@unavatar/core 3.7.67 → 3.7.69

package/package.json

@@ -2,7 +2,7 @@
   "name": "@unavatar/core",
   "description": "Get unified user avatar from social networks, including Instagram, SoundCloud, Telegram, Twitter, YouTube & more.",
   "homepage": "https://unavatar.io",
-  "version": "3.7.67",
+  "version": "3.7.69",
   "main": "src/index.js",
   "exports": {
     ".": "./src/index.js",

package/src/providers/instagram.js

@@ -2,10 +2,12 @@
 
 const randomCrawlerAgent = require('../util/crawler-agent')
 
+const isBlocked = $ => $('title').text().includes('Login')
+
 module.exports = ({ createHtmlProvider, getOgImage }) =>
   createHtmlProvider({
     name: 'instagram',
     url: input => `https://www.instagram.com/${input}`,
-    getter: getOgImage,
+    getter: $ => !isBlocked($) && getOgImage($),
     htmlOpts: () => ({ headers: { 'user-agent': randomCrawlerAgent() } })
   })

package/src/providers/patreon.js

@@ -2,9 +2,37 @@
 
 const { $jsonld } = require('@metascraper/helpers')
 
-module.exports = ({ createHtmlProvider }) =>
+const unescapeUnicode = str =>
+  str.replace(/\\u([0-9a-fA-F]{4})/g, (_, hex) =>
+    String.fromCharCode(parseInt(hex, 16))
+  )
+
+const getRscAvatar = $ => {
+  let url
+  $('script').each((_, el) => {
+    const text = $(el).html() || ''
+    if (!text.includes('avatarPhotoImageUrls')) return
+    const match = text.match(
+      /avatarPhotoImageUrls[\s\S]*?\\"original\\":\\"((?:[^\\"]|\\u[0-9a-fA-F]{4})+)/
+    )
+    if (match) {
+      url = unescapeUnicode(match[1])
+      return false
+    }
+  })
+  return url
+}
+
+const getAvatar = $ =>
+  $jsonld('mainEntity.image.contentUrl')($) || getRscAvatar($)
+
+const factory = ({ createHtmlProvider }) =>
   createHtmlProvider({
     name: 'patreon',
     url: username => `https://www.patreon.com/${username}`,
-    getter: $ => $jsonld('mainEntity.image.thumbnailUrl')($)
+    getter: getAvatar
   })
+
+factory.getAvatar = getAvatar
+
+module.exports = factory

package/src/util/html-provider.js

@@ -18,9 +18,21 @@ const createEmptyProviderValueError = ({ provider, statusCode }) =>
     message: 'Empty value returned by the provider.'
   })
 
+
 const getOgImage = $ => $('meta[property="og:image"]').attr('content')
 
 module.exports = ({ PROXY_TIMEOUT, getHTML, onFetchHTML }) => {
+  /**
+   * @param {object} opts
+   * @param {string} opts.name - Provider identifier used in logs and metrics.
+   * @param {(input: string) => string | Promise<string>} opts.url - Builds the URL to fetch for a given input.
+   * @param {($: cheerio.CheerioAPI) => string | false | undefined} opts.getter
+   *   Extracts the avatar URL from the fetched HTML.
+   *   - `string`    — avatar URL found (success).
+   *   - `undefined`  — avatar not found (normal failure, no retry).
+   *   - `false`     — page is blocked; error.blocked will be set to true.
+   * @param {() => object} [opts.htmlOpts] - Returns extra options merged into the fetch call.
+   */
   const createHtmlProvider = ({ name, url, getter, htmlOpts }) => {
     const provider = async function ({ input, req = {}, res = {} }) {
       const providerUrl = await url(input)
@@ -57,8 +69,9 @@ module.exports = ({ PROXY_TIMEOUT, getHTML, onFetchHTML }) => {
             statusCode
           })
           error.html = attempt.lastHtml
+          if (result === false) error.blocked = true
 
-          log.error({ statusCode })
+          log.error({ statusCode, status: result === false ? 'blocked' : undefined })
 
           throw error
         }