使用 Next.js 和 Sanity 建立現代部落格：逐步指南

沒有 CMS 的部落格可能會導致無盡的挫折感和浪費時間。 Sanity.io 簡化了整個流程，讓您專注於您的內容。

在今天的文章中，您將學習如何使用 Sanity CMS 和 Next.js 14 建立部落格。

在本指南結束時，您將擁有一個功能齊全且易於管理並運行的部落格。

先決條件

要閱讀本文，您必須具備以下技能：

1. HTML、CSS 與 JavaScript 知識
2. Next.js 和 TypeScript 基礎
3. Tailwind CSS 的基本了解
4. 並且 Node.js 安裝在您的電腦上。

什麼是理智？

Sanity 是一個無頭 CMS，它使內容管理變得簡單且有效率。透過 Sanity Studio 直覺的儀表板，您可以按照自己想要的方式輕鬆建立、編輯和組織內容。

Sanity 還提供靈活的 API 和 Webhook 支持，讓您可以完全控制內容的交付方式和地點。無論是網站、行動應用程式或任何其他平台，Sanity 都能確保您的內容始終是最新的且易於存取。

初始化 Next.js 項目

我們正在將 Sanity 整合到 Next.js 專案中。所以，我們需要先建立一個next.js專案。

要建立 next.js 項目，請執行以下命令：

npx create-next-app@latest

按照終端機上的說明進行操作並選擇名稱，之後即可使用預設建議。

這將產生一個簡單的 Next.js 專案。

現在，讓我們在程式碼編輯器上開啟專案。

cd sanity-blog
code .

現在執行 dev 指令在 Localhost:3000 上開啟項目

npm run dev

設定 Sanity Studio

Sanity Studio 是您管理內容的儀表板。

您可以獨立建置Studio並部署。但我們會將 Studio 嵌入 Next.js 專案中。易於維護和使用。

因此，我們將建立一個 Sanity 項目，然後將其整合到我們的 Next.js 專案中。

執行此指令來初始化 Sanity 專案。

npm create sanity@latest

當您執行此命令時，系統會要求您登入 Sanity。如果您已經有帳戶，請選擇提供者並登入您的帳戶。
如果您沒有帳戶，請建立帳戶並再次執行安裝命令。

執行此命令後，它會詢問您一系列問題來設定您的專案。

您可以使用預設選項。

我們只需要項目名稱，其餘的都不重要。

Looks like you already have a Sanity-account. Sweet!

✔ Fetching existing projects
? Select project to use Create new project
? Your project name: sanity-blog
Your content will be stored in a dataset that can be public or private, depending on
whether you want to query your content with or without authentication.
The default dataset configuration has a public dataset named "production".
? Use the default dataset configuration? Yes
✔ Creating dataset
? Project output path: /home/amrin/Desktop/writing/sanity-blog
? Select project template Clean project with no predefined schema types
? Do you want to use TypeScript? Yes
✔ Bootstrapping files from template
✔ Resolving latest module versions
✔ Creating default project files
? Package manager to use for installing dependencies? npm
Running 'npm install --legacy-peer-deps'

安裝依賴項

在我們將 sanity studio 整合到 Next.js 部落格之前，我們需要安裝這些依賴項。

npm install sanity next-sanity --save

將 Sanity 整合到 Next.js 專案中

要將 Sanity 整合到 Next.js 中，我們需要專案名稱和專案 ID。我們可以從 Sanity 儀表板中取得該資訊。

前往 sanity.io/manage 你會看到那裡的所有物品。

點擊項目標題即可查看詳細資訊。

你會看到類似這樣的內容：

繼續複製專案名稱和 ProjectID 並將其新增至您的 .env 檔案

NEXT_PUBLIC_SANITY_PROJECT_TITLE = "";
NEXT_PUBLIC_SANITY_PROJECT_ID = "";

現在在專案資料夾的根目錄下建立設定檔。並命名為sanity.config.ts

import { defineConfig } from "sanity";
import {structureTool} from "sanity/structure";
import schemas from "@/sanity/schemas";

const config = defineConfig({
  projectId: process.env.NEXT_PUBLIC_SANITY_PROJECT_ID as string,
  title: process.env.NEXT_PUBLIC_SANITY_PROJECT_TITLE as string,
  dataset: "production",
  apiVersion: "2023-06-18",
  basePath: "/admin",
  plugins: [structureTool()],
  schema: { types: schemas },
});

export default config;

