Dealing with Negative numbers that use ()

Options
tgb417
tgb417 Dataiku DSS Core Designer, Dataiku DSS & SQL, Dataiku DSS ML Practitioner, Dataiku DSS Core Concepts, Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Frontrunner 2022 Finalist, Frontrunner 2022 Winner, Dataiku Frontrunner Awards 2021 Participant, Frontrunner 2022 Participant, Neuron 2023 Posts: 1,595 Neuron

I have a dataset coming in with negative numbers being represented in an accounting fashion with leading and trailing parentheses. For example (200) equals -200.

Anyone have a quick way to clean that up in a visual recipie?

The Extract Number Processor does not resolve this issue.

I'm using DSS 8.0.4.

Best Answer

  • Ignacio_Toledo
    Ignacio_Toledo Dataiku DSS Core Designer, Dataiku DSS Core Concepts, Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Frontrunner 2022 Finalist, Frontrunner 2022 Winner, Dataiku Frontrunner Awards 2021 Participant, Frontrunner 2022 Participant, Neuron 2023 Posts: 411 Neuron
    Answer ✓
    Options

    Hi @tgb417

    I just tried this approach, with a find and replace step, and it does work for me:

    idea.png

    Hope it helps!

Answers

  • tgb417
    tgb417 Dataiku DSS Core Designer, Dataiku DSS & SQL, Dataiku DSS ML Practitioner, Dataiku DSS Core Concepts, Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Frontrunner 2022 Finalist, Frontrunner 2022 Winner, Dataiku Frontrunner Awards 2021 Participant, Frontrunner 2022 Participant, Neuron 2023 Posts: 1,595 Neuron
    edited July 17
    Options

    I ended up writing this complicated formula.

    if(contains(strval("Sum of Total Received"), '('), 
    concat("-",substring(strval("Sum of Total Received"),1,length(strval("Sum of Total Received"))-1)),
    numval("Sum of Total Received"))

    However, that is sort of a pain. Anyone have a faster way to deal with this challange.

  • tgb417
    tgb417 Dataiku DSS Core Designer, Dataiku DSS & SQL, Dataiku DSS ML Practitioner, Dataiku DSS Core Concepts, Neuron 2020, Neuron, Registered, Dataiku Frontrunner Awards 2021 Finalist, Neuron 2021, Neuron 2022, Frontrunner 2022 Finalist, Frontrunner 2022 Winner, Dataiku Frontrunner Awards 2021 Participant, Frontrunner 2022 Participant, Neuron 2023 Posts: 1,595 Neuron
    Options

    @Ignacio_Toledo
    , This is good, you can do this in just one recipe step. My approach took 4 steps per column to get the names of columns back the way they were using my method.

    And even better you can do this on multiple columns with the same step.

    Bravo!

    I've tested and can reproduce your approach.

  • CoreyS
    CoreyS Dataiker Alumni, Dataiku DSS Core Designer, Dataiku DSS Core Concepts, Registered Posts: 1,150 ✭✭✭✭✭✭✭✭✭
    Options

    Based off of this wonderful thread we have published a Knowledge Base article: Dealing with Accounting-style negative numbers

    Thank you both for your help!

Setup Info
    Tags
      Help me…