/**
* Find words for a more-like-this query former.
*
* @param docNum the id of the lucene document from which to find terms
*/
private PriorityQueue<ScoreTerm> retrieveTerms(int docNum) throws IOException {
Map<String, Int> termFreqMap = new HashMap<>();
for (String fieldName : fieldNames) {
final Fields vectors = ir.getTermVectors(docNum);
final Terms vector;
if (vectors != null) {
vector = vectors.terms(fieldName);
} else {
vector = null;
}
// field does not store term vector info
if (vector == null) {
StoredDocument d = ir.document(docNum);
StorableField[] fields = d.getFields(fieldName);
for (StorableField field : fields) {
final String stringValue = field.stringValue();
if (stringValue != null) {
addTermFrequencies(new StringReader(stringValue), termFreqMap, fieldName);
}
}
} else {
addTermFrequencies(termFreqMap, vector);
}
}
return createQueue(termFreqMap);
}
/**
* Given a {@link org.apache.solr.schema.SchemaField}, create one or more {@link
* org.apache.lucene.index.IndexableField} instances
*
* @param field the {@link org.apache.solr.schema.SchemaField}
* @param value The value to add to the field
* @param boost The boost to apply
* @return An array of {@link org.apache.lucene.index.IndexableField}
* @see #createField(SchemaField, Object, float)
* @see #isPolyField()
*/
public List<StorableField> createFields(SchemaField field, Object value, float boost) {
StorableField f = createField(field, value, boost);
if (field.hasDocValues() && f.fieldType().docValuesType() == null) {
// field types that support doc values should either override createField
// to return a field with doc values or extend createFields if this can't
// be done in a single field instance (see StrField for example)
throw new UnsupportedOperationException(
"This field type does not support doc values: " + this);
}
return f == null ? Collections.<StorableField>emptyList() : Collections.singletonList(f);
}
private static SolrInputDocument toSolrInputDocument(StoredDocument doc, IndexSchema schema) {
SolrInputDocument out = new SolrInputDocument();
for (StorableField f : doc.getFields()) {
String fname = f.name();
SchemaField sf = schema.getFieldOrNull(f.name());
Object val = null;
if (sf != null) {
if (!sf.stored() || schema.isCopyFieldTarget(sf)) continue;
val = sf.getType().toObject(f); // object or external string?
} else {
val = f.stringValue();
if (val == null) val = f.numericValue();
if (val == null) val = f.binaryValue();
if (val == null) val = f;
}
// todo: how to handle targets of copy fields (including polyfield sub-fields)?
out.addField(fname, val);
}
return out;
}
private static SolrDocument toSolrDoc(StoredDocument doc, IndexSchema schema) {
SolrDocument out = new SolrDocument();
for (StorableField f : doc.getFields()) {
// Make sure multivalued fields are represented as lists
Object existing = out.get(f.name());
if (existing == null) {
SchemaField sf = schema.getFieldOrNull(f.name());
// don't return copyField targets
if (sf != null && schema.isCopyFieldTarget(sf)) continue;
if (sf != null && sf.multiValued()) {
List<Object> vals = new ArrayList<Object>();
vals.add(f);
out.setField(f.name(), vals);
} else {
out.setField(f.name(), f);
}
} else {
out.addField(f.name(), f);
}
}
return out;
}
@Override
public void write(TextResponseWriter writer, String name, StorableField f) throws IOException {
writer.writeStr(name, f.stringValue(), true);
}
/** Given the stored field, return the indexed form */
public String storedToIndexed(StorableField f) {
// right now, the transformation of single valued fields like SortableInt
// is done when the Field is created, not at analysis time... this means
// that the indexed form is the same as the stored field form.
return f.stringValue();
}
/**
* Convert the stored-field format to an external (string, human readable) value
*
* @see #toInternal
*/
public String toExternal(StorableField f) {
// currently used in writing XML of the search result (but perhaps
// a more efficient toXML(IndexableField f, Writer w) should be used
// in the future.
return f.stringValue();
}