This directory contains unit and integration tests for the shell scripts in the embabel-learning workspace.

📖 Important: Read ARCHITECTURE.md before writing tests! It explains how to properly use the configuration system and avoid hardcoding values.

📖 Important: Read ARCHITECTURE.md before writing tests! It explains how to properly use the configuration system and avoid hardcoding values.

test/
├── README.md                    # This file
├── run-tests.sh                 # Main test runner
├── helpers/
│   └── test-framework.sh       # Simple test framework (self-contained)
├── unit/
│   ├── test-config-loader.sh   # Tests for config-loader.sh
│   ├── test-safety-checks.sh   # Tests for safety-checks.sh
│   └── test-sync-discord.sh   # Tests for discord-sync/sync-discord.sh
└── integration/
    └── (future integration tests)

We use a self-contained, lightweight test framework inspired by shunit2. No external dependencies required!

• assertTrue "message" command - Asserts command succeeds
• assertFalse "message" command - Asserts command fails
• assertEquals expected actual "message" - Asserts equality
• assertNotEquals expected actual "message" - Asserts inequality
• assertContains haystack needle "message" - Asserts substring exists
• assertNotContains haystack needle "message" - Asserts substring doesn't exist
• assertFileExists file "message" - Asserts file exists
• assertFileNotExists file "message" - Asserts file doesn't exist
• assertDirectoryExists dir "message" - Asserts directory exists

cd test
./run-tests.sh

cd test/unit
bash test-config-loader.sh

The test framework shows detailed output by default. To see more details, check the test file directly.

⚠️ CRITICAL: Before writing tests, read ARCHITECTURE.md to understand:

• How to use the configuration system (never hardcode values!)
• How TEST_UPSTREAM_ORG works
• Best practices for test structure

#!/bin/bash
# Unit tests for my-script.sh

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
TEST_FRAMEWORK="$SCRIPT_DIR/../helpers/test-framework.sh"
source "$TEST_FRAMEWORK"

# Setup function (runs before each test)
setUp() {
    # Create test environment
    TEST_DIR="/tmp/test-$$"
    mkdir -p "$TEST_DIR"
}

# Teardown function (runs after each test)
tearDown() {
    # Cleanup
    rm -rf "$TEST_DIR"
}

# Test functions (must start with "test")
testMyFunctionWorks() {
    # Your test code here
    assertTrue "Function should work" my_function
}

testMyFunctionFails() {
    assertFalse "Function should fail with bad input" my_function "bad_input"
}

# Run tests if executed directly
if [ "${0##*/}" = "test-my-script.sh" ]; then
    resetCounters
    runTests "$0"
fi

1. Load test framework - Source test-framework.sh
2. Set up test environment - Use setUp() function
3. Write test functions - Functions starting with test
4. Clean up - Use tearDown() function
5. Run tests - Call runTests if executed directly

• ✅ config-loader.sh - Configuration loading, defaults, warnings
• ✅ safety-checks.sh - Commit/push blocking, repo detection
• ✅ sync-discord.sh - Discord sync script argument parsing, validation, path resolution

• monitor-embabel.sh - Monitoring functionality
• sync-upstream.sh - Sync operations
• view-pr.sh - PR viewing
• list-embabel-repos.sh - Repository listing
• Integration tests for common workflows

Tests use temporary directories (/tmp/embabel-learning-test-*) to avoid polluting the actual workspace. All test directories are cleaned up automatically.

• Each test file runs in its own subshell
• Test directories are created with unique IDs ($$)
• Cleanup happens automatically in tearDown()

To run tests in CI/CD:

# Install dependencies (none required!)
# Run tests
cd test && ./run-tests.sh

The test framework is self-contained and doesn't require any external tools.

When adding new scripts:

1. Write tests first (TDD approach) or
2. Add tests alongside the script
3. Update this README with test coverage
4. Ensure all tests pass before committing

Solution: Make test files executable:

chmod +x test/**/*.sh

Solution: Make sure you're running from the test directory:

cd test
./run-tests.sh

Solution: Ensure test-framework.sh is executable and in helpers/:

ls -l test/helpers/test-framework.sh
chmod +x test/helpers/test-framework.sh

The test framework (test-framework.sh) provides:

• Assertion functions - Various assert functions for testing
• Test statistics - Automatic counting of passed/failed tests
• Color output - Green for pass, red for fail
• Summary reporting - Automatic summary at end
• Error handling - Proper exit codes for CI/CD

No external dependencies - just pure bash!