2016-11-16 02:56:29 +11:00
|
|
|
# frozen_string_literal: true
|
|
|
|
|
2016-02-24 22:57:29 +11:00
|
|
|
class ProcessFeedService < BaseService
|
2016-02-21 08:53:20 +11:00
|
|
|
def call(body, account)
|
|
|
|
xml = Nokogiri::XML(body)
|
2016-11-14 05:12:40 +11:00
|
|
|
xml.encoding = 'utf-8'
|
2016-11-08 11:32:34 +11:00
|
|
|
|
2017-04-08 21:26:03 +10:00
|
|
|
update_author(body, account)
|
2016-11-08 11:32:34 +11:00
|
|
|
process_entries(xml, account)
|
2016-03-25 12:13:30 +11:00
|
|
|
end
|
2016-02-21 08:53:20 +11:00
|
|
|
|
2016-03-25 12:13:30 +11:00
|
|
|
private
|
2016-02-29 00:26:26 +11:00
|
|
|
|
2017-04-08 21:26:03 +10:00
|
|
|
def update_author(body, account)
|
2017-04-06 05:41:50 +10:00
|
|
|
RemoteProfileUpdateWorker.perform_async(account.id, body.force_encoding('UTF-8'), true)
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2016-11-08 11:32:34 +11:00
|
|
|
def process_entries(xml, account)
|
2016-12-01 07:32:11 +11:00
|
|
|
xml.xpath('//xmlns:entry', xmlns: TagManager::XMLNS).reverse_each.map { |entry| ProcessEntry.new.call(entry, account) }.compact
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-03-16 20:46:15 +11:00
|
|
|
|
2016-11-08 11:32:34 +11:00
|
|
|
class ProcessEntry
|
|
|
|
def call(xml, account)
|
|
|
|
@account = account
|
2017-07-11 21:37:05 +10:00
|
|
|
@fetched = Activity.new(xml)
|
|
|
|
|
|
|
|
return unless [:activity, :note, :comment].include?(@fetched.type)
|
|
|
|
|
|
|
|
klass = case @fetched.verb
|
|
|
|
when :post
|
|
|
|
PostActivity
|
|
|
|
when :share
|
|
|
|
ShareActivity
|
|
|
|
when :delete
|
|
|
|
DeletionActivity
|
|
|
|
else
|
|
|
|
return
|
|
|
|
end
|
|
|
|
|
|
|
|
@fetched = klass.new(xml, account)
|
|
|
|
@fetched.perform
|
2016-11-19 09:19:38 +11:00
|
|
|
rescue ActiveRecord::RecordInvalid => e
|
|
|
|
Rails.logger.debug "Nothing was saved for #{id} because: #{e}"
|
|
|
|
nil
|
2016-03-25 12:13:30 +11:00
|
|
|
end
|
2016-02-25 10:17:01 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class Activity
|
|
|
|
def initialize(xml, account = nil)
|
|
|
|
@xml = xml
|
|
|
|
@account = account
|
2017-05-04 12:34:57 +10:00
|
|
|
end
|
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def verb
|
|
|
|
raw = @xml.at_xpath('./activity:verb', activity: TagManager::AS_XMLNS).content
|
|
|
|
TagManager::VERBS.key(raw)
|
|
|
|
rescue
|
|
|
|
:post
|
|
|
|
end
|
2017-05-23 03:35:48 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def type
|
|
|
|
raw = @xml.at_xpath('./activity:object-type', activity: TagManager::AS_XMLNS).content
|
|
|
|
TagManager::TYPES.key(raw)
|
|
|
|
rescue
|
|
|
|
:activity
|
|
|
|
end
|
2016-11-06 01:20:05 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def id
|
|
|
|
@xml.at_xpath('./xmlns:id', xmlns: TagManager::XMLNS).content
|
2017-06-07 20:28:16 +10:00
|
|
|
end
|
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def url
|
|
|
|
link = @xml.at_xpath('./xmlns:link[@rel="alternate"]', xmlns: TagManager::XMLNS)
|
|
|
|
link.nil? ? nil : link['href']
|
|
|
|
end
|
2017-05-23 03:35:48 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
private
|
2017-01-21 04:31:49 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def find_status(uri)
|
|
|
|
if TagManager.instance.local_id?(uri)
|
|
|
|
local_id = TagManager.instance.unique_tag_to_local_id(uri, 'Status')
|
|
|
