source: trunk/gli/src/org/greenstone/gatherer/msm/GreenstoneArchiveParser.java@ 6877

/**
 *#########################################################################
 *
 * A component of the Gatherer application, part of the Greenstone digital
 * library suite from the New Zealand Digital Library Project at the
 * University of Waikato, New Zealand.
 *
 * <BR><BR>
 *
 * Author: John Thompson, Greenstone Digital Library, University of Waikato
 *
 * <BR><BR>
 *
 * Copyright (C) 1999 New Zealand Digital Library Project
 *
 * <BR><BR>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * <BR><BR>
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * <BR><BR>
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 *########################################################################
 */
package org.greenstone.gatherer.msm;

import java.io.*;
import java.net.*;
import java.util.*;
import org.greenstone.gatherer.Configuration;
import org.greenstone.gatherer.Dictionary;
import org.greenstone.gatherer.Gatherer;
import org.greenstone.gatherer.collection.Collection;
import org.greenstone.gatherer.collection.CollectionManager;
import org.greenstone.gatherer.file.FileNode;
import org.greenstone.gatherer.msm.ElementWrapper;
import org.greenstone.gatherer.msm.MetadataSet;
import org.greenstone.gatherer.msm.MetadataSetManager;
import org.greenstone.gatherer.msm.MSMUtils;
import org.greenstone.gatherer.shell.GShell;
import org.greenstone.gatherer.shell.GShellProgressMonitor;
import org.greenstone.gatherer.util.StaticStrings;
import org.greenstone.gatherer.util.Utility;
import org.greenstone.gatherer.valuetree.GValueModel;
import org.greenstone.gatherer.valuetree.GValueNode;
import org.w3c.dom.*;

public class GreenstoneArchiveParser {

    private GShell shell;

    static final String ignore_list[] = {"assocfilepath", "gsdl",  "Identifier", "URL"}; //"Source",

    public GreenstoneArchiveParser(GShellProgressMonitor progress, GShell shell) {
    // We can only extract metadata if an extracted metadata set exists in our collection.
    if(Gatherer.c_man.msm.getSet(Utility.EXTRACTED_METADATA_NAMESPACE) != null) {
        this.shell = shell;
        // The very firstist thing we do is remove any existing extracted metadata
        if(Gatherer.c_man != null && Gatherer.c_man.getCollection() != null) {
        Gatherer.c_man.getCollection().getGDM().removeExtractedMetadata();
        }

        // Determine the collection archive directory.
        File archive_directory = new File(Gatherer.c_man.getCollectionArchive());
        // For each of the hash coded directories within.
        File document_directories[] = archive_directory.listFiles();
        for(int i = 0; i < document_directories.length; i++) {
        // Find the doc.xml file within
        if(document_directories[i].isDirectory()) {
            File document_file = new File(document_directories[i], "doc.xml");
            // Then extract the metadata from it.
            if(document_file.exists()) {
            int count = extractMetadata(document_file);
            // Display a pretty progress message.
            String[] args = new String[2];
            args[0] = document_directories[i].getName();
            args[1] = String.valueOf(count);
            shell.fireMessage(GShell.IMPORT, shell.typeAsString(GShell.IMPORT) + "> " + Dictionary.get("GShell.Extracted", args), GShell.OK, null);
            args = null;
            progress.increment();
            }
        }
        }
    }
    // All done. Outta here like a bald man.
    }

    private int extractMetadata(File file) {
    int count = 0;
    // Retrieve the DOM of the file.
    Document document = Utility.parse(file, false);

