在日常开发中，GraphQL Schema 设计最佳实践的使用频率越来越高。本文系统地讲解其用法、原理和优化策略。

快速上手

实际项目中的用法会更复杂一些：

import { ApolloClient, InMemoryCache, createHttpLink } from '@apollo/client'
import { setContext } from '@apollo/client/link/context'

const httpLink = createHttpLink({ uri: '/graphql' })
const authLink = setContext((_, { headers }) => {
  const token = localStorage.getItem('token')
  return { headers: { ...headers, authorization: `Bearer ${token}` } }
})

const client = new ApolloClient({
  link: authLink.concat(httpLink),
  cache: new InMemoryCache({
    typePolicies: {
      Query: {
        fields: {
          users: { keyArgs: ['filter'], merge: (e, i) => ({ ...i, edges: [...(e?.edges||[]), ...i.edges] }) }
        }
      }
    }
  })
})

通过这种方式，代码的可测试性和可扩展性都得到了提升。

内部原理

以下是一个完整的示例：

import { ApolloClient, InMemoryCache, createHttpLink } from '@apollo/client'
import { setContext } from '@apollo/client/link/context'

const httpLink = createHttpLink({ uri: '/graphql' })
const authLink = setContext((_, { headers }) => {
  const token = localStorage.getItem('token')
  return { headers: { ...headers, authorization: `Bearer ${token}` } }
})

const client = new ApolloClient({
  link: authLink.concat(httpLink),
  cache: new InMemoryCache({
    typePolicies: {
      Query: {
        fields: {
          users: { keyArgs: ['filter'], merge: (e, i) => ({ ...i, edges: [...(e?.edges||[]), ...i.edges] }) }
        }
      }
    }
  })
})

注意边界条件处理，这在生产环境中至关重要。

业务实战

关键在于理解核心逻辑：

import { ApolloClient, InMemoryCache, createHttpLink } from '@apollo/client'
import { setContext } from '@apollo/client/link/context'

const httpLink = createHttpLink({ uri: '/graphql' })
const authLink = setContext((_, { headers }) => {
  const token = localStorage.getItem('token')
  return { headers: { ...headers, authorization: `Bearer ${token}` } }
})

const client = new ApolloClient({
  link: authLink.concat(httpLink),
  cache: new InMemoryCache({
    typePolicies: {
      Query: {
        fields: {
          users: { keyArgs: ['filter'], merge: (e, i) => ({ ...i, edges: [...(e?.edges||[]), ...i.edges] }) }
        }
      }
    }
  })
})

性能优化需要结合具体场景，不是所有情况都需要过度优化。

性能对比

我们可以通过以下方式来改进：

import { ApolloClient, InMemoryCache, createHttpLink } from '@apollo/client'
import { setContext } from '@apollo/client/link/context'

const httpLink = createHttpLink({ uri: '/graphql' })
const authLink = setContext((_, { headers }) => {
  const token = localStorage.getItem('token')
  return { headers: { ...headers, authorization: `Bearer ${token}` } }
})

const client = new ApolloClient({
  link: authLink.concat(httpLink),
  cache: new InMemoryCache({
    typePolicies: {
      Query: {
        fields: {
          users: { keyArgs: ['filter'], merge: (e, i) => ({ ...i, edges: [...(e?.edges||[]), ...i.edges] }) }
        }
      }
    }
  })
})

这套方案已经在线上稳定运行了半年以上，经过了实际验证。

小结

• 代码示例仅供参考，需根据业务场景调整
• GraphQL Schema 设计最佳实践不是银弹，需要根据项目规模和技术栈选择
• 理解底层原理比记住 API 更重要
• 生产环境使用前务必做好兼容性验证
• 团队协作中约定和文档比技术本身更重要