Guia de início rápido: usar o construtor de API de dados com SQL

Neste início rápido, cria endpoints REST e GraphQL para uma base de dados SQL local usando o Data API builder (DAB). Escolha o seu motor de base de dados para começar.

Pré-requisitos

Docker(opcional se já tiveres uma base de dados)
.NET 8 (ou mais recente)

Instalar a CLI do construtor de API de dados

Instala o pacote Microsoft.DataApiBuilder do NuGet como ferramenta .NET.

Use dotnet tool install para instalar a versão mais recente do Microsoft.DataApiBuilder com o --global argumento.
```
dotnet tool install --global Microsoft.DataApiBuilder
```
Observação

Se o pacote já estiver instalado, atualize-o usando dotnet tool update.
```
dotnet tool update --global Microsoft.DataApiBuilder
```
Verifique se a ferramenta está instalada com dotnet tool list utilizando o argumento --global.
```
dotnet tool list --global
```

Extrair a imagem da base de dados

Sugestão

Já tem uma base de dados? Avance para Criar e popular a base de dados, execute o script SQL para o seu motor e depois prossiga para Configurar o Data API Builder com a sua própria string de ligação.

Descarregue a imagem Docker para o seu motor de base de dados. Este passo pode demorar alguns minutos, dependendo da velocidade da sua ligação.

docker pull mcr.microsoft.com/mssql/server:2025-latest

docker pull postgres:16

docker pull mysql:8

Iniciar a base de dados

Executa uma instância de base de dados local no Docker.

docker run --name dab-mssql --env "ACCEPT_EULA=Y" --env "MSSQL_SA_PASSWORD=P@ssw0rd1" --publish 1433:1433 --detach mcr.microsoft.com/mssql/server:2025-latest

Sugestão

Se a porta 1433 já estiver em uso (por exemplo, por uma instalação local do SQL Server), mude --publish para uma porta host diferente, como 1434:1433, e atualize Server=localhost,1433 para Server=localhost,1434 nas etapas seguintes.

Verifique se o motor da base de dados está pronto antes de executar o próximo comando.

docker exec dab-mssql /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P "P@ssw0rd1" -C -Q "SELECT 1"

Se isto devolver um erro, espere alguns segundos e tente novamente.

docker run --name dab-postgres --env "POSTGRES_PASSWORD=P@ssw0rd1" --publish 5432:5432 --detach postgres:16

docker run --name dab-mysql --env "MYSQL_ROOT_PASSWORD=P@ssw0rd1" --publish 3306:3306 --detach mysql:8

Verifique se o motor da base de dados está pronto antes de executar o próximo comando.

docker exec dab-mysql mysql -uroot -pP@ssw0rd1 -e "SELECT 1"

Se isto devolver um erro, espere alguns segundos e tente novamente.

Criar e semear a base de dados

Cria uma base de dados e uma todos tabela, depois adiciona dados de exemplo. Se estiveres a usar Docker, não é necessário nenhum cliente SQL —docker exec executa os comandos diretamente dentro do contentor. Se estiveres a usar a tua própria base de dados, executa o script SQL na tua ferramenta preferida.

Cria a base de dados.

docker exec dab-mssql /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P "P@ssw0rd1" -C -Q "CREATE DATABASE todos;"

Crie a tabela e adicione dados de exemplo.

docker exec dab-mssql /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P "P@ssw0rd1" -C -d todos -Q "CREATE TABLE dbo.todos (id int PRIMARY KEY, title nvarchar(100) NOT NULL, completed bit NOT NULL DEFAULT 0); INSERT INTO dbo.todos VALUES (1, 'Walk the dog', 0), (2, 'Feed the fish', 0), (3, 'Comb the cat', 1);"

Sugestão

Usar o seu próprio SQL Server? Execute este script diretamente:

CREATE DATABASE todos;
GO
USE todos;
GO
CREATE TABLE dbo.todos (id int PRIMARY KEY, title nvarchar(100) NOT NULL, completed bit NOT NULL DEFAULT 0);
INSERT INTO dbo.todos VALUES (1, 'Walk the dog', 0), (2, 'Feed the fish', 0), (3, 'Comb the cat', 1);

Cria a base de dados.

docker exec dab-postgres psql -U postgres -c "CREATE DATABASE todos;"

Crie a tabela e adicione dados de exemplo.

docker exec dab-postgres psql -U postgres -d todos -c "CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed boolean NOT NULL DEFAULT false); INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);"

