What they never tell you about GitHub

GitHub is brilliant…. There are loads of tutorials, video and articles out there, recently I’ve spent time filling gaps in my Git version control knowledge and something struck me as odd. Every tutorial should make it clear right from the start and reminders throughout their presentations that you

DO NOT STORE PASSWORDS, KEYS, USERNAMES, CREDENTIALS, ETC. ON GIT HUB

Why?

Because there are bastards out there who are scanning git repositories for anything that could help access secured areas or consume resources they should not.

Git does scan your repo’s and will warn you but its not in real time so be warned. Still not convinced? The image is of a free Amazon Web Service account that was set up as a learning exercise the keys where saved in a Git Repo and over the course of 3 days a free account had racked up nearly $10,000US .

What can you do, if you accidentally commit a password, key, etc? Delete the branch? Roll back?

No!

The best advice and this is also given by Git themselves is to consider your data compromised and your accounts that use the data hacked. Once again, rolling back and deleting does not work.

So reset your passwords/key etc and don’t just add a 1 at the end or swap out vowels for 12345.

How do you manage password and environment values?

Have a Google for git .gitignore or visit dotIgnore on GitHub

JavaScript fetch()

In a previous post I was refreshing my memory with vanilla JavaScript  after years of jQuery-ing, especially AJAX request and the

XMLHttpRequest()

My eyes rolled when I was reading more about it then  I found fetch()  ( I know its been about since 2015 and now almost universally adopted).

Fetch does what XMLHttpRequest does but in a more elegant way and DOES NOT need additional libraries as it is bundled with JavaScript. It also makes use of “promises” .

const url = "https://randomuser.me/api/?results=10";

fetch(url)
.then(function(response) {
 return response.json();
})
.then(function(data) {
 console.log(data.results);
})
.catch(function() {
 console.log("Booo");
});

The above code queries the URL, then parses the string as JSON, then does something with the data.

It also catches errors.

Good eh?

Its looks to be easy to expand and do all manner of fetching. For example submitting a form

var form = new FormData(commentForm);
fetch( url, { method: "POST", body: form })

You can then chain .then() promises to get the desired outcomes.

For more information

A quick introduction video from Google

From Google Working with the Fetch API

Jake Archibald’s post Thats so Fetch

Scotch post The Fetch API

 

 

 

 

 

strpos() not working?

PHP strpos() not working as expected? It could be a  “non-strict” comparison problem.

When using strpos() to determine whether a substring exists within a string the  results can be misleading: Remember FALSE == 0?

Consider the following:

$quote = 'Dave rocks';
 
if (strpos($quote, 'Dave')) {
    echo 'Dave is awesome.';
} else {
    echo 'Dave is not awesome.';
}

strpos() Returns position is 0 ( zero ) that is evaluated as FALSE so, “Dave is not awesome”.

Much better. In this case adding the strict comparison === ( 3 equals ) to the “if” statement asks if strpos() returns a number and  is not strictly FALSE. So, “Dave is awesome”

$quote = 'Dave rocks';
 
if (strpos($quote, 'Dave') !== FALSE) {
    echo 'Dave is awesome.';
} else {
    echo 'Dave is not awesome.';
}

For more see PHP.net

 

 

 

New querySelector() and querySelectorAll()

Catching up with JavaScript after years of using jQuery I found these two shortcuts for replacing longer getElementById() methods. NOTE: the querySelector is passed an ID in the CSS format

<button id="our-button">Add New Item</button>

<script>
    // These are both the same
    var ourButton = document.getElementById("our-button");
    var ourButton = document.querySelector("#our-button");
</script>
<ul id="our-list>
    <li>A thing</li>
    <li>Another thing</li>
    <li>A new thing</li>
</ul>


<script>
// These are both the same
   var listItems = document.getElementById('#our-list').getElementsByTagName("li");
   var listItems = document.querySelectorAll("#our-list li");
</script>

 

 

Javascript AJAX without jQuery

I learnt this stuff years ago but with the rise of jQuery I like many others had become lazy. Add the fact that so many browsers handled things slightly different going down the library route had its benefits.

It is 2017 and things have improved many jQuery(ish) approaches have been adopted by Javascript and browsers have become more similar.

