@stablelib/utf8 - npm Package Compare versions

Comparing version 1.0.1 to 1.0.2

lib/utf8.bench.js

@@ -5,9 +5,38 @@ "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-var utf8_1 = require("./utf8");
-var benchmark_1 = require("@stablelib/benchmark");
-var text = "\u2014 \u0415h bien, mon prince. G\u00EAnes et Lucques ne sont plus que des apanages,\ndes \u043F\u043E\u043C\u0435\u0441\u0442\u044C\u044F, de la famille Buonaparte.  Non, je  vous pr\u00E9viens, que si vous\nne  me dites pas, que nous avons la guerre, si vous vous permettez encore de\npallier  toutes les infamies, toutes les  atrocit\u00E9s  de cet  Antichrist  (ma\nparole, j'y  crois) -- je  ne  vous  connais plus, vous n'\u00EAtes plus mon ami,\nvous n'\u00EAtes  plus  \u043C\u043E\u0439  \u0432\u0435\u0440\u043D\u044B\u0439  \u0440\u0430\u0431,  comme  vous  dites.  [1]  \u041D\u0443,\n\u0437\u0434\u0440\u0430\u0432\u0441\u0442\u0432\u0443\u0439\u0442\u0435, \u0437\u0434\u0440\u0430\u0432\u0441\u0442\u0432\u0443\u0439\u0442\u0435.  Je vois  que  je  vous fais  peur, [2]\n\u0441\u0430\u0434\u0438\u0442\u0435\u0441\u044C \u0438 \u0440\u0430\u0441\u0441\u043A\u0430\u0437\u044B\u0432\u0430\u0439\u0442\u0435.\n     \u0422\u0430\u043A \u0433\u043E\u0432\u043E\u0440\u0438\u043B\u0430 \u0432 \u0438\u044E\u043B\u0435 1805 \u0433\u043E\u0434\u0430 \u0438\u0437\u0432\u0435\u0441\u0442\u043D\u0430\u044F \u0410\u043D\u043D\u0430 \u041F\u0430\u0432\u043B\u043E\u0432\u043D\u0430 \u0428\u0435\u0440\u0435\u0440, \u0444\u0440\u0435\u0439\u043B\u0438\u043D\u0430 \u0438\n\u043F\u0440\u0438\u0431\u043B\u0438\u0436\u0435\u043D\u043D\u0430\u044F  \u0438\u043C\u043F\u0435\u0440\u0430\u0442\u0440\u0438\u0446\u044B  \u041C\u0430\u0440\u0438\u0438  \u0424\u0435\u043E\u0434\u043E\u0440\u043E\u0432\u043D\u044B,  \u0432\u0441\u0442\u0440\u0435\u0447\u0430\u044F \u0432\u0430\u0436\u043D\u043E\u0433\u043E \u0438  \u0447\u0438\u043D\u043E\u0432\u043D\u043E\u0433\u043E\n\u043A\u043D\u044F\u0437\u044F  \u0412\u0430\u0441\u0438\u043B\u0438\u044F,  \u043F\u0435\u0440\u0432\u043E\u0433\u043E  \u043F\u0440\u0438\u0435\u0445\u0430\u0432\u0448\u0435\u0433\u043E  \u043D\u0430  \u0435\u0435 \u0432\u0435\u0447\u0435\u0440. \u0410\u043D\u043D\u0430  \u041F\u0430\u0432\u043B\u043E\u0432\u043D\u0430  \u043A\u0430\u0448\u043B\u044F\u043B\u0430\n\u043D\u0435\u0441\u043A\u043E\u043B\u044C\u043A\u043E  \u0434\u043D\u0435\u0439, \u0443  \u043D\u0435\u0435 \u0431\u044B\u043B \u0433\u0440\u0438\u043F\u043F, \u043A\u0430\u043A \u043E\u043D\u0430 \u0433\u043E\u0432\u043E\u0440\u0438\u043B\u0430 (\u0433\u0440\u0438\u043F\u043F  \u0431\u044B\u043B \u0442\u043E\u0433\u0434\u0430  \u043D\u043E\u0432\u043E\u0435\n\u0441\u043B\u043E\u0432\u043E, \u0443\u043F\u043E\u0442\u0440\u0435\u0431\u043B\u044F\u0432\u0448\u0435\u0435\u0441\u044F \u0442\u043E\u043B\u044C\u043A\u043E  \u0440\u0435\u0434\u043A\u0438\u043C\u0438).  \u0412 \u0437\u0430\u043F\u0438\u0441\u043E\u0447\u043A\u0430\u0445, \u0440\u0430\u0437\u043E\u0441\u043B\u0430\u043D\u043D\u044B\u0445  \u0443\u0442\u0440\u043E\u043C \u0441\n\u043A\u0440\u0430\u0441\u043D\u044B\u043C \u043B\u0430\u043A\u0435\u0435\u043C, \u0431\u044B\u043B\u043E \u043D\u0430\u043F\u0438\u0441\u0430\u043D\u043E \u0431\u0435\u0437 \u0440\u0430\u0437\u043B\u0438\u0447\u0438\u044F \u0432\u043E \u0432\u0441\u0435\u0445:\n\n     \"Si vous n'avez rien de mieux \u00E0 faire, M. le comte (\u0438\u043B\u0438 mon prince), et\nsi la perspective de passer la soir\u00E9e chez une pauvre malade ne vous effraye\npas  trop,  je serai charm\u00E9e de vous  voir chez moi  entre 7 et  10  heures.\nAnnette Scherer\".[3]\n\n     \u2014 Dieu, quelle virulente sortie [4] -- \u043E\u0442\u0432\u0435\u0447\u0430\u043B,  \u043D\u0438\u0441\u043A\u043E\u043B\u044C\u043A\u043E \u043D\u0435\n\u0441\u043C\u0443\u0442\u044F\u0441\u044C  \u0442\u0430\u043A\u043E\u044E \u0432\u0441\u0442\u0440\u0435\u0447\u0435\u0439, \u0432\u043E\u0448\u0435\u0434\u0448\u0438\u0439  \u043A\u043D\u044F\u0437\u044C,  \u0432  \u043F\u0440\u0438\u0434\u0432\u043E\u0440\u043D\u043E\u043C, \u0448\u0438\u0442\u043E\u043C  \u043C\u0443\u043D\u0434\u0438\u0440\u0435,  \u0432\n\u0447\u0443\u043B\u043A\u0430\u0445,  \uD83D\uDE06 \uD83D\uDE4F\uD83C\uDFFD  \uD83D\uDCB9 \u0431\u0430\u0448\u043C\u0430\u043A\u0430\u0445,  \u043F\u0440\u0438  \u0437\u0432\u0435\u0437\u0434\u0430\u0445,  \u0441  \u0441\u0432\u0435\u0442\u043B\u044B\u043C \u0432\u044B\u0440\u0430\u0436\u0435\u043D\u0438\u0435\u043C \u043F\u043B\u043E\u0441\u043A\u043E\u0433\u043E  \u043B\u0438\u0446\u0430.