下面列出了org.apache.lucene.index.IndexOptions#DOCS_AND_FREQS 实例代码,或者点击链接到github查看源代码,也可以在右侧发表评论。
private static IndexOptions getIndexOptions(IndexInput input, byte b) throws IOException {
switch (b) {
case 0:
return IndexOptions.NONE;
case 1:
return IndexOptions.DOCS;
case 2:
return IndexOptions.DOCS_AND_FREQS;
case 3:
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
case 4:
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
default:
// BUG
throw new CorruptIndexException("invalid IndexOptions byte: " + b, input);
}
}
private static IndexOptions getIndexOptions(IndexInput input, byte b) throws IOException {
switch (b) {
case 0:
return IndexOptions.NONE;
case 1:
return IndexOptions.DOCS;
case 2:
return IndexOptions.DOCS_AND_FREQS;
case 3:
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
case 4:
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
default:
// BUG
throw new CorruptIndexException("invalid IndexOptions byte: " + b, input);
}
}
public void testSameNormsAsBM25() {
BooleanSimilarity sim1 = new BooleanSimilarity();
BM25Similarity sim2 = new BM25Similarity();
sim2.setDiscountOverlaps(true);
for (int iter = 0; iter < 100; ++iter) {
final int length = TestUtil.nextInt(random(), 1, 100);
final int position = random().nextInt(length);
final int numOverlaps = random().nextInt(length);
final int maxTermFrequency = 1;
final int uniqueTermCount = 1;
FieldInvertState state = new FieldInvertState(Version.LATEST.major, "foo", IndexOptions.DOCS_AND_FREQS, position, length, numOverlaps, 100, maxTermFrequency, uniqueTermCount);
assertEquals(
sim2.computeNorm(state),
sim1.computeNorm(state),
0f);
}
}
public void testSameNormsAsBM25() {
ClassicSimilarity sim1 = new ClassicSimilarity();
BM25Similarity sim2 = new BM25Similarity();
sim2.setDiscountOverlaps(true);
for (int iter = 0; iter < 100; ++iter) {
final int length = TestUtil.nextInt(random(), 1, 1000);
final int position = random().nextInt(length);
final int numOverlaps = random().nextInt(length);
final int maxTermFrequency = 1;
final int uniqueTermCount = 1;
FieldInvertState state = new FieldInvertState(Version.LATEST.major, "foo", IndexOptions.DOCS_AND_FREQS, position, length, numOverlaps, 100, maxTermFrequency, uniqueTermCount);
assertEquals(
sim2.computeNorm(state),
sim1.computeNorm(state),
0f);
}
}
@Override
public IndexOptions indexOptions() {
if (!indexed()) {
return IndexOptions.NONE;
}
IndexOptions options = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
if (omitTermFreqAndPositions()) {
options = IndexOptions.DOCS;
} else if (omitPositions()) {
options = IndexOptions.DOCS_AND_FREQS;
} else if (storeOffsetsWithPositions()) {
options = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
}
return options;
}
/**
* Utility method to create a {@link org.apache.lucene.document.FieldType}
* based on the {@link SchemaField}
*/
public static org.apache.lucene.document.FieldType createFieldType(SchemaField field) {
if (!field.indexed() && !field.stored()) {
log.trace("Ignoring unindexed/unstored field: {}", field);
return null;
}
org.apache.lucene.document.FieldType newType = new org.apache.lucene.document.FieldType();
newType.setTokenized(field.isTokenized());
newType.setStored(field.stored());
newType.setOmitNorms(field.omitNorms());
IndexOptions options = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
if (field.omitTermFreqAndPositions()) {
options = IndexOptions.DOCS;
} else if (field.omitPositions()) {
options = IndexOptions.DOCS_AND_FREQS;
} else if (field.storeOffsetsWithPositions()) {
options = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
}
newType.setIndexOptions(options);
newType.setStoreTermVectors(field.storeTermVector());
newType.setStoreTermVectorOffsets(field.storeTermOffsets());
newType.setStoreTermVectorPositions(field.storeTermPositions());
newType.setStoreTermVectorPayloads(field.storeTermPayloads());
return newType;
}
private static IndexOptions nodeIndexOptionValue(final Object propNode) {
final String value = propNode.toString();
if (INDEX_OPTIONS_OFFSETS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
} else if (INDEX_OPTIONS_POSITIONS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
} else if (INDEX_OPTIONS_FREQS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS;
} else if (INDEX_OPTIONS_DOCS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS;
} else {
throw new ElasticsearchParseException("failed to parse index option [{}]", value);
}
}