So here goes. First off a GET request to retrieve a Chuck Norris joke

<script>
var url = "http://api.icndb.com/jokes/random";
var request = new XMLHttpRequest();

request.open('GET', url, true);

request.onload = function() {
 if (request.status >= 200 && request.status < 400) {
 // Success!

printJokes(request.responseText);

} else {
 // We reached our target server, but it returned an error
 console.log('There was a problem. Status:' + request.status);

}
};

request.onerror = function() {
 console.log('There was a problem!');
};

request.send();


var printJokes = function(jokes){

JSON.parse(jokes, (key, value) => {
 key == 'joke' ?
 console.log(value) : ''
 });

}
</script>

 

 

 

Toggling in Javascript, PHP and possibly other environments

For a long time if I wanted to change the state of a variable between true and false (toggling). In PHP I would do something like…

$switch = true;
if($switch == true){
    $switch = false;
}else{
    $switch = true
}

That worked fine, then there came Ternary shorthand

$switch = true;
$switch = ($switch == true ? false : true);

Not bad eh? Much easier to read. Then came this

$switch = true;
$switch = !$switch;

If you read it out loud it sounds like, ‘Switch is not true?’ So the answer would be either

True = $switch is false

Or

False = $switch is true

Javascript reminder #1 Spread syntax

Okay, a quick reminder on Spread Syntax

If you have a Javascript function that you call with

average(1, 2, 3);

function average(x, y, z){
// Average of the three numbers
}

You are limited to just three properties/arguments. Not very useful if you have more or less numbers you want to find an average. Using Spread Syntax we don’t have to worry about the number of values passed

function average(...args) {
  var sum = 0;
  for (let value of args) {
    sum += value;
  }
  return sum / args.length;
}

average(2, 3, 4, 5); // 3.5

The Spread Syntax ( the three dots{name}) includes all the unknown values.  It gets better still

var averageMethod = 'mode';
var roundTo = 2; /Decimal places
doThing( averageMethod, [2,4,3,6,10,19], roundTo);

function doThing(method, ...valuesToAverage, roundTo)
{
// Calculate averages
}

And Spread Syntax can be used when creating arrays

var parts = ['shoulders', 'knees']; 
var lyrics = ['head', ...parts, 'and', 'toes']; 
// ["head", "shoulders", "knees", "and", "toes"]

Good eh?

 

Composer – Init, Install and update

Assuming that composer is installed on your machine what next?

If you do not have a composer.json file lets create one. In your command line ‘cd’ into your projects root folder. Then type

composer init

and answer the questions. Easy.

Look at your projects root directory and you will see a composer.json file. This is where you can add/remove packages, set namespaces, set files that are always load, run scripts, and lots of other good stuff.

Do you see a vendor/ directory? No? At this stage type

composer install

This actually ignores the .json file and reads or creates a composer.lock file then installs all the packages and dependencies required ( as well a bunch of other stuff – see the end of the post). The lock file is important when there are multiple developers/users it ensures that we/you are all using the exact same versions of installed packages.

Now lets pretend that you have cloned a project from GitHub that uses composer. You wont see a vendor/ directory but you will have a lock file.

composer install

Now you have a vendor/ directory and all is right with the world.

If you try and  run install again it won’t do anything ( try it if you want )

If you want to add packages use the

composer require name\package

If you want to update the composer files and logs

composer update

Updates your dependencies to the latest version according to composer.json, and updates the composer.lock file.

“A bunch of other stuff”

Composer can do a whole lot more. Including

  • Running scripts after installing, updating, etc
  • Autoload files. For example helper files with common functions
  • Pulling packages from GitHub. Great if you are writing your own
  • Lets you have a development environment and production sets of packages. E.g. so you can have PHPUnit just on your local machine.

Laravel Validation of unique fields

Laravel Validation of unique fields e.g email

Basic server side validation in Laravel is ridiculously easy. The following inside a controller will validate the passed values and redirect back if it fails.

$this->validate($request, [

            'firstName' => 'required',

            'lastName' => 'required',

            'email' => 'required|unique:users',

            'password' => 'required|confirmed',

            ]);

You can see what fields are required; the email has to be unique in the database users table and the password is required and must match a field named ‘password_comfirm’. When this validation passes you can then add the new user.

What about UPDATING or (PUT/PATCH) the user form is returned, the firstName has changed and the email remains the same. The original validation code will not pass and return an error of ‘email already in use’.

Solution

$this->validate($request, [

            'firstName' => 'required',

            'lastName' => 'required',

            'email' => 'required|unique:users,email,'.$id.'',

            ]);

The email’s unique rule now looks up the users table, looks for the email address except where the id column matches $id

For more information: https://laravel.com/docs/5.4/validation#rule-unique

 

PHP Don’t Abbreviate

Simply put, please do not abbreviate in your code.

Always code as if the guy who ends up maintaining your code will be a
violent psychopath who knows where you live.  Code for readability. John F Wood

If others have to read it they will have no idea what an abbreviation means and when you come back to it in 6 months time will you remember what this does?

<?php
function delHisFile($p) { 
// Do something
 }

Or

function sort(array $a) {
 for ($i = 0, $c = count($a);  $i < $c;  $i++) { 
// Do something
 }
}

How does this read?

<?php
function deleteHistoryFiles($path) { 
// Do something
 }


function sort(array $array) {
 for ($index = 0, $array_count = count($array); $index < $array_count;  $index++) { 
// Do something
 }
}

When I started out coding for some reason I thought variables, functions, methods and class names that where long and descriptive would have an adverse effect on performance ( I know daft eh) .