summary refs log tree commit diff
path: root/tests
diff options
context:
space:
mode:
authorPatrick Cloke <clokep@users.noreply.github.com>2022-01-24 08:58:18 -0500
committerGitHub <noreply@github.com>2022-01-24 08:58:18 -0500
commit807efd26aec9b65c6a2f02d10fd139095a5b3387 (patch)
tree86b81a5bbd5fc8fa3985c9d2a77de3ad01b24320 /tests
parentMerge tag 'v1.51.0rc2' into develop (diff)
downloadsynapse-807efd26aec9b65c6a2f02d10fd139095a5b3387.tar.xz
Support rendering previews with data: URLs in them (#11767)
Images which are data URLs will no longer break URL
previews and will properly be "downloaded" and
thumbnailed.
Diffstat (limited to '')
-rw-r--r--tests/rest/media/v1/test_html_preview.py (renamed from tests/test_preview.py)34
-rw-r--r--tests/rest/media/v1/test_url_preview.py81
-rw-r--r--tests/server.py2
3 files changed, 107 insertions, 10 deletions
diff --git a/tests/test_preview.py b/tests/rest/media/v1/test_html_preview.py
index 46e02f483f..a4b57e3d1f 100644
--- a/tests/test_preview.py
+++ b/tests/rest/media/v1/test_html_preview.py
@@ -16,10 +16,11 @@ from synapse.rest.media.v1.preview_html import (
     _get_html_media_encodings,
     decode_body,
     parse_html_to_open_graph,
+    rebase_url,
     summarize_paragraphs,
 )
 
-from . import unittest
+from tests import unittest
 
 try:
     import lxml
@@ -447,3 +448,34 @@ class MediaEncodingTestCase(unittest.TestCase):
             'text/html; charset="invalid"',
         )
         self.assertEqual(list(encodings), ["utf-8", "cp1252"])
+
+
+class RebaseUrlTestCase(unittest.TestCase):
+    def test_relative(self):
+        """Relative URLs should be resolved based on the context of the base URL."""
+        self.assertEqual(
+            rebase_url("subpage", "https://example.com/foo/"),
+            "https://example.com/foo/subpage",
+        )
+        self.assertEqual(
+            rebase_url("sibling", "https://example.com/foo"),
+            "https://example.com/sibling",
+        )
+        self.assertEqual(
+            rebase_url("/bar", "https://example.com/foo/"),
+            "https://example.com/bar",
+        )
+
+    def test_absolute(self):
+        """Absolute URLs should not be modified."""
+        self.assertEqual(
+            rebase_url("https://alice.com/a/", "https://example.com/foo/"),
+            "https://alice.com/a/",
+        )
+
+    def test_data(self):
+        """Data URLs should not be modified."""
+        self.assertEqual(
+            rebase_url("data:,Hello%2C%20World%21", "https://example.com/foo/"),
+            "data:,Hello%2C%20World%21",
+        )
diff --git a/tests/rest/media/v1/test_url_preview.py b/tests/rest/media/v1/test_url_preview.py
index 16e904f15b..53f6186213 100644
--- a/tests/rest/media/v1/test_url_preview.py
+++ b/tests/rest/media/v1/test_url_preview.py
@@ -12,9 +12,11 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import base64
 import json
 import os
 import re
+from urllib.parse import urlencode
 
 from twisted.internet._resolver import HostResolution
 from twisted.internet.address import IPv4Address, IPv6Address
@@ -23,6 +25,7 @@ from twisted.test.proto_helpers import AccumulatingProtocol
 
 from synapse.config.oembed import OEmbedEndpointConfig
 from synapse.rest.media.v1.preview_url_resource import IMAGE_CACHE_EXPIRY_MS
+from synapse.types import JsonDict
 from synapse.util.stringutils import parse_and_validate_mxc_uri
 
 from tests import unittest
@@ -142,6 +145,14 @@ class URLPreviewTests(unittest.HomeserverTestCase):
     def create_test_resource(self):
         return self.hs.get_media_repository_resource()
 
