public Fragmenter getFragmenter(String fieldName, SolrParams params) {
numRequests++;
if (defaults != null) {
params = new DefaultSolrParams(params, defaults);
}
int fragsize =
params.getFieldInt(
fieldName, HighlightParams.FRAGSIZE, LuceneRegexFragmenter.DEFAULT_FRAGMENT_SIZE);
int increment =
params.getFieldInt(
fieldName, HighlightParams.INCREMENT, LuceneRegexFragmenter.DEFAULT_INCREMENT_GAP);
float slop =
params.getFieldFloat(fieldName, HighlightParams.SLOP, LuceneRegexFragmenter.DEFAULT_SLOP);
int maxchars =
params.getFieldInt(
fieldName,
HighlightParams.MAX_RE_CHARS,
LuceneRegexFragmenter.DEFAULT_MAX_ANALYZED_CHARS);
String rawpat =
params.getFieldParam(
fieldName, HighlightParams.PATTERN, LuceneRegexFragmenter.DEFAULT_PATTERN_RAW);
Pattern p = rawpat == defaultPatternRaw ? defaultPattern : Pattern.compile(rawpat);
if (fragsize <= 0) {
return new NullFragmenter();
}
return new LuceneRegexFragmenter(fragsize, increment, slop, maxchars, p);
}
private void doHighlightingByFastVectorHighlighter(
FastVectorHighlighter highlighter,
FieldQuery fieldQuery,
SolrQueryRequest req,
NamedList docSummaries,
int docId,
Document doc,
String fieldName)
throws IOException {
SolrParams params = req.getParams();
SolrFragmentsBuilder solrFb = getSolrFragmentsBuilder(fieldName, params);
String[] snippets =
highlighter.getBestFragments(
fieldQuery,
req.getSearcher().getIndexReader(),
docId,
fieldName,
params.getFieldInt(fieldName, HighlightParams.FRAGSIZE, 100),
params.getFieldInt(fieldName, HighlightParams.SNIPPETS, 1),
getFragListBuilder(fieldName, params),
getFragmentsBuilder(fieldName, params),
solrFb.getPreTags(params, fieldName),
solrFb.getPostTags(params, fieldName),
getEncoder(fieldName, params));
if (snippets != null && snippets.length > 0) docSummaries.add(fieldName, snippets);
else alternateField(docSummaries, params, doc, fieldName);
}
private PivotFacetField(ResponseBuilder rb, PivotFacetValue parent, String fieldName) {
field = fieldName;
parentValue = parent;
// facet params
SolrParams parameters = rb.req.getParams();
facetFieldMinimumCount = parameters.getFieldInt(field, FacetParams.FACET_PIVOT_MINCOUNT, 1);
facetFieldOffset = parameters.getFieldInt(field, FacetParams.FACET_OFFSET, 0);
facetFieldLimit = parameters.getFieldInt(field, FacetParams.FACET_LIMIT, 100);
String defaultSort =
(facetFieldLimit > 0) ? FacetParams.FACET_SORT_COUNT : FacetParams.FACET_SORT_INDEX;
facetFieldSort = parameters.getFieldParam(field, FacetParams.FACET_SORT, defaultSort);
valueCollection =
new PivotFacetFieldValueCollection(
facetFieldMinimumCount, facetFieldOffset, facetFieldLimit, facetFieldSort);
if ((facetFieldLimit < 0)
||
// TODO: possible refinement issue if limit=0 & mincount=0 & missing=true
// (ie: we only want the missing count for this field)
(facetFieldLimit <= 0 && facetFieldMinimumCount == 0)
|| (facetFieldSort.equals(FacetParams.FACET_SORT_INDEX) && facetFieldMinimumCount <= 0)) {
// in any of these cases, there's no need to refine this level of the pivot
needRefinementAtThisLevel = false;
}
}
private void alternateField(
NamedList docSummaries, SolrParams params, Document doc, String fieldName) {
String alternateField = params.getFieldParam(fieldName, HighlightParams.ALTERNATE_FIELD);
