feat(data-ana): 完整实现 data-ana 数据分析服务

包含 CDC consumer、analytics/mastery/warning service、grpc server、repository、ClickHouse DDL 等

services/data-ana/src/data_ana/config.py

@@ -1,46 +1,97 @@
-"""配置管理."""
+"""配置管理（pydantic-settings 完整配置项）.
 
-from pydantic_settings import BaseSettings
+对齐 02-architecture-design.md §15 配置清单.
+环境变量前缀 DATA_ANA_（如 DATA_ANA_HTTP_PORT=3006）.
+"""
+
+from pydantic_settings import BaseSettings, SettingsConfigDict
 
 
 class Settings(BaseSettings):
     """应用配置.
 
-    ClickHouse 连接参数为可选：当 clickhouse_host 为空字符串时，
-    服务进入降级模式（查询方法返回 None / 空数据），保证服务可启动。
-
-    Kafka 连接参数为可选：当 kafka_brokers 为空字符串时，
-    CDC 消费者不启动（降级模式），保证服务可启动。
+    所有外部依赖（ClickHouse / Kafka / Redis / iam gRPC）均为可选：
+    未配置或不可达时服务进入降级模式，返回骨架数据 + details.degraded: true.
     """
 
-    port: int = 3006
-    # ClickHouse 连接（可选：留空则降级模式）
+    # 服务
+    service_name: str = "data-ana"
+    http_port: int = 3006
+    grpc_port: int = 50055
+    log_level: str = "INFO"
+    dev_mode: bool = False
+
+    # ClickHouse（可选：留空则降级模式）
     clickhouse_host: str = ""
     clickhouse_port: int = 8123
     clickhouse_database: str = "edu_analytics"
     clickhouse_user: str = ""
     clickhouse_password: str = ""
-    # 可观测性
-    otel_endpoint: str = "http://localhost:4318"
-    log_level: str = "info"
-    # 开发模式开关
-    dev_mode: bool = False
-    # Kafka brokers（CDC 消费；留空则不启动消费者）
-    # 主机访问用 localhost:9092，容器内访问用 kafka:29092
-    kafka_brokers: str = ""
-    # CDC 消费组 id
-    kafka_group_id: str = "data-ana-cdc-consumer"
-    # 要消费的 CDC topic（Debezium 默认命名：<prefix>.<database>.<table>）
-    # 用逗号分隔多个 topic
+    clickhouse_connect_timeout_ms: int = 3000
+    clickhouse_query_timeout_s: int = 3  # P4 退出标准 5s，查询 3s 超时降级
+
+    # Kafka
+    kafka_brokers: str = ""  # 留空则不启动 CDC 消费者
+    kafka_consumer_group: str = "data-ana-cdc"
     kafka_cdc_topics: str = (
         "edu-cdc.next_edu_cloud.core_edu_grades,"
         "edu-cdc.next_edu_cloud.core_edu_exams,"
-        "edu-cdc.next_edu_cloud.classes"
+        "edu-cdc.next_edu_cloud.core_edu_homework_submissions,"
+        "edu-cdc.next_edu_cloud.core_edu_attendance,"
+        "edu-cdc.next_edu_cloud.classes,"
+        "edu-cdc.next_edu_cloud.iam_users,"
+        "edu-cdc.next_edu_cloud.content_knowledge_points"
     )
-    # 消费者自动偏移重置策略（earliest / latest）
-    kafka_auto_offset_reset: str = "earliest"
+    kafka_mastery_topic: str = "edu.insight.mastery.updated"
+    kafka_warning_topic: str = "edu.insight.mastery.updated"  # 复用 mastery topic（总裁裁决 §2.11）
+    kafka_ai_usage_topic: str = "edu.insight.ai.usage"
+    kafka_enable_auto_commit: bool = False  # v2: 手动 commit（at-least-once）
+    kafka_auto_offset_reset: str = "latest"
+    kafka_producer_transactional_id: str = "data-ana-producer"
 
-    model_config = {"env_file": ".env", "env_prefix": ""}
+    # iam gRPC
+    iam_grpc_endpoint: str = ""  # 留空则使用降级兜底（按 role 映射 DataScope）
+    iam_grpc_timeout_s: int = 2
+    datascope_cache_ttl_s: int = 300  # 5min
+
+    # Redis
+    redis_url: str = ""  # 留空则跳过缓存
+    redis_pool_size: int = 10
+    redis_socket_timeout_ms: int = 200
+
+    # OTel
+    otel_endpoint: str = "http://localhost:4318"
+    otel_service_name: str = "data-ana"
+
+    # 掌握度算法
+    mastery_window_size: int = 5
+    mastery_decay_base: float = 0.6
+    mastery_forgetting_half_life_days: int = 30
+    mastery_min_samples: int = 3
+
+    # 预警阈值
+    warning_low_mastery_threshold: float = 0.4
+    warning_critical_mastery_threshold: float = 0.2
+    warning_score_drop_percent: float = 0.2
+    warning_absent_per_week: int = 3
+
+    # 降级
+    degraded_mode_enabled: bool = True
+
+    # 向后兼容：旧代码引用 settings.port / settings.kafka_group_id
+    @property
+    def port(self) -> int:
+        return self.http_port
+
+    @property
+    def kafka_group_id(self) -> str:
+        return self.kafka_consumer_group
+
+    model_config = SettingsConfigDict(
+        env_file=".env",
+        env_prefix="DATA_ANA_",
+        extra="ignore",
+    )
 
 
 settings = Settings()