node-html-parser 6.1.15-0 → 7.0.1

package/CHANGELOG.md

@@ -2,6 +2,29 @@
 
 All notable changes to this project will be documented in this file. See [standard-version](https://github.com/conventional-changelog/standard-version) for commit guidelines.
 
+### [7.0.1](https://github.com/taoqf/node-fast-html-parser/compare/v7.0.0...v7.0.1) (2024-12-26)
+
+
+### Bug Fixes
+
+* upgrade node version ([efd77ff](https://github.com/taoqf/node-fast-html-parser/commit/efd77ff93593922512b12216984ba778b2f46593))
+
+## [7.0.0](https://github.com/taoqf/node-fast-html-parser/compare/v6.1.14...v7.0.0) (2024-12-26)
+
+
+### ⚠ BREAKING CHANGES
+
+* fix #277, for that change estarget to es6
+
+### Features
+
+* fix [#277](https://github.com/taoqf/node-fast-html-parser/issues/277), for that change estarget to es6 ([432a3e7](https://github.com/taoqf/node-fast-html-parser/commit/432a3e71ba219e76188bcc2e89e525e40911d164))
+
+
+### Bug Fixes
+
+* add tests for [#227](https://github.com/taoqf/node-fast-html-parser/issues/227) ([5856ee2](https://github.com/taoqf/node-fast-html-parser/commit/5856ee2ef2a0dfef43d75f6d4d13c37c213f25cf))
+
 ### [6.1.14](https://github.com/taoqf/node-fast-html-parser/compare/v6.1.13...v6.1.14) (2024-05-14)
 
 

package/README.md

@@ -6,7 +6,7 @@ DOM tree, with element query support.
 Per the design, it intends to parse massive HTML files in lowest price, thus the
 performance is the top priority.  For this reason, some malformatted HTML may not
 be able to parse correctly, but most usual errors are covered (eg. HTML4 style
-no closing `<li>`, `<td>` etc).
+no closing `<td>` etc).
 
