Examples with ColumnStatistics io.trino.orc.metadata.statistics.ColumnStatistics used on opensource projects

Example 1 with ColumnStatistics

use of io.trino.orc.metadata.statistics.ColumnStatistics in project trino by trinodb.

the class ListColumnWriter method getIndexStreams.

@Override
public List<StreamDataOutput> getIndexStreams(CompressedMetadataWriter metadataWriter) throws IOException {
    checkState(closed);
    ImmutableList.Builder<RowGroupIndex> rowGroupIndexes = ImmutableList.builder();
    List<LongStreamCheckpoint> lengthCheckpoints = lengthStream.getCheckpoints();
    Optional<List<BooleanStreamCheckpoint>> presentCheckpoints = presentStream.getCheckpoints();
    for (int i = 0; i < rowGroupColumnStatistics.size(); i++) {
        int groupId = i;
        ColumnStatistics columnStatistics = rowGroupColumnStatistics.get(groupId);
        LongStreamCheckpoint lengthCheckpoint = lengthCheckpoints.get(groupId);
        Optional<BooleanStreamCheckpoint> presentCheckpoint = presentCheckpoints.map(checkpoints -> checkpoints.get(groupId));
        List<Integer> positions = createArrayColumnPositionList(compressed, lengthCheckpoint, presentCheckpoint);
        rowGroupIndexes.add(new RowGroupIndex(positions, columnStatistics));
    }
    Slice slice = metadataWriter.writeRowIndexes(rowGroupIndexes.build());
    Stream stream = new Stream(columnId, StreamKind.ROW_INDEX, slice.length(), false);
    ImmutableList.Builder<StreamDataOutput> indexStreams = ImmutableList.builder();
    indexStreams.add(new StreamDataOutput(slice, stream));
    indexStreams.addAll(elementWriter.getIndexStreams(metadataWriter));
    indexStreams.addAll(elementWriter.getBloomFilters(metadataWriter));
    return indexStreams.build();
}

Example 2 with ColumnStatistics

use of io.trino.orc.metadata.statistics.ColumnStatistics in project trino by trinodb.

the class ListColumnWriter method finishRowGroup.

@Override
public Map<OrcColumnId, ColumnStatistics> finishRowGroup() {
    checkState(!closed);
    ColumnStatistics statistics = new ColumnStatistics((long) nonNullValueCount, 0, null, null, null, null, null, null, null, null, null);
    rowGroupColumnStatistics.add(statistics);
    nonNullValueCount = 0;
    ImmutableMap.Builder<OrcColumnId, ColumnStatistics> columnStatistics = ImmutableMap.builder();
    columnStatistics.put(columnId, statistics);
    columnStatistics.putAll(elementWriter.finishRowGroup());
    return columnStatistics.buildOrThrow();
}

Example 3 with ColumnStatistics

use of io.trino.orc.metadata.statistics.ColumnStatistics in project trino by trinodb.

the class SliceDirectColumnWriter method getIndexStreams.

@Override
public List<StreamDataOutput> getIndexStreams(CompressedMetadataWriter metadataWriter) throws IOException {
    checkState(closed);
    ImmutableList.Builder<RowGroupIndex> rowGroupIndexes = ImmutableList.builder();
    List<LongStreamCheckpoint> lengthCheckpoints = lengthStream.getCheckpoints();
    List<ByteArrayStreamCheckpoint> dataCheckpoints = dataStream.getCheckpoints();
    Optional<List<BooleanStreamCheckpoint>> presentCheckpoints = presentStream.getCheckpoints();
    for (int i = 0; i < rowGroupColumnStatistics.size(); i++) {
        int groupId = i;
        ColumnStatistics columnStatistics = rowGroupColumnStatistics.get(groupId);
        LongStreamCheckpoint lengthCheckpoint = lengthCheckpoints.get(groupId);
        ByteArrayStreamCheckpoint dataCheckpoint = dataCheckpoints.get(groupId);
        Optional<BooleanStreamCheckpoint> presentCheckpoint = presentCheckpoints.map(checkpoints -> checkpoints.get(groupId));
        List<Integer> positions = createSliceColumnPositionList(compressed, lengthCheckpoint, dataCheckpoint, presentCheckpoint);
        rowGroupIndexes.add(new RowGroupIndex(positions, columnStatistics));
    }
    Slice slice = metadataWriter.writeRowIndexes(rowGroupIndexes.build());
    Stream stream = new Stream(columnId, StreamKind.ROW_INDEX, slice.length(), false);
    return ImmutableList.of(new StreamDataOutput(slice, stream));
}

Example 4 with ColumnStatistics

use of io.trino.orc.metadata.statistics.ColumnStatistics in project trino by trinodb.

the class DecimalColumnWriter method finishRowGroup.

@Override
public Map<OrcColumnId, ColumnStatistics> finishRowGroup() {
    checkState(!closed);
    ColumnStatistics statistics;
    if (type.isShort()) {
        statistics = shortDecimalStatisticsBuilder.buildColumnStatistics();
        shortDecimalStatisticsBuilder = new ShortDecimalStatisticsBuilder(type.getScale());
    } else {
        statistics = longDecimalStatisticsBuilder.buildColumnStatistics();
        longDecimalStatisticsBuilder = new LongDecimalStatisticsBuilder();
    }
    rowGroupColumnStatistics.add(statistics);
    return ImmutableMap.of(columnId, statistics);
}

Example 5 with ColumnStatistics

use of io.trino.orc.metadata.statistics.ColumnStatistics in project trino by trinodb.

the class FloatColumnWriter method getBloomFilters.

@Override
public List<StreamDataOutput> getBloomFilters(CompressedMetadataWriter metadataWriter) throws IOException {
    List<BloomFilter> bloomFilters = rowGroupColumnStatistics.stream().map(ColumnStatistics::getBloomFilter).filter(Objects::nonNull).collect(toImmutableList());
    if (!bloomFilters.isEmpty()) {
        Slice slice = metadataWriter.writeBloomFilters(bloomFilters);
        Stream stream = new Stream(columnId, StreamKind.BLOOM_FILTER_UTF8, slice.length(), false);
        return ImmutableList.of(new StreamDataOutput(slice, stream));
    }
    return ImmutableList.of();
}