Sugestão

Usar o seu próprio servidor PostgreSQL? Execute este script diretamente:

CREATE DATABASE todos;
\c todos
CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed boolean NOT NULL DEFAULT false);
INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);

Cria a base de dados.

docker exec dab-mysql mysql -uroot -pP@ssw0rd1 -e "CREATE DATABASE todos;"

Crie a tabela e adicione dados de exemplo.

docker exec dab-mysql mysql -uroot -pP@ssw0rd1 -e "USE todos; CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed bool NOT NULL DEFAULT false); INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);"

Sugestão

Usar o seu próprio servidor MySQL? Execute este script diretamente:

CREATE DATABASE todos;
USE todos;
CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed bool NOT NULL DEFAULT false);
INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);

Configurar construtor de APIs de Dados

Cria um ficheiro de configuração DAB e adiciona uma entidade Todo .

Sugestão

Usar a tua própria base de dados? Substitua a corda dab init de ligação pela sua própria:

SQL Server:Server=<host>,<port>;Database=todos;User Id=<user>;Password=<password>;TrustServerCertificate=true;Encrypt=true;
PostgreSQL:Host=<host>;Port=5432;Database=todos;User ID=<user>;Password=<password>;
MySQL:Server=<host>;Port=3306;Database=todos;User=<user>;Password=<password>;

Inicialize a configuração.

dab init --database-type "mssql" --host-mode "Development" --connection-string "Server=localhost,1433;Database=todos;User Id=sa;Password=P@ssw0rd1;TrustServerCertificate=true;Encrypt=true;"

Adiciona a entidade Todo .

dab add Todo --source "dbo.todos" --permissions "anonymous:*"

Inicialize a configuração.

dab init --database-type "postgresql" --host-mode "Development" --connection-string "Host=localhost;Port=5432;Database=todos;User ID=postgres;Password=P@ssw0rd1;"

Adiciona a entidade Todo .

dab add Todo --source "public.todos" --permissions "anonymous:*"

Inicialize a configuração.

dab init --database-type "mysql" --host-mode "Development" --connection-string "Server=localhost;Port=3306;Database=todos;User=root;Password=P@ssw0rd1;"

Adiciona a entidade Todo .

dab add Todo --source "todos" --permissions "anonymous:*"

O seu dab-config.json ficheiro deverá agora parecer semelhante ao seguinte exemplo:

{
  "$schema": "https://github.com/Azure/data-api-builder/releases/download/vmajor.minor.patch/dab.draft.schema.json",
  "data-source": {
    "database-type": "mssql",
    "connection-string": "Server=localhost,1433;Database=todos;User Id=sa;Password=P@ssw0rd1;TrustServerCertificate=true;Encrypt=true;"
  },
  "runtime": {
    "rest": {
      "enabled": true
    },
    "graphql": {
      "enabled": true
    },
    "host": {
      "mode": "development",
      "cors": {
        "origins": ["*"]
      }
    }
  },
  "entities": {
    "Todo": {
      "source": "dbo.todos",
      "permissions": [
        {
          "role": "anonymous",
          "actions": [
            "*"
          ]
        }
      ]
    }
  }
}

{
  "$schema": "https://github.com/Azure/data-api-builder/releases/download/vmajor.minor.patch/dab.draft.schema.json",
  "data-source": {
    "database-type": "postgresql",
    "connection-string": "Host=localhost;Port=5432;Database=todos;User ID=postgres;Password=P@ssw0rd1;"
  },
  "runtime": {
    "rest": {
      "enabled": true
    },
    "graphql": {
      "enabled": true
    },
    "host": {
      "mode": "development",
      "cors": {
        "origins": ["*"]
      }
    }
  },
  "entities": {
    "Todo": {
      "source": "public.todos",
      "permissions": [
        {
          "role": "anonymous",
          "actions": [
            "*"
          ]
        }
      ]
    }
  }
}

{
  "$schema": "https://github.com/Azure/data-api-builder/releases/download/vmajor.minor.patch/dab.draft.schema.json",
  "data-source": {
    "database-type": "mysql",
    "connection-string": "Server=localhost;Port=3306;Database=todos;User=root;Password=P@ssw0rd1;"
  },
  "runtime": {
    "rest": {
      "enabled": true
    },
    "graphql": {
      "enabled": true
    },
    "host": {
      "mode": "development",
      "cors": {
        "origins": ["*"]
      }
    }
  },
  "entities": {
    "Todo": {
      "source": "todos",
      "permissions": [
        {
          "role": "anonymous",
          "actions": [
            "*"
          ]
        }
      ]
    }
  }
}

Sugestão

Poderá saltar os comandos dab init e dab add e criar diretamente o ficheiro dab-config.json com o conteúdo aqui mostrado.

