Skip to content

GEMINI.md

Latest commit

 

History

History
78 lines (64 loc) · 3.84 KB

GEMINI.md

File metadata and controls

78 lines (64 loc) · 3.84 KB

Project: Playwright Java Test Automation Architecture

Context & Tech Stack

  • Build Tool: Gradle (using build.gradle)
  • Language: Java 17+
  • Browser Automation Library: Playwright
  • Architecture: Data-driven test automation using POM to model and reuse web page functionality.
  • Base Package: io.github.tahanima
  • Test Runner: JUnit 5 with custom annotations (@TestDataSource, @Validation, @Smoke, @Regression).

Project Mapping

  • Pages: src/main/java/io/github/tahanima/ui/page/ – Encapsulate low-level page actions and locators.
  • Components: src/main/java/io/github/tahanima/ui/component/ – Reusable UI elements (Header, SideNav, etc.).
  • TestData POJOs: src/main/java/io/github/tahanima/testdata/ – Business logic layer that interprets CSV data and interacts with Pages.
  • Tests: src/test/java/io/github/tahanima/e2e/ – High-level orchestration; must extend BaseTest.
  • Test Data Files: src/test/resources/testdata/ – CSV files for data-driven inputs.
  • Utilities:
    • BrowserManager.java – Browser lifecycle management.
    • CsvLoader.java – CSV loading for test data.
  • Factories:
    • BasePageFactory.java – Centralized page initialization.
    • BrowserFactory.java – Abstracts browser setup and creation.
  • Reporting: AllureManager.java – Custom logging, screenshots, and test reporting.

Development Rules

  1. Instantiation: Do not use new LoginPage(). Always use the BasePageFactory to initialize pages.
  2. Logic Placement:
    • Pages/Components: Handle locators and atomic actions (e.g., typeUsername).
    • TestData POJOs: Handle business workflows (e.g., loginWith(LoginTestData data)).
    • Tests: Handle assertions and test flow. Minimal direct Playwright API calls here.
  3. Stability & Waits:
    • No Hard Sleeps: Never use Thread.sleep(). Rely on Playwright’s auto-waiting or page.waitForCondition().
    • Web-First Assertions: Use PlaywrightAssertions.assertThat() for UI states to enable automatic retries.
  4. Configuration: Use ConfigurationManager to access config.properties. Do not hardcode URLs or credentials.
  5. Data Loading: Use the custom @TestDataSource annotation for CSV data-driven tests.
  6. Naming Conventions:
    • Pages → *Page.java
    • Components → *Component.java
    • Test Data POJOs → *TestData.java (must extend BaseTestData).

Locator Strategy (Priority Order)

  1. page.getByRole() (Prioritize accessibility-first selectors).
  2. page.getByLabel() / page.getByPlaceholder().
  3. page.getByTestId() (Use when data-testid attributes are available).
  4. CSS/XPath (Use only as a last resort for complex structural selectors).

Test Data Flow

  • CSV files → TestDataArgumentsProvider → Test method parameters → Test execution.
  • POJOs (testdata package) represent structured test data for readability and business logic reuse.

Running Tests

  • Run all tests: ./gradlew clean test
  • Run a specific test class: ./gradlew test --tests *LoginTest
  • Generate Allure report: ./gradlew allureServe

Available Skills (AI Toolbelt)

  1. code-reviewer:
    • Trigger: When providing new code or asking for architectural feedback.
    • Focus: Enforcing layer separation (Page vs. TestData vs. E2E) and checking Factory usage.
  2. healing-agent:
    • Trigger: When reporting a TimeoutError, broken locator, or execution failure.
    • Focus: Converting brittle selectors to the "Locator Strategy" priority list above.

Usage Guidelines

  • Skill Activation: If a task matches the triggers above, ask for permission to load the skill using /skills load <name>.
  • Contextual Awareness: Before proposing code changes, refer to src/main/resources/config.properties for environment-specific settings.