private ScoredCandidates<Container> scoreContainers(
Multiset<String> parents, int children, ResultDescription desc) {
Builder<Container> candidates = DefaultScoredCandidates.fromSource(NAME);
ResolvedContent containers = resolver.findByCanonicalUris(parents.elementSet());
for (Multiset.Entry<String> parent : parents.entrySet()) {
Maybe<Identified> possibledContainer = containers.get(parent.getElement());
if (possibledContainer.hasValue()) {
Identified identified = possibledContainer.requireValue();
if (identified instanceof Container) {
Container container = (Container) identified;
Score score = score(parent.getCount(), children);
candidates.addEquivalent(container, score);
desc.appendText(
"%s: scored %s (%s)", container.getCanonicalUri(), score, container.getTitle());
} else {
desc.appendText("%s: %s not container", parent, identified.getClass().getSimpleName());
}
} else {
desc.appendText("%s: missing", parent);
}
}
return candidates.build();
}
private void computeViolationsPerRules(DecoratorContext context) {
Map<RulePriority, Multiset<Rule>> rulesPerSeverity = Maps.newHashMap();
for (Violation violation : context.getViolations()) {
Multiset<Rule> rulesBag = initRules(rulesPerSeverity, violation.getSeverity());
rulesBag.add(violation.getRule());
}
for (RulePriority severity : RulePriority.values()) {
Metric metric = SeverityUtils.severityToViolationMetric(severity);
Collection<Measure> children = context.getChildrenMeasures(MeasuresFilters.rules(metric));
for (Measure child : children) {
RuleMeasure childRuleMeasure = (RuleMeasure) child;
Rule rule = childRuleMeasure.getRule();
if (rule != null && MeasureUtils.hasValue(childRuleMeasure)) {
Multiset<Rule> rulesBag = initRules(rulesPerSeverity, severity);
rulesBag.add(rule, childRuleMeasure.getIntValue());
}
}
Multiset<Rule> rulesBag = rulesPerSeverity.get(severity);
if (rulesBag != null) {
for (Multiset.Entry<Rule> entry : rulesBag.entrySet()) {
RuleMeasure measure =
RuleMeasure.createForRule(metric, entry.getElement(), (double) entry.getCount());
measure.setSeverity(severity);
context.saveMeasure(measure);
}
}
}
}
public static void main(String args[]) {
// create a multiset collection
Multiset<String> multiset = HashMultiset.create();
multiset.add("a");
multiset.add("b");
multiset.add("c");
multiset.add("d");
multiset.add("a");
multiset.add("b");
multiset.add("c");
multiset.add("b");
multiset.add("b");
multiset.add("b");
// print the occurrence of an element
System.out.println("Occurrence of 'b' : " + multiset.count("b"));
// print the total size of the multiset
System.out.println("Total Size : " + multiset.size());
// get the distinct elements of the multiset as set
Set<String> set = multiset.elementSet();
// display the elements of the set
System.out.println("Set [");
for (String s : set) {
System.out.println(s);
}
System.out.println("]");
// display all the elements of the multiset using iterator
Iterator<String> iterator = multiset.iterator();
System.out.println("MultiSet [");
while (iterator.hasNext()) {
System.out.println(iterator.next());
}
System.out.println("]");
// display the distinct elements of the multiset with their occurrence count
System.out.println("MultiSet [");
for (Multiset.Entry<String> entry : multiset.entrySet()) {
System.out.println("Element: " + entry.getElement() + ", Occurrence(s): " + entry.getCount());
}
System.out.println("]");
// remove extra occurrences
multiset.remove("b", 2);
// print the occurrence of an element
System.out.println("Occurence of 'b' : " + multiset.count("b"));
}
@Override
public void printResults() {
Multiset<String> storage = (Multiset<String>) this.storage;
System.out.println(category);
for (Multiset.Entry<String> entry : storage.entrySet()) {
System.out.printf("%s:%s\n", entry.getElement(), entry.getCount());
}
super.printResults();
}
public static void printResult(
Ordering<Multiset.Entry<Word>> comparator, int top, Multiset<Word> result) {
System.out.println("**********sort result**************");
System.out.println("\tword\t\tcount");
