dtenedor commented on code in PR #59640:
URL: https://github.com/apache/spark/pull/59640#discussion_r4382014323
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/rules.scala:
##########
@@ -293,6 +295,8 @@ case class PreprocessTableCreation(catalog: SessionCatalog)
extends Rule[Logical
c.copy(tableDesc = normalizedTable, query = Some(reorderedQuery))
} else {
+ checkFileTypeHasQualifier(tableDesc.schema)
Review Comment:
Sounds good. Could you link the follow-up JIRA here or in the PR
description? This is the one remaining path that can store a bare FILE in a
table, so it should block enabling `spark.sql.fileType.enabled` by default.
##########
sql/core/src/test/scala/org/apache/spark/sql/FileTypeRestrictionsSuite.scala:
##########
@@ -0,0 +1,357 @@
+/*
+ * 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
+
+import org.apache.spark.SparkConf
+import org.apache.spark.api.python.{PythonBroadcast, PythonEvalType,
SimplePythonFunction}
+import org.apache.spark.broadcast.Broadcast
+import org.apache.spark.sql.api.java.UDF1
+import org.apache.spark.sql.catalyst.TableIdentifier
+import org.apache.spark.sql.catalyst.catalog.{CatalogStorageFormat,
CatalogTable, CatalogTableType}
+import org.apache.spark.sql.connector.catalog.InMemoryCatalog
+import org.apache.spark.sql.execution.python.{UserDefinedPythonFunction,
UserDefinedPythonTableFunction}
+import org.apache.spark.sql.functions.{col, udf}
+import org.apache.spark.sql.test.SharedSparkSession
+import org.apache.spark.sql.types.{DataType, FileType, IntegerType, MapType,
StructType}
+
+/**
+ * Tests the rules on where the FILE type can be used: table columns must be
FILE EXTERNAL,
+ * routines declare a plain FILE, and FILE has no defined ordering.
+ */
+class FileTypeRestrictionsSuite extends QueryTest with SharedSparkSession {
+
+ override protected def sparkConf: SparkConf =
+ super.sparkConf.set("spark.sql.catalog.testcat",
classOf[InMemoryCatalog].getName)
+
+ // A FILE without a qualifier, at the top level and nested.
+ private val unqualifiedFileTypes = Seq(
+ "FILE",
+ "ARRAY<FILE>",
+ "MAP<STRING, FILE>",
+ "STRUCT<f1: FILE>",
+ "STRUCT<f1: INT, f2: STRUCT<f3: ARRAY<FILE>>>")
+
+ private val externalFile = "SELECT CAST(NULL AS FILE EXTERNAL) AS f"
+
+ test("a table column definition requires FILE EXTERNAL") {
+ Seq("testcat.t", "t").foreach { table =>
+ val provider = if (table.startsWith("testcat")) "" else "USING parquet"
+ unqualifiedFileTypes.foreach { dataType =>
+ withTable(table) {
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"CREATE TABLE $table (id INT, c $dataType) $provider")
+ },
+ condition = "FILE_TYPE_MISSING_QUALIFIER")
+ }
+ }
+ }
+ withTable("testcat.t") {
+ sql("CREATE TABLE testcat.t (id INT, c STRUCT<f1: ARRAY<FILE
EXTERNAL>>)")
+ }
+ }
+
+ test("REPLACE TABLE requires FILE EXTERNAL") {
+ withTable("testcat.t") {
+ sql("CREATE TABLE testcat.t (id INT)")
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql("REPLACE TABLE testcat.t (id INT, c FILE)")
+ },
+ condition = "FILE_TYPE_MISSING_QUALIFIER")
+ }
+ }
+
+ test("ALTER TABLE ADD COLUMNS and REPLACE COLUMNS require FILE EXTERNAL") {
+ withTable("testcat.t") {
+ sql("CREATE TABLE testcat.t (id INT)")
+ unqualifiedFileTypes.foreach { dataType =>
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"ALTER TABLE testcat.t ADD COLUMN c $dataType")
+ },
+ condition = "FILE_TYPE_MISSING_QUALIFIER")
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"ALTER TABLE testcat.t REPLACE COLUMNS (id INT, c $dataType)")
+ },
+ condition = "FILE_TYPE_MISSING_QUALIFIER")
+ }
+ sql("ALTER TABLE testcat.t ADD COLUMN c FILE EXTERNAL")
+ }
+ }
+
+ test("CREATE TABLE AS SELECT cannot store a FILE without a qualifier") {
+ Seq("testcat.t", "t").foreach { table =>
+ val provider = if (table.startsWith("testcat")) "" else "USING parquet"
+ withTable(table) {
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"CREATE TABLE $table $provider AS SELECT 1 AS id, CAST(NULL
AS FILE) AS f")
+ },
+ condition = "FILE_TYPE_UNKNOWN_WRITE",
+ parameters = Map("colNames" -> "[`f`]"))
+ }
+ }
+ }
+
+ test("a FILE column cannot be a partition or clustering column") {
+ Seq(
+ ("PARTITIONED BY (f)", "PARTITION_COLUMN"),
+ ("PARTITIONED BY (bucket(4, f))", "PARTITION_COLUMN"),
+ ("CLUSTER BY (f)", "CLUSTERING_COLUMN")).foreach { case (clause, usage)
=>
+ Seq("testcat.t", "t").foreach { table =>
+ // Bucketing is only a V2 transform.
