author | attila |
Wed, 05 Jun 2013 10:44:32 +0200 | |
changeset 17981 | 9b8e085aa1fe |
parent 17523 | cb4a7c901e0d |
child 18317 | 2f5434c9c9fd |
permissions | -rw-r--r-- |
16147 | 1 |
/* |
16151 | 2 |
* Copyright (c) 2010, 2013, Oracle and/or its affiliates. All rights reserved. |
16147 | 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 |
|
7 |
* published by the Free Software Foundation. Oracle designates this |
|
8 |
* particular file as subject to the "Classpath" exception as provided |
|
9 |
* by Oracle in the LICENSE file that accompanied this code. |
|
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 |
* |
|
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. |
|
24 |
*/ |
|
25 |
||
26 |
package jdk.nashorn.internal.parser; |
|
27 |
||
28 |
import static jdk.nashorn.internal.parser.TokenType.COLON; |
|
29 |
import static jdk.nashorn.internal.parser.TokenType.COMMARIGHT; |
|
30 |
import static jdk.nashorn.internal.parser.TokenType.EOF; |
|
31 |
import static jdk.nashorn.internal.parser.TokenType.ESCSTRING; |
|
32 |
import static jdk.nashorn.internal.parser.TokenType.RBRACE; |
|
33 |
import static jdk.nashorn.internal.parser.TokenType.RBRACKET; |
|
34 |
import static jdk.nashorn.internal.parser.TokenType.STRING; |
|
35 |
||
36 |
import java.util.ArrayList; |
|
37 |
import java.util.List; |
|
38 |
import jdk.nashorn.internal.ir.LiteralNode; |
|
39 |
import jdk.nashorn.internal.ir.Node; |
|
40 |
import jdk.nashorn.internal.ir.ObjectNode; |
|
41 |
import jdk.nashorn.internal.ir.PropertyNode; |
|
42 |
import jdk.nashorn.internal.ir.UnaryNode; |
|
43 |
import jdk.nashorn.internal.runtime.ErrorManager; |
|
44 |
import jdk.nashorn.internal.runtime.Source; |
|
45 |
||
46 |
/** |
|
47 |
* Parses JSON text and returns the corresponding IR node. This is derived from the objectLiteral production of the main parser. |
|
48 |
* |
|
49 |
* See: 15.12.1.2 The JSON Syntactic Grammar |
|
50 |
*/ |
|
51 |
public class JSONParser extends AbstractParser { |
|
52 |
||
53 |
/** |
|
54 |
* Constructor |
|
55 |
* @param source the source |
|
56 |
* @param errors the error manager |
|
57 |
* @param strict are we in strict mode |
|
58 |
*/ |
|
59 |
public JSONParser(final Source source, final ErrorManager errors, final boolean strict) { |
|
60 |
super(source, errors, strict); |
|
61 |
} |
|
62 |
||
63 |
/** |
|
64 |
* Implementation of the Quote(value) operation as defined in the ECMA script spec |
|
65 |
* It wraps a String value in double quotes and escapes characters within in |
|
66 |
* |
|
67 |
* @param value string to quote |
|
68 |
* |
|
69 |
* @return quoted and escaped string |
|
70 |
*/ |
|
71 |
public static String quote(final String value) { |
|
72 |
||
73 |
final StringBuilder product = new StringBuilder(); |
|
74 |
||
75 |
product.append("\""); |
|
76 |
||
77 |
for (final char ch : value.toCharArray()) { |
|
78 |
// TODO: should use a table? |
|
79 |
switch (ch) { |
|
80 |
case '\\': |
|
81 |
product.append("\\\\"); |
|
82 |
break; |
|
83 |
case '"': |
|
84 |
product.append("\\\""); |
|
85 |
break; |
|
86 |
case '\b': |
|
87 |
product.append("\\b"); |
|
88 |
break; |
|
89 |
case '\f': |
|
90 |
product.append("\\f"); |
|
91 |
break; |
|
92 |
case '\n': |
|
93 |
product.append("\\n"); |
|
94 |
break; |
|
95 |
case '\r': |
|
96 |
product.append("\\r"); |
|
97 |
break; |
|
98 |
case '\t': |
|
99 |
product.append("\\t"); |
|
100 |
break; |
|
101 |
default: |
|
102 |
if (ch < ' ') { |
|
103 |
product.append(Lexer.unicodeEscape(ch)); |
|
104 |
break; |
|
105 |
} |
|
106 |
||
107 |
product.append(ch); |
|
108 |
break; |
|
109 |
} |
|
110 |
} |
|
111 |
||
112 |
product.append("\""); |
|
113 |
||
114 |
return product.toString(); |
|
115 |
} |
|
116 |
||
117 |
/** |
|
118 |
