Sorbet: how do you add a signature for a dynamically generated method? - sorbet

I'm using https://github.com/kenn/active_flag and https://github.com/chanzuckerberg/sorbet-rails
This is what its rbi looks like:
module ActiveFlag
extend ActiveSupport::Concern
end
class ActiveFlag::Definition
def column; end
def human(key, options = nil); end
def humans; end
def initialize(column, keys, klass); end
def keys; end
def maps; end
def pairs; end
def set_all!(key); end
def to_array(integer); end
def to_i(arg); end
def to_value(instance, integer); end
def unset_all!(key); end
end
class ActiveFlag::Railtie < Rails::Railtie
end
class ActiveFlag::Value < Set
def method_missing(symbol, *args, &block); end
def raw; end
def set!(key, options = nil); end
def set(key); end
def set?(key); end
def to_human; end
def to_s; end
def unset!(key, options = nil); end
def unset(key); end
def unset?(key); end
def with(instance, definition); end
end
module ActiveFlag::ClassMethods
def flag(column, keys); end
end
class ActiveRecord::Base
extend ActiveFlag::ClassMethods
end
The last bit (extending AR::Base) I added, the rest srb rbi gems generated automatically.
To actually use Active Flag, I then do this in my model:
flag :visible_to, [:employee, :manager, :admin]
visible_to is an integer column. Sorbet Rails has already typed it as such:
sig { returns(Integer) }
def visible_to; end
sig { params(value: Integer).void }
def visible_to=(value); end
sig { returns(T::Boolean) }
def visible_to?; end
I could change this definition, but it's an autogenerated file and I assume the changes will get lost. So the next thing I tried was adding a sig directly above where the method gets used:
sig { returns(ActiveFlag::Value) }
def visible_to; end
flag :visible_to, [:employee, :manager, :admin]
The problem here is that Sorbet fails because the method I've "defined" doesn't return anything. Which I know is fine, because it'll get overriden when flag is called (it uses define_method internally), but I don't know how to communicate this to Sorbet.
Returning value that does not conform to method result type https://srb.help/7005
54 | def visible_to; end
^^^^^^^^^^^^^^^^^^^
Expected ActiveFlag::Value
Method visible_to has return type ActiveFlag::Value
54 | def visible_to; end
^^^^^^^^^^^^^^
Got NilClass originating from:
54 | def visible_to; e
So my question is. What's the best way to tell Sorbet that this method will return an ActiveFlag::Value once it gets defined, ideally without making a manual change to an autogenerated file?
btw. I tried to see how types for enum in Rails are handled... it doesn't look like that's been done on sorbet-typed yet. Possibly for the same reason.

When you want to override a generated rbi file you can create a second rbi file for the same class in your workspace and move the methods in there. Sorbet will handle merging multiple definitions files for you. We keep these files in a separate sorbet/custom directory next to the generated files; others keep the rbi files next to the rb files in their application directory.
As to enums, Sorbet does not natively support enum literal types so that's likely why.

You could implement a custom plugin for sorbet-rails to generate methods added by active_flag and remove the sigs generated wrongly. Here is the documentation for it:
https://github.com/chanzuckerberg/sorbet-rails/blob/master/README.md#extending-model-generation-task-with-custom-plugins

Related

How to stop a helper method from applying to a specific controller?

