Use WagtailTestUtils.get_soup() more widely

wagtail/admin/tests/pages/test_create_page.py

@@ -2,7 +2,6 @@ import datetime
 import unittest
 from unittest import mock
 
-from bs4 import BeautifulSoup
 from django.contrib.auth.models import Group, Permission
 from django.http import HttpRequest, HttpResponse
 from django.test import TestCase
@@ -1022,7 +1021,7 @@ class TestPageCreation(WagtailTestUtils, TestCase):
             )
         )
 
-        html = BeautifulSoup(response.content, "html5lib")
+        html = self.get_soup(response.content)
 
         actual_attrs = html.find("input", {"name": "title"}).attrs
 

wagtail/admin/tests/test_dbwhitelister.py

@@ -1,22 +1,22 @@
-from bs4 import BeautifulSoup
 from django.test import TestCase
 
 from wagtail.admin.rich_text.converters.editor_html import EditorHTMLConverter
+from wagtail.test.utils import WagtailTestUtils
 
 
-class TestDbWhitelisterMethods(TestCase):
+class TestDbWhitelisterMethods(WagtailTestUtils, TestCase):
     def setUp(self):
         self.whitelister = EditorHTMLConverter().whitelister
 
     def test_clean_tag_node_div(self):
-        soup = BeautifulSoup("<div>foo</div>", "html5lib")
+        soup = self.get_soup("<div>foo</div>", "html5lib")
         tag = soup.div
         self.assertEqual(tag.name, "div")
         self.whitelister.clean_tag_node(soup, tag)
         self.assertEqual(tag.name, "p")
 
     def test_clean_tag_node_with_data_embedtype(self):
-        soup = BeautifulSoup(
+        soup = self.get_soup(
             '<p><a data-embedtype="image" data-id=1 data-format="left" data-alt="bar" irrelevant="baz">foo</a></p>',
             "html5lib",
         )
@@ -27,7 +27,7 @@ class TestDbWhitelisterMethods(TestCase):
         )
 
     def test_clean_tag_node_with_data_linktype(self):
-        soup = BeautifulSoup(
+        soup = self.get_soup(
             '<a data-linktype="document" data-id="1" irrelevant="baz">foo</a>',
             "html5lib",
         )
@@ -36,13 +36,13 @@ class TestDbWhitelisterMethods(TestCase):
         self.assertEqual(str(tag), '<a id="1" linktype="document">foo</a>')
 
     def test_clean_tag_node(self):
-        soup = BeautifulSoup('<a irrelevant="baz">foo</a>', "html5lib")
+        soup = self.get_soup('<a irrelevant="baz">foo</a>', "html5lib")
         tag = soup.a
         self.whitelister.clean_tag_node(soup, tag)
         self.assertEqual(str(tag), "<a>foo</a>")
 
 
-class TestDbWhitelister(TestCase):
+class TestDbWhitelister(WagtailTestUtils, TestCase):
     def setUp(self):
         self.whitelister = EditorHTMLConverter().whitelister
 
@@ -52,7 +52,7 @@ class TestDbWhitelister(TestCase):
         (necessary because we can't guarantee the order that attributes are output in)
         """
         self.assertEqual(
-            BeautifulSoup(str1, "html5lib"), BeautifulSoup(str2, "html5lib")
+            self.get_soup(str1, "html5lib"), self.get_soup(str2, "html5lib")
         )
 
     def test_page_link_is_rewritten(self):

wagtail/admin/tests/test_edit_handlers.py

@@ -3,7 +3,6 @@ from functools import wraps
 from typing import Any, List, Mapping, Optional
 from unittest import mock
 
-from bs4 import BeautifulSoup
 from django import forms
 from django.conf import settings
 from django.contrib.auth import get_user_model
@@ -2223,7 +2222,7 @@ class TestTitleFieldPanel(WagtailTestUtils, TestCase):
             instance=instance,
         )
         html = bound_edit_handler.render_form_content()
-        return BeautifulSoup(html, "html5lib")
+        return self.get_soup(html)
 
     @clear_edit_handler(Page)
     def test_default_page_content_panels_uses_title_field(self):

wagtail/admin/tests/test_rich_text.py

@@ -1,6 +1,5 @@
 import unittest
 
-from bs4 import BeautifulSoup
 from django.conf import settings
 from django.test import SimpleTestCase, TestCase
 from django.test.utils import override_settings
@@ -431,11 +430,11 @@ class TestDraftailWithAdditionalFeatures(
         self.assertNotContains(response, '"type": "ITALIC"')
 
