Release v0.6.0

Quality & Testing Improvements

This release focuses on significantly improving code quality, test coverage, and developer experience. We've completed all P0 (high priority) and P1 (medium priority) improvements from our TODO list.

Test Coverage Increased by 11.7%

Total tests: 248 → 277 (+29 new tests)

All core modules now have comprehensive test coverage:

• ✅ Provider validation: 9 tests
• ✅ Git repository operations: 14 tests
• ✅ Review command: 6 tests

Real API Health Checks for Providers

The gcop config validate command has been significantly enhanced to perform real API connectivity checks instead of just checking if API keys are non-empty.

Before v0.6.0:

$ gcop config validate
✓ Configuration loaded
✓ Provider 'claude' configured
# Only checked if API key exists, didn't test actual connectivity

After v0.6.0:

$ gcop config validate
[1/2] Validating configuration...
[2/2] Testing provider connection...
✓ Provider 'claude' validated successfully
# Actually sends a test request to verify API works!

What's new:

1. Claude/OpenAI Providers: Send minimal test requests (max_tokens=1) to verify API connectivity

   • Detects invalid API keys (401 errors)
   • Detects rate limits (429 errors)
   • Minimal cost: only 1 token per validation

2. Ollama Provider: Checks /api/tags health endpoint

   • Verifies Ollama server is running
   • Checks if configured model is downloaded
   • Provides helpful suggestion: ollama pull <model> if model not found

3. FallbackProvider: Validates all configured providers

   • Shows status for each provider in fallback chain
   • Aggregates results and warns if all providers fail

4. Better Error Messages: Enhanced error output with actionable suggestions

   ✗ Validation failed: Claude API validation failed: Invalid API key
   
   💡 Suggestion: Check your ANTHROPIC_API_KEY environment variable or
      api_key in config file (~/.config/gcop/config.toml)

Impact:

• Catch configuration issues before running gcop commit
• Save time debugging API connectivity problems
• Better onboarding experience for new users

Git Repository Test Coverage

Added comprehensive test suite for Git repository operations (tests/git_repository_test.rs with 14 tests):

Edge Cases Covered:

• ✅ Empty repositories (no commits yet)
• ✅ Large file size limits (default 10MB, custom limits)
• ✅ Paths with spaces and Unicode characters
• ✅ First commit in repository (no parent)
• ✅ Invalid commit hashes and range formats
• ✅ Detached HEAD state
• ✅ Concurrent test safety (using serial_test crate)

Example Test:

#[test]
fn test_get_file_content_exceeds_max_size() {
    // Creates 11MB file (exceeds default 10MB limit)
    // Expects: InvalidInput error with "File too large" message
}

These tests ensure robustness when dealing with unusual repository states and prevent regressions.

Review Command Tests

Added integration tests for the gcop review command (tests/review_command_test.rs with 6 tests):

Coverage:

• ✅ All 4 target types: --changes, --commit <hash>, --range <range>, --file <path>
• ✅ Error handling: empty diff, LLM failures
• ✅ Correct routing to Git operations

Architectural Improvement:

Refactored review.rs to support dependency injection:

// Public API (unchanged)
pub async fn run(cli: &Cli, config: &AppConfig, ...) -> Result<()>

// Internal implementation (testable with mocks)
pub async fn run_internal(
    config: &AppConfig,
    git: &dyn GitOperations,  // ← Injectable for tests
    llm: &dyn LLMProvider,     // ← Injectable for tests
) -> Result<()>

This pattern maintains backward compatibility while enabling thorough testing with mock objects.

MSRV (Minimum Supported Rust Version) Fixed

Rust 1.92.0 or higher is now officially required (Rust 2024 edition).

What's added:

1. rust-toolchain.toml: Ensures consistent Rust version across all environments

   [toolchain]
   channel = "stable"
   components = ["rustfmt", "clippy"]

2. Cargo.toml: Declares MSRV for crates.io and cargo

   rust-version = "1.92.0"

3. CI/CD: Added MSRV check job in GitHub Actions

   • Ensures code compiles with minimum Rust version
   • Prevents accidental use of newer Rust features

4. Documentation: Updated all installation guides (English/Chinese)

   • README.md
   • README_ZH.md
   • docs/guide/installation.md
   • docs/zh/guide/installation.md

Why this matters:

• Clear expectations for contributors
• Prevents "it works on my machine" issues
• Maintains compatibility with Rust 2024 edition requirements

Technical Details

Provider Validation Implementation

Each provider implements the validate() method from the LLMProvider trait:

async fn validate(&self) -> Result<()> {
    // Claude/OpenAI: Send test request
    let test_request = Request {
        model: self.model.clone(),
        max_tokens: 1,  // Minimal cost
        ...
    };

    let response = self.client.post(&self.endpoint)
        .json(&test_request)
        .send()
        .await?;

    if !response.status().is_success() {
        return Err(GcopError::LlmApi { ... });
    }

    Ok(())
}

For Ollama, we use the /api/tags endpoint:

async fn validate(&self) -> Result<()> {
    let tags_response = self.client
        .get(&health_endpoint)
        .send()
        .await?;

    // Check if configured model exists
    if !models.iter().any(|m| m.name.starts_with(&self.model)) {
        return Err(GcopError::Config(
            "Model not found. Run 'ollama pull <model>' first."
        ));
    }

    Ok(())
}

Test Infrastructure

New tests use:

• mockito: HTTP mocking for provider tests
• mockall: Trait mocking for Git/LLM operations
• tempfile: Temporary directories for Git repository tests
• serial_test: Serialized execution for tests that modify global state
• tokio::test: Async test support

Upgrade Notes

Upgrading from v0.5.1 requires no action. This release is fully backward compatible.

What stays the same:

• All public APIs remain unchanged
• Configuration file format unchanged
• Command-line interface unchanged

What improves:

• gcop config validate now performs real API checks
• Better error messages and suggestions
• More robust with comprehensive test coverage

Breaking Changes

None. This is a minor version bump (0.5.1 → 0.6.0) with new features and improvements, but no breaking changes.

Installation

# Homebrew (macOS/Linux)
brew tap AptS-1547/gcop-rs
brew upgrade gcop-rs

# pipx (Python users, recommended)
pipx upgrade gcop-rs

# cargo-binstall (no compilation)
cargo binstall gcop-rs

# cargo install (compile from source)
cargo install gcop-rs --force

Or download pre-built binaries from Releases.

What's Next

Remaining low-priority improvements (P2):

• UI module testability improvements
• Diff syntax highlighting with syntect

See TODO.md for details.

Statistics

• Files changed: 22
• Lines added: +1,726
• Lines removed: -64
• New test files: 3
• New tests: 29
• Test coverage increase: 248 → 277 (+11.7%)

Contributors

• AptS-1547 (Yuhan Bian / 卞雨涵)
• AptS-1548 (48)

Feedback

If you have any issues or suggestions, please submit an Issue.

---

Full Changelog: v0.5.1...v0.6.0