Stub ActiveRecord::Relation with ActiveRecord objects - ruby

I'm not testing a Rails app. Just getting that out of the way.
I'm testing a library that connects to a relatively active server, restricting records by timestamp. These returned records change as time goes on, making testing other restrictions more complicated. I need to stub out the ActiveRecord::where method to return my own custom relation with objects I create to meet the criteria I need.
Something like
relation = double(ActiveRecord::Relation)
relation.stub(:[]).and_return( [MyClass.new(...), MyClass.new(...), ...] )
MyClass.stub(:where).and_return( relation )
is what I'd like, but that doesn't work. I need it to be an ActiveRecord::Relation because I need to be able to call ActiveRecord::where and ActiveRecord::select on the object in the code.
Edit 2014-01-28
In lib/call.rb
class Call < ActiveRecord::Base
class << self
def sales start_time, end_time
restricted_records = records(start_time, end_time, :agent_id)
#other code
end
#other methods
private
def records start_time, end_time, *select
# I'm leaving in commented code so you can see why I want the ActiveRecord::Relation object, not an Array
calls = Call.where("ts BETWEEN '#{start_time}' AND '#{end_time}'") #.select(select)
raise calls.inspect
#.to_a.map(&:serializable_hash).map {|record| symbolize(record)}
end
end
end
In spec/call_spec.rb
require 'spec_helper'
require 'call.rb'
describe Call do
let(:period_start) { Time.now - 60 }
let(:period_end) { Time.now }
describe "::sales" do
before do
relation = Call.all
relation.stub(:[]).and_return( [Call.new(queue: "12345")] )
Call.stub(:where).and_return( relation )
end
subject { Call.sales(period_start, period_end) }
it "restricts results to my custom object" do
subject
end
end
end
Output from test:
RuntimeError:
#<ActiveRecord::Relation [ #an array containing all the actual Call records, not my object ]>

ActiveRecord::Relation is a class and :[] is an instance method of that class. You're stubbing a method of the class itself, so it's not going to be invoked by any of the Rails code.
If you want MyClass.where to return a relation with just the :[] stubbed, you'll have to create a Relation instance first, as in:
relation = MyClass.all
relation.stub(:[]).and_return( [MyClass.new(...), MyClass.new(...), ...] )
MyClass.stub(:where).and_return( relation )
However, note that in order to get to your returned array in this context, you'll need to do:
MyClass.where("ignored parameters")["ignored parameters"]
Further, if you subsequently call where on relation, you'll return a new instance of Relation which will no longer be stubbed.

Update 2022
The previous upvoted answer is wholly incorrect since does not work with indexing, .to_a, .first, .last, .any?, .none?, any pretty much every other method.
Instead, you can mock the records contained within a relation by stubbing its records method.
custom_records = ["a", "b", "c"]
relation = Model.all
relation.stub(:records).and_return(custom_records)
allow(Model).to receive(:where).and_return(relation)
# Later ...
records = Model.where('1 + 1 = 2') # content of the query doesn't matter, .where is mocked
records.first # => "a"
records.last # => "c"
records.to_a # => ["a", "b", "c"]
records.any? { |x| x == "b" } # => true
Most of the methods will work, but there are a few exceptions that will need to be stubbed separately.
.count - directly invokes a SELECT COUNT(*) SQL query, which bypasses our records mock. Fix:
relation.stub(:count).and_return(custom_records.count)
.exists? - directly invokes another SQL query, again bypassing our records mock. Fix:
relation.stub(:exists?).and_return(custom_records.present?)
others - There are probably other methods that you might need to stub (depending on if your code uses those methods), you can stub each method as-needed
Furthermore you can mock the return value of a has_many relation (which was my actual use case when googling this question) by doing
allow(record).to receive(:related_records).and_wrap_original do |original, *args, &block|
relation = original.call(*args, &block)
relation.stub(:records).and_return(my_custom_array_of_related_records)
relation
end

Related

Metrics/AbcSize Too High: How do I decrease the ABC in this method?

