Pentaho

 View Only

 How to populate fields after Transformation Executor automatically?

  • Pentaho
  • Pentaho
Gert Wieland's profile image
Gert Wieland posted 01-09-2018 05:23

I'm trying to use a sub (child) transformation, and then continue to edit the data stream in the parent transformation. Everything works, except I'm not able to pull the resulting columns automatically into any subsequent operator after the Transformation Exectutor (TE). Here's where I'm at:

transformation

The child transformation simply pulls some data from a database. The parent transformation calls the child transformation and runs it for several databases (configured in the Data Grid operator). On top of the screenshot I put the configuration of the TE. When I rightclick and "preview" the TE, I get to see correct results. If I add columns in the child transformation, they show properly.

My problem is that I can't seem to get any operator (for example a simple select operator) to populate with the fields provided by the child operator. I read the TE wiki, but I've not been able to to figure it out. My goal is to simply add a "select" operator which populates the field list based on the fields from the child transformation (when clicking "Get fields to select").

I'd really appreciate some help, I've tried "everything" (except the right thing apparently )


#Pentaho
Michael Ratner's profile image
Michael Ratner posted 01-09-2018 06:15

Hi Gert,

Thank you for your question. It seems to be related to Pentaho, and we have a dedicate space for all Pentaho-related questions and other posts. The space you posted in is just for general feedback and questions on how to use the Community, etc., as could be seen here:

about

Your question simply won't get the audience it deserves, since members receive notifications only for spaces they "follow".

I will move your post now, but below is how you can find the Pentaho area to post your questions:

pentaho

Thanks,

-Michael

Gert Wieland's profile image
Gert Wieland posted 01-09-2018 14:26

Hi Michael,

Thank you very much for that - and sorry for the wrong posting, I'll make sure to post future questions in the appropriate area.

Much appreciated!

Gert

Data Conversion's profile image
Data Conversion posted 01-09-2018 16:58

Gert,

I don't see any way to automatically do what you are talking about...  But that sounds like a great Feature Request (http://jira.pentaho.com )

One way to do what you're talking about quickly:

  1. Open the transformation that you're going to execute.
  2. Right Click on the final step in the transformation
  3. Select Output Fields
  4. Click in the Row Grid
  5. Press Ctrl-A (Select All)
  6. Press Ctrl-C (Copy)
  7. Return to your Transformation Executor configuration (Result Rows Step)
  8. Click on the number 1 (in the row grid, but not in any cell) - this should highlight the entire row
  9. Press Ctrl-V (Paste)

NOTE: You won't need a Get Rows from Results step from my reading of the documentation.

Gert Wieland's profile image
Gert Wieland posted 01-09-2018 18:38

Hi Gotn O'Guts,

Thank you very much for your suggestion - and particularly for confirming I didn't miss anything, but that indeed it's not working the way I thought it would. Your suggestion is a helpful workaround since it avoids re-typing the field names.

Nevertheless, I thought this would be a solution to break down complex transformations into blocks/ sections, and combine them within a meta-transformation. Since that's not the case, I changed my design so the final step in the child transformation writes the results into a database so I don't need to read the fields in subsequent operators.

I'll still put in the feature request (thanks for the link) because I think it could be helpful.

Thanks again for your feedback, much appreciated,

Gert

Related Content