entities 2.0.3 → 3.0.1

package/lib/decode.d.ts

@@ -1,7 +1,15 @@
-export declare const decodeXML: (str: string) => string;
-export declare const decodeHTMLStrict: (str: string) => string;
-export interface MapType {
-    [key: string]: string;
+import htmlDecodeTree from "./generated/decode-data-html";
+import xmlDecodeTree from "./generated/decode-data-xml";
+export { htmlDecodeTree, xmlDecodeTree };
+export declare enum BinTrieFlags {
+    HAS_VALUE = 32768,
+    BRANCH_LENGTH = 32512,
+    MULTI_BYTE = 128,
+    JUMP_TABLE = 127
 }
-export declare const decodeHTML: (str: string) => string;
+export declare const JUMP_OFFSET_BASE: number;
+export declare function determineBranch(decodeTree: Uint16Array, current: number, nodeIdx: number, char: number): number;
+export declare function decodeHTML(str: string): string;
+export declare function decodeHTMLStrict(str: string): string;
+export declare function decodeXML(str: string): string;
 //# sourceMappingURL=decode.d.ts.map

package/lib/decode.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"decode.d.ts","sourceRoot":"","sources":["../src/decode.ts"],"names":[],"mappings":"AAKA,eAAO,MAAM,SAAS,QAeL,MAAM,WAf0B,CAAC;AAClD,eAAO,MAAM,gBAAgB,QAcZ,MAAM,WAdoC,CAAC;AAE5D,MAAM,WAAW,OAAO;IACpB,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CAAC;CACzB;AAeD,eAAO,MAAM,UAAU,QAyBN,MAAM,WACnB,CAAC"}
+{"version":3,"file":"decode.d.ts","sourceRoot":"","sources":["../src/decode.ts"],"names":[],"mappings":"AAAA,OAAO,cAAc,MAAM,8BAA8B,CAAC;AAC1D,OAAO,aAAa,MAAM,6BAA6B,CAAC;AAIxD,OAAO,EAAE,cAAc,EAAE,aAAa,EAAE,CAAC;AAczC,oBAAY,YAAY;IACpB,SAAS,QAAwB;IACjC,aAAa,QAAwB;IACrC,UAAU,MAAwB;IAClC,UAAU,MAAwB;CACrC;AAED,eAAO,MAAM,gBAAgB,QAAqB,CAAC;AAmGnD,wBAAgB,eAAe,CAC3B,UAAU,EAAE,WAAW,EACvB,OAAO,EAAE,MAAM,EACf,OAAO,EAAE,MAAM,EACf,IAAI,EAAE,MAAM,GACb,MAAM,CA0CR;AAKD,wBAAgB,UAAU,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CAE9C;AAED,wBAAgB,gBAAgB,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CAEpD;AAED,wBAAgB,SAAS,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CAE7C"}

