git.squeep.com Git - akkoma/blob - test/pleroma/html_test.exs

   1 # Pleroma: A lightweight social networking server
   2 # Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
   3 # SPDX-License-Identifier: AGPL-3.0-only
   4
   5 defmodule Pleroma.HTMLTest do
   6   alias Pleroma.HTML
   7   alias Pleroma.Object
   8   alias Pleroma.Web.CommonAPI
   9   use Pleroma.DataCase, async: true
  10
  11   import Pleroma.Factory
  12
  13   @html_sample """
  14     <b>this is in bold</b>
  15     <p>this is a paragraph</p>
  16     this is a linebreak<br />
  17     this is a link with allowed "rel" attribute: <a href="http://example.com/" rel="tag">example.com</a>
  18     this is a link with not allowed "rel" attribute: <a href="http://example.com/" rel="tag noallowed">example.com</a>
  19     this is an image: <img src="http://example.com/image.jpg"><br />
  20     <script>alert('hacked')</script>
  21   """
  22
  23   @html_onerror_sample """
  24   <img src="http://example.com/image.jpg" onerror="alert('hacked')">
  25   """
  26
  27   @html_span_class_sample """
  28   <span class="animate-spin">hi</span>
  29   """
  30
  31   @html_span_microformats_sample """
  32   <span class="h-card"><a class="u-url mention">@<span>foo</span></a></span>
  33   """
  34
  35   @html_span_invalid_microformats_sample """
  36   <span class="h-card"><a class="u-url mention animate-spin">@<span>foo</span></a></span>
  37   """
  38
  39   describe "StripTags scrubber" do
  40     test "works as expected" do
  41       expected = """
  42         this is in bold
  43         this is a paragraph
  44         this is a linebreak
  45         this is a link with allowed &quot;rel&quot; attribute: example.com
  46         this is a link with not allowed &quot;rel&quot; attribute: example.com
  47         this is an image:
  48         alert(&#39;hacked&#39;)
  49       """
  50
  51       assert expected == HTML.strip_tags(@html_sample)
  52     end
  53
  54     test "does not allow attribute-based XSS" do
  55       expected = "\n"
  56
  57       assert expected == HTML.strip_tags(@html_onerror_sample)
  58     end
  59   end
  60
  61   describe "TwitterText scrubber" do
  62     test "normalizes HTML as expected" do
  63       expected = """
  64         this is in bold
  65         <p>this is a paragraph</p>
  66         this is a linebreak<br/>
  67         this is a link with allowed &quot;rel&quot; attribute: <a href="http://example.com/" rel="tag">example.com</a>
  68         this is a link with not allowed &quot;rel&quot; attribute: <a href="http://example.com/">example.com</a>
  69         this is an image: <img src="http://example.com/image.jpg"/><br/>
  70         alert(&#39;hacked&#39;)
  71       """
  72
  73       assert expected == HTML.filter_tags(@html_sample, Pleroma.HTML.Scrubber.TwitterText)
  74     end
  75
  76     test "does not allow attribute-based XSS" do
  77       expected = """
  78       <img src="http://example.com/image.jpg"/>
  79       """
  80
  81       assert expected == HTML.filter_tags(@html_onerror_sample, Pleroma.HTML.Scrubber.TwitterText)
  82     end
  83
  84     test "does not allow spans with invalid classes" do
  85       expected = """
  86       <span>hi</span>
  87       """
  88
  89       assert expected ==
  90                HTML.filter_tags(@html_span_class_sample, Pleroma.HTML.Scrubber.TwitterText)
  91     end
  92
  93     test "does allow microformats" do
  94       expected = """
  95       <span class="h-card"><a class="u-url mention">@<span>foo</span></a></span>
  96       """
  97
  98       assert expected ==
  99                HTML.filter_tags(@html_span_microformats_sample, Pleroma.HTML.Scrubber.TwitterText)
 100     end
 101
 102     test "filters invalid microformats markup" do
 103       expected = """
 104       <span class="h-card"><a>@<span>foo</span></a></span>
 105       """
 106
 107       assert expected ==
 108                HTML.filter_tags(
 109                  @html_span_invalid_microformats_sample,
 110                  Pleroma.HTML.Scrubber.TwitterText
 111                )
 112     end
 113   end
 114
 115   describe "default scrubber" do
 116     test "normalizes HTML as expected" do
 117       expected = """
 118         <b>this is in bold</b>
 119         <p>this is a paragraph</p>
 120         this is a linebreak<br/>
 121         this is a link with allowed &quot;rel&quot; attribute: <a href="http://example.com/" rel="tag">example.com</a>
 122         this is a link with not allowed &quot;rel&quot; attribute: <a href="http://example.com/">example.com</a>
 123         this is an image: <img src="http://example.com/image.jpg"/><br/>
