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 /*
 142             // We're reading in items at level zero, which must mean we want to specify
 143             // our own 'books', so chuck out the first (empty) one that AddBook already
 144             // created
 145             if (m_firstTime && (m_Level == 0) && (m_ItemsCnt > 0))
 146             {
 147                 delete[] m_Items[0].m_Name;
 148                 delete[] m_Items[0].m_Page;
 149                 m_ItemsCnt --;
 150             }
 151             else
 152             {
 153                 if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
 154                     m_Items = (wxHtmlContentsItem*) realloc(m_Items, (m_ItemsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 155             }
 156 */
 157             if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
 158                 m_Items = (wxHtmlContentsItem*) realloc(m_Items,
 159                                        (m_ItemsCnt + wxHTML_REALLOC_STEP) *
 160                                        sizeof(wxHtmlContentsItem));
 161
 162             m_Items[m_ItemsCnt].m_Level = m_Level;
 163             m_Items[m_ItemsCnt].m_ID = m_ID;
 164             m_Items[m_ItemsCnt].m_Page = new wxChar[m_Page.Length() + 1];
 165             wxStrcpy(m_Items[m_ItemsCnt].m_Page, m_Page.c_str());
 166             m_Items[m_ItemsCnt].m_Name = new wxChar [m_Name.Length() + 1];
 167             wxStrcpy(m_Items[m_ItemsCnt].m_Name, m_Name.c_str());
 168             m_Items[m_ItemsCnt].m_Book = m_Book;
 169             m_ItemsCnt++;
 170         }
 171
 172         return TRUE;
 173     }
 174     else
 175     { // "PARAM"
 176         if (m_Name == wxEmptyString && tag.GetParam(wxT("NAME")) == wxT("Name"))
 177         {
 178             m_Name = tag.GetParam(wxT("VALUE"));
 179             if (m_Name.Find(wxT('&')) != -1)
 180             {
 181 #define ESCSEQ(escape, subst)  \
 182                   { _T("&") _T(escape) _T(";"), _T("&") _T(escape) _T(" "), _T("&") _T(escape), _T(subst) }
 183         static wxChar* substitutions[][4] =
 184                         {
 185                 ESCSEQ("quot", "\""),
 186                 ESCSEQ("#34", "\""),
 187                 ESCSEQ("lt", "<"),
 188                 ESCSEQ("#60", "<"),
 189                 ESCSEQ("gt", ">"),
 190                 ESCSEQ("#62", ">"),
 191
 192                 ESCSEQ("#94", "^"), /* ^ */
 193
 194                 ESCSEQ("nbsp", " "),
 195                 ESCSEQ("#32", " "),
 196                 ESCSEQ("iexcl", "!"),
 197                 ESCSEQ("#33", "!"),
 198                 ESCSEQ("cent", "¢"/* ¢ */),
 199                 ESCSEQ("#162", "¢"/* ¢ */),
 200
 201                 ESCSEQ("trade", "(TM)"),
 202                 ESCSEQ("#153", "(TM)"),
 203
 204                 ESCSEQ("yen", "¥"),
 205                 ESCSEQ("#165", "¥"),
 206                 ESCSEQ("brkbar", "¦"),
 207                 ESCSEQ("#166", "¦"),
 208                 ESCSEQ("sect", "§"),
 209                 ESCSEQ("#167", "§"),
 210                 ESCSEQ("uml", "¨"),
 211                 ESCSEQ("#168", "¨"),
 212
 213                 ESCSEQ("copy", "©"), /* © */
 214                 ESCSEQ("#169", "©"),
 215                 ESCSEQ("ordf", "ª"),
 216                 ESCSEQ("#170", "ª"),
 217                 ESCSEQ("laquo", "«"), /* « */
 218                 ESCSEQ("#171", "«"),
 219                 ESCSEQ("not", "¬"),
 220                 ESCSEQ("#172", "¬"),
 221
 222                 ESCSEQ("reg", "®"), /* ® */
 223                 ESCSEQ("#174", "®"),
 224
 225                 ESCSEQ("deg", "°"), /* ° */
 226                 ESCSEQ("#176", "°"),
 227                 ESCSEQ("plusm", "±"), /* ± */
 228                 ESCSEQ("#177", "±"),
 229
 230                 ESCSEQ("acute", "´"),
 231                 ESCSEQ("#180", "´"),
 232                 ESCSEQ("macron", "¯"),
 233                 ESCSEQ("#175", "¯"),
