Working toward a safer standard library for the next generation

NextStd is an experimental effort to explore safer, well-defined, and modern alternatives to traditional C standard library components. The project focuses on correctness, explicit behavior, and performance, while remaining practical for real-world systems and embedded use.

This is a work in progress. APIs, guarantees, and scope may evolve as the project matures. All core logic is backed by Rust, safely exposed to C via FFI.

A memory-safe, crash-proof alternative to <stdio.h> that completely eliminates format string vulnerabilities. It uses C11 _Generic routing to automatically detect data types at compile time.

• Output (ns_print, ns_println): Safely prints int, float, double , char*, and custom NextStd types without %d or %s specifiers.
• Input (ns_read): Safely reads int, float, and double. It gracefully defaults to 0 on invalid user input instead of panicking, and includes strict NULL pointer safeguards.

A modern alternative to standard C strings (char*) designed to eliminate buffer overflows, $O(N)$ length calculations, and double-free vulnerabilities.

• Small String Optimization (SSO): Strings under 24 bytes are allocated directly on the stack, resulting in zero heap fragmentation (ideal for embedded constraints).
• Automatic Heap Scaling: Strings 24 bytes and larger seamlessly scale to the heap, with memory lifecycles safely managed by the Rust backend.
• Core Functions: ns_string_new(), ns_string_free()

NextStd bridges the gap between modern safety and legacy compatibility:

1. Rust Backend: Handles memory allocation, bounds checking, and core logic.
2. C Frontend: Exposes a clean, macro-driven API that feels completely native to C developers.