X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/981e62aa85597e36b40d89274bcf1203e5ae5f67..062dfc9a96dc9e796c53544f41ff92dc47f26e82:/src/html/htmlfilt.cpp diff --git a/src/html/htmlfilt.cpp b/src/html/htmlfilt.cpp index 199b91c6c4..0baea69675 100644 --- a/src/html/htmlfilt.cpp +++ b/src/html/htmlfilt.cpp @@ -1,33 +1,38 @@ ///////////////////////////////////////////////////////////////////////////// -// 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 ///////////////////////////////////////////////////////////////////////////// - -#ifdef __GNUG__ -#pragma implementation -#endif - #include "wx/wxprec.h" -#include "wx/defs.h" -#if wxUSE_HTML - -#ifdef __BORDLANDC__ -#pragma hdrstop +#ifdef __BORLANDC__ + #pragma hdrstop #endif -#ifndef WXPRECOMP -#include "wx/wx.h" +#if wxUSE_HTML && wxUSE_STREAMS + +#ifndef WX_PRECOMP + #include "wx/log.h" + #include "wx/intl.h" #endif +#include "wx/strconv.h" +#include "wx/sstream.h" #include "wx/html/htmlfilt.h" #include "wx/html/htmlwin.h" +// utility function: read entire contents of an wxInputStream into a wxString +// +// TODO: error handling? +static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv) +{ + wxStringOutputStream out(&str, conv); + s->Read(out); +} /* @@ -46,7 +51,7 @@ IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterPlainText, wxHtmlFilter) bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const { - return TRUE; + return true; } @@ -54,19 +59,15 @@ 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; } @@ -101,26 +102,18 @@ 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) @@ -130,7 +123,7 @@ bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const // This is true in most case but some page can return: // "text/html; char-encoding=...." // So we use Find instead - return (file.GetMimeType().Find(_("text/html")) == 0); + return (file.GetMimeType().Find(wxT("text/html")) == 0); } @@ -138,25 +131,56 @@ 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) return wxEmptyString; - src = new char[s -> GetSize() + 1]; - src[s -> GetSize()] = 0; - s -> Read(src, s -> GetSize()); - doc = src; - delete[] src; - - // add meta tag if we obtained this through http: - if (file.GetMimeType().Find(_T("; charset=")) == 0) + if (s == NULL) + { + wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str()); + return wxEmptyString; + } + + // 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(wxT("; charset="))) != wxNOT_FOUND) { - wxString s(_T(""); - return s; + wxString charset = file.GetMimeType().Mid(charsetPos + 10); + wxCSConv conv(charset); + ReadString(doc, s, conv); } + else + { + size_t size = s->GetSize(); + wxCharBuffer buf( size+1 ); + s->Read( buf.data(), size ); + *(buf.data() + size) = 0; + wxString tmpdoc( buf, wxConvISO8859_1); + + wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc); + if (charset.empty()) + doc = tmpdoc; + else + { + wxCSConv conv(charset); + doc = wxString( buf, conv ); + } + } +#else // !wxUSE_UNICODE + ReadString(doc, s, wxConvLibc); + // add meta tag if we obtained this through http: + if (!file.GetMimeType().empty()) + { + wxString hdr; + wxString mime = file.GetMimeType(); + hdr.Printf(wxT(""), mime.c_str()); + return hdr+doc; + } +#endif - else return doc; + return doc; } @@ -173,7 +197,7 @@ class wxHtmlFilterModule : public wxModule { wxHtmlWindow::AddFilter(new wxHtmlFilterHTML); wxHtmlWindow::AddFilter(new wxHtmlFilterImage); - return TRUE; + return true; } virtual void OnExit() {} };