Add unit test suite and code coverage enforcement

[mmccarty]

Summary

• Add comprehensive unit test suite (53 tests) covering all modules: CLI entry points, config, doctor checks (GPU, CUDA, memory, NVLink), debug commands, and doctor orchestration
• Add pytest-cov dependency and configure 95% minimum coverage threshold with automatic reporting
• Fix pytest test path configuration ([tool.pytest] → [tool.pytest.ini_options], tests → rapids_cli/tests)
• Add CLAUDE.md with project architecture, development commands, code style, and testing guidance
• Add SPDX license headers to dependencies.yaml

Test plan

• pytest passes all 53 tests
• Coverage meets 95% threshold
• pre-commit run --all-files passes

🤖 Generated with Claude Code

Summary by CodeRabbit

Release Notes

• Documentation

  • Added developer guidance for the repository.

• Tests

  • Expanded test coverage for CLI commands, configuration validation, debug output, doctor checks, GPU detection, memory diagnostics, and NVLink status.

• Chores

  • Updated project dependencies and testing configuration with coverage reporting and metrics.

[jacobtomlinson]

@coderabbitai full review

[coderabbitai]

✅ Actions performed

Full review triggered.

[coderabbitai]

Walkthrough

This PR adds comprehensive test coverage infrastructure to the RAPIDS CLI project, including pytest-cov dependency addition, coverage configuration to pyproject.toml, seven new test modules covering CLI/config/debug/doctor/GPU/memory/NVLink functionality, and documentation guidance (CLAUDE.md).

Changes

Cohort / File(s)	Summary
Documentation & Configuration CLAUDE.md, dependencies.yaml, pyproject.toml	Adds Claude Code guidance documentation, introduces pytest-cov dependency, refactors pytest config to [tool.pytest.ini_options] with updated testpaths, and adds coverage configuration sections with source/omit/exclude patterns and 95% coverage threshold.
CLI Test Module rapids_cli/tests/test_cli.py	New test suite validating CLI help output, subcommand invocation (doctor/debug), exit codes, verbose/dry-run/filter behavior, and run_debug output formats using CliRunner with patched implementations.
Configuration Test Module rapids_cli/tests/test_config.py	New test suite verifying config object structure, presence of required sections (min_supported_versions, valid_subcommands, os_requirements), and nested configuration keys.
Debug Test Module rapids_cli/tests/test_debug.py	New test suite covering CUDA driver version gathering, installed package enumeration, command output fallback logic, tool detection, and run_debug output in both console and JSON formats with mocked NVML/filesystem interactions.
Doctor Test Module rapids_cli/tests/test_doctor.py	New test suite validating doctor_check across pass/fail scenarios, verbose discovery output, dry-run behavior, check filtering, warning handling, and ImportError suppression using mocked entry points.
GPU Test Module rapids_cli/tests/test_gpu.py	New test suite for GPU detection and compute capability validation, covering success paths, no-GPU scenarios, NVML errors, and compute capability sufficiency checks.
Memory Test Module rapids_cli/tests/test_memory.py	New test suite validating system/GPU memory detection for single and multiple GPU configurations, and memory-to-GPU ratio validation with warning conditions, using mocked psutil and pynvml calls.
NVLink Test Module rapids_cli/tests/test_nvlink.py	New test suite verifying NVLink status checks for multi-GPU configurations, single-GPU unsupported scenarios, and error handling for missing GPUs and unsupported NVML features.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~18 minutes

Poem

🐰 A testing feast of grandeur built with care,
Seven modules dancing through the digital air,
With mocks and patches woven tight and true,
Coverage climbs to ninety-five or new! 🎯
The doctor checks, the GPUs aligned—
A robust CLI, so finely designed! ✨

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 70.37% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title directly and clearly describes the main changes: adding a comprehensive unit test suite with code coverage enforcement (95% threshold).

