src/html/helpdata.cpp

   1 /////////////////////////////////////////////////////////////////////////////
   2 // Name:        helpdata.cpp
   3 // Purpose:     wxHtmlHelpData
   4 // Notes:       Based on htmlhelp.cpp, implementing a monolithic
   5 //              HTML Help controller class,  by Vaclav Slavik
   6 // Author:      Harm van der Heijden and Vaclav Slavik
   7 // RCS-ID:      $Id$
   8 // Copyright:   (c) Harm van der Heijden and Vaclav Slavik
   9 // Licence:     wxWindows licence
  10 /////////////////////////////////////////////////////////////////////////////
  11
  12 #ifdef __GNUG__
  13 #pragma implementation
  14 #endif
  15
  16 // For compilers that support precompilation, includes "wx.h".
  17 #include "wx/wxprec.h"
  18
  19 #ifdef __BORLANDC__
  20 #pragma hdrstop
  21 #endif
  22
  23 #include "wx/defs.h"
  24
  25 #if wxUSE_HTML && wxUSE_STREAMS
  26
  27 #ifndef WXPRECOMP
  28 #include "wx/wx.h"
  29 #endif
  30
  31 #include "wx/html/helpdata.h"
  32 #include "wx/tokenzr.h"
  33 #include "wx/wfstream.h"
  34 #include "wx/busyinfo.h"
  35 #include "wx/encconv.h"
  36 #include "wx/fontmap.h"
  37 #include "wx/log.h"
  38 #include "wx/html/htmlpars.h"
  39 #include "wx/html/htmldefs.h"
  40
  41 #include "wx/arrimpl.cpp"
  42 WX_DEFINE_OBJARRAY(wxHtmlBookRecArray)
  43
  44 //-----------------------------------------------------------------------------
  45 // static helper functions
  46 //-----------------------------------------------------------------------------
  47
  48 // Reads one line, stores it into buf and returns pointer to new line or NULL.
  49 static char* ReadLine(char *line, char *buf)
  50 {
  51     char *writeptr = buf, *readptr = line;
  52
  53     while (*readptr != 0 && *readptr != '\r' && *readptr != '\n') *(writeptr++) = *(readptr++);
  54     *writeptr = 0;
  55     while (*readptr == '\r' || *readptr == '\n') readptr++;
  56     if (*readptr == 0) return NULL;
  57     else return readptr;
  58 }
  59
  60
  61
  62 static int LINKAGEMODE IndexCompareFunc(const void *a, const void *b)
  63 {
  64     return wxStrcmp(((wxHtmlContentsItem*)a)->m_Name, ((wxHtmlContentsItem*)b)->m_Name);
  65 }
  66
  67
  68 //-----------------------------------------------------------------------------
  69 // HP_Parser
  70 //-----------------------------------------------------------------------------
  71
  72 class HP_Parser : public wxHtmlParser
  73 {
  74     public:
  75         void AddText(const char* WXUNUSED(text)) { }
  76         wxObject* GetProduct() { return NULL; }
  77 };
  78
  79
  80 //-----------------------------------------------------------------------------
  81 // HP_TagHandler
  82 //-----------------------------------------------------------------------------
  83
  84 class HP_TagHandler : public wxHtmlTagHandler
  85 {
  86     private:
  87         wxString m_Name, m_Page;
  88         int m_Level;
  89         int m_ID;
  90         int m_Index;
  91         wxHtmlContentsItem *m_Items;
  92         int m_ItemsCnt;
  93         wxHtmlBookRecord *m_Book;
  94
  95     public:
  96         HP_TagHandler(wxHtmlBookRecord *b) : wxHtmlTagHandler()
  97                 { m_Book = b; m_Items = NULL; m_ItemsCnt = 0; m_Name = m_Page = wxEmptyString;
  98                       m_Level = 0; m_ID = -1; }
  99         wxString GetSupportedTags() { return wxT("UL,OBJECT,PARAM"); }
 100         bool HandleTag(const wxHtmlTag& tag);
 101         void WriteOut(wxHtmlContentsItem*& array, int& size);
 102         void ReadIn(wxHtmlContentsItem* array, int size);
 103 };
 104
 105
 106 bool HP_TagHandler::HandleTag(const wxHtmlTag& tag)
 107 {
 108     if (tag.GetName() == wxT("UL"))
 109     {
 110         m_Level++;
 111         ParseInner(tag);
 112         m_Level--;
 113         return TRUE;
 114     }
 115     else if (tag.GetName() == wxT("OBJECT"))
 116     {
 117         m_Name = m_Page = wxEmptyString;
 118         ParseInner(tag);
 119
 120 #if 0
 121          if (!m_Page.IsEmpty())
