969 },
 757             { wxT("omicron"),959 },
 758             { wxT("oplus"),8853 },
 759             { wxT("or"),8744 },
 760             { wxT("ordf"),170 },
 761             { wxT("ordm"),186 },
 762             { wxT("oslash"),248 },
 763             { wxT("otilde"),245 },
 764             { wxT("otimes"),8855 },
 765             { wxT("ouml"),246 },
 766             { wxT("para"),182 },
 767             { wxT("part"),8706 },
 768             { wxT("permil"),8240 },
 769             { wxT("perp"),8869 },
 770             { wxT("phi"),966 },
 771             { wxT("pi"),960 },
 772             { wxT("piv"),982 },
 773             { wxT("plusmn"),177 },
 774             { wxT("pound"),163 },
 775             { wxT("prime"),8242 },
 776             { wxT("prod"),8719 },
 777             { wxT("prop"),8733 },
 778             { wxT("psi"),968 },
 779             { wxT("quot"),34 },
 780             { wxT("rArr"),8658 },
 781             { wxT("radic"),8730 },
 782             { wxT("rang"),9002 },
 783             { wxT("raquo"),187 },
 784             { wxT("rarr"),8594 },
 785             { wxT("rceil"),8969 },
 786             { wxT("rdquo"),8221 },
 787             { wxT("real"),8476 },
 788             { wxT("reg"),174 },
 789             { wxT("rfloor"),8971 },
 790             { wxT("rho"),961 },
 791             { wxT("rlm"),8207 },
 792             { wxT("rsaquo"),8250 },
 793             { wxT("rsquo"),8217 },
 794             { wxT("sbquo"),8218 },
 795             { wxT("scaron"),353 },
 796             { wxT("sdot"),8901 },
 797             { wxT("sect"),167 },
 798             { wxT("shy"),173 },
 799             { wxT("sigma"),963 },
 800             { wxT("sigmaf"),962 },
 801             { wxT("sim"),8764 },
 802             { wxT("spades"),9824 },
 803             { wxT("sub"),8834 },
 804             { wxT("sube"),8838 },
 805             { wxT("sum"),8721 },
 806             { wxT("sup"),8835 },
 807             { wxT("sup1"),185 },
 808             { wxT("sup2"),178 },
 809             { wxT("sup3"),179 },
 810             { wxT("supe"),8839 },
 811             { wxT("szlig"),223 },
 812             { wxT("tau"),964 },
 813             { wxT("there4"),8756 },
 814             { wxT("theta"),952 },
 815             { wxT("thetasym"),977 },
 816             { wxT("thinsp"),8201 },
 817             { wxT("thorn"),254 },
 818             { wxT("tilde"),732 },
 819             { wxT("times"),215 },
 820             { wxT("trade"),8482 },
 821             { wxT("uArr"),8657 },
 822             { wxT("uacute"),250 },
 823             { wxT("uarr"),8593 },
 824             { wxT("ucirc"),251 },
 825             { wxT("ugrave"),249 },
 826             { wxT("uml"),168 },
 827             { wxT("upsih"),978 },
 828             { wxT("upsilon"),965 },
 829             { wxT("uuml"),252 },
 830             { wxT("weierp"),8472 },
 831             { wxT("xi"),958 },
 832             { wxT("yacute"),253 },
 833             { wxT("yen"),165 },
 834             { wxT("yuml"),255 },
 835             { wxT("zeta"),950 },
 836             { wxT("zwj"),8205 },
 837             { wxT("zwnj"),8204 },
 838             {NULL, 0}};
 839         static size_t substitutions_cnt = 0;
 840
 841         if (substitutions_cnt == 0)
 842             while (substitutions[substitutions_cnt].code != 0)
 843                 substitutions_cnt++;
 844
 845         wxHtmlEntityInfo *info;
 846         info = (wxHtmlEntityInfo*) bsearch(entity.c_str(), substitutions,
 847                                            substitutions_cnt,
 848                                            sizeof(wxHtmlEntityInfo),
 849                                            wxHtmlEntityCompare);
 850         if (info)
 851             code = info->code;
 852     }
 853
 854     if (code == 0)
 855         return 0;
 856     else
 857         return GetCharForCode(code);
 858 }
 859
 860 wxFSFile *wxHtmlParser::OpenURL(wxHtmlURLType WXUNUSED(type),
 861                                 const wxString& url) const
 862 {
 863     return m_FS ? m_FS->OpenFile(url) : NULL;
 864
 865 }
 866
 867
 868 //-----------------------------------------------------------------------------
 869 // wxHtmlParser::ExtractCharsetInformation
 870 //-----------------------------------------------------------------------------
 871
 872 class wxMetaTagParser : public wxHtmlParser
 873 {
 874 public:
 875     wxMetaTagParser() { }
 876
 877     wxObject* GetProduct() { return NULL; }
 878
 879 protected:
 880     virtual void AddText(const wxChar* WXUNUSED(txt)) {}
 881
 882     DECLARE_NO_COPY_CLASS(wxMetaTagParser)
 883 };
 884
 885 class wxMetaTagHandler : public wxHtmlTagHandler
 886 {
 887 public:
 888     wxMetaTagHandler(wxString *retval) : wxHtmlTagHandler(), m_retval(retval) {}
 889     wxString GetSupportedTags() { return wxT("META,BODY"); }
 890     bool HandleTag(const wxHtmlTag& tag);
 891
 892 private:
 893     wxString *m_retval;
 894
 895     DECLARE_NO_COPY_CLASS(wxMetaTagHandler)
 896 };
 897
 898 bool wxMetaTagHandler::HandleTag(const wxHtmlTag& tag)
 899 {
 900     if (tag.GetName() == _T("BODY"))
 901     {
 902         m_Parser->StopParsing();
 903         return false;
 904     }
 905
 906     if (tag.HasParam(_T("HTTP-EQUIV")) &&
 907         tag.GetParam(_T("HTTP-EQUIV")).IsSameAs(_T("Content-Type"), false) &&
 908         tag.HasParam(_T("CONTENT")))
 909     {
 910         wxString content = tag.GetParam(_T("CONTENT")).Lower();
 911         if (content.Left(19) == _T("text/html; charset="))
 912         {
 913             *m_retval = content.Mid(19);
 914             m_Parser->StopParsing();
 915         }
 916     }
 917     return false;
 918 }
 919
 920
 921 /*static*/
 922 wxString wxHtmlParser::ExtractCharsetInformation(const wxString& markup)
 923 {
 924     wxString charset;
 925     wxMetaTagParser *parser = new wxMetaTagParser();
 926     if(parser)
 927     {
 928         parser->AddTagHandler(new wxMetaTagHandler(&charset));
 929         parser->Parse(markup);
 930         delete parser;
 931     }
 932     return charset;
 933 }
 934
 935 #endif