Files
ksmithandClaude Sonnet 5.5 698940763f Bound OPML uploads in size, depth, count and broker payload (R2-S5, #37)
An OPML upload was read fully into memory, parsed into a whole tree,
walked recursively, and (for large imports) embedded in a Celery
message. Nothing bounded raw size, ZIP entries, nesting depth, outline
count or feed count.

Uploads are capped at 5 MB (checked on the reported size before reading)
and ZIPs at 50 entries, with the decompressed member bounded by what it
actually yields rather than only its header. Parsing now streams with
iterparse and aborts the moment a limit is crossed — 24 levels deep,
20,000 outlines, 5,000 feeds — so a hostile file never builds a huge or
deep tree; a 50,000-level nesting bomb is rejected cleanly. Limit errors
are shown to the user. Large imports go to the worker by cache reference
(deleted once read, expires after an hour) instead of as the whole
document inside the broker message.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-01 11:58:49 -07:00

220 lines
8.9 KiB
Python

"""OPML import/export.
Import parses with defusedxml since an uploaded .opml file is untrusted
user input (XXE risk with a plain XML parser). Export uses the stdlib's
plain ElementTree since the output is entirely app-generated — no
untrusted input, no XXE concern there.
"""
from __future__ import annotations
import xml.etree.ElementTree as ET
import zipfile
from datetime import datetime
from datetime import timezone as dt_timezone
from io import BytesIO
import defusedxml.ElementTree as DefusedET
from apps.articles.models import SubscriptionTag, Tag
from apps.core.urlutils import http_url_or_empty
from .models import Feed, Subscription
# FreshRSS's own "Export" produces a .zip (an .opml.xml plus a starred-
# articles .json we don't import) rather than a bare .opml file — accept
# that directly instead of making people unzip it themselves first.
MAX_ZIPPED_OPML_SIZE = 20 * 1024 * 1024
# An upload is untrusted: bound everything an attacker controls (R2-S5).
# A real export of thousands of feeds is well under a megabyte.
MAX_UPLOAD_BYTES = 5 * 1024 * 1024
MAX_ZIP_ENTRIES = 50
MAX_OPML_DEPTH = 24
MAX_OPML_OUTLINES = 20_000
MAX_OPML_FEEDS = 5_000
def extract_opml_bytes(uploaded_bytes: bytes) -> bytes:
"""If the upload is a zip (e.g. FreshRSS's export), returns the bytes
of the .opml/.opml.xml member inside it; otherwise returns the input
unchanged, assuming it's already a bare OPML file."""
if not zipfile.is_zipfile(BytesIO(uploaded_bytes)):
return uploaded_bytes
with zipfile.ZipFile(BytesIO(uploaded_bytes)) as archive:
if len(archive.infolist()) > MAX_ZIP_ENTRIES:
raise ValueError("The uploaded zip contains too many files.")
candidates = [
info
for info in archive.infolist()
if not info.is_dir() and info.filename.lower().endswith((".opml", ".opml.xml"))
]
if not candidates:
raise ValueError("No .opml file found inside the uploaded zip.")
member = min(candidates, key=lambda info: info.file_size)
if member.file_size > MAX_ZIPPED_OPML_SIZE:
raise ValueError("The .opml file inside the zip is implausibly large.")
# file_size is just a header field an attacker controls, so bound
# what actually decompresses too, not only what it claims.
with archive.open(member) as handle:
data = handle.read(MAX_ZIPPED_OPML_SIZE + 1)
if len(data) > MAX_ZIPPED_OPML_SIZE:
raise ValueError("The .opml file inside the zip is implausibly large.")
return data
def parse_opml(file_obj) -> list[dict]:
"""Returns a flat list of {feed_url, title, site_url, tag_names}.
A feed's tags come from two places, combined: every enclosing
category/folder outline's name (so a feed nested two levels deep picks
up both as separate tags — tags aren't limited to one level the way
the old folder scheme was), and a comma-separated "category" attribute
on the feed's own outline, the convention export_opml below writes and
many other readers use for a flat multi-category representation.
"""
entries: list[dict] = []
# Streamed with iterparse instead of building the whole tree first, so
# a hostile file (absurd nesting, millions of outlines) is rejected as
# soon as it crosses a limit rather than after it has used the memory.
in_body = False
