Skip to content

my_log.md

Latest commit

 

History

History
96 lines (69 loc) · 3.07 KB

my_log.md

File metadata and controls

96 lines (69 loc) · 3.07 KB

At its core, a shell is just a REPL — Read-Eval-Print Loop — that runs commands for the user.

A shell is fundamentally just a loop that:

  • Reads a line of input
  • Parses it into a command and argument
  • Creates a process ( fork() )
  • Runs the command( execvp() )
  • Waits for it finish( wait() )
  • Repeats

Imagine:

shell $ echo hello hello $ ls -l total 4 -rw-r--r-- 1 user user 0 Jul 17 file.txt $ exit`

Each line:

  • You typed something (input)

  • Shell split it up (tokenizing)

  • Shell created a child process

  • Shell replaced that process image with echo or ls

  • Waited for it to complete

  • Printed output

  • Repeated

That’s it.

So your first goal: Build a loop that takes a command, executes it, waits for it to finish, and repeats.

Plan of Action:

  1. Basic Input and exit
  2. Parse input into arguments
  3. fork() and exec()
  4. Error Handling and edge cases
  5. Path prompt + current directory
  6. Built in commands

How to take INPUT from the user?

Use fgets(). Why? Why not scanf(). Coz it breaks on spaces, also it is vulnerable to overflows and other unwanted things, you wouldn't imagine using it for usage in a shell.

fgets() : fgets(input, 100, stdin); // reads at most 99 chars + \0 { fgets() adds a \0 at the end to terminate }

Also there's one more problem with fgets(). It capture the newline character too, when ENTER is pressed. To get rid of it, we use strcspn() strcspn(s1, s2) returns the index of the first character in s1 that exists in s2 So we find it using this: strcspn(input, "\n");, and then replace it with '\0'

Now after we've taken an input, we'll need to take action based on it. For that, we'll first have to compare first. This is where one must walk consciously, or else would just trip We use strcmp(s1, s2) for it. It return a value: 0 - if s1 and s2 are same; >0 - if s1 > s2; <0 - if s2>s1 lexicographically. [It compares ASCII values, character-by-character]

Now, the next step is tokenization of the command.

Implementing Tokenization

We use strtok() to tokenize a string, into separate words.

argv[i] = strtok(input, " ");

The first call to strtok() splits input using " " (space) as a delimiter.
It returns a pointer to the first token.

char input[] = "ls -l /home"; char *token = strtok(input, " "); while (token != NULL) { printf("%s\n", token); token = strtok(NULL, " "); // this makes it to keep performing it on the same input} argv[i] = NULL; // Very important for exec

Well the last statement as we can see, states that it is very important for exec(). Why? Coz exec() functions expect the argument array ( typically named argv), to follow a very specific format:

A null-terminated array of character pointers (char *argv[]), where:

  • argv[0] is the name of the command,
  • argv[1] to argv[n-1] are the arguments,
  • and argv[n] is NULL, indicating the end of the argument list.`

So, the last line is required to tell execvp() where the argument list ends.