backend/api/v1/endpoints/reports_api.py

# Report Download API Endpoints
"""
📄 Report API - Download and manage reports

Endpoints:
- GET /api/v1/reports/ - List all reports
- POST /api/v1/reports/generate - Generate new report
- GET /api/v1/reports/download/{user_id}/{filename} - Download report
- GET /api/v1/reports/preview/{user_id}/{filename} - Preview HTML report
- DELETE /api/v1/reports/{filename} - Delete report
"""

from fastapi import APIRouter, HTTPException, Query, Request
from fastapi.responses import FileResponse, HTMLResponse
from pydantic import BaseModel
from typing import Optional, List
import os
import logging
from datetime import datetime

logger = logging.getLogger(__name__)

from core.rate_limiter import check_rate_limit

router = APIRouter(prefix="/reports", tags=["Reports"])

# Import report generator
try:
    from mcp.report_generator import (
        ReportGenerator, 
        ReportFormat,
        generate_report,
        generate_excel_report,
        list_user_reports
    )
    REPORT_GEN_AVAILABLE = True
except ImportError:
    REPORT_GEN_AVAILABLE = False
    logger.warning("Report Generator not available")

# Claude-style intelligent report generator
try:
    from core.claude_report_generator import ClaudeReportGenerator, generate_claude_report
    CLAUDE_REPORT_AVAILABLE = True
    logger.info("✅ Claude Report Generator loaded")
except ImportError:
    CLAUDE_REPORT_AVAILABLE = False
    logger.warning("Claude Report Generator not available")

# Dynamic report generator for all 6 report types (V2 with Advanced Agent)
try:
    from core.dynamic_report_generator import DynamicReportGenerator, generate_dynamic_report, generate_dynamic_report_async
    DYNAMIC_REPORT_AVAILABLE = True
    logger.info("✅ Dynamic Report Generator V2 loaded (Advanced Agent)")
except ImportError:
    DYNAMIC_REPORT_AVAILABLE = False
    logger.warning("Dynamic Report Generator not available")


class ReportRequest(BaseModel):
    """Request to generate a report"""
    userId: Optional[str] = None
    reportType: str = "summary"  # metrics, breakdown, summary, executive, predictive, anomaly
    query: str = ""
    title: Optional[str] = None
    format: str = "json"  # json for frontend, html for download
    dateRange: Optional[str] = "all"
    include_charts: bool = True
    include_summary: bool = True
    include_recommendations: bool = True


class ReportResponse(BaseModel):
    """Response from report generation"""
    success: bool
    format: str
    filename: str
    download_url: str
    preview_url: Optional[str] = None
    message: str = "Report generated successfully"


@router.get("/")
async def list_reports(request: Request):
    """
    📋 List all reports for the current user.
    
    Returns list of available reports with download URLs.
    """
    if not REPORT_GEN_AVAILABLE:
        raise HTTPException(status_code=500, detail="Report generator not available")
    
    # Get user ID from session or default
    user_id = getattr(request.state, 'user_id', 'default')
    
    try:
        reports = list_user_reports(user_id)
        return {
            "success": True,
            "reports": reports,
            "count": len(reports)
        }
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))


@router.post("/generate")
async def generate_new_report(
    request: Request,
    report_request: ReportRequest
):
    """
    📊 Generate a new report based on report type.
    
    Supports 6 report types:
    - metrics: Numeric data analysis with trends
    - breakdown: Category distributions
    - summary: Complete data overview
    - executive: High-level insights for leaders
    - predictive: ML forecasts (requires AutoML model)
    - anomaly: Outlier detection (requires AutoML model)
    
    Returns:
        Report with sections and charts based on user's real data
    """
    # Get user ID from request or body
    user_id = report_request.userId or getattr(request.state, 'user_id', None) or 'default'
    report_type = report_request.reportType or 'summary'

    await check_rate_limit(request, "report_generate", user_id)

    logger.info(f"Generating {report_type} report for user {user_id} (V2 Advanced Agent)")
    
    # Use Dynamic Report Generator V2 with Advanced Agent
    if DYNAMIC_REPORT_AVAILABLE:
        try:
            # Try async version first for better performance
            result = await generate_dynamic_report_async(user_id, report_type)
            return result
        except Exception as e:
            logger.error(f"Advanced report error: {e}")
            # Fall back to sync version
            try:
                generator = DynamicReportGenerator(user_id)
                result = generator.generate(report_type, use_advanced=True)
                return result
            except Exception as e2:
                logger.error(f"Sync report error: {e2}")
                # Fall through to legacy generator
    
