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         bool m_firstTime; // For checking if we're adding sections at level zero, so we 'delete' the first one
  95
  96     public:
  97         HP_TagHandler(wxHtmlBookRecord *b) : wxHtmlTagHandler()
  98                 { m_Book = b; m_Items = NULL; m_ItemsCnt = 0; m_Name = m_Page = wxEmptyString;
  99                       m_Level = 0; m_ID = -1; m_firstTime = TRUE; }
 100         wxString GetSupportedTags() { return wxT("UL,OBJECT,PARAM"); }
 101         bool HandleTag(const wxHtmlTag& tag);
 102         void WriteOut(wxHtmlContentsItem*& array, int& size);
 103         void ReadIn(wxHtmlContentsItem* array, int size);
 104 };
 105
 106
 107 bool HP_TagHandler::HandleTag(const wxHtmlTag& tag)
 108 {
 109     if (tag.GetName() == wxT("UL"))
 110     {
 111         m_Level++;
 112         ParseInner(tag);
 113         m_Level--;
 114         return TRUE;
 115     }
 116     else if (tag.GetName() == wxT("OBJECT"))
 117     {
 118         m_Name = m_Page = wxEmptyString;
 119         ParseInner(tag);
 120
 121         if (tag.GetParam("TYPE") == "text/sitemap")
 122
 123         // if (!m_Page.IsEmpty())
 124         /* Valid HHW's file may contain only two object tags:
 125
 126            <OBJECT type="text/site properties">
 127                <param name="ImageType" value="Folder">
 128            </OBJECT>
 129
 130            or
 131
 132                <OBJECT type="text/sitemap">
 133                        <param name="Name" value="main page">
 134                        <param name="Local" value="another.htm">
 135                    </OBJECT>
 136
 137            We're interested in the latter. !m_Page.IsEmpty() is valid
 138            condition because text/site properties does not contain Local param
 139         */
 140         {
 141             // We're reading in items at level zero, which must mean we want to specify
 142             // our own 'books', so chuck out the first (empty) one that AddBook already
 143             // created
 144             if (m_firstTime && (m_Level == 0) && (m_ItemsCnt > 0))
 145             {
 146                 m_ItemsCnt --;
 147             }
 148             else
 149             {
 150                 if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
 151                     m_Items = (wxHtmlContentsItem*) realloc(m_Items, (m_ItemsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 152             }
 153             m_Items[m_ItemsCnt].m_Level = m_Level;
 154             m_Items[m_ItemsCnt].m_ID = m_ID;
 155             m_Items[m_ItemsCnt].m_Page = new wxChar[m_Page.Length() + 1];
 156             wxStrcpy(m_Items[m_ItemsCnt].m_Page, m_Page.c_str());
 157             m_Items[m_ItemsCnt].m_Name = new wxChar [m_Name.Length() + 1];
 158             wxStrcpy(m_Items[m_ItemsCnt].m_Name, m_Name.c_str());
 159             m_Items[m_ItemsCnt].m_Book = m_Book;
 160             m_ItemsCnt++;
 161
 162             m_firstTime = FALSE;
 163         }
 164
 165         return TRUE;
 166     }
 167     else
 168     { // "PARAM"
 169         if (m_Name == wxEmptyString && tag.GetParam(wxT("NAME")) == wxT("Name"))
 170         {
 171             m_Name = tag.GetParam(wxT("VALUE"));
 172             if (m_Name.Find(wxT('&')) != -1)
 173             {
 174 #define ESCSEQ(escape, subst)  \
