2
|
1 |
/*
|
5506
|
2 |
* Copyright (c) 1998, 2000, Oracle and/or its affiliates. All rights reserved.
|
2
|
3 |
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
|
|
4 |
*
|
|
5 |
* This code is free software; you can redistribute it and/or modify it
|
|
6 |
* under the terms of the GNU General Public License version 2 only, as
|
5506
|
7 |
* published by the Free Software Foundation. Oracle designates this
|
2
|
8 |
* particular file as subject to the "Classpath" exception as provided
|
5506
|
9 |
* by Oracle in the LICENSE file that accompanied this code.
|
2
|
10 |
*
|
|
11 |
* This code is distributed in the hope that it will be useful, but WITHOUT
|
|
12 |
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
|
|
13 |
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
|
|
14 |
* version 2 for more details (a copy is included in the LICENSE file that
|
|
15 |
* accompanied this code).
|
|
16 |
*
|
|
17 |
* You should have received a copy of the GNU General Public License version
|
|
18 |
* 2 along with this work; if not, write to the Free Software Foundation,
|
|
19 |
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
|
|
20 |
*
|
5506
|
21 |
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
|
|
22 |
* or visit www.oracle.com if you need additional information or have any
|
|
23 |
* questions.
|
2
|
24 |
*/
|
|
25 |
|
|
26 |
package javax.swing.text.html.parser;
|
|
27 |
|
|
28 |
/**
|
|
29 |
* A content model state. This is basically a list of pointers to
|
|
30 |
* the BNF expression representing the model (the ContentModel).
|
|
31 |
* Each element in a DTD has a content model which describes the
|
|
32 |
* elements that may occur inside, and the order in which they can
|
|
33 |
* occur.
|
|
34 |
* <p>
|
|
35 |
* Each time a token is reduced a new state is created.
|
|
36 |
* <p>
|
|
37 |
* See Annex H on page 556 of the SGML handbook for more information.
|
|
38 |
*
|
|
39 |
* @see Parser
|
|
40 |
* @see DTD
|
|
41 |
* @see Element
|
|
42 |
* @see ContentModel
|
|
43 |
* @author Arthur van Hoff
|
|
44 |
*/
|
|
45 |
class ContentModelState {
|
|
46 |
ContentModel model;
|
|
47 |
long value;
|
|
48 |
ContentModelState next;
|
|
49 |
|
|
50 |
/**
|
|
51 |
* Create a content model state for a content model.
|
|
52 |
*/
|
|
53 |
public ContentModelState(ContentModel model) {
|
|
54 |
this(model, null, 0);
|
|
55 |
}
|
|
56 |
|
|
57 |
/**
|
|
58 |
* Create a content model state for a content model given the
|
|
59 |
* remaining state that needs to be reduce.
|
|
60 |
*/
|
|
61 |
ContentModelState(Object content, ContentModelState next) {
|
|
62 |
this(content, next, 0);
|
|
63 |
}
|
|
64 |
|
|
65 |
/**
|
|
66 |
* Create a content model state for a content model given the
|
|
67 |
* remaining state that needs to be reduce.
|
|
68 |
*/
|
|
69 |
ContentModelState(Object content, ContentModelState next, long value) {
|
|
70 |
this.model = (ContentModel)content;
|
|
71 |
this.next = next;
|
|
72 |
this.value = value;
|
|
73 |
}
|
|
74 |
|
|
75 |
/**
|
|
76 |
* Return the content model that is relevant to the current state.
|
|
77 |
*/
|
|
78 |
public ContentModel getModel() {
|
|
79 |
ContentModel m = model;
|
|
80 |
for (int i = 0; i < value; i++) {
|
|
81 |
if (m.next != null) {
|
|
82 |
m = m.next;
|
|
83 |
} else {
|
|
84 |
return null;
|
|
85 |
}
|
|
86 |
}
|
|
87 |
return m;
|
|
88 |
}
|
|
89 |
|
|
90 |
/**
|
|
91 |
* Check if the state can be terminated. That is there are no more
|
|
92 |
* tokens required in the input stream.