 122         /* Valid HHW's file may contain only two object tags:
 123
 124            <OBJECT type="text/site properties">
 125                <param name="ImageType" value="Folder">
 126            </OBJECT>
 127
 128            or
 129
 130                <OBJECT type="text/sitemap">
 131                        <param name="Name" value="main page">
 132                        <param name="Local" value="another.htm">
 133                    </OBJECT>
 134
 135            We're interested in the latter. !m_Page.IsEmpty() is valid
 136            condition because text/site properties does not contain Local param
 137         */
 138 #endif
 139         if (tag.GetParam(wxT("TYPE")) == wxT("text/sitemap"))
 140         {
 141             if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
 142                 m_Items = (wxHtmlContentsItem*) realloc(m_Items,
 143                                        (m_ItemsCnt + wxHTML_REALLOC_STEP) *
 144                                        sizeof(wxHtmlContentsItem));
 145
 146             m_Items[m_ItemsCnt].m_Level = m_Level;
 147             m_Items[m_ItemsCnt].m_ID = m_ID;
 148             m_Items[m_ItemsCnt].m_Page = new wxChar[m_Page.Length() + 1];
 149             wxStrcpy(m_Items[m_ItemsCnt].m_Page, m_Page.c_str());
 150             m_Items[m_ItemsCnt].m_Name = new wxChar [m_Name.Length() + 1];
 151             wxStrcpy(m_Items[m_ItemsCnt].m_Name, m_Name.c_str());
 152             m_Items[m_ItemsCnt].m_Book = m_Book;
 153             m_ItemsCnt++;
 154         }
 155
 156         return TRUE;
 157     }
 158     else
 159     { // "PARAM"
 160         if (m_Name == wxEmptyString && tag.GetParam(wxT("NAME")) == wxT("Name"))
 161         {
 162             m_Name = tag.GetParam(wxT("VALUE"));
 163             if (m_Name.Find(wxT('&')) != -1)
 164             {
 165 #define ESCSEQ(escape, subst)  \
 166                   { _T("&") _T(escape) _T(";"), _T("&") _T(escape) _T(" "), _T("&") _T(escape), _T(subst) }
 167         static wxChar* substitutions[][4] =
 168                         {
 169                 ESCSEQ("quot", "\""),
 170                 ESCSEQ("#34", "\""),
 171                 ESCSEQ("lt", "<"),
 172                 ESCSEQ("#60", "<"),
 173                 ESCSEQ("gt", ">"),
 174                 ESCSEQ("#62", ">"),
 175
 176                 ESCSEQ("#94", "^"), /* ^ */
 177
 178                 ESCSEQ("nbsp", " "),
 179                 ESCSEQ("#32", " "),
 180                 ESCSEQ("iexcl", "!"),
 181                 ESCSEQ("#33", "!"),
 182                 ESCSEQ("cent", "¢"/* ¢ */),
 183                 ESCSEQ("#162", "¢"/* ¢ */),
 184
 185                 ESCSEQ("trade", "(TM)"),
 186                 ESCSEQ("#153", "(TM)"),
 187
 188                 ESCSEQ("yen", "¥"),
 189                 ESCSEQ("#165", "¥"),
 190                 ESCSEQ("brkbar", "¦"),
 191                 ESCSEQ("#166", "¦"),
 192                 ESCSEQ("sect", "§"),
 193                 ESCSEQ("#167", "§"),
 194                 ESCSEQ("uml", "¨"),
 195                 ESCSEQ("#168", "¨"),
 196
 197                 ESCSEQ("copy", "©"), /* © */
 198                 ESCSEQ("#169", "©"),
 199                 ESCSEQ("ordf", "ª"),
 200                 ESCSEQ("#170", "ª"),
 201                 ESCSEQ("laquo", "«"), /* « */
 202                 ESCSEQ("#171", "«"),
 203                 ESCSEQ("not", "¬"),
 204                 ESCSEQ("#172", "¬"),
 205
 206                 ESCSEQ("reg", "®"), /* ® */
 207                 ESCSEQ("#174", "®"),
 208
 209                 ESCSEQ("deg", "°"), /* ° */
 210                 ESCSEQ("#176", "°"),
 211                 ESCSEQ("plusm", "±"), /* ± */
 212                 ESCSEQ("#177", "±"),
 213
 214                 ESCSEQ("acute", "´"),
 215                 ESCSEQ("#180", "´"),
 216                 ESCSEQ("macron", "¯"),
 217                 ESCSEQ("#175", "¯"),
 218                 ESCSEQ("micro", "µ"), /* µ */
 219                 ESCSEQ("#181", "µ"),
