You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@airflow.apache.org by "Mario Measic (Jira)" <ji...@apache.org> on 2019/11/04 14:24:00 UTC

[jira] [Updated] (AIRFLOW-5847) SqoopHook import methods improvement

     [ https://issues.apache.org/jira/browse/AIRFLOW-5847?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Mario Measic updated AIRFLOW-5847:
----------------------------------
    Description: 
From the current definition of the import_query method in SqoopHook class, we are unable to specify JDBC URI (through a connect option).

The connect_str is being built through a _prepare_command([https://github.com/apache/airflow/blob/master/airflow/contrib/hooks/sqoop_hook.py#L118]):

In the case of Oracle and JDBC Oracle driver, one should provide a service name or a SID in the following format: _jdbc:oracle:thin:@//<host>:<port>/ServiceName_ 

The current implementation provides a schema name in lieu of service name or SID. Additionally, I would love to understand how does Sqoop import handles, in this case, a connection string that is of format: _host:port/schema_

Writing this as a placeholder since there can be a plenty of work here, depending on the hook type (source).

  was:
From the current definition of the import_query method in SqoopHook class, we are unable to specify JDBC URI (through a connect option).

The connect_str is being built through a _prepare_command([https://github.com/apache/airflow/blob/master/airflow/contrib/hooks/sqoop_hook.py#L118]):

In the case of Oracle and JDBC Oracle driver, one should provide a service name or a SID in the following format: _jdbc:oracle:thin:@//<host>:<port>/ServiceName_ 

The current implementation provides a schema name in lieu of service name or SID. Additionally, I would love to understand how does Sqoop import handles, in this case, a connection string that is of format: _host:port/schema_

Writing this as a placeholder since there can be a plenty of work here, depending on the hook type (source).

 


> SqoopHook import methods improvement
> ------------------------------------
>
>                 Key: AIRFLOW-5847
>                 URL: https://issues.apache.org/jira/browse/AIRFLOW-5847
>             Project: Apache Airflow
>          Issue Type: Improvement
>          Components: contrib, hooks
>    Affects Versions: 1.10.6
>            Reporter: Mario Measic
>            Assignee: Mario Measic
>            Priority: Major
>
> From the current definition of the import_query method in SqoopHook class, we are unable to specify JDBC URI (through a connect option).
> The connect_str is being built through a _prepare_command([https://github.com/apache/airflow/blob/master/airflow/contrib/hooks/sqoop_hook.py#L118]):
> In the case of Oracle and JDBC Oracle driver, one should provide a service name or a SID in the following format: _jdbc:oracle:thin:@//<host>:<port>/ServiceName_ 
> The current implementation provides a schema name in lieu of service name or SID. Additionally, I would love to understand how does Sqoop import handles, in this case, a connection string that is of format: _host:port/schema_
> Writing this as a placeholder since there can be a plenty of work here, depending on the hook type (source).



--
This message was sent by Atlassian Jira
(v8.3.4#803005)