* Public parsed method - start lexing a new token stream for |
|
119 |
* a JSON script |
|
120 |
* |
|
121 |
* @return the JSON literal |
|
122 |
*/ |
|
123 |
public Node parse() { |
|
124 |
stream = new TokenStream(); |
|
125 |
||
126 |
lexer = new Lexer(source, stream) { |
|
127 |
||
128 |
@Override |
|
129 |
protected boolean skipComments() { |
|
130 |
return false; |
|
131 |
} |
|
132 |
||
133 |
@Override |
|
134 |
protected boolean isStringDelimiter(final char ch) { |
|
135 |
return ch == '\"'; |
|
136 |
} |
|
137 |
||
138 |
@Override |
|
139 |
protected boolean isWhitespace(final char ch) { |
|
140 |
return Lexer.isJsonWhitespace(ch); |
|
141 |
} |
|
142 |
||
143 |
@Override |
|
144 |
protected boolean isEOL(final char ch) { |
|
145 |
return Lexer.isJsonEOL(ch); |
|
146 |
} |
|
147 |
}; |
|
148 |
||
149 |
k = -1; |
|
150 |
||
151 |
next(); |
|
152 |
||
153 |
final Node resultNode = jsonLiteral(); |
|
154 |
expect(EOF); |
|
155 |
||
156 |
return resultNode; |
|
157 |
} |
|
158 |
||
159 |
@SuppressWarnings("fallthrough") |
|
160 |
private LiteralNode<?> getStringLiteral() { |
|
161 |
final LiteralNode<?> literal = getLiteral(); |
|
162 |
final String str = (String)literal.getValue(); |
|
163 |
||
164 |
for (int i = 0; i < str.length(); i++) { |
|
165 |
final char ch = str.charAt(i); |
|
166 |
switch (ch) { |
|
167 |
default: |
|
168 |
if (ch > 0x001f) { |
|
169 |
break; |
|
170 |
} |
|
171 |
case '"': |
|
172 |
case '\\': |
|
17233 | 173 |
throw error(AbstractParser.message("unexpected.token", str)); |
16147 | 174 |
} |
175 |
} |
|
176 |
||
177 |
return literal; |
|
178 |
} |
|
179 |
||
180 |
/** |
|
181 |
* Parse a JSON literal from the token stream |
|
182 |
* @return the JSON literal as a Node |
|
183 |
*/ |
|
184 |
private Node jsonLiteral() { |
|
185 |
final long literalToken = token; |
|
186 |
||
187 |
switch (type) { |
|
188 |
case STRING: |
|
189 |
return getStringLiteral(); |
|
190 |
case ESCSTRING: |
|
191 |
case DECIMAL: |
|
192 |
case FLOATING: |
|
193 |
return getLiteral(); |
|
194 |
case FALSE: |
|
195 |
next(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
196 |
return LiteralNode.newInstance(literalToken, finish, false); |
16147 | 197 |
case TRUE: |
198 |
next(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
199 |
return LiteralNode.newInstance(literalToken, finish, true); |
16147 | 200 |
case NULL: |
201 |
next(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
202 |
return LiteralNode.newInstance(literalToken, finish); |
16147 | 203 |
case LBRACKET: |
204 |
return arrayLiteral(); |
|
205 |
case LBRACE: |
|
206 |
return objectLiteral(); |
|
207 |
/* |
|
208 |
* A.8.1 JSON Lexical Grammar |
|
209 |
* |
|
210 |
* JSONNumber :: See 15.12.1.1 |
|
211 |
* -opt DecimalIntegerLiteral JSONFractionopt ExponentPartopt |
|
212 |
*/ |
|
213 |
case SUB: |
|
214 |
next(); |
|
215 |
||
216 |
final long realToken = token; |
|
217 |
final Object value = getValue(); |
|
218 |
||
219 |
if (value instanceof Number) { |
|
220 |
next(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
221 |
return new UnaryNode(literalToken, LiteralNode.newInstance(realToken, finish, (Number)value)); |
16147 | 222 |
} |
223 |
||
17233 | 224 |
throw error(AbstractParser.message("expected", "number", type.getNameOrType())); |
16147 | 225 |
default: |
226 |
break; |
|
227 |
} |
|
228 |
||
17233 | 229 |
throw error(AbstractParser.message("expected", "json literal", type.getNameOrType())); |
16147 | 230 |
} |
231 |
||
232 |
/** |
|
233 |
* Parse an array literal from the token stream |
|
234 |
* @return the array literal as a Node |
|
235 |
*/ |
|
236 |
private Node arrayLiteral() { |
|
237 |
// Unlike JavaScript array literals, elison is not permitted in JSON. |
|
238 |
||
239 |