 220                 ESCSEQ("para", "¶"), /* ¶ */
 221                 ESCSEQ("#182", "¶"),
 222
 223                 ESCSEQ("ordm", "º"), /* º */
 224                 ESCSEQ("#186", "º"),
 225                 ESCSEQ("raquo", "»"), /* » */
 226                 ESCSEQ("#187", "»"),
 227
 228                 ESCSEQ("iquest", "¿"), /* ¿ */
 229                 ESCSEQ("#191", "¿"),
 230                 ESCSEQ("Agrave", "\300"/* À */),
 231                 ESCSEQ("#193", "\300"/* À */),
 232
 233                 ESCSEQ("Acirc", "\302"/* Â */),
 234                 ESCSEQ("Atilde", "\303"/* Ã */),
 235                 ESCSEQ("Auml", "\304"/* Ä */),
 236                 ESCSEQ("Aring", " "),
 237                 ESCSEQ("AElig", " "),
 238                 ESCSEQ("Ccedil", "\347"/* ç */),
 239                 ESCSEQ("Egrave", "\310"/* È */),
 240                 ESCSEQ("Eacute", "\311"/* É */),
 241                 ESCSEQ("Ecirc", "\312"/* Ê */),
 242                 ESCSEQ("Euml", "\313"/* Ë */),
 243                 ESCSEQ("Igrave", "\314"/* Ì */),
 244
 245                 ESCSEQ("Icirc", "\316"/* Î */),
 246                 ESCSEQ("Iuml", "\317"/* Ï */),
 247
 248                 ESCSEQ("Ntilde", "\321"/* Ñ */),
 249                 ESCSEQ("Ograve", "\322"/* Ò */),
 250
 251                 ESCSEQ("Ocirc", "\324"/* Ô */),
 252                 ESCSEQ("Otilde", "\325"/* Õ */),
 253                 ESCSEQ("Ouml", "\326"/* Ö */),
 254
 255                 ESCSEQ("Oslash", " "),
 256                 ESCSEQ("Ugrave", "\331"/* Ù */),
 257
 258                 ESCSEQ("Ucirc", " "),
 259                 ESCSEQ("Uuml", "\334"/* Ü */),
 260
 261                 ESCSEQ("szlig", "\247"/* § */),
 262                 ESCSEQ("agrave;","à"),
 263                 ESCSEQ("aacute", "\341"/* á */),
 264                 ESCSEQ("acirc", "\342"/* â */),
 265                 ESCSEQ("atilde", "\343"/* ã */),
 266                 ESCSEQ("auml", "\344"/* ä */),
 267                 ESCSEQ("aring", "a"),
 268                 ESCSEQ("aelig", "ae"),
 269                 ESCSEQ("ccedil", "\347"/* ç */),
 270                 ESCSEQ("egrave", "\350"/* è */),
 271                 ESCSEQ("eacute", "\351"/* é */),
 272                 ESCSEQ("ecirc", "\352"/* ê */),
 273                 ESCSEQ("euml", "\353"/* ë */),
 274                 ESCSEQ("igrave", "\354"/* ì */),
 275                 ESCSEQ("iacute", "\355"/* í */),
 276                 ESCSEQ("icirc", " "),
 277                 ESCSEQ("iuml", "\357"/* ï */),
 278                 ESCSEQ("eth", " "),
 279                 ESCSEQ("ntilde", "\361"/* ñ */),
 280                 ESCSEQ("ograve", "\362"/* ò */),
 281                 ESCSEQ("oacute", "\363"/* ó */),
 282                 ESCSEQ("ocirc", "\364"/* ô */),
 283                 ESCSEQ("otilde", "\365"/* õ */),
 284                 ESCSEQ("ouml", "\366"/* ö */),
 285                 ESCSEQ("divide", " "),
 286                 ESCSEQ("oslash", " "),
 287                 ESCSEQ("ugrave", "\371"/* ù */),
 288                 ESCSEQ("uacute", "\372"/* ú */),
 289                 ESCSEQ("ucirc", "\373"/* û */),
 290                 ESCSEQ("uuml", "\374"/* ü */),
 291
 292                 ESCSEQ("yuml", ""),
 293
 294                 /* this one should ALWAYS stay the last one!!! */
 295                 ESCSEQ("amp", "&"),
 296                 ESCSEQ("#38", "&"),
 297
 298                 { NULL, NULL, NULL }
 299                 };
 300
 301                 for (int i = 0; substitutions[i][0] != NULL; i++)
 302                 {
 303                     m_Name.Replace(substitutions[i][0], substitutions[i][3], TRUE);
 304                     m_Name.Replace(substitutions[i][1], substitutions[i][3], TRUE);
 305                     m_Name.Replace(substitutions[i][2], substitutions[i][3], TRUE);
 306                 }
 307             }
 308         }
 309         if (tag.GetParam(wxT("NAME")) == wxT("Local")) m_Page = tag.GetParam(wxT("VALUE"));
 310         if (tag.GetParam(wxT("NAME")) == wxT("ID")) tag.ScanParam(wxT("VALUE"), wxT("%i"), &m_ID);