以下是設定檔中內容的快速概述：

先匯入需要的函數和檔案。然後定義配置。此配置有很多選項：

projectId:這是您之前建立的 Sanity 專案 ID。

標題：您的 Sanity 專案標題。

資料集： 為 Studio 定義資料集。

basePath: Studio 的路徑。我們使用 /admin 路由來存取 Studio。您可以選擇任何您想要的路線。

架構：這些是內容的架構。架構定義文件的外觀以及它所包含的欄位。我們將為貼文和作者、類別等提供架構。

我們還沒有架構，我們稍後會創建它。

設定工作室

要建立工作室，我們首先需要一條路線。導航至 src/app，然後建立路由組並將其命名為 studio。 我們正在對其進行分組，以將網站與 studio 路由分開。

Inside the studio create an admin folder and inside that add a catch-all route.

└── (studio)
    ├── admin
        └── [[...index]]
            └── page.tsx

Include this code in the admin route. We are getting the sanity.config we created earlier and NextStudio from sanity Studio to initialize the Studio.

"use client";

import config from "../../../../../sanity.config";
import { NextStudio } from "next-sanity/studio";

export default function AdminPage() {
  return <NextStudio config={config} />;
}

We are almost done setting up the studio.
Lastly, we need to write the schemas for the content. After that, we can take a look into the studio.

Create The Schema

A Schema defines the structure of a document in the Studio. We define schema with properties.

Some of the properties are required and some are not.

The common properties are:

name: Name of the Schema. We will use this name to fetch the data.

title: Human readable title for the Schema. It will be visible in the Studio.

type: A valid document type.

fields: An array of all the properties of the document. If it’s a post schema the fields will have properties like Title, slug, body, meta description, etc. These properties will show up as input fields on the Studio.

Since we are building a blog we will have multiple Schemas such as:

• Post
• Author
• Category

To learn more about Sanity Schema Visit the documentation.

Post Schema

Create a folder named sanity inside the src directory.

Inside that create another folder named schemas and create index.ts and post.ts file

Here’s what the post Schema looks like.

const post = {
  name: "post",
  title: "Post",
  type: "document",
  fields: [
    {
      name: "title",
      title: "Title",
      type: "string",
      validation: (Rule: any) => Rule.required(),
    },
    {
      name: "metadata",
      title: "Metadata",
      type: "string",
      validation: (Rule: any) => Rule.required(),
    },
    {
      name: "slug",
      title: "Slug",
      type: "slug",
      options: {
        source: "title",
        unique: true,
        slugify: (input: any) => {
          return input
            .toLowerCase()
            .replace(/\s+/g, "-")
            .replace(/[^\w-]+/g, "");
        },
      },
      validation: (Rule: any) =>
        Rule.required().custom((fields: any) => {
          if (
            fields?.current !== fields?.current?.toLowerCase() ||
            fields?.current.split(" ").includes("")
          ) {
            return "Slug must be lowercase and not be included space";
          }
          return true;
        }),
    },
    {
      name: "tags",
      title: "Tags",
      type: "array",
      validation: (Rule: any) => Rule.required(),
      of: [
        {
          type: "string",
          validation: (Rule: any) =>
            Rule.custom((fields: any) => {
              if (
                fields !== fields.toLowerCase() ||
                fields.split(" ").includes("")
              ) {
                return "Tags must be lowercase and not be included space";
              }
              return true;
            }),
        },
      ],
    },
    {
      name: "author",
      title: "Author",
      type: "reference",
      to: { type: "author" },
      validation: (Rule: any) => Rule.required(),
    },
    {
      name: "mainImage",
      title: "Main image",
      type: "image",
      options: {
        hotspot: true,
      },
      // validation: (Rule: any) => Rule.required(),
    },
    {
      name: "publishedAt",
      title: "Published at",
      type: "datetime",
      validation: (Rule: any) => Rule.required(),
    },
    {
      name: "body",
      title: "Body",
      type: "blockContent",
      validation: (Rule: any) => Rule.required(),
    },
  ],

  preview: {
    select: {
      title: "title",
      author: "author.name",
      media: "mainImage",
    },
    prepare(selection: any) {
      const { author } = selection;
      return Object.assign({}, selection, {
        subtitle: author && `by ${author}`,
      });
    },
  },
};
export default post;

Copy the schema over to the post.ts file.

To save time we are not going to see the other schemas, you can get them from the repository.

