|
129
|
1 /*******************************************************************************
|
|
|
2 * Copyright (c) 2000, 2008 IBM Corporation and others.
|
|
|
3 * All rights reserved. This program and the accompanying materials
|
|
|
4 * are made available under the terms of the Eclipse Public License v1.0
|
|
|
5 * which accompanies this distribution, and is available at
|
|
|
6 * http://www.eclipse.org/legal/epl-v10.html
|
|
|
7 *
|
|
|
8 * Contributors:
|
|
|
9 * IBM Corporation - initial API and implementation
|
|
|
10 * Port to the D programming language:
|
|
|
11 * Frank Benoit <benoit@tionex.de>
|
|
|
12 *******************************************************************************/
|
|
|
13 module dwtx.jface.internal.text.html.HTML2TextReader;
|
|
|
14
|
|
|
15 import dwt.dwthelper.utils;
|
|
|
16
|
|
|
17 import java.io.IOException;
|
|
|
18 import java.io.PushbackReader;
|
|
|
19 import java.io.Reader;
|
|
|
20 import java.util.HashMap;
|
|
|
21 import java.util.HashSet;
|
|
|
22 import java.util.Map;
|
|
|
23 import java.util.Set;
|
|
|
24
|
|
|
25 import dwt.DWT;
|
|
|
26 import dwt.custom.StyleRange;
|
|
|
27 import dwtx.jface.text.TextPresentation;
|
|
|
28
|
|
|
29
|
|
|
30 /**
|
|
|
31 * Reads the text contents from a reader of HTML contents and translates
|
|
|
32 * the tags or cut them out.
|
|
|
33 * <p>
|
|
|
34 * Moved into this package from <code>dwtx.jface.internal.text.revisions</code>.</p>
|
|
|
35 */
|
|
|
36 public class HTML2TextReader : SubstitutionTextReader {
|
|
|
37
|
|
|
38 private static final String EMPTY_STRING= ""; //$NON-NLS-1$
|
|
|
39 private static final Map fgEntityLookup;
|
|
|
40 private static final Set fgTags;
|
|
|
41
|
|
|
42 static {
|
|
|
43
|
|
|
44 fgTags= new HashSet();
|
|
|
45 fgTags.add("b"); //$NON-NLS-1$
|
|
|
46 fgTags.add("br"); //$NON-NLS-1$
|
|
|
47 fgTags.add("br/"); //$NON-NLS-1$
|
|
|
48 fgTags.add("div"); //$NON-NLS-1$
|
|
|
49 fgTags.add("h1"); //$NON-NLS-1$
|
|
|
50 fgTags.add("h2"); //$NON-NLS-1$
|
|
|
51 fgTags.add("h3"); //$NON-NLS-1$
|
|
|
52 fgTags.add("h4"); //$NON-NLS-1$
|
|
|
53 fgTags.add("h5"); //$NON-NLS-1$
|
|
|
54 fgTags.add("p"); //$NON-NLS-1$
|
|
|
55 fgTags.add("dl"); //$NON-NLS-1$
|
|
|
56 fgTags.add("dt"); //$NON-NLS-1$
|
|
|
57 fgTags.add("dd"); //$NON-NLS-1$
|
|
|
58 fgTags.add("li"); //$NON-NLS-1$
|
|
|
59 fgTags.add("ul"); //$NON-NLS-1$
|
|
|
60 fgTags.add("pre"); //$NON-NLS-1$
|
|
|
61 fgTags.add("head"); //$NON-NLS-1$
|
|
|
62
|
|
|
63 fgEntityLookup= new HashMap(7);
|
|
|
64 fgEntityLookup.put("lt", "<"); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
65 fgEntityLookup.put("gt", ">"); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
66 fgEntityLookup.put("nbsp", " "); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
67 fgEntityLookup.put("amp", "&"); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
68 fgEntityLookup.put("circ", "^"); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
69 fgEntityLookup.put("tilde", "~"); //$NON-NLS-2$ //$NON-NLS-1$
|
|
|
70 fgEntityLookup.put("quot", "\""); //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
71 }
|
|
|
72
|
|
|
73 private int fCounter= 0;
|
|
|
74 private TextPresentation fTextPresentation;
|
|
|
75 private int fBold= 0;
|
|
|
76 private int fStartOffset= -1;
|
|
|
77 private bool fInParagraph= false;
|
|
|
78 private bool fIsPreformattedText= false;
|
|
|
79 private bool fIgnore= false;
|
|
|
80 private bool fHeaderDetected= false;
|
|
|
81
|
|
|
82 /**
|
|
|
83 * Transforms the HTML text from the reader to formatted text.
