diff-numerics verions 2.0.0

Release v2.0.0 - Initial Re-Release

Release Date: February 12, 2026

🎉 First Stable Release

diff-numerics v2.0.0 is a high-performance numerical file comparison tool built with modern C++17, designed for scientific computing, data validation, and automated testing workflows.

✨ Features

Core Functionality

• Intelligent numerical comparison with configurable relative tolerance and absolute threshold
• Scientific notation support for exponential formats
• Column-selective comparison using 1-based indexing
• Comment line filtering with configurable prefixes

Output Options

• Side-by-side view with aligned columns and intelligent separators
• Unified diff format (traditional < / > style)
• Quiet mode for scripting and CI/CD integration
• Summary-only mode for quick file equality checks

Visualization

• Fine-grained digit colorization - highlights only the differing digits
• Full value highlighting - colors entire differing numbers in red
• ANSI-aware formatting - proper terminal width calculations

Quality & Performance

• Modern C++17 implementation with STL algorithms and std::from_chars
• Clean modular architecture with comprehensive documentation
• Full GoogleTest suite with 18+ test cases
• CMake build system with installation support

diff-numerics

Initial Release

This is the first official release of diff-numerics.

Features

• Compare numerical data files with configurable tolerance and threshold.
• Side-by-side and standard diff output modes.
• Option to suppress common lines for focused diffs.
• Highlight only differing digits with color (ANSI escape codes).
• Compare specific columns by index.
• Command-line interface with robust error handling.
• Quiet and summary output modes for scripting and automation.

Getting Started

1. Build the project using CMake or Makefile.
2. Run the diff-numerics binary with two data files as arguments.

Example:

./diff-numerics file1.dat file2.dat

Notes

• See the README for detailed usage instructions and options.
• Please report issues or feature requests via GitHub Issues.

Thank you for trying out diff-numerics!