error class name changes, more details to error message

apache · Oct 29, 2022 · e5140b0 · e5140b0
1 parent dd63be8
commit e5140b0
Show file tree

Hide file tree

Showing 6 changed files with 99 additions and 94 deletions.
diff --git a/connector/protobuf/src/main/scala/org/apache/spark/sql/protobuf/ProtobufDeserializer.scala b/connector/protobuf/src/main/scala/org/apache/spark/sql/protobuf/ProtobufDeserializer.scala
@@ -195,7 +195,8 @@ private[sql] class ProtobufDeserializer(
         (updater, ordinal, value) =>
           val byte_array = value match {
             case s: ByteString => s.toByteArray
-            case _ => throw QueryCompilationErrors.invalidByteStringFormatError()
+            case unsupported =>
+              throw QueryCompilationErrors.invalidByteStringFormatError(unsupported)
           }
           updater.set(ordinal, byte_array)
 

diff --git a/connector/protobuf/src/main/scala/org/apache/spark/sql/protobuf/utils/ProtobufUtils.scala b/connector/protobuf/src/main/scala/org/apache/spark/sql/protobuf/utils/ProtobufUtils.scala
@@ -201,7 +201,7 @@ private[sql] object ProtobufUtils extends Logging {
       fileDescriptorSet = DescriptorProtos.FileDescriptorSet.parseFrom(dscFile)
     } catch {
       case ex: InvalidProtocolBufferException =>
-        throw QueryCompilationErrors.descrioptorParseError(ex)
+        throw QueryCompilationErrors.descrioptorParseError(descFilePath, ex)
       case ex: IOException =>
         throw QueryCompilationErrors.cannotFindDescriptorFileError(descFilePath, ex)
     }
@@ -214,7 +214,7 @@ private[sql] object ProtobufUtils extends Logging {
       fileDescriptorList
     } catch {
       case e: Descriptors.DescriptorValidationException =>
-        throw QueryCompilationErrors.failedParsingDescriptorError(e)
+        throw QueryCompilationErrors.failedParsingDescriptorError(descFilePath, e)
     }
   }
 

diff --git a/connector/protobuf/src/test/scala/org/apache/spark/sql/protobuf/ProtobufSerdeSuite.scala b/connector/protobuf/src/test/scala/org/apache/spark/sql/protobuf/ProtobufSerdeSuite.scala
@@ -71,7 +71,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
         protoFile,
         Deserializer,
         fieldMatch,
-        errorClass = "PROTOBUF_TYPE_TO_CATALYST_TYPE_ERROR",
+        errorClass = "CANNOT_CONVERT_PROTOBUF_MESSAGE_TYPE_TO_SQL_TYPE",
         params = Map(
           "protobufType" -> "MissMatchTypeInRoot",
           "toType" -> toSQLType(CATALYST_STRUCT)))
@@ -80,7 +80,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
         protoFile,
         Serializer,
         fieldMatch,
-        errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+        errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
         params = Map(
           "protobufType" -> "MissMatchTypeInRoot",
           "toType" -> toSQLType(CATALYST_STRUCT)))
@@ -98,15 +98,15 @@ class ProtobufSerdeSuite extends SharedSparkSession {
       protoFile,
       Serializer,
       BY_NAME,
-      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
       params = Map(
         "protobufType" -> "FieldMissingInProto",
         "toType" -> toSQLType(CATALYST_STRUCT)))
 