package/lib/decode.js

@@ -3,52 +3,143 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.decodeHTML = exports.decodeHTMLStrict = exports.decodeXML = void 0;
-var entities_json_1 = __importDefault(require("./maps/entities.json"));
-var legacy_json_1 = __importDefault(require("./maps/legacy.json"));
-var xml_json_1 = __importDefault(require("./maps/xml.json"));
+exports.decodeXML = exports.decodeHTMLStrict = exports.decodeHTML = exports.determineBranch = exports.JUMP_OFFSET_BASE = exports.BinTrieFlags = exports.xmlDecodeTree = exports.htmlDecodeTree = void 0;
+var decode_data_html_1 = __importDefault(require("./generated/decode-data-html"));
+exports.htmlDecodeTree = decode_data_html_1.default;
+var decode_data_xml_1 = __importDefault(require("./generated/decode-data-xml"));
+exports.xmlDecodeTree = decode_data_xml_1.default;
 var decode_codepoint_1 = __importDefault(require("./decode_codepoint"));
-exports.decodeXML = getStrictDecoder(xml_json_1.default);
-exports.decodeHTMLStrict = getStrictDecoder(entities_json_1.default);
-function getStrictDecoder(map) {
-    var keys = Object.keys(map).join("|");
-    var replace = getReplacer(map);
-    keys += "|#[xX][\\da-fA-F]+|#\\d+";
-    var re = new RegExp("&(?:" + keys + ");", "g");
-    return function (str) { return String(str).replace(re, replace); };
+var BinTrieFlags;
+(function (BinTrieFlags) {
+    BinTrieFlags[BinTrieFlags["HAS_VALUE"] = 32768] = "HAS_VALUE";
+    BinTrieFlags[BinTrieFlags["BRANCH_LENGTH"] = 32512] = "BRANCH_LENGTH";
+    BinTrieFlags[BinTrieFlags["MULTI_BYTE"] = 128] = "MULTI_BYTE";
+    BinTrieFlags[BinTrieFlags["JUMP_TABLE"] = 127] = "JUMP_TABLE";
+})(BinTrieFlags = exports.BinTrieFlags || (exports.BinTrieFlags = {}));
+exports.JUMP_OFFSET_BASE = 48 /* ZERO */ - 1;
+function getDecoder(decodeTree) {
+    return function decodeHTMLBinary(str, strict) {
+        var ret = "";
+        var lastIdx = 0;
+        var strIdx = 0;
+        while ((strIdx = str.indexOf("&", strIdx)) >= 0) {
+            ret += str.slice(lastIdx, strIdx);
+            lastIdx = strIdx;
+            // Skip the "&"
+            strIdx += 1;
+            // If we have a numeric entity, handle this separately.
+            if (str.charCodeAt(strIdx) === 35 /* NUM */) {
+                // Skip the leading "&#". For hex entities, also skip the leading "x".
+                var start = strIdx + 1;
+                var base = 10;
+                var cp = str.charCodeAt(start);
+                if ((cp | 32 /* To_LOWER_BIT */) === 120 /* LOWER_X */) {
+                    base = 16;
+                    strIdx += 1;
+                    start += 1;
+                }
+                while (((cp = str.charCodeAt(++strIdx)) >= 48 /* ZERO */ &&
+                    cp <= 57 /* NINE */) ||
+                    (base === 16 &&
+                        (cp | 32 /* To_LOWER_BIT */) >= 97 /* LOWER_A */ &&
+                        (cp | 32 /* To_LOWER_BIT */) <= 102 /* LOWER_F */))
+                    ;
+                if (start !== strIdx) {
+                    var entity = str.substring(start, strIdx);
+                    var parsed = parseInt(entity, base);
+                    if (str.charCodeAt(strIdx) === 59 /* SEMI */) {
+                        strIdx += 1;
+                    }
+                    else if (strict) {
+                        continue;
+                    }
+                    ret += decode_codepoint_1.default(parsed);
+                    lastIdx = strIdx;
+                }
+                continue;
+            }
+            var result = null;
+            var excess = 1;
+            var treeIdx = 0;
+            var current = decodeTree[treeIdx];
+            for (; strIdx < str.length; strIdx++, excess++) {
+                treeIdx = determineBranch(decodeTree, current, treeIdx + 1, str.charCodeAt(strIdx));
+                if (treeIdx < 0)
+                    break;
+                current = decodeTree[treeIdx];
