public TermInfo collect(String term) throws IOException {
TermInfo info = new TermInfo();
BytesRef luceneTerm = new BytesRef(term.getBytes());
// this gives documents in which the term is found, but no offset information can be retrieved
PostingsEnum postings =
MultiFields.getTermDocsEnum(indexReader, ngramInfoFieldname, luceneTerm);
// now go through each document
int docId = postings.nextDoc();
while (docId != PostingsEnum.NO_MORE_DOCS) {
// get the term vector for that document.
TermsEnum it = indexReader.getTermVector(docId, ngramInfoFieldname).iterator();
// find the term of interest
it.seekExact(luceneTerm);
// get its posting info. this will contain offset info
PostingsEnum postingsInDoc = it.postings(null, PostingsEnum.OFFSETS);
postingsInDoc.nextDoc();
Document doc = indexReader.document(docId);
String id = doc.get(idFieldname);
JATEDocument jd = new JATEDocument(id);
Set<int[]> offsets = new HashSet<>();
int totalFreq = postingsInDoc.freq();
for (int i = 0; i < totalFreq; i++) {
postingsInDoc.nextPosition();
offsets.add(new int[] {postingsInDoc.startOffset(), postingsInDoc.endOffset()});
}
info.getOffsets().put(jd, offsets);
docId = postings.nextDoc();
}
return info;
}
private FieldLookup loadFieldData(String name) {
FieldLookup data = cachedFieldData.get(name);
if (data == null) {
MappedFieldType fieldType = mapperService.smartNameFieldType(name, types);
if (fieldType == null) {
throw new IllegalArgumentException(
"No field found for ["
+ name
+ "] in mapping with types "
+ Arrays.toString(types)
+ "");
}
data = new FieldLookup(fieldType);
cachedFieldData.put(name, data);
}
if (data.fields() == null) {
String fieldName = data.fieldType().names().indexName();
fieldVisitor.reset(fieldName);
try {
reader.document(docId, fieldVisitor);
fieldVisitor.postProcess(data.fieldType());
data.fields(
ImmutableMap.of(name, fieldVisitor.fields().get(data.fieldType().names().indexName())));
} catch (IOException e) {
throw new ElasticsearchParseException("failed to load field [{}]", e, name);
}
}
return data;
}
private Set<Uid> getShardDocUIDs(final IndexShard shard) throws IOException {
shard.refresh("get_uids");
try (Engine.Searcher searcher = shard.acquireSearcher("test")) {
Set<Uid> ids = new HashSet<>();
for (LeafReaderContext leafContext : searcher.reader().leaves()) {
LeafReader reader = leafContext.reader();
Bits liveDocs = reader.getLiveDocs();
for (int i = 0; i < reader.maxDoc(); i++) {
if (liveDocs == null || liveDocs.get(i)) {
Document uuid = reader.document(i, Collections.singleton(UidFieldMapper.NAME));
ids.add(Uid.createUid(uuid.get(UidFieldMapper.NAME)));
}
}
}
return ids;
}
}