for (Multiset.Entry<Word> wordEntry : comparator.greatestOf(result.entrySet(), top)) {
System.out.printf("\t%s\t%6d\n", wordEntry.getElement().getWord(), wordEntry.getCount());
}
System.out.println("**********done...**************");
}
private static <K> String formatEntries(
Set<Multiset.Entry<K>> entries, Converter<K> keyConverter) {
StringBuilder sb = new StringBuilder();
boolean first = true;
for (Multiset.Entry<K> entry : entries) {
if (!first) {
sb.append(PAIR_SEPARATOR);
}
sb.append(keyConverter.format(entry.getElement()));
sb.append(FIELD_SEPARATOR);
sb.append(newIntegerConverter().format(entry.getCount()));
first = false;
}
return sb.toString();
}
@Override
public Set<Multiset.Entry<E>> create(Object... entries) {
List<Object> contents = new ArrayList<Object>();
Set<E> elements = new HashSet<E>();
for (Object o : entries) {
@SuppressWarnings("unchecked")
Multiset.Entry<E> entry = (Entry<E>) o;
checkArgument(
elements.add(entry.getElement()), "Duplicate keys not allowed in EntrySetGenerator");
for (int i = 0; i < entry.getCount(); i++) {
contents.add(entry.getElement());
}
}
return ((Multiset<E>) gen.create(contents.toArray())).entrySet();
}
/**
* @param orConditions StatisticsQueryOrConditions<StatisticsQueryCondition>
* @param statisticsStorage
* @param scoringExps Set of experiments that have at least one non-zero score for
* statisticsQuery. This is used retrieving efos to be displayed in heatmap when no query efvs
* exist (c.f. atlasStatisticsQueryService.getScoringAttributesForGenes())
* @return Multiset<Integer> containing experiment counts corresponding to all attributes in each
* StatisticsQueryCondition in orConditions
*/
private static Multiset<Integer> getScoresForOrConditions(
final StatisticsQueryOrConditions<StatisticsQueryCondition> orConditions,
StatisticsStorage statisticsStorage,
Set<ExperimentInfo> scoringExps) {
Multiset<Integer> scores = HashMultiset.create();
for (StatisticsQueryCondition orCondition : orConditions.getConditions()) {
orCondition.setBioEntityIdRestrictionSet(orConditions.getBioEntityIdRestrictionSet());
scores.addAll(scoreQuery(orCondition, statisticsStorage, scoringExps));
}
// Now apply orConditions' min experiments restriction to scores
Multiset<Integer> qualifyingScores = HashMultiset.create();
for (Multiset.Entry<Integer> entry : scores.entrySet()) {
if (entry.getCount() >= orConditions.getMinExperiments()) {
qualifyingScores.setCount(entry.getElement(), entry.getCount());
}
}
return qualifyingScores;
}
private void snapshotTimers(DataOutputView out) throws IOException {
out.writeInt(watermarkTimersQueue.size());
for (Timer<K, W> timer : watermarkTimersQueue) {
keySerializer.serialize(timer.key, out);
windowSerializer.serialize(timer.window, out);
out.writeLong(timer.timestamp);
}
out.writeInt(processingTimeTimers.size());
for (Timer<K, W> timer : processingTimeTimers) {
keySerializer.serialize(timer.key, out);
windowSerializer.serialize(timer.window, out);
out.writeLong(timer.timestamp);
}
out.writeInt(processingTimeTimerTimestamps.entrySet().size());
for (Multiset.Entry<Long> timerTimestampCounts : processingTimeTimerTimestamps.entrySet()) {
out.writeLong(timerTimestampCounts.getElement());
out.writeInt(timerTimestampCounts.getCount());
}
}
private void checkAttributeNamesForDuplicates(ValueType type, Protoclass protoclass) {
if (!type.attributes.isEmpty()) {
Multiset<String> attributeNames = HashMultiset.create(type.attributes.size());
for (ValueAttribute attribute : type.attributes) {
attributeNames.add(attribute.name());
}
List<String> duplicates = Lists.newArrayList();
for (Multiset.Entry<String> entry : attributeNames.entrySet()) {
if (entry.getCount() > 1) {
duplicates.add(entry.getElement());
}
}
if (!duplicates.isEmpty()) {
protoclass
.report()
.error(
"Duplicate attribute names %s. You should check if correct @Value.Style applied",
duplicates);
}
}
}
public void MagicProcessor() {
if (modelChanges == null) {
modelChanges = new ArrayList<ModelChange>();
return;
} else if (modelChanges.size() == 0) return;
// else go!
