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