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                 /* this one should ALWAYS stay the last one!!! */
 304                 ESCSEQ("amp", "&"),
 305                 ESCSEQ("#38", "&"),
 306
 307                 { NULL, NULL, NULL }
 308                 };
 309
 310                 for (int i = 0; substitutions[i][0] != NULL; i++)
 311                 {
 312                     m_Name.Replace(substitutions[i][0], substitutions[i][3], TRUE);
 313                     m_Name.Replace(substitutions[i][1], substitutions[i][3], TRUE);
 314                     m_Name.Replace(substitutions[i][2], substitutions[i][3], TRUE);
 315                 }
 316             }
 317         }
 318         if (tag.GetParam(wxT("NAME")) == wxT("Local")) m_Page = tag.GetParam(wxT("VALUE"));
 319         if (tag.GetParam(wxT("NAME")) == wxT("ID")) tag.ScanParam(wxT("VALUE"), wxT("%i"), &m_ID);
 320         return FALSE;
 321     }
 322 }
 323
 324
 325
 326 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
 327 {
 328     array = m_Items;
 329     size = m_ItemsCnt;
 330     m_Items = NULL;
 331     m_ItemsCnt = 0;
 332 }
 333
 334 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
 335 {
 336     m_Items = array;
 337     m_ItemsCnt = size;
 338 }
 339
 340
 341
 342
 343 //-----------------------------------------------------------------------------
 344 // wxHtmlHelpData
 345 //-----------------------------------------------------------------------------
 346
 347 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
 348
 349 wxHtmlHelpData::wxHtmlHelpData()
 350 {
 351     m_TempPath = wxEmptyString;
 352
 353     m_Contents = NULL;
 354     m_ContentsCnt = 0;
 355     m_Index = NULL;
 356     m_IndexCnt = 0;
 357 }
 358
 359 wxHtmlHelpData::~wxHtmlHelpData()
 360 {
 361     int i;
 362
 363     m_BookRecords.Empty();
 364     if (m_Contents)
 365     {
 366         for (i = 0; i < m_ContentsCnt; i++)
 367             {
 368             delete[] m_Contents[i].m_Page;
 369             delete[] m_Contents[i].m_Name;
 370         }
 371         free(m_Contents);
 372     }
 373     if (m_Index)
 374     {
 375         for (i = 0; i < m_IndexCnt; i++)
 376             {
 377             delete[] m_Index[i].m_Page;
 378             delete[] m_Index[i].m_Name;
 379         }
 380         free(m_Index);
 381     }
 382 }
 383
 384 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
 385 {
 386     wxFSFile *f;
 387     char *buf;
 388     int sz;
 389     wxString string;
 390
 391     HP_Parser parser;
 392     HP_TagHandler *handler = new HP_TagHandler(book);
 393     parser.AddTagHandler(handler);
 394
 395     f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
 396     if (f)
 397     {
 398         sz = f->GetStream()->GetSize();
 399         buf = new char[sz + 1];
 400         buf[sz] = 0;
 401         f->GetStream()->Read(buf, sz);
 402         delete f;
 403         handler->ReadIn(m_Contents, m_ContentsCnt);
 404         parser.Parse(buf);
 405         handler->WriteOut(m_Contents, m_ContentsCnt);
 406         delete[] buf;
 407     }
 408     else
 409         wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
 410
 411     f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
 412     if (f)
 413     {
 414         sz = f->GetStream()->GetSize();
 415         buf = new char[sz + 1];
 416         buf[sz] = 0;
 417         f->GetStream()->Read(buf, sz);
 418         delete f;
 419         handler->ReadIn(m_Index, m_IndexCnt);
 420         parser.Parse(buf);
 421         handler->WriteOut(m_Index, m_IndexCnt);
 422         delete[] buf;
 423     }
 424     else if (!indexfile.IsEmpty())
 425         wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
 426     return TRUE;
 427 }
 428
 429
 430
 431
 432 #if wxUSE_UNICODE
 433
 434 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
 435 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
 436
 437 #else
 438
 439 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
 440 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
 441
 442 #endif
 443
 444
 445 #define CURRENT_CACHED_BOOK_VERSION     1
 446
 447 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
 448 {
 449     int i, st;
 450     wxInt32 x;
 451     wxInt32 version;
 452
 453     /* load header - version info : */
 454
 455     f->Read(&x, sizeof(x));
 456     version = wxINT32_SWAP_ON_BE(x);
 457
 458     if (version != CURRENT_CACHED_BOOK_VERSION)
 459     {
 460         wxLogError(_("Incorrect version of HTML help book"));
 461         return FALSE;
 462         // NOTE: when adding new version, please ensure backward compatibility!