\n\u041E\u043D \u0433\u043E\u0432\u043E\u0440\u0438\u043B \u043D\u0430 \u0442\u043E\u043C \u0438\u0437\u044B\u0441\u043A\u0430\u043D\u043D\u043E\u043C \u0444\u0440\u0430\u043D\u0446\u0443\u0437\u0441\u043A\u043E\u043C \u044F\u0437\u044B\u043A\u0435, \u043D\u0430  \u043A\u043E\u0442\u043E\u0440\u043E\u043C \u043D\u0435 \u0442\u043E\u043B\u044C\u043A\u043E \u0433\u043E\u0432\u043E\u0440\u0438\u043B\u0438,\n\u043D\u043E \u0438  \u0434\u0443\u043C\u0430\u043B\u0438  \u043D\u0430\u0448\u0438  \u0434\u0435\u0434\u044B, \u0438 \u0441 \u0442\u0435\u043C\u0438 \u0442\u0438\u0445\u0438\u043C\u0438, \u043F\u043E\u043A\u0440\u043E\u0432\u0438\u0442\u0435\u043B\u044C\u0441\u0442\u0432\u0435\u043D\u043D\u044B\u043C\u0438 \u0438\u043D\u0442\u043E\u043D\u0430\u0446\u0438\u044F\u043C\u0438,\n\u043A\u043E\u0442\u043E\u0440\u044B\u0435  \u0441\u0432\u043E\u0439\u0441\u0442\u0432\u0435\u043D\u043D\u044B  \u0441\u043E\u0441\u0442\u0430\u0440\u0435\u0432\u0448\u0435\u043C\u0443\u0441\u044F  \u0432  \u0441\u0432\u0435\u0442\u0435  \u0438  \u043F\u0440\u0438  \u0434\u0432\u043E\u0440\u0435  \u0437\u043D\u0430\u0447\u0438\u0442\u0435\u043B\u044C\u043D\u043E\u043C\u0443\n\u0447\u0435\u043B\u043E\u0432\u0435\u043A\u0443. \u041E\u043D \u043F\u043E\u0434\u043E\u0448\u0435\u043B \u043A \u0410\u043D\u043D\u0435  \u041F\u0430\u0432\u043B\u043E\u0432\u043D\u0435, \u043F\u043E\u0446\u0435\u043B\u043E\u0432\u0430\u043B \u0435\u0435 \u0440\u0443\u043A\u0443,  \u043F\u043E\u0434\u0441\u0442\u0430\u0432\u0438\u0432 \u0435\u0439 \u0441\u0432\u043E\u044E\n\u043D\u0430\u0434\u0443\u0448\u0435\u043D\u043D\u0443\u044E \u0438 \u0441\u0438\u044F\u044E\u0449\u0443\u044E \u043B\u044B\u0441\u0438\u043D\u0443, \u0438 \u043F\u043E\u043A\u043E\u0439\u043D\u043E \u0443\u0441\u0435\u043B\u0441\u044F \u043D\u0430 \u0434\u0438\u0432\u0430\u043D\u0435.\n\n\u6230\u722D\u8207\u548C\u5E73 \u5217\u592B\u00B7\u6258\u723E\u65AF\u6CF0 \uD835\uDFD8\uD835\uDFD9\uD835\uDFDA\uD835\uDFDB\uD835\uDFDC\uD835\uDFDD\uD835\uDFDE\uD835\uDFDF\uD835\uDFE0\uD835\uDFE1";
-var enc = utf8_1.encode(text);
+const utf8_1 = require("./utf8");
+const benchmark_1 = require("@stablelib/benchmark");
+const text = `— Еh bien, mon prince. Gênes et Lucques ne sont plus que des apanages,
+des поместья, de la famille Buonaparte.  Non, je  vous préviens, que si vous
