case insensitive HHP files and fixed buffer overflow vulnerability
[wxWidgets.git] / 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/intl.h"
29 #include "wx/log.h"
30 #endif
31
32 #include "wx/html/helpdata.h"
33 #include "wx/tokenzr.h"
34 #include "wx/wfstream.h"
35 #include "wx/busyinfo.h"
36 #include "wx/encconv.h"
37 #include "wx/fontmap.h"
38 #include "wx/log.h"
39 #include "wx/html/htmlpars.h"
40 #include "wx/html/htmldefs.h"
41 #include "wx/filename.h"
42
43 #include "wx/arrimpl.cpp"
44 WX_DEFINE_OBJARRAY(wxHtmlBookRecArray)
45
46 //-----------------------------------------------------------------------------
47 // static helper functions
48 //-----------------------------------------------------------------------------
49
50 // Reads one line, stores it into buf and returns pointer to new line or NULL.
51 static char* ReadLine(char *line, char *buf, size_t bufsize)
52 {
53 char *writeptr = buf;
54 char *endptr = buf + bufsize - 1;
55 char *readptr = line;
56
57 while (*readptr != 0 && *readptr != '\r' && *readptr != '\n' &&
58 writeptr != endptr)
59 *(writeptr++) = *(readptr++);
60 *writeptr = 0;
61 while (*readptr == '\r' || *readptr == '\n')
62 readptr++;
63 if (*readptr == 0)
64 return NULL;
65 else
66 return readptr;
67 }
68
69
70
71 extern "C" int LINKAGEMODE
72 wxHtmlHelpIndexCompareFunc(const void *a, const void *b)
73 {
74 return wxStricmp(((wxHtmlContentsItem*)a)->m_Name, ((wxHtmlContentsItem*)b)->m_Name);
75 }
76
77
78 //-----------------------------------------------------------------------------
79 // HP_Parser
80 //-----------------------------------------------------------------------------
81
82 class HP_Parser : public wxHtmlParser
83 {
84 public:
85 wxObject* GetProduct() { return NULL; }
86 protected:
87 virtual void AddText(const wxChar* WXUNUSED(txt)) {}
88 };
89
90
91 //-----------------------------------------------------------------------------
92 // HP_TagHandler
93 //-----------------------------------------------------------------------------
94
95 class HP_TagHandler : public wxHtmlTagHandler
96 {
97 private:
98 wxString m_Name, m_Page;
99 int m_Level;
100 int m_ID;
101 int m_Index;
102 wxHtmlContentsItem *m_Items;
103 int m_ItemsCnt;
104 wxHtmlBookRecord *m_Book;
105
106 public:
107 HP_TagHandler(wxHtmlBookRecord *b) : wxHtmlTagHandler()
108 { m_Book = b; m_Items = NULL; m_ItemsCnt = 0; m_Name = m_Page = wxEmptyString;
109 m_Level = 0; m_ID = -1; }
110 wxString GetSupportedTags() { return wxT("UL,OBJECT,PARAM"); }
111 bool HandleTag(const wxHtmlTag& tag);
112 void WriteOut(wxHtmlContentsItem*& array, int& size);
113 void ReadIn(wxHtmlContentsItem* array, int size);
114 };
115
116
117 bool HP_TagHandler::HandleTag(const wxHtmlTag& tag)
118 {
119 if (tag.GetName() == wxT("UL"))
120 {
121 m_Level++;
122 ParseInner(tag);
123 m_Level--;
124 return TRUE;
125 }
126 else if (tag.GetName() == wxT("OBJECT"))
127 {
128 m_Name = m_Page = wxEmptyString;
129 ParseInner(tag);
130
131 #if 0
132 if (!m_Page.IsEmpty())
133 /* Valid HHW's file may contain only two object tags:
134
135 <OBJECT type="text/site properties">
136 <param name="ImageType" value="Folder">
137 </OBJECT>
138
139 or
140
141 <OBJECT type="text/sitemap">
142 <param name="Name" value="main page">