Inicie a API

Use dab start para executar a ferramenta e criar endpoints de API para a sua entidade.

dab start

A saída deve incluir o endereço da API em execução.

      Successfully completed runtime initialization.
info: Microsoft.Hosting.Lifetime[14]
      Now listening on: <http://localhost:5000>

Sugestão

Neste exemplo, o aplicativo está sendo executado na localhost porta 5000. Seu aplicativo em execução pode ter um endereço e uma porta diferentes.

Testar a API

Abra o seu navegador e navegue até ao endpoint REST da entidade Todo .
```
http://localhost:5000/api/Todo
```

A resposta JSON deve incluir os três itens de tarefas pendentes.

{
  "value": [
    { "id": 1, "title": "Walk the dog", "completed": false },
    { "id": 2, "title": "Feed the fish", "completed": false },
    { "id": 3, "title": "Comb the cat", "completed": true }
  ]
}

Navegue até à página de documentação do Swagger em /swagger.
```
http://localhost:5000/swagger
```

Criar um aplicativo Web

Mostre as suas tarefas num navegador usando um ficheiro HTML simples. Crie um ficheiro nomeado todo.html usando o endpoint REST ou GraphQL.

REST
GraphQL

<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <title>Todo App</title>
  <style>
    body { font-family: sans-serif; max-width: 400px; margin: 2rem auto; }
    li.done { text-decoration: line-through; color: gray; }
    #error { color: red; }
  </style>
</head>
<body>
  <h1>Todos</h1>
  <ul id="list"></ul>
  <p id="error"></p>
  <script>
    fetch('http://localhost:5000/api/Todo')
      .then(r => r.json())
      .then(data => {
        const ul = document.getElementById('list');
        data.value.forEach(todo => {
          const li = document.createElement('li');
          li.textContent = todo.title;
          if (todo.completed) li.className = 'done';
          ul.appendChild(li);
        });
      })
      .catch(() => {
        document.getElementById('error').textContent =
          'Could not reach the API. Make sure DAB is running on http://localhost:5000.';
      });
  </script>
</body>
</html>

<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <title>Todo App</title>
  <style>
    body { font-family: sans-serif; max-width: 400px; margin: 2rem auto; }
    li.done { text-decoration: line-through; color: gray; }
    #error { color: red; }
  </style>
</head>
<body>
  <h1>Todos</h1>
  <ul id="list"></ul>
  <p id="error"></p>
  <script>
    fetch('http://localhost:5000/graphql', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ query: '{ todos { items { id title completed } } }' })
    })
      .then(r => r.json())
      .then(data => {
        const ul = document.getElementById('list');
        data.data.todos.items.forEach(todo => {
          const li = document.createElement('li');
          li.textContent = todo.title;
          if (todo.completed) li.className = 'done';
          ul.appendChild(li);
        });
      })
      .catch(() => {
        document.getElementById('error').textContent =
          'Could not reach the API. Make sure DAB is running on http://localhost:5000.';
      });
  </script>
</body>
</html>

Abra todo.html no seu navegador. A página recolhe todos os itens de tarefas pendentes e exibe-os como uma lista, com os itens concluídos mostrados em riscado.

Importante

A cors definição na sua configuração permite que este ficheiro HTML — aberto no seu sistema de ficheiros local — chame a API. Sem isso, o navegador bloqueia o pedido.

Limpeza

Pare e remova o contentor Docker quando terminar.

docker stop dab-mssql && docker rm dab-mssql

docker stop dab-postgres && docker rm dab-postgres

docker stop dab-mysql && docker rm dab-mysql

Usa o GitHub Copilot para recriar este quickstart

Abre o espaço de trabalho onde queres criar o exemplo no Visual Studio Code, muda o GitHub Copilot para o modo agente e cola este prompt.

You are GitHub Copilot running in agent mode. Recreate the Data API builder basic SQL quickstart as a complete, runnable local project in the current VS Code workspace under `quickstart-00-basic-sql`. Build a local Docker-based sample that starts one database engine, creates and seeds a `todos` database, configures Data API builder (DAB), exposes REST, GraphQL, and MCP endpoints, adds MCP Inspector for DAB MCP testing, and creates a small static web app that calls DAB. Keep the implementation minimal, but make the web interface neat and approachable: responsive layout, accessible labels, clear loading and error states, and simple styling that is polished rather than austere.