 311         return FALSE;
 312     }
 313 }
 314
 315
 316
 317 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
 318 {
 319     array = m_Items;
 320     size = m_ItemsCnt;
 321     m_Items = NULL;
 322     m_ItemsCnt = 0;
 323 }
 324
 325 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
 326 {
 327     m_Items = array;
 328     m_ItemsCnt = size;
 329 }
 330
 331
 332
 333
 334 //-----------------------------------------------------------------------------
 335 // wxHtmlHelpData
 336 //-----------------------------------------------------------------------------
 337
 338 wxString wxHtmlBookRecord::GetFullPath(const wxString &page) const
 339 {
 340     if (wxIsAbsolutePath(page))
 341         return page;
 342     else
 343         return m_BasePath + page;
 344 }
 345
 346
 347
 348 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
 349
 350 wxHtmlHelpData::wxHtmlHelpData()
 351 {
 352     m_TempPath = wxEmptyString;
 353
 354     m_Contents = NULL;
 355     m_ContentsCnt = 0;
 356     m_Index = NULL;
 357     m_IndexCnt = 0;
 358 }
 359
 360 wxHtmlHelpData::~wxHtmlHelpData()
 361 {
 362     int i;
 363
 364     m_BookRecords.Empty();
 365     if (m_Contents)
 366     {
 367         for (i = 0; i < m_ContentsCnt; i++)
 368             {
 369             delete[] m_Contents[i].m_Page;
 370             delete[] m_Contents[i].m_Name;
 371         }
 372         free(m_Contents);
 373     }
 374     if (m_Index)
 375     {
 376         for (i = 0; i < m_IndexCnt; i++)
 377             {
 378             delete[] m_Index[i].m_Page;
 379             delete[] m_Index[i].m_Name;
 380         }
 381         free(m_Index);
 382     }
 383 }
 384
 385 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
 386 {
 387     wxFSFile *f;
 388     char *buf;
 389     int sz;
 390     wxString string;
 391
 392     HP_Parser parser;
 393     HP_TagHandler *handler = new HP_TagHandler(book);
 394     parser.AddTagHandler(handler);
 395
 396     f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
 397     if (f)
 398     {
 399         sz = f->GetStream()->GetSize();
 400         buf = new char[sz + 1];
 401         buf[sz] = 0;
 402         f->GetStream()->Read(buf, sz);
 403         delete f;
 404         handler->ReadIn(m_Contents, m_ContentsCnt);
 405         parser.Parse(buf);
 406         handler->WriteOut(m_Contents, m_ContentsCnt);
 407         delete[] buf;
 408     }
 409     else
 410         wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
 411
 412     f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
 413     if (f)
 414     {
 415         sz = f->GetStream()->GetSize();
 416         buf = new char[sz + 1];
 417         buf[sz] = 0;
 418         f->GetStream()->Read(buf, sz);
 419         delete f;
 420         handler->ReadIn(m_Index, m_IndexCnt);
 421         parser.Parse(buf);
 422         handler->WriteOut(m_Index, m_IndexCnt);
 423         delete[] buf;
 424     }
 425     else if (!indexfile.IsEmpty())
 426         wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
 427     return TRUE;
 428 }
 429
 430
 431
 432
 433 #if wxUSE_UNICODE
 434
 435 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
 436 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
 437
 438 #else
 439
 440 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
 441 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
 442
 443 #endif
 444
 445
 446 #define CURRENT_CACHED_BOOK_VERSION     1
 447
 448 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
 449 {
 450     int i, st;
 451     wxInt32 x;
 452     wxInt32 version;
 453
 454     /* load header - version info : */
 455
 456     f->Read(&x, sizeof(x));
 457     version = wxINT32_SWAP_ON_BE(x);
 458
 459     if (version != CURRENT_CACHED_BOOK_VERSION)
 460     {
 461         wxLogError(_("Incorrect version of HTML help book"));
 462         return FALSE;
 463         // NOTE: when adding new version, please ensure backward compatibility!
