[SPARK-46576][SQL] Improve error messages for unsupported data source…

… save mode ### What changes were proposed in this pull request? This PR renames the error class `_LEGACY_ERROR_TEMP_1308` to `UNSUPPORTED_DATA_SOURCE_SAVE_MODE` and improves its error messages. ### Why are the changes needed? To make the error more user-friendly. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? New unit tests ### Was this patch authored or co-authored using generative AI tooling? No Closes apache#44576 from allisonwang-db/spark-46576-unsupported-save-mode. Authored-by: allisonwang-db <[email protected]> Signed-off-by: Max Gekk <[email protected]>
AndrewXWei · Jan 4, 2024 · 69c4687 · 69c4687
1 parent 59d147a
commit 69c4687
Show file tree

Hide file tree

Showing 6 changed files with 34 additions and 18 deletions.
diff --git a/common/utils/src/main/resources/error/error-classes.json b/common/utils/src/main/resources/error/error-classes.json
@@ -3588,6 +3588,12 @@
     ],
     "sqlState" : "0A000"
   },
+  "UNSUPPORTED_DATA_SOURCE_SAVE_MODE" : {
+    "message" : [
+      "The data source '<source>' cannot be written in the <createMode> mode. Please use either the \"Append\" or \"Overwrite\" mode instead."
+    ],
+    "sqlState" : "0A000"
+  },
   "UNSUPPORTED_DATA_TYPE_FOR_DATASOURCE" : {
     "message" : [
       "The <format> datasource doesn't support the column <columnName> of the type <columnType>."
@@ -5403,11 +5409,6 @@
       "There is a 'path' option set and save() is called with a path parameter. Either remove the path option, or call save() without the parameter. To ignore this check, set '<config>' to 'true'."
     ]
   },
-  "_LEGACY_ERROR_TEMP_1308" : {
-    "message" : [
-      "TableProvider implementation <source> cannot be written with <createMode> mode, please use Append or Overwrite modes instead."
-    ]
-  },
   "_LEGACY_ERROR_TEMP_1309" : {
     "message" : [
       "insertInto() can't be used together with partitionBy(). Partition columns have already been defined for the table. It is not necessary to use partitionBy()."

diff --git a/connector/kafka-0-10-sql/src/test/scala/org/apache/spark/sql/kafka010/KafkaSinkSuite.scala b/connector/kafka-0-10-sql/src/test/scala/org/apache/spark/sql/kafka010/KafkaSinkSuite.scala
@@ -557,7 +557,7 @@ class KafkaSinkBatchSuiteV2 extends KafkaSinkBatchSuiteBase {
 
   test("batch - unsupported save modes") {
     testUnsupportedSaveModes((mode) =>
-      Seq(s"cannot be written with ${mode.name} mode", "does not support truncate"))
+      Seq(s"cannot be written in the \"${mode.name}\" mode", "does not support truncate"))
   }
 
   test("generic - write big data with small producer buffer") {

diff --git a/docs/sql-error-conditions.md b/docs/sql-error-conditions.md
@@ -2332,6 +2332,12 @@ Unsupported data source type for direct query on files: `<dataSourceType>`
 
 Unsupported data type `<typeName>`.
 
+### UNSUPPORTED_DATA_SOURCE_SAVE_MODE
+
+[SQLSTATE: 0A000](sql-error-conditions-sqlstates.html#class-0A-feature-not-supported)
+
+The data source '`<source>`' cannot be written in the `<createMode>` mode. Please use either the "Append" or "Overwrite" mode instead.
+
 ### UNSUPPORTED_DATA_TYPE_FOR_DATASOURCE
 
 [SQLSTATE: 0A000](sql-error-conditions-sqlstates.html#class-0A-feature-not-supported)

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/errors/QueryCompilationErrors.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/errors/QueryCompilationErrors.scala
@@ -3193,10 +3193,10 @@ private[sql] object QueryCompilationErrors extends QueryErrorsBase with Compilat
 
   def writeWithSaveModeUnsupportedBySourceError(source: String, createMode: String): Throwable = {
     new AnalysisException(
-      errorClass = "_LEGACY_ERROR_TEMP_1308",
+      errorClass = "UNSUPPORTED_DATA_SOURCE_SAVE_MODE",
       messageParameters = Map(
         "source" -> source,
-        "createMode" -> createMode))
+        "createMode" -> toDSOption(createMode)))
   }
 
   def partitionByDoesNotAllowedWhenUsingInsertIntoError(): Throwable = {

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/connector/DataSourceV2Suite.scala b/sql/core/src/test/scala/org/apache/spark/sql/connector/DataSourceV2Suite.scala
@@ -454,10 +454,10 @@ class DataSourceV2Suite extends QueryTest with SharedSparkSession with AdaptiveS
               .write.format(cls.getName)
               .option("path", path).mode("ignore").save()
           },
-          errorClass = "_LEGACY_ERROR_TEMP_1308",
+          errorClass = "UNSUPPORTED_DATA_SOURCE_SAVE_MODE",
           parameters = Map(
             "source" -> cls.getName,
-            "createMode" -> "Ignore"
+            "createMode" -> "\"Ignore\""
           )
         )
 
@@ -467,10 +467,10 @@ class DataSourceV2Suite extends QueryTest with SharedSparkSession with AdaptiveS
               .write.format(cls.getName)
               .option("path", path).mode("error").save()
           },
-          errorClass = "_LEGACY_ERROR_TEMP_1308",
+          errorClass = "UNSUPPORTED_DATA_SOURCE_SAVE_MODE",
           parameters = Map(
             "source" -> cls.getName,
-            "createMode" -> "ErrorIfExists"
+            "createMode" -> "\"ErrorIfExists\""
           )
         )
       }

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/python/PythonDataSourceSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/python/PythonDataSourceSuite.scala
@@ -622,12 +622,21 @@ class PythonDataSourceSuite extends QueryTest with SharedSparkSession {
     }
 
     withClue("without mode") {
-      val error = intercept[AnalysisException] {
-        spark.range(1).write.format(dataSourceName).save()
-      }
-      // TODO: improve this error message.
-      assert(error.getMessage.contains("TableProvider implementation SimpleDataSource " +
-        "cannot be written with ErrorIfExists mode, please use Append or Overwrite modes instead."))
+      checkError(
+        exception = intercept[AnalysisException] {
+          spark.range(1).write.format(dataSourceName).save()
+        },
+        errorClass = "UNSUPPORTED_DATA_SOURCE_SAVE_MODE",
+        parameters = Map("source" -> "SimpleDataSource", "createMode" -> "\"ErrorIfExists\""))
+    }
+
+    withClue("with unsupported mode") {
+      checkError(
+        exception = intercept[AnalysisException] {
+          spark.range(1).write.format(dataSourceName).mode("ignore").save()
+        },
+        errorClass = "UNSUPPORTED_DATA_SOURCE_SAVE_MODE",
+        parameters = Map("source" -> "SimpleDataSource", "createMode" -> "\"Ignore\""))
     }
 
     withClue("invalid mode") {