Source repository guidance: no dedicated Azure-Samples repository currently appears for this basic SQL quickstart. However, https://github.com/Azure-Samples/dab-2.0-quickstart-web_anon-api_anon-db_sql_auth is very close and features such as the database and the web site can be reused. If internet access is available, review that site and reuse shared file patterns when they match this local Docker quickstart. Otherwise, implement from this article and the current Data API builder docs. Do not invent a different architecture or add extra services beyond this prompt.

Minimize user interaction. Use the defaults in this prompt and make reasonable best guesses for noncritical choices. Do not ask for a root folder or project folder name; use the current VS Code workspace and the default subfolder. Ask only when you need approval for resource changes, secrets, permissions, materially higher cost, external account choices, or an ambiguous requirement that affects the architecture.

Start with a short plan and proceed with safe defaults before you create files or run commands. Use SQL Server, the default `todos` schema and seed data, SQL Commander, the listed non-default host ports, and local Docker only unless the user explicitly asks for a different database engine or an Azure extension. Ask only these questions if the values aren't already available from the environment or prior context:

- If you want an Azure extension, which Azure subscription, primary region, fallback region, and resource group should it use? Default fallback region: `westus2`.

Show a short checklist before implementation. Include phases for project scaffold, Docker Compose, database initialization, DAB configuration, web app, MCP Inspector, validation, and cleanup. Proceed with local files and local Docker validation without asking for extra confirmation. Do not create Azure resources for this quickstart unless the user explicitly asks for an Azure extension and approves the exact Azure command set.

After you start, continue working without asking status-check questions. If a command, build, container, endpoint, or validation step fails, inspect the error, adjust the project, rerun the step, and continue. Keep iterating until the sample runs end-to-end or you hit a blocker that requires user action.

Use cost-first defaults. The default solution is local Docker only with no Azure cost. If the user asks for an Azure extension, choose the cheapest option that satisfies the selected database engine: use a free Azure SQL database offer when SQL Server is selected and the subscription and region support it; otherwise choose the lowest-cost Azure database option that supports the selected SQL Server, PostgreSQL, or MySQL scenario. Use Azure Container Apps consumption, minimal CPU and memory, Basic Azure Container Registry, minimal Log Analytics retention, and no always-on or dedicated plans unless required. Prioritize finishing the project. Treat regional provisioning limits as expected adjustment points, not failures: if the primary region can't provision a required service or free SQL option, use the approved fallback region such as `westus2`, and continue the deployment. Ask the user only when both the primary and fallback regions can't satisfy the requirements, when a change would materially increase cost, when a new permission is required, or when you need approval for Azure commands that create or change resources beyond the already-approved plan.

Verify prerequisites and report only missing items: Docker Desktop running, .NET SDK, DAB CLI, and a shell that can run Docker Compose. Use the DAB CLI docs while building: https://learn.microsoft.com/azure/data-api-builder/command-line/.

Use these docs during implementation:

- DAB CLI reference: https://learn.microsoft.com/azure/data-api-builder/command-line/
- `dab init`: https://learn.microsoft.com/azure/data-api-builder/command-line/dab-init
- `dab add`: https://learn.microsoft.com/azure/data-api-builder/command-line/dab-add
- `dab validate`: https://learn.microsoft.com/azure/data-api-builder/command-line/dab-validate
- `dab start`: https://learn.microsoft.com/azure/data-api-builder/command-line/dab-start
- DAB MCP overview: https://learn.microsoft.com/azure/data-api-builder/mcp/overview
- DAB configuration: https://learn.microsoft.com/azure/data-api-builder/configuration/

Create this structure under the sample folder:

- `docker-compose.yml` for the selected database, DAB, MCP Inspector, and the web app.
- `.env` for local passwords and connection strings.
- `.gitignore` with `.env`, `**/bin`, and `**/obj`.
- `database/` for selected-engine initialization scripts.
- `data-api/dab-config.json` for DAB configuration.
- `web-app/` for static HTML, CSS, and JavaScript.
- `mcp-inspector/README.md` with the auto-connect URL.
- `README.md` with run, validation, troubleshooting, and cleanup steps.

