src/html/helpdata.cpp

   1 /////////////////////////////////////////////////////////////////////////////
   2 // Name:        helpdata.cpp
   3 // Purpose:     wxHtmlHelpData
   4 // Notes:       Based on htmlhelp.cpp, implementing a monolithic
   5 //              HTML Help controller class,  by Vaclav Slavik
   6 // Author:      Harm van der Heijden and Vaclav Slavik
   7 // RCS-ID:      $Id$
   8 // Copyright:   (c) Harm van der Heijden and Vaclav Slavik
   9 // Licence:     wxWindows licence
  10 /////////////////////////////////////////////////////////////////////////////
  11
  12 #ifdef __GNUG__
  13 #pragma implementation
  14 #endif
  15
  16 // For compilers that support precompilation, includes "wx.h".
  17 #include "wx/wxprec.h"
  18
  19 #ifdef __BORLANDC__
  20 #pragma hdrstop
  21 #endif
  22
  23 #include "wx/defs.h"
  24
  25 #if wxUSE_HTML && wxUSE_STREAMS
  26
  27 #ifndef WXPRECOMP
  28 #include "wx/wx.h"
  29 #endif
  30
  31 #include "wx/html/helpdata.h"
  32 #include "wx/tokenzr.h"
  33 #include "wx/wfstream.h"
  34 #include "wx/busyinfo.h"
  35 #include "wx/encconv.h"
  36 #include "wx/fontmap.h"
  37 #include "wx/log.h"
  38 #include "wx/html/htmlpars.h"
  39 #include "wx/html/htmldefs.h"
  40
  41 #include "wx/arrimpl.cpp"
  42 WX_DEFINE_OBJARRAY(wxHtmlBookRecArray)
  43
  44 //-----------------------------------------------------------------------------
  45 // static helper functions
  46 //-----------------------------------------------------------------------------
  47
  48 // Reads one line, stores it into buf and returns pointer to new line or NULL.
  49 static char* ReadLine(char *line, char *buf)
  50 {
  51     char *writeptr = buf, *readptr = line;
  52
  53     while (*readptr != 0 && *readptr != '\r' && *readptr != '\n') *(writeptr++) = *(readptr++);
  54     *writeptr = 0;
  55     while (*readptr == '\r' || *readptr == '\n') readptr++;
  56     if (*readptr == 0) return NULL;
  57     else return readptr;
  58 }
  59
  60
  61
  62 static int LINKAGEMODE IndexCompareFunc(const void *a, const void *b)
  63 {
  64     return wxStrcmp(((wxHtmlContentsItem*)a)->m_Name, ((wxHtmlContentsItem*)b)->m_Name);
  65 }
  66
  67
  68 //-----------------------------------------------------------------------------
  69 // HP_Parser
  70 //-----------------------------------------------------------------------------
  71
  72 class HP_Parser : public wxHtmlParser
  73 {
  74     public:
  75         void AddText(const char* WXUNUSED(text)) { }
  76         wxObject* GetProduct() { return NULL; }
  77 };
  78
  79
  80 //-----------------------------------------------------------------------------
  81 // HP_TagHandler
  82 //-----------------------------------------------------------------------------
  83
  84 class HP_TagHandler : public wxHtmlTagHandler
  85 {
  86     private:
  87         wxString m_Name, m_Page;
  88         int m_Level;
  89         int m_ID;
  90         int m_Index;
  91         wxHtmlContentsItem *m_Items;
  92         int m_ItemsCnt;
  93         wxHtmlBookRecord *m_Book;
  94
  95     public:
  96         HP_TagHandler(wxHtmlBookRecord *b) : wxHtmlTagHandler()
  97                 { m_Book = b; m_Items = NULL; m_ItemsCnt = 0; m_Name = m_Page = wxEmptyString;
  98                       m_Level = 0; m_ID = -1; }
  99         wxString GetSupportedTags() { return wxT("UL,OBJECT,PARAM"); }
 100         bool HandleTag(const wxHtmlTag& tag);
 101         void WriteOut(wxHtmlContentsItem*& array, int& size);
 102         void ReadIn(wxHtmlContentsItem* array, int size);
 103 };
 104
 105
 106 bool HP_TagHandler::HandleTag(const wxHtmlTag& tag)
 107 {
 108     if (tag.GetName() == wxT("UL"))
 109     {
 110         m_Level++;
 111         ParseInner(tag);
 112         m_Level--;
 113         return TRUE;
 114     }
 115     else if (tag.GetName() == wxT("OBJECT"))
 116     {
 117         m_Name = m_Page = wxEmptyString;
 118         ParseInner(tag);
 119
 120 #if 0
 121          if (!m_Page.IsEmpty())
