From 981ae74e5cda458608d64a1aa3b62cb62a8fc4f6 Mon Sep 17 00:00:00 2001
From: rubenwardy <rw@rubenwardy.com>
Date: Wed, 22 Jan 2020 22:10:02 +0000
Subject: [PATCH] Improve markdown escaping

Fixes #118
---
 app/__init__.py                  |  6 ++-
 app/blueprints/users/profile.py  |  4 +-
 app/markdown.py                  | 63 ++++++++++++++++++++++++++++++++
 app/templates/packages/view.html |  2 +-
 app/templates/todo/topics.html   |  2 +-
 requirements.txt                 |  4 +-
 6 files changed, 74 insertions(+), 7 deletions(-)
 create mode 100644 app/markdown.py

diff --git a/app/__init__.py b/app/__init__.py
index 478f0d8..d064087 100644
--- a/app/__init__.py
+++ b/app/__init__.py
@@ -20,7 +20,6 @@ from flask_user import *
 from flask_gravatar import Gravatar
 import flask_menu as menu
 from flask_mail import Mail
-from flaskext.markdown import Markdown
 from flask_github import GitHub
 from flask_wtf.csrf import CsrfProtect
 from flask_flatpages import FlatPages
@@ -35,7 +34,6 @@ app.config.from_pyfile(os.environ["FLASK_CONFIG"])
 r = redis.Redis.from_url(app.config["REDIS_URL"])
 
 menu.Menu(app=app)
-markdown = Markdown(app, extensions=["fenced_code"], safe_mode=True, output_format="html5")
 github = GitHub(app)
 csrf = CsrfProtect(app)
 mail = Mail(app)
@@ -59,6 +57,10 @@ if not app.debug and app.config["MAIL_UTILS_ERROR_SEND_TO"]:
 	register_mail_error_handler(app, mail)
 
 
+from .markdown import init_app
+init_app(app)
+
+
 @babel.localeselector
 def get_locale():
 	return request.accept_languages.best_match(app.config['LANGUAGES'].keys())
diff --git a/app/blueprints/users/profile.py b/app/blueprints/users/profile.py
index 37ea992..0036fa9 100644
--- a/app/blueprints/users/profile.py
+++ b/app/blueprints/users/profile.py
@@ -18,7 +18,7 @@
 from flask import *
 from flask_user import *
 from flask_login import login_user, logout_user
-from app import markdown
+from app.markdown import render_markdown
 from . import bp
 from app.models import *
 from flask_wtf import FlaskForm
@@ -153,7 +153,7 @@ def send_email(username):
 	form = SendEmailForm(request.form)
 	if form.validate_on_submit():
 		text = form.text.data
-		html = markdown(text)
+		html = render_markdown(text)
 		task = sendEmailRaw.delay([user.email], form.subject.data, text, html)
 		return redirect(url_for("tasks.check", id=task.id, r=next_url))
 
diff --git a/app/markdown.py b/app/markdown.py
new file mode 100644
index 0000000..370d3bb
--- /dev/null
+++ b/app/markdown.py
@@ -0,0 +1,63 @@
+import bleach
+from markdown import Markdown
+from flask import Markup
+
+# Whitelist source: MIT
+#
+# https://github.com/Wenzil/mdx_bleach/blob/master/mdx_bleach/whitelist.py
+
+"""
+Default whitelist of allowed HTML tags. Any other HTML tags will be escaped or
+stripped from the text. This applies to the html output that Markdown produces.
+"""
+ALLOWED_TAGS = [
+	'ul',
+	'ol',
+	'li',
+	'p',
+	'pre',
+	'code',
+	'blockquote',
+	'h1',
+	'h2',
+	'h3',
+	'h4',
+	'h5',
+	'h6',
+	'hr',
+	'br',
+	'strong',
+	'em',
+	'a',
+	'img'
+]
+
+"""
+Default whitelist of attributes. It allows the href and title attributes for <a>
+tags and the src, title and alt attributes for <img> tags. Any other attribute
+will be stripped from its tag.
+"""
+ALLOWED_ATTRIBUTES = {
+	'a': ['href', 'title'],
+	'img': ['src', 'title', 'alt']
+}
+
+"""
+If you allow tags that have attributes containing a URI value
+(like the href attribute of an anchor tag,) you may want to adapt
+the accepted protocols. The default list only allows http, https and mailto.
+"""
+ALLOWED_PROTOCOLS = ['http', 'https', 'mailto']
+
+
+md = Markdown(extensions=["fenced_code"], output_format="html5")
+
+def render_markdown(source):
+	return bleach.clean(md.convert(source), \
+		tags=ALLOWED_TAGS, attributes=ALLOWED_ATTRIBUTES, \
+		styles=[], protocols=ALLOWED_PROTOCOLS)
+
+def init_app(app):
+	@app.template_filter()
+	def markdown(source):
+		return Markup(render_markdown(source))
diff --git a/app/templates/packages/view.html b/app/templates/packages/view.html
index 0a4b2bf..77b84da 100644
--- a/app/templates/packages/view.html
+++ b/app/templates/packages/view.html
@@ -364,7 +364,7 @@
 				</ul>
 			</div>
 
-			<div class="card my-4"">
+			<div class="card my-4">
 				<div class="card-header">
 					{% if package.approved and package.checkPerm(current_user, "CREATE_THREAD") %}
 						<div class="btn-group float-right">
diff --git a/app/templates/todo/topics.html b/app/templates/todo/topics.html
index e51f4ec..87e142f 100644
--- a/app/templates/todo/topics.html
+++ b/app/templates/todo/topics.html
@@ -60,7 +60,7 @@ Topics to be Added
 			{% set perc = 100 * (total - topic_count) /  total %}
 			<div class="progress-bar bg-success" role="progressbar"
 				style="width: {{ perc }}%" aria-valuenow="{{ perc }}" aria-valuemin="0" aria-valuemax="100"></div>
-		</div>	
+		</div>
 	{% else %}
 		<p>
 			The forum topic crawler needs to run at least once for this section to work.
diff --git a/requirements.txt b/requirements.txt
index 9473f9f..104755b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,7 +2,6 @@ Flask~=1.1
 Flask-FlatPages~=0.7
 Flask-Gravatar~=0.5
 Flask-Login~=0.4.1
-Flask-Markdown~=0.3
 Flask-Menu~=0.7
 Flask-Migrate~=2.3
 Flask-SQLAlchemy~=2.3
@@ -11,6 +10,9 @@ Flask-Babel
 GitHub-Flask~=3.2
 SQLAlchemy-Searchable~=1.1
 
+markdown ~= 3.1
+bleach ~= 3.1
+
 beautifulsoup4~=4.6
 celery~=4.4
 kombu~=4.6