// =============================================================================
// DataMate 知识图谱 - Neo4j Schema 初始化脚本
// Schema 版本：1.0.0
// 更新日期：2026-02-17
//
// 使用方式：
//   1. 通过 Cypher Shell 执行：
//      cat schema.cypher | cypher-shell -u neo4j -p <password>
//   2. 或在 Neo4j Browser 中逐段执行
//
// 注意：
//   - 所有索引和约束使用 IF NOT EXISTS，可重复执行
//   - 约束自动创建对应索引，无需重复创建
//   - 关系属性索引需要 Neo4j Enterprise Edition，社区版使用属性内联匹配
// =============================================================================

// =============================================================================
// 第 1 部分：节点约束
// =============================================================================

// Entity 节点 ID 唯一性约束（自动创建索引）
CREATE CONSTRAINT entity_id_unique IF NOT EXISTS
FOR (n:Entity) REQUIRE n.id IS UNIQUE;

// =============================================================================
// 第 2 部分：节点索引
// =============================================================================

// graph_id 索引 —— 多租户隔离的核心索引，所有查询都会带上 graph_id
CREATE INDEX entity_graph_id IF NOT EXISTS
FOR (n:Entity) ON (n.graph_id);

// type 索引 —— 按实体类型过滤
CREATE INDEX entity_type IF NOT EXISTS
FOR (n:Entity) ON (n.type);

// name 索引 —— 按名称搜索
CREATE INDEX entity_name IF NOT EXISTS
FOR (n:Entity) ON (n.name);

// source_id 索引 —— MySQL → Neo4j 同步时按源 ID 查找
CREATE INDEX entity_source_id IF NOT EXISTS
FOR (n:Entity) ON (n.source_id);

// 复合索引：(graph_id, type) —— 查询某图谱内指定类型的实体
CREATE INDEX entity_graph_id_type IF NOT EXISTS
FOR (n:Entity) ON (n.graph_id, n.type);

// 复合索引：(graph_id, id) —— 精确查找实体（最常用查询路径）
CREATE INDEX entity_graph_id_id IF NOT EXISTS
FOR (n:Entity) ON (n.graph_id, n.id);

// 复合索引：(graph_id, source_id) —— 同步时按源 ID 查找
CREATE INDEX entity_graph_id_source_id IF NOT EXISTS
FOR (n:Entity) ON (n.graph_id, n.source_id);

// created_at 索引 —— 按创建时间排序
CREATE INDEX entity_created_at IF NOT EXISTS
FOR (n:Entity) ON (n.created_at);

// =============================================================================
// 第 3 部分：全文索引（用于模糊搜索）
// =============================================================================

// Entity name + description 全文索引
CREATE FULLTEXT INDEX entity_fulltext IF NOT EXISTS
FOR (n:Entity) ON EACH [n.name, n.description];

// =============================================================================
// 第 3.1 部分：SyncHistory 约束和索引（同步元数据节点）
// =============================================================================

// (graph_id, sync_id) 唯一约束 —— 防止 syncId 碰撞产生重复记录
CREATE CONSTRAINT sync_history_graph_sync_unique IF NOT EXISTS
FOR (h:SyncHistory) REQUIRE (h.graph_id, h.sync_id) IS UNIQUE;

// (graph_id, started_at) 索引 —— 加速按时间范围查询同步历史
CREATE INDEX sync_history_graph_started IF NOT EXISTS
FOR (h:SyncHistory) ON (h.graph_id, h.started_at);

// (graph_id, status, started_at) 索引 —— 加速按状态+时间的过滤查询
CREATE INDEX sync_history_graph_status_started IF NOT EXISTS
FOR (h:SyncHistory) ON (h.graph_id, h.status, h.started_at);

// =============================================================================
// 第 4 部分：关系属性说明
// =============================================================================

