- Explain how to diagram a request/response cycle for a user interaction
- Create and use a diagram to follow the flow of data from a user interaction
- Use a client to make HTTP requests to a server
- Evaluate the differences using a core library, a 3rd party library, a command line app, a web or desktop application, or the browser as a client to make HTTP requests.
Fork and clone this repository to your local machine
$ git clone git@github.com:[username]/js-dom-requests-responses.git && cd js-dom-requests-responses
$ npm ci # to install dependencies
$ node app.js # run the server
# then point your browser to `localhost:3000`- It's a quick way to figure out and cement your understanding of how a process works.
- It's much quicker than writing the code that enacts the process.
- It's much easier to iterate on a process model than the code that enacts the process.
- It's much easier to add detail to a process model than to add it to code.
- It's a great way to communicate and discuss a process with another person.
While your server is running...
Model each of the processes in the processes to model section below.
-
Prioritise the clarity and detail of your model over getting through all the processes listed below.
-
Make sure that your models include the following:
- The order that things happen in.
- The HTTP requests and the data they carry. (Resources for this at the bottom of this readme.)
- The HTTP responses and the data they carry. (Resources for this at the bottom of this readme.)
- The movement of requests between client and server.
- Mouse clicks.
- What is displayed in the browser.
You can use chrome dev-tools to monitor what is happening on the network.
A user visits http://localhost:3000 and they are shown this HTML:
<h1>Welcome!</h1>
<a href="/name">Name Me!</a>
<h3>Cats!</h3>
<img width="500px" src="https://i.inews.co.uk/content/uploads/2021/11/PRI_209429745-e1636480034415.jpg" alt="beauty1">
<img width="500px" src="cat.png" alt="beauty2">
Don't forget to model the image requests and responses.
A user tries to visit http://localhost:3000/, but mistypes the url as http://localhost:3000/whoops in their browser.
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Error</title>
</head>
<body>
<pre>Cannot GET /whooops</pre>
</body>
</html>A user navigates to the homepage, and then clicks the link 'Name me!' and is shown this HTML:
<h1 class="main"></h1>
<p>Using default form submission (GET)</p>
<form action="/name" method="get">
<input type="text" name="name">
<input type="submit" value="go!">
</form>
<p>Using default form submission (POST)</p>
<form action="/name" method="post">
<input type="text" name="name">
<input type="submit" value="go!">
</form>
<p>Using AJAX</p>
<form class="name-form">
<input class="name-input" type="text"></input>
<input type="submit" value="go!"></input>
</form>
<script src="https://cdn.jsdelivr.net/npm/axios/dist/axios.min.js"></script>
<script src="http-requests.js"></script>
Don't forget to model the script requests and responses.
Don't forget to model each of the three form submissions - each interaction is different.
Make GET and POST requests to the Name API endpoint
- Use the Fetch API from your browser console
- Use the Axios API from a nodejs script
- Use your browser directly - can it make both kinds of requests?
- Use a command line app like
curl,httpie - Use an application like insomnia
fetch("http://localhost:3000/api/name")
.then(res => res.json())
.then(data => console.log(data))axios.get(`http://localhost:3000/api/name`)
.then(res => console.log(res.data))curl localhost:3000/api/name{
"name":"Ed"
}
const opts = {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({name: "Ed"})
}
fetch(`http://localhost:3000/api/name`, opts)
.then(res => res.json())
.then(data => console.log(data))axios.post(`http://localhost:3000/api/name`, {
name: "Ed"
})
.then(res => console.log(res.data))curl -X POST -d "name=Ed" localhost:3000/api/name{
"name":"Ed"
}