[
https://issues.apache.org/jira/browse/HIVE-15277?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15751777#comment-15751777
]
Jesus Camacho Rodriguez commented on HIVE-15277:
------------------------------------------------
[~bslim], some of the test failures (age=1) seem related to this patch. Could
you take a look?
Change in _LineageLogger_ causes changes in lineage golden files. It is better
to tackle that in a follow-up and remove it from this patch, as it is not part
of this issue.
> Teach Hive how to create/delete Druid segments
> -----------------------------------------------
>
> Key: HIVE-15277
> URL: https://issues.apache.org/jira/browse/HIVE-15277
> Project: Hive
> Issue Type: Sub-task
> Components: Druid integration
> Affects Versions: 2.2.0
> Reporter: slim bouguerra
> Assignee: slim bouguerra
> Attachments: HIVE-15277.2.patch, HIVE-15277.patch, HIVE-15277.patch,
> HIVE-15277.patch, HIVE-15277.patch, HIVE-15277.patch, HIVE-15277.patch,
> file.patch
>
>
> We want to extend the DruidStorageHandler to support CTAS queries.
> In this implementation Hive will generate druid segment files and insert the
> metadata to signal the handoff to druid.
> The syntax will be as follows:
> {code:sql}
> CREATE TABLE druid_table_1
> STORED BY 'org.apache.hadoop.hive.druid.DruidStorageHandler'
> TBLPROPERTIES ("druid.datasource" = "datasourcename")
> AS <select `timecolumn` as `___time`, `dimension1`,`dimension2`, `metric1`,
> `metric2`....>;
> {code}
> This statement stores the results of query <input_query> in a Druid
> datasource named 'datasourcename'. One of the columns of the query needs to
> be the time dimension, which is mandatory in Druid. In particular, we use the
> same convention that it is used for Druid: there needs to be a the column
> named '__time' in the result of the executed query, which will act as the
> time dimension column in Druid. Currently, the time column dimension needs to
> be a 'timestamp' type column.
> metrics can be of type long, double and float while dimensions are strings.
> Keep in mind that druid has a clear separation between dimensions and
> metrics, therefore if you have a column in hive that contains number and need
> to be presented as dimension use the cast operator to cast as string.
> This initial implementation interacts with Druid Meta data storage to
> add/remove the table in druid, user need to supply the meta data config as
> --hiveconf hive.druid.metadata.password=XXX --hiveconf
> hive.druid.metadata.username=druid --hiveconf
> hive.druid.metadata.uri=jdbc:mysql://host/druid
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