// gets all model deltas and processes!
// System.out.println("[MAGIC] Update match set based on model change started...");
long start = System.currentTimeMillis();
Set<PQuery> affecteds = new HashSet<PQuery>();
for (ModelChange change : modelChanges) {
if (change instanceof EFeatureChange)
affecteds.addAll(
LookaheadMatcherTreat.RelativeSet.get(((EFeatureChange) change).getChangedFeature()));
if (change instanceof EClassChange)
affecteds.addAll(
LookaheadMatcherTreat.RelativeSet.get(((EClassChange) change).getChange()));
if (change instanceof EDataTypeChange)
affecteds.addAll(
LookaheadMatcherTreat.RelativeSet.get(((EDataTypeChange) change).getChange()));
}
ArrayList<ModelDelta> deltas = new ArrayList<ModelDelta>();
for (PQuery affectedQuery : affecteds) {
ArrayList<AheadStructure> cachedStructures =
LookaheadMatcherTreat.GodSetStructures.get(affectedQuery);
// deliver deltas for pattern!
for (ModelChange change : modelChanges) {
for (AheadStructure aSn : cachedStructures) {
for (AxisConstraint rC : aSn.SearchedConstraints) {
if (rC instanceof RelationConstraint && change instanceof EFeatureChange) {
EFeatureChange changenow = (EFeatureChange) change;
if (((RelationConstraint) rC).getEdge().equals(changenow.getChangedFeature()))
rC.putToMailbox(change);
} else if (rC instanceof TypeConstraint && change instanceof EDataTypeChange) {
EDataTypeChange changenow = (EDataTypeChange) change;
if (((TypeConstraint) rC).getType().equals(changenow.getChange()))
rC.putToMailbox(change);
}
if (rC instanceof TypeConstraint && change instanceof EClassChange) {
EClassChange changenow = (EClassChange) change;
if (((TypeConstraint) rC).getType().equals(changenow.getChange()))
rC.putToMailbox(change);
}
}
}
}
for (ModelChange change : modelChanges) {
// process this change: first remove all deltas from constraints with this change
for (AheadStructure aSn : cachedStructures) {
for (AxisConstraint rC : aSn.SearchedConstraints) {
if (rC.hasMailboxContent()) {
if (rC.getMailboxContent().contains(change)) rC.removeFromMailbox(change);
}
}
}
// apply modelchange:
HashMap<PVariable, Object> knownLocalAndParameters = new HashMap<PVariable, Object>();
for (AheadStructure aSn : cachedStructures) {
// find all relationConstraints
for (AxisConstraint rC : aSn.SearchedConstraints) {
if (rC instanceof RelationConstraint && change instanceof EFeatureChange) {
EFeatureChange changenow = (EFeatureChange) change;
if (((RelationConstraint) rC).getEdge().equals(changenow.getChangedFeature())) {
// affected relaconstraint's lookvariables should be bound!!
knownLocalAndParameters.put(
((RelationConstraint) rC).getSource(), changenow.getHost());
knownLocalAndParameters.put(
((RelationConstraint) rC).getTarget(), changenow.getInstance());
}
} else if (rC instanceof TypeConstraint && change instanceof EDataTypeChange) {
EDataTypeChange changenow = (EDataTypeChange) change;
if (((TypeConstraint) rC).getType().equals(changenow.getChange())) {
// affected typeconstraint's lookvariable should be bound!!
knownLocalAndParameters.put(
((TypeConstraint) rC).getTypedVariable(), changenow.getInstance());
}
}
if (rC instanceof TypeConstraint && change instanceof EClassChange) {
EClassChange changenow = (EClassChange) change;
if (((TypeConstraint) rC).getType().equals(changenow.getChange())) {
// affected typeconstraint's lookvariable should be bound!!
