NOTE: This documentation refers to v1 folder. For basic understanding of fork and execvp, refer to v0.

How this Custom Shell Works

This project is basically a mini shell — like a tiny version of bash or zsh. A shell is nothing but a program that takes commands from the user, runs them, and shows the result.

Let’s go step by step:

---

1. The Loop (The Heart of the Shell)

The shell keeps running inside a while loop.

• It shows a prompt ($ ).
• It waits for you to type something.
• It tries to understand what you typed.
• Then it runs it.
• And then it goes back to showing the prompt again.

This repeats forever, until you type exit.

---

2. Reading What You Type

We use fgets() to read the command you type. For example, if you type:

ls -l

fgets will grab that whole line as a string.

But fgets also captures the Enter key (the newline \n). So, before doing anything, we remove the newline. Otherwise, the command would look like "ls\n" instead of "ls", and Linux wouldn’t understand it.

---

3. Splitting the Command (Tokenizing)

When you type a command, it can have multiple parts. For example:

ls -l /home

This actually has 3 parts:

1. ls (the program to run)
2. -l (an option)
3. /home (a folder name)

To handle this, we split the input string by spaces. So, "ls -l /home" becomes an array:

args[0] = "ls"
args[1] = "-l"
args[2] = "/home"
args[3] = NULL

This is exactly the format that Linux system calls expect.

---

4. Forking (Creating a Child)

The shell itself cannot directly become ls. If it does, then the shell would be gone after one command.

So instead, we create a child process using fork().

• The parent (our shell) stays alive.
• The child is a copy of the parent, but inside the child we immediately replace it with the real program you wanted (ls, cat, etc.) using execvp().

So the child turns into the program you typed, while the parent waits.

---

5. Waiting for the Child

After the child starts running, the parent shell uses waitpid() to pause until the child is done.

This is why when you type ls, you see the output first, and only after that you get back the $ prompt.

---

6. Exiting

If you type exit, the shell just breaks out of the loop and ends the program.

This way you can close the shell gracefully.

---

7. Memory Safety

• Every time we read a command, we allocate memory (malloc).
• After using it, we free it so the program doesn’t waste memory.
• We also handle resizing of the arguments array in case the command has many parts.

---

Summary

My shell is like a middleman:

• It listens to what you type.
• It splits it into understandable parts.
• It creates a child process to run the program.
• It waits for the child to finish.
• Then it comes back and asks you again: “What next?”

That’s literally all a shell does — but now I have my own working version of it 🎉