if (alternateField != null && alternateField.length() > 0) {
String[] altTexts = doc.getValues(alternateField);
if (altTexts != null && altTexts.length > 0) {
int alternateFieldLen =
params.getFieldInt(fieldName, HighlightParams.ALTERNATE_FIELD_LENGTH, 0);
if (alternateFieldLen <= 0) {
docSummaries.add(fieldName, altTexts);
} else {
List<String> altList = new ArrayList<String>();
int len = 0;
for (String altText : altTexts) {
altList.add(
len + altText.length() > alternateFieldLen
? new String(altText.substring(0, alternateFieldLen - len))
: altText);
len += altText.length();
if (len >= alternateFieldLen) break;
}
docSummaries.add(fieldName, altList);
}
}
}
}
private void alternateField(
NamedList docSummaries, SolrParams params, Document doc, String fieldName) {
String alternateField = params.getFieldParam(fieldName, HighlightParams.ALTERNATE_FIELD);
if (alternateField != null && alternateField.length() > 0) {
IndexableField[] docFields = doc.getFields(alternateField);
List<String> listFields = new ArrayList<String>();
for (IndexableField field : docFields) {
if (field.binaryValue() == null) listFields.add(field.stringValue());
}
String[] altTexts = listFields.toArray(new String[listFields.size()]);
if (altTexts != null && altTexts.length > 0) {
Encoder encoder = getEncoder(fieldName, params);
int alternateFieldLen =
params.getFieldInt(fieldName, HighlightParams.ALTERNATE_FIELD_LENGTH, 0);
List<String> altList = new ArrayList<String>();
int len = 0;
for (String altText : altTexts) {
if (alternateFieldLen <= 0) {
altList.add(encoder.encodeText(altText));
} else {
altList.add(
len + altText.length() > alternateFieldLen
? encoder.encodeText(new String(altText.substring(0, alternateFieldLen - len)))
: encoder.encodeText(altText));
len += altText.length();
if (len >= alternateFieldLen) break;
}
}
docSummaries.add(fieldName, altList);
}
}
}
public Fragmenter getFragmenter(String fieldName, SolrParams params) {
numRequests++;
params = SolrParams.wrapDefaults(params, defaults);
int fragsize = params.getFieldInt(fieldName, HighlightParams.FRAGSIZE, 100);
return (fragsize <= 0) ? new NullFragmenter() : new LuceneGapFragmenter(fragsize);
}
private void doHighlightingByHighlighter(
Query query,
SolrQueryRequest req,
NamedList docSummaries,
int docId,
Document doc,
String fieldName)
throws IOException {
SolrParams params = req.getParams();
String[] docTexts = doc.getValues(fieldName);
// according to Document javadoc, doc.getValues() never returns null. check empty instead of
// null
if (docTexts.length == 0) return;
SolrIndexSearcher searcher = req.getSearcher();
IndexSchema schema = searcher.getSchema();
TokenStream tstream = null;
int numFragments = getMaxSnippets(fieldName, params);
boolean mergeContiguousFragments = isMergeContiguousFragments(fieldName, params);
String[] summaries = null;
List<TextFragment> frags = new ArrayList<TextFragment>();
TermOffsetsTokenStream tots = null; // to be non-null iff we're using TermOffsets optimization
try {
TokenStream tvStream =
TokenSources.getTokenStream(searcher.getIndexReader(), docId, fieldName);
if (tvStream != null) {
tots = new TermOffsetsTokenStream(tvStream);
}
} catch (IllegalArgumentException e) {
// No problem. But we can't use TermOffsets optimization.
