Object: remove context_id field

30 to 70% of the objects in the object table are simple JSON objects
containing a single field, 'id', being the context's ID. The reason for
the creation of an object per context seems to be an old relic from the
StatusNet era, and has only been used nowadays as an helper for threads
in Pleroma-FE via the `pleroma.conversation_id` field in status views.
An object per context was created, and its numerical ID (table column)
was used and stored as 'context_id' in the object and activity along
with the full 'context' URI/string.

This commit removes this field and stops creation of objects for each
context, which will also allow incoming activities to use activity IDs
as contexts, something which was not possible before, or would have been
very broken under most circumstances.

The `pleroma.conversation_id` field has been reimplemented in a way to
maintain backwards-compatibility by calculating a CRC32 of the full
context URI/string in the object, instead of relying on the row ID for
the created context object.

14 files changed, 9 insertions, 102 deletions

diff --git a/lib/pleroma/object.ex b/lib/pleroma/object.ex
index fe264b5e0..c214a79c5 100644
--- a/lib/pleroma/object.ex
+++ b/lib/pleroma/object.ex
@@ -208,10 +208,6 @@ defmodule Pleroma.Object do
     end
   end
 
-  def context_mapping(context) do
-    Object.change(%Object{}, %{data: %{"id" => context}})
-  end
-
   def make_tombstone(%Object{data: %{"id" => id, "type" => type}}, deleted \\ DateTime.utc_now()) do
     %ObjectTombstone{
       id: id,
diff --git a/lib/pleroma/web/activity_pub/object_validators/article_note_page_validator.ex b/lib/pleroma/web/activity_pub/object_validators/article_note_page_validator.ex
index 57c8d1dc0..c5fb94034 100644
--- a/lib/pleroma/web/activity_pub/object_validators/article_note_page_validator.ex
+++ b/lib/pleroma/web/activity_pub/object_validators/article_note_page_validator.ex
@@ -94,7 +94,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.ArticleNotePageValidator do
   defp validate_data(data_cng) do
     data_cng
     |> validate_inclusion(:type, ["Article", "Note", "Page"])
-    |> validate_required([:id, :actor, :attributedTo, :type, :context, :context_id])
+    |> validate_required([:id, :actor, :attributedTo, :type, :context])
     |> CommonValidations.validate_any_presence([:cc, :to])
     |> CommonValidations.validate_fields_match([:actor, :attributedTo])
     |> CommonValidations.validate_actor_presence()
diff --git a/lib/pleroma/web/activity_pub/object_validators/common_fixes.ex b/lib/pleroma/web/activity_pub/object_validators/common_fixes.ex
index 4f8c083eb..c7e292bec 100644
--- a/lib/pleroma/web/activity_pub/object_validators/common_fixes.ex
+++ b/lib/pleroma/web/activity_pub/object_validators/common_fixes.ex
@@ -22,14 +22,12 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.CommonFixes do
   end
 
   def fix_object_defaults(data) do
-    %{data: %{"id" => context}, id: context_id} =
-      Utils.create_context(data["context"] || data["conversation"])
+    context = Utils.maybe_create_context(data["context"] || data["conversation"])
 
     %User{follower_address: follower_collection} = User.get_cached_by_ap_id(data["attributedTo"])
 
     data
     |> Map.put("context", context)
-    |> Map.put("context_id", context_id)
     |> cast_and_filter_recipients("to", follower_collection)
     |> cast_and_filter_recipients("cc", follower_collection)
     |> cast_and_filter_recipients("bto", follower_collection)
diff --git a/lib/pleroma/web/activity_pub/object_validators/event_validator.ex b/lib/pleroma/web/activity_pub/object_validators/event_validator.ex
index 0e99f2037..ab204f69a 100644
--- a/lib/pleroma/web/activity_pub/object_validators/event_validator.ex
+++ b/lib/pleroma/web/activity_pub/object_validators/event_validator.ex
@@ -62,7 +62,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.EventValidator do
   defp validate_data(data_cng) do
     data_cng
     |> validate_inclusion(:type, ["Event"])
-    |> validate_required([:id, :actor, :attributedTo, :type, :context, :context_id])
+    |> validate_required([:id, :actor, :attributedTo, :type, :context])
     |> CommonValidations.validate_any_presence([:cc, :to])
     |> CommonValidations.validate_fields_match([:actor, :attributedTo])
     |> CommonValidations.validate_actor_presence()
diff --git a/lib/pleroma/web/activity_pub/object_validators/question_validator.ex b/lib/pleroma/web/activity_pub/object_validators/question_validator.ex
index 9412be4bc..ce3305142 100644
--- a/lib/pleroma/web/activity_pub/object_validators/question_validator.ex
+++ b/lib/pleroma/web/activity_pub/object_validators/question_validator.ex
@@ -80,7 +80,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.QuestionValidator do
   defp validate_data(data_cng) do
     data_cng
     |> validate_inclusion(:type, ["Question"])
-    |> validate_required([:id, :actor, :attributedTo, :type, :context, :context_id])
+    |> validate_required([:id, :actor, :attributedTo, :type, :context])
     |> CommonValidations.validate_any_presence([:cc, :to])
     |> CommonValidations.validate_fields_match([:actor, :attributedTo])
     |> CommonValidations.validate_actor_presence()
diff --git a/lib/pleroma/web/activity_pub/utils.ex b/lib/pleroma/web/activity_pub/utils.ex
index 9cde7805c..d3b7d804f 100644
--- a/lib/pleroma/web/activity_pub/utils.ex
+++ b/lib/pleroma/web/activity_pub/utils.ex
@@ -154,22 +154,7 @@ defmodule Pleroma.Web.ActivityPub.Utils do
     Notification.get_notified_from_activity(%Activity{data: object}, false)
   end
 
