Nextjs 14 Form Submit with tRPC

Introdution

In this article,we are gonna enumerate some kinds of form submitting with trpc concretely, globally in detail. If need, please see 《NextJs 14 从入门到精通之tRPC快速入门》

Prepare trpc base enviroment

Form component typically is used in Client Component, you can submmit with http request or call server action directly. So we start along this thought

We need to create tRPC appRouter at first

bun add @trpc/server@next @trpc/client@next

ts 复制代码
import { initTRPC } from '@trpc/server';
import { db } from '@/server/db';

export const createTRPCContext = async (opts: {
  req?: NextRequest;
}) => {
  // get session
  const session = { user: {userId: 'xxx'}, sessionId: 'xxxx' };

  return {
    db,
    session,
    ...opts,
  };
};

const t = initTRPC.context<typeof createTRPCContext>().create()

const publicProcedure = t.procedure;

const protectedProcedure = t.procedure.use(({ ctx, next }) => {
    if (!ctx.session || !ctx.session.userId) {
      throw new TRPCError({ code: 'UNAUTHORIZED' });
    }

    return next({
      ctx: {
        // infers the `session` as non-nullable
        session: {
          ...ctx.session,
          userId: ctx.session.userId,
        },
      },
    });
});

export const appRouter = t.router({
    createUser: protectedProcedure
        .input(z.object({
          name: z.string(),
          password: z.string()
        }))
        .mutation(async ({ ctx, input }) => {
            const { userId } = ctx;
            const data = await ctx.db.user.create({
              data: {...input, userId }
            })
            return { status: 200, data, message: 'success' }
        })
})

Http Request with tRPC

Firstly, make tRPC appRouter match NextJs api-router

ts 复制代码
// src/app/api/trpc/[...trpc]/route.ts
import { fetchRequestHandler } from '@trpc/server/adapters/fetch';
import { appRouter } from '@/server/api/root';
import { createTRPCContext } from '@/server/api/trpc';

const handler = (req: NextRequest) =>
  fetchRequestHandler({
    endpoint: '/api/trpc',
    req,
    router: appRouter,
    createContext: () => createTRPCContext(),
    onError: ({ path, error }) => {
      console.log('------', error);
    },
  });

export { handler as GET, handler as POST };

Secondly, Create tRPC Client with react-query

bun add @trpc/react-query@next @tanstack/react-query

tsx 复制代码
// tRPCReact.ts
import { QueryClient, QueryClientProvider } from '@tanstack/react-query';
import { createTRPCReact } from '@trpc/react-query';
import { unstable_httpBatchStreamLink } from '@trpc/client';
import { useState } from 'react';

const createQueryClient = () => new QueryClient();

let clientQueryClientSingleton: QueryClient | undefined = undefined;

const getQueryClient = () => {
  if (typeof window === 'undefined') {
    return createQueryClient();
  }
  return (clientQueryClientSingleton ??= createQueryClient());
};

export const api = createTRPCReact<AppRouter>();

export function TRPCReactProvider(props: { children: React.ReactNode }) {
  const queryClient = getQueryClient();

  const [trpcClient] = useState(() =>
    api.createClient({
      links: [
        unstable_httpBatchStreamLink({
          url: 'http:localhost:3000/api/trpc',
        }),
      ],
    })
  );

  return (
    <QueryClientProvider client={queryClient}>
      <api.Provider client={trpcClient} queryClient={queryClient}>
        {props.children}
      </api.Provider>
    </QueryClientProvider>
  );
}

Thirdly, used in form component

  • To use react-query, your component must be wrapped with TRPCReactProvider before.
ts 复制代码
export default function RootLayout({ children }) {
    return (
        <html lang="en">
            <body>
                <TRPCReactProvider>{children}</TRPCReactProvider>
            </body>
        </html>
    )
}
  • In form component.
tsx 复制代码
'use client';
import { api } from 'tRPCReact';
export function Form() {
    const create = api.createUser.useMutation({
      onSuccess: () => {}
      onError: (e) => {}
    })
    
    return (
        <form>
            {/* ... */}
            <button onClick={() => create.mutate({
                name, password
            })} >Submit</button>
        </form>
    )
}

Server Action with tRPC's createCaller

NextJs 14 supports calling server action directly for form submmiting. And similarly,tRPC supports server side calls. so we can do following this:

