1 |
7542
|
tao
|
/**
|
2 |
8138
|
tao
|
* Copyright: 2013 Regents of the University of California and the
|
3 |
7542
|
tao
|
* National Center for Ecological Analysis and Synthesis
|
4 |
|
|
*
|
5 |
|
|
* This program is free software; you can redistribute it and/or modify
|
6 |
|
|
* it under the terms of the GNU General Public License as published by
|
7 |
|
|
* the Free Software Foundation; either version 2 of the License, or
|
8 |
|
|
* (at your option) any later version.
|
9 |
|
|
*
|
10 |
|
|
* This program is distributed in the hope that it will be useful,
|
11 |
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
12 |
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
13 |
|
|
* GNU General Public License for more details.
|
14 |
|
|
*
|
15 |
|
|
* You should have received a copy of the GNU General Public License
|
16 |
|
|
* along with this program; if not, write to the Free Software
|
17 |
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
18 |
|
|
*/
|
19 |
|
|
package edu.ucsb.nceas.metacat.index;
|
20 |
|
|
|
21 |
7555
|
tao
|
import java.io.ByteArrayInputStream;
|
22 |
8343
|
tao
|
import java.io.FileInputStream;
|
23 |
7786
|
tao
|
import java.io.FileNotFoundException;
|
24 |
7546
|
tao
|
import java.io.IOException;
|
25 |
|
|
import java.io.InputStream;
|
26 |
8893
|
tao
|
import java.net.MalformedURLException;
|
27 |
7546
|
tao
|
import java.util.ArrayList;
|
28 |
7800
|
tao
|
import java.util.Calendar;
|
29 |
8893
|
tao
|
import java.util.Collection;
|
30 |
8464
|
leinfelder
|
import java.util.Date;
|
31 |
7546
|
tao
|
import java.util.HashMap;
|
32 |
|
|
import java.util.Iterator;
|
33 |
7542
|
tao
|
import java.util.List;
|
34 |
7546
|
tao
|
import java.util.Map;
|
35 |
|
|
import java.util.Set;
|
36 |
7542
|
tao
|
|
37 |
|
|
import javax.xml.parsers.DocumentBuilder;
|
38 |
|
|
import javax.xml.parsers.DocumentBuilderFactory;
|
39 |
7546
|
tao
|
import javax.xml.parsers.ParserConfigurationException;
|
40 |
7542
|
tao
|
import javax.xml.xpath.XPath;
|
41 |
7546
|
tao
|
import javax.xml.xpath.XPathExpressionException;
|
42 |
7542
|
tao
|
import javax.xml.xpath.XPathFactory;
|
43 |
|
|
|
44 |
7546
|
tao
|
import org.apache.commons.codec.EncoderException;
|
45 |
8893
|
tao
|
import org.apache.commons.collections.CollectionUtils;
|
46 |
7546
|
tao
|
import org.apache.commons.io.output.ByteArrayOutputStream;
|
47 |
7786
|
tao
|
import org.apache.commons.lang.StringUtils;
|
48 |
7546
|
tao
|
import org.apache.commons.logging.Log;
|
49 |
|
|
import org.apache.commons.logging.LogFactory;
|
50 |
7604
|
tao
|
import org.apache.solr.client.solrj.SolrQuery;
|
51 |
7542
|
tao
|
import org.apache.solr.client.solrj.SolrServer;
|
52 |
7546
|
tao
|
import org.apache.solr.client.solrj.SolrServerException;
|
53 |
7604
|
tao
|
import org.apache.solr.client.solrj.response.QueryResponse;
|
54 |
7547
|
tao
|
import org.apache.solr.client.solrj.response.UpdateResponse;
|
55 |
7604
|
tao
|
import org.apache.solr.common.SolrDocument;
|
56 |
|
|
import org.apache.solr.common.SolrDocumentList;
|
57 |
7546
|
tao
|
import org.apache.solr.common.SolrInputDocument;
|
58 |
8464
|
leinfelder
|
import org.apache.solr.schema.IndexSchema;
|
59 |
7542
|
tao
|
import org.dataone.cn.indexer.XMLNamespaceConfig;
|
60 |
8464
|
leinfelder
|
import org.dataone.cn.indexer.convert.SolrDateConverter;
|
61 |
9028
|
leinfelder
|
import org.dataone.cn.indexer.parser.BaseXPathDocumentSubprocessor;
|
62 |
9060
|
leinfelder
|
import org.dataone.cn.indexer.parser.IDocumentDeleteSubprocessor;
|
63 |
7542
|
tao
|
import org.dataone.cn.indexer.parser.IDocumentSubprocessor;
|
64 |
7546
|
tao
|
import org.dataone.cn.indexer.parser.SolrField;
|
65 |
|
|
import org.dataone.cn.indexer.solrhttp.SolrDoc;
|
66 |
|
|
import org.dataone.cn.indexer.solrhttp.SolrElementField;
|
67 |
7733
|
tao
|
import org.dataone.service.exceptions.NotFound;
|
68 |
|
|
import org.dataone.service.exceptions.NotImplemented;
|
69 |
7786
|
tao
|
import org.dataone.service.exceptions.ServiceFailure;
|
70 |
7733
|
tao
|
import org.dataone.service.exceptions.UnsupportedType;
|
71 |
7815
|
leinfelder
|
import org.dataone.service.types.v1.Event;
|
72 |
7577
|
tao
|
import org.dataone.service.types.v1.Identifier;
|
73 |
8826
|
leinfelder
|
import org.dataone.service.types.v2.SystemMetadata;
|
74 |
8464
|
leinfelder
|
import org.dataone.service.util.DateTimeMarshaller;
|
75 |
7555
|
tao
|
import org.dataone.service.util.TypeMarshaller;
|
76 |
8023
|
tao
|
import org.dspace.foresite.OREParserException;
|
77 |
7555
|
tao
|
import org.jibx.runtime.JiBXException;
|
78 |
7546
|
tao
|
import org.w3c.dom.Document;
|
79 |
|
|
import org.xml.sax.SAXException;
|
80 |
7542
|
tao
|
|
81 |
7828
|
leinfelder
|
import edu.ucsb.nceas.metacat.common.index.event.IndexEvent;
|
82 |
8464
|
leinfelder
|
import edu.ucsb.nceas.metacat.common.query.SolrQueryServiceController;
|
83 |
7800
|
tao
|
import edu.ucsb.nceas.metacat.index.event.EventlogFactory;
|
84 |
7711
|
tao
|
import edu.ucsb.nceas.metacat.index.resourcemap.ResourceMapSubprocessor;
|
85 |
|
|
|
86 |
7542
|
tao
|
/**
|
87 |
|
|
* A class does insert, update and remove indexes to a SOLR server
|
88 |
|
|
* @author tao
|
89 |
|
|
*
|
90 |
|
|
*/
|
91 |
|
|
public class SolrIndex {
|
92 |
7591
|
leinfelder
|
|
93 |
7604
|
tao
|
public static final String ID = "id";
|
94 |
|
|
private static final String IDQUERY = ID+":*";
|
95 |
7542
|
tao
|
private List<IDocumentSubprocessor> subprocessors = null;
|
96 |
9060
|
leinfelder
|
private List<IDocumentDeleteSubprocessor> deleteSubprocessors = null;
|
97 |
|
|
|
98 |
7542
|
tao
|
private SolrServer solrServer = null;
|
99 |
|
|
private XMLNamespaceConfig xmlNamespaceConfig = null;
|
100 |
7546
|
tao
|
private List<SolrField> sysmetaSolrFields = null;
|
101 |
7542
|
tao
|
|
102 |
|
|
private static DocumentBuilderFactory documentBuilderFactory = null;
|
103 |
|
|
private static DocumentBuilder builder = null;
|
104 |
|
|
|
105 |
|
|
private static XPathFactory xpathFactory = null;
|
106 |
|
|
private static XPath xpath = null;
|
107 |
7546
|
tao
|
Log log = LogFactory.getLog(SolrIndex.class);
|
108 |
7542
|
tao
|
|
109 |
7546
|
tao
|
static {
|
110 |
|
|
documentBuilderFactory = DocumentBuilderFactory.newInstance();
|
111 |
|
|
documentBuilderFactory.setNamespaceAware(true);
|
112 |
|
|
try {
|
113 |
|
|
builder = documentBuilderFactory.newDocumentBuilder();
|
114 |
|
|
} catch (ParserConfigurationException e) {
|
115 |
|
|
e.printStackTrace();
|
116 |
|
|
}
|
117 |
|
|
xpathFactory = XPathFactory.newInstance();
|
118 |
|
|
xpath = xpathFactory.newXPath();
|
119 |
|
|
}
|
120 |
|
|
|
121 |
7542
|
tao
|
/**
|
122 |
|
|
* Constructor
|
123 |
7548
|
tao
|
* @throws SAXException
|
124 |
|
|
* @throws IOException
|
125 |
7542
|
tao
|
*/
|
126 |
8161
|
tao
|
public SolrIndex(XMLNamespaceConfig xmlNamespaceConfig, List<SolrField> sysmetaSolrFields)
|
127 |
7548
|
tao
|
throws XPathExpressionException, ParserConfigurationException, IOException, SAXException {
|
128 |
|
|
this.xmlNamespaceConfig = xmlNamespaceConfig;
|
129 |
|
|
this.sysmetaSolrFields = sysmetaSolrFields;
|
130 |
|
|
init();
|
131 |
8125
|
tao
|
}
|
132 |
7542
|
tao
|
|
133 |
7546
|
tao
|
private void init() throws ParserConfigurationException, XPathExpressionException {
|
134 |
|
|
xpath.setNamespaceContext(xmlNamespaceConfig);
|
135 |
|
|
initExpressions();
|
136 |
|
|
}
|
137 |
|
|
|
138 |
|
|
private void initExpressions() throws XPathExpressionException {
|
139 |
|
|
for (SolrField field : sysmetaSolrFields) {
|
140 |
|
|
field.initExpression(xpath);
|
141 |
|
|
}
|
142 |
|
|
|
143 |
|
|
}
|
144 |
7542
|
tao
|
|
145 |
7546
|
tao
|
|
146 |
7542
|
tao
|
/**
|
147 |
|
|
* Get the list of the Subprocessors in this index.