-  def create_context(context) do
-    context = context || generate_id("contexts")
-
-    # Ecto has problems accessing the constraint inside the jsonb,
-    # so we explicitly check for the existed object before insert
-    object = Object.get_cached_by_ap_id(context)
-
-    with true <- is_nil(object),
-         changeset <- Object.context_mapping(context),
-         {:ok, inserted_object} <- Repo.insert(changeset) do
-      inserted_object
-    else
-      _ ->
-        object
-    end
-  end
+  def maybe_create_context(context), do: context || generate_id("contexts")
 
   @doc """
   Enqueues an activity for federation if it's local
@@ -201,18 +186,16 @@ defmodule Pleroma.Web.ActivityPub.Utils do
     |> Map.put_new("id", "pleroma:fakeid")
     |> Map.put_new_lazy("published", &make_date/0)
     |> Map.put_new("context", "pleroma:fakecontext")
-    |> Map.put_new("context_id", -1)
     |> lazy_put_object_defaults(true)
   end
 
   def lazy_put_activity_defaults(map, _fake?) do
-    %{data: %{"id" => context}, id: context_id} = create_context(map["context"])
+    context = maybe_create_context(map["context"])
 
     map
     |> Map.put_new_lazy("id", &generate_activity_id/0)
     |> Map.put_new_lazy("published", &make_date/0)
     |> Map.put_new("context", context)
-    |> Map.put_new("context_id", context_id)
     |> lazy_put_object_defaults(false)
   end
 
@@ -226,7 +209,6 @@ defmodule Pleroma.Web.ActivityPub.Utils do
       |> Map.put_new("id", "pleroma:fake_object_id")
       |> Map.put_new_lazy("published", &make_date/0)
       |> Map.put_new("context", activity["context"])
-      |> Map.put_new("context_id", activity["context_id"])
       |> Map.put_new("fake", true)
 
     %{activity | "object" => object}
@@ -239,7 +221,6 @@ defmodule Pleroma.Web.ActivityPub.Utils do
       |> Map.put_new_lazy("id", &generate_object_id/0)
       |> Map.put_new_lazy("published", &make_date/0)
       |> Map.put_new("context", activity["context"])
-      |> Map.put_new("context_id", activity["context_id"])
 
     %{activity | "object" => object}
   end
diff --git a/lib/pleroma/web/common_api/utils.ex b/lib/pleroma/web/common_api/utils.ex
index ce850b038..052bd7770 100644
--- a/lib/pleroma/web/common_api/utils.ex
+++ b/lib/pleroma/web/common_api/utils.ex
@@ -449,35 +449,6 @@ defmodule Pleroma.Web.CommonAPI.Utils do
 