 175                   { _T("&") _T(escape) _T(";"), _T("&") _T(escape) _T(" "), _T("&") _T(escape), _T(subst) }
 176         static wxChar* substitutions[][4] =
 177                         {
 178                 ESCSEQ("quot", "\""),
 179                 ESCSEQ("#34", "\""),
 180                 ESCSEQ("lt", "<"),
 181                 ESCSEQ("#60", "<"),
 182                 ESCSEQ("gt", ">"),
 183                 ESCSEQ("#62", ">"),
 184
 185                 ESCSEQ("#94", "^"), /* ^ */
 186
 187                 ESCSEQ("nbsp", " "),
 188                 ESCSEQ("#32", " "),
 189                 ESCSEQ("iexcl", "!"),
 190                 ESCSEQ("#33", "!"),
 191                 ESCSEQ("cent", "¢"/* ¢ */),
 192                 ESCSEQ("#162", "¢"/* ¢ */),
 193
 194                 ESCSEQ("trade", "(TM)"),
 195                 ESCSEQ("#153", "(TM)"),
 196
 197                 ESCSEQ("yen", "¥"),
 198                 ESCSEQ("#165", "¥"),
 199                 ESCSEQ("brkbar", "¦"),
 200                 ESCSEQ("#166", "¦"),
 201                 ESCSEQ("sect", "§"),
 202                 ESCSEQ("#167", "§"),
 203                 ESCSEQ("uml", "¨"),
 204                 ESCSEQ("#168", "¨"),
 205
 206                 ESCSEQ("copy", "©"), /* © */
 207                 ESCSEQ("#169", "©"),
 208                 ESCSEQ("ordf", "ª"),
 209                 ESCSEQ("#170", "ª"),
 210                 ESCSEQ("laquo", "«"), /* « */
 211                 ESCSEQ("#171", "«"),
 212                 ESCSEQ("not", "¬"),
 213                 ESCSEQ("#172", "¬"),
 214
 215                 ESCSEQ("reg", "®"), /* ® */
 216                 ESCSEQ("#174", "®"),
 217
 218                 ESCSEQ("deg", "°"), /* ° */
 219                 ESCSEQ("#176", "°"),
 220                 ESCSEQ("plusm", "±"), /* ± */
 221                 ESCSEQ("#177", "±"),
 222
 223                 ESCSEQ("acute", "´"),
 224                 ESCSEQ("#180", "´"),
 225                 ESCSEQ("macron", "¯"),
 226                 ESCSEQ("#175", "¯"),
 227                 ESCSEQ("micro", "µ"), /* µ */
 228                 ESCSEQ("#181", "µ"),
 229                 ESCSEQ("para", "¶"), /* ¶ */
 230                 ESCSEQ("#182", "¶"),
 231
 232                 ESCSEQ("ordm", "º"), /* º */
 233                 ESCSEQ("#186", "º"),
 234                 ESCSEQ("raquo", "»"), /* » */
 235                 ESCSEQ("#187", "»"),
 236
 237                 ESCSEQ("iquest", "¿"), /* ¿ */
 238                 ESCSEQ("#191", "¿"),
 239                 ESCSEQ("Agrave", "\300"/* À */),
 240                 ESCSEQ("#193", "\300"/* À */),
 241
 242                 ESCSEQ("Acirc", "\302"/* Â */),
 243                 ESCSEQ("Atilde", "\303"/* Ã */),
 244                 ESCSEQ("Auml", "\304"/* Ä */),
 245                 ESCSEQ("Aring", " "),
 246                 ESCSEQ("AElig", " "),
 247                 ESCSEQ("Ccedil", "\347"/* ç */),
 248                 ESCSEQ("Egrave", "\310"/* È */),
 249                 ESCSEQ("Eacute", "\311"/* É */),
 250                 ESCSEQ("Ecirc", "\312"/* Ê */),
 251                 ESCSEQ("Euml", "\313"/* Ë */),
 252                 ESCSEQ("Igrave", "\314"/* Ì */),
 253
 254                 ESCSEQ("Icirc", "\316"/* Î */),
 255                 ESCSEQ("Iuml", "\317"/* Ï */),
 256
 257                 ESCSEQ("Ntilde", "\321"/* Ñ */),
