How to use datasets with duplicate column names in a dataflow

user022825
user022825 Member
in Magic ETL

Using a JIRA connector, I import a dataset, which has a 1000 columns. There are two columns which have duplicate names. Consequently, I cannot use the dataset in a Domo SQL transform as it gives me a duplicate column error. I also cannot write each column name individually as there are 1000 columns (and I need 999). 

 

Is there a way to use my dataset in a transform, while catering to the duplicate column issue?

 

One solution could be selecting all columns in my table except one. Is there a way to do that?

Best Answer

  • jaeW_at_Onyx
    jaeW_at_Onyx Coach
    Answer ✓

    Is this a Domo connector or one that you wrote?  

     

    I'm surprised that Domo didn't auto_adjust the duplicate column name OR fail the ingest.

     

    If it's a connector you wrote, consider reshaping the data to have more rows and fewer columns.  That's going to be a real pain to build analysis against.

     

    If i had to guess you've got something like 'one row per project or ticket' and then flattened the data such that all the attribute_values are going across in columns but very few of the columns are actually populated as you scan through the rows. 

     

    Also, 1000 is a REALLY convenient number for a connector ... are you sure it's not accidentally truncating data because it ran into a limit?

     

    Short answer I would take a closer look at the connector and see if i can't reshape the data before bringing it into Domo.

     

     

    Jae Wilson
    Check out my 🎥 Domo Training YouTube Channel 👨‍💻

    **Say "Thanks" by clicking the ❤️ in the post that helped you.
    **Please mark the post that solves your problem by clicking on "Accept as Solution"

Answers

  • jaeW_at_Onyx
    jaeW_at_Onyx Coach
    Answer ✓

    Is this a Domo connector or one that you wrote?  

     

    I'm surprised that Domo didn't auto_adjust the duplicate column name OR fail the ingest.

     

    If it's a connector you wrote, consider reshaping the data to have more rows and fewer columns.  That's going to be a real pain to build analysis against.

     

    If i had to guess you've got something like 'one row per project or ticket' and then flattened the data such that all the attribute_values are going across in columns but very few of the columns are actually populated as you scan through the rows. 

     

    Also, 1000 is a REALLY convenient number for a connector ... are you sure it's not accidentally truncating data because it ran into a limit?

     

    Short answer I would take a closer look at the connector and see if i can't reshape the data before bringing it into Domo.

     

     

    Jae Wilson
    Check out my 🎥 Domo Training YouTube Channel 👨‍💻

    **Say "Thanks" by clicking the ❤️ in the post that helped you.
    **Please mark the post that solves your problem by clicking on "Accept as Solution"
  • user022825
    user022825 Member

    Hello,

    Thanks for the reply. I am using the JIRA Rest API Domo Connector and you are right, it is a one row per ticket dataset.

     

    I was also expecting the connector to auto adjust the column names. I wasn't needing the duplicated columns so I was able to include only the ones I need in the import using the Filter type = Include and Fields options.

    Thanks!

Categories