General

Decomposing a project using Nx - Part 1

Published Apr 8, 2021

Updated Feb 14, 2023

7 min read

This article was written over 18 months ago and may contain information that is out of date. Some content may be relevant but please refer to the relevant official documentation or available resources for the latest information.

Decomposing a project using Nx - Part 1

Working on a large codebase brings multiple challenges that we need to deal with. One of them is how to manage the repository structure and keep it as clean and maintainable as possible. There are multiple different factors that can be considered when talking about project maintainability, and one of them, that is fundamental in my opinion, is how we structure the project.

When it comes to managing large scale project which may consist of many modules, or even separate applications, a Nx Workspace based mono-repository is a good candidate for managing such a project. If you don't know what an Nx Workspace is, I encourage you to read my previous article where I introduce it along with monorepo fundamentals.

In this article series, I'll show you:

2 approaches for decomposing a project
How they can help you to better manage your project's codebase
What tools Nx Workspace provides us with that help us enforce boundaries within a project

Modules vs libraries

It is a well-known good practice, especially when working with a complex Web Application, to divide the functionality into separate, self-contained, and, when possible, reusable modules. This is a great principle and many modern CLIs (ie. Angular, Nest) provide us with tooling for creating such modules with ease, so we don't waste time creating additional module structure by hand.

Of course, we could take it a step further, and instead of just creating a separate module, create a whole separate library instead. This seems to be a bit of an overkill at first, but when we consider that Nx CLI provides us with just as easy a way of creating a library as we did for a module, it doesn't feel so daunting anymore. With that in mind, let's consider what the benefits of creating a separate library instead of just a module are:

libs may result in faster builds
- nx affected command will run the lint, test, build, or any other target only for the libraries that were affected by a given change
- with buildable libs and incremental builds, we can scale our repo even further
libs enable us to enforce stricter boundaries
code sharing and minimizing bundle size is easier with libs
- we can extract and publish reusable parts of our codebase
- with small and focused libraries we only import small pieces into the application (in the case of multi-app monorepo)

Decomposition strategies - horizontal

In this article, I want to focus on horizontal decomposition strategy, which is great not only for large, enterprise projects, but for smaller applications as well. Horizontal decomposition focuses on splitting the project into layers that are focused on a single technical functionality aspect of the module. A good example of libraries type in this case is:

application layer
feature layer
business logic layer
api/data access layer
presentational components layer

As you may see in this example layering concept, each of the library types has a specific responsibility that can be encapsulated. I have created an example application that demonstrates how the aforementioned decomposition can be applied into even a simple example app. You can find the source code on my repository. Please check out the post/nx-decomposition-p1 branch to get the code related to this post. This application allows a user to see a list of photos and like or dislike them. It is a very simple use case, but even here, we can distinguish few layers of code:

photo-fe - frontend application top layer
photo-feature-list - this is a feature layer. It collects data from data-access layer, and displays it using ui presentational components.
photo-data-access - this is a layer responsible for accessing and storing the data. This is where we include calls to the API and store the received data using NgRx store.
photo-ui - this library contains all the presentational components necessary to display the list of photos
photo-api-model, photo-model - those are libraries that contain data model structure used either in the API (it's shared by FE and BE applications), and the internal frontend model. API and internal models are the same now but this approach gives us the flexibility to, for example, stop API breaking changes from affecting the whole FE application. To achieve this we could just convert from API to internal model, and vice-versa.

This application decomposition allows for easier modifications of internal layer implementation. As long as we keep the interface intact, we can add additional levels of necessary logic, and not worry about affecting other layers. This way we can split responsibility between team members or whole teams.