 234                 ESCSEQ("micro", "µ"), /* µ */
 235                 ESCSEQ("#181", "µ"),
 236                 ESCSEQ("para", "¶"), /* ¶ */
 237                 ESCSEQ("#182", "¶"),
 238
 239                 ESCSEQ("ordm", "º"), /* º */
 240                 ESCSEQ("#186", "º"),
 241                 ESCSEQ("raquo", "»"), /* » */
 242                 ESCSEQ("#187", "»"),
 243
 244                 ESCSEQ("iquest", "¿"), /* ¿ */
 245                 ESCSEQ("#191", "¿"),
 246                 ESCSEQ("Agrave", "\300"/* À */),
 247                 ESCSEQ("#193", "\300"/* À */),
 248
 249                 ESCSEQ("Acirc", "\302"/* Â */),
 250                 ESCSEQ("Atilde", "\303"/* Ã */),
 251                 ESCSEQ("Auml", "\304"/* Ä */),
 252                 ESCSEQ("Aring", " "),
 253                 ESCSEQ("AElig", " "),
 254                 ESCSEQ("Ccedil", "\347"/* ç */),
 255                 ESCSEQ("Egrave", "\310"/* È */),
 256                 ESCSEQ("Eacute", "\311"/* É */),
 257                 ESCSEQ("Ecirc", "\312"/* Ê */),
 258                 ESCSEQ("Euml", "\313"/* Ë */),
 259                 ESCSEQ("Igrave", "\314"/* Ì */),
 260
 261                 ESCSEQ("Icirc", "\316"/* Î */),
 262                 ESCSEQ("Iuml", "\317"/* Ï */),
 263
 264                 ESCSEQ("Ntilde", "\321"/* Ñ */),
 265                 ESCSEQ("Ograve", "\322"/* Ò */),
 266
 267                 ESCSEQ("Ocirc", "\324"/* Ô */),
 268                 ESCSEQ("Otilde", "\325"/* Õ */),
 269                 ESCSEQ("Ouml", "\326"/* Ö */),
 270
 271                 ESCSEQ("Oslash", " "),
 272                 ESCSEQ("Ugrave", "\331"/* Ù */),
 273
 274                 ESCSEQ("Ucirc", " "),
 275                 ESCSEQ("Uuml", "\334"/* Ü */),
 276
 277                 ESCSEQ("szlig", "\247"/* § */),
 278                 ESCSEQ("agrave;","à"),
 279                 ESCSEQ("aacute", "\341"/* á */),
 280                 ESCSEQ("acirc", "\342"/* â */),
 281                 ESCSEQ("atilde", "\343"/* ã */),
 282                 ESCSEQ("auml", "\344"/* ä */),
 283                 ESCSEQ("aring", "a"),
 284                 ESCSEQ("aelig", "ae"),
 285                 ESCSEQ("ccedil", "\347"/* ç */),
 286                 ESCSEQ("egrave", "\350"/* è */),
 287                 ESCSEQ("eacute", "\351"/* é */),
 288                 ESCSEQ("ecirc", "\352"/* ê */),
 289                 ESCSEQ("euml", "\353"/* ë */),
 290                 ESCSEQ("igrave", "\354"/* ì */),
 291                 ESCSEQ("iacute", "\355"/* í */),
 292                 ESCSEQ("icirc", " "),
 293                 ESCSEQ("iuml", "\357"/* ï */),
 294                 ESCSEQ("eth", " "),
 295                 ESCSEQ("ntilde", "\361"/* ñ */),
 296                 ESCSEQ("ograve", "\362"/* ò */),
 297                 ESCSEQ("oacute", "\363"/* ó */),
 298                 ESCSEQ("ocirc", "\364"/* ô */),
 299                 ESCSEQ("otilde", "\365"/* õ */),
 300                 ESCSEQ("ouml", "\366"/* ö */),
 301                 ESCSEQ("divide", " "),
 302                 ESCSEQ("oslash", " "),
 303                 ESCSEQ("ugrave", "\371"/* ù */),
 304                 ESCSEQ("uacute", "\372"/* ú */),
 305                 ESCSEQ("ucirc", "\373"/* û */),
 306                 ESCSEQ("uuml", "\374"/* ü */),
 307
 308                 ESCSEQ("yuml", ""),
 309
 310                 /* this one should ALWAYS stay the last one!!! */
 311                 ESCSEQ("amp", "&"),
 312                 ESCSEQ("#38", "&"),
 313
 314                 { NULL, NULL, NULL }
 315                 };
 316
 317                 for (int i = 0; substitutions[i][0] != NULL; i++)
 318                 {
 319                     m_Name.Replace(substitutions[i][0], substitutions[i][3], TRUE);
 320                     m_Name.Replace(substitutions[i][1], substitutions[i][3], TRUE);
 321                     m_Name.Replace(substitutions[i][2], substitutions[i][3], TRUE);
 322                 }
 323             }
 324         }
