I have a module of following
module SimpleTask
def task1
end
def task2
end
def task3
end
end
And I have a model which requires only task2 method of module SimpleTask.
I know including SimpleTask in my model with include SimpleTask would do the job.
But I wonder if I can only include specific task2 method in my model.
It sounds like you need to refactor #task2 into a separate module (e.g., BaseTask). Then you can easily include only BaseTask where you only need #task2.
module BaseTask
def task2
...
end
end
module SimpleTask
include BaseTask
def task1
...
end
def task3
...
end
end
It's hard to help much more without a more concrete question (such as interdependence between the methods of SimpleTask, etc.
You could do some meta-programming where you include SimpleTask and then undefine the methods you don't want, but that's pretty ugly IMO.
You could add
module SimpleTask
def task1
end
def task2
end
def task3
end
module_function :task2
end
So that you can call the method like a class method on the module as well as having it as an instance
method in the places you do want all three methods, ie:
class Foo
include SimpleTask
end #=> Foo.new.task2
class LessFoo
def only_needs_the_one_method
SimpleTask.task2
end
end #=> LessFoo.new.only_needs_the_one_method
Or, if there's really no shared state in the module and you don't mind always using the module name itself, you can just declare all the methods class-level like so:
module SimpleTask
def self.task1
end
def self.task2
end
def self.task3
end
end
class Foo
include SimpleTask # Does, more or less nothing now
def do_something
SimpleTask.task1
end
end
#=> Foo.new.task2 #=> "task2 not a method or variable in Foo"
#=> Foo.new.do_something does, however, work
class LessFoo
def only_needs_the_one_method
SimpleTask.task2
end
end #=> LessFoo.new.only_needs_the_one_method works as well in this case
But you'd have to change all the callers in that case.
I'm going to steal an example from delegate.rb, it restricts what it includes
...
class Delegator < BasicObject
kernel = ::Kernel.dup
kernel.class_eval do
[:to_s,:inspect,:=~,:!~,:===,:<=>,:eql?,:hash].each do |m|
undef_method m
end
end
include kernel
...
becomes
module PreciseInclude
def include_except(mod, *except)
the_module = mod.dup
the_module.class_eval do
except.each do |m|
remove_method m # was undef_method, that prevents parent calls
end
end
include the_module
end
end
class Foo
extend PreciseInclude
include_except(SimpleTask, :task1, :task2)
end
Foo.instance_methods.grep(/task/) => [:task3]
you can always flip it so instead of include it becomes include_only
The catch is that remove_method won't work for nested modules, and using undef will prevent searching the entire hierarchy for that method.
A simple solution for this is
define_method :task2, SimpleTask.instance_method(:task2)
Related
How to define an original name scope in module/class with Ruby
I want to implement class like the following:
module SomeModule
extend OriginalNameScope
scope(:some) do
def method1
puts 1
end
def method2
puts 2
end
end
end
class SomeClass
include SomeModule
end
c = SomeClass.new
# I want to call methods like the following:
c.some_method1
c.some_method2
How to implement the OriginalNameScope module? I found out to get the method definitions in this method, but I don't know how to redefine methods with a prefix scope.
module OriginalNameScope
def scope(name, &method_definition)
puts method_definition.class
# => Proc
end
end
This is actually just a combination of some simple standard Ruby metaprogramming patterns and idioms:
module OriginalNameScope
def scope(name)
singleton_class.prepend(Module.new do
define_method(:method_added) do |meth|
if name && !#__recursion_guard__
#__recursion_guard__ = meth
method = instance_method(meth)
undef_method(meth)
define_method(:"#{name}_#{meth}") do |*args, &block|
method.bind(self).(*args, &block)
end
end
#__recursion_guard__ = nil
super(meth)
end
end)
yield
end
end
I just slapped this together, there's probably a lot that can be improved (e.g. use Refinements) and simplified.
Consider:
module A
def self.a; puts "a"; end;
def aa; A.a; end;
end
include A
aa
a
aa works but not a. Is there a trick to get A.a available as is a just as include Math makes log available as Math.log? I suspect the way is to write an a method for each self.a method, but is there's a trick to avoid that?
Might you be looking for extend self?
module A
def a; puts "a"; end;
extend self
end
A.a
include A
a
Based on other people's answers, I think what I want is:
module A
module Functions
def a
puts 'a'
end
end
extend Functions
end
# a is a module function of A
A.a
# Explicitly include these functions into one's instance
include A::Functions
a
Now one can include A without polluting their instance space with methods...
unless explicitly doing so with include A::Functions.
There is a trick if you have access to the modules source, and there's still a trick if you don't. If you do, this would be the module A:
module A
def a
puts 'a!'
end
module_function :a
end
All of these will call a:
A.a
include A; a
Even if you don't have access to the module's source, this is still possible with a little (lot) of metaprogramming:
SomeMod.instance_methods(false).each {|method| SomeMod.__send__(:module_function, method.to_sym)}
This only works if the methods are defined as instance methods only in the original module.
If you want do define them as class methods and only make the instance when included:
module A
def self.a
puts 'a'
end
def self.included(klass)
A.singleton_class.instance_methods(false).each do |m|
klass.__send__(:define_method, m.to_sym) do |*args|
A.__send__(m.to_sym, *args)
end
end
end
end
possibly I'm not explaining the concept very well, but I'm looking to add class methods to a series of ruby classes to enable them to hold class specific information which will then be called by individual instance methods of the classes.
I can make it work, but it is a bit ugly. Can anyone as it requires 2 modules, one included and the other extended (see example code below).
Can anyone think of a more elegant way of implementing this functionality ?