 
-class TestPageLinkHandler(TestCase):
+class TestPageLinkHandler(WagtailTestUtils, TestCase):
     fixtures = ["test.json"]
 
     def test_get_db_attributes(self):
-        soup = BeautifulSoup('<a data-id="test-id">foo</a>', "html5lib")
+        soup = self.get_soup('<a data-id="test-id">foo</a>')
         tag = soup.a
         result = PageLinkHandler.get_db_attributes(tag)
         self.assertEqual(result, {"id": "test-id"})

wagtail/admin/tests/test_userbar.py

@@ -1,6 +1,5 @@
 import json
 
-from bs4 import BeautifulSoup
 from django.contrib.auth.models import AnonymousUser, Permission
 from django.template import Context, Template
 from django.test import TestCase
@@ -193,7 +192,7 @@ class TestAccessibilityCheckerConfig(WagtailTestUtils, TestCase):
     def get_script(self):
         template = Template("{% load wagtailuserbar %}{% wagtailuserbar %}")
         content = template.render(Context({"request": self.request}))
-        soup = BeautifulSoup(content, "html.parser")
+        soup = self.get_soup(content)
 
         # Should include the configuration as a JSON script with the specific id
         return soup.find("script", id="accessibility-axe-configuration")

wagtail/documents/tests/test_rich_text.py

@@ -1,4 +1,3 @@
-from bs4 import BeautifulSoup
 from django.test import TestCase
 
 from wagtail.documents import get_document_model
@@ -9,13 +8,14 @@ from wagtail.documents.rich_text.editor_html import (
     DocumentLinkHandler as EditorHtmlDocumentLinkHandler,
 )
 from wagtail.fields import RichTextField
+from wagtail.test.utils import WagtailTestUtils
 
 
-class TestEditorHtmlDocumentLinkHandler(TestCase):
+class TestEditorHtmlDocumentLinkHandler(WagtailTestUtils, TestCase):
     fixtures = ["test.json"]
 
     def test_get_db_attributes(self):
-        soup = BeautifulSoup('<a data-id="test-id">foo</a>', "html5lib")
+        soup = self.get_soup('<a data-id="test-id">foo</a>')
         tag = soup.a
         result = EditorHtmlDocumentLinkHandler.get_db_attributes(tag)
         self.assertEqual(result, {"id": "test-id"})

wagtail/embeds/tests/test_rich_text.py

@@ -1,6 +1,5 @@
 from unittest.mock import patch
 
-from bs4 import BeautifulSoup
 from django.test import TestCase, override_settings
 
 from wagtail.embeds.exceptions import EmbedNotFoundException
@@ -10,11 +9,12 @@ from wagtail.embeds.rich_text.editor_html import (
     MediaEmbedHandler as EditorHtmlMediaEmbedHandler,
 )
 from wagtail.rich_text import expand_db_html
+from wagtail.test.utils import WagtailTestUtils
 
 
-class TestEditorHtmlMediaEmbedHandler(TestCase):
+class TestEditorHtmlMediaEmbedHandler(WagtailTestUtils, TestCase):
     def test_get_db_attributes(self):
-        soup = BeautifulSoup('<b data-url="test-url">foo</b>', "html5lib")
+        soup = self.get_soup('<b data-url="test-url">foo</b>')
         tag = soup.b
         result = EditorHtmlMediaEmbedHandler.get_db_attributes(tag)
         self.assertEqual(result, {"url": "test-url"})

wagtail/images/tests/test_rich_text.py

@@ -1,4 +1,3 @@
-from bs4 import BeautifulSoup
 from django.test import TestCase
 
 from wagtail.fields import RichTextField
@@ -13,9 +12,8 @@ from .utils import Image, get_test_image_file
 
 class TestEditorHtmlImageEmbedHandler(WagtailTestUtils, TestCase):
     def test_get_db_attributes(self):
-        soup = BeautifulSoup(
+        soup = self.get_soup(
             '<b data-id="test-id" data-format="test-format" data-alt="test-alt">foo</b>',
-            "html5lib",
         )
         tag = soup.b
         result = EditorHtmlImageEmbedHandler.get_db_attributes(tag)

