xref: /OpenGrok/opengrok-indexer/src/main/jflex/analysis/document/TroffFullTokenizer.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) 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.document;
26 
27 import java.util.Locale;
28 import org.opengrok.indexer.analysis.JFlexSymbolMatcher;
29 %%
30 
31 %public
32 %class TroffFullTokenizer
33 %extends JFlexSymbolMatcher
34 %unicode
35 %int
36 %include ../CommonLexer.lexh
37 %caseless
38 %char
39 
40 //WhiteSpace     = [ \t\f\r]+|\n
41 Identifier = [a-zA-Z_] [a-zA-Z0-9_]*
42 Number = [0-9]+|[0-9]+\.[0-9]+| "0[xX]" [0-9a-fA-F]+
43 Printable = [\@\$\%\^\&\-+=\?\.\:]
44 
45 %%
46 ^\.(EQ|in|sp|ne|rt|br|pn|ds|de|if|ig|el|ft|hy|ie|ll|ps|rm|ta|ti|NH|DT|EE)[^\n]* {}
47 ^.[a-zA-Z]{1,2} {}
48 \\f[ABCIR]  {}
49 ^"...\\\"" {}
50 
51 \\&.        {
52     onSymbolMatched(".", yychar); return yystate();
53 }
54 
55 {Identifier}|{Number}|{Printable} {
56     onSymbolMatched(yytext().toLowerCase(Locale.ROOT), yychar);
57     return yystate();
58 }
59 
60 [^]    {}
61