Intro to Backend Dev
Web development is often split into two categories: backend and frontend. It's easy for beginners to get confused between the two, especially because there is often some overlap.
Backend development involves writing code and processes that run only on a web server, whereas frontend code runs in the user's web browser.
Frontend code is publicly visible and includes a website's HTML, CSS and JavaScript resources, while backend code is almost always hidden from the user. Some examples of backend tasks include:
Fetching and preparing data from a database
Authenticating a user when they log into a secure system
Performing complex calculations or CPU-intensive tasks
Due to security and performance reasons, it's easy to see why these tasks shouldn't happen on the user's device.
The backend code for ubyssey.ca is mainly concerned with fetching data from a database (usually news articles), preparing the data and sending it back to the user. This process looks something like this:
A user sends a request to a URL (e.g. https://www.ubyssey.ca/news/max-holmes-wins-vp-academic/​
The web server receives the request and parses any relevant information (e.g. the article slug "max-holmes-wins-vp-academic")
The server then loads the article data from the database
After loading the data, it is converted into an HTML document that is returned to the user's web browser
The rest of this tutorial will be covering the above steps in more detail.
The backend code for ubyssey.ca is written in the Python programming language. We choose to use Python for several reasons, the main one being that its fairly easy to learn and many people are already familiar with it. More specifically, our platform is built on top of the popular Django web framework, so much of the heavy lifting (routing, authentication, database access) is already handled for us.
If you want to learn more about Python, check out the resources page. However, this tutorial is designed for beginners who have never used Python before.
A Universal Resource Locator (URL), often called a web address, is what a web browsers uses to access resources stored on a web server. Think of it like a mailing address for a web page. The first part of a URL, called the hostname (e.g. ww.ubyssey.ca), is converted to an IP address (e.g. 216.239.32.21) through something called a DNS lookup, which tells the browser which server to send the request to. The server is then responsible for "routing" the remaining portion of the URL (e.g. /news/max-holmes-wins-vp-academic/), which is often the first task that a web backend must take care of.
A web server should be able to understand the structure of a URL and send incoming requests to the right place. If the server can't understand the URL or if the URL is invalid, a proper error message should be returned.
Luckily for us, Django has a set of tools that make this process relatively straightforward. We simply tell Django the structure of a URL and it will route incoming requests to functions that we define. For this tutorial, we'll be focusing on the homepage, which happens to have the simplest URL pattern:
# ubyssey/urls.py​from django.conf.urls import url​theme = UbysseyTheme()​urlpatterns = [url(r'^$', theme.home),​# More patterns below...]
URL patterns are defined using regular expressions, which give us a way to match multiple strings against a defined structure. For example, here's the pattern for article URLs, which always contain a section name, followed by a slash, and then the article slug:
url(r'^(?P<section>[-\w]+)/(?P<slug>[-\w]+)/$', theme.article)​'/news/max-holmes-wins-vp-academic/' # Match'/sports/fall-softball-classic-2017/' # Match'/this-is-not-an-article/' # Not a match!
You'll notice that Django's url helper takes a second parameter, which is the method to send the request to. In the homepage example, if the URL matches an empty string, the request is routed to theme.home . The definition for theme.home lives in ubyssey/views/main.py:
def home(self, request):frontpage = ArticleHelper.get_frontpage(sections=('news', 'culture', 'opinion', 'sports', 'features', 'science'),max_days=7)​frontpage_ids = [int(a.id) for a in frontpage[:2]]​sections = ArticleHelper.get_frontpage_sections(exclude=frontpage_ids)​try:articles = {'primary': frontpage[0],'secondary': frontpage[1],'thumbs': frontpage[2:4],'bullets': frontpage[4:6],​}except IndexError:raise Exception('Not enough articles to populate the frontpage!')​popular = ArticleHelper.get_popular()[:5]​blog = ArticleHelper.get_frontpage(section='blog', limit=5)​title = '%s - UBC\'s official student newspaper' % self.SITE_TITLE​context = {'title': title,'meta': {'title': title,'description': 'Weekly student newspaper of the University of British Columbia.','url': self.SITE_URL},'articles': articles,'sections': sections,'popular': popular,'blog': blog,'day_of_week': datetime.now().weekday(),}​return render(request, 'homepage/base.html', context)
The above code might look a little daunting if you're new to Python, but don't worry -- we'll be focusing mainly on the last several lines, specifically the definition for context and the final return statement.
Before we go any further, you should take a minute to set up a code editor if you haven't already. We recommend using something like Atom or VS Code, but you can use vim or emacs if you're into that. Once you have an editor installed, load the ubyssey.ca folder so that all the files are in one place.
Hint: In Atom, you can click File > Add Project Folder and select the ubyssey.ca folder.
The home method is responsible for sending an HTML response back to the user, but before it can do that, it has to fetch and prepare all the data it needs. Most of the calls to ArticleHelper fetch data from our database, which is then added to the context structure at the end of the method.
To keep things simple for this tutorial, we won't be fetching data from the database, but we will be adding data to the context structure.
The context data structure is a primitive type called a dictionary, which is just Python's version of a hash table. It simply creates a map of keys to values:
my_dict = {'name': 'John','age': 22,}​print 'My name is ' + my_dict['name']# > My name is John​print 'My age is ' + str(my_dict['age'])# > My age is 22
The data that we need to generate, in this case, is a random number between 1 and 10. Here's how you'd do that in Python:
import random​rand_num = random.randint(1, 10)
Extend the context data structure to include a random number from 1 to 10.
The final line in the home method makes a call to the render method, which (you guessed it!) renders the final output of the homepage. You'll notice that it takes three arguments:
request- the Django request object, stores information about the incoming requesthomepage/base.html- the template file to usecontext- the context to pass to the template
In this case, the render method loads the homepage/base.html template and returns the result of rendering that template with the given context.
This simple example shows how a Django template reads data from the context:
context = {'name': 'Sarah',}​return render(request, 'my-template.html', context)
In a Django template, all of the keys defined in the context dictionary become available as named variables. To print one of these variables, simple wrap its name in double curly braces:
<strong>Hi, my name is {{ name }}!</strong>
This will produce:
Hi, my name is Sarah!
Edit templates/homepage/base.html so that it prints the following at the top of the page:
I completed the getting started tutorial! YOUR RANDOM NUMBER HERE
Hint: add your code directly after this line.
Now if you reload the homepage, you should see your text!