|
148 |
|
|
* @return the list of the Subprocessors.
|
149 |
|
|
*/
|
150 |
|
|
public List<IDocumentSubprocessor> getSubprocessors() {
|
151 |
|
|
return subprocessors;
|
152 |
|
|
}
|
153 |
|
|
|
154 |
|
|
/**
|
155 |
|
|
* Set the list of Subprocessors.
|
156 |
|
|
* @param subprocessorList the list will be set.
|
157 |
|
|
*/
|
158 |
|
|
public void setSubprocessors(List<IDocumentSubprocessor> subprocessorList) {
|
159 |
7546
|
tao
|
for (IDocumentSubprocessor subprocessor : subprocessorList) {
|
160 |
9028
|
leinfelder
|
if (subprocessor instanceof BaseXPathDocumentSubprocessor) {
|
161 |
|
|
((BaseXPathDocumentSubprocessor)subprocessor).initExpression(xpath);
|
162 |
9018
|
leinfelder
|
}
|
163 |
7546
|
tao
|
}
|
164 |
7542
|
tao
|
this.subprocessors = subprocessorList;
|
165 |
|
|
}
|
166 |
7546
|
tao
|
|
167 |
9060
|
leinfelder
|
public List<IDocumentDeleteSubprocessor> getDeleteSubprocessors() {
|
168 |
|
|
return deleteSubprocessors;
|
169 |
|
|
}
|
170 |
|
|
|
171 |
|
|
public void setDeleteSubprocessors(
|
172 |
|
|
List<IDocumentDeleteSubprocessor> deleteSubprocessors) {
|
173 |
|
|
this.deleteSubprocessors = deleteSubprocessors;
|
174 |
|
|
}
|
175 |
|
|
|
176 |
|
|
/**
|
177 |
7546
|
tao
|
* Generate the index for the given information
|
178 |
|
|
* @param id
|
179 |
7555
|
tao
|
* @param systemMetadata
|
180 |
7546
|
tao
|
* @param dataStream
|
181 |
|
|
* @return
|
182 |
|
|
* @throws IOException
|
183 |
|
|
* @throws SAXException
|
184 |
|
|
* @throws ParserConfigurationException
|
185 |
|
|
* @throws XPathExpressionException
|
186 |
7555
|
tao
|
* @throws JiBXException
|
187 |
7711
|
tao
|
* @throws SolrServerException
|
188 |
7546
|
tao
|
* @throws EncoderException
|
189 |
7733
|
tao
|
* @throws UnsupportedType
|
190 |
|
|
* @throws NotFound
|
191 |
|
|
* @throws NotImplemented
|
192 |
7546
|
tao
|
*/
|
193 |
8766
|
leinfelder
|
private Map<String, SolrDoc> process(String id, SystemMetadata systemMetadata, String objectPath)
|
194 |
7546
|
tao
|
throws IOException, SAXException, ParserConfigurationException,
|
195 |
7733
|
tao
|
XPathExpressionException, JiBXException, EncoderException, SolrServerException, NotImplemented, NotFound, UnsupportedType{
|
196 |
7546
|
tao
|
|
197 |
|
|
// Load the System Metadata document
|
198 |
7555
|
tao
|
ByteArrayOutputStream systemMetadataOutputStream = new ByteArrayOutputStream();
|
199 |
|
|
TypeMarshaller.marshalTypeToOutputStream(systemMetadata, systemMetadataOutputStream);
|
200 |
|
|
ByteArrayInputStream systemMetadataStream = new ByteArrayInputStream(systemMetadataOutputStream.toByteArray());
|
201 |
|
|
Document sysMetaDoc = generateXmlDocument(systemMetadataStream);
|
202 |
7546
|
tao
|
if (sysMetaDoc == null) {
|
203 |
|
|
log.error("Could not load System metadata for ID: " + id);
|
204 |
|
|
return null;
|
205 |
|
|
}
|
206 |
|
|
|
207 |
|
|
// Extract the field values from the System Metadata
|
208 |
|
|
List<SolrElementField> sysSolrFields = processSysmetaFields(sysMetaDoc, id);
|
209 |
|
|
SolrDoc indexDocument = new SolrDoc(sysSolrFields);
|
210 |
|
|
Map<String, SolrDoc> docs = new HashMap<String, SolrDoc>();
|
211 |
|
|
docs.put(id, indexDocument);
|
212 |
9018
|
leinfelder
|
|
213 |
|
|
// get the format id for this object
|
214 |
|
|
String formatId = indexDocument.getFirstFieldValue(SolrElementField.FIELD_OBJECTFORMAT);
|
215 |
7546
|
tao
|
|
216 |
|
|
// Determine if subprocessors are available for this ID
|
217 |
|
|
if (subprocessors != null) {
|
218 |
9018
|
leinfelder
|
// for each subprocessor loaded from the spring config
|
219 |
|
|
for (IDocumentSubprocessor subprocessor : subprocessors) {
|
220 |
|
|
// Does this subprocessor apply?
|
221 |
|
|
if (subprocessor.canProcess(formatId)) {
|
222 |
|
|
// if so, then extract the additional information from the
|
223 |
|
|
// document.
|
224 |
|
|
try {
|
225 |
|
|
// docObject = the resource map document or science
|
226 |
|
|
// metadata document.
|
227 |
|
|
// note that resource map processing touches all objects
|
228 |
|
|
// referenced by the resource map.
|
229 |
|
|
FileInputStream dataStream = new FileInputStream(objectPath);
|
230 |
|
|
if (!dataStream.getFD().valid()) {
|
231 |
|
|
log.error("Could not load OBJECT file for ID,Path=" + id + ", "
|
232 |
|
|
+ objectPath);
|
233 |
|
|
//throw new Exception("Could not load OBJECT for ID " + id );
|
234 |
|
|
} else {
|
235 |
|
|
docs = subprocessor.processDocument(id, docs, dataStream);
|
236 |
|
|
}
|
237 |
|
|
} catch (Exception e) {
|
238 |
9368
|
tao
|
e.printStackTrace();
|
239 |
9018
|
leinfelder
|
log.error(e.getMessage(), e);
|
240 |
|
|
throw new SolrServerException(e.getMessage());
|
241 |
|
|
}
|
242 |
|
|
}
|
243 |
|
|
}
|
244 |
7546
|
tao
|
}
|
245 |
|
|
|
246 |
|
|
// TODO: in the XPathDocumentParser class in d1_cn_index_process module,
|
247 |
|
|
// merge is only for resource map. We need more work here.