  • Go back to the file of appRouter, export default createCallerFactory
ts 复制代码
// ...
const t = initTRPC.context<typeof createTRPCContext>().create()
// ...

export const createCallerFactory = t.createCallerFactory;
  • Create trpc caller
ts 复制代码
// tRPCServer
const createCaller = createCallerFactory(appRouter);
export const api = createCaller(() => createTRPCContext());
  • Used in form component
tsx 复制代码
'use client';
import { api } from 'tRPCServer';

export function Form() {
    const create = async (formData, FormData) {
        'use server'
        try {
            api.createUser(formData)
        } catch (e) {}
    }

    return (
        <form action={create}>
            {/* ... */}
        </form>
    )
}

Server Action with tRPC's experimental_caller

This is an experimental feature, we should not better to use it. it's based on top of procedure, likely a middleware, finally return a new procedure.

And we will not use the previous defined appRouter. we should re-write the inner logic.

so we can do by following:

  • Export default createCallerFactory
ts 复制代码
// ....
const t = initTRPC.context<typeof createTRPCContext>().create();
// ....
// experimental: form action
export const createActionCaller = protectedProcedure.experimental_caller;
  • Create a file for createActionCaller
ts 复制代码
// tRPCAction
import { experimental_nextAppDirCaller } from '@trpc/server/adapters/next-app-dir';

export const tRPCAction = createActionCaller(
  experimental_nextAppDirCaller({
    createContext: () => createTRPCContext()
  })
);
  • Define a Server Action as a file and write business logic
ts 复制代码
'use server';
import { tRPCAction } from 'tRPCAction';
import { z } from 'zod';

export const createUserAction = tRPCAction
        .input(z.object({
          name: z.string(),
          password: z.string()
        }))
        .mutation(async ({ ctx, input }) => {
            const { userId } = ctx;
            const data = await ctx.db.user.create({
              data: {...input, userId }
            })
            return { status: 200, data, message: 'success' }
        })
  • Used in form component
tsx 复制代码
'use client';
import { createUserAction } from 'actions';

export function Form() {
    return (
        <form action={createUserAction}>
            {/* ... */}
        </form>
    )
}

Summarize

In Next.js 14, using server Actions to submit a form and call tRPC's createCaller is a straightforward way. Both offer different ways to invoke server-side code, depending on your needs and development style. Here is an explanation of the two ways:

  • createCaller is a way to call the server-side tRPC endpoint directly. It is suitable for situations where you need to call the tRPC endpoint directly in server-side code, such as during server Actions or other server-side processing.

  • experimental_caller provides an experimental method for calling tRPC methods on the server side. It is a more straightforward API, but because it is experimental, it may change in future releases

In fact, I suggest to use createCaller, because this way is used directly and once appRouter logic is defined. we can use server side call and server action.

相关推荐
Csvn19 分钟前
OpenSpec 详细使用教程
前端
明月_清风19 分钟前
加密解密系统完全指南:原理剖析与 Go 实践
后端
之歆1 小时前
Day19_LESS 完全指南——从入门到工程实践
前端·css·less
小江的记录本1 小时前
【JVM虚拟机】垃圾回收GC:垃圾收集器:CMS:核心原理、回收流程、优缺点、废弃原因(附《思维导图》+《面试高频考点清单》)
java·jvm·后端·python·spring·面试·maven
云水一下2 小时前
HTML5 从入门到精通:实战收官——从零搭建完整静态网站,综合运用所有知识
前端·html5
不总是2 小时前
Windows 系统 Node.js 免安装版(zip)安装与配置教程(2026 最新)
前端·windows·node.js
冬奇Lab2 小时前
每日一个开源项目(第105篇):Twenty - 跳出 Salesforce 的圈套,定义现代开源 CRM
前端·后端·开源
zhangyao9403303 小时前
开发pc端时,表格的高度怎么设置才能铺满页面
前端·javascript·elementui
ServBay3 小时前
月之暗面 Kimi Code 0.4.0 发布,终端 AI 编码助手全面采用 TypeScript,实现毫秒级启动
后端·aigc·ai编程
小江的记录本3 小时前
【JVM虚拟机】垃圾回收GC:垃圾回收算法:标记-清除、标记-复制、标记-整理、分代收集(附《思维导图》+《面试高频考点清单》)
java·jvm·后端·python·算法·安全·面试