knownLocalAndParameters.put(
((TypeConstraint) rC).getTypedVariable(), changenow.getInstance());
}
}
}
}
// manual satisfy and clone cachedStructures (createNew* clones input):
ArrayList<AheadStructure> newStructs = null;
isModified = false;
if (change instanceof EFeatureChange) {
EFeatureChange changenow = (EFeatureChange) change;
newStructs =
createNewFromOldRelaC(
changenow.getHost(),
changenow.getInstance(),
changenow.getChangedFeature(),
cachedStructures);
} else if (change instanceof EDataTypeChange) {
EDataTypeChange changenow = (EDataTypeChange) change;
newStructs =
createNewFromOldTypeC(
false, changenow.getChange(), changenow.getInstance(), cachedStructures);
}
if (change instanceof EClassChange) {
EClassChange changenow = (EClassChange) change;
newStructs =
createNewFromOldTypeC(
false, changenow.getChange(), changenow.getInstance(), cachedStructures);
}
if (isModified) {
// the new matches that'll appear in matching based on manually satisfied structure
Multiset<LookaheadMatching> newbies_toExamine =
(new LookaheadMatcherInterface(this.navHelper))
.searchChangesAll(
treat.getIncQueryEngine(),
affectedQuery,
newStructs,
knownLocalAndParameters,
new TreatConstraintEnumerator(this.navHelper));
// a new map to store a matching and whether it is added or removed
HashMultimap<LookaheadMatching, Boolean> newMatchingsAndChange = HashMultimap.create();
// iterate over multiset and create delta
for (com.google.common.collect.Multiset.Entry<LookaheadMatching> inners :
newbies_toExamine.entrySet()) {
for (int pi = 0; pi < inners.getCount(); pi++)
newMatchingsAndChange.put(inners.getElement(), change.isAddition());
}
// delta needed to propagate the changes
if (newMatchingsAndChange.size() > 0) {
ModelDelta d = new ModelDelta(affectedQuery, newMatchingsAndChange);
deltas.add(d);
}
}
}
}
// apply deltas
for (ModelDelta delta : deltas) {
// System.out.println("Propagate a delta: " + delta.getPattern().getFullyQualifiedName());
AdvancedDeltaProcessor.getInstance().ReceiveDelta(delta);
}
AdvancedDeltaProcessor.getInstance().ProcessReceivedDeltaSet();
// System.out.println("[MAGIC] Update match set based on model change ended! Time:" +
// Long.toString(System.currentTimeMillis() - start));
// finally:
modelChanges = new ArrayList<ModelChange>();
}
@Override
public void merge(final Iterator<DataFile> inFileIterator, DataFile outFile) throws IOException {
final Multiset<String> counts = HashMultiset.create();
final Set<String> emptyCounts = new HashSet<>();
while (inFileIterator.hasNext()) {
// Get input file
final DataFile inFile = inFileIterator.next();
EoulsanLogger.getLogger().info("Merge " + inFile.getName() + " to " + outFile.getName());
boolean first = true;
try (BufferedReader reader = new BufferedReader(new InputStreamReader(inFile.open()))) {
String line = null;
while ((line = reader.readLine()) != null) {
// Do no handle header
if (first) {
first = false;
continue;
}
final int tabPos = line.indexOf('\t');
// Do not handle empty or invalid lines
if (tabPos == -1) {
continue;
}
try {
final String id = line.substring(0, tabPos).trim();
final int count = Integer.parseInt(line.substring(tabPos).trim());
if (count == 0) {
emptyCounts.add(id);
}
counts.add(id, count);
} catch (NumberFormatException e) {
// Do not handle parsing errors
}
}
}
}
// Write the result file
try (Writer writer = new OutputStreamWriter(outFile.create())) {
writer.write(ExpressionSplitter.EXPRESSION_FILE_HEADER);
// Write the non empty counts
for (Multiset.Entry<String> e : counts.entrySet()) {
final String id = e.getElement();
// Remove the id from empty counts
emptyCounts.remove(id);
// Write the entry
writer.write(id + '\t' + e.getCount() + '\n');
}
// Write the empty counts
for (String id : emptyCounts) {
writer.write(id + "\t0\n");
}
}
}
public static void main(String[] args) throws Exception {
if (args.length < 5) {
System.out.println(
"Arguments: [model] [label index] [dictionnary] [document frequency] [Customer description]");
return;
}
String modelPath = args[0];
String labelIndexPath = args[1];
String dictionaryPath = args[2];