+ne  me dites pas, que nous avons la guerre, si vous vous permettez encore de
+pallier  toutes les infamies, toutes les  atrocités  de cet  Antichrist  (ma
+parole, j'y  crois) -- je  ne  vous  connais plus, vous n'êtes plus mon ami,
+vous n'êtes  plus  мой  верный  раб,  comme  vous  dites.  [1]  Ну,
+здравствуйте, здравствуйте.  Je vois  que  je  vous fais  peur, [2]
+садитесь и рассказывайте.
+     Так говорила в июле 1805 года известная Анна Павловна Шерер, фрейлина и
+приближенная  императрицы  Марии  Феодоровны,  встречая важного и  чиновного
+князя  Василия,  первого  приехавшего  на  ее вечер. Анна  Павловна  кашляла
+несколько  дней, у  нее был грипп, как она говорила (грипп  был тогда  новое
+слово, употреблявшееся только  редкими).  В записочках, разосланных  утром с
+красным лакеем, было написано без различия во всех:
+
+     "Si vous n'avez rien de mieux à faire, M. le comte (или mon prince), et
+si la perspective de passer la soirée chez une pauvre malade ne vous effraye
+pas  trop,  je serai charmée de vous  voir chez moi  entre 7 et  10  heures.
+Annette Scherer".[3]
+
+     — Dieu, quelle virulente sortie [4] -- отвечал,  нисколько не
+смутясь  такою встречей, вошедший  князь,  в  придворном, шитом  мундире,  в
+чулках,  😆 🙏🏽  💹 башмаках,  при  звездах,  с  светлым выражением плоского  лица.
+Он говорил на том изысканном французском языке, на  котором не только говорили,
+но и  думали  наши  деды, и с теми тихими, покровительственными интонациями,
+которые  свойственны  состаревшемуся  в  свете  и  при  дворе  значительному
+человеку. Он подошел к Анне  Павловне, поцеловал ее руку,  подставив ей свою
+надушенную и сияющую лысину, и покойно уселся на диване.
+
+戰爭與和平 列夫·托爾斯泰 𝟘𝟙𝟚𝟛𝟜𝟝𝟞𝟟𝟠𝟡`;
+const enc = (0, utf8_1.encode)(text);
 // Benchmarks report MiB/s for bytes (UTF-8 encoded).
-benchmark_1.report("utf8 encode", benchmark_1.benchmark(function () { return utf8_1.encode(text); }, enc.length));
-benchmark_1.report("utf8 decode", benchmark_1.benchmark(function () { return utf8_1.decode(enc); }, enc.length));
+(0, benchmark_1.report)("utf8 encode", (0, benchmark_1.benchmark)(() => (0, utf8_1.encode)(text), enc.length));
+(0, benchmark_1.report)("utf8 decode", (0, benchmark_1.benchmark)(() => (0, utf8_1.decode)(enc), enc.length));
 //# sourceMappingURL=utf8.bench.js.map

