fix(website): escape </script> in embedded filter URLs JSON

`| safe` bypasses Jinja autoescape. If a category name ever contained
"</script>", the literal substring would close the script block early,
leaking JSON content into the DOM and creating an XSS vector. Replace
"</" with "<\\/" (still valid JSON) and pass ensure_ascii=False so
non-ASCII names render readably. Also add a group_path() helper to
parallel category_path()/subcategory_path() and reuse category_urls
when seeding filter_urls.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Vinta Chen
2026-05-03 00:40:52 +08:00
parent e0e7fc9168
commit 704332271b
2 changed files with 36 additions and 5 deletions
+7 -5
View File
@@ -92,6 +92,10 @@ def category_public_url(category: ParsedSection) -> str:
return f"{SITE_URL}categories/{category['slug']}/"
def group_path(group_slug: str) -> str:
return f"/categories/{group_slug}/"
def group_public_url(group_slug: str) -> str:
return f"{SITE_URL}categories/{group_slug}/"
@@ -315,11 +319,9 @@ def build(repo_root: Path) -> None:
entries = sort_entries(entries)
category_urls = {cat["name"]: category_path(cat) for cat in categories}
filter_urls: dict[str, str] = {}
for cat in categories:
filter_urls[cat["name"]] = category_path(cat)
filter_urls: dict[str, str] = dict(category_urls)
for group in parsed_groups:
filter_urls[group["name"]] = f"/categories/{group['slug']}/"
filter_urls[group["name"]] = group_path(group["slug"])
for entry in entries:
for sub in entry.get("subcategories", []):
filter_urls[sub["value"]] = sub["url"]
@@ -348,7 +350,7 @@ def build(repo_root: Path) -> None:
build_date=build_date.strftime("%B %d, %Y"),
sponsors=sponsors,
category_urls=category_urls,
filter_urls_json=json.dumps(filter_urls, sort_keys=True),
filter_urls_json=json.dumps(filter_urls, sort_keys=True, ensure_ascii=False).replace("</", "<\\/"),
),
encoding="utf-8",
)
+29
View File
@@ -655,6 +655,35 @@ class TestBuild:
assert data["AI & ML"] == "/categories/ai-ml/"
assert data["Machine Learning > Classical"] == "/categories/machine-learning/classical/"
def test_filter_urls_json_escapes_closing_script_tag(self, tmp_path):
readme = textwrap.dedent("""\
# T
---
## Sneaky </script><script>x=1</script>
- [a](https://example.com) - A.
# Contributing
Done.
""")
self._copy_real_templates(tmp_path)
(tmp_path / "README.md").write_text(readme, encoding="utf-8")
build(tmp_path)
site = tmp_path / "website" / "output"
index_html = (site / "index.html").read_text(encoding="utf-8")
marker = '<script type="application/json" id="filter-urls">'
start = index_html.index(marker) + len(marker)
end = index_html.index("</script>", start)
block = index_html[start:end]
assert "</script>" not in block
data = json.loads(block)
assert any("Sneaky" in key for key in data)
def test_build_creates_group_pages(self, tmp_path):
readme = textwrap.dedent("""\
# T