📖 42-Webserv

About · Running · Configuration · Key features

💡 About the project

This is when you finally understand why a URL starts with HTTP.

This project is about writing a non-blocking HTTP/1.1 server in C++, designed to handle multiple simultaneous connections efficiently using asynchronous I/O. The server adheres to the HTTP/1.1 protocol and supports essential features such as GET, POST, DELETE, and error handling, ensuring responsive communication without stalling the main execution thread. Additionally, the server integrates non-blocking CGI execution, supporting various languages such as Python, Perl, PHP, and binary executables. Multi threading implementation, each server instance runs on its own thread.

🚀 Running the server

make run

It will automatically compile the project and run it.

./webserv [configuration file]

Will launch the previously compiled project with make and run it. (configuration file must be a valid .conf file see section below)

💻 Configurating the server

It will allow you to setup the following parameters :

• name -- define the server name
• listen:port -- to configure on which port the server will be listening (only one listen per server block)
• methods -- to set the authorized HTTP request (GET | POST | DELETE)
• location -- to set parameters for a specific route
• index -- to serve a default file if none match the requested URI
• root -- to route the requested URI to another directory
• upload -- to set a specific directory for upload requests
• listing -- to list all files present in the targeted directory (ON | OFF)
• error -- to define the page to display when an error occurs, add a default page if none is provided in the .conf file.
• cgi -- to configure a cgi to execute for a specific file extenstion (.php | .py | .pl)
• redirection -- to define a redirection to a location or to an another server

{
    name server
    listen 0.0.0.0:8080
    methods GET POST
    body_size 1000000000000
    root ./rsrcs
    upload /upload
    index index.html
    error 404 404.html
    location /cookie
        root ./rsrcs/cookie
        index cookie.html
        methods GET
    location /cgi
        root ./rsrcs/cgi
        listing ON
        cgi .py .php
        methods GET POST
    location /intra
        root ./rsrcs/intra
        index intra.html
        methods GET
    location /listing
        listing ON
        error 403 403.html
        error 404 404.html
        root ./rsrcs/listing
        methods GET
    location /listing/403.html
        root ./rsrcs/listing/403.html
    location /listing/404.html
        root ./rsrcs/listing/404.html
    location /listing/1/2/3
        redirection /listing
    location /upload
        root ./rsrcs/upload
        methods GET DELETE
    location /redirection
        redirection https://fr.wikipedia.org/
}

---

📜 Key Features

• Configuration Flexibility
  -- Accepts a custom configuration file or defaults to a predefined path

• Non-Blocking Architecture
  -- Uses a single select() (or equivalent) per server for all I/O operations, including listening and client communication

• Efficient Resource Management
  -- Handles all read/write operations via select(), optimizing resource usage and preventing blocking

• HTTP Methods
  -- Supports GET, POST, and DELETE

• File Uploads
  -- Allows clients to upload files directly to the server

• Static File Serving
  -- Serves static websites

• Accurate HTTP Responses
  -- Delivers precise status codes along with a specified error page or a default one

• CGI Integration
  -- Supports CGI execution (e.g., PHP, Python, Perl, ...)

• Multi-Port Listening
  -- Listens on multiple ports as defined in the configuration file

• Browser Compatibility
  -- Compliant with HTTP 1.1 standards.

• Handle Cookies
  -- A small cookie clicker has been made to show the implementation of the cookies

• Keep-alive connection
  -- Connection is kept alive if possible

• Time out
  -- Timeout for CGI and basic requests

---

---

This group project was done by MVPee (mvan-pee) and Dan (dspilleb)