Merge branch 'main' into close-tags

Move tests of option closeAllByClose to test/tests/unclosedtags.js

Author: ig3

.github/workflows/build.yml

@@ -17,13 +17,13 @@ jobs:
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
 
       - name: Determine Yarn Cache Path
         id: yarn-cache-dir-path
-        run: echo "::set-output name=dir::$(yarn cache dir)"
+        run: echo "dir=$(yarn cache dir)" >> $GITHUB_OUTPUT
 
-      - uses: actions/cache@v1
+      - uses: actions/cache@v4
         id: yarn-cache # use this to check for `cache-hit` (`steps.yarn-cache.outputs.cache-hit != 'true'`)
         with:
           path: ${{ steps.yarn-cache-dir-path.outputs.dir }}
@@ -32,7 +32,7 @@ jobs:
             ${{ runner.os }}-yarn-
 
       - name: Use Node.js ${{ matrix.node-version }}
-        uses: actions/setup-node@v1
+        uses: actions/setup-node@v4
         with:
           node-version: ${{ matrix.node-version }}
 

.github/workflows/publish.yml

@@ -11,10 +11,10 @@ jobs:
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
 
-      - name: Setup Node.js 22.x to publish to npmjs.org
-        uses: actions/setup-node@v1
+      - name: Setup Node.js to publish to npmjs.org
+        uses: actions/setup-node@v4
         with:
           node-version: "18.x"
           registry-url: "https://registry.npmjs.org"

CHANGELOG.md

@@ -2,6 +2,16 @@
 
 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.
 
+### [7.0.2](https://github.com/taoqf/node-fast-html-parser/compare/v7.0.1...v7.0.2) (2026-01-07)
+
+
+### Bug Fixes
+
+* [#227](https://github.com/taoqf/node-fast-html-parser/issues/227) ([51528c4](https://github.com/taoqf/node-fast-html-parser/commit/51528c41ef2648d6c4dc1aecd14ee9d2b0083c4f))
+* [#294](https://github.com/taoqf/node-fast-html-parser/issues/294) Closing tag is missing but valid HTML is still not parseable ([950865f](https://github.com/taoqf/node-fast-html-parser/commit/950865fab5f4df7853b36712869b71c90f4d3a1b))
+* add missing dev dependency: yarn ([6d73ea3](https://github.com/taoqf/node-fast-html-parser/commit/6d73ea37c48f4170c35907869ba410c5122a9a1f))
+* test valid.js ([a81fc46](https://github.com/taoqf/node-fast-html-parser/commit/a81fc46fab2507615b0362150d62568a6f52ee4e))
+
 ### [7.0.1](https://github.com/taoqf/node-fast-html-parser/compare/v7.0.0...v7.0.1) (2024-12-26)
 
 

README.md

@@ -43,6 +43,7 @@ import { parse } from 'node-html-parser';
 
 const root = parse('<ul id="list"><li>Hello World</li></ul>');
 
+// parse() adds a wrapper node, so the input data's first node is the root's first child node
 console.log(root.firstChild.structure);
 // ul#list
 //   li
@@ -74,7 +75,7 @@ var root = HTMLParser.parse('<ul id="list"><li>Hello World</li></ul>');
 
 ### parse(data[, options])
 
-Parse the data provided, and return the root of the generated DOM.
+Parse the data provided, wrap the result in a new node, and return the root of the generated DOM.
 
 - **data**, data to parse
 - **options**, parse options
@@ -350,11 +351,11 @@ Get all child elements, so all child nodes of type HTMLELement.
 
 ### firstChild
 
-Get first child node. `undefined` if the node has no children.
+Get first child node of the wrapper node added by `parse()`. `undefined` if the node has no children.
 
 ### lastChild
 
-Get last child node. `undefined` if the node has no children.
+Get last child node of the wrapper node added by `parse()`. `undefined` if the node has no children.
 
 ### firstElementChild
 

package.json

@@ -1,6 +1,6 @@
 {
 	"name": "node-html-parser",
-	"version": "7.0.1",
+	"version": "7.0.2",
 	"description": "A very fast HTML parser, generating a simplified DOM, with basic element query support.",
 	"main": "dist/index.js",
 	"types": "dist/index.d.ts",
@@ -87,7 +87,8 @@
 		"standard-version": "^9.5.0",
 		"travis-cov": "latest",
 		"ts-node": "^10.9.1",
-		"typescript": "latest"
+		"typescript": "latest",
+		"yarn": "^1.22.22"
 	},
 	"config": {
 		"blanket": {

src/nodes/html.ts

@@ -139,8 +139,8 @@ export default class HTMLElement extends Node {
 	private _attrs: Attributes;
 	private _rawAttrs: RawAttributes;
 	private _parseOptions: Partial<Options>;
+	private _id: string;
 	public rawTagName: string; // there is not friend funciton in es
-	public id: string;
 	public classList: DOMTokenList;
 