 464     }
 465
 466     /* load contents : */
 467
 468     f->Read(&x, sizeof(x));
 469     st = m_ContentsCnt;
 470     m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
 471     m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
 472                                                (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
 473                                                wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 474     for (i = st; i < m_ContentsCnt; i++)
 475     {
 476         f->Read(&x, sizeof(x));
 477         m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
 478         f->Read(&x, sizeof(x));
 479         m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
 480         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 481         m_Contents[i].m_Name = new wxChar[x];
 482         READ_STRING(f, m_Contents[i].m_Name, x);
 483         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 484         m_Contents[i].m_Page = new wxChar[x];
 485         READ_STRING(f, m_Contents[i].m_Page, x);
 486         m_Contents[i].m_Book = book;
 487     }
 488
 489     /* load index : */
 490
 491     f->Read(&x, sizeof(x));
 492     st = m_IndexCnt;
 493     m_IndexCnt += wxINT32_SWAP_ON_BE(x);
 494     m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
 495                                                      wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 496     for (i = st; i < m_IndexCnt; i++)
 497     {
 498         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 499         m_Index[i].m_Name = new wxChar[x];
 500         READ_STRING(f, m_Index[i].m_Name, x);
 501         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 502         m_Index[i].m_Page = new wxChar[x];
 503         READ_STRING(f, m_Index[i].m_Page, x);
 504         m_Index[i].m_Book = book;
 505     }
 506     return TRUE;
 507 }
 508
 509
 510 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
 511 {
 512     int i;
 513     wxInt32 x;
 514
 515     /* save header - version info : */
 516
 517     x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
 518     f->Write(&x, sizeof(x));
 519
 520     /* save contents : */
 521
 522     x = 0;
 523     for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
 524     x = wxINT32_SWAP_ON_BE(x);
 525     f->Write(&x, sizeof(x));
 526     for (i = 0; i < m_ContentsCnt; i++)
 527     {
 528         if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
 529         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
 530         f->Write(&x, sizeof(x));
 531         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
 532         f->Write(&x, sizeof(x));
 533         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
 534         f->Write(&x, sizeof(x));
 535         WRITE_STRING(f, m_Contents[i].m_Name, x);
 536         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
 537         f->Write(&x, sizeof(x));
 538         WRITE_STRING(f, m_Contents[i].m_Page, x);
 539     }
 540
 541     /* save index : */
 542
 543     x = 0;
 544     for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
 545     x = wxINT32_SWAP_ON_BE(x);
 546     f->Write(&x, sizeof(x));
 547     for (i = 0; i < m_IndexCnt; i++)
 548     {
 549         if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
 550         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
 551         f->Write(&x, sizeof(x));
 552         WRITE_STRING(f, m_Index[i].m_Name, x);
 553         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
 554         f->Write(&x, sizeof(x));
 555         WRITE_STRING(f, m_Index[i].m_Page, x);
 556     }
 557     return TRUE;
 558 }
 559
 560
 561 void wxHtmlHelpData::SetTempDir(const wxString& path)
 562 {
 563     if (path == wxEmptyString) m_TempPath = path;
 564     else
 565     {
 566         if (wxIsAbsolutePath(path)) m_TempPath = path;
 567         else m_TempPath = wxGetCwd() + _T("/") + path;
 568
 569         if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
 570             m_TempPath << _T('/');
 571     }
 572 }
 573
 574
 575
 576 static wxString SafeFileName(const wxString& s)
 577 {
 578     wxString res(s);
 579     res.Replace(wxT("#"), wxT("_"));
 580     res.Replace(wxT(":"), wxT("_"));
 581     res.Replace(wxT("\\"), wxT("_"));
 582     res.Replace(wxT("/"), wxT("_"));
 583     return res;
 584 }
 585
