use of io.druid.segment.Cursor in project druid by druid-io.
the class SelectQueryEngine method process.
public Sequence<Result<SelectResultValue>> process(final SelectQuery query, final Segment segment) {
final StorageAdapter adapter = segment.asStorageAdapter();
if (adapter == null) {
throw new ISE("Null storage adapter found. Probably trying to issue a query against a segment being memory unmapped.");
}
// at the point where this code is called, only one datasource should exist.
String dataSource = Iterables.getOnlyElement(query.getDataSource().getNames());
final Iterable<DimensionSpec> dims;
if (query.getDimensions() == null || query.getDimensions().isEmpty()) {
dims = DefaultDimensionSpec.toSpec(adapter.getAvailableDimensions());
} else {
dims = query.getDimensions();
}
final Iterable<String> metrics;
if (query.getMetrics() == null || query.getMetrics().isEmpty()) {
metrics = adapter.getAvailableMetrics();
} else {
metrics = query.getMetrics();
}
List<Interval> intervals = query.getQuerySegmentSpec().getIntervals();
Preconditions.checkArgument(intervals.size() == 1, "Can only handle a single interval, got[%s]", intervals);
// should be rewritten with given interval
final String segmentId = DataSegmentUtils.withInterval(dataSource, segment.getIdentifier(), intervals.get(0));
final Filter filter = Filters.convertToCNFFromQueryContext(query, Filters.toFilter(query.getDimensionsFilter()));
return QueryRunnerHelper.makeCursorBasedQuery(adapter, query.getQuerySegmentSpec().getIntervals(), filter, query.getVirtualColumns(), query.isDescending(), query.getGranularity(), new Function<Cursor, Result<SelectResultValue>>() {
@Override
public Result<SelectResultValue> apply(Cursor cursor) {
final SelectResultValueBuilder builder = new SelectResultValueBuilder(cursor.getTime(), query.getPagingSpec(), query.isDescending());
final LongColumnSelector timestampColumnSelector = cursor.makeLongColumnSelector(Column.TIME_COLUMN_NAME);
final List<ColumnSelectorPlus<SelectColumnSelectorStrategy>> selectorPlusList = Arrays.asList(DimensionHandlerUtils.createColumnSelectorPluses(STRATEGY_FACTORY, Lists.newArrayList(dims), cursor));
for (DimensionSpec dimSpec : dims) {
builder.addDimension(dimSpec.getOutputName());
}
final Map<String, ObjectColumnSelector> metSelectors = Maps.newHashMap();
for (String metric : metrics) {
final ObjectColumnSelector metricSelector = cursor.makeObjectColumnSelector(metric);
metSelectors.put(metric, metricSelector);
builder.addMetric(metric);
}
final PagingOffset offset = query.getPagingOffset(segmentId);
cursor.advanceTo(offset.startDelta());
int lastOffset = offset.startOffset();
for (; !cursor.isDone() && offset.hasNext(); cursor.advance(), offset.next()) {
final Map<String, Object> theEvent = singleEvent(EventHolder.timestampKey, timestampColumnSelector, selectorPlusList, metSelectors);
builder.addEntry(new EventHolder(segmentId, lastOffset = offset.current(), theEvent));
}
builder.finished(segmentId, lastOffset);
return builder.build();
}
});
}
use of io.druid.segment.Cursor in project druid by druid-io.
the class DimExtractionTopNAlgorithm method scanAndAggregate.
@Override
public void scanAndAggregate(TopNParams params, Aggregator[][] rowSelector, Map<Comparable, Aggregator[]> aggregatesStore, int numProcessed) {
final Cursor cursor = params.getCursor();
final ColumnSelectorPlus<TopNColumnSelectorStrategy> selectorPlus = params.getSelectorPlus();
selectorPlus.getColumnSelectorStrategy().dimExtractionScanAndAggregate(query, selectorPlus.getSelector(), cursor, rowSelector, aggregatesStore);
}
use of io.druid.segment.Cursor in project druid by druid-io.
the class IncrementalIndexStorageAdapter method makeCursors.
