Code Monkey home page Code Monkey logo

python-p4-request-response-cycle's Introduction

The Request-Response Cycle

Learning Goals

  • Build and run a Flask application on your computer.
  • Manipulate and test the structure of a request object.

Key Vocab

  • Web Framework: software that is designed to support the development of web applications. Web frameworks provide built-in tools for generating web servers, turning Python objects into HTML, and more.
  • Extension: a package or module that adds functionality to a Flask application that it does not have by default.
  • Request: an attempt by one machine to contact another over the internet.
  • Client: an application or machine that accesses services being provided by a server through the internet.
  • Web Server: a combination of software and hardware that uses Hypertext Transfer Protocol (HTTP) and other protocols to respond to requests made over the internet.
  • Web Server Gateway Interface (WSGI): an interface between web servers and applications.
  • Template Engine: software that takes in strings with tokenized values, replacing the tokens with their values as output in a web browser.

Introduction

Web browsers cannot execute Python code. To make our views show up in the browser, Flask needs to translate HTTP requests into Python objects and new Python objects into HTTP responses. There are plenty of strategies to do this by hand, but Flask makes our lives easier with Werkzeug- you'll find that Flask generates WSGI maps for your application with little to no manual configuration on your part.

Let's take a look under the hood at the request-response cycle in Flask applications.


Application and Request Contexts

When a Flask application gets a request from the browser, it has to pass some specific objects to the view function that will respond to that request. One example is the request object itself, which contains the HTTP request from the browser.

You may have noticed in the past few lessons, though, that view functions do not take request objects as arguments. Since every view function needs access to a request object- among other things- Flask manages requests through contexts.

Let's take a look at a simple view function that uses a request object:

# server/app.py

from flask import Flask, request

app = Flask(__name__)

@app.route('/')
def index():
    host = request.headers.get('Host')
    return f'<h1>The host for this page is {host}</h1>'

if __name__ == '__main__':
    app.run(port=5555, debug=True)

Remember to set the environment variables in the server/ directory as well:

$ export FLASK_APP=app.py
$ export FLASK_RUN_PORT=5555

Running this app and navigating to 127.0.0.1:5555, you should see the following:

Simple web page with the text "The host for this page is 127.0.0.1:5555"

This might seem a bit odd. While we did import request, we didn't assign it any attributes that would tell it about the activity on our server. It almost seems like a global variable- one that is set outside of our application instances and views that provides an unchanging set of information. That wouldn't make much sense as an implementation, though: our application will be handling many requests, often overlapping with one another while multiple users are visiting our website.

Flask generates a context for requests after receiving a request and before our application runs. When request is called, we have access to all of that request data without having to do any configuration or include a request argument to our view.

If you're testing request for a specific app outside of any views, you will have to generate this context manually. This only requires two commands:

# example only- no need to write this out

from app import app
from flask import request

request_context = app.test_request_context()
request_context.push()

NOTE: If you are working in a debugging shell, it is a good idea to clear out your request_context object with its pop() method before moving onto a new request.

As the name test_request_context() suggests, this only provides test data. You won't use this in your applications, but you might see it in the tests for Phase 4's labs.

Flask also has an application context. This works very much the same way as a request context: when a request is received, Flask generates the application context. When the application instance is accessed, the application context becomes available to our app and views. This provides information on the application we're working on and is accessible through the flask.current_app object.

# server/app.py

from flask import Flask, request, current_app

app = Flask(__name__)

@app.route('/')
def index():
    host = request.headers.get('Host')
    appname = current_app.name
    return f'''<h1>The host for this page is {host}</h1>
               <h2>The name of this application is {appname}</h2>'''

if __name__ == '__main__':
    app.run(port=5555, debug=True)

Flask also provides us two unique objects that allow us to manipulate request data more effectively:

  • g is an object that can be used to store anything that you want to store globally for the lifetime of a request. It is reset with each new request.
  • session is a dictionary object that can be used to hold onto values between multiple requests.

Handling Requests

Now that we've seen how a request object is created, let's take a look at how requests are used by Flask applications.

The URL Map

Every time an application receives a request, it has to decide which view to run to handle its data. Routes were helpful for defining where to serve our application's views, but now we need to use our application's URL map to find the right view from a URL. The URL map is just as it sounds: a dictionary that maps URLs to the views that serve the client at those URLs. Every time we use the @app.route decorator, a new mapping is added to the URL map.

We can view the URL map for ourselves in an ipdb shell. Run python debug.py from the base directory for this lesson and enter the following:

$ ipdb> app.url_map
# => Map([<Rule '/static/<filename>' (GET, HEAD, OPTIONS) -> static>,
 <Rule '/' (GET, HEAD, OPTIONS) -> index>])

The first Rule we see exists by default: it is a special route that gives us access to static files through the client. The second, referring to the '/' URL, describes the index() view that we made above.

GET, HEAD, and OPTIONS are request methods that are accepted through the routes. All HTTP (and HTTPS) requests must be issued with a request method to indicate what task the server is meant to carry out. GET requests a representation of the resource- this is the most common method your requests carry when you're surfing the web. HEAD is the same as GET, but requests that we leave out the body of the response. OPTIONS requests a list of the HTTP methods that a resource will accept. Flask will include all three by default- but you can remove them or add more later on if you'd like!

For more on HTTP request methods, visit Mozilla HTTP Request Methods.