143 <param name="Local" value="another.htm">
144 </OBJECT>
145
146 We're interested in the latter. !m_Page.IsEmpty() is valid
147 condition because text/site properties does not contain Local param
148 */
149 #endif
150 if (tag.GetParam(wxT("TYPE")) == wxT("text/sitemap"))
151 {
152 if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0)
153 m_Items = (wxHtmlContentsItem*) realloc(m_Items,
154 (m_ItemsCnt + wxHTML_REALLOC_STEP) *
155 sizeof(wxHtmlContentsItem));
156
157 m_Items[m_ItemsCnt].m_Level = m_Level;
158 m_Items[m_ItemsCnt].m_ID = m_ID;
159 m_Items[m_ItemsCnt].m_Page = new wxChar[m_Page.Length() + 1];
160 wxStrcpy(m_Items[m_ItemsCnt].m_Page, m_Page.c_str());
161 m_Items[m_ItemsCnt].m_Name = new wxChar [m_Name.Length() + 1];
162 wxStrcpy(m_Items[m_ItemsCnt].m_Name, m_Name.c_str());
163 m_Items[m_ItemsCnt].m_Book = m_Book;
164 m_ItemsCnt++;
165 }
166
167 return TRUE;
168 }
169 else
170 { // "PARAM"
171 if (m_Name == wxEmptyString && tag.GetParam(wxT("NAME")) == wxT("Name"))
172 m_Name = tag.GetParam(wxT("VALUE"));
173 if (tag.GetParam(wxT("NAME")) == wxT("Local"))
174 m_Page = tag.GetParam(wxT("VALUE"));
175 if (tag.GetParam(wxT("NAME")) == wxT("ID"))
176 tag.GetParamAsInt(wxT("VALUE"), &m_ID);
177 return FALSE;
178 }
179 }
180
181
182
183 void HP_TagHandler::WriteOut(wxHtmlContentsItem*& array, int& size)
184 {
185 array = m_Items;
186 size = m_ItemsCnt;
187 m_Items = NULL;
188 m_ItemsCnt = 0;
189 }
190
191 void HP_TagHandler::ReadIn(wxHtmlContentsItem* array, int size)
192 {
193 m_Items = array;
194 m_ItemsCnt = size;
195 }
196
197
198
199
200 //-----------------------------------------------------------------------------
201 // wxHtmlHelpData
202 //-----------------------------------------------------------------------------
203
204 wxString wxHtmlBookRecord::GetFullPath(const wxString &page) const
205 {
206 if (wxIsAbsolutePath(page))
207 return page;
208 else
209 return m_BasePath + page;
210 }
211
212
213
214 IMPLEMENT_DYNAMIC_CLASS(wxHtmlHelpData, wxObject)
215
216 wxHtmlHelpData::wxHtmlHelpData()
217 {
218 m_TempPath = wxEmptyString;
219
220 m_Contents = NULL;
221 m_ContentsCnt = 0;
222 m_Index = NULL;
223 m_IndexCnt = 0;
224 }
225
226 wxHtmlHelpData::~wxHtmlHelpData()
227 {
228 int i;
229
230 m_BookRecords.Empty();
231 if (m_Contents)
232 {
233 for (i = 0; i < m_ContentsCnt; i++)
234 {
235 delete[] m_Contents[i].m_Page;
236 delete[] m_Contents[i].m_Name;
237 }
238 free(m_Contents);
239 }
240 if (m_Index)
241 {
242 for (i = 0; i < m_IndexCnt; i++)
243 {
244 delete[] m_Index[i].m_Page;
245 delete[] m_Index[i].m_Name;
246 }
247 free(m_Index);
248 }
249 }
250
251 bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, const wxString& indexfile, const wxString& contentsfile)
252 {
253 wxFSFile *f;
254 char *buf;
255 int sz;
256 wxString string;
257
258 HP_Parser parser;
259 HP_TagHandler *handler = new HP_TagHandler(book);
260 parser.AddTagHandler(handler);
261
262 f = ( contentsfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(contentsfile) );
263 if (f)
264 {
265 sz = f->GetStream()->GetSize();
266 buf = new char[sz + 1];
267 buf[sz] = 0;
268 f->GetStream()->Read(buf, sz);
269 delete f;
270 handler->ReadIn(m_Contents, m_ContentsCnt);
271 parser.Parse(buf);
272 handler->WriteOut(m_Contents, m_ContentsCnt);