// Neo4j 社区版不支持关系属性索引。
// 所有关系查询通过节点索引定位后，在关系上使用属性内联匹配：
//   -[r:RELATED_TO {graph_id: $graphId, relation_type: $type}]->
//
// 如果使用 Neo4j Enterprise Edition，可取消以下注释创建关系索引：
//
// CREATE INDEX rel_graph_id IF NOT EXISTS
// FOR ()-[r:RELATED_TO]-() ON (r.graph_id);
//
// CREATE INDEX rel_relation_type IF NOT EXISTS
// FOR ()-[r:RELATED_TO]-() ON (r.relation_type);
//
// CREATE INDEX rel_id IF NOT EXISTS
// FOR ()-[r:RELATED_TO]-() ON (r.id);

// =============================================================================
// 第 5 部分：示例数据（可选，用于验证 Schema）
// =============================================================================

// 以下示例数据使用固定的 graph_id，用于开发和测试环境。
// 生产环境中 graph_id 由应用层生成和管理。

// --- 创建示例组织 ---
CREATE (org:Entity {
  id: '00000000-0000-0000-0000-000000000001',
  name: '数据工程部',
  type: 'Org',
  description: '负责数据采集、清洗和标注',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'MANUAL',
  confidence: 1.0,
  properties_json: '{"org_code":"DE","level":1,"member_count":15}',
  created_at: datetime()
});

// --- 创建示例用户 ---
CREATE (user:Entity {
  id: '00000000-0000-0000-0000-000000000002',
  name: '张三',
  type: 'User',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"username":"zhangsan","email":"zhangsan@example.com","role":"USER","enabled":true}',
  created_at: datetime()
});

// --- 创建示例数据集（源） ---
CREATE (ds1:Entity {
  id: '00000000-0000-0000-0000-000000000010',
  name: '用户行为日志-原始',
  type: 'Dataset',
  description: '原始用户行为埋点数据',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_id: '100',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"dataset_type":"TEXT","status":"ACTIVE","category":"用户行为","format":"JSON","record_count":2000000,"size_bytes":3221225472}',
  created_at: datetime()
});

// --- 创建示例数据集（清洗后） ---
CREATE (ds2:Entity {
  id: '00000000-0000-0000-0000-000000000011',
  name: '用户行为日志-清洗后',
  type: 'Dataset',
  description: '经过去重和格式标准化的用户行为数据',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_id: '101',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"dataset_type":"TEXT","status":"ACTIVE","category":"用户行为","format":"JSON","record_count":1500000,"size_bytes":2147483648,"version":1}',
  created_at: datetime()
});

// --- 创建示例字段 ---
CREATE (f1:Entity {
  id: '00000000-0000-0000-0000-000000000020',
  name: 'user_id',
  type: 'Field',
  description: '用户唯一标识符',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"data_type":"STRING","nullable":false,"is_primary_key":true}',
  created_at: datetime()
});

CREATE (f2:Entity {
  id: '00000000-0000-0000-0000-000000000021',
  name: 'event_type',
  type: 'Field',
  description: '事件类型',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"data_type":"STRING","nullable":false,"sample_values":["click","view","purchase"]}',
  created_at: datetime()
});

CREATE (f3:Entity {
  id: '00000000-0000-0000-0000-000000000022',
  name: 'user_id',
  type: 'Field',
  description: '用户唯一标识符（清洗后）',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"data_type":"STRING","nullable":false,"is_primary_key":true}',
  created_at: datetime()
});

// --- 创建示例工作流 ---
CREATE (wf:Entity {
  id: '00000000-0000-0000-0000-000000000030',
  name: '文本去重清洗管道',
  type: 'Workflow',
  description: 'SimHash去重 + 格式标准化 + 空值过滤',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"workflow_type":"CLEANING","status":"ACTIVE","version":"1.0","operator_count":3}',
  created_at: datetime()
});

// --- 创建示例作业 ---
CREATE (job:Entity {
  id: '00000000-0000-0000-0000-000000000040',
  name: '清洗作业-20260215-001',
  type: 'Job',
  description: '用户行为日志去重清洗',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_id: '500',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"job_type":"CLEANING","status":"COMPLETED","started_at":"2026-02-15T10:00:00","completed_at":"2026-02-15T10:35:00","duration_seconds":2100,"input_count":2000000,"output_count":1500000}',
  created_at: datetime()
});

