[GLUTEN-1632][CH]Daily Update Clickhouse Version (20240104)

backends-clickhouse/src/test/scala/io/glutenproject/execution/GlutenClickHouseTPCDSParquetAQESuite.scala

@@ -70,7 +70,7 @@ class GlutenClickHouseTPCDSParquetAQESuite
                           |  where ss_quantity between 1 and 20
                           |  and ss_sold_date_sk = 2452635
                           |""".stripMargin) { _ => }
-    assert(result(0).getDouble(0) == 379.21313271604936)
+    AlmostEqualsIsRel(379.21313271604936, result.head.getDouble(0), DBL_RELAX_EPSILON)
   }
 
   test("test select avg(int), avg(long)") {

backends-clickhouse/src/test/scala/io/glutenproject/execution/GlutenClickHouseTPCDSParquetColumnarShuffleAQESuite.scala

@@ -63,7 +63,7 @@ class GlutenClickHouseTPCDSParquetColumnarShuffleAQESuite
                           |  where ss_quantity between 1 and 20
                           |  and ss_sold_date_sk = 2452635
                           |""".stripMargin) { _ => }
-    assert(result(0).getDouble(0) == 379.21313271604936)
+    AlmostEqualsIsRel(379.21313271604936, result.head.getDouble(0), DBL_RELAX_EPSILON)
   }
 
   test("test select avg(int), avg(long)") {

backends-clickhouse/src/test/scala/io/glutenproject/execution/GlutenClickHouseTPCDSParquetColumnarShuffleSuite.scala

@@ -60,7 +60,7 @@ class GlutenClickHouseTPCDSParquetColumnarShuffleSuite extends GlutenClickHouseT
                           |  where ss_quantity between 1 and 20
                           |  and ss_sold_date_sk = 2452635
                           |""".stripMargin) { _ => }
-    assert(result(0).getDouble(0) == 379.21313271604936)
+    AlmostEqualsIsRel(379.21313271604936, result.head.getDouble(0), DBL_RELAX_EPSILON)
   }
 
   test("test select avg(int), avg(long)") {

backends-clickhouse/src/test/scala/io/glutenproject/execution/GlutenClickHouseTPCDSParquetSuite.scala

@@ -76,7 +76,7 @@ class GlutenClickHouseTPCDSParquetSuite extends GlutenClickHouseTPCDSAbstractSui
                           |  where ss_quantity between 1 and 20
                           |  and ss_sold_date_sk = 2452635
                           |""".stripMargin) { _ => }
-    assert(result(0).getDouble(0) == 379.21313271604936)
+    AlmostEqualsIsRel(379.21313271604936, result.head.getDouble(0), DBL_RELAX_EPSILON)
   }
 
   test("test select avg(int), avg(long)") {

backends-clickhouse/src/test/scala/io/glutenproject/execution/GlutenClickHouseWholeStageTransformerSuite.scala

@@ -16,4 +16,20 @@
  */
 package io.glutenproject.execution
 
-abstract class GlutenClickHouseWholeStageTransformerSuite extends WholeStageTransformerSuite {}
+abstract class GlutenClickHouseWholeStageTransformerSuite extends WholeStageTransformerSuite {
+
+  val DBL_EPSILON = 2.2204460492503131e-16
+  val DBL_RELAX_EPSILON: Double = Math.pow(10, -11)
+  val FLT_EPSILON = 1.19209290e-07f
+  def AlmostEqualsIsRel(expected: Double, actual: Double, EPSILON: Double = DBL_EPSILON): Unit = {
+    val diff = Math.abs(expected - actual)
+    val epsilon = EPSILON * Math.max(Math.abs(expected), Math.abs(actual))
+    if (diff > epsilon) {
+      fail(s"""
+              |expected: $expected
+              |actual:   $actual
+              | abs(expected-expected) ~ epsilon = $diff ~ $epsilon
+              |""".stripMargin)
+    }
+  }
+}

cpp-ch/clickhouse.version

@@ -1,3 +1,3 @@
 CH_ORG=Kyligence
-CH_BRANCH=rebase_ch/20240103
-CH_COMMIT=9c039962813
+CH_BRANCH=rebase_ch/20240104
+CH_COMMIT=7ed09d2c7ac

cpp-ch/local-engine/Storages/ch_parquet/OptimizedParquetBlockInputFormat.cpp

@@ -45,7 +45,7 @@ OptimizedParquetBlockInputFormat::OptimizedParquetBlockInputFormat(ReadBuffer &
 {
 }
 
-Chunk OptimizedParquetBlockInputFormat::generate()
+Chunk OptimizedParquetBlockInputFormat::read()
 {
     Chunk res;
     block_missing_values.clear();
@@ -62,7 +62,7 @@ Chunk OptimizedParquetBlockInputFormat::generate()
     std::shared_ptr<arrow::Table> table;
     arrow::Status read_status = file_reader->ReadRowGroup(row_group_current, column_indices, &table);
     if (!read_status.ok())
-        throw ParsingException(ErrorCodes::CANNOT_READ_ALL_DATA, "Error while reading Parquet data: {}", read_status.ToString());
+        throw Exception(ErrorCodes::CANNOT_READ_ALL_DATA, "Error while reading Parquet data: {}", read_status.ToString());
 
     if (format_settings.use_lowercase_column_name)
         table = *table->RenameColumns(column_names);

cpp-ch/local-engine/Storages/ch_parquet/OptimizedParquetBlockInputFormat.h

@@ -48,7 +48,7 @@ class OptimizedParquetBlockInputFormat : public IInputFormat
     const BlockMissingValues & getMissingValues() const override;
 
 private:
-    Chunk generate() override;
+    Chunk read() override;
 
 protected:
     void prepareReader();

cpp-ch/local-engine/tests/gtest_ch_join.cpp

@@ -199,6 +199,7 @@ TEST(TestJoin, StorageJoinFromReadBufferTest)
     auto table_join = std::make_shared<TableJoin>(SizeLimits(), false, JoinKind::Left, JoinStrictness::All, cols);
     auto join_storage = std::shared_ptr<StorageJoinFromReadBuffer>(new StorageJoinFromReadBuffer( // NOLINT
         *in,
+        2048, // Even if you don't know the number of rows, passing an arbitrary value is fine
         cols,
         false,
         table_join,