Nx workspace comes with a great toolset for managing dependencies between the internal libraries. A great starting point to get a grasp of the repository structure is to visualize the repository structure, and its dependencies. The following command will show us all libraries within a monorepo and dependencies between those libraries:

nx dep-graph

It will open a dependency graph in a browser. From the left side menu, you can choose which projects you want to include in the visualization. After clicking Select all, you should see the following graph:

You can read more about dependecy graph here:

Enforce boundaries

As you may see in the dependency graph above, our application layer is accessing only certain other parts/libraries. As the project grows, we would probably like to make sure that the code still follows a given structure. I.e. we would not like UI presentational components to access any data access functionality of the application. Their only responsibility should be to display the provided data, and propagate user's interactions via output properties. This is where Nx tags comes in very handy. We can assign each library its own set of predefined tags, and then create boundaries base on those tags. For this example application, let's define the following set of tags:

type:application
type:feature
type:data-access
type:ui
type:model
type:api-model
type:be

Now, within the nx.json file, we can assign those tags to specific libraries to reflect its intent:

  "projects": {
    "photo-api-model": {
      "tags": [
        "type:api-model"
      ]
    },
    "photo-data-access": {
      "tags": [
        "type:data-access"
      ]
    },
    "photo-feature-list": {
      "tags": [
        "type:feature"
      ]
    },
    "photo-model": {
      "tags": [
        "type:model"
      ]
    },
    "photo-ui": {
      "tags": [
        "type:ui"
      ]
    },
    "photo-fe": {
      "tags": [
        "type:app"
      ]
    },
    "photo-api": {
      "tags": [
        "type:be"
      ]
    }
  }

Now that we have our tags defined, we can use either an ESLint or TSLint rule provided by Nrwl Nx to restrict access between libraries. Those rules are named @nrwl/nx/enforce-module-boundaries and nx-enforce-module-boundaries for ESLint and TSLint respectively. Let's define our allowed libraries anteriactions as follows:

type:application - can only access type:feature libraries
type:feature - can only access type:data-access, type:model, type:ui libraries
type:data-access - can only access type:api-model, type:model libraries
type:ui - can only access type:ui, type:model libraries
type:model - can not access other libraries
type:api-model - can not access other libraries
type:be - can only access type:api-model libraries

To enforce those constraints we can add each of the rules mentioned above to the @nrwl/nx/enforce-module-boundaries, or nx-enforce-module-boundaries configuration. Let's open the top level .eslintrc.json or .tslint.json files, and replace the default configuration with the following one:

"@nrwl/nx/enforce-module-boundaries": [
  "error",
  {
    "enforceBuildableLibDependency": true,
    "allow": [],
    "depConstraints": [
      {
        "sourceTag": "type:app",
        "onlyDependOnLibsWithTags": ["type:feature"]
      },
      {
        "sourceTag": "type:feature",
        "onlyDependOnLibsWithTags": ["type:data-access","type:model", "type:ui"]
      },
      {
        "sourceTag": "type:data-access",
        "onlyDependOnLibsWithTags": ["type:api-model", "type:model"]
      },
      {
        "sourceTag": "type:ui",
        "onlyDependOnLibsWithTags": ["type:ui", "type:model"]
      },
      {
        "sourceTag": "type:be",
        "onlyDependOnLibsWithTags": ["type:api-model"]
      }

    ]
  }
]

For type:model and type:api-model, we can either not include any configuration, or explicitly add configuration with an empty array of allowed tags:

{
  "sourceTag": "type:model",
  "onlyDependOnLibsWithTags": []
},
{
  "sourceTag": "type:api-model",
  "onlyDependOnLibsWithTags": []
}

Now, you can run the following command to verify that all constraints are met:

nx run-many --target=lint --all

You can set up the CI to run this check for all the PRs to the repository, and therefore, avoid including code that does not follow the architectural pattern that you decided for your project.

If any of the aforementioned constrains were violated, the linting process would produce an error like this

A project tagged with "type:data-access" can only depend on projects tagged with "type:api-model" or "type:model".

This gives a clear message on what the problem is, and tells the developer that they are trying to do something that should not be done.

You can read more about Nx tags & constraints in the documentation.

Conclusion

When designing a software solution that is expected to grow and be maintained for a long time, it is crucial to create an architecture that will support that goal. Composing an application out of well-defined and separated horizontal layers is a great tool that can be applied to a variety of projects - even the smaller ones. Nx comes with a built-in generic mechanism that allows system architects to impose their architectural decisions on a project and prevent unrestrained access between libraries. Additionally, with a help of Nx CLI, it is just as fast and easy to create new libraries as with creating a new module. So why not take advantage of it?

