Hello, world.

This is CS 50 on Twitch.

My name is Colton Ogden.

Apologies for the microphone being muted.

I turn it up briefly and forgot to turn it back on.

Um, this is part two of our stream on Solitaire, also known as Kalon Dyke.

And it's a popular card game, or at least was popular in Canada and the United States.

Last week, I'm gonna transition in my laptop.

We ended things with things with a board sort of like this with a little bit of grid grid layout in the background, some cards.

And this was just kind of a test to make sure that we had our our deck actually working with random car draws indigent to the ability to click and drag a card around there.

Just a sprite backing rather.

But you know, things are sort of moving in the right direction.

But we do have a long ways to go.

For example, today some of the things that I'd like to bite off our one actually laying out the board in a way that makes sense normally in solitaire, things air laid out such that and we'll look at Google images in a second.

But cards are laid out here in a sequence at the bottom, this bottom row saying with one card than two cars than three cards and four cards, et cetera.

With only the top card being visible, you get a deck up here in the top right, which is important.

That's where you draw cards from to put them into more piles on and then at the top, which you can't actually see.

Because the current algorithm is laying the cards out a TTE, the very top left.

You're actually stacking all the cards of each suit from the bottom to the top from Ace to the King.

So to answer our meats question, the topic of discussion is solitaire.

So we're using love to D today on the framework.

Love to D is a really nice framework for Tootie game development.

You gotta love to d dot org's.

We talked about this in the last Dream episode.

Definitely check that out if you haven't already seen it, where we implement everything from scratch, including parsing a spreadsheet of all our cards, laying out the grid, creating a deck class and a card class which we can then shuffle the deck and draw cards from which is very important because this is sort of the backbone of our whole game.

But more on that in part, one part to today is gonna be much more about the actual algorithms of laying out the board, getting to click and drag run piles of cards, and we'll see just how much we can implement before we get to potentially apart.

Three.

If we're very lucky, we'll finish apart, too, But I don't think so.

It's a rather media game, so thanks everyone for tuning in.

I got as Lee in the chat Male cat, uh, meet Microsoft India.

Pekka Mika to that chat cat Male can asked.

Well, I talk about machine learning.

Probably not machine learning today.

I'm not a machine learning expert, but, uh, would be nice to get somebody in the future who is more well versed in machine learning, such as?

Maybe Nick Wang, who did a prior stream.

If you haven't really watched the stream on binary classification with Nick Wong, definitely check it out.

It's on YouTube, probably not on twitch anymore.

It was one of our earlier twitches, and there was only saved for 30 days.

But yes, we are indeed continuing to make solitary meet Microsoft.

India says please share the link.

Yes, absolutely.

So I'll share two links in the chat.

Here.

Let me just pull up the repo name.

So I remember 100% what it was.

I have the code from last week on Get hub.

So if your are tuning in and you want to clone the code if you end up cloning the code Now, if you're watching this on YouTube, chances are the code's not gonna be the same as this, right at this exact moment in time.

Luckily, you can download arbitrary commit if you want to.

Unfortunately, for some reason, get up is taking Ah, long time toe load.

I'm not entirely sure why, but looks like it's working.

I saved it as solitaire stream, So I'm gonna take this into the chat here.

Hopefully, this doesn't cause any issues.

I just eliminated the Mac Rose on this machine so that I could actually type stuff this week.

Um, get up dot com slash colton Oscar P slash solitaire stream.

Take that and that should allow you to download the repo.

Uh, if you don't know what love to D is and what's.

I just ended up switching my view to the wrong one again.

If you don't know what Love to D is, you follow that link.

Downloaded.

It's a cross platform framework for making Tootie Games Institute.

Ian left to be done or unfortunately, doesn't let you make three D games.

I'm waiting for the day.

Love three D comes out.

I think that would be amazing development, but, alas, it is not a real thing yet.

So solitaire for those unfamiliar, let me just pull up an image of the solid terror starting layout so we can continue and sort of have a visual goal of where we're going next.

This is where we're going next.

It's not going quite like this.

This is a little bit fancier texture work done on it, but you can see that the grid is actually very similar to what we laid out last week.

You have seven rose, seven columns.

Sorry.

In the bottom row here, these are your stacks tableaus goes of different terminology that I'm not 100% clear on.

And you can notice that the cards start off on the left, with one card being drawn and that one's face up.

The second column has two cards, the top of which is face up the bottom would, which is face down the 3rd 1 The top card is still face up, but every card is face down and you can see the pattern continues on until the seventh column.

So you get all of these sort of areas to start laying out cards from left to right, and they increase in going toe in that direction, going towards the right one at a time at the very top left.

This is where the actual game is won, and we haven't covered this in too much detail yet.

But by stacking every card of each suits, remember, therefore suits ace diamonds, spades, clubs.

So by stacking every card, every a soccer star, every diamond card, every heart's card, every club's card, every spades card from a stick king.

That's how you actually win the game.

And so what you're looking for when you first start is, of course, an ace, because if I if I find an ace while I can stack in a CE here or an ace here an ace here, depending on which suit it is right.

And that gives you the ability to then lay a two on top of that and then a three on top of that.

And by the time you finally make your way up to King, you've won the game.

If you get all four of them out and to get new cards, which you can lay out either here on any of these piles or on these four winning piles, you have the deck here that you can draw cards from.

