feat: improve nymag.com extractor to grab deks from features

.babelrc

@@ -11,7 +11,8 @@
         "./cleaners": "cleaners",
         "./resource": "resource",
         "./extractors": "extractors",
-        "./test-helpers.js": "test-helpers"
+        "./test-helpers.js": "test-helpers",
+        "./iris.js": "iris"
       }
     }]
   ]

TODO.md

@@ -2,11 +2,11 @@ TODO:
 - Complete response:
   - add excerpt
   - add word count
-  - add total pages
-  - add rendered pages
 - Test if .is method is faster than regex methods
 
 DONE:
+x add total pages
+x add rendered pages
 x add canonicalUrl
 x add domain
 x Separate constants into activity-specific folders (dom, scoring)

src/extractors/custom/nymag.com/index.js

@@ -38,6 +38,7 @@ const NYMagExtractor = {
 
   title: {
     selectors: [
+      'h1.lede-feature-title',
       'h1.headline-primary',
       'h1',
     ],
@@ -46,6 +47,13 @@ const NYMagExtractor = {
   author: {
     selectors: [
       '.by-authors',
+      '.lede-feature-author',
+    ],
+  },
+
+  dek: {
+    selectors: [
+      '.lede-feature-teaser',
     ],
   },
 

src/extractors/custom/nymag.com/index.test.js

@@ -0,0 +1,21 @@
+import assert from 'assert'
+import fs from 'fs'
+
+import RootExtractor from 'extractors/root-extractor'
+import Resource from 'extractors/root-extractor'
+import Iris from 'iris'
+import NYMagExtractor from 'extractors/custom/nymag.com'
+
+describe('NYMagExtractor', () => {
+  it('works with a feature story', async () => {
+    const html = fs.readFileSync('./fixtures/nymag.com/ailes.html')
+    const uri = 'http://nymag.com/daily/intelligencer/2016/09/how-fox-news-women-took-down-roger-ailes.html'
+
+    const { dek, title, author } = await Iris.parse(uri, html)
+    const actualDek = 'How Fox News women took down the most powerful, and predatory, man in media.'
+
+    assert.equal(dek, actualDek)
+    assert.equal(title, 'The Revenge of Roger’s Angels')
+    assert.equal(author, 'Gabriel Sherman')
+  })
+})