/**
* 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);
}
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;
}
