1 |
9420
|
leinfelder
|
package edu.ucsb.nceas.metacat.index.annotation;
|
2 |
|
|
|
3 |
|
|
import java.io.IOException;
|
4 |
|
|
import java.io.InputStream;
|
5 |
|
|
import java.util.Iterator;
|
6 |
|
|
import java.util.List;
|
7 |
|
|
import java.util.Map;
|
8 |
|
|
|
9 |
|
|
import javax.xml.xpath.XPathExpressionException;
|
10 |
|
|
|
11 |
|
|
import org.apache.commons.codec.EncoderException;
|
12 |
|
|
import org.apache.commons.logging.Log;
|
13 |
|
|
import org.apache.commons.logging.LogFactory;
|
14 |
|
|
import org.dataone.cn.indexer.annotation.AnnotatorSubprocessor;
|
15 |
|
|
import org.dataone.cn.indexer.solrhttp.SolrDoc;
|
16 |
|
|
import org.dataone.cn.indexer.solrhttp.SolrElementField;
|
17 |
|
|
|
18 |
|
|
import edu.ucsb.nceas.metacat.index.resourcemap.ResourceMapSubprocessor;
|
19 |
|
|
|
20 |
|
|
public class MetacatAnnotatorSubprocessor extends AnnotatorSubprocessor {
|
21 |
|
|
|
22 |
9424
|
leinfelder
|
private static Log log = LogFactory.getLog(MetacatAnnotatorSubprocessor.class);
|
23 |
9420
|
leinfelder
|
|
24 |
|
|
|
25 |
|
|
@Override
|
26 |
|
|
public Map<String, SolrDoc> processDocument(String annotationId, Map<String, SolrDoc> docs,
|
27 |
|
|
InputStream is) throws Exception {
|
28 |
|
|
|
29 |
|
|
// check for annotations, and add them if found
|
30 |
|
|
SolrDoc annotations = parseAnnotation(is);
|
31 |
|
|
if (annotations != null) {
|
32 |
|
|
String referencedPid = annotations.getIdentifier();
|
33 |
|
|
SolrDoc referencedDoc = docs.get(referencedPid);
|
34 |
|
|
|
35 |
|
|
// make sure we have a reference for the document we annotating
|
36 |
9424
|
leinfelder
|
boolean referenceExists = true;
|
37 |
9420
|
leinfelder
|
if (referencedDoc == null) {
|
38 |
|
|
try {
|
39 |
|
|
referencedDoc = ResourceMapSubprocessor.getSolrDoc(referencedPid);
|
40 |
|
|
} catch (Exception e) {
|
41 |
|
|
log.error("Unable to retrieve solr document: " + referencedPid
|
42 |
|
|
+ ". Exception attempting to communicate with solr server.", e);
|
43 |
|
|
}
|
44 |
|
|
|
45 |
9424
|
leinfelder
|
|
46 |
9420
|
leinfelder
|
if (referencedDoc == null) {
|
47 |
|
|
referencedDoc = new SolrDoc();
|
48 |
9424
|
leinfelder
|
referenceExists = false;
|
49 |
9420
|
leinfelder
|
}
|
50 |
|
|
docs.put(referencedPid, referencedDoc);
|
51 |
|
|
}
|
52 |
|
|
|
53 |
|
|
// make sure we say we annotate the object
|
54 |
|
|
SolrDoc annotationDoc = docs.get(annotationId);
|
55 |
|
|
if (annotationDoc != null) {
|
56 |
|
|
annotationDoc.addField(new SolrElementField(FIELD_ANNOTATES, referencedPid));
|
57 |
|
|
}
|
58 |
|
|
|
59 |
|
|
// add the annotations to the referenced document
|
60 |
|
|
Iterator<SolrElementField> annotationIter = annotations.getFieldList().iterator();
|
61 |
|
|
while (annotationIter.hasNext()) {
|
62 |
|
|
SolrElementField annotation = annotationIter.next();