String documentFrequencyPath = args[3];
String carsPath = args[4];
Configuration configuration = new Configuration();
// model is a matrix (wordId, labelId) => probability score
NaiveBayesModel model = NaiveBayesModel.materialize(new Path(modelPath), configuration);
StandardNaiveBayesClassifier classifier = new StandardNaiveBayesClassifier(model);
// labels is a map label => classId
Map<Integer, String> labels =
BayesUtils.readLabelIndex(configuration, new Path(labelIndexPath));
Map<String, Integer> dictionary = readDictionnary(configuration, new Path(dictionaryPath));
Map<Integer, Long> documentFrequency =
readDocumentFrequency(configuration, new Path(documentFrequencyPath));
// analyzer used to extract word from tweet
Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_43);
int labelCount = labels.size();
int documentCount = documentFrequency.get(-1).intValue();
System.out.println("Number of labels: " + labelCount);
System.out.println("Number of documents in training set: " + documentCount);
BufferedReader reader = new BufferedReader(new FileReader(carsPath));
while (true) {
String line = reader.readLine();
if (line == null) {
break;
}
String[] tokens = line.split("\t", 47);
String cmplid = tokens[0];
String cdescr = tokens[19];
System.out.println("Complaint id: " + cmplid + "\t" + cdescr);
Multiset<String> words = ConcurrentHashMultiset.create();
// extract words from complaint description
TokenStream ts = analyzer.tokenStream("text", new StringReader(cdescr));
CharTermAttribute termAtt = ts.addAttribute(CharTermAttribute.class);
ts.reset();
int wordCount = 0;
while (ts.incrementToken()) {
if (termAtt.length() > 0) {
String word = ts.getAttribute(CharTermAttribute.class).toString();
Integer wordId = dictionary.get(word);
// if the word is not in the dictionary, skip it
if (wordId != null) {
words.add(word);
wordCount++;
}
}
}
// create vector wordId => weight using tfidf
Vector vector = new RandomAccessSparseVector(1000);
TFIDF tfidf = new TFIDF();
for (Multiset.Entry<String> entry : words.entrySet()) {
String word = entry.getElement();
int count = entry.getCount();
Integer wordId = dictionary.get(word);
Long freq = documentFrequency.get(wordId);
double tfIdfValue = tfidf.calculate(count, freq.intValue(), wordCount, documentCount);
vector.setQuick(wordId, tfIdfValue);
}
// With the classifier, we get one score for each label
// The label with the highest score is the one the tweet is more likely to
// be associated to
Vector resultVector = classifier.classifyFull(vector);
double bestScore = -Double.MAX_VALUE;
int bestCategoryId = -1;
for (Element element : resultVector.all()) {
int categoryId = element.index();
double score = element.get();
if (score > bestScore) {
bestScore = score;
bestCategoryId = categoryId;
}
System.out.print(" " + labels.get(categoryId) + ": " + score);
}
System.out.println(" => " + labels.get(bestCategoryId));
}
analyzer.close();
reader.close();
}
/** Transform this raw statement into a CreateTableStatement. */
public ParsedStatement.Prepared prepare() throws RequestValidationException {
// Column family name
if (!columnFamily().matches("\\w+"))
throw new InvalidRequestException(
String.format(
"\"%s\" is not a valid column family name (must be alphanumeric character only: [0-9A-Za-z]+)",
columnFamily()));
if (columnFamily().length() > Schema.NAME_LENGTH)
throw new InvalidRequestException(
String.format(
"Column family names shouldn't be more than %s characters long (got \"%s\")",
Schema.NAME_LENGTH, columnFamily()));
for (Multiset.Entry<ColumnIdentifier> entry : definedNames.entrySet())
if (entry.getCount() > 1)
throw new InvalidRequestException(
String.format("Multiple definition of identifier %s", entry.getElement()));
properties.validate();
CreateTableStatement stmt =
new CreateTableStatement(cfName, properties, ifNotExists, staticColumns);
Map<ByteBuffer, CollectionType> definedCollections = null;
for (Map.Entry<ColumnIdentifier, CQL3Type> entry : definitions.entrySet()) {
ColumnIdentifier id = entry.getKey();
CQL3Type pt = entry.getValue();