Handle secrets first. Add `.env` to `.gitignore` before writing passwords. Use `DATABASE_PASSWORD` and `DATABASE_CONNECTION_STRING`. Never print secret values. Use `@env('DATABASE_CONNECTION_STRING')` in `dab-config.json`. Avoid `$` in Docker Compose passwords because Compose treats `$` as variable interpolation.

Use Docker Compose, not raw `docker run`, for the generated project. Containers must talk by service name, not `localhost`. Mount `data-api/dab-config.json` into DAB read-only at `/App/dab-config.json`. Use health checks and `depends_on` so DAB starts after the selected database is healthy.

Implement database initialization explicitly. For PostgreSQL and MySQL, mount selected-engine scripts into `/docker-entrypoint-initdb.d` for first-run initialization. For SQL Server, add a one-shot init service or setup script that waits for the `db` service to become healthy and then runs `sqlcmd` to create the `todos` database, table, and seed rows. Do not assume the database health check creates the database or schema.

Use one of these selected-engine configurations.

SQL Server:

```yaml
services:
  db:
    image: mcr.microsoft.com/mssql/server:2025-latest
    environment:
      ACCEPT_EULA: "Y"
      MSSQL_SA_PASSWORD: ${DATABASE_PASSWORD}
    ports:
      - "14330:1433"
    healthcheck:
      test: /opt/mssql-tools18/bin/sqlcmd -S localhost -U sa -P "${DATABASE_PASSWORD}" -C -Q "SELECT 1" || exit 1
      interval: 10s
      timeout: 5s
      retries: 10
```

PostgreSQL:

```yaml
services:
  db:
    image: postgres:16
    environment:
      POSTGRES_PASSWORD: ${DATABASE_PASSWORD}
    ports:
      - "54320:5432"
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U postgres"]
      interval: 10s
      timeout: 5s
      retries: 10
```

MySQL:

```yaml
services:
  db:
    image: mysql:8
    environment:
      MYSQL_ROOT_PASSWORD: ${DATABASE_PASSWORD}
    ports:
      - "33060:3306"
    healthcheck:
      test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
      interval: 10s
      timeout: 5s
      retries: 10
```

Use the selected database engine only. Do not scaffold all three engines unless the user asks for a matrix sample.

Use the matching schema and connection details.

SQL Server:

```sql
CREATE DATABASE todos;
GO
USE todos;
GO
CREATE TABLE dbo.todos (id int PRIMARY KEY, title nvarchar(100) NOT NULL, completed bit NOT NULL DEFAULT 0);
INSERT INTO dbo.todos VALUES (1, 'Walk the dog', 0), (2, 'Feed the fish', 0), (3, 'Comb the cat', 1);
```

```text
DATABASE_CONNECTION_STRING=Server=db;Database=todos;User Id=sa;Password=<password>;TrustServerCertificate=true;Encrypt=true;
```

PostgreSQL:

```sql
CREATE DATABASE todos;
\c todos
CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed boolean NOT NULL DEFAULT false);
INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);
```

```text
DATABASE_CONNECTION_STRING=Host=db;Port=5432;Database=todos;User ID=postgres;Password=<password>;
```

MySQL:

```sql
CREATE DATABASE todos;
USE todos;
CREATE TABLE todos (id int PRIMARY KEY, title varchar(100) NOT NULL, completed bool NOT NULL DEFAULT false);
INSERT INTO todos VALUES (1, 'Walk the dog', false), (2, 'Feed the fish', false), (3, 'Comb the cat', true);
```

```text
DATABASE_CONNECTION_STRING=Server=db;Port=3306;Database=todos;User=root;Password=<password>;
```

Use the DAB CLI workflow for the selected engine and validate after each config change.

SQL Server:

```dotnetcli
dab init --config data-api/dab-config.json --database-type mssql --host-mode Development --connection-string "@env('DATABASE_CONNECTION_STRING')" --rest.enabled true --graphql.enabled true --mcp.enabled true
dab add Todo --config data-api/dab-config.json --source dbo.todos --source.type table --permissions "anonymous:*" --mcp.dml-tools true
dab validate --config data-api/dab-config.json
```

PostgreSQL:

```dotnetcli
dab init --config data-api/dab-config.json --database-type postgresql --host-mode Development --connection-string "@env('DATABASE_CONNECTION_STRING')" --rest.enabled true --graphql.enabled true --mcp.enabled true
dab add Todo --config data-api/dab-config.json --source public.todos --source.type table --permissions "anonymous:*" --mcp.dml-tools true
dab validate --config data-api/dab-config.json
```

