Files
Evection/apps/feeds/test_opml.py
T
ksmithandClaude Sonnet 5.5 41f99e9405 Allow only http(s) URLs from feeds and OPML, at storage and render (R2-S1, #33)
A feed's channel link, an entry's link and an OPML htmlUrl were stored
as given and rendered in href attributes (bulk writes skip model
validation), so a malicious feed could supply a javascript: or data:
link. CSP is a second layer, not the control.

apps.core.urlutils accepts only absolute http(s) URLs. Fetching stores
channel/entry links through it — relative links now resolve against the
feed's own URL instead of being stored as dead relative paths — OPML
import skips non-http(s) feed URLs and blanks unsafe htmlUrl, and the
add-feed form rejects non-http(s) (URLField also allows ftp). A
safe_http_url template filter guards the two places a feed-supplied URL
becomes a link (the article's "Original" link and the dashboard), so
values already in the database from before this fix are neutralized too.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-01 11:57:23 -07:00

213 lines
7.5 KiB
Python

import zipfile
from io import BytesIO
import pytest
from django.contrib.auth import get_user_model
from apps.articles.models import SubscriptionTag, Tag
from .models import Feed, Subscription
from .opml import export_opml, extract_opml_bytes, import_opml, parse_opml
User = get_user_model()
SIMPLE_OPML = b"""<?xml version="1.0" encoding="UTF-8"?>
<opml version="2.0">
<head><title>Subscriptions</title></head>
<body>
<outline text="Tech">
<outline type="rss" text="Example Blog" title="Example Blog"
xmlUrl="https://example.com/feed.xml" htmlUrl="https://example.com"/>
</outline>
<outline type="rss" text="Untagged Feed" title="Untagged Feed"
xmlUrl="https://untagged.example.com/feed.xml"/>
</body>
</opml>
"""
NESTED_OPML = b"""<?xml version="1.0" encoding="UTF-8"?>
<opml version="2.0">
<body>
<outline text="Parent">
<outline text="Child">
<outline type="rss" text="Deeply nested feed"
xmlUrl="https://nested.example.com/feed.xml"/>
</outline>
</outline>
</body>
</opml>
"""
CATEGORY_ATTR_OPML = b"""<?xml version="1.0" encoding="UTF-8"?>
<opml version="2.0">
<body>
<outline type="rss" text="Multi-tag feed" category="Tech,Backend"
xmlUrl="https://multi.example.com/feed.xml"/>
</body>
</opml>
"""
MALFORMED_XML = b"<opml><body><outline"
@pytest.fixture
def alice(db):
return User.objects.create_user("alice", password="pw-alice-123")
def _zip_of(filename: str, content: bytes) -> bytes:
buf = BytesIO()
with zipfile.ZipFile(buf, "w") as archive:
archive.writestr(filename, content)
return buf.getvalue()
class TestExtractOPMLBytes:
def test_bare_opml_is_returned_unchanged(self):
assert extract_opml_bytes(SIMPLE_OPML) == SIMPLE_OPML
def test_extracts_opml_from_a_freshrss_style_export_zip(self):
# FreshRSS's own "Export" produces exactly this: an .opml.xml plus
# a starred-articles .json we don't need for feed import.
zipped = _zip_of("feeds_2026-10-30.opml.xml", SIMPLE_OPML)
with zipfile.ZipFile(BytesIO(zipped), "a") as archive:
archive.writestr("starred_2026-10-30.json", b'{"items": []}')
assert extract_opml_bytes(zipped) == SIMPLE_OPML
def test_zip_with_no_opml_member_raises(self):
zipped = _zip_of("starred_2026-10-30.json", b'{"items": []}')
with pytest.raises(ValueError):
extract_opml_bytes(zipped)
class TestParseOPML:
def test_parses_tagged_and_untagged_entries(self):
entries = parse_opml(BytesIO(SIMPLE_OPML))
assert len(entries) == 2
tagged = next(e for e in entries if e["feed_url"] == "https://example.com/feed.xml")
assert tagged["tag_names"] == ["Tech"]
assert tagged["title"] == "Example Blog"
untagged = next(
e for e in entries if e["feed_url"] == "https://untagged.example.com/feed.xml"
)
assert untagged["tag_names"] == []
def test_nesting_produces_a_tag_per_level(self):
entries = parse_opml(BytesIO(NESTED_OPML))
assert len(entries) == 1
