[wxWidgets.git] / src / html / 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
/////////////////////////////////////////////////////////////////////////////

#include "wx/wxprec.h"

#ifdef __BORLANDC__
    #pragma hdrstop
#endif

#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);
}

/*

There is code for several default filters:

*/

IMPLEMENT_ABSTRACT_CLASS(wxHtmlFilter, wxObject)

//--------------------------------------------------------------------------------
// wxHtmlFilterPlainText
//          filter for text/plain or uknown
//--------------------------------------------------------------------------------

IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterPlainText, wxHtmlFilter)

bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const
{
    return true;
}


wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const
{
    wxInputStream *s = file.GetStream();
    wxString doc, doc2;

    if (s == NULL) return wxEmptyString;
    ReadString(doc, s, wxConvISO8859_1);

    doc.Replace(wxT("&"), wxT("&amp;"), true);
    doc.Replace(wxT("<"), wxT("&lt;"), true);
    doc.Replace(wxT(">"), wxT("&gt;"), true);
    doc2 = wxT("<HTML><BODY><PRE>\n") + doc + wxT("\n</PRE></BODY></HTML>");
    return doc2;
}


//--------------------------------------------------------------------------------
// wxHtmlFilterImage
//          filter for image/*
//--------------------------------------------------------------------------------

class wxHtmlFilterImage : public wxHtmlFilter
{
    DECLARE_DYNAMIC_CLASS(wxHtmlFilterImage)

    public:
        virtual bool CanRead(const wxFSFile& file) const;
        virtual wxString ReadFile(const wxFSFile& file) const;
};

IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterImage, wxHtmlFilter)


bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const
{
    return (file.GetMimeType().Left(6) == wxT("image/"));
}


wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const
{
    wxString res = wxT("<HTML><BODY><IMG SRC=\"") + file.GetLocation() + wxT("\"></BODY></HTML>");
    return res;
}


//--------------------------------------------------------------------------------
// wxHtmlFilterHTML
//          filter for text/html
//--------------------------------------------------------------------------------


IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter)

bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const
{
//    return (file.GetMimeType() == "text/html");
// This is true in most case but some page can return:
// "text/html; char-encoding=...."
// So we use Find instead
  return (file.GetMimeType().Find(wxT("text/html")) == 0);
}


wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const
{
    wxInputStream *s = file.GetStream();
    wxString doc;

    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 <meta> tags. In ANSI mode, we don't
    //     do it as it is done by wxHtmlParser (for this reason, we add <meta>
    //     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
    {
        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(_T("<meta http-equiv=\"Content-Type\" content=\"%s\">"), mime.c_str());
        return hdr+doc;
    }
#endif

    return doc;
}


///// Module:

class wxHtmlFilterModule : public wxModule
{
    DECLARE_DYNAMIC_CLASS(wxHtmlFilterModule)

    public:
        virtual bool OnInit()
        {
            wxHtmlWindow::AddFilter(new wxHtmlFilterHTML);
            wxHtmlWindow::AddFilter(new wxHtmlFilterImage);
            return true;
        }
        virtual void OnExit() {}
};

IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterModule, wxModule)

