Benchmark: Normalize digits

Script Preparation code:

var str = "１５０７２１";
var DIGITS = new Map([
  ["0", "0"],
  ["1", "1"],
  ["2", "2"],
  ["3", "3"],
  ["4", "4"],
  ["5", "5"],
  ["6", "6"],
  ["7", "7"],
  ["8", "8"],
  ["9", "9"],
  ["\uFF10", "0"], // Fullwidth digit 0
  ["\uFF11", "1"], // Fullwidth digit 1
  ["\uFF12", "2"], // Fullwidth digit 2
  ["\uFF13", "3"], // Fullwidth digit 3
  ["\uFF14", "4"], // Fullwidth digit 4
  ["\uFF15", "5"], // Fullwidth digit 5
  ["\uFF16", "6"], // Fullwidth digit 6
  ["\uFF17", "7"], // Fullwidth digit 7
  ["\uFF18", "8"], // Fullwidth digit 8
  ["\uFF19", "9"], // Fullwidth digit 9
  ["\u0660", "0"], // Arabic-indic digit 0
  ["\u0661", "1"], // Arabic-indic digit 1
  ["\u0662", "2"], // Arabic-indic digit 2
  ["\u0663", "3"], // Arabic-indic digit 3
  ["\u0664", "4"], // Arabic-indic digit 4
  ["\u0665", "5"], // Arabic-indic digit 5
  ["\u0666", "6"], // Arabic-indic digit 6
  ["\u0667", "7"], // Arabic-indic digit 7
  ["\u0668", "8"], // Arabic-indic digit 8
  ["\u0669", "9"], // Arabic-indic digit 9
  ["\u06F0", "0"], // Eastern-Arabic digit 0
  ["\u06F1", "1"], // Eastern-Arabic digit 1
  ["\u06F2", "2"], // Eastern-Arabic digit 2
  ["\u06F3", "3"], // Eastern-Arabic digit 3
  ["\u06F4", "4"], // Eastern-Arabic digit 4
  ["\u06F5", "5"], // Eastern-Arabic digit 5
  ["\u06F6", "6"], // Eastern-Arabic digit 6
  ["\u06F7", "7"], // Eastern-Arabic digit 7
  ["\u06F8", "8"], // Eastern-Arabic digit 8
  ["\u06F9", "9"], // Eastern-Arabic digit 9
]);

var DIGITz = {
  "0": "0",
  "1": "1",
  "2": "2",
  "3": "3",
  "4": "4",
  "5": "5",
  "6": "6",
  "7": "7",
  "8": "8",
  "9": "9",
  "\uFF10": "0", // Fullwidth digit 0
  "\uFF11": "1", // Fullwidth digit 1
  "\uFF12": "2", // Fullwidth digit 2
  "\uFF13": "3", // Fullwidth digit 3
  "\uFF14": "4", // Fullwidth digit 4
  "\uFF15": "5", // Fullwidth digit 5
  "\uFF16": "6", // Fullwidth digit 6
  "\uFF17": "7", // Fullwidth digit 7
  "\uFF18": "8", // Fullwidth digit 8
  "\uFF19": "9", // Fullwidth digit 9
  "\u0660": "0", // Arabic-indic digit 0
  "\u0661": "1", // Arabic-indic digit 1
  "\u0662": "2", // Arabic-indic digit 2
  "\u0663": "3", // Arabic-indic digit 3
  "\u0664": "4", // Arabic-indic digit 4
  "\u0665": "5", // Arabic-indic digit 5
  "\u0666": "6", // Arabic-indic digit 6
  "\u0667": "7", // Arabic-indic digit 7
  "\u0668": "8", // Arabic-indic digit 8
  "\u0669": "9", // Arabic-indic digit 9
  "\u06F0": "0", // Eastern-Arabic digit 0
  "\u06F1": "1", // Eastern-Arabic digit 1
  "\u06F2": "2", // Eastern-Arabic digit 2
  "\u06F3": "3", // Eastern-Arabic digit 3
  "\u06F4": "4", // Eastern-Arabic digit 4
  "\u06F5": "5", // Eastern-Arabic digit 5
  "\u06F6": "6", // Eastern-Arabic digit 6
  "\u06F7": "7", // Eastern-Arabic digit 7
  "\u06F8": "8", // Eastern-Arabic digit 8
  "\u06F9": "9", // Eastern-Arabic digit 9
};

Tests:

Map
let result = ''; for (let i = 0, n = str.length; i < n; ++i) { result += DIGITS.get(str[i]); }
one liner
str.normalize("NFKC").replace(/[٠-٩۰-۹]/g,a=>a.charCodeAt(0)&15);
one liner swapped
str.replace(/[٠-٩۰-۹]/g,a=>a.charCodeAt(0)&15).normalize("NFKC");
ES6 for loop
let result = ''; for (let letter of str) { result += DIGITS.get(letter); }
Using object
let result = ''; for (let letter of str) { result += DIGITz[letter]; }
regex normalize
str.replace(/[\uFF10-\uFF19]/g,a=>a.charCodeAt(a.charCodeAt(0)-0xfee0)).replace(/[٠-٩۰-۹]/g,a=>a.charCodeAt(0)&15);
regex normalize swapped
str.replace(/[٠-٩۰-۹]/g,a=>a.charCodeAt(0)&15).replace(/[\uFF10-\uFF19]/g,a=>a.charCodeAt(a.charCodeAt(0)-0xfee0));

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
Map
one liner
one liner swapped
ES6 for loop
Using object
regex normalize
regex normalize swapped

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: 2 years ago)

User agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/102.0.5005.61 Safari/537.36

Browser/OS: Chrome 102 on Windows

View result in a separate tab

Test name	Executions per second
Map	1099992.9 Ops/sec
one liner	2655456.5 Ops/sec
one liner swapped	2604336.0 Ops/sec
ES6 for loop	1732553.9 Ops/sec
Using object	1558265.4 Ops/sec
regex normalize	1535058.1 Ops/sec
regex normalize swapped	1612221.9 Ops/sec

Autogenerated LLM Summary (model llama3.2:3b, generated 2 months ago):

LLMs can make mistakes. Check important info.

It seems like you're trying to analyze the benchmark results of some code transformations related to string processing in JavaScript.

To provide a helpful answer, I'll need to parse and analyze the provided data. However, since the actual data is not included in your text, I'll assume that you want me to extract relevant information from the given "Benchmark result" string.

From the string, it appears that there are four test cases:

1. **one liner**: With two sub-cases: `one liner` and `one liner swapped`.
2. **ES6 for loop**.
3. **Using object**.
4. **regex normalize** and its swapped version.

To answer your question, I'll need to know what you're specifically looking for in the benchmark results. Are you interested in:

* The execution time for each test case?
* The best-performing test case?
* A comparison between the different approaches (e.g., which one is faster)?
* Something else?

Please clarify your requirements, and I'll do my best to provide a helpful answer based on the provided data.

Related benchmarks:

Normalize digits (version: 7)

Comparing performance of: Map vs one liner vs one liner swapped vs ES6 for loop vs Using object vs regex normalize vs regex normalize swapped

Created: 2 years ago by: Registered User

Jump to the latest result

Map

one liner

one liner swapped

ES6 for loop

Using object

regex normalize

regex normalize swapped

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 2 months ago):