xref: /OpenGrok/opengrok-indexer/src/main/jflex/analysis/json/JsonSymbolTokenizer.lex (revision d219b4cea555a12b602d2d5518daa22134ad4879)
1 /*
2  * CDDL HEADER START
3  *
4  * The contents of this file are subject to the terms of the
5  * Common Development and Distribution License (the "License").
6  * You may not use this file except in compliance with the License.
7  *
8  * See LICENSE.txt included in this distribution for the specific
9  * language governing permissions and limitations under the License.
10  *
11  * When distributing Covered Code, include this CDDL HEADER in each
12  * file and include the License file at LICENSE.txt.
13  * If applicable, add the following below this CDDL HEADER, with the
14  * fields enclosed by brackets "[]" replaced with your own identifying
15  * information: Portions Copyright [yyyy] [name of copyright owner]
16  *
17  * CDDL HEADER END
18  */
19 
20 /*
21  * Copyright (c) 2017, 2021, Oracle and/or its affiliates. All rights reserved.
22  * Portions Copyright (c) 2017, Chris Fraire <cfraire@me.com>.
23  */
24 
25 /*
26  * Gets Json symbols - ignores comments, strings, keywords
27  */
28 
29 package org.opengrok.indexer.analysis.json;
30 
31 import org.opengrok.indexer.analysis.JFlexSymbolMatcher;
32 %%
33 %public
34 %class JsonSymbolTokenizer
35 %extends JFlexSymbolMatcher
36 %unicode
37 %buffer 32766
38 %int
39 %include ../CommonLexer.lexh
40 %char
41 
42 %state STRING
43 
44 %include Json.lexh
45 %%
46 
47 //TODO improve per http://www.ecma-international.org/publications/files/ECMA-ST/ECMA-404.pdf
48 
49 <YYINITIAL> {
50 {Identifier} {
51     String id = yytext();
52                 if(!Consts.kwd.contains(id)){
53                         onSymbolMatched(id, yychar);
54                         return yystate();
55 		}
56  }
57 
58  {Number}        {}
59 
60  \"     { yybegin(STRING); }
61 }
62 
63 <STRING> {
64  \\[\"\\]    {}
65  \"     { yybegin(YYINITIAL); }
66 }
67 
68 <YYINITIAL, STRING> {
69 [^]    {}
70 }
71