lib/utf8.js

@@ -5,7 +5,8 @@ "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
+exports.decode = exports.encodedLength = exports.encode = void 0;
 /**
  * Package utf8 implements UTF-8 encoding and decoding.
  */
-var INVALID_UTF16 = "utf8: invalid string";
-var INVALID_UTF8 = "utf8: invalid source encoding";
+const INVALID_UTF16 = "utf8: invalid string";
+const INVALID_UTF8 = "utf8: invalid source encoding";
 /**
@@ -17,8 +18,11 @@ * Encodes the given string into UTF-8 byte array.
     // Calculate result length and allocate output array.
-    // encodedLength() also validates string and throws errors,
+    // encodedLength() validates string and throws errors,
     // so we don't need repeat validation here.
-    var arr = new Uint8Array(encodedLength(s));
-    var pos = 0;
-    for (var i = 0; i < s.length; i++) {
-        var c = s.charCodeAt(i);
+    const arr = new Uint8Array(encodedLength(s));
+    let pos = 0;
+    for (let i = 0; i < s.length; i++) {
+        let c = s.charCodeAt(i);
+        if (c >= 0xd800 && c <= 0xdbff) {
+            c = ((c - 0xd800) << 10) + (s.charCodeAt(++i) - 0xdc00) + 0x10000;
+        }
         if (c < 0x80) {
@@ -28,19 +32,15 @@ arr[pos++] = c;
         else if (c < 0x800) {
-            arr[pos++] = 0xc0 | c >> 6;
-            arr[pos++] = 0x80 | c & 0x3f;
+            arr[pos++] = 0xc0 | (c >> 6);
+            arr[pos++] = 0x80 | (c & 0x3f);
         }
-        else if (c < 0xd800) {
-            arr[pos++] = 0xe0 | c >> 12;
-            arr[pos++] = 0x80 | (c >> 6) & 0x3f;
-            arr[pos++] = 0x80 | c & 0x3f;
+        else if (c < 0x10000) {
+            arr[pos++] = 0xe0 | (c >> 12);
+            arr[pos++] = 0x80 | ((c >> 6) & 0x3f);
+            arr[pos++] = 0x80 | (c & 0x3f);
         }
         else {
-            i++; // get one more character
-            c = (c & 0x3ff) << 10;
-            c |= s.charCodeAt(i) & 0x3ff;
-            c += 0x10000;
-            arr[pos++] = 0xf0 | c >> 18;
-            arr[pos++] = 0x80 | (c >> 12) & 0x3f;
-            arr[pos++] = 0x80 | (c >> 6) & 0x3f;
-            arr[pos++] = 0x80 | c & 0x3f;
+            arr[pos++] = 0xf0 | (c >> 18);
+            arr[pos++] = 0x80 | ((c >> 12) & 0x3f);
+            arr[pos++] = 0x80 | ((c >> 6) & 0x3f);
+            arr[pos++] = 0x80 | (c & 0x3f);
         }
@@ -56,5 +56,17 @@ }
 function encodedLength(s) {
-    var result = 0;
-    for (var i = 0; i < s.length; i++) {
-        var c = s.charCodeAt(i);
+    let result = 0;
+    for (let i = 0; i < s.length; i++) {
+        let c = s.charCodeAt(i);
+        if (c >= 0xd800 && c <= 0xdbff) {
+            // surrogate pair
+            if (i === s.length - 1) {
+                throw new Error(INVALID_UTF16);
+            }
+            i++;
+            const c2 = s.charCodeAt(i);
+            if (c2 < 0xdc00 || c2 > 0xdfff) {
+                throw new Error(INVALID_UTF16);
+            }
+            c = ((c - 0xd800) << 10) + (c2 - 0xdc00) + 0x10000;
+        }
         if (c < 0x80) {
@@ -66,15 +78,8 @@ result += 1;
         }
-        else if (c < 0xd800) {
+        else if (c < 0x10000) {
             result += 3;
         }
-        else if (c <= 0xdfff) {
-            if (i >= s.length - 1) {
-                throw new Error(INVALID_UTF16);
-            }
-            i++; // "eat" next character
+        else {
             result += 4;
         }
-        else {
-            throw new Error(INVALID_UTF16);
-        }
     }
@@ -89,7 +94,7 @@ return result;
 function decode(arr) {
-    var chars = [];
-    for (var i = 0; i < arr.length; i++) {
-        var b = arr[i];
+    const chars = [];
+    for (let i = 0; i < arr.length; i++) {
+        let b = arr[i];
         if (b & 0x80) {
-            var min = void 0;
+            let min;
             if (b < 0xe0) {
@@ -100,3 +105,3 @@ // Need 1 more byte.
                 }
-                var n1 = arr[++i];
+                const n1 = arr[++i];
                 if ((n1 & 0xc0) !== 0x80) {
@@ -113,4 +118,4 @@ throw new Error(INVALID_UTF8);
                 }
-                var n1 = arr[++i];
-                var n2 = arr[++i];
+                const n1 = arr[++i];
+                const n2 = arr[++i];
                 if ((n1 & 0xc0) !== 0x80 || (n2 & 0xc0) !== 0x80) {
@@ -127,5 +132,5 @@ throw new Error(INVALID_UTF8);
                 }
-                var n1 = arr[++i];
-                var n2 = arr[++i];
-                var n3 = arr[++i];
+                const n1 = arr[++i];
+                const n2 = arr[++i];
+                const n3 = arr[++i];
                 if ((n1 & 0xc0) !== 0x80 || (n2 & 0xc0) !== 0x80 || (n3 & 0xc0) !== 0x80) {
@@ -132,0 +137,0 @@ throw new Error(INVALID_UTF8);

