Merge pull request #59 from fmacpro/development

get site icon url functionality

index.js

@@ -170,6 +170,15 @@ const articleParser = async function (options, socket) {
     article.mobile = await page.screenshot({ encoding: 'base64', type: 'jpeg', quality: 60 })
   }
 
+  // Evaluate site icon url
+  if (options.enabled.includes('siteicon')) {
+    socket.emit('parse:status', 'Evaluating site icon')
+    article.siteicon = await page.evaluate(() => {
+      var j = window.$
+      return j('link[rel~="icon"]').prop('href')
+    })
+  }
+
   // Evaluate meta
   socket.emit('parse:status', 'Evaluating Meta Data')
 

package.json

@@ -1,6 +1,6 @@
 {
   "name": "horseman-article-parser",
-  "version": "0.8.53",
+  "version": "0.8.54",
   "description": "Web Page Inspection Tool. Sentiment Analysis, Keyword Extraction, Named Entity Recognition & Spell Check",
   "main": "index.js",
   "scripts": {

test.js

@@ -11,7 +11,7 @@ const testPlugin = function (Doc, world) {
 
 const options = {
   url: 'https://www.theguardian.com/commentisfree/2021/jan/07/what-happened-in-washington-dc-is-happening-around-the-world',
-  enabled: ['lighthouse', 'screenshot', 'links', 'sentiment', 'entities', 'spelling', 'keywords'],
+  enabled: ['lighthouse', 'screenshot', 'links', 'sentiment', 'entities', 'spelling', 'keywords', 'siteicon'],
   nlp: {
     plugins: [testPlugin]
   }
@@ -24,6 +24,7 @@ parser.parseArticle(options)
       excerpt: article.excerpt,
       metadescription: article.meta.description.text,
       url: article.url,
+      siteicon: article.siteicon,
       sentiment: { score: article.sentiment.score, comparative: article.sentiment.comparative },
       keyphrases: article.processed.keyphrases,
       keywords: article.processed.keywords,