|
248 |
|
|
for (SolrDoc mergeDoc : docs.values()) {
|
249 |
|
|
if (!mergeDoc.isMerged()) {
|
250 |
7711
|
tao
|
mergeWithIndexedDocument(mergeDoc);
|
251 |
7546
|
tao
|
}
|
252 |
|
|
}
|
253 |
|
|
|
254 |
|
|
//SolrElementAdd addCommand = getAddCommand(new ArrayList<SolrDoc>(docs.values()));
|
255 |
|
|
|
256 |
|
|
return docs;
|
257 |
|
|
}
|
258 |
|
|
|
259 |
7711
|
tao
|
/**
|
260 |
|
|
* Merge updates with existing solr documents
|
261 |
|
|
*
|
262 |
|
|
* This method appears to re-set the data package field data into the
|
263 |
|
|
* document about to be updated in the solr index. Since packaging
|
264 |
|
|
* information is derived from the package document (resource map), this
|
265 |
|
|
* information is not present when processing a document contained in a data
|
266 |
|
|
* package. This method replaces those values from the existing solr index
|
267 |
|
|
* record for the document being processed. -- sroseboo, 1-18-12
|
268 |
|
|
*
|
269 |
|
|
* @param indexDocument
|
270 |
|
|
* @return
|
271 |
|
|
* @throws IOException
|
272 |
|
|
* @throws EncoderException
|
273 |
|
|
* @throws XPathExpressionException
|
274 |
|
|
* @throws SAXException
|
275 |
|
|
* @throws ParserConfigurationException
|
276 |
|
|
* @throws SolrServerException
|
277 |
7733
|
tao
|
* @throws UnsupportedType
|
278 |
|
|
* @throws NotFound
|
279 |
|
|
* @throws NotImplemented
|
280 |
7711
|
tao
|
*/
|
281 |
|
|
// TODO:combine merge function with resourcemap merge function
|
282 |
|
|
|
283 |
|
|
private SolrDoc mergeWithIndexedDocument(SolrDoc indexDocument) throws IOException,
|
284 |
7733
|
tao
|
EncoderException, XPathExpressionException, SolrServerException, ParserConfigurationException, SAXException, NotImplemented, NotFound, UnsupportedType {
|
285 |
7711
|
tao
|
List<String> ids = new ArrayList<String>();
|
286 |
|
|
ids.add(indexDocument.getIdentifier());
|
287 |
|
|
List<SolrDoc> indexedDocuments = ResourceMapSubprocessor.getSolrDocs(ids);
|
288 |
|
|
SolrDoc indexedDocument = indexedDocuments == null || indexedDocuments.size() <= 0 ? null
|
289 |
|
|
: indexedDocuments.get(0);
|
290 |
8464
|
leinfelder
|
|
291 |
|
|
IndexSchema indexSchema = SolrQueryServiceController.getInstance().getSchema();
|
292 |
|
|
|
293 |
7711
|
tao
|
if (indexedDocument == null || indexedDocument.getFieldList().size() <= 0) {
|
294 |
|
|
return indexDocument;
|
295 |
|
|
} else {
|
296 |
|
|
for (SolrElementField field : indexedDocument.getFieldList()) {
|
297 |
|
|
if ((field.getName().equals(SolrElementField.FIELD_ISDOCUMENTEDBY)
|
298 |
|
|
|| field.getName().equals(SolrElementField.FIELD_DOCUMENTS) || field
|
299 |
|
|
.getName().equals(SolrElementField.FIELD_RESOURCEMAP))
|
300 |
|
|
&& !indexDocument.hasFieldWithValue(field.getName(), field.getValue())) {
|
301 |
|
|
indexDocument.addField(field);
|
302 |
8464
|
leinfelder
|
} else if (!indexSchema.isCopyFieldTarget(indexSchema.getField(field.getName())) && !indexDocument.hasField(field.getName())) {
|
303 |
|
|
indexDocument.addField(field);
|
304 |
7711
|
tao
|
}
|
305 |
|
|
}
|
306 |
|
|
|
307 |
|
|
indexDocument.setMerged(true);
|
308 |
|
|
return indexDocument;
|
309 |
|
|
}
|
310 |
|
|
}
|
311 |
|
|
|
312 |
7546
|
tao
|
/*
|
313 |
|
|
* Generate a Document from the InputStream
|
314 |
|
|
*/
|
315 |
|
|
private Document generateXmlDocument(InputStream smdStream) throws SAXException {
|
316 |
|
|
Document doc = null;
|
317 |
|
|
|
318 |
|
|
try {
|
319 |
|
|
doc = builder.parse(smdStream);
|
320 |
|
|
} catch (IOException e) {
|
321 |
|
|
log.error(e.getMessage(), e);
|
322 |
|
|
}
|
323 |
|
|
|
324 |
|
|
return doc;
|
325 |
|
|
}
|
326 |
|
|
|
327 |
|
|
/*
|
328 |
|
|
* Index the fields of the system metadata
|
329 |
|
|
*/
|
330 |
|
|
private List<SolrElementField> processSysmetaFields(Document doc, String identifier) {
|
331 |
|
|
|
332 |
|
|
List<SolrElementField> fieldList = new ArrayList<SolrElementField>();
|
333 |
|
|
// solrFields is the list of fields defined in the application context
|
334 |
|
|
|
335 |
|
|
for (SolrField field : sysmetaSolrFields) {
|
336 |
|
|
try {
|
337 |
|
|
// the field.getFields method can return a single value or
|
338 |
|
|
// multiple values for multi-valued fields
|
339 |
|
|
// or can return multiple SOLR document fields.
|
340 |
|
|
fieldList.addAll(field.getFields(doc, identifier));
|
341 |
|
|
} catch (Exception e) {
|
342 |
|
|
e.printStackTrace();
|
343 |
|
|
}
|
344 |
|
|
}
|
345 |
|
|
return fieldList;
|
346 |
|
|
|
347 |
|
|
}
|
348 |
|
|
|
349 |
|
|
/**
|
350 |
7577
|
tao
|
* Check the parameters of the insert or update methods.
|
351 |
|
|
* @param pid
|
352 |
|
|
* @param systemMetadata
|
353 |
|
|
* @param data
|
354 |
|
|
* @throws SolrServerException
|
355 |
|
|
*/
|
356 |
8766
|
leinfelder
|
private void checkParams(Identifier pid, SystemMetadata systemMetadata, String objectPath) throws SolrServerException {
|
357 |
8343
|
tao
|
if(pid == null || pid.getValue() == null || pid.getValue().trim().equals("")) {
|
358 |
7577
|
tao
|
throw new SolrServerException("The identifier of the indexed document should not be null or blank.");
|
359 |
|
|
}
|
360 |
|
|
if(systemMetadata == null) {
|
361 |
8343
|
tao
|
throw new SolrServerException("The system metadata of the indexed document "+pid.getValue()+ " should not be null.");
|
362 |
7577
|
tao
|
}
|
363 |
8766
|
leinfelder
|
if(objectPath == null) {
|
364 |
8343
|
tao
|
throw new SolrServerException("The indexed document itself for pid "+pid.getValue()+" should not be null.");
|
365 |
7577
|
tao
|
}
|
366 |
|
|
}
|
367 |
|
|
|
368 |
|
|
/**
|
369 |
7627
|
tao
|
* Insert the indexes for a document.
|
370 |
7546
|
tao
|
* @param pid the id of this document
|
371 |
|
|
* @param systemMetadata the system metadata associated with the data object
|
372 |
8766
|
leinfelder
|
* @param data the path to the object file itself
|
373 |
7546
|
tao
|
* @throws SolrServerException
|
374 |
7555
|
tao
|
* @throws JiBXException
|
375 |
7711
|
tao
|
* @throws EncoderException
|
376 |
7733
|
tao
|
* @throws UnsupportedType
|
377 |
|
|
* @throws NotFound
|
378 |
|
|
* @throws NotImplemented
|
379 |
7546
|
tao
|
*/
|
380 |
8766
|
leinfelder
|
private synchronized void insert(Identifier pid, SystemMetadata systemMetadata, String objectPath)
|
381 |
7546
|
tao
|
throws IOException, SAXException, ParserConfigurationException,
|
382 |
7733
|
tao
|
XPathExpressionException, SolrServerException, JiBXException, EncoderException, NotImplemented, NotFound, UnsupportedType {
|
383 |
8766
|
leinfelder
|
checkParams(pid, systemMetadata, objectPath);
|
384 |
|
|
Map<String, SolrDoc> docs = process(pid.getValue(), systemMetadata, objectPath);
|
385 |
7546
|
tao
|
|
386 |
|
|
//transform the Map to the SolrInputDocument which can be used by the solr server
|
387 |
|
|
if(docs != null) {
|
388 |
|
|
Set<String> ids = docs.keySet();
|
389 |
|
|
for(String id : ids) {
|
390 |
|
|
if(id != null) {
|
391 |
|
|
SolrDoc doc = docs.get(id);
|
392 |
7786
|
tao
|
insertToIndex(doc);
|
393 |
|
|
}
|
394 |
|
|
|
395 |
|
|
}
|
396 |
|
|
}
|
397 |
|
|
}
|
398 |
|
|
|
399 |
8464
|
leinfelder
|
/**
|
400 |
|
|
* Adds the given fields to the solr index for the given pid, preserving the index values
|
401 |
|
|
* that previously existed
|
402 |
|
|
* @param pid
|
403 |
|
|
* @param fields
|
404 |
|
|
*/
|
405 |
|
|
public void insertFields(Identifier pid, Map<String, List<Object>> fields) {
|
406 |
|
|
|
407 |
|
|
try {
|
408 |
|
|
// copy the original values already indexed for this document