I have recently started using Rubocop to "standardise" my code, and it has helped me optimise a lot of my code, as well as help me learn a lot of Ruby "tricks". I understand that I should use my own judgement and disable Cops where necessary, but I have found myself quite stuck with the below code:
def index
if params[:filters].present?
if params[:filters][:deleted].blank? || params[:filters][:deleted] == "false"
# if owned is true, then we don't need to filter by admin
params[:filters][:admin] = nil if params[:filters][:admin].present? && params[:filters][:owned] == "true"
# if admin is true, then must not filter by owned if false
params[:filters][:owned] = nil if params[:filters][:owned].present? && params[:filters][:admin] == "false"
companies_list =
case params[:filters][:admin]&.to_b
when true
current_user.admin_companies
when false
current_user.non_admin_companies
end
if params[:filters][:owned].present?
companies_list ||= current_user.companies
if params[:filters][:owned].to_b
companies_list = companies_list.where(owner: current_user)
else
companies_list = companies_list.where.not(owner: current_user)
end
end
else
# Filters for deleted companies
companies_list = {}
end
end
companies_list ||= current_user.companies
response = { data: companies_list.alphabetical.as_json(current_user: current_user) }
json_response(response)
end
Among others, the error that I'm getting is the following:
C: Metrics/AbcSize: Assignment Branch Condition size for index is too high. [<13, 57, 16> 60.61/15]
I understand the maths behind it, but I don't know how to simplify this code to achieve the same result.
Could someone please give me some guidance on this?
Thanks in advance.
Well first and foremost, is this code fully tested, including all the myriad conditions? It's so complex that refactoring will surely be disastrous unless the test suite is rigorous. So, write a comprehensive test suite if you don't already have one. If there's already a test suite, make sure it tests all the conditions.
Second, apply the "fat model skinny controller" paradigm. So move all the complexity into a model, let's call it CompanyFilter
def index
companies_list = CompanyFilter.new(current_user, params).list
response = { data: companies_list.alphabetical.as_json(current_user: current_user) }
json_response(response)
end
and move all those if/then/else statements into the CompanyFilter#list method
tests still pass? great, you'll still get the Rubocop warnings, but related to the CompanyFilter class.
Now you need to untangle all the conditions. It's a bit hard for me to understand what's going on, but it looks as if it should be reducible to a single case statement, with 5 possible outcomes. So the CompanyFilter class might look something like this:
class CompanyFilter
attr_accessors :current_user, :params
def initialize(current_user, params)
#current_user = current_user
#params = params
end
def list
case
when no_filter_specified
{}
when user_is_admin
#current_user.admin_companies
when user_is_owned
# etc
when # other condition
# etc
end
end
private
def no_filter_specified
#params[:filter].blank?
end
def user_is_admin
# returns boolean based on params hash
end
def user_is_owned
# returns boolean based on params hash
end
end
tests still passing? perfect! [Edit] Now you can move most of your controller tests into a model test for the CompanyFilter class.
Finally I would define all the different companies_list queries as scopes on the Company model, e.g.
class Company < ApplicationRecord
# some examples, I don't know what's appropriate in this app
scope :for_user, ->(user){ where("...") }
scope :administered_by, ->(user){ where("...") }
end
When composing database scopes ActiveRecord::SpawnMethods#merge is your friend.
Post.where(title: 'How to use .merge')
.merge(Post.where(published: true))
While it doesn't look like much it lets you programatically compose scopes without overelying on mutating assignment and if/else trees. You can for example compose an array of conditions and merge them together into a single ActiveRecord::Relation object with Array#reduce:
[Post.where(title: 'foo'), Post.where(author: 'bar')].reduce(&:merge)
# => SELECT "posts".* FROM "posts" WHERE "posts"."title" = $1 AND "posts"."author" = $2 LIMIT $3
So lets combine that with a skinny controllers approach where you handle filtering in a seperate object:
class ApplicationFilter
include ActiveModel::Attributes
include ActiveModel::AttributeAssignment
attr_accessor :user
def initialize(**attributes)
super()
assign_attributes(attributes)
end
# A convenience method to both instanciate and apply the filters
def self.call(user, params, scope: model_class.all)
return scope unless params[:filters].present?
scope.merge(
new(
permit_params(params).merge(user: user)
).to_scope
)
end
def to_scope
filters.map { |filter| apply_filter(filter) }
.compact
.select {|f| f.respond_to?(:merge) }
.reduce(&:merge)
end
private
# calls a filter_by_foo method if present or
# defaults to where(key => value)
def apply_filter(attribute)
if respond_to? "filter_by_#{attribute}"
send("filter_by_#{attribute}")
else
self.class.model_class.where(
attribute => send(attribute)
)
end
end
# Convention over Configuration is sexy.
def self.model_class
name.chomp("Filter").constantize
end
# filters the incoming params hash based on the attributes of this filter class
def self.permit_params
params.permit(filters).reject{ |k,v| v.blank? }
end
# provided for modularity
def self.filters
attribute_names
end
end
This uses some of the goodness provided by Rails to setup objects with attributes that will dynamically handle filtering attributes. It looks at the list of attributes you have declared and then slices those off the params and applies a method for that filter if present.
We can then write a concrete implementation:
class CompanyFilter < ApplicationFilter
attribute :admin, :boolean, default: false
attribute :owned, :boolean
private
def filter_by_admin
if admin
user.admin_companies
else
user.non_admin_companies
end
end
# this should be refactored to use an assocation on User
def filter_by_owned
case owned
when nil
nil
when true
Company.where(owner: user)
when false
Company.where.not(owner: user)
end
end
end
And you can call it with:
# scope is optional
#companies = CompanyFilter.call(current_user, params), scope: current_user.companies)