   def get_report_statuses(_, _), do: {:ok, nil}
 
-  # DEPRECATED mostly, context objects are now created at insertion time.
-  def context_to_conversation_id(context) do
-    with %Object{id: id} <- Object.get_cached_by_ap_id(context) do
-      id
-    else
-      _e ->
-        changeset = Object.context_mapping(context)
-
-        case Repo.insert(changeset) do
-          {:ok, %{id: id}} ->
-            id
-
-          # This should be solved by an upsert, but it seems ecto
-          # has problems accessing the constraint inside the jsonb.
-          {:error, _} ->
-            Object.get_cached_by_ap_id(context).id
-        end
-    end
-  end
-
-  def conversation_id_to_context(id) do
-    with %Object{data: %{"id" => context}} <- Repo.get(Object, id) do
-      context
-    else
-      _e ->
-        {:error, dgettext("errors", "No such conversation")}
-    end
-  end
-
   def validate_character_limit("" = _full_payload, [] = _attachments) do
     {:error, dgettext("errors", "Cannot post an empty status without attachments")}
   end
diff --git a/lib/pleroma/web/mastodon_api/views/status_view.ex b/lib/pleroma/web/mastodon_api/views/status_view.ex
index 1ebfd6740..31a0c420f 100644
--- a/lib/pleroma/web/mastodon_api/views/status_view.ex
+++ b/lib/pleroma/web/mastodon_api/views/status_view.ex
@@ -61,7 +61,7 @@ defmodule Pleroma.Web.MastodonAPI.StatusView do
     do: context_id
 
   defp get_context_id(%{data: %{"context" => context}}) when is_binary(context),
-    do: Utils.context_to_conversation_id(context)
+    do: :erlang.crc32(context)
 
   defp get_context_id(_), do: nil
 
diff --git a/test/pleroma/web/activity_pub/activity_pub_test.exs b/test/pleroma/web/activity_pub/activity_pub_test.exs
index 4d7e76266..b8d73ea10 100644
--- a/test/pleroma/web/activity_pub/activity_pub_test.exs
+++ b/test/pleroma/web/activity_pub/activity_pub_test.exs
@@ -554,7 +554,6 @@ defmodule Pleroma.Web.ActivityPub.ActivityPubTest do
       assert activity.data["ok"] == data["ok"]
       assert activity.data["id"] == given_id
       assert activity.data["context"] == "blabla"
-      assert activity.data["context_id"]
     end
 
     test "adds a context when none is there" do
@@ -576,8 +575,6 @@ defmodule Pleroma.Web.ActivityPub.ActivityPubTest do
 
       assert is_binary(activity.data["context"])
       assert is_binary(object.data["context"])
-      assert activity.data["context_id"]
-      assert object.data["context_id"]
     end
 
     test "adds an id to a given object if it lacks one and is a note and inserts it to the object database" do
@@ -1612,7 +1609,7 @@ defmodule Pleroma.Web.ActivityPub.ActivityPubTest do
                })
 
       assert Repo.aggregate(Activity, :count, :id) == 1
-      assert Repo.aggregate(Object, :count, :id) == 2
+      assert Repo.aggregate(Object, :count, :id) == 1
       assert Repo.aggregate(Notification, :count, :id) == 0
     end
   end
diff --git a/test/pleroma/web/activity_pub/transmogrifier/question_handling_test.exs b/test/pleroma/web/activity_pub/transmogrifier/question_handling_test.exs
index d22ec400d..d31070546 100644
--- a/test/pleroma/web/activity_pub/transmogrifier/question_handling_test.exs
+++ b/test/pleroma/web/activity_pub/transmogrifier/question_handling_test.exs
@@ -33,8 +33,6 @@ defmodule Pleroma.Web.ActivityPub.Transmogrifier.QuestionHandlingTest do
     assert object.data["context"] ==
              "tag:mastodon.sdf.org,2019-05-10:objectId=15095122:objectType=Conversation"
 
-    assert object.data["context_id"]
-
     assert object.data["anyOf"] == []
 
     assert Enum.sort(object.data["oneOf"]) ==
@@ -68,7 +66,6 @@ defmodule Pleroma.Web.ActivityPub.Transmogrifier.QuestionHandlingTest do
     reply_object = Object.normalize(reply_activity, fetch: false)
 
     assert reply_object.data["context"] == object.data["context"]
-    assert reply_object.data["context_id"] == object.data["context_id"]
   end
 
   test "Mastodon Question activity with HTML tags in plaintext" do
diff --git a/test/pleroma/web/activity_pub/transmogrifier_test.exs b/test/pleroma/web/activity_pub/transmogrifier_test.exs
index 335fe1a30..b254e5591 100644
--- a/test/pleroma/web/activity_pub/transmogrifier_test.exs
+++ b/test/pleroma/web/activity_pub/transmogrifier_test.exs
@@ -227,7 +227,6 @@ defmodule Pleroma.Web.ActivityPub.TransmogrifierTest do
       assert is_nil(modified["object"]["like_count"])
       assert is_nil(modified["object"]["announcements"])
       assert is_nil(modified["object"]["announcement_count"])
-      assert is_nil(modified["object"]["context_id"])
       assert is_nil(modified["object"]["generator"])
     end
 