273 delete[] buf;
274 }
275 else
276 wxLogError(_("Cannot open contents file: %s"), contentsfile.c_str());
277
278 f = ( indexfile.IsEmpty() ? (wxFSFile*) NULL : fsys.OpenFile(indexfile) );
279 if (f)
280 {
281 sz = f->GetStream()->GetSize();
282 buf = new char[sz + 1];
283 buf[sz] = 0;
284 f->GetStream()->Read(buf, sz);
285 delete f;
286 handler->ReadIn(m_Index, m_IndexCnt);
287 parser.Parse(buf);
288 handler->WriteOut(m_Index, m_IndexCnt);
289 delete[] buf;
290 }
291 else if (!indexfile.IsEmpty())
292 wxLogError(_("Cannot open index file: %s"), indexfile.c_str());
293 return TRUE;
294 }
295
296
297
298
299 #if wxUSE_UNICODE
300
301 #define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f->Read(&tmpc, 1); s[i] = (wxChar)tmpc;} }
302 #define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f->Write(&tmpc, 1);} }
303
304 #else
305
306 #define READ_STRING(f, s, lng) f->Read(s, lng * sizeof(char));
307 #define WRITE_STRING(f, s, lng) f->Write(s, lng * sizeof(char));
308
309 #endif
310
311
312 #define CURRENT_CACHED_BOOK_VERSION 1
313
314 bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f)
315 {
316 int i, st;
317 wxInt32 x;
318 wxInt32 version;
319
320 /* load header - version info : */
321
322 f->Read(&x, sizeof(x));
323 version = wxINT32_SWAP_ON_BE(x);
324
325 if (version != CURRENT_CACHED_BOOK_VERSION)
326 {
327 wxLogError(_("Incorrect version of HTML help book"));
328 return FALSE;
329 // NOTE: when adding new version, please ensure backward compatibility!
330 }
331
332 /* load contents : */
333
334 f->Read(&x, sizeof(x));
335 st = m_ContentsCnt;
336 m_ContentsCnt += wxINT32_SWAP_ON_BE(x);
337 m_Contents = (wxHtmlContentsItem*) realloc(m_Contents,
338 (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) *
339 wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
340 for (i = st; i < m_ContentsCnt; i++)
341 {
342 f->Read(&x, sizeof(x));
343 m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x);
344 f->Read(&x, sizeof(x));
345 m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x);
346 f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
347 m_Contents[i].m_Name = new wxChar[x];
348 READ_STRING(f, m_Contents[i].m_Name, x);
349 f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
350 m_Contents[i].m_Page = new wxChar[x];
351 READ_STRING(f, m_Contents[i].m_Page, x);
352 m_Contents[i].m_Book = book;
353 }
354
355 /* load index : */
356
357 f->Read(&x, sizeof(x));
358 st = m_IndexCnt;
359 m_IndexCnt += wxINT32_SWAP_ON_BE(x);
360 m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) *
361 wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem));
362 for (i = st; i < m_IndexCnt; i++)
363 {
364 f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
365 m_Index[i].m_Name = new wxChar[x];
366 READ_STRING(f, m_Index[i].m_Name, x);
367 f->Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x);
368 m_Index[i].m_Page = new wxChar[x];
369 READ_STRING(f, m_Index[i].m_Page, x);
370 m_Index[i].m_Book = book;
371 }
372 return TRUE;
373 }
374
375
376 bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f)
377 {
378 int i;
379 wxInt32 x;
380
381 /* save header - version info : */
382
383 x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION);
