"""
Pydantic models for API data validation.

Defines request and response schemas with validation rules.
"""

from datetime import datetime
from typing import Optional

from pydantic import BaseModel, ConfigDict, Field, field_serializer, field_validator


class IssueInput(BaseModel):
    """Input model for GitHub issue or pull request classification."""

    issue_text: str = Field(
        ...,
        min_length=1,
        description="Issue title text",
        examples=["Fix bug in authentication module"],
    )
    issue_description: Optional[str] = Field(
        default=None,
        description="Issue body text",
        examples=["The authentication module fails when handling expired tokens"],
    )
    repo_name: Optional[str] = Field(
        default=None, description="Repository name", examples=["user/repo-name"]
    )
    pr_number: Optional[int] = Field(
        default=None, ge=1, description="Pull request number", examples=[123]
    )
    created_at: Optional[datetime] = Field(
        default=None, description="Issue creation timestamp", examples=["2024-01-15T10:30:00Z"]
    )
    author_name: Optional[str] = Field(
        default=None, description="Issue author username", examples=["johndoe"]
    )

    @field_validator("issue_text", "issue_description")
    @classmethod
    def clean_text(cls, v: Optional[str]) -> Optional[str]:
        """Validate and clean text fields."""
        if v is None:
            return v
        v = v.strip()
        if not v:
            raise ValueError("Text cannot be empty or whitespace only")
        return v

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "issue_text": "Add support for OAuth authentication",
                "issue_description": "Implement OAuth 2.0 flow for third-party providers",
                "repo_name": "myorg/myproject",
                "pr_number": 456,
                "author_name": "developer123",
            }
        }
    )


class SkillPrediction(BaseModel):
    """Single skill prediction with confidence score."""

    skill_name: str = Field(
        ...,
        description="Name of the predicted skill (domain/subdomain)",
        examples=["Language/Java", "DevOps/CI-CD"],
    )
    confidence: float = Field(
        ..., ge=0.0, le=1.0, description="Confidence score (0.0 to 1.0)", examples=[0.85]
    )

    model_config = ConfigDict(
        json_schema_extra={"example": {"skill_name": "Language/Java", "confidence": 0.92}}
    )


class PredictionResponse(BaseModel):
    """Response model for skill classification predictions."""

    predictions: list[SkillPrediction] = Field(
        default_factory=list, description="List of predicted skills with confidence scores"
    )
    num_predictions: int = Field(
        ..., ge=0, description="Total number of predicted skills", examples=[5]
    )
    model_version: str = Field(default="1.0.0", description="Model version", examples=["1.0.0"])
    processing_time_ms: Optional[float] = Field(
        default=None, ge=0.0, description="Processing time in milliseconds", examples=[125.5]
    )

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "predictions": [
                    {"skill_name": "Language/Java", "confidence": 0.92},
                    {"skill_name": "DevOps/CI-CD", "confidence": 0.78},
                ],
                "num_predictions": 2,
                "model_version": "1.0.0",
                "processing_time_ms": 125.5,
            }
        }
    )


class BatchIssueInput(BaseModel):
    """Input model for batch prediction."""

    issues: list[IssueInput] = Field(
        ...,
        min_length=1,
        max_length=100,
        description="Issues to classify (max 100)",
    )

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "issues": [
                    {
                        "issue_text": "Fix authentication bug",
                        "issue_description": "Users cannot login with OAuth",
                    },
                    {
                        "issue_text": "Add database migration",
                        "issue_description": "Create migration for new user table",
                    },
                ]
            }
        }
    )


class BatchPredictionResponse(BaseModel):
    """Response model for batch predictions."""

    results: list[PredictionResponse] = Field(
        default_factory=list, description="Prediction results, one per issue"
    )
    total_issues: int = Field(..., ge=0, description="Number of issues processed", examples=[2])
    total_processing_time_ms: Optional[float] = Field(
        default=None, ge=0.0, description="Processing time in milliseconds", examples=[250.0]
    )

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "results": [
                    {
                        "predictions": [{"skill_name": "Language/Java", "confidence": 0.92}],
                        "num_predictions": 1,
                        "model_version": "1.0.0",
                    }
                ],
                "total_issues": 2,
                "total_processing_time_ms": 250.0,
            }
        }
    )


class ErrorResponse(BaseModel):
    """Error response model."""

    error: str = Field(..., description="Error message", examples=["Invalid input"])
    detail: Optional[str] = Field(
        default=None, description="Detailed error", examples=["Field 'issue_text' is required"]
    )
    timestamp: datetime = Field(default_factory=datetime.now, description="Error timestamp")

    @field_serializer("timestamp")
    def serialize_timestamp(self, value: datetime) -> str:
        return value.isoformat()

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "error": "Validation Error",
                "detail": "issue_text: field required",
                "timestamp": "2024-01-15T10:30:00Z",
            }
        }
    )


class HealthCheckResponse(BaseModel):
    """Health check response model."""

    status: str = Field(default="healthy", description="Service status", examples=["healthy"])
    model_loaded: bool = Field(..., description="Model ready status", examples=[True])
    version: str = Field(default="1.0.0", description="API version", examples=["1.0.0"])
    timestamp: datetime = Field(default_factory=datetime.now, description="Timestamp")


class PredictionRecord(PredictionResponse):
    """Extended prediction model with metadata from MLflow."""

    run_id: str = Field(..., description="MLflow Run ID")
    timestamp: datetime = Field(..., description="Prediction timestamp")
    input_text: Optional[str] = Field(default="", description="Input text classified")

    model_config = ConfigDict(
        json_schema_extra={
            "example": {
                "predictions": [
                    {"skill_name": "Language/Java", "confidence": 0.92},
                    {"skill_name": "DevOps/CI-CD", "confidence": 0.78},
                ],
                "num_predictions": 2,
                "model_version": "1.0.0",
                "processing_time_ms": 125.5,
                "run_id": "a1b2c3d4e5f6",
                "timestamp": "2024-01-15T10:30:00Z",
                "input_text": "Fix bug in authentication module",
            }
        }
    )