|
409 |
|
|
SolrQuery query = new SolrQuery("id:\"" + pid.getValue() + "\"");
|
410 |
|
|
QueryResponse res = solrServer.query(query);
|
411 |
|
|
SolrDoc doc = new SolrDoc();
|
412 |
8503
|
leinfelder
|
|
413 |
|
|
// include existing values if they exist
|
414 |
8761
|
leinfelder
|
IndexSchema indexSchema = SolrQueryServiceController.getInstance().getSchema();
|
415 |
|
|
|
416 |
8503
|
leinfelder
|
if (res.getResults().size() > 0) {
|
417 |
|
|
SolrDocument orig = res.getResults().get(0);
|
418 |
|
|
for (String fieldName: orig.getFieldNames()) {
|
419 |
|
|
// don't transfer the copyTo fields, otherwise there are errors
|
420 |
|
|
if (indexSchema.isCopyFieldTarget(indexSchema.getField(fieldName))) {
|
421 |
|
|
continue;
|
422 |
|
|
}
|
423 |
|
|
for (Object value: orig.getFieldValues(fieldName)) {
|
424 |
|
|
String stringValue = value.toString();
|
425 |
|
|
// special handling for dates in ISO 8601
|
426 |
|
|
if (value instanceof Date) {
|
427 |
|
|
stringValue = DateTimeMarshaller.serializeDateToUTC((Date)value);
|
428 |
|
|
SolrDateConverter converter = new SolrDateConverter();
|
429 |
|
|
stringValue = converter.convert(stringValue);
|
430 |
|
|
}
|
431 |
|
|
SolrElementField field = new SolrElementField(fieldName, stringValue);
|
432 |
|
|
log.debug("Adding field: " + fieldName);
|
433 |
|
|
doc.addField(field);
|
434 |
|
|
}
|
435 |
|
|
}
|
436 |
8464
|
leinfelder
|
}
|
437 |
|
|
|
438 |
|
|
// add the additional fields we are trying to include in the index
|
439 |
|
|
for (String fieldName: fields.keySet()) {
|
440 |
|
|
List<Object> values = fields.get(fieldName);
|
441 |
|
|
for (Object value: values) {
|
442 |
8756
|
leinfelder
|
if (!doc.hasFieldWithValue(fieldName, value.toString())) {
|
443 |
8761
|
leinfelder
|
if (indexSchema.getField(fieldName).multiValued()) {
|
444 |
|
|
doc.addField(new SolrElementField(fieldName, value.toString()));
|
445 |
|
|
} else {
|
446 |
|
|
doc.updateOrAddField(fieldName, value.toString());
|
447 |
|
|
}
|
448 |
8756
|
leinfelder
|
}
|
449 |
8464
|
leinfelder
|
}
|
450 |
|
|
}
|
451 |
|
|
|
452 |
8580
|
leinfelder
|
// make sure there is an id in the solrdoc so it is added to the index
|
453 |
|
|
if (!doc.hasField(ID)) {
|
454 |
|
|
doc.updateOrAddField(ID, pid.getValue());
|
455 |
|
|
}
|
456 |
|
|
|
457 |
8464
|
leinfelder
|
// insert the whole thing
|
458 |
|
|
insertToIndex(doc);
|
459 |
|
|
} catch (Exception e) {
|
460 |
|
|
String error = "SolrIndex.insetFields - could not update the solr index: " + e.getMessage();
|
461 |
|
|
writeEventLog(null, pid, error);
|
462 |
|
|
log.error(error, e);
|
463 |
|
|
}
|
464 |
|
|
|
465 |
|
|
}
|
466 |
|
|
|
467 |
7786
|
tao
|
/*
|
468 |
|
|
* Insert a SolrDoc to the solr server.
|
469 |
|
|
*/
|
470 |
|
|
private synchronized void insertToIndex(SolrDoc doc) throws SolrServerException, IOException {
|
471 |
|
|
if(doc != null ) {
|
472 |
|
|
SolrInputDocument solrDoc = new SolrInputDocument();
|
473 |
|
|
List<SolrElementField> list = doc.getFieldList();
|
474 |
|
|
if(list != null) {
|
475 |
|
|
//solrDoc.addField(METACATPIDFIELD, pid);
|
476 |
|
|
Iterator<SolrElementField> iterator = list.iterator();
|
477 |
|
|
while (iterator.hasNext()) {
|
478 |
|
|
SolrElementField field = iterator.next();
|
479 |
|
|
if(field != null) {
|
480 |
|
|
String value = field.getValue();
|
481 |
|
|
String name = field.getName();
|
482 |
|
|
//System.out.println("add name/value pair - "+name+"/"+value);
|
483 |
|
|
solrDoc.addField(name, value);
|
484 |
7546
|
tao
|
}
|
485 |
|
|
}
|
486 |
|
|
}
|
487 |
7786
|
tao
|
if(!solrDoc.isEmpty()) {
|
488 |
7856
|
tao
|
/*IndexEvent event = new IndexEvent();
|
489 |
7800
|
tao
|
event.setDate(Calendar.getInstance().getTime());
|
490 |
|
|
Identifier pid = new Identifier();
|
491 |
|
|
pid.setValue(doc.getIdentifier());
|
492 |
7856
|
tao
|
event.setIdentifier(pid);*/
|
493 |
7800
|
tao
|
try {
|
494 |
7801
|
tao
|
UpdateResponse response = solrServer.add(solrDoc);
|
495 |
|
|
solrServer.commit();
|
496 |
7805
|
tao
|
/*event.setType(IndexEvent.SUCCESSINSERT);
|
497 |
7801
|
tao
|
event.setDescription("Successfully insert the solr index for the id "+pid.getValue());
|
498 |
|
|
try {
|
499 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
500 |
|
|
} catch (Exception e) {
|
501 |
7802
|
tao
|
log.error("SolrIndex.insertToIndex - IndexEventLog can't log the index inserting event :"+e.getMessage());
|
502 |
7805
|
tao
|
}*/
|
503 |
7801
|
tao
|
} catch (SolrServerException e) {
|
504 |
7856
|
tao
|
/*event.setAction(Event.CREATE);
|
505 |
7801
|
tao
|
event.setDescription("Failed to insert the solr index for the id "+pid.getValue()+" since "+e.getMessage());
|
506 |
|
|
try {
|
507 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
508 |
|
|
} catch (Exception ee) {
|
509 |
7802
|
tao
|
log.error("SolrIndex.insertToIndex - IndexEventLog can't log the index inserting event :"+ee.getMessage());
|
510 |
7856
|
tao
|
}*/
|
511 |
7801
|
tao
|
throw e;
|
512 |
|
|
} catch (IOException e) {
|
513 |
7856
|
tao
|
/*event.setAction(Event.CREATE);
|
514 |
7801
|
tao
|
event.setDescription("Failed to insert the solr index for the id "+pid.getValue()+" since "+e.getMessage());
|
515 |
|
|
try {
|
516 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
517 |
|
|
} catch (Exception ee) {
|
518 |
7802
|
tao
|
log.error("SolrIndex.insertToIndex - IndexEventLog can't log the index inserting event :"+ee.getMessage());
|
519 |
7856
|
tao
|
}*/
|
520 |
7801
|
tao
|
throw e;
|
521 |
|
|
|
522 |
7800
|
tao
|
}
|
523 |
7786
|
tao
|
//System.out.println("=================the response is:\n"+response.toString());
|
524 |
|
|
}
|
525 |
7546
|
tao
|
}
|
526 |
|
|
}
|
527 |
7577
|
tao
|
|
528 |
|
|
/**
|
529 |
7627
|
tao
|
* Update the solr index. This method handles the three scenarios:
|
530 |
8343
|
tao
|
* 1. Remove an existing doc - if the the system metadata shows the value of the archive is true,
|
531 |
7627
|
tao
|
* remove the index for the previous version(s) and generate new index for the doc.
|
532 |
8343
|
tao
|
* 2. Add a new doc - if the system metadata shows the value of the archive is false, generate the
|
533 |
7627
|
tao
|
* index for the doc.
|
534 |
7577
|
tao
|
*/
|
535 |
8343
|
tao
|
public void update(Identifier pid, SystemMetadata systemMetadata) {
|
536 |
8893
|
tao
|
if(systemMetadata==null || pid==null) {
|
537 |
|
|
log.error("SolrIndex.update - the systemMetadata or pid is null. So nothing will be indexed.");
|
538 |
|
|
return;
|
539 |
|
|
}
|
540 |
8343
|
tao
|
String objectPath = null;
|
541 |
|
|
try {
|
542 |
8893
|
tao
|
if(!systemMetadata.getArchived()) {
|
543 |
|
|
objectPath = DistributedMapsFactory.getObjectPathMap().get(pid);
|
544 |
|
|
}
|
545 |
8766
|
leinfelder
|
update(pid, systemMetadata, objectPath);
|
546 |
8343
|
tao
|
EventlogFactory.createIndexEventLog().remove(pid);
|
547 |
|
|
} catch (Exception e) {
|
548 |
|
|
String error = "SolrIndex.update - could not update the solr index since " + e.getMessage();
|
549 |
|
|
writeEventLog(systemMetadata, pid, error);
|
550 |
|
|
log.error(error, e);
|
551 |
|
|
}
|
552 |
7577
|
tao
|
}
|
553 |
8893
|
tao
|
|
554 |
7603
|
tao
|
|
555 |
8343
|
tao
|
/**
|
556 |
|
|
* Update the solr index. This method handles the three scenarios:
|
557 |
|
|
* 1. Remove an existing doc - if the the system metadata shows the value of the archive is true,
|
558 |
|
|
* remove the index for the previous version(s) and generate new index for the doc.