 258                 ESCSEQ("Ograve", "\322"/* Ò */),
 259
 260                 ESCSEQ("Ocirc", "\324"/* Ô */),
 261                 ESCSEQ("Otilde", "\325"/* Õ */),
 262                 ESCSEQ("Ouml", "\326"/* Ö */),
 263
 264                 ESCSEQ("Oslash", " "),
 265                 ESCSEQ("Ugrave", "\331"/* Ù */),
 266
 267                 ESCSEQ("Ucirc", " "),
 268                 ESCSEQ("Uuml", "\334"/* Ü */),
 269
 270                 ESCSEQ("szlig", "\247"/* § */),
 271                 ESCSEQ("agrave;","à"),
 272                 ESCSEQ("aacute", "\341"/* á */),
 273                 ESCSEQ("acirc", "\342"/* â */),
 274                 ESCSEQ("atilde", "\343"/* ã */),
 275                 ESCSEQ("auml", "\344"/* ä */),
 276                 ESCSEQ("aring", "a"),
 277                 ESCSEQ("aelig", "ae"),
 278                 ESCSEQ("ccedil", "\347"/* ç */),
 279                 ESCSEQ("egrave", "\350"/* è */),
 280                 ESCSEQ("eacute", "\351"/* é */),
 281                 ESCSEQ("ecirc", "\352"/* ê */),
 282                 ESCSEQ("euml", "\353"/* ë */),
 283                 ESCSEQ("igrave", "\354"/* ì */),
 284                 ESCSEQ("iacute", "\355"/* í */),
 285                 ESCSEQ("icirc", " "),
 286                 ESCSEQ("iuml", "\357"/* ï */),
 287                 ESCSEQ("eth", " "),
 288                 ESCSEQ("ntilde", "\361"/* ñ */),
 289                 ESCSEQ("ograve", "\362"/* ò */),
 290                 ESCSEQ("oacute", "\363"/* ó */),
 291                 ESCSEQ("ocirc", "\364"/* ô */),
 292                 ESCSEQ("otilde", "\365"/* õ */),
 293                 ESCSEQ("ouml", "\366"/* ö */),
 294                 ESCSEQ("divide", " "),
 295                 ESCSEQ("oslash", " "),
 296                 ESCSEQ("ugrave", "\371"/* ù */),
 297                 ESCSEQ("uacute", "\372"/* ú */),
 298                 ESCSEQ("ucirc", "\373"/* û */),
 299                 ESCSEQ("uuml", "\374"/* ü */),
 300
 301                 ESCSEQ("yuml", ""),
 302
 303                 ESCSEQ("#60", "<"),
 304                 ESCSEQ("#62", ">"),
 305
 306                 /* this one should ALWAYS stay the last one!!! */
 307                 ESCSEQ("amp", "&"),
 308                 ESCSEQ("#38", "&"),
 309
 310                 { NULL, NULL, NULL }
 311                 };
 312
 313                 for (int i = 0; substitutions[i][0] != NULL; i++)
 314                 {
 315                     m_Name.Replace(substitutions[i][0], substitutions[i][3], TRUE);
 316                     m_Name.Replace(substitutions[i][1], substitutions[i][3], TRUE);
 317                     m_Name.Replace(substitutions[i][2], substitutions[i][3], TRUE);
 318                 }
 319             }
 320         }
 321         if (tag.GetParam(wxT("NAME")) == wxT("Local")) m_Page = tag.GetParam(wxT("VALUE"));
 322         if (tag.GetParam(wxT("NAME")) == wxT("ID")) tag.ScanParam(wxT("VALUE"), wxT("%i"), &m_ID);
 323         return FALSE;
 324     }
 325 }
 326
 327
 328
 329 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
 330 {
 331     array = m_Items;
 332     size = m_ItemsCnt;
 333     m_Items = NULL;
 334     m_ItemsCnt = 0;
 335 }
 336
 337 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
 338 {
 339     m_Items = array;
 340     m_ItemsCnt = size;
 341 }
 342
 343
 344
 345
