In August, I came up with an algorithmic method for matching wildcards in Rust and had no trouble making it UTF-8-ready. I realized the technique would work for other programming languages and set out to make it happen. But what would it take to make sure a UTF-8-ready implementation for C/C++ would be as safe from memory management errors and leaks as with newer languages? A combination of tests and tools has made that possible.

The overall result is a story about portability. Nowadays most talk of portability centers on cross-platform portability: the sort that matters to software of the accounting or word processing variety. This story is about cross-language portabiility: the sort that matters to techniques, data structures, and algorithms.

The story has lots of details and a happy ending. You can find the tested, drop-in-and use implementations for Go, for Swift, and for C/C++ on GitHub.