You may not choose to build this yourself from scratch but I have seen code libraries that will return a tuple in this way and so what we're going to cover is how you can use a tuple as a dictionary key and that may sound a little bit strange up until this point of course whenever we work with dictionaries the keys were always string values.

But now we can actually add multiple values inside of each one of those keys. So what I'm going to do for an example let's imagine that we had an e-commerce site and we wanted to keep track of the various packages that we had. And we also wanted to include some metadata into the Keys so I'm going to create a variable here called Priority index and I'm going to place a dictionary inside of it.

priority_index = {
}

Now the very first package I want it to have a weight of one and then because this is a tuple I can pass in another value. And so this package is called the premier package and then from there, I can add a colon and then put some product IDs like 1, 34, and 12 in a list.

priority_index = {
  (1, 'premier'): [1, 34, 12],

Now I can duplicate this and have a dictionary. The second package is also going to have a weight of 1 and then it's going to be called the MVP package and then we can pass an 84, 22, and 24 just made-up product IDs in a real application those would be coming in from the database.

Now in this last item this last package I'm going to give it a weight of 2 and this is just going to be called let's just call it the standard package and this is going to have its own set of ID's so it's a 93, 81, and 3.

priority_index = {
  (1, 'premier'): [1, 34, 12],
  (1, 'mvp'): [84, 22, 24],
  (2, 'standard'): [93, 81, 3],
}

So what we have here is a very interesting kind of data structure because we're combining all three of the popular Python data structures. We have a dictionary, we have a tuple that is being used as a key and then the value is the list. And so what this allows us to do is actually to add metadata directly into our key.

Usually, our keys are just strings and they have to be unique but what we have here is a situation where we can actually have multiple items in the key itself and so the way we can work with this is let's say I want to print this out. I'm going to convert it to a list and then I'm going to call my priority index dictionary and then call keys on it. And remember we need to convert this to a list because if we simply call keys on priority index it's just going to give us that view where we want to have a list that we can actually work with.

So now if I run this code you can see that we have the ability now to access these keys

You could use all of the standard ways of grabbing them so you could grab items and anything that we were using before you could use it with our traditional dictionaries. You can use here the difference is instead of a set of strings. Now we get a set of tuples so say that you were building this into an application and you wanted to know what your top packages were. Or say that you only wanted to feature your packages that had a weight of 1.

Then you could loop over which we're going to learn about looping later on in the course you can loop over these keys grab the values and only select the items that have this weight of 1 and then it will ignore all the rest of them so it's a very cool way of right inside of the key itself to have some additional information. So in summary that is how you can utilize tuples in Python as your dictionary keys.

Code

priority_index = {
  (1, 'premier'): [1, 34, 12],
  (1, 'mvp'): [84, 22, 24],
  (2, 'standard'): [93, 81, 3],
}

print(list(priority_index.keys()))