Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
-
None
-
None
Description
After https://github.com/apache/zeppelin/pull/1360 merged, Zeppelin doesn't support "mapping from %jdbc(prefix) to %prefix" after 0.7.0 version. Instead, we can use %interpreter_alias as a JDBC interpreter prefix. You can find this information in http://zeppelin.apache.org/docs/0.7.0-SNAPSHOT/install/upgrade.html#upgrading-from-zeppelin-06-to-07.
Actually %jdbc(prefix) itself works at the moment as http://zeppelin.apache.org/docs/0.7.0-SNAPSHOT/interpreter/jdbc.html#multiple-connections describes. But auto-prefix for %jdbc(prefix) doesn't work. After running the paragraph with %jdbc(prefix, newly appended paragraph is not including %jdbc(prefix). It’s quite uncomfortable to type %jdbc(prefix) to every paragraphs’ header. To avoid this inconvenience, we need to use %interpreter_alias for JDBC interpreter prefix. Please see the attached gif image files.
For example, if a user wants to use %hive, he needs to create a new JDBC interpreter(the interpreter name should be “hive” not “jdbc") then set default.driver with hive driver. Currently there is no guide for this in http://zeppelin.apache.org/docs/0.7.0-SNAPSHOT/interpreter/jdbc.html page. Moreover the user wants to connect multiple jdbc drivers(let’s say he needs to connect “mysql” at the same time), he needs to create another JDBC interpreter for Mysql connection. Then he can use %mysql.
So http://zeppelin.apache.org/docs/0.7.0-SNAPSHOT/interpreter/jdbc.html#multiple-connections this page needs to be updated as well.
Attachments
Attachments
Issue Links
- links to