// --- 创建示例标注任务 ---
CREATE (lt:Entity {
  id: '00000000-0000-0000-0000-000000000050',
  name: '情感分析标注-批次1',
  type: 'LabelTask',
  description: '用户评论情感标注（正面/负面/中性）',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_id: '600',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"task_mode":"MANUAL","data_type":"text","labeling_type":"sentiment_analysis","status":"IN_PROGRESS","progress":30.0,"template_name":"情感分析"}',
  created_at: datetime()
});

// --- 创建示例知识集 ---
CREATE (ks:Entity {
  id: '00000000-0000-0000-0000-000000000060',
  name: '用户行为分析知识库',
  type: 'KnowledgeSet',
  description: '从用户行为数据中提取的业务规则和洞察',
  graph_id: '11111111-1111-1111-1111-111111111111',
  source_type: 'SYNC',
  confidence: 1.0,
  properties_json: '{"status":"PUBLISHED","domain":"用户行为","business_line":"数据分析","sensitivity":"INTERNAL","item_count":85}',
  created_at: datetime()
});

// =============================================================================
// 第 6 部分：示例关系
// =============================================================================

// HAS_FIELD：源数据集 → 字段
MATCH (ds1:Entity {id: '00000000-0000-0000-0000-000000000010'})
MATCH (f1:Entity {id: '00000000-0000-0000-0000-000000000020'})
CREATE (ds1)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000001',
  relation_type: 'HAS_FIELD',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"ordinal":0,"required":true}',
  created_at: datetime()
}]->(f1);

MATCH (ds1:Entity {id: '00000000-0000-0000-0000-000000000010'})
MATCH (f2:Entity {id: '00000000-0000-0000-0000-000000000021'})
CREATE (ds1)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000002',
  relation_type: 'HAS_FIELD',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"ordinal":1,"required":true}',
  created_at: datetime()
}]->(f2);

// HAS_FIELD：清洗后数据集 → 字段
MATCH (ds2:Entity {id: '00000000-0000-0000-0000-000000000011'})
MATCH (f3:Entity {id: '00000000-0000-0000-0000-000000000022'})
CREATE (ds2)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000003',
  relation_type: 'HAS_FIELD',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"ordinal":0,"required":true}',
  created_at: datetime()
}]->(f3);

// DERIVED_FROM：清洗后数据集 → 源数据集
MATCH (ds2:Entity {id: '00000000-0000-0000-0000-000000000011'})
MATCH (ds1:Entity {id: '00000000-0000-0000-0000-000000000010'})
CREATE (ds2)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000004',
  relation_type: 'DERIVED_FROM',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"derivation_type":"CLEANING","job_id":"00000000-0000-0000-0000-000000000040","transformation":"SimHash去重 + 空值过滤"}',
  created_at: datetime()
}]->(ds1);

// TRIGGERS：工作流 → 作业
MATCH (wf:Entity {id: '00000000-0000-0000-0000-000000000030'})
MATCH (job:Entity {id: '00000000-0000-0000-0000-000000000040'})
CREATE (wf)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000005',
  relation_type: 'TRIGGERS',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"trigger_type":"MANUAL","triggered_at":"2026-02-15T10:00:00"}',
  created_at: datetime()
}]->(job);

// USES_DATASET：作业 → 源数据集
MATCH (job:Entity {id: '00000000-0000-0000-0000-000000000040'})
MATCH (ds1:Entity {id: '00000000-0000-0000-0000-000000000010'})
CREATE (job)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000006',
  relation_type: 'USES_DATASET',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"usage_role":"INPUT"}',
  created_at: datetime()
}]->(ds1);

// PRODUCES：作业 → 清洗后数据集
MATCH (job:Entity {id: '00000000-0000-0000-0000-000000000040'})
MATCH (ds2:Entity {id: '00000000-0000-0000-0000-000000000011'})
CREATE (job)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000007',
  relation_type: 'PRODUCES',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"output_type":"PRIMARY"}',
  created_at: datetime()
}]->(ds2);