wagtail/test/utils/form_data.py

@@ -10,6 +10,8 @@ from django.http import QueryDict
 
 from wagtail.admin.rich_text import get_rich_text_editor_widget
 
+from .wagtail_tests import WagtailTestUtils
+
 
 def _nested_form_data(data):
     if isinstance(data, dict):
@@ -186,7 +188,7 @@ def _querydict_from_form(form: bs4.Tag, exclude_csrf: bool = True) -> QueryDict:
 def querydict_from_html(
     html: str, form_id: str = None, form_index: int = 0, exclude_csrf: bool = True
 ) -> QueryDict:
-    soup = bs4.BeautifulSoup(html, "html5lib")
+    soup = WagtailTestUtils.get_soup(html)
     if form_id is not None:
         form = soup.find("form", attrs={"id": form_id})
         if form is None:

wagtail/tests/test_whitelist.py

@@ -1,6 +1,6 @@
-from bs4 import BeautifulSoup
 from django.test import TestCase
 
+from wagtail.test.utils import WagtailTestUtils
 from wagtail.whitelist import (
     Whitelister,
     allow_without_attributes,
@@ -26,9 +26,9 @@ class TestCheckUrl(TestCase):
         self.assertFalse(bool(check_url("jav\tascript:alert('XSS')")))
 
 
-class TestAttributeRule(TestCase):
+class TestAttributeRule(WagtailTestUtils, TestCase):
     def setUp(self):
-        self.soup = BeautifulSoup('<b foo="bar">baz</b>', "html5lib")
+        self.soup = self.get_soup('<b foo="bar">baz</b>', "html5lib")
 
     def test_no_rule_for_attr(self):
         """
@@ -87,7 +87,7 @@ class TestAttributeRule(TestCase):
         Test that attribute_rule() with will drop all
         attributes.
         """
-        soup = BeautifulSoup(
+        soup = self.get_soup(
             '<b foo="bar" baz="quux" snowman="barbecue"></b>', "html5lib"
         )
         tag = soup.b
@@ -95,7 +95,7 @@ class TestAttributeRule(TestCase):
         self.assertEqual(str(tag), "<b></b>")
 
 
-class TestWhitelister(TestCase):
+class TestWhitelister(WagtailTestUtils, TestCase):
     def setUp(self):
         self.whitelister = Whitelister()
 
@@ -103,7 +103,7 @@ class TestWhitelister(TestCase):
         """
         Unknown node should remove a node from the parent document
         """
-        soup = BeautifulSoup("<foo><bar>baz</bar>quux</foo>", "html5lib")
+        soup = self.get_soup("<foo><bar>baz</bar>quux</foo>", "html5lib")
         tag = soup.foo
         self.whitelister.clean_unknown_node("", soup.bar)
         self.assertEqual(str(tag), "<foo>quux</foo>")
@@ -113,7 +113,7 @@ class TestWhitelister(TestCase):
         <b> tags are allowed without attributes. This remains true
         when tags are nested.
         """
-        soup = BeautifulSoup('<b><b class="delete me">foo</b></b>', "html5lib")
+        soup = self.get_soup('<b><b class="delete me">foo</b></b>', "html5lib")
         tag = soup.b
         self.whitelister.clean_tag_node(tag, tag)
         self.assertEqual(str(tag), "<b><b>foo</b></b>")
@@ -122,19 +122,19 @@ class TestWhitelister(TestCase):
         """
         <foo> tags should be removed, even when nested.
         """
-        soup = BeautifulSoup("<b><foo>bar</foo></b>", "html5lib")
+        soup = self.get_soup("<b><foo>bar</foo></b>", "html5lib")
         tag = soup.b
         self.whitelister.clean_tag_node(tag, tag)
         self.assertEqual(str(tag), "<b>bar</b>")
 
     def test_clean_string_node_does_nothing(self):
-        soup = BeautifulSoup("<b>bar</b>", "html5lib")
+        soup = self.get_soup("<b>bar</b>", "html5lib")
         string = soup.b.string
         self.whitelister.clean_string_node(string, string)
         self.assertEqual(str(string), "bar")
 
     def test_clean_node_does_not_change_navigable_strings(self):
-        soup = BeautifulSoup("<b>bar</b>", "html5lib")
+        soup = self.get_soup("<b>bar</b>", "html5lib")
         string = soup.b.string
         self.whitelister.clean_node(string, string)
         self.assertEqual(str(string), "bar")