From 05cd37097c134d559550adca6cd17cf7578be94b Mon Sep 17 00:00:00 2001 From: ThibG Date: Sat, 15 Jul 2017 17:24:35 +0200 Subject: [PATCH] Optimize uri normalization (#4212) * Add dependency on idn-ruby to speed up URI normalization * Use normalized_host instead of normalize.host when applicable When we are only interested in the normalized host, calling normalized_host avoids normalizing the other components of the URI as well as creating a new object --- Gemfile | 1 + Gemfile.lock | 2 ++ app/lib/tag_manager.rb | 2 +- app/services/concerns/author_extractor.rb | 2 +- app/services/fetch_remote_status_service.rb | 4 ++-- app/workers/pubsubhubbub/delivery_worker.rb | 2 +- 6 files changed, 8 insertions(+), 5 deletions(-) diff --git a/Gemfile b/Gemfile index 5d5ddfae109..a6c2b2d65c6 100644 --- a/Gemfile +++ b/Gemfile @@ -36,6 +36,7 @@ gem 'htmlentities', '~> 4.3' gem 'http', '~> 2.2' gem 'http_accept_language', '~> 2.1' gem 'httplog', '~> 0.99' +gem 'idn-ruby', require: 'idn' gem 'kaminari', '~> 1.0' gem 'link_header', '~> 0.0' gem 'mime-types', '~> 3.1' diff --git a/Gemfile.lock b/Gemfile.lock index daef3e1ad86..f637c9bbe47 100644 --- a/Gemfile.lock +++ b/Gemfile.lock @@ -208,6 +208,7 @@ GEM parser (>= 2.2.3.0) rainbow (~> 2.2) terminal-table (>= 1.5.1) + idn-ruby (0.1.0) jmespath (1.3.1) json (2.1.0) jsonapi-renderer (0.1.2) @@ -528,6 +529,7 @@ DEPENDENCIES http_accept_language (~> 2.1) httplog (~> 0.99) i18n-tasks (~> 0.9) + idn-ruby kaminari (~> 1.0) letter_opener (~> 1.4) letter_opener_web (~> 1.3) diff --git a/app/lib/tag_manager.rb b/app/lib/tag_manager.rb index f1a2234dc44..5f87a2a48dc 100644 --- a/app/lib/tag_manager.rb +++ b/app/lib/tag_manager.rb @@ -70,7 +70,7 @@ class TagManager uri = Addressable::URI.new uri.host = domain.gsub(/[\/]/, '') - uri.normalize.host + uri.normalized_host end def same_acct?(canonical, needle) diff --git a/app/services/concerns/author_extractor.rb b/app/services/concerns/author_extractor.rb index 00fe1c66393..867d6dc2597 100644 --- a/app/services/concerns/author_extractor.rb +++ b/app/services/concerns/author_extractor.rb @@ -14,7 +14,7 @@ module AuthorExtractor return nil if username.blank? || uri.blank? - domain = Addressable::URI.parse(uri).normalize.host + domain = Addressable::URI.parse(uri).normalized_host acct = "#{username}@#{domain}" end diff --git a/app/services/fetch_remote_status_service.rb b/app/services/fetch_remote_status_service.rb index 4cfd33d90a0..6ac31e4d8b5 100644 --- a/app/services/fetch_remote_status_service.rb +++ b/app/services/fetch_remote_status_service.rb @@ -24,7 +24,7 @@ class FetchRemoteStatusService < BaseService xml.encoding = 'utf-8' account = author_from_xml(xml.at_xpath('/xmlns:entry', xmlns: TagManager::XMLNS)) - domain = Addressable::URI.parse(url).normalize.host + domain = Addressable::URI.parse(url).normalized_host return nil unless !account.nil? && confirmed_domain?(domain, account) @@ -39,6 +39,6 @@ class FetchRemoteStatusService < BaseService end def confirmed_domain?(domain, account) - account.domain.nil? || domain.casecmp(account.domain).zero? || domain.casecmp(Addressable::URI.parse(account.remote_url).normalize.host).zero? + account.domain.nil? || domain.casecmp(account.domain).zero? || domain.casecmp(Addressable::URI.parse(account.remote_url).normalized_host).zero? end end diff --git a/app/workers/pubsubhubbub/delivery_worker.rb b/app/workers/pubsubhubbub/delivery_worker.rb index 05d160cf7c7..2e1101b93cc 100644 --- a/app/workers/pubsubhubbub/delivery_worker.rb +++ b/app/workers/pubsubhubbub/delivery_worker.rb @@ -43,7 +43,7 @@ class Pubsubhubbub::DeliveryWorker end def host - Addressable::URI.parse(subscription.callback_url).normalize.host + Addressable::URI.parse(subscription.callback_url).normalized_host end def headers