001 /*
002 * Copyright (C) 2012 eXo Platform SAS.
003 *
004 * This is free software; you can redistribute it and/or modify it
005 * under the terms of the GNU Lesser General Public License as
006 * published by the Free Software Foundation; either version 2.1 of
007 * the License, or (at your option) any later version.
008 *
009 * This software is distributed in the hope that it will be useful,
010 * but WITHOUT ANY WARRANTY; without even the implied warranty of
011 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
012 * Lesser General Public License for more details.
013 *
014 * You should have received a copy of the GNU Lesser General Public
015 * License along with this software; if not, write to the Free
016 * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
017 * 02110-1301 USA, or see the FSF site: http://www.fsf.org.
018 */
019 package org.crsh.cli.impl.tokenizer;
020
021 import org.crsh.cli.impl.line.LineParser;
022 import org.crsh.cli.impl.line.Quoting;
023
024 import java.util.LinkedList;
025
026 /**
027 * @author Julien Viet
028 */
029 class Automaton extends LineParser.Visitor {
030
031 /** . */
032 private Status status = Status.WHITESPACE;
033
034 /** . */
035 private final StringBuilder buffer = new StringBuilder();
036
037 /** . */
038 final LinkedList<Token> tokens = new LinkedList<Token>();
039
040 /** . */
041 private int from = 0;
042
043 /** . */
044 private int lastWhitespace = 0;
045
046 /** . */
047 private final CharSequence s;
048
049 Automaton(CharSequence s) {
050 this.s = s;
051 }
052
053 void close() {
054 if (buffer.length() > 0) {
055 if (status == Status.WHITESPACE) {
056 next(lastWhitespace);
057 } else {
058 next(s.length());
059 }
060 }
061 }
062
063 private void next(int current) {
064 Token token;
065 switch (status) {
066 case WHITESPACE:
067 token = new Token.Whitespace(from, s.subSequence(from, current).toString());
068 break;
069 case WORD:
070 token = new Token.Literal.Word(from, s.subSequence(from, current).toString(), buffer.toString());
071 break;
072 case SHORT_OPTION:
073 token = new Token.Literal.Option.Short(from, s.subSequence(from, current).toString(), buffer.toString());
074 break;
075 case LONG_OPTION:
076 token = new Token.Literal.Option.Long(from, s.subSequence(from, current).toString(), buffer.toString());
077 break;
078 default:
079 throw new AssertionError();
080 }
081 tokens.add(token);
082 status = Status.WHITESPACE;
083 buffer.setLength(0);
084 from = current;
085 }
086
087 @Override
088 public void onChar(int index, Quoting quoting, boolean backslash, char c) {
089 if (Character.isWhitespace(c) && quoting == null && !backslash) {
090 lastWhitespace = index + 1;
091 if (status != Status.WHITESPACE) {
092 next(index);
093 }
094 buffer.append(c);
095 } else {
096 switch (status) {
097 case WHITESPACE:
098 if (buffer.length() > 0) {
099 next(lastWhitespace);
100 }
101 buffer.append(c);
102 if (c == '-') {
103 status = Status.SHORT_OPTION;
104 } else {
105 status = Status.WORD;
106 }
107 break;
108 case WORD:
109 buffer.append(c);
110 break;
111 case SHORT_OPTION:
112 if (c == '-') {
113 buffer.append('-');
114 status = Status.LONG_OPTION;
115 } else if (Character.isLetter(c)) {
116 buffer.append(c);
117 } else {
118 status = Status.WORD;
119 buffer.append(c);
120 }
121 break;
122 case LONG_OPTION:
123 if (c == '-') {
124 if (buffer.length() > 2) {
125 buffer.append(c);
126 } else {
127 status = Status.WORD;
128 buffer.append(c);
129 }
130 } else if (Character.isLetter(c)) {
131 buffer.append(c);
132 } else {
133 status = Status.WORD;
134 buffer.append(c);
135 }
136 }
137 }
138 }
139
140 enum Status { WHITESPACE, WORD, SHORT_OPTION, LONG_OPTION }
141
142 }