if (pt.isCollection()) {
if (definedCollections == null)
definedCollections = new HashMap<ByteBuffer, CollectionType>();
definedCollections.put(id.key, (CollectionType) pt.getType());
}
stmt.columns.put(id, pt.getType()); // we'll remove what is not a column below
}
if (keyAliases.isEmpty())
throw new InvalidRequestException("No PRIMARY KEY specifed (exactly one required)");
else if (keyAliases.size() > 1)
throw new InvalidRequestException("Multiple PRIMARY KEYs specifed (exactly one required)");
List<ColumnIdentifier> kAliases = keyAliases.get(0);
List<AbstractType<?>> keyTypes = new ArrayList<AbstractType<?>>(kAliases.size());
for (ColumnIdentifier alias : kAliases) {
stmt.keyAliases.add(alias.key);
AbstractType<?> t = getTypeAndRemove(stmt.columns, alias);
if (t instanceof CounterColumnType)
throw new InvalidRequestException(
String.format("counter type is not supported for PRIMARY KEY part %s", alias));
if (staticColumns.contains(alias))
throw new InvalidRequestException(
String.format("Static column %s cannot be part of the PRIMARY KEY", alias));
keyTypes.add(t);
}
stmt.keyValidator =
keyTypes.size() == 1 ? keyTypes.get(0) : CompositeType.getInstance(keyTypes);
// Dense means that no part of the comparator stores a CQL column name. This means
// COMPACT STORAGE with at least one columnAliases (otherwise it's a thrift "static" CF).
stmt.isDense = useCompactStorage && !columnAliases.isEmpty();
// Handle column aliases
if (columnAliases.isEmpty()) {
if (useCompactStorage) {
// There should remain some column definition since it is a non-composite "static" CF
if (stmt.columns.isEmpty())
throw new InvalidRequestException(
"No definition found that is not part of the PRIMARY KEY");
if (definedCollections != null)
throw new InvalidRequestException(
"Collection types are not supported with COMPACT STORAGE");
stmt.comparator = CFDefinition.definitionType;
} else {
List<AbstractType<?>> types =
new ArrayList<AbstractType<?>>(definedCollections == null ? 1 : 2);
types.add(CFDefinition.definitionType);
if (definedCollections != null)
types.add(ColumnToCollectionType.getInstance(definedCollections));
stmt.comparator = CompositeType.getInstance(types);
}
} else {
// If we use compact storage and have only one alias, it is a
// standard "dynamic" CF, otherwise it's a composite
if (useCompactStorage && columnAliases.size() == 1) {
if (definedCollections != null)
throw new InvalidRequestException(
"Collection types are not supported with COMPACT STORAGE");
ColumnIdentifier alias = columnAliases.get(0);
stmt.columnAliases.add(alias.key);
stmt.comparator = getTypeAndRemove(stmt.columns, alias);
if (stmt.comparator instanceof CounterColumnType)
throw new InvalidRequestException(
String.format("counter type is not supported for PRIMARY KEY part %s", alias));
if (staticColumns.contains(alias))
throw new InvalidRequestException(
String.format("Static column %s cannot be part of the PRIMARY KEY", alias));
} else {
List<AbstractType<?>> types = new ArrayList<AbstractType<?>>(columnAliases.size() + 1);
for (ColumnIdentifier t : columnAliases) {
stmt.columnAliases.add(t.key);
AbstractType<?> type = getTypeAndRemove(stmt.columns, t);
if (type instanceof CounterColumnType)
throw new InvalidRequestException(
String.format("counter type is not supported for PRIMARY KEY part %s", t));
if (staticColumns.contains(t))
throw new InvalidRequestException(
String.format("Static column %s cannot be part of the PRIMARY KEY", t));
types.add(type);
}
if (useCompactStorage) {
if (definedCollections != null)
throw new InvalidRequestException(
"Collection types are not supported with COMPACT STORAGE");
} else {
// For sparse, we must add the last UTF8 component
// and the collection type if there is one
types.add(CFDefinition.definitionType);
if (definedCollections != null)
types.add(ColumnToCollectionType.getInstance(definedCollections));
}
if (types.isEmpty())
throw new IllegalStateException("Nonsensical empty parameter list for CompositeType");
stmt.comparator = CompositeType.getInstance(types);
}
}
if (!staticColumns.isEmpty()) {