 346 //-----------------------------------------------------------------------------
 347 // wxHtmlHelpData
 348 //-----------------------------------------------------------------------------
 349
 350 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
 351
 352 wxHtmlHelpData::wxHtmlHelpData()
 353 {
 354     m_TempPath = wxEmptyString;
 355
 356     m_Contents = NULL;
 357     m_ContentsCnt = 0;
 358     m_Index = NULL;
 359     m_IndexCnt = 0;
 360 }
 361
 362 wxHtmlHelpData::~wxHtmlHelpData()
 363 {
 364     int i;
 365
 366     m_BookRecords.Empty();
 367     if (m_Contents)
 368     {
 369         for (i = 0; i < m_ContentsCnt; i++)
 370             {
 371             delete[] m_Contents[i].m_Page;
 372             delete[] m_Contents[i].m_Name;
 373         }
 374         free(m_Contents);
 375     }
 376     if (m_Index)
 377     {
 378         for (i = 0; i < m_IndexCnt; i++)
 379             {
 380             delete[] m_Index[i].m_Page;
 381             delete[] m_Index[i].m_Name;
 382         }
 383         free(m_Index);
 384     }
 385 }
 386
 387 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
 388 {
 389     wxFSFile *f;
 390     char *buf;
 391     int sz;
 392     wxString string;
 393
 394     HP_Parser parser;
 395     HP_TagHandler *handler = new HP_TagHandler(book);
 396     parser.AddTagHandler(handler);
 397
 398     f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
 399     if (f)
 400     {
 401         sz = f->GetStream()->GetSize();
 402         buf = new char[sz + 1];
 403         buf[sz] = 0;
 404         f->GetStream()->Read(buf, sz);
 405         delete f;
 406         handler->ReadIn(m_Contents, m_ContentsCnt);
 407         parser.Parse(buf);
 408         handler->WriteOut(m_Contents, m_ContentsCnt);
 409         delete[] buf;
 410     }
 411     else
 412         wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
 413
 414     f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
 415     if (f)
 416     {
 417         sz = f->GetStream()->GetSize();
 418         buf = new char[sz + 1];
 419         buf[sz] = 0;
 420         f->GetStream()->Read(buf, sz);
 421         delete f;
 422         handler->ReadIn(m_Index, m_IndexCnt);
 423         parser.Parse(buf);
 424         handler->WriteOut(m_Index, m_IndexCnt);
 425         delete[] buf;
 426     }
 427     else if (!indexfile.IsEmpty())
 428         wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
 429     return TRUE;
 430 }
 431
 432
 433
 434
 435 #if wxUSE_UNICODE
 436
 437 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
 438 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
 439
 440 #else
 441
 442 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
 443 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
 444
 445 #endif
 446
 447
 448 #define CURRENT_CACHED_BOOK_VERSION     1
 449
 450 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
 451 {
 452     int i, st;
 453     wxInt32 x;
 454     wxInt32 version;
 455
 456     /* load header - version info : */
 457
 458     f->Read(&x, sizeof(x));
 459     version = wxINT32_SWAP_ON_BE(x);
 460
 461     if (version != CURRENT_CACHED_BOOK_VERSION)
 462     {
 463         wxLogError(_("Incorrect version of HTML help book"));
 464         return FALSE;
 465         // NOTE: when adding new version, please ensure backward compatibility!