@Override
public Sequence<Cursor> makeCursors(final Filter filter, final Interval interval, final VirtualColumns virtualColumns, final Granularity gran, final boolean descending) {
if (index.isEmpty()) {
return Sequences.empty();
}
Interval actualIntervalTmp = interval;
final Interval dataInterval = new Interval(getMinTime().getMillis(), gran.bucketEnd(getMaxTime()).getMillis());
if (!actualIntervalTmp.overlaps(dataInterval)) {
return Sequences.empty();
}
if (actualIntervalTmp.getStart().isBefore(dataInterval.getStart())) {
actualIntervalTmp = actualIntervalTmp.withStart(dataInterval.getStart());
}
if (actualIntervalTmp.getEnd().isAfter(dataInterval.getEnd())) {
actualIntervalTmp = actualIntervalTmp.withEnd(dataInterval.getEnd());
}
final Interval actualInterval = actualIntervalTmp;
Iterable<Interval> iterable = gran.getIterable(actualInterval);
if (descending) {
iterable = Lists.reverse(ImmutableList.copyOf(iterable));
}
return Sequences.map(Sequences.simple(iterable), new Function<Interval, Cursor>() {
EntryHolder currEntry = new EntryHolder();
@Override
public Cursor apply(@Nullable final Interval interval) {
final long timeStart = Math.max(interval.getStartMillis(), actualInterval.getStartMillis());
return new Cursor() {
private final ValueMatcher filterMatcher = makeFilterMatcher(filter, this);
private Iterator<Map.Entry<IncrementalIndex.TimeAndDims, Integer>> baseIter;
private Iterable<Map.Entry<IncrementalIndex.TimeAndDims, Integer>> cursorIterable;
private boolean emptyRange;
final DateTime time;
int numAdvanced = -1;
boolean done;
{
cursorIterable = index.getFacts().timeRangeIterable(descending, timeStart, Math.min(actualInterval.getEndMillis(), gran.increment(interval.getStart()).getMillis()));
emptyRange = !cursorIterable.iterator().hasNext();
time = gran.toDateTime(interval.getStartMillis());
reset();
}
@Override
public DateTime getTime() {
return time;
}
@Override
public void advance() {
if (!baseIter.hasNext()) {
done = true;
return;
}
while (baseIter.hasNext()) {
BaseQuery.checkInterrupted();
currEntry.set(baseIter.next());
if (filterMatcher.matches()) {
return;
}
}
if (!filterMatcher.matches()) {
done = true;
}
}
@Override
public void advanceUninterruptibly() {
if (!baseIter.hasNext()) {
done = true;
return;
}
while (baseIter.hasNext()) {
if (Thread.currentThread().isInterrupted()) {
return;
}
currEntry.set(baseIter.next());
if (filterMatcher.matches()) {
return;
}
}
if (!filterMatcher.matches()) {
done = true;
}
}
@Override
public void advanceTo(int offset) {
int count = 0;
while (count < offset && !isDone()) {
advance();
count++;
}
}
@Override
public boolean isDone() {
return done;
}
@Override
public boolean isDoneOrInterrupted() {
return isDone() || Thread.currentThread().isInterrupted();
}
@Override
public void reset() {
baseIter = cursorIterable.iterator();
if (numAdvanced == -1) {
numAdvanced = 0;
} else {
Iterators.advance(baseIter, numAdvanced);
}
BaseQuery.checkInterrupted();
boolean foundMatched = false;
while (baseIter.hasNext()) {
currEntry.set(baseIter.next());
if (filterMatcher.matches()) {
foundMatched = true;
break;
}
numAdvanced++;
}
done = !foundMatched && (emptyRange || !baseIter.hasNext());
}
@Override
public DimensionSelector makeDimensionSelector(DimensionSpec dimensionSpec) {
if (virtualColumns.exists(dimensionSpec.getDimension())) {
return virtualColumns.makeDimensionSelector(dimensionSpec, this);
}
return dimensionSpec.decorate(makeDimensionSelectorUndecorated(dimensionSpec));
}
private DimensionSelector makeDimensionSelectorUndecorated(DimensionSpec dimensionSpec) {
final String dimension = dimensionSpec.getDimension();
final ExtractionFn extractionFn = dimensionSpec.getExtractionFn();
