Thanks again for the thoughtful feedback on my original post — much appreciated.
I’ve made a few updates to chonkify, my tiny (~1kB) utility for chunking anything (arrays, strings, buffers, async iterables, etc.):
- Full source now included (no more minified-only dist)
- Clarified string behavior:
-- chonk() uses UTF-16 code points (like JS)
-- new chonkGraphemes() splits by grapheme clusters (so emoji stay intact)
- Updated README with clear examples and tradeoffs
I still think grapheme support is cool — just not the default anymore.
JS strings are based on UTF-16, but real-world text (emojis, Hindi, Thai, etc.) often needs grapheme-aware handling. Now you can choose between "what JS does" and "what users expect".
This is still very much a pet project — I built it primarily for myself and don’t claim any production-grade seriousness. That said, I’d be thrilled if it turns out useful to anyone else.
Would love more feedback!