search: don't index invalid cells (#52418)

combo/apps/search/utils.py

@@ -71,6 +71,8 @@ def index_site():
                 for v in validity_info_list
                 if v.object_id == cell.pk and v.content_type.model_class() == cell.__class__
             ]
+            if cell.is_hidden_because_invalid():
+                continue
             cell_type = ContentType.objects.get_for_model(cell)
             indexed_cell = IndexedCell(cell_type=cell_type, cell_pk=cell.id)
             try:

combo/data/models.py

@@ -1083,15 +1083,17 @@ class CellBase(six.with_metaclass(CellMeta, models.Model)):
                 return True
         return False
 
+    def is_hidden_because_invalid(self):
+        validity_info = self.get_validity_info()
+        return (
+            validity_info is not None
+            and validity_info.invalid_datetime
+            and validity_info.invalid_datetime <= now()
+        )
+
     def is_visible(self, user=None, check_validity_info=True):
-        if check_validity_info:
-            validity_info = self.get_validity_info()
-            if (
-                validity_info is not None
-                and validity_info.invalid_datetime
-                and validity_info.invalid_datetime <= now()
-            ):
-                return False
+        if check_validity_info and self.is_hidden_because_invalid():
+            return False
         return element_is_visible(self, user=user)
 
     def is_relevant(self, context):

tests/test_search.py

@@ -1,3 +1,4 @@
+import datetime
 import json
 import os
 import pytest
@@ -12,6 +13,7 @@ from django.test.client import RequestFactory
 from django.test.utils import CaptureQueriesContext
 from django.urls import reverse
 from django.utils.http import urlencode
+from django.utils.timezone import now
 
 from requests.exceptions import ConnectionError
 from requests.models import Response
@@ -1155,6 +1157,28 @@ def test_index_site_inactive_placeholder(app):
     assert IndexedCell.objects.count() == 0
 
 
+def test_index_site_invalid_cell(app):
+    page = Page.objects.create(title='page', slug='example-page')
+    cell = TextCell.objects.create(page=page, placeholder='content', text='<p>foobar</p>', order=0)
+
+    index_site()
+    assert IndexedCell.objects.count() == 1
+
+    # invalid cell since just now
+    cell.mark_as_invalid(reason_code='foobar')
+    assert cell.is_visible() is True
+    index_site()
+    assert IndexedCell.objects.count() == 1
+
+    # invalid cell since two days
+    validity_info = cell.get_validity_info()
+    validity_info.invalid_since = now() - datetime.timedelta(days=2)
+    validity_info.save()
+    assert cell.is_visible() is False
+    index_site()
+    assert IndexedCell.objects.count() == 0
+
+
 def test_index_site_num_queries(settings, app):
     group = Group.objects.create(name='plop')
     for i in range(0, 10):