In case you have any questions, you can always tweet or DM me @ktrz. I'm always happy to help!

This Dot is a consultancy dedicated to guiding companies through their modernization and digital transformation journeys. Specializing in replatforming, modernizing, and launching new initiatives, we stand out by taking true ownership of your engineering projects.

We love helping teams with projects that have missed their deadlines or helping keep your strategic digital initiatives on course. Check out our case studies and our clients that trust us with their engineering.

About the author(s)

Chris Trześniewski
Software Engineer, ultra-marathoner. He is passionate about functional programming in JavaScript and loves working with RxJS. In his free time he likes to go jogging.
@ktrz__@ktrz

Learning Paths for Next.JS Developers with Ankita Kulkarni

In this video, Rob Ocel and co-hosts Tracy Lee, Adam Rackis, and Danny Thompson talk with tech educator Ankita Kulkarni about her journey from engineering leader to full-time educator. Ankita shares insights on teaching Next.js, bridging practical knowledge gaps, and helping developers tackle real-world challenges. They discuss Next.js as a React-based framework, its benefits, and the challenges it presents for beginners. Chapters - Introduction to the Podcast and Guests 00:01 - Meet Ankita Kulkarni, Tech Educator 00:26 - Ankita's Transition to Full-Time Education 01:41 - Teaching Practical Knowledge in Next.js 03:19 - Effective Methods for Teaching Next.js 05:27 - Challenges of Being a Full-Time Educator 07:48 - Balancing Broad and Specific Examples 09:54 - Embracing Mistakes as a Teaching Tool 12:13 - Pair Programming and Mentorship 14:00 - Discussion on Next.js and Framework Adoption 16:48 - Advantages and Challenges of Next.js 18:12 - Choosing the Right Framework for Your Needs 20:35 - Impact of Next.js in React Documentation 22:26 - Learning Paths for New Developers 23:24 - The Rise of Full-Stack Web Development 25:09 - Benefits of Frameworks Abstracting Complexity 26:27 - OpenNext and Deployment Flexibility 28:06 - Ankita's Excitement for New Next.js Features 30:35 - The Future of Next.js Without Vercel 32:16 - Final Thoughts and Where to Find Everyone Online 34:21 Follow Ankita Kulkarni on Social Media Twitter: https://x.com/kulkarniankita9 YouTube: https://www.youtube.com/@kulkarniankita Sponsored by This Dot: thisdot.co...

