Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
45 changes: 33 additions & 12 deletions packages/metascraper-x/src/index.js
Original file line number Diff line number Diff line change
Expand Up @@ -25,30 +25,51 @@ const test = memoizeOne(url =>
['twitter.com', 'x.com'].includes(parseUrl(url).domain)
)

const getAuthorName = memoizeOne(ogTitle =>
ogTitle?.split(' on X:')[0].split(' (@')[0].trim()
)

module.exports = ({ resolveUrl = url => url } = {}) => {
const rules = {
author: [
toAuthor($ => {
const author = $('meta[property="og:title"]').attr('content')
return author?.includes(' on X') ? author.split(' on X')[0] : author
})
],
title: [
toTitle(
($, url) => `@${new URL(url).pathname.toString().split('/')[1]} on X`
toAuthor($ =>
getAuthorName($('meta[property="og:title"]').attr('content'))
)
],
url: [
toUrl($ =>
$('link[rel="canonical"]').attr('href')?.replace('twitter.com', 'x.com')
)
title: [
toTitle(($, url) => {
const username = new URL(url).pathname.split('/')[1]
const authorName = getAuthorName(
$('meta[property="og:title"]').attr('content')
)
return authorName
? `${authorName} (@${username}) on X`
: `@${username} on X`
})
],
url: [toUrl($ => $('link[rel="canonical"]').attr('href'))],
description: [
toDescription(async $ => {
let description =
$jsonld('mainEntity.description')($) ||
$('meta[property="og:description"]').attr('content')

if (!description) {
const ogTitle = $('meta[property="og:title"]').attr('content')
if (ogTitle?.includes(' on X: "')) {
description = ogTitle.split(' on X: "')[1].split('" / X')[0]
const urls = getUrls(description)
if (urls.length > 1) {
const lastUrl = urls[urls.length - 1]
if (description.endsWith(lastUrl)) {
description = description.replace(lastUrl, '').trim()
}
}
}
}

if (!description) return

const urls = getUrls(description)
const resolvedUrls = await Promise.all(urls.map(resolveUrl))

