fix(SimplyRecipes): instruction extraction

Author: nerdstep

src/plugins/__tests__/html-stripper.processor.test.ts

@@ -9,7 +9,6 @@ describe('HtmlStripperPlugin', () => {
     expect(plugin.shouldProcess('title')).toBe(true)
     expect(plugin.shouldProcess('instructions')).toBe(true)
     expect(plugin.shouldProcess('ingredients')).toBe(true)
-    expect(plugin.shouldProcess('description')).toBe(false)
     expect(plugin.shouldProcess('category')).toBe(false)
   })
 
@@ -54,7 +53,6 @@ describe('HtmlStripperPlugin', () => {
   })
 
   it('returns value unchanged for non-target fields', () => {
-    expect(plugin.process('description', '<b>desc</b>')).toBe('<b>desc</b>')
     expect(plugin.process('category', new Set(['<b>cat</b>']))).toEqual(
       new Set(['<b>cat</b>']),
     )

src/scrapers/simplyrecipes.ts

@@ -1,9 +1,37 @@
 import { AbstractScraper } from '@/abstract-scraper'
+import type { RecipeFields } from '@/types/recipe.interface'
+import { normalizeString } from '@/utils/parsing'
 
 export class SimplyRecipes extends AbstractScraper {
   static host() {
     return 'simplyrecipes.com'
   }
 
-  extractors = {}
+  extractors = {
+    instructions: this.instructions.bind(this),
+  }
+
+  /**
+   * Scrape and normalize each step under
+   * div.structured-project__steps > ol > li
+   */
+  protected instructions(): RecipeFields['instructions'] {
+    // select all <li> under the steps container
+    const items = this.$('div.structured-project__steps ol li').toArray()
+
+    if (items.length === 0) {
+      return new Set()
+    }
+
+    const steps = items
+      .map((el) => {
+        // clone & strip images/figures before grabbing text
+        const $clone = this.$(el).clone()
+        $clone.find('img, picture, figure').remove()
+        return normalizeString($clone.text())
+      })
+      .filter((text) => text.length > 0)
+
+    return new Set(steps)
+  }
 }