 122         /* Valid HHW's file may contain only two object tags:
 123
 124            <OBJECT type="text/site properties">
 125                <param name="ImageType" value="Folder">
 126            </OBJECT>
 127
 128            or
 129
 130                <OBJECT type="text/sitemap">
 131                        <param name="Name" value="main page">
 132                        <param name="Local" value="another.htm">
 133                    </OBJECT>
 134
 135            We're interested in the latter. !m_Page.IsEmpty() is valid
 136            condition because text/site properties does not contain Local param
 137         */
 138 #endif
 139         if (tag.GetParam(wxT("TYPE")) == wxT("text/sitemap"))
 140         {
 141             if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
 142                 m_Items = (wxHtmlContentsItem*) realloc(m_Items,
 143                                        (m_ItemsCnt + wxHTML_REALLOC_STEP) *
 144                                        sizeof(wxHtmlContentsItem));
 145
 146             m_Items[m_ItemsCnt].m_Level = m_Level;
 147             m_Items[m_ItemsCnt].m_ID = m_ID;
 148             m_Items[m_ItemsCnt].m_Page = new wxChar[m_Page.Length() + 1];
 149             wxStrcpy(m_Items[m_ItemsCnt].m_Page, m_Page.c_str());
 150             m_Items[m_ItemsCnt].m_Name = new wxChar [m_Name.Length() + 1];
 151             wxStrcpy(m_Items[m_ItemsCnt].m_Name, m_Name.c_str());
 152             m_Items[m_ItemsCnt].m_Book = m_Book;
 153             m_ItemsCnt++;
 154         }
 155
 156         return TRUE;
 157     }
 158     else
 159     { // "PARAM"
 160         if (m_Name == wxEmptyString && tag.GetParam(wxT("NAME")) == wxT("Name"))
 161         {
 162             m_Name = tag.GetParam(wxT("VALUE"));
 163             if (m_Name.Find(wxT('&')) != -1)
 164             {
 165 #define ESCSEQ(escape, subst)  \
 166                   { _T("&") _T(escape) _T(";"), _T("&") _T(escape) _T(" "), _T("&") _T(escape), _T(subst) }
 167         static wxChar* substitutions[][4] =
 168                         {
 169                 ESCSEQ("quot", "\""),
 170                 ESCSEQ("#34", "\""),
 171                 ESCSEQ("#8220", "\""),
 172                 ESCSEQ("#8221", "\""),
 173                 ESCSEQ("lt", "<"),
 174                 ESCSEQ("#60", "<"),
 175                 ESCSEQ("gt", ">"),
 176                 ESCSEQ("#62", ">"),
 177
 178                 ESCSEQ("#94", "^"), /* ^ */
 179
 180                 ESCSEQ("nbsp", " "),
 181                 ESCSEQ("#32", " "),
 182                 ESCSEQ("iexcl", "!"),
 183                 ESCSEQ("#33", "!"),
 184                 ESCSEQ("cent", "¢"/* ¢ */),
 185                 ESCSEQ("#162", "¢"/* ¢ */),
 186
 187                 ESCSEQ("trade", "(TM)"),
 188                 ESCSEQ("#153", "(TM)"),
 189                 ESCSEQ("#8482", "(TM)"),
 190
 191                 ESCSEQ("yen", "¥"),
 192                 ESCSEQ("#165", "¥"),
 193                 ESCSEQ("brkbar", "¦"),
 194                 ESCSEQ("#166", "¦"),
 195                 ESCSEQ("sect", "§"),
 196                 ESCSEQ("#167", "§"),
 197                 ESCSEQ("uml", "¨"),
 198                 ESCSEQ("#168", "¨"),
 199
 200                 ESCSEQ("copy", "©"), /* © */
 201                 ESCSEQ("#169", "©"),
 202                 ESCSEQ("ordf", "ª"),
 203                 ESCSEQ("#170", "ª"),
 204                 ESCSEQ("laquo", "«"), /* « */
 205                 ESCSEQ("#171", "«"),
 206                 ESCSEQ("not", "¬"),
 207                 ESCSEQ("#172", "¬"),
 208
 209                 ESCSEQ("reg", "®"), /* ® */
 210                 ESCSEQ("#174", "®"),
 211
 212                 ESCSEQ("deg", "°"), /* ° */
 213                 ESCSEQ("#176", "°"),
 214                 ESCSEQ("plusm", "±"), /* ± */
 215                 ESCSEQ("#177", "±"),
 216
 217                 ESCSEQ("acute", "´"),
 218                 ESCSEQ("#180", "´"),
 219                 ESCSEQ("macron", "¯"),
 220                 ESCSEQ("#175", "¯"),
 221                 ESCSEQ("micro", "µ"), /* µ */
