drill-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Victoria Markman (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (DRILL-2621) When view is created on top of parquet file, underlying data types should be automatically converted to SQL type
Date Mon, 30 Mar 2015 16:01:53 GMT

     [ https://issues.apache.org/jira/browse/DRILL-2621?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Victoria Markman updated DRILL-2621:
------------------------------------
    Description: 
Since parquet types are known to us, we need to make sure to expose these types without making
user explicitly cast each column.

"create view v1 as select c_integer from j1; " produces view with column of ANY data type.

{code}
0: jdbc:drill:schema=dfs> describe v1;
+-------------+------------+-------------+
| COLUMN_NAME | DATA_TYPE  | IS_NULLABLE |
+-------------+------------+-------------+
| c_integer   | ANY        | YES         |
+-------------+------------+-------------+
1 row selected (0.091 seconds)
{code}

I think we need to extend "CREATE VIEW" document for GA explaining this fact. 

  was:
Since parquet types are known to us, we need to make sure to expose these types without making
user explicitly cast each column.

"create view v1 as select c_integer from j1; " produces view with column of ANY data type.

{code}
0: jdbc:drill:schema=dfs> describe v1;
+-------------+------------+-------------+
| COLUMN_NAME | DATA_TYPE  | IS_NULLABLE |
+-------------+------------+-------------+
| c_integer   | ANY        | YES         |
+-------------+------------+-------------+
1 row selected (0.091 seconds)
{code}

I think we need to expend "CREATE VIEW" document for GA explaining this fact. 


> When view is created on top of parquet file, underlying data types should be automatically
converted to SQL type
> ----------------------------------------------------------------------------------------------------------------
>
>                 Key: DRILL-2621
>                 URL: https://issues.apache.org/jira/browse/DRILL-2621
>             Project: Apache Drill
>          Issue Type: New Feature
>            Reporter: Victoria Markman
>
> Since parquet types are known to us, we need to make sure to expose these types without
making user explicitly cast each column.
> "create view v1 as select c_integer from j1; " produces view with column of ANY data
type.
> {code}
> 0: jdbc:drill:schema=dfs> describe v1;
> +-------------+------------+-------------+
> | COLUMN_NAME | DATA_TYPE  | IS_NULLABLE |
> +-------------+------------+-------------+
> | c_integer   | ANY        | YES         |
> +-------------+------------+-------------+
> 1 row selected (0.091 seconds)
> {code}
> I think we need to extend "CREATE VIEW" document for GA explaining this fact. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message