Compare commits

..
Author SHA1 Message Date
ksmithandClaude Sonnet 5 17e7f63e61 Update link/author/summary/published_at on refetch, not just title (T6)
A changed guid's article only ever had its title/content_html/
content_hash refreshed on a subsequent poll — legitimate feed edits to
an entry's link, author, summary, or publication date were silently
ignored forever, which can affect navigation, ordering, and rules that
key off those fields.

content_hash is now computed over every mutable field feedparser gives
us for an entry (title, link, author, summary, content_html,
published_at), and all of them are written together when the hash
changes. Per-user read/star state is untouched either way, same as
before.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-10-31 14:44:15 -06:00
ksmithandClaude Sonnet 5 0bfe9c8076 Enqueue the initial feed fetch instead of running it inline (T3)
add_feed ran fetch_and_ingest synchronously during the POST request, so
a slow or large feed tied up a Gunicorn worker for the duration of the
fetch (up to the 15s HTTP timeout, plus parse/ingest time) and a user
could repeat it with distinct URLs to reduce request capacity.

The subscription is now created synchronously as before, but the first
fetch is dispatched via fetch_feed.delay() — the same task the periodic
poll uses, so the save/error bookkeeping is unchanged. The user sees an
immediate redirect with a message that the fetch is in progress rather
than waiting on it in the request/response cycle.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-10-31 14:44:09 -06:00
4 changed files with 85 additions and 27 deletions