|
|
|
84 *
|
|
|
85 * @param reader the reader
|
|
|
86 * @param presentation If not <code>null</code>, formattings will be applied to
|
|
|
87 * the presentation.
|
|
|
88 */
|
|
|
89 public HTML2TextReader(Reader reader, TextPresentation presentation) {
|
|
|
90 super(new PushbackReader(reader));
|
|
|
91 fTextPresentation= presentation;
|
|
|
92 }
|
|
|
93
|
|
|
94 public int read() throws IOException {
|
|
|
95 int c= super.read();
|
|
|
96 if (c !is -1)
|
|
|
97 ++ fCounter;
|
|
|
98 return c;
|
|
|
99 }
|
|
|
100
|
|
|
101 protected void startBold() {
|
|
|
102 if (fBold is 0)
|
|
|
103 fStartOffset= fCounter;
|
|
|
104 ++ fBold;
|
|
|
105 }
|
|
|
106
|
|
|
107 protected void startPreformattedText() {
|
|
|
108 fIsPreformattedText= true;
|
|
|
109 setSkipWhitespace(false);
|
|
|
110 }
|
|
|
111
|
|
|
112 protected void stopPreformattedText() {
|
|
|
113 fIsPreformattedText= false;
|
|
|
114 setSkipWhitespace(true);
|
|
|
115 }
|
|
|
116
|
|
|
117 protected void stopBold() {
|
|
|
118 -- fBold;
|
|
|
119 if (fBold is 0) {
|
|
|
120 if (fTextPresentation !is null) {
|
|
|
121 fTextPresentation.addStyleRange(new StyleRange(fStartOffset, fCounter - fStartOffset, null, null, DWT.BOLD));
|
|
|
122 }
|
|
|
123 fStartOffset= -1;
|
|
|
124 }
|
|
|
125 }
|
|
|
126
|
|
|
127 /*
|
|
|
128 * @see dwtx.jdt.internal.ui.text.SubstitutionTextReader#computeSubstitution(int)
|
|
|
129 */
|
|
|
130 protected String computeSubstitution(int c) throws IOException {
|
|
|
131
|
|
|
132 if (c is '<')
|
|
|
133 return processHTMLTag();
|
|
|
134 else if (fIgnore)
|
|
|
135 return EMPTY_STRING;
|
|
|
136 else if (c is '&')
|
|
|
137 return processEntity();
|
|
|
138 else if (fIsPreformattedText)
|
|
|
139 return processPreformattedText(c);
|
|
|
140
|
|
|
141 return null;
|
|
|
142 }
|
|
|
143
|
|
|
144 private String html2Text(String html) {
|
|
|
145
|
|
|
146 if (html is null || html.length() is 0)
|
|
|
147 return EMPTY_STRING;
|
|
|
148
|
|
|
149 html= html.toLowerCase();
|
|
|
150
|
|
|
151 String tag= html;
|
|
|
152 if ('/' is tag.charAt(0))
|
|
|
153 tag= tag.substring(1);
|
|
|
154
|
|
|
155 if (!fgTags.contains(tag))
|
|
|
156 return EMPTY_STRING;
|
|
|
157
|
|
|
158
|
|
|
159 if ("pre".equals(html)) { //$NON-NLS-1$
|
|
|
160 startPreformattedText();
|
|
|
161 return EMPTY_STRING;
|
|
|
162 }
|
|
|
163
|
|
|
164 if ("/pre".equals(html)) { //$NON-NLS-1$
|
|
|
165 stopPreformattedText();
|
|
|
166 return EMPTY_STRING;
|
|
|
167 }
|
|
|
168
|
|
|
169 if (fIsPreformattedText)
|
|
|
170 return EMPTY_STRING;
|
|
|
171
|
|
|
172 if ("b".equals(html)) { //$NON-NLS-1$
