String

Stock surface

Wrenlift implements the full stock Wren String API — concatenation, count, indexOf, contains, startsWith, endsWith, slice (s[a..b]), codePointAt, byteAt, split, trim, replace, and the iteration protocol. Strings are immutable; every "mutator" returns a new string.

This page covers what Wrenlift adds on top of that surface. Everything else is exactly as Wren spec documents it.

fromCodePoint

Construct a one-character string from a Unicode code-point integer:

System.print(String.fromCodePoint(65))       // "A"
System.print(String.fromCodePoint(0x1F44B))  // "👋"

Combine with 0..n ranges to build small UTF-8 buffers without ByteArray:

// Build a 4-byte UTF-8 payload of 0x41 0x42 0x43 0x44 (ABCD):
var s = ""
for (cp in [0x41, 0x42, 0x43, 0x44]) {
  s = s + String.fromCodePoint(cp)
}

Throws Fiber.abort on invalid code points (negative, > 0x10FFFF, or in the surrogate gap 0xD800..=0xDFFF).

Byte iteration

Stock Wren iterates a string by code-point. Wrenlift keeps that, and adds a byte-level view through the .bytes accessor:

var s = "héllo"
System.print(s.count)        // 5  — code points
System.print(s.bytes.count)  // 6  — bytes (é is 2 bytes in UTF-8)

for (b in s.bytes) {
  System.print(b)            // 104, 195, 169, 108, 108, 111
}

s.bytes is a lightweight view, not a copy — constructing it doesn't allocate a backing buffer. For a real, mutable buffer, copy through ByteArray.fromString(s).

Typed-array interop

Strings and typed arrays bridge cleanly in both directions:

Direction	Call
String → bytes	ByteArray.fromString(s) (UTF-8 snapshot)
Bytes → String	String.fromBytes(byteArray) (decode UTF-8 — aborts on invalid sequences)

var b = ByteArray.fromString("hi 👋")
System.print(b.count)              // 6

var s = String.fromBytes(b)
System.print(s)                    // "hi 👋"