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

Tip

Issue Planner is now in beta. Read the docs and try it out! Share your feedback on Discord.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@CLAUDE.md`:
- Around line 29-30: The README example references a non-existent test file
`rapids_cli/tests/test_cuda.py`; update the example command to point to an
actual test file (e.g., `rapids_cli/tests/test_gpu.py` or one of `test_cli.py`,
`test_config.py`, `test_debug.py`, `test_doctor.py`, `test_memory.py`,
`test_nvlink.py`) so the documented pytest invocation works; replace
`test_cuda.py` with a real test filename in the CLAUDE.md example line.

🧹 Nitpick comments (5)

CLAUDE.md (1)

100-100: Nit: "CLI interface" is redundant.

"CLI" already stands for "Command-Line Interface", so "CLI interface" expands to "Command-Line Interface interface."

Proposed fix

-- `rich` and `rich-click` for terminal output and CLI interface
+- `rich` and `rich-click` for terminal output and CLI

rapids_cli/tests/test_nvlink.py (1)

34-38: Unescaped regex metacharacters in match pattern.

The dots in "GPU not found. Please ensure GPUs are installed." are regex metacharacters that match any character. While unlikely to cause false matches here, use re.escape() or escape the dots for correctness.

Proposed fix

+import re
+
 def test_check_nvlink_status_no_gpu():
     import pynvml
 
     with patch("pynvml.nvmlInit", side_effect=pynvml.NVMLError(1)):
         with pytest.raises(
-            ValueError, match="GPU not found. Please ensure GPUs are installed."
+            ValueError, match=re.escape("GPU not found. Please ensure GPUs are installed.")
         ):
             check_nvlink_status(verbose=False)

rapids_cli/tests/test_debug.py (1)

84-88: Broadly patching pathlib.Path.glob and pathlib.Path.read_text is fragile.

These patches affect all Path instances process-wide, not just the ones in debug.py. If pytest internals, coverage tooling, or rich use Path.glob or Path.read_text during the test, it could cause subtle failures. Consider patching at the module level (e.g., rapids_cli.debug.debug.Path) or using more targeted mocking if issues arise.

rapids_cli/tests/test_doctor.py (1)

101-117: Unused capsys parameter and weak assertion — consider verifying only the filtered check ran.

capsys is declared but never used (Ruff ARG001). More importantly, the test only asserts result is True but doesn't verify that only the cudf entry point was loaded/executed (and cuml was skipped). Without that, the test doesn't actually confirm filtering behavior — it only confirms no crash.

Consider calling mock_ep2.load.assert_not_called() and removing the unused capsys, or using it to verify relevant output.

Proposed improvement

-def test_doctor_check_with_filters(capsys):
+def test_doctor_check_with_filters():
     """Test doctor_check with filters."""
     mock_ep1 = MagicMock()
     mock_ep1.name = "cudf_check"
     mock_ep1.value = "cudf.module:check"
     mock_ep1.load.return_value = mock_passing_check

     mock_ep2 = MagicMock()
     mock_ep2.name = "cuml_check"
     mock_ep2.value = "cuml.module:check"
     mock_ep2.load.return_value = mock_passing_check

     with patch(
         "rapids_cli.doctor.doctor.entry_points", return_value=[mock_ep1, mock_ep2]
     ):
         result = doctor_check(verbose=False, dry_run=False, filters=["cudf"])
         assert result is True
+        mock_ep1.load.assert_called_once()
+        mock_ep2.load.assert_not_called()

rapids_cli/tests/test_memory.py (1)

75-82: Use a raw string for the match pattern to avoid unescaped regex metacharacters.

The match argument is a regex pattern, and the dots in "GPU not found. Please ensure GPUs are installed." are metacharacters that match any character. While unlikely to cause a false positive here, using a raw string with escaped dots is more correct. Flagged by Ruff (RUF043).

Proposed fix

         with pytest.raises(
-            ValueError, match="GPU not found. Please ensure GPUs are installed."
+            ValueError, match=r"GPU not found\. Please ensure GPUs are installed\."
         ):

[jameslamb]

Packaging changes look good to me. I guess pyyaml must previously have been getting into the environment some other way, but the dependency is needed because rapids-cli has direct usage:

rapids-cli/rapids_cli/config.py

Line 7 in ba0763c

import yaml

I didn't review the other changes, will leave that to Jacob and Naty.