|
return Status.find_by(id: local_id)
|
2016-11-09 05:37:08 +11:00
|
|
|
end
|
2016-09-10 04:04:34 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
Status.find_by(uri: uri)
|
2017-05-23 03:35:48 +10:00
|
|
|
end
|
2016-11-21 19:56:01 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def redis
|
|
|
|
Redis.current
|
2017-06-06 08:09:14 +10:00
|
|
|
end
|
2017-07-11 21:37:05 +10:00
|
|
|
end
|
2017-06-06 08:09:14 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class CreationActivity < Activity
|
|
|
|
def perform
|
|
|
|
if redis.exists("delete_upon_arrival:#{@account.id}:#{id}")
|
|
|
|
Rails.logger.debug "Delete for status #{id} was queued, ignoring"
|
|
|
|
return
|
|
|
|
end
|
2017-05-17 08:41:15 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
Rails.logger.debug "Creating remote status #{id}"
|
|
|
|
# Return early if status already exists in db
|
|
|
|
status = find_status(id)
|
|
|
|
|
|
|
|
return [status, false] unless status.nil?
|
|
|
|
|
|
|
|
return [nil, false] if @account.suspended?
|
|
|
|
|
|
|
|
status = Status.create!(
|
|
|
|
uri: id,
|
|
|
|
url: url,
|
|
|
|
account: @account,
|
|
|
|
reblog: reblog,
|
|
|
|
text: content,
|
|
|
|
spoiler_text: content_warning,
|
|
|
|
created_at: published,
|
|
|
|
reply: thread?,
|
|
|
|
language: content_language,
|
|
|
|
visibility: visibility_scope,
|
|
|
|
conversation: converstation_to_persistent&.first,
|
|
|
|
thread: thread? ? find_status(thread.first) : nil
|
|
|
|
)
|
|
|
|
|
|
|
|
save_mentions(status)
|
|
|
|
save_hashtags(status)
|
|
|
|
save_media(status)
|
|
|
|
|
|
|
|
if thread? && status.thread.nil?
|
|
|
|
Rails.logger.debug "Trying to attach #{status.id} (#{id}) to #{thread.first}"
|
|
|
|
ThreadResolveWorker.perform_async(status.id, thread.second)
|
|
|
|
end
|
2017-05-17 08:41:15 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
Rails.logger.debug "Queuing remote status #{status.id} (#{id}) for distribution"
|
2017-05-17 08:41:15 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
LinkCrawlWorker.perform_async(status.id) unless status.spoiler_text?
|
|
|
|
DistributionWorker.perform_async(status.id)
|
2016-02-29 07:22:56 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
[status, true]
|
2017-03-14 02:34:15 +11:00
|
|
|
end
|
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def content
|
|
|
|
@xml.at_xpath('./xmlns:content', xmlns: TagManager::XMLNS).content
|
|
|
|
end
|
2017-03-14 02:34:15 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def content_language
|
|
|
|
@xml.at_xpath('./xmlns:content', xmlns: TagManager::XMLNS)['xml:lang']&.presence || 'en'
|
|
|
|
end
|
2017-05-04 12:34:57 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def content_warning
|
|
|
|
@xml.at_xpath('./xmlns:summary', xmlns: TagManager::XMLNS)&.content || ''
|
2017-05-04 12:34:57 +10:00
|
|
|
end
|
2016-10-15 05:14:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def visibility_scope
|
|
|
|
@xml.at_xpath('./mastodon:scope', mastodon: TagManager::MTDN_XMLNS)&.content&.to_sym || :public
|
|
|
|
end
|
2016-02-29 07:22:56 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def published
|
|
|
|
@xml.at_xpath('./xmlns:published', xmlns: TagManager::XMLNS).content
|
|
|
|
end
|
2017-04-28 01:06:47 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def thread?
