In Turnkey, we've reimagined the concept of "tests" from the DG desktop client into two distinct parts: datasets and rules. This split offers several advantages. First, it allows for more control over who creates datasets versus rules, empowering admins to manage datasets while enabling business users to create rules. Secondly, it facilitates reusability, as the same dataset can be utilized across multiple rules, saving time and streamlining workflows.
When you create a dataset, you're essentially defining the data you'll later use for your rules. The process begins by selecting a data provider, which establishes the connection to your actual data, such as a database. This ensures that the software knows where to fetch the data from when you query it. The next step involves providing an SQL statement to specify which tables or data you want to use for your rules.
Creating a data provider and dataset are crucial steps before crafting rules to track exceptions or anomalies. For instance, consider a sales dataset containing information about customer transactions. Before creating a rule to ensure that only valid sales entries are imported into our systems, it's essential to first identify the dataset that will be used to detect exceptions.
By simplifying and separating the dataset creation process from rule creation, we aim to enhance user control and efficiency in managing data and implementing rules to ensure data quality.
