Mocking and faking in Laravel - Step-by-Step Execution

Concept Flow - Mocking and faking

Start Test

↓

Create Fake/Mock

↓

Inject into Code

↓

Run Code Using Fake/Mock

↓

Verify Behavior or Output

↓

Test Pass/Fail

↓

End

This flow shows how Laravel tests use mocks or fakes to replace real parts, run code, then check results.

Execution Sample

Laravel

<?php
use Illuminate\Support\Facades\Notification;

public function testUserReceivesNotification()
{
    Notification::fake();
    $user = User::factory()->create();
    $user->notify(new WelcomeNotification());
    Notification::assertSentTo($user, WelcomeNotification::class);
}

This test fakes notifications, triggers one, then checks if it was sent to the user.

Execution Table

Step	Action	Effect	State Change	Verification
1	Call Notification::fake()	Replaces real notification with fake	Notification facade now uses fake	No notifications sent yet
2	Create user with factory	User object created in database	User exists in DB	User ready for notification
3	Call $user->notify(new WelcomeNotification())	Notification sent via fake	Fake records notification sent	Notification recorded but not actually sent
4	Call Notification::assertSentTo($user, WelcomeNotification::class)	Checks if notification was sent to user	No state change	Assertion passes if notification recorded
5	Test ends	Test passes if assertion true	Test complete	Test result: PASS

💡 Test ends after assertion confirms notification was sent to user

Variable Tracker

Variable	Start	After Step 2	After Step 3	Final
Notification facade	Real notification system	Fake notification system active	Fake records notification sent	Fake notification system active
$user	Not created	User object created	User object exists	User object exists
Notifications sent	0	0	1 (WelcomeNotification)	1 (WelcomeNotification)

Key Moments - 3 Insights

Why do we call Notification::fake() before sending notifications?

Calling Notification::fake() replaces the real notification system with a fake one that records notifications instead of sending them. This is shown in execution_table step 1 and 3.

Does Notification::assertSentTo() send a notification?

No, assertSentTo() only checks if a notification was recorded as sent by the fake. It does not send anything. See execution_table step 4.

What happens if we don't fake notifications in the test?

Without faking, notifications would actually send, which is not good for tests. Also, assertSentTo() would fail because no fake records exist. This is why step 1 is important.

Visual Quiz - 3 Questions

Look at the execution table, what is the state of the Notification facade after step 2?

A It uses the fake notification system

B It uses the real notification system

C It has sent one notification

D It is uninitialized

💡 Hint

Check the 'State Change' column at step 2 in the execution_table

At which step does the test verify if the notification was sent?

A Step 1

B Step 4

C Step 3

D Step 5

💡 Hint

Look at the 'Verification' column to find when assertion happens

If Notification::fake() was not called, what would happen to the test?

A Test would pass anyway

B Notification::assertSentTo() would record notifications

C Notification would be sent for real and test might fail

D User would not be created

💡 Hint

Refer to key_moments about why faking is needed before sending notifications

Concept Snapshot

Laravel Mocking and Faking Cheat Sheet:
- Use Notification::fake() to replace real notifications with a fake.
- Run code that triggers notifications.
- Use Notification::assertSentTo() to check if notification was sent.
- Fakes record calls without side effects.
- Always fake before triggering to avoid real actions.

Full Transcript

In Laravel testing, mocking and faking help replace real services with test doubles. For example, Notification::fake() swaps the real notification system with a fake that records notifications instead of sending them. Then, when code triggers notifications, the fake records them. Finally, tests use assertions like Notification::assertSentTo() to verify notifications were sent to the right users. This approach avoids side effects and makes tests reliable. The execution flow starts with faking, then creating data, triggering notifications, asserting, and ending the test. Variables like the Notification facade switch from real to fake, and notifications sent count changes from zero to one. Key points include always faking before sending and understanding that assertions check recorded calls, not send notifications themselves.