- Practice defining instance methods on a class
- Practice defining instance methods that use the
self
keyword - Learn about monkey patching
In this lab, you'll be adding a few instance methods to Ruby's String
class.
We generally want to avoid altering built-in classes (especially if we are
working with other people), but in this case, we're not overwriting any methods,
and it's a good exercise in OO programming. The practice of adding methods to or
altering Ruby's built-in classes is called monkey patching.
The String
class is just like any other class that we might define, such as a
Dog
class or a Person
class. The String
class, however, is native to, or
included in, Ruby. So, we can add or change methods in the String
class just
like we would in any of the classes that we define ourselves.
Monkey patching is the practice of adding methods to or altering Ruby's built-in classes.
Monkey patching is dangerous, and should almost never be used in real
application. What if, for example, you decide to monkey patch Ruby's String
class to produce a quick-fix that shortens a certain section of code in your
program? Then, months later, you run into major bugs as a result, or some of
your collaborators don't know about your monkey patch and develop bugs of their
own that they don't know the origin of? For reasons like these, monkey patching
should be considered very, very carefully. We're going to do it today, just for
fun (and to learn about self
), but you want to avoid doing it when working on
your own programs.
To see what monkey patching looks like in action, open up IRB. Let's try to call a method on a string that definitely doesn't exist in Ruby:
"hello".report_on_self
# NoMethodError (undefined method `report_on_self' for "hello":String)
We can monkey patch the String
class to add an instance method, just like
we'd add an instance method to one of our own custom classes. Run this in IRB as
well:
class String
def report_on_self
"Self is: #{self}"
end
end
Then use the #report_on_self
method on any string:
"hello".report_on_self
# => "Self is: hello"
"this string".report_on_self
# => "Self is: this string"
You'll be coding your solution in lib/count_sentences.rb
. We'll be
monkey-patching the built-in Ruby String
class.
Define a instance method #sentence?
that returns true
if the string you are
calling it on ends in a period and false
if it does not.
Top-Tip: In Ruby, there's a convention to name methods that return
true
orfalse
with a question mark (?
) at the end, like theArray#include?
orArray#any?
methods.
You will need to use the self
keyword in the body of this method to refer to
the string on which the method is being called. You can use the #end_with?
method on a string to determine what its final character is.
For example:
"HI!".end_with?("!")
# => true
Call the end_with?
method on self
inside of your #sentence?
method.
After implementing the #sentence?
method, you should be able to call it
on any string:
"The weather outside is frightful.".sentence?
# true
"But the fire is so delightful".sentence?
# false
This instance method should use the self
keyword to refer to the string on
which it is being called. This method should return true if a string ends with a
question mark and false if it does not.
This instance method should use the self
keyword to refer to the string on
which it is being called. This method should return true if a string ends with
an exclamation mark and false if it does not.
What we'd like to be able to do is call a count_sentences
method on a string,
and get back a, well, count of sentences in that string. In other words:
"This is a string! It has three sentences. Right?".count_sentences
# => 3
Your #count_sentences
method should use the self
keyword to refer to the
string on which it is called.
Think about the steps you need to go through to enact the desired behavior:
#split
the string on any and all periods, question marks and exclamation marks. You can do this in a few different ways, but it may help to use a Regular Expression.#count
the number of elements that results from that#split
.
Remember to consider edge cases, such as the following sentence:
This, well, is a sentence. This is too!! And so is this, I think? Woo...
What would happen if we split this sentence on the punctuation characters? We would end up with an array that contains empty strings as well as strings containing sentences. How would you eliminate empty strings from an array?
We recommend placing require 'pry'
on the top of the file and placing a
binding.pry
inside of the #count_sentences
method to help you get this test
passing.