 	/**
@@ -185,7 +185,7 @@ export default class HTMLElement extends Node {
 		super(parentNode, range);
 		this.rawTagName = tagName;
 		this.rawAttrs = rawAttrs || '';
-		this.id = keyAttrs.id || '';
+		this._id = keyAttrs.id || '';
 		this.childNodes = [];
 		this._parseOptions = _parseOptions;
 		this.classList = new DOMTokenList(
@@ -248,6 +248,13 @@ export default class HTMLElement extends Node {
 		return this.voidTag.isVoidElement(this.localName);
 	}
 
+	public get id() {
+		return this._id;
+	}
+	public set id(newid: string) {
+		this.setAttribute('id', newid);
+	}
+
 	/**
 	 * Get escpaed (as-it) text value of current node and its children.
 	 * @return {string} text content
@@ -417,7 +424,7 @@ export default class HTMLElement extends Node {
 			res.push('  '.repeat(indention) + str);
 		}
 		function dfs(node: HTMLElement) {
-			const idStr = node.id ? `#${node.id}` : '';
+			const idStr = node._id ? `#${node._id}` : '';
 			const classStr = node.classList.length ? `.${node.classList.value.join('.')}` : ''; // eslint-disable-line @typescript-eslint/no-unsafe-member-access, @typescript-eslint/no-unsafe-member-access, @typescript-eslint/restrict-template-expressions, @typescript-eslint/no-unsafe-call
 			write(`${node.rawTagName}${idStr}${classStr}`);
 			indention++;
@@ -565,7 +572,7 @@ export default class HTMLElement extends Node {
 			}
 
 			if (child.nodeType === NodeType.ELEMENT_NODE) {
-				if (child.id === id) {
+				if (child._id === id) {
 					return child;
 				}
 
@@ -716,9 +723,9 @@ export default class HTMLElement extends Node {
 				return `${name}=${val}`;
 			})
 			.join(' ');
-		// Update this.id
+		// Update this._id
 		if (key === 'id') {
-			this.id = '';
+			this._id = '';
 		}
 		return this;
 	}
@@ -765,9 +772,9 @@ export default class HTMLElement extends Node {
 				return `${name}=${val}`;
 			})
 			.join(' ');
-		// Update this.id
+		// Update this._id
 		if (key === 'id') {
-			this.id = value;
+			this._id = value;
 		}
 		return this;
 	}
@@ -793,6 +800,10 @@ export default class HTMLElement extends Node {
 				return `${name}=${this.quoteAttribute(String(val))}`;
 			})
 			.join(' ');
+		// Update this._id
+		if ('id' in attributes) {
+			this._id = attributes['id'];
+		}
 		return this;
 	}
 
@@ -1006,6 +1017,9 @@ const kElementsClosedByClosing = {
 	th: { tr: true, table: true, TR: true, TABLE: true },
 	TH: { tr: true, table: true, TR: true, TABLE: true },
 } as Record<string, Record<string, boolean>>;
+const kElementsClosedByClosingExcept = {
+	p: { a: true, audio: true, del: true, ins: true, map: true, noscript: true, video: true },
+} as Record<string, Record<string, boolean>>;
 
 export interface Options {
 	lowerCaseTagName?: boolean;
@@ -1192,22 +1206,44 @@ export function base_parse(data: string, options = {} as Partial<Options>) {
 							continue;
 						}
 					}
-					if (options.closeAllByClosing === true) {
-						// If tag was opened, close all nested tags
-						let i;
-						for (i = stack.length - 2; i >= 0; i--) {
-							if (stack[i].rawTagName === tagName) break;
-						}
-						if (i >= 0) {
-							while (stack.length > i) {
+					const openTag =
+						currentParent.rawTagName ?
+							currentParent.rawTagName.toLowerCase() :
+							'';
+					if (kElementsClosedByClosingExcept[openTag]) {
+						const closingTag = tagName.toLowerCase();
+						if (stack.length > 1) {
+							const possibleContainer = stack[stack.length - 2];
+							if (
+								possibleContainer &&
+								possibleContainer.rawTagName &&
+							  possibleContainer.rawTagName.toLowerCase() === closingTag &&
+								!kElementsClosedByClosingExcept[openTag][closingTag]
+							) {
 								// Update range end for closed tag
 								(<[number, number]>currentParent.range)[1] = createRange(-1, Math.max(lastTextPos, tagEndPos))[1];
 								stack.pop();
 								currentParent = arr_back(stack);
+								continue;
 							}
-							continue;
 						}
 					}
+          if (options.closeAllByClosing === true) {
+            // If tag was opened, close all nested tags
+            let i;
+            for (i = stack.length - 2; i >= 0; i--) {
+              if (stack[i].rawTagName === tagName) break;
+            }
+            if (i >= 0) {
+              while (stack.length > i) {
+								// Update range end for closed tag
+								(<[number, number]>currentParent.range)[1] = createRange(-1, Math.max(lastTextPos, tagEndPos))[1];
+								stack.pop();
+								currentParent = arr_back(stack);
+              }
+              continue;
+            }
+          }
 					// Use aggressive strategy to handle unmatching markups.
 					break;
 				}

