Introduction

Welcome to this journey into the world of web servers and the Flask framework! In the previous weeks, you’ve successfully set up a web server using GitHub Pages, converting Jupyter Notebooks into Markdown for a seamless online presentation. Today, we’ll take that knowledge to the next level as we dive into creating your very own web server using Flask.

Understanding Web Servers

What is a Web Server?

Traditionally, we had librarians at libraries that would help you find books or information. Today in the digital world, thousands upon thousands of home pages, search engines, and digital archives have been built using web servers.

GitHub Pages vs. Flask

You’ve already experienced a form of web server through GitHub Pages. Think of GitHub Pages as a library that has established rules for publishing Markdown notes and Jupyter Notebooks neatly on a bookshelf.

Now, let’s introduce Flask, your personal web server. Flask can create and manage any type of content, including customizing everything according to your preferences, and even serve additional information (like a database with APIs).

The Flask Framework Flask is a micro web framework written in Python. It’s designed to be minimal and easy to use, making it perfect for building web applications, APIs, and, yes, even your web server. Today, we will start with the basics of Flask and see how it empowers you to create and manage web content.

Our Goals for Today

Here’s what we’ll accomplish in this session:

  • Create a minimal Flask server.
  • Explore the Python/Flask process.
  • Access data from our Flask server using Python.
  • Access data from our Flask server using JavaScript.
  • Learn how to stop the Python/Flask process gracefully.

Install required libraries

These libraries are required to run and interact with the Python web server.

!pip install flask flask-cors requests

Requirement already satisfied: flask in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (3.1.0)
Requirement already satisfied: flask-cors in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (5.0.0)
Requirement already satisfied: requests in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (2.32.3)
Requirement already satisfied: Werkzeug>=3.1 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from flask) (3.1.3)
Requirement already satisfied: Jinja2>=3.1.2 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from flask) (3.1.4)
Requirement already satisfied: blinker>=1.9 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from flask) (1.9.0)
Requirement already satisfied: click>=8.1.3 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from flask) (8.1.7)
Requirement already satisfied: itsdangerous>=2.2 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from flask) (2.2.0)
Requirement already satisfied: urllib3<3,>=1.21.1 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from requests) (2.2.2)
Requirement already satisfied: charset-normalizer<4,>=2 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from requests) (3.3.2)
Requirement already satisfied: certifi>=2017.4.17 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from requests) (2024.8.30)
Requirement already satisfied: idna<4,>=2.5 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from requests) (3.8)
Requirement already satisfied: MarkupSafe>=2.0 in /home/anvayv/nighthawk/anvay_2025/venv/lib/python3.10/site-packages (from Jinja2>=3.1.2->flask) (2.1.5)

Start Web Server

This Python code provides a simple server with an accessible API.

Note: Jupyter magic commmand %%python --bg that follows runs the server in background. This enables us to continue interacting with the subsequent Notebook cells.

%%python --bg

from flask import Flask, jsonify
from flask_cors import CORS

# initialize a flask application (app)
app = Flask(__name__)
CORS(app, supports_credentials=True, origins='*')  # Allow all origins (*)

# ... your existing Flask

# add an api endpoint to flask app
@app.route('/api/anvay')
def get_anvay():
    # start a list, to be used like a information database
    InfoDb = []

    # add a row to list, an Info record
    InfoDb.append({
        "FirstName": "Anvay",
        "LastName": "Vahia",
        "DOB": "January 29",
        "Residence": "North Pole",
        "Email": "anvayv22800@stu.powayusd.com",
        "Owns_Cars": ["None"]
    })
    
    return jsonify(InfoDb)

# add an HTML endpoint to flask app
@app.route('/')
def say_hello():
    html_content = """
    <html>
    <head>
        <title>Hellox</title>
    </head>
    <body>
        <h2>Hello, World!</h2>
    </body>
    </html>
    """
    return html_content

if __name__ == '__main__':
    # starts flask server on default port, http://127.0.0.1:5001
    app.run(port=5001)

Explore the Python/Flask process with Linux

This script discovers the running flask process on your machine using Linux commands.

  1. lsof - list open files
  2. lsof and awk return the process id, so ps can list details, the vericle bar is called a pipe. A pipe flows output from one command to the next.
  3. curl is a Linux utiltity that is easiest way to test if web server is responding
%%script bash

# After app.run(), see the the Python open files on port 5001
echo "Python open files on port 5001" 
lsof -i :5001
# see the the Python process 
echo
echo "Python process"
lsof -i :5001 | awk '/Python/ {print $2}' | xargs ps
# show ontent of the Python server using curl
echo
echo "Content of the Python root endpoint (aka /), using curl",  
curl http://localhost:5001/



Python open files on port 5001
COMMAND  PID   USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
python  2833 anvayv    3u  IPv4  44288      0t0  TCP localhost:5001 (LISTEN)

Python process


your 131072x1 screen size is bogus. expect trouble


    PID TTY          TIME CMD
    414 pts/0    00:00:00 sh
    424 pts/0    00:00:00 sh
    532 pts/0    00:00:00 sh
    544 pts/0    00:00:08 node
    769 pts/0    00:00:05 node
    905 pts/0    00:00:24 node
    957 pts/0    00:00:00 node
   1165 pts/0    00:00:01 pet
   1327 pts/0    00:00:00 node
   1341 pts/0    00:00:03 node
   1566 pts/0    00:00:01 python
   1583 pts/0    00:00:09 node
   1772 pts/0    00:00:00 node
   2833 pts/0    00:00:00 python
   2857 pts/0    00:00:00 bash
   2866 pts/0    00:00:00 xargs
   2867 pts/0    00:00:00 ps

Content of the Python root endpoint (aka /), using curl,


  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   135  100   135    0     0  46503      0 --:--:-- --:--:-- --:--:-- 67500



    <html>
    <head>
        <title>Hellox</title>
    </head>
    <body>
        <h2>Hello, World!</h2>
    </body>
    </html>

Access data from our Flask server using Python

The code block below shows alternate ways to access the Web Server.

