private static TInfo parseTerm(FunctionQParser fp) throws SyntaxError {
TInfo tinfo = new TInfo();
tinfo.indexedField = tinfo.field = fp.parseArg();
tinfo.val = fp.parseArg();
tinfo.indexedBytes = new BytesRef();
FieldType ft = fp.getReq().getSchema().getFieldTypeNoEx(tinfo.field);
if (ft == null) ft = new StrField();
if (ft instanceof TextField) {
// need to do analysis on the term
String indexedVal = tinfo.val;
Query q =
ft.getFieldQuery(fp, fp.getReq().getSchema().getFieldOrNull(tinfo.field), tinfo.val);
if (q instanceof TermQuery) {
Term term = ((TermQuery) q).getTerm();
tinfo.indexedField = term.field();
indexedVal = term.text();
}
UnicodeUtil.UTF16toUTF8(indexedVal, 0, indexedVal.length(), tinfo.indexedBytes);
} else {
ft.readableToIndexed(tinfo.val, tinfo.indexedBytes);
}
return tinfo;
}
/**
* Returns a Query instance for doing searches against a field.
*
* @param parser The {@link org.apache.solr.search.QParser} calling the method
* @param field The {@link org.apache.solr.schema.SchemaField} of the field to search
* @param externalVal The String representation of the value to search
* @return The {@link org.apache.lucene.search.Query} instance. This implementation returns a
* {@link org.apache.lucene.search.TermQuery} but overriding queries may not
*/
public Query getFieldQuery(QParser parser, SchemaField field, String externalVal) {
BytesRefBuilder br = new BytesRefBuilder();
readableToIndexed(externalVal, br);
if (field.hasDocValues() && !field.indexed()) {
// match-only
return getRangeQuery(parser, field, externalVal, externalVal, true, true);
} else {
return new TermQuery(new Term(field.getName(), br));
}
}
@Override
public void process(ResponseBuilder rb) throws IOException {
SolrParams params = rb.req.getParams();
if (!params.getBool(TermsParams.TERMS, false)) return;
String[] fields = params.getParams(TermsParams.TERMS_FIELD);
NamedList<Object> termsResult = new SimpleOrderedMap<>();
rb.rsp.add("terms", termsResult);
if (fields == null || fields.length == 0) return;
int limit = params.getInt(TermsParams.TERMS_LIMIT, 10);
if (limit < 0) {
limit = Integer.MAX_VALUE;
}
String lowerStr = params.get(TermsParams.TERMS_LOWER);
String upperStr = params.get(TermsParams.TERMS_UPPER);
boolean upperIncl = params.getBool(TermsParams.TERMS_UPPER_INCLUSIVE, false);
boolean lowerIncl = params.getBool(TermsParams.TERMS_LOWER_INCLUSIVE, true);
boolean sort =
!TermsParams.TERMS_SORT_INDEX.equals(
params.get(TermsParams.TERMS_SORT, TermsParams.TERMS_SORT_COUNT));
int freqmin = params.getInt(TermsParams.TERMS_MINCOUNT, 1);
int freqmax = params.getInt(TermsParams.TERMS_MAXCOUNT, UNLIMITED_MAX_COUNT);
if (freqmax < 0) {
freqmax = Integer.MAX_VALUE;
}
String prefix = params.get(TermsParams.TERMS_PREFIX_STR);
String regexp = params.get(TermsParams.TERMS_REGEXP_STR);
Pattern pattern = regexp != null ? Pattern.compile(regexp, resolveRegexpFlags(params)) : null;
boolean raw = params.getBool(TermsParams.TERMS_RAW, false);
final AtomicReader indexReader = rb.req.getSearcher().getAtomicReader();
Fields lfields = indexReader.fields();
for (String field : fields) {
NamedList<Integer> fieldTerms = new NamedList<>();
termsResult.add(field, fieldTerms);
Terms terms = lfields == null ? null : lfields.terms(field);
if (terms == null) {
// no terms for this field
continue;
}
FieldType ft = raw ? null : rb.req.getSchema().getFieldTypeNoEx(field);
if (ft == null) ft = new StrField();
// prefix must currently be text
BytesRef prefixBytes = prefix == null ? null : new BytesRef(prefix);
BytesRef upperBytes = null;
if (upperStr != null) {
upperBytes = new BytesRef();
ft.readableToIndexed(upperStr, upperBytes);
}
BytesRef lowerBytes;
if (lowerStr == null) {
// If no lower bound was specified, use the prefix
lowerBytes = prefixBytes;
} else {
lowerBytes = new BytesRef();
if (raw) {
// TODO: how to handle binary? perhaps we don't for "raw"... or if the field exists
// perhaps we detect if the FieldType is non-character and expect hex if so?
lowerBytes = new BytesRef(lowerStr);
} else {
lowerBytes = new BytesRef();
ft.readableToIndexed(lowerStr, lowerBytes);
}
}
TermsEnum termsEnum = terms.iterator(null);
BytesRef term = null;
if (lowerBytes != null) {
if (termsEnum.seekCeil(lowerBytes) == TermsEnum.SeekStatus.END) {
termsEnum = null;
} else {
term = termsEnum.term();
// Only advance the enum if we are excluding the lower bound and the lower Term actually
// matches
if (lowerIncl == false && term.equals(lowerBytes)) {
term = termsEnum.next();
}
}
} else {
// position termsEnum on first term
term = termsEnum.next();
}
int i = 0;
BoundedTreeSet<CountPair<BytesRef, Integer>> queue =
(sort ? new BoundedTreeSet<CountPair<BytesRef, Integer>>(limit) : null);
CharsRef external = new CharsRef();
while (term != null && (i < limit || sort)) {
boolean externalized = false; // did we fill in "external" yet for this term?