+                // If the branch is a value, store it and continue
+                if (current & BinTrieFlags.HAS_VALUE) {
+                    // If we have a legacy entity while parsing strictly, just skip the number of bytes
+                    if (strict && str.charCodeAt(strIdx) !== 59 /* SEMI */) {
+                        // No need to consider multi-byte values, as the legacy entity is always a single byte
+                        treeIdx += 1;
+                    }
+                    else {
+                        // If this is a surrogate pair, combine the higher bits from the node with the next byte
+                        result =
+                            current & BinTrieFlags.MULTI_BYTE
+                                ? String.fromCharCode(decodeTree[++treeIdx], decodeTree[++treeIdx])
+                                : String.fromCharCode(decodeTree[++treeIdx]);
+                        excess = 0;
+                    }
+                }
+            }
+            if (result != null) {
+                ret += result;
+                lastIdx = strIdx - excess + 1;
+            }
+        }
+        return ret + str.slice(lastIdx);
+    };
 }
-var sorter = function (a, b) { return (a < b ? 1 : -1); };
-exports.decodeHTML = (function () {
-    var legacy = Object.keys(legacy_json_1.default).sort(sorter);
-    var keys = Object.keys(entities_json_1.default).sort(sorter);
-    for (var i = 0, j = 0; i < keys.length; i++) {
-        if (legacy[j] === keys[i]) {
-            keys[i] += ";?";
-            j++;
+function determineBranch(decodeTree, current, nodeIdx, char) {
+    if (current <= 128) {
+        return char === current ? nodeIdx : -1;
+    }
+    var branchCount = (current & BinTrieFlags.BRANCH_LENGTH) >> 8;
+    if (branchCount === 0) {
+        return -1;
+    }
+    if (branchCount === 1) {
+        return char === decodeTree[nodeIdx] ? nodeIdx + 1 : -1;
+    }
+    var jumpOffset = current & BinTrieFlags.JUMP_TABLE;
+    if (jumpOffset) {
+        var value = char - exports.JUMP_OFFSET_BASE - jumpOffset;
+        return value < 0 || value > branchCount
+            ? -1
+            : decodeTree[nodeIdx + value] - 1;
+    }
+    // Binary search for the character.
+    var lo = nodeIdx;
+    var hi = lo + branchCount - 1;
+    while (lo <= hi) {
+        var mid = (lo + hi) >>> 1;
+        var midVal = decodeTree[mid];
+        if (midVal < char) {
+            lo = mid + 1;
+        }
+        else if (midVal > char) {
+            hi = mid - 1;
         }
         else {
-            keys[i] += ";";
+            return decodeTree[mid + branchCount];
         }
     }
-    var re = new RegExp("&(?:" + keys.join("|") + "|#[xX][\\da-fA-F]+;?|#\\d+;?)", "g");
-    var replace = getReplacer(entities_json_1.default);
-    function replacer(str) {
-        if (str.substr(-1) !== ";")
-            str += ";";
-        return replace(str);
-    }
-    //TODO consider creating a merged map
-    return function (str) { return String(str).replace(re, replacer); };
-})();
-function getReplacer(map) {
-    return function replace(str) {
-        if (str.charAt(1) === "#") {
-            var secondChar = str.charAt(2);
-            if (secondChar === "X" || secondChar === "x") {
-                return decode_codepoint_1.default(parseInt(str.substr(3), 16));
-            }
-            return decode_codepoint_1.default(parseInt(str.substr(2), 10));
-        }
-        return map[str.slice(1, -1)];
-    };
+    return -1;
+}
+exports.determineBranch = determineBranch;
+var htmlDecoder = getDecoder(decode_data_html_1.default);
+var xmlDecoder = getDecoder(decode_data_xml_1.default);
+function decodeHTML(str) {
+    return htmlDecoder(str, false);
+}
+exports.decodeHTML = decodeHTML;
+function decodeHTMLStrict(str) {
+    return htmlDecoder(str, true);
+}
+exports.decodeHTMLStrict = decodeHTMLStrict;
+function decodeXML(str) {
+    return xmlDecoder(str, true);
 }
+exports.decodeXML = decodeXML;