  1. Import requests and use it to obtain response from endpoints
  2. The response is a Python object that contains status codes and data
  3. The data can be in different forms, we will be focussed on JSON responses in Full-Stack.
import requests
from IPython.display import HTML, display

# call api root endpoint (aka '/'), often called home page
response = requests.get('http://127.0.0.1:5001/')
# output response in different forms
print("Print Status Message:", response)
print("\nPrint Raw HTML:\n", response.text)
display(HTML(response.text))

# call unknown api endpoint
response = requests.get('http://127.0.0.1:5001/unknown-page')
print("Print Status Message:", response)

Print Status Message: <Response [200]>

Print Raw HTML:
 
    <html>
    <head>
        <title>Hellox</title>
    </head>
    <body>
        <h2>Hello, World!</h2>
    </body>
    </html>
Hellox

Hello, World!

Print Status Message: <Response [404]>

import requests
# an api endpoint most commonly returns JSON data
response = requests.get('http://127.0.0.1:5001/api/data')
response.json()

[{'DOB': 'October 21',
  'Email': 'jmortensen@powayusd.com',
  'FirstName': 'John',
  'LastName': 'Mortensen',
  'Owns_Cars': ['2015-Fusion',
   '2011-Ranger',
   '2003-Excursion',
   '1997-F350',
   '1969-Cadillac'],
  'Residence': 'San Diego'},
 {'DOB': 'February 27',
  'Email': 'slopez@powayusd.com',
  'FirstName': 'Shane',
  'LastName': 'Lopez',
  'Owns_Cars': ['2021-Insight'],
  'Residence': 'San Diego'}]

Access data from our Flask server using JavaScript

This sample is very similar to Full-Stack as the JavaScript is running through Jupyter and the Web server is a Python Process running on our machine (local server).

  1. HTML is used to setup basics of a table
  2. The script block, has javascript fetch that passes endpoint (url) and options. The options are critical to communicating request requirements.
  3. Similar to python examples, data is extracted and that data is written to the document, which is what is viewable to the user as the page is rendered. Headings are static in the document, but rows are dynamically extracted according to the information contained in the server.
%%html

<h1>Access data from our Flask server using JavaScript</h1>

<p>This code extracts data "live" from a local Web Server with JavaScript fetch.  Additionally, it formats the data into a table.</p>

<!-- Head contains information to Support the Document -->


<!-- HTML table fragment for page -->
<table id="demo" class="table">
  <thead>
      <tr>
          <th>First Name</th>
          <th>Last Name</th>
          <th>Residence</th>
      </tr>
  </thead>
  <tbody id="result">
    <!-- javascript generated data -->
  </tbody>
</table>

<script>
  (function() {
    let resultContainer = document.getElementById("result");

    let url = "http://127.0.0.1:5001/api/data";

    let options = {
      method: 'GET', 
      mode: 'cors',
      cache: 'default', 
      credentials: 'include',
      headers: {
        'Content-Type': 'application/json',
      },
    };

    fetch(url, options)
      .then(response => {
        if (response.status !== 200) {
          console.error(response.status);
          return;
        }
        response.json().then(data => {
          console.log(data);
          for (const row of data) {
            const tr = document.createElement("tr");
            const firstname = document.createElement("td");
            const lastname = document.createElement("td");
            const residence = document.createElement("td");

            firstname.innerHTML = row.FirstName;
            lastname.innerHTML = row.LastName;
            residence.innerHTML = row.Residence;

            tr.appendChild(firstname);
            tr.appendChild(lastname);
            tr.appendChild(residence);

            resultContainer.appendChild(tr);
          }
        });
      });
  })();

</script>

Access data from our Flask server using JavaScript

This code extracts data "live" from a local Web Server with JavaScript fetch. Additionally, it formats the data into a table.

First Name Last Name Residence

Stop the Python/Flask process

This script ends Python/Flask process using pipes to obtain the python process. Then echo the python process to kill -9.

%%script bash

python_ps=$(lsof -i :5001 | awk '/python/ {print $2}')
echo "Killing python process with PID: $python_ps"
echo $python_ps | xargs kill -9

Killing python process with PID: 2833

Hacks

Edit, stop and start the web server.

  • Add to the Home Page
  • Add your own information to the Web API
  • Use from Template to start your own Team Flask project https://github.com/nighthawkcoders/flocker_backend

Tech Talk Notes:

  • BACKEND (not FRONTEND)
  • Front matter in YAML, runs Github pages
  • Living in San Diego
    • TECHHHHHHH
      • Why? Dont freeze your toes off in Utah
      • YOU NEED TO LEARN TEECHHHHH
  • Understand web systems to part of the digital community
    • Top 5 companies are all TECHH

API

  • Enpoints in the /api directory
  • NEED to know RESTFul APIs and what they do

Bash Commands

  • cURL, gets text from HTTP (Hyper Text Transfer Protocol)
  • lsof, finding running PID’s and ports