RegistryLoader


1   /*
2    * RegistryLoader.java
3    *
4    * Version: $Revision: 1.9 $
5    *
6    * Date: $Date: 2005/11/16 21:40:50 $
7    *
8    * Copyright (c) 2002-2005, Hewlett-Packard Company and Massachusetts
9    * Institute of Technology.  All rights reserved.
10   *
11   * Redistribution and use in source and binary forms, with or without
12   * modification, are permitted provided that the following conditions are
13   * met:
14   *
15   * - Redistributions of source code must retain the above copyright
16   * notice, this list of conditions and the following disclaimer.
17   *
18   * - Redistributions in binary form must reproduce the above copyright
19   * notice, this list of conditions and the following disclaimer in the
20   * documentation and/or other materials provided with the distribution.
21   *
22   * - Neither the name of the Hewlett-Packard Company nor the name of the
23   * Massachusetts Institute of Technology nor the names of their
24   * contributors may be used to endorse or promote products derived from
25   * this software without specific prior written permission.
26   *
27   * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
28   * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
29   * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
30   * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
31   * HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
32   * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
33   * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
34   * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
35   * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
36   * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
37   * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
38   * DAMAGE.
39   */
40  package org.dspace.administer;
41  
42  import java.io.File  ;
43  import java.io.IOException  ;
44  import java.sql.SQLException  ;
45  
46  import javax.xml.parsers.DocumentBuilder  ;
47  import javax.xml.parsers.DocumentBuilderFactory  ;
48  import javax.xml.parsers.ParserConfigurationException  ;
49  import javax.xml.transform.TransformerException  ;
50  
51  import org.apache.log4j.Logger;
52  import org.apache.xpath.XPathAPI;
53  import org.dspace.authorize.AuthorizeException;
54  import org.dspace.content.BitstreamFormat;
55  import org.dspace.content.MetadataField;
56  import org.dspace.content.MetadataSchema;
57  import org.dspace.content.NonUniqueMetadataException;
58  import org.dspace.core.Context;
59  import org.dspace.core.LogManager;
60  import org.w3c.dom.Document  ;
61  import org.w3c.dom.Node  ;
62  import org.w3c.dom.NodeList  ;
63  import org.xml.sax.SAXException  ;
64  
65  /**
66   * Loads the bitstream format and Dublin Core type registries into the database.
67   * Intended for use as a command-line tool.
68   * <P>
69   * Example usage:
70   * <P>
71   * <code>RegistryLoader -bitstream bitstream-formats.xml</code>
72   * <P>
73   * <code>RegistryLoader -dc dc-types.xml</code>
74   * 
75   * @author Robert Tansley
76   * @version $Revision: 1.9 $
77   */
78  public class RegistryLoader
79  {
80      /** log4j category */
81      private static Logger log = Logger.getLogger(RegistryLoader.class);
82  
83      /**
84       * For invoking via the command line
85       * 
86       * @param argv
87       *            command-line arguments
88       */
89      public static void main(String  [] argv) throws Exception  
90      {
91          String   usage = "Usage: " + RegistryLoader.class.getName()
92                  + " (-bitstream | -dc) registry-file.xml";
93  
94          Context context = null;
95  
96          try
97          {
98              context = new Context();
99  
100             // Can't update registries anonymously, so we need to turn off
101             // authorisation
102             context.setIgnoreAuthorization(true);
103 
104             // Work out what we're loading
105             if (argv[0].equalsIgnoreCase("-bitstream"))
106             {
107                 RegistryLoader.loadBitstreamFormats(context, argv[1]);
108             }
109             else if (argv[0].equalsIgnoreCase("-dc"))
110             {
111                 loadDublinCoreTypes(context, argv[1]);
112             }
113             else
114             {
115                 System.err.println(usage);
116             }
117 
118             context.complete();
119 
120             System.exit(0);
121         }
122         catch (ArrayIndexOutOfBoundsException   ae)
123         {
124             System.err.println(usage);
125 
126             if (context != null)
127             {
128                 context.abort();
129             }
130 
131             System.exit(1);
132         }
133         catch (Exception   e)
134         {
135             log.fatal(LogManager.getHeader(context, "error_loading_registries",
136                     ""), e);
137 
138             if (context != null)
139             {
140                 context.abort();
141             }
142 
143             System.exit(1);
144         }
145     }
146 
147     /**
148      * Load Bitstream Format metadata
149      * 
150      * @param context
151      *            DSpace context object
152      * @param filename
153      *            the filename of the XML file to load