 222                 ESCSEQ("#181", "µ"),
 223                 ESCSEQ("para", "¶"), /* ¶ */
 224                 ESCSEQ("#182", "¶"),
 225
 226                 ESCSEQ("ordm", "º"), /* º */
 227                 ESCSEQ("#186", "º"),
 228                 ESCSEQ("raquo", "»"), /* » */
 229                 ESCSEQ("#187", "»"),
 230
 231                 ESCSEQ("iquest", "¿"), /* ¿ */
 232                 ESCSEQ("#191", "¿"),
 233                 ESCSEQ("Agrave", "\300"/* À */),
 234                 ESCSEQ("#193", "\300"/* À */),
 235
 236                 ESCSEQ("Acirc", "\302"/* Â */),
 237                 ESCSEQ("Atilde", "\303"/* Ã */),
 238                 ESCSEQ("Auml", "\304"/* Ä */),
 239                 ESCSEQ("Aring", " "),
 240                 ESCSEQ("AElig", " "),
 241                 ESCSEQ("Ccedil", "\347"/* ç */),
 242                 ESCSEQ("Egrave", "\310"/* È */),
 243                 ESCSEQ("Eacute", "\311"/* É */),
 244                 ESCSEQ("Ecirc", "\312"/* Ê */),
 245                 ESCSEQ("Euml", "\313"/* Ë */),
 246                 ESCSEQ("Igrave", "\314"/* Ì */),
 247
 248                 ESCSEQ("Icirc", "\316"/* Î */),
 249                 ESCSEQ("Iuml", "\317"/* Ï */),
 250
 251                 ESCSEQ("Ntilde", "\321"/* Ñ */),
 252                 ESCSEQ("Ograve", "\322"/* Ò */),
 253
 254                 ESCSEQ("Ocirc", "\324"/* Ô */),
 255                 ESCSEQ("Otilde", "\325"/* Õ */),
 256                 ESCSEQ("Ouml", "\326"/* Ö */),
 257
 258                 ESCSEQ("Oslash", " "),
 259                 ESCSEQ("Ugrave", "\331"/* Ù */),
 260
 261                 ESCSEQ("Ucirc", " "),
 262                 ESCSEQ("Uuml", "\334"/* Ü */),
 263
 264                 ESCSEQ("szlig", "\247"/* § */),
 265                 ESCSEQ("agrave","à"),
 266                 ESCSEQ("aacute", "\341"/* á */),
 267                 ESCSEQ("acirc", "\342"/* â */),
 268                 ESCSEQ("atilde", "\343"/* ã */),
 269                 ESCSEQ("auml", "\344"/* ä */),
 270                 ESCSEQ("aring", "a"),
 271                 ESCSEQ("aelig", "ae"),
 272                 ESCSEQ("ccedil", "\347"/* ç */),
 273                 ESCSEQ("egrave", "\350"/* è */),
 274                 ESCSEQ("eacute", "\351"/* é */),
 275                 ESCSEQ("ecirc", "\352"/* ê */),
 276                 ESCSEQ("euml", "\353"/* ë */),
 277                 ESCSEQ("igrave", "\354"/* ì */),
 278                 ESCSEQ("iacute", "\355"/* í */),
 279                 ESCSEQ("icirc", " "),
 280                 ESCSEQ("iuml", "\357"/* ï */),
 281                 ESCSEQ("eth", " "),
 282                 ESCSEQ("ntilde", "\361"/* ñ */),
 283                 ESCSEQ("ograve", "\362"/* ò */),
 284                 ESCSEQ("oacute", "\363"/* ó */),
 285                 ESCSEQ("ocirc", "\364"/* ô */),
 286                 ESCSEQ("otilde", "\365"/* õ */),
 287                 ESCSEQ("ouml", "\366"/* ö */),
 288                 ESCSEQ("divide", " "),
 289                 ESCSEQ("oslash", " "),
 290                 ESCSEQ("ugrave", "\371"/* ù */),
 291                 ESCSEQ("uacute", "\372"/* ú */),
 292                 ESCSEQ("ucirc", "\373"/* û */),
 293                 ESCSEQ("uuml", "\374"/* ü */),
 294
 295                 ESCSEQ("yuml", ""),
 296
 297                 /* this one should ALWAYS stay the last one!!! */
 298                 ESCSEQ("amp", "&"),
 299                 ESCSEQ("#38", "&"),
 300
 301                 { NULL, NULL, NULL }
 302                 };
 303
 304                 for (int i = 0; substitutions[i][0] != NULL; i++)
 305                 {
 306                     m_Name.Replace(substitutions[i][0], substitutions[i][3], TRUE);
 307                     m_Name.Replace(substitutions[i][1], substitutions[i][3], TRUE);
 308                     m_Name.Replace(substitutions[i][2], substitutions[i][3], TRUE);
 309                 }
 310             }
 311         }
 312         if (tag.GetParam(wxT("NAME")) == wxT("Local")) m_Page = tag.GetParam(wxT("VALUE"));
