From 42c4427cb15a54ca4e7703c41a4521be19beb937 Mon Sep 17 00:00:00 2001
From: Ted Johansson <ted@discourse.org>
Date: Tue, 4 Mar 2025 17:29:38 +0800
Subject: [PATCH] DEV: Add rake task to bulk delete posts (#31576)

This PR adds a destroy:posts rake task that can be used to hard-delete a list of posts. Useful for dealing with large amounts of spam that has been soft deleted and needs to go.

Notes:

Works on both non-deleted and soft-deleted posts. (We might want to change this to work on only soft-deleted posts?)
Works exclusively on post IDs. We can't mix topic and post IDs as they might clash, and we have no way of resolving that ambiguity.
Accepts either a rake-style array of IDs or, more conveniently, you can pipe the argument in through STDIN.
Added a confirmation step since it's a fairly destructive operation.
---
 app/services/destroy_task.rb        | 28 ++++++++++++++++++++++++++++
 config/locales/server.en.yml        |  1 +
 lib/tasks/destroy.rake              | 13 +++++++++++++
 spec/lib/tasks/destroy_rake_spec.rb | 21 +++++++++++++++++++++
 spec/services/destroy_task_spec.rb  | 24 ++++++++++++++++++++++++
 spec/support/rake_context.rb        | 26 ++++++++++++++++++++++++++
 6 files changed, 113 insertions(+)
 create mode 100644 spec/lib/tasks/destroy_rake_spec.rb
 create mode 100644 spec/support/rake_context.rb

diff --git a/app/services/destroy_task.rb b/app/services/destroy_task.rb
index fda78c9583d..1b43115cc6d 100644
--- a/app/services/destroy_task.rb
+++ b/app/services/destroy_task.rb
@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require "highline/import"
+
 class DestroyTask
   def initialize(io = STDOUT)
     @io = io
@@ -46,6 +48,32 @@ class DestroyTask
     categories.each { |c| @io.puts destroy_topics(c.slug, c.parent_category&.slug) }
   end
 
+  def destroy_posts(post_ids, require_confirmation: true)
+    posts = Post.with_deleted.where(id: post_ids)
+
+    @io.puts "There are #{posts.count} posts to delete"
+
+    if posts.count < post_ids.size
+      @io.puts "Couldn't find the following posts:"
+      @io.puts "  #{post_ids.map(&:to_i) - posts.pluck(:id)}"
+    end
+
+    if require_confirmation
+      confirm_destroy = ask("Are you sure? (Y/n)")
+      exit 1 if confirm_destroy.downcase != "y"
+    end
+
+    posts.find_each do |post|
+      @io.puts "Destroying post #{post.id}"
+      @io.puts PostDestroyer.new(
+                 Discourse.system_user,
+                 post,
+                 context: I18n.t("staff_action_logs.cli_bulk_post_delete"),
+                 force_destroy: true,
+               ).destroy
+    end
+  end
+
   def destroy_private_messages
     Topic
       .where(archetype: "private_message")
diff --git a/config/locales/server.en.yml b/config/locales/server.en.yml
index e7e6d2e209b..5975c6a081b 100644
--- a/config/locales/server.en.yml
+++ b/config/locales/server.en.yml
@@ -5560,6 +5560,7 @@ en:
       revoked: Revoked
       restored: Restored
     bulk_user_delete: "deleted in a bulk delete operation"
+    cli_bulk_post_delete: "Bulk deleted from rake task"
 