Custom to_json for nested complex objects in Ruby

I'm new to Ruby and having a little trouble json. I have inherited my classes with custom made JSONable class, as explained HERE in this answer. I have customized it according to my need, but I couldn't figure out how to make it work with custom nested (complex) objects, according to my requirement. I have following scenario.
First Class:
class Option < JSONable
def IncludeAll=(includeAll) #bool
#includeAll = includeAll
end
def IncludeAddress=(includeAddress) #bool
#includeAddress= includeAddress
end
......
Second Class:
class Search < JSONable
def CustomerId=(customerId)
#customerId = customerId
end
def identifier=(identifier)
#identifier = identifier
end
def Options=(options) #This is expected to be of Class Option, declared above
#options = options
end
Third Class:
class Request < JSONable
def DateTimeStamp=(dateTimeStamp)
#dateTimeStamp = dateTimeStamp
end
def SDKVersion=(sDKVersion)
#sDKVersion = sDKVersion
end
def RequestMessage=(requestMessage) #This is of type Search, declared above
#requestMessage = requestMessage
end
I call it as:
search = Search.new
searchOpts = Options.new
request = Request.new
search.identifier = identifier
searchOpts.IncludeAll = false
searchOpts.IncludeAddress = true
search.Options = searchOpts #setting nested level2 property here
//THE MOST OUTER CLASS OBJECT
request.SDKVersion = "xyz"
request.RequestMessage = search #setting nested level1
My ultimate goal is to send this request object to an API, after converting it to JSON. so i call to_json on request object as:
request.to_json
But here, suggested solution in that post (JSONable) fails in this case, as it can't convert the nested complex objects request.search and request.search.Options to Json.
(gives error: in 'to_json': wrong number of arguments (1 for 0) (ArgumentError)')
What I tried:
class JSONable
def to_json
hash = {}
self.instance_variables.each do |var|
#hash[var] = self.instance_variable_get var #tried to apply following check
if((self.instance_variable_get var).instance_of? Options ||((varVal).instance_of? Search))
varVal = self.instance_variable_get var
hash[var] = varVal.to_json #convert inner object to json
else
hash[var] = self.instance_variable_get var
end
end
hash.to_json
end
.....
This converts the nested model without any problem, but it messes up the 3rd level json. The result is as following:
{"DateTimeStamp":"121212","SDKVersion":"1.5","Culture":"en","RequestMessage":"{\"identifier\":\"851848913\",\"Options\":\"{\\\"IncludeAll\\\":true,\\\"IncludeAssociatedEntities\\\":true,\\\"IncludeAddress\\\":true,\\\"IncludePaymentInstructions\\\":true}\"}"}
And API doesn't respond. It seems as it messes up the boolean variables, which should be something like:
"SearchOption":"{\"IncludeAll\":true,\"IncludeAssociatedEntities\":true,\...
but it gives:
"SearchOption\":\"{\\\"IncludeAll\\\":true,\\\"IncludeAssociatedEntities\\\":true,\\\"Includ...
So the API logic can't cast it to corresponding bool objects anymore. JSON validator also fails to validate this result, i checked online
Questions:
How can I avoid this, and produce valid JSON in this case?
How can I apply generic check to in my JSONable class to check if the object is of some custom class / complex object.
(currently i have checked only for specific classes as:)
if((self.instance_variable_get var).instance_of? Options ||((varVal).instance_of? Search))
Other Info:
It works fine for all complex objects, having no nested objects
API is developed in .NET
I'm not using Rails, its a Ruby console app (I'm new to Ruby)
The answer you referred is dated “Dec 2010.” JSON library is included in ruby stdlib for years already and it perfectly converts Hash instances to json. That said, you just need to construct hashes out of your objects and then call JSON.dump on the resulting hash. I have no idea what JSONable is and you definitely do not need it. Introduce some base class, let’s call it Base:
class Base
def to_h
instance_variables.map do |iv|
value = instance_variable_get(:"##{iv}")
[
iv.to_s[1..-1], # name without leading `#`
case value
when Base then value.to_h # Base instance? convert deeply
when Array # Array? convert elements
value.map do |e|
e.respond_to?(:to_h) ? e.to_h : e
end
else value # seems to be non-convertable, put as is
end
]
end.to_h
end
end
Now just derive your classes from Base to make them respond to to_h, define all your instance variables as you did, and call:
require 'json'
JSON.dump request.to_h # request.to_h.to_json should work as well
The above should produce the nested JSON, hashes are happily converted to json by this library automagically.