+    def _assert_small_png(self, json_body: JsonDict) -> None:
+        """Assert properties from the SMALL_PNG test image."""
+        self.assertTrue(json_body["og:image"].startswith("mxc://"))
+        self.assertEqual(json_body["og:image:height"], 1)
+        self.assertEqual(json_body["og:image:width"], 1)
+        self.assertEqual(json_body["og:image:type"], "image/png")
+        self.assertEqual(json_body["matrix:image:size"], 67)
+
     def test_cache_returns_correct_type(self):
         self.lookups["matrix.org"] = [(IPv4Address, "10.1.2.3")]
 
@@ -569,6 +580,66 @@ class URLPreviewTests(unittest.HomeserverTestCase):
             server.data,
         )
 
+    def test_data_url(self):
+        """
+        Requesting to preview a data URL is not supported.
+        """
+        self.lookups["matrix.org"] = [(IPv4Address, "10.1.2.3")]
+
+        data = base64.b64encode(SMALL_PNG).decode()
+
+        query_params = urlencode(
+            {
+                "url": f'<html><head><img src="data:image/png;base64,{data}" /></head></html>'
+            }
+        )
+
+        channel = self.make_request(
+            "GET",
+            f"preview_url?{query_params}",
+            shorthand=False,
+        )
+        self.pump()
+
+        self.assertEqual(channel.code, 500)
+
+    def test_inline_data_url(self):
+        """
+        An inline image (as a data URL) should be parsed properly.
+        """
+        self.lookups["matrix.org"] = [(IPv4Address, "10.1.2.3")]
+
+        data = base64.b64encode(SMALL_PNG)
+
+        end_content = (
+            b"<html><head>" b'<img src="data:image/png;base64,%s" />' b"</head></html>"
+        ) % (data,)
+
+        channel = self.make_request(
+            "GET",
+            "preview_url?url=http://matrix.org",
+            shorthand=False,
+            await_result=False,
+        )
+        self.pump()
+
+        client = self.reactor.tcpClients[0][2].buildProtocol(None)
+        server = AccumulatingProtocol()
+        server.makeConnection(FakeTransport(client, self.reactor))
+        client.makeConnection(FakeTransport(server, self.reactor))
+        client.dataReceived(
+            (
+                b"HTTP/1.0 200 OK\r\nContent-Length: %d\r\n"
+                b'Content-Type: text/html; charset="utf8"\r\n\r\n'
+            )
+            % (len(end_content),)
+            + end_content
+        )
+
+        self.pump()
+        self.assertEqual(channel.code, 200)
+        self._assert_small_png(channel.json_body)
+
     def test_oembed_photo(self):
         """Test an oEmbed endpoint which returns a 'photo' type which redirects the preview to a new URL."""
         self.lookups["publish.twitter.com"] = [(IPv4Address, "10.1.2.3")]
@@ -626,10 +697,7 @@ class URLPreviewTests(unittest.HomeserverTestCase):
         self.assertEqual(channel.code, 200)
         body = channel.json_body
         self.assertEqual(body["og:url"], "http://twitter.com/matrixdotorg/status/12345")
-        self.assertTrue(body["og:image"].startswith("mxc://"))
-        self.assertEqual(body["og:image:height"], 1)
-        self.assertEqual(body["og:image:width"], 1)
-        self.assertEqual(body["og:image:type"], "image/png")
+        self._assert_small_png(body)
 
     def test_oembed_rich(self):
         """Test an oEmbed endpoint which returns HTML content via the 'rich' type."""
@@ -820,10 +888,7 @@ class URLPreviewTests(unittest.HomeserverTestCase):
         self.assertEqual(
             body["og:url"], "http://www.twitter.com/matrixdotorg/status/12345"
         )
-        self.assertTrue(body["og:image"].startswith("mxc://"))
-        self.assertEqual(body["og:image:height"], 1)
-        self.assertEqual(body["og:image:width"], 1)
-        self.assertEqual(body["og:image:type"], "image/png")
+        self._assert_small_png(body)
 
     def _download_image(self):
         """Downloads an image into the URL cache.
diff --git a/tests/server.py b/tests/server.py
index a0cd14ea45..82990c2eb9 100644
--- a/tests/server.py
+++ b/tests/server.py
@@ -313,7 +313,7 @@ def make_request(
     req = request(channel, site)
     req.content = BytesIO(content)
     # Twisted expects to be at the end of the content when parsing the request.
-    req.content.seek(SEEK_END)
+    req.content.seek(0, SEEK_END)
 
     if access_token:
         req.requestHeaders.addRawHeader(
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-22 15:58:22 +0000