 463     }
 464
 465     /* load contents : */
 466
 467     f->Read(&x, sizeof(x));
 468     st = m_ContentsCnt;
 469     m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
 470     m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
 471                                                (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
 472                                                wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 473     for (i = st; i < m_ContentsCnt; i++)
 474     {
 475         f->Read(&x, sizeof(x));
 476         m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
 477         f->Read(&x, sizeof(x));
 478         m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
 479         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 480         m_Contents[i].m_Name = new wxChar[x];
 481         READ_STRING(f, m_Contents[i].m_Name, x);
 482         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 483         m_Contents[i].m_Page = new wxChar[x];
 484         READ_STRING(f, m_Contents[i].m_Page, x);
 485         m_Contents[i].m_Book = book;
 486     }
 487
 488     /* load index : */
 489
 490     f->Read(&x, sizeof(x));
 491     st = m_IndexCnt;
 492     m_IndexCnt += wxINT32_SWAP_ON_BE(x);
 493     m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
 494                                                      wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 495     for (i = st; i < m_IndexCnt; i++)
 496     {
 497         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 498         m_Index[i].m_Name = new wxChar[x];
 499         READ_STRING(f, m_Index[i].m_Name, x);
 500         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 501         m_Index[i].m_Page = new wxChar[x];
 502         READ_STRING(f, m_Index[i].m_Page, x);
 503         m_Index[i].m_Book = book;
 504     }
 505     return TRUE;
 506 }
 507
 508
 509 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
 510 {
 511     int i;
 512     wxInt32 x;
 513
 514     /* save header - version info : */
 515
 516     x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
 517     f->Write(&x, sizeof(x));
 518
 519     /* save contents : */
 520
 521     x = 0;
 522     for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
 523     x = wxINT32_SWAP_ON_BE(x);
 524     f->Write(&x, sizeof(x));
 525     for (i = 0; i < m_ContentsCnt; i++)
 526     {
 527         if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
 528         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
 529         f->Write(&x, sizeof(x));
 530         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
 531         f->Write(&x, sizeof(x));
 532         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
 533         f->Write(&x, sizeof(x));
 534         WRITE_STRING(f, m_Contents[i].m_Name, x);
 535         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
 536         f->Write(&x, sizeof(x));
 537         WRITE_STRING(f, m_Contents[i].m_Page, x);
 538     }
 539
 540     /* save index : */
 541
 542     x = 0;
 543     for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
 544     x = wxINT32_SWAP_ON_BE(x);
 545     f->Write(&x, sizeof(x));
 546     for (i = 0; i < m_IndexCnt; i++)
 547     {
 548         if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
 549         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
 550         f->Write(&x, sizeof(x));
 551         WRITE_STRING(f, m_Index[i].m_Name, x);
 552         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
 553         f->Write(&x, sizeof(x));
 554         WRITE_STRING(f, m_Index[i].m_Page, x);
 555     }
 556     return TRUE;
 557 }
 558
 559
 560 void wxHtmlHelpData::SetTempDir(const wxString& path)
 561 {
 562     if (path == wxEmptyString) m_TempPath = path;
 563     else
 564     {
 565         if (wxIsAbsolutePath(path)) m_TempPath = path;
 566         else m_TempPath = wxGetCwd() + _T("/") + path;
 567
 568         if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
 569             m_TempPath << _T('/');
 570     }
 571 }
 572
 573
 574
 575 static wxString SafeFileName(const wxString& s)
 576 {
 577     wxString res(s);
 578     res.Replace(wxT("#"), wxT("_"));
 579     res.Replace(wxT(":"), wxT("_"));
 580     res.Replace(wxT("\\"), wxT("_"));
 581     res.Replace(wxT("/"), wxT("_"));
 582     return res;
 583 }
 584
 585 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
 586                                   wxFontEncoding encoding,
 587                                   const wxString& title, const wxString& contfile,
 588                                   const wxString& indexfile, const wxString& deftopic,
 589                                   const wxString& path)
 590 {
 591     wxFileSystem fsys;
 592     wxFSFile *fi;
 593     wxHtmlBookRecord *bookr;
 594
 595     int IndexOld = m_IndexCnt,
 596         ContentsOld = m_ContentsCnt;
 597
 598     if (! path.IsEmpty())
 599         fsys.ChangePathTo(path, TRUE);
 600
 601     bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
 602
 603     if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
 604         m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 605     m_Contents[m_ContentsCnt].m_Level = 0;
 606     m_Contents[m_ContentsCnt].m_ID = 0;
 607     m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
 608     wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
 609     m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
 610     wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
 611     m_Contents[m_ContentsCnt].m_Book = bookr;
 612
 613     // store the contents index for later
 614     int cont_start = m_ContentsCnt++;
 615
 616     // Try to find cached binary versions:
 617     // 1. save file as book, but with .hhp.cached extension
 618     // 2. same as 1. but in temp path
 619     // 3. otherwise or if cache load failed, load it from MS.
