Count occurrences of registered string patterns in a chunked UTF-8 byte stream (or JS string series) using minimal memory.
See SPEC.md for the full specification.
# Deno
deno add jsr:@axhxrx/utf8-count
# Bun / Node
bunx jsr add @axhxrx/utf8-countimport { OccurrenceCounter } from '@axhxrx/utf8-count'
const counter = new OccurrenceCounter(['hello', 'world'])
counter.appendString('hello wor');
counter.appendString('ld, hell');
counter.appendString('o again')
counter.count('hello') // 2
counter.count('world') // 1
// UTF-8 byte input, including multi-byte codepoints split across chunks:
const c = new OccurrenceCounter(['café'])
c.appendBytes(new Uint8Array([0x63, 0x61, 0x66, 0xc3])) // "caf" + first byte of 'é'
c.appendBytes(new Uint8Array([0xa9])) // rest of 'é'
c.count('café') // 1./bin/count.ts README.md count byte count string "new Uint8Array"
README.md: 1,485 bytes in 1 chunk (0.84 ms, caseSensitive=true)
"count" 20
"byte" 8
"count" 20
"string" 4
"new Uint8Array" 4
Or, from JSR:
deno run -R https://jsr.io/@axhxrx/utf8-count/0.1.0/bin/count.ts \
--ignore-case \
README.md \
deno bun node
README.md: 1,600 bytes in 1 chunk (2.83 ms, caseSensitive=false)
"deno" 7
"bun" 6
"node" 5
bun test
deno test -A
node --testDeno, Bun, Node 24.2+
MIT
-
2026-04-18 💄 0.1.2 — beautify README
-
2026-04-18 🩹 0.1.1 — fix bug in README
-
2026-04-18 📦 0.1.0 — initial release
以上
