paint-brush
How to Find the Stinky Parts of Your Code [Part XXIV]by@mcsee
691 reads
691 reads

How to Find the Stinky Parts of Your Code [Part XXIV]

by Maximiliano ContieriOctober 19th, 2022
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

There are likely many instances where it could be edited or improved. Most of these smells are just hints of something that might be wrong. They are not required fixed per se… (You should look into it though.) The smell of code smells is a classic. It smells because of many instances of things that could be wrong. The smell is not required to be fixed, but you should look at it more closely. It's a classic classic. Code smells are a classic, but there are many ways to improve your code smell.

People Mentioned

Mention Thumbnail
Mention Thumbnail

Company Mentioned

Mention Thumbnail
featured image - How to Find the Stinky Parts of Your Code [Part XXIV]
Maximiliano Contieri HackerNoon profile picture

Code smells Are a Classic.

It smells because there are likely many instances where it could be edited or improved.


Most of these smells are just hints of something that might be wrong. They are not required fixed per se… (You should look into it though.)


Previous Code Smells

Let's continue...


Code Smell 116 - Variables Declared With 'var'

Var, Let, Const: are all the same, aren't they?


TL;DR: Choose wisely your variable names, scope, and mutability.

Problems

Solutions

  1. Declare const all variables unless you need to change them.

Context

Most languages don't need variable declarations.


Some other languages allow us to state mutability.


We should be strict and explicit with our declarations.

Sample Code

Wrong

var pi = 3.14
var universeAgeInYears = 13.800.000.000

pi = 3.1415 // no error
universeAgeInYears = 13.800.000.001 // no error

Right

const pi = 3.14 //Value cannot mutate or change 
let universeAgeInYears = 13.800.000.000 //Value can change

pi = 3.1415 // error. cannot define
universeAgeInYears = 13.800.000.001 // no error

Detection

  • [x]Manual

With mutation testing, by forcing a 'const' declaration, we can check if a value remains constant and be more declarative by explicitly enforcing it.

Tags

  • Mutability
  • Javascript

Conclusion

Readability is always very important.


We need to explicitly state our intentions and usages.

Relations

Code Smell 86 - Mutable Const Arrays

More Info


Just as it is a good practice to make all fields private unless they need greater visibility, it is a good practice to make all fields final unless they need to be mutable.


Brian Goetz

Software Engineering Great Quotes


Code Smell 117 - Unrealistic Data

Programmers are lazy and seldom try to learn from real business domains.


TL;DR: Use real case scenarios and real data (when possible)

Problems

  • Bijection Violation
  • Bad test use cases
  • Readability

Solutions

  1. Change test data for a real one.
  2. Use MAPPER to map real entities and real data.

Context

In the past, developers used to fake domain data.


We considered Hello Word a good practice and we tested it with abstract data.


We developed using a waterfall model very far from real users.


With bijection and MAPPER techniques, DDD and TDD, user acceptance testing became more important.


Using Agile methodologies, we need to test with real-world data.


If we find an error in a production system, we need to add a case covering the exact mistake with real data.

Sample Code

Wrong

class BookCartTestCase(unittest.TestCase):
    def setUp(self):
        self.cart = Cart()

    def test_add_book(self):
       self.cart.add_item('xxxxx', 3, 10)
        #This is not a real example

       self.assertEqual(self.cart.total, 30, msg='Book Cart total not correct after adding books')
       self.assertEqual(self.cart.items['xxxxx'], 3, msg='Quantity of items not correct after adding book')
 
    def test_remove_item(self):
        self.cart.add_item('fgdfhhfhhh', 3, 10)
        self.cart.remove_item('fgdfhhfhrhh', 2, 10)    
        #We made a typo since example is not a real one
        self.assertEqual(self.cart.total, 10, msg='Book Cart total not correct after removing book')
        self.assertEqual(self.cart.items['fgdfhhfhhh'], 1, msg='Quantity of books not correct after removing book')

Right

class BookCartTestCase(unittest.TestCase):
    def setUp(self):
        self.cart = Cart()

    def test_add_book(self):
       self.cart.add_item('Harry Potter', 3, 10)

       self.assertEqual(self.cart.total, 30, msg='Book Cart total not correct after adding books')
       self.assertEqual(self.cart.items['Harry Potter'], 3, msg='Quantity of items not correct after adding book')

    #We don't reuse same example. 
    #We use a new REAL book
    def test_remove_item(self):
        self.cart.add_item('Divergent', 3, 10)
        self.cart.remove_item('Divergent', 2, 10)    
        self.assertEqual(self.cart.total, 10, msg='Book Cart total not correct after removing book')
        self.assertEqual(self.cart.items['Divergent'], 1, msg='Quantity of books not correct after removing book')

Detection

  • [x]Manual

This is a semantic smell.

Exceptions

In some domains and under regulation, we cannot use real data.


We should fake it with meaningful data.

Tags

  • Testing

Conclusion

Code comments are a code smell.


Reading tests is the only way to learn how the software behaves.


We need to be extra explicit on our tests.

Relations

Code Smell 05 - Comment Abusers

More Info

Credits

Photo by Hofmann Natalia on Unsplash


Thanks to Curtis Einsmann


You do not really understand something unless you can explain it to your grandmother.

Albert Einstein

Software Engineering Great Quotes


Code Smell 118 - Return False

Checking for a boolean condition to return a boolean value is awkward


TL;DR: Don't return explicit booleans. Most boolean usages are code smells.

