public Statistics<?> build() { Statistics<?> stats = createStats(type); if (min != null && max != null) { stats.setMinMaxFromBytes(min, max); } stats.num_nulls = this.numNulls; return stats; } }
public Statistics<?> build() { Statistics<?> stats = createStats(type); if (min != null && max != null) { stats.setMinMaxFromBytes(min, max); } stats.num_nulls = this.numNulls; return stats; } }
public static org.apache.parquet.column.statistics.Statistics<?> readStats(Statistics statistics, PrimitiveTypeName type) { org.apache.parquet.column.statistics.Statistics<?> stats = org.apache.parquet.column.statistics.Statistics.getStatsBasedOnType(type); if (statistics != null) { if (statistics.isSetMax() && statistics.isSetMin()) { stats.setMinMaxFromBytes(statistics.min.array(), statistics.max.array()); } stats.setNumNulls(statistics.null_count); } return stats; }
public static org.apache.parquet.column.statistics.Statistics<?> readStats(Statistics statistics, PrimitiveTypeName type) { org.apache.parquet.column.statistics.Statistics<?> stats = org.apache.parquet.column.statistics.Statistics.getStatsBasedOnType(type); if (statistics != null) { if (statistics.isSetMax() && statistics.isSetMin()) { stats.setMinMaxFromBytes(statistics.min.array(), statistics.max.array()); } stats.setNumNulls(statistics.null_count); } return stats; }
stat.setNumNulls(0); byte[] val = implicitColValues.get(columnName).getBytes(); stat.setMinMaxFromBytes(val, val); statMap.put(path, new ColumnStatistics(stat, type));
@Test public void testMatchesWithStatistics() throws ParquetCorruptionException { String value = "Test"; ColumnDescriptor columnDescriptor = new ColumnDescriptor(new String[] {"path"}, BINARY, 0, 0); RichColumnDescriptor column = new RichColumnDescriptor(columnDescriptor, new PrimitiveType(OPTIONAL, BINARY, "Test column")); TupleDomain<ColumnDescriptor> effectivePredicate = getEffectivePredicate(column, createVarcharType(255), utf8Slice(value)); TupleDomainParquetPredicate parquetPredicate = new TupleDomainParquetPredicate(effectivePredicate, singletonList(column)); Statistics stats = getStatsBasedOnType(column.getType()); stats.setNumNulls(1L); stats.setMinMaxFromBytes(value.getBytes(), value.getBytes()); assertTrue(parquetPredicate.matches(2, singletonMap(column, stats), ID, true)); }
stat.setNumNulls(0); byte[] val = implicitColValues.get(columnName).getBytes(); stat.setMinMaxFromBytes(val, val); statMap.put(field, new ColumnStatistics(stat, type));