154      */
155     public static void loadBitstreamFormats(Context context, String   filename)
156             throws SQLException  , IOException  , ParserConfigurationException  ,
157             SAXException  , TransformerException  , AuthorizeException
158     {
159         Document   document = loadXML(filename);
160 
161         // Get the nodes corresponding to formats
162         NodeList   typeNodes = XPathAPI.selectNodeList(document,
163                 "dspace-bitstream-types/bitstream-type");
164 
165         // Add each one as a new format to the registry
166         for (int i = 0; i < typeNodes.getLength(); i++)
167         {
168             Node   n = typeNodes.item(i);
169             loadFormat(context, n);
170         }
171 
172         log.info(LogManager.getHeader(context, "load_bitstream_formats",
173                 "number_loaded=" + typeNodes.getLength()));
174     }
175 
176     /**
177      * Process a node in the bitstream format registry XML file. The node must
178      * be a "bitstream-type" node
179      * 
180      * @param context
181      *            DSpace context object
182      * @param node
183      *            the node in the DOM tree
184      */
185     private static void loadFormat(Context context, Node   node)
186             throws SQLException  , IOException  , TransformerException  ,
187             AuthorizeException
188     {
189         // Get the values
190         String   mimeType = getElementData(node, "mimetype");
191         String   shortDesc = getElementData(node, "short_description");
192         String   desc = getElementData(node, "description");
193 
194         String   supportLevelString = getElementData(node, "support_level");
195         int supportLevel = Integer.parseInt(supportLevelString);
196 
197         String   internalString = getElementData(node, "internal");
198         boolean internal = new Boolean  (internalString).booleanValue();
199 
200         String  [] extensions = getRepeatedElementData(node, "extension");
201 
202         // Create the format object
203         BitstreamFormat format = BitstreamFormat.create(context);
204 
205         // Fill it out with the values
206         format.setMIMEType(mimeType);
207         format.setShortDescription(shortDesc);
208         format.setDescription(desc);
209         format.setSupportLevel(supportLevel);
210         format.setInternal(internal);
211         format.setExtensions(extensions);
212 
213         // Write to database
214         format.update();
215     }
216 
217     /**
218      * Load Dublin Core types
219      * 
220      * @param context
221      *            DSpace context object
222      * @param filename
223      *            the filename of the XML file to load
224      * @throws NonUniqueMetadataException
225      */
226     public static void loadDublinCoreTypes(Context context, String   filename)
227             throws SQLException  , IOException  , ParserConfigurationException  ,
228             SAXException  , TransformerException  , AuthorizeException,
229             NonUniqueMetadataException
230     {
231         Document   document = loadXML(filename);
232 
233         // Get the nodes corresponding to formats
234         NodeList   typeNodes = XPathAPI.selectNodeList(document,
235                 "/dspace-dc-types/dc-type");
236 
237         // Add each one as a new format to the registry
238         for (int i = 0; i < typeNodes.getLength(); i++)
239         {
240             Node   n = typeNodes.item(i);
241             loadDCType(context, n);
242         }
243 
244         log.info(LogManager.getHeader(context, "load_dublin_core_types",
245                 "number_loaded=" + typeNodes.getLength()));
246     }
247 
248     /**
249      * Process a node in the bitstream format registry XML file. The node must
250      * be a "bitstream-type" node
251      * 
252      * @param context
253      *            DSpace context object
254      * @param node
255      *            the node in the DOM tree
256      * @throws NonUniqueMetadataException
257      */
258     private static void loadDCType(Context context, Node   node)
259             throws SQLException  , IOException  , TransformerException  ,
260             AuthorizeException, NonUniqueMetadataException
261     {
262         // Get the values
263         String   schema = getElementData(node, "schema");
264         String   element = getElementData(node, "element");
265         String   qualifier = getElementData(node, "qualifier");
266         String   scopeNote = getElementData(node, "scope_note");
267 
268         // If the schema is not provided default to DC
269         if (schema == null)
270         {
271             schema = MetadataSchema.DC_SCHEMA;
272         }
273 
274         // Find the matching schema object
275         MetadataSchema schemaObj = MetadataSchema.find(context, schema);
276         
277         MetadataField field = new MetadataField();
278         field.setSchemaID(schemaObj.getSchemaID());
279         field.setElement(element);
280         field.setQualifier(qualifier);
281         field.setScopeNote(scopeNote);
282         field.create(context);
283     }
284 
285     // ===================== XML Utility Methods =========================
286 
287     /**
288      * Load in the XML from file.