 586 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
 587                                   wxFontEncoding encoding,
 588                                   const wxString& title, const wxString& contfile,
 589                                   const wxString& indexfile, const wxString& deftopic,
 590                                   const wxString& path)
 591 {
 592     wxFileSystem fsys;
 593     wxFSFile *fi;
 594     wxHtmlBookRecord *bookr;
 595
 596     int IndexOld = m_IndexCnt,
 597         ContentsOld = m_ContentsCnt;
 598
 599     if (! path.IsEmpty())
 600         fsys.ChangePathTo(path, TRUE);
 601
 602     bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
 603
 604     if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
 605         m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 606     m_Contents[m_ContentsCnt].m_Level = 0;
 607     m_Contents[m_ContentsCnt].m_ID = 0;
 608     m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
 609     wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
 610     m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
 611     wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
 612     m_Contents[m_ContentsCnt].m_Book = bookr;
 613
 614     // store the contents index for later
 615     int cont_start = m_ContentsCnt++;
 616
 617     // Try to find cached binary versions:
 618     // 1. save file as book, but with .hhp.cached extension
 619     // 2. same as 1. but in temp path
 620     // 3. otherwise or if cache load failed, load it from MS.
 621
 622     fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
 623
 624     if (fi == NULL ||
 625           fi->GetModificationTime() < bookfile.GetModificationTime() ||
 626           !LoadCachedBook(bookr, fi->GetStream()))
 627     {
 628         if (fi != NULL) delete fi;
 629         fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
 630         if (m_TempPath == wxEmptyString || fi == NULL ||
 631             fi->GetModificationTime() < bookfile.GetModificationTime() ||
 632             !LoadCachedBook(bookr, fi->GetStream()))
 633         {
 634             LoadMSProject(bookr, fsys, indexfile, contfile);
 635             if (m_TempPath != wxEmptyString)
 636             {
 637                 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
 638                                                   SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
 639                 SaveCachedBook(bookr, outs);
 640                 delete outs;
 641             }
 642         }
 643     }
 644
 645     if (fi != NULL) delete fi;
 646
 647     // Now store the contents range
 648     bookr->SetContentsRange(cont_start, m_ContentsCnt);
 649
 650     // Convert encoding, if neccessary:
 651     if (encoding != wxFONTENCODING_SYSTEM)
 652     {
 653         wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
 654         if (a.GetCount() != 0 && a[0] != encoding)
 655         {
 656             int i;
 657             wxEncodingConverter conv;
 658             conv.Init(encoding, a[0]);
 659
 660             for (i = IndexOld; i < m_IndexCnt; i++)
 661                 conv.Convert(m_Index[i].m_Name);
 662             for (i = ContentsOld; i < m_ContentsCnt; i++)
 663                 conv.Convert(m_Contents[i].m_Name);
 664         }
 665     }
 666
 667     m_BookRecords.Add(bookr);
 668     if (m_IndexCnt > 0)
 669         qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), IndexCompareFunc);
 670
 671     return TRUE;
 672 }
 673
 674
 675 bool wxHtmlHelpData::AddBook(const wxString& book)
 676 {
 677     if (book.Right(4).Lower() == wxT(".zip") ||
 678         book.Right(4).Lower() == wxT(".htb") /*html book*/)
 679
 680     {
 681         wxFileSystem fsys;
 682         wxString s;
 683         bool rt = FALSE;
 684
 685         s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
 686         while (!s.IsEmpty())
 687         {
 688             if (AddBook(s)) rt = TRUE;
 689             s = fsys.FindNext();
 690         }
 691
 692         return rt;
 693     }
 694
 695
 696     else
 697     {
 698         wxFSFile *fi;
 699         wxFileSystem fsys;
 700         wxInputStream *s;
 701         wxString bookFull;
 702
 703         int sz;
 704         char *buff, *lineptr;
 705         char linebuf[300];
 706
 707         wxString title = _("noname"),
 708                  safetitle,
 709                  start = wxEmptyString,
