discourse/app/jobs/scheduled/clean_up_uploads.rb

# frozen_string_literal: true

module Jobs
  class CleanUpUploads < ::Jobs::Scheduled
    every 1.hour

    def execute(args)
      grace_period = [SiteSetting.clean_orphan_uploads_grace_period_hours, 1].max

      # always remove invalid upload records
      Upload
        .by_users
        .where("retain_hours IS NULL OR created_at < current_timestamp - interval '1 hour' * retain_hours")
        .where("created_at < ?", grace_period.hour.ago)
        .where(url: "")
        .find_each(&:destroy!)

      return unless SiteSetting.clean_up_uploads?

      if c = last_cleanup
        return if (Time.zone.now.to_i - c) < (grace_period / 2).hours
      end

      base_url = Discourse.store.internal? ? Discourse.store.relative_base_url : Discourse.store.absolute_base_url
      s3_hostname = URI.parse(base_url).hostname
      s3_cdn_hostname = URI.parse(SiteSetting.Upload.s3_cdn_url || "").hostname

      result = Upload.by_users
        .where("uploads.retain_hours IS NULL OR uploads.created_at < current_timestamp - interval '1 hour' * uploads.retain_hours")
        .where("uploads.created_at < ?", grace_period.hour.ago)
        .where("uploads.access_control_post_id IS NULL")
        .joins(<<~SQL)
          LEFT JOIN site_settings ss
          ON NULLIF(ss.value, '')::integer = uploads.id
          AND ss.data_type = #{SiteSettings::TypeSupervisor.types[:upload].to_i}
        SQL
        .joins("LEFT JOIN post_uploads pu ON pu.upload_id = uploads.id")
        .joins("LEFT JOIN users u ON u.uploaded_avatar_id = uploads.id")
        .joins("LEFT JOIN user_avatars ua ON ua.gravatar_upload_id = uploads.id OR ua.custom_upload_id = uploads.id")
        .joins("LEFT JOIN user_profiles up ON up.profile_background_upload_id = uploads.id OR up.card_background_upload_id = uploads.id")
        .joins("LEFT JOIN categories c ON c.uploaded_logo_id = uploads.id OR c.uploaded_background_id = uploads.id")
        .joins("LEFT JOIN custom_emojis ce ON ce.upload_id = uploads.id")
        .joins("LEFT JOIN theme_fields tf ON tf.upload_id = uploads.id")
        .joins("LEFT JOIN user_exports ue ON ue.upload_id = uploads.id")
        .joins("LEFT JOIN groups g ON g.flair_upload_id = uploads.id")
        .joins("LEFT JOIN badges b ON b.image_upload_id = uploads.id")
        .where("pu.upload_id IS NULL")
        .where("u.uploaded_avatar_id IS NULL")
        .where("ua.gravatar_upload_id IS NULL AND ua.custom_upload_id IS NULL")
        .where("up.profile_background_upload_id IS NULL AND up.card_background_upload_id IS NULL")
        .where("c.uploaded_logo_id IS NULL AND c.uploaded_background_id IS NULL")
        .where("ce.upload_id IS NULL")
        .where("tf.upload_id IS NULL")
        .where("ue.upload_id IS NULL")
        .where("g.flair_upload_id IS NULL")
        .where("b.image_upload_id IS NULL")
        .where("ss.value IS NULL")

      if SiteSetting.selectable_avatars.present?
        result = result.where.not(id: SiteSetting.selectable_avatars.map(&:id))
      end

      result.find_each do |upload|
        if upload.sha1.present?
          encoded_sha = Base62.encode(upload.sha1.hex)
          next if ReviewableQueuedPost.pending.where("payload->>'raw' LIKE '%#{upload.sha1}%' OR payload->>'raw' LIKE '%#{encoded_sha}%'").exists?
          next if Draft.where("data LIKE '%#{upload.sha1}%' OR data LIKE '%#{encoded_sha}%'").exists?
          next if ThemeSetting.where(data_type: ThemeSetting.types[:upload]).where("value LIKE ?", "%#{upload.sha1}%").exists?
          upload.destroy
        else
          upload.delete
        end
      end

      self.last_cleanup = Time.zone.now.to_i
    end

    def last_cleanup=(v)
      Discourse.redis.setex(last_cleanup_key, 7.days.to_i, v.to_s)
    end

    def last_cleanup
      v = Discourse.redis.get(last_cleanup_key)
      v ? v.to_i : v
    end

    def reset_last_cleanup!
      Discourse.redis.del(last_cleanup_key)
    end

    protected

    def last_cleanup_key
      "LAST_UPLOAD_CLEANUP"
    end

  end
end