Files
AIclinicalresearch/backend/compare_dc_asl.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

96 lines
2.1 KiB
TypeScript

import { PrismaClient } from '@prisma/client';
const prisma = new PrismaClient();
async function getTableColumns(schema: string, tableName: string): Promise<any[]> {
return prisma.$queryRawUnsafe(`
SELECT column_name, data_type, is_nullable, column_default
FROM information_schema.columns
WHERE table_schema = '${schema}' AND table_name = '${tableName}'
ORDER BY ordinal_position
`);
}
async function main() {
console.log('🔍 DC 和 ASL 模块表结构对比\n');
console.log('=' .repeat(70));
// DC 模块的表
const dcTables = [
'dc_extraction_items',
'dc_extraction_tasks',
'dc_health_checks',
'dc_templates',
'dc_tool_c_ai_history',
'dc_tool_c_sessions'
];
// ASL 模块的表
const aslTables = [
'fulltext_screening_results',
'fulltext_screening_tasks',
'literatures',
'screening_projects',
'screening_results',
'screening_tasks'
];
console.log('\n📋 DC 模块 (dc_schema) 当前表结构:\n');
for (const table of dcTables) {
console.log(`\n--- dc_schema.${table} ---`);
try {
const cols = await getTableColumns('dc_schema', table);
if (cols.length === 0) {
console.log(' ❌ 表不存在');
} else {
cols.forEach((c: any) => {
console.log(` ${c.column_name}: ${c.data_type} ${c.is_nullable === 'NO' ? 'NOT NULL' : ''}`);
});
}
} catch (e) {
console.log(' ❌ 查询失败');
}
}
console.log('\n\n📋 ASL 模块 (asl_schema) 当前表结构:\n');
for (const table of aslTables) {
console.log(`\n--- asl_schema.${table} ---`);
try {
const cols = await getTableColumns('asl_schema', table);
if (cols.length === 0) {
console.log(' ❌ 表不存在');
} else {
cols.forEach((c: any) => {
console.log(` ${c.column_name}: ${c.data_type} ${c.is_nullable === 'NO' ? 'NOT NULL' : ''}`);
});
}
} catch (e) {
console.log(' ❌ 查询失败');
}
}
console.log('\n' + '=' .repeat(70));
}
main()
.catch(console.error)
.finally(() => prisma.$disconnect());