Skip to content
This repository was archived by the owner on Apr 26, 2024. It is now read-only.

Try to recover from unknown media encodings. #9164

Merged
merged 11 commits into from
Jan 26, 2021
Merged

Try to recover from unknown media encodings. #9164

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
a92f7b5
Try to recover from unknown character encodings.
clokep Jan 19, 2021
3dec5b8
Share some common code.
clokep Jan 19, 2021
4d255b6
Newsfragment
clokep Jan 19, 2021
89a43fe
More error handling.
clokep Jan 21, 2021
fd51df9
Add tests.
clokep Jan 21, 2021
2b02f44
Typo
clokep Jan 21, 2021
465c8e2
Lint.
clokep Jan 21, 2021
58abad9
Params -> Args
clokep Jan 21, 2021
911eec6
Add a missing type hint.
clokep Jan 25, 2021
f629933
Try to simplify code.
clokep Jan 25, 2021
5f61cda
Merge remote-tracking branch 'origin/develop' into clokep/media-encoding
clokep Jan 25, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 3 additions & 7 deletions synapse/rest/media/v1/preview_url_resource.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -717,32 +717,28 @@ def decode_and_calc_og(
parser = etree.HTMLParser(recover=True, encoding=request_encoding)
except LookupError:
# blindly consider the encoding as utf-8.
try:
parser = etree.HTMLParser(recover=True, encoding="utf-8")
except Exception as e:
logger.warning("Unable to create fallback HTML parser: %s" % (e,))
return {}
parser = etree.HTMLParser(recover=True, encoding="utf-8")
except Exception as e:
logger.warning("Unable to create HTML parser: %s" % (e,))
return {}

# Attempt to parse the body. If this fails, log and return no metadata.
try:
tree = etree.fromstring(body, parser)
return _calc_og(tree, media_uri)
except UnicodeDecodeError:
# blindly try decoding the body as utf-8, which seems to fix
# the charset mismatches on https://google.com
try:
tree = etree.fromstring(body.decode("utf-8", "ignore"), parser)
return _calc_og(tree, media_uri)
except Exception as e:
logger.warning("Failed to parse HTML body as UTF-8: %s" % (e,))
return {}
except Exception as e:
logger.warning("Failed to parse HTML body: %s" % (e,))
return {}

return _calc_og(tree, media_uri)


def _calc_og(tree, media_uri: str) -> Dict[str, Optional[str]]:
# suck our tree into lxml and define our OG response.
Expand Down
27 changes: 27 additions & 0 deletions tests/test_preview.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -261,3 +261,30 @@ def test_empty(self):
html = ""
og = decode_and_calc_og(html, "http://example.com/test.html")
self.assertEqual(og, {})

def test_invalid_encoding(self):
"""An invalid character encoding should be ignored and treated as UTF-8, if possible."""
html = """
<html>
<head><title>Foo</title></head>
<body>
Some text.
</body>
</html>
"""
og = decode_and_calc_og(html, "http://example.com/test.html", "invalid-encoding")
self.assertEqual(og, {"og:title": "Foo", "og:description": "Some text."})

def test_invalid_encoding2(self):
"""A body which doesn't match the sent character encoding."""
# Note that this contains an invalid UTF-8 sequence in the title.
html = b"""
<html>
<head><title>\xff\xff Foo</title></head>
<body>
Some text.
</body>
</html>
"""
og = decode_and_calc_og(html, "http://example.com/test.html")
self.assertEqual(og, {"og:title": "ÿÿ Foo", "og:description": "Some text."})