Nov 12, 2024

2 mins

NextJSJavaScript

Introduction to Zod for Data Validation

As web developers, we're often working with data from external sources like APIs we don't control or user inputs submitted to our backends. We can't always rely on this data to take the form we expect, and we can encounter unexpected errors when it deviates from expectations. But with the Zod library, we can define what our data ought to look like and parse the incoming data against those defined schemas. This lets us work with that data confidently, or to quickly throw an error when it isn't correct. Why use Zod? TypeScript is great for letting us define the shape of our data in our code. It helps us write more correct code the first time around by warning us if we are doing something we shouldn't. But TypeScript can't do everything for us. For example, we can define a variable as a string or a number, but we can't say "a string that starts with user_id_ and is 20 characters long" or "an integer between 1 and 5". There are limits to how much TypeScript can narrow down our data for us. Also, TypeScript is a tool for us developers. When we compile our code, our types are not available to the vanilla JavaScript. JavaScript can't validate that the data we actually use in our code matches what we thought we'd get when we wrote our TypeScript types unless you're willing to manually write code to perform those checks. This is where we can reach for a tool like Zod. With Zod, we can write data schemas. These schemas, in the simplest scenarios, look very much like TypeScript types. But we can do more with Zod than we can with TypeScript alone. Zod schemas let us create additional rules for data parsing and validation. A 20-character string that starts with user_id_? It's z.string().startsWith('user_id_').length(20). An integer between 1 and 5 inclusive? It's z.number().int().gte(1).lte(5). Zod's primitives give us many extra functions to be more specific about *exactly* what data we expect. Unlike TypeScript, Zod schemas aren't removed on compilation to JavaScript—we still have access to them! If our app receives some data, we can verify that it matches the expected shape by passing it to your Zod schema's parse function. You'll either get back your data in exactly the shape you defined in your schema, or Zod will give you an error to tell you what was wrong. Zod schemas aren't a replacement for TypeScript; rather, they are an excellent complement. Once we've defined our Zod schema, it's simple to derive a TypeScript type from it and to use that type as we normally would. But when we really need to be sure our data conforms to the schema, we can always parse the data with our schema for that extra confidence. Defining Data Schemas Zod schemas are the variables that define our expectations for the shape of our data, validate those expectations, and transform the data if necessary to match our desired shape. It's easy to start with simple schemas, and to add complexity as required. Zod provides different functions that represent data structures and related validation options, which can be combined to create larger schemas. In many cases, you'll probably be building a schema for a data object with properties of some primitive type. For example, here's a schema that would validate a JavaScript object representing an order for a pizza: ` Zod provides a number of primitives for defining schemas that line up with JavaScript primitives: string, number, bigint, boolean, date, symbol, undefined, and null. It also includes primitives void, any, unknown, and never for additional typing information. In addition to basic primitives, Zod can define object, array, and other native data structure schemas, as well as schemas for data structures not natively part of JavaScript like tuple and enum. The documentation contains considerable detail on the available data structures and how to use them. Parsing and Validating Data with Schemas With Zod schemas, you're not only telling your program what data should look like; you're also creating the tools to easily verify that the incoming data matches the schema definitions. This is where Zod really shines, as it greatly simplifies the process of validating data like user inputs or third party API responses. Let's say you're writing a website form to register new users. At a minimum, you'll need to make sure the new user's email address is a valid email address. For a password, we'll ask for something at least 8 characters long and including one letter, one number, and one special character. (Yes, this is not really the best way to write strong passwords; but for the sake of showing off how Zod works, we're going with it.) We'll also ask the user to confirm their password by typing it twice. First, let's create a Zod schema to model these inputs: ` So far, this schema is pretty basic. It's only making sure that whatever the user types as an email is an email, and it's checking that the password is at least 8 characters long. But it is *not* checking if password and confirmPassword match, nor checking for the complexity requirements. Let's enhance our schema to fix that! ` By adding refine with a custom validation function, we have been able to verify that the passwords match. If they don't, parsing will give us an error to let us know that the data was invalid. We can also chain refine functions to add checks for our password complexity rules: ` Here we've chained multiple refine functions. You could alternatively use superRefine, which gives you even more fine grained control. Now that we've built out our schema and added refinements for extra validation, we can parse some user inputs. Let's see two test cases: one that's bound to fail, and one that will succeed. ` There are two main ways we can use our schema to validate our data: parse and safeParse. The main difference is that parse will throw an error if validation fails, while safeParse will return an object with a success property of either true or false, and either a data property with your parsed data or an error property with the details of a ZodError explaining why the parsing failed. In the case of our example data, userInput2 will parse just fine and return the data for you to use. But userInput1 will create a ZodError listing all of the ways it has failed validation. ` ` We can use these error messages to communicate to the user how they need to fix their form inputs if validation fails. Each error in the list describes the validation failure and gives us a human readable message to go with it. You'll notice that the validation errors for checking for a valid email and for checking password length have a lot of details, but we've got three items at the end of the error list that don't really tell us anything useful: just a custom error of Invalid input. The first is from our refine checking if the passwords match, and the second two are from our refine functions checking for password complexity (numbers and special characters). Let's modify our refine functions so that these errors are useful! We'll add our own error parameters to customize the message we get back and the path to the data that failed validation. ` Now, our error messages from failures in refine are informative! You can figure out which form fields aren't validating from the path, and then display the messages next to form fields to let the user know how to remedy the error. ` By giving our refine checks a custom path and message, we can make better use of the returned errors. In this case, we can highlight specific problem form fields for the user and give them the message about what is wrong. Integrating with TypeScript Integrating Zod with TypeScript is very easy. Using z.infer<typeof YourSchema> will allow you to avoid writing extra TypeScript types that merely reflect the intent of your Zod schemas. You can create a type from any Zod schema like so: ` Using a TypeScript type derived from a Zod schema does *not* give you any extra level of data validation at the type level beyond what TypeScript is capable of. If you create a type from z.string.min(3).max(20), the TypeScript type will still just be string. And when compiled to JavaScript, even that will be gone! That's why you still need to use parse/safeParse on incoming data to validate it before proceeding as if it really does match your requirements. A common pattern with inferring types from Zod schemas is to use the same name for both. Because the schema is a variable, there's no name conflict if the type uses the same name. However, I find that this can lead to confusing situations when trying to import one or the other—my personal preference is to name the Zod schema with Schema at the end to make it clear which is which. Conclusion Zod is an excellent tool for easily and confidently asserting that the data you're working with is exactly the sort of data you were expecting. It gives us the ability to assert at runtime that we've got what we wanted, and allows us to then craft strategies to handle what happens if that data is wrong. Combined with the ability to infer TypeScript types from Zod schemas, it lets us write and run more reliable code with greater confidence....