 325         if (tag.GetParam(wxT("NAME")) == wxT("Local")) m_Page = tag.GetParam(wxT("VALUE"));
 326         if (tag.GetParam(wxT("NAME")) == wxT("ID")) tag.ScanParam(wxT("VALUE"), wxT("%i"), &m_ID);
 327         return FALSE;
 328     }
 329 }
 330
 331
 332
 333 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
 334 {
 335     array = m_Items;
 336     size = m_ItemsCnt;
 337     m_Items = NULL;
 338     m_ItemsCnt = 0;
 339 }
 340
 341 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
 342 {
 343     m_Items = array;
 344     m_ItemsCnt = size;
 345 }
 346
 347
 348
 349
 350 //-----------------------------------------------------------------------------
 351 // wxHtmlHelpData
 352 //-----------------------------------------------------------------------------
 353
 354 wxString wxHtmlBookRecord::GetFullPath(const wxString &page) const
 355 {
 356     if (wxIsAbsolutePath(page))
 357         return page;
 358     else
 359         return m_BasePath + page;
 360 }
 361
 362
 363
 364 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
 365
 366 wxHtmlHelpData::wxHtmlHelpData()
 367 {
 368     m_TempPath = wxEmptyString;
 369
 370     m_Contents = NULL;
 371     m_ContentsCnt = 0;
 372     m_Index = NULL;
 373     m_IndexCnt = 0;
 374 }
 375
 376 wxHtmlHelpData::~wxHtmlHelpData()
 377 {
 378     int i;
 379
 380     m_BookRecords.Empty();
 381     if (m_Contents)
 382     {
 383         for (i = 0; i < m_ContentsCnt; i++)
 384             {
 385             delete[] m_Contents[i].m_Page;
 386             delete[] m_Contents[i].m_Name;
 387         }
 388         free(m_Contents);
 389     }
 390     if (m_Index)
 391     {
 392         for (i = 0; i < m_IndexCnt; i++)
 393             {
 394             delete[] m_Index[i].m_Page;
 395             delete[] m_Index[i].m_Name;
 396         }
 397         free(m_Index);
 398     }
 399 }
 400
 401 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
 402 {
 403     wxFSFile *f;
 404     char *buf;
 405     int sz;
 406     wxString string;
 407
 408     HP_Parser parser;
 409     HP_TagHandler *handler = new HP_TagHandler(book);
 410     parser.AddTagHandler(handler);
 411
 412     f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
 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_Contents, m_ContentsCnt);
 421         parser.Parse(buf);
 422         handler->WriteOut(m_Contents, m_ContentsCnt);
 423         delete[] buf;
 424     }
 425     else
 426         wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
 427
 428     f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
 429     if (f)
 430     {
 431         sz = f->GetStream()->GetSize();
 432         buf = new char[sz + 1];
 433         buf[sz] = 0;
 434         f->GetStream()->Read(buf, sz);
 435         delete f;
 436         handler->ReadIn(m_Index, m_IndexCnt);
 437         parser.Parse(buf);
 438         handler->WriteOut(m_Index, m_IndexCnt);
 439         delete[] buf;
 440     }
 441     else if (!indexfile.IsEmpty())
 442         wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
 443     return TRUE;
 444 }
 445
 446
 447
 448
 449 #if wxUSE_UNICODE
 450
 451 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
 452 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
 453
 454 #else
 455
 456 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
 457 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
 458
 459 #endif
 460
 461
 462 #define CURRENT_CACHED_BOOK_VERSION     1
 463
 464 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
 465 {
 466     int i, st;
 467     wxInt32 x;
 468     wxInt32 version;
 469
 470     /* load header - version info : */
 471
 472     f->Read(&x, sizeof(x));
 473     version = wxINT32_SWAP_ON_BE(x);
 474
 475     if (version != CURRENT_CACHED_BOOK_VERSION)
 476     {
 477         wxLogError(_("Incorrect version of HTML help book"));
 478         return FALSE;
 479         // NOTE: when adding new version, please ensure backward compatibility!
