From 6d22fcb21a6f0bafc206123d23af501697fe6114 Mon Sep 17 00:00:00 2001
From: Magnus Hagander <magnus@hagander.net>
Date: Mon, 16 Jun 2025 11:25:19 +0200
Subject: [PATCH] Round of since/ and before/ URLs to whole days

Previously we'd generate links for each individual email, when clicking
the link on the emails. This generates a huge number of URLs that
contain basically the same thing, which has an adverse effect on both
caching and (stupid) bots. So round it off to even days which puts at
least some level of a cap on it.

Also, when a hit comes in that specifies the full hour, redirect it back
to the rounded-off value.
---
 django/archives/mailarchives/models.py                      | 4 ++++
 django/archives/mailarchives/templates/_message.html        | 2 +-
 .../mailarchives/templates/datelist_topandbottom.html       | 4 ++--
 django/archives/mailarchives/views.py                       | 6 ++++++
 4 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/django/archives/mailarchives/models.py b/django/archives/mailarchives/models.py
index 44c4469..4a6a055 100644
--- a/django/archives/mailarchives/models.py
+++ b/django/archives/mailarchives/models.py
@@ -41,6 +41,10 @@ class Message(models.Model):
     def shortdate(self):
         return self.date.strftime("%Y%m%d%H%M")
 
+    @property
+    def dateonly(self):
+        return self.date.strftime("%Y%m%d")
+
     def from_name_only(self):
         try:
             return parseaddr(self.mailfrom)[0]
diff --git a/django/archives/mailarchives/templates/_message.html b/django/archives/mailarchives/templates/_message.html
index b4f42af..5b7404a 100644
--- a/django/archives/mailarchives/templates/_message.html
+++ b/django/archives/mailarchives/templates/_message.html
@@ -52,7 +52,7 @@
       <th scope="row">Lists:</th>
       <td>
         {%for l in lists %}
-          <span class="listname"><a href="/https/git.postgresql.org/list/{{l.listname}}/since/{{msg.shortdate}}">{{l.listname}}</a></span>
+          <span class="listname"><a href="/https/git.postgresql.org/list/{{l.listname}}/since/{{msg.dateonly}}0000">{{l.listname}}</a></span>
         {%endfor%}
       </td>
     </tr>
diff --git a/django/archives/mailarchives/templates/datelist_topandbottom.html b/django/archives/mailarchives/templates/datelist_topandbottom.html
index cd620ca..ccac52a 100644
--- a/django/archives/mailarchives/templates/datelist_topandbottom.html
+++ b/django/archives/mailarchives/templates/datelist_topandbottom.html
@@ -1,11 +1,11 @@
 {%if messages%}
 <h3>Browse Archives</h3>
 {%with messages|first as firstmsg%}
-<a href="/https/git.postgresql.org/list/{{list.listname}}/before/{{firstmsg.shortdate}}" rel="nofollow">Prev</a>
+<a href="/https/git.postgresql.org/list/{{list.listname}}/before/{{firstmsg.dateonly}}0000" rel="nofollow">Prev</a>
 {%endwith%}
 |
 {%with messages|last as lastmsg%}
-<a href="/https/git.postgresql.org/list/{{list.listname}}/since/{{lastmsg.shortdate}}" rel="nofollow">Next</a>
+<a href="/https/git.postgresql.org/list/{{list.listname}}/since/{{lastmsg.dateonly}}0000" rel="nofollow">Next</a>
 {%endwith%}
 {%endif%}
 
diff --git a/django/archives/mailarchives/views.py b/django/archives/mailarchives/views.py
index 1fc1a8a..c1118ac 100644
--- a/django/archives/mailarchives/views.py
+++ b/django/archives/mailarchives/views.py
@@ -331,6 +331,9 @@ def datelistsincetime(request, listname, year, month, day, hour, minute):
     ensure_list_permissions(request, l)
 
     try:
+        if int(hour) != 0 or int(minute) != 0:
+            # "round off" timestamps to the whole day, to reduce the number of unique urls
+            return HttpResponseRedirect("{}{}{}0000".format(year, month, day))
         d = datetime(int(year), int(month), int(day), int(hour), int(minute))
     except ValueError:
         raise Http404("Invalid date format, not found")
@@ -352,6 +355,9 @@ def datelistbeforetime(request, listname, year, month, day, hour, minute):
     ensure_list_permissions(request, l)
 
     try:
+        if int(hour) != 0 or int(minute) != 0:
+            # "round off" timestamps to the whole day, to reduce the number of unique urls
+            return HttpResponseRedirect("{}{}{}0000".format(year, month, day))
         d = datetime(int(year), int(month), int(day), int(hour), int(minute))
     except ValueError:
         raise Http404("Invalid date format, not found")
-- 
2.39.5