package/lib/decode_codepoint.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"decode_codepoint.d.ts","sourceRoot":"","sources":["../src/decode_codepoint.ts"],"names":[],"mappings":"AAGA,MAAM,CAAC,OAAO,UAAU,eAAe,CAAC,SAAS,EAAE,MAAM,UAmBxD"}
+{"version":3,"file":"decode_codepoint.d.ts","sourceRoot":"","sources":["../src/decode_codepoint.ts"],"names":[],"mappings":"AAmDA,MAAM,CAAC,OAAO,UAAU,eAAe,CAAC,SAAS,EAAE,MAAM,GAAG,MAAM,CAMjE"}

package/lib/decode_codepoint.js

@@ -1,24 +1,54 @@
 "use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
+// Adapted from https://github.com/mathiasbynens/he/blob/36afe179392226cf1b6ccdb16ebbb7a5a844d93a/src/he.js#L106-L134
 Object.defineProperty(exports, "__esModule", { value: true });
-var decode_json_1 = __importDefault(require("./maps/decode.json"));
-// modified version of https://github.com/mathiasbynens/he/blob/master/src/he.js#L94-L119
+var decodeMap = new Map([
+    [0, 65533],
+    [128, 8364],
+    [130, 8218],
+    [131, 402],
+    [132, 8222],
+    [133, 8230],
+    [134, 8224],
+    [135, 8225],
+    [136, 710],
+    [137, 8240],
+    [138, 352],
+    [139, 8249],
+    [140, 338],
+    [142, 381],
+    [145, 8216],
+    [146, 8217],
+    [147, 8220],
+    [148, 8221],
+    [149, 8226],
+    [150, 8211],
+    [151, 8212],
+    [152, 732],
+    [153, 8482],
+    [154, 353],
+    [155, 8250],
+    [156, 339],
+    [158, 382],
+    [159, 376],
+]);
+var fromCodePoint = 
+// eslint-disable-next-line @typescript-eslint/no-unnecessary-condition, node/no-unsupported-features/es-builtins
+String.fromCodePoint ||
+    function (codePoint) {
+        var output = "";
+        if (codePoint > 0xffff) {
+            codePoint -= 0x10000;
+            output += String.fromCharCode(((codePoint >>> 10) & 0x3ff) | 0xd800);
+            codePoint = 0xdc00 | (codePoint & 0x3ff);
+        }
+        output += String.fromCharCode(codePoint);
+        return output;
+    };
 function decodeCodePoint(codePoint) {
+    var _a;
     if ((codePoint >= 0xd800 && codePoint <= 0xdfff) || codePoint > 0x10ffff) {
         return "\uFFFD";
     }
-    if (codePoint in decode_json_1.default) {
-        codePoint = decode_json_1.default[codePoint];
-    }
-    var output = "";
-    if (codePoint > 0xffff) {
-        codePoint -= 0x10000;
-        output += String.fromCharCode(((codePoint >>> 10) & 0x3ff) | 0xd800);
-        codePoint = 0xdc00 | (codePoint & 0x3ff);
-    }
-    output += String.fromCharCode(codePoint);
-    return output;
+    return fromCodePoint((_a = decodeMap.get(codePoint)) !== null && _a !== void 0 ? _a : codePoint);
 }
 exports.default = decodeCodePoint;

package/lib/encode-trie.d.ts

@@ -0,0 +1,8 @@
+export declare const getCodePoint: (str: string, index: number) => number;
+export declare function encodeHTMLTrieRe(regExp: RegExp, str: string): string;
+export interface TrieNode {
+    value?: string;
+    next?: Map<number, TrieNode>;
+}
+export declare function getTrie(map: Record<string, string>): Map<number, TrieNode>;
+//# sourceMappingURL=encode-trie.d.ts.map

package/lib/encode-trie.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"encode-trie.d.ts","sourceRoot":"","sources":["../src/encode-trie.ts"],"names":[],"mappings":"AAYA,eAAO,MAAM,YAAY,QAGT,MAAM,SAAS,MAAM,KAAG,MAQD,CAAC;AAIxC,wBAAgB,gBAAgB,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,MAAM,CAkCpE;AAED,MAAM,WAAW,QAAQ;IACrB,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,IAAI,CAAC,EAAE,GAAG,CAAC,MAAM,EAAE,QAAQ,CAAC,CAAC;CAChC;AAED,wBAAgB,OAAO,CAAC,GAAG,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,GAAG,GAAG,CAAC,MAAM,EAAE,QAAQ,CAAC,CAmB1E"}

