Introduce CodePointString

README.md

@@ -1,5 +1,8 @@
 TypeScript library for handling source code strings without having to deal with intricacies of JS's UTF16 encoding.
 
+# CodePointString
+A wrapper for a string that's just an array of codepoints. There's no newline or offset tracking to the original string.
+
 # SourceText
 A sane, UTF-16-safe string wrapper specifically designed for parsing source code, tracking line numbers, and generating CLI error messages.
 Think of it as a fat wrapper for a string that understand more info about the string like line structure.

src/index.ts

@@ -68,6 +68,44 @@ export type CodePointSpan = {
   end: CodePointIndex,
 }
 
+// === CodePointString ===
+export class CodePointString {
+  readonly codePoints: readonly CodePoint[];
+
+  constructor(source: string) {
+    const codePointsInternal: CodePoint[] = [];
+    let i = 0;
+    while (i < source.length) {
+      const char = source.codePointAt(i) as CodePoint;
+      codePointsInternal.push(char);
+
+      const size =(char > 0xFFFF ? 2 : 1);
+      i += size;
+    }
+    this.codePoints = Object.freeze(codePointsInternal);
+  }
+
+  static makeFromString(s: string): CodePointString {
+    return new CodePointString(s);
+  }
+
+  codePointAt(index: CodePointIndex): CodePoint {
+    return this.codePoints[index];
+  }
+
+  get length(): CodePointIndex {
+    return this.codePoints.length;
+  }
+
+  toString(): string {
+    let result = "";
+    for (const cp of this.codePoints) {
+      result += String.fromCodePoint(cp);
+    }
+    return result;
+  }
+}
+
 // === Source Text ===
 // TODO:
 // @deprecated and say to use `SourceText.makeFromString` instead.