Move readability HTML cleanup from request-time to fetch-time

Changes:
 - Run readability on post content/original_content when fetching feeds, not on every post detail render — eliminates wasted CPU on cached page hits.
 - Remove the now-unused clean_html template filter and its tests.
 - Drop redundant .all() in the index view.

Author: matagus

planet/managers.py

@@ -4,6 +4,7 @@
 from django.apps import apps
 from django.db import models, transaction
 from django.utils import timezone
+from readability import Document
 
 from planet.utils import md5_hash, normalize_language, to_datetime
 
@@ -179,9 +180,11 @@ def create_from(self, entry_data, feed):
             return None
 
         try:
-            post.content = entry_data.summary
+            raw_content = entry_data.summary
         except AttributeError:
-            pass
+            raw_content = ""
+        if raw_content:
+            post.content = Document(raw_content).summary(html_partial=True)
 
         try:
             language = entry_data.summary_detail.language

planet/templates/planet/posts/detail.html

@@ -30,7 +30,7 @@
             {% include "planet/posts/full_details.html" %}
         </div>
         <div class="row">
-            <p>{{ post.original_content|default:post.content|clean_html }}</p>
+            <p>{{ post.original_content|default:post.content|safe }}</p>
         </div>
     </div>
 {% endblock %}

planet/templatetags/planet_tags.py

@@ -1,25 +1,14 @@
 import re
 
 from django import template
-from django.template.defaultfilters import stringfilter
 from django.utils.html import strip_tags
-from django.utils.safestring import mark_safe
-from readability import Document
 
 from planet.models import Author, Blog, Feed, Post
 from planet.settings import PLANET_CONFIG
 
 register = template.Library()
 
 
-@register.filter
-@stringfilter
-def clean_html(html):
-    if not html:
-        return ""
-    return mark_safe(Document(html).summary(html_partial=True))
-
-
 @register.simple_tag
 def get_first_paragraph(body):
     if body is None:

planet/utils.py

@@ -53,7 +53,7 @@ def fetch_post_content(url):
             raw = resp.read()
             charset = resp.headers.get_content_charset() or "utf-8"
         html = raw.decode(charset, errors="replace")
-        return Document(html).summary()
+        return Document(html).summary(html_partial=True)
     except Exception as exc:
         logger.warning("Failed to fetch original content for %s: %s", url, exc)
         return None

planet/views.py

@@ -7,7 +7,7 @@
 
 
 def index(request):
-    posts = Post.objects.all().with_relations()
+    posts = Post.objects.with_relations()
     return render(request, "planet/posts/list.html", {"posts": posts})