mirror of
https://github.com/glitch-soc/mastodon.git
synced 2024-11-28 02:50:14 -05:00
40e5d2303b
to_s method of HTTP::Response keeps blocking while it receives the whole content, no matter how it is big. This means it may waste time to receive unacceptably large files. It may also consume memory and disk in the process. This solves the inefficency by checking response length while receiving.
90 lines
3.1 KiB
Ruby
90 lines
3.1 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
class FetchAtomService < BaseService
|
|
include JsonLdHelper
|
|
|
|
def call(url)
|
|
return if url.blank?
|
|
|
|
result = process(url)
|
|
|
|
# retry without ActivityPub
|
|
result ||= process(url) if @unsupported_activity
|
|
|
|
result
|
|
rescue OpenSSL::SSL::SSLError => e
|
|
Rails.logger.debug "SSL error: #{e}"
|
|
nil
|
|
rescue HTTP::ConnectionError => e
|
|
Rails.logger.debug "HTTP ConnectionError: #{e}"
|
|
nil
|
|
end
|
|
|
|
private
|
|
|
|
def process(url, terminal = false)
|
|
@url = url
|
|
perform_request { |response| process_response(response, terminal) }
|
|
end
|
|
|
|
def perform_request(&block)
|
|
accept = 'text/html'
|
|
accept = 'application/activity+json, application/ld+json, application/atom+xml, ' + accept unless @unsupported_activity
|
|
|
|
Request.new(:get, @url).add_headers('Accept' => accept).perform(&block)
|
|
end
|
|
|
|
def process_response(response, terminal = false)
|
|
return nil if response.code != 200
|
|
|
|
if response.mime_type == 'application/atom+xml'
|
|
[@url, { prefetched_body: response.body_with_limit }, :ostatus]
|
|
elsif ['application/activity+json', 'application/ld+json; profile="https://www.w3.org/ns/activitystreams"'].include?(response.mime_type)
|
|
body = response.body_with_limit
|
|
json = body_to_json(body)
|
|
if supported_context?(json) && json['type'] == 'Person' && json['inbox'].present?
|
|
[json['id'], { prefetched_body: body, id: true }, :activitypub]
|
|
elsif supported_context?(json) && json['type'] == 'Note'
|
|
[json['id'], { prefetched_body: body, id: true }, :activitypub]
|
|
else
|
|
@unsupported_activity = true
|
|
nil
|
|
end
|
|
elsif !terminal
|
|
link_header = response['Link'] && parse_link_header(response)
|
|
|
|
if link_header&.find_link(%w(rel alternate))
|
|
process_link_headers(link_header)
|
|
elsif response.mime_type == 'text/html'
|
|
process_html(response)
|
|
end
|
|
end
|
|
end
|
|
|
|
def process_html(response)
|
|
page = Nokogiri::HTML(response.body_with_limit)
|
|
|
|
json_link = page.xpath('//link[@rel="alternate"]').find { |link| ['application/activity+json', 'application/ld+json; profile="https://www.w3.org/ns/activitystreams"'].include?(link['type']) }
|
|
atom_link = page.xpath('//link[@rel="alternate"]').find { |link| link['type'] == 'application/atom+xml' }
|
|
|
|
result ||= process(json_link['href'], terminal: true) unless json_link.nil? || @unsupported_activity
|
|
result ||= process(atom_link['href'], terminal: true) unless atom_link.nil?
|
|
|
|
result
|
|
end
|
|
|
|
def process_link_headers(link_header)
|
|
json_link = link_header.find_link(%w(rel alternate), %w(type application/activity+json)) || link_header.find_link(%w(rel alternate), ['type', 'application/ld+json; profile="https://www.w3.org/ns/activitystreams"'])
|
|
atom_link = link_header.find_link(%w(rel alternate), %w(type application/atom+xml))
|
|
|
|
result ||= process(json_link.href, terminal: true) unless json_link.nil? || @unsupported_activity
|
|
result ||= process(atom_link.href, terminal: true) unless atom_link.nil?
|
|
|
|
result
|
|
end
|
|
|
|
def parse_link_header(response)
|
|
LinkHeader.parse(response['Link'].is_a?(Array) ? response['Link'].first : response['Link'])
|
|
end
|
|
end
|