Fix frontmatter parsing with utf8 bom (#12664)

.changeset/unlucky-wasps-refuse.md

@@ -0,0 +1,5 @@
+---
+'@astrojs/markdown-remark': patch
+---
+
+Fixes frontmatter parsing if file is encoded in UTF8 with BOM

packages/markdown/remark/src/frontmatter.ts

@@ -11,9 +11,10 @@ export function isFrontmatterValid(frontmatter: Record<string, any>) {
 }
 
 // Capture frontmatter wrapped with `---`, including any characters and new lines within it.
-// Only capture if it exists near the top of the file (whitespaces between the start of file and
-// the start of `---` are allowed)
-const frontmatterRE = /^\s*---([\s\S]*?\n)---/;
+// Only capture if `---` exists near the top of the file, including:
+// 1. Start of file (including if has BOM encoding)
+// 2. Start of file with any whitespace (but `---` must still start on a new line)
+const frontmatterRE = /(?:^\uFEFF?|^\s*\n)---([\s\S]*?\n)---/;
 export function extractFrontmatter(code: string): string | undefined {
 	return frontmatterRE.exec(code)?.[1];
 }

packages/markdown/remark/test/frontmatter.test.js

@@ -2,16 +2,21 @@ import assert from 'node:assert/strict';
 import { describe, it } from 'node:test';
 import { extractFrontmatter, parseFrontmatter } from '../dist/index.js';
 
+const bom = '\uFEFF';
+
 describe('extractFrontmatter', () => {
 	it('works', () => {
 		const yaml = `\nfoo: bar\n`;
 		assert.equal(extractFrontmatter(`---${yaml}---`), yaml);
-		assert.equal(extractFrontmatter(`  ---${yaml}---`), yaml);
+		assert.equal(extractFrontmatter(`${bom}---${yaml}---`), yaml);
 		assert.equal(extractFrontmatter(`\n---${yaml}---`), yaml);
 		assert.equal(extractFrontmatter(`\n  \n---${yaml}---`), yaml);
 		assert.equal(extractFrontmatter(`---${yaml}---\ncontent`), yaml);
+		assert.equal(extractFrontmatter(`${bom}---${yaml}---\ncontent`), yaml);
 		assert.equal(extractFrontmatter(`\n\n---${yaml}---\n\ncontent`), yaml);
 		assert.equal(extractFrontmatter(`\n  \n---${yaml}---\n\ncontent`), yaml);
+		assert.equal(extractFrontmatter(` ---${yaml}---`), undefined);
+		assert.equal(extractFrontmatter(`---${yaml} ---`), undefined);
 		assert.equal(extractFrontmatter(`text\n---${yaml}---\n\ncontent`), undefined);
 	});
 });
@@ -24,10 +29,10 @@ describe('parseFrontmatter', () => {
 			rawFrontmatter: yaml,
 			content: '',
 		});
-		assert.deepEqual(parseFrontmatter(`  ---${yaml}---`), {
+		assert.deepEqual(parseFrontmatter(`${bom}---${yaml}---`), {
 			frontmatter: { foo: 'bar' },
 			rawFrontmatter: yaml,
-			content: '  ',
+			content: bom,
 		});
 		assert.deepEqual(parseFrontmatter(`\n---${yaml}---`), {
 			frontmatter: { foo: 'bar' },
@@ -44,6 +49,11 @@ describe('parseFrontmatter', () => {
 			rawFrontmatter: yaml,
 			content: '\ncontent',
 		});
+		assert.deepEqual(parseFrontmatter(`${bom}---${yaml}---\ncontent`), {
+			frontmatter: { foo: 'bar' },
+			rawFrontmatter: yaml,
+			content: `${bom}\ncontent`,
+		});
 		assert.deepEqual(parseFrontmatter(`\n\n---${yaml}---\n\ncontent`), {
 			frontmatter: { foo: 'bar' },
 			rawFrontmatter: yaml,
@@ -54,6 +64,16 @@ describe('parseFrontmatter', () => {
 			rawFrontmatter: yaml,
 			content: '\n  \n\n\ncontent',
 		});
+		assert.deepEqual(parseFrontmatter(` ---${yaml}---`), {
+			frontmatter: {},
+			rawFrontmatter: '',
+			content: ` ---${yaml}---`,
+		});
+		assert.deepEqual(parseFrontmatter(`---${yaml} ---`), {
+			frontmatter: {},
+			rawFrontmatter: '',
+			content: `---${yaml} ---`,
+		});
 		assert.deepEqual(parseFrontmatter(`text\n---${yaml}---\n\ncontent`), {
 			frontmatter: {},
 			rawFrontmatter: '',