[
https://issues.apache.org/jira/browse/HIVE-5355?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13807058#comment-13807058
]
Jason Dere commented on HIVE-5355:
----------------------------------
Hey Xuefu, for these changes:
{noformat}
case Types.VARCHAR:
- if (columnAttributes != null) {
- return columnAttributes.precision;
- }
- return Integer.MAX_VALUE; // hive has no max limit for strings
+ return columnPrecision(columnType, columnAttributes);
{noformat}
String columns end up getting translated to JDBC varchar type. So changing
this would result in a NPE if the user tries to call getColumnDisplaySize() for
string columns, because I think the columnAttributes are null for non-qualified
Hive types.
Not sure if you have to worry about a similar issue for decimal if you are
connecting to a pre-HIVE-3976 Hive server, are newer clients meant to be
backward compatible with older servers?
> JDBC support for decimal precision/scale
> ----------------------------------------
>
> Key: HIVE-5355
> URL: https://issues.apache.org/jira/browse/HIVE-5355
> Project: Hive
> Issue Type: Task
> Components: Serializers/Deserializers
> Affects Versions: 0.10.0
> Reporter: Xuefu Zhang
> Assignee: Xuefu Zhang
> Fix For: 0.13.0
>
> Attachments: HIVE-5355.1.patch, HIVE-5355.2.patch, HIVE-5355.patch
>
>
> A subtask of HIVE-3976.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