// stop if the prefix doesn't match
if (prefixBytes != null && !StringHelper.startsWith(term, prefixBytes)) break;
if (pattern != null) {
// indexed text or external text?
// TODO: support "raw" mode?
ft.indexedToReadable(term, external);
externalized = true;
if (!pattern.matcher(external).matches()) {
term = termsEnum.next();
continue;
}
}
if (upperBytes != null) {
int upperCmp = term.compareTo(upperBytes);
// if we are past the upper term, or equal to it (when don't include upper) then stop.
if (upperCmp > 0 || (upperCmp == 0 && !upperIncl)) break;
}
// This is a good term in the range. Check if mincount/maxcount conditions are satisfied.
int docFreq = termsEnum.docFreq();
if (docFreq >= freqmin && docFreq <= freqmax) {
// add the term to the list
if (sort) {
queue.add(new CountPair<>(BytesRef.deepCopyOf(term), docFreq));
} else {
// TODO: handle raw somehow
if (!externalized) {
ft.indexedToReadable(term, external);
}
fieldTerms.add(external.toString(), docFreq);
i++;
}
}
term = termsEnum.next();
}
if (sort) {
for (CountPair<BytesRef, Integer> item : queue) {
if (i >= limit) break;
ft.indexedToReadable(item.key, external);
fieldTerms.add(external.toString(), item.val);
i++;
}
}
}
}
@Override
public void process(ResponseBuilder rb) throws IOException {
SolrQueryRequest req = rb.req;
SolrQueryResponse rsp = rb.rsp;
SolrParams params = req.getParams();
if (!params.getBool(COMPONENT_NAME, true)) {
return;
}
String val = params.get("getVersions");
if (val != null) {
processGetVersions(rb);
return;
}
val = params.get("getUpdates");
if (val != null) {
processGetUpdates(rb);
return;
}
String id[] = params.getParams("id");
String ids[] = params.getParams("ids");
if (id == null && ids == null) {
return;
}
String[] allIds = id == null ? new String[0] : id;
if (ids != null) {
List<String> lst = new ArrayList<String>();
for (String s : allIds) {
lst.add(s);
}
for (String idList : ids) {
lst.addAll(StrUtils.splitSmart(idList, ",", true));
}
allIds = lst.toArray(new String[lst.size()]);
}
SolrCore core = req.getCore();
SchemaField idField = core.getLatestSchema().getUniqueKeyField();
FieldType fieldType = idField.getType();
SolrDocumentList docList = new SolrDocumentList();
UpdateLog ulog = core.getUpdateHandler().getUpdateLog();
RefCounted<SolrIndexSearcher> searcherHolder = null;
DocTransformer transformer = rsp.getReturnFields().getTransformer();
if (transformer != null) {
TransformContext context = new TransformContext();
context.req = req;
transformer.setContext(context);
}
try {
SolrIndexSearcher searcher = null;
BytesRef idBytes = new BytesRef();
for (String idStr : allIds) {
fieldType.readableToIndexed(idStr, idBytes);
if (ulog != null) {
Object o = ulog.lookup(idBytes);
if (o != null) {
// should currently be a List<Oper,Ver,Doc/Id>
List entry = (List) o;
assert entry.size() >= 3;
int oper = (Integer) entry.get(0) & UpdateLog.OPERATION_MASK;
switch (oper) {
case UpdateLog.ADD:
SolrDocument doc =
toSolrDoc(
(SolrInputDocument) entry.get(entry.size() - 1), core.getLatestSchema());
if (transformer != null) {
transformer.transform(doc, -1); // unknown docID
}
docList.add(doc);
break;
case UpdateLog.DELETE:
break;
default:
throw new SolrException(
SolrException.ErrorCode.SERVER_ERROR, "Unknown Operation! " + oper);
}
continue;
}
}
// didn't find it in the update log, so it should be in the newest searcher opened
if (searcher == null) {
searcherHolder = core.getRealtimeSearcher();
searcher = searcherHolder.get();
}
// SolrCore.verbose("RealTimeGet using searcher ", searcher);
int docid = searcher.getFirstMatch(new Term(idField.getName(), idBytes));
if (docid < 0) continue;
StoredDocument luceneDocument = searcher.doc(docid);
SolrDocument doc = toSolrDoc(luceneDocument, core.getLatestSchema());
if (transformer != null) {
transformer.transform(doc, docid);
}
docList.add(doc);
}
} finally {
if (searcherHolder != null) {
searcherHolder.decref();
}
}
// if the client specified a single id=foo, then use "doc":{
// otherwise use a standard doclist
if (ids == null && allIds.length <= 1) {
// if the doc was not found, then use a value of null.
rsp.add("doc", docList.size() > 0 ? docList.get(0) : null);
} else {
docList.setNumFound(docList.size());
rsp.add("response", docList);
}
}