001    /*
002     * Sonar, open source software quality management tool.
003     * Copyright (C) 2009 SonarSource SA
004     * mailto:contact AT sonarsource DOT com
005     *
006     * Sonar is free software; you can redistribute it and/or
007     * modify it under the terms of the GNU Lesser General Public
008     * License as published by the Free Software Foundation; either
009     * version 3 of the License, or (at your option) any later version.
010     *
011     * Sonar is distributed in the hope that it will be useful,
012     * but WITHOUT ANY WARRANTY; without even the implied warranty of
013     * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
014     * Lesser General Public License for more details.
015     *
016     * You should have received a copy of the GNU Lesser General Public
017     * License along with Sonar; if not, write to the Free Software
018     * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02
019     */
020    package org.sonar.squid.recognizer;
021    
022    import java.util.ArrayList;
023    import java.util.List;
024    
025    public class CodeRecognizer {
026    
027      private LanguageFootprint language;
028      private double            threshold;
029    
030      public CodeRecognizer(double threshold, LanguageFootprint language) {
031        this.language = language;
032        this.threshold = threshold;
033      }
034    
035      public final double recognition(String line) {
036        double probability = 0;
037        for (Detector pattern : language.getDetectors()) {
038          probability = 1 - ((1 - probability) * (1 - pattern.recognition(line)));
039        }
040        return probability;
041      }
042    
043      public final List<String> extractCodeLines(List<String> lines) {
044        ArrayList<String> codeLines = new ArrayList<String>();
045        for (String line : lines) {
046          if (recognition(line) >= threshold) {
047            codeLines.add(line);
048          }
049        }
050        return codeLines;
051      }
052    
053      public final boolean isLineOfCode(String line) {
054        return recognition(line) - threshold > 0;
055      }
056    }