if (dimension.equals(Column.TIME_COLUMN_NAME)) {
DimensionSelector selector = new SingleScanTimeDimSelector(makeLongColumnSelector(dimension), extractionFn, descending);
return selector;
}
final IncrementalIndex.DimensionDesc dimensionDesc = index.getDimension(dimensionSpec.getDimension());
if (dimensionDesc == null) {
// not a dimension, column may be a metric
ColumnCapabilities capabilities = getColumnCapabilities(dimension);
if (capabilities == null) {
return NullDimensionSelector.instance();
}
if (capabilities.getType() == ValueType.LONG) {
return new LongWrappingDimensionSelector(makeLongColumnSelector(dimension), extractionFn);
}
if (capabilities.getType() == ValueType.FLOAT) {
return new FloatWrappingDimensionSelector(makeFloatColumnSelector(dimension), extractionFn);
}
// if we can't wrap the base column, just return a column of all nulls
return NullDimensionSelector.instance();
} else {
final DimensionIndexer indexer = dimensionDesc.getIndexer();
return indexer.makeDimensionSelector(dimensionSpec, currEntry, dimensionDesc);
}
}
@Override
public FloatColumnSelector makeFloatColumnSelector(String columnName) {
if (virtualColumns.exists(columnName)) {
return virtualColumns.makeFloatColumnSelector(columnName, this);
}
final Integer dimIndex = index.getDimensionIndex(columnName);
if (dimIndex != null) {
final IncrementalIndex.DimensionDesc dimensionDesc = index.getDimension(columnName);
final DimensionIndexer indexer = dimensionDesc.getIndexer();
return indexer.makeFloatColumnSelector(currEntry, dimensionDesc);
}
final Integer metricIndexInt = index.getMetricIndex(columnName);
if (metricIndexInt == null) {
return ZeroFloatColumnSelector.instance();
}
final int metricIndex = metricIndexInt;
return new FloatColumnSelector() {
@Override
public float get() {
return index.getMetricFloatValue(currEntry.getValue(), metricIndex);
}
@Override
public void inspectRuntimeShape(RuntimeShapeInspector inspector) {
inspector.visit("index", index);
}
};
}
@Override
public LongColumnSelector makeLongColumnSelector(String columnName) {
if (virtualColumns.exists(columnName)) {
return virtualColumns.makeLongColumnSelector(columnName, this);
}
if (columnName.equals(Column.TIME_COLUMN_NAME)) {
class TimeLongColumnSelector implements LongColumnSelector {
@Override
public long get() {
return currEntry.getKey().getTimestamp();
}
@Override
public void inspectRuntimeShape(RuntimeShapeInspector inspector) {
}
}
return new TimeLongColumnSelector();
}
final Integer dimIndex = index.getDimensionIndex(columnName);
if (dimIndex != null) {
final IncrementalIndex.DimensionDesc dimensionDesc = index.getDimension(columnName);
final DimensionIndexer indexer = dimensionDesc.getIndexer();
return indexer.makeLongColumnSelector(currEntry, dimensionDesc);
}
final Integer metricIndexInt = index.getMetricIndex(columnName);
if (metricIndexInt == null) {
return ZeroLongColumnSelector.instance();
}
final int metricIndex = metricIndexInt;
return new LongColumnSelector() {
@Override
public long get() {
return index.getMetricLongValue(currEntry.getValue(), metricIndex);
}
@Override
public void inspectRuntimeShape(RuntimeShapeInspector inspector) {
inspector.visit("index", index);
}
};
}
@Override
public ObjectColumnSelector makeObjectColumnSelector(String column) {
if (virtualColumns.exists(column)) {
return virtualColumns.makeObjectColumnSelector(column, this);
}
if (column.equals(Column.TIME_COLUMN_NAME)) {
return new ObjectColumnSelector<Long>() {
@Override
public Class classOfObject() {
return Long.TYPE;
}
@Override
public Long get() {
return currEntry.getKey().getTimestamp();
}
};
}
final Integer metricIndexInt = index.getMetricIndex(column);
if (metricIndexInt != null) {