It's also called the Stock.

And so if I draw, for example, in a CE right here, I could put that in one of the piles.

Or if I draw a, for example, a um and it goes, um, lower value down.

If I were to draw a three a black three, I could put it on top of this red four, right, because the cars that you lay out all of these, this is an important rule of solitaire.

When you're laying cards out on your seven piles, they need to alternate in color, and they need to go from top to bottom.

They need to go from King down to down to ace.

So this four, for example, I could put a black three either a club or a spade three on this four.

I could put in a CE on this, too, but I'd probably want to put the ace on one of these piles here so I could start building up my, um my winning piles, right.

Wanna put the ace first?

That's the card you're looking for.

I could put a black eight right here.

I could put a red Queen here.

I could put a red six here, a red seven year red to here.

And this is how you build up your piles that you can move around eventually.

If you build all your piles in descending sort of order amongst your seven tableaus, you could very easily take cards from any of these piles and then put them in your winning piles.

And so this is a refresher on how the mechanics of solitaire work and so that we have a clear picture of where we're going as we implement our game.

An important thing that I think we should first think about implementing is populating these seven columns down here in this bottom row also probably want to get rid of this drawing these five random cards on.

Then maybe also just rendering a card backing here.

And that will be representative of our deck tux.

Man, thank you very much for tuning into the stream.

And thank you very much, too.

Kay's Oxy and South Africa for the follows right before the stream started, uh, again, any questions on the rules or what we've done so far?

Definitely the minute chat.

I'll be watching that as we go along twitch TV slash CS 50 TV as well.

All right, so let's take a look here.

So what we did last time was we implemented the game board, and what this was was just are basically just drawing the screen green.

And then those rectangles, though the grid layout for where we want to place our cards.

And that's essentially the backbone of where we're going to end up placing stuff.

And we're probably gonna want some notion of a card pile later on, which has a bunch of cards in it.

And those cards are gonna have to enforce their order and their colorization.

So we're gonna need to worry about that in a little bit.

What I want to do is I want to get rid of the sort of arbitrary rendering that we already have.

So what I think I did was in the deck class.

I made a render function here that just drew the 1st 5 cards in the deck.

And I don't really want that.

I don't want to draw those five cards anymore, so I'm actually just going to pass on that and re render it and notice that now we're no longer drawing those five cards.

And we indeed have our grid layout, as we saw previously, right?

I also probably don't need this random card that I can click around.

I could get rid of that pretty safely we're implementing we implemented is such that I can click to start the drag, and then it'll just follow my mouse and then click again to a new drag.

Our undrafted, the card and smart 13 18.

Thank you very much.

Oh, thank you for cool hairstyle.

Thank you very much for the car.

For the confident.

I appreciate that.

Um, so I think in card, actually, we may have been doing that in here?

Let's see.

It might have been in Maine, actually.

Let's make sure.

Yeah, I think Yeah, we we were rendering a queen of Hearts.

We we chose to render a queen of hearts just as a test to make sure card class was working.

You can see here in my directory, and this needs to be shrunk down a little bit.

I can get rid of some of these, but notice that I have a in my source directory in the repo.

And if you've cloned the Rebo, then you'll see this.

I have a card class, which is just representative of what a card is, which stores a suit and its stores of face value, which are two important things that we need to take into consideration for algorithm will need to check to see one.

When we place a winning card down in one of our top tables up here, we want to enforce that.

Each of these piles is just one of the same suit, right hearts, diamonds, clubs and spades.

When we want to actually stack them, we need to worry about the face value.

Whether it's a 34 Queen King, we'll do that, not only here in the winning piles section of our code.

We're also taking into consideration the suit, but we're also going to need to take into consideration.

Um, well, actually, we need to take into consideration the suit and the face value when we lay it down on the tableaus, too.

Because remember, we need to enforce a red black sort of progression of cards down the line.

Now it doesn't need to necessarily be one suit or the other a specific one.

I just need to be one of two if it's a red card, if it's a heart or a diamond, it card needs to be clubs or spades in order for us to lay it down and vice versa.

So that's just one thing that we need to take into consideration.

So that's why we have a card class.

The card class maintains a reference to its face, its suit, whether it's hidden, whether it's picked up and it's X and Y value so that we can render it to a given place on the screen.

There's also a deck class, which just maintains a list of cards that are all shuffled.

We wrote a shuffle algorithm.

Last time the, um, somebody refresh my memory on what the shuffle of, uh, it was Yates Shuffle.

What was it, Barry?

It's truffle, uh, and also known as the New Shuffle.

I think that was what Fisher Yates shuffle.

That's what it is, essentially just taking a bunch of random, basically indexing into our deck at random, putting it into another temporary deck and then eliminating that from the prior deck, which saved us a little bit of Spar City.

We don't need toe keep rule.

Re rolling are random ization if we end up messing up sort of how we take cards out of the deck and also shout it to David for tuning into the stream.

Um, in a tux.

My looks like Tech Man's about to say something for the card placement within the rose as well, so that was what we did with the deck class.

Very important, we'll be using that to draw individual cards for our stock later on in the game we needed to dependencies module so that we could actually not only specify where are textures are laid out, but also to slice them apart and store reference to the quads that we used in the last the last video, which were just rectangles that represent a chunk of our texture because we don't want to draw our entire spreadsheet recall, and we'll pull this up here.