Changes to tootctl statuses remove (#11267)

* Fix comments in statuses_cli

The comment for the part of query keeping statuses which have replies was
incorrect, explaining the part of the query kept favourited statuses instead

* Keep statuses favourited by local users in tootctl statuses remove

* Do delete old replies that are not direct interactions with local users

* Skip reblogs of local statuses instead of all reblogs

* Optimize some queries

* Keep old statuses with recent reblogs

* Add option to clear toots from followed accounts too

1 files changed, 21 insertions, 11 deletions

diff --git a/lib/mastodon/statuses_cli.rb b/lib/mastodon/statuses_cli.rb
index 4d9af0a54..74f15de5f 100644
--- a/lib/mastodon/statuses_cli.rb
+++ b/lib/mastodon/statuses_cli.rb
@@ -13,6 +13,7 @@ module Mastodon
     end
 
     option :days, type: :numeric, default: 90
+    option :clean_followed, type: :boolean
     desc 'remove', 'Remove unreferenced statuses'
     long_desc <<~LONG_DESC
       Remove statuses that are not referenced by local user activity, such as
@@ -34,17 +35,26 @@ module Mastodon
 
       say('Beginning removal... This might take a while...')
 
-      Status.remote
-            .where('id < ?', max_id)
-            .where(reblog_of_id: nil)                                                                                                                                                                                              # Skip reblogs
-            .where(in_reply_to_id: nil)                                                                                                                                                                                            # Skip replies
-            .where('id NOT IN (SELECT status_pins.status_id FROM status_pins WHERE statuses.id = status_id)')                                                                                                                      # Skip statuses that are pinned on profiles
-            .where('id NOT IN (SELECT mentions.status_id FROM mentions WHERE statuses.id = mentions.status_id AND mentions.account_id IN (SELECT accounts.id FROM accounts WHERE domain IS NULL))')                                # Skip statuses that mention local accounts
-            .where('id NOT IN (SELECT statuses1.in_reply_to_id FROM statuses AS statuses1 WHERE statuses.id = statuses1.in_reply_to_id)')                                                                                          # Skip statuses favourited by local accounts
-            .where('id NOT IN (SELECT statuses1.reblog_of_id FROM statuses AS statuses1 WHERE statuses.id = statuses1.reblog_of_id AND statuses1.account_id IN (SELECT accounts.id FROM accounts WHERE accounts.domain IS NULL))') # Skip statuses reblogged by local accounts
-            .where('account_id NOT IN (SELECT follows.target_account_id FROM follows WHERE statuses.account_id = follows.target_account_id)')                                                                                      # Skip accounts followed by local accounts
-            .in_batches
-            .delete_all
+      scope = Status.remote.where('id < ?', max_id)
+      # Skip reblogs of local statuses
+      scope = scope.where('reblog_of_id NOT IN (SELECT statuses1.id FROM statuses AS statuses1 WHERE statuses1.id = statuses.reblog_of_id AND (statuses1.uri IS NULL OR statuses1.local))')
+      # Skip statuses that are pinned on profiles
+      scope = scope.where('id NOT IN (SELECT status_pins.status_id FROM status_pins WHERE statuses.id = status_id)')
+      # Skip statuses that mention local accounts
+      scope = scope.where('id NOT IN (SELECT mentions.status_id FROM mentions WHERE statuses.id = mentions.status_id AND mentions.account_id IN (SELECT accounts.id FROM accounts WHERE domain IS NULL))')
+      # Skip statuses which have replies
+      scope = scope.where('id NOT IN (SELECT statuses1.in_reply_to_id FROM statuses AS statuses1 WHERE statuses.id = statuses1.in_reply_to_id)')
+      # Skip statuses reblogged by local accounts or with recent boosts
+      scope = scope.where('id NOT IN (SELECT statuses1.reblog_of_id FROM statuses AS statuses1 WHERE statuses.id = statuses1.reblog_of_id AND (statuses1.uri IS NULL OR statuses1.local OR statuses1.id >= ?))', max_id)
+      # Skip statuses favourited by local users
+      scope = scope.where('id NOT IN (SELECT favourites.status_id FROM favourites WHERE statuses.id = favourites.status_id AND favourites.account_id IN (SELECT accounts.id FROM accounts WHERE domain IS NULL))')
+
+      unless options[:clean_followed]
+        # Skip accounts followed by local accounts
+        scope = scope.where('account_id NOT IN (SELECT follows.target_account_id FROM follows WHERE statuses.account_id = follows.target_account_id)')
+      end
+
+      scope.in_batches.delete_all
 
       say('Beginning removal of now-orphaned media attachments to free up disk space...')