feat: add inline rules at the end (#163)

microlinkhq · Apr 3, 2019 · 936749a · 936749a
1 parent bd21875
commit 936749a
Show file tree

Hide file tree

Showing 3 changed files with 22 additions and 68 deletions.
diff --git a/README.md b/README.md
@@ -277,7 +277,9 @@ The HTML markup for extracting the content.
 
 Type: `Array`
 
-You can pass additional rules on execution time. These rules will be merged with your loaded rules.
+You can pass additional rules to add on execution time. 
+
+These rules will be merged with your loaded [`rules`](#rules) at the beginning.
 
 ## Benchmark
 

diff --git a/packages/metascraper/src/merge-rules.js b/packages/metascraper/src/merge-rules.js
@@ -2,13 +2,17 @@
 
 const { cloneDeep, concat, first, findIndex, forEach, chain } = require('lodash')
 
-module.exports = (rules, baseRules) => chain(rules)
-  .reduce((acc, rules) => {
-    forEach(rules, (rule, propName) => {
-      const index = findIndex(acc, item => first(item) === propName)
-      if (index !== -1) acc[index][1] = concat(acc[index][1], rule)
-      else acc.push([propName, rule])
-    })
-    return acc
-  }, cloneDeep(baseRules))
-  .value()
+module.exports = (rules, baseRules) =>
+  chain(rules)
+    .reduce((acc, rules) => {
+      forEach(rules, (rule, propName) => {
+        // find the rules associated with `propName`
+        const index = findIndex(acc, item => first(item) === propName)
+        // if `propName` has more rule, add the new rule from the end
+        if (index !== -1) acc[index][1] = concat(rule, ...acc[index][1])
+        // otherwise, create an array of rules
+        else acc.push([propName, rule])
+      })
+      return acc
+    }, cloneDeep(baseRules))
+    .value()
diff --git a/packages/metascraper/test/unit/merge-rules.js b/packages/metascraper/test/unit/merge-rules.js
@@ -49,76 +49,24 @@ it('add a new rule for a prop that exists', async () => {
     <meta charset="UTF-8">
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
     <meta http-equiv="X-UA-Compatible" content="ie=edge">
+    <meta property="og:image" content="http://ia.media-imdb.com/images/rock.jpg" />
     <title>Document</title>
   </head>
   <body>
-    <div class="logos">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-    </div>
-
-    <img class="main-logo" href="https://microlink.io/logo.png">
+    <img id="logo" src="https://microlink.io/logo.png">
     <p>Hello World </p>
   </body>
   </html>
   `
 
   const rules = [
     {
-      foo: [() => 'bar']
+      image: [({ htmlDom: $ }) => $('#logo').attr('src')]
     }
   ]
 
-  const metascraper = require('../..')([
-    {
-      foo: [() => false, () => false, () => false]
-    }
-  ])
-
-  const meta = await metascraper({ url, html, rules })
-  should(meta.foo).be.equal('bar')
-})
-
-it('rules are added from the end', async () => {
-  const url = 'https://microlink.io'
-
-  const html = `
-  <!DOCTYPE html>
-  <html lang="en">
-  <head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <meta http-equiv="X-UA-Compatible" content="ie=edge">
-    <title>Document</title>
-  </head>
-  <body>
-    <div class="logos">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-      <img class="logo" href="https://microlink.io/logo.png">
-    </div>
-
-    <img class="main-logo" href="https://microlink.io/logo.png">
-    <p>Hello World </p>
-  </body>
-  </html>
-  `
-
-  const rules = [
-    {
-      foo: [() => 'bar']
-    }
-  ]
-
-  const metascraper = require('../..')([
-    {
-      foo: [() => false, () => false, () => 'baz']
-    }
-  ])
+  const metascraper = require('../..')([require('metascraper-image')()])
 
   const meta = await metascraper({ url, html, rules })
-  should(meta.foo).be.equal('baz')
+  should(meta.image).be.equal('https://microlink.io/logo.png')
 })