Files
AIclinicalresearch/backend/check_iit_asl_data.ts
HaHafeng 40c2f8e148 feat(rag): Complete RAG engine implementation with pgvector
Major Features:
- Created ekb_schema (13th schema) with 3 tables: KB/Document/Chunk
- Implemented EmbeddingService (text-embedding-v4, 1024-dim vectors)
- Implemented ChunkService (smart Markdown chunking)
- Implemented VectorSearchService (multi-query + hybrid search)
- Implemented RerankService (qwen3-rerank)
- Integrated DeepSeek V3 QueryRewriter for cross-language search
- Python service: Added pymupdf4llm for PDF-to-Markdown conversion
- PKB: Dual-mode adapter (pgvector/dify/hybrid)

Architecture:
- Brain-Hand Model: Business layer (DeepSeek) + Engine layer (pgvector)
- Cross-language support: Chinese query matches English documents
- Small Embedding (1024) + Strong Reranker strategy

Performance:
- End-to-end latency: 2.5s
- Cost per query: 0.0025 RMB
- Accuracy improvement: +20.5% (cross-language)

Tests:
- test-embedding-service.ts: Vector embedding verified
- test-rag-e2e.ts: Full pipeline tested
- test-rerank.ts: Rerank quality validated
- test-query-rewrite.ts: Cross-language search verified
- test-pdf-ingest.ts: Real PDF document tested (Dongen 2003.pdf)

Documentation:
- Added 05-RAG-Engine-User-Guide.md
- Added 02-Document-Processing-User-Guide.md
- Updated system status documentation

Status: Production ready
2026-01-21 20:24:29 +08:00

86 lines
3.1 KiB
TypeScript

import { PrismaClient } from '@prisma/client';
const prisma = new PrismaClient();
async function main() {
console.log('🔍 检查 IIT 和 ASL 模块的数据\n');
console.log('=' .repeat(60));
// IIT 模块
console.log('\n📋 IIT 模块 (iit_schema):\n');
const iitTables = [
{ name: 'projects', query: 'SELECT COUNT(*) as count FROM iit_schema.projects' },
{ name: 'audit_logs', query: 'SELECT COUNT(*) as count FROM iit_schema.audit_logs' },
{ name: 'pending_actions', query: 'SELECT COUNT(*) as count FROM iit_schema.pending_actions' },
{ name: 'task_runs', query: 'SELECT COUNT(*) as count FROM iit_schema.task_runs' },
{ name: 'user_mappings', query: 'SELECT COUNT(*) as count FROM iit_schema.user_mappings' },
];
for (const t of iitTables) {
const result: any = await prisma.$queryRawUnsafe(t.query);
const count = Number(result[0].count);
console.log(` ${t.name}: ${count} 条记录 ${count > 0 ? '✅' : '(空)'}`);
}
// 如果有数据,显示一些详情
const iitProjects: any[] = await prisma.$queryRaw`SELECT id, name, status, created_at FROM iit_schema.projects LIMIT 5`;
if (iitProjects.length > 0) {
console.log('\n 最近的 IIT 项目:');
iitProjects.forEach(p => console.log(` - ${p.name} (${p.status}) @ ${p.created_at}`));
}
// ASL 模块(智能文献筛选)
console.log('\n\n📋 ASL 模块 - 智能文献筛选 (asl_schema):\n');
const aslTables = [
{ name: 'screening_projects', query: 'SELECT COUNT(*) as count FROM asl_schema.screening_projects' },
{ name: 'literatures', query: 'SELECT COUNT(*) as count FROM asl_schema.literatures' },
{ name: 'screening_tasks', query: 'SELECT COUNT(*) as count FROM asl_schema.screening_tasks' },
{ name: 'screening_results', query: 'SELECT COUNT(*) as count FROM asl_schema.screening_results' },
{ name: 'fulltext_screening_tasks', query: 'SELECT COUNT(*) as count FROM asl_schema.fulltext_screening_tasks' },
{ name: 'fulltext_screening_results', query: 'SELECT COUNT(*) as count FROM asl_schema.fulltext_screening_results' },
];
for (const t of aslTables) {
const result: any = await prisma.$queryRawUnsafe(t.query);
const count = Number(result[0].count);
console.log(` ${t.name}: ${count} 条记录 ${count > 0 ? '✅' : '(空)'}`);
}
// 如果有数据,显示一些详情
const aslProjects: any[] = await prisma.$queryRaw`SELECT id, project_name, status, created_at FROM asl_schema.screening_projects LIMIT 5`;
if (aslProjects.length > 0) {
console.log('\n 最近的 ASL 项目:');
aslProjects.forEach(p => console.log(` - ${p.project_name} (${p.status}) @ ${p.created_at}`));
}
const literatures: any[] = await prisma.$queryRaw`SELECT id, title, stage FROM asl_schema.literatures LIMIT 5`;
if (literatures.length > 0) {
console.log('\n 最近的文献:');
literatures.forEach(l => console.log(` - ${l.title?.substring(0, 50)}... (${l.stage})`));
}
console.log('\n' + '=' .repeat(60));
}
main()
.catch(console.error)
.finally(() => prisma.$disconnect());