Merge branch '294' of github.com:ig3/node-html-parser into ig3-294

taoqf · taoqf · commit 562528629cd8 · 2026-01-07T15:15:22.000+08:00
diff --git a/package.json b/package.json
@@ -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": {
diff --git a/src/nodes/html.ts b/src/nodes/html.ts
@@ -1017,6 +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;
@@ -1202,6 +1205,28 @@ export function base_parse(data: string, options = {} as Partial<Options>) {
 							continue;
 						}
 					}
+					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;
+							}
+						}
+					}
 					// Use aggressive strategy to handle unmatching markups.
 					break;
 				}
diff --git a/test/tests/issues/294.js b/test/tests/issues/294.js
@@ -0,0 +1,12 @@
+const { parse, valid } = require('@test/test-target');
+
+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);
+    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);
+	});
+});
diff --git a/test/tests/valid.js b/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 () {
diff --git a/yarn.lock b/yarn.lock