|
|
|
173 startBold();
|
|
|
174 return EMPTY_STRING;
|
|
|
175 }
|
|
|
176
|
|
|
177 if ((html.length() > 1 && html.charAt(0) is 'h' && Character.isDigit(html.charAt(1))) || "dt".equals(html)) { //$NON-NLS-1$
|
|
|
178 startBold();
|
|
|
179 return EMPTY_STRING;
|
|
|
180 }
|
|
|
181
|
|
|
182 if ("dl".equals(html)) //$NON-NLS-1$
|
|
|
183 return LINE_DELIM;
|
|
|
184
|
|
|
185 if ("dd".equals(html)) //$NON-NLS-1$
|
|
|
186 return "\t"; //$NON-NLS-1$
|
|
|
187
|
|
|
188 if ("li".equals(html)) //$NON-NLS-1$
|
|
|
189 // FIXME: this hard-coded prefix does not work for RTL languages, see https://bugs.eclipse.org/bugs/show_bug.cgi?id=91682
|
|
|
190 return LINE_DELIM + HTMLMessages.getString("HTML2TextReader.listItemPrefix"); //$NON-NLS-1$
|
|
|
191
|
|
|
192 if ("/b".equals(html)) { //$NON-NLS-1$
|
|
|
193 stopBold();
|
|
|
194 return EMPTY_STRING;
|
|
|
195 }
|
|
|
196
|
|
|
197 if ("p".equals(html)) { //$NON-NLS-1$
|
|
|
198 fInParagraph= true;
|
|
|
199 return LINE_DELIM;
|
|
|
200 }
|
|
|
201
|
|
|
202 if ("br".equals(html) || "br/".equals(html) || "div".equals(html)) //$NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$
|
|
|
203 return LINE_DELIM;
|
|
|
204
|
|
|
205 if ("/p".equals(html)) { //$NON-NLS-1$
|
|
|
206 bool inParagraph= fInParagraph;
|
|
|
207 fInParagraph= false;
|
|
|
208 return inParagraph ? EMPTY_STRING : LINE_DELIM;
|
|
|
209 }
|
|
|
210
|
|
|
211 if ((html.startsWith("/h") && html.length() > 2 && Character.isDigit(html.charAt(2))) || "/dt".equals(html)) { //$NON-NLS-1$ //$NON-NLS-2$
|
|
|
212 stopBold();
|
|
|
213 return LINE_DELIM;
|
|
|
214 }
|
|
|
215
|
|
|
216 if ("/dd".equals(html)) //$NON-NLS-1$
|
|
|
217 return LINE_DELIM;
|
|
|
218
|
|
|
219 if ("head".equals(html) && !fHeaderDetected) { //$NON-NLS-1$
|
|
|
220 fHeaderDetected= true;
|
|
|
221 fIgnore= true;
|
|
|
222 return EMPTY_STRING;
|
|
|
223 }
|
|
|
224
|
|
|
225 if ("/head".equals(html) && fHeaderDetected && fIgnore) { //$NON-NLS-1$
|
|
|
226 fIgnore= false;
|
|
|
227 return EMPTY_STRING;
|
|
|
228 }
|
|
|
229
|
|
|
230 return EMPTY_STRING;
|
|
|
231 }
|
|
|
232
|
|
|
233 /*
|
|
|
234 * A '<' has been read. Process a html tag
|
|
|
235 */
|
|
|
236 private String processHTMLTag() throws IOException {
|
|
|
237
|
|
|
238 StringBuffer buf= new StringBuffer();
|
|
|
239 int ch;
|
|
|
240 do {
|
|
|
241
|
|
|
242 ch= nextChar();
|
|
|
243
|
|
|
244 while (ch !is -1 && ch !is '>') {
|
|
|
245 buf.append(Character.toLowerCase((char) ch));
|
|
|
246 ch= nextChar();
|
|
|
247 if (ch is '"'){
|
|
|
248 buf.append(Character.toLowerCase((char) ch));