 313         if (tag.GetParam(wxT("NAME")) == wxT("ID")) tag.ScanParam(wxT("VALUE"), wxT("%i"), &m_ID);
 314         return FALSE;
 315     }
 316 }
 317
 318
 319
 320 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
 321 {
 322     array = m_Items;
 323     size = m_ItemsCnt;
 324     m_Items = NULL;
 325     m_ItemsCnt = 0;
 326 }
 327
 328 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
 329 {
 330     m_Items = array;
 331     m_ItemsCnt = size;
 332 }
 333
 334
 335
 336
 337 //-----------------------------------------------------------------------------
 338 // wxHtmlHelpData
 339 //-----------------------------------------------------------------------------
 340
 341 wxString wxHtmlBookRecord::GetFullPath(const wxString &page) const
 342 {
 343     if (wxIsAbsolutePath(page))
 344         return page;
 345     else
 346         return m_BasePath + page;
 347 }
 348
 349
 350
 351 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
 352
 353 wxHtmlHelpData::wxHtmlHelpData()
 354 {
 355     m_TempPath = wxEmptyString;
 356
 357     m_Contents = NULL;
 358     m_ContentsCnt = 0;
 359     m_Index = NULL;
 360     m_IndexCnt = 0;
 361 }
 362
 363 wxHtmlHelpData::~wxHtmlHelpData()
 364 {
 365     int i;
 366
 367     m_BookRecords.Empty();
 368     if (m_Contents)
 369     {
 370         for (i = 0; i < m_ContentsCnt; i++)
 371             {
 372             delete[] m_Contents[i].m_Page;
 373             delete[] m_Contents[i].m_Name;
 374         }
 375         free(m_Contents);
 376     }
 377     if (m_Index)
 378     {
 379         for (i = 0; i < m_IndexCnt; i++)
 380             {
 381             delete[] m_Index[i].m_Page;
 382             delete[] m_Index[i].m_Name;
 383         }
 384         free(m_Index);
 385     }
 386 }
 387
 388 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
 389 {
 390     wxFSFile *f;
 391     char *buf;
 392     int sz;
 393     wxString string;
 394
 395     HP_Parser parser;
 396     HP_TagHandler *handler = new HP_TagHandler(book);
 397     parser.AddTagHandler(handler);
 398
 399     f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
 400     if (f)
 401     {
 402         sz = f->GetStream()->GetSize();
 403         buf = new char[sz + 1];
 404         buf[sz] = 0;
 405         f->GetStream()->Read(buf, sz);
 406         delete f;
 407         handler->ReadIn(m_Contents, m_ContentsCnt);
 408         parser.Parse(buf);
 409         handler->WriteOut(m_Contents, m_ContentsCnt);
 410         delete[] buf;
 411     }
 412     else
 413         wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
 414
 415     f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
 416     if (f)
 417     {
 418         sz = f->GetStream()->GetSize();
 419         buf = new char[sz + 1];
 420         buf[sz] = 0;
 421         f->GetStream()->Read(buf, sz);
 422         delete f;
 423         handler->ReadIn(m_Index, m_IndexCnt);
 424         parser.Parse(buf);
 425         handler->WriteOut(m_Index, m_IndexCnt);
 426         delete[] buf;
 427     }
 428     else if (!indexfile.IsEmpty())
 429         wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
 430     return TRUE;
 431 }
 432
 433
 434
 435
 436 #if wxUSE_UNICODE
 437
 438 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
 439 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
 440
 441 #else
 442
 443 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
 444 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
 445
 446 #endif
 447
 448
 449 #define CURRENT_CACHED_BOOK_VERSION     1
 450
 451 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
 452 {
 453     int i, st;
 454     wxInt32 x;
 455     wxInt32 version;
 456
 457     /* load header - version info : */
 458
 459     f->Read(&x, sizeof(x));
 460     version = wxINT32_SWAP_ON_BE(x);
 461
 462     if (version != CURRENT_CACHED_BOOK_VERSION)
 463     {
 464         wxLogError(_("Incorrect version of HTML help book"));
 465         return FALSE;
 466         // NOTE: when adding new version, please ensure backward compatibility!
