Inam Ul Haq
Posted on May 23, 2020
HTTP is a protocol used for WWW(World Wide Web) which serve us functional websites, web applications, and web APIs. HTTP clients (internet browsers) use TCP/IP or UDP networking protocol with HTTP to provide us the greatness and usefulness of WWW.
Why we need TCP/IP or UDP with HTTP?
HTTP only defines rules for communication between server and client(internet browser) but for communication, we need to connect to other computer first.
TCP/IP and UDP are two popular networking protocols used to connect computers together, so they can communicate. Both of these networking protocols are implemented in most of the Operating Systems out there. All the DNS and sockets stuff happens in TCP/IP and UDP protocols not in HTTP.
HTTP clients leverage these protocols to connect to different servers (like Facebook's) and then talks to them in HTTP. TCP/IP is most popular among others, so I'll be using that for rest of the post.
Introduction to HTTP
HTTP is a client-server architected protocol where client and server can ask each other to do specific tasks using agreed upon language(HTTP). HTTP defines words and manners for communication between client and server e.g. take a look at following HTTP request:
GET / HTTP/1.1
User-Agent: Mozilla/4.0 (compatible; MSIE5.01; Windows NT)
Host: www.google.com
Accept-Language: en-us
Accept-Encoding: gzip, deflate
Connection: Keep-Alive
This request is a GET request, to the www.google.com Host, asking for the / document (document is term used in HTTP for a web page), using 1.1 version of HTTP, along with other instructions/information for the server.
Now server will respond with a HTTP response, according to the request. Response then will be interpreted by the HTTP client. Take a look at the following HTTP response:
HTTP/1.1 200 OK
Content-Encoding: gzip
Accept-Ranges: bytes
Age: 322343
Cache-Control: max-age=604800
Content-Type: text/html; charset=UTF-8
Date: Sat, 23 May 2020 05:20:54 GMT
Etag: "3147526947"
Expires: Sat, 30 May 2020 05:20:54 GMT
Last-Modified: Thu, 17 Oct 2019 07:18:26 GMT
Server: ECS (bsa/EB18)
Vary: Accept-Encoding
X-Cache: HIT
Content-Length: 648
<!doctype html>
<html>
...
</html>
You can notice there are two kinds of data in the HTTP response; lines of colon separated key:value pairs and raw data; separated by a line break.
Colon separated key:value pairs are called a headers(header for singular). Some standard headers are for instructions(like Connection) and others are for information(like User-Agent).
Raw data is called body of the request or response. body can contain any kind of raw data e.g. html, json, binary files, and form data.
Both request and response can contain HTTP defined standard and non-standard headers and a body.
Most of the time, servers and clients interpret body as instructed in the headers. For example, the response above has a Content-Type: text/html; charset=UTF-8 header which instructs HTTP client to interpret body of the response as HTML which results in a rendered HTML page.
HTTP transaction lifecycle
backdrop: you use internet browser to request www.google.com
- Browser uses
TCP/IPprotocol, provided by the Operating System, to connect to www.google.com server - Browser sends HTTP request to the server using connection made in previous step
- Server receives HTTP request
- Server interpret HTTP request and respond with HTTP response
- Browser receives HTTP response
- Browser interpret HTTP response (e.g. if it's HTML; render it)
browser is an HTTP client
server in the post means a web/HTTP server(there are other kinds of servers too)
If you didn't get any concept/term from the post, ask me in the comments section.
You can follow me on Twitter
Checkout GetResume, a Free Resume Builder
Posted on May 23, 2020
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.