Reduce complexity of status data URL query

Get rid of all the junk trying to only return URLs that have been
checked in the last 24 hours; it just isn't worth it. Instead, do that
screening only in the views that need it, namely the HTML status page.

Signed-off-by: Dan McGee <dan@archlinux.org>

2 files changed, 9 insertions, 10 deletions

diff --git a/mirrors/utils.py b/mirrors/utils.py
index 4484fa24..8edceb9b 100644
--- a/mirrors/utils.py
+++ b/mirrors/utils.py
@@ -118,20 +118,16 @@ def annotate_url(url, url_data):
 def get_mirror_statuses(cutoff=DEFAULT_CUTOFF, mirror_id=None, show_all=False):
     cutoff_time = now() - cutoff
 
-    valid_urls = MirrorUrl.objects.filter(
-            logs__check_time__gte=cutoff_time).distinct()
-
+    urls = MirrorUrl.objects.select_related(
+            'mirror', 'protocol').order_by('mirror__id', 'url')
     if mirror_id:
-        valid_urls = valid_urls.filter(mirror_id=mirror_id)
+        urls = urls.filter(mirror_id=mirror_id)
     if not show_all:
-        valid_urls = valid_urls.filter(active=True, mirror__active=True,
+        urls = urls.filter(active=True, mirror__active=True,
                 mirror__public=True)
 
-    url_data = status_data(cutoff, mirror_id)
-    urls = MirrorUrl.objects.select_related('mirror', 'protocol').filter(
-            id__in=valid_urls).order_by('mirror__id', 'url')
-
     if urls:
+        url_data = status_data(cutoff, mirror_id)
         urls = [annotate_url(url, url_data.get(url.id, {})) for url in urls]
         last_check = max([u.last_check for u in urls if u.last_check])
         num_checks = max([u.check_count for u in urls])
diff --git a/mirrors/views.py b/mirrors/views.py
index 1a9741ed..90787763 100644
--- a/mirrors/views.py
+++ b/mirrors/views.py
@@ -245,7 +245,10 @@ def status(request, tier=None):
         if tier is not None and url.mirror.tier != tier:
             continue
         # split them into good and bad lists based on delay
-        if not url.delay or url.delay > bad_timedelta:
+        if url.completion_pct is None:
+            # skip URLs that have never been checked
+            continue
+        elif not url.delay or url.delay > bad_timedelta:
             bad_urls.append(url)
         else:
             good_urls.append(url)