 467     }
 468
 469     /* load contents : */
 470
 471     f->Read(&x, sizeof(x));
 472     st = m_ContentsCnt;
 473     m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
 474     m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
 475                                                (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
 476                                                wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 477     for (i = st; i < m_ContentsCnt; i++)
 478     {
 479         f->Read(&x, sizeof(x));
 480         m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
 481         f->Read(&x, sizeof(x));
 482         m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
 483         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 484         m_Contents[i].m_Name = new wxChar[x];
 485         READ_STRING(f, m_Contents[i].m_Name, x);
 486         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 487         m_Contents[i].m_Page = new wxChar[x];
 488         READ_STRING(f, m_Contents[i].m_Page, x);
 489         m_Contents[i].m_Book = book;
 490     }
 491
 492     /* load index : */
 493
 494     f->Read(&x, sizeof(x));
 495     st = m_IndexCnt;
 496     m_IndexCnt += wxINT32_SWAP_ON_BE(x);
 497     m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
 498                                                      wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
 499     for (i = st; i < m_IndexCnt; i++)
 500     {
 501         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 502         m_Index[i].m_Name = new wxChar[x];
 503         READ_STRING(f, m_Index[i].m_Name, x);
 504         f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
 505         m_Index[i].m_Page = new wxChar[x];
 506         READ_STRING(f, m_Index[i].m_Page, x);
 507         m_Index[i].m_Book = book;
 508     }
 509     return TRUE;
 510 }
 511
 512
 513 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
 514 {
 515     int i;
 516     wxInt32 x;
 517
 518     /* save header - version info : */
 519
 520     x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
 521     f->Write(&x, sizeof(x));
 522
 523     /* save contents : */
 524
 525     x = 0;
 526     for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
 527     x = wxINT32_SWAP_ON_BE(x);
 528     f->Write(&x, sizeof(x));
 529     for (i = 0; i < m_ContentsCnt; i++)
 530     {
 531         if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
 532         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
 533         f->Write(&x, sizeof(x));
 534         x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
 535         f->Write(&x, sizeof(x));
 536         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
 537         f->Write(&x, sizeof(x));
 538         WRITE_STRING(f, m_Contents[i].m_Name, x);
 539         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
 540         f->Write(&x, sizeof(x));
 541         WRITE_STRING(f, m_Contents[i].m_Page, x);
 542     }
 543
 544     /* save index : */
 545
 546     x = 0;
 547     for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
 548     x = wxINT32_SWAP_ON_BE(x);
 549     f->Write(&x, sizeof(x));
 550     for (i = 0; i < m_IndexCnt; i++)
 551     {
 552         if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
 553         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
 554         f->Write(&x, sizeof(x));
 555         WRITE_STRING(f, m_Index[i].m_Name, x);
 556         x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
 557         f->Write(&x, sizeof(x));
 558         WRITE_STRING(f, m_Index[i].m_Page, x);
 559     }
 560     return TRUE;
 561 }
 562
 563
 564 void wxHtmlHelpData::SetTempDir(const wxString& path)
 565 {
 566     if (path == wxEmptyString) m_TempPath = path;
 567     else
 568     {
 569         if (wxIsAbsolutePath(path)) m_TempPath = path;
 570         else m_TempPath = wxGetCwd() + _T("/") + path;
 571
 572         if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
 573             m_TempPath << _T('/');
 574     }
 575 }
 576
 577
 578
 579 static wxString SafeFileName(const wxString& s)
 580 {
 581     wxString res(s);
 582     res.Replace(wxT("#"), wxT("_"));
 583     res.Replace(wxT(":"), wxT("_"));
 584     res.Replace(wxT("\\"), wxT("_"));
 585     res.Replace(wxT("/"), wxT("_"));
 586     return res;
 587 }
 588