 620
 621     fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
 622
 623     if (fi == NULL ||
 624           fi->GetModificationTime() < bookfile.GetModificationTime() ||
 625           !LoadCachedBook(bookr, fi->GetStream()))
 626     {
 627         if (fi != NULL) delete fi;
 628         fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
 629         if (m_TempPath == wxEmptyString || fi == NULL ||
 630             fi->GetModificationTime() < bookfile.GetModificationTime() ||
 631             !LoadCachedBook(bookr, fi->GetStream()))
 632         {
 633             LoadMSProject(bookr, fsys, indexfile, contfile);
 634             if (m_TempPath != wxEmptyString)
 635             {
 636                 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
 637                                                   SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
 638                 SaveCachedBook(bookr, outs);
 639                 delete outs;
 640             }
 641         }
 642     }
 643
 644     if (fi != NULL) delete fi;
 645
 646     // Now store the contents range
 647     bookr->SetContentsRange(cont_start, m_ContentsCnt);
 648
 649     // Convert encoding, if neccessary:
 650     if (encoding != wxFONTENCODING_SYSTEM)
 651     {
 652         wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
 653         if (a.GetCount() != 0 && a[0] != encoding)
 654         {
 655             int i;
 656             wxEncodingConverter conv;
 657             conv.Init(encoding, a[0]);
 658
 659             for (i = IndexOld; i < m_IndexCnt; i++)
 660                 conv.Convert(m_Index[i].m_Name);
 661             for (i = ContentsOld; i < m_ContentsCnt; i++)
 662                 conv.Convert(m_Contents[i].m_Name);
 663         }
 664     }
 665
 666     m_BookRecords.Add(bookr);
 667     if (m_IndexCnt > 0)
 668         qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), IndexCompareFunc);
 669
 670     return TRUE;
 671 }
 672
 673
 674 bool wxHtmlHelpData::AddBook(const wxString& book)
 675 {
 676     if (book.Right(4).Lower() == wxT(".zip") ||
 677         book.Right(4).Lower() == wxT(".htb") /*html book*/)
 678
 679     {
 680         wxFileSystem fsys;
 681         wxString s;
 682         bool rt = FALSE;
 683
 684         s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
 685         while (!s.IsEmpty())
 686         {
 687             if (AddBook(s)) rt = TRUE;
 688             s = fsys.FindNext();
 689         }
 690
 691         return rt;
 692     }
 693
 694
 695     else
 696     {
 697         wxFSFile *fi;
 698         wxFileSystem fsys;
 699         wxInputStream *s;
 700         wxString bookFull;
 701
 702         int sz;
 703         char *buff, *lineptr;
 704         char linebuf[300];
 705
 706         wxString title = _("noname"),
 707                  safetitle,
 708                  start = wxEmptyString,
 709                  contents = wxEmptyString,
 710                  index = wxEmptyString,
 711                  charset = wxEmptyString;
 712
 713         if (wxIsAbsolutePath(book)) bookFull = book;
 714         else bookFull = wxGetCwd() + "/" + book;
 715
 716         fi = fsys.OpenFile(bookFull);
 717         if (fi == NULL)
 718         {
 719             wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
 720             return FALSE;
 721         }
 722         fsys.ChangePathTo(bookFull);
 723         s = fi->GetStream();
 724         sz = s->GetSize();
 725         buff = new char[sz + 1];
 726         buff[sz] = 0;
 727         s->Read(buff, sz);
 728         lineptr = buff;
 729
 730         do {
 731             lineptr = ReadLine(lineptr, linebuf);
 732
 733             if (strstr(linebuf, "Title=") == linebuf)
 734                 title = linebuf + strlen("Title=");
 735             if (strstr(linebuf, "Default topic=") == linebuf)
 736                 start = linebuf + strlen("Default topic=");