   reviewables:
     already_handled: "Thanks, but we've already reviewed that post and determined it does not need to be flagged again."
diff --git a/lib/tasks/destroy.rake b/lib/tasks/destroy.rake
index 1708c701eee..5a6d9d6c227 100644
--- a/lib/tasks/destroy.rake
+++ b/lib/tasks/destroy.rake
@@ -50,3 +50,16 @@ task "destroy:categories" => :environment do |t, args|
   puts "Going to delete these categories: #{categories}"
   categories.each { |id| destroy_task.destroy_category(id, true) }
 end
+
+# Hard delete a list of posts by ID. Pass a comma
+# separated list either as a rake argument or through
+# STDIN, e.g. through a pipe.
+#
+#   Example: rake destroy:posts[4,8,15,16,23,42]
+#   Example: cat post_ids.txt | rake destroy:posts
+desc "Destroy a list of posts given by ID"
+task "destroy:posts" => :environment do |_task, args|
+  post_ids = args.extras.empty? ? STDIN.read.strip.split(",") : args.extras
+  puts "Going to delete these posts: #{post_ids}"
+  DestroyTask.new.destroy_posts(post_ids)
+end
diff --git a/spec/lib/tasks/destroy_rake_spec.rb b/spec/lib/tasks/destroy_rake_spec.rb
new file mode 100644
index 00000000000..e07999661d2
--- /dev/null
+++ b/spec/lib/tasks/destroy_rake_spec.rb
@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+describe "destroy:posts" do # rubocop:disable RSpec/DescribeClass
+  subject(:task) { subject }
+
+  include_context "in a rake task"
+
+  # No console output in test suite, thanks.
+  before { STDOUT.stubs(:puts) }
+
+  it "accepts a list of post IDs piped through STDIN" do
+    destroy_task = instance_spy(DestroyTask)
+    DestroyTask.stubs(:new).returns(destroy_task)
+
+    STDIN.stubs(:read).returns("1,2,3\n")
+
+    task.invoke
+
+    expect(destroy_task).to have_received(:destroy_posts).with(%w[1 2 3])
+  end
+end
diff --git a/spec/services/destroy_task_spec.rb b/spec/services/destroy_task_spec.rb
index e4fccb804d8..140685c5c63 100644
--- a/spec/services/destroy_task_spec.rb
+++ b/spec/services/destroy_task_spec.rb
@@ -67,6 +67,30 @@ RSpec.describe DestroyTask do
     end
   end
 
+  describe "#destroy_posts" do
+    let(:task) { DestroyTask.new(StringIO.new) }
+
+    let!(:t1) { Fabricate(:topic) }
+    let!(:t2) { Fabricate(:topic) }
+
+    let!(:p1) { Fabricate(:post, topic: t1) }
+    let!(:p2) { Fabricate(:post, topic: t1) }
+    let!(:p3) { Fabricate(:post, topic: t2) }
+
+    before { p2.trash! }
+
+    it "destroys posts listed and creates staff action logs" do
+      expect { task.destroy_posts([p2.id, p3.id], require_confirmation: false) }.to change {
+        Post.with_deleted.count
+      }.by(-2).and change { UserHistory.pluck(:action) }.from([]).to(
+              [
+                UserHistory.actions[:delete_post_permanently],
+                UserHistory.actions[:delete_topic_permanently],
+              ],
+            )
+    end
+  end
+
   describe "private messages" do
     let!(:pm) { Fabricate(:private_message_post) }
     let!(:pm2) { Fabricate(:private_message_post) }
diff --git a/spec/support/rake_context.rb b/spec/support/rake_context.rb
new file mode 100644
index 00000000000..ef4dd5ae428
--- /dev/null
+++ b/spec/support/rake_context.rb
@@ -0,0 +1,26 @@
+# frozen_string_literal: true
+
+require "rake"
+
+shared_context "in a rake task" do
+  subject { rake[task_name] }
+
+  let(:rake) { Rake::Application.new }
+  let(:task_name) { self.class.top_level_description }
+  let(:task_path) { "lib/tasks/#{task_name.split(":").first}" }
+
+  def loaded_files_excluding_current_rake_file
+    $".reject { |file| file == Rails.root.join("#{task_path}.rake").to_s }
+  end
+
+  before do
+    Rake.application = rake
+    Rake.application.rake_require(
+      task_path,
+      [Rails.root.to_s],
+      loaded_files_excluding_current_rake_file,
+    )
+
+    Rake::Task.define_task(:environment)
+  end
+end