From 7a5063d10564bd2bb0142c532a845b963ca533fd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Julius=20H=C3=A4rtl?= <jus@bitgrid.net>
Date: Tue, 2 Jul 2019 20:22:20 +0200
Subject: Use text node to serialize plain text
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Julius Härtl <jus@bitgrid.net>
---
 src/EditorFactory.js        | 16 +++++++++++++---
 src/tests/markdown.spec.js  |  3 +++
 src/tests/plaintext.spec.js | 18 ++++++++++++++++--
 3 files changed, 32 insertions(+), 5 deletions(-)

diff --git a/src/EditorFactory.js b/src/EditorFactory.js
index 3a991f01b..ec9a8835e 100644
--- a/src/EditorFactory.js
+++ b/src/EditorFactory.js
@@ -101,6 +101,9 @@ const createEditor = ({ content, onUpdate, extensions, enableRichEditing, langua
 const markdownit = MarkdownIt('commonmark', { html: false, breaks: false })
 	.enable('strikethrough')
 
+const SerializeException = (message) => {
+	this.message = message
+}
 const createMarkdownSerializer = (_nodes, _marks) => {
 	const nodes = Object
 		.entries(_nodes)
@@ -124,9 +127,16 @@ const createMarkdownSerializer = (_nodes, _marks) => {
 }
 
 const serializePlainText = (tiptap) => {
-	const tmp = document.createElement('div')
-	tmp.innerHTML = tiptap.getHTML()
-	return tmp.textContent || tmp.innerText || ''
+	const doc = tiptap.getJSON()
+
+	if (doc.content.length !== 1 || doc.content[0].content.length !== 1) {
+		throw new SerializeException('Failed to serialize document to plain text')
+	}
+	const codeBlock = doc.content[0].content[0]
+	if (codeBlock.type !== 'text') {
+		throw new SerializeException('Failed to serialize document to plain text')
+	}
+	return codeBlock.text
 }
 
 export default createEditor
diff --git a/src/tests/markdown.spec.js b/src/tests/markdown.spec.js
index a191d15c1..cf011929b 100644
--- a/src/tests/markdown.spec.js
+++ b/src/tests/markdown.spec.js
@@ -75,6 +75,9 @@ describe('Markdown though editor', () => {
   test('images', () => {
     expect(markdownThroughEditor('![test](foo)')).toBe('![test](foo)')
   })
+  test('special characters', () => {
+    expect(markdownThroughEditor('"\';&.-#><')).toBe('"\';&.-#><')
+  })
 })
 
 describe('Markdown serializer from html', () => {
diff --git a/src/tests/plaintext.spec.js b/src/tests/plaintext.spec.js
index 19f87d40f..a139a4a86 100644
--- a/src/tests/plaintext.spec.js
+++ b/src/tests/plaintext.spec.js
@@ -1,5 +1,6 @@
 import { markdownit, createEditor, createMarkdownSerializer, serializePlainText } from './../EditorFactory';
 import spec from "./fixtures/spec"
+import xssFuzzVectors from './fixtures/xssFuzzVectors';
 
 const escapeHTML = (s) => {
   return s.toString()
@@ -19,7 +20,7 @@ const plaintextThroughEditor = (markdown) => {
   return serializePlainText(tiptap) || 'failed'
 }
 
-describe('Markdown though editor', () => {
+describe('commonmark as plaintext', () => {
   // FIXME: Those two tests currently fail as trailing whitespace seems to be stripped,
   // if it occurs in the first line which is empty otherwise.
   const skippedMarkdownTests = [
@@ -34,7 +35,8 @@ describe('Markdown though editor', () => {
       expect(plaintextThroughEditor(entry.markdown)).toBe(entry.markdown)
     })
   })
-
+})
+describe('markdown as plaintext', () => {
   test('headlines', () => {
     expect(plaintextThroughEditor('# Test')).toBe('# Test')
     expect(plaintextThroughEditor('## Test')).toBe('## Test')
@@ -66,3 +68,15 @@ describe('Markdown though editor', () => {
     expect(plaintextThroughEditor('![test](foo)')).toBe('![test](foo)')
   })
 })
+
+describe('html as plain text', () => {
+  test('link', () => {
+    expect(plaintextThroughEditor('<a>sdf</a>')).toBe('<a>sdf</a>')
+    expect(plaintextThroughEditor('<a href="foobar">sdf</a>')).toBe('<a href="foobar">sdf</a>')
+
+  })
+  test('special characters', () => {
+    expect(plaintextThroughEditor('"\';&.-#><')).toBe('"\';&.-#><')
+    expect(plaintextThroughEditor(xssFuzzVectors)).toBe(xssFuzzVectors)
+  })
+})
-- 
cgit v1.2.3