sql-to-go

根据 MySQL CREATE TABLE 语句生成 Go 结构体：解析表名与列定义，将 MySQL 类型映射为 Go 类型，把 snake_case 列名转换为 CamelCase 字段名，并按主键、自增、非空、唯一、默认值、可空等约束生成 form、json、xorm 标签；可参考附带的类型映射文档。

weirubo

热度 49.53

最近更新 2026-05-11 17:32:08

分类：开发依赖：纯提示词应用场景：代码生成

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add sql-to-go

Metadata

name: sql-to-go
description: Convert MySQL CREATE TABLE statements to Go structs with form, json, and xorm tags. Use when converting SQL DDL to Go struct definitions, generating ORM models from database schemas, or creating Go types from MySQL tables. Supports MySQL data type mapping, snake_case to CamelCase conversion, and automatic tag generation.

SQL to Go Struct

Quick Start

Convert a MySQL CREATE TABLE statement to a Go struct with tags:

Parse the CREATE TABLE statement to extract table name, columns, and constraints
Map each MySQL column type to corresponding Go type
Convert snake_case column names to CamelCase field names
Generate struct with form, json, and xorm tags

Type Mapping

For complete MySQL to Go type mappings, see type_mappings.md.

Quick Reference

MySQL	Go	Notes
`INT`	`int`
`BIGINT`	`int64`
`VARCHAR`	`string`
`TEXT`	`string`
`DATETIME`	`time.Time`
`TIMESTAMP`	`time.Time`
`BOOL` / `BIT(1)`	`bool`
`DECIMAL`	`float64`	Or custom type

Column Name Conversion

Convert snake_case to CamelCase:

user_name    → UserName
created_at   → CreatedAt
user_id      → UserID
api_key      → APIKey
is_active    → IsActive

Special cases - acronyms remain uppercase:

id → ID
url → URL
uri → URI

Tag Generation

Generate three tags for each field:

form: HTTP form parameter name (snake_case)
json: JSON field name (snake_case)
xorm: XORM ORM column name and constraints

XORM Tag Patterns

Constraint	Tag
Primary key	`xorm:"pk"`
Auto increment	`xorm:"autoincr"`
Not null	`xorm:"not null"`
Unique	`xorm:"unique"`
Default value	`xorm:"default 'value'"`
Nullable	`xorm:"nullable"`
Index	`xorm:"index"`
Comment	`xorm:"comment 'text'"`

Combine constraints: xorm:"'user_id' pk autoincr"

Example

Input

CREATE TABLE users (
    id BIGINT UNSIGNED AUTO_INCREMENT PRIMARY KEY,
    username VARCHAR(50) NOT NULL UNIQUE,
    email VARCHAR(100) NOT NULL,
    age INT DEFAULT 0,
    created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
    is_active BOOLEAN DEFAULT TRUE,
    bio TEXT
) COMMENT='User table';

Output

package models

import "time"

// User represents the users table
type User struct {
    ID        uint64    `form:"id" json:"id" xorm:"'id' pk autoincr"`
    Username  string    `form:"username" json:"username" xorm:"'username' unique not null"`
    Email     string    `form:"email" json:"email" xorm:"'email' not null"`
    Age       int       `form:"age" json:"age" xorm:"'age' default 0"`
    CreatedAt time.Time `form:"created_at" json:"created_at" xorm:"'created_at'"`
    IsActive  bool      `form:"is_active" json:"is_active" xorm:"'is_active' default true"`
    Bio       string    `form:"bio" json:"bio" xorm:"'bio'"`
}

Workflow

Extract table info: Get table name, column definitions, constraints
Apply type mappings: See type_mappings.md for reference
Generate field name: Convert snake_case column to CamelCase
Add tags: Generate form/json/xorm tags based on column attributes
Handle special cases: Primary keys, indexes, defaults, comments

Nullable Columns

For nullable columns (allows NULL), use pointer types:

// MySQL: phone VARCHAR(20) NULL
Phone *string `form:"phone" json:"phone" xorm:"'phone' nullable"`

Struct Comment

Use table comment as struct comment, or generate from table name:

// User represents the users table