tweaks

Author: gfx

benchmark/decode-string.ts

@@ -5,7 +5,7 @@ import { utf8EncodeJs, utf8Count, utf8DecodeJs, utf8DecodeTD } from "../src/util
 import Benchmark from "benchmark";
 
 for (const baseStr of ["A", "あ", "🌏"]) {
-  const dataSet = [10, 100, 200, 1_000, 10_000, 100_000].map((n) => {
+  const dataSet = [10, 100, 500, 1_000].map((n) => {
     return baseStr.repeat(n);
   });
 
@@ -14,7 +14,7 @@ for (const baseStr of ["A", "あ", "🌏"]) {
     const bytes = new Uint8Array(new ArrayBuffer(byteLength));
     utf8EncodeJs(str, bytes, 0);
 
-    console.log(`\n## string "${baseStr}" x ${str.length} (byteLength=${byteLength})\n`);
+    console.log(`\n## string "${baseStr}" (strLength=${str.length}, byteLength=${byteLength})\n`);
 
     const suite = new Benchmark.Suite();
 

benchmark/encode-string.ts

@@ -5,15 +5,15 @@ import { utf8EncodeJs, utf8Count, utf8EncodeTE } from "../src/utils/utf8";
 import Benchmark from "benchmark";
 
 for (const baseStr of ["A", "あ", "🌏"]) {
-  const dataSet = [10, 100, 200, 1_000, 10_000, 100_000].map((n) => {
+  const dataSet = [10, 30, 50, 100].map((n) => {
     return baseStr.repeat(n);
   });
 
   for (const str of dataSet) {
     const byteLength = utf8Count(str);
     const buffer = new Uint8Array(byteLength);
 
-    console.log(`\n## string "${baseStr}" x ${str.length} (byteLength=${byteLength})\n`);
+    console.log(`\n## string "${baseStr}" (strLength=${str.length}, byteLength=${byteLength})\n`);
 
     const suite = new Benchmark.Suite();
 

src/utils/utf8.ts

@@ -1,5 +1,3 @@
-/* eslint-disable @typescript-eslint/no-unnecessary-condition */
-import { UINT32_MAX } from "./int";
 
 export function utf8Count(str: string): number {
   const strLength = str.length;
@@ -88,9 +86,14 @@ export function utf8EncodeJs(str: string, output: Uint8Array, outputOffset: numb
 // https://encoding.spec.whatwg.org/
 // and available in all the modern browsers:
 // https://caniuse.com/textencoder
+// They are available in Node.js since v12 LTS as well:
+// https://nodejs.org/api/globals.html#textencoder
 
 const sharedTextEncoder = new TextEncoder();
-const TEXT_ENCODER_THRESHOLD = 200;
+
+// This threshold should be determined by benchmarking, which might vary in engines and input data.
+// Run `npx ts-node benchmark/encode-string.ts` for details.
+const TEXT_ENCODER_THRESHOLD = 50;
 
 export function utf8EncodeTE(str: string, output: Uint8Array, outputOffset: number): void {
   sharedTextEncoder.encodeInto(str, output.subarray(outputOffset));
@@ -156,6 +159,9 @@ export function utf8DecodeJs(bytes: Uint8Array, inputOffset: number, byteLength:
 }
 
 const sharedTextDecoder = new TextDecoder();
+
+// This threshold should be determined by benchmarking, which might vary in engines and input data.
+// Run `npx ts-node benchmark/decode-string.ts` for details.
 const TEXT_DECODER_THRESHOLD = 200;
 
 export function utf8DecodeTD(bytes: Uint8Array, inputOffset: number, byteLength: number): string {