 480     }
 481
 482     /* load contents : */
 483
 484     f->Read(&x, sizeof(x));
 485     st = m_ContentsCnt;
 486     m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
 487     m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
 488                                                (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
 489                                                wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 490     for (i = st; i < m_ContentsCnt; i++)
 491     {
 492         f->Read(&x, sizeof(x));
 493         m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
 494         f->Read(&x, sizeof(x));
 495         m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
 496         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 497         m_Contents[i].m_Name = new wxChar[x];
 498         READ_STRING(f, m_Contents[i].m_Name, x);
 499         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 500         m_Contents[i].m_Page = new wxChar[x];
 501         READ_STRING(f, m_Contents[i].m_Page, x);
 502         m_Contents[i].m_Book = book;
 503     }
 504
 505     /* load index : */
 506
 507     f->Read(&x, sizeof(x));
 508     st = m_IndexCnt;
 509     m_IndexCnt += wxINT32_SWAP_ON_BE(x);
 510     m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
 511                                                      wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 512     for (i = st; i < m_IndexCnt; i++)
 513     {
 514         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 515         m_Index[i].m_Name = new wxChar[x];
 516         READ_STRING(f, m_Index[i].m_Name, x);
 517         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 518         m_Index[i].m_Page = new wxChar[x];
 519         READ_STRING(f, m_Index[i].m_Page, x);
 520         m_Index[i].m_Book = book;
 521     }
 522     return TRUE;
 523 }
 524
 525
 526 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
 527 {
 528     int i;
 529     wxInt32 x;
 530
 531     /* save header - version info : */
 532
 533     x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
 534     f->Write(&x, sizeof(x));
 535
 536     /* save contents : */
 537
 538     x = 0;
 539     for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
 540     x = wxINT32_SWAP_ON_BE(x);
 541     f->Write(&x, sizeof(x));
 542     for (i = 0; i < m_ContentsCnt; i++)
 543     {
 544         if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
 545         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
 546         f->Write(&x, sizeof(x));
 547         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
 548         f->Write(&x, sizeof(x));
 549         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
 550         f->Write(&x, sizeof(x));
 551         WRITE_STRING(f, m_Contents[i].m_Name, x);
 552         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
 553         f->Write(&x, sizeof(x));
 554         WRITE_STRING(f, m_Contents[i].m_Page, x);
 555     }
 556
 557     /* save index : */
 558
 559     x = 0;
 560     for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
 561     x = wxINT32_SWAP_ON_BE(x);
 562     f->Write(&x, sizeof(x));
 563     for (i = 0; i < m_IndexCnt; i++)
 564     {
 565         if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
 566         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
 567         f->Write(&x, sizeof(x));
 568         WRITE_STRING(f, m_Index[i].m_Name, x);
 569         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
 570         f->Write(&x, sizeof(x));
 571         WRITE_STRING(f, m_Index[i].m_Page, x);
 572     }
 573     return TRUE;
 574 }
 575
 576
 577 void wxHtmlHelpData::SetTempDir(const wxString& path)
 578 {
 579     if (path == wxEmptyString) m_TempPath = path;
 580     else
 581     {
 582         if (wxIsAbsolutePath(path)) m_TempPath = path;
 583         else m_TempPath = wxGetCwd() + _T("/") + path;
 584
 585         if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
 586             m_TempPath << _T('/');
 587     }
 588 }
 589
 590
 591
 592 static wxString SafeFileName(const wxString& s)
 593 {
 594     wxString res(s);
 595     res.Replace(wxT("#"), wxT("_"));
 596     res.Replace(wxT(":"), wxT("_"));
 597     res.Replace(wxT("\\"), wxT("_"));
 598     res.Replace(wxT("/"), wxT("_"));
 599     return res;
 600 }
 601