// ASSIGNED_TO：标注任务 → 用户
MATCH (lt:Entity {id: '00000000-0000-0000-0000-000000000050'})
MATCH (user:Entity {id: '00000000-0000-0000-0000-000000000002'})
CREATE (lt)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000008',
  relation_type: 'ASSIGNED_TO',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"assigned_at":"2026-02-14T09:00:00","role":"EXECUTOR"}',
  created_at: datetime()
}]->(user);

// USES_DATASET：标注任务 → 清洗后数据集
MATCH (lt:Entity {id: '00000000-0000-0000-0000-000000000050'})
MATCH (ds2:Entity {id: '00000000-0000-0000-0000-000000000011'})
CREATE (lt)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000009',
  relation_type: 'USES_DATASET',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"usage_role":"INPUT"}',
  created_at: datetime()
}]->(ds2);

// BELONGS_TO：用户 → 组织
MATCH (user:Entity {id: '00000000-0000-0000-0000-000000000002'})
MATCH (org:Entity {id: '00000000-0000-0000-0000-000000000001'})
CREATE (user)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000010',
  relation_type: 'BELONGS_TO',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"membership_type":"PRIMARY","since":"2025-03-01T00:00:00"}',
  created_at: datetime()
}]->(org);

// BELONGS_TO：源数据集 → 组织
MATCH (ds1:Entity {id: '00000000-0000-0000-0000-000000000010'})
MATCH (org:Entity {id: '00000000-0000-0000-0000-000000000001'})
CREATE (ds1)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000011',
  relation_type: 'BELONGS_TO',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 1.0,
  source_id: '',
  properties_json: '{"membership_type":"PRIMARY"}',
  created_at: datetime()
}]->(org);

// IMPACTS：源字段 → 清洗后字段
MATCH (f1:Entity {id: '00000000-0000-0000-0000-000000000020'})
MATCH (f3:Entity {id: '00000000-0000-0000-0000-000000000022'})
CREATE (f1)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000012',
  relation_type: 'IMPACTS',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 0.95,
  source_id: '',
  properties_json: '{"impact_type":"DIRECT","job_id":"00000000-0000-0000-0000-000000000040"}',
  created_at: datetime()
}]->(f3);

// SOURCED_FROM：知识集 → 清洗后数据集
MATCH (ks:Entity {id: '00000000-0000-0000-0000-000000000060'})
MATCH (ds2:Entity {id: '00000000-0000-0000-0000-000000000011'})
CREATE (ks)-[:RELATED_TO {
  id: '00000000-0000-0000-0000-100000000013',
  relation_type: 'SOURCED_FROM',
  graph_id: '11111111-1111-1111-1111-111111111111',
  weight: 1.0,
  confidence: 0.85,
  source_id: '',
  properties_json: '{"extraction_method":"LLM","extracted_at":"2026-02-16T14:30:00","item_count":85}',
  created_at: datetime()
}]->(ds2);

// =============================================================================
// 第 7 部分：验证查询
// =============================================================================

// 验证节点数量
// MATCH (n:Entity {graph_id: '11111111-1111-1111-1111-111111111111'})
// RETURN n.type AS type, count(*) AS count
// ORDER BY count DESC;

// 验证关系数量
// MATCH (:Entity {graph_id: '11111111-1111-1111-1111-111111111111'})
//       -[r:RELATED_TO {graph_id: '11111111-1111-1111-1111-111111111111'}]->
//       (:Entity {graph_id: '11111111-1111-1111-1111-111111111111'})
// RETURN r.relation_type AS type, count(*) AS count
// ORDER BY count DESC;

// 验证端到端血缘
// MATCH path = (ds2:Entity {name: '用户行为日志-清洗后'})
//              -[:RELATED_TO *1..5]->
//              (origin:Entity)
// WHERE ALL(r IN relationships(path) WHERE r.graph_id = '11111111-1111-1111-1111-111111111111')
// RETURN path;

// =============================================================================
// 第 8 部分：清理示例数据（可选）
// =============================================================================

// 如需清理示例数据，执行以下语句：
// MATCH (n:Entity {graph_id: '11111111-1111-1111-1111-111111111111'})
// DETACH DELETE n;