v1.0.0

Release v1.0.0: Initial Release (The Core Engine)

The inaugural release of RingDetect-HPC! This version establishes the core architecture: a lightweight C/Python frontend for memory management paired with a blisteringly fast Fortran 2008 backend for heavy graph theory computation.

✨ Core Capabilities

• O(N) Spatial Hashing: Instantly builds adjacency matrices without the traditional $O(N^2)$ distance bottleneck.
• OpenMP Parallelization: Deeply parallelized Depth-First Search (DFS) allows for rapid cycle perception across massive macromolecular systems.
• Geometric Planarity Checking: Computes 3D normal vectors on the fly to detect if a cycle is geometrically planar (useful for identifying aromaticity and conformation).
• Periodic Boundary Conditions (PBC): Fully supports the Minimum Image Convention (MIC) for continuous crystal lattices via the -c flag.
• Zero-Allocation Search: Uses in-place array mutation during the DFS to completely eliminate RAM allocation locks.
• Dual Interface: Ships with a standalone, zero-dependency C executable and a fast Python ctypes wrapper for zero-copy memory pointer passing.
• Multi-Format Parsing: Natively reads standard coordinates from XYZ, PDB, MOL, CSV, and RAW files.
• JSON Export: Output results in strict JSON format (-j) for easy downstream parsing into Pandas DataFrames or web tools.