Search in sources :

Example 1 with DataCachingSelectiveColumnReader

use of io.prestosql.orc.reader.DataCachingSelectiveColumnReader in project hetu-core by openlookeng.

the class AbstractOrcRecordReader method advanceToNextRowGroup.

private boolean advanceToNextRowGroup() throws IOException {
    nextRowInGroup = 0;
    if (currentRowGroup >= 0) {
        if (rowGroupStatisticsValidation.isPresent()) {
            OrcWriteValidation.StatisticsValidation statisticsValidation = rowGroupStatisticsValidation.get();
            long offset = stripes.get(currentStripe).getOffset();
            writeValidation.get().validateRowGroupStatistics(orcDataSource.getId(), offset, currentRowGroup, statisticsValidation.build().get());
            statisticsValidation.reset();
        }
    }
    while (!rowGroups.hasNext() && currentStripe < stripes.size()) {
        advanceToNextStripe();
        currentRowGroup = -1;
    }
    if (!rowGroups.hasNext()) {
        currentGroupRowCount = 0;
        return false;
    }
    currentRowGroup++;
    RowGroup localCurrentRowGroup = rowGroups.next();
    currentGroupRowCount = localCurrentRowGroup.getRowCount();
    if (localCurrentRowGroup.getMinAverageRowBytes() > 0) {
        maxBatchSize = toIntExact(min(maxBatchSize, max(1, maxBlockBytes / localCurrentRowGroup.getMinAverageRowBytes())));
    }
    currentPosition = currentStripePosition + localCurrentRowGroup.getRowOffset();
    filePosition = stripeFilePositions.get(currentStripe) + localCurrentRowGroup.getRowOffset();
    // give reader data streams from row group
    InputStreamSources rowGroupStreamSources = localCurrentRowGroup.getStreamSources();
    for (AbstractColumnReader columnReader : columnReaders) {
        if (columnReader != null) {
            if (columnReader instanceof CachingColumnReader || columnReader instanceof ResultCachingSelectiveColumnReader || columnReader instanceof DataCachingSelectiveColumnReader) {
                StreamSourceMeta streamSourceMeta = new StreamSourceMeta();
                streamSourceMeta.setDataSourceId(orcDataSource.getId());
                streamSourceMeta.setLastModifiedTime(orcDataSource.getLastModifiedTime());
                streamSourceMeta.setStripeOffset(stripes.get(currentStripe).getOffset());
                streamSourceMeta.setRowGroupOffset(localCurrentRowGroup.getRowOffset());
                streamSourceMeta.setRowCount(localCurrentRowGroup.getRowCount());
                rowGroupStreamSources.setStreamSourceMeta(streamSourceMeta);
            }
            columnReader.startRowGroup(rowGroupStreamSources);
        }
    }
    return true;
}
Also used : InputStreamSources(io.prestosql.orc.stream.InputStreamSources) ResultCachingSelectiveColumnReader(io.prestosql.orc.reader.ResultCachingSelectiveColumnReader) DataCachingSelectiveColumnReader(io.prestosql.orc.reader.DataCachingSelectiveColumnReader) AbstractColumnReader(io.prestosql.orc.reader.AbstractColumnReader) StreamSourceMeta(io.prestosql.orc.stream.StreamSourceMeta) CachingColumnReader(io.prestosql.orc.reader.CachingColumnReader)

Aggregations

AbstractColumnReader (io.prestosql.orc.reader.AbstractColumnReader)1 CachingColumnReader (io.prestosql.orc.reader.CachingColumnReader)1 DataCachingSelectiveColumnReader (io.prestosql.orc.reader.DataCachingSelectiveColumnReader)1 ResultCachingSelectiveColumnReader (io.prestosql.orc.reader.ResultCachingSelectiveColumnReader)1 InputStreamSources (io.prestosql.orc.stream.InputStreamSources)1 StreamSourceMeta (io.prestosql.orc.stream.StreamSourceMeta)1