 589 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
 590                                   wxFontEncoding encoding,
 591                                   const wxString& title, const wxString& contfile,
 592                                   const wxString& indexfile, const wxString& deftopic,
 593                                   const wxString& path)
 594 {
 595     wxFileSystem fsys;
 596     wxFSFile *fi;
 597     wxHtmlBookRecord *bookr;
 598
 599     int IndexOld = m_IndexCnt,
 600         ContentsOld = m_ContentsCnt;
 601
 602     if (! path.IsEmpty())
 603         fsys.ChangePathTo(path, TRUE);
 604
 605     bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
 606
 607     if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
 608         m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
 609     m_Contents[m_ContentsCnt].m_Level = 0;
 610     m_Contents[m_ContentsCnt].m_ID = 0;
 611     m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
 612     wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
 613     m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
 614     wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
 615     m_Contents[m_ContentsCnt].m_Book = bookr;
 616
 617     // store the contents index for later
 618     int cont_start = m_ContentsCnt++;
 619
 620     // Try to find cached binary versions:
 621     // 1. save file as book, but with .hhp.cached extension
 622     // 2. same as 1. but in temp path
 623     // 3. otherwise or if cache load failed, load it from MS.
 624
 625     fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
 626
 627     if (fi == NULL ||
 628           fi->GetModificationTime() < bookfile.GetModificationTime() ||
 629           !LoadCachedBook(bookr, fi->GetStream()))
 630     {
 631         if (fi != NULL) delete fi;
 632         fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
 633         if (m_TempPath == wxEmptyString || fi == NULL ||
 634             fi->GetModificationTime() < bookfile.GetModificationTime() ||
 635             !LoadCachedBook(bookr, fi->GetStream()))
 636         {
 637             LoadMSProject(bookr, fsys, indexfile, contfile);
 638             if (m_TempPath != wxEmptyString)
 639             {
 640                 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
 641                                                   SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
 642                 SaveCachedBook(bookr, outs);
 643                 delete outs;
 644             }
 645         }
 646     }
 647
 648     if (fi != NULL) delete fi;
 649
 650     // Now store the contents range
 651     bookr->SetContentsRange(cont_start, m_ContentsCnt);
 652
 653     // Convert encoding, if neccessary:
 654     if (encoding != wxFONTENCODING_SYSTEM)
 655     {
 656         wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
 657         if (a.GetCount() != 0 && a[0] != encoding)
 658         {
 659             int i;
 660             wxEncodingConverter conv;
 661             conv.Init(encoding, a[0]);
 662
 663             for (i = IndexOld; i < m_IndexCnt; i++)
 664                 conv.Convert(m_Index[i].m_Name);
 665             for (i = ContentsOld; i < m_ContentsCnt; i++)
 666                 conv.Convert(m_Contents[i].m_Name);
 667         }
 668     }
 669
 670     m_BookRecords.Add(bookr);
 671     if (m_IndexCnt > 0)
 672         qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), IndexCompareFunc);
 673
 674     return TRUE;
 675 }
 676
 677
 678 bool wxHtmlHelpData::AddBook(const wxString& book)
 679 {
 680     if (book.Right(4).Lower() == wxT(".zip") ||
 681         book.Right(4).Lower() == wxT(".htb") /*html book*/)
 682
 683     {
 684         wxFileSystem fsys;
 685         wxString s;
 686         bool rt = FALSE;
 687
 688         s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
 689         while (!s.IsEmpty())
 690         {
 691             if (AddBook(s)) rt = TRUE;
 692             s = fsys.FindNext();
 693         }
 694
 695         return rt;
 696     }
 697
 698
 699     else
 700     {
 701         wxFSFile *fi;
 702         wxFileSystem fsys;
 703         wxInputStream *s;
 704         wxString bookFull;
 705
 706         int sz;
 707         char *buff, *lineptr;
 708         char linebuf[300];
 709
 710         wxString title = _("noname"),
 711                  safetitle,
 712                  start = wxEmptyString,
