stop remark from decoding HTML entities

src/components/Widgets/Markdown/serializers/__tests__/remarkAllowHtmlEntities.spec.js

@@ -0,0 +1,24 @@
+import unified from 'unified';
+import markdownToRemark from 'remark-parse';
+import remarkAllowHtmlEntities from '../remarkAllowHtmlEntities';
+
+const process = markdown => {
+  const mdast = unified().use(markdownToRemark).use(remarkAllowHtmlEntities).parse(markdown);
+
+  /**
+   * The MDAST will look like:
+   *
+   * { type: 'root', children: [
+   *   { type: 'paragraph', children: [
+   *     // results here
+   *   ]}
+   * ]}
+   */
+  return mdast.children[0].children[0].value;
+};
+
+describe('remarkAllowHtmlEntities', () => {
+  it('should not decode HTML entities', () => {
+    expect(process('<div>')).toEqual('<div>');
+  });
+});

src/components/Widgets/Markdown/serializers/__tests__/slate.spec.js

@@ -8,7 +8,7 @@ describe('slate', () => {
     expect(process('a\n')).toEqual('a\n');
   });
 
-  xit('should not decode encoded html entities in inline code', () => {
+  it('should not decode encoded html entities in inline code', () => {
     expect(process('<code>&lt;div&gt;</code>')).toEqual('<code>&lt;div&gt;</code>\n');
   });
 

src/components/Widgets/Markdown/serializers/index.js

@@ -18,6 +18,7 @@ import remarkImagesToText from './remarkImagesToText';
 import remarkShortcodes from './remarkShortcodes';
 import remarkEscapeMarkdownEntities from './remarkEscapeMarkdownEntities';
 import remarkStripTrailingBreaks from './remarkStripTrailingBreaks';
+import remarkAllowHtmlEntities from './remarkAllowHtmlEntities';
 import slateToRemark from './slateRemark';
 import registry from '../../../../lib/registry';
 
@@ -66,6 +67,7 @@ export const markdownToRemark = markdown => {
   const parsed = unified()
     .use(markdownToRemarkPlugin, { fences: true, commonmark: true })
     .use(markdownToRemarkRemoveTokenizers, { inlineTokenizers: ['url'] })
+    .use(remarkAllowHtmlEntities)
     .parse(markdown);
 
   /**

src/components/Widgets/Markdown/serializers/remarkAllowHtmlEntities.js

@@ -0,0 +1,59 @@
+export default function remarkAllowHtmlEntities() {
+  this.Parser.prototype.inlineTokenizers.text = text;
+
+  /**
+   * This is a port of the `remark-parse` text tokenizer, adapted to exclude
+   * HTML entity decoding.
+   */
+  function text(eat, value, silent) {
+    var self = this;
+    var methods;
+    var tokenizers;
+    var index;
+    var length;
+    var subvalue;
+    var position;
+    var tokenizer;
+    var name;
+    var min;
+    var now;
+
+    /* istanbul ignore if - never used (yet) */
+    if (silent) {
+      return true;
+    }
+
+    methods = self.inlineMethods;
+    length = methods.length;
+    tokenizers = self.inlineTokenizers;
+    index = -1;
+    min = value.length;
+
+    while (++index < length) {
+      name = methods[index];
+
+      if (name === 'text' || !tokenizers[name]) {
+        continue;
+      }
+
+      tokenizer = tokenizers[name].locator;
+
+      if (!tokenizer) {
+        eat.file.fail('Missing locator: `' + name + '`');
+      }
+
+      position = tokenizer.call(self, value, 1);
+
+      if (position !== -1 && position < min) {
+        min = position;
+      }
+    }
+
+    subvalue = value.slice(0, min);
+
+    eat(subvalue)({
+      type: 'text',
+      value: subvalue,
+    });
+  }
+};