You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Hemanth Umashankar (Jira)" <ji...@apache.org> on 2026/02/17 09:36:00 UTC

[jira] [Assigned] (HIVE-29464) Rethink MapWork.aliasToPartnInfo

     [ https://issues.apache.org/jira/browse/HIVE-29464?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hemanth Umashankar reassigned HIVE-29464:
-----------------------------------------

    Assignee: Hemanth Umashankar

> Rethink MapWork.aliasToPartnInfo
> --------------------------------
>
>                 Key: HIVE-29464
>                 URL: https://issues.apache.org/jira/browse/HIVE-29464
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: László Bodor
>            Assignee: Hemanth Umashankar
>            Priority: Major
>
> 1. getter for a mutable collection is extensively used:
> {code}
> public Map<String, PartitionDesc> getAliasToPartnInfo()
> { return aliasToPartnInfo; }
> {code}
> so we don't have an idea when partitions are actually added to this collection
> 2. setter is not used at all
> {code}
> public void setAliasToPartnInfo(
> LinkedHashMap<String, PartitionDesc> aliasToPartnInfo)
> {     this.aliasToPartnInfo = aliasToPartnInfo; }
> {code}
> 3. some callers of this method only relies to TableDesc objects, which is usually the one and only TableDesc for thousands of PartitionDesc objects, shall we consider introducing another method for retrieving TableDesc
> e.g. [KafkaDagCredentialSupplier|https://github.com/apache/hive/blob/399200af7cb11cf6ee3329ebdabe17792e5e7e85/kafka-handler/src/java/org/apache/hadoop/hive/kafka/KafkaDagCredentialSupplier.java#L62-L69]
> 4. TODO: check if it is possible to have multiple TableDesc for the same MapWork, and design the TableDesc getter accordingly



--
This message was sent by Atlassian Jira
(v8.20.10#820010)