// Only CQL3 tables can have static columns
if (useCompactStorage)
throw new InvalidRequestException(
"Static columns are not supported in COMPACT STORAGE tables");
// Static columns only make sense if we have at least one clustering column. Otherwise
// everything is static anyway
if (columnAliases.isEmpty())
throw new InvalidRequestException(
"Static columns are only useful (and thus allowed) if the table has at least one clustering column");
}
if (useCompactStorage && !stmt.columnAliases.isEmpty()) {
if (stmt.columns.isEmpty()) {
// The only value we'll insert will be the empty one, so the default validator don't
// matter
stmt.defaultValidator = BytesType.instance;
// We need to distinguish between
// * I'm upgrading from thrift so the valueAlias is null
// * I've defined my table with only a PK (and the column value will be empty)
// So, we use an empty valueAlias (rather than null) for the second case
stmt.valueAlias = ByteBufferUtil.EMPTY_BYTE_BUFFER;
} else {
if (stmt.columns.size() > 1)
throw new InvalidRequestException(
String.format(
"COMPACT STORAGE with composite PRIMARY KEY allows no more than one column not part of the PRIMARY KEY (got: %s)",
StringUtils.join(stmt.columns.keySet(), ", ")));
Map.Entry<ColumnIdentifier, AbstractType> lastEntry =
stmt.columns.entrySet().iterator().next();
stmt.defaultValidator = lastEntry.getValue();
stmt.valueAlias = lastEntry.getKey().key;
stmt.columns.remove(lastEntry.getKey());
}
} else {
// For compact, we are in the "static" case, so we need at least one column defined. For
// non-compact however, having
// just the PK is fine since we have CQL3 row marker.
if (useCompactStorage && stmt.columns.isEmpty())
throw new InvalidRequestException(
"COMPACT STORAGE with non-composite PRIMARY KEY require one column not part of the PRIMARY KEY, none given");
// There is no way to insert/access a column that is not defined for non-compact storage, so
// the actual validator don't matter much (except that we want to recognize counter CF as
// limitation apply to them).
stmt.defaultValidator =
!stmt.columns.isEmpty()
&& (stmt.columns.values().iterator().next() instanceof CounterColumnType)
? CounterColumnType.instance
: BytesType.instance;
}
// If we give a clustering order, we must explicitly do so for all aliases and in the order of
// the PK
if (!definedOrdering.isEmpty()) {
if (definedOrdering.size() > columnAliases.size())
throw new InvalidRequestException(
"Only clustering key columns can be defined in CLUSTERING ORDER directive");
int i = 0;
for (ColumnIdentifier id : definedOrdering.keySet()) {
ColumnIdentifier c = columnAliases.get(i);
if (!id.equals(c)) {
if (definedOrdering.containsKey(c))
throw new InvalidRequestException(
String.format(
"The order of columns in the CLUSTERING ORDER directive must be the one of the clustering key (%s must appear before %s)",
c, id));
else
throw new InvalidRequestException(
String.format("Missing CLUSTERING ORDER for column %s", c));
}
++i;
}
}
return new ParsedStatement.Prepared(stmt);
}
/**
* The core scoring method for statistics queries
*
* @param statisticsQuery query to be peformed on statisticsStorage
* @param statisticsStorage core data for Statistics qeries
* @param scoringExps an out parameter.
* <p>- If null, experiment counts result of statisticsQuery should be returned. if - If
* non-null, it serves as a flag that an optimised statisticsQuery should be performed to just
* collect Experiments for which non-zero counts exist for Statistics query. A typical call
* scenario in this case is just one efv per statisticsQuery, in which we can both: 1. check
* if the efv Attribute itself is a scoring one 2. map this Attribute and Experimeants in
* scoringExps to efo terms - via the reverse mapping efv-experiment-> efo term in EfoIndex
* (c.f. atlasStatisticsQueryService.getScoringAttributesForGenes())
* @return Multiset of aggregated experiment counts, where the set of scores genes is intersected
* across statisticsQuery.getConditions(), and union-ed across attributes within each
* condition in statisticsQuery.getConditions().
