Fix alter column stats

apache · Nov 16, 2023 · 2d72713 · 2d72713
1 parent fd6a2cb
commit 2d72713
Show file tree

Hide file tree

Showing 2 changed files with 45 additions and 12 deletions.
diff --git a/fe/fe-core/src/main/java/org/apache/doris/statistics/StatisticsRepository.java b/fe/fe-core/src/main/java/org/apache/doris/statistics/StatisticsRepository.java
@@ -19,7 +19,6 @@
 
 import org.apache.doris.analysis.AlterColumnStatsStmt;
 import org.apache.doris.analysis.TableName;
-import org.apache.doris.catalog.Column;
 import org.apache.doris.catalog.Env;
 import org.apache.doris.catalog.Partition;
 import org.apache.doris.common.AnalysisException;
@@ -35,6 +34,8 @@
 import org.apache.logging.log4j.LogManager;
 import org.apache.logging.log4j.Logger;
 
+import java.nio.charset.StandardCharsets;
+import java.util.Base64;
 import java.util.Collection;
 import java.util.Collections;
 import java.util.HashMap;
@@ -227,7 +228,6 @@ public static void alterColumnStatistics(AlterColumnStatsStmt alterColumnStatsSt
  String dataSize = alterColumnStatsStmt.getValue(StatsType.DATA_SIZE);
  ColumnStatisticBuilder builder = new ColumnStatisticBuilder();
  String colName = alterColumnStatsStmt.getColumnName();
- Column column = objects.table.getColumn(colName);
  if (rowCount != null) {
  builder.setCount(Double.parseDouble(rowCount));
  }
@@ -239,14 +239,6 @@ public static void alterColumnStatistics(AlterColumnStatsStmt alterColumnStatsSt
  if (nullCount != null) {
  builder.setNumNulls(Double.parseDouble(nullCount));
  }
- if (min != null) {
- builder.setMinExpr(StatisticsUtil.readableValue(column.getType(), min));
- builder.setMinValue(StatisticsUtil.convertToDouble(column.getType(), min));
- }
- if (max != null) {
- builder.setMaxExpr(StatisticsUtil.readableValue(column.getType(), max));
- builder.setMaxValue(StatisticsUtil.convertToDouble(column.getType(), max));
- }
  if (dataSize != null) {
  double size = Double.parseDouble(dataSize);
  double rows = Double.parseDouble(rowCount);
@@ -269,8 +261,10 @@ public static void alterColumnStatistics(AlterColumnStatsStmt alterColumnStatsSt
  params.put("count", String.valueOf(columnStatistic.count));
  params.put("ndv", String.valueOf(columnStatistic.ndv));
  params.put("nullCount", String.valueOf(columnStatistic.numNulls));
- params.put("min", min == null ? "NULL" : min);
- params.put("max", max == null ? "NULL" : max);
+ params.put("min", min == null ? "NULL" :
+ Base64.getEncoder().encodeToString(min.getBytes(StandardCharsets.UTF_8)));
+ params.put("max", max == null ? "NULL" :
+ Base64.getEncoder().encodeToString(max.getBytes(StandardCharsets.UTF_8)));
  params.put("dataSize", String.valueOf(columnStatistic.dataSize));
 
  if (partitionIds.isEmpty()) {

diff --git a/regression-test/suites/external_table_p2/hive/test_hive_statistic.groovy b/regression-test/suites/external_table_p2/hive/test_hive_statistic.groovy
@@ -293,6 +293,45 @@ suite("test_hive_statistic", "p2,external,hive,external_remote,external_remote_h
  assertEquals(result.size(), 1)
  assertEquals(result[0][6], "N/A")
  assertEquals(result[0][7], "N/A")
+
+ sql """use tpch1_parquet;"""
+ sql """drop stats region"""
+ sql """alter table region modify column r_comment set stats ('row_count'='5.0', 'ndv'='5.0', 'num_nulls'='0.0', 'data_size'='330.0', 'min_value'='ges. thinly even pinto beans ca', 'max_value'='uickly special accounts cajole carefully blithely close requests. carefully final asymptotes haggle furiousl');"""
+ sql """alter table region modify column r_name set stats ('row_count'='5.0', 'ndv'='5.0', 'num_nulls'='0.0', 'data_size'='34.0', 'min_value'='AFRICA', 'max_value'='MIDDLE EAST');"""
+ sql """alter table region modify column r_regionkey set stats ('row_count'='5.0', 'ndv'='5.0', 'num_nulls'='0.0', 'data_size'='20.0', 'min_value'='0', 'max_value'='4');"""
+ result = sql """show column stats region(r_regionkey)"""
+ assertEquals(result.size(), 1)
+ assertEquals(result[0][0], "r_regionkey")
+ assertEquals(result[0][1], "5.0")
+ assertEquals(result[0][2], "5.0")
+ assertEquals(result[0][3], "0.0")
+ assertEquals(result[0][4], "20.0")
+ assertEquals(result[0][5], "4.0")
+ assertEquals(result[0][6], "0")
+ assertEquals(result[0][7], "4")
+
+ result = sql """show column stats region(r_comment)"""
+ assertEquals(result.size(), 1)
+ assertEquals(result[0][0], "r_comment")
+ assertEquals(result[0][1], "5.0")
+ assertEquals(result[0][2], "5.0")
+ assertEquals(result[0][3], "0.0")
+ assertEquals(result[0][4], "330.0")
+ assertEquals(result[0][5], "66.0")
+ assertEquals(result[0][6], "\'ges. thinly even pinto beans ca\'")
+ assertEquals(result[0][7], "\'uickly special accounts cajole carefully blithely close requests. carefully final asymptotes haggle furiousl\'")
+
+ result = sql """show column stats region(r_name)"""
+ assertEquals(result.size(), 1)
+ assertEquals(result[0][0], "r_name")
+ assertEquals(result[0][1], "5.0")
+ assertEquals(result[0][2], "5.0")
+ assertEquals(result[0][3], "0.0")
+ assertEquals(result[0][4], "34.0")
+ assertEquals(result[0][5], "6.8")
+ assertEquals(result[0][6], "\'AFRICA\'")
+ assertEquals(result[0][7], "\'MIDDLE EAST\'")
+
  sql """drop catalog ${catalog_name}"""
  }
 }