 466     }
 467
 468     /* load contents : */
 469
 470     f->Read(&x, sizeof(x));
 471     st = m_ContentsCnt;
 472     m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
 473     m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
 474                                                (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
 475                                                wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 476     for (i = st; i < m_ContentsCnt; i++)
 477     {
 478         f->Read(&x, sizeof(x));
 479         m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
 480         f->Read(&x, sizeof(x));
 481         m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
 482         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 483         m_Contents[i].m_Name = new wxChar[x];
 484         READ_STRING(f, m_Contents[i].m_Name, x);
 485         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 486         m_Contents[i].m_Page = new wxChar[x];
 487         READ_STRING(f, m_Contents[i].m_Page, x);
 488         m_Contents[i].m_Book = book;
 489     }
 490
 491     /* load index : */
 492
 493     f->Read(&x, sizeof(x));
 494     st = m_IndexCnt;
 495     m_IndexCnt += wxINT32_SWAP_ON_BE(x);
 496     m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
 497                                                      wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 498     for (i = st; i < m_IndexCnt; i++)
 499     {
 500         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 501         m_Index[i].m_Name = new wxChar[x];
 502         READ_STRING(f, m_Index[i].m_Name, x);
 503         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 504         m_Index[i].m_Page = new wxChar[x];
 505         READ_STRING(f, m_Index[i].m_Page, x);
 506         m_Index[i].m_Book = book;
 507     }
 508     return TRUE;
 509 }
 510
 511
 512 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
 513 {
 514     int i;
 515     wxInt32 x;
 516
 517     /* save header - version info : */
 518
 519     x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
 520     f->Write(&x, sizeof(x));
 521
 522     /* save contents : */
 523
 524     x = 0;
 525     for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
 526     x = wxINT32_SWAP_ON_BE(x);
 527     f->Write(&x, sizeof(x));
 528     for (i = 0; i < m_ContentsCnt; i++)
 529     {
 530         if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
 531         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
 532         f->Write(&x, sizeof(x));
 533         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
 534         f->Write(&x, sizeof(x));
 535         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
 536         f->Write(&x, sizeof(x));
 537         WRITE_STRING(f, m_Contents[i].m_Name, x);
 538         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
 539         f->Write(&x, sizeof(x));
 540         WRITE_STRING(f, m_Contents[i].m_Page, x);
 541     }
 542
 543     /* save index : */
 544
 545     x = 0;
 546     for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
 547     x = wxINT32_SWAP_ON_BE(x);
 548     f->Write(&x, sizeof(x));
 549     for (i = 0; i < m_IndexCnt; i++)
 550     {
 551         if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
 552         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
 553         f->Write(&x, sizeof(x));
 554         WRITE_STRING(f, m_Index[i].m_Name, x);
 555         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
 556         f->Write(&x, sizeof(x));
 557         WRITE_STRING(f, m_Index[i].m_Page, x);
 558     }
 559     return TRUE;
 560 }
 561
 562
 563 void wxHtmlHelpData::SetTempDir(const wxString& path)
 564 {
 565     if (path == wxEmptyString) m_TempPath = path;
 566     else
 567     {
 568         if (wxIsAbsolutePath(path)) m_TempPath = path;
 569         else m_TempPath = wxGetCwd() + _T("/") + path;
 570
 571         if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
 572             m_TempPath << _T('/');
 573     }
 574 }
 575
 576
 577
 578 static wxString SafeFileName(const wxString& s)
 579 {
 580     wxString res(s);
 581     res.Replace(wxT("#"), wxT("_"));
 582     res.Replace(wxT(":"), wxT("_"));
 583     res.Replace(wxT("\\"), wxT("_"));
 584     res.Replace(wxT("/"), wxT("_"));
 585     return res;
 586 }
 587
 588 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
 589                                   wxFontEncoding encoding,
 590                                   const wxString& title, const wxString& contfile,
 591                                   const wxString& indexfile, const wxString& deftopic,
 592                                   const wxString& path)
 593 {
 594     wxFileSystem fsys;
 595     wxFSFile *fi;
 596     wxHtmlBookRecord *bookr;
 597
 598     int IndexOld = m_IndexCnt,
 599         ContentsOld = m_ContentsCnt;
 600
 601     if (! path.IsEmpty())
 602         fsys.ChangePathTo(path, TRUE);
 603
 604     bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
 605
 606     if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
 607         m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 608     m_Contents[m_ContentsCnt].m_Level = 0;
 609     m_Contents[m_ContentsCnt].m_ID = 0;
 610     m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
 611     wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
 612     m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
 613     wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
 614     m_Contents[m_ContentsCnt].m_Book = bookr;
 615
 616     // store the contents index for later
 617     int cont_start = m_ContentsCnt++;
 618
 619     // Try to find cached binary versions:
 620     // 1. save file as book, but with .hhp.cached extension
 621     // 2. same as 1. but in temp path
 622     // 3. otherwise or if cache load failed, load it from MS.
