cURL and wget: Using cURL or wget to make requests to an endpoint

As web developers, we often have to hit API endpoints manually to check the response before we build a frontend for it. Our first instinct might be to reach for a GUI tool like Postman or Insomnia. However, there’s an easier way using cURL that is already available on most Linux based systems that we develop on.

Using cURL

cURL works like a conventional Linux program and its output can be piped to other programs or redirected to files. So, it can be used to save something like JSON responses to a file to inspect later.

Example of cURL requests are:

# GET request to <url>
curl <url>

# POST request to <url>
curl --request POST <url>

# POST request with body. body is form-urlencoded
curl --request POST --data "some data" <url>

# POST request with header and JSON body
curl --header "Content-Type: application/json" 
  --request POST 
  --data '{"username":"xyz","password":"xyz"}' 
  https://example.com/login

Common cURL flags to know

• --request - specify the request type such as GET, POST, PUT, DELETE, etc.
• --data - data being passed in body
• --cookie - send cookie from string/file
• --location - follow redirects

PRO TIP: To reverse engineer the cURL command for a particular request our browser makes, we can get it from the network tab of the Chrome browser console like below.

Use case for wget

I use cURL for almost all API request purposes. The only time I find myself using wget is when I need to recursively download a website or a specific directory on a domain. This is something cURL can’t do. The syntax for this looks like:

wget --recursive <URL_to_download>