0.8.1

Variable Overview

Overview

Ananas Analytics Desktop provides a variable system to help you parameterize your analysis.

For example, in our first data flow guide, we analyzed the top 10 countries with most footplayers with a simple SQL transform:

SELECT 
    count(1) AS players,
    Nationality
FROM PCOLLECTION
GROUP BY Nationality
ORDER BY players DESC
LIMIT 10

What if we want to analyze top 20 countries?

One solution would be to modify the SQL from LIMIT 10 to LIMIT 20, and rerun the analysis. But when you have complex data flow with multiple steps, finding the right step to modify turns out to be a tricky task. What if you want to change such parameter multiple times per day? isn't it much easier to put all the parameters in a single place?

Ananas Analytisc Desktop Variable tab helps you to create new variables that you can use in your data analysis flow.

variable tab

Add a new variable

In variable tab (left sidebar), click + button leads you to a new variable editor.

new variable

By applying a unique variable name (please avoid white space in the variable), and choose the variable type.

There are 4 types variables:

  • String
  • Number
  • Date
  • Credential

You can also provide an optional description in the editor.

Use variable in your step settings

To reference a variable in your step settings, you can use the following format:

${variable_name}

For example, in our Fifaplayer example, we can replace LIMIT 10 to LIMIT ${nationality_num}

SELECT 
    count(1) AS players,
    Nationality
FROM PCOLLECTION
GROUP BY Nationality
ORDER BY players DESC
LIMIT ${nationality_num}

On the right sidebar -> variable tab, you can review current available variables. Above Test and RUN button, there would be a variable field ready for you to fill the parameter value.

apply variable in step

Now you can use the destination and visualization steps as an ad-hoc query interface.

ad hoc query with variable

Predefine Variables

There are several predefined variables

EXECUTE_DATE

Type date. The timestamp, when your analysis job start to execute.

PROJECT_PATH

Type String. The absolute project path. This variable is useful when you want to reference resources in your project folder.

Variable Modifier

The simplest use case is to reference your variable directly in the form ${variable_name}. You can also apply modifiers to your variables. For example, you can format a date type variable to year-month-day format with following:

${date_var?string['YYYY-MM-dd']}

More details on how to use modifiers see FreeMarker Built-in reference

Tips

Variables can be used not only in step settings but also Execution Engine settings

← Google DataflowOverview →