X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/683646599939f5113215254870b15d03e830458e..0158f58424124c608703e1325f623a3473f3baa6:/src/html/helpdata.cpp?ds=sidebyside diff --git a/src/html/helpdata.cpp b/src/html/helpdata.cpp index d4bf792b8f..7c3616de5f 100644 --- a/src/html/helpdata.cpp +++ b/src/html/helpdata.cpp @@ -32,6 +32,9 @@ #include "wx/tokenzr.h" #include "wx/wfstream.h" #include "wx/busyinfo.h" +#include "wx/encconv.h" +#include "wx/fontmap.h" +#include "wx/log.h" #include "wx/html/htmlpars.h" #include "wx/html/htmldefs.h" @@ -55,18 +58,6 @@ static char* ReadLine(char *line, char *buf) } -static wxString SafeFileName(const wxString& s) -{ - wxString res = s; - res.Replace(wxT(":"), wxT("_"), TRUE); - res.Replace(wxT(" "), wxT("_"), TRUE); - res.Replace(wxT("/"), wxT("_"), TRUE); - res.Replace(wxT("\\"), wxT("_"), TRUE); - res.Replace(wxT("#"), wxT("_"), TRUE); - res.Replace(wxT("."), wxT("_"), TRUE); - return res; -} - static int LINKAGEMODE IndexCompareFunc(const void *a, const void *b) { @@ -112,7 +103,7 @@ class HP_TagHandler : public wxHtmlTagHandler bool HP_TagHandler::HandleTag(const wxHtmlTag& tag) { - if (tag.GetName() == "UL") { + if (tag.GetName() == wxT("UL")) { m_Level++; ParseInner(tag); m_Level--; @@ -121,7 +112,27 @@ bool HP_TagHandler::HandleTag(const wxHtmlTag& tag) else if (tag.GetName() == wxT("OBJECT")) { m_Name = m_Page = wxEmptyString; ParseInner(tag); - if (m_Page != wxEmptyString) { + + if (!m_Page.IsEmpty()) + /* should be 'if (tag.GetParam("TYPE") == "text/sitemap")' + but this works fine. Valid HHW's file may contain only two + object tags: + + + + + + or + + + + + + + We're interested in the latter. !m_Page.IsEmpty() is valid + condition because text/site properties does not contain Local param + */ + { if (m_ItemsCnt % wxHTML_REALLOC_STEP == 0) m_Items = (wxHtmlContentsItem*) realloc(m_Items, (m_ItemsCnt + wxHTML_REALLOC_STEP) * sizeof(wxHtmlContentsItem)); m_Items[m_ItemsCnt].m_Level = m_Level; @@ -133,6 +144,7 @@ bool HP_TagHandler::HandleTag(const wxHtmlTag& tag) m_Items[m_ItemsCnt].m_Book = m_Book; m_ItemsCnt++; } + return TRUE; } else { // "PARAM" @@ -210,7 +222,7 @@ bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, c HP_TagHandler *handler = new HP_TagHandler(book); parser.AddTagHandler(handler); - f = ( contentsfile.IsEmpty() ? 0 : fsys.OpenFile(contentsfile) ); + f = ( contentsfile.IsEmpty() ? NULL : fsys.OpenFile(contentsfile) ); if (f) { sz = f -> GetStream() -> GetSize(); buf = new char[sz + 1]; @@ -222,8 +234,10 @@ bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, c handler -> WriteOut(m_Contents, m_ContentsCnt); delete[] buf; } + else + wxLogError(_("Cannot open contents file: %s"), contentsfile.mb_str()); - f = ( indexfile.IsEmpty() ? 0 : fsys.OpenFile(indexfile) ); + f = ( indexfile.IsEmpty() ? NULL : fsys.OpenFile(indexfile) ); if (f) { sz = f -> GetStream() -> GetSize(); buf = new char[sz + 1]; @@ -235,32 +249,66 @@ bool wxHtmlHelpData::LoadMSProject(wxHtmlBookRecord *book, wxFileSystem& fsys, c handler -> WriteOut(m_Index, m_IndexCnt); delete[] buf; } + else if (!indexfile.IsEmpty()) + wxLogError(_("Cannot open index file: %s"), indexfile.mb_str()); return TRUE; } + + +#if wxUSE_UNICODE + +#define READ_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { f -> Read(&tmpc, 1); s[i] = (wxChar)tmpc;} } +#define WRITE_STRING(f, s, lng) { char tmpc; for (int i = 0; i < lng; i++) { tmpc = (char)s[i]; f -> Write(&tmpc, 1);} } + +#else + +#define