Skip to main content

Variables in Robot Framework

Variables in Robot Framework help make tests dynamic, reusable, and maintainable. They allow you to avoid hardcoding values and manage test data efficiently.

Why Variables Are Important​

Without variables:

  • Tests become hard to maintain
  • Data changes require code changes
  • Reuse becomes difficult

With variables:

  • Tests are flexible
  • Configuration is centralized
  • CI/CD execution is easier

Types of Variables​

Robot Framework supports three main variable types.

1. Scalar Variables​

  • Hold a single value
  • Syntax: ${variable}
${URL}    https://example.com
${BROWSER}    chrome

2. List Variables​

  • Hold multiple values
  • Syntax: @{list}
@{CREDENTIALS}    admin    secret

Accessing list values:

${CREDENTIALS}[0]

3. Dictionary Variables​

  • Store key-value pairs
  • Syntax: &{dict}
&{USER}    username=admin    password=secret

Accessing dictionary values:

${USER}[username]

Variable Scope​

Robot Framework variables have different scopes.

Test Scope​

  • Available only within a single test case

Suite Scope​

  • Available across all tests in a suite

Global Scope​

  • Available across all suites

Choosing correct scope avoids data leakage issues.

Variable Files​

Variables can be stored in:

  • .robot variable files
  • Python files
  • YAML / JSON files

Used for:

  • Environment data
  • Credentials
  • URLs
  • Config values

Overriding Variables at Runtime​

Variables can be overridden during execution:

robot --variable BROWSER:chrome tests/

This is commonly used in CI/CD pipelines.

Common Mistakes ❌​

  • Hardcoding values
  • Using global variables unnecessarily
  • Mixing test data with logic

Best Practices ✅​

  • Externalize environment-specific data
  • Use dictionaries for structured data
  • Limit global variable usage
  • Use meaningful variable names

Key Takeaways​

  • Variables make tests flexible
  • Three types: scalar, list, dictionary
  • Scope control is critical
  • Variable files improve maintainability