+ if (table.startsWith("testcat") || !clause.contains("bucket")) {
+ val provider = if (table.startsWith("testcat")) "" else "USING
parquet"
+ withTable(table) {
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"CREATE TABLE $table (id INT, f FILE EXTERNAL) $provider
$clause")
+ },
+ condition = s"FILE_TYPE_UNSUPPORTED_COLUMN_USAGE.$usage",
+ parameters = Map("colName" -> "`f`", "dataType" -> "\"FILE
EXTERNAL\""))
+ }
+ }
+ }
+ }
+ Seq("testcat.t", "t").foreach { table =>
+ withTable(table) {
+ if (table.startsWith("testcat")) {
+ sql(s"CREATE TABLE $table (id INT, f FILE EXTERNAL) CLUSTER BY (id)")
+ } else {
+ // No V1 data source can store a FILE yet, so the table is created
in the catalog.
+ spark.sessionState.catalog.createTable(CatalogTable(
+ identifier = TableIdentifier(table),
+ tableType = CatalogTableType.MANAGED,
+ storage = CatalogStorageFormat.empty,
+ schema = new StructType().add("id", IntegerType).add("f",
FileType.external),
+ provider = Some("parquet")), ignoreIfExists = false)
+ }
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"ALTER TABLE $table CLUSTER BY (f)")
+ },
+ condition = "FILE_TYPE_UNSUPPORTED_COLUMN_USAGE.CLUSTERING_COLUMN",
+ parameters = Map("colName" -> "`f`", "dataType" -> "\"FILE
EXTERNAL\""))
+ }
+ }
+ }
+
+ test("a column cannot be altered to a FILE") {
+ withTable("testcat.t") {
+ sql("CREATE TABLE testcat.t (id INT, s STRING, f FILE EXTERNAL)")
+ Seq("s", "f").foreach { column =>
+ val e = intercept[AnalysisException] {
+ sql(s"ALTER TABLE testcat.t ALTER COLUMN $column TYPE FILE")
+ }
+ assert(e.getCondition === "NOT_SUPPORTED_CHANGE_COLUMN")
+ }
+ }
+ }
+
+ test("a FILE cannot be a map key") {
+ checkError(
+ exception = intercept[AnalysisException] {
+ sql(s"SELECT map(f, 1) FROM ($externalFile)")
+ },
+ condition = "DATATYPE_MISMATCH.INVALID_MAP_KEY_TYPE",
+ parameters = Map("sqlExpr" -> "\"map(f, 1)\"", "keyType" -> "\"FILE
EXTERNAL\""),
+ context = ExpectedContext(fragment = "map(f, 1)", start = 7, stop = 15))
+ sql(s"SELECT map(1, f) FROM ($externalFile)")
+ }
+
+ test("a FILE cannot be a grouping expression") {
+ val query = s"SELECT count(*) FROM ($externalFile) GROUP BY f"
+ checkError(
+ exception = intercept[AnalysisException](sql(query)),
+ condition = "GROUP_EXPRESSION_TYPE_IS_NOT_ORDERABLE",
+ parameters = Map("sqlExpr" -> "\"f\"", "dataType" -> "\"FILE
EXTERNAL\""),
+ context = ExpectedContext(
+ fragment = "f", start = query.length - 1, stop = query.length - 1))
+ }
+
+ test("a FILE cannot be in INTERSECT, EXCEPT or DISTINCT") {
Review Comment:
Thanks, most of these are in now. Two items don't hold yet: the V1 CTAS
cases with `PARTITIONED BY` and `CLUSTER BY` never run because of the new guard
in the partition test, and the query-time Python checks don't need a Python
worker. I've opened new threads for both.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