 710                  contents = wxEmptyString,
 711                  index = wxEmptyString,
 712                  charset = wxEmptyString;
 713
 714         if (wxIsAbsolutePath(book)) bookFull = book;
 715         else bookFull = wxGetCwd() + "/" + book;
 716
 717         fi = fsys.OpenFile(bookFull);
 718         if (fi == NULL)
 719         {
 720             wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
 721             return FALSE;
 722         }
 723         fsys.ChangePathTo(bookFull);
 724         s = fi->GetStream();
 725         sz = s->GetSize();
 726         buff = new char[sz + 1];
 727         buff[sz] = 0;
 728         s->Read(buff, sz);
 729         lineptr = buff;
 730
 731         do {
 732             lineptr = ReadLine(lineptr, linebuf);
 733
 734             if (strstr(linebuf, "Title=") == linebuf)
 735                 title = linebuf + strlen("Title=");
 736             if (strstr(linebuf, "Default topic=") == linebuf)
 737                 start = linebuf + strlen("Default topic=");
 738             if (strstr(linebuf, "Index file=") == linebuf)
 739                 index = linebuf + strlen("Index file=");
 740             if (strstr(linebuf, "Contents file=") == linebuf)
 741                 contents = linebuf + strlen("Contents file=");
 742             if (strstr(linebuf, "Charset=") == linebuf)
 743                 charset = linebuf + strlen("Charset=");
 744         } while (lineptr != NULL);
 745         delete[] buff;
 746
 747         wxFontEncoding enc;
 748         if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
 749         else enc = wxTheFontMapper->CharsetToEncoding(charset);
 750         bool rtval = AddBookParam(*fi, enc,
 751                                   title, contents, index, start, fsys.GetPath());
 752         delete fi;
 753         return rtval;
 754     }
 755 }
 756
 757 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
 758 {
 759     int cnt;
 760     int i;
 761     wxFileSystem fsys;
 762     wxFSFile *f;
 763     wxString url(wxEmptyString);
 764
 765     /* 1. try to open given file: */
 766
 767     cnt = m_BookRecords.GetCount();
 768     for (i = 0; i < cnt; i++)
 769     {
 770         f = fsys.OpenFile(m_BookRecords[i].GetFullPath(x));
 771         if (f)
 772             {
 773             url = m_BookRecords[i].GetFullPath(x);
 774             delete f;
 775             return url;
 776         }
 777     }
 778
 779
 780     /* 2. try to find a book: */
 781
 782     for (i = 0; i < cnt; i++)
 783     {
 784         if (m_BookRecords[i].GetTitle() == x)
 785             {
 786             url = m_BookRecords[i].GetFullPath(m_BookRecords[i].GetStart());
 787             return url;
 788         }
 789     }
 790
 791     /* 3. try to find in contents: */
 792
 793     cnt = m_ContentsCnt;
 794     for (i = 0; i < cnt; i++)
 795     {
 796         if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
 797             {
 798             url = m_Contents[i].GetFullPath();
 799             return url;
 800         }
 801     }
 802
 803
 804     /* 4. try to find in index: */
 805
 806     cnt = m_IndexCnt;
 807     for (i = 0; i < cnt; i++)
 808     {
 809         if (wxStrcmp(m_Index[i].m_Name, x) == 0)
 810             {
 811             url = m_Index[i].GetFullPath();
 812             return url;
 813         }
 814     }
 815
 816     return url;
 817 }
 818
 819 wxString wxHtmlHelpData::FindPageById(int id)
 820 {
 821     int i;
 822     wxString url(wxEmptyString);
 823
 824     for (i = 0; i < m_ContentsCnt; i++)
 825     {
 826         if (m_Contents[i].m_ID == id)
 827             {
 828             url = m_Contents[i].GetFullPath();
 829             return url;
 830         }
 831     }
 832
 833     return url;
 834 }
 835
 836 //----------------------------------------------------------------------------------
 837 // wxHtmlSearchStatus functions
 838 //----------------------------------------------------------------------------------
 839
 840 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
 841                                        bool case_sensitive, bool whole_words_only,
 842                                        const wxString& book)
 843 {
 844     m_Data = data;
 845     m_Keyword = keyword;
 846     wxHtmlBookRecord* bookr = NULL;
 847     if (book != wxEmptyString)
 848     {
 849         // we have to search in a specific book. Find it first