     assertFailedConversionMessage(protoFile,
       Serializer,
       BY_NAME,
-      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
       params = Map(
         "protobufType" -> "FieldMissingInProto",
         "toType" -> toSQLType(nonnullCatalyst)))
@@ -124,7 +124,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
         Deserializer,
         fieldMatch,
         catalyst,
-        errorClass = "PROTOBUF_TYPE_TO_CATALYST_TYPE_ERROR",
+        errorClass = "CANNOT_CONVERT_PROTOBUF_MESSAGE_TYPE_TO_SQL_TYPE",
         params = Map(
           "protobufType" -> "MissMatchTypeInDeepNested",
           "toType" -> toSQLType(catalyst)))
@@ -134,7 +134,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
         Serializer,
         fieldMatch,
         catalyst,
-        errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+        errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
         params = Map(
           "protobufType" -> "MissMatchTypeInDeepNested",
           "toType" -> toSQLType(catalyst)))
@@ -149,7 +149,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
       protoFile,
       Serializer,
       BY_NAME,
-      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
       params = Map(
         "protobufType" -> "FieldMissingInSQLRoot",
         "toType" -> toSQLType(CATALYST_STRUCT)))
@@ -166,7 +166,7 @@ class ProtobufSerdeSuite extends SharedSparkSession {
       protoNestedFile,
       Serializer,
       BY_NAME,
-      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE",
+      errorClass = "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE",
       params = Map(
         "protobufType" -> "FieldMissingInSQLNested",
         "toType" -> toSQLType(CATALYST_STRUCT)))
@@ -196,10 +196,10 @@ class ProtobufSerdeSuite extends SharedSparkSession {
 
     val expectMsg = serdeFactory match {
       case Deserializer =>
-        s"[PROTOBUF_TYPE_TO_CATALYST_TYPE_ERROR] Unable to convert" +
+        s"[CANNOT_CONVERT_PROTOBUF_MESSAGE_TYPE_TO_SQL_TYPE] Unable to convert" +
           s" ${protoSchema.getName} of Protobuf to SQL type ${toSQLType(catalystSchema)}."
       case Serializer =>
-        s"[UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE] Unable to convert SQL type" +
+        s"[UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE] Unable to convert SQL type" +
           s" ${toSQLType(catalystSchema)} to Protobuf type ${protoSchema.getName}."
     }
 