#endif
Commit	Line	Data
5526e819	1	/////////////////////////////////////////////////////////////////////////////
93763ad5	2	// Name: src/html/htmlfilt.cpp
5526e819 VS	3	// Purpose: wxHtmlFilter - input filter for translating into HTML format
5526e819 VS	4	// Author: Vaclav Slavik
69941f05	5	// RCS-ID: $Id$
5526e819	6	// Copyright: (c) 1999 Vaclav Slavik
65571936	7	// Licence: wxWindows licence
5526e819 VS	8	/////////////////////////////////////////////////////////////////////////////
5526e819 VS	9
4dcaf11a	10	#include "wx/wxprec.h"
5526e819	11
2b5f62a0	12	#ifdef __BORLANDC__
93763ad5	13	#pragma hdrstop
5526e819 VS	14	#endif
5526e819 VS	15
93763ad5 WS	16	#if wxUSE_HTML && wxUSE_STREAMS
93763ad5 WS	17
b4f4d3dd	18	#ifndef WX_PRECOMP
04dbb646 VZ	19	#include "wx/log.h"
04dbb646 VZ	20	#include "wx/intl.h"
5526e819 VS	21	#endif
5526e819 VS	22
2b5f62a0	23	#include "wx/strconv.h"
63482fd5	24	#include "wx/sstream.h"
69941f05	25	#include "wx/html/htmlfilt.h"
4dcaf11a	26	#include "wx/html/htmlwin.h"
5526e819	27
63482fd5 VZ	28	// utility function: read entire contents of an wxInputStream into a wxString
	29	//
	30	// TODO: error handling?
2b5f62a0	31	static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv)
eb37e1d2	32	{
63482fd5 VZ	33	wxStringOutputStream out(&str, conv);
63482fd5 VZ	34	s->Read(out);
eb37e1d2	35	}
5526e819 VS	36
	37	/*
	38
	39	There is code for several default filters:
	40
	41	*/
	42
	43	IMPLEMENT_ABSTRACT_CLASS(wxHtmlFilter, wxObject)
	44
	45	//--------------------------------------------------------------------------------
	46	// wxHtmlFilterPlainText
	47	// filter for text/plain or uknown
	48	//--------------------------------------------------------------------------------
	49
	50	IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterPlainText, wxHtmlFilter)
	51
a4c97004	52	bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const
5526e819	53	{
d1da8872	54	return true;
5526e819 VS	55	}
	56
	57
	58
420ec58a	59	wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const
5526e819 VS	60	{
5526e819 VS	61	wxInputStream *s = file.GetStream();
5526e819 VS	62	wxString doc, doc2;
	63
	64	if (s == NULL) return wxEmptyString;
2b5f62a0	65	ReadString(doc, s, wxConvISO8859_1);
5526e819	66
d1da8872 WS	67	doc.Replace(wxT("&"), wxT("&"), true);
	68	doc.Replace(wxT("<"), wxT("<"), true);
	69	doc.Replace(wxT(">"), wxT(">"), true);
2b5f62a0	70	doc2 = wxT("<HTML><BODY><PRE>\n") + doc + wxT("\n</PRE></BODY></HTML>");
5526e819 VS	71	return doc2;
	72	}
	73
	74
	75
	76
	77
	78	//--------------------------------------------------------------------------------
	79	// wxHtmlFilterImage
	80	// filter for image/*
	81	//--------------------------------------------------------------------------------
	82
	83	class wxHtmlFilterImage : public wxHtmlFilter
	84	{
	85	DECLARE_DYNAMIC_CLASS(wxHtmlFilterImage)
	86
	87	public:
420ec58a VS	88	virtual bool CanRead(const wxFSFile& file) const;
420ec58a VS	89	virtual wxString ReadFile(const wxFSFile& file) const;
5526e819 VS	90	};
	91
	92	IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterImage, wxHtmlFilter)
	93
	94
	95
420ec58a	96	bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const
5526e819	97	{
0413cec5	98	return (file.GetMimeType().Left(6) == wxT("image/"));
5526e819 VS	99	}
	100
	101
	102
420ec58a	103	wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const
5526e819	104	{
2b5f62a0 VZ	105	wxString res = wxT("<HTML><BODY><IMG SRC=\"") + file.GetLocation() + wxT("\"></BODY></HTML>");
2b5f62a0 VZ	106	return res;
5526e819 VS	107	}
	108
	109
	110
	111
	112	//--------------------------------------------------------------------------------
2b5f62a0 VZ	113	// wxHtmlFilterHTML
2b5f62a0 VZ	114	// filter for text/html
5526e819 VS	115	//--------------------------------------------------------------------------------
5526e819 VS	116
5526e819 VS	117
	118	IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter)
	119
420ec58a	120	bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const
5526e819	121	{
f61815af GL	122	// return (file.GetMimeType() == "text/html");
	123	// This is true in most case but some page can return:
	124	// "text/html; char-encoding=...."
	125	// So we use Find instead
77611ad4	126	return (file.GetMimeType().Find(wxT("text/html")) == 0);
5526e819 VS	127	}
	128
	129
	130
420ec58a	131	wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const
5526e819 VS	132	{
5526e819 VS	133	wxInputStream *s = file.GetStream();
5526e819 VS	134	wxString doc;
5526e819 VS	135
04dbb646	136	if (s == NULL)
f3c82859	137	{
f6bcfd97	138	wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str());
f3c82859 VS	139	return wxEmptyString;
f3c82859 VS	140	}
5526e819	141
d1da8872	142	// NB: We convert input file to wchar_t here in Unicode mode, based on
2b5f62a0 VZ	143	// either Content-Type header or <meta> tags. In ANSI mode, we don't
	144	// do it as it is done by wxHtmlParser (for this reason, we add <meta>
	145	// tag if we used Content-Type header).
	146	#if wxUSE_UNICODE
d1da8872	147	int charsetPos;
2b5f62a0 VZ	148	if ((charsetPos = file.GetMimeType().Find(_T("; charset="))) != wxNOT_FOUND)
	149	{
	150	wxString charset = file.GetMimeType().Mid(charsetPos + 10);
	151	wxCSConv conv(charset);
	152	ReadString(doc, s, conv);
	153	}
	154	else
	155	{
08f744fb RR	156	size_t size = s->GetSize();
	157	wxCharBuffer buf( size+1 );
	158	s->Read( buf.data(), size );
	159	*(buf.data() + size) = 0;
	160	wxString tmpdoc( buf, wxConvISO8859_1);
	161
2b5f62a0 VZ	162	wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc);
	163	if (charset.empty())
	164	doc = tmpdoc;
	165	else
	166	{
	167	wxCSConv conv(charset);
08f744fb	168	doc = wxString( buf, conv );
2b5f62a0 VZ	169	}
	170	}
	171	#else // !wxUSE_UNICODE
	172	ReadString(doc, s, wxConvLibc);
04dbb646	173	// add meta tag if we obtained this through http:
2b5f62a0	174	if (!file.GetMimeType().empty())
981e62aa	175	{
2b5f62a0 VZ	176	wxString hdr;
	177	wxString mime = file.GetMimeType();
	178	hdr.Printf(_T("<meta http-equiv=\"Content-Type\" content=\"%s\">"), mime.c_str());
	179	return hdr+doc;
981e62aa	180	}
2b5f62a0	181	#endif
981e62aa	182
3ca6a5f0	183	return doc;
5526e819 VS	184	}
	185
	186
	187
	188
	189	///// Module:
	190
	191	class wxHtmlFilterModule : public wxModule
	192	{
	193	DECLARE_DYNAMIC_CLASS(wxHtmlFilterModule)
	194
	195	public:
	196	virtual bool OnInit()
	197	{
	198	wxHtmlWindow::AddFilter(new wxHtmlFilterHTML);
	199	wxHtmlWindow::AddFilter(new wxHtmlFilterImage);
d1da8872	200	return true;
5526e819 VS	201	}
	202	virtual void OnExit() {}
	203	};
	204
	205	IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterModule, wxModule)
	206
8dd71e2b	207	#endif