]>
Commit | Line | Data |
---|---|---|
5526e819 | 1 | ///////////////////////////////////////////////////////////////////////////// |
69941f05 | 2 | // Name: htmlfilt.cpp |
5526e819 VS |
3 | // Purpose: wxHtmlFilter - input filter for translating into HTML format |
4 | // Author: Vaclav Slavik | |
69941f05 | 5 | // RCS-ID: $Id$ |
5526e819 | 6 | // Copyright: (c) 1999 Vaclav Slavik |
65571936 | 7 | // Licence: wxWindows licence |
5526e819 VS |
8 | ///////////////////////////////////////////////////////////////////////////// |
9 | ||
4dcaf11a | 10 | #include "wx/wxprec.h" |
5526e819 | 11 | |
314260fb | 12 | #include "wx/defs.h" |
f6bcfd97 | 13 | #if wxUSE_HTML && wxUSE_STREAMS |
5526e819 | 14 | |
2b5f62a0 | 15 | #ifdef __BORLANDC__ |
5526e819 VS |
16 | #pragma hdrstop |
17 | #endif | |
18 | ||
19 | #ifndef WXPRECOMP | |
04dbb646 VZ |
20 | #include "wx/log.h" |
21 | #include "wx/intl.h" | |
5526e819 VS |
22 | #endif |
23 | ||
2b5f62a0 | 24 | #include "wx/strconv.h" |
69941f05 | 25 | #include "wx/html/htmlfilt.h" |
4dcaf11a | 26 | #include "wx/html/htmlwin.h" |
5526e819 | 27 | |
eb37e1d2 | 28 | // utility function: read a wxString from a wxInputStream |
2b5f62a0 | 29 | static void ReadString(wxString& str, wxInputStream* s, wxMBConv& conv) |
eb37e1d2 MB |
30 | { |
31 | size_t streamSize = s->GetSize(); | |
32 | ||
2b5f62a0 | 33 | if (streamSize == ~(size_t)0) |
eb37e1d2 MB |
34 | { |
35 | const size_t bufSize = 4095; | |
36 | char buffer[bufSize+1]; | |
37 | size_t lastRead; | |
38 | ||
39 | do | |
40 | { | |
41 | s->Read(buffer, bufSize); | |
42 | lastRead = s->LastRead(); | |
43 | buffer[lastRead] = 0; | |
2b5f62a0 | 44 | str.Append(wxString(buffer, conv)); |
eb37e1d2 | 45 | } |
2b5f62a0 | 46 | while (lastRead == bufSize); |
eb37e1d2 MB |
47 | } |
48 | else | |
49 | { | |
50 | char* src = new char[streamSize+1]; | |
51 | s->Read(src, streamSize); | |
52 | src[streamSize] = 0; | |
2b5f62a0 VZ |
53 | str = wxString(src, conv); |
54 | delete[] src; | |
eb37e1d2 MB |
55 | } |
56 | } | |
5526e819 VS |
57 | |
58 | /* | |
59 | ||
60 | There is code for several default filters: | |
61 | ||
62 | */ | |
63 | ||
64 | IMPLEMENT_ABSTRACT_CLASS(wxHtmlFilter, wxObject) | |
65 | ||
66 | //-------------------------------------------------------------------------------- | |
67 | // wxHtmlFilterPlainText | |
68 | // filter for text/plain or uknown | |
69 | //-------------------------------------------------------------------------------- | |
70 | ||
71 | IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterPlainText, wxHtmlFilter) | |
72 | ||
a4c97004 | 73 | bool wxHtmlFilterPlainText::CanRead(const wxFSFile& WXUNUSED(file)) const |
5526e819 | 74 | { |
d1da8872 | 75 | return true; |
5526e819 VS |
76 | } |
77 | ||
78 | ||
79 | ||
420ec58a | 80 | wxString wxHtmlFilterPlainText::ReadFile(const wxFSFile& file) const |
5526e819 VS |
81 | { |
82 | wxInputStream *s = file.GetStream(); | |
5526e819 VS |
83 | wxString doc, doc2; |
84 | ||
85 | if (s == NULL) return wxEmptyString; | |
2b5f62a0 | 86 | ReadString(doc, s, wxConvISO8859_1); |
5526e819 | 87 | |
d1da8872 WS |
88 | doc.Replace(wxT("&"), wxT("&"), true); |
89 | doc.Replace(wxT("<"), wxT("<"), true); | |
90 | doc.Replace(wxT(">"), wxT(">"), true); | |
2b5f62a0 | 91 | doc2 = wxT("<HTML><BODY><PRE>\n") + doc + wxT("\n</PRE></BODY></HTML>"); |
5526e819 VS |
92 | return doc2; |
93 | } | |
94 | ||
95 | ||
96 | ||
97 | ||
98 | ||
99 | //-------------------------------------------------------------------------------- | |
100 | // wxHtmlFilterImage | |
101 | // filter for image/* | |
102 | //-------------------------------------------------------------------------------- | |
103 | ||
104 | class wxHtmlFilterImage : public wxHtmlFilter | |
105 | { | |
106 | DECLARE_DYNAMIC_CLASS(wxHtmlFilterImage) | |
107 | ||
108 | public: | |
420ec58a VS |
109 | virtual bool CanRead(const wxFSFile& file) const; |
110 | virtual wxString ReadFile(const wxFSFile& file) const; | |
5526e819 VS |
111 | }; |
112 | ||