package/lib/encode-trie.js

@@ -0,0 +1,77 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.getTrie = exports.encodeHTMLTrieRe = exports.getCodePoint = void 0;
+var entities_json_1 = __importDefault(require("./maps/entities.json"));
+function isHighSurrugate(c) {
+    return (c & 64512 /* Mask */) === 55296 /* High */;
+}
+// For compatibility with node < 4, we wrap `codePointAt`
+exports.getCodePoint = 
+// eslint-disable-next-line @typescript-eslint/no-unnecessary-condition
+String.prototype.codePointAt != null
+    ? function (str, index) { return str.codePointAt(index); }
+    : // http://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
+        function (c, index) {
+            return isHighSurrugate(c.charCodeAt(index))
+                ? (c.charCodeAt(index) - 55296 /* High */) * 0x400 +
+                    c.charCodeAt(index + 1) -
+                    0xdc00 +
+                    0x10000
+                : c.charCodeAt(index);
+        };
+var htmlTrie = getTrie(entities_json_1.default);
+function encodeHTMLTrieRe(regExp, str) {
+    var _a;
+    var ret = "";
+    var lastIdx = 0;
+    var match;
+    while ((match = regExp.exec(str)) !== null) {
+        var i = match.index;
+        var char = str.charCodeAt(i);
+        var next = htmlTrie.get(char);
+        if (next) {
+            if (next.next != null && i + 1 < str.length) {
+                var value = (_a = next.next.get(str.charCodeAt(i + 1))) === null || _a === void 0 ? void 0 : _a.value;
+                if (value != null) {
+                    ret += str.substring(lastIdx, i) + value;
+                    regExp.lastIndex += 1;
+                    lastIdx = i + 2;
+                    continue;
+                }
+            }
+            ret += str.substring(lastIdx, i) + next.value;
+            lastIdx = i + 1;
+        }
+        else {
+            ret += str.substring(lastIdx, i) + "&#x" + exports.getCodePoint(str, i).toString(16) + ";";
+            // Increase by 1 if we have a surrogate pair
+            lastIdx = regExp.lastIndex += Number(isHighSurrugate(char));
+        }
+    }
+    return ret + str.substr(lastIdx);
+}
+exports.encodeHTMLTrieRe = encodeHTMLTrieRe;
+function getTrie(map) {
+    var _a, _b, _c, _d;
+    var trie = new Map();
+    for (var _i = 0, _e = Object.keys(map); _i < _e.length; _i++) {
+        var value = _e[_i];
+        var key = map[value];
+        // Resolve the key
+        var lastMap = trie;
+        for (var i = 0; i < key.length - 1; i++) {
+            var char = key.charCodeAt(i);
+            var next = (_a = lastMap.get(char)) !== null && _a !== void 0 ? _a : {};
+            lastMap.set(char, next);
+            lastMap = (_b = next.next) !== null && _b !== void 0 ? _b : (next.next = new Map());
+        }
+        var val = (_c = lastMap.get(key.charCodeAt(key.length - 1))) !== null && _c !== void 0 ? _c : {};
+        (_d = val.value) !== null && _d !== void 0 ? _d : (val.value = "&" + value + ";");
+        lastMap.set(key.charCodeAt(key.length - 1), val);
+    }
+    return trie;
+}
+exports.getTrie = getTrie;

package/lib/encode.d.ts

@@ -1,4 +1,46 @@
-export declare const encodeXML: (data: string) => string;
-export declare const encodeHTML: (data: string) => string;
-export declare function escape(data: string): string;
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in XML
+ * documents using XML entities.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `ü`) will be used.
+ */
+export declare function encodeXML(str: string): string;
+/**
+ * Encodes all entities and non-ASCII characters in the input.
+ *
+ * This includes characters that are valid ASCII characters in HTML documents.
+ * For example `#` will be encoded as `#`. To get a more compact output,
+ * consider using the `encodeNonAsciiHTML` function.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `ü`) will be used.
+ */
+export declare function encodeHTML(data: string): string;
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in HTML
+ * documents using HTML entities.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `ü`) will be used.
+ */
+export declare function encodeNonAsciiHTML(data: string): string;
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in XML
+ * documents using numeric hexadecimal reference (eg. `ü`).
+ *
+ * Have a look at `escapeUTF8` if you want a more concise output at the expense
+ * of reduced transportability.
+ *
+ * @param data String to escape.
+ */
+export declare const escape: typeof encodeXML;
+/**
+ * Encodes all characters not valid in XML documents using XML entities.
+ *
+ * Note that the output will be character-set dependent.
+ *
+ * @param data String to escape.
+ */
+export declare function escapeUTF8(data: string): string;
 //# sourceMappingURL=encode.d.ts.map