 713                  contents = wxEmptyString,
 714                  index = wxEmptyString,
 715                  charset = wxEmptyString;
 716
 717         if (wxIsAbsolutePath(book)) bookFull = book;
 718         else bookFull = wxGetCwd() + "/" + book;
 719
 720         fi = fsys.OpenFile(bookFull);
 721         if (fi == NULL)
 722         {
 723             wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
 724             return FALSE;
 725         }
 726         fsys.ChangePathTo(bookFull);
 727         s = fi->GetStream();
 728         sz = s->GetSize();
 729         buff = new char[sz + 1];
 730         buff[sz] = 0;
 731         s->Read(buff, sz);
 732         lineptr = buff;
 733
 734         do {
 735             lineptr = ReadLine(lineptr, linebuf);
 736
 737             if (strstr(linebuf, "Title=") == linebuf)
 738                 title = linebuf + strlen("Title=");
 739             if (strstr(linebuf, "Default topic=") == linebuf)
 740                 start = linebuf + strlen("Default topic=");
 741             if (strstr(linebuf, "Index file=") == linebuf)
 742                 index = linebuf + strlen("Index file=");
 743             if (strstr(linebuf, "Contents file=") == linebuf)
 744                 contents = linebuf + strlen("Contents file=");
 745             if (strstr(linebuf, "Charset=") == linebuf)
 746                 charset = linebuf + strlen("Charset=");
 747         } while (lineptr != NULL);
 748         delete[] buff;
 749
 750         wxFontEncoding enc;
 751         if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
 752         else enc = wxTheFontMapper->CharsetToEncoding(charset);
 753         bool rtval = AddBookParam(*fi, enc,
 754                                   title, contents, index, start, fsys.GetPath());
 755         delete fi;
 756         return rtval;
 757     }
 758 }
 759
 760 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
 761 {
 762     int cnt;
 763     int i;
 764     wxFileSystem fsys;
 765     wxFSFile *f;
 766     wxString url(wxEmptyString);
 767
 768     /* 1. try to open given file: */
 769
 770     cnt = m_BookRecords.GetCount();
 771     for (i = 0; i < cnt; i++)
 772     {
 773         f = fsys.OpenFile(m_BookRecords[i].GetFullPath(x));
 774         if (f)
 775             {
 776             url = m_BookRecords[i].GetFullPath(x);
 777             delete f;
 778             return url;
 779         }
 780     }
 781
 782
 783     /* 2. try to find a book: */
 784
 785     for (i = 0; i < cnt; i++)
 786     {
 787         if (m_BookRecords[i].GetTitle() == x)
 788             {
 789             url = m_BookRecords[i].GetFullPath(m_BookRecords[i].GetStart());
 790             return url;
 791         }
 792     }
 793
 794     /* 3. try to find in contents: */
 795
 796     cnt = m_ContentsCnt;
 797     for (i = 0; i < cnt; i++)
 798     {
 799         if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
 800             {
 801             url = m_Contents[i].GetFullPath();
 802             return url;
 803         }
 804     }
 805
 806
 807     /* 4. try to find in index: */
 808
 809     cnt = m_IndexCnt;
 810     for (i = 0; i < cnt; i++)
 811     {
 812         if (wxStrcmp(m_Index[i].m_Name, x) == 0)
 813             {
 814             url = m_Index[i].GetFullPath();
 815             return url;
 816         }
 817     }
 818
 819     return url;
 820 }
 821
 822 wxString wxHtmlHelpData::FindPageById(int id)
 823 {
 824     int i;
 825     wxString url(wxEmptyString);
 826
 827     for (i = 0; i < m_ContentsCnt; i++)
 828     {
 829         if (m_Contents[i].m_ID == id)
 830             {
 831             url = m_Contents[i].GetFullPath();
 832             return url;
 833         }
 834     }
 835
 836     return url;
 837 }
 838
 839 //----------------------------------------------------------------------------------
 840 // wxHtmlSearchStatus functions
 841 //----------------------------------------------------------------------------------
 842
 843 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
 844                                        bool case_sensitive, bool whole_words_only,
 845                                        const wxString& book)
 846 {
 847     m_Data = data;
 848     m_Keyword = keyword;
 849     wxHtmlBookRecord* bookr = NULL;
 850     if (book != wxEmptyString)
 851     {
 852         // we have to search in a specific book. Find it first