 602 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
 603                                   wxFontEncoding encoding,
 604                                   const wxString& title, const wxString& contfile,
 605                                   const wxString& indexfile, const wxString& deftopic,
 606                                   const wxString& path)
 607 {
 608     wxFileSystem fsys;
 609     wxFSFile *fi;
 610     wxHtmlBookRecord *bookr;
 611
 612     int IndexOld = m_IndexCnt,
 613         ContentsOld = m_ContentsCnt;
 614
 615     if (! path.IsEmpty())
 616         fsys.ChangePathTo(path, TRUE);
 617
 618     bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
 619
 620     if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
 621         m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 622     m_Contents[m_ContentsCnt].m_Level = 0;
 623     m_Contents[m_ContentsCnt].m_ID = 0;
 624     m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
 625     wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
 626     m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
 627     wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
 628     m_Contents[m_ContentsCnt].m_Book = bookr;
 629
 630     // store the contents index for later
 631     int cont_start = m_ContentsCnt++;
 632
 633     // Try to find cached binary versions:
 634     // 1. save file as book, but with .hhp.cached extension
 635     // 2. same as 1. but in temp path
 636     // 3. otherwise or if cache load failed, load it from MS.
 637
 638     fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
 639
 640     if (fi == NULL ||
 641           fi->GetModificationTime() < bookfile.GetModificationTime() ||
 642           !LoadCachedBook(bookr, fi->GetStream()))
 643     {
 644         if (fi != NULL) delete fi;
 645         fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
 646         if (m_TempPath == wxEmptyString || fi == NULL ||
 647             fi->GetModificationTime() < bookfile.GetModificationTime() ||
 648             !LoadCachedBook(bookr, fi->GetStream()))
 649         {
 650             LoadMSProject(bookr, fsys, indexfile, contfile);
 651             if (m_TempPath != wxEmptyString)
 652             {
 653                 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
 654                                                   SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
 655                 SaveCachedBook(bookr, outs);
 656                 delete outs;
 657             }
 658         }
 659     }
 660
 661     if (fi != NULL) delete fi;
 662
 663     // Now store the contents range
 664     bookr->SetContentsRange(cont_start, m_ContentsCnt);
 665
 666     // Convert encoding, if neccessary:
 667     if (encoding != wxFONTENCODING_SYSTEM)
 668     {
 669         wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
 670         if (a.GetCount() != 0 && a[0] != encoding)
 671         {
 672             int i;
 673             wxEncodingConverter conv;
 674             conv.Init(encoding, a[0]);
 675
 676             for (i = IndexOld; i < m_IndexCnt; i++)
 677                 conv.Convert(m_Index[i].m_Name);
 678             for (i = ContentsOld; i < m_ContentsCnt; i++)
 679                 conv.Convert(m_Contents[i].m_Name);
 680         }
 681     }
 682
 683     m_BookRecords.Add(bookr);
 684     if (m_IndexCnt > 0)
 685         qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), IndexCompareFunc);
 686
 687     return TRUE;
 688 }
 689
 690
 691 bool wxHtmlHelpData::AddBook(const wxString& book)
 692 {
 693     if (book.Right(4).Lower() == wxT(".zip") ||
 694         book.Right(4).Lower() == wxT(".htb") /*html book*/)
 695
 696     {
 697         wxFileSystem fsys;
 698         wxString s;
 699         bool rt = FALSE;
 700
 701         s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
 702         while (!s.IsEmpty())
 703         {
 704             if (AddBook(s)) rt = TRUE;
 705             s = fsys.FindNext();
 706         }
 707
 708         return rt;
 709     }
 710
 711
 712     else
 713     {
 714         wxFSFile *fi;
 715         wxFileSystem fsys;
 716         wxInputStream *s;
 717         wxString bookFull;
 718
 719         int sz;
 720         char *buff, *lineptr;
 721         char linebuf[300];
 722
 723         wxString title = _("noname"),
 724                  safetitle,
 725                  start = wxEmptyString,