}
for (int j = 0; j < docTexts.length; j++) {
if (tots != null) {
// if we're using TermOffsets optimization, then get the next
// field value's TokenStream (i.e. get field j's TokenStream) from tots:
tstream = tots.getMultiValuedTokenStream(docTexts[j].length());
} else {
// fall back to analyzer
tstream = createAnalyzerTStream(schema, fieldName, docTexts[j]);
}
Highlighter highlighter;
if (Boolean.valueOf(req.getParams().get(HighlightParams.USE_PHRASE_HIGHLIGHTER, "true"))) {
// TODO: this is not always necessary - eventually we would like to avoid this wrap
// when it is not needed.
tstream = new CachingTokenFilter(tstream);
// get highlighter
highlighter = getPhraseHighlighter(query, fieldName, req, (CachingTokenFilter) tstream);
// after highlighter initialization, reset tstream since construction of highlighter already
// used it
tstream.reset();
} else {
// use "the old way"
highlighter = getHighlighter(query, fieldName, req);
}
int maxCharsToAnalyze =
params.getFieldInt(
fieldName, HighlightParams.MAX_CHARS, Highlighter.DEFAULT_MAX_CHARS_TO_ANALYZE);
if (maxCharsToAnalyze < 0) {
highlighter.setMaxDocCharsToAnalyze(docTexts[j].length());
} else {
highlighter.setMaxDocCharsToAnalyze(maxCharsToAnalyze);
}
try {
TextFragment[] bestTextFragments =
highlighter.getBestTextFragments(
tstream, docTexts[j], mergeContiguousFragments, numFragments);
for (int k = 0; k < bestTextFragments.length; k++) {
if ((bestTextFragments[k] != null) && (bestTextFragments[k].getScore() > 0)) {
frags.add(bestTextFragments[k]);
}
}
} catch (InvalidTokenOffsetsException e) {
throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, e);
}
}
// sort such that the fragments with the highest score come first
Collections.sort(
frags,
new Comparator<TextFragment>() {
public int compare(TextFragment arg0, TextFragment arg1) {
return Math.round(arg1.getScore() - arg0.getScore());
}
});
// convert fragments back into text
// TODO: we can include score and position information in output as snippet attributes
if (frags.size() > 0) {
ArrayList<String> fragTexts = new ArrayList<String>();
for (TextFragment fragment : frags) {
if ((fragment != null) && (fragment.getScore() > 0)) {
fragTexts.add(fragment.toString());
}
if (fragTexts.size() >= numFragments) break;
}
summaries = fragTexts.toArray(new String[0]);
if (summaries.length > 0) docSummaries.add(fieldName, summaries);
}
// no summeries made, copy text from alternate field
if (summaries == null || summaries.length == 0) {
alternateField(docSummaries, params, doc, fieldName);
}
}
/**
* Return the max number of snippets for this field. If this has not been configured for this
* field, fall back to the configured default or the solr default.
*
* @param fieldName The name of the field
* @param params The params controlling Highlighting
*/
protected int getMaxSnippets(String fieldName, SolrParams params) {
return params.getFieldInt(fieldName, HighlightParams.SNIPPETS, 1);
}
/**
* Returns a list of terms in the specified field along with the corresponding count of documents
* in the set that match that constraint. This method uses the FilterCache to get the intersection
* count between <code>docs</code> and the DocSet for each term in the filter.
*
* @see FacetParams#FACET_LIMIT
* @see FacetParams#FACET_ZEROS
* @see FacetParams#FACET_MISSING
*/
public NamedList<Integer> getFacetTermEnumCounts(
SolrIndexSearcher searcher,
DocSet docs,
String field,
int offset,
int limit,
int mincount,
boolean missing,
String sort,
String prefix,
String contains,
boolean ignoreCase,
SolrParams params)
throws IOException {
/* :TODO: potential optimization...
* cache the Terms with the highest docFreq and try them first
* don't enum if we get our max from them
*/
// Minimum term docFreq in order to use the filterCache for that term.
