X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/04dbb6467be8f564f380bd9a1106fbdecbd26a98..b9d880d6382e24156cdce4ae0390c843cc6208d9:/src/html/htmlfilt.cpp diff --git a/src/html/htmlfilt.cpp b/src/html/htmlfilt.cpp index b28da27368..025a19e155 100644 --- a/src/html/htmlfilt.cpp +++ b/src/html/htmlfilt.cpp @@ -1,34 +1,58 @@ ///////////////////////////////////////////////////////////////////////////// -// Name: htmlfilt.cpp +// Name: src/html/htmlfilt.cpp // Purpose: wxHtmlFilter - input filter for translating into HTML format // Author: Vaclav Slavik // RCS-ID: $Id$ // Copyright: (c) 1999 Vaclav Slavik -// Licence: wxWindows Licence +// Licence: wxWindows licence ///////////////////////////////////////////////////////////////////////////// +#include "wx/wxprec.h" -#ifdef __GNUG__ -#pragma implementation +#ifdef __BORLANDC__ + #pragma hdrstop #endif -#include "wx/wxprec.h" - -#include "wx/defs.h" #if wxUSE_HTML && wxUSE_STREAMS -#ifdef __BORDLANDC__ -#pragma hdrstop -#endif - #ifndef WXPRECOMP #include "wx/log.h" #include "wx/intl.h" #endif +#include "wx/strconv.h" #include "wx/html/htmlfilt.h" #include "wx/html/htmlwin.h" +// utility function: read a wxString from a wxInputStream +static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv) +{ + size_t streamSize = s->GetSize(); + + if (streamSize == ~(size_t)0) + { + const size_t bufSize = 4095; + char buffer[bufSize+1]; + size_t lastRead; + + do + { + s->Read(buffer, bufSize); + lastRead = s->LastRead(); + buffer[lastRead] = 0; + str.Append(wxString(buffer, conv)); + } + while (lastRead == bufSize); + } + else + { + char* src = new char[streamSize+1]; + s->Read(src, streamSize); + src[streamSize] = 0; + str = wxString(src, conv); + delete[] src; + } +} /* @@ -47,7 +71,7 @@ IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterPlainText, wxHtmlFilter) bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const { - return TRUE; + return true; } @@ -55,19 +79,15 @@ bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const { wxInputStream *s = file.GetStream(); - char *src; wxString doc, doc2; if (s == NULL) return wxEmptyString; - src = new char[s->GetSize()+1]; - src[s->GetSize()] = 0; - s->Read(src, s->GetSize()); - doc = src; - delete [] src; - - doc.Replace(wxT("<"), wxT("<"), TRUE); - doc.Replace(wxT(">"), wxT(">"), TRUE); - doc2 = "
\n" + doc + "\n"; + ReadString(doc, s, wxConvISO8859_1); + + doc.Replace(wxT("&"), wxT("&"), true); + doc.Replace(wxT("<"), wxT("<"), true); + doc.Replace(wxT(">"), wxT(">"), true); + doc2 = wxT("
\n") + doc + wxT("\n"); return doc2; } @@ -102,26 +122,18 @@ bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const { - return (""); + wxString res = wxT(""); + return res; } //-------------------------------------------------------------------------------- -// wxHtmlFilterPlainText -// filter for text/plain or uknown +// wxHtmlFilterHTML +// filter for text/html //-------------------------------------------------------------------------------- -class wxHtmlFilterHTML : public wxHtmlFilter -{ - DECLARE_DYNAMIC_CLASS(wxHtmlFilterHTML) - - public: - virtual bool CanRead(const wxFSFile& file) const; - virtual wxString ReadFile(const wxFSFile& file) const; -}; - IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter) @@ -139,7 +151,6 @@ bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const { wxInputStream *s = file.GetStream(); - char *src; wxString doc; if (s == NULL) @@ -147,19 +158,43 @@ wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str()); return wxEmptyString; } - src = new char[s->GetSize() + 1]; - src[s->GetSize()] = 0; - s->Read(src, s->GetSize()); - doc = src; - delete[] src; + // NB: We convert input file to wchar_t here in Unicode mode, based on + // either Content-Type header or tags. In ANSI mode, we don't + // do it as it is done by wxHtmlParser (for this reason, we add + // tag if we used Content-Type header). +#if wxUSE_UNICODE + int charsetPos; + if ((charsetPos = file.GetMimeType().Find(_T("; charset="))) != wxNOT_FOUND) + { + wxString charset = file.GetMimeType().Mid(charsetPos + 10); + wxCSConv conv(charset); + ReadString(doc, s, conv); + } + else + { + wxString tmpdoc; + ReadString(tmpdoc, s, wxConvISO8859_1); + wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc); + if (charset.empty()) + doc = tmpdoc; + else + { + wxCSConv conv(charset); + doc = wxString(tmpdoc.mb_str(wxConvISO8859_1), conv); + } + } +#else // !wxUSE_UNICODE + ReadString(doc, s, wxConvLibc); // add meta tag if we obtained this through http: - if (file.GetMimeType().Find(_T("; charset=")) == 0) + if (!file.GetMimeType().empty()) { - wxString s(_T(""); - return s; + wxString hdr; + wxString mime = file.GetMimeType(); + hdr.Printf(_T(""), mime.c_str()); + return hdr+doc; } +#endif return doc; } @@ -178,7 +213,7 @@ class wxHtmlFilterModule : public wxModule { wxHtmlWindow::AddFilter(new wxHtmlFilterHTML); wxHtmlWindow::AddFilter(new wxHtmlFilterImage); - return TRUE; + return true; } virtual void OnExit() {} };