lib/utf8.test.js

@@ -5,7 +5,7 @@ "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-var hex = require("@stablelib/hex");
-var utf8_1 = require("./utf8");
-describe("utf8", function () {
-    it("should encode and decode strings", function () {
-        var tests = [
+const hex = require("@stablelib/hex");
+const utf8_1 = require("./utf8");
+describe("utf8", () => {
+    it("should encode and decode strings", () => {
+        const tests = [
             "abcdef",
@@ -17,11 +17,12 @@ "☺☻☹",
             "test 测试 тест",
-            "𝟘𝟙𝟚𝟛𝟜𝟝𝟞𝟟𝟠𝟡"
+            "𝟘𝟙𝟚𝟛𝟜𝟝𝟞𝟟𝟠𝟡",
+            "❤️"
         ];
-        var encoded = tests.map(utf8_1.encode);
-        var decoded = encoded.map(utf8_1.decode);
+        const encoded = tests.map(utf8_1.encode);
+        const decoded = encoded.map(utf8_1.decode);
         expect(decoded).toEqual(tests);
     });
-    it("should not decode malformed bytes", function () {
+    it("should not decode malformed bytes", () => {
         // Source: https://hsivonen.fi/broken-utf-8/test.html
-        var tests = [
+        const tests = [
             // Non-shortest forms for lowest single-byte (U+0000)
@@ -102,6 +103,6 @@ "C0 80",
         ];
-        tests.forEach(function (s, i) {
-            var b = hex.decode(s.replace(/ /g, ""));
-            expect(function () {
-                var x = utf8_1.decode(b);
+        tests.forEach((s, i) => {
+            const b = hex.decode(s.replace(/ /g, ""));
+            expect(() => {
+                const x = (0, utf8_1.decode)(b);
                 // The following will only run in case of unsuccessful test:
@@ -112,9 +113,9 @@ console.log(i, "should not have decoded", s, "to", x);
     });
-    it("should decode a huge string", function () {
-        var s = "";
-        for (var i = 0; i < 1024 * 1024; i++) {
+    it("should decode a huge string", () => {
+        let s = "";
+        for (let i = 0; i < 1024 * 1024; i++) {
             s += "это test";
         }
-        var enc = utf8_1.encode(s);
-        var dec = utf8_1.decode(enc);
+        const enc = (0, utf8_1.encode)(s);
+        const dec = (0, utf8_1.decode)(enc);
         expect(dec).toEqual(s);
@@ -121,0 +122,0 @@ });

package.json

 {
   "name": "@stablelib/utf8",
-  "version": "1.0.1",
+  "version": "1.0.2",
   "description": "UTF-8 encoder and decoder",
@@ -25,3 +25,3 @@ "main": "./lib/utf8.js",
   },
-  "gitHead": "03dadf27703120d54e6be8436525228ee1c4299b"
+  "gitHead": "5243520e343c217b6a751464dec1bc980cb510d8"
 }

tsconfig.json

 {
+    "extends": "../../configs/tsconfig.json",
     "compilerOptions": {
-        "target": "es5",
-        "module": "commonjs",
-        "strict": true,
-        "noUnusedParameters": true,
-        "noImplicitReturns": true,
-        "noUnusedLocals": true,
-        "removeComments": false,
-        "preserveConstEnums": true,
-        "moduleResolution": "node",
-        "newLine": "LF",
-        "sourceMap": true,
-        "declaration": true,
         "outDir": "lib",
-        "lib": [
-            "es5",
-            "es2015.promise",
-            "dom",
-            "scripthost"
-        ]
     },
@@ -23,0 +6,0 @@ "exclude": [

utf8.test.ts

@@ -16,3 +16,4 @@ // Copyright (C) 2016 Dmitry Chestnykh
             "test 测试 тест",
-            "𝟘𝟙𝟚𝟛𝟜𝟝𝟞𝟟𝟠𝟡"
+            "𝟘𝟙𝟚𝟛𝟜𝟝𝟞𝟟𝟠𝟡",
+            "❤️"
         ];
@@ -19,0 +20,0 @@ const encoded = tests.map(encode);