int minDfFilterCache = global.getFieldInt(field, FacetParams.FACET_ENUM_CACHE_MINDF, 0);
// make sure we have a set that is fast for random access, if we will use it for that
DocSet fastForRandomSet = docs;
if (minDfFilterCache > 0 && docs instanceof SortedIntDocSet) {
SortedIntDocSet sset = (SortedIntDocSet) docs;
fastForRandomSet = new HashDocSet(sset.getDocs(), 0, sset.size());
}
IndexSchema schema = searcher.getSchema();
LeafReader r = searcher.getLeafReader();
FieldType ft = schema.getFieldType(field);
boolean sortByCount = sort.equals("count") || sort.equals("true");
final int maxsize = limit >= 0 ? offset + limit : Integer.MAX_VALUE - 1;
final BoundedTreeSet<CountPair<BytesRef, Integer>> queue =
sortByCount ? new BoundedTreeSet<CountPair<BytesRef, Integer>>(maxsize) : null;
final NamedList<Integer> res = new NamedList<>();
int min = mincount - 1; // the smallest value in the top 'N' values
int off = offset;
int lim = limit >= 0 ? limit : Integer.MAX_VALUE;
BytesRef prefixTermBytes = null;
if (prefix != null) {
String indexedPrefix = ft.toInternal(prefix);
prefixTermBytes = new BytesRef(indexedPrefix);
}
Fields fields = r.fields();
Terms terms = fields == null ? null : fields.terms(field);
TermsEnum termsEnum = null;
SolrIndexSearcher.DocsEnumState deState = null;
BytesRef term = null;
if (terms != null) {
termsEnum = terms.iterator();
// TODO: OPT: if seek(ord) is supported for this termsEnum, then we could use it for
// facet.offset when sorting by index order.
if (prefixTermBytes != null) {
if (termsEnum.seekCeil(prefixTermBytes) == TermsEnum.SeekStatus.END) {
termsEnum = null;
} else {
term = termsEnum.term();
}
} else {
// position termsEnum on first term
term = termsEnum.next();
}
}
PostingsEnum postingsEnum = null;
CharsRefBuilder charsRef = new CharsRefBuilder();
if (docs.size() >= mincount) {
while (term != null) {
if (prefixTermBytes != null && !StringHelper.startsWith(term, prefixTermBytes)) break;
if (contains == null || contains(term.utf8ToString(), contains, ignoreCase)) {
int df = termsEnum.docFreq();
// If we are sorting, we can use df>min (rather than >=) since we
// are going in index order. For certain term distributions this can
// make a large difference (for example, many terms with df=1).
if (df > 0 && df > min) {
int c;
if (df >= minDfFilterCache) {
// use the filter cache
if (deState == null) {
deState = new SolrIndexSearcher.DocsEnumState();
deState.fieldName = field;
deState.liveDocs = r.getLiveDocs();
deState.termsEnum = termsEnum;
deState.postingsEnum = postingsEnum;
}
c = searcher.numDocs(docs, deState);
postingsEnum = deState.postingsEnum;
} else {
// iterate over TermDocs to calculate the intersection
// TODO: specialize when base docset is a bitset or hash set (skipDocs)? or does it
// matter for this?
// TODO: do this per-segment for better efficiency (MultiDocsEnum just uses base class
// impl)
// TODO: would passing deleted docs lead to better efficiency over checking the
// fastForRandomSet?