    Gatherer.println("Parsed greenstone archive document: " + file.getAbsolutePath());
    // If we successfully parsed the document, then it is time to search through the DOM for the Metadata tags.
    if(document != null) {
        String file_path = null;
        Element archive_element = document.getDocumentElement();
        // Retrieve the initial Section element
        NodeList section_elements = archive_element.getElementsByTagName(StaticStrings.SECTION_ELEMENT);
        // It may be the case that we have some sort of mutant freak archive file which doesn't have any section elements.
        if(section_elements.getLength() < 1) {
        return count; 
        }
        Element section_element = (Element) section_elements.item(0);       
        section_elements = null;
        // Retrieve all of the Metadata sections.
        NodeList metadata_elements = section_element.getElementsByTagName(StaticStrings.METADATA_ELEMENT);
        section_element = null;
        // We first zip through the retrieved metadata, and if we encounter the element 'SourceSegment' - a sure sign this collection came from a bibliographic type file - we break out of extracted metadata parsing as no sense could be made of the data extracted anyway (plus we suffer a death of thirty-thousand pointy bits of metadata!)
        for(int i = 0; i < metadata_elements.getLength(); i++) {
        Element metadata_element = (Element) metadata_elements.item(i);
        String name = metadata_element.getAttribute(StaticStrings.NAME_ATTRIBUTE);
        if(name.equalsIgnoreCase(StaticStrings.SOURCESEGMENT_VALUE)) {
            return 0;
        }
        }
        // Now for each Metadata entry retrieved...
        for(int i = 0; i < metadata_elements.getLength(); i++) {
        Element metadata_element = (Element) metadata_elements.item(i);
        String name = metadata_element.getAttribute(StaticStrings.NAME_ATTRIBUTE);
        // There is also a special case when the metadata name is gsdlsourcefilename, as we use this to find the FileRecord we want to add metadata to.
        if(name.equals("gsdlsourcefilename")) {
            file_path = MSMUtils.getValue(metadata_element);
        }
        else {
            // Check if its name starts with, or is equal to, one of the values in our ignore list, and if so ignore this metadata.
            boolean ignore = false;
            for(int j = 0; !ignore && j < ignore_list.length; j++) {
            ignore = name.startsWith(ignore_list[j]);
            }
            // Otherwise ensure the metadata is present in our collection.
            if(!ignore && file_path != null) {
            // If we successfully retrieved a record we can continue.
            if(file_path != null) {
                // We now retrieve the appropriate element. If no such element exists we create a new one in the greenstone mds. Remember that no element in the greenstone mds has an associated value tree, so it is perfect for metadata elements with a small number of repeated values but where the values have no relation between files (such as encoding, where many files will be iso_8859_1, but if you change one you don't intend to change them all).
                ElementWrapper element = Gatherer.c_man.msm.getElement(name);
                if(element == null) {
                MetadataSet extracted_mds = Gatherer.c_man.msm.getSet(Utility.EXTRACTED_METADATA_NAMESPACE);
                if(extracted_mds != null) {
                    element = extracted_mds.addElement(name, Gatherer.config.getLanguage());
                }
                }
                // If we successfully retrieved an element (and we should have) we can continue.
                // WARNING!! There is one known exception - MARC records. Adding the extracted elements is all good, but adding the extracted metadata causes the whole thing to collapse in a pile of unhappy.
                if(element != null && !file_path.endsWith(StaticStrings.MARC_EXTENSION) && (element.getNamespace().equals("") || element.getNamespace().equals(Utility.EXTRACTED_METADATA_NAMESPACE))) {
                // Retrieve the metadata for the current file
                File target_file = new File(file_path);
                String value = "";
                try {
                    value = Utility.decodeGreenstone(URLDecoder.decode(MSMUtils.getValue(metadata_element), "UTF-8"));
                }
                catch(IllegalArgumentException error) { // ****
                    value = Utility.decodeGreenstone(MSMUtils.getValue(metadata_element));
                }
                catch(UnsupportedEncodingException error) {
                    Gatherer.printStackTrace(error);
                }
                // If we successfully retrieved a value we can continue.
                if(value != null) {
                    // Create a new metadata object.
                    GValueModel value_tree = Gatherer.c_man.msm.getValueTree(element);
                    GValueNode value_node = null;
                    if(value_tree != null) {
                    value_node = value_tree.getValue(value);
                    }
                    else {
                    value_node = new GValueNode(element.toString(), value);
                    }
                    Metadata metadata = new Metadata(element, value_node);
                    element.inc();
                    ///ystem.err.println("Adding extracted metadata: " + metadata);
                    Gatherer.c_man.getCollection().msm.fireMetadataChanged(new MSMEvent(this, System.currentTimeMillis(), target_file, null, metadata));
                    count++;
                    // All done. On to next metadata.
                }
                value = null;
                target_file = null;
                }
                else {
                Gatherer.println("Cannot retrieve metadata element " + name);
                }
            }
            }
        }
        }
    }
    return count;
    }

    static final String metadata_ignore_list[] = {"assocfilepath", "gsdl", "Identifier","URL"};

    static public ArrayList extractMetadataElements(File archive_directory) {
    ArrayList extracted_metadata_elements = new ArrayList();
    File document_directories[] = archive_directory.listFiles();
    for(int i = 0; i < document_directories.length; i++) {
        // Find the doc.xml file within
        if(document_directories[i].isDirectory()) {
        File document_file = new File(document_directories[i], "doc.xml");
        // Then extract the metadata from it.
        if(document_file.exists()) {
            try {
            Document document = Utility.parse(document_file, false);
            // Retrieve all of the Metadata sections.
            Element archive_element = document.getDocumentElement();
            NodeList metadata_elements = archive_element.getElementsByTagName("Metadata");
            // Now for each Metadata entry retrieved...
            for(int j = 0; j < metadata_elements.getLength(); j++) {
                Element metadata_element = (Element) metadata_elements.item(j);
                String name = metadata_element.getAttribute("name");
                // Check if its name starts with, or is equal to, one of the values in our ignore list, and if so ignore this metadata.
                boolean ignore = false;
                for(int k = 0; !ignore && k < metadata_ignore_list.length; k++) {
                ignore = name.startsWith(metadata_ignore_list[k]);
                }
                if(!ignore && !extracted_metadata_elements.contains(name)) {
                extracted_metadata_elements.add(name);
                }
                name = null;
                metadata_element = null;
            }
            metadata_elements = null;
            archive_element = null;
            document = null;
            }
            catch (Exception error) {
            Gatherer.printStackTrace(error);
            }
        }
        document_file = null;
        }
    }
    document_directories = null;
    return extracted_metadata_elements;
    }
}