draw title from first H1 only (#23266)

Part of #1218

script/search/parse-page-sections-into-records.js

@@ -9,7 +9,7 @@ const ignoredHeadingSlugs = ['in-this-article', 'further-reading', 'prerequisite
 
 export default function parsePageSectionsIntoRecords(page) {
   const { href, $, languageCode } = page
-  const title = $('h1').text().trim()
+  const title = $('h1').first().text().trim()
   const breadcrumbsArray = $('[data-search=breadcrumbs] nav.breadcrumbs a')
     .map((i, el) => {
       return $(el).text().trim().replace('/', '').replace(/\s+/g, ' ')

tests/unit/search/fixtures/page-with-multiple-h1s.html

@@ -0,0 +1,18 @@
+<div data-search="breadcrumbs">
+  <nav class="breadcrumbs">
+    <a href="#">GitHub Actions</a>
+    <a href="#">actions learning path</a>
+    <a href="#">I am the page title</a>
+  </nav>
+</div>
+
+<h1>I am the page title</h1>
+
+<div data-search="lead">
+  <p>This is an introduction to the article.</p>
+</div>
+
+<div data-search="article-body">
+  <h1>A heading 1 inside the body</h1>
+  <p>This won't be ignored.</p>
+</div>

tests/unit/search/parse-page-sections-into-records.js

@@ -18,6 +18,10 @@ const fixtures = {
     path.join(__dirname, 'fixtures/page-without-body.html'),
     'utf8'
   ),
+  pageMultipleH1s: await fs.readFile(
+    path.join(__dirname, 'fixtures/page-with-multiple-h1s.html'),
+    'utf8'
+  ),
 }
 
 describe('search parsePageSectionsIntoRecords module', () => {
@@ -78,4 +82,12 @@ describe('search parsePageSectionsIntoRecords module', () => {
 
     expect(record).toEqual(expected)
   })
+
+  test('only picks up the first h1 for the title', () => {
+    const html = fixtures.pageMultipleH1s
+    const $ = cheerio.load(html)
+    const href = '/example/href'
+    const record = parsePageSectionsIntoRecords({ href, $, languageCode: 'en' })
+    expect(record.title).toEqual('I am the page title')
+  })
 })