 726                  contents = wxEmptyString,
 727                  index = wxEmptyString,
 728                  charset = wxEmptyString;
 729
 730         if (wxIsAbsolutePath(book)) bookFull = book;
 731         else bookFull = wxGetCwd() + "/" + book;
 732
 733         fi = fsys.OpenFile(bookFull);
 734         if (fi == NULL)
 735         {
 736             wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
 737             return FALSE;
 738         }
 739         fsys.ChangePathTo(bookFull);
 740         s = fi->GetStream();
 741         sz = s->GetSize();
 742         buff = new char[sz + 1];
 743         buff[sz] = 0;
 744         s->Read(buff, sz);
 745         lineptr = buff;
 746
 747         do {
 748             lineptr = ReadLine(lineptr, linebuf);
 749
 750             if (strstr(linebuf, "Title=") == linebuf)
 751                 title = linebuf + strlen("Title=");
 752             if (strstr(linebuf, "Default topic=") == linebuf)
 753                 start = linebuf + strlen("Default topic=");
 754             if (strstr(linebuf, "Index file=") == linebuf)
 755                 index = linebuf + strlen("Index file=");
 756             if (strstr(linebuf, "Contents file=") == linebuf)
 757                 contents = linebuf + strlen("Contents file=");
 758             if (strstr(linebuf, "Charset=") == linebuf)
 759                 charset = linebuf + strlen("Charset=");
 760         } while (lineptr != NULL);
 761         delete[] buff;
 762
 763         wxFontEncoding enc;
 764         if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
 765         else enc = wxTheFontMapper->CharsetToEncoding(charset);
 766         bool rtval = AddBookParam(*fi, enc,
 767                                   title, contents, index, start, fsys.GetPath());
 768         delete fi;
 769         return rtval;
 770     }
 771 }
 772
 773 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
 774 {
 775     int cnt;
 776     int i;
 777     wxFileSystem fsys;
 778     wxFSFile *f;
 779     wxString url(wxEmptyString);
 780
 781     /* 1. try to open given file: */
 782
 783     cnt = m_BookRecords.GetCount();
 784     for (i = 0; i < cnt; i++)
 785     {
 786         f = fsys.OpenFile(m_BookRecords[i].GetFullPath(x));
 787         if (f)
 788             {
 789             url = m_BookRecords[i].GetFullPath(x);
 790             delete f;
 791             return url;
 792         }
 793     }
 794
 795
 796     /* 2. try to find a book: */
 797
 798     for (i = 0; i < cnt; i++)
 799     {
 800         if (m_BookRecords[i].GetTitle() == x)
 801             {
 802             url = m_BookRecords[i].GetFullPath(m_BookRecords[i].GetStart());
 803             return url;
 804         }
 805     }
 806
 807     /* 3. try to find in contents: */
 808
 809     cnt = m_ContentsCnt;
 810     for (i = 0; i < cnt; i++)
 811     {
 812         if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
 813             {
 814             url = m_Contents[i].GetFullPath();
 815             return url;
 816         }
 817     }
 818
 819
 820     /* 4. try to find in index: */
 821
 822     cnt = m_IndexCnt;
 823     for (i = 0; i < cnt; i++)
 824     {
 825         if (wxStrcmp(m_Index[i].m_Name, x) == 0)
 826             {
 827             url = m_Index[i].GetFullPath();
 828             return url;
 829         }
 830     }
 831
 832     return url;
 833 }
 834
 835 wxString wxHtmlHelpData::FindPageById(int id)
 836 {
 837     int i;
 838     wxString url(wxEmptyString);
 839
 840     for (i = 0; i < m_ContentsCnt; i++)
 841     {
 842         if (m_Contents[i].m_ID == id)
 843             {
 844             url = m_Contents[i].GetFullPath();
 845             return url;
 846         }
 847     }
 848
 849     return url;
 850 }
 851
 852 //----------------------------------------------------------------------------------
 853 // wxHtmlSearchStatus functions
 854 //----------------------------------------------------------------------------------
 855
 856 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
 857                                        bool case_sensitive, bool whole_words_only,
 858                                        const wxString& book)
 859 {
 860     m_Data = data;
 861     m_Keyword = keyword;
 862     wxHtmlBookRecord* bookr = NULL;
 863     if (book != wxEmptyString)
 864     {
 865         // we have to search in a specific book. Find it first