 623
 624     fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
 625
 626     if (fi == NULL ||
 627           fi->GetModificationTime() < bookfile.GetModificationTime() ||
 628           !LoadCachedBook(bookr, fi->GetStream()))
 629     {
 630         if (fi != NULL) delete fi;
 631         fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
 632         if (m_TempPath == wxEmptyString || fi == NULL ||
 633             fi->GetModificationTime() < bookfile.GetModificationTime() ||
 634             !LoadCachedBook(bookr, fi->GetStream()))
 635         {
 636             LoadMSProject(bookr, fsys, indexfile, contfile);
 637             if (m_TempPath != wxEmptyString)
 638             {
 639                 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
 640                                                   SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
 641                 SaveCachedBook(bookr, outs);
 642                 delete outs;
 643             }
 644         }
 645     }
 646
 647     if (fi != NULL) delete fi;
 648
 649     // Now store the contents range
 650     bookr->SetContentsRange(cont_start, m_ContentsCnt);
 651
 652     // Convert encoding, if neccessary:
 653     if (encoding != wxFONTENCODING_SYSTEM)
 654     {
 655         wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
 656         if (a.GetCount() != 0 && a[0] != encoding)
 657         {
 658             int i;
 659             wxEncodingConverter conv;
 660             conv.Init(encoding, a[0]);
 661
 662             for (i = IndexOld; i < m_IndexCnt; i++)
 663                 conv.Convert(m_Index[i].m_Name);
 664             for (i = ContentsOld; i < m_ContentsCnt; i++)
 665                 conv.Convert(m_Contents[i].m_Name);
 666         }
 667     }
 668
 669     m_BookRecords.Add(bookr);
 670     if (m_IndexCnt > 0)
 671         qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), IndexCompareFunc);
 672
 673     return TRUE;
 674 }
 675
 676
 677 bool wxHtmlHelpData::AddBook(const wxString& book)
 678 {
 679     if (book.Right(4).Lower() == wxT(".zip") ||
 680         book.Right(4).Lower() == wxT(".htb") /*html book*/)
 681
 682     {
 683         wxFileSystem fsys;
 684         wxString s;
 685         bool rt = FALSE;
 686
 687         s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
 688         while (!s.IsEmpty())
 689         {
 690             if (AddBook(s)) rt = TRUE;
 691             s = fsys.FindNext();
 692         }
 693
 694         return rt;
 695     }
 696
 697
 698     else
 699     {
 700         wxFSFile *fi;
 701         wxFileSystem fsys;
 702         wxInputStream *s;
 703         wxString bookFull;
 704
 705         int sz;
 706         char *buff, *lineptr;
 707         char linebuf[300];
 708
 709         wxString title = _("noname"),
 710                  safetitle,
 711                  start = wxEmptyString,
 712                  contents = wxEmptyString,
 713                  index = wxEmptyString,
 714                  charset = wxEmptyString;
 715
 716         if (wxIsAbsolutePath(book)) bookFull = book;
 717         else bookFull = wxGetCwd() + "/" + book;
 718
 719         fi = fsys.OpenFile(bookFull);
 720         if (fi == NULL)
 721         {
 722             wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
 723             return FALSE;
 724         }
 725         fsys.ChangePathTo(bookFull);
 726         s = fi->GetStream();
 727         sz = s->GetSize();
 728         buff = new char[sz + 1];
 729         buff[sz] = 0;
 730         s->Read(buff, sz);
 731         lineptr = buff;
 732
 733         do {
 734             lineptr = ReadLine(lineptr, linebuf);
 735
 736             if (strstr(linebuf, "Title=") == linebuf)
 737                 title = linebuf + strlen("Title=");
 738             if (strstr(linebuf, "Default topic=") == linebuf)
 739                 start = linebuf + strlen("Default topic=");
