Fix resolving accounts sometimes creating duplicate records for a given AP id (#15364)

* Fix ResolveAccountService accepting mismatching acct: URI

* Set attributes that should be updated regardless of suspension

* Fix key fetching

* Automatically merge remote accounts with duplicate `uri`

* Add tests

* Add "tootctl accounts fix-duplicates"

Finds duplicate accounts sharing a same ActivityPub `id`, re-fetch them and
merge them under the canonical `acct:` URI.

Co-authored-by: Claire <claire.github-309c@sitedethib.com>

4 files changed, 44 insertions, 21 deletions

diff --git a/app/services/activitypub/fetch_remote_account_service.rb b/app/services/activitypub/fetch_remote_account_service.rb
index e5bd0c47c..9d01f5386 100644
--- a/app/services/activitypub/fetch_remote_account_service.rb
+++ b/app/services/activitypub/fetch_remote_account_service.rb
@@ -28,7 +28,7 @@ class ActivityPub::FetchRemoteAccountService < BaseService
 
     return unless only_key || verified_webfinger?
 
-    ActivityPub::ProcessAccountService.new.call(@username, @domain, @json, only_key: only_key)
+    ActivityPub::ProcessAccountService.new.call(@username, @domain, @json, only_key: only_key, verified_webfinger: !only_key)
   rescue Oj::ParseError
     nil
   end
diff --git a/app/services/activitypub/process_account_service.rb b/app/services/activitypub/process_account_service.rb
index 4cb8e09db..6afeb92d6 100644
--- a/app/services/activitypub/process_account_service.rb
+++ b/app/services/activitypub/process_account_service.rb
@@ -28,6 +28,8 @@ class ActivityPub::ProcessAccountService < BaseService
         update_account
         process_tags
         process_attachments
+
+        process_duplicate_accounts! if @options[:verified_webfinger]
       else
         raise Mastodon::RaceConditionError
       end
@@ -69,34 +71,42 @@ class ActivityPub::ProcessAccountService < BaseService
     @account.protocol            = :activitypub
 
     set_suspension!
+    set_immediate_protocol_attributes!
+    set_fetchable_key! unless @account.suspended? && @account.suspension_origin_local?
     set_immediate_attributes! unless @account.suspended?
-    set_fetchable_attributes! unless @options[:only_keys] || @account.suspended?
+    set_fetchable_attributes! unless @options[:only_key] || @account.suspended?
 
     @account.save_with_optional_media!
   end
 
-  def set_immediate_attributes!
+  def set_immediate_protocol_attributes!
     @account.inbox_url               = @json['inbox'] || ''
     @account.outbox_url              = @json['outbox'] || ''
     @account.shared_inbox_url        = (@json['endpoints'].is_a?(Hash) ? @json['endpoints']['sharedInbox'] : @json['sharedInbox']) || ''
     @account.followers_url           = @json['followers'] || ''
-    @account.featured_collection_url = @json['featured'] || ''
-    @account.devices_url             = @json['devices'] || ''
     @account.url                     = url || @uri
     @account.uri                     = @uri
+    @account.actor_type              = actor_type
+  end
+
+  def set_immediate_attributes!
+    @account.featured_collection_url = @json['featured'] || ''
+    @account.devices_url             = @json['devices'] || ''
     @account.display_name            = @json['name'] || ''
     @account.note                    = @json['summary'] || ''
     @account.locked                  = @json['manuallyApprovesFollowers'] || false
     @account.fields                  = property_values || {}
     @account.also_known_as           = as_array(@json['alsoKnownAs'] || []).map { |item| value_or_id(item) }
-    @account.actor_type              = actor_type
     @account.discoverable            = @json['discoverable'] || false
   end
 
+  def set_fetchable_key!
+    @account.public_key        = public_key || ''
+  end
+
   def set_fetchable_attributes!
     @account.avatar_remote_url = image_url('icon')  || '' unless skip_download?
     @account.header_remote_url = image_url('image') || '' unless skip_download?
-    @account.public_key        = public_key || ''
     @account.statuses_count    = outbox_total_items    if outbox_total_items.present?
     @account.following_count   = following_total_items if following_total_items.present?
     @account.followers_count   = followers_total_items if followers_total_items.present?
@@ -140,6 +150,12 @@ class ActivityPub::ProcessAccountService < BaseService
     VerifyAccountLinksWorker.perform_async(@account.id)
   end
 
+  def process_duplicate_accounts!
+    return unless Account.where(uri: @account.uri).where.not(id: @account.id).exists?
+
+    AccountMergingWorker.perform_async(@account.id)
+  end
+
   def actor_type
     if @json['type'].is_a?(Array)
       @json['type'].find { |type| ActivityPub::FetchRemoteAccountService::SUPPORTED_TYPES.include?(type) }
diff --git a/app/services/resolve_account_service.rb b/app/services/resolve_account_service.rb
index 74b0b82d0..3301aaf51 100644
--- a/app/services/resolve_account_service.rb
+++ b/app/services/resolve_account_service.rb
@@ -49,7 +49,7 @@ class ResolveAccountService < BaseService
     # Now it is certain, it is definitely a remote account, and it
     # either needs to be created, or updated from fresh data
 
-    process_account!
+    fetch_account!
   rescue Webfinger::Error, Oj::ParseError => e
     Rails.logger.debug "Webfinger query for #{@uri} failed: #{e}"
     nil
@@ -104,16 +104,12 @@ class ResolveAccountService < BaseService
     acct.gsub(/\Aacct:/, '').split('@')
   end
 
-  def process_account!
+  def fetch_account!
     return unless activitypub_ready?
 
     RedisLock.acquire(lock_options) do |lock|
       if lock.acquired?
-        @account = Account.find_remote(@username, @domain)
-
-        next if actor_json.nil?
-
-        @account = ActivityPub::ProcessAccountService.new.call(@username, @domain, actor_json)
+        @account = ActivityPub::FetchRemoteAccountService.new.call(actor_url)
       else
         raise Mastodon::RaceConditionError
       end
@@ -136,13 +132,6 @@ class ResolveAccountService < BaseService
     @actor_url ||= @webfinger.link('self', 'href')
   end
 
-  def actor_json
-    return @actor_json if defined?(@actor_json)
-
-    json        = fetch_resource(actor_url, false)
-    @actor_json = supported_context?(json) && equals_or_includes_any?(json['type'], ActivityPub::FetchRemoteAccountService::SUPPORTED_TYPES) ? json : nil
-  end
-
   def gone_from_origin?
     @gone
   end
diff --git a/app/workers/account_merging_worker.rb b/app/workers/account_merging_worker.rb
new file mode 100644
index 000000000..8c234e7ac
--- /dev/null
+++ b/app/workers/account_merging_worker.rb
@@ -0,0 +1,18 @@
+# frozen_string_literal: true
+
+class AccountMergingWorker
+  include Sidekiq::Worker
+
+  sidekiq_options queue: 'pull'
+
+  def perform(account_id)
+    account = Account.find(account_id)
+
+    return true if account.nil? || account.local?
+
+    Account.where(uri: account.uri).where.not(id: account.id).find_each do |duplicate|
+      account.merge_with!(duplicate)
+      duplicate.destroy
+    end
+  end
+end