Expand Down
2 changes: 1 addition & 1 deletion packages/metascraper-x/test/fixtures/post-og.html
Original file line number Diff line number Diff line change
@@ -1,2 +1,2 @@
<!DOCTYPE html><html dir="ltr" lang="en"><head><meta charset="utf-8" /><meta name="viewport" content="width=device-width,initial-scale=1,maximum-scale=1,user-scalable=0,viewport-fit=cover" /><link rel="preconnect" href="//abs.twimg.com" /><link rel="dns-prefetch" href="//abs.twimg.com" /><link rel="preconnect" href="//api.twitter.com" /><link rel="dns-prefetch" href="//api.twitter.com" /><link rel="preconnect" href="//api.x.com" /><link rel="dns-prefetch" href="//api.x.com" /><link rel="preconnect" href="//pbs.twimg.com" /><link rel="dns-prefetch" href="//pbs.twimg.com" /><link rel="preconnect" href="//t.co" /><link rel="dns-prefetch" href="//t.co" /><link rel="preconnect" href="//video.twimg.com" /><link rel="dns-prefetch" href="//video.twimg.com" /><meta http-equiv="onion-location" content="https://twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid.onion/" /><meta property="fb:app_id" content="2231777543" /><meta content="X (formerly Twitter)" property="og:site_name" /><meta name="google-site-verification" content="600dQ0pZYsH2xOFt4hYmf5f5NpjCbWE_qk5Y04dErYM" /><meta name="facebook-domain-verification" content="x6sdcc8b5ju3bh8nbm59eswogvg6t1" /><meta name="mobile-web-app-capable" content="yes" /><meta name="apple-mobile-web-app-title" content="Twitter" /><meta name="apple-mobile-web-app-status-bar-style" content="white" /><link rel="search" type="application/opensearchdescription+xml" href="/os-x.xml" title="X">
<link rel="search" type="application/opensearchdescription+xml" href="/os-grok.xml" title="Grok"><link rel="manifest" href="/manifest.json" crossOrigin="use-credentials" /><link rel="search" type="application/opensearchdescription+xml" href="/os-x.xml" title="X"><link rel="search" type="application/opensearchdescription+xml" href="/os-grok.xml" title="Grok"><link rel="shortcut icon" href="//abs.twimg.com/favicons/twitter.3.ico"><meta name="theme-color" media="(prefers-color-scheme: light)" content="#FFFFFF" /><meta name="theme-color" media="(prefers-color-scheme: dark)" content="#000000" /><script type="application/ld+json">{"@context":"https://schema.org","@type":"WebSite","url":"https://twitter.com/","potentialAction":{"@type":"SearchAction","query-input":"required name=search_term_string","target":{"@type":"EntryPoint","urlTemplate":"https://twitter.com/search?q={search_term_string}&ref_src=twcamp%5Eseo_searchbox%7Ctwsrc%5Eseo"}}}</script><meta http-equiv="origin-trial" content="AlpCmb40F5ZjDi9ZYe+wnr/V8MF+XmY41K4qUhoq+2mbepJTNd3q4CRqlACfnythEPZqcjryfAS1+ExS0FFRcA8AAABmeyJvcmlnaW4iOiJodHRwczovL3R3aXR0ZXIuY29tOjQ0MyIsImZlYXR1cmUiOiJMYXVuY2ggSGFuZGxlciIsImV4cGlyeSI6MTY1NTI1MTE5OSwiaXNTdWJkb21haW4iOnRydWV9" /><style>html,body{height: 100%;}::cue{white-space:normal}</style><meta content="article" property="og:type" /><meta content="https://twitter.com/UaSmart/status/934106870834454529" property="og:url" /><meta content="SmartUA (@UaSmart) on X" property="og:title" /><meta content="Y terminamos el dia con Cultura de empresa con @patoroco, @flopezluis, Katia, Angélica en @codemotion_es #codemotion2017" property="og:description" /><meta content="https://pbs.twimg.com/media/DPadOKpXcAIL-NW.jpg:large" property="og:image" /><meta content="Y terminamos el dia con Cultura de empresa con @patoroco, @flopezluis, Katia, Angélica en @codemotion_es #codemotion2017" name="description" /><meta content="SmartUA (@UaSmart) on X" name="title" /><meta content="twitter://status?id=934106870834454529" property="al:ios:url" /><meta content="333903271" property="al:ios:app_store_id" /><meta content="X" property="al:ios:app_name" /><meta content="twitter://status?id=934106870834454529" property="al:android:url" /><meta content="com.twitter.android" property="al:android:package" /><meta content="X" property="al:android:app_name" /></head><body style="background-color: #FFFFFF;"><div id="react-root" style="height:100%;display:flex;"></div>
<link rel="search" type="application/opensearchdescription+xml" href="/os-grok.xml" title="Grok"><link rel="manifest" href="/manifest.json" crossOrigin="use-credentials" /><link rel="search" type="application/opensearchdescription+xml" href="/os-x.xml" title="X"><link rel="search" type="application/opensearchdescription+xml" href="/os-grok.xml" title="Grok"><link rel="shortcut icon" href="//abs.twimg.com/favicons/twitter.3.ico"><meta name="theme-color" media="(prefers-color-scheme: light)" content="#FFFFFF" /><meta name="theme-color" media="(prefers-color-scheme: dark)" content="#000000" /><script type="application/ld+json">{"@context":"https://schema.org","@type":"WebSite","url":"https://twitter.com/","potentialAction":{"@type":"SearchAction","query-input":"required name=search_term_string","target":{"@type":"EntryPoint","urlTemplate":"https://twitter.com/search?q={search_term_string}&ref_src=twcamp%5Eseo_searchbox%7Ctwsrc%5Eseo"}}}</script><meta http-equiv="origin-trial" content="AlpCmb40F5ZjDi9ZYe+wnr/V8MF+XmY41K4qUhoq+2mbepJTNd3q4CRqlACfnythEPZqcjryfAS1+ExS0FFRcA8AAABmeyJvcmlnaW4iOiJodHRwczovL3R3aXR0ZXIuY29tOjQ0MyIsImZlYXR1cmUiOiJMYXVuY2ggSGFuZGxlciIsImV4cGlyeSI6MTY1NTI1MTE5OSwiaXNTdWJkb21haW4iOnRydWV9" /><style>html,body{height: 100%;}::cue{white-space:normal}</style><meta content="article" property="og:type" /><meta content="https://twitter.com/realDonaldTrump/status/1222907250383245320" property="og:url" /><meta content="Donald J. Trump (@realDonaldTrump) on X" property="og:title" /><meta content="“Schiff blasted for not focusing on California homeless.” @foxandfriends His District is in terrible shape. He is a corrupt pol who only dreams of the Impeachment Hoax. In my opinion he is mentally deranged!" property="og:description" /><meta content="https://pbs.twimg.com/profile_images/874276197357596672/kUuht00m_200x200.jpg" property="og:image" /><meta content="“Schiff blasted for not focusing on California homeless.” @foxandfriends His District is in terrible shape. He is a corrupt pol who only dreams of the Impeachment Hoax. In my opinion he is mentally deranged!" name="description" /><meta content="Donald J. Trump (@realDonaldTrump) on X" name="title" /><meta content="twitter://status?id=1222907250383245320" property="al:ios:url" /><meta content="333903271" property="al:ios:app_store_id" /><meta content="X" property="al:ios:app_name" /><meta content="twitter://status?id=1222907250383245320" property="al:android:url" /><meta content="com.twitter.android" property="al:android:package" /><meta content="X" property="al:android:app_name" /></head><body style="background-color: #FFFFFF;"><div id="react-root" style="height:100%;display:flex;"></div>
Loading
Loading