depth = 0
outline_count = 0
open_feed_outlines = 0 # >0 while inside a feed outline: its children are ignored
pushed_name: list[bool] = [] # per open category outline: did it add a tag name?
ancestor_names: list[str] = []
for event, elem in DefusedET.iterparse(file_obj, events=("start", "end")):
if event == "start":
depth += 1
if depth > MAX_OPML_DEPTH:
raise ValueError("The OPML file is nested too deeply.")
if elem.tag == "body":
in_body = True
elif elem.tag == "outline" and in_body:
outline_count += 1
if outline_count > MAX_OPML_OUTLINES:
raise ValueError("The OPML file contains too many entries.")
if open_feed_outlines:
open_feed_outlines += 1
continue
raw_xml_url = elem.get("xmlUrl")
if raw_xml_url:
open_feed_outlines = 1
xml_url = http_url_or_empty(raw_xml_url)
if not xml_url:
continue # a feed URL that isn't http(s) can never be fetched
tag_names = list(ancestor_names)
for name in (elem.get("category") or "").split(","):
name = name.strip()
if name and name not in tag_names:
tag_names.append(name)
entries.append(
{
"feed_url": xml_url,
"title": elem.get("title") or elem.get("text") or "",
"site_url": http_url_or_empty(elem.get("htmlUrl")),
"tag_names": tag_names,
}
)
if len(entries) > MAX_OPML_FEEDS:
raise ValueError("The OPML file contains too many feeds.")
else:
nested_name = elem.get("title") or elem.get("text")
pushed_name.append(bool(nested_name))
if nested_name:
ancestor_names.append(nested_name)
else:
depth -= 1
if elem.tag == "body":
in_body = False
elif elem.tag == "outline" and in_body:
if open_feed_outlines:
open_feed_outlines -= 1
elif pushed_name:
if pushed_name.pop():
ancestor_names.pop()
elem.clear()
return entries
def import_opml(user, file_obj) -> dict:
"""Creates/reuses Feeds and Subscriptions for every entry, tagging each
subscription per its entry's tag_names. Does not fetch any feed
content — newly created Feeds have no last_fetched_at, so the next
poll_due_feeds beat cycle picks them up naturally."""
entries = parse_opml(file_obj)
tag_cache: dict[str, Tag] = {}
created_count = 0
already_subscribed_count = 0
for entry in entries:
feed, _ = Feed.objects.get_or_create(
feed_url=entry["feed_url"][:1000],
defaults={
"title": entry["title"][:500],
"site_url": entry["site_url"][:1000],
},
)
subscription, created = Subscription.objects.get_or_create(
user=user,
feed=feed,
defaults={"custom_title": ""},
)
if created:
created_count += 1
else:
already_subscribed_count += 1
for tag_name in entry["tag_names"]:
tag = tag_cache.get(tag_name)
if tag is None:
tag, _ = Tag.objects.get_or_create(user=user, name=tag_name[:100])
tag_cache[tag_name] = tag
SubscriptionTag.objects.get_or_create(user=user, subscription=subscription, tag=tag)
return {
"total": len(entries),
"created": created_count,
"already_subscribed": already_subscribed_count,
}
def export_opml(user) -> bytes:
"""A flat list of outlines (no folder nesting) with each feed's tags
joined into a comma-separated "category" attribute — the standard OPML
way to represent multi-category membership without picking one tag to
nest under, which a feed with several tags wouldn't fit."""
root = ET.Element("opml", version="2.0")
head = ET.SubElement(root, "head")
ET.SubElement(head, "title").text = "Evection subscriptions"
ET.SubElement(head, "dateCreated").text = datetime.now(dt_timezone.utc).strftime(
"%a, %d %b %Y %H:%M:%S %z"
)
body = ET.SubElement(root, "body")
subscriptions = (
Subscription.objects.filter(user=user).select_related("feed").order_by("order", "id")
)
tag_names_by_subscription: dict[int, list[str]] = {}
for row in (
SubscriptionTag.objects.filter(user=user).select_related("tag").order_by("tag__name")
):
tag_names_by_subscription.setdefault(row.subscription_id, []).append(row.tag.name)
for subscription in subscriptions:
attrs = {
"type": "rss",
"text": subscription.display_title,
"title": subscription.display_title,
"xmlUrl": subscription.feed.feed_url,
"htmlUrl": subscription.feed.site_url or "",
}
tag_names = tag_names_by_subscription.get(subscription.pk)
if tag_names:
attrs["category"] = ",".join(tag_names)
ET.SubElement(body, "outline", attrs)
return ET.tostring(root, encoding="utf-8", xml_declaration=True)