The_Great_Archive/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2026-09-14 11:06:13 +05:00
Code Issues Actions Packages Projects Releases Wiki Activity

Keep replay aliases off control-plane hosts

Browse Source
This commit is contained in:
Nick Sweeting 2026-08-01 22:54:40 -07:00
parent f0abdd2bc3
commit 9efb80ac39
2 changed files with 37 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
archivebox/core/middleware.py
View File

@ -19,6 +19,7 @@ from archivebox.config.common import get_config
from archivebox.config.version import get_COMMIT_HASH
from archivebox.core.routes_util import (
build_admin_url,
build_original_url,
build_snapshot_url,
build_web_url,
get_admin_host,
@ -214,6 +215,10 @@ def HostRoutingMiddleware(get_response):
snapshot_path_re = re.compile(
r"^/(?P<username>[^/]+)/(?P<date>\d{4}(?:\d{2})?(?:\d{2})?)/(?P<domain>[^/]+)/(?P<snapshot_id>[0-9a-fA-F-]{8,36})(?:/(?P<path>.*))?$",
)
snapshot_replay_path_re = re.compile(
r"^/snapshot/(?P<snapshot_id>[0-9a-fA-F-]{8,36})(?:/(?P<path>.*))?$",
)
original_replay_path_re = re.compile(r"^/original/(?P<domain>[^/]+)(?:/(?P<path>.*))?$")
def middleware(request):
if request.path in {"/health", "/health/"}:
@ -263,6 +268,29 @@ def HostRoutingMiddleware(get_response):
if config.USES_SUBDOMAIN_ROUTING and not config.BASE_URL:
return get_response(request)
if config.USES_SUBDOMAIN_ROUTING:
snapshot_replay_match = snapshot_replay_path_re.match(request.path)
if snapshot_replay_match:
target = build_snapshot_url(
snapshot_replay_match.group("snapshot_id"),
(snapshot_replay_match.group("path") or "").strip("/"),
request=request,
)
if request.META.get("QUERY_STRING"):
target = f"{target}?{request.META['QUERY_STRING']}"
return redirect(target)
original_replay_match = original_replay_path_re.match(request.path)
if original_replay_match:
target = build_original_url(
original_replay_match.group("domain"),
(original_replay_match.group("path") or "").strip("/"),
request=request,
)
if request.META.get("QUERY_STRING"):
target = f"{target}?{request.META['QUERY_STRING']}"
return redirect(target)
if not config.USES_SUBDOMAIN_ROUTING:
if host_matches(request_host, listen_host):
return get_response(request)

9
archivebox/tests/test_urls.py
View File

@ -443,6 +443,15 @@ class TestUrlRouting:
assert resp.status_code in (301, 302)
assert resp["Location"] == f"http://{snapshot_host}"
for control_host in (admin_host, web_host):
resp = client.get(f"/snapshot/{snapshot.id}/index.jsonl?download=1", HTTP_HOST=control_host)
assert resp.status_code in (301, 302)
assert resp["Location"] == f"http://{snapshot_host}/index.jsonl?download=1"
resp = client.get(f"/original/{snapshot.domain}/index.html", HTTP_HOST=control_host)
assert resp.status_code in (301, 302)
assert resp["Location"] == f"http://{original_host}/index.html"
resp = client.get("/static/jquery.min.js", HTTP_HOST=snapshot_host)
assert resp.status_code == 200
assert "javascript" in (resp.headers.get("Content-Type") or "")