dash-em

dash-em is a specialized string manipulation library written in C, designed to remove em-dashes (U+2014) from UTF-8 encoded text. The project is explicitly described as a "meme" or "joke" library, intentionally over-engineered for a task that could be performed with a simple regular expression. However, the implementation is technically robust, utilizing Single Instruction, Multiple Data (SIMD) vectorization to achieve high throughput. The library supports multiple CPU instruction sets, including AVX-512, AVX2, SSE4.2, and ARM NEON, automatically selecting the most appropriate path based on the host hardware. This allows for significant performance gains, with benchmarks showing throughput speeds ranging from 1.27 GB/s to over 42 GB/s, depending on the density of em-dashes in the input and the specific CPU architecture. The core C library is dependency-free and includes micro-optimizations such as loop unrolling and cache prefetching. In addition to the core C implementation, dash-em provides bindings for a wide range of programming languages, including JavaScript/TypeScript, Python, Go, Rust, Java, C#/.NET, PHP, Ruby, Swift, Kotlin, R, Dart, Scala, Perl, Lua, Haskell, Elixir, Zig, and Objective-C. It also compiles to WebAssembly (wasm32 and WASI) for browser and serverless environments. The API is consistent across languages, typically offering a `remove()` function that takes a string or buffer and returns the cleaned output. JavaScript bindings also support zero-copy and in-place operations via `removeBuffer()` and `removeBufferInPlace()`. Installation is straightforward, with packages available for standard package managers (e.g., npm, pip, cargo) and a CMake build system for the core library. The project includes comprehensive test suites and continuous integration pipelines to ensure correctness and performance across all supported platforms and bindings.