diff --git a/core/src/main/resources/error/error-classes.json b/core/src/main/resources/error/error-classes.json
@@ -17,29 +17,49 @@
     ],
     "sqlState" : "22005"
   },
+  "CANNOT_CONSTRUCT_PROTOBUF_DESCRIPTOR" : {
+    "message" : [
+      "Error constructing FileDescriptor for <descFilePath>"
+    ]
+  },
+  "CANNOT_CONVERT_PROTOBUF_MESSAGE_TYPE_TO_SQL_TYPE" : {
+    "message" : [
+      "Unable to convert <protobufType> of Protobuf to SQL type <toType>."
+    ]
+  },
+  "CANNOT_CONVERT_SQL_TYPE_TO_PROTOBUF_FIELD_TYPE" : {
+    "message" : [
+      "Cannot convert SQL <sqlColumn> to Protobuf <protobufColumn> because schema is incompatible (protobufType = <protobufType>, sqlType = <sqlType>)."
+    ]
+  },
   "CANNOT_DECODE_URL" : {
     "message" : [
       "Cannot decode url : <url>."
     ],
     "sqlState" : "42000"
   },
-  "CANNOT_FIND_PROTOBUF_DESCRIPTOR_FILE_ERROR" : {
-    "message" : [
-      "Error reading Protobuf descriptor file at path: <filePath>"
-    ]
-  },
   "CANNOT_INFER_DATE" : {
     "message" : [
       "Cannot infer date in schema inference when LegacyTimeParserPolicy is \"LEGACY\". Legacy Date formatter does not support strict date format matching which is required to avoid inferring timestamps and other non-date entries to date."
     ],
     "sqlState" : "22007"
   },
+  "CANNOT_LOAD_PROTOBUF_CLASS" : {
+    "message" : [
+      "Could not load Protobuf class with name <protobufClassName><message>"
+    ]
+  },
   "CANNOT_PARSE_DECIMAL" : {
     "message" : [
       "Cannot parse decimal"
     ],
     "sqlState" : "42000"
   },
+  "CANNOT_PARSE_PROTOBUF_DESCRIPTOR" : {
+    "message" : [
+      "Error parsing file <descFilePath> descriptor byte[] into Descriptor object"
+    ]
+  },
   "CANNOT_PARSE_TIMESTAMP" : {
     "message" : [
       "<message>. If necessary set <ansiConfig> to \"false\" to bypass this error."
@@ -70,11 +90,6 @@
     ],
     "sqlState" : "22005"
   },
-  "CATALYST_TYPE_TO_PROTOBUF_ENUM_TYPE_ERROR" : {
-    "message" : [
-      "Cannot convert SQL <sqlColumn> to Protobuf <protobufColumn> because <data> cannot be written since it's not defined in ENUM <enumString>"
-    ]
-  },
   "COLUMN_NOT_IN_GROUP_BY_CLAUSE" : {
     "message" : [
       "The expression <expression> is neither present in the group by, nor is it an aggregate function. Add to group by or wrap in `first()` (or `first_value()`) if you don't care which value you get."
@@ -527,9 +542,9 @@
       "Invalid bucket file: <path>"
     ]
   },
-  "INVALID_BYTE_STRING_ERROR" : {
+  "INVALID_BYTE_STRING" : {
     "message" : [
-      "Invalid ByteString format"
+      "The expected format is ByteString, but was <unsupported> (<class>)."
     ]
   },
   "INVALID_COLUMN_OR_FIELD_DATA_TYPE" : {
@@ -576,6 +591,11 @@
       "<value> is an invalid property value, please use quotes, e.g. SET <key>=<value>"
     ]
   },
+  "INVALID_PROTOBUF_MESSAGE_TYPE" : {
+    "message" : [
+      "<protobufClassName> is not a Protobuf message type"
+    ]
+  },
   "INVALID_SQL_SYNTAX" : {
     "message" : [
       "Invalid SQL syntax: <inputString>"
@@ -594,7 +614,7 @@
       }
     }
   },
-  "MALFORMED_PROTOBUF_MESSAGE_ERROR" : {
+  "MALFORMED_PROTOBUF_MESSAGE" : {
     "message" : [
       "Malformed Protobuf messages are detected in message deserialization. Parse Mode: <failFastMode>. To process malformed protobuf message as null result, try setting the option 'mode' as 'PERMISSIVE'."
     ]
@@ -649,22 +669,22 @@
     ],
     "sqlState" : "42000"
   },
-  "NO_CATALYST_TYPE_IN_PROTOBUF_SCHEMA" : {
-    "message" : [
-      "Cannot find <catalystFieldPath> in Protobuf schema"
-    ]
-  },
   "NO_HANDLER_FOR_UDAF" : {
     "message" : [
       "No handler for UDAF '<functionName>'. Use sparkSession.udf.register(...) instead."
     ]
   },
-  "NO_PROTOBUF_MESSAGE_TYPE_ERROR" : {
+  "NO_PROTOBUF_MESSAGE_TYPE" : {
     "message" : [
       "No MessageTypes returned, <descriptorName>"
     ]
   },