postingsEnum = termsEnum.postings(postingsEnum, PostingsEnum.NONE);
c = 0;
if (postingsEnum instanceof MultiPostingsEnum) {
MultiPostingsEnum.EnumWithSlice[] subs =
((MultiPostingsEnum) postingsEnum).getSubs();
int numSubs = ((MultiPostingsEnum) postingsEnum).getNumSubs();
for (int subindex = 0; subindex < numSubs; subindex++) {
MultiPostingsEnum.EnumWithSlice sub = subs[subindex];
if (sub.postingsEnum == null) continue;
int base = sub.slice.start;
int docid;
while ((docid = sub.postingsEnum.nextDoc()) != DocIdSetIterator.NO_MORE_DOCS) {
if (fastForRandomSet.exists(docid + base)) c++;
}
}
} else {
int docid;
while ((docid = postingsEnum.nextDoc()) != DocIdSetIterator.NO_MORE_DOCS) {
if (fastForRandomSet.exists(docid)) c++;
}
}
}
if (sortByCount) {
if (c > min) {
BytesRef termCopy = BytesRef.deepCopyOf(term);
queue.add(new CountPair<>(termCopy, c));
if (queue.size() >= maxsize) min = queue.last().val;
}
} else {
if (c >= mincount && --off < 0) {
if (--lim < 0) break;
ft.indexedToReadable(term, charsRef);
res.add(charsRef.toString(), c);
}
}
}
}
term = termsEnum.next();
}
}
if (sortByCount) {
for (CountPair<BytesRef, Integer> p : queue) {
if (--off >= 0) continue;
if (--lim < 0) break;
ft.indexedToReadable(p.key, charsRef);
res.add(charsRef.toString(), p.val);
}
}
if (missing) {
res.add(null, getFieldMissingCount(searcher, docs, field));
}
return res;
}
/**
* Term counts for use in field faceting that resepcts the specified mincount - if mincount is
* null, the "zeros" param is consulted for the appropriate backcompat default
*
* @see FacetParams#FACET_ZEROS
*/
private NamedList<Integer> getTermCounts(String field, Integer mincount, ParsedParams parsed)
throws IOException {
final SolrParams params = parsed.params;
final DocSet docs = parsed.docs;
final int threads = parsed.threads;
int offset = params.getFieldInt(field, FacetParams.FACET_OFFSET, 0);
int limit = params.getFieldInt(field, FacetParams.FACET_LIMIT, 100);
if (limit == 0) return new NamedList<>();
if (mincount == null) {
Boolean zeros = params.getFieldBool(field, FacetParams.FACET_ZEROS);
// mincount = (zeros!=null && zeros) ? 0 : 1;
mincount = (zeros != null && !zeros) ? 1 : 0;
// current default is to include zeros.
}
boolean missing = params.getFieldBool(field, FacetParams.FACET_MISSING, false);
// default to sorting if there is a limit.
String sort =
params.getFieldParam(
field,
FacetParams.FACET_SORT,
limit > 0 ? FacetParams.FACET_SORT_COUNT : FacetParams.FACET_SORT_INDEX);
String prefix = params.getFieldParam(field, FacetParams.FACET_PREFIX);
String contains = params.getFieldParam(field, FacetParams.FACET_CONTAINS);
boolean ignoreCase = params.getFieldBool(field, FacetParams.FACET_CONTAINS_IGNORE_CASE, false);
NamedList<Integer> counts;
SchemaField sf = searcher.getSchema().getField(field);
FieldType ft = sf.getType();
// determine what type of faceting method to use
final String methodStr = params.getFieldParam(field, FacetParams.FACET_METHOD);
FacetMethod method = null;
if (FacetParams.FACET_METHOD_enum.equals(methodStr)) {
method = FacetMethod.ENUM;
} else if (FacetParams.FACET_METHOD_fcs.equals(methodStr)) {
method = FacetMethod.FCS;
} else if (FacetParams.FACET_METHOD_fc.equals(methodStr)) {
method = FacetMethod.FC;
}
if (method == FacetMethod.ENUM && TrieField.getMainValuePrefix(ft) != null) {
// enum can't deal with trie fields that index several terms per value
method = sf.multiValued() ? FacetMethod.FC : FacetMethod.FCS;
}
if (method == null && ft instanceof BoolField) {
// Always use filters for booleans... we know the number of values is very small.
method = FacetMethod.ENUM;
}
final boolean multiToken = sf.multiValued() || ft.multiValuedFieldCache();
if (ft.getNumericType() != null && !sf.multiValued()) {
// the per-segment approach is optimal for numeric field types since there
// are no global ords to merge and no need to create an expensive
// top-level reader
method = FacetMethod.FCS;
}
if (method == null) {
// TODO: default to per-segment or not?