|
559 |
|
|
* 2. Add a new doc - if the system metadata shows the value of the archive is false, generate the
|
560 |
|
|
* index for the doc.
|
561 |
|
|
* @param pid
|
562 |
|
|
* @param systemMetadata
|
563 |
|
|
* @param data
|
564 |
|
|
* @throws SolrServerException
|
565 |
|
|
* @throws ServiceFailure
|
566 |
|
|
* @throws XPathExpressionException
|
567 |
|
|
* @throws NotImplemented
|
568 |
|
|
* @throws NotFound
|
569 |
|
|
* @throws UnsupportedType
|
570 |
|
|
* @throws IOException
|
571 |
|
|
* @throws SAXException
|
572 |
|
|
* @throws ParserConfigurationException
|
573 |
|
|
* @throws OREParserException
|
574 |
|
|
* @throws JiBXException
|
575 |
|
|
* @throws EncoderException
|
576 |
|
|
*/
|
577 |
9060
|
leinfelder
|
void update(Identifier pid, SystemMetadata systemMetadata, String objectPath) throws Exception {
|
578 |
8893
|
tao
|
//checkParams(pid, systemMetadata, objectPath);
|
579 |
|
|
if(systemMetadata==null || pid==null) {
|
580 |
|
|
log.error("SolrIndex.update - the systemMetadata or pid is null. So nothing will be indexed.");
|
581 |
|
|
return;
|
582 |
|
|
}
|
583 |
8722
|
leinfelder
|
boolean isArchive = systemMetadata.getArchived() != null && systemMetadata.getArchived();
|
584 |
8343
|
tao
|
if(isArchive ) {
|
585 |
|
|
//delete the index for the archived objects
|
586 |
|
|
remove(pid.getValue(), systemMetadata);
|
587 |
|
|
log.info("SolrIndex.update============================= archive the idex for the identifier "+pid);
|
588 |
|
|
} else {
|
589 |
|
|
//generate index for either add or update.
|
590 |
8766
|
leinfelder
|
insert(pid, systemMetadata, objectPath);
|
591 |
8343
|
tao
|
log.info("SolrIndex.update============================= insert index for the identifier "+pid);
|
592 |
|
|
}
|
593 |
|
|
}
|
594 |
7627
|
tao
|
|
595 |
7786
|
tao
|
|
596 |
|
|
|
597 |
|
|
/*
|
598 |
|
|
* Is the pid a resource map
|
599 |
|
|
*/
|
600 |
8343
|
tao
|
private boolean isDataPackage(String pid, SystemMetadata sysmeta) throws FileNotFoundException, ServiceFailure {
|
601 |
7786
|
tao
|
boolean isDataPackage = false;
|
602 |
8343
|
tao
|
//SystemMetadata sysmeta = DistributedMapsFactory.getSystemMetadata(pid);
|
603 |
7786
|
tao
|
if(sysmeta != null) {
|
604 |
8352
|
tao
|
isDataPackage = IndexGeneratorTimerTask.isResourceMap(sysmeta.getFormatId());
|
605 |
7786
|
tao
|
}
|
606 |
|
|
return isDataPackage;
|
607 |
|
|
}
|
608 |
|
|
|
609 |
|
|
private boolean isPartOfDataPackage(String pid) throws XPathExpressionException, NotImplemented, NotFound, UnsupportedType, SolrServerException, IOException, ParserConfigurationException, SAXException {
|
610 |
|
|
SolrDoc dataPackageIndexDoc = ResourceMapSubprocessor.getSolrDoc(pid);
|
611 |
|
|
if (dataPackageIndexDoc != null) {
|
612 |
|
|
String resourceMapId = dataPackageIndexDoc
|
613 |
|
|
.getFirstFieldValue(SolrElementField.FIELD_RESOURCEMAP);
|
614 |
|
|
return StringUtils.isNotEmpty(resourceMapId);
|
615 |
|
|
} else {
|
616 |
|
|
return false;
|
617 |
|
|
}
|
618 |
|
|
}
|
619 |
8343
|
tao
|
/**
|
620 |
|
|
* Remove the indexed associated with specified pid.
|
621 |
|
|
* @param pid the pid which the indexes are associated with
|
622 |
|
|
* @throws IOException
|
623 |
|
|
* @throws SolrServerException
|
624 |
|
|
* @throws ParserConfigurationException
|
625 |
|
|
* @throws SAXException
|
626 |
|
|
* @throws UnsupportedType
|
627 |
|
|
* @throws NotFound
|
628 |
|
|
* @throws NotImplemented
|
629 |
|
|
* @throws XPathExpressionException
|
630 |
|
|
* @throws ServiceFailure
|
631 |
|
|
* @throws OREParserException
|
632 |
|
|
*/
|
633 |
9060
|
leinfelder
|
private void remove(String pid, SystemMetadata sysmeta) throws Exception {
|
634 |
8343
|
tao
|
if (isDataPackage(pid, sysmeta)) {
|
635 |
|
|
removeDataPackage(pid);
|
636 |
|
|
} else if (isPartOfDataPackage(pid)) {
|
637 |
|
|
removeFromDataPackage(pid);
|
638 |
|
|
} else {
|
639 |
|
|
removeFromIndex(pid);
|
640 |
|
|
}
|
641 |
|
|
}
|
642 |
|
|
|
643 |
|
|
/*
|
644 |
8893
|
tao
|
* Remove the resource map from the solr index. It doesn't only remove the index for itself and also
|
645 |
|
|
* remove the relationship for the related metadata and data objects.
|
646 |
8864
|
tao
|
*/
|
647 |
9060
|
leinfelder
|
private void removeDataPackage(String pid) throws Exception {
|
648 |
8893
|
tao
|
removeFromIndex(pid);
|
649 |
|
|
List<SolrDoc> docsToUpdate = getUpdatedSolrDocsByRemovingResourceMap(pid);
|
650 |
|
|
if (docsToUpdate != null && !docsToUpdate.isEmpty()) {
|
651 |
|
|
//SolrElementAdd addCommand = new SolrElementAdd(docsToUpdate);
|
652 |
|
|
//httpService.sendUpdate(solrIndexUri, addCommand);
|
653 |
|
|
for(SolrDoc doc : docsToUpdate) {
|
654 |
|
|
removeFromIndex(doc.getIdentifier());
|
655 |
|
|
insertToIndex(doc);
|
656 |
|
|
}
|
657 |
|
|
}
|
658 |
|
|
|
659 |
8864
|
tao
|
}
|
660 |
8893
|
tao
|
|
661 |
8864
|
tao
|
/*
|
662 |
8893
|
tao
|
* Get the list of the solr doc which need to be updated because the removal of the resource map
|
663 |
8864
|
tao
|
*/
|
664 |
8893
|
tao
|
private List<SolrDoc> getUpdatedSolrDocsByRemovingResourceMap(String resourceMapId)
|
665 |
|
|
throws UnsupportedType, NotFound, SolrServerException, ParserConfigurationException, SAXException, MalformedURLException, IOException, XPathExpressionException {
|
666 |
|
|
List<SolrDoc> updatedSolrDocs = null;
|
667 |
|
|
if (resourceMapId != null && !resourceMapId.trim().equals("")) {
|
668 |
|
|
/*List<SolrDoc> docsContainResourceMap = httpService.getDocumentsByResourceMap(
|
669 |
|
|
solrQueryUri, resourceMapId);*/
|
670 |
|
|
List<SolrDoc> docsContainResourceMap = ResourceMapSubprocessor.getDocumentsByResourceMap(resourceMapId);
|
671 |
|
|
updatedSolrDocs = removeResourceMapRelationship(docsContainResourceMap,
|
672 |
|
|
resourceMapId);
|
673 |
|
|
}
|
674 |
|
|
return updatedSolrDocs;
|
675 |
8864
|
tao
|
}
|
676 |
8893
|
tao
|
|
677 |
|
|
/*
|
678 |
|
|
* Get the list of the solr doc which need to be updated because the removal of the resource map
|
679 |
|
|
*/
|
680 |
|
|
private List<SolrDoc> removeResourceMapRelationship(List<SolrDoc> docsContainResourceMap,
|
681 |
|
|
String resourceMapId) throws XPathExpressionException, IOException {
|
682 |
|
|
List<SolrDoc> totalUpdatedSolrDocs = new ArrayList<SolrDoc>();
|
683 |
|
|
if (docsContainResourceMap != null && !docsContainResourceMap.isEmpty()) {
|
684 |
|
|
for (SolrDoc doc : docsContainResourceMap) {
|
685 |
|
|
List<SolrDoc> updatedSolrDocs = new ArrayList<SolrDoc>();
|
686 |
|
|
List<String> resourceMapIdStrs = doc
|
687 |
|
|
.getAllFieldValues(SolrElementField.FIELD_RESOURCEMAP);
|
688 |
|
|
List<String> dataIdStrs = doc
|
689 |
|
|
.getAllFieldValues(SolrElementField.FIELD_DOCUMENTS);
|
690 |
|
|
List<String> metadataIdStrs = doc
|
691 |
|
|
.getAllFieldValues(SolrElementField.FIELD_ISDOCUMENTEDBY);
|
692 |
|
|
if ((dataIdStrs == null || dataIdStrs.isEmpty())
|
693 |
|
|
&& (metadataIdStrs == null || metadataIdStrs.isEmpty())) {
|
694 |
|
|
// only has resourceMap field, doesn't have either documentBy or documents fields.
