Timo Martinson•9mo ago

How to start using data validation in kv?

Hey fellows! I want to dive into data validation. Any experience or suggestions? Feel free to post a commented version of the repository I provided here:

import Post from '@/modules/post/model.ts'

import { setData, getData, listData, removeData } from '@/helpers/data.ts'

type PostParams = {
    blogSlug?: string
    postSlug?: string
    postInput?: Post
}

class PostRepository {
    static async createPost({ postInput, blogSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postInput?.slug as string]
        const result = await setData({ key, value: postInput })
        if (result.ok) return await getData({ key })
    }

    static async listPosts({ blogSlug }: PostParams) {
        const prefix = ['blog', blogSlug as string, 'post']
        return await listData({ prefix })
    }

    static async showPost({ blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        return await getData({ key })
    }

    static async updatePost({ postInput, blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        return await setData({ key, value: postInput })
    }

    static async deletePost({ blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        await removeData({ key })
    }
}

export default PostRepository

import Post from '@/modules/post/model.ts'

import { setData, getData, listData, removeData } from '@/helpers/data.ts'

type PostParams = {
    blogSlug?: string
    postSlug?: string
    postInput?: Post
}

class PostRepository {
    static async createPost({ postInput, blogSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postInput?.slug as string]
        const result = await setData({ key, value: postInput })
        if (result.ok) return await getData({ key })
    }

    static async listPosts({ blogSlug }: PostParams) {
        const prefix = ['blog', blogSlug as string, 'post']
        return await listData({ prefix })
    }

    static async showPost({ blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        return await getData({ key })
    }

    static async updatePost({ postInput, blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        return await setData({ key, value: postInput })
    }

    static async deletePost({ blogSlug, postSlug }: PostParams) {
        const key = ['blog', blogSlug as string, 'post', postSlug as string]
        await removeData({ key })
    }
}

export default PostRepository

Thank you!

4 Replies

Timo MartinsonOP•9mo ago

This is my model:

type Post = {
    title: string
    slug: string
    content: string
    draft: boolean
    publishedAt: Date
}

type Post = {
    title: string
    slug: string
    content: string
    draft: boolean
    publishedAt: Date
}

Deno AI Helper•9mo ago

Hello there! 🦕 To start using data validation in Deno KV, you can follow a few strategies as suggested by the documentation. When you're dealing with TypeScript, you can use interfaces or classes to describe the shape and behavior of objects in your programs, which can serve as a form of data validation. Here's a step-by-step guide based on the documentation: 1. Define Your Data Models: You've already defined a Post model, which is great. Ensure that your model accurately represents the data structure you want to store in Deno KV. 2. Type Assertions: When retrieving objects from Deno KV, you can use type assertions to inform TypeScript about the shape of an object. For example:

   const post = await kv.get(["blog", "my-blog", "post", "my-post"]) as Post;

   const post = await kv.get(["blog", "my-blog", "post", "my-post"]) as Post;

3. Type Parameters: Alternatively, you can specify an optional type parameter for the get method to automatically cast the returned value to the specified type:

   const post = await kv.get<Post>(["blog", "my-blog", "post", "my-post"]);

   const post = await kv.get<Post>(["blog", "my-blog", "post", "my-post"]);

4. Service Layer: For more complex validation that involves business logic, you can create a set of functions or a service layer that operates on your data transfer objects (DTOs). This layer can ensure that the data being passed around is valid and not merely typed correctly. 5. Validation Functions: Develop pure functions that can validate your DTOs. These functions can check for required fields, data formats, and other business rules before you save or retrieve data from Deno KV. 6. Use the Service Layer in Your Repository: In your PostRepository, you can use these validation functions to validate postInput before calling setData and after retrieving data with getData. Here's an example of how you might

Deno AI Helper•9mo ago

integrate a simple validation function into your PostRepository:

import Post from '@/modules/post/model.ts'
import { validatePost } from '@/validators/postValidator.ts' // This is a hypothetical validator function you would create

// ... rest of your repository code ...

class PostRepository {
    static async createPost({ postInput, blogSlug }: PostParams) {
        if (!validatePost(postInput)) {
            throw new Error('Invalid post data');
        }
        // ... rest of the createPost method ...
    }

    // ... rest of the repository methods ...
}

export default PostRepository

import Post from '@/modules/post/model.ts'
import { validatePost } from '@/validators/postValidator.ts' // This is a hypothetical validator function you would create

// ... rest of your repository code ...

class PostRepository {
    static async createPost({ postInput, blogSlug }: PostParams) {
        if (!validatePost(postInput)) {
            throw new Error('Invalid post data');
        }
        // ... rest of the createPost method ...
    }

    // ... rest of the repository methods ...
}

export default PostRepository

Remember, the actual implementation of your validation logic will depend on the specific requirements of your application. The above steps are a general guide to get you started with data validation in Deno KV as suggested by the documentation. Keep in mind that Deno KV is currently in beta and the APIs are subject to change. Always ensure you're using the --unstable flag when working with Deno KV. Happy coding! 🦕

Helpful Links: Data Modeling in TypeScript | Deno Docs Deno Docs How to Build a Blog with Fresh Creating a CRUD API | Fresh docs

Leokuma•9mo ago

This might be a useful resource: https://github.com/oliver-oloughlin/kvdex

GitHub

GitHub - oliver-oloughlin/kvdex: A high-level abstraction layer for...

A high-level abstraction layer for Deno KV with zero third-party dependencies by default 🦕🗝️ - oliver-oloughlin/kvdex