Github user bzz commented on the issue:

    https://github.com/apache/zeppelin/pull/928
  
    @rawkintrevo thank you for addressing the feedback promptly.
    
    I tried examples and everything worked very well!
    
    One thing though - on the second run of `add_mahout.py` I got
    
    ```
    $python scripts/mahout/add_mahout.py
    --zeppelin_home not specified, using .../zeppelin
    ZEPPELIN_HOME validated
    apache-mahout-distribution-0.12.2.tar.gz found, skipping download
    created new interpreter 'sparkMahout' from interpreter 'spark
    created new interpreter 'flinkMahout' from interpreter 'flink
    updating 'sparkMahout' with Apache Mahout dependencies and settings
    updating 'flinkMahout' with Apache Mahout dependencies and settings
    'export MAHOUT_HOME=...' already exists in zeppelin_env.sh, not appending
    restarting Apache Zeppelin to load new interpreters...
    Zeppelin is not running
    Zeppelin start                                             [  OK  ]
    
---------------------------------------------------------------------------------------------------------------
    all done! Thanks for using Apache Mahout
    bye
    ```
    
    It didn't add more interpreters indeed but, do you think in such case, from 
user experience perspective it could be better just to stop the execution and 
notify user in the CLI that he already has it configured, nothing was touched 
and skip starting Zeppelin, etc? 
    
    One last thing, shall we may be rename script to 
`add_mahout_interpreter.py` to make its purpos more obvious?
    
    One possible further improvement (in the further PRs) could be - 
auto-import things from `org.apache.mahout` and define 
`SparkDistributedContext` so user do not have to do that every time. But right 
now Zeppelin lacks the means to do so, without writing an interpreter impl.
    
    Looks awesome to me, let's merge after final updates, if there is no 
further discussion!


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

Reply via email to