@@ -242,7 +241,6 @@ defmodule Pleroma.Web.ActivityPub.TransmogrifierTest do
       assert is_nil(modified["object"]["like_count"])
       assert is_nil(modified["object"]["announcements"])
       assert is_nil(modified["object"]["announcement_count"])
-      assert is_nil(modified["object"]["context_id"])
       assert is_nil(modified["object"]["likes"])
     end
 
diff --git a/test/pleroma/web/activity_pub/utils_test.exs b/test/pleroma/web/activity_pub/utils_test.exs
index 447621718..e42893849 100644
--- a/test/pleroma/web/activity_pub/utils_test.exs
+++ b/test/pleroma/web/activity_pub/utils_test.exs
@@ -429,7 +429,6 @@ defmodule Pleroma.Web.ActivityPub.UtilsTest do
       object = Object.normalize(note_activity, fetch: false)
       res = Utils.lazy_put_activity_defaults(%{"context" => object.data["id"]})
       assert res["context"] == object.data["id"]
-      assert res["context_id"] == object.id
       assert res["id"]
       assert res["published"]
     end
@@ -437,7 +436,6 @@ defmodule Pleroma.Web.ActivityPub.UtilsTest do
     test "returns map with fake id and published data" do
       assert %{
                "context" => "pleroma:fakecontext",
-               "context_id" => -1,
                "id" => "pleroma:fakeid",
                "published" => _
              } = Utils.lazy_put_activity_defaults(%{}, true)
@@ -454,13 +452,11 @@ defmodule Pleroma.Web.ActivityPub.UtilsTest do
         })
 
       assert res["context"] == object.data["id"]
-      assert res["context_id"] == object.id
       assert res["id"]
       assert res["published"]
       assert res["object"]["id"]
       assert res["object"]["published"]
       assert res["object"]["context"] == object.data["id"]
-      assert res["object"]["context_id"] == object.id
     end
   end
 
diff --git a/test/pleroma/web/common_api/utils_test.exs b/test/pleroma/web/common_api/utils_test.exs
index 5b2019969..33ede0f04 100644
--- a/test/pleroma/web/common_api/utils_test.exs
+++ b/test/pleroma/web/common_api/utils_test.exs
@@ -273,22 +273,6 @@ defmodule Pleroma.Web.CommonAPI.UtilsTest do
     end
   end
 
-  describe "context_to_conversation_id" do
-    test "creates a mapping object" do
-      conversation_id = Utils.context_to_conversation_id("random context")
-      object = Object.get_by_ap_id("random context")
-
-      assert conversation_id == object.id
-    end
-
-    test "returns an existing mapping for an existing object" do
-      {:ok, object} = Object.context_mapping("random context") |> Repo.insert()
-      conversation_id = Utils.context_to_conversation_id("random context")
-
-      assert conversation_id == object.id
-    end
-  end
-
   describe "formats date to asctime" do
     test "when date is in ISO 8601 format" do
       date = DateTime.utc_now() |> DateTime.to_iso8601()
@@ -517,17 +501,6 @@ defmodule Pleroma.Web.CommonAPI.UtilsTest do
     end
   end
 
-  describe "conversation_id_to_context/1" do
-    test "returns id" do
-      object = insert(:note)
-      assert Utils.conversation_id_to_context(object.id) == object.data["id"]
-    end
-
-    test "returns error if object not found" do
-      assert Utils.conversation_id_to_context("123") == {:error, "No such conversation"}
-    end
-  end
-
   describe "maybe_notify_mentioned_recipients/2" do
     test "returns recipients when activity is not `Create`" do
       activity = insert(:like_activity)
diff --git a/test/pleroma/web/mastodon_api/views/status_view_test.exs b/test/pleroma/web/mastodon_api/views/status_view_test.exs
index 5d81c92b9..fa71f86f2 100644
--- a/test/pleroma/web/mastodon_api/views/status_view_test.exs
+++ b/test/pleroma/web/mastodon_api/views/status_view_test.exs
@@ -226,7 +226,7 @@ defmodule Pleroma.Web.MastodonAPI.StatusViewTest do
     object_data = Object.normalize(note, fetch: false).data
     user = User.get_cached_by_ap_id(note.data["actor"])
 
-    convo_id = Utils.context_to_conversation_id(object_data["context"])
+    convo_id = :erlang.crc32(object_data["context"])
 
     status = StatusView.render("show.json", %{activity: note})