Core Components: - PDFStorageService with Dify/OSS adapters - LLM12FieldsService with Nougat-first + dual-model + 3-layer JSON parsing - PromptBuilder for dynamic prompt assembly - MedicalLogicValidator with 5 rules + fault tolerance - EvidenceChainValidator for citation integrity - ConflictDetectionService for dual-model comparison Prompt Engineering: - System Prompt (6601 chars, Section-Aware strategy) - User Prompt template (PICOS context injection) - JSON Schema (12 fields constraints) - Cochrane standards (not loaded in MVP) Key Innovations: - 3-layer JSON parsing (JSON.parse + json-repair + code block extraction) - Promise.allSettled for dual-model fault tolerance - safeGetFieldValue for robust field extraction - Mixed CN/EN token calculation Integration Tests: - integration-test.ts (full test) - quick-test.ts (quick test) - cached-result-test.ts (fault tolerance test) Documentation Updates: - Development record (Day 2-3 summary) - Quality assurance strategy (full-text screening) - Development plan (progress update) - Module status (v1.1 update) - Technical debt (10 new items) Test Results: - JSON parsing success rate: 100% - Medical logic validation: 5/5 passed - Dual-model parallel processing: OK - Cost per PDF: CNY 0.10 Files: 238 changed, 14383 insertions(+), 32 deletions(-) Docs: docs/03-涓氬姟妯″潡/ASL-AI鏅鸿兘鏂囩尞/05-寮€鍙戣褰?2025-11-22_Day2-Day3_LLM鏈嶅姟涓庨獙璇佺郴缁熷紑鍙?md
157 lines
4.6 KiB
PL/PgSQL
157 lines
4.6 KiB
PL/PgSQL
-- ========================================
|
||
-- 002-migrate-platform.sql
|
||
-- ========================================
|
||
-- 目的:迁移platform_schema(用户表)
|
||
-- 迁移表:1个(users)
|
||
-- 预计时间:15分钟
|
||
-- 作者:AI助手
|
||
-- 日期:2025-11-09
|
||
-- ========================================
|
||
|
||
-- 前置条件:
|
||
-- 1. 已执行 001-create-all-10-schemas.sql
|
||
-- 2. public.users 表存在且有数据
|
||
|
||
BEGIN;
|
||
|
||
-- ========================================
|
||
-- 第一步:创建platform_schema.users表
|
||
-- ========================================
|
||
|
||
CREATE TABLE IF NOT EXISTS platform_schema.users (
|
||
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
||
email VARCHAR(255) UNIQUE NOT NULL,
|
||
password VARCHAR(255) NOT NULL,
|
||
name VARCHAR(255),
|
||
avatar_url VARCHAR(500),
|
||
role VARCHAR(50) NOT NULL DEFAULT 'user',
|
||
status VARCHAR(50) DEFAULT 'active',
|
||
kb_quota INT DEFAULT 3,
|
||
kb_used INT DEFAULT 0,
|
||
trial_ends_at TIMESTAMP,
|
||
is_trial BOOLEAN DEFAULT true,
|
||
last_login_at TIMESTAMP,
|
||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||
);
|
||
|
||
-- ========================================
|
||
-- 第二步:创建索引
|
||
-- ========================================
|
||
|
||
CREATE INDEX IF NOT EXISTS idx_platform_users_email ON platform_schema.users(email);
|
||
CREATE INDEX IF NOT EXISTS idx_platform_users_role ON platform_schema.users(role);
|
||
CREATE INDEX IF NOT EXISTS idx_platform_users_status ON platform_schema.users(status);
|
||
CREATE INDEX IF NOT EXISTS idx_platform_users_created_at ON platform_schema.users(created_at);
|
||
|
||
-- ========================================
|
||
-- 第三步:迁移数据
|
||
-- ========================================
|
||
|
||
-- 从public.users迁移数据到platform_schema.users
|
||
INSERT INTO platform_schema.users (
|
||
id, email, password, name, avatar_url,
|
||
role, status, kb_quota, kb_used,
|
||
trial_ends_at, is_trial, last_login_at,
|
||
created_at, updated_at
|
||
)
|
||
SELECT
|
||
id, email, password, name, avatar_url,
|
||
role, status, kb_quota, kb_used,
|
||
trial_ends_at, is_trial, last_login_at,
|
||
created_at, updated_at
|
||
FROM public.users
|
||
ON CONFLICT (id) DO NOTHING; -- 如果已存在则跳过(支持重复执行)
|
||
|
||
-- ========================================
|
||
-- 第四步:数据验证
|
||
-- ========================================
|
||
|
||
DO $$
|
||
DECLARE
|
||
public_count INTEGER;
|
||
platform_count INTEGER;
|
||
BEGIN
|
||
-- 统计原表数据量
|
||
SELECT COUNT(*) INTO public_count FROM public.users;
|
||
|
||
-- 统计新表数据量
|
||
SELECT COUNT(*) INTO platform_count FROM platform_schema.users;
|
||
|
||
RAISE NOTICE '原表 public.users 数据量:%', public_count;
|
||
RAISE NOTICE '新表 platform_schema.users 数据量:%', platform_count;
|
||
|
||
-- 验证数据一致性
|
||
IF public_count = platform_count THEN
|
||
RAISE NOTICE '✅ 数据迁移成功:数据量完全一致';
|
||
ELSE
|
||
RAISE WARNING '⚠️ 警告:数据量不一致!预期 %,实际 %', public_count, platform_count;
|
||
END IF;
|
||
|
||
-- 验证email唯一性
|
||
IF (SELECT COUNT(DISTINCT email) FROM platform_schema.users) = platform_count THEN
|
||
RAISE NOTICE '✅ Email唯一性校验通过';
|
||
ELSE
|
||
RAISE WARNING '⚠️ 警告:Email存在重复';
|
||
END IF;
|
||
END $$;
|
||
|
||
-- ========================================
|
||
-- 第五步:对比验证(抽样检查)
|
||
-- ========================================
|
||
|
||
-- 对比前5条数据
|
||
SELECT 'public.users' AS source, id, email, name, role, created_at
|
||
FROM public.users
|
||
ORDER BY created_at DESC
|
||
LIMIT 5;
|
||
|
||
SELECT 'platform_schema.users' AS source, id, email, name, role, created_at
|
||
FROM platform_schema.users
|
||
ORDER BY created_at DESC
|
||
LIMIT 5;
|
||
|
||
COMMIT;
|
||
|
||
-- ========================================
|
||
-- 执行结果统计(可单独运行)
|
||
-- ========================================
|
||
|
||
SELECT
|
||
'platform_schema.users' AS table_name,
|
||
COUNT(*) AS total_count,
|
||
COUNT(DISTINCT email) AS unique_emails,
|
||
COUNT(CASE WHEN role = 'admin' THEN 1 END) AS admin_count,
|
||
COUNT(CASE WHEN role = 'user' THEN 1 END) AS user_count,
|
||
COUNT(CASE WHEN status = 'active' THEN 1 END) AS active_count,
|
||
MIN(created_at) AS first_user_date,
|
||
MAX(created_at) AS last_user_date
|
||
FROM platform_schema.users;
|
||
|
||
-- ========================================
|
||
-- 后续步骤说明
|
||
-- ========================================
|
||
-- 注意:public.users表暂时保留,不删除
|
||
-- 原因:
|
||
-- 1. 其他Schema的表(aia, pkb)会引用platform_schema.users
|
||
-- 2. 所有迁移完成并验证后,再决定是否删除public.users
|
||
-- 3. 删除前需确保所有外键已更新
|
||
-- ========================================
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|