|
695 |
|
|
// so we only remove the resource map field.
|
696 |
|
|
doc.removeFieldsWithValue(SolrElementField.FIELD_RESOURCEMAP, resourceMapId);
|
697 |
|
|
updatedSolrDocs.add(doc);
|
698 |
|
|
} else if ((dataIdStrs != null && !dataIdStrs.isEmpty())
|
699 |
|
|
&& (metadataIdStrs == null || metadataIdStrs.isEmpty())) {
|
700 |
|
|
//The solr doc is for a metadata object since the solr doc documents data files
|
701 |
|
|
updatedSolrDocs = removeAggregatedItems(resourceMapId, doc, resourceMapIdStrs,
|
702 |
|
|
dataIdStrs, SolrElementField.FIELD_DOCUMENTS);
|
703 |
|
|
} else if ((dataIdStrs == null || dataIdStrs.isEmpty())
|
704 |
|
|
&& (metadataIdStrs != null && !metadataIdStrs.isEmpty())) {
|
705 |
|
|
//The solr doc is for a data object since it documentedBy elements.
|
706 |
|
|
updatedSolrDocs = removeAggregatedItems(resourceMapId, doc, resourceMapIdStrs,
|
707 |
|
|
metadataIdStrs, SolrElementField.FIELD_ISDOCUMENTEDBY);
|
708 |
|
|
} else if ((dataIdStrs != null && !dataIdStrs.isEmpty())
|
709 |
|
|
&& (metadataIdStrs != null && !metadataIdStrs.isEmpty())){
|
710 |
|
|
// both metadata and data for one object
|
711 |
|
|
List<SolrDoc> solrDocsRemovedDocuments = removeAggregatedItems(resourceMapId, doc, resourceMapIdStrs,
|
712 |
|
|
dataIdStrs, SolrElementField.FIELD_DOCUMENTS);
|
713 |
|
|
List<SolrDoc> solrDocsRemovedDocumentBy = removeAggregatedItems(resourceMapId, doc, resourceMapIdStrs,
|
714 |
|
|
metadataIdStrs, SolrElementField.FIELD_ISDOCUMENTEDBY);
|
715 |
|
|
updatedSolrDocs = mergeUpdatedSolrDocs(solrDocsRemovedDocumentBy, solrDocsRemovedDocuments);
|
716 |
|
|
}
|
717 |
|
|
//move them to the final result
|
718 |
|
|
if(updatedSolrDocs != null) {
|
719 |
|
|
for(SolrDoc updatedDoc: updatedSolrDocs) {
|
720 |
|
|
totalUpdatedSolrDocs.add(updatedDoc);
|
721 |
|
|
}
|
722 |
|
|
}
|
723 |
|
|
|
724 |
|
|
}
|
725 |
|
|
|
726 |
|
|
}
|
727 |
|
|
return totalUpdatedSolrDocs;
|
728 |
|
|
}
|
729 |
8864
|
tao
|
|
730 |
|
|
/*
|
731 |
8893
|
tao
|
* Process the list of ids of the documentBy/documents in a slor doc.
|
732 |
8343
|
tao
|
*/
|
733 |
8893
|
tao
|
private List<SolrDoc> removeAggregatedItems(String targetResourceMapId, SolrDoc doc,
|
734 |
|
|
List<String> resourceMapIdsInDoc, List<String> aggregatedItemsInDoc, String fieldNameRemoved) {
|
735 |
|
|
List<SolrDoc> updatedSolrDocs = new ArrayList<SolrDoc>();
|
736 |
|
|
if (doc != null && resourceMapIdsInDoc != null && aggregatedItemsInDoc != null
|
737 |
|
|
&& fieldNameRemoved != null) {
|
738 |
|
|
if (resourceMapIdsInDoc.size() == 1) {
|
739 |
|
|
//only has one resource map. remove the resource map. also remove the documentBy
|
740 |
|
|
doc.removeFieldsWithValue(SolrElementField.FIELD_RESOURCEMAP, targetResourceMapId);
|
741 |
|
|
doc.removeAllFields(fieldNameRemoved);
|
742 |
|
|
updatedSolrDocs.add(doc);
|
743 |
|
|
} else if (resourceMapIdsInDoc.size() > 1) {
|
744 |
|
|
//we have multiple resource maps. We should match them.
|
745 |
|
|
Map<String, String> ids = matchResourceMapsAndItems(doc.getIdentifier(),
|
746 |
|
|
targetResourceMapId, resourceMapIdsInDoc, aggregatedItemsInDoc, fieldNameRemoved);
|
747 |
|
|
if (ids != null) {
|
748 |
|
|
for (String id : ids.keySet()) {
|
749 |
|
|
doc.removeFieldsWithValue(fieldNameRemoved, id);
|
750 |
|
|
}
|
751 |
|
|
}
|
752 |
|
|
doc.removeFieldsWithValue(SolrElementField.FIELD_RESOURCEMAP,
|
753 |
|
|
targetResourceMapId);
|
754 |
|
|
updatedSolrDocs.add(doc);
|
755 |
|
|
/*if (aggregatedItemsInDoc.size() > 1) {
|
756 |
|
|
|
757 |
7786
|
tao
|
|
758 |
8893
|
tao
|
} else {
|
759 |
|
|
//multiple resource map aggregate same metadata and data. Just remove the resource map
|
760 |
|
|
doc.removeFieldsWithValue(SolrElementField.FIELD_RESOURCEMAP,
|
761 |
|
|
targetResourceMapId);
|
762 |
|
|
updatedSolrDocs.add(doc);
|
763 |
|
|
}*/
|
764 |
8343
|
tao
|
}
|
765 |
8893
|
tao
|
}
|
766 |
|
|
return updatedSolrDocs;
|
767 |
|
|
}
|
768 |
8343
|
tao
|
|
769 |
8893
|
tao
|
/*
|
770 |
|
|
* Return a map of mapping aggregation id map the target resourceMapId.
|
771 |
|
|
* This will look the aggregation information in another side - If the targetId
|
772 |
|
|
* is a metadata object, we will look the data objects which it describes; If
|
773 |
|
|
* the targetId is a data object, we will look the metadata object which documents it.
|
774 |
|
|
*/
|
775 |
|
|
private Map<String, String> matchResourceMapsAndItems(String targetId,
|
776 |
|
|
String targetResourceMapId, List<String> originalResourceMaps, List<String> aggregatedItems, String fieldName) {
|
777 |
|
|
Map<String, String> map = new HashMap<String, String>();
|
778 |
|
|
if (targetId != null && targetResourceMapId != null && aggregatedItems != null
|
779 |
|
|
&& fieldName != null) {
|
780 |
|
|
String newFieldName = null;
|
781 |
|
|
if (fieldName.equals(SolrElementField.FIELD_ISDOCUMENTEDBY)) {
|
782 |
|
|
newFieldName = SolrElementField.FIELD_DOCUMENTS;
|
783 |
|
|
} else if (fieldName.equals(SolrElementField.FIELD_DOCUMENTS)) {
|
784 |
|
|
newFieldName = SolrElementField.FIELD_ISDOCUMENTEDBY;
|
785 |
|
|
}
|
786 |
|
|
if (newFieldName != null) {
|
787 |
|
|
for (String item : aggregatedItems) {
|
788 |
|
|
SolrDoc doc = null;
|
789 |
|
|
try {
|
790 |
|
|
doc = getDocumentById(item);
|
791 |
|
|
List<String> fieldValues = doc.getAllFieldValues(newFieldName);
|
792 |
|
|
List<String> resourceMapIds = doc
|
793 |
|
|
.getAllFieldValues(SolrElementField.FIELD_RESOURCEMAP);
|
794 |
|
|
if ((fieldValues != null && fieldValues.contains(targetId))
|
795 |
|
|
&& (resourceMapIds != null && resourceMapIds
|
796 |
|
|
.contains(targetResourceMapId))) {
|
797 |
|
|
//okay, we found the target aggregation item id and the resource map id
|
798 |
|
|
//in this solr doc. However, we need check if another resource map with different
|
799 |
|
|
//id but specify the same relationship. If we have the id(s), we should not
|
800 |
|
|
// remove the documents( or documentBy) element since we need to preserve the
|
801 |
|
|
// relationship for the remain resource map.
