[jira] [Work logged] (HIVE-26556) Iceberg: Properties set in HiveIcebergSerde are not propagated to jobconf

["ASF GitHub Bot (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-26556?focusedWorklogId=814241&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-814241 ]

ASF GitHub Bot logged work on HIVE-26556:
-----------------------------------------

                Author: ASF GitHub Bot
            Created on: 06/Oct/22 06:40
            Start Date: 06/Oct/22 06:40
    Worklog Time Spent: 10m 
      Work Description: ayushtkn commented on code in PR #3617:
URL: https://github.com/apache/hive/pull/3617#discussion_r988605068


##########
iceberg/iceberg-handler/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergSerDe.java:
##########
@@ -130,13 +131,14 @@ public void initialize(@Nullable Configuration configuration, Properties serDePr
       }
     }
 
-    this.projectedSchema = projectedSchema(configuration, serDeProperties.getProperty(Catalogs.NAME), tableSchema);
+    this.projectedSchema =
+        projectedSchema(configuration, serDeProperties.getProperty(Catalogs.NAME), tableSchema, jobConfs);
 
     // Currently ClusteredWriter is used which requires that records are ordered by partition keys.
     // Here we ensure that SortedDynPartitionOptimizer will kick in and do the sorting.
     // TODO: remove once we have both Fanout and ClusteredWriter available: HIVE-25948
-    HiveConf.setIntVar(configuration, HiveConf.ConfVars.HIVEOPTSORTDYNAMICPARTITIONTHRESHOLD, 1);
-    HiveConf.setVar(configuration, HiveConf.ConfVars.DYNAMICPARTITIONINGMODE, "nonstrict");
+    jobConfs.put(HiveConf.ConfVars.HIVEOPTSORTDYNAMICPARTITIONTHRESHOLD.varname, "1");
+    jobConfs.put(HiveConf.ConfVars.DYNAMICPARTITIONINGMODE.varname, "nonstrict");

Review Comment:
   I don't think there is any problem there, the caller isn't explicitly passing, it is {{SessionState.getSessionConf()}}, and ideally the values which we are setting aren't of any use to session & all other SerDe use it as a Read-Only configuration.
   And moreover I don't think we should change the session conf for no reason here, not sure if it can even bother further queries or so... 





Issue Time Tracking
-------------------

    Worklog Id:     (was: 814241)
    Time Spent: 40m  (was: 0.5h)

> Iceberg: Properties set in HiveIcebergSerde are not propagated to jobconf
> -------------------------------------------------------------------------
>
>                 Key: HIVE-26556
>                 URL: https://issues.apache.org/jira/browse/HIVE-26556
>             Project: Hive
>          Issue Type: Bug
>            Reporter: Ayush Saxena
>            Assignee: Ayush Saxena
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> Some hive properties (ex. InputFormatConfig.CASE_SENSITIVE) are not propagated to the jobconf. This scenario can be reproduced by running TestHiveIcebergSelects#testScanTableCaseInsensitive test method. 



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