Load the schemas

Open up the index.ts file and add this code snippet.

import author from "./author";
import blockContent from "./blockContent";
import category from "./category";
import post from "./post";

const schemas = [post, author, category, blockContent];

export default schemas;

We are importing all the schema in this file and creating an array to load the schema on the studio.

Now you can add posts from the studio.

To create a new post, go to localhost:3000/admin you will see all the schemas there. Go ahead and create a few posts.

Query the content with GROQ

We integrated the Studio and created a few posts. Now we need a way to fetch those posts and render them on the home page.

We will use GROQ to do exactly that. GROQ is a query language designed to query large schema-less JSON data collection. With GROQ expressive filtering we can fetch data the way we want to use it. It can join and fetch from multiple documents.

To start using GROQ we need to create the config file and the queries.

Go ahead and create these files inside the sanity folder.

└── sanity
    ├── config
    │   └── client-config.ts
    ├── sanity-query.ts
    ├── sanity-utils.ts

Paste this code into the client-config.ts file.

import { ClientPerspective } from "next-sanity";

const config = {
    projectId: process.env.NEXT_PUBLIC_SANITY_PROJECT_ID as string,
    dataset: "production",
    apiVersion: "2023-03-09",
    useCdn: false,
    token: process.env.SANITY_API_KEY as string,
    perspective: 'published' as ClientPerspective,
};

export default config;

This is the config for fetching the data with the GROQ query.

Here’s a quick break-down of the config options:

apiVersion: It’s the Sanity API version. You can use the current date.

useCDN: Used to disable edge cache. We are setting it to false as we will integrate webhook. It will deliver updated data.

token: Sanity API key. Required for webhook integration. (We will integrate webhook in the next section)

perspective: To determine the document status. If it’s set to published it will only query the published documents otherwise it will fetch all the document drafts and published.

Now we will write the queries. We are going to keep the Queries and the Fetch functions in separate files.

Here are the queries, copy these into the sanity-query.ts file.

import { groq } from "next-sanity";
const postData = `{
  title,
  metadata,
  slug,
  tags,
  author->{
    _id,
    name,
    slug,
    image,
    bio
  },
  mainImage,
  publishedAt,
  body
}`;

export const postQuery = groq`*[_type == "post"] ${postData}`;

export const postQueryBySlug = groq`*[_type == "post" && slug.current == $slug][0] ${postData}`;

export const postQueryByTag = groq`*[_type == "post" && $slug in tags[]->slug.current] ${postData}`;

export const postQueryByAuthor = groq`*[_type == "post" && author->slug.current == $slug] ${postData}`;

export const postQueryByCategory = groq`*[_type == "post" && category->slug.current == $slug] ${postData}`;

At the top, is the postData object, which defines all the properties we want to fetch.

Then the actual queries. Each query has the query first then the postData object.

These queries are self-descriptive, but for clarity here’s a quick explanation for the postQueryBySlug:

_type: The name of the document.

slug.current: Checking the slug of each of the documents if it matches with $slug (We will pass $slug with the fetch function).

postData: Filtering out the data we want to get i.e. the title, mainImage, and description.

We will use these queries to fetch the data from Sanity Studio.

Copy these codes into the sanity-utils.ts file.

import ImageUrlBuilder from "@sanity/image-url";
import { createClient, type QueryParams } from "next-sanity";
import clientConfig from "./config/client-config";
import { postQuery, postQueryBySlug } from "./sanity-query";
import { Blog } from "@/types/blog";

export const client = createClient(clientConfig);
export function imageBuilder(source: string) {
  return ImageUrlBuilder(clientConfig).image(source);
}

export async function sanityFetch<QueryResponse>({
  query,
  qParams,
  tags,
}: {
  query: string,
  qParams: QueryParams,
  tags: string[],
}): Promise<QueryResponse> {
  return (
    client.fetch <
    QueryResponse >
    (query,
    qParams,
    {
      cache: "force-cache",
      next: { tags },
    })
  );
}

export const getPosts = async () => {
  const data: Blog[] = await sanityFetch({
    query: postQuery,
    qParams: {},
    tags: ["post", "author", "category"],
  });
  return data;
};

export const getPostBySlug = async (slug: string) => {
  const data: Blog = await sanityFetch({
    query: postQueryBySlug,
    qParams: { slug },
    tags: ["post", "author", "category"],
  });

  return data;
};

Here’s a quick overview of what’s going on here:

client: creating a Sanity client with the config we created earlier. It will be used to fetch the data with GROQ.

imageBuilder: To use the post image. The images are provided from sanity cdn and it requires all these configs.

sanityFetch: It’s the function to fetch data with cache. ( We could just use fetch too but we are configuring this now so that we can just add the webhook and we are good to go. )

Create the type for the post

Since we are using typescript we need to write the Type for the post. You can see we are using Blog type on the query functions.

