Files
AIRegulation-Deployment/scripts/05_init_db.sh

125 lines
5.3 KiB
Bash
Raw Normal View History

2026-04-23 09:58:47 +08:00
#!/usr/bin/env bash
# ══════════════════════════════════════════════════
# 05_init_db.sh
# 初始化数据库PostgreSQL Schema + Milvus Collections + Neo4j Constraints
# 用法bash scripts/05_init_db.sh
# 前提postgres / milvus / neo4j 已运行且健康
# ══════════════════════════════════════════════════
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
PROJECT_DIR="$(dirname "$SCRIPT_DIR")"
cd "$PROJECT_DIR"
RED='\033[0;31m'; GREEN='\033[0;32m'; YELLOW='\033[1;33m'; BLUE='\033[0;34m'; NC='\033[0m'
info() { echo -e "${BLUE}[INFO]${NC} $*"; }
ok() { echo -e "${GREEN}[OK]${NC} $*"; }
warn() { echo -e "${YELLOW}[WARN]${NC} $*"; }
error() { echo -e "${RED}[ERROR]${NC} $*"; exit 1; }
source .env 2>/dev/null || true
POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-compliance123}
NEO4J_PASSWORD=${NEO4J_PASSWORD:-neo4j123}
# ── Step 1PostgreSQL Schema ───────────────────
info "Step 1/3初始化 PostgreSQL Schema..."
if docker compose ps postgres | grep -q "healthy"; then
docker compose exec -T postgres psql \
-U compliance -d compliance_db \
-f /docker-entrypoint-initdb.d/01_init_schema.sql \
2>&1 | tail -5 || warn "SQL 可能部分已存在IF NOT EXISTS这是正常的"
ok "PostgreSQL Schema 初始化完成"
else
error "PostgreSQL 未运行请先执行bash scripts/03_start_infra.sh"
fi
# ── Step 2Milvus Collections ──────────────────
info "Step 2/3初始化 Milvus Collections..."
if docker compose ps milvus | grep -q "healthy"; then
docker compose run --rm --no-deps compliance-backend \
python3 -c "
import asyncio
from pymilvus import connections, Collection, CollectionSchema, FieldSchema, DataType, utility
connections.connect(host='milvus', port='19530')
print('Milvus 连接成功')
def create_collection(name, description):
if utility.has_collection(name):
print(f' Collection {name} 已存在,跳过')
return
fields = [
FieldSchema(name='id', dtype=DataType.VARCHAR, is_primary=True, max_length=128),
FieldSchema(name='file_id', dtype=DataType.VARCHAR, max_length=128),
FieldSchema(name='workspace_id', dtype=DataType.VARCHAR, max_length=128),
FieldSchema(name='chunk_idx', dtype=DataType.INT64),
FieldSchema(name='content', dtype=DataType.VARCHAR, max_length=65535),
FieldSchema(name='dense_vec', dtype=DataType.FLOAT_VECTOR, dim=1024), # BGE-M3 dense
FieldSchema(name='metadata', dtype=DataType.JSON),
]
schema = CollectionSchema(fields, description=description)
col = Collection(name, schema)
# 创建向量索引HNSW适合调研阶段
index_params = {
'metric_type': 'COSINE',
'index_type': 'HNSW',
'params': {'M': 16, 'efConstruction': 200}
}
col.create_index('dense_vec', index_params)
col.load()
print(f' Collection {name} 创建完成')
create_collection('regulation_chunks', '法规条款向量库')
create_collection('doc_chunks', '企业文档向量库')
create_collection('case_library', '行业案例库')
print('Milvus 初始化完成')
" 2>&1
ok "Milvus Collections 初始化完成"
else
error "Milvus 未运行请先执行bash scripts/03_start_infra.sh"
fi
# ── Step 3Neo4j 约束和索引 ────────────────────
info "Step 3/3初始化 Neo4j 约束和索引..."
sleep 5 # Neo4j 可能还在预热
docker compose exec -T neo4j cypher-shell \
-u neo4j -p "$NEO4J_PASSWORD" \
--format plain <<'CYPHER'
// 节点约束(唯一性)
CREATE CONSTRAINT regulation_id IF NOT EXISTS
FOR (r:Regulation) REQUIRE r.id IS UNIQUE;
CREATE CONSTRAINT clause_id IF NOT EXISTS
FOR (c:Clause) REQUIRE c.id IS UNIQUE;
CREATE CONSTRAINT obligation_id IF NOT EXISTS
FOR (o:Obligation) REQUIRE o.id IS UNIQUE;
// 全文索引(模糊查询)
CREATE FULLTEXT INDEX regulation_fulltext IF NOT EXISTS
FOR (r:Regulation) ON EACH [r.title, r.code, r.domain];
CREATE FULLTEXT INDEX clause_fulltext IF NOT EXISTS
FOR (c:Clause) ON EACH [c.content, c.title];
// 插入示例节点(验证连通性)
MERGE (d:Domain {name: 'vehicle_safety', label: '车辆安全法规'});
MERGE (d:Domain {name: 'data_security', label: '数据安全法规'});
MERGE (d:Domain {name: 'ehs', label: 'EHS安全法规'});
MERGE (d:Domain {name: 'carbon', label: '碳排放法规'});
RETURN '初始化完成' AS result;
CYPHER
ok "Neo4j 约束和索引初始化完成"
echo ""
echo -e "${GREEN}══════════════════════════════════════════${NC}"
echo -e "${GREEN} 数据库初始化完成!${NC}"
echo -e "${GREEN}══════════════════════════════════════════${NC}"
echo ""
echo " PostgreSQL: 所有表已创建"
echo " Milvus: regulation_chunks / doc_chunks / case_library"
echo " Neo4j: 约束 + 全文索引 + 基础域节点"
echo ""
echo "下一步bash scripts/06_start_all.sh"