 740             if (strstr(linebuf, "Index file=") == linebuf)
 741                 index = linebuf + strlen("Index file=");
 742             if (strstr(linebuf, "Contents file=") == linebuf)
 743                 contents = linebuf + strlen("Contents file=");
 744             if (strstr(linebuf, "Charset=") == linebuf)
 745                 charset = linebuf + strlen("Charset=");
 746         } while (lineptr != NULL);
 747         delete[] buff;
 748
 749         wxFontEncoding enc;
 750         if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
 751         else enc = wxTheFontMapper->CharsetToEncoding(charset);
 752         bool rtval = AddBookParam(*fi, enc,
 753                                   title, contents, index, start, fsys.GetPath());
 754         delete fi;
 755         return rtval;
 756     }
 757 }
 758
 759 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
 760 {
 761     int cnt;
 762     int i;
 763     wxFileSystem fsys;
 764     wxFSFile *f;
 765     wxString url(wxEmptyString);
 766
 767     /* 1. try to open given file: */
 768
 769     cnt = m_BookRecords.GetCount();
 770     for (i = 0; i < cnt; i++)
 771     {
 772         f = fsys.OpenFile(m_BookRecords[i].GetBasePath() + x);
 773         if (f)
 774             {
 775             url = m_BookRecords[i].GetBasePath() + x;
 776             delete f;
 777             return url;
 778         }
 779     }
 780
 781
 782     /* 2. try to find a book: */
 783
 784     for (i = 0; i < cnt; i++)
 785     {
 786         if (m_BookRecords[i].GetTitle() == x)
 787             {
 788             url = m_BookRecords[i].GetBasePath() + m_BookRecords[i].GetStart();
 789             return url;
 790         }
 791     }
 792
 793     /* 3. try to find in contents: */
 794
 795     cnt = m_ContentsCnt;
 796     for (i = 0; i < cnt; i++)
 797     {
 798         if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
 799             {
 800             url = m_Contents[i].m_Book->GetBasePath() + m_Contents[i].m_Page;
 801             return url;
 802         }
 803     }
 804
 805
 806     /* 4. try to find in index: */
 807
 808     cnt = m_IndexCnt;
 809     for (i = 0; i < cnt; i++)
 810     {
 811         if (wxStrcmp(m_Index[i].m_Name, x) == 0)
 812             {
 813             url = m_Index[i].m_Book->GetBasePath() + m_Index[i].m_Page;
 814             return url;
 815         }
 816     }
 817
 818     return url;
 819 }
 820
 821 wxString wxHtmlHelpData::FindPageById(int id)
 822 {
 823     int i;
 824     wxString url(wxEmptyString);
 825
 826     for (i = 0; i < m_ContentsCnt; i++)
 827     {
 828         if (m_Contents[i].m_ID == id)
 829             {
 830             url = m_Contents[i].m_Book->GetBasePath() + m_Contents[i].m_Page;
 831             return url;
 832         }
 833     }
 834
 835     return url;
 836 }
 837
 838 //----------------------------------------------------------------------------------
 839 // wxHtmlSearchStatus functions
 840 //----------------------------------------------------------------------------------
 841
 842 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
 843                                        bool case_sensitive, bool whole_words_only,
 844                                        const wxString& book)
 845 {
 846     m_Data = data;
 847     m_Keyword = keyword;
 848     wxHtmlBookRecord* bookr = NULL;
 849     if (book != wxEmptyString)
 850     {
 851         // we have to search in a specific book. Find it first
 852         int i, cnt = data->m_BookRecords.GetCount();
 853         for (i = 0; i < cnt; i++)
 854             if (data->m_BookRecords[i].GetTitle() == book)
 855                 {
 856                 bookr = &(data->m_BookRecords[i]);
 857                 m_CurIndex = bookr->GetContentsStart();
 858                 m_MaxIndex = bookr->GetContentsEnd();
 859                 break;
 860             }
 861         // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