|
|
93 |
* @return true if the model can terminate without further input
|
|
94 |
*/
|
|
95 |
public boolean terminate() {
|
|
96 |
switch (model.type) {
|
|
97 |
case '+':
|
|
98 |
if ((value == 0) && !(model).empty()) {
|
|
99 |
return false;
|
|
100 |
}
|
|
101 |
case '*':
|
|
102 |
case '?':
|
|
103 |
return (next == null) || next.terminate();
|
|
104 |
|
|
105 |
case '|':
|
|
106 |
for (ContentModel m = (ContentModel)model.content ; m != null ; m = m.next) {
|
|
107 |
if (m.empty()) {
|
|
108 |
return (next == null) || next.terminate();
|
|
109 |
}
|
|
110 |
}
|
|
111 |
return false;
|
|
112 |
|
|
113 |
case '&': {
|
|
114 |
ContentModel m = (ContentModel)model.content;
|
|
115 |
|
|
116 |
for (int i = 0 ; m != null ; i++, m = m.next) {
|
|
117 |
if ((value & (1L << i)) == 0) {
|
|
118 |
if (!m.empty()) {
|
|
119 |
return false;
|
|
120 |
}
|
|
121 |
}
|
|
122 |
}
|
|
123 |
return (next == null) || next.terminate();
|
|
124 |
}
|
|
125 |
|
|
126 |
case ',': {
|
|
127 |
ContentModel m = (ContentModel)model.content;
|
|
128 |
for (int i = 0 ; i < value ; i++, m = m.next);
|
|
129 |
|
|
130 |
for (; (m != null) && m.empty() ; m = m.next);
|
|
131 |
if (m != null) {
|
|
132 |
return false;
|
|
133 |
}
|
|
134 |
return (next == null) || next.terminate();
|
|
135 |
}
|
|
136 |
|
|
137 |
default:
|
|
138 |
return false;
|
|
139 |
}
|
|
140 |
}
|
|
141 |
|
|
142 |
/**
|
|
143 |
* Check if the state can be terminated. That is there are no more
|
|
144 |
* tokens required in the input stream.
|
|
145 |
* @return the only possible element that can occur next
|
|
146 |
*/
|
|
147 |
public Element first() {
|
|
148 |
switch (model.type) {
|
|
149 |
case '*':
|
|
150 |
case '?':
|
|
151 |
case '|':
|
|
152 |
case '&':
|
|
153 |
return null;
|
|
154 |
|
|
155 |
case '+':
|
|
156 |
return model.first();
|
|
157 |
|
|
158 |
case ',': {
|
|
159 |
ContentModel m = (ContentModel)model.content;
|
|
160 |
for (int i = 0 ; i < value ; i++, m = m.next);
|
|
161 |
return m.first();
|
|
162 |
}
|
|
163 |
|
|
164 |
default:
|
|
165 |
return model.first();
|
|
166 |
}
|
|
167 |
}
|
|
168 |
|
|
169 |
/**
|
|
170 |
* Advance this state to a new state. An exception is thrown if the
|
|
171 |
* token is illegal at this point in the content model.
