Files
Evection/apps/feeds/opml.py
T
ksmithandClaude Sonnet 5 f818b11041 Accept a FreshRSS "Export" .zip directly for OPML import
FreshRSS's Export feature produces a .zip (an .opml.xml plus a starred-
articles .json this app doesn't import) rather than a bare .opml file —
uploading that zip as-is failed with "That file doesn't look like a valid
OPML export" since defusedxml can't parse a zip as XML.

apps.feeds.opml.extract_opml_bytes detects a zip upload and pulls the
.opml/.opml.xml member out of it before parsing; a bare .opml upload is
returned unchanged, so existing behavior is untouched. Verified against
the reporter's actual FreshRSS export (18 feeds across 6 categories) both
directly through extract_opml_bytes/parse_opml and as a real upload
through the running app.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-10-30 21:45:16 -06:00

173 lines
6.5 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 .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
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:
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.")
return archive.read(member)
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.
"""
tree = DefusedET.parse(file_obj)
root = tree.getroot()
body = root.find("body")
entries: list[dict] = []
if body is None:
return entries
def walk(container, ancestor_names):
for outline in container.findall("outline"):
xml_url = outline.get("xmlUrl")
if xml_url:
tag_names = list(ancestor_names)
category_attr = outline.get("category")
if category_attr:
for name in category_attr.split(","):
name = name.strip()
if name and name not in tag_names:
tag_names.append(name)
entries.append(
{
"feed_url": xml_url,
"title": outline.get("title") or outline.get("text") or "",
"site_url": outline.get("htmlUrl") or "",
"tag_names": tag_names,
}
)
else:
nested_name = outline.get("title") or outline.get("text")
next_ancestors = ancestor_names + [nested_name] if nested_name else ancestor_names
walk(outline, next_ancestors)
walk(body, [])
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)