Project Solution: Twitter Class Diagram
Welcome to the solution for the Twitter class diagram.
Guide Tasks
  • Read Tutorial
  • Watch Guide Video
Video locked
This video is viewable to users with a Bottega Bootcamp license

Now, this is going to look very similar to what we implemented in the UML training course if you went through that, but I made a number of modifications. And the main reason is that I didn't want to give you the full answer to the project and so this is a slightly modified version that has all of the same class names but I made some changes that I would personally use in a project if I was tasked with building Twitter.

large

Now there is a lot of content here. This is definitely one of the larger class diagrams that you'll go through in this entire course, but this is great training because this is a pretty standard kind of setup for building an application. There are some important components here that I want to point out. There are also some optional elements here that I want you to be aware of because there are going to be many times where an organization has their own version of UML. They have their own type of syntax and their own way of doing things. And that's perfectly fine. I write my own UML slightly different than any other developer that I've met. There are always going to be slight changes. And so what I want to do is point out what the standard elements are and then show you why I've made some of my own creative decisions, because they work for me.

Let's start off with the tweet class.

Now I'm not going to go through all of the attributes for every single one of the classes. Like I mentioned in the instruction video, this is something that I'm leaving up to you based off of what you personally have discovered when you've researched Twitter. But just for this one we'll mention that there are elements such as an ID, text for the tweet, a reference to who the author is, if it was retweeted or not, when it was created, and different elements like that. You may notice that I don't have the data types for each one of these. Usually, I'll only put the data type when I feel like it's a custom one that really needs to have some additional explanation. For example, I know that text is going to be of type text or a varchar or something like that. Being able to add that in there doesn't really make the process of development any more efficient. So I don't like spending time on things that have no real benefit. And so that's just one example and you'll see that throughout this entire course.

And that's a reason why if you asked 10 developers and software engineers who are experienced in using UML and you task them with this project, you're going to get ten different variations of that project. That's one reason why I want to show you what some of those types of variations are, because when you go to work you're going to experience that, and I don't want you to think that there's only one right way of doing things.

Now with all that being said, there are some elements that really do need to be inside of every class diagram, so we'll focus on those. One of the most common is being able to demonstrate multiplicity. Remember, multiplicity is the practice of being able to show how an association is going to be defined. So our tweet has the potential to have many "likes." If you go to Twitter and you look at a tweet, you will see that it does have the ability to have multiple people who have liked that tweet. In our diagram, you can see that a tweet has many Likes and the way that that is defined is with the association arrow, and we have "1" followed by a star. What that means is that one tweet has the ability to have many likes.

Now there are many elements of a class diagram that are optional. However the one part that I'm going to stress is to make sure that you have your multiplicities correct. The reason for that is if you were to hand me a UML diagram and say, "Go build an entire database based off of this diagram," if the UML multiplicity items like this are set up properly it's going to make it very easy for me to go and build out all of those models, because I'm going to know that each like is going to have to have a reference to the tweet. So that's going to be a foreign key relationship right there. If this is set up incorrectly then what's going to happen is it's going to cause more problems for the developers. They're going to build it out and then only later realize that they had to have a foreign key inside of like. And so if you're passing this off to another developer they're not going to like you very much if you do not set up that component because you're creating more work for them.

Now, remember, inside of any type of class diagram, you have the name, followed by the attributes, followed by any operations. Typically, in my own approach, I don't spend a lot of time adding the names of the operations. And the reason for that is because it's been my experience that it just creates a lot of boilerplate work--in other words if I'm building out a new(), a create(), an edit(), and an update() action inside of a class then I already know as a developer that I'm going to add those in. It would just waste time in adding them in at this stage. I'll usually just throw in some of the operations that may be slightly different than standard ones, just to give myself or the developer that I'm handing this off to an idea of some of those custom components.

But one thing that I do think is very important is to list off the correct attributes, because what I will do as a developer is, when I take in a class diagram, I will go and generate my models based specifically off of this list. So I'll go create a tweet model that would automatically have an ID, but I would say it needs a text, an author reference, a retweet reference, a created, a like, hashtag, etc. etc.. And so that is something that I would literally copy and paste those elements and those names into my own generators. So I spend more time thinking about the right way to set up these attributes and less on these types of methods.

So that is tweet and like. We're not going to go over every single one of these elements because you're going to notice there are some incredibly similar types of components, so I want to focus on what makes them different. And then you can always reference the master file if you have any questions about any of the other components.

