Ingester.java

Go to the documentation of this file.

/*
 * Autopsy Forensic Browser
 *
 * Copyright 2011-2017 Basis Technology Corp.
 * Contact: carrier <at> sleuthkit <dot> org
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.sleuthkit.autopsy.keywordsearch;

import java.io.BufferedReader;
import java.util.HashMap;
import java.util.Map;
import java.util.logging.Level;
import org.apache.solr.client.solrj.SolrServerException;
import org.apache.solr.common.SolrInputDocument;
import org.openide.util.NbBundle;
import org.sleuthkit.autopsy.coreutils.Logger;
import org.sleuthkit.autopsy.datamodel.ContentUtils;
import org.sleuthkit.autopsy.ingest.IngestJobContext;
import org.sleuthkit.autopsy.keywordsearch.Chunker.Chunk;
import org.sleuthkit.datamodel.AbstractFile;
import org.sleuthkit.datamodel.BlackboardArtifact;
import org.sleuthkit.datamodel.DerivedFile;
import org.sleuthkit.datamodel.Directory;
import org.sleuthkit.datamodel.File;
import org.sleuthkit.datamodel.LayoutFile;
import org.sleuthkit.datamodel.LocalFile;
import org.sleuthkit.datamodel.SlackFile;
import org.sleuthkit.datamodel.SleuthkitItemVisitor;
import org.sleuthkit.datamodel.SleuthkitVisitableItem;
import org.sleuthkit.datamodel.TskCoreException;

//JMTODO: Should this class really be a singleton?
class Ingester {

    private static final Logger logger = Logger.getLogger(Ingester.class.getName());
    private volatile boolean uncommitedIngests = false;
    private final Server solrServer = KeywordSearch.getServer();
    private static final SolrFieldsVisitor SOLR_FIELDS_VISITOR = new SolrFieldsVisitor();
    private static Ingester instance;
    private static final int SINGLE_READ_CHARS = 512;

    private Ingester() {
    }

    public static synchronized Ingester getDefault() {
        if (instance == null) {
            instance = new Ingester();
        }
        return instance;
    }

    //JMTODO: this is probably useless
    @Override
    @SuppressWarnings("FinalizeDeclaration")
    protected void finalize() throws Throwable {
        super.finalize();

        // Warn if files might have been left uncommited.
        if (uncommitedIngests) {
            logger.warning("Ingester was used to add files that it never committed."); //NON-NLS
        }
    }

    void indexMetaDataOnly(AbstractFile file) throws IngesterException {
        indexChunk("", file.getName(), getContentFields(file));
    }

    void indexMetaDataOnly(BlackboardArtifact artifact) throws IngesterException {
        indexChunk("", new ArtifactTextExtractor().getName(artifact), getContentFields(artifact));
    }

    private Map<String, String> getContentFields(SleuthkitVisitableItem item) {
        return item.accept(SOLR_FIELDS_VISITOR);
    }

    < T extends SleuthkitVisitableItem> boolean indexText(TextExtractor< T> extractor, T source, IngestJobContext context) throws Ingester.IngesterException {
        final long sourceID = extractor.getID(source);
        final String sourceName = extractor.getName(source);

        int numChunks = 0; //unknown until chunking is done

        if (extractor.isDisabled()) {
            /* some Extractors, notable the strings extractor, have options
             * which can be configured such that no extraction should be done */
            return true;
        }

        Map<String, String> fields = getContentFields(source);
        //Get a reader for the content of the given source
        try (BufferedReader reader = new BufferedReader(extractor.getReader(source));) {
            Chunker chunker = new Chunker(reader);
            for (Chunk chunk : chunker) {
                String chunkId = Server.getChunkIdString(sourceID, numChunks + 1);
                fields.put(Server.Schema.ID.toString(), chunkId);
                fields.put(Server.Schema.CHUNK_SIZE.toString(), String.valueOf(chunk.getBaseChunkLength()));
                try {
                    //add the chunk text to Solr index
                    indexChunk(chunk.toString(), sourceName, fields);
                    numChunks++;
                } catch (Ingester.IngesterException ingEx) {
                    extractor.logWarning("Ingester had a problem with extracted string from file '" //NON-NLS
                            + sourceName + "' (id: " + sourceID + ").", ingEx);//NON-NLS

                    throw ingEx; //need to rethrow to signal error and move on
                }
            }
            if (chunker.hasException()) {
                extractor.logWarning("Error chunking content from " + sourceID + ": " + sourceName, chunker.getException());
                return false;
            }
        } catch (Exception ex) {
            extractor.logWarning("Unexpected error, can't read content stream from " + sourceID + ": " + sourceName, ex);//NON-NLS
            return false;
        } finally {
            //after all chunks, index just the meta data, including the  numChunks, of the parent file
            fields.put(Server.Schema.NUM_CHUNKS.toString(), Integer.toString(numChunks));
            //reset id field to base document id
            fields.put(Server.Schema.ID.toString(), Long.toString(sourceID));
            //"parent" docs don't have chunk_size
            fields.remove(Server.Schema.CHUNK_SIZE.toString());
            indexChunk(null, sourceName, fields);
        }

        return true;
    }

    private void indexChunk(String chunk, String sourceName, Map<String, String> fields) throws IngesterException {
        if (fields.get(Server.Schema.IMAGE_ID.toString()) == null) {
            //JMTODO: actually if the we couldn't get the image id it is set to -1,
            // but does this really mean we don't want to index it?

