Add feature to automatically delete old toots (#16529)

* Add account statuses cleanup policy model

* Record last inspected toot to delete to speed up successive calls to statuses_to_delete

* Add service to cleanup a given account's statuses within a budget

* Add worker to go through account policies and delete old toots

* Fix last inspected status id logic

All existing statuses older or equal to last inspected status id must be
kept by the current policy. This is an invariant that must be kept so that
resuming deletion from the last inspected status remains sound.

* Add tests

* Refactor scheduler and add tests

* Add user interface

* Add support for discriminating based on boosts/favs

* Add UI support for min_reblogs and min_favs, rework UI

* Address first round of review comments

* Replace Snowflake#id_at_start with with_random parameter

* Add tests

* Add tests for StatusesCleanupController

* Rework settings page

* Adjust load-avoiding mechanisms

* Please CodeClimate

1 files changed, 127 insertions, 0 deletions

diff --git a/spec/workers/scheduler/accounts_statuses_cleanup_scheduler_spec.rb b/spec/workers/scheduler/accounts_statuses_cleanup_scheduler_spec.rb
new file mode 100644
index 000000000..8f20725c8
--- /dev/null
+++ b/spec/workers/scheduler/accounts_statuses_cleanup_scheduler_spec.rb
@@ -0,0 +1,127 @@
+require 'rails_helper'
+
+describe Scheduler::AccountsStatusesCleanupScheduler do
+  subject { described_class.new }
+
+  let!(:account1)  { Fabricate(:account, domain: nil) }
+  let!(:account2)  { Fabricate(:account, domain: nil) }
+  let!(:account3)  { Fabricate(:account, domain: nil) }
+  let!(:account4)  { Fabricate(:account, domain: nil) }
+  let!(:remote)    { Fabricate(:account) }
+
+  let!(:policy1)   { Fabricate(:account_statuses_cleanup_policy, account: account1) }
+  let!(:policy2)   { Fabricate(:account_statuses_cleanup_policy, account: account3) }
+  let!(:policy3)   { Fabricate(:account_statuses_cleanup_policy, account: account4, enabled: false) }
+
+  let(:queue_size)       { 0 }
+  let(:queue_latency)    { 0 }
+  let(:process_set_stub) do
+    [
+      {
+        'concurrency' => 2,
+        'queues' => ['push', 'default'],
+      },
+    ]
+  end
+  let(:retry_size) { 0 }
+
+  before do
+    queue_stub = double
+    allow(queue_stub).to receive(:size).and_return(queue_size)
+    allow(queue_stub).to receive(:latency).and_return(queue_latency)
+    allow(Sidekiq::Queue).to receive(:new).and_return(queue_stub)
+    allow(Sidekiq::ProcessSet).to receive(:new).and_return(process_set_stub)
+
+    sidekiq_stats_stub = double
+    allow(sidekiq_stats_stub).to receive(:retry_size).and_return(retry_size)
+    allow(Sidekiq::Stats).to receive(:new).and_return(sidekiq_stats_stub)
+
+    # Create a bunch of old statuses
+    10.times do
+      Fabricate(:status, account: account1, created_at: 3.years.ago)
+      Fabricate(:status, account: account2, created_at: 3.years.ago)
+      Fabricate(:status, account: account3, created_at: 3.years.ago)
+      Fabricate(:status, account: account4, created_at: 3.years.ago)
+      Fabricate(:status, account: remote, created_at: 3.years.ago)
+    end
+
+    # Create a bunch of newer statuses
+    5.times do
+      Fabricate(:status, account: account1, created_at: 3.minutes.ago)
+      Fabricate(:status, account: account2, created_at: 3.minutes.ago)
+      Fabricate(:status, account: account3, created_at: 3.minutes.ago)
+      Fabricate(:status, account: account4, created_at: 3.minutes.ago)
+      Fabricate(:status, account: remote, created_at: 3.minutes.ago)
+    end
+  end
+
+  describe '#under_load?' do
+    context 'when nothing is queued' do
+      it 'returns false' do
+        expect(subject.under_load?).to be false
+      end
+    end
+
+    context 'when numerous jobs are queued' do
+      let(:queue_size)    { 5 }
+      let(:queue_latency) { 120 }
+
+      it 'returns true' do
+        expect(subject.under_load?).to be true
+      end
+    end
+
+    context 'when there is a huge amount of jobs to retry' do
+      let(:retry_size) { 1_000_000 }
+
+      it 'returns true' do
+        expect(subject.under_load?).to be true
+      end
+    end
+  end
+
+  describe '#get_budget' do
+    context 'on a single thread' do
+      let(:process_set_stub) { [ { 'concurrency' => 1, 'queues' => ['push', 'default'] } ] }
+
+      it 'returns a low value' do
+        expect(subject.compute_budget).to be < 10
+      end
+    end
+
+    context 'on a lot of threads' do
+      let(:process_set_stub) do
+        [
+          { 'concurrency' => 2, 'queues' => ['push', 'default'] },
+          { 'concurrency' => 2, 'queues' => ['push'] },
+          { 'concurrency' => 2, 'queues' => ['push'] },
+          { 'concurrency' => 2, 'queues' => ['push'] },
+        ]
+      end
+
+      it 'returns a larger value' do
+        expect(subject.compute_budget).to be > 10
+      end
+    end
+  end
+
+  describe '#perform' do
+    context 'when the budget is lower than the number of toots to delete' do
+      it 'deletes as many statuses as the given budget' do
+        expect { subject.perform }.to change { Status.count }.by(-subject.compute_budget)
+      end
+
+      it 'does not delete from accounts with no cleanup policy' do
+        expect { subject.perform }.to_not change { account2.statuses.count }
+      end
+
+      it 'does not delete from accounts with disabled cleanup policies' do
+        expect { subject.perform }.to_not change { account4.statuses.count }
+      end
+
+      it 'eventually deletes every deletable toot' do
+        expect { subject.perform; subject.perform; subject.perform; subject.perform }.to change { Status.count }.by(-20)
+      end
+    end
+  end
+end