|
63 |
9424
|
leinfelder
|
// only skip merge field if there was an existing record
|
64 |
|
|
if (referenceExists && !this.getFieldsToMerge().contains(annotation.getName())) {
|
65 |
9420
|
leinfelder
|
log.debug("SKIPPING field (not in fieldsToMerge): " + annotation.getName());
|
66 |
|
|
continue;
|
67 |
|
|
}
|
68 |
|
|
referencedDoc.addField(annotation);
|
69 |
|
|
log.debug("ADDING annotation to " + referencedPid + ": " + annotation.getName()
|
70 |
|
|
+ "=" + annotation.getValue());
|
71 |
|
|
}
|
72 |
|
|
} else {
|
73 |
|
|
log.warn("Annotations were not found when parsing: " + annotationId);
|
74 |
|
|
}
|
75 |
|
|
// return the collection that we have augmented
|
76 |
|
|
return docs;
|
77 |
|
|
}
|
78 |
|
|
|
79 |
|
|
/**
|
80 |
|
|
* Merge updates with existing solr documents
|
81 |
|
|
*
|
82 |
|
|
* @param indexDocument
|
83 |
|
|
* @return
|
84 |
|
|
* @throws IOException
|
85 |
|
|
* @throws EncoderException
|
86 |
|
|
* @throws XPathExpressionException
|
87 |
|
|
*/
|
88 |
|
|
public SolrDoc mergeWithIndexedDocument(SolrDoc indexDocument) throws IOException,
|
89 |
|
|
EncoderException, XPathExpressionException {
|
90 |
|
|
|
91 |
|
|
return mergeWithIndexedDocument(indexDocument, getFieldsToMerge());
|
92 |
|
|
}
|
93 |
|
|
|
94 |
|
|
/**
|
95 |
|
|
* Inspired by SubprocessorUtility method, but works with embedded solr server
|
96 |
|
|
* @param indexDocument
|
97 |
|
|
* @param fieldsToMerge
|
98 |
|
|
* @return
|
99 |
|
|
* @throws IOException
|
100 |
|
|
* @throws EncoderException
|
101 |
|
|
* @throws XPathExpressionException
|
102 |
|
|
*/
|
103 |
|
|
private SolrDoc mergeWithIndexedDocument(SolrDoc indexDocument, List<String> fieldsToMerge)
|
104 |
|
|
throws IOException, EncoderException, XPathExpressionException {
|
105 |
|
|
|
106 |
|
|
log.debug("about to merge indexed document with new doc to insert for pid: "
|
107 |
|
|
+ indexDocument.getIdentifier());
|
108 |
|
|
SolrDoc solrDoc = null;
|
109 |
|
|
try {
|
110 |
|
|
solrDoc = ResourceMapSubprocessor.getSolrDoc(indexDocument.getIdentifier());
|
111 |
|
|
} catch (Exception e) {
|
112 |
|
|
log.error("Could not retrieve existing index document: " + indexDocument.getIdentifier(), e);
|
113 |
|
|
}
|
114 |
|
|
if (solrDoc != null) {
|
115 |
|
|
log.debug("found existing doc to merge for pid: " + indexDocument.getIdentifier());
|
116 |
|
|
for (SolrElementField field : solrDoc.getFieldList()) {
|
117 |
|
|
if (fieldsToMerge.contains(field.getName())
|
118 |
|
|
&& !indexDocument.hasFieldWithValue(field.getName(), field.getValue())) {
|
119 |
|
|
indexDocument.addField(field);
|
120 |
|
|
log.debug("merging field: " + field.getName() + " with value: "
|
121 |
|
|
+ field.getValue());
|
122 |
|
|
}
|
123 |
|
|
}
|
124 |
|
|
}
|
125 |
|
|
return indexDocument;
|
126 |
|
|
}
|
127 |
|
|
|
128 |
|
|
}
|