 737             if (strstr(linebuf, "Index file=") == linebuf)
 738                 index = linebuf + strlen("Index file=");
 739             if (strstr(linebuf, "Contents file=") == linebuf)
 740                 contents = linebuf + strlen("Contents file=");
 741             if (strstr(linebuf, "Charset=") == linebuf)
 742                 charset = linebuf + strlen("Charset=");
 743         } while (lineptr != NULL);
 744         delete[] buff;
 745
 746         wxFontEncoding enc;
 747         if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
 748         else enc = wxTheFontMapper->CharsetToEncoding(charset);
 749         bool rtval = AddBookParam(*fi, enc,
 750                                   title, contents, index, start, fsys.GetPath());
 751         delete fi;
 752         return rtval;
 753     }
 754 }
 755
 756 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
 757 {
 758     int cnt;
 759     int i;
 760     wxFileSystem fsys;
 761     wxFSFile *f;
 762     wxString url(wxEmptyString);
 763
 764     /* 1. try to open given file: */
 765
 766     cnt = m_BookRecords.GetCount();
 767     for (i = 0; i < cnt; i++)
 768     {
 769         f = fsys.OpenFile(m_BookRecords[i].GetBasePath() + x);
 770         if (f)
 771             {
 772             url = m_BookRecords[i].GetBasePath() + x;
 773             delete f;
 774             return url;
 775         }
 776     }
 777
 778
 779     /* 2. try to find a book: */
 780
 781     for (i = 0; i < cnt; i++)
 782     {
 783         if (m_BookRecords[i].GetTitle() == x)
 784             {
 785             url = m_BookRecords[i].GetBasePath() + m_BookRecords[i].GetStart();
 786             return url;
 787         }
 788     }
 789
 790     /* 3. try to find in contents: */
 791
 792     cnt = m_ContentsCnt;
 793     for (i = 0; i < cnt; i++)
 794     {
 795         if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
 796             {
 797             url = m_Contents[i].m_Book->GetBasePath() + m_Contents[i].m_Page;
 798             return url;
 799         }
 800     }
 801
 802
 803     /* 4. try to find in index: */
 804
 805     cnt = m_IndexCnt;
 806     for (i = 0; i < cnt; i++)
 807     {
 808         if (wxStrcmp(m_Index[i].m_Name, x) == 0)
 809             {
 810             url = m_Index[i].m_Book->GetBasePath() + m_Index[i].m_Page;
 811             return url;
 812         }
 813     }
 814
 815     return url;
 816 }
 817
 818 wxString wxHtmlHelpData::FindPageById(int id)
 819 {
 820     int i;
 821     wxString url(wxEmptyString);
 822
 823     for (i = 0; i < m_ContentsCnt; i++)
 824     {
 825         if (m_Contents[i].m_ID == id)
 826             {
 827             url = m_Contents[i].m_Book->GetBasePath() + m_Contents[i].m_Page;
 828             return url;
 829         }
 830     }
 831
 832     return url;
 833 }
 834
 835 //----------------------------------------------------------------------------------
 836 // wxHtmlSearchStatus functions
 837 //----------------------------------------------------------------------------------
 838
 839 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
 840                                        bool case_sensitive, bool whole_words_only,
 841                                        const wxString& book)
 842 {
 843     m_Data = data;
 844     m_Keyword = keyword;
 845     wxHtmlBookRecord* bookr = NULL;
 846     if (book != wxEmptyString)
 847     {
 848         // we have to search in a specific book. Find it first
 849         int i, cnt = data->m_BookRecords.GetCount();
 850         for (i = 0; i < cnt; i++)
 851             if (data->m_BookRecords[i].GetTitle() == book)
 852                 {
 853                 bookr = &(data->m_BookRecords[i]);
 854                 m_CurIndex = bookr->GetContentsStart();
 855                 m_MaxIndex = bookr->GetContentsEnd();
 856                 break;
 857             }
 858         // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