Nov 15, 2024

7 mins

JavaScriptZod

Introduction to Directives Composition API in Angular

In version 15, Angular introduced a new directives composition API that allows developers to compose existing directives into new more complex directives or components. This allows us to encapsulate behaviors into smaller directives and reuse them across the application more easily. In this article, we will explore the new API, and see how we can use it in our own components. All the examples from this article (and more) can be found on Stackblitz here. Starting point In the article, I will use two simple directives as an example * HighlightDirective - a directive borrowed from Angular's Getting started guide. This directive will change an element's background color whenever the element hovers. ` Fig. 1 * BorderDirective - a similar directive that will apply a border of a specified color to the element whenever it hovers ` Fig. 2 We can now easily apply our directives to any element we want ie. a paragraph: ` Fig. 3 However, if we wanted to apply both highlighting and border on hover we would need to add both directives explicitly: ` Fig. 4 With the new directives composition API, we can easily create another directive that composes behaviors of our 2 directives. Host Directives Angular 15 added a new property to the @Directive and @Component decorators. In this property, we can specify an array of different directives that we want our new component or directive to apply on a host element. We can do it as follows: ` As you can see in the above example, by just defining the hostDirectives property containing our highlight and border directives, we created a new directive that composes both behaviors into one directive. We can now achieve the same result as in Fig. 4 by using just a single directive: ` Fig. 5 Passing inputs and outputs Our newly composed directive works nicely already, but there is a problem. How do we pass properties to the directives that are defined in the hostDirectives array? They are not passed by default, but we can configure them to do so pretty easily by using an extended syntax: ` Fig. 6 This syntax takes exposes the "inner" directives\ color input from the HighlightAndBorderDirective`, and passes them down to both highlight and border directives. ` Fig. 7 This works, but we ended up with a border and highlight color both being blue. Luckily Angular's API allows us to easily redefine the properties' names using the : syntax. So let's remap out properties to highlightColor and borderColor so that the names don't collide with each other: ` Fig. 8 Now we can control both colors individually: ` Fig. 9 We could apply the same approach to mapping directive's outputs eg. ` Fig. 10 or ` Fig. 11 Adding host directives to a component Similarly to composing a directive out of other directives, we can apply the same approach to adding behavior to components using hostDirectives API. This way, we could, for example, create a more specialized component or just apply the behavior of the directive to a whole host element: ` Fig. 12 This component will render the paragraph, and apply both directives' behavior to the host element: ` Fig. 13 Just like we did for the directive, we can also expose and remap the directives inputs using the extended syntax. But if we would like to access and modify the directives inputs from within our component, we can also do that. This is where Angular's dependency injection comes in handy. We can inject the host directives via a constructor just like we would do for a service. After we have the directives instances available, we can modify them ie. in the ngOnInit lifecycle hook: ` Fig. 14 With this change, the code from Fig. 13 will use lightcoral as a background color and red as a border color. Performance Note While this API gives us a powerful tool-set for reusing behaviors across different components, it can impact the performance of our application if used excessively. For each instance of a given composed component Angular will create objects of the component class itself as well as an instance of each directive that it is composed of. If the component appears only a couple of times in the application. then it won't make a significant difference. However, if we create, for example, a composed checkbox component that appears hundreds of times in the app, this may have a noticeable performance impact. Please make sure you use this pattern with caution, and profile your application in order to find the right composition pattern for your application. Summary As I have shown in the above examples, the directives composition API can be a quite useful but easy-to-use tool for extracting behaviors into smaller directives and combining them into more complex behaviors. In case you have any questions, you can always tweet or DM me at @ktrz. I'm always happy to help!...