I have a helper_method that allows links to escape from a subdomain. However it is impacting my videos_controller, as it essentially seems to negate the 'current_event' method when not in the events controlller.
I've tried several dozen different ways over the last 4 days to make it so I can still escape my links from the subdomain, but still allow the videos_controller to work.
I think the best way to achieve this is to exclude the videos_controller from the helper method, but I'm not sure how (or if it is actually the best way forward - I'm obviously a noob!) Any suggestions please?! Relevant code below:
module UrlHelper
def url_for(options = nil)
if request.subdomain.present? and request.subdomain.downcase != 'www' and !options.nil? and options.is_a?(Hash) and options.has_key? :only_path and options[:only_path]
options[:only_path] = false
end
super
end
end
Videos_controller
def new
if current_event?
#video = current_event.videos.new
else
#video = Video.new
end
end
def create
if current_event.present?
#video = current_event.videos.new(params[:video])
#video.user_id = current_user.id
key = get_key_from_the_cloud
#video.key = key
else
#video = current_user.videos.new(params[:video])
#video.user_id = current_user.id
key = get_key_from_the_cloud
#video.key = key
end
if #video.save
flash[:success] = "Video uploaded!"
redirect_to root_url(subdomain: => current_event.name)
else
flash[:error] = "#{#video.errors.messages}"
render :new
end
end
current_event method
def current_event
if request.subdomain.present?
#event = Event.find_by_name(request.subdomain)
end
end
Did you take a look at this post yet?
You might want to create a new function test that only does something like
module UrlHelper
def test
puts "Test is called"
end
end
If that works you know its not including that fails but it has to be the method.
Otherwise you know the module is not included and you can narrow down the search.

ActiveRecord: Is it possible to get the number of DB queries executed in addition to total time in the Rails log?

For every request, I get this in the logs:
Completed 200 OK in 854ms (Views: 1.0ms | ActiveRecord: 17.0ms)
Is it possible to get it to also include the number of queries?
Something like:
Completed 200 OK in 854ms (Views: 1.0ms | ActiveRecord: 17.0ms | Queries: 10)
Ideally, I'd like all the "cached" ones to show up in that count too. Ie, even if the "cache" is saving me from "N+1" queries from hitting the DB, I still want to know I have a problem.
I'm fine with monkeypatching / manually editing something, since I really want this just for my dev box.
(If this can be made civilizedly so I can have it in production, that's even better, but if not, I'm fine with just having a manually modified Rails in my own machine)
Thanks!
Daniel
I know the ThinkingSphinx gem does something quite like this, adding the time spent running Sphinx queries to the summary in the log. You can probably do something similar ( maybe by making your own gem, since I bet other people would appreciate this functionality) to make the number of queries appear.
I haven't really looked hard at how it works, but it looks like modifications to ActionController and LogSubscriber are responsible:
lib/thinking_sphinx/action_controller.rb:
module ThinkingSphinx
module ActionController
extend ActiveSupport::Concern
protected
attr_internal :query_runtime
def cleanup_view_runtime
log_subscriber = ThinkingSphinx::ActiveRecord::LogSubscriber
query_runtime_pre_render = log_subscriber.reset_runtime
runtime = super
query_runtime_post_render = log_subscriber.reset_runtime
self.query_runtime = query_runtime_pre_render + query_runtime_post_render
runtime - query_runtime_post_render
end
def append_info_to_payload(payload)
super
payload[:query_runtime] = query_runtime
end
module ClassMethods
def log_process_action(payload)
messages, query_runtime = super, payload[:query_runtime]
messages << ("Sphinx: %.1fms" % query_runtime.to_f) if query_runtime
messages
end
end
end
end
lib/thinking_sphinx/active_record/log_subscriber.rb:
require 'active_support/log_subscriber'
module ThinkingSphinx
module ActiveRecord
class LogSubscriber < ActiveSupport::LogSubscriber
def self.runtime=(value)
Thread.current['thinking_sphinx_query_runtime'] = value
end
def self.runtime
Thread.current['thinking_sphinx_query_runtime'] ||= 0
end
def self.reset_runtime
rt, self.runtime = runtime, 0
rt
end
def initialize
super
#odd_or_even = false
end
def query(event)
self.class.runtime += event.duration
return unless logger.debug?
identifier = color('Sphinx Query (%.1fms)' % event.duration, GREEN, true)
query = event.payload[:query]
query = color query, nil, true if odd?
debug " #{identifier} #{query}"
end
def message(event)
return unless logger.debug?
identifier = color 'Sphinx', GREEN, true
message = event.payload[:message]
message = color message, nil, true if odd?
debug " #{identifier} #{message}"
end
def odd?
#odd_or_even = !#odd_or_even
end
def logger
return #logger if defined? #logger
self.logger = ::ActiveRecord::Base.logger
end
def logger=(logger)
#logger = logger
end
attach_to :thinking_sphinx
end
end
end
I hope this helps.