# Unlike the old single-folder scheme, each enclosing level becomes
# its own tag rather than collapsing to the nearest one.
assert entries[0]["tag_names"] == ["Parent", "Child"]
def test_category_attribute_produces_multiple_tags(self):
entries = parse_opml(BytesIO(CATEGORY_ATTR_OPML))
assert len(entries) == 1
assert entries[0]["tag_names"] == ["Tech", "Backend"]
def test_malformed_xml_raises(self):
with pytest.raises(Exception):
parse_opml(BytesIO(MALFORMED_XML))
@pytest.mark.django_db
class TestImportOPML:
def test_creates_feeds_tags_and_subscriptions(self, alice):
result = import_opml(alice, BytesIO(SIMPLE_OPML))
assert result == {"total": 2, "created": 2, "already_subscribed": 0}
assert Tag.objects.filter(user=alice, name="Tech").exists()
tagged_sub = Subscription.objects.get(
user=alice, feed__feed_url="https://example.com/feed.xml"
)
assert SubscriptionTag.objects.filter(
user=alice, subscription=tagged_sub, tag__name="Tech"
).exists()
untagged_sub = Subscription.objects.get(
user=alice, feed__feed_url="https://untagged.example.com/feed.xml"
)
assert not SubscriptionTag.objects.filter(user=alice, subscription=untagged_sub).exists()
def test_reimporting_the_same_file_does_not_duplicate(self, alice):
import_opml(alice, BytesIO(SIMPLE_OPML))
result = import_opml(alice, BytesIO(SIMPLE_OPML))
assert result == {"total": 2, "created": 0, "already_subscribed": 2}
assert Subscription.objects.filter(user=alice).count() == 2
assert Tag.objects.filter(user=alice, name="Tech").count() == 1
def test_reuses_existing_feed_across_users(self, alice):
bob = User.objects.create_user("bob", password="pw-bob-123")
import_opml(alice, BytesIO(SIMPLE_OPML))
import_opml(bob, BytesIO(SIMPLE_OPML))
assert Feed.objects.filter(feed_url="https://example.com/feed.xml").count() == 1
assert (
Subscription.objects.filter(feed__feed_url="https://example.com/feed.xml").count() == 2
)
@pytest.mark.django_db
class TestExportOPML:
def test_export_round_trips_through_import(self, alice):
import_opml(alice, BytesIO(SIMPLE_OPML))
exported = export_opml(alice)
assert b'category="Tech"' in exported
bob = User.objects.create_user("bob", password="pw-bob-123")
result = import_opml(bob, BytesIO(exported))
assert result["created"] == 2
tagged_sub = Subscription.objects.get(
user=bob, feed__feed_url="https://example.com/feed.xml"
)
assert SubscriptionTag.objects.filter(
user=bob, subscription=tagged_sub, tag__name="Tech"
).exists()
untagged_sub = Subscription.objects.get(
user=bob, feed__feed_url="https://untagged.example.com/feed.xml"
)
assert not SubscriptionTag.objects.filter(user=bob, subscription=untagged_sub).exists()
def test_export_represents_multiple_tags_as_comma_separated_category(self, alice):
feed = Feed.objects.create(feed_url="https://multi.example.com/feed.xml")
subscription = Subscription.objects.create(user=alice, feed=feed)
for name in ["Tech", "Backend"]:
tag, _ = Tag.objects.get_or_create(user=alice, name=name)
SubscriptionTag.objects.create(user=alice, subscription=subscription, tag=tag)
exported = export_opml(alice)
# Alphabetical (export_opml orders by tag name), not insertion order.
assert b'category="Backend,Tech"' in exported
def test_export_with_no_subscriptions(self, alice):
exported = export_opml(alice)
assert b"<opml" in exported
assert b"<outline" not in exported
class TestUnsafeOPMLUrls:
def test_non_http_feed_urls_are_skipped_and_unsafe_html_urls_blanked(self):
from io import BytesIO
from apps.feeds.opml import parse_opml
xml = b"""<?xml version="1.0"?><opml version="2.0"><head/><body>
<outline text="Bad" xmlUrl="javascript:alert(1)"/>
<outline text="FTP" xmlUrl="ftp://example.com/feed.xml"/>
<outline text="Good" xmlUrl="https://example.com/feed.xml" htmlUrl="javascript:alert(2)"/>
</body></opml>"""
entries = parse_opml(BytesIO(xml))
assert [e["feed_url"] for e in entries] == ["https://example.com/feed.xml"]
assert entries[0]["site_url"] == ""