384 f->Write(&x, sizeof(x));
385
386 /* save contents : */
387
388 x = 0;
389 for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++;
390 x = wxINT32_SWAP_ON_BE(x);
391 f->Write(&x, sizeof(x));
392 for (i = 0; i < m_ContentsCnt; i++)
393 {
394 if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue;
395 x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level);
396 f->Write(&x, sizeof(x));
397 x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID);
398 f->Write(&x, sizeof(x));
399 x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1);
400 f->Write(&x, sizeof(x));
401 WRITE_STRING(f, m_Contents[i].m_Name, x);
402 x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1);
403 f->Write(&x, sizeof(x));
404 WRITE_STRING(f, m_Contents[i].m_Page, x);
405 }
406
407 /* save index : */
408
409 x = 0;
410 for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++;
411 x = wxINT32_SWAP_ON_BE(x);
412 f->Write(&x, sizeof(x));
413 for (i = 0; i < m_IndexCnt; i++)
414 {
415 if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue;
416 x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1);
417 f->Write(&x, sizeof(x));
418 WRITE_STRING(f, m_Index[i].m_Name, x);
419 x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1);
420 f->Write(&x, sizeof(x));
421 WRITE_STRING(f, m_Index[i].m_Page, x);
422 }
423 return TRUE;
424 }
425
426
427 void wxHtmlHelpData::SetTempDir(const wxString& path)
428 {
429 if (path == wxEmptyString) m_TempPath = path;
430 else
431 {
432 if (wxIsAbsolutePath(path)) m_TempPath = path;
433 else m_TempPath = wxGetCwd() + _T("/") + path;
434
435 if (m_TempPath[m_TempPath.Length() - 1] != _T('/'))
436 m_TempPath << _T('/');
437 }
438 }
439
440
441
442 static wxString SafeFileName(const wxString& s)
443 {
444 wxString res(s);
445 res.Replace(wxT("#"), wxT("_"));
446 res.Replace(wxT(":"), wxT("_"));
447 res.Replace(wxT("\\"), wxT("_"));
448 res.Replace(wxT("/"), wxT("_"));
449 return res;
450 }
451
452 bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile,
453 wxFontEncoding encoding,
454 const wxString& title, const wxString& contfile,
455 const wxString& indexfile, const wxString& deftopic,
456 const wxString& path)
457 {
458 wxFileSystem fsys;
459 wxFSFile *fi;
460 wxHtmlBookRecord *bookr;
461
462 int IndexOld = m_IndexCnt,
463 ContentsOld = m_ContentsCnt;
464
465 if (! path.IsEmpty())
466 fsys.ChangePathTo(path, TRUE);
467
468 bookr = new wxHtmlBookRecord(fsys.GetPath(), title, deftopic);
469
470 if (m_ContentsCnt % wxHTML_REALLOC_STEP == 0)
471 m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem));
472 m_Contents[m_ContentsCnt].m_Level = 0;
473 m_Contents[m_ContentsCnt].m_ID = 0;
474 m_Contents[m_ContentsCnt].m_Page = new wxChar[deftopic.Length() + 1];
475 wxStrcpy(m_Contents[m_ContentsCnt].m_Page, deftopic.c_str());
476 m_Contents[m_ContentsCnt].m_Name = new wxChar [title.Length() + 1];
477 wxStrcpy(m_Contents[m_ContentsCnt].m_Name, title.c_str());
478 m_Contents[m_ContentsCnt].m_Book = bookr;
479
480 // store the contents index for later
481 int cont_start = m_ContentsCnt++;
482
483 // Try to find cached binary versions:
484 // 1. save file as book, but with .hhp.cached extension
485 // 2. same as 1. but in temp path
486 // 3. otherwise or if cache load failed, load it from MS.