 859         wxASSERT(bookr);
 860     }
 861     if (! bookr)
 862     {
 863         // no book specified; search all books
 864         m_CurIndex = 0;
 865         m_MaxIndex = m_Data->m_ContentsCnt;
 866     }
 867     m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
 868     m_Active = (m_CurIndex < m_MaxIndex);
 869     m_LastPage = NULL;
 870 }
 871
 872 bool wxHtmlSearchStatus::Search()
 873 {
 874     wxFSFile *file;
 875     int i = m_CurIndex;  // shortcut
 876     bool found = FALSE;
 877     wxChar *thepage;
 878
 879     if (!m_Active)
 880     {
 881         // sanity check. Illegal use, but we'll try to prevent a crash anyway
 882         wxASSERT(m_Active);
 883         return FALSE;
 884     }
 885
 886     m_Name = wxEmptyString;
 887     m_ContentsItem = NULL;
 888     thepage = m_Data->m_Contents[i].m_Page;
 889
 890     m_Active = (++m_CurIndex < m_MaxIndex);
 891     // check if it is same page with different anchor:
 892     if (m_LastPage != NULL)
 893     {
 894         wxChar *p1, *p2;
 895         for (p1 = thepage, p2 = m_LastPage;
 896              *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
 897
 898         m_LastPage = thepage;
 899
 900         if (*p1 == 0 || *p1 == _T('#'))
 901             return FALSE;
 902     }
 903     else m_LastPage = thepage;
 904
 905     wxFileSystem fsys;
 906     file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetBasePath() + thepage);
 907     if (file)
 908     {
 909         if (m_Engine.Scan(file->GetStream())) {
 910             m_Name = m_Data->m_Contents[i].m_Name;
 911             m_ContentsItem = m_Data->m_Contents + i;
 912             found = TRUE;
 913         }
 914         delete file;
 915     }
 916     return found;
 917 }
 918
 919
 920
 921
 922
 923
 924
 925
 926 //--------------------------------------------------------------------------------
 927 // wxSearchEngine
 928 //--------------------------------------------------------------------------------
 929
 930 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
 931 {
 932     m_CaseSensitive = case_sensitive;
 933     m_WholeWords = whole_words_only;
 934     if (m_Keyword) delete[] m_Keyword;
 935     m_Keyword = new wxChar[keyword.Length() + 1];
 936     wxStrcpy(m_Keyword, keyword.c_str());
 937
 938     if (!m_CaseSensitive)
 939     {
 940         for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
 941             {
 942             if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
 943                 m_Keyword[i] += wxT('a') - wxT('A');
 944                 }
 945         }
 946 }
 947
 948
 949
 950 #define WHITESPACE(c)  (c == ' ' || c == '\n' || c == '\r' || c == '\t')
 951
 952 bool wxSearchEngine::Scan(wxInputStream *stream)
 953 {
 954     wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
 955
 956     int i, j;
 957     int lng = stream ->GetSize();
 958     int wrd = wxStrlen(m_Keyword);
 959     bool found = FALSE;
 960     char *buf = new char[lng + 1];
 961     stream->Read(buf, lng);
 962     buf[lng] = 0;
 963
 964     if (!m_CaseSensitive)
 965         for (i = 0; i < lng; i++)
 966             if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
 967
 968     if (m_WholeWords)
 969     {
 970         for (i = 0; i < lng - wrd; i++)
 971             {
 972             if (WHITESPACE(buf[i])) continue;
 973             j = 0;
 974             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 975             if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
 976         }
 977     }
 978
 979     else
 980     {
 981         for (i = 0; i < lng - wrd; i++)
 982             {
 983             j = 0;
 984             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 985             if (j == wrd) { found = TRUE; break; }
 986         }
 987     }
 988
 989     delete[] buf;
 990     return found;
 991 }
 992
 993
 994
 995
 996 #endif