Please explain the behaviour of the shell (supposed to be csh)

Question

On my Linux box, echo $SHELL results in /bin/csh. So I assume my default shell is c-shell. I am trying to understand the behavior of a shell script.
scenario 1 - script contains -

echo $1 $2 $3
echo $*
echo $argv[1] $argv[2]
$argv[3]
echo $argv[*]
echo $#argv

output -

arg1 arg2 arg3
arg1 arg2 arg3 arg4
[1] [2]
./test.sh: line 4: [3]: command not found
[*]
4argv

• So clearly the c-shell is not able to execute the last 4 lines in the script which should be executed by csh.

However - when I add the shebang line #!/bin/csh at the top of the script, it prints all the output correctly.

Question - Why is the csh not executing the last 4 lines correctly in first scenario and why do I explicitly have to include the shebang line?

Answer 1

When the file starts with #!/bin/csh, you explicitly tell the kernel to run the script with that program. But if there's no #! at the start of your file, execution fails – the exec() syscall fails – and your shell attempts to run the script using an even older method, by passing the file directly to /bin/sh.

The Bourne shell, sh, implemented this behavior way before #! appeared, and so scripts written this way expect to be run under sh or at least a compatible shell. csh is far from compatible, so it runs such scripts under sh. Your $SHELL is irrelevant.

tcsh:

Because many systems use either the standard version 6 or ver- sion 7 shells whose shell scripts are not compatible with this shell, the shell uses such a 'standard' shell to execute a script whose first character is not a '#', i.e., that does not start with a comment.