487
488 fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached"));
489
490 if (fi == NULL ||
491 fi->GetModificationTime() < bookfile.GetModificationTime() ||
492 !LoadCachedBook(bookr, fi->GetStream()))
493 {
494 if (fi != NULL) delete fi;
495 fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached"));
496 if (m_TempPath == wxEmptyString || fi == NULL ||
497 fi->GetModificationTime() < bookfile.GetModificationTime() ||
498 !LoadCachedBook(bookr, fi->GetStream()))
499 {
500 LoadMSProject(bookr, fsys, indexfile, contfile);
501 if (m_TempPath != wxEmptyString)
502 {
503 wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath +
504 SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached"));
505 SaveCachedBook(bookr, outs);
506 delete outs;
507 }
508 }
509 }
510
511 if (fi != NULL) delete fi;
512
513 // Now store the contents range
514 bookr->SetContentsRange(cont_start, m_ContentsCnt);
515
516 // Convert encoding, if neccessary:
517 if (encoding != wxFONTENCODING_SYSTEM)
518 {
519 wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding);
520 if (a.GetCount() != 0 && a[0] != encoding)
521 {
522 int i;
523 wxEncodingConverter conv;
524 conv.Init(encoding, a[0]);
525
526 for (i = IndexOld; i < m_IndexCnt; i++)
527 conv.Convert(m_Index[i].m_Name);
528 for (i = ContentsOld; i < m_ContentsCnt; i++)
529 conv.Convert(m_Contents[i].m_Name);
530 }
531 }
532
533 m_BookRecords.Add(bookr);
534 if (m_IndexCnt > 0)
535 qsort(m_Index, m_IndexCnt, sizeof(wxHtmlContentsItem), wxHtmlHelpIndexCompareFunc);
536
537 return TRUE;
538 }
539
540
541 bool wxHtmlHelpData::AddBook(const wxString& book)
542 {
543 if (book.Right(4).Lower() == wxT(".zip") ||
544 book.Right(4).Lower() == wxT(".htb") /*html book*/)
545
546 {
547 wxFileSystem fsys;
548 wxString s;
549 bool rt = FALSE;
550
551 s = fsys.FindFirst(book + wxT("#zip:") + wxT("*.hhp"), wxFILE);
552 while (!s.IsEmpty())
553 {
554 if (AddBook(s)) rt = TRUE;
555 s = fsys.FindNext();
556 }
557
558 return rt;
559 }
560
561
562 else
563 {
564 wxFSFile *fi;
565 wxFileSystem fsys;
566 wxInputStream *s;
567 wxString bookFull;
568
569 wxString title = _("noname"),
570 safetitle,
571 start = wxEmptyString,
572 contents = wxEmptyString,
573 index = wxEmptyString,
574 charset = wxEmptyString;
575
576 #if defined(__WXMAC__) && !defined(__DARWIN__)
577 if (wxIsAbsolutePath(book)) bookFull = book;
578 else bookFull = wxGetCwd() + book; // no slash or dot
579 wxFileName fn( bookFull );
580 bookFull = fn.GetFullPath( wxPATH_UNIX );
581 #else
582 if (wxIsAbsolutePath(book)) bookFull = book;
583 else bookFull = wxGetCwd() + "/" + book;
584 #endif
585
586 fi = fsys.OpenFile(bookFull);
587 if (fi == NULL)
588 {
589 wxLogError(_("Cannot open HTML help book: %s"), bookFull.c_str());
590 return FALSE;
591 }
592 fsys.ChangePathTo(bookFull);
593 s = fi->GetStream();
594
595 int sz;
596 char *buff, *lineptr;
597 char linebuf[300];
598
599 sz = s->GetSize();
600 buff = new char[sz + 1];
601 buff[sz] = 0;
602 s->Read(buff, sz);
603 lineptr = buff;
604
605 do
606 {
607 lineptr = ReadLine(lineptr, linebuf, 300);
608
609 for (char *ch = linebuf; *ch != '\0' && *ch != '='; ch++)
610 *ch = tolower(*ch);
