lukasandClaude Opus 5.5
02011f23f6
<title> is not an allowed tag, so nh3 unwrapped it and kept its text — "mailer", "Email template" or the subject rendered before the content. It is now dropped with its content, both in the sanitized HTML and in the text derived from HTML-only mail (snippet and search). SANITIZER_VERSION goes to 4 so stored bodies re-sanitize from the raw blob on next render. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
102 lines
3.5 KiB
Python
102 lines
3.5 KiB
Python
"""HTML sanitization (SPEC §6.4, §10.4, I-8)."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import uuid
|
|
|
|
from app.indexer.sanitize import sanitize_html
|
|
|
|
MID = uuid.uuid4()
|
|
|
|
|
|
def test_scripts_and_handlers_are_removed() -> None:
|
|
html = '<div onclick="evil()">hi<script>alert(1)</script></div>'
|
|
result = sanitize_html(html, MID)
|
|
assert "<script" not in result.html.lower()
|
|
assert "onclick" not in result.html.lower()
|
|
assert "hi" in result.html
|
|
|
|
|
|
def test_document_title_is_not_rendered() -> None:
|
|
# Newsletters put "mailer" or the subject in <title>; unwrapping the tag would
|
|
# show it as a stray line above the body. Hidden preheaders keep their style.
|
|
html = (
|
|
"<html><head><title>mailer</title></head><body>"
|
|
'<span style="display:none">preview</span><p>Hi</p></body></html>'
|
|
)
|
|
result = sanitize_html(html, MID)
|
|
assert "mailer" not in result.html
|
|
assert '<span style="display:none">preview</span>' in result.html
|
|
assert "<p>Hi</p>" in result.html
|
|
|
|
|
|
def test_remote_image_is_blocked_and_flagged() -> None:
|
|
html = '<p>see <img src="http://tracker.example/pixel.gif"> this</p>'
|
|
result = sanitize_html(html, MID)
|
|
assert result.has_remote_images
|
|
assert "tracker.example" not in result.html
|
|
|
|
|
|
def test_cid_image_is_rewritten_to_endpoint() -> None:
|
|
html = '<img src="cid:logo123">'
|
|
result = sanitize_html(html, MID)
|
|
assert f"/api/messages/{MID}/cid/logo123" in result.html
|
|
assert not result.has_remote_images
|
|
|
|
|
|
def test_data_image_is_preserved() -> None:
|
|
html = '<img src="data:image/png;base64,iVBORw0KGgo=">'
|
|
result = sanitize_html(html, MID)
|
|
assert "data:image/png" in result.html
|
|
assert not result.has_remote_images
|
|
|
|
|
|
def test_safe_links_survive_with_rel() -> None:
|
|
html = '<a href="https://example.com">link</a>'
|
|
result = sanitize_html(html, MID)
|
|
assert "https://example.com" in result.html
|
|
assert "javascript:" not in result.html.lower()
|
|
|
|
|
|
def test_javascript_uri_is_stripped() -> None:
|
|
html = '<a href="javascript:alert(1)">x</a>'
|
|
result = sanitize_html(html, MID)
|
|
assert "javascript:" not in result.html.lower()
|
|
|
|
|
|
def test_sanitizer_version_reported() -> None:
|
|
result = sanitize_html("<p>hi</p>", MID)
|
|
assert result.sanitizer_v >= 1
|
|
|
|
|
|
def test_style_block_and_classes_are_kept() -> None:
|
|
# Mailers (e.g. Fastmail invites) put layout in a <style> block + classes.
|
|
html = '<style>.banner{color:#333}</style><div class="banner">hi</div>'
|
|
result = sanitize_html(html, MID)
|
|
assert "<style" in result.html.lower()
|
|
assert ".banner" in result.html
|
|
assert 'class="banner"' in result.html
|
|
|
|
|
|
def test_style_block_script_css_is_dropped() -> None:
|
|
html = '<style>.x{width:expression(alert(1))}</style><p class="x">hi</p>'
|
|
result = sanitize_html(html, MID)
|
|
assert "expression(" not in result.html.lower()
|
|
# the block is nuked but the element/class survive
|
|
assert 'class="x"' in result.html
|
|
|
|
|
|
def test_style_block_import_and_url_stripped_when_remote_blocked() -> None:
|
|
html = "<style>@import url('http://evil/x.css'); .b{background:url(http://evil/p.gif)}</style>"
|
|
result = sanitize_html(html, MID)
|
|
assert "@import" not in result.html.lower()
|
|
assert "evil" not in result.html.lower()
|
|
assert result.has_remote_images
|
|
|
|
|
|
def test_script_content_still_stripped_with_style_allowed() -> None:
|
|
html = "<script>alert(1)</script><style>.a{color:red}</style>"
|
|
result = sanitize_html(html, MID)
|
|
assert "alert(1)" not in result.html
|
|
assert ".a{color:red}" in result.html
|