Skip to content

specio3 v0.1.3 - High-Performance SPC File Reader

Choose a tag to compare

@WillCallahan WillCallahan released this 08 Aug 04:58
· 4 commits to main since this release

Overview

specio3 is a high-performance Python library for reading Galactic SPC (Spectral Data Collection) files. Built with C++ for speed and wrapped with pybind11 for seamless Python integration.

✨ Key Features

  • 🚀 High-Performance C++ Core: Spectral file parsing implemented in optimized C++ with pybind11 bindings
  • 📊 Comprehensive SPC Format Support:
    • Y-only (evenly spaced X-axis)
    • XY (explicit X values)
    • XYY (shared X-axis, multiple Y spectra)
    • XYXY (per-spectrum X-axis)
  • 🐍 Pythonic API: Clean, well-typed interface with comprehensive documentation
  • 📈 NumPy Integration: Returns data as numpy arrays for seamless scientific computing
  • 🛡️ Robust Error Handling: Comprehensive validation with clear error messages
  • 🔄 Cross-Platform: Supports Linux, macOS, and Windows
  • 📦 Easy Installation: Available on PyPI with pre-built wheels

🏆 Performance

3.6x faster than spectrochempy - the reference Python implementation:

File Size specio3 spectrochempy Speedup
103b4anh.spc 131 KB 2.4 ms 8.1 ms 3.4x faster
087b4ana.spc 258 KB 4.1 ms 14.2 ms 3.4x faster
040b4ana.spc 131 KB 2.0 ms 7.8 ms 3.9x faster

📦 Installation

pip install specio3

🔧 Quick Start

import specio3
import matplotlib.pyplot as plt

# Read an SPC file
spectra = specio3.read_spc('your_file.spc')

# Get the first spectrum
x_values, y_values = spectra[0]

# Plot the spectrum
plt.figure(figsize=(10, 6))
plt.plot(x_values, y_values)
plt.xlabel('Wavenumber (cm⁻¹)')
plt.ylabel('Intensity')
plt.title('Spectral Data')
plt.show()

print(f"Loaded {len(spectra)} spectra")
print(f"First spectrum has {len(x_values)} data points")

🔄 What's New in v0.1.3

  • Performance: Optimized GitHub Actions wheel builds (5.6x faster CI/CD)
  • Security: Comprehensive security audit and vulnerability fixes
  • Organization: Improved project structure with docs/ and scripts/ directories
  • Testing: Enhanced test reliability and removed .DS_Store from test output
  • CI/CD: Fixed wheel platform tags and deployment issues
  • Documentation: Updated API documentation and examples

🛠️ Technical Details

Supported File Types

  • Y-only: Evenly spaced X-axis with generated X values
  • XY: Explicit X and Y values for single spectrum
  • XYY: Shared X-axis with multiple Y spectra
  • XYXY: Individual X-axis per spectrum

API Reference

  • read_spc(path: str) -> List[Tuple[NDArray[np.float64], NDArray[np.float64]]]
  • Returns list of (x_array, y_array) tuples
  • Comprehensive error handling for file access and format issues

🧪 Testing & Quality

  • Comprehensive Test Suite: Tested against real SPC files from EPA Spectra dataset
  • Multi-Platform CI/CD: Automated testing on Linux, macOS, and Windows
  • Python 3.10+ Support: Compatible with Python 3.10, 3.11, and 3.12
  • Type Safety: Full type hints for optimal IDE support
  • Security Audited: Comprehensive security analysis included

🤝 Contributing

We welcome contributions! See our Contributing Guidelines for details.

📄 License

MIT License - see LICENSE file for details.

🙏 Acknowledgments

  • Built with pybind11 for Python-C++ bindings
  • Uses NumPy for efficient array operations
  • Tested with spectral data from EPA Spectra dataset
  • Performance benchmarks compared against spectrochempy

Made with ❤️ for the spectroscopy community

For support, documentation, and discussions, visit our GitHub repository.