test/tests/issues/294.js

@@ -1,20 +1,12 @@
 const { parse, valid } = require('@test/test-target');
 
-describe('issue 294 Closing tag is missing but valid HTML is still not parseable', function () {
-	it('invalid HTML missing closing tag should not parse', function () {
-		const content = '<body><main class=h-entry><p>hello</main></body>';
-		valid(content).should.equal(false);
+describe('issue 294 Closing tag is missing but valid HTML still not parsable', function () {
+	it('Valid HTML missing closing p tag should parse', function () {
+    const content = '<body><main class=h-entry><p>hello</main></body>';
+    valid(content).should.equal(true);
 		const root = parse(content);
-		const list = root.querySelectorAll('.h-entry');
-		list.length.should.equal(0);
-	});
-	it('invalid HTML missing closing tag should parse', function () {
-		const content = '<body><main class=h-entry><p>hello</main></body>';
-		valid(content).should.equal(false);
-		const root = parse(content, {
-      closeAllByClosing: true
-    });
-		const list = root.querySelectorAll('.h-entry');
-		list.length.should.equal(1);
+    root.outerHTML.should.equal('<body><main class=h-entry><p>hello</p></main></body>');
+    const list = root.querySelectorAll('.h-entry');
+    list.length.should.equal(1);
 	});
 });

test/tests/unclosedtags.js

@@ -0,0 +1,16 @@
+const { parse, valid } = require('@test/test-target');
+
+describe('Unclosed tags', function () {
+	it('Unclosed tags should be closed at end of parent element', function () {
+    const html = '<p>not bold<b>bold</p><p>more</p>';
+    valid(html, { closeAllByClosing: true }).should.be.true();
+    const root = parse(html, { closeAllByClosing: true });
+    root.outerHTML.should.equal('<p>not bold<b>bold</b></p><p>more</p>');
+	});
+	it('Nested unclosed tags should be closed at end of parent element', function () {
+    const html = '<p>not bold<b>bold<i>bold italic</p><p>more</p>';
+    valid(html, { closeAllByClosing: true }).should.be.true();
+    const root = parse(html, { closeAllByClosing: true });
+    root.outerHTML.should.equal('<p>not bold<b>bold<i>bold italic</i></b></p><p>more</p>');
+  });
+});

test/tests/valid.js

@@ -20,9 +20,25 @@ describe('parseWithValidation', function () {
 		result.should.eql(false);
 	})
 
-	it('hillcrestpartyrentals.html  should return Object with valid: false.  not closing <p> tag on line 476', function () {
+	// #294: Closing tag is missing but valid HTML is still not parseable
+	//
+	// Tag omission in text/html:
+	// A p element's end tag can be omitted if the p element is immediately
+	// followed by an address, article, aside, blockquote, details, dialog,
+	// div, dl, fieldset, figcaption, figure, footer, form, h1, h2, h3, h4,
+	// h5, h6, header, hgroup, hr, main, menu, nav, ol, p, pre, search,
+	// section, table, or ul element, or if there is no more content in the
+	// parent element and the parent element is an HTML element that is not
+	// an a, audio, del, ins, map, noscript, or video element, or an
+	// autonomous custom element.
+	//
+	// Based on this, hillcrestpartyrentals.html is in fact valid HTML. All
+	// the p elements missing close tags are contained within td elements
+	// and, therefore, should be closed when there is no more content in the
+	// parent td element (i.e. at the `</td>`).
+	it('hillcrestpartyrentals.html  should return Object with valid: true.  not closing <p> tag on line 476', function () {
 		const result = valid(fs.readFileSync(__dirname + '/../assets/html/hillcrestpartyrentals.html').toString());
-		result.should.eql(false);
+		result.should.eql(true);
 	})
 
 	it('google.html  should return Object with valid: true', function () {