Fix parsing authors from page metadata

packages/api/src/utils/parser.ts

@@ -374,7 +374,7 @@ export const parsePageMetadata = (html: string): Metadata | undefined => {
 
     const author =
       window.document
-        .querySelector("head meta[property='author']")
+        .querySelector("head meta[name='author']")
         ?.getAttribute('content') || undefined
 
     // TODO: we should be able to apply the JSONLD metadata

packages/api/test/utils/parser.test.ts

@@ -3,7 +3,7 @@ import { expect } from 'chai'
 import 'chai/register-should'
 import { JSDOM } from 'jsdom'
 import fs from 'fs'
-import { findNewsletterUrl, isProbablyNewsletter } from '../../src/utils/parser'
+import { findNewsletterUrl, isProbablyNewsletter, parsePageMetadata } from '../../src/utils/parser'
 
 const load = (path: string): string => {
   return fs.readFileSync(path, 'utf8')
@@ -38,3 +38,14 @@ describe('findNewsletterUrl', async () => {
   })
 })
 
+describe('parseMetadata', async () => {
+  it('gets author, title, image, description', async () => {
+    const html = load('./test/utils/data/substack-post.html')
+    const metadata = await parsePageMetadata(html)
+    expect(metadata?.author).to.deep.equal('Omnivore')
+    expect(metadata?.title).to.deep.equal('Code Block Syntax Highlighting')
+    expect(metadata?.previewImage).to.deep.equal('https://cdn.substack.com/image/fetch/w_1200,h_600,c_fill,f_jpg,q_auto:good,fl_progressive:steep,g_auto/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2F2ab1f7e8-2ca7-4011-8ccb-43d0b3bd244f_1490x2020.png')
+    expect(metadata?.description).to.deep.equal('Highlighted <code> in Omnivore')
+
+  })
+})