Create a blog.ts file inside the types folder and copy this type:

import { PortableTextBlock } from "sanity";

export type Author = {
  name: string,
  image: string,
  bio?: string,
  slug: {
    current: string,
  },
  _id?: number | string,
  _ref?: number | string,
};

export type Blog = {
  _id: number,
  title: string,
  slug: any,
  metadata: string,
  body: PortableTextBlock[],
  mainImage: any,
  author: Author,
  tags: string[],
  publishedAt: string,
};

All the types are normal, the PortableTextBlock is from sanity. It defines the type of the post body.

All the setup is done!

Let’s fetch the posts and render them on our Next.js project.

Render the content on the Next.js project

First, we will fetch all the posts and create the blog page. Then we will fetch the post by slug for the single post page.

Post Archive

Create the Blog component app/components/Blog/index.ts and add this code.

import { Blog } from "@/types/blog";
import Link from "next/link";
import React from "react";

const BlogItem = ({ blog }: { blog: Blog }) => {
  return (
    <Link
      href={`/blog/${blog.slug.current}`}
      className="block p-5 bg-white rounded-lg border border-gray-200 shadow-md hover:bg-gray-100 my-8"
    >
      <article>
        <h3 className="mb-1 text-2xl font-bold tracking-tight text-gray-700">
          {blog.title}
        </h3>
        <p className="mb-3 font-normal text-sm text-gray-600">
          {new Date(blog.publishedAt).toDateString()}
        </p>

        <p className="mb-3 font-normal text-gray-600">
          {blog.metadata.slice(0, 140)}...
        </p>
      </article>
    </Link>
  );
};

export default BlogItem;

Remove all the styles and code from globals.css (keep the tailwind utils) file and page.tsx file

Now add this to the page.tsx file inside (site)

import { getPosts } from "@/sanity/sanity-utils";
import BlogItem from "@/components/Blog";

export default async function Home() {
  const posts = await getPosts();

  return (
    <div className="py-5">
      {posts?.length > 0 ? (
        posts.map((post: any) => <BlogItem key={post._id} blog={post} />)
      ) : (
        <p>No posts found</p>
      )}
    </div>
  );
}

First import the getPosts function and BlogItem. Inside the Home component fetch the posts and render them.

We need a navbar to navigate between pages.

To save time I already created the Header file and loaded it inside the layout.tsx file.

Check out the source code and copy the Header component from there.

import Header from "@/components/Header";

export default function RootLayout({
  children,
}: Readonly<{
  children: React.ReactNode,
}>) {
  return (
    <>
      <Header />
      <main className="max-w-[1000px] mx-auto px-10 md:px-24">{children}</main>
    </>
  );
}

This is how it looks:

Single post

Now we need to create a single post page so that we can read the post.

Create the single post page inside blog/[slug]/page.tsx and paste this code snippet.

import React from "react";
import { getPostBySlug } from "@/sanity/sanity-utils";
import RenderBodyContent from "@/components/Blog/RenderBodyContent";

const SingleBlogPage = async ({ params }: { params: any }) => {
  const post = await getPostBySlug(params.slug);

  return (
    <article className="my-10">
      <div className="mb-5">
        <h1 className="text-3xl py-2">{post.title}</h1>
        <p className="pb-1">
          <span className="font-medium">Published:</span>
          {new Date(post.publishedAt).toDateString()}
          <span className="font-medium pl-2">by </span>
          {post.author.name}
        </p>

        <p>{post.metadata}</p>
      </div>

      <article className="prose lg:prose-xl">
        <RenderBodyContent post={post} />
      </article>
    </article>
  );
};

export default SingleBlogPage;

First import getPostBySlug and RenderBodyContent (we will create it in a while).

Fetch the post by slug and render the post with RenderBodyContent.