-  "NO_UDF_INTERFACE_ERROR" : {
+  "NO_SQL_TYPE_IN_PROTOBUF_SCHEMA" : {
+    "message" : [
+      "Cannot find <catalystFieldPath> in Protobuf schema"
+    ]
+  },
+  "NO_UDF_INTERFACE" : {
     "message" : [
       "UDF class <className> doesn't implement any UDF interface"
     ]
@@ -736,32 +756,22 @@
     ],
     "sqlState" : "42000"
   },
-  "PROTOBUF_CLASS_LOAD_ERROR" : {
-    "message" : [
-      "Could not load Protobuf class with name <protobufClassName><message>"
-    ]
-  },
-  "PROTOBUF_DEPENDENCY_ERROR" : {
+  "PROTOBUF_DEPENDENCY_NOT_FOUND" : {
     "message" : [
       "Could not find dependency: <dependencyName>"
     ]
   },
-  "PROTOBUF_DESCRIPTOR_ERROR" : {
-    "message" : [
-      "Error parsing descriptor byte[] into Descriptor object"
-    ]
-  },
-  "PROTOBUF_DESCRIPTOR_PARSING_ERROR" : {
+  "PROTOBUF_DESCRIPTOR_FILE_NOT_FOUND" : {
     "message" : [
-      "Error constructing FileDescriptor"
+      "Error reading Protobuf descriptor file at path: <filePath>"
     ]
   },
-  "PROTOBUF_FIELD_MISSING_ERROR" : {
+  "PROTOBUF_FIELD_MISSING" : {
     "message" : [
       "Searching for <field> in Protobuf schema at <protobufSchema> gave <matchSize> matches. Candidates: <matches>"
     ]
   },
-  "PROTOBUF_FIELD_MISSING_IN_CATALYST_SCHEMA" : {
+  "PROTOBUF_FIELD_MISSING_IN_SQL_SCHEMA" : {
     "message" : [
       "Found <field> in Protobuf schema but there is no match in the SQL schema"
     ]
@@ -771,29 +781,24 @@
       "Type mismatch encountered for field: <field>"
     ]
   },
-  "PROTOBUF_MESSAGE_TYPE_ERROR" : {
+  "PROTOBUF_FIELD_TYPE_TO_SQL_TYPE_ERROR" : {
     "message" : [
-      "<protobufClassName> is not a Protobuf message type"
+      "Cannot convert Protobuf <protobufColumn> to SQL <sqlColumn> because schema is incompatible (protobufType = <protobufType>, sqlType = <sqlType>)."
     ]
   },
-  "PROTOBUF_RECURSION_ERROR" : {
+  "PROTOBUF_MESSAGE_NOT_FOUND" : {
     "message" : [
-      "Found recursive reference in Protobuf schema, which can not be processed by Spark: <fieldDescriptor>"
+      "Unable to locate Message <messageName> in Descriptor"
     ]
   },
-  "PROTOBUF_TYPE_NOT_SUPPORT_ERROR" : {
+  "PROTOBUF_TYPE_NOT_SUPPORT" : {
     "message" : [
       "Protobuf type not yet supported: <protobufType>."
     ]
   },
-  "PROTOBUF_TYPE_TO_CATALYST_TYPE_ERROR" : {
-    "message" : [
-      "Unable to convert <protobufType> of Protobuf to SQL type <toType>."
-    ]
-  },
-  "PROTOBUF_TYPE_TO_SQL_TYPE_ERROR" : {
+  "RECURSIVE_PROTOBUF_SCHEMA" : {
     "message" : [
-      "Cannot convert Protobuf <protobufColumn> to SQL <sqlColumn> because schema is incompatible (protobufType = <protobufType>, sqlType = <sqlType>)."
+      "Found recursive reference in Protobuf schema, which can not be processed by Spark: <fieldDescriptor>"
     ]
   },
   "RENAME_SRC_PATH_NOT_FOUND" : {
@@ -850,9 +855,9 @@
     ],
     "sqlState" : "22023"
   },
-  "SQL_TYPE_TO_PROTOBUF_TYPE_ERROR" : {
+  "SQL_TYPE_TO_PROTOBUF_ENUM_TYPE_ERROR" : {
     "message" : [
-      "Cannot convert SQL <sqlColumn> to Protobuf <protobufColumn> because schema is incompatible (protobufType = <protobufType>, sqlType = <sqlType>)."
+      "Cannot convert SQL <sqlColumn> to Protobuf <protobufColumn> because <data> cannot be written since it's not defined in ENUM <enumString>"
     ]
   },
   "TABLE_OR_VIEW_ALREADY_EXISTS" : {
@@ -887,16 +892,11 @@
       "Unable to acquire <requestedBytes> bytes of memory, got <receivedBytes>"
     ]
   },
-  "UNABLE_TO_CONVERT_TO_PROTOBUF_TYPE" : {
+  "UNABLE_TO_CONVERT_TO_PROTOBUF_MESSAGE_TYPE" : {
     "message" : [
       "Unable to convert SQL type <toType> to Protobuf type <protobufType>."
     ]
   },
-  "UNABLE_TO_LOCATE_PROTOBUF_MESSAGE_ERROR" : {
-    "message" : [
-      "Unable to locate Message <messageName> in Descriptor"
-    ]
-  },
   "UNKNOWN_PROTOBUF_MESSAGE_TYPE" : {
     "message" : [
       "Attempting to treat <descriptorName> as a Message, but it was <containingType>"