import base64
import os
from fastapi import FastAPI, Header, HTTPException
from fastapi.middleware.cors import CORSMiddleware
from pydantic import BaseModel
from typing import Optional, List

# Importing your internal logic
from src.processor import get_text_from_base64 
from src.analyzer import run_analysis

app = FastAPI(
    title="Intelligent Document Processing API",
    description="Backend Engine for AI-powered Summary, Entity Extraction, and Sentiment Analysis.",
    version="1.0.0"
)

# 1. ADD CORS MIDDLEWARE HERE
# This allows the task site's "Test" button to talk to your API
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"], 
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

# Configuration
API_KEY = os.getenv("API_KEY", "sk_track2_987654321")

# --- Data Models ---
class AnalysisRequest(BaseModel):
    fileName: str
    fileType: str 
    fileBase64: str

class EntityResults(BaseModel):
    names: List[str]
    organizations: List[str]
    dates: List[str]
    amounts: List[str]

# Simplified to match the task site's expected fields exactly
class AnalysisResponse(BaseModel):
    fileName: str
    summary: str
    entities: EntityResults
    sentiment: str

# 2. ADD ROOT ROUTE HERE

# --- API Endpoint ---
@app.post("/api/document-analyze", response_model=AnalysisResponse)
async def analyze_document(
    data: AnalysisRequest,
    x_api_key: Optional[str] = Header(None)
):
    # 1. Security Check
    if x_api_key != API_KEY:
        raise HTTPException(status_code=401, detail="Invalid API Key")

    try:
        # 2. Extract & Decode
        file_b64 = data.fileBase64
        
        # Base64 Padding Fix
        missing_padding = len(file_b64) % 4
        if missing_padding:
            file_b64 += '=' * (4 - missing_padding)
        
        try:
            file_bytes = base64.b64decode(file_b64)
        except Exception:
            raise HTTPException(status_code=400, detail="Corrupted file data")

        # 3. OCR / Extraction
        text = get_text_from_base64(file_bytes, data.fileType.lower())
        
        if not text or len(text.strip()) < 10:
            raise ValueError("Document is empty or text is unreadable (OCR Failed)")

        # 4. AI Analysis
        analysis = run_analysis(text)
        
        # 5. Return Structured Response (Matched to Task Site)
        return {
            "fileName": data.fileName,
            "summary": analysis["summary"],
            "entities": {
                "names": list(set(analysis["entities"].get("names", []))),
                "organizations": list(set(analysis["entities"].get("organizations", []))),
                "dates": list(set(analysis["entities"].get("dates", []))),
                "amounts": list(set(analysis["entities"].get("amounts", [])))
            },
            "sentiment": analysis["sentiment"] 
        }

    except ValueError as ve:
        raise HTTPException(status_code=422, detail=str(ve))
    except Exception as e:
        print(f"🔥 DEPLOYMENT ERROR: {str(e)}") 
        raise HTTPException(status_code=500, detail="Internal Server Error during AI analysis")