|
|
|
|
!@xml.at_xpath('./thr:in-reply-to', thr: TagManager::THR_XMLNS).nil?
|
|
|
|
end
|
2017-04-28 01:06:47 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def thread
|
|
|
|
thr = @xml.at_xpath('./thr:in-reply-to', thr: TagManager::THR_XMLNS)
|
|
|
|
[thr['ref'], thr['href']]
|
|
|
|
end
|
2017-04-28 01:06:47 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
private
|
2017-01-28 02:55:06 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def converstation_to_persistent
|
|
|
|
uri = @xml.at_xpath('./ostatus:conversation', ostatus: TagManager::OS_XMLNS)&.attribute('ref')&.content
|
|
|
|
return if uri.nil?
|
2016-11-08 11:32:34 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
if TagManager.instance.local_id?(uri)
|
|
|
|
local_id = TagManager.instance.unique_tag_to_local_id(uri, 'Conversation')
|
|
|
|
return Conversation.find_by(id: local_id)
|
|
|
|
end
|
2016-11-09 05:37:08 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
found = Conversation.find_by(uri: uri)
|
|
|
|
found ? [found, false] : [Conversation.create!(uri: uri), true]
|
|
|
|
end
|
2016-12-06 08:59:30 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def save_mentions(parent)
|
|
|
|
processed_account_ids = []
|
2016-11-08 11:32:34 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
@xml.xpath('./xmlns:link[@rel="mentioned"]', xmlns: TagManager::XMLNS).each do |link|
|
|
|
|
next if [TagManager::TYPES[:group], TagManager::TYPES[:collection]].include? link['ostatus:object-type']
|
2016-09-30 05:28:21 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
mentioned_account = account_from_href(link['href'])
|
2016-02-29 07:22:56 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
next if mentioned_account.nil? || processed_account_ids.include?(mentioned_account.id)
|
2017-05-13 03:09:21 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
mentioned_account.mentions.where(status: parent).first_or_create(status: parent)
|
2017-05-13 03:09:21 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
# So we can skip duplicate mentions
|
|
|
|
processed_account_ids << mentioned_account.id
|
|
|
|
end
|
|
|
|
end
|
2017-05-13 03:09:21 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def save_hashtags(parent)
|
|
|
|
tags = @xml.xpath('./xmlns:category', xmlns: TagManager::XMLNS).map { |category| category['term'] }.select(&:present?)
|
|
|
|
ProcessHashtagsService.new.call(parent, tags)
|
2016-02-29 07:22:56 +11:00
|
|
|
end
|
2016-11-08 11:32:34 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def save_media(parent)
|
|
|
|
do_not_download = DomainBlock.find_by(domain: parent.account.domain)&.reject_media?
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
@xml.xpath('./xmlns:link[@rel="enclosure"]', xmlns: TagManager::XMLNS).each do |link|
|
|
|
|
next unless link['href']
|
2016-10-15 05:14:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
media = MediaAttachment.where(status: parent, remote_url: link['href']).first_or_initialize(account: parent.account, status: parent, remote_url: link['href'])
|
|
|
|
parsed_url = Addressable::URI.parse(link['href']).normalize
|
2016-09-06 02:39:53 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
next if !%w(http https).include?(parsed_url.scheme) || parsed_url.host.empty?
|
2016-09-30 05:28:21 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
media.save
|
2016-09-06 02:39:53 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
next if do_not_download
|
2016-02-25 03:23:59 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
begin
|
|
|
|
media.file_remote_url = link['href']
|
|
|
|
media.save!