MySQL:

```dotnetcli
dab init --config data-api/dab-config.json --database-type mysql --host-mode Development --connection-string "@env('DATABASE_CONNECTION_STRING')" --rest.enabled true --graphql.enabled true --mcp.enabled true
dab add Todo --config data-api/dab-config.json --source todos --source.type table --permissions "anonymous:*" --mcp.dml-tools true
dab validate --config data-api/dab-config.json
```

Use this DAB container pattern in Compose:

```yaml
  data-api:
    image: mcr.microsoft.com/azure-databases/data-api-builder:latest
    environment:
      DATABASE_CONNECTION_STRING: ${DATABASE_CONNECTION_STRING}
    ports:
      - "5000:5000"
    volumes:
      - ./data-api/dab-config.json:/App/dab-config.json:ro
    depends_on:
      db:
        condition: service_healthy
```

Configure DAB CORS before you start the browser-based web app. Do not leave `runtime.host.cors.origins` as `[]`. Set it to include the exact web app origin, including scheme and port, such as `http://localhost:8000` for this Docker Compose web app. Keep `allow-credentials` set to `false` unless the sample explicitly uses browser credentials or cookies. Direct REST, GraphQL, or Swagger requests can succeed even when the browser blocks JavaScript fetch calls, so browser-origin CORS must be configured and validated separately.

Add MCP Inspector with the auto-connect URL. Use Streamable HTTP and omit auth only for local development.

```yaml
  mcp-inspector:
    image: ghcr.io/modelcontextprotocol/inspector:latest
    environment:
      HOST: 0.0.0.0
      MCP_AUTO_OPEN_ENABLED: "false"
      DANGEROUSLY_OMIT_AUTH: "true"
    ports:
      - "6274:6274"
      - "6277:6277"
    depends_on:
      - data-api
```

```text
http://localhost:6274/?transport=streamable-http&serverUrl=http%3A%2F%2Fdata-api%3A5000%2Fmcp
```

Use the Compose service name `data-api` in the MCP Inspector auto-connect URL because MCP Inspector runs in the Compose network. Also document a host-side MCP URL for VS Code or direct browser testing:

```text
http://localhost:5000/mcp
```

For SQL Server only, include SQL Commander if the user wants a database browser. Use env var `ConnectionStrings__db` and include `TrustServerCertificate=true`.

```yaml
  sql-commander:
    image: jerrynixon/sql-commander:latest
    environment:
      ConnectionStrings__db: ${DATABASE_CONNECTION_STRING}
    ports:
      - "8080:8080"
    depends_on:
      db:
        condition: service_healthy
```

Build the static web app with minimal code and a polished UI. It should show the todo list, loading state, empty state, error state, API base URL, and quick links to REST, GraphQL, Swagger, and MCP Inspector. Keep dependencies minimal; use plain HTML, CSS, and JavaScript unless the user asks for a framework.

Validate before reporting success:

- `docker compose up -d` starts the selected database, DAB, MCP Inspector, and web app.
- The selected database health check passes.
- The `todos` database and table exist with three seeded rows.
- A direct database query returns the three seeded todo rows.
- `dab validate --config data-api/dab-config.json` exits with code 0.
- DAB `/health` returns a 2xx response.
- REST returns the three todo rows at `http://localhost:5000/api/Todo`.
- GraphQL returns the three todo rows.
- Swagger opens at `http://localhost:5000/swagger`.
- A browser-origin request from the web app origin, for example `http://localhost:8000`, receives an `Access-Control-Allow-Origin` response header that matches that origin.
- MCP Inspector opens with the auto-connect URL and can list DAB tools.
- The web site returns a successful HTTP response.
- The web app displays the todo rows and looks neat, not austere.
- `README.md` includes run, validation, troubleshooting, and cleanup steps.

Do not report final URLs, asset locations, or a success summary until you directly verify database connectivity and query results, a 2xx DAB health response, and a successful web site response. This validation ensures the sample works without requiring the developer to check.

Próximo passo

pontos finais REST

Comentários

Esta página foi útil?

Last updated on 2026-06-04

Guia de início rápido: usar o construtor de API de dados com SQL

Pré-requisitos

Instalar a CLI do construtor de API de dados

Extrair a imagem da base de dados

Iniciar a base de dados

Criar e semear a base de dados

Configurar construtor de APIs de Dados

Inicie a API

Testar a API

Criar um aplicativo Web

Limpeza

Usa o GitHub Copilot para recriar este quickstart

Próximo passo

Comentários

Recursos adicionais