 124         alert(&#39;hacked&#39;)
 125       """
 126
 127       assert expected == HTML.filter_tags(@html_sample, Pleroma.HTML.Scrubber.Default)
 128     end
 129
 130     test "does not allow attribute-based XSS" do
 131       expected = """
 132       <img src="http://example.com/image.jpg"/>
 133       """
 134
 135       assert expected == HTML.filter_tags(@html_onerror_sample, Pleroma.HTML.Scrubber.Default)
 136     end
 137
 138     test "does not allow spans with invalid classes" do
 139       expected = """
 140       <span>hi</span>
 141       """
 142
 143       assert expected == HTML.filter_tags(@html_span_class_sample, Pleroma.HTML.Scrubber.Default)
 144     end
 145
 146     test "does allow microformats" do
 147       expected = """
 148       <span class="h-card"><a class="u-url mention">@<span>foo</span></a></span>
 149       """
 150
 151       assert expected ==
 152                HTML.filter_tags(@html_span_microformats_sample, Pleroma.HTML.Scrubber.Default)
 153     end
 154
 155     test "filters invalid microformats markup" do
 156       expected = """
 157       <span class="h-card"><a>@<span>foo</span></a></span>
 158       """
 159
 160       assert expected ==
 161                HTML.filter_tags(
 162                  @html_span_invalid_microformats_sample,
 163                  Pleroma.HTML.Scrubber.Default
 164                )
 165     end
 166   end
 167
 168   describe "extract_first_external_url_from_object" do
 169     test "extracts the url" do
 170       user = insert(:user)
 171
 172       {:ok, activity} =
 173         CommonAPI.post(user, %{
 174           status:
 175             "I think I just found the best github repo https://github.com/komeiji-satori/Dress"
 176         })
 177
 178       object = Object.normalize(activity, fetch: false)
 179       {:ok, url} = HTML.extract_first_external_url_from_object(object)
 180       assert url == "https://github.com/komeiji-satori/Dress"
 181     end
 182
 183     test "skips mentions" do
 184       user = insert(:user)
 185       other_user = insert(:user)
 186
 187       {:ok, activity} =
 188         CommonAPI.post(user, %{
 189           status:
 190             "@#{other_user.nickname} install misskey! https://github.com/syuilo/misskey/blob/develop/docs/setup.en.md"
 191         })
 192
 193       object = Object.normalize(activity, fetch: false)
 194       {:ok, url} = HTML.extract_first_external_url_from_object(object)
 195
 196       assert url == "https://github.com/syuilo/misskey/blob/develop/docs/setup.en.md"
 197
 198       refute url == other_user.ap_id
 199     end
 200
 201     test "skips hashtags" do
 202       user = insert(:user)
 203
 204       {:ok, activity} =
 205         CommonAPI.post(user, %{
 206           status: "#cofe https://www.pixiv.net/member_illust.php?mode=medium&illust_id=72255140"
 207         })
 208
 209       object = Object.normalize(activity, fetch: false)
 210       {:ok, url} = HTML.extract_first_external_url_from_object(object)
 211
 212       assert url == "https://www.pixiv.net/member_illust.php?mode=medium&illust_id=72255140"
 213     end
 214
 215     test "skips microformats hashtags" do
 216       user = insert(:user)
 217
 218       {:ok, activity} =
 219         CommonAPI.post(user, %{
 220           status:
 221             "<a href=\"https://pleroma.gov/tags/cofe\" rel=\"tag\">#cofe</a> https://www.pixiv.net/member_illust.php?mode=medium&illust_id=72255140",
 222           content_type: "text/html"
 223         })
 224
 225       object = Object.normalize(activity, fetch: false)
 226       {:ok, url} = HTML.extract_first_external_url_from_object(object)
 227
 228       assert url == "https://www.pixiv.net/member_illust.php?mode=medium&illust_id=72255140"
 229     end
 230
 231     test "does not crash when there is an HTML entity in a link" do
 232       user = insert(:user)
 233
 234       {:ok, activity} = CommonAPI.post(user, %{status: "\"http://cofe.com/?boomer=ok&foo=bar\""})
 235
 236       object = Object.normalize(activity, fetch: false)
 237
 238       assert {:ok, nil} = HTML.extract_first_external_url_from_object(object)
 239     end
 240
 241     test "skips attachment links" do
 242       user = insert(:user)
 243
 244       {:ok, activity} =
 245         CommonAPI.post(user, %{
 246           status:
 247             "<a href=\"https://pleroma.gov/media/d24caa3a498e21e0298377a9ca0149a4f4f8b767178aacf837542282e2d94fb1.png?name=image.png\" class=\"attachment\">image.png</a>"
 248         })
 249
 250       object = Object.normalize(activity, fetch: false)
 251
 252       assert {:ok, nil} = HTML.extract_first_external_url_from_object(object)
 253     end
 254   end
 255 end