method = FacetMethod.FC;
}
if (method == FacetMethod.FCS && multiToken) {
// only fc knows how to deal with multi-token fields
method = FacetMethod.FC;
}
if (method == FacetMethod.ENUM && sf.hasDocValues()) {
// only fc can handle docvalues types
method = FacetMethod.FC;
}
if (params.getFieldBool(field, GroupParams.GROUP_FACET, false)) {
counts =
getGroupedCounts(
searcher,
docs,
field,
multiToken,
offset,
limit,
mincount,
missing,
sort,
prefix,
contains,
ignoreCase);
} else {
assert method != null;
switch (method) {
case ENUM:
assert TrieField.getMainValuePrefix(ft) == null;
counts =
getFacetTermEnumCounts(
searcher,
docs,
field,
offset,
limit,
mincount,
missing,
sort,
prefix,
contains,
ignoreCase,
params);
break;
case FCS:
assert !multiToken;
if (ft.getNumericType() != null && !sf.multiValued()) {
// force numeric faceting
if (prefix != null && !prefix.isEmpty()) {
throw new SolrException(
ErrorCode.BAD_REQUEST,
FacetParams.FACET_PREFIX + " is not supported on numeric types");
}
if (contains != null && !contains.isEmpty()) {
throw new SolrException(
ErrorCode.BAD_REQUEST,
FacetParams.FACET_CONTAINS + " is not supported on numeric types");
}
counts =
NumericFacets.getCounts(
searcher, docs, field, offset, limit, mincount, missing, sort);
} else {
PerSegmentSingleValuedFaceting ps =
new PerSegmentSingleValuedFaceting(
searcher,
docs,
field,
offset,
limit,
mincount,
missing,
sort,
prefix,
contains,
ignoreCase);
Executor executor = threads == 0 ? directExecutor : facetExecutor;
ps.setNumThreads(threads);
counts = ps.getFacetCounts(executor);
}
break;
case FC:
counts =
DocValuesFacets.getCounts(
searcher,
docs,
field,
offset,
limit,
mincount,
missing,
sort,
prefix,
contains,
ignoreCase);
break;
default:
throw new AssertionError();
}
}
return counts;
}
private void doHighlightingByHighlighter(
Query query,
SolrQueryRequest req,
NamedList docSummaries,
int docId,
Document doc,
String fieldName)
throws IOException {
final SolrIndexSearcher searcher = req.getSearcher();
final IndexSchema schema = searcher.getSchema();
// TODO: Currently in trunk highlighting numeric fields is broken (Lucene) -
// so we disable them until fixed (see LUCENE-3080)!
// BEGIN: Hack
final SchemaField schemaField = schema.getFieldOrNull(fieldName);
if (schemaField != null
&& ((schemaField.getType() instanceof org.apache.solr.schema.TrieField)
|| (schemaField.getType() instanceof org.apache.solr.schema.TrieDateField))) return;
// END: Hack
SolrParams params = req.getParams();
IndexableField[] docFields = doc.getFields(fieldName);
List<String> listFields = new ArrayList<String>();
for (IndexableField field : docFields) {
listFields.add(field.stringValue());
}
// preserve order of values in a multiValued list
boolean preserveMulti = params.getFieldBool(fieldName, HighlightParams.PRESERVE_MULTI, false);
String[] docTexts = (String[]) listFields.toArray(new String[listFields.size()]);
// according to Document javadoc, doc.getValues() never returns null. check empty instead of
// null
if (docTexts.length == 0) return;
TokenStream tstream = null;
int numFragments = getMaxSnippets(fieldName, params);
boolean mergeContiguousFragments = isMergeContiguousFragments(fieldName, params);
String[] summaries = null;
List<TextFragment> frags = new ArrayList<TextFragment>();
TermOffsetsTokenStream tots = null; // to be non-null iff we're using TermOffsets optimization
TokenStream tvStream =
TokenSources.getTokenStreamWithOffsets(searcher.getIndexReader(), docId, fieldName);
