Lookup or Joiner using Contains in the Match

Mahulima
Mahulima Contributor

Is it possible to do below check

I have this in one channel

Want to check and populate 'Issue Pending' column value from other channel where it has comma separated values

Answers

  • Josh Boxer
    Josh Boxer Administrator

    Sorry not sure what you are trying to achieve. If you can explain more clearly what does the data look like now vs what you want the data to look like at the end.

  • Mahulima
    Mahulima Contributor

    I have the Dataset 1 as

    Dataset 2 as

    I want to populate Issue Pending column from Dataset 2 based on this :

    if Rule Name column value is one of the value in Failed Rules column in Dataset 2 then populate the Issue pending column as return. In my final result I want Dataset 1 value along with Issue pending column value from Dataset 2

  • Josh Boxer
    Josh Boxer Administrator
    edited January 28

    Ok, there are possibly a few different ways to achieve this, but whenever you have a comma separated list you should think about using the List Functions. Turn Dataset1 (Rule Name) into a second comma separated list then compare both:

    https://docs.experianaperture.io/data-quality/hosted-aperture-data-studio/data-studio-objects/functions/#list-transformation~native-functions

    Lookup step is useful to turn Dataset1 into a comma separated list. Use Transform to add a new column containing a constant value say 'A', then a Lookup step that feeds in the same Transform step twice. Lookup type 'All values', Match type 'Exact', Lookup column 'constant value A'. Finally a Group step to dedupe newly created list column down to a single row.

  • Mahulima
    Mahulima Contributor

    Dataset 1 - if I convert that as comma separated that will not give me the actual result.

    suppose it has 3 different values in the column the comma separated will look like XX,YY,ZZ and it Dataset 2 it may be XX,AA or YY,BB or it may be just XX . it is dynamic and its not fixed value . and all 3 given scenario should match

  • Mahulima
    Mahulima Contributor

    i think it would be beneficial if i can make dataset 2 data in different rows which are comma separated

    also list interaction is only available with transformation step

  • Mahulima
    Mahulima Contributor

    @Josh Boxer

    ultimately I want Dataset 1 value along with the another column value from dataset 2

    but the solution you provided is not matching my need

    if I convert Dataset set in one single row it wont match the value in dataset 2 as I explained before

  • Josh Boxer
    Josh Boxer Administrator

    Ok then I still dont understand what you are trying to achieve since you never mentioned wanting other columns. Maybe someone else will be able to understand better.

  • Mahulima
    Mahulima Contributor

    @Josh Boxer

    Dataset 1 has

    Dataset 2

    I just want to now Populate "Issue Pending" Column from Dataset 2

    where "Failed Rules" in Dataset 1 has either exact value of the Rule name column in Dataset 1 or one of those is Present there. Also Issue Pending Column value is not expected same for all rows, its just a sample data thats why its same

  • Looks like you are trying to create a report that's a combination of the results from 'Validate' step and the contents of an Issue List. Correct?

  • Mahulima
    Mahulima Contributor

    @James T Sidebotham validate step out put as it shows in the SS of Dataset 1 but contents of list the 2nd dataset that is not exactly true, as the output of the issue list is further calculated and created like that. But yes overall true

  • OK, I am still not sure what the final report would look like, or what it would expose / solve problem. I am curious though. I use the outputs from 'Validate' step and 'Issue List' exports to present to clients, but never tried to merge them.