611
612 if (strstr(linebuf, "title=") == linebuf)
613 title = linebuf + strlen("title=");
614 if (strstr(linebuf, "default topic=") == linebuf)
615 start = linebuf + strlen("default topic=");
616 if (strstr(linebuf, "index file=") == linebuf)
617 index = linebuf + strlen("index file=");
618 if (strstr(linebuf, "contents file=") == linebuf)
619 contents = linebuf + strlen("contents file=");
620 if (strstr(linebuf, "charset=") == linebuf)
621 charset = linebuf + strlen("charset=");
622 } while (lineptr != NULL);
623 delete[] buff;
624
625 wxFontEncoding enc;
626 if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM;
627 else enc = wxTheFontMapper->CharsetToEncoding(charset);
628 bool rtval = AddBookParam(*fi, enc,
629 title, contents, index, start, fsys.GetPath());
630 delete fi;
631 return rtval;
632 }
633 }
634
635 wxString wxHtmlHelpData::FindPageByName(const wxString& x)
636 {
637 int cnt;
638 int i;
639 wxFileSystem fsys;
640 wxFSFile *f;
641 wxString url(wxEmptyString);
642
643 /* 1. try to open given file: */
644
645 cnt = m_BookRecords.GetCount();
646 for (i = 0; i < cnt; i++)
647 {
648 f = fsys.OpenFile(m_BookRecords[i].GetFullPath(x));
649 if (f)
650 {
651 url = m_BookRecords[i].GetFullPath(x);
652 delete f;
653 return url;
654 }
655 }
656
657
658 /* 2. try to find a book: */
659
660 for (i = 0; i < cnt; i++)
661 {
662 if (m_BookRecords[i].GetTitle() == x)
663 {
664 url = m_BookRecords[i].GetFullPath(m_BookRecords[i].GetStart());
665 return url;
666 }
667 }
668
669 /* 3. try to find in contents: */
670
671 cnt = m_ContentsCnt;
672 for (i = 0; i < cnt; i++)
673 {
674 if (wxStrcmp(m_Contents[i].m_Name, x) == 0)
675 {
676 url = m_Contents[i].GetFullPath();
677 return url;
678 }
679 }
680
681
682 /* 4. try to find in index: */
683
684 cnt = m_IndexCnt;
685 for (i = 0; i < cnt; i++)
686 {
687 if (wxStrcmp(m_Index[i].m_Name, x) == 0)
688 {
689 url = m_Index[i].GetFullPath();
690 return url;
691 }
692 }
693
694 return url;
695 }
696
697 wxString wxHtmlHelpData::FindPageById(int id)
698 {
699 int i;
700 wxString url(wxEmptyString);
701
702 for (i = 0; i < m_ContentsCnt; i++)
703 {
704 if (m_Contents[i].m_ID == id)
705 {
706 url = m_Contents[i].GetFullPath();
707 return url;
708 }
709 }
710
711 return url;
712 }
713
714 //----------------------------------------------------------------------------------
715 // wxHtmlSearchStatus functions
716 //----------------------------------------------------------------------------------
717
718 wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword,
719 bool case_sensitive, bool whole_words_only,
720 const wxString& book)
721 {
722 m_Data = data;
723 m_Keyword = keyword;
724 wxHtmlBookRecord* bookr = NULL;
725 if (book != wxEmptyString)
726 {
727 // we have to search in a specific book. Find it first
728 int i, cnt = data->m_BookRecords.GetCount();
729 for (i = 0; i < cnt; i++)
730 if (data->m_BookRecords[i].GetTitle() == book)
731 {
732 bookr = &(data->m_BookRecords[i]);
733 m_CurIndex = bookr->GetContentsStart();
734 m_MaxIndex = bookr->GetContentsEnd();
735 break;
736 }
737 // check; we won't crash if the book doesn't exist, but it's Bad Anyway.