final int metricIndex = metricIndexInt;
final Class classOfObject = index.getMetricClass(column);
return new ObjectColumnSelector() {
@Override
public Class classOfObject() {
return classOfObject;
}
@Override
public Object get() {
return index.getMetricObjectValue(currEntry.getValue(), metricIndex);
}
};
}
IncrementalIndex.DimensionDesc dimensionDesc = index.getDimension(column);
if (dimensionDesc == null) {
return null;
} else {
final int dimensionIndex = dimensionDesc.getIndex();
final DimensionIndexer indexer = dimensionDesc.getIndexer();
return new ObjectColumnSelector<Object>() {
@Override
public Class classOfObject() {
return Object.class;
}
@Override
public Object get() {
IncrementalIndex.TimeAndDims key = currEntry.getKey();
if (key == null) {
return null;
}
Object[] dims = key.getDims();
if (dimensionIndex >= dims.length) {
return null;
}
return indexer.convertUnsortedEncodedKeyComponentToActualArrayOrList(dims[dimensionIndex], DimensionIndexer.ARRAY);
}
};
}
}
@Nullable
@Override
public ColumnCapabilities getColumnCapabilities(String columnName) {
if (virtualColumns.exists(columnName)) {
return virtualColumns.getColumnCapabilities(columnName);
}
return index.getCapabilities(columnName);
}
};
}
});
}
use of io.druid.segment.Cursor in project druid by druid-io.
the class DumpSegment method runDump.
private void runDump(final Injector injector, final QueryableIndex index) throws IOException {
final ObjectMapper objectMapper = injector.getInstance(Key.get(ObjectMapper.class, Json.class));
final QueryableIndexStorageAdapter adapter = new QueryableIndexStorageAdapter(index);
final List<String> columnNames = getColumnsToInclude(index);
final DimFilter filter = filterJson != null ? objectMapper.readValue(filterJson, DimFilter.class) : null;
final Sequence<Cursor> cursors = adapter.makeCursors(Filters.toFilter(filter), index.getDataInterval().withChronology(ISOChronology.getInstanceUTC()), VirtualColumns.EMPTY, Granularities.ALL, false);
withOutputStream(new Function<OutputStream, Object>() {
@Override
public Object apply(final OutputStream out) {
final Sequence<Object> sequence = Sequences.map(cursors, new Function<Cursor, Object>() {
@Override
public Object apply(Cursor cursor) {
final List<ObjectColumnSelector> selectors = Lists.newArrayList();
for (String columnName : columnNames) {
selectors.add(makeSelector(columnName, index.getColumn(columnName), cursor));
}
while (!cursor.isDone()) {
final Map<String, Object> row = Maps.newLinkedHashMap();
for (int i = 0; i < columnNames.size(); i++) {
final String columnName = columnNames.get(i);
final Object value = selectors.get(i).get();
if (timeISO8601 && columnNames.get(i).equals(Column.TIME_COLUMN_NAME)) {
row.put(columnName, new DateTime(value, DateTimeZone.UTC).toString());
} else {
row.put(columnName, value);
}
}
try {
out.write(objectMapper.writeValueAsBytes(row));
out.write('\n');
} catch (IOException e) {
throw Throwables.propagate(e);
}
cursor.advance();
}
return null;
}
});
evaluateSequenceForSideEffects(sequence);
return null;
}
});
}
use of io.druid.segment.Cursor in project druid by druid-io.
the class FilterPartitionBenchmark method timeFilterHalf.
@Benchmark
@BenchmarkMode(Mode.AverageTime)
@OutputTimeUnit(TimeUnit.MICROSECONDS)
public void timeFilterHalf(Blackhole blackhole) throws Exception {
StorageAdapter sa = new QueryableIndexStorageAdapter(qIndex);
Sequence<Cursor> cursors = makeCursors(sa, timeFilterHalf);
Sequence<List<Long>> longListSeq = readCursorsLong(cursors, blackhole);
List<Long> strings = Sequences.toList(Sequences.limit(longListSeq, 1), Lists.<List<Long>>newArrayList()).get(0);
for (Long st : strings) {
blackhole.consume(st);
}
}
Aggregations