 862         wxASSERT(bookr);
 863     }
 864     if (! bookr)
 865     {
 866         // no book specified; search all books
 867         m_CurIndex = 0;
 868         m_MaxIndex = m_Data->m_ContentsCnt;
 869     }
 870     m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
 871     m_Active = (m_CurIndex < m_MaxIndex);
 872     m_LastPage = NULL;
 873 }
 874
 875 bool wxHtmlSearchStatus::Search()
 876 {
 877     wxFSFile *file;
 878     int i = m_CurIndex;  // shortcut
 879     bool found = FALSE;
 880     wxChar *thepage;
 881
 882     if (!m_Active)
 883     {
 884         // sanity check. Illegal use, but we'll try to prevent a crash anyway
 885         wxASSERT(m_Active);
 886         return FALSE;
 887     }
 888
 889     m_Name = wxEmptyString;
 890     m_ContentsItem = NULL;
 891     thepage = m_Data->m_Contents[i].m_Page;
 892
 893     m_Active = (++m_CurIndex < m_MaxIndex);
 894     // check if it is same page with different anchor:
 895     if (m_LastPage != NULL)
 896     {
 897         wxChar *p1, *p2;
 898         for (p1 = thepage, p2 = m_LastPage;
 899              *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
 900
 901         m_LastPage = thepage;
 902
 903         if (*p1 == 0 || *p1 == _T('#'))
 904             return FALSE;
 905     }
 906     else m_LastPage = thepage;
 907
 908     wxFileSystem fsys;
 909     file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetBasePath() + thepage);
 910     if (file)
 911     {
 912         if (m_Engine.Scan(file->GetStream())) {
 913             m_Name = m_Data->m_Contents[i].m_Name;
 914             m_ContentsItem = m_Data->m_Contents + i;
 915             found = TRUE;
 916         }
 917         delete file;
 918     }
 919     return found;
 920 }
 921
 922
 923
 924
 925
 926
 927
 928
 929 //--------------------------------------------------------------------------------
 930 // wxSearchEngine
 931 //--------------------------------------------------------------------------------
 932
 933 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
 934 {
 935     m_CaseSensitive = case_sensitive;
 936     m_WholeWords = whole_words_only;
 937     if (m_Keyword) delete[] m_Keyword;
 938     m_Keyword = new wxChar[keyword.Length() + 1];
 939     wxStrcpy(m_Keyword, keyword.c_str());
 940
 941     if (!m_CaseSensitive)
 942     {
 943         for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
 944             {
 945             if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
 946                 m_Keyword[i] += wxT('a') - wxT('A');
 947                 }
 948         }
 949 }
 950
 951
 952
 953 #define WHITESPACE(c)  (c == ' ' || c == '\n' || c == '\r' || c == '\t')
 954
 955 bool wxSearchEngine::Scan(wxInputStream *stream)
 956 {
 957     wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
 958
 959     int i, j;
 960     int lng = stream ->GetSize();
 961     int wrd = wxStrlen(m_Keyword);
 962     bool found = FALSE;
 963     char *buf = new char[lng + 1];
 964     stream->Read(buf, lng);
 965     buf[lng] = 0;
 966
 967     if (!m_CaseSensitive)
 968         for (i = 0; i < lng; i++)
 969             if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
 970
 971     if (m_WholeWords)
 972     {
 973         for (i = 0; i < lng - wrd; i++)
 974             {
 975             if (WHITESPACE(buf[i])) continue;
 976             j = 0;
 977             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 978             if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
 979         }
 980     }
 981
 982     else
 983     {
 984         for (i = 0; i < lng - wrd; i++)
 985             {
 986             j = 0;
 987             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 988             if (j == wrd) { found = TRUE; break; }
 989         }
 990     }
 991
 992     delete[] buf;
 993     return found;
 994 }
 995
 996
 997
 998
 999 #endif