001    /*
002     * SonarQube, open source software quality management tool.
003     * Copyright (C) 2008-2014 SonarSource
004     * mailto:contact AT sonarsource DOT com
005     *
006     * SonarQube 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     * SonarQube 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 License
017     * along with this program; if not, write to the Free Software Foundation,
018     * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
019     */
020    package org.sonar.xoo.lang;
021    
022    import com.google.common.base.Splitter;
023    import com.google.common.collect.Lists;
024    import org.apache.commons.io.FileUtils;
025    import org.sonar.api.CoreProperties;
026    import org.sonar.api.batch.fs.FilePredicates;
027    import org.sonar.api.batch.fs.InputFile;
028    import org.sonar.api.batch.sensor.Sensor;
029    import org.sonar.api.batch.sensor.SensorContext;
030    import org.sonar.api.batch.sensor.SensorDescriptor;
031    import org.sonar.api.batch.sensor.duplication.DuplicationTokenBuilder;
032    import org.sonar.xoo.Xoo;
033    
034    import java.io.File;
035    import java.io.IOException;
036    import java.util.List;
037    
038    /**
039     * Tokenize xoo files (separator is whitespace) for duplication detection
040     */
041    public class XooTokenizerSensor implements Sensor {
042    
043      private void computeTokens(InputFile inputFile, SensorContext context) {
044        DuplicationTokenBuilder tokenBuilder = context.duplicationTokenBuilder(inputFile);
045        File ioFile = inputFile.file();
046        int lineId = 0;
047        try {
048          for (String line : FileUtils.readLines(ioFile)) {
049            lineId++;
050            for (String token : Splitter.on(" ").split(line)) {
051              tokenBuilder.addToken(lineId, token);
052            }
053          }
054          tokenBuilder.done();
055        } catch (IOException e) {
056          throw new IllegalStateException("Unable to read file " + ioFile, e);
057        }
058      }
059    
060      @Override
061      public void describe(SensorDescriptor descriptor) {
062        descriptor
063          .name("Xoo Tokenizer Sensor")
064          .workOnLanguages(Xoo.KEY)
065          .workOnFileTypes(InputFile.Type.MAIN);
066      }
067    
068      @Override
069      public void execute(SensorContext context) {
070        String[] cpdExclusions = context.settings().getStringArray(CoreProperties.CPD_EXCLUSIONS);
071        FilePredicates p = context.fileSystem().predicates();
072        List<InputFile> sourceFiles = Lists.newArrayList(context.fileSystem().inputFiles(p.and(
073          p.hasType(InputFile.Type.MAIN),
074          p.hasLanguage(Xoo.KEY),
075          p.doesNotMatchPathPatterns(cpdExclusions)
076          )));
077        if (sourceFiles.isEmpty()) {
078          return;
079        }
080        for (InputFile file : sourceFiles) {
081          computeTokens(file, context);
082        }
083      }
084    }