Add core data models and the feed fetch pipeline

Implements Folder/Feed/Subscription (feeds app) and Article/
UserArticleState/Tag/ArticleTag (articles app), preserving the
shared-Feed/shared-Article dedup design: a feed is fetched once
regardless of subscriber count, and per-user read/star/save/hide state
is created lazily rather than eagerly per subscriber.

Adds the fetch pipeline: feedparser-based parsing with conditional GET
(ETag/Last-Modified), nh3 HTML sanitization, guid-based dedup with
content-hash change detection that never resets a user's read/star
state, and exponential backoff on fetch errors respecting a configured
minimum interval floor. Wired into Celery via poll_due_feeds/fetch_feed
tasks, plus a minimal "add feed by URL" view that fetches synchronously
so new subscribers see results immediately.

Verified live against a real feed (hnrss.org) for ingestion, 304
dedup, and error backoff, in addition to the unit/integration test
suite (28 tests, all passing).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
ksmithandClaude Sonnet 5 committed 2026-10-30 16:03:20 -06:00
1 parent c635cd9330
commit d200ab80f7
19 files changed
+1355 -8

No files matched your search

+11
View File
@@ -0,0 +1,11 @@
{% extends "base.html" %}
{% block title %}Add feed — Evection{% endblock %}
{% block content %}
<h1>Add a feed</h1>
<form method="post">
{% csrf_token %}
{{ form.as_p }}
<button type="submit">Subscribe</button>
</form>
<p><a href="{% url 'reader:stream' %}">Back to stream</a></p>
{% endblock %}
+1 -1
View File
@@ -3,5 +3,5 @@
{% block content %}
<h1>Your stream</h1>
<p>No feeds yet. The reading UI lands in a later milestone.</p>
<p><a href="{% url 'accounts:logout' %}">Log out</a></p>
<p><a href="{% url 'feeds:add_feed' %}">Add a feed</a> &middot; <a href="{% url 'accounts:logout' %}">Log out</a></p>
{% endblock %}