|
802 |
|
|
boolean hasDuplicateIds = false;
|
803 |
|
|
if(originalResourceMaps != null) {
|
804 |
|
|
for(String id :resourceMapIds) {
|
805 |
|
|
if (originalResourceMaps.contains(id) && !id.equals(targetResourceMapId)) {
|
806 |
|
|
hasDuplicateIds = true;
|
807 |
|
|
break;
|
808 |
|
|
}
|
809 |
|
|
}
|
810 |
|
|
}
|
811 |
|
|
if(!hasDuplicateIds) {
|
812 |
|
|
map.put(item, targetResourceMapId);
|
813 |
|
|
}
|
814 |
|
|
|
815 |
|
|
}
|
816 |
|
|
} catch (Exception e) {
|
817 |
|
|
log.warn("SolrIndex.matchResourceMapsAndItems - can't get the solrdoc for the id "
|
818 |
|
|
+ item + " since " + e.getMessage());
|
819 |
|
|
}
|
820 |
|
|
}
|
821 |
|
|
}
|
822 |
|
|
}
|
823 |
|
|
return map;
|
824 |
|
|
}
|
825 |
8343
|
tao
|
|
826 |
8893
|
tao
|
/*
|
827 |
|
|
* Get the solr index doc from the index server for the given id.
|
828 |
|
|
*/
|
829 |
|
|
private SolrDoc getDocumentById(String id) throws NotImplemented, NotFound, UnsupportedType,
|
830 |
|
|
SolrServerException, ParserConfigurationException, SAXException, XPathExpressionException, IOException {
|
831 |
|
|
SolrDoc doc = ResourceMapSubprocessor.getSolrDoc(id);
|
832 |
|
|
return doc;
|
833 |
|
|
}
|
834 |
|
|
|
835 |
|
|
/*
|
836 |
|
|
* Merge two list of updated solr docs. removedDocumentBy has the correct information about documentBy element.
|
837 |
|
|
* removedDocuments has the correct information about the documents element.
|
838 |
|
|
* So we go through the two list and found the two docs having the same identifier.
|
839 |
|
|
* Get the list of the documents value from the one in the removedDoucments (1).
|
840 |
|
|
* Remove all values of documents from the one in the removedDocumentBy.
|
841 |
|
|
* Then copy the list of documents value from (1) to to the one in the removedDocumentBy.
|
842 |
|
|
*/
|
843 |
|
|
private List<SolrDoc> mergeUpdatedSolrDocs(List<SolrDoc>removedDocumentBy, List<SolrDoc>removedDocuments) {
|
844 |
|
|
List<SolrDoc> mergedDocuments = new ArrayList<SolrDoc>();
|
845 |
|
|
if(removedDocumentBy == null || removedDocumentBy.isEmpty()) {
|
846 |
|
|
mergedDocuments = removedDocuments;
|
847 |
|
|
} else if (removedDocuments == null || removedDocuments.isEmpty()) {
|
848 |
|
|
mergedDocuments = removedDocumentBy;
|
849 |
|
|
} else {
|
850 |
|
|
int sizeOfDocBy = removedDocumentBy.size();
|
851 |
|
|
int sizeOfDocs = removedDocuments.size();
|
852 |
|
|
for(int i=sizeOfDocBy-1; i>= 0; i--) {
|
853 |
|
|
SolrDoc docInRemovedDocBy = removedDocumentBy.get(i);
|
854 |
|
|
for(int j= sizeOfDocs-1; j>=0; j--) {
|
855 |
|
|
SolrDoc docInRemovedDocs = removedDocuments.get(j);
|
856 |
|
|
if(docInRemovedDocBy.getIdentifier().equals(docInRemovedDocs.getIdentifier())) {
|
857 |
|
|
//find the same doc in both list. let's merge them.
|
858 |
|
|
//first get all the documents element from the docWithDocs(it has the correct information about the documents element)
|
859 |
|
|
List<String> idsInDocuments = docInRemovedDocs.getAllFieldValues(SolrElementField.FIELD_DOCUMENTS);
|
860 |
|
|
docInRemovedDocBy.removeAllFields(SolrElementField.FIELD_DOCUMENTS);//clear out any documents element in docInRemovedDocBy
|
861 |
|
|
//add the Documents element from the docInRemovedDocs if it has any.
|
862 |
|
|
// The docInRemovedDocs has the correct information about the documentBy. Now it copied the correct information of the documents element.
|
863 |
|
|
// So docInRemovedDocs has both correct information about the documentBy and documents elements.
|
864 |
|
|
if(idsInDocuments != null) {
|
865 |
|
|
for(String id : idsInDocuments) {
|
866 |
|
|
if(id != null && !id.trim().equals("")) {
|
867 |
|
|
docInRemovedDocBy.addField(new SolrElementField(SolrElementField.FIELD_DOCUMENTS, id));
|
868 |
|
|
}
|
869 |
|
|
|
870 |
|
|
}
|
871 |
|
|
}
|
872 |
|
|
//intersect the resource map ids.
|
873 |
|
|
List<String> resourceMapIdsInWithDocs = docInRemovedDocs.getAllFieldValues(SolrElementField.FIELD_RESOURCEMAP);
|
874 |
|
|
List<String> resourceMapIdsInWithDocBy = docInRemovedDocBy.getAllFieldValues(SolrElementField.FIELD_RESOURCEMAP);
|
875 |
|
|
docInRemovedDocBy.removeAllFields(SolrElementField.FIELD_RESOURCEMAP);
|
876 |
|
|
Collection resourceMapIds = CollectionUtils.union(resourceMapIdsInWithDocs, resourceMapIdsInWithDocBy);
|
877 |
|
|
if(resourceMapIds != null) {
|
878 |
|
|
for(Object idObj : resourceMapIds) {
|
879 |
|
|
String id = (String)idObj;
|
880 |
|
|
docInRemovedDocBy.addField(new SolrElementField(SolrElementField.FIELD_RESOURCEMAP, id));
|
881 |
|
|
}
|
882 |
|
|
}
|
883 |
|
|
//we don't need do anything about the documentBy elements since the docInRemovedDocBy has the correct information.
|
884 |
|
|
mergedDocuments.add(docInRemovedDocBy);
|
885 |
|
|
//delete the two documents from the list
|
886 |
|
|
removedDocumentBy.remove(i);
|
887 |
|
|
removedDocuments.remove(j);
|
888 |
|
|
break;
|
889 |
8343
|
tao
|
}
|
890 |
8893
|
tao
|
|
891 |
8343
|
tao
|
}
|
892 |
|
|
}
|
893 |
8893
|
tao
|
// when we get there, if the two lists are empty, this will be a perfect merge. However, if something are left. we
|
894 |
|
|
//just put them in.
|
895 |
|
|
for(SolrDoc doc: removedDocumentBy) {
|
896 |
|
|
mergedDocuments.add(doc);
|
897 |
|
|
}
|
898 |
|
|
for(SolrDoc doc: removedDocuments) {
|
899 |
|
|
mergedDocuments.add(doc);
|
900 |
|
|
}
|
901 |
8343
|
tao
|
}
|
902 |
8893
|
tao
|
return mergedDocuments;
|
903 |
8343
|
tao
|
}
|
904 |
8893
|
tao
|
|
905 |
8343
|
tao
|
|
906 |
8344
|
tao
|
/*
|
907 |
|
|
* Remove a pid which is part of resource map.