*/
public static Multiset<Integer> scoreQuery(
StatisticsQueryCondition statisticsQuery,
final StatisticsStorage statisticsStorage,
Set<ExperimentInfo> scoringExps) {
// gatherScoringExpsOnly -> experiment counts should be calculated for statisticsQuery
// !gatherScoringExpsOnly -> scoring experiments should be collected (into scoringExps) only
boolean gatherScoringExpsOnly = scoringExps != null;
Set<StatisticsQueryOrConditions<StatisticsQueryCondition>> andStatisticsQueryConditions =
statisticsQuery.getConditions();
Multiset<Integer> results = null;
if (andStatisticsQueryConditions.isEmpty()) { // End of recursion
Set<Integer> bioEntityIdRestrictionSet = statisticsQuery.getBioEntityIdRestrictionSet();
Set<EfAttribute> attributes = statisticsQuery.getAttributes();
if (attributes.isEmpty()) {
// No attributes were provided - we have to use pre-computed scores across all attributes
Multiset<Integer> scoresAcrossAllEfos =
statisticsStorage.getScoresAcrossAllEfos(statisticsQuery.getStatisticsType());
results = intersect(scoresAcrossAllEfos, bioEntityIdRestrictionSet);
} else {
results = HashMultiset.create();
setQueryExperiments(statisticsQuery, statisticsStorage);
// For each experiment in the query, traverse through all attributes and add all gene
// indexes into one ConciseSet. This way a gene can score
// only once for a single experiment - across all OR attributes in this query. Once all
// attributes have been traversed for a single experiment,
// add ConciseSet to Multiset results
for (ExperimentInfo exp : statisticsQuery.getExperiments()) {
FastSet statsForExperiment = new FastSet();
for (EfAttribute attr : attributes) {
Map<ExperimentInfo, ConciseSet> expsToStats =
getStatisticsForAttribute(
statisticsQuery.getStatisticsType(), attr, statisticsStorage);
if (expsToStats != null) {
if (expsToStats.isEmpty()) {
log.debug(
"Failed to retrieve stats for stat: "
+ statisticsQuery.getStatisticsType()
+ " and attr: "
+ attr);
} else {
if (expsToStats.get(exp) != null) {
if (!gatherScoringExpsOnly) {
statsForExperiment.addAll(
intersect(expsToStats.get(exp), bioEntityIdRestrictionSet));
} else if (containsAtLeastOne(expsToStats.get(exp), bioEntityIdRestrictionSet)) {
// exp contains at least one non-zero score for at least one gene index in
// bioEntityIdRestrictionSet -> add it to scoringExps
scoringExps.add(exp);
}
} else {
log.debug(
"Failed to retrieve stats for stat: "
+ statisticsQuery.getStatisticsType()
+ " exp: "
+ exp.getAccession()
+ " and attr: "
+ attr);
}
}
}
}
if (!gatherScoringExpsOnly) {
results.addAll(statsForExperiment);
}
}
}
} else {
// run over all AND conditions, do "OR" inside (cf. scoreOrStatisticsQueryConditions()) ,
// "AND"'ing over the whole thing
for (StatisticsQueryOrConditions<StatisticsQueryCondition> orConditions :
andStatisticsQueryConditions) {
// Pass gene restriction set down to orConditions
orConditions.setGeneRestrictionSet(statisticsQuery.getBioEntityIdRestrictionSet());
// process OR conditions
Multiset<Integer> condGenes =
getScoresForOrConditions(orConditions, statisticsStorage, scoringExps);
if (results == null) results = condGenes;
else {
Iterator<Multiset.Entry<Integer>> resultGenes = results.entrySet().iterator();
while (resultGenes.hasNext()) {
Multiset.Entry<Integer> entry = resultGenes.next();
if (!condGenes.contains(
entry.getElement())) // AND operation between different top query conditions
resultGenes.remove();
else
// for all gene ids belonging to intersection of all conditions seen so far, we
// accumulate experiment counts
results.setCount(
entry.getElement(), entry.getCount() + condGenes.count(entry.getElement()));
}
}
}
}
if (results == null) {
results = HashMultiset.create();
}
return results;
}