if (tvStream != null) {
tots = new TermOffsetsTokenStream(tvStream);
}
for (int j = 0; j < docTexts.length; j++) {
if (tots != null) {
// if we're using TermOffsets optimization, then get the next
// field value's TokenStream (i.e. get field j's TokenStream) from tots:
tstream = tots.getMultiValuedTokenStream(docTexts[j].length());
} else {
// fall back to analyzer
tstream = createAnalyzerTStream(schema, fieldName, docTexts[j]);
}
int maxCharsToAnalyze =
params.getFieldInt(
fieldName, HighlightParams.MAX_CHARS, Highlighter.DEFAULT_MAX_CHARS_TO_ANALYZE);
Highlighter highlighter;
if (Boolean.valueOf(req.getParams().get(HighlightParams.USE_PHRASE_HIGHLIGHTER, "true"))) {
if (maxCharsToAnalyze < 0) {
tstream = new CachingTokenFilter(tstream);
} else {
tstream = new CachingTokenFilter(new OffsetLimitTokenFilter(tstream, maxCharsToAnalyze));
}
// get highlighter
highlighter = getPhraseHighlighter(query, fieldName, req, (CachingTokenFilter) tstream);
// after highlighter initialization, reset tstream since construction of highlighter already
// used it
tstream.reset();
} else {
// use "the old way"
highlighter = getHighlighter(query, fieldName, req);
}
if (maxCharsToAnalyze < 0) {
highlighter.setMaxDocCharsToAnalyze(docTexts[j].length());
} else {
highlighter.setMaxDocCharsToAnalyze(maxCharsToAnalyze);
}
try {
TextFragment[] bestTextFragments =
highlighter.getBestTextFragments(
tstream, docTexts[j], mergeContiguousFragments, numFragments);
for (int k = 0; k < bestTextFragments.length; k++) {
if (preserveMulti) {
if (bestTextFragments[k] != null) {
frags.add(bestTextFragments[k]);
}
} else {
if ((bestTextFragments[k] != null) && (bestTextFragments[k].getScore() > 0)) {
frags.add(bestTextFragments[k]);
}
}
}
} catch (InvalidTokenOffsetsException e) {
throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, e);
}
}
// sort such that the fragments with the highest score come first
if (!preserveMulti) {
Collections.sort(
frags,
new Comparator<TextFragment>() {
@Override
public int compare(TextFragment arg0, TextFragment arg1) {
return Math.round(arg1.getScore() - arg0.getScore());
}
});
}
// convert fragments back into text
// TODO: we can include score and position information in output as snippet attributes
if (frags.size() > 0) {
ArrayList<String> fragTexts = new ArrayList<String>();
for (TextFragment fragment : frags) {
if (preserveMulti) {
if (fragment != null) {
fragTexts.add(fragment.toString());
}
} else {
if ((fragment != null) && (fragment.getScore() > 0)) {
fragTexts.add(fragment.toString());
}
}
if (fragTexts.size() >= numFragments && !preserveMulti) break;
}
summaries = fragTexts.toArray(new String[0]);
if (summaries.length > 0) docSummaries.add(fieldName, summaries);
}
// no summeries made, copy text from alternate field
if (summaries == null || summaries.length == 0) {
alternateField(docSummaries, params, doc, fieldName);
}
}
/** Called by {@link org.apache.solr.request.SimpleFacets} to compute heatmap facets. */
public static NamedList<Object> getHeatmapForField(
String fieldKey, String fieldName, ResponseBuilder rb, SolrParams params, DocSet docSet)
throws IOException {
// get the strategy from the field type
final SchemaField schemaField = rb.req.getSchema().getField(fieldName);
final FieldType type = schemaField.getType();
final PrefixTreeStrategy strategy;
final DistanceUnits distanceUnits;
// note: the two instanceof conditions is not ideal, versus one. If we start needing to add more
// then refactor.