|
|
172 |
* @return next state after reducing a token
|
|
173 |
*/
|
|
174 |
public ContentModelState advance(Object token) {
|
|
175 |
switch (model.type) {
|
|
176 |
case '+':
|
|
177 |
if (model.first(token)) {
|
|
178 |
return new ContentModelState(model.content,
|
|
179 |
new ContentModelState(model, next, value + 1)).advance(token);
|
|
180 |
}
|
|
181 |
if (value != 0) {
|
|
182 |
if (next != null) {
|
|
183 |
return next.advance(token);
|
|
184 |
} else {
|
|
185 |
return null;
|
|
186 |
}
|
|
187 |
}
|
|
188 |
break;
|
|
189 |
|
|
190 |
case '*':
|
|
191 |
if (model.first(token)) {
|
|
192 |
return new ContentModelState(model.content, this).advance(token);
|
|
193 |
}
|
|
194 |
if (next != null) {
|
|
195 |
return next.advance(token);
|
|
196 |
} else {
|
|
197 |
return null;
|
|
198 |
}
|
|
199 |
|
|
200 |
case '?':
|
|
201 |
if (model.first(token)) {
|
|
202 |
return new ContentModelState(model.content, next).advance(token);
|
|
203 |
}
|
|
204 |
if (next != null) {
|
|
205 |
return next.advance(token);
|
|
206 |
} else {
|
|
207 |
return null;
|
|
208 |
}
|
|
209 |
|
|
210 |
case '|':
|
|
211 |
for (ContentModel m = (ContentModel)model.content ; m != null ; m = m.next) {
|
|
212 |
if (m.first(token)) {
|
|
213 |
return new ContentModelState(m, next).advance(token);
|
|
214 |
}
|
|
215 |
}
|
|
216 |
break;
|
|
217 |
|
|
218 |
case ',': {
|
|
219 |
ContentModel m = (ContentModel)model.content;
|
|
220 |
for (int i = 0 ; i < value ; i++, m = m.next);
|
|
221 |
|
|
222 |
if (m.first(token) || m.empty()) {
|
|
223 |
if (m.next == null) {
|
|
224 |
return new ContentModelState(m, next).advance(token);
|
|
225 |
} else {
|
|
226 |
return new ContentModelState(m,
|
|
227 |
new ContentModelState(model, next, value + 1)).advance(token);
|
|
228 |
}
|
|
229 |
}
|
|
230 |
break;
|
|
231 |
}
|
|
232 |
|
|
233 |
case '&': {
|
|
234 |
ContentModel m = (ContentModel)model.content;
|
|
235 |
boolean complete = true;
|
|
236 |
|
|
237 |
for (int i = 0 ; m != null ; i++, m = m.next) {
|
|
238 |
if ((value & (1L << i)) == 0) {
|
|
239 |
if (m.first(token)) {
|
|
240 |
return new ContentModelState(m,
|
|
241 |
new ContentModelState(model, next, value | (1L << i))).advance(token);
|
|
242 |
}
|
|
243 |
if (!m.empty()) {
|
|
244 |
complete = false;
|
|
245 |
}
|
|
246 |
}
|
|
247 |
}
|
|
248 |
if (complete) {
|
|
249 |
if (next != null) {
|
|
250 |
return next.advance(token);
|
|
251 |
} else {
|
|
252 |
return null;
|
|
253 |
}
|
|
254 |
}
|
|
255 |
break;
|
|
256 |
}
|
|
257 |
|
|
258 |
default:
|
|
259 |
if (model.content == token) {
|
|
260 |
if (next == null && (token instanceof Element) &&
|
|
261 |
((Element)token).content != null) {
|
|
262 |
return new ContentModelState(((Element)token).content);
|
|
263 |
}
|
|
264 |
return next;
|
|
265 |
}
|
|
266 |
// PENDING: Currently we don't correctly deal with optional start
|
|
267 |
// tags. This can most notably be seen with the 4.01 spec where
|
|
268 |
// TBODY's start and end tags are optional.
|
|
269 |
// Uncommenting this and the PENDING in ContentModel will
|
|
270 |
// correctly skip the omit tags, but the delegate is not notified.
|
|
271 |
// Some additional API needs to be added to track skipped tags,
|
|
272 |
// and this can then be added back.
|
|
273 |
/*
|
|
274 |
if ((model.content instanceof Element)) {
|
|
275 |
Element e = (Element)model.content;
|
|
276 |
|
|
277 |
if (e.omitStart() && e.content != null) {
|
|
278 |
return new ContentModelState(e.content, next).advance(
|
|
279 |
token);
|
|
280 |
}
|
|
281 |
}
|
|
282 |
*/
|
|
283 |
}
|
|
284 |
|
|
285 |
// We used to throw this exception at this point. However, it
|
|
286 |
// was determined that throwing this exception was more expensive
|
|
287 |
// than returning null, and we could not justify to ourselves why
|
|
288 |
// it was necessary to throw an exception, rather than simply
|
|
289 |
// returning null. I'm leaving it in a commented out state so
|
|
290 |
// that it can be easily restored if the situation ever arises.
|
|
291 |
//
|
|
292 |
// throw new IllegalArgumentException("invalid token: " + token);
|
|
293 |
return null;
|
|
294 |
}
|
|
295 |
}
|