738 wxASSERT(bookr);
739 }
740 if (! bookr)
741 {
742 // no book specified; search all books
743 m_CurIndex = 0;
744 m_MaxIndex = m_Data->m_ContentsCnt;
745 }
746 m_Engine.LookFor(keyword, case_sensitive, whole_words_only);
747 m_Active = (m_CurIndex < m_MaxIndex);
748 m_LastPage = NULL;
749 }
750
751 bool wxHtmlSearchStatus::Search()
752 {
753 wxFSFile *file;
754 int i = m_CurIndex; // shortcut
755 bool found = FALSE;
756 wxChar *thepage;
757
758 if (!m_Active)
759 {
760 // sanity check. Illegal use, but we'll try to prevent a crash anyway
761 wxASSERT(m_Active);
762 return FALSE;
763 }
764
765 m_Name = wxEmptyString;
766 m_ContentsItem = NULL;
767 thepage = m_Data->m_Contents[i].m_Page;
768
769 m_Active = (++m_CurIndex < m_MaxIndex);
770 // check if it is same page with different anchor:
771 if (m_LastPage != NULL)
772 {
773 wxChar *p1, *p2;
774 for (p1 = thepage, p2 = m_LastPage;
775 *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {}
776
777 m_LastPage = thepage;
778
779 if (*p1 == 0 || *p1 == _T('#'))
780 return FALSE;
781 }
782 else m_LastPage = thepage;
783
784 wxFileSystem fsys;
785 file = fsys.OpenFile(m_Data->m_Contents[i].m_Book->GetFullPath(thepage));
786 if (file)
787 {
788 if (m_Engine.Scan(file->GetStream()))
789 {
790 m_Name = m_Data->m_Contents[i].m_Name;
791 m_ContentsItem = m_Data->m_Contents + i;
792 found = TRUE;
793 }
794 delete file;
795 }
796 return found;
797 }
798
799
800
801
802
803
804
805
806 //--------------------------------------------------------------------------------
807 // wxSearchEngine
808 //--------------------------------------------------------------------------------
809
810 void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only)
811 {
812 m_CaseSensitive = case_sensitive;
813 m_WholeWords = whole_words_only;
814 if (m_Keyword) delete[] m_Keyword;
815 m_Keyword = new wxChar[keyword.Length() + 1];
816 wxStrcpy(m_Keyword, keyword.c_str());
817
818 if (!m_CaseSensitive)
819 {
820 for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--)
821 {
822 if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z')))
823 m_Keyword[i] += wxT('a') - wxT('A');
824 }
825 }
826 }
827
828
829
830 #define WHITESPACE(c) (c == ' ' || c == '\n' || c == '\r' || c == '\t')
831
832 bool wxSearchEngine::Scan(wxInputStream *stream)
833 {
834 wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!"));
835
836 int i, j;
837 int lng = stream ->GetSize();
838 int wrd = wxStrlen(m_Keyword);
839 bool found = FALSE;
840 char *buf = new char[lng + 1];
841 stream->Read(buf, lng);
842 buf[lng] = 0;
843
844 if (!m_CaseSensitive)
845 for (i = 0; i < lng; i++)
846 if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A';
847
848 if (m_WholeWords)
849 {
850 for (i = 0; i < lng - wrd; i++)
851 {
852 if (WHITESPACE(buf[i])) continue;
853 j = 0;
854 while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
855 if (j == wrd && WHITESPACE(buf[i + j])) { found = TRUE; break; }
856 }
857 }
858
859 else
860 {
861 for (i = 0; i < lng - wrd; i++)
862 {
863 j = 0;
864 while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++;
865 if (j == wrd) { found = TRUE; break; }
866 }
867 }
868
869 delete[] buf;
870 return found;
871 }
872
873
874
875 #endif