"""
Pydantic schemas for Overlay API responses
"""

from datetime import datetime
from typing import Any, Dict, List, Optional
from pydantic import BaseModel, Field


class OverlayFeatures(BaseModel):
    headline_burst_z: Optional[float] = None
    youtube_influence_z: Optional[float] = None
    wiki_attention_z: Optional[float] = None
    theme_heat_z: Optional[float] = None
    crowding_stress_z: Optional[float] = None


class OverlaySourcePresence(BaseModel):
    yahoo: bool = False
    youtube: bool = False
    wikimedia: bool = False
    google_trends: bool = False
    finra: bool = False


class YahooSourceDetail(BaseModel):
    headline_count_6h: int = 0
    headline_count_24h: int = 0
    publisher_breadth_24h: int = 0


class YouTubeSourceDetail(BaseModel):
    mentions_24h: int = 0
    weighted_views_24h: float = 0.0


class WikiSourceDetail(BaseModel):
    page_views_1d: Optional[int] = None
    page_views_7d_avg: Optional[float] = None


class FinraSourceDetail(BaseModel):
    short_volume_ratio: Optional[float] = None
    short_volume_spike_zscore: Optional[float] = None


class OverlaySourceDetails(BaseModel):
    yahoo: Optional[YahooSourceDetail] = None
    youtube: Optional[YouTubeSourceDetail] = None
    wikimedia: Optional[WikiSourceDetail] = None
    finra: Optional[FinraSourceDetail] = None


class OverlayMetadata(BaseModel):
    feature_version: str = "v1"
    data_freshness: Optional[datetime] = None
    next_update_expected: Optional[datetime] = None


class OverlayScoreResponse(BaseModel):
    symbol: str
    as_of_ts: Optional[datetime] = None
    overlay_score: float = 0.0
    overlay_confidence: float = 0.0
    overlay_band: Optional[str] = None
    hold_extension_hint: Optional[str] = None
    add_on_eligibility: Optional[bool] = None
    features: OverlayFeatures = Field(default_factory=OverlayFeatures)
    source_presence: OverlaySourcePresence = Field(default_factory=OverlaySourcePresence)
    source_details: OverlaySourceDetails = Field(default_factory=OverlaySourceDetails)
    metadata: OverlayMetadata = Field(default_factory=OverlayMetadata)


class BulkOverlayResponse(BaseModel):
    results: List[OverlayScoreResponse]
    total_count: int
    metadata: Dict[str, Any] = Field(default_factory=dict)


class OverlayTopMover(BaseModel):
    symbol: str
    overlay_score: float
    overlay_band: Optional[str] = None
    as_of_ts: Optional[datetime] = None


class TopMoversResponse(BaseModel):
    top_movers: List[OverlayTopMover]
    total_count: int
    metadata: Dict[str, Any] = Field(default_factory=dict)


class HeadlineItem(BaseModel):
    title: str
    publisher: Optional[str] = None
    published_at: datetime
    article_guid: str


class HeadlinesResponse(BaseModel):
    symbol: str
    headlines: List[HeadlineItem]
    headline_count_6h: int = 0
    headline_count_24h: int = 0
    publisher_breadth_24h: int = 0
    metadata: Dict[str, Any] = Field(default_factory=dict)


class VideoItem(BaseModel):
    video_id: str
    channel_id: str
    title: str
    view_count: int = 0
    comment_count: int = 0
    published_at: Optional[datetime] = None
    channel_weight: float = 0.5


class YouTubeResponse(BaseModel):
    symbol: str
    videos: List[VideoItem]
    mentions_24h: int = 0
    weighted_views_24h: float = 0.0
    metadata: Dict[str, Any] = Field(default_factory=dict)


class WikiPageviewPoint(BaseModel):
    date: datetime
    views: int
    page_title: str


class WikiResponse(BaseModel):
    symbol: str
    pageviews: List[WikiPageviewPoint]
    views_1d: Optional[int] = None
    views_7d_avg: Optional[float] = None
    metadata: Dict[str, Any] = Field(default_factory=dict)


class CrowdingResponse(BaseModel):
    symbol: str
    short_volume_ratio: Optional[float] = None
    short_volume_spike_zscore: Optional[float] = None
    crowding_stress_z: Optional[float] = None
    metadata: Dict[str, Any] = Field(default_factory=dict)


class TrendPoint(BaseModel):
    observed_at: datetime
    interest_value: int
    topic_id: str
    topic_label: Optional[str] = None


class TrendsResponse(BaseModel):
    symbol: str
    trends: List[TrendPoint]
    theme_heat_z: Optional[float] = None
    metadata: Dict[str, Any] = Field(default_factory=dict)


class OverlayHistoryPoint(BaseModel):
    as_of_ts: datetime
    overlay_score: float
    overlay_confidence: float
    overlay_band: Optional[str] = None


class OverlayHistoryResponse(BaseModel):
    symbol: str
    history: List[OverlayHistoryPoint]
    metadata: Dict[str, Any] = Field(default_factory=dict)


class SourceHealthItem(BaseModel):
    source: str
    last_collected_at: Optional[datetime] = None
    status: str = "unknown"
    success_rate_24h: Optional[float] = None
    records_24h: int = 0


class AdminHealthResponse(BaseModel):
    overlay_enabled: bool
    sources: List[SourceHealthItem]
    last_pipeline_run: Optional[datetime] = None
    metadata: Dict[str, Any] = Field(default_factory=dict)


class TriggerPipelineResponse(BaseModel):
    status: str
    message: str
    job_ids: List[str] = Field(default_factory=list)


class JobLogEntry(BaseModel):
    id: str
    job_type: str
    status: str
    started_at: datetime
    completed_at: Optional[datetime] = None
    records_processed: int = 0
    error_message: Optional[str] = None


class JobLogResponse(BaseModel):
    logs: List[JobLogEntry]
    total_count: int
    metadata: Dict[str, Any] = Field(default_factory=dict)