READ_STRING(f, s, lng) f -> Read(s, lng * sizeof(char)); +#define WRITE_STRING(f, s, lng) f -> Write(s, lng * sizeof(char)); + +#endif + + +#define CURRENT_CACHED_BOOK_VERSION 1 + bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f) { int i, st; - int x; + wxInt32 x; + wxInt32 version; + + /* load header - version info : */ + f -> Read(&x, sizeof(x)); + version = wxINT32_SWAP_ON_BE(x); + + if (version != CURRENT_CACHED_BOOK_VERSION) + { + wxLogError(_("Incorrect version of HTML help book")); + return FALSE; + // NOTE: when adding new version, please ensure backward compatibility! + } + /* load contents : */ f -> Read(&x, sizeof(x)); st = m_ContentsCnt; - m_ContentsCnt += x; - m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) * wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem)); + m_ContentsCnt += wxINT32_SWAP_ON_BE(x); + m_Contents = (wxHtmlContentsItem*) realloc(m_Contents, + (m_ContentsCnt / wxHTML_REALLOC_STEP + 1) * + wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem)); for (i = st; i < m_ContentsCnt; i++) { f -> Read(&x, sizeof(x)); - m_Contents[i].m_Level = x; - f -> Read(&x, sizeof(x)); - m_Contents[i].m_ID = x; + m_Contents[i].m_Level = wxINT32_SWAP_ON_BE(x); f -> Read(&x, sizeof(x)); + m_Contents[i].m_ID = wxINT32_SWAP_ON_BE(x); + f -> Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x); m_Contents[i].m_Name = new wxChar[x]; - f -> Read(m_Contents[i].m_Name, x*sizeof(wxChar)); - f -> Read(&x, sizeof(x)); + READ_STRING(f, m_Contents[i].m_Name, x); + f -> Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x); m_Contents[i].m_Page = new wxChar[x]; - f -> Read(m_Contents[i].m_Page, x*sizeof(wxChar)); + READ_STRING(f, m_Contents[i].m_Page, x); m_Contents[i].m_Book = book; } @@ -268,15 +316,16 @@ bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f) f -> Read(&x, sizeof(x)); st = m_IndexCnt; - m_IndexCnt += x; - m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) * wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem)); + m_IndexCnt += wxINT32_SWAP_ON_BE(x); + m_Index = (wxHtmlContentsItem*) realloc(m_Index, (m_IndexCnt / wxHTML_REALLOC_STEP + 1) * + wxHTML_REALLOC_STEP * sizeof(wxHtmlContentsItem)); for (i = st; i < m_IndexCnt; i++) { - f -> Read(&x, sizeof(x)); + f -> Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x); m_Index[i].m_Name = new wxChar[x]; - f -> Read(m_Index[i].m_Name, x*sizeof(wxChar)); - f -> Read(&x, sizeof(x)); + READ_STRING(f, m_Index[i].m_Name, x); + f -> Read(&x, sizeof(x)); x = wxINT32_SWAP_ON_BE(x); m_Index[i].m_Page = new wxChar[x]; - f -> Read(m_Index[i].m_Page, x*sizeof(wxChar)); + READ_STRING(f, m_Index[i].m_Page, x); m_Index[i].m_Book = book; } return TRUE; @@ -286,40 +335,47 @@ bool wxHtmlHelpData::LoadCachedBook(wxHtmlBookRecord *book, wxInputStream *f) bool wxHtmlHelpData::SaveCachedBook(wxHtmlBookRecord *book, wxOutputStream *f) { int i; - int x; + wxInt32 x; + + /* save header - version info : */ + + x = wxINT32_SWAP_ON_BE(CURRENT_CACHED_BOOK_VERSION); + f -> Write(&x, sizeof(x)); /* save contents : */ x = 0; for (i = 0; i < m_ContentsCnt; i++) if (m_Contents[i].m_Book == book && m_Contents[i].m_Level > 0) x++; + x = wxINT32_SWAP_ON_BE(x); f -> Write(&x, sizeof(x)); for (i = 0; i < m_ContentsCnt; i++) { if (m_Contents[i].m_Book != book || m_Contents[i].m_Level == 0) continue; - x = m_Contents[i].m_Level; + x = wxINT32_SWAP_ON_BE(m_Contents[i].m_Level); f -> Write(&x, sizeof(x)); - x = m_Contents[i].m_ID; + x = wxINT32_SWAP_ON_BE(m_Contents[i].m_ID); f -> Write(&x, sizeof(x)); - x = wxStrlen(m_Contents[i].m_Name) + 1; + x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Name) + 1); f -> Write(&x, sizeof(x)); - f -> Write(m_Contents[i].m_Name, x*sizeof(wxChar)); - x = wxStrlen(m_Contents[i].m_Page) + 1; + WRITE_STRING(f, m_Contents[i].m_Name, x); + x = wxINT32_SWAP_ON_BE(wxStrlen(m_Contents[i].m_Page) + 1); f -> Write(&x, sizeof(x)); - f -> Write(m_Contents[i].m_Page, x*sizeof(wxChar)); + WRITE_STRING(f, m_Contents[i].m_Page, x); } /* save index : */ x = 0; for (i = 0; i < m_IndexCnt; i++) if (m_Index[i].m_Book == book && m_Index[i].m_Level > 0) x++; + x = wxINT32_SWAP_ON_BE(x); f -> Write(&x, sizeof(x)); for (i = 0; i < m_IndexCnt; i++) { if (m_Index[i].m_Book != book || m_Index[i].m_Level == 0) continue; - x = wxStrlen(m_Index[i].m_Name) + 1; + x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Name) + 1); f -> Write(&x, sizeof(x)); - f -> Write(m_Index[i].m_Name, x*sizeof(wxChar)); - x = wxStrlen(m_Index[i].m_Page) + 1; + WRITE_STRING(f, m_Index[i].m_Name, x); + x = wxINT32_SWAP_ON_BE(wxStrlen(m_Index[i].m_Page) + 1); f -> Write(&x, sizeof(x)); - f -> Write(m_Index[i].m_Page, x*sizeof(wxChar)); + WRITE_STRING(f, m_Index[i].m_Page, x); } return TRUE; } @@ -330,22 +386,37 @@ void wxHtmlHelpData::SetTempDir(const wxString& path) if (path == wxEmptyString) m_TempPath = path; else { if (wxIsAbsolutePath(path)) m_TempPath = path; - else m_TempPath = wxGetCwd() + "/" + path; + else m_TempPath = wxGetCwd() + _T("/") + path; - if (m_TempPath[m_TempPath.Length() - 1] != '/') - m_TempPath << "/"; + if (m_TempPath[m_TempPath.Length() - 1] != _T('/')) + m_TempPath << _T('/'); } } -bool wxHtmlHelpData::AddBookParam(const wxString& title, const wxString& contfile, + +static wxString SafeFileName(const wxString& s) +{ + wxString res(s); + res.Replace(wxT("#"), wxT("_")); + res.Replace(wxT(":"), wxT("_")); + res.Replace(wxT("\\"), wxT("_")); + res.Replace(wxT("/"), wxT("_")); + return res; +} + +bool wxHtmlHelpData::AddBookParam(const wxFSFile& bookfile, + wxFontEncoding encoding, + const wxString& title, const wxString& contfile, const wxString& indexfile, const wxString& deftopic, const wxString& path) { wxFileSystem fsys; wxFSFile *fi; wxHtmlBookRecord *bookr; - wxString safetitle; + + int IndexOld = m_IndexCnt, + ContentsOld = m_ContentsCnt; if (! path.IsEmpty()) fsys.ChangePathTo(path, TRUE); @@ -366,23 +437,54 @@ bool wxHtmlHelpData::AddBookParam(const wxString& title, const wxString& contfil int cont_start = m_ContentsCnt++; // Try to find cached binary versions: - safetitle = SafeFileName(title); - fi = fsys.OpenFile(safetitle + wxT(".cached")); - if (fi == NULL) fi = fsys.OpenFile(m_TempPath + safetitle + wxT(".cached")); - if ((fi == NULL) || (m_TempPath == wxEmptyString)) { - LoadMSProject(bookr, fsys, indexfile, contfile); - if (m_TempPath != wxEmptyString) { - wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath + safetitle + wxT(".cached")); - SaveCachedBook(bookr, outs); - delete outs; + // 1. save file as book, but with .hhp.cached extension + // 2. same as 1. but in temp path + // 3. otherwise or if cache load failed, load it from MS. + + fi = fsys.OpenFile(bookfile.GetLocation() + wxT(".cached")); + + if (fi == NULL || + fi -> GetModificationTime() < bookfile.GetModificationTime() || + !LoadCachedBook(bookr, fi -> GetStream())) + { + if (fi != NULL) delete fi; + fi = fsys.OpenFile(m_TempPath + wxFileNameFromPath(bookfile.GetLocation()) + wxT(".cached")); + if (m_TempPath == wxEmptyString || fi == NULL || + fi -> GetModificationTime() < bookfile.GetModificationTime() || + !LoadCachedBook(bookr, fi -> GetStream())) + { + LoadMSProject(bookr, fsys, indexfile, contfile); + if (m_TempPath != wxEmptyString) + { + wxFileOutputStream *outs = new wxFileOutputStream(m_TempPath + + SafeFileName(wxFileNameFromPath(bookfile.GetLocation())) + wxT(".cached")); + SaveCachedBook(bookr, outs); + delete outs; + } } - } else { - LoadCachedBook(bookr, fi -> GetStream()); - delete fi; } + + if (fi != NULL) delete fi; // Now store the contents range bookr->SetContentsRange(cont_start, m_ContentsCnt); + + // Convert encoding, if neccessary: + if (encoding != wxFONTENCODING_SYSTEM) + { + wxFontEncodingArray a = wxEncodingConverter::GetPlatformEquivalents(encoding); + if (a.GetCount() != 0 && a[0] != encoding) + { + int i; + wxEncodingConverter conv; + conv.Init(encoding, a[0]); + + for (i = IndexOld; i < m_IndexCnt; i++) + conv.Convert(m_Index[i].m_Name); + for (i = ContentsOld; i < m_ContentsCnt; i++) + conv.Convert(m_Contents[i].m_Name); + } + } m_BookRecords.Add(bookr); if (m_IndexCnt > 0) @@ -425,15 +527,21 @@ bool wxHtmlHelpData::AddBook(const wxString& book) char linebuf[300]; wxString title = _("noname"), - safetitle, - start = wxEmptyString, - contents = wxEmptyString, index = wxEmptyString; + safetitle, + start = wxEmptyString, + contents = wxEmptyString, + index = wxEmptyString, + charset = wxEmptyString; if (wxIsAbsolutePath(book)) bookFull = book; else bookFull = wxGetCwd() + "/" + book; fi = fsys.OpenFile(bookFull); - if (fi == NULL) return FALSE; + if (fi == NULL) + { + wxLogError(_("Cannot open HTML help book: %s"), bookFull.mb_str()); + return FALSE; + } fsys.ChangePathTo(bookFull); s = fi -> GetStream(); sz = s -> GetSize(); @@ -441,7 +549,6 @@ bool wxHtmlHelpData::AddBook(const wxString& book) buff[sz] = 0; s -> Read(buff, sz); lineptr = buff; - delete fi; do { lineptr = ReadLine(lineptr, linebuf); @@ -454,10 +561,18 @@ bool wxHtmlHelpData::AddBook(const wxString& book) index = linebuf + strlen("Index file="); if (strstr(linebuf, "Contents file=") == linebuf) contents = linebuf + strlen("Contents file="); + if (strstr(linebuf, "Charset=") == linebuf) + charset = linebuf + strlen("Charset="); } while (lineptr != NULL); delete[] buff; - - return AddBookParam(title, contents, index, start, fsys.GetPath()); + + wxFontEncoding enc; + if (charset == wxEmptyString) enc = wxFONTENCODING_SYSTEM; + else enc = wxTheFontMapper -> CharsetToEncoding(charset); + bool rtval = AddBookParam(*fi, enc, + title, contents, index, start, fsys.GetPath()); + delete fi; + return rtval; } } @@ -535,6 +650,7 @@ wxString wxHtmlHelpData::FindPageById(int id) //---------------------------------------------------------------------------------- wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& keyword, + bool case_sensitive, bool whole_words_only, const wxString& book) { m_Data = data; @@ -558,45 +674,54 @@ wxHtmlSearchStatus::wxHtmlSearchStatus(wxHtmlHelpData* data, const wxString& key m_CurIndex = 0; m_MaxIndex = m_Data->m_ContentsCnt; } - m_Engine.LookFor(keyword); + m_Engine.LookFor(keyword, case_sensitive, whole_words_only); m_Active = (m_CurIndex < m_MaxIndex); - m_LastPage = wxEmptyString; + m_LastPage = NULL; } bool wxHtmlSearchStatus::Search() { - wxFileSystem fsys; wxFSFile *file; int i = m_CurIndex; // shortcut bool found = FALSE; + wxChar *thepage; - if (! m_Active) { - // sanity check. Illegal use, but we'll