|
|
|
|
rescue ActiveRecord::RecordInvalid
|
|
|
|
next
|
|
|
|
end
|
|
|
|
end
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def account_from_href(href)
|
|
|
|
url = Addressable::URI.parse(href).normalize
|
2017-04-17 02:01:48 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
if TagManager.instance.web_domain?(url.host)
|
|
|
|
Account.find_local(url.path.gsub('/users/', ''))
|
|
|
|
else
|
|
|
|
Account.where(uri: href).or(Account.where(url: href)).first || FetchRemoteAccountService.new.call(href)
|
|
|
|
end
|
2017-04-17 02:01:48 +10:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class ShareActivity < CreationActivity
|
|
|
|
def perform
|
|
|
|
status, just_created = super
|
|
|
|
NotifyService.new.call(reblog.account, status) if reblog&.account&.local? && just_created
|
|
|
|
status
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-02-25 03:23:59 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def object
|
|
|
|
@xml.at_xpath('.//activity:object', activity: TagManager::AS_XMLNS)
|
|
|
|
end
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
private
|
Fix #24 - Thread resolving for remote statuses
This is a big one, so let me enumerate:
Accounts as well as stream entry pages now contain Link headers that
reference the Atom feed and Webfinger URL for the former and Atom entry
for the latter. So you only need to HEAD those resources to get that
information, no need to download and parse HTML <link>s.
ProcessFeedService will now queue ThreadResolveWorker for each remote
status that it cannot find otherwise. Furthermore, entries are now
processed in reverse order (from bottom to top) in case a newer entry
references a chronologically previous one.
ThreadResolveWorker uses FetchRemoteStatusService to obtain a status
and attach the child status it was queued for to it.
FetchRemoteStatusService looks up the URL, first with a HEAD, tests
if it's an Atom feed, in which case it processes it directly. Next
for Link headers to the Atom feed, in which case that is fetched
and processed. Lastly if it's HTML, it is checked for <link>s to the Atom
feed, and if such is found, that is fetched and processed. The account for
the status is derived from author/name attribute in the XML and the hostname
in the URL (domain). FollowRemoteAccountService and ProcessFeedService
are used.
This means that potentially threads are resolved recursively until a dead-end
is encountered, however it is performed asynchronously over background jobs,
so it should be ok.
2016-09-21 09:34:14 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def status
|
|
|
|
reblog && super
|
|
|
|
end
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def reblog
|
|
|
|
return @reblog if defined? @reblog
|
2016-02-29 07:22:56 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
original_status = RemoteActivity.new(object).perform
|
|
|
|
@reblog = original_status.reblog? ? original_status.reblog : original_status
|
|
|
|
end
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-02-25 03:23:59 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class PostActivity < CreationActivity
|
|
|
|
def perform
|
|
|
|
status, just_created = super
|
2017-04-17 04:32:17 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
if just_created
|
|
|
|
status.mentions.includes(:account).each do |mention|
|
|
|
|
mentioned_account = mention.account
|
|
|
|
next unless mentioned_account.local?
|
|
|
|
NotifyService.new.call(mentioned_account, mention)
|
|
|
|
end
|
|
|
|
end
|
2017-01-25 10:49:08 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
status
|
|
|
|
end
|
2017-02-12 01:10:22 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
private
|
2016-02-25 03:23:59 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
def reblog
|
|
|
|
nil
|
|
|
|
end
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2016-02-24 11:28:53 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class DeletionActivity < Activity
|
|
|
|
def perform
|
|
|
|
Rails.logger.debug "Deleting remote status #{id}"
|
|
|
|
status = Status.find_by(uri: id, account: @account)
|
2016-02-24 13:05:40 +11:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
if status.nil?
|
|
|
|
redis.setex("delete_upon_arrival:#{@account.id}:#{id}", 6 * 3_600, id)
|
|
|
|
else
|
|
|
|
RemoveStatusService.new.call(status)
|
|
|
|
end
|
|
|
|
end
|
2016-11-08 11:32:34 +11:00
|
|
|
end
|
2017-05-04 12:34:57 +10:00
|
|
|
|
2017-07-11 21:37:05 +10:00
|
|
|
class RemoteActivity < Activity
|
|
|
|
def perform
|
|
|
|
find_status(id) || FetchRemoteStatusService.new.call(url)
|
|
|
|
end
|
2017-05-04 12:34:57 +10:00
|
|
|
end
|
2016-09-20 08:39:03 +10:00
|
|
|
end
|
2016-02-21 08:53:20 +11:00
|
|
|
end
|