No files matched your search

+33 -8
View File
@@ -97,8 +97,15 @@ def sanitize_html(raw_html: str) -> str:
return cleaned[:MAX_CONTENT_HTML_CHARS]
def _content_hash(title: str, content_html: str) -> str:
return hashlib.sha256(f"{title}\n{content_html}".encode("utf-8")).hexdigest()
def _content_hash(
title: str, link: str, author: str, summary: str, content_html: str, published_at
) -> str:
"""Covers every mutable Article field an entry carries (T6) — a feed
editing its link/author/summary/published date, not just title or
body, must still be detected as a change worth re-saving."""
published_key = published_at.isoformat() if published_at else ""
payload = "\n".join([title, link, author, summary, content_html, published_key])
return hashlib.sha256(payload.encode("utf-8")).hexdigest()
def _entry_guid(entry) -> str:
@@ -213,20 +220,24 @@ def fetch_and_ingest(feed: Feed) -> FetchResult:
for entry in entries:
guid = _entry_guid(entry)
title = (getattr(entry, "title", "") or "")[:1000]
link = (getattr(entry, "link", "") or "")[:2000]
author = (getattr(entry, "author", "") or "")[:500]
summary = (getattr(entry, "summary", "") or "")[:5000]
content_html = sanitize_html(_entry_content_html(entry))
content_hash = _content_hash(title, content_html)
published_at = _entry_published_at(entry)
content_hash = _content_hash(title, link, author, summary, content_html, published_at)
article, created = Article.objects.get_or_create(
feed=feed,
guid=guid,
defaults={
"link": (getattr(entry, "link", "") or "")[:2000],
"link": link,
"title": title,
"author": (getattr(entry, "author", "") or "")[:500],
"author": author,
"content_html": content_html,
"summary": (getattr(entry, "summary", "") or "")[:5000],
"summary": summary,
"content_hash": content_hash,
"published_at": _entry_published_at(entry),
"published_at": published_at,
},
)
@@ -234,9 +245,23 @@ def fetch_and_ingest(feed: Feed) -> FetchResult:
new_ids.append(article.pk)
elif article.content_hash != content_hash:
article.title = title
article.link = link
article.author = author
article.summary = summary
article.content_html = content_html
article.published_at = published_at
article.content_hash = content_hash
article.save(update_fields=["title", "content_html", "content_hash"])
article.save(
update_fields=[
"title",
"link",
"author",
"summary",
"content_html",
"published_at",
"content_hash",
]
)
updated_ids.append(article.pk)
return FetchResult(
+36
View File
@@ -1,3 +1,5 @@
import datetime as dt
import pytest
import requests
import responses
@@ -204,6 +206,40 @@ def test_refetching_updated_content_does_not_reset_user_state():
assert ArticleTag.objects.filter(user=user, article=article, tag=starred_tag).exists()
@pytest.mark.django_db
@responses.activate
def test_refetching_picks_up_link_author_summary_and_published_at_changes():
"""Previously only title/content_html/content_hash were updated on
refetch (T6) — a feed correcting its link, author, summary, or
publication date without touching the title went unnoticed forever."""
feed = Feed.objects.create(feed_url=FEED_URL)
responses.add(responses.GET, FEED_URL, body=GOOD_FEED_XML, status=200)
fetch_and_ingest(feed)
feed.save()
article = Article.objects.get(feed=feed)
assert article.link == "https://example.com/1"
assert article.author == ""
edited_xml = GOOD_FEED_XML.replace(
"<link>https://example.com/1</link>",
"<link>https://example.com/1-moved</link>",
).replace(
"<pubDate>Mon, 01 Jan 2024 00:00:00 GMT</pubDate>",
"<pubDate>Tue, 02 Jan 2024 00:00:00 GMT</pubDate>\n <author>jane@example.com</author>",
)
responses.add(responses.GET, FEED_URL, body=edited_xml, status=200)
result = fetch_and_ingest(feed)
feed.save()
assert result.updated_article_ids == [article.pk]
article.refresh_from_db()
assert article.link == "https://example.com/1-moved"
assert article.author == "jane@example.com"
assert article.published_at == dt.datetime(2024, 1, 2, tzinfo=dt.timezone.utc)
def test_sanitize_html_caps_stored_output_length():
huge = "<p>" + ("a" * (MAX_CONTENT_HTML_CHARS + 10_000)) + "</p>"
cleaned = sanitize_html(huge)
+8 -3
View File
@@ -54,7 +54,11 @@ GOOD_FEED_XML = """<?xml version="1.0"?>
@pytest.mark.django_db
@responses.activate
def test_add_feed_view_subscribes_and_fetches_immediately(client, two_users):
def test_add_feed_view_subscribes_and_enqueues_the_initial_fetch(client, two_users):
"""The subscription is created synchronously; the first fetch is
dispatched via fetch_feed.delay() (T3) rather than run inline — in
tests CELERY_TASK_ALWAYS_EAGER still runs it immediately, so the
article it ingests is visible by the time the response returns."""
alice, _ = two_users
client.login(username="alice", password="pw-alice-123")
responses.add(responses.GET, "https://example.com/feed.xml", body=GOOD_FEED_XML, status=200)
@@ -62,9 +66,10 @@ def test_add_feed_view_subscribes_and_fetches_immediately(client, two_users):
response = client.post(reverse("feeds:add_feed"), {"feed_url": "https://example.com/feed.xml"})
assert response.status_code == 302
assert Subscription.objects.filter(
subscription = Subscription.objects.get(
user=alice, feed__feed_url="https://example.com/feed.xml"
).exists()
)
assert subscription.feed.articles.filter(guid="guid-1").exists()
@pytest.mark.django_db
+8 -16
View File
@@ -12,8 +12,7 @@ from apps.articles.models import SubscriptionTag, Tag
from .forms import AddFeedForm, EditSubscriptionForm, OPMLImportForm
from .models import Feed, Subscription
from .opml import export_opml, extract_opml_bytes, import_opml, parse_opml
from .services.fetcher import fetch_and_ingest
from .tasks import import_opml_task
from .tasks import fetch_feed, import_opml_task
OPML_ASYNC_THRESHOLD = 50
@@ -39,22 +38,15 @@ def add_feed(request):
user=request.user, subscription=subscription, tag=tag
)
# Fetch synchronously so the user sees results immediately; the
# periodic Celery poll takes over from here for subsequent fetches.
result = fetch_and_ingest(feed)
feed.save()
if result.status == Feed.STATUS_ERROR:
messages.warning(
request,
f"Subscribed to {feed_url}, but the initial fetch failed: {result.error_message}",
)
else:
article_count = len(result.new_article_ids)
# Enqueue the initial fetch instead of running it inline (T3) — a
# slow or large feed would otherwise tie up a Gunicorn worker for
# the duration of the request; fetch_feed handles the save/error
# bookkeeping the same way the periodic poll's dispatch does.
fetch_feed.delay(feed.pk)
messages.success(
request,
f"Subscribed to '{subscription.display_title}' — "
f"{article_count} article(s) fetched.",
f"Subscribed to {feed_url}. Fetching its articles now — "
"refresh in a few seconds if you don't see them yet.",
)
return redirect("reader:stream")