Request Hooks

As you build out a variety of Flask web applications, you will notice that there are many tasks that you want to carry out before or after most of your view functions. This could be as simple as generating a reminder message or as complex as multi-factor authentication- either way, you'll want to handle these with hooks.

Hooks are best implemented as decorators. There are four types of hooks:

  1. @app.before_request: runs a function before each request.
  2. @app.before_first_request: runs a function before the first request (but not subsequent requests).
  3. @app.after_request: runs a function after each request.
  4. @app.teardown_request: runs a function after each request, even if an error has occurred.

Let's set up a hook so that our views all know where our application files are located:

# server/app.py

import os

from flask import Flask, request, current_app, g

app = Flask(__name__)

@app.before_request
def app_path():
    g.path = os.path.abspath(os.getcwd())

@app.route('/')
def index():
    host = request.headers.get('Host')
    appname = current_app.name
    return f'''<h1>The host for this page is {host}</h1>
            <h2>The name of this application is {appname}</h2>
            <h3>The path of this application on the user's device is {g.path}</h3>'''

if __name__ == '__main__':
    app.run(port=5555, debug=True)

After you restart your server, you should see that our hook has been run and g modified such that index() now knows where it lives on your computer. (Now it can find its way home if it ever gets lost!)

Index page from before with h3 text beneath that says "The path of this application on the user's device is /Users/benbotsford/Documents/new-curriculum/intro-to-flask/python-p4-request-response-cycle"


Creating Responses

When a view function spins up, Flask gets ready for an HTTP response as a return value. This can be a simple string, a multi-line string of HTML, or a combination of strings and codes.

An important part of any response is the HTTP status code. Flask sets this to 200 by default, which indicates that the request successfully reached the specified resource and an appropriate response was generated. When we need to send a different status code, we can simply add this as a second return value after the response body:

# index() in server/app.py

@app.route('/')
def index():
    host = request.headers.get('Host')
    appname = current_app.name
    return f'''<h1>The host for this page is {host}</h1>
            <h2>The name of this application is {appname}</h2>
            <h3>The path of this application on the user's device is {g.path}</h3>''', \
            202

202 is the "Accepted" status code. This signifies that a request has been received by the server, but that the server has not done anything about it yet. We could also return 204 if there were no content on the page, or 404 if the URL was not found.

For more on HTTP status codes, visit the Mozilla documentation.

There is a third, optional argument that can be added in to create headers for our response. This is simply a dictionary with keys for the header attributes mapped to their respective values.

Response Objects

For a more object-oriented approach to responses, you can use Flask's make_response() function. This takes 1-3 arguments in the same format as our earlier response: a body string, a status code, and a headers dictionary, respectively.

# index() in server/app.py
...
from flask import make_response
...

@app.route('/')
def index():
    host = request.headers.get('Host')
    appname = current_app.name
    response_body = f'''
        <h1>The host for this page is {host}</h1>
        <h2>The name of this application is {appname}</h2>
        <h3>The path of this application on the user's device is {g.path}</h3>
    '''

    status_code = 200
    headers = {}

    return make_response(response_body, status_code, headers)

This won't change what you see in the browser, but it will make your code cleaner and easier to replicate (even automate!) in other views.

For more on response objects, visit API Flask Documentation.

Special Responses

There are specific cases where your response is meant to do something other than display an HTML body in the browser. Two cases are most common: redirects and aborts.

The redirect() function is usually delivered with a "301: Moved Permanently" or "302: Found" status code. These signify that the URL for the resource on our server has been changed. Rather than torture the user with trying to find the new URL, we can redirect them to the current URL for the resource.

redirect() is very simple: it takes one argument, the URL for the relocated resource.

# example only

from flask import redirect

@app.route('/reginald-kenneth-dwight')
def index():
    return redirect('names.com/elton-john')

If you navigate to the above link, you'll notice a "404: Not Found" status code! This means that the resource does not exist for the website in question. If we want to inform users of this error in our applications, we need to use Flask's abort() function:

# example only

...
from flask import abort, make_response
...

@app.route('/<stage_name>')
def get_name(stage_name):
    match = session.query('StageName').filter(StageName.name == stage_name)[0]
    if not match:
        abort(404)
    return make_response(f'<h1>{stage_name} is an existing stage name!</h1>')

Conclusion

This has been a brief introduction to requests and responses in Flask. These messages allow our application to communicate with a client through the server. Flask does most of the work for us when setting up contexts and URL maps, but parsing requests and generating responses effectively will allow you to do much more with Flask than you've seen so far. Now that we've discussed communication with the frontend, let's get started on communication with the backend through SQLAlchemy.

If you don't fully understand every concept yet, don't worry! Check below to make sure your code matches ours and look back at this lesson if you're having trouble with contexts, requests, or responses in the future.


Solution Code

#!/usr/bin/env python3

import os

from flask import Flask, request, current_app, g, make_response

app = Flask(__name__)

@app.before_request
def app_path():
    g.path = os.path.abspath(os.getcwd())

@app.route('/')
def index():
    host = request.headers.get('Host')
    appname = current_app.name
    response_body = f'''
        <h1>The host for this page is {host}</h1>
        <h2>The name of this application is {appname}</h2>
        <h3>The path of this application on the user's device is {g.path}</h3>
    '''

    status_code = 200
    headers = {}

    return make_response(response_body, status_code, headers)

if __name__ == '__main__':
    app.run(port=5555, debug=True)

Resources

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.