// Capture LBRACKET token. |
|
240 |
final long arrayToken = token; |
|
241 |
// LBRACKET tested in caller. |
|
242 |
next(); |
|
243 |
||
244 |
Node result = null; |
|
245 |
// Prepare to accummulating elements. |
|
246 |
final List<Node> elements = new ArrayList<>(); |
|
247 |
||
248 |
loop: |
|
249 |
while (true) { |
|
250 |
switch (type) { |
|
251 |
case RBRACKET: |
|
252 |
next(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
253 |
result = LiteralNode.newInstance(arrayToken, finish, elements); |
16147 | 254 |
break loop; |
255 |
||
256 |
case COMMARIGHT: |
|
257 |
next(); |
|
258 |
break; |
|
259 |
||
260 |
default: |
|
261 |
// Add expression element. |
|
262 |
elements.add(jsonLiteral()); |
|
263 |
// Comma between array elements is mandatory in JSON. |
|
264 |
if (type != COMMARIGHT && type != RBRACKET) { |
|
17233 | 265 |
throw error(AbstractParser.message("expected", ", or ]", type.getNameOrType())); |
16147 | 266 |
} |
267 |
break; |
|
268 |
} |
|
269 |
} |
|
270 |
||
271 |
return result; |
|
272 |
} |
|
273 |
||
274 |
/** |
|
275 |
* Parse an object literal from the token stream |
|
276 |
* @return the object literal as a Node |
|
277 |
*/ |
|
278 |
private Node objectLiteral() { |
|
279 |
// Capture LBRACE token. |
|
280 |
final long objectToken = token; |
|
281 |
// LBRACE tested in caller. |
|
282 |
next(); |
|
283 |
||
284 |
// Prepare to accumulate elements. |
|
17981
9b8e085aa1fe
8015955: ObjectNode.elements should be stronger typed
attila
parents:
17523
diff
changeset
|
285 |
final List<PropertyNode> elements = new ArrayList<>(); |
16147 | 286 |
|
287 |
// Create a block for the object literal. |
|
288 |
loop: |
|
289 |
while (true) { |
|
290 |
switch (type) { |
|
291 |
case RBRACE: |
|
292 |
next(); |
|
293 |
break loop; |
|
294 |
||
295 |
case COMMARIGHT: |
|
296 |
next(); |
|
297 |
break; |
|
298 |
||
299 |
default: |
|
300 |
// Get and add the next property. |
|
17981
9b8e085aa1fe
8015955: ObjectNode.elements should be stronger typed
attila
parents:
17523
diff
changeset
|
301 |
final PropertyNode property = propertyAssignment(); |
16147 | 302 |
elements.add(property); |
303 |
||
304 |
// Comma between property assigments is mandatory in JSON. |
|
305 |
if (type != RBRACE && type != COMMARIGHT) { |
|
17233 | 306 |
throw error(AbstractParser.message("expected", ", or }", type.getNameOrType())); |
16147 | 307 |
} |
308 |
break; |
|
309 |
} |
|
310 |
} |
|
311 |
||
312 |
// Construct new object literal. |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
313 |
return new ObjectNode(objectToken, finish, elements); |
16147 | 314 |
} |
315 |
||
316 |
/** |
|
317 |
* Parse a property assignment from the token stream |
|
318 |
* @return the property assignment as a Node |
|
319 |
*/ |
|
17981
9b8e085aa1fe
8015955: ObjectNode.elements should be stronger typed
attila
parents:
17523
diff
changeset
|
320 |
private PropertyNode propertyAssignment() { |
16147 | 321 |
// Capture firstToken. |
322 |
final long propertyToken = token; |
|
323 |
LiteralNode<?> name = null; |
|
324 |
||
325 |
if (type == STRING) { |
|
326 |
name = getStringLiteral(); |
|
327 |
} else if (type == ESCSTRING) { |
|
328 |
name = getLiteral(); |
|
329 |
} |
|
330 |
||
331 |
if (name != null) { |
|
332 |
expect(COLON); |
|
333 |
final Node value = jsonLiteral(); |
|
17523
cb4a7c901e0d
8013913: Removed Source field from all nodes except FunctionNode in order to save footprint
lagergren
parents:
17233
diff
changeset
|
334 |
return new PropertyNode(propertyToken, value.getFinish(), name, value, null, null); |
16147 | 335 |
} |
336 |
||
337 |
// Raise an error. |
|
17233 | 338 |
throw error(AbstractParser.message("expected", "string", type.getNameOrType())); |
16147 | 339 |
} |
340 |
||
341 |
} |