 ## Install
 
@@ -186,7 +186,6 @@ Node --|> TextNode
 Node ..> ClassList
 ```
 
-
 ## HTMLElement Methods
 
 ### trimRight()
@@ -316,7 +315,6 @@ Clone a node.
 
 Get element by it's ID.
 
-
 ## HTMLElement Properties
 
 ### text

package/dist/index.js

@@ -4,21 +4,20 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.NodeType = exports.TextNode = exports.Node = exports.valid = exports.CommentNode = exports.HTMLElement = exports.parse = void 0;
-var comment_1 = __importDefault(require("./nodes/comment"));
+const comment_1 = __importDefault(require("./nodes/comment"));
 exports.CommentNode = comment_1.default;
-var html_1 = __importDefault(require("./nodes/html"));
+const html_1 = __importDefault(require("./nodes/html"));
 exports.HTMLElement = html_1.default;
-var node_1 = __importDefault(require("./nodes/node"));
+const node_1 = __importDefault(require("./nodes/node"));
 exports.Node = node_1.default;
-var text_1 = __importDefault(require("./nodes/text"));
+const text_1 = __importDefault(require("./nodes/text"));
 exports.TextNode = text_1.default;
-var type_1 = __importDefault(require("./nodes/type"));
+const type_1 = __importDefault(require("./nodes/type"));
 exports.NodeType = type_1.default;
-var parse_1 = __importDefault(require("./parse"));
-var valid_1 = __importDefault(require("./valid"));
+const parse_1 = __importDefault(require("./parse"));
+const valid_1 = __importDefault(require("./valid"));
 exports.valid = valid_1.default;
-function parse(data, options) {
-    if (options === void 0) { options = {}; }
+function parse(data, options = {}) {
     return (0, parse_1.default)(data, options);
 }
 exports.default = parse;

package/dist/main.js

@@ -535,7 +535,11 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
             if (attr == null) {
                 return 'null';
             }
-            return JSON.stringify(attr.replace(/"/g, '"')).replace(/\\t/g, '\t').replace(/\\n/g, '\n').replace(/\\r/g, '\r').replace(/\\/g, '');
+            return JSON.stringify(attr.replace(/"/g, '"'))
+                .replace(/\\t/g, '\t')
+                .replace(/\\n/g, '\n')
+                .replace(/\\r/g, '\r')
+                .replace(/\\/g, '');
         };
         /**
          * Remove Child element from childNodes array
@@ -738,8 +742,7 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
                 return child === _this;
             });
             resetParent([this], null);
-            resetParent(content, parent);
-            parent.childNodes = __spreadArray(__spreadArray(__spreadArray([], parent.childNodes.slice(0, idx), true), content, true), parent.childNodes.slice(idx + 1), true);
+            parent.childNodes = __spreadArray(__spreadArray(__spreadArray([], parent.childNodes.slice(0, idx), true), resetParent(content, parent), true), parent.childNodes.slice(idx + 1), true);
             return this;
         };
         Object.defineProperty(HTMLElement.prototype, "outerHTML", {
@@ -826,10 +829,12 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
             });
             this.childNodes.length = o;
             // remove whitespace between attributes
-            var attrs = Object.keys(this.rawAttributes).map(function (key) {
+            var attrs = Object.keys(this.rawAttributes)
+                .map(function (key) {
                 var val = _this.rawAttributes[key];
                 return "".concat(key, "=").concat(JSON.stringify(val));
-            }).join(' ');
+            })
+                .join(' ');
             this.rawAttrs = attrs;
             delete this._rawAttrs;
             return this;
@@ -921,7 +926,6 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
                     if (child.id === id) {
                         return child;
                     }
-                    ;
                     // if children are existing push the current status to the stack and keep searching for elements in the level below
                     if (child.childNodes.length > 0) {
                         stack.push(index);
@@ -1036,7 +1040,7 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
                 }
                 var attrs = {};
                 if (this.rawAttrs) {
-                    var re = /([a-zA-Z()[\]#@$.?:][a-zA-Z0-9-_:()[\]#]*)(?:\s*=\s*((?:'[^']*')|(?:"[^"]*")|\S+))?/g;
+                    var re = /([a-zA-Z()[\]#@$.?:][a-zA-Z0-9-._:()[\]#]*)(?:\s*=\s*((?:'[^']*')|(?:"[^"]*")|\S+))?/g;
                     var match = void 0;
                     while ((match = re.exec(this.rawAttrs))) {
                         var key = match[1];
@@ -1372,8 +1376,10 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
         return HTMLElement;
     }(node_2.default));
     exports.default = HTMLElement;
+    // #xB7 | [#xC0-#xD6] | [#xD8-#xF6] | [#xF8-#x37D] | [#x37F-#x1FFF] | [#x200C-#x200D] | [#x203F-#x2040] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF]
     // https://html.spec.whatwg.org/multipage/custom-elements.html#valid-custom-element-name