package/lib/encode.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"encode.d.ts","sourceRoot":"","sources":["../src/encode.ts"],"names":[],"mappings":"AAKA,eAAO,MAAM,SAAS,SAmEJ,MAAM,WAnEoC,CAAC;AAO7D,eAAO,MAAM,UAAU,SA4DL,MAAM,WA5DuC,CAAC;AAoEhE,wBAAgB,MAAM,CAAC,IAAI,EAAE,MAAM,UAIlC"}
+{"version":3,"file":"encode.d.ts","sourceRoot":"","sources":["../src/encode.ts"],"names":[],"mappings":"AAgBA;;;;;;GAMG;AACH,wBAAgB,SAAS,CAAC,GAAG,EAAE,MAAM,GAAG,MAAM,CA0B7C;AAED;;;;;;;;;GASG;AACH,wBAAgB,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAE/C;AACD;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAEvD;AAoCD;;;;;;;;GAQG;AACH,eAAO,MAAM,MAAM,kBAAY,CAAC;AAEhC;;;;;;GAMG;AACH,wBAAgB,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAkB/C"}

package/lib/encode.js

@@ -3,71 +3,124 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.escape = exports.encodeHTML = exports.encodeXML = void 0;
+exports.escapeUTF8 = exports.escape = exports.encodeNonAsciiHTML = exports.encodeHTML = exports.encodeXML = void 0;
 var xml_json_1 = __importDefault(require("./maps/xml.json"));
-var inverseXML = getInverseObj(xml_json_1.default);
-var xmlReplacer = getInverseReplacer(inverseXML);
-exports.encodeXML = getInverse(inverseXML, xmlReplacer);
+var encode_trie_1 = require("./encode-trie");
 var entities_json_1 = __importDefault(require("./maps/entities.json"));
-var inverseHTML = getInverseObj(entities_json_1.default);
-var htmlReplacer = getInverseReplacer(inverseHTML);
-exports.encodeHTML = getInverse(inverseHTML, htmlReplacer);
-function getInverseObj(obj) {
-    return Object.keys(obj)
-        .sort()
-        .reduce(function (inverse, name) {
-        inverse[obj[name]] = "&" + name + ";";
-        return inverse;
-    }, {});
-}
-function getInverseReplacer(inverse) {
-    var single = [];
-    var multiple = [];
-    for (var _i = 0, _a = Object.keys(inverse); _i < _a.length; _i++) {
-        var k = _a[_i];
-        if (k.length === 1) {
-            // Add value to single array
-            single.push("\\" + k);
+var htmlReplacer = getCharRegExp(entities_json_1.default, true);
+var xmlReplacer = getCharRegExp(xml_json_1.default, true);
+var xmlInvalidChars = getCharRegExp(xml_json_1.default, false);
+var xmlCodeMap = new Map(Object.keys(xml_json_1.default).map(function (k) { return [
+    xml_json_1.default[k].charCodeAt(0),
+    "&" + k + ";",
+]; }));
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in XML
+ * documents using XML entities.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `&#xfc;`) will be used.
+ */
+function encodeXML(str) {
+    var ret = "";
+    var lastIdx = 0;
+    var match;
+    while ((match = xmlReplacer.exec(str)) !== null) {
+        var i = match.index;
+        var char = str.charCodeAt(i);
+        var next = xmlCodeMap.get(char);
+        if (next) {
+            ret += str.substring(lastIdx, i) + next;
+            lastIdx = i + 1;
         }
         else {
-            // Add value to multiple array
-            multiple.push(k);
+            ret += str.substring(lastIdx, i) + "&#x" + encode_trie_1.getCodePoint(str, i).toString(16) + ";";
+            // Increase by 1 if we have a surrogate pair
+            lastIdx = xmlReplacer.lastIndex += Number((char & 65408) === 0xd800);
         }
     }
+    return ret + str.substr(lastIdx);
+}
+exports.encodeXML = encodeXML;
+/**
+ * Encodes all entities and non-ASCII characters in the input.
+ *
+ * This includes characters that are valid ASCII characters in HTML documents.
+ * For example `#` will be encoded as `&num;`. To get a more compact output,
+ * consider using the `encodeNonAsciiHTML` function.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `&#xfc;`) will be used.
+ */
+function encodeHTML(data) {
+    return encode_trie_1.encodeHTMLTrieRe(htmlReplacer, data);
+}
+exports.encodeHTML = encodeHTML;
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in HTML
+ * documents using HTML entities.
+ *