 853         int i, cnt = data->m_BookRecords.GetCount();
 854         for (i = 0; i < cnt; i++)
 855             if (data->m_BookRecords[i].GetTitle() == book)
 856                 {
 857                 bookr = &(data->m_BookRecords[i]);
 858                 m_CurIndex = bookr->GetContentsStart();
 859                 m_MaxIndex = bookr->GetContentsEnd();
 860                 break;
 861             }
 862         // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
 863         wxASSERT(bookr);
 864     }
 865     if (! bookr)
 866     {
 867         // no book specified; search all books
 868         m_CurIndex = 0;
 869         m_MaxIndex = m_Data->m_ContentsCnt;
 870     }
 871     m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
 872     m_Active = (m_CurIndex < m_MaxIndex);
 873     m_LastPage = NULL;
 874 }
 875
 876 bool wxHtmlSearchStatus::Search()
 877 {
 878     wxFSFile *file;
 879     int i = m_CurIndex;  // shortcut
 880     bool found = FALSE;
 881     wxChar *thepage;
 882
 883     if (!m_Active)
 884     {
 885         // sanity check. Illegal use, but we'll try to prevent a crash anyway
 886         wxASSERT(m_Active);
 887         return FALSE;
 888     }
 889
 890     m_Name = wxEmptyString;
 891     m_ContentsItem = NULL;
 892     thepage = m_Data->m_Contents[i].m_Page;
 893
 894     m_Active = (++m_CurIndex < m_MaxIndex);
 895     // check if it is same page with different anchor:
 896     if (m_LastPage != NULL)
 897     {
 898         wxChar *p1, *p2;
 899         for (p1 = thepage, p2 = m_LastPage;
 900              *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
 901
 902         m_LastPage = thepage;
 903
 904         if (*p1 == 0 || *p1 == _T('#'))
 905             return FALSE;
 906     }
 907     else m_LastPage = thepage;
 908
 909     wxFileSystem fsys;
 910     file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetFullPath(thepage));
 911     if (file)
 912     {
 913         if (m_Engine.Scan(file->GetStream()))
 914         {
 915             m_Name = m_Data->m_Contents[i].m_Name;
 916             m_ContentsItem = m_Data->m_Contents + i;
 917             found = TRUE;
 918         }
 919         delete file;
 920     }
 921     return found;
 922 }
 923
 924
 925
 926
 927
 928
 929
 930
 931 //--------------------------------------------------------------------------------
 932 // wxSearchEngine
 933 //--------------------------------------------------------------------------------
 934
 935 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
 936 {
 937     m_CaseSensitive = case_sensitive;
 938     m_WholeWords = whole_words_only;
 939     if (m_Keyword) delete[] m_Keyword;
 940     m_Keyword = new wxChar[keyword.Length() + 1];
 941     wxStrcpy(m_Keyword, keyword.c_str());
 942
 943     if (!m_CaseSensitive)
 944     {
 945         for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
 946             {
 947             if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
 948                 m_Keyword[i] += wxT('a') - wxT('A');
 949                 }
 950         }
 951 }
 952
 953
 954
 955 #define WHITESPACE(c)  (c == ' ' || c == '\n' || c == '\r' || c == '\t')
 956
 957 bool wxSearchEngine::Scan(wxInputStream *stream)
 958 {
 959     wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
 960
 961     int i, j;
 962     int lng = stream ->GetSize();
 963     int wrd = wxStrlen(m_Keyword);
 964     bool found = FALSE;
 965     char *buf = new char[lng + 1];
 966     stream->Read(buf, lng);
 967     buf[lng] = 0;
 968
 969     if (!m_CaseSensitive)
 970         for (i = 0; i < lng; i++)
 971             if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
 972
 973     if (m_WholeWords)
 974     {
 975         for (i = 0; i < lng - wrd; i++)
 976             {
 977             if (WHITESPACE(buf[i])) continue;
 978             j = 0;
 979             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 980             if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
 981         }
 982     }
 983
 984     else
 985     {
 986         for (i = 0; i < lng - wrd; i++)
 987             {
 988             j = 0;
 989             while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
 990             if (j == wrd) { found = TRUE; break; }
 991         }
 992     }
 993
 994     delete[] buf;
 995     return found;
 996 }
 997
 998
 999
1000 #endif