Render body content

It’s a custom component to render the post body.
Create RenderBodyContent.tsx file inside the components/Blog folder*.*

import config from "@/sanity/config/client-config";
import { Blog } from "@/types/blog";
import { PortableText } from "@portabletext/react";
import { getImageDimensions } from "@sanity/asset-utils";
import urlBuilder from "@sanity/image-url";
import Image from "next/image";

// lazy-loaded image component
const ImageComponent = ({ value, isInline }: any) => {
  const { width, height } = getImageDimensions(value);
  return (
    <div className="my-10 overflow-hidden rounded-[15px]">
      <Image
        src={
          urlBuilder(config)
            .image(value)
            .fit("max")
            .auto("format")
            .url() as string
        }
        width={width}
        height={height}
        alt={value.alt || "blog image"}
        loading="lazy"
        style={{
          display: isInline ? "inline-block" : "block",
          aspectRatio: width / height,
        }}
      />
    </div>
  );
};

const components = {
  types: {
    image: ImageComponent,
  },
};

const RenderBodyContent = ({ post }: { post: Blog }) => {
  return (
    <>
      <PortableText value={post?.body as any} components={components} />
    </>
  );
};

export default RenderBodyContent;

This component will handle special types differently. We are only handling Images here.

You can include code blocks, embeds, and many more. You can find more information on Sanity plugins on Sanity.

Here’s what it looks like.

To make the post look like this install the tailwind/typography plugin and load that inside the tailwind.config.ts file.

npm install @tailwindcss/typography

Webhook Integration

We will integrate Sanity webhook to fetch the data on time. Otherwise, you will have to deploy the site every time you write a post.

We already added revalidation on the fetch functions. Right now we need to generate the Keys and create the Webhook endpoint.

Generate the API key

Go to sanity.io/manage and navigate to API→Tokens then click on the Add API token button.

Give your API a name then choose Editor and save.

You will get an API key and save it on the env file

SANITY_API_KEY = "YOUR_API_KEY";

Create the Webhook endpoint

First, create the webhook endpoint app/api/revalidate/route.ts and add this code snippet.

import { revalidateTag } from "next/cache";
import { type NextRequest, NextResponse } from "next/server";
import { parseBody } from "next-sanity/webhook";

export async function POST(req: NextRequest) {
  try {
    const { body, isValidSignature } = await parseBody<{
      _type: string;
      slug?: string | undefined;
    }>(req, process.env.NEXT_PUBLIC_SANITY_HOOK_SECRET);

    if (!isValidSignature) {
      return new Response("Invalid Signature", { status: 401 });
    }

    if (!body?._type) {
      return new Response("Bad Request", { status: 400 });
    }

    revalidateTag(body._type);
    return NextResponse.json({
      status: 200,
      revalidated: true,
      now: Date.now(),
      body,
    });
  } catch (error: any) {
    console.error(error);
    return new Response(error.message, { status: 500 });
  }
}

We are using tag-based revalidation in this webhook.

This endpoint will be called by the webhook every time you create, delete, or update a document from Sanity Studio.

Generate the Webhook Secret

Navigate to sanity.io/manage API→Webhooks. Click on the Create Webhook button.

You will get a modal with a form. Fill in the form with the following pieces of information:

Name: Name of the Webhook

Description: Short description of what the webhook does (This is an optional field).

URL: Set the URL to https://YOUR_SITE_URL/api/revalidate

Dataset: Choose your desired dataset or leave the default value.

Trigger on: Set the hook to trigger on "Create", "Update", and "Delete".

Filter: Leave this field blank.

Projections: Set the projections to {_type, "slug": slug.current}
Status: Check the enable webhook box.

HTTP Method: POST.

Leave HTTP headers, API version, and Draft as default.

Secret: Give your webhook a unique secret and copy it.

Hit save to create your webhook.

Save your webhook in the .env file under this variable name.

SANITY_HOOK_SECRET=YOUR_SECRET

Testing the webhook: Go ahead and change the content of an Article and publish it. After that hard reload your website you should see the changes in real time.

Note: You can test webhook from the live site or you can choose tools like ngrok to expose the localhost URL and use that to test it.

Conclusion

That’s it you built a blog with Sanity CMS. Congrats! ?

Even though this guide looks so long, it’s just the beginning. Sanity has more features and options, you can build cool things.
It’s impossible to cover everything in a single article.

I will suggest you to checkout these resources to learn more and improve your blog

• Sanity docs
• Code highlighter

• Sanity Plugins

• Source Code

Connect With Me

I hope you enjoyed the post, if you want to stay conntected with me checkout my socials.
Would love to talk to you!

Twitter/x

Github

LinkedIn

Happy Coding.

以上是使用 Next.js 和 Sanity 建立現代部落格：逐步指南的詳細內容。更多資訊請關注PHP中文網其他相關文章！