xref: /OpenGrok/opengrok-indexer/src/main/jflex/analysis/plain/PlainFullTokenizer.lex (revision 731916a808817deb91f063ac93b2a723abef5ef0)
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) 2005, 2021, Oracle and/or its affiliates. All rights reserved.
22  * Portions Copyright (c) 2017-2018, Chris Fraire <cfraire@me.com>.
23  */
24 
25 package org.opengrok.indexer.analysis.plain;
26 
27 import java.util.Locale;
28 import org.opengrok.indexer.analysis.JFlexSymbolMatcher;
29 %%
30 %public
31 %class PlainFullTokenizer
32 %extends JFlexSymbolMatcher
33 %unicode
34 %buffer 32766
35 %int
36 %include ../CommonLexer.lexh
37 %caseless
38 %char
39 
40 //WhiteSpace     = [ \t\f\r]+|\n
41 Identifier = [a-zA-Z\p{Letter}_] [a-zA-Z\p{Letter}0-9\p{Number}_]*
42 Number = [0-9]+|[0-9]+\.[0-9]+| "0[xX]" [0-9a-fA-F]+
43 // No letters in the following, so no toLowerCase(Locale.ROOT) needed.
44 Printable = [\@\$\%\^\&\-+=\?\.\:!\[\]\{\}*\/\|#\<\>\(\),;~]
45 
46 %%
47 {Identifier}|{Number}|{Printable} {
48     onSymbolMatched(yytext().toLowerCase(Locale.ROOT), yychar);
49     return yystate();
50 }
51 [^]    {}
52