try to prevent a crash anyway -#if !defined(__VISAGECPP__) -wxASSERT(0); -#else -wxASSERT(m_Active); -#endif - return FALSE; + if (!m_Active) { + // sanity check. Illegal use, but we'll try to prevent a crash anyway + wxASSERT(m_Active); + return FALSE; } - m_ContentsItem = NULL; m_Name = wxEmptyString; + m_ContentsItem = NULL; + thepage = m_Data->m_Contents[i].m_Page; - file = fsys.OpenFile(m_Data->m_Contents[i].m_Book -> GetBasePath() + - m_Data->m_Contents[i].m_Page); - if (file) { - if (m_LastPage != file->GetLocation()) { - m_LastPage = file->GetLocation(); - if (m_Engine.Scan(file -> GetStream())) { - m_Name = m_Data->m_Contents[i].m_Name; - m_ContentsItem = m_Data->m_Contents + i; - found = TRUE; - } + m_Active = (++m_CurIndex < m_MaxIndex); + // check if it is same page with different anchor: + if (m_LastPage != NULL) + { + wxChar *p1, *p2; + for (p1 = thepage, p2 = m_LastPage; + *p1 != 0 && *p1 != _T('#') && *p1 == *p2; p1++, p2++) {} + + m_LastPage = thepage; + + if (*p1 == 0 || *p1 == _T('#')) + return FALSE; + } + else m_LastPage = thepage; + + wxFileSystem fsys; + file = fsys.OpenFile(m_Data->m_Contents[i].m_Book -> GetBasePath() + thepage); + if (file) + { + if (m_Engine.Scan(file -> GetStream())) { + m_Name = m_Data->m_Contents[i].m_Name; + m_ContentsItem = m_Data->m_Contents + i; + found = TRUE; } delete file; } - m_Active = (++m_CurIndex < m_MaxIndex); return found; } @@ -611,21 +736,27 @@ wxASSERT(m_Active); // wxSearchEngine //-------------------------------------------------------------------------------- -void wxSearchEngine::LookFor(const wxString& keyword) +void wxSearchEngine::LookFor(const wxString& keyword, bool case_sensitive, bool whole_words_only) { + m_CaseSensitive = case_sensitive; + m_WholeWords = whole_words_only; if (m_Keyword) delete[] m_Keyword; m_Keyword = new wxChar[keyword.Length() + 1]; wxStrcpy(m_Keyword, keyword.c_str()); - for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--) - if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z'))) - m_Keyword[i] += wxT('a') - wxT('A'); + + if (!m_CaseSensitive) + for (int i = wxStrlen(m_Keyword) - 1; i >= 0; i--) + if ((m_Keyword[i] >= wxT('A')) && (m_Keyword[i] <= wxT('Z'))) + m_Keyword[i] += wxT('a') - wxT('A'); } +#define WHITESPACE(c) (c == ' ' || c == '\n' || c == '\r' || c == '\t') + bool wxSearchEngine::Scan(wxInputStream *stream) { - wxASSERT_MSG(m_Keyword != NULL, _("wxSearchEngine::LookFor must be called before scanning!")); + wxASSERT_MSG(m_Keyword != NULL, wxT("wxSearchEngine::LookFor must be called before scanning!")); int i, j; int lng = stream ->GetSize(); @@ -635,13 +766,27 @@ bool wxSearchEngine::Scan(wxInputStream *stream) stream -> Read(buf, lng); buf[lng] = 0; - for (i = 0; i < lng; i++) - if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A'; + if (!m_CaseSensitive) + for (i = 0; i < lng; i++) + if ((buf[i] >= 'A') && (buf[i] <= 'Z')) buf[i] += 'a' - 'A'; - for (i = 0; i < lng - wrd; i++) { - j = 0; - while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++; - if (j == wrd) {found = TRUE; break; } + if (m_WholeWords) + { + for (i = 0; i < lng - wrd; i++) { + if (WHITESPACE(buf[i])) continue; + j = 0; + while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++; + if (j == wrd && WHITESPACE(buf[i + j])) {found = TRUE; break; } + } + } + + else + { + for (i = 0; i < lng - wrd; i++) { + j = 0; + while ((j < wrd) && (buf[i + j] == m_Keyword[j])) j++; + if (j == wrd) {found = TRUE; break; } + } } delete[] buf;