            //skip the file, image id unknown
            //JMTODO: does this need to ne internationalized?
            String msg = NbBundle.getMessage(Ingester.class,
                    "Ingester.ingest.exception.unknownImgId.msg", sourceName); //JMTODO: does this need to ne internationalized?
            logger.log(Level.SEVERE, msg);
            throw new IngesterException(msg);
        }

        //Make a SolrInputDocument out of the field map
        SolrInputDocument updateDoc = new SolrInputDocument();
        for (String key : fields.keySet()) {
            updateDoc.addField(key, fields.get(key));
        }
        //add the content to the SolrInputDocument
        //JMTODO: can we just add it to the field map before passing that in?
        updateDoc.addField(Server.Schema.CONTENT.toString(), chunk);

        try {
            //TODO: consider timeout thread, or vary socket timeout based on size of indexed content
            solrServer.addDocument(updateDoc);
            uncommitedIngests = true;

        } catch (KeywordSearchModuleException | NoOpenCoreException ex) {
            //JMTODO: does this need to be internationalized?
            throw new IngesterException(
                    NbBundle.getMessage(Ingester.class, "Ingester.ingest.exception.err.msg", sourceName), ex);
        }
    }

    void commit() {
        try {
            solrServer.commit();
            uncommitedIngests = false;
        } catch (NoOpenCoreException | SolrServerException ex) {
            logger.log(Level.WARNING, "Error commiting index", ex); //NON-NLS

        }
    }

    static private class SolrFieldsVisitor extends SleuthkitItemVisitor.Default<Map<String, String>> {

        @Override
        protected Map<String, String> defaultVisit(SleuthkitVisitableItem svi) {
            return new HashMap<>();
        }

        @Override
        public Map<String, String> visit(File f) {
            return getCommonAndMACTimeFields(f);
        }

        @Override
        public Map<String, String> visit(DerivedFile df) {
            return getCommonAndMACTimeFields(df);
        }

        @Override
        public Map<String, String> visit(Directory d) {
            return getCommonAndMACTimeFields(d);
        }

        @Override
        public Map<String, String> visit(LayoutFile lf) {
            // layout files do not have times
            return getCommonFields(lf);
        }

        @Override
        public Map<String, String> visit(LocalFile lf) {
            return getCommonAndMACTimeFields(lf);
        }

        @Override
        public Map<String, String> visit(SlackFile f) {
            return getCommonAndMACTimeFields(f);
        }

        private Map<String, String> getCommonAndMACTimeFields(AbstractFile file) {
            Map<String, String> params = getCommonFields(file);
            params.put(Server.Schema.CTIME.toString(), ContentUtils.getStringTimeISO8601(file.getCtime(), file));
            params.put(Server.Schema.ATIME.toString(), ContentUtils.getStringTimeISO8601(file.getAtime(), file));
            params.put(Server.Schema.MTIME.toString(), ContentUtils.getStringTimeISO8601(file.getMtime(), file));
            params.put(Server.Schema.CRTIME.toString(), ContentUtils.getStringTimeISO8601(file.getCrtime(), file));
            return params;
        }

        private Map<String, String> getCommonFields(AbstractFile af) {
            Map<String, String> params = new HashMap<>();
            params.put(Server.Schema.ID.toString(), Long.toString(af.getId()));
            try {
                params.put(Server.Schema.IMAGE_ID.toString(), Long.toString(af.getDataSource().getId()));
            } catch (TskCoreException ex) {
                logger.log(Level.SEVERE, "Could not get data source id to properly index the file " + af.getId(), ex); //NON-NLS
                params.put(Server.Schema.IMAGE_ID.toString(), Long.toString(-1));
            }
            params.put(Server.Schema.FILE_NAME.toString(), af.getName());
            return params;
        }

        @Override
        public Map<String, String> visit(BlackboardArtifact artifact) {
            Map<String, String> params = new HashMap<>();
            params.put(Server.Schema.ID.toString(), Long.toString(artifact.getArtifactID()));
            try {
                params.put(Server.Schema.IMAGE_ID.toString(), Long.toString(ArtifactTextExtractor.getDataSource(artifact).getId()));
            } catch (TskCoreException ex) {
                logger.log(Level.SEVERE, "Could not get data source id to properly index the artifact " + artifact.getArtifactID(), ex); //NON-NLS
                params.put(Server.Schema.IMAGE_ID.toString(), Long.toString(-1));
            }
            return params;
        }
    }

    static class IngesterException extends Exception {

        private static final long serialVersionUID = 1L;

        IngesterException(String message, Throwable ex) {
            super(message, ex);
        }

        IngesterException(String message) {
            super(message);
        }
    }
}