Avoid rewrite html charset

Kikobeats · Kikobeats · commit 9159dd2f5826 · 2018-11-17T22:08:17.000+01:00
diff --git a/index.js b/index.js
@@ -4,9 +4,6 @@ const jschardet = require('jschardet')
 const isBuffer = require('is-buffer')
 const iconv = require('iconv-lite')
 const charset = require('charset')
-const he = require('he')
-
-const REGEX_CHARSET = /charset=["]*([^>"\s]+)/i
 
 const inferredEncoding = content => {
   const charset = jschardet.detect(content)
@@ -26,13 +23,6 @@ module.exports = targetEncoding => {
   return (buffer, contentType) => {
     if (!isBuffer(buffer)) throw new TypeError('content should be a buffer.')
     const encoding = getEncoding(buffer, contentType)
-
-    const str = iconv
-      .decode(buffer, encoding)
-      .replace(REGEX_CHARSET, targetEncoding)
-      .toString()
-
-    // Ensure to resolve Base64 entities
-    return he.decode(str)
+    return iconv.decode(buffer, encoding)
   }
 }
diff --git a/package.json b/package.json
@@ -26,11 +26,9 @@
   ],
   "dependencies": {
     "charset": "~1.0.1",
-    "he": "~1.2.0",
     "iconv-lite": "~0.4.19",
     "is-buffer": "~2.0.0",
-    "jschardet": "~1.6.0",
-    "to-buffer": "~1.1.0"
+    "jschardet": "~1.6.0"
   },
   "devDependencies": {
     "chai": "latest",
diff --git a/test/fixtures/base64.html b/test/fixtures/base64.html
diff --git a/test/index.js b/test/index.js
@@ -6,70 +6,33 @@ const fs = require('fs')
 
 const toUTF8 = require('..')('utf-8')
 
-describe('Encoding Converter', function () {
-  it('properly decodes Shift-JIS html documents', function () {
+describe('encoding conversion', function () {
+  it('Shift-JIS', function () {
     const buffer = loadExample('51242_54045.html')
     const output = toUTF8(buffer, 'text/html')
 
     expect(output).to.contain('或る日の小せん')
   })
 
-  it('properly decodes Windows-1250 html documents', function () {
+  it('Windows-1250', function () {
     const buffer = loadExample('rp.pl.html')
     const output = toUTF8(buffer, 'windows-1250')
 
     expect(output).to.contain('majątków')
   })
 
-  it('guesses encoding even without meta tags or content type', function () {
-    const buffer = loadExample('shiftjis.no.meta.html')
-    const output = toUTF8(buffer, 'text/html')
-
-    expect(output).to.contain('次常用國字標準字體表')
-  })
-
-  it('works for documents which already are UTF-8', function () {
+  it('UTF-8', function () {
     const buffer = loadExample('utf8.with.meta.html')
     const output = toUTF8(buffer, 'text/html')
 
     expect(output).to.contain('日本語')
   })
 
-  it('Replace charset from the original buffer', function () {
-    const buffer = loadExample('51242_54045.html')
+  it('inferred', function () {
+    const buffer = loadExample('shiftjis.no.meta.html')
     const output = toUTF8(buffer, 'text/html')
 
-    expect(output).to.contain(
-      '<meta http-equiv="Content-Type" content="text/html;utf-8" />'
-    )
-  })
-
-  it.only('Decode Base64', function () {
-    const buffer = loadExample('base64.html')
-    const output = toUTF8(buffer)
-    expect(output.trim()).to.deep.equal(
-      `
-    <html>
-
-<head></head>
-
-<body>
-  <pre style="word-wrap: break-word; white-space: pre-wrap;"><!DOCTYPE html>
-  <html lang="en">
-  <head>
-    <meta 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>
-    <a href="https://httpbin-org.herokuapp.com/redirect/3"></a>
-  </body>
-  </html></pre>
-</body>
-
-</html>`.trim()
-    )
+    expect(output).to.contain('次常用國字標準字體表')
   })
 })
 
