Tag: php

What is Middleware and how to create one in Laravel?

It’s best to envision middleware as a series of “layers” for HTTP requests that must pass through before they hit your application. Each layer can examine the request and even reject it entirely.

Middleware provides a convenient mechanism for inspecting and filtering HTTP requests entering your application. It’s best to envision middleware as a series of “layers” for HTTP requests that must pass through before they hit your application. Each layer can examine the request and even reject it entirely.

For example, Laravel includes a middleware that verifies the authenticity of the user of your application. If the user is not authenticated, the middleware will redirect the user to your application’s login screen. However, if the user is authenticated, the middleware will allow the request to proceed further into the application.

To perform different tasks, we can develop many middlewares besides authentication. For example, a logging middleware might log all incoming requests to your application. 

Laravel framework has included many middlewares, including middleware for authentication and CSRF protection. All of these middlewares are located in the app/Http/Middleware directory.

To create a middleware, we can use the following command,

php artisan make:middleware <middleware-name>

For example, if we want to create a middleware for checking transactions, we can run the following command,

php artisan make:middleware CheckTransaction

 After successful execution of the command, a middleware class will be created under the app/Http/Middleware directory.

In this class, we can define methods to check transactions. If the transaction is not completed, we can redirect the user back to the failed transaction page. However, on the successful transactions, we can allow users to proceed to the next page.

<?php
 
namespace App\Http\Middleware;
 
use Closure;
 
class CheckTransaction
{
    /**
     * Handle an incoming request.
     *
     * @param  \Illuminate\Http\Request  $request
     * @param  \Closure  $next
     * @return mixed
     */
    public function handle($request, Closure $next)
    {
        if ($request->input('status') !== 'completed') {
            return redirect('transaction-failed');
        }
 
        return $next($request);
    }
}

As you can see, if the transaction status does not set to “completed”, the middleware will return an HTTP redirect to the client; otherwise, the request will be passed further into the application.

To pass the request deeper into the application (allowing the middleware to “pass”), you should call the $next callback with the $request.

Finding bugs using PHPStan as a Static Analyzer

With PHP being an interpreted language it has a downside when it comes to finding bugs in your code. It will not show you errors in your software until you actually run it. PHPStan tries to solve this problem by doing static analysis on your code. It was recently created by Ondrej Mirtes.

Running PHPStan will tell you about bugs in your codebase almost instantly (yes, it’s very fast). At the time of writing this article, PHPStan currently checks your code on:

  • The existence of classes and interfaces in an instance of, catch type hints, other language constructs, and even annotations. PHP does not do this and just stays silent instead.
  • Existence of variables while respecting scopes of branches and loops.
  • Existence and visibility of called methods and functions.
  • Existence and visibility of accessed properties and constants.
  • Correct types assigned to properties.
  • The correct number and types of parameters are passed to constructors, methods, and functions.
  • Correct types returned from methods and functions.
  • The correct number of parameters passed to sprintf/printf calls is based on format strings.
  • Useless casts like (string) ‘foo’.
  • Unused constructor parameters – they can either be deleted or the author forgot to use them in the class code.
  • That only objects are passed to the clone keyword.

As you can see, it contains a lot of useful checks which will warn you of potential bugs before you even run your code.

Installing PHPStan

Installing PHPStan is as easy as including it in your project through composer:

$ composer require --dev phpstan/phpstan

We can now run PHPStan from the base directory of our project:

$ vendor/bin/phpstan analyze -l 4 src

A breakdown of this command:

  • vendor/bin/phpstan is the executable
  • analyze tells PHPStan to analyze all files in the given directories
  • -l 4 means that we want to analyse on the most strict level
  • src is the directory we want to analyse

Try running this in your own project and see what kind of potential errors are living in your codebase.

Integrating PHPStan into CI

It’s super easy to use PHPStan in Continuous Integration. For most of my personal projects, I use TravisCI. Since we’ve included PHPStan as a dev-dependency in our composer.json file we just have to add the PHPStan executable to the scripts that the CI-software needs to run.

For TravisCI, this means just changing the default script in a .travis.yml like this:

language: php
php:
  - '8.0'
install: composer install

# Simply add these lines
script:
    - vendor/bin/phpunit
    - vendor/bin/phpstan analyse src tests --level=4

The default script that TravisCI runs for PHP projects is simply phpunit. Now we’ve added PHPStan to it. If PHPStan finds any errors within your project, the build will fail.

Error handling in PHP (Part 1)

Error handling is an important part of any developer as it provides vital flaws about the program developed by the developer. So, it becomes very crucial to learn the techniques to manage it.

As a developer, we have been told that you should not show errors on the production server because of the security risk due to the path displayed by the PHP errors displayed on the screen. So, we add the following code for the production server,

ini_set('error_reporting', 0);
error_reporting(0);

ini_set('display_errors', FALSE);

But, without error logs, developers cannot able to know actual problems or flaws in the system. So, rather than hiding errors, developers should store them in the log files. We can achieve this using the following code,

ini_set('error_reporting', E_ALL);
error_reporting(E_ALL);
ini_set('log_errors', TRUE);
ini_set('html_errors', FALSE);
ini_set('error_log', LOG_PATH.'error.log');
ini_set('display_errors', FALSE);

This way, we can manage error logs and hide errors on the production server. We can manage separate log files for the different modules of the project.

How to get the last executed query in PHP CodeIgniter?

Are you wanted to get the last executed SQL query in the CodeIgniter project? then, I will help to get the latest query in CodeIgniter

We can get the last executed query using the last_query() function of the inbuilt db class of the CodeIgniter. This function can be used with simple syntax like $this->db->last_query() to see SQL statements of last executed query in PHP CodeIgniter app. You have to simple code that functions after the main query that you wanted to check.

Here is a simple function code which can be added in any controller of the CodeIgniter project and also output for the last query:

Example:

public function check_query_function() {

    $sql = $this->db->get("products");
  
    $query = $this->db->last_query();
   
    echo "<pre>";
    print_r($query);
    exit;
}

Output:

SELECT * FROM `products`

Best alternative to set a PostgreSQL schema using PHP PDO

To set the PostgreSQL DB connection, schema parameter is not to be included

$Conn = new PDO('pgsql:host=localhost;port=5432;dbname=db', 'user', 'pass');
$result = $Conn->exec('SET search_path TO accountschema');
if ( ! $result) {
die('Failed to set schema: ' . $Conn->errorMsg());
}

Is this a good practice? Is there a better way to do this?

In order to specify the default schema you should set the search_path instead.

$Conn->exec('SET search_path TO accountschema');

You can also set the default search_path per database user and in that case the above statement becomes redundant.

ALTER USER user SET search_path TO accountschema;