"""
FastAPI Backend Server for AuthorCheck AI Detection
Uses DeBERTa model for AI-generated text detection
"""

from fastapi import FastAPI, HTTPException
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel, Field
from typing import List, Optional
import uvicorn
from model_handler import AIDetectionModelHandler
import logging

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
)
logger = logging.getLogger(__name__)

# Initialize FastAPI app
app = FastAPI(
    title="AuthorCheck API",
    description="AI-powered text analysis and detection API",
    version="1.0.0"
)

# Configure CORS
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

# Initialize model handler
model_handler = AIDetectionModelHandler()


# Request/Response Models
class AnalysisRequest(BaseModel):
    text: str = Field(..., min_length=1, description="Text to analyze")
    analysisTypes: Optional[List[str]] = Field(
        default=["all"],
        description="Types of analysis to perform"
    )
    model: Optional[str] = Field(
        default="ai-detector",
        description="Model to use for analysis"
    )


class EmotionData(BaseModel):
    emotion: str
    score: float
    intensity: str


class AdvancedSentiment(BaseModel):
    emotions: List[EmotionData]
    confidence: float
    context: str
    avg_polarity: Optional[float] = None
    polarity_variance: Optional[float] = None


class TopicData(BaseModel):
    topic: str
    relevance: float
    keywords: List[str]


class WritingStyle(BaseModel):
    tone: str
    formality: str
    complexity: str
    style: List[str]
    audience: str
    sentiment_consistency: Optional[str] = None


class Insight(BaseModel):
    type: str
    title: str
    description: str
    suggestion: Optional[str] = None


class PlagiarismRisk(BaseModel):
    score: int
    level: str
    details: str


class ContentQuality(BaseModel):
    overall: int
    clarity: int
    coherence: int
    engagement: int
    originality: int


class AnalysisResponse(BaseModel):
    advancedSentiment: AdvancedSentiment
    topics: List[TopicData]
    writingStyle: WritingStyle
    insights: List[Insight]
    plagiarismRisk: PlagiarismRisk
    contentQuality: ContentQuality
    aiOrHuman: str
    aiOrHumanConfidence: float
    aiOrHumanExplanation: str


# API Endpoints
@app.get("/")
async def root():
    """Root endpoint - API health check"""
    return {
        "status": "online",
        "message": "AuthorCheck API is running",
        "version": "1.0.0"
    }


@app.get("/health")
async def health_check():
    """Health check endpoint"""
    try:
        model_loaded = model_handler.is_loaded()
        return {
            "status": "healthy" if model_loaded else "degraded",
            "model_loaded": model_loaded,
            "model_type": "DeBERTa AI Detector"
        }
    except Exception as e:
        logger.error(f"Health check failed: {e}")
        return {
            "status": "unhealthy",
            "error": str(e)
        }


@app.post("/api/analyze")
async def analyze_text(request: AnalysisRequest):
    """
    Analyze text using the DeBERTa AI detection model
    
    Returns comprehensive analysis including:
    - AI vs Human detection
    - Sentiment analysis
    - Topic detection
    - Writing style analysis
    - Content quality metrics
    """
    try:
        if not request.text or len(request.text.strip()) == 0:
            raise HTTPException(status_code=400, detail="Text cannot be empty")
        
        # Check text length for meaningful analysis (50-7000 words)
        word_count = len(request.text.split())
        if word_count < 50:
            raise HTTPException(
                status_code=400,
                detail="Text is too short for analysis. Please provide at least 50 words for accurate AI detection and sentiment analysis."
            )
        
        if word_count > 7000:
            raise HTTPException(
                status_code=400,
                detail="Text is too long for analysis. Maximum 7,000 words allowed."
            )
        
        # Perform AI detection using the model
        logger.info(f"Analyzing text of length: {len(request.text)}")
        analysis_result = model_handler.analyze_text(request.text)
        
        logger.info(f"Analysis complete: {analysis_result['aiOrHuman']} ({analysis_result['aiOrHumanConfidence']:.2f}%)")
        
        return analysis_result
        
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Analysis error: {e}", exc_info=True)
        raise HTTPException(
            status_code=500,
            detail=f"Analysis failed: {str(e)}"
        )


@app.post("/api/detect")
async def detect_ai(request: AnalysisRequest):
    """
    Simple endpoint for AI detection only
    Returns just the AI/Human classification
    """
    try:
        if not request.text or len(request.text.strip()) == 0:
            raise HTTPException(status_code=400, detail="Text cannot be empty")
        
        # Check text length (50-7000 words)
        word_count = len(request.text.split())
        if word_count < 50:
            raise HTTPException(
                status_code=400,
                detail="Text is too short. Please provide at least 50 words."
            )
        elif word_count > 7000:
            raise HTTPException(
                status_code=400,
                detail="Text is too long. Maximum 7,000 words allowed."
            )
        
        result = model_handler.detect_ai(request.text)
        
        return {
            "text": request.text[:100] + "..." if len(request.text) > 100 else request.text,
            "classification": result["classification"],
            "prediction": result.get("prediction", result["classification"]),
            "probability": result["probability"],
            "confidence": result["confidence"],
            "explanation": result["explanation"],
            "mixed_analysis": result.get("mixed_analysis", None),
            "modelProcessingTime": result.get("modelProcessingTime", None)
        }
        
    except HTTPException:
        raise
    except Exception as e:
        logger.error(f"Detection error: {e}", exc_info=True)
        raise HTTPException(
            status_code=500,
            detail=f"Detection failed: {str(e)}"
        )


@app.get("/api/model/info")
async def model_info():
    """Get information about the loaded model"""
    try:
        return model_handler.get_model_info()
    except Exception as e:
        logger.error(f"Model info error: {e}")
        raise HTTPException(status_code=500, detail=str(e))


# Run the server
if __name__ == "__main__":
    uvicorn.run(
        "app:app",
        host="0.0.0.0",
        port=8000,
        reload=True,
        log_level="info"
    )