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