]> git.saurik.com Git - wxWidgets.git/blob - src/html/htmlfilt.cpp
Fix crash in wxDC::GetMultiLineTextExtent() after last commit.
[wxWidgets.git] / src / html / htmlfilt.cpp
1 /////////////////////////////////////////////////////////////////////////////
2 // Name: src/html/htmlfilt.cpp
3 // Purpose: wxHtmlFilter - input filter for translating into HTML format
4 // Author: Vaclav Slavik
5 // RCS-ID: $Id$
6 // Copyright: (c) 1999 Vaclav Slavik
7 // Licence: wxWindows licence
8 /////////////////////////////////////////////////////////////////////////////
9
10 #include "wx/wxprec.h"
11
12 #ifdef __BORLANDC__
13 #pragma hdrstop
14 #endif
15
16 #if wxUSE_HTML && wxUSE_STREAMS
17
18 #ifndef WX_PRECOMP
19 #include "wx/log.h"
20 #include "wx/intl.h"
21 #endif
22
23 #include "wx/strconv.h"
24 #include "wx/sstream.h"
25 #include "wx/html/htmlfilt.h"
26 #include "wx/html/htmlwin.h"
27
28 // utility function: read entire contents of an wxInputStream into a wxString
29 //
30 // TODO: error handling?
31 static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv)
32 {
33 wxStringOutputStream out(&str, conv);
34 s->Read(out);
35 }
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
52 bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const
53 {
54 return true;
55 }
56
57
58
59 wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const
60 {
61 wxInputStream *s = file.GetStream();
62 wxString doc, doc2;
63
64 if (s == NULL) return wxEmptyString;
65 ReadString(doc, s, wxConvISO8859_1);
66
67 doc.Replace(wxT("&"), wxT("&"), true);
68 doc.Replace(wxT("<"), wxT("&lt;"), true);
69 doc.Replace(wxT(">"), wxT("&gt;"), true);
70 doc2 = wxT("<HTML><BODY><PRE>\n") + doc + wxT("\n</PRE></BODY></HTML>");
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:
88 virtual bool CanRead(const wxFSFile& file) const;
89 virtual wxString ReadFile(const wxFSFile& file) const;
90 };
91
92 IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterImage, wxHtmlFilter)
93
94
95
96 bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const
97 {
98 return (file.GetMimeType().Left(6) == wxT("image/"));
99 }
100
101
102
103 wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const
104 {
105 wxString res = wxT("<HTML><BODY><IMG SRC=\"") + file.GetLocation() + wxT("\"></BODY></HTML>");
106 return res;
107 }
108
109
110
111
112 //--------------------------------------------------------------------------------
113 // wxHtmlFilterHTML
114 // filter for text/html
115 //--------------------------------------------------------------------------------
116
117
118 IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter)
119
120 bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const
121 {
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
126 return (file.GetMimeType().Find(wxT("text/html")) == 0);
127 }
128
129
130
131 wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const
132 {
133 wxInputStream *s = file.GetStream();
134 wxString doc;
135
136 if (s == NULL)
137 {
138 wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str());
139 return wxEmptyString;
140 }
141
142 // NB: We convert input file to wchar_t here in Unicode mode, based on
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
147 int charsetPos;
148 if ((charsetPos = file.GetMimeType().Find(wxT("; 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 {
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
162 wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc);
163 if (charset.empty())
164 doc = tmpdoc;
165 else
166 {
167 wxCSConv conv(charset);
168 doc = wxString( buf, conv );
169 }
170 }
171 #else // !wxUSE_UNICODE
172 ReadString(doc, s, wxConvLibc);
173 // add meta tag if we obtained this through http:
174 if (!file.GetMimeType().empty())
175 {
176 wxString hdr;
177 wxString mime = file.GetMimeType();
178 hdr.Printf(wxT("<meta http-equiv=\"Content-Type\" content=\"%s\">"), mime.c_str());
179 return hdr+doc;
180 }
181 #endif
182
183 return doc;
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);
200 return true;
201 }
202 virtual void OnExit() {}
203 };
204
205 IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterModule, wxModule)
206
207 #endif