Problems

  • Declarativeness
  • Ninja Code
  • Implementative solutions

Solutions

  1. Return a boolean proposition instead of checking a negation.
  2. The answer must be a business logic formula, not an algorithm.

Context

When dealing with boolean formulas, it is more readable to show a business boolean formula than introduce a negated IF clause.


Programmers tend to return accidental implementative solutions instead of real business rules.

Sample Code

Wrong

function canWeMoveOn() {
  if (work.hasPendingTasks())
    return false;
  else
    return true;
}

Right

function canWeMoveOn() {
  return !work.hasPendingTasks();
}

Detection

  • [x]Automatic

Based on syntax trees, we can safely refactor the code.

Tags

  • Boolean

Conclusion

Beware of returning booleans.

After the return, you will need an If statement which is also a code smell.

Relations

Code Smell 115 - Return True

Code Smell 101 - Comparison Against Booleans

Code Smell 24 - Boolean Coercions

Code Smell 62 - Flag Variables

Code Smell 102 - Arrow Code

Code Smell 51 - Double Negatives

More Info

Credits

Photo by Morgan Housel on Unsplash

Thanks to Nico K. for this suggestion.


It's not at all important to get it right the first time. It's vitally important to get it right the last time.

Andrew Hunt

Software Engineering Great Quotes


Code Smell 119 - Stairs Code

Nested boolean conditions express a business rule. Not an IF

TL;DR: Avoid checking for boolean expressions and returning an explicit boolean.

Problems

  • Declarativeness
  • Ninja Code
  • Readability
  • Arrow Code

Solutions

  1. Return a boolean business formula value.

Context

When dealing with boolean formulas, it is more readable to show a business boolean formula than a stair of boolean checks followed by returning an explicit true/false;

Sample Code

Wrong

def is_platypus(self):
    if self.is_mammal():
        if self.has_fur():
            if self.has_beak():
                if self.has_tail():
                    if self.can_swim():
                        return True
    return False

# This is also wrong since it is polluted with IFs and not readable by a biologist
def is_platypus(self):
    if not self.is_mammal():
        return False
    if not self.has_fur():
        return False
    if not self.has_beak():
        return False
    if not self.has_tail():
        return False
    if not self.can_swim():
        return False 
    return True

Right

def is_platypus(self):
    return self.is_mammal() && self.has_fur() && self.has_beak() && self.has_tail() && self.can_swim()
  
# We can even group conditions according to animal taxonomies

Detection

  • [x]Automatic

Based on syntax trees, we can safely refactor the code removing the explicit boolean value.

Tags

  • Boolean

Conclusion

Beware of returning booleans.

After the return, you will need an If statement which is also a code smell.

Relations

Code Smell 115 - Return True

Code Smell 118 - Return False

Code Smell 101 - Comparison Against Booleans

Code Smell 24 - Boolean Coercions

Code Smell 62 - Flag Variables

Code Smell 102 - Arrow Code

Code Smell 80 - Nested Try/Catch

More Info

Credits

Photo by Jukan Tateisi on Unsplash

Thanks again to Nico K. for this suggestion.


The real hero of programming is the one who writes negative code.

Douglas McIlroy

Software Engineering Great Quotes


Code Smell 120 - Sequential IDs

Most IDS are code smells. Sequential IDs are also a vulnerability

TL;DR: Don't expose obvious consecutive IDs.

Problems

Solutions

  1. Use non-obvious keys.
  2. Use dark keys or UUIDs.

Context

IDs are a problem when dealing with domain objects.

IDs do not exist in the real-world so, they break our bijection.

We should only use IDs when exposing internal resources to the outer world beyond system boundaries.

These are always accidental problems and should not interfere with our models.

Sample Code

Wrong

class Book {
    private Long bookId; // book knows its ID
    private List<Long> authorIds; // book knows author IDs
}

Book harryPotter = new Book(1, {2});
Book cleanCode = new Book(2, {4});
Book donQuixote = new Book(3, {5});

// We can scrape from now on.

Right

class Author {    
    // .. Author protocol
}

class Book {    
    private List<Author> authors; // book knows authors
    // No strange behavior. just what a book can do
    // Real books don't know about IDs
    // ISBN is accidental to a book. Readers don't care
}

class BookResource {    
    private Book resource; // The resource knows the underlying book
    private id; // The id is the link we provide to external world
}

Book harryPotter = new Book(new Author('J. K. Rowling'));
Book cleanCode = new Book(new Author('Robert Martin'))
Book donQuixote = new Book(new Author('Miguel Cervantes'));
                             
BookResource harryPotterResource = new BookResource(harryPotter, UUID.randomUUID());                             

// Books don't know they id. Just the resource does

Detection

  • [x]Automatic

We can use Pentesting techniques against our system to detect this smell.

Tags

  • Security

Conclusion

In case we need to expose internal objects to the external world, we should use non-obvious IDs.

In this way, we can detect (and block) brute force attacks monitoring the traffic and 404 errors.

More Info

Credits

Photo by Max Bender on Unsplash

Thanks @davidkroell for the KSUID tip.


The only truly secure system is one that is powered off, cast in a block of concrete and sealed in a lead-lined room with armed guards.

Gene Spafford

Software Engineering Great Quotes


And that’s all for now…


The next article will explain 5 more code smells!