Thanks
Steve
This module is extended to give class methods but adding an instance member to each class it is included in
module My1
def my_methods (*sym_array)
#my_methods=sym_array
end
def method_list
#my_methods
end
end
This module is included to give instance methods
module My2
def foo
self.class.method_list.each { |m| self.send m }
end
end
Now use the modules - the ugliness is having to use an include and extend statement to allow me to pass a set of symbols to a class method which will then be implemented in an
instance
class Foo
extend My1
include My2
my_methods :baz
def baz
puts "Baz!"
end
end
class Bar
extend My1
include My2
my_methods :frodo
def frodo
puts "Frodo!"
end
end
class Wibble < Bar
extend My1
include My2
my_methods :wobble
def wobble
puts "Wobble!"
end
end
Here is the required output - note that each class has its own instance #my_methods so the behaviour is different for the derived class Wibble < Bar
f=Foo.new
b=Bar.new
w=Wibble.new
f.foo #=> "Bar!"
b.foo #=> "Frodo!"
w.foo #=> "Wobble!"
When a module is included, a hook is called on it. You can use that to do the extend you want.
module M1
def self.included(base)
base.extend(M2)
end
end
People often call that second module M1::ClassMethods. If you're using rails, ActiveSupport::Concern encapsulates this pattern
I would suggest to use a hook from module instead:
module MyModule
def self.included(klass)
klass.extend ClassMethods
end
def foo
self.class.method_list.each{ |m| self.send m }
end
module ClassMethods
attr_reader :method_list
def my_methods(*sym_array)
#method_list = sym_array
end
end
end
So it simplifies to call include only a module whenever you want the functionality to given classes.
class Foo
include MyModule
my_methods :baz
def baz
puts "Baz!"
end
end
Is it possible to make this work, without having to include the module at the end of the class and just include it at the top?
module VerboseJob
def self.included(job_class)
class << job_class
alias_method :original_perform, :perform
def perform(*args)
JobLogger.verbose { original_perform(*args) }
end
end
end
end
class HelloJob
include VerboseJob
def self.perform(arg1, arg2)
puts "Job invoked with #{arg1} and #{arg2}"
end
end
What I want to happen, is for HelloJob.perform to actually invoke VerboseJob.perform (which then calls the original method inside a block). Because the module here is included at the top of the class, this doesn't work, since perform isn't yet defined. Moving the include to the end does work, but is there a way that's a bit more forgiving? I like to keep all the included modules at the top of my class definitions.
I'm sort of looking for some method that is called on Module or Class when it has been fully loaded, instead of as it is interpreted by the runtime.
Here is a rather roundabout/hackish way of doing it that I came up with by deferring the definition of the wrapper method until the original method had been defined:
module A
def self.included(base)
base.class_eval do
def self.singleton_method_added(name)
##ran||=false
if name==:perform && !##ran
##ran=true
class<<self
alias_method :original_perform, :perform
def perform(*args)
puts "Hello"
original_perform(*args)
end
end
end
end
end
end
end
class B
include A
def self.perform
puts "Foobar"
end
end
B.perform
Edit:
d11wtq simplified this to the much cleaner:
module VerboseJob
module ClassMethods
def wrap_perform!
class << self
def perform_with_verbose(*args)
JobLogger.verbose { perform_without_verbose(*args) }
end
alias_method_chain :perform, :verbose \
unless instance_method(:perform) == instance_method(:perform_with_verbose)
end
end
def singleton_method_added(name)
wrap_perform! if name == :perform
end
end
def self.included(job_class)
job_class.extend ClassMethods
job_class.wrap_perform! if job_class.respond_to?(:perform)
end
end
Assuming that you want all of your classes defined before you want perform run, you may want to use Kernel#at_exit:
Converts block to a Proc object (and therefore binds it at the point
of call) and registers it for execution when the program exits. If
multiple handlers are registered, they are executed in reverse order
of registration.
def do_at_exit(str1)
at_exit { print str1 }
end
at_exit { puts "cruel world" }
do_at_exit("goodbye ")
exit
produces:
goodbye cruel world
You may also want to look at how unit testing frameworks, such as Test::Unit or MiniTest, handle delaying the running of tasks.
I need to define the constant in the module that use the method from the class that includes this module:
module B
def self.included(base)
class << base
CONST = self.find
end
end
end
class A
def self.find
"AAA"
end
include B
end
puts A::CONST
But the compiler gives the error on the 4th line.
Is there any other way to define the constant?
The more idiomatic way to achieve this in Ruby is:
module B
def self.included(klass)
klass.class_eval <<-ruby_eval
CONST = find
ruby_eval
# note that the block form of class_eval won't work
# because you can't assign a constant inside a method
end
end
class A
def self.find
"AAA"
end
include B
end
puts A::CONST
What you were doing (class << base) actually puts you into the context of A's metaclass, not A itself. The find method is on A itself, not its metaclass. The thing to keep in mind is that classes are themselves objects, and so have their own metaclasses.
To try to make it clearer:
class Human
def parent
# this method is on the Human class and available
# to all instances of Human.
end
class << self
def build
# this method is on the Human metaclass, and
# available to its instance, Human itself.
end
# the "self" here is Human's metaclass, so build
# cannot be called.
end
def self.build
# exactly the same as the above
end
build # the "self" here is Human itself, so build can
# be called
end
Not sure if that helps, but if you don't understand it, you can still use the class_eval idiom above.
In your specific case.
module B
def self.included(base)
base.const_set("CONST", base.find)
end
end
class A
def self.find
"AAA"
end
include B
end
puts A::CONST
Despite it works, it's a little bit messy. Are you sure you can't follow a different way to achieve your goal?
module B
def self.included(base)
class << base
CONST = self.find
end
end
end
class A
class << self
def self.find
"AAA"
end
end
include B
end
then the compiler error is fixed, pls try.