discourse/app/jobs/regular/process_post.rb
Daniel Waterworth 76bb01c31c DEV: Increase process_post lock timeout
The process_post job uses CookedPostProcessor which also uses a
DistributedMutex. There's no good reason for the timeout of the outer
lock to be smaller than the timeout of the inner lock.
2020-04-03 13:34:44 +01:00

61 lines
2.0 KiB
Ruby

# frozen_string_literal: true
require 'image_sizer'
module Jobs
class ProcessPost < ::Jobs::Base
def execute(args)
DistributedMutex.synchronize("process_post_#{args[:post_id]}", validity: 10.minutes) do
post = Post.find_by(id: args[:post_id])
# two levels of deletion
return unless post.present? && post.topic.present?
orig_cooked = post.cooked
recooked = nil
if args[:cook].present?
cooking_options = args[:cooking_options] || {}
cooking_options[:topic_id] = post.topic_id
recooked = post.cook(post.raw, cooking_options.symbolize_keys)
post.update_columns(cooked: recooked, baked_at: Time.zone.now, baked_version: Post::BAKED_VERSION)
end
cp = CookedPostProcessor.new(post, args)
cp.post_process(bypass_bump: args[:bypass_bump], new_post: args[:new_post])
# If we changed the document, save it
cooked = cp.html
if cooked != (recooked || orig_cooked)
if orig_cooked.present? && cooked.blank?
# TODO stop/restart the worker if needed, let's gather a few here first
Rails.logger.warn("Cooked post processor in FATAL state, bypassing. You need to urgently restart sidekiq\norig: #{orig_cooked}\nrecooked: #{recooked}\ncooked: #{cooked}\npost id: #{post.id}")
else
post.update_column(:cooked, cp.html)
extract_links(post)
post.publish_change_to_clients! :revised
end
end
if !post.user&.staff? && !post.user&.staged?
s = post.cooked
s << " #{post.topic.title}" if post.post_number == 1
if !args[:bypass_bump] && WordWatcher.new(s).should_flag?
PostActionCreator.create(Discourse.system_user, post, :inappropriate)
end
end
end
end
# onebox may have added some links, so extract them now
def extract_links(post)
TopicLink.extract_from(post)
QuotedPost.extract_from(post)
end
end
end