289      * 
290      * @param filename
291      *            the filename to load from
292      * 
293      * @return the DOM representation of the XML file
294      */
295     private static Document   loadXML(String   filename) throws IOException  ,
296             ParserConfigurationException  , SAXException  
297     {
298         DocumentBuilder   builder = DocumentBuilderFactory.newInstance()
299                 .newDocumentBuilder();
300 
301         return builder.parse(new File  (filename));
302     }
303 
304     /**
305      * Get the CDATA of a particular element. For example, if the XML document
306      * contains:
307      * <P>
308      * <code>
309      * &lt;foo&gt;&lt;mimetype&gt;application/pdf&lt;/mimetype&gt;&lt;/foo&gt;
310      * </code>
311      * passing this the <code>foo</code> node and <code>mimetype</code> will
312      * return <code>application/pdf</code>.
313      * </P>
314      * Why this isn't a core part of the XML API I do not know...
315      * 
316      * @param parentElement
317      *            the element, whose child element you want the CDATA from
318      * @param childName
319      *            the name of the element you want the CDATA from
320      * 
321      * @return the CDATA as a <code>String</code>
322      */
323     private static String   getElementData(Node   parentElement, String   childName)
324             throws TransformerException  
325     {
326         // Grab the child node
327         Node   childNode = XPathAPI.selectSingleNode(parentElement, childName);
328 
329         if (childNode == null)
330         {
331             // No child node, so no values
332             return null;
333         }
334 
335         // Get the #text
336         Node   dataNode = childNode.getFirstChild();
337 
338         if (dataNode == null)
339         {
340             return null;
341         }
342 
343         // Get the data
344         String   value = dataNode.getNodeValue().trim();
345 
346         return value;
347     }
348 
349     /**
350      * Get repeated CDATA for a particular element. For example, if the XML
351      * document contains:
352      * <P>
353      * <code>
354      * &lt;foo&gt;
355      *   &lt;bar&gt;val1&lt;/bar&gt;
356      *   &lt;bar&gt;val2&lt;/bar&gt;
357      * &lt;/foo&gt;
358      * </code>
359      * passing this the <code>foo</code> node and <code>bar</code> will
360      * return <code>val1</code> and <code>val2</code>.
361      * </P>
362      * Why this also isn't a core part of the XML API I do not know...
363      * 
364      * @param parentElement
365      *            the element, whose child element you want the CDATA from
366      * @param childName
367      *            the name of the element you want the CDATA from
368      * 
369      * @return the CDATA as a <code>String</code>
370      */
371     private static String  [] getRepeatedElementData(Node   parentElement,
372             String   childName) throws TransformerException  
373     {
374         // Grab the child node
375         NodeList   childNodes = XPathAPI.selectNodeList(parentElement, childName);
376 
377         String  [] data = new String  [childNodes.getLength()];
378 
379         for (int i = 0; i < childNodes.getLength(); i++)
380         {
381             // Get the #text node
382             Node   dataNode = childNodes.item(i).getFirstChild();
383 
384             // Get the data
385             data[i] = dataNode.getNodeValue().trim();
386         }
387 
388         return data;
389     }
390 }
391
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Free Books Free Magazines
Popular Tags