Skip to content

Dataset

The Dataset is the concept responsible for defining the test data used during scenario execution in Probato. It allows the same Script to be executed multiple times with different data sets, in a native and declarative way.

In Probato’s mental model, the Dataset exists to separate data from execution logic, avoiding complex conditionals and promoting reuse.

The Role of the Dataset in Probato

The Dataset is responsible for:

  • Providing external data for Script execution
  • Enabling automatic data-driven execution
  • Avoiding data-based conditional logic
  • Keeping test code simple, readable, and reusable

The Dataset answers the question: With which data will the scenario be executed?

Where the Dataset Fits in the Mental Model

In Probato’s conceptual flow, the Dataset is always associated with a Script.

Conceptual model
Suite
 ├── SQL (global state / feature preconditions)
 ├── NoSQL (global state / feature preconditions)
 └── Script
      ├── Dataset (execution data)
      ├── SQL (scenario-specific state)
      ├── NoSQL (scenario-specific state)
      ├── Precondition
      │     └── Page Object
      │           ├── Action
      │           └── Param
      ├── Procedure
      │     └── Page Object
      │           ├── Action
      │           └── Param
      └── Postcondition
            └── Page Object
                  ├── Action
                  └── Param

The Dataset is never associated directly with a Procedure or a Page Object.

Dataset Characteristics

In Probato, a Dataset has the following characteristics:

  • External to the test code
  • Strongly typed
  • Resolved before Procedure execution
  • Independent of test logic

Each Dataset entry generates an independent execution of the Script, ensuring isolation and predictability.

Data Models

The data defined in a Dataset is mapped to data models.

These models:

  • Represent the Dataset structure
  • Are automatically injected into the Procedure
  • Ensure type safety and execution clarity

The Procedure receives only the resolved data model, without knowing the data’s origin or physical format.

Benefits of Using Datasets

Proper use of Datasets enables:

  • Greater test coverage without code duplication
  • Simpler, more declarative scenarios
  • Clear separation between data and behavior
  • Easier test maintenance and evolution

What Should NOT Be in a Dataset

To maintain separation of responsibilities, a Dataset must not:

  • Contain execution logic
  • Define business rules
  • Change application state
  • Depend on execution context

A Dataset must be only a data source.

Best Practices

  • Keep Datasets small and focused
  • Create one Dataset per scenario type
  • Avoid overly generic Datasets
  • Name data models clearly and semantically

Next Step

After understanding the Dataset, the next concept is Database, responsible for defining the application state before scenario execution.

➡️ Continue to Database.