 850         int i, cnt = data->m_BookRecords.GetCount();
 851         for (i = 0; i < cnt; i++)
 852             if (data->m_BookRecords[i].GetTitle() == book)
 853                 {
 854                 bookr = &(data->m_BookRecords[i]);
 855                 m_CurIndex = bookr->GetContentsStart();
 856                 m_MaxIndex = bookr->GetContentsEnd();
 857                 break;
 858             }
 859         // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
 860         wxASSERT(bookr);
 861     }
 862     if (! bookr)
 863     {
 864         // no book specified; search all books
 865         m_CurIndex = 0;
 866         m_MaxIndex = m_Data->m_ContentsCnt;
 867     }
 868     m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
 869     m_Active = (m_CurIndex < m_MaxIndex);
 870     m_LastPage = NULL;
 871 }
 872
 873 bool wxHtmlSearchStatus::Search()
 874 {
 875     wxFSFile *file;
 876     int i = m_CurIndex;  // shortcut
 877     bool found = FALSE;
 878     wxChar *thepage;
 879
 880     if (!m_Active)
 881     {
 882         // sanity check. Illegal use, but we'll try to prevent a crash anyway
 883         wxASSERT(m_Active);
 884         return FALSE;
 885     }
 886
 887     m_Name = wxEmptyString;
 888     m_ContentsItem = NULL;
 889     thepage = m_Data->m_Contents[i].m_Page;
 890
 891     m_Active = (++m_CurIndex < m_MaxIndex);
 892     // check if it is same page with different anchor:
 893     if (m_LastPage != NULL)
 894     {
 895         wxChar *p1, *p2;
 896         for (p1 = thepage, p2 = m_LastPage;
 897              *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
 898
 899         m_LastPage = thepage;
 900
 901         if (*p1 == 0 || *p1 == _T('#'))
 902             return FALSE;
 903     }
 904     else m_LastPage = thepage;
 905
 906     wxFileSystem fsys;
 907     file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetFullPath(thepage));
 908     if (file)
 909     {
 910         if (m_Engine.Scan(file->GetStream()))
 911         {
 912             m_Name = m_Data->m_Contents[i].m_Name;
 913             m_ContentsItem = m_Data->m_Contents + i;
 914             found = TRUE;
 915         }
 916         delete file;
 917     }
 918     return found;
 919 }
 920
 921
 922
 923
 924
 925
 926
 927
 928 //--------------------------------------------------------------------------------
 929 // wxSearchEngine
 930 //--------------------------------------------------------------------------------
 931
 932 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
 933 {
 934     m_CaseSensitive = case_sensitive;
 935     m_WholeWords = whole_words_only;
 936     if (m_Keyword) delete[] m_Keyword;
 937     m_Keyword = new wxChar[keyword.Length() + 1];
 938     wxStrcpy(m_Keyword, keyword.c_str());
 939
 940     if (!m_CaseSensitive)
 941     {
 942         for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
 943             {
 944             if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
 945                 m_Keyword[i] += wxT('a') - wxT('A');
 946                 }
 947         }
 948 }
 949
 950
 951
 952 #define WHITESPACE(c)  (c == ' ' || c == '\n' || c == '\r' || c == '\t')
 953
 954 bool wxSearchEngine::Scan(wxInputStream *stream)
 955 {
 956     wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
 957
 958     int i, j;
 959     int lng = stream ->GetSize();
 960     int wrd = wxStrlen(m_Keyword);
 961     bool found = FALSE;
 962     char *buf = new char[lng + 1];
 963     stream->Read(buf, lng);
 964     buf[lng] = 0;
 965
 966     if (!m_CaseSensitive)
 967         for (i = 0; i < lng; i++)
 968             if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
 969
 970     if (m_WholeWords)
 971     {
 972         for (i = 0; i < lng - wrd; i++)
 973             {
 974             if (WHITESPACE(buf[i])) continue;
 975             j = 0;
 976             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 977             if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
 978         }
 979     }
 980
 981     else
 982     {
 983         for (i = 0; i < lng - wrd; i++)
 984             {
 985             j = 0;
 986             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 987             if (j == wrd) { found = TRUE; break; }
 988         }
 989     }
 990
 991     delete[] buf;
 992     return found;
 993 }
 994
 995
 996
 997 #endif