# scheme/statistics.py
# 统计分析模块的请求/响应模型（含 2.7 通用高级筛选器）
from typing import Any, List, Literal, Union

from pydantic import BaseModel, Field


# ============================================================
# 2.7.1 通用高级筛选器：支持 AND/OR 逻辑组合与嵌套 sub_rules
# ============================================================
class FilterRule(BaseModel):
    """单条筛选规则"""
    field: str = Field(..., description="字段名，如 age/gender/salary/class_name")
    operator: Literal[">", "<", "=", "!=", ">=", "<=", "like", "in"] = Field(
        ..., description="比较操作符"
    )
    value: Any = Field(..., description="比较值；operator=in 时必须是列表")


class FilterGroup(BaseModel):
    """逻辑组合规则：可嵌套 sub_rules 实现任意深度的 AND/OR 组合"""
    logic: Literal["AND", "OR"] = Field("AND", description="组合方式")
    sub_rules: List[Union["FilterGroup", "FilterRule"]] = Field(
        ..., min_length=1, description="子规则（可继续嵌套 FilterGroup）"
    )


# Pydantic v2 递归模型需要显式 rebuild
FilterGroup.model_rebuild()


class FilterRequest(BaseModel):
    """高级筛选请求体"""
    model: Literal["student"] = Field(..., description="目标模型（当前支持 student）")
    rules: List[Union[FilterGroup, FilterRule]] = Field(
        ..., min_length=1, description="筛选规则列表（顶层默认 AND 连接）"
    )


class FilterResponse(BaseModel):
    total: int
    items: List[dict] = Field(..., description="命中的学生记录")


# ============================================================
# 2.6 统计分析响应模型
# ============================================================
class ClassGenderStat(BaseModel):
    """2.6.1 多维度班级统计：总人数 + 性别分布"""
    class_id: int
    class_name: str
    total: int
    male: int
    female: int


class AllAboveStudent(BaseModel):
    """2.6.2 每次考试都在分数线以上的学生"""
    stu_id: int
    stu_name: str
    class_name: str
    exam_count: int = Field(..., description="参加考核次数")
    min_score: float = Field(..., description="最低分（>= 分数线）")
    scores: List[dict] = Field(..., description="各次成绩明细 [{exam_id, score}]")


class FailStudent(BaseModel):
    """2.6.2 不及格次数 >= N 的学生"""
    stu_id: int
    stu_name: str
    class_name: str
    fail_count: int = Field(..., description="不及格次数")
    fail_details: List[dict] = Field(..., description="不及格成绩明细 [{exam_id, score}]")


class ClassExamAvg(BaseModel):
    """2.6.2 每次考试每个班级的平均分（支持动态排序）"""
    exam_id: int
    class_id: int
    class_name: str
    avg_score: float


class TopSalaryStudent(BaseModel):
    """2.6.3 薪资 Top N"""
    stu_id: int
    stu_name: str
    class_name: str
    job_time: Any = Field(None, description="offer 下发时间")
    company_name: str
    salary: float


class EmploymentDuration(BaseModel):
    """2.6.3 单个学生就业时长（天）"""
    stu_id: int
    stu_name: str
    class_name: str
    employment_open_time: Any
    job_time: Any = Field(None, description="未拿到 offer 为 null")
    duration_days: int = Field(..., description="offer下发时间 - 就业开放时间；未拿到 offer 为 -1")


class ClassAvgDuration(BaseModel):
    """2.6.3 每个班级平均就业时长"""
    class_id: int
    class_name: str
    opened_count: int = Field(..., description="进入就业阶段人数（有开放时间）")
    offered_count: int = Field(..., description="已拿到 offer 人数")
    avg_duration_days: float = Field(..., description="平均就业时长（仅统计已拿到 offer 的学生）")


class ScoreVolatility(BaseModel):
    """2.7.2 成绩波动分析：最大分差 Top N"""
    stu_id: int
    stu_name: str
    class_name: str
    max_score: float
    min_score: float
    diff: float = Field(..., description="最大分差 = 最高分 - 最低分")


class EmploymentFunnel(BaseModel):
    """2.7.2 班级就业漏斗：总人数 -> 已就业 -> 高薪(>10k) -> 就业率"""
    class_id: int
    class_name: str
    total: int = Field(..., description="班级总人数")
    employed: int = Field(..., description="已就业人数")
    high_salary: int = Field(..., description="高薪人数（薪资 > 10000）")
    employment_rate: float = Field(..., description="就业率（百分比，保留2位）")