+ * If a character has no equivalent entity, a
+ * numeric hexadecimal reference (eg. `&#xfc;`) will be used.
+ */
+function encodeNonAsciiHTML(data) {
+    return encode_trie_1.encodeHTMLTrieRe(xmlReplacer, data);
+}
+exports.encodeNonAsciiHTML = encodeNonAsciiHTML;
+function getCharRegExp(map, nonAscii) {
+    // Collect the start characters of all entities
+    var chars = Object.keys(map)
+        .map(function (k) { return "\\" + map[k].charAt(0); })
+        .filter(function (v) { return !nonAscii || v.charCodeAt(1) < 128; })
+        .sort(function (a, b) { return a.charCodeAt(1) - b.charCodeAt(1); })
+        // Remove duplicates
+        .filter(function (v, i, a) { return v !== a[i + 1]; });
     // Add ranges to single characters.
-    single.sort();
-    for (var start = 0; start < single.length - 1; start++) {
+    for (var start = 0; start < chars.length - 1; start++) {
         // Find the end of a run of characters
         var end = start;
-        while (end < single.length - 1 &&
-            single[end].charCodeAt(1) + 1 === single[end + 1].charCodeAt(1)) {
+        while (end < chars.length - 1 &&
+            chars[end].charCodeAt(1) + 1 === chars[end + 1].charCodeAt(1)) {
             end += 1;
         }
         var count = 1 + end - start;
         // We want to replace at least three characters
         if (count < 3)
             continue;
-        single.splice(start, count, single[start] + "-" + single[end]);
+        chars.splice(start, count, chars[start] + "-" + chars[end]);
     }
-    multiple.unshift("[" + single.join("") + "]");
-    return new RegExp(multiple.join("|"), "g");
+    return new RegExp("[" + chars.join("") + (nonAscii ? "\\x80-\\uFFFF" : "") + "]", "g");
 }
-var reNonASCII = /(?:[\x80-\uD7FF\uE000-\uFFFF]|[\uD800-\uDBFF][\uDC00-\uDFFF]|[\uD800-\uDBFF](?![\uDC00-\uDFFF])|(?:[^\uD800-\uDBFF]|^)[\uDC00-\uDFFF])/g;
-function singleCharReplacer(c) {
-    // eslint-disable-next-line @typescript-eslint/no-non-null-assertion
-    return "&#x" + c.codePointAt(0).toString(16).toUpperCase() + ";";
-}
-function getInverse(inverse, re) {
-    return function (data) {
-        return data
-            .replace(re, function (name) { return inverse[name]; })
-            .replace(reNonASCII, singleCharReplacer);
-    };
-}
-var reXmlChars = getInverseReplacer(inverseXML);
-function escape(data) {
-    return data
-        .replace(reXmlChars, singleCharReplacer)
-        .replace(reNonASCII, singleCharReplacer);
+/**
+ * Encodes all non-ASCII characters, as well as characters not valid in XML
+ * documents using numeric hexadecimal reference (eg. `&#xfc;`).
+ *
+ * Have a look at `escapeUTF8` if you want a more concise output at the expense
+ * of reduced transportability.
+ *
+ * @param data String to escape.
+ */
+exports.escape = encodeXML;
+/**
+ * Encodes all characters not valid in XML documents using XML entities.
+ *
+ * Note that the output will be character-set dependent.
+ *
+ * @param data String to escape.
+ */
+function escapeUTF8(data) {
+    var match;
+    var lastIdx = 0;
+    var result = "";
+    while ((match = xmlInvalidChars.exec(data))) {
+        if (lastIdx !== match.index) {
+            result += data.substring(lastIdx, match.index);
+        }
+        // We know that this chararcter will be in `inverseXML`
+        result += xmlCodeMap.get(match[0].charCodeAt(0));
+        // Every match will be of length 1
+        lastIdx = match.index + 1;
+    }
+    return result + data.substring(lastIdx);
 }
-exports.escape = escape;
+exports.escapeUTF8 = escapeUTF8;

package/lib/generated/decode-data-html.d.ts

@@ -0,0 +1,3 @@
+declare const _default: Uint16Array;
+export default _default;
+//# sourceMappingURL=decode-data-html.d.ts.map

package/lib/generated/decode-data-html.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"decode-data-html.d.ts","sourceRoot":"","sources":["../../src/generated/decode-data-html.ts"],"names":[],"mappings":";AAEA,wBAAox9E"}