-    var kMarkupPattern = /<!--[\s\S]*?-->|<(\/?)([a-zA-Z][-.:0-9_a-zA-Z]*)((?:\s+[^>]*?(?:(?:'[^']*')|(?:"[^"]*"))?)*)\s*(\/?)>/g;
+    var kMarkupPattern = /<!--[\s\S]*?-->|<(\/?)([a-zA-Z][-.:0-9_a-zA-Z@\xB7\xC0-\xD6\xD8-\xF6\u00F8-\u03A1\u03A3-\u03D9\u03DB-\u03EF\u03F7-\u03FF\u0400-\u04FF\u0500-\u052F\u1D00-\u1D2B\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1E9B\u1F00-\u1F15\u1F18-\u1F1D\u1F20-\u1F45\u1F48-\u1F4D\u1F50-\u1F57\u1F59\u1F5B\u1F5D\u1F5F-\u1F7D\u1F80-\u1FB4\u1FB6-\u1FBC\u1FBE\u1FC2-\u1FC4\u1FC6-\u1FCC\u1FD0-\u1FD3\u1FD6-\u1FDB\u1FE0-\u1FEC\u1FF2-\u1FF4\u1FF6-\u1FFC\u2126\u212A-\u212B\u2132\u214E\u2160-\u2188\u2C60-\u2C7F\uA722-\uA787\uA78B-\uA78E\uA790-\uA7AD\uA7B0-\uA7B7\uA7F7-\uA7FF\uAB30-\uAB5A\uAB5C-\uAB5F\uAB64-\uAB65\uFB00-\uFB06\uFB13-\uFB17\uFF21-\uFF3A\uFF41-\uFF5A\x37F-\u1FFF\u200C-\u200D\u203F-\u2040\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD]*)((?:\s+[^>]*?(?:(?:'[^']*')|(?:"[^"]*"))?)*)\s*(\/?)>/gu;
+    // const kMarkupPattern = /<!--[\s\S]*?-->|<(\/?)([a-zA-Z][-.:0-9_a-zA-Z]*)((?:\s+[^>]*?(?:(?:'[^']*')|(?:"[^"]*"))?)*)\s*(\/?)>/g;
     var kAttributePattern = /(?:^|\s)(id|class)\s*=\s*((?:'[^']*')|(?:"[^"]*")|\S+)/gi;
     var kElementsClosedByOpening = {
         li: { li: true, LI: true },
@@ -1574,11 +1580,10 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
      * Parse a chuck of HTML source.
      */
     function parse(data, options) {
-        var _a, _b;
         if (options === void 0) { options = {}; }
         var stack = base_parse(data, options);
         var root = stack[0];
-        while (stack.length > 1) {
+        var _loop_1 = function () {
             // Handle each error elements.
             var last = stack.pop();
             var oneBefore = (0, back_1.default)(stack);
@@ -1588,7 +1593,9 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
                     // this is wrong, becouse this will put the H3 outside the current right position which should be inside the current Html Element, see issue 152 for more info
                     if (options.parseNoneClosedTags !== true) {
                         oneBefore.removeChild(last);
-                        (_a = oneBefore.parentNode.childNodes).push.apply(_a, last.childNodes);
+                        last.childNodes.forEach(function (child) {
+                            oneBefore.parentNode.appendChild(child);
+                        });
                         stack.pop();
                     }
                 }
@@ -1598,20 +1605,24 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
                     // eslint-disable-next-line no-lonely-if
                     if (options.parseNoneClosedTags !== true) {
                         oneBefore.removeChild(last);
-                        (_b = oneBefore.childNodes).push.apply(_b, last.childNodes);
+                        last.childNodes.forEach(function (child) {
+                            oneBefore.appendChild(child);
+                        });
                     }
                 }
             }
             else {
                 // If it's final element just skip.
             }
+        };
+        while (stack.length > 1) {
+            _loop_1();
         }
         // response.childNodes.forEach((node) => {
         // 	if (node instanceof HTMLElement) {
         // 		node.parentNode = null;
         // 	}
         // });
-        resetParent(root.childNodes, root, true);
         return root;
     }
     exports.parse = parse;
@@ -1628,13 +1639,10 @@ define("nodes/html", ["require", "exports", "css-select", "he", "back", "matcher
             return val;
         });
     }
-    function resetParent(nodes, parent, recursive) {
-        if (recursive === void 0) { recursive = false; }
-        nodes.forEach(function (node) {
+    function resetParent(nodes, parent) {
+        return nodes.map(function (node) {
             node.parentNode = parent;
-            if (recursive && node instanceof HTMLElement) {
-                resetParent(node.childNodes, node, true);
-            }
+            return node;
         });
     }
 });

package/dist/matcher.js

@@ -3,7 +3,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
-var type_1 = __importDefault(require("./nodes/type"));
+const type_1 = __importDefault(require("./nodes/type"));
 function isTag(node) {
     return node && node.nodeType === type_1.default.ELEMENT_NODE;
 }
@@ -23,10 +23,10 @@ function getText(node) {
     return node.text;
 }
 function removeSubsets(nodes) {
-    var idx = nodes.length;
-    var node;
-    var ancestor;
-    var replace;
+    let idx = nodes.length;
+    let node;
+    let ancestor;
+    let replace;
     // Check if each node (or one of its ancestors) is already contained in the
     // array.
     while (--idx > -1) {
@@ -50,26 +50,26 @@ function removeSubsets(nodes) {
     return nodes;
 }
 function existsOne(test, elems) {
-    return elems.some(function (elem) {
+    return elems.some((elem) => {
         return isTag(elem) ? test(elem) || existsOne(test, getChildren(elem)) : false;