 866         int i, cnt = data->m_BookRecords.GetCount();
 867         for (i = 0; i < cnt; i++)
 868             if (data->m_BookRecords[i].GetTitle() == book)
 869                 {
 870                 bookr = &(data->m_BookRecords[i]);
 871                 m_CurIndex = bookr->GetContentsStart();
 872                 m_MaxIndex = bookr->GetContentsEnd();
 873                 break;
 874             }
 875         // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
 876         wxASSERT(bookr);
 877     }
 878     if (! bookr)
 879     {
 880         // no book specified; search all books
 881         m_CurIndex = 0;
 882         m_MaxIndex = m_Data->m_ContentsCnt;
 883     }
 884     m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
 885     m_Active = (m_CurIndex < m_MaxIndex);
 886     m_LastPage = NULL;
 887 }
 888
 889 bool wxHtmlSearchStatus::Search()
 890 {
 891     wxFSFile *file;
 892     int i = m_CurIndex;  // shortcut
 893     bool found = FALSE;
 894     wxChar *thepage;
 895
 896     if (!m_Active)
 897     {
 898         // sanity check. Illegal use, but we'll try to prevent a crash anyway
 899         wxASSERT(m_Active);
 900         return FALSE;
 901     }
 902
 903     m_Name = wxEmptyString;
 904     m_ContentsItem = NULL;
 905     thepage = m_Data->m_Contents[i].m_Page;
 906
 907     m_Active = (++m_CurIndex < m_MaxIndex);
 908     // check if it is same page with different anchor:
 909     if (m_LastPage != NULL)
 910     {
 911         wxChar *p1, *p2;
 912         for (p1 = thepage, p2 = m_LastPage;
 913              *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
 914
 915         m_LastPage = thepage;
 916
 917         if (*p1 == 0 || *p1 == _T('#'))
 918             return FALSE;
 919     }
 920     else m_LastPage = thepage;
 921
 922     wxFileSystem fsys;
 923     file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetFullPath(thepage));
 924     if (file)
 925     {
 926         if (m_Engine.Scan(file->GetStream()))
 927         {
 928             m_Name = m_Data->m_Contents[i].m_Name;
 929             m_ContentsItem = m_Data->m_Contents + i;
 930             found = TRUE;
 931         }
 932         delete file;
 933     }
 934     return found;
 935 }
 936
 937
 938
 939
 940
 941
 942
 943
 944 //--------------------------------------------------------------------------------
 945 // wxSearchEngine
 946 //--------------------------------------------------------------------------------
 947
 948 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
 949 {
 950     m_CaseSensitive = case_sensitive;
 951     m_WholeWords = whole_words_only;
 952     if (m_Keyword) delete[] m_Keyword;
 953     m_Keyword = new wxChar[keyword.Length() + 1];
 954     wxStrcpy(m_Keyword, keyword.c_str());
 955
 956     if (!m_CaseSensitive)
 957     {
 958         for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
 959             {
 960             if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
 961                 m_Keyword[i] += wxT('a') - wxT('A');
 962                 }
 963         }
 964 }
 965
 966
 967
 968 #define WHITESPACE(c)  (c == ' ' || c == '\n' || c == '\r' || c == '\t')
 969
 970 bool wxSearchEngine::Scan(wxInputStream *stream)
 971 {
 972     wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
 973
 974     int i, j;
 975     int lng = stream ->GetSize();
 976     int wrd = wxStrlen(m_Keyword);
 977     bool found = FALSE;
 978     char *buf = new char[lng + 1];
 979     stream->Read(buf, lng);
 980     buf[lng] = 0;
 981
 982     if (!m_CaseSensitive)
 983         for (i = 0; i < lng; i++)
 984             if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
 985
 986     if (m_WholeWords)
 987     {
 988         for (i = 0; i < lng - wrd; i++)
 989             {
 990             if (WHITESPACE(buf[i])) continue;
 991             j = 0;
 992             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 993             if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
 994         }
 995     }
 996
 997     else
 998     {
 999         for (i = 0; i < lng - wrd; i++)
1000             {
1001             j = 0;
1002             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
1003             if (j == wrd) { found = TRUE; break; }
1004         }
1005     }
1006
1007     delete[] buf;
1008     return found;
1009 }
1010
1011
1012
1013 #endif