Now one of the types of syntaxes that I use that is different than the UML specification is for join tables. So, if we look at say the TweetHashtag table or the Retweet class, what these are is--they're not as much classes. Depending on your programming language they may be called classes, but what they really are are join tables. So, in this case, a tweet is joined to a hashtag class via this join table. And the reason why I like to use this dotted line syntax is because UML is visual and that's the entire point of the language. So if I am looking at a class diagram I like to be able to know where my classes are, like tweet, or like, or image, or hashtag, and I like there to be some type of clear visual distinction between those classes and the database tables that join those classes.

In other words, TweetHashtag is not going to have any operations. It's not going to have a set of methods. All it's doing is allowing me to have navigability between a tweet and a hashtag. If you remember what navigability is, it allows me to look up a tweet and then run a method that goes in and queries all of the hashtags. So that's what the TweetHashtag class allows me to do. And whenever I'm designing a system, if I know where the join tables are, I know that that's going to help me make several decisions.

One of the first and most important decisions is it's going to tell me that I can't create this class in this database table until the tweet class and the hashtag class have been created. That may not sound like a big deal, however, it actually is, because this tells me the order that I need to build each one of these models out and that I know I'm going to have to create a tweet class before I can do this TweetHashtag one just because it has to have a reference to a tweet and to a hashtag. So that's definitely one key reason, but then it also is simply helpful to understand that the functionality inside of these types of elements is going to be very minimal. It simply is a way that we can join classes together.

Now, for a tweet to retweet, this one's a little bit different because this is what's called a self-referential table. What we're doing is, we don't want to create an entire class for retweets, because, if you think about it, a retweet is simply a tweet. So this could be something, if you really like using inheritance, you could do something where retweet is a type of tweet, and then that would be able to give you access to it. But in the case of the way I typically build out systems I would simply make this a join table that can reference itself so that I will know which objects are tweets and which ones are retweets, simply based on what is inside of this join table.

And notice how this has a reference to retweeted, so, which item got retweeted, along with the new tweet that got created. So, in other words, to take a practical example, if I'm following a friend I see them make a post and I click on the little retweet button, then it's going to generate a new retweet record in the database that's going to have a reference to my tweet along with the source one that was retweeted. And another option that you could have added in here is "content," because Twitter in later years gave the ability not only to retweet but also to add notes associated with it.

That's another nice thing about using join tables, is they are not only there to keep track of IDs, they can also keep track of additional content that brings those together.

There is one other element I want to focus on and that is "following."

So we have our user class. It's pretty basic. It takes in an ID, name, time zone, city, and then some security settings, and then it has this following element. Now this is something that is very tricky for junior developers, and so being able to understand the way this works is going to help you out a lot when it comes to either going out and getting a job or building this type of functionality into your system. And this is not just for social networks. It's definitely something that you'll need to use if you're building out a social network, but at the end of the day there are very few applications, especially large production applications, that don't have to have this self referential kind of set up in some form or another.

We can see that we have our following join table, and we have users. So the tricky part here is we have a user that needs to be able to follow another user. But how do you keep track of that? Having this type of join table is a great way of doing it. So, if you have User A who wants to follow User B then what you can do is create this database table and be able to say, this following user is where we're going to store the ID of who is being followed. And then we have follower. So we're going to store the user id of the person who is performing the following action. And the nice thing about this is, because we have this reference right here, and a user hasMany followings, and following hasMany users, we can run clean database queries to perform actions, such as being able to render a list of all the users that one user is following.

And you're also able to perform tasks such as being able to create a feed. If you've ever wondered about how, when you're on Twitter, and you have all of these various tweets coming in from people you follow, it's accomplished using this type of setup. And what it's called is a self referential table. Notice, we're not joining items together. If you've never implemented this before I definitely recommend for you to research the concept of self referential tables but then also go and implement it in code and be able to learn how that works. This is going to be something that you're probably going to have to implement at some time or another.

So those are some of the key elements that are included in this solution. I definitely recommend for you to explore the entire source file and be able to see the way that I implemented it. And just like every other solution in this project, yours is going to look different than mine. It needs to have the same core components, it needs to have the full list of class names, and, in this case, one of the most important things is really focus when you're analyzing this and comparing it with your solution, focus on these multiplicities. Focus on those to make sure that you have the table set up correctly, so that you know which table and which class hasMany of another element, and when you have items such as when a class belongs to another class.

Remember the point of this entire course: to give you the ability to solve problems. Wwhen you're tasked with building out an application, and especially if it's a non-trivial type of app, that can be a very scary thing if you try to jump into the code right away. What this course is aiming to do is to give you an in-between step to be able to say, let's take a step back and let's model the system so that we can understand what we need to go build. Because the smaller of chunks that you can break a problem into the easier it starts to become. So by building out a properly structured class diagram, what you're going to be able to do is take this diagram and instantly convert it into code, you're going to be able to go and build your entire database just based off of everything that you put inside of this diagram.