Passing the parameter of an Object into an Object method

This a simplified version of what I am trying to solve:
In Ruby 2.0.0, Rails 4.0.0, Activerecord
Segment has_many Sales
Sale.find(1).bid = 1
Sale.find(1).ask = 2
Sale.find(2).bid = 10
Sale.find(2).ask = 20
etc
When I write this method:
class Segment
def add_stuff(param)
sales.map{ |s| s.param }.inject(:+)
end
end
Question: How to pass in bid or ask as param and interpolate that param properly within the block? Thanks.
Never use map on active record association, when you can use pluck or other querying method like, for example sum:
def add_staff(param)
sales.sum(param)
end
Just for the completeness, here is how to do it with map - note however that map is much slower and error prone than querying methods:
def add_staff(param)
sales.map {|s| s[param]} # if param is name of the column on sales model.
end
Or more general, where param is name of association or custom (non-column) method
def add_staff(param)
sales.map(&param) # Or map {|s| s.send(param) }
end

Static local variables for methods in Ruby?

I have this:
def valid_attributes
{ :email => "some_#{rand(9999)}#thing.com" }
end
For Rspec testing right? But I would like to do something like this:
def valid_attributes
static user_id = 0
user_id += 1
{ :email => "some_#{user_id}#thing.com" }
end
I don't want user_id to be accessible from anywhere but that method,
is this possible with Ruby?
This is a closure case. Try this
lambda {
user_id = 0
self.class.send(:define_method, :valid_attributes) do
user_id += 1
{ :email => "some_#{user_id}#thing.com" }
end
}.call
Wrapping everything in lambda allows the variables defined within lambda to only exist in the scope. You can add other methods also. Good luck!
This answer is a little larger in scope than your question, but I think it gets at the root of what you're trying to do, and will be the easiest and most maintainable.
I think what you're really looking for here is factories. Try using something like factory_girl, which will make a lot of testing much easier.
First, you'd set up a factory to create whatever type of object it is you're testing, and use a sequence for the email attribute:
FactoryGirl.define do
factory :model do
sequence(:email) {|n| "person#{n}#example.com" }
# include whatever else is required to make your model valid
end
end
Then, when you need valid attributes, you can use
Factory.attributes_for(:model)
You can also use Factory.create and Factory.build to create saved and unsaved instances of the model.
There's explanation of a lot more of the features in the getting started document, as well as instructions on how to add factories to your project.
You can use a closure:
def validator_factory
user_id = 0
lambda do
user_id += 1
{ :email => "some_#{user_id}#thing.com" }
end
end
valid_attributes = validator_factory
valid_attributes.call #=> {:email=>"some_1#thing.com"}
valid_attributes.call #=> {:email=>"some_2#thing.com"}
This way user_id won't be accessible outside.
I'd use an instance variable:
def valid_attributes
#user_id ||= 0
#user_id += 1
{ :email => "some_#{#user_id}#thing.com" }
end
The only variables Ruby has are local variables, instance variables, class variables and global variables. None of them fit what you're after.
What you probably need is a singleton that stores the user_id, and gives you a new ID number each time. Otherwise, your code won't be thread-safe.

ActiveRecord and using reject method

I have a model that fetches all the games from a particular city. When I get those games I want to filter them and I would like to use the reject method, but I'm running into an error I'm trying to understand.
# STEP 1 - Model
class Matches < ActiveRecord::Base
def self.total_losses(cities)
reject{ |a| cities.include?(a.winner) }.count
end
end
# STEP 2 - Controller
#games = Matches.find_matches_by("Toronto")
# GOOD! - Returns ActiveRecord::Relation
# STEP 3 - View
cities = ["Toronto", "NYC"]
#games.total_losses(cities)
# FAIL - undefined method reject for #<Class:0x00000101ee6360>
# STEP 3 - View
cities = ["Toronto", "NYC"]
#games.reject{ |a| cities.include?(a.winner) }.count
# PASSES - it returns a number.
Why does reject fail in my model but not in my view ?
The difference is the object you are calling reject on. In the view, #games is an array of Active Record objects, so calling #games.reject uses Array#reject. In your model, you're calling reject on self in a class method, meaning it's attempting to call Matches.reject, which doesn't exist. You need to fetch records first, like this:
def self.total_losses(cities)
all.reject { |a| cities.include(a.winner) }.count
end

Resources