feat(dc/tool-c): 完成AI代码生成服务(Day 3 MVP)

核心功能:
- 新增AICodeService(550行):AI代码生成核心服务
- 新增AIController(257行):4个API端点
- 新增dc_tool_c_ai_history表:存储对话历史
- 实现自我修正机制:最多3次智能重试
- 集成LLMFactory:复用通用能力层
- 10个Few-shot示例:覆盖Level 1-4场景

技术优化:
- 修复NaN序列化问题(Python端转None)
- 修复数据传递问题(从Session获取真实数据)
- 优化System Prompt(明确环境信息)
- 调整Few-shot示例(移除import语句)

测试结果:
- 通过率:9/11(81.8%) 达到MVP标准
- 成功场景:缺失值处理、编码、分箱、BMI、筛选、填补、统计、分类
- 待优化:数值清洗、智能去重(已记录技术债务TD-C-006)

API端点:
- POST /api/v1/dc/tool-c/ai/generate(生成代码)
- POST /api/v1/dc/tool-c/ai/execute(执行代码)
- POST /api/v1/dc/tool-c/ai/process(生成并执行,一步到位)
- GET /api/v1/dc/tool-c/ai/history/:sessionId(对话历史)

文档更新:
- 新增Day 3开发完成总结(770行)
- 新增复杂场景优化技术债务(TD-C-006)
- 更新工具C当前状态文档
- 更新技术债务清单

影响范围:
- backend/src/modules/dc/tool-c/*(新增2个文件,更新1个文件)
- backend/scripts/create-tool-c-ai-history-table.mjs(新增)
- backend/prisma/schema.prisma(新增DcToolCAiHistory模型)
- extraction_service/services/dc_executor.py(NaN序列化修复)
- docs/03-业务模块/DC-数据清洗整理/*(5份文档更新)

Breaking Changes: 无

总代码行数:+950行

Refs: #Tool-C-Day3
This commit is contained in:
2025-12-07 16:21:32 +08:00
parent 2348234013
commit f01981bf78
68 changed files with 6257 additions and 17 deletions

View File

@@ -0,0 +1,142 @@
/**
* 创建 Tool C Session 表
*
* 执行方式node scripts/create-tool-c-table.js
*/
const { PrismaClient } = require('@prisma/client');
const fs = require('fs');
const path = require('path');
const prisma = new PrismaClient();
async function createToolCTable() {
console.log('========================================');
console.log('开始创建 Tool C Session 表');
console.log('========================================\n');
try {
// 1. 检查表是否已存在
console.log('[1/4] 检查表是否已存在...');
const checkResult = await prisma.$queryRawUnsafe(`
SELECT EXISTS (
SELECT FROM information_schema.tables
WHERE table_schema = 'dc_schema'
AND table_name = 'dc_tool_c_sessions'
) as exists
`);
const tableExists = checkResult[0].exists;
if (tableExists) {
console.log('✅ 表已存在: dc_schema.dc_tool_c_sessions');
console.log('\n是否需要重新创建(这将删除现有数据)');
console.log('如需重新创建,请手动执行: DROP TABLE dc_schema.dc_tool_c_sessions CASCADE;\n');
return;
}
console.log('✅ 表不存在,准备创建\n');
// 2. 创建表
console.log('[2/4] 创建表 dc_tool_c_sessions...');
await prisma.$executeRawUnsafe(`
CREATE TABLE dc_schema.dc_tool_c_sessions (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id VARCHAR(255) NOT NULL,
file_name VARCHAR(500) NOT NULL,
file_key VARCHAR(500) NOT NULL,
total_rows INTEGER NOT NULL,
total_cols INTEGER NOT NULL,
columns JSONB NOT NULL,
encoding VARCHAR(50),
file_size INTEGER NOT NULL,
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
expires_at TIMESTAMP NOT NULL
)
`);
console.log('✅ 表创建成功\n');
// 3. 创建索引
console.log('[3/4] 创建索引...');
await prisma.$executeRawUnsafe(`
CREATE INDEX idx_dc_tool_c_sessions_user_id ON dc_schema.dc_tool_c_sessions(user_id)
`);
await prisma.$executeRawUnsafe(`
CREATE INDEX idx_dc_tool_c_sessions_expires_at ON dc_schema.dc_tool_c_sessions(expires_at)
`);
console.log('✅ 索引创建成功\n');
// 4. 添加注释
console.log('[4/4] 添加表注释...');
await prisma.$executeRawUnsafe(`
COMMENT ON TABLE dc_schema.dc_tool_c_sessions IS 'Tool C (科研数据编辑器) Session会话表'
`);
await prisma.$executeRawUnsafe(`
COMMENT ON COLUMN dc_schema.dc_tool_c_sessions.file_key IS 'OSS存储路径: dc/tool-c/sessions/{timestamp}-{fileName}'
`);
await prisma.$executeRawUnsafe(`
COMMENT ON COLUMN dc_schema.dc_tool_c_sessions.columns IS '列名数组 ["age", "gender", "diagnosis"]'
`);
await prisma.$executeRawUnsafe(`
COMMENT ON COLUMN dc_schema.dc_tool_c_sessions.expires_at IS '过期时间创建后10分钟'
`);
console.log('✅ 注释添加成功\n');
// 5. 验证表创建
console.log('========================================');
console.log('验证表结构');
console.log('========================================\n');
const columns = await prisma.$queryRawUnsafe(`
SELECT column_name, data_type, is_nullable
FROM information_schema.columns
WHERE table_schema = 'dc_schema'
AND table_name = 'dc_tool_c_sessions'
ORDER BY ordinal_position
`);
console.log('表结构:');
console.table(columns);
const indexes = await prisma.$queryRawUnsafe(`
SELECT indexname, indexdef
FROM pg_indexes
WHERE schemaname = 'dc_schema'
AND tablename = 'dc_tool_c_sessions'
`);
console.log('\n索引:');
console.table(indexes);
console.log('\n========================================');
console.log('🎉 Tool C Session 表创建成功!');
console.log('========================================\n');
console.log('表名: dc_schema.dc_tool_c_sessions');
console.log(`列数: ${columns.length}`);
console.log(`索引数: ${indexes.length}\n`);
} catch (error) {
console.error('\n❌ 创建表失败:', error.message);
console.error('\n详细错误:');
console.error(error);
process.exit(1);
} finally {
await prisma.$disconnect();
}
}
// 执行
createToolCTable()
.then(() => {
console.log('脚本执行完成');
process.exit(0);
})
.catch((error) => {
console.error('脚本执行失败:', error);
process.exit(1);
});