[ 
https://issues.apache.org/jira/browse/HIVE-5204?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13757301#comment-13757301
 ] 

Edward Capriolo commented on HIVE-5204:
---------------------------------------

This seems odd. Are these constants used anywhere else? I feel like other 
serde's might reference these directly.
{quote}
-    registerNumericType(serdeConstants.TINYINT_TYPE_NAME, 1);
-    registerNumericType(serdeConstants.SMALLINT_TYPE_NAME, 2);
-    registerNumericType(serdeConstants.INT_TYPE_NAME, 3);
-    registerNumericType(serdeConstants.BIGINT_TYPE_NAME, 4);
-    registerNumericType(serdeConstants.FLOAT_TYPE_NAME, 5);
-    registerNumericType(serdeConstants.DOUBLE_TYPE_NAME, 6);
-    registerNumericType(serdeConstants.DECIMAL_TYPE_NAME, 7);
-    registerNumericType(serdeConstants.STRING_TYPE_NAME, 8);
+    registerNumericType(PrimitiveCategory.BYTE, 1);
+    registerNumericType(PrimitiveCategory.SHORT, 2);
+    registerNumericType(PrimitiveCategory.INT, 3);
+    registerNumericType(PrimitiveCategory.LONG, 4);
+    registerNumericType(PrimitiveCategory.FLOAT, 5);
+    registerNumericType(PrimitiveCategory.DOUBLE, 6);
+    registerNumericType(PrimitiveCategory.DECIMAL, 7);
+    registerNumericType(PrimitiveCategory.STRING, 8);
{quote}
                
> Change type compatibility methods to use PrimitiveCategory rather than 
> TypeInfo
> -------------------------------------------------------------------------------
>
>                 Key: HIVE-5204
>                 URL: https://issues.apache.org/jira/browse/HIVE-5204
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Jason Dere
>            Assignee: Jason Dere
>         Attachments: HIVE-5204.1.patch
>
>
> The type compatibility methods in the FunctionRegistry (getCommonClass, 
> implicitConvertable) compare TypeInfo objects directly when its doing its 
> type compatibility logic. This won't work as well with qualified types 
> (varchar, char, decimal), because we will need different TypeInfo objects to 
> represent varchar(5) and varchar(10), and the equality comparisons won't work 
> anymore. We can change this logic to look at the PrimitiveCategory for the 
> TypeInfo instead.
> NO PRECOMMIT TESTS - dependent on changes in HIVE-5203

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to