|
|
|
249 ch= nextChar();
|
|
|
250 while (ch !is -1 && ch !is '"'){
|
|
|
251 buf.append(Character.toLowerCase((char) ch));
|
|
|
252 ch= nextChar();
|
|
|
253 }
|
|
|
254 }
|
|
|
255 if (ch is '<' && !isInComment(buf)) {
|
|
|
256 unread(ch);
|
|
|
257 return '<' + buf.toString();
|
|
|
258 }
|
|
|
259 }
|
|
|
260
|
|
|
261 if (ch is -1)
|
|
|
262 return null;
|
|
|
263
|
|
|
264 if (!isInComment(buf) || isCommentEnd(buf)) {
|
|
|
265 break;
|
|
|
266 }
|
|
|
267 // unfinished comment
|
|
|
268 buf.append((char) ch);
|
|
|
269 } while (true);
|
|
|
270
|
|
|
271 return html2Text(buf.toString());
|
|
|
272 }
|
|
|
273
|
|
|
274 private static bool isInComment(StringBuffer buf) {
|
|
|
275 return buf.length() >= 3 && "!--".equals(buf.substring(0, 3)); //$NON-NLS-1$
|
|
|
276 }
|
|
|
277
|
|
|
278 private static bool isCommentEnd(StringBuffer buf) {
|
|
|
279 int tagLen= buf.length();
|
|
|
280 return tagLen >= 5 && "--".equals(buf.substring(tagLen - 2)); //$NON-NLS-1$
|
|
|
281 }
|
|
|
282
|
|
|
283 private String processPreformattedText(int c) {
|
|
|
284 if (c is '\r' || c is '\n')
|
|
|
285 fCounter++;
|
|
|
286 return null;
|
|
|
287 }
|
|
|
288
|
|
|
289
|
|
|
290 private void unread(int ch) throws IOException {
|
|
|
291 ((PushbackReader) getReader()).unread(ch);
|
|
|
292 }
|
|
|
293
|
|
|
294 protected String entity2Text(String symbol) {
|
|
|
295 if (symbol.length() > 1 && symbol.charAt(0) is '#') {
|
|
|
296 int ch;
|
|
|
297 try {
|
|
|
298 if (symbol.charAt(1) is 'x') {
|
|
|
299 ch= Integer.parseInt(symbol.substring(2), 16);
|
|
|
300 } else {
|
|
|
301 ch= Integer.parseInt(symbol.substring(1), 10);
|
|
|
302 }
|
|
|
303 return EMPTY_STRING + (char)ch;
|
|
|
304 } catch (NumberFormatException e) {
|
|
|
305 }
|
|
|
306 } else {
|
|
|
307 String str= (String) fgEntityLookup.get(symbol);
|
|
|
308 if (str !is null) {
|
|
|
309 return str;
|
|
|
310 }
|
|
|
311 }
|
|
|
312 return "&" + symbol; // not found //$NON-NLS-1$
|
|
|
313 }
|
|
|
314
|
|
|
315 /*
|
|
|
316 * A '&' has been read. Process a entity
|
|
|
317 */
|
|
|
318 private String processEntity() throws IOException {
|
|
|
319 StringBuffer buf= new StringBuffer();
|
|
|
320 int ch= nextChar();
|
|
|
321 while (Character.isLetterOrDigit((char)ch) || ch is '#') {
|
|
|
322 buf.append((char) ch);
|
|
|
323 ch= nextChar();
|
|
|
324 }
|
|
|
325
|
|
|
326 if (ch is ';')
|
|
|
327 return entity2Text(buf.toString());
|
|
|
328
|
|
|
329 buf.insert(0, '&');
|
|
|
330 if (ch !is -1)
|
|
|
331 buf.append((char) ch);
|
|
|
332 return buf.toString();
|
|
|
333 }
|
|
|
334 }
|
