voonhous commented on code in PR #19163:
URL: https://github.com/apache/hudi/pull/19163#discussion_r3870771411


##########
hudi-spark-datasource/hudi-spark/src/test/scala/org/apache/spark/sql/hudi/ddl/TestAlterTableColumnCoverage.scala:
##########
@@ -0,0 +1,111 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.hudi.ddl
+
+import org.apache.spark.sql.catalyst.TableIdentifier
+import org.apache.spark.sql.hudi.common.HoodieSparkSqlTestBase
+
+/**
+ * Column add / rename / drop / position / comment coverage for
+ * [[org.apache.spark.sql.hudi.command.AlterTableCommand]] driven through the 
schema-on-read
+ * (schema evolution) path. Uses an unpartitioned table so the 
`commitWithSchema` data-schema
+ * derivation takes the no-partition-column branch, complementing the 
partitioned coverage in
+ * [[TestSpark3DDL]].
+ */
+class TestAlterTableColumnCoverage extends HoodieSparkSqlTestBase {
+
+  test("Test alter table add/rename/drop/position/comment on unpartitioned 
table") {
+    withSQLConf("hoodie.schema.on.read.enable" -> "true") {
+      withTempDir { tmp =>
+        val tableName = generateTableName
+        spark.sql(
+          s"""
+             |create table $tableName (
+             |  id int,
+             |  name string,
+             |  price double,
+             |  ts long
+             |) using hudi
+             | location '${tmp.getCanonicalPath}'
+             | tblproperties (type = 'cow', primaryKey = 'id', preCombineField 
= 'ts')
+           """.stripMargin)
+        spark.sql(s"insert into $tableName values (1, 'a1', 10.0, 1000)")
+
+        // ADD columns (first alter -> empty history schema branch), with a 
comment and an
+        // explicit position.
+        spark.sql(s"alter table $tableName add columns (age int comment 'the 
age' after name)")
+        spark.sql(s"insert into $tableName values (2, 'a2', 25, 20.0, 2000)")
+        checkAnswer(s"select id, name, age, price, ts from $tableName order by 
id")(
+          Seq(1, "a1", null, 10.0, 1000L),
+          Seq(2, "a2", 25, 20.0, 2000L)
+        )
+        val addedSchema = spark.sessionState.catalog
+          .getTableMetadata(TableIdentifier(tableName)).schema
+        assert(addedSchema.exists(_.name == "age"))
+        assert(addedSchema.find(_.name == 
"age").get.getComment().contains("the age"))
+
+        // ALTER column type widening (int -> long); second alter -> non-empty 
history branch.
+        spark.sql(s"alter table $tableName alter column age type long")
+        
assert(spark.sessionState.catalog.getTableMetadata(TableIdentifier(tableName))
+          .schema.find(_.name == "age").get.dataType.typeName == "long")
+
+        checkAnswer(s"select id, name, age from $tableName order by id")(
+          Seq(1, "a1", null),
+          Seq(2, "a2", 25L)
+        )
+
+        // ALTER column comment on an existing column.
+        spark.sql(s"alter table $tableName alter column price comment 'unit 
price'")
+        
assert(spark.sessionState.catalog.getTableMetadata(TableIdentifier(tableName))
+          .schema.find(_.name == "price").get.getComment().contains("unit 
price"))
+
+        // NOTE: RENAME COLUMN and DROP COLUMN (applyDeleteAction) are not 
exercised here. On this

Review Comment:
   Filed as #19766, with a repro against `d291efccaad2` (same code on master): 
both RENAME and DROP throw `MissingSchemaFieldException` from 
`commitWithSchema:280` under defaults and pass with `column.drop=true`. The 
NOTE is gone with the file (line-26 thread); the PR description links the issue.
   



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to