if ((type instanceof AbstractSpatialPrefixTreeFieldType)) {
AbstractSpatialPrefixTreeFieldType rptType = (AbstractSpatialPrefixTreeFieldType) type;
strategy = (PrefixTreeStrategy) rptType.getStrategy(fieldName);
distanceUnits = rptType.getDistanceUnits();
} else if (type instanceof RptWithGeometrySpatialField) {
RptWithGeometrySpatialField rptSdvType = (RptWithGeometrySpatialField) type;
strategy = rptSdvType.getStrategy(fieldName).getIndexStrategy();
distanceUnits = rptSdvType.getDistanceUnits();
} else {
// FYI we support the term query one too but few people use that one
throw new SolrException(
SolrException.ErrorCode.BAD_REQUEST,
"heatmap field needs to be of type "
+ SpatialRecursivePrefixTreeFieldType.class
+ " or "
+ RptWithGeometrySpatialField.class);
}
final SpatialContext ctx = strategy.getSpatialContext();
// get the bbox (query Rectangle)
String geomStr = params.getFieldParam(fieldKey, FacetParams.FACET_HEATMAP_GEOM);
final Shape boundsShape =
geomStr == null ? ctx.getWorldBounds() : SpatialUtils.parseGeomSolrException(geomStr, ctx);
// get the grid level (possibly indirectly via distErr or distErrPct)
final int gridLevel;
Integer gridLevelObj = params.getFieldInt(fieldKey, FacetParams.FACET_HEATMAP_LEVEL);
final int maxGridLevel = strategy.getGrid().getMaxLevels();
if (gridLevelObj != null) {
gridLevel = gridLevelObj;
if (gridLevel <= 0 || gridLevel > maxGridLevel) {
throw new SolrException(
SolrException.ErrorCode.BAD_REQUEST,
FacetParams.FACET_HEATMAP_LEVEL + " should be > 0 and <= " + maxGridLevel);
}
} else {
// SpatialArgs has utility methods to resolve a 'distErr' from optionally set distErr &
// distErrPct. Arguably that
// should be refactored to feel less weird than using it like this.
SpatialArgs spatialArgs =
new SpatialArgs(
SpatialOperation.Intersects /*ignored*/,
boundsShape == null ? ctx.getWorldBounds() : boundsShape);
final Double distErrObj = params.getFieldDouble(fieldKey, FacetParams.FACET_HEATMAP_DIST_ERR);
if (distErrObj != null) {
// convert distErr units based on configured units
spatialArgs.setDistErr(distErrObj * distanceUnits.multiplierFromThisUnitToDegrees());
}
spatialArgs.setDistErrPct(
params.getFieldDouble(fieldKey, FacetParams.FACET_HEATMAP_DIST_ERR_PCT));
double distErr = spatialArgs.resolveDistErr(ctx, DEFAULT_DIST_ERR_PCT);
if (distErr <= 0) {
throw new SolrException(
SolrException.ErrorCode.BAD_REQUEST,
FacetParams.FACET_HEATMAP_DIST_ERR_PCT
+ " or "
+ FacetParams.FACET_HEATMAP_DIST_ERR
+ " should be > 0 or instead provide "
+ FacetParams.FACET_HEATMAP_LEVEL
+ "="
+ maxGridLevel
+ " if you insist on maximum detail");
}
// The SPT (grid) can lookup a grid level satisfying an error distance constraint
gridLevel = strategy.getGrid().getLevelForDistance(distErr);
}
// Compute!
final HeatmapFacetCounter.Heatmap heatmap;
try {
heatmap =
HeatmapFacetCounter.calcFacets(
strategy,
rb.req.getSearcher().getTopReaderContext(),
docSet.getTopFilter(),
boundsShape,
gridLevel,
params.getFieldInt(
fieldKey, FacetParams.FACET_HEATMAP_MAX_CELLS, 100_000) // will throw if exceeded
);
} catch (IllegalArgumentException e) { // e.g. too many cells
throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e.toString(), e);
}
// Populate response
NamedList<Object> result = new NamedList<>();
result.add("gridLevel", gridLevel);
result.add("columns", heatmap.columns);
result.add("rows", heatmap.rows);
result.add("minX", heatmap.region.getMinX());
result.add("maxX", heatmap.region.getMaxX());
result.add("minY", heatmap.region.getMinY());
result.add("maxY", heatmap.region.getMaxY());
boolean hasNonZero = false;
for (int count : heatmap.counts) {
if (count > 0) {
hasNonZero = true;
break;
}
}
formatCountsAndAddToNL(
fieldKey,
rb,
params,
heatmap.columns,
heatmap.rows,
hasNonZero ? heatmap.counts : null,
result);
return result;
}