utf8.ts

@@ -17,3 +17,3 @@ // Copyright (C) 2016 Dmitry Chestnykh
     // Calculate result length and allocate output array.
-    // encodedLength() also validates string and throws errors,
+    // encodedLength() validates string and throws errors,
     // so we don't need repeat validation here.
@@ -25,21 +25,19 @@ const arr = new Uint8Array(encodedLength(s));
         let c = s.charCodeAt(i);
+        if (c >= 0xd800 && c <= 0xdbff) {
+            c = ((c - 0xd800) << 10) + (s.charCodeAt(++i) - 0xdc00) + 0x10000;
+        }
         if (c < 0x80) {
             arr[pos++] = c;
         } else if (c < 0x800) {
-            arr[pos++] = 0xc0 | c >> 6;
-            arr[pos++] = 0x80 | c & 0x3f;
-        } else if (c < 0xd800) {
-            arr[pos++] = 0xe0 | c >> 12;
-            arr[pos++] = 0x80 | (c >> 6) & 0x3f;
-            arr[pos++] = 0x80 | c & 0x3f;
+            arr[pos++] = 0xc0 | (c >> 6);
+            arr[pos++] = 0x80 | (c & 0x3f);
+        } else if (c < 0x10000) {
+            arr[pos++] = 0xe0 | (c >> 12);
+            arr[pos++] = 0x80 | ((c >> 6) & 0x3f);
+            arr[pos++] = 0x80 | (c & 0x3f);
         } else {
-            i++; // get one more character
-            c = (c & 0x3ff) << 10;
-            c |= s.charCodeAt(i) & 0x3ff;
-            c += 0x10000;
-
-            arr[pos++] = 0xf0 | c >> 18;
-            arr[pos++] = 0x80 | (c >> 12) & 0x3f;
-            arr[pos++] = 0x80 | (c >> 6) & 0x3f;
-            arr[pos++] = 0x80 | c & 0x3f;
+            arr[pos++] = 0xf0 | (c >> 18);
+            arr[pos++] = 0x80 | ((c >> 12) & 0x3f);
+            arr[pos++] = 0x80 | ((c >> 6) & 0x3f);
+            arr[pos++] = 0x80 | (c & 0x3f);
         }
@@ -57,3 +55,17 @@ }
     for (let i = 0; i < s.length; i++) {
-        const c = s.charCodeAt(i);
+        let c = s.charCodeAt(i);
+
+        if (c >= 0xd800 && c <= 0xdbff) {
+            // surrogate pair
+            if (i === s.length - 1) {
+                throw new Error(INVALID_UTF16);
+            }
+            i++;
+            const c2 = s.charCodeAt(i);
+            if (c2 < 0xdc00 || c2 > 0xdfff) {
+                throw new Error(INVALID_UTF16);
+            }
+            c = ((c - 0xd800) << 10) + (c2 - 0xdc00) + 0x10000;
+        }
+
         if (c < 0x80) {
@@ -63,12 +75,6 @@ result += 1;
             result += 2;
-        } else if (c < 0xd800) {
+        } else if (c < 0x10000) {
             result += 3;
-        } else if (c <= 0xdfff) {
-            if (i >= s.length - 1) {
-                throw new Error(INVALID_UTF16);
-            }
-            i++; // "eat" next character
+        } else {
             result += 4;
-        } else {
-            throw new Error(INVALID_UTF16);
         }
@@ -75,0 +81,0 @@ }

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Long strings

Supply chain risk

Contains long string literals, which may be a sign of obfuscated or packed code.

Found 1 instance in 1 package

Minified code

Quality

This package contains minified code. This may be harmless in some cases where minified code is included in packaged libraries, however packages on npm should not minify code.

Found 1 instance in 1 package

Improved metrics

Number of low quality alerts

0

decreased by-100%

Number of low supply chain risk alerts

1

decreased by-50%

Worsened metrics

Total package byte prevSize

34544

decreased by-7.96%

Lines of code

626

decreased by-0.32%

No dependency changes