|
908 |
|
|
*/
|
909 |
9060
|
leinfelder
|
private void removeFromDataPackage(String pid) throws Exception {
|
910 |
8343
|
tao
|
SolrDoc indexedDoc = ResourceMapSubprocessor.getSolrDoc(pid);
|
911 |
|
|
removeFromIndex(pid);
|
912 |
|
|
List<SolrDoc> docsToUpdate = new ArrayList<SolrDoc>();
|
913 |
|
|
|
914 |
|
|
List<String> documents = indexedDoc.getAllFieldValues(SolrElementField.FIELD_DOCUMENTS);
|
915 |
|
|
for (String documentsValue : documents) {
|
916 |
|
|
SolrDoc solrDoc = ResourceMapSubprocessor.getSolrDoc(documentsValue);
|
917 |
|
|
solrDoc.removeFieldsWithValue(SolrElementField.FIELD_ISDOCUMENTEDBY, pid);
|
918 |
|
|
removeFromIndex(documentsValue);
|
919 |
|
|
insertToIndex(solrDoc);
|
920 |
|
|
}
|
921 |
|
|
|
922 |
|
|
List<String> documentedBy = indexedDoc
|
923 |
|
|
.getAllFieldValues(SolrElementField.FIELD_ISDOCUMENTEDBY);
|
924 |
|
|
for (String documentedByValue : documentedBy) {
|
925 |
|
|
SolrDoc solrDoc = ResourceMapSubprocessor.getSolrDoc(documentedByValue);
|
926 |
8345
|
tao
|
solrDoc.removeFieldsWithValue(SolrElementField.FIELD_DOCUMENTS, pid);
|
927 |
8343
|
tao
|
//docsToUpdate.add(solrDoc);
|
928 |
|
|
removeFromIndex(documentedByValue);
|
929 |
|
|
insertToIndex(solrDoc);
|
930 |
|
|
}
|
931 |
|
|
|
932 |
|
|
//SolrElementAdd addCommand = new SolrElementAdd(docsToUpdate);
|
933 |
|
|
//httpService.sendUpdate(solrIndexUri, addCommand);
|
934 |
|
|
}
|
935 |
|
|
|
936 |
|
|
/*
|
937 |
|
|
* Remove a pid from the solr index
|
938 |
|
|
*/
|
939 |
9060
|
leinfelder
|
private synchronized void removeFromIndex(String identifier) throws Exception {
|
940 |
|
|
|
941 |
|
|
|
942 |
|
|
Map<String, SolrDoc> docs = new HashMap<String, SolrDoc>();
|
943 |
|
|
|
944 |
|
|
for (IDocumentDeleteSubprocessor deleteSubprocessor : deleteSubprocessors) {
|
945 |
|
|
docs.putAll(deleteSubprocessor.processDocForDelete(identifier, docs));
|
946 |
|
|
}
|
947 |
|
|
List<SolrDoc> docsToUpdate = new ArrayList<SolrDoc>();
|
948 |
|
|
List<String> idsToIndex = new ArrayList<String>();
|
949 |
|
|
for (String idToUpdate : docs.keySet()) {
|
950 |
|
|
if (docs.get(idToUpdate) != null) {
|
951 |
|
|
docsToUpdate.add(docs.get(idToUpdate));
|
952 |
|
|
} else {
|
953 |
|
|
idsToIndex.add(idToUpdate);
|
954 |
|
|
}
|
955 |
|
|
}
|
956 |
|
|
|
957 |
|
|
// update the docs we have
|
958 |
|
|
for (SolrDoc docToUpdate : docsToUpdate) {
|
959 |
|
|
insertToIndex(docToUpdate);
|
960 |
|
|
}
|
961 |
|
|
|
962 |
|
|
// delete this one
|
963 |
|
|
deleteDocFromIndex(identifier);
|
964 |
|
|
|
965 |
|
|
// index the rest
|
966 |
|
|
for (String idToIndex : idsToIndex) {
|
967 |
|
|
Identifier pid = new Identifier();
|
968 |
|
|
pid.setValue(idToIndex);
|
969 |
|
|
SystemMetadata sysMeta = DistributedMapsFactory.getSystemMetadata(idToIndex);
|
970 |
|
|
if (SolrDoc.visibleInIndex(sysMeta)) {
|
971 |
|
|
String objectPath = DistributedMapsFactory.getObjectPathMap().get(pid);
|
972 |
|
|
insert(pid, sysMeta, objectPath);
|
973 |
|
|
}
|
974 |
|
|
}
|
975 |
|
|
|
976 |
|
|
}
|
977 |
|
|
|
978 |
|
|
private void deleteDocFromIndex(String pid) throws Exception {
|
979 |
|
|
if (pid != null && !pid.trim().equals("")) {
|
980 |
8343
|
tao
|
/*IndexEvent event = new IndexEvent();
|
981 |
|
|
event.setDate(Calendar.getInstance().getTime());
|
982 |
|
|
Identifier identifier = new Identifier();
|
983 |
|
|
identifier.setValue(pid);
|
984 |
|
|
event.setIdentifier(identifier);*/
|
985 |
|
|
try {
|
986 |
|
|
solrServer.deleteById(pid);
|
987 |
|
|
solrServer.commit();
|
988 |
|
|
/*event.setType(IndexEvent.SUCCESSDELETE);
|
989 |
|
|
event.setDescription("Successfully remove the solr index for the id "+identifier.getValue());
|
990 |
|
|
try {
|
991 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
992 |
|
|
} catch (Exception e) {
|
993 |
|
|
log.error("SolrIndex.removeFromIndex - IndexEventLog can't log the index deleting event :"+e.getMessage());
|
994 |
|
|
}*/
|
995 |
|
|
} catch (SolrServerException e) {
|
996 |
|
|
/*event.setAction(Event.DELETE);
|
997 |
|
|
event.setDescription("Failurely remove the solr index for the id "+identifier.getValue()+" since "+e.getMessage());
|
998 |
|
|
try {
|
999 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
1000 |
|
|
} catch (Exception ee) {
|
1001 |
|
|
log.error("SolrIndex.removeFromIndex - IndexEventLog can't log the index deleting event :"+ee.getMessage());
|
1002 |
|
|
}*/
|
1003 |
|
|
throw e;
|
1004 |
|
|
|
1005 |
|
|
} catch (IOException e) {
|
1006 |
|
|
/*event.setAction(Event.DELETE);
|
1007 |
|
|
event.setDescription("Failurely remove the solr index for the id "+identifier.getValue()+" since "+e.getMessage());
|
1008 |
|
|
try {
|
1009 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
1010 |
|
|
} catch (Exception ee) {
|
1011 |
|
|
log.error("SolrIndex.removeFromIndex - IndexEventLog can't log the index deleting event :"+ee.getMessage());
|
1012 |
|
|
}*/
|
1013 |
|
|
throw e;
|
1014 |
|
|
}
|
1015 |
|
|
|
1016 |
|
|
}
|
1017 |
9060
|
leinfelder
|
|
1018 |
8343
|
tao
|
}
|
1019 |
|
|
|
1020 |
7569
|
tao
|
/**
|
1021 |
|
|
* Get the solrServer
|
1022 |
|
|
* @return
|
1023 |
|
|
*/
|
1024 |
7604
|
tao
|
public SolrServer getSolrServer() {
|
1025 |
7569
|
tao
|
return solrServer;
|
1026 |
|
|
}
|
1027 |
|
|
|
1028 |
|
|
/**
|
1029 |
7604
|
tao
|
* Set the solrServer.
|
1030 |
7569
|
tao
|
* @param solrServer
|
1031 |
|
|
*/
|
1032 |
7604
|
tao
|
public void setSolrServer(SolrServer solrServer) {
|
1033 |
7569
|
tao
|
this.solrServer = solrServer;
|
1034 |
|
|
}
|
1035 |
7604
|
tao
|
|
1036 |
|
|
/**
|
1037 |
7606
|
tao
|
* Get all indexed ids in the solr server.
|
1038 |
|
|
* @return an empty list if there is no index.
|
1039 |
7604
|
tao
|
* @throws SolrServerException
|
1040 |
|
|
*/
|
1041 |
|
|
public List<String> getSolrIds() throws SolrServerException {
|
1042 |
|
|
List<String> list = new ArrayList<String>();
|
1043 |
|
|
SolrQuery query = new SolrQuery(IDQUERY);
|
1044 |
|
|
query.setRows(Integer.MAX_VALUE);
|
1045 |
|
|
query.setFields(ID);
|
1046 |
|
|
QueryResponse response = solrServer.query(query);
|
1047 |
|
|
SolrDocumentList docs = response.getResults();
|
1048 |
|
|
if(docs != null) {
|
1049 |
|
|
for(SolrDocument doc :docs) {
|
1050 |
|
|
String identifier = (String)doc.getFieldValue(ID);
|
1051 |
|
|
//System.out.println("======================== "+identifier);
|
1052 |
|
|
list.add(identifier);
|
1053 |
|
|
}
|
1054 |
|
|
}
|
1055 |
|
|
return list;
|
1056 |
|
|
}
|
1057 |
8343
|
tao
|
|
1058 |
|
|
private void writeEventLog(SystemMetadata systemMetadata, Identifier pid, String error) {
|
1059 |
|
|
IndexEvent event = new IndexEvent();
|
1060 |
|
|
event.setIdentifier(pid);
|
1061 |
|
|
event.setDate(Calendar.getInstance().getTime());
|
1062 |
|
|
String action = null;
|
1063 |
|
|
if (systemMetadata == null ) {
|
1064 |
|
|
action = Event.CREATE.xmlValue();
|
1065 |
|
|
event.setAction(Event.CREATE);
|
1066 |
|
|
}
|
1067 |
|
|
else if(systemMetadata.getArchived()) {
|
1068 |
|
|
action = Event.DELETE.xmlValue();
|
1069 |
|
|
event.setAction(Event.DELETE);
|
1070 |
|
|
} else {
|
1071 |
|
|
action = Event.CREATE.xmlValue();
|
1072 |
|
|
event.setAction(Event.CREATE);
|
1073 |
|
|
}
|
1074 |
|
|
event.setDescription("Failed to "+action+"the solr index for the id "+pid.getValue()+" since "+error);
|
1075 |
|
|
try {
|
1076 |
|
|
EventlogFactory.createIndexEventLog().write(event);
|
1077 |
|
|
} catch (Exception ee) {
|
1078 |
|
|
log.error("SolrIndex.insertToIndex - IndexEventLog can't log the index inserting event :"+ee.getMessage());
|
1079 |
|
|
}
|
1080 |
|
|
}
|
1081 |
7542
|
tao
|
}
|