public TermVectorLeafReader(String field, Terms terms) {
fields = new Fields() {
@Override
public Iterator<String> iterator() {
return Collections.singletonList(field).iterator();
}
@Override
public Terms terms(String fld) throws IOException {
if (!field.equals(fld)) {
return null;
}
return terms;
}
@Override
public int size() {
return 1;
}
};
IndexOptions indexOptions;
if (!terms.hasFreqs()) {
indexOptions = IndexOptions.DOCS;
} else if (!terms.hasPositions()) {
indexOptions = IndexOptions.DOCS_AND_FREQS;
} else if (!terms.hasOffsets()) {
indexOptions = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
} else {
indexOptions = IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
}
FieldInfo fieldInfo = new FieldInfo(field, 0,
true, true, terms.hasPayloads(),
indexOptions, DocValuesType.NONE, -1, Collections.emptyMap(), 0, 0, 0, false);
fieldInfos = new FieldInfos(new FieldInfo[]{fieldInfo});
}
public void testDiscountOverlapsBoost() throws IOException {
BM25Similarity expected = new BM25Similarity();
SimilarityBase actual = new DFRSimilarity(new BasicModelIne(), new AfterEffectB(), new NormalizationH2());
expected.setDiscountOverlaps(false);
actual.setDiscountOverlaps(false);
FieldInvertState state = new FieldInvertState(Version.LATEST.major, "foo", IndexOptions.DOCS_AND_FREQS);
state.setLength(5);
state.setNumOverlap(2);
assertEquals(expected.computeNorm(state), actual.computeNorm(state));
expected.setDiscountOverlaps(true);
actual.setDiscountOverlaps(true);
assertEquals(expected.computeNorm(state), actual.computeNorm(state));
}
private static IndexOptions nodeIndexOptionValue(final Object propNode) {
final String value = propNode.toString();
if (INDEX_OPTIONS_OFFSETS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS;
} else if (INDEX_OPTIONS_POSITIONS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS_AND_POSITIONS;
} else if (INDEX_OPTIONS_FREQS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS_AND_FREQS;
} else if (INDEX_OPTIONS_DOCS.equalsIgnoreCase(value)) {
return IndexOptions.DOCS;
} else {
throw new ElasticsearchParseException("failed to parse index option [{}]", value);
}
}
@Override
public TextFieldMapper build(BuilderContext context) {
if (positionIncrementGap != POSITION_INCREMENT_GAP_USE_ANALYZER) {
if (fieldType.indexOptions().compareTo(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS) < 0) {
throw new IllegalArgumentException("Cannot set position_increment_gap on field ["
+ name + "] without positions enabled");
}
fieldType.setIndexAnalyzer(new NamedAnalyzer(fieldType.indexAnalyzer(), positionIncrementGap));
fieldType.setSearchAnalyzer(new NamedAnalyzer(fieldType.searchAnalyzer(), positionIncrementGap));
fieldType.setSearchQuoteAnalyzer(new NamedAnalyzer(fieldType.searchQuoteAnalyzer(), positionIncrementGap));
}
setupFieldType(context);
PrefixFieldMapper prefixMapper = null;
if (prefixFieldType != null) {
if (fieldType().isSearchable() == false) {
throw new IllegalArgumentException("Cannot set index_prefixes on unindexed field [" + name() + "]");
}
// Copy the index options of the main field to allow phrase queries on
// the prefix field.
if (context.indexCreatedVersion().onOrAfter(Version.ES_V_6_5_1)) {
if (fieldType.indexOptions() == IndexOptions.DOCS_AND_FREQS) {
// frequencies are not needed because prefix queries always use a constant score
prefixFieldType.setIndexOptions(IndexOptions.DOCS);
} else {
prefixFieldType.setIndexOptions(fieldType.indexOptions());
}
} else if (fieldType.indexOptions() == IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS) {
prefixFieldType.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
}
if (fieldType.storeTermVectorOffsets()) {
prefixFieldType.setStoreTermVectorOffsets(true);
}
prefixFieldType.setAnalyzer(fieldType.indexAnalyzer());
prefixMapper = new PrefixFieldMapper(prefixFieldType, context.indexSettings());
}
if (fieldType().indexPhrases) {
if (fieldType().isSearchable() == false) {
throw new IllegalArgumentException("Cannot set index_phrases on unindexed field [" + name() + "]");
}
if (fieldType.indexOptions().compareTo(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS) < 0) {
throw new IllegalArgumentException("Cannot set index_phrases on field [" + name() + "] if positions are not enabled");
}
}
return new TextFieldMapper(
name,
position,
defaultExpression,
fieldType(),
defaultFieldType,
positionIncrementGap,
prefixMapper,
context.indexSettings(),
multiFieldsBuilder.build(this, context),
copyTo
);
}