     });
 }
 function getSiblings(node) {
-    var parent = getParent(node);
+    const parent = getParent(node);
     return parent ? getChildren(parent) : [];
 }
 function hasAttrib(elem, name) {
     return getAttributeValue(elem, name) !== undefined;
 }
 function findOne(test, elems) {
-    var elem = null;
-    for (var i = 0, l = elems === null || elems === void 0 ? void 0 : elems.length; i < l && !elem; i++) {
-        var el = elems[i];
+    let elem = null;
+    for (let i = 0, l = elems === null || elems === void 0 ? void 0 : elems.length; i < l && !elem; i++) {
+        const el = elems[i];
         if (test(el)) {
             elem = el;
         }
         else {
-            var childs = getChildren(el);
+            const childs = getChildren(el);
             if (childs && childs.length > 0) {
                 elem = findOne(test, childs);
             }
@@ -78,29 +78,29 @@ function findOne(test, elems) {
     return elem;
 }
 function findAll(test, nodes) {
-    var result = [];
-    for (var i = 0, j = nodes.length; i < j; i++) {
+    let result = [];
+    for (let i = 0, j = nodes.length; i < j; i++) {
         if (!isTag(nodes[i]))
             continue;
         if (test(nodes[i]))
             result.push(nodes[i]);
-        var childs = getChildren(nodes[i]);
+        const childs = getChildren(nodes[i]);
         if (childs)
             result = result.concat(findAll(test, childs));
     }
     return result;
 }
 exports.default = {
-    isTag: isTag,
-    getAttributeValue: getAttributeValue,
-    getName: getName,
-    getChildren: getChildren,
-    getParent: getParent,
-    getText: getText,
-    removeSubsets: removeSubsets,
-    existsOne: existsOne,
-    getSiblings: getSiblings,
-    hasAttrib: hasAttrib,
-    findOne: findOne,
-    findAll: findAll
+    isTag,
+    getAttributeValue,
+    getName,
+    getChildren,
+    getParent,
+    getText,
+    removeSubsets,
+    existsOne,
+    getSiblings,
+    hasAttrib,
+    findOne,
+    findAll
 };

package/dist/nodes/comment.js

@@ -1,57 +1,33 @@
 "use strict";
-var __extends = (this && this.__extends) || (function () {
-    var extendStatics = function (d, b) {
-        extendStatics = Object.setPrototypeOf ||
-            ({ __proto__: [] } instanceof Array && function (d, b) { d.__proto__ = b; }) ||
-            function (d, b) { for (var p in b) if (Object.prototype.hasOwnProperty.call(b, p)) d[p] = b[p]; };
-        return extendStatics(d, b);
-    };
-    return function (d, b) {
-        if (typeof b !== "function" && b !== null)
-            throw new TypeError("Class extends value " + String(b) + " is not a constructor or null");
-        extendStatics(d, b);
-        function __() { this.constructor = d; }
-        d.prototype = b === null ? Object.create(b) : (__.prototype = b.prototype, new __());
-    };
-})();
 var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
-var node_1 = __importDefault(require("./node"));
-var type_1 = __importDefault(require("./type"));
-var CommentNode = /** @class */ (function (_super) {
-    __extends(CommentNode, _super);
-    function CommentNode(rawText, parentNode, range, rawTagName) {
-        if (parentNode === void 0) { parentNode = null; }
-        if (rawTagName === void 0) { rawTagName = '!--'; }
-        var _this = _super.call(this, parentNode, range) || this;
-        _this.rawText = rawText;
-        _this.rawTagName = rawTagName;
+const node_1 = __importDefault(require("./node"));
+const type_1 = __importDefault(require("./type"));
+class CommentNode extends node_1.default {
+    clone() {
+        return new CommentNode(this.rawText, null, undefined, this.rawTagName);
+    }
+    constructor(rawText, parentNode = null, range, rawTagName = '!--') {
+        super(parentNode, range);
+        this.rawText = rawText;
+        this.rawTagName = rawTagName;
         /**
          * Node Type declaration.
          * @type {Number}
          */
-        _this.nodeType = type_1.default.COMMENT_NODE;
-        return _this;
+        this.nodeType = type_1.default.COMMENT_NODE;
     }
-    CommentNode.prototype.clone = function () {
-        return new CommentNode(this.rawText, null, undefined, this.rawTagName);
-    };
-    Object.defineProperty(CommentNode.prototype, "text", {
-        /**
-         * Get unescaped text value of current node and its children.
-         * @return {string} text content
-         */
-        get: function () {
-            return this.rawText;
-        },
-        enumerable: false,
-        configurable: true
-    });
-    CommentNode.prototype.toString = function () {
-        return "<!--".concat(this.rawText, "-->");
-    };
-    return CommentNode;
-}(node_1.default));
+    /**
+     * Get unescaped text value of current node and its children.
+     * @return {string} text content
+     */
+    get text() {
+        return this.rawText;
+    }
+    toString() {
+        return `<!--${this.rawText}-->`;
+    }
+}
 exports.default = CommentNode;