Rails 3 – add action in controller's from before_filter

I am trying to add a mixin to my controller dynamically depending on the request parameters like so :
# Controller
class QuantitiesController < Admin::BaseController
before_filter :extend_input_method, only: [:create, :new]
def extend_input_method
input_method = params[:input_method]
if input_method
send(:extend, "InputMethod::#{input_method.classify}".constantize)
end
end
end
# Mixin that gets included in the controller
module InputMethod::Single
include InputMethod::Helpers
def new
puts "CALLED #new" # Debug information
load_recent_entries
quantity
end
def create
#quantity = scoped_by_subject.new(process_attributes)
if #quantity.save
save_success
else
load_recent_entries
save_error
end
end
end
The new method never gets called but my template gets rendered without raising an exception, even if action_name is new and respond_to?("new") is true after extending the instance.
I'd like to understand why this isn't working and how I can achieve something similar.
This is the solution I came up with. It works for my needs.
class QuantitiesController < Admin::BaseController
before_filter :extend_input_method, only: [:create, :new]
def new
_new
end
def create
_create
end
private
def extend_input_method
input_method = params[:input_method]
extend(Dep.get("InputMethod::#{input_method.classify}")) if input_method
end
end
module InputMethod::Single
include InputMethod::Helpers
def _new
# Do stuff...
end
def _create
# Do stuff...
end
end

Rails 3: Trying to extend Action Mailer with a module

Trying to rewrite an old alias_method_chain to add a filter on outgoing emails, and it isn't working. I'm pretty sure I've leaving something out/missing something, but I don't know what.
This file is in /lib/outgoing_mail_filter.rb, which is loaded with config/initializers/required.rb
Here's the old code that worked under Rails 2:
class ActionMailer::Base
def deliver_with_recipient_filter!(mail = #mail)
unless 'production' == Rails.env
mail.to = mail.to.to_a.delete_if do |to|
!(to.ends_with?('some_domain.com'))
end
end
unless mail.to.blank?
deliver_without_recipient_filter!(mail)
end
end
alias_method_chain 'deliver!'.to_sym, :recipient_filter
end
And here's my current attempt at re-writing it:
class ActionMailer::Base
module RecipientFilter
def deliver(mail = #mail)
super
unless 'production' == Rails.env
mail.to = mail.to.to_a.delete_if do |to|
!(to.ends_with?('some_domain.com'))
end
end
unless mail.to.blank?
deliver(mail)
end
end
end
include RecipientFilter
end
When I run my tests, it doesn't even look like this is being called or anything. Any help is appreciated
I'm using mail_safe to rewrite emails in the development environment, highly recommended. You could look into it for inspiration if it doesn't fit your bill, the code is very simple.
The following code is extracted from /lib/mail_safe/rails3_hook.rb and should do what you want:
require 'mail'
module MailSafe
class MailInterceptor
def self.delivering_email(mail)
# replace the following line with your code
# and don't forget to return the mail object at the end
MailSafe::AddressReplacer.replace_external_addresses(mail) if mail
end
::Mail.register_interceptor(self)
end
end
Alternate version, registering with ActionMailer::Base instead of Mail (thanks to Kevin Whitaker for letting me know it's possible):
module MailSafe
class MailInterceptor
def self.delivering_email(mail)
# replace the following line with your code
# and don't forget to return the mail object at the end
MailSafe::AddressReplacer.replace_external_addresses(mail) if mail
end
::ActionMailer::Base.register_interceptor(self)
end
end

