Hi,

OK, there's a few concepts you'd need to get to grips with in order to write a solution to this problem. You've not said what shell or platform you're using, so I'm going to assume you're using Bash on Linux here, as that's probably the most likely possibility.

Firstly: knowing how many parameters a script has passed to it. This is actually fairly straightforward, and is handled by the Bash built-in variable $#. Take a look at this one-line example script, and try running it with different numbers of arguments. You should soon get a feel for how this works.

Code:

#!/bin/bash
echo "You provided $# arguments to this script."

Next, testing things. There will be times (such as in your problem statement here) when you need to know how many arguments there are, or if they're equivalent to each other. Bash has a few ways of doing this, the simplest and most straightforward for your purposes being the if statement.

For instance, try running this second small example script with different numbers of arguments. Take particular note of what happens if you run it with three arguments.

Code:

#!/bin/bash

if [ "$#" == "3" ]
then
        echo "You entered three arguments - that's my lucky number, thank you !"
else
        echo "Hmm.  That's an OK number of arguments, but it's not my favourite. Thanks for trying though."
fi

Here, we're checking to see if the number of arguments ($#) is equal (==) to the number 3. If it is, we display a special message. If it's anything other than three (in other words, if it's something else), we display another message instead.

Now, Bash can test for all kinds of things besides whether one thing is equal to another thing. Of particular interest to you should be the conditional tests it can carry out to check if something is a file, or a directory, or if it exists, and so on. If you look at the man page for Bash in the "CONDITIONAL EXPRESSIONS" section, you'll find a whole list of these.

For example, if I wanted to check to see if the second argument passed to my script actually existed on the filesystem, I could do something like this.

Code:

#!/bin/bash

if [ -e "$2" ]
then
        echo "Hey, look !  The second argument you gave me exists !  Let's take a look at what it is:"
        /bin/ls -ld "$2"
fi

Here, we check to see if the second argument ($2) exists (the meaning of the -e conditional expression). If it does, we print a message, and run the command /bin/ls -ld on it.

The last thing you'll probably need to understand to fully solve this problem is that conditional expressions (and plenty else besides) can be chained together using logic such that at least one of them must be true, or that they all must be true. Take a look at the "Compound Commands" section of the Bash man page to read how they all work.

I'll leave you with one last example though of a situation where we want to check that two particular things are the case at the same time.

Code:

#!/bin/bash
if [ "$1" == "dog" ] && [ "$2" == "cat" ]
then
        echo "Those are my two favourite animals too !  We have a lot in common !"
else
        echo "Sorry, but those aren't my two favourite animals."
fi

Here, if the first argument ($1) and (the meaning of && here) the second argument ($2) are equal to "dog" and "cat" respectively, we print a special message. If they're not both set to exactly those things (and they both must be, due to our use of && here) then we print another message instead.

Anyway, hope all this helps give you some pointers on how to get started, and which documentation is best to look at to familiarise yourself with the Bash tests you'll need to know about.