Jan 4, 2023

4 mins

TypeScriptDirectiveAngular

Build Facial Recognition and Chatbot AIs using TypeScript with Jack Herrington

In this JS Drop, Danny Thompson is joined by YouTuber Jack Herrington to explore a unique TypeScript and AI project that lets you recognize TV show characters in real time. Just point your camera at a character, and instantly get their details or even chat with them as if they were real! Jack walks through the tech, explaining how client-side face recognition and server-side AI work together to make this possible using the Versal AI library. They discuss prompt engineering, building efficient APIs, and ensuring smooth, interactive AI responses. Jack also shares potential real-world applications, from entertainment to security. Chapters - 0:32 Project Overview – Jack explains the AI-powered character recognition project - 2:17 Setting Up the Project – Overview of how the application is structured and initial setup - 3:04 Client-Side AI – How face detection and character recognition work on the client side - 5:12 Switching to Server Side – Jack demonstrates server-side AI and setting up API endpoints - 8:20 Explaining AI Tooling – How tools and prompts are used to give context to the AI - 10:01 Detailed Prompt Structure – Breaking down the prompt and character context for AI responses - 12:40 Client-Server Interaction – Using the Versal AI library to manage streaming responses - 15:09 Handling Character Data – Training the AI on specific character images and details - 18:15 Practical Use Cases – Discussing potential real-world applications for the face recognition tool - 21:34 Challenges and Lessons Learned – Jack shares obstacles he faced and how he overcame them - 25:45 Building the API – Tips and considerations for creating reliable API endpoints - 28:40 Handling User Inputs – Testing unexpected questions and how the AI responds - 32:00 Using Advanced AI Models – Jack talks about choosing GPT-4 and issues with smaller models - 35:47 Introducing ProNextJS.dev – Jack discusses his new Next.js course, covering advanced topics - 37:20 Closing Thoughts – Danny and Jack wrap up with final thoughts and a link to the GitHub repo Follow Jack Herrington on Social Media Twitter: https://x.com/jherr Linkedin: https://www.linkedin.com/in/jherr/ YouTube: https://www.youtube.com/@jherr Sponsored by This Dot: thisdot.co...

Nov 11, 2024

2 mins

AIArtificial Intelligence

Let's innovate together!

We're ready to be your trusted technical partners in your digital innovation journey.

Whether it's modernization or custom software solutions, our team of experts can guide you through best practices and how to build scalable, performant software that lasts.

Decomposing a project using Nx - Part 1

Decomposing a project using Nx - Part 1

Modules vs libraries

Decomposition strategies - horizontal

Enforce boundaries

Conclusion

Chris Trześniewski

You might also like

Learning Paths for Next.JS Developers with Ankita Kulkarni

Introduction to Zod for Data Validation

Introduction to Directives Composition API in Angular

Build Facial Recognition and Chatbot AIs using TypeScript with Jack Herrington

Let's innovate together!

You might also like

Learning Paths for Next.JS Developers with Ankita Kulkarni

Introduction to Zod for Data Validation

Introduction to Directives Composition API in Angular

Build Facial Recognition and Chatbot AIs using TypeScript with Jack Herrington