Rails 3 - Building forms from Serialized Data

I've been working on a rails project where I am needed to serialize permissions for user roles and store in the database. As far as that goes I'm all good. Now my problem comes when I want to modify the serialized data from a rails generated form.
I acted on instinct and tried with the expected behavior.
That would be to use something like this:
f.check_box :permissions_customer_club_events_read
But as no getters or setters exist for the serialized data, this doesn't work (obviously :p). Now I wonder how I would go about tackling this problem and the only thing that comes to mind is dynamically generating getter and setter methods from my serialized hash.
Example:
def permissions_customer_club_events_read=(val)
permissions[:customer][:club][:events][:read] = val
end
def permissions_customer_club_events_read
permissions[:customer][:club][:events][:read]
end
Anyone understand what I'm getting at?
Here is my Model:
class User::Affiliation::Role < ActiveRecord::Base
require 'yajl'
class YajlCoder
def dump data
Yajl.dump data
end
def load data
return unless data
Yajl.load data
end
end
serialize :permissions, YajlCoder.new
after_initialize :init
def init
## Sets base permission structure ##
self.permissions ||= YAML.load_file("#{Rails.root}/config/permissions.yml")
end
end
I suggest you have a look at something like attr_bucket. Ostensibly, this can be used to solve some inheritance annoyances, but it will also solve your problem for you. Here is the essence.
It looks like you know what all your permissions are, but you want to serialize all of them into the same database field. But within your actual rails app, you want to treat all your permissions as if they were totally separate fields. This is exactly what a solution like attr_bucket will let you do. Let's take your example, you would do something like this:
class User::Affiliation::Role < ActiveRecord::Base
attr_bucket :permissions => [:permissions_customer_club_events_read, :permissions_customer_club_events_write, :permission_do_crazy_things]
after_initialize :init
def init
## Sets base permission structure ##
self.permissions ||= YAML.load_file("#{Rails.root}/config/permissions.yml")
end
end
Now you will be able to use permissions_customer_club_events_read, permissions_customer_club_events_write, permission_do_crazy_things as if they were separate database fields (this includes using them in forms etc.), but when you actually save your objects all those fields would get 'bucketed' together and serialized into the :permissions field.
The only caveat is the serialization mechanism, I believe attr_bucket will serialize everything using YAML, whereas you were using JSON. If this doesn't matter then you're golden, otherwise you might need to patch attr_bucket to use json instead of YAML which should be pretty straight forward.
Sorry if I did not understand the question ;)
You could have a customdata module, included in your model, and use method_missing:
module CustomData
def self.included(base)
base.instance_eval do
after_save :save_data
end
def method_missing(method, *args, &block)
if method.to_s =~ /^data_/
data[method] ? data[method] : nil
else
super
end
end
def data
#data ||= begin
#get and return your data
end
end
private
def save_data
end
end
With this method, you would have to use f.check_box :data_permissions_customer_club_events_read
It's not really complete, but I hope you get the idea ;)
attr_bucket seems like a good solution too.
This worked out for me in the end, this is how I solved it.
serialize :permissions, YajlCoder.new
after_initialize :init
def init
self.permissions ||= YAML.load_file("#{Rails.root}/config/permissions.yml")['customer']
build_attributes_from self.permissions, :permissions
end
private
def build_attributes_from store, prefix, path=[]
store.each do |k,v|
if v.class == Hash
build_attributes_from v, prefix, ( path + [k] )
else
create_attr_accessors_from prefix, ( path + [k] )
end
end
end
def create_attr_accessors_from prefix, path=[]
method_name = prefix.to_s + "_" + path.join('_')
class << self
self
end.send :define_method, method_name do
self.permissions.dig(:path => path)
end
class << self
self
end.send :define_method, "#{method_name}=" do |value|
self.permissions.dig(:path => path, :value => value)
end
end
And some monkey patching for hashes...
class Hash
def dig(args={})
path = args[:path].to_enum || []
value = args[:value] || nil
if value == nil
path.inject(self) do |location, key|
location.respond_to?(:keys) ? location[key] : nil
end
else
path.inject(self) do |location, key|
location[key] = ( location[key].class == Hash ) ? location[key] : value
end
end
end
end
Now getter and setter methods are generated for all of the serialized fields.