113 | IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterImage, wxHtmlFilter) | |
114 | ||
115 | ||
116 | ||
420ec58a | 117 | bool wxHtmlFilterImage::CanRead(const wxFSFile& file) const |
5526e819 | 118 | { |
0413cec5 | 119 | return (file.GetMimeType().Left(6) == wxT("image/")); |
5526e819 VS |
120 | } |
121 | ||
122 | ||
123 | ||
420ec58a | 124 | wxString wxHtmlFilterImage::ReadFile(const wxFSFile& file) const |
5526e819 | 125 | { |
2b5f62a0 VZ |
126 | wxString res = wxT("<HTML><BODY><IMG SRC=\"") + file.GetLocation() + wxT("\"></BODY></HTML>"); |
127 | return res; | |
5526e819 VS |
128 | } |
129 | ||
130 | ||
131 | ||
132 | ||
133 | //-------------------------------------------------------------------------------- | |
2b5f62a0 VZ |
134 | // wxHtmlFilterHTML |
135 | // filter for text/html | |
5526e819 VS |
136 | //-------------------------------------------------------------------------------- |
137 | ||
5526e819 VS |
138 | |
139 | IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterHTML, wxHtmlFilter) | |
140 | ||
420ec58a | 141 | bool wxHtmlFilterHTML::CanRead(const wxFSFile& file) const |
5526e819 | 142 | { |
f61815af GL |
143 | // return (file.GetMimeType() == "text/html"); |
144 | // This is true in most case but some page can return: | |
145 | // "text/html; char-encoding=...." | |
146 | // So we use Find instead | |
77611ad4 | 147 | return (file.GetMimeType().Find(wxT("text/html")) == 0); |
5526e819 VS |
148 | } |
149 | ||
150 | ||
151 | ||
420ec58a | 152 | wxString wxHtmlFilterHTML::ReadFile(const wxFSFile& file) const |
5526e819 VS |
153 | { |
154 | wxInputStream *s = file.GetStream(); | |
5526e819 VS |
155 | wxString doc; |
156 | ||
04dbb646 | 157 | if (s == NULL) |
f3c82859 | 158 | { |
f6bcfd97 | 159 | wxLogError(_("Cannot open HTML document: %s"), file.GetLocation().c_str()); |
f3c82859 VS |
160 | return wxEmptyString; |
161 | } | |
5526e819 | 162 | |
d1da8872 | 163 | // NB: We convert input file to wchar_t here in Unicode mode, based on |
2b5f62a0 VZ |
164 | // either Content-Type header or <meta> tags. In ANSI mode, we don't |
165 | // do it as it is done by wxHtmlParser (for this reason, we add <meta> | |
166 | // tag if we used Content-Type header). | |
167 | #if wxUSE_UNICODE | |
d1da8872 | 168 | int charsetPos; |
2b5f62a0 VZ |
169 | if ((charsetPos = file.GetMimeType().Find(_T("; charset="))) != wxNOT_FOUND) |
170 | { | |
171 | wxString charset = file.GetMimeType().Mid(charsetPos + 10); | |
172 | wxCSConv conv(charset); | |
173 | ReadString(doc, s, conv); | |
174 | } | |
175 | else | |
176 | { | |
177 | wxString tmpdoc; | |
178 | ReadString(tmpdoc, s, wxConvISO8859_1); | |
179 | wxString charset = wxHtmlParser::ExtractCharsetInformation(tmpdoc); | |
180 | if (charset.empty()) | |
181 | doc = tmpdoc; | |
182 | else | |
183 | { | |
184 | wxCSConv conv(charset); | |
185 | doc = wxString(tmpdoc.mb_str(wxConvISO8859_1), conv); | |
186 | } | |
187 | } | |
188 | #else // !wxUSE_UNICODE | |
189 | ReadString(doc, s, wxConvLibc); | |
04dbb646 | 190 | // add meta tag if we obtained this through http: |
2b5f62a0 | 191 | if (!file.GetMimeType().empty()) |
981e62aa | 192 | { |
2b5f62a0 VZ |
193 | wxString hdr; |
194 | wxString mime = file.GetMimeType(); | |
195 | hdr.Printf(_T("<meta http-equiv=\"Content-Type\" content=\"%s\">"), mime.c_str()); | |
196 | return hdr+doc; | |
981e62aa | 197 | } |
2b5f62a0 | 198 | #endif |
981e62aa | 199 | |
3ca6a5f0 | 200 | return doc; |
5526e819 VS |
201 | } |
202 | ||
203 | ||
204 | ||
205 | ||
206 | ///// Module: | |
207 | ||
208 | class wxHtmlFilterModule : public wxModule | |
209 | { | |
210 | DECLARE_DYNAMIC_CLASS(wxHtmlFilterModule) | |
211 | ||
212 | public: | |
213 | virtual bool OnInit() | |
214 | { | |
215 | wxHtmlWindow::AddFilter(new wxHtmlFilterHTML); | |
216 | wxHtmlWindow::AddFilter(new wxHtmlFilterImage); | |
d1da8872 | 217 | return true; |
5526e819 VS |
218 | } |
219 | virtual void OnExit() {} | |
220 | }; | |
221 | ||
222 | IMPLEMENT_DYNAMIC_CLASS(wxHtmlFilterModule, wxModule) | |
223 | ||
8dd71e2b | 224 | #endif |