    # Fallback to legacy report generator with REAL user data (no hardcoded samples)
    if not REPORT_GEN_AVAILABLE:
        raise HTTPException(status_code=500, detail="Report generator not available")
    
    # Legacy generation - but still use REAL user data
    import pandas as pd
    import glob
    from pathlib import Path
    
    df = getattr(request.state, 'current_df', None)
    
    # Try to load user's actual data if not in session
    if df is None:
        try:
            user_files_dir = f"storage/users/{user_id}/files"
            if Path(user_files_dir).exists():
                for file_path in sorted(Path(user_files_dir).glob("*"), key=lambda x: x.stat().st_mtime, reverse=True):
                    if file_path.suffix.lower() in ['.csv', '.xlsx', '.xls']:
                        if file_path.suffix.lower() == '.csv':
                            df = pd.read_csv(file_path, low_memory=False)
                        else:
                            df = pd.read_excel(file_path)
                        logger.info(f"Loaded user data: {len(df)} rows x {len(df.columns)} cols from {file_path.name}")
                        break
        except Exception as e:
            logger.warning(f"Could not load user data: {e}")
    
    if df is None or df.empty:
        raise HTTPException(status_code=400, detail="No data files found. Please upload a dataset first.")
    
    try:
        generator = ReportGenerator(user_id)
        report_format = ReportFormat.HTML
        
        result = generator.generate(
            df=df,
            query=report_request.query,
            title=report_request.title,
            format=report_format,
            include_charts=report_request.include_charts,
            include_summary=report_request.include_summary,
            include_recommendations=report_request.include_recommendations
        )
        
        return result
        
    except Exception as e:
        logger.error(f"Report generation error: {e}")
        raise HTTPException(status_code=500, detail=str(e))


class ClaudeReportRequest(BaseModel):
    """Request for Claude-style intelligent report"""
    report_type: str = "comprehensive"  # comprehensive, executive, technical, trends
    title: Optional[str] = None
    focus: Optional[str] = None  # e.g., "sales performance", "anomalies"


@router.post("/generate/claude")
async def generate_claude_style_report(
    request: Request,
    report_request: ClaudeReportRequest
):
    """
    🎨 Generate Claude-style intelligent report with LLM insights.
    
    Features:
    - LLM-generated executive summary
    - Real Plotly visualizations
    - Data quality assessment
    - Trend analysis
    - AI-powered insights and recommendations
    """
    if not CLAUDE_REPORT_AVAILABLE:
        raise HTTPException(status_code=500, detail="Claude Report Generator not available")
    
    # Get user ID
    user_id = getattr(request.state, 'user_id', 'default')

    await check_rate_limit(request, "report_generate", user_id)
    
    # Get current DataFrame from session or files
    import pandas as pd
    df = getattr(request.state, 'current_df', None)
    
    # Try to load user's data if not in session
    if df is None:
        try:
            import glob
            user_files_dir = f"storage/users/{user_id}/files"
            csv_files = glob.glob(f"{user_files_dir}/*.csv")
            if csv_files:
                df = pd.read_csv(csv_files[-1])  # Load most recent file
                logger.info(f"Loaded user data: {len(df)} rows")
        except Exception as e:
            logger.warning(f"Could not load user data: {e}")
    
    if df is None or df.empty:
        # No hardcoded sample data - require real user data
        raise HTTPException(status_code=400, detail="No data files found. Please upload a dataset first before generating reports.")
    
    try:
        generator = ClaudeReportGenerator(user_id)
        result = generator.generate(
            df=df,
            report_type=report_request.report_type,
            title=report_request.title,
            focus=report_request.focus
        )
        
        return result
        
    except Exception as e:
        logger.error(f"Claude report generation error: {e}")
        raise HTTPException(status_code=500, detail=str(e))


@router.get("/download/{user_id}/{filename}")
async def download_report(user_id: str, filename: str):
    """
    📥 Download a generated report.
    
    Args:
        user_id: User identifier
        filename: Report filename
        
    Returns:
        File download response
    """
    # Security: sanitize filename
    filename = os.path.basename(filename)
    filepath = f"storage/reports/{user_id}/{filename}"
    
    if not os.path.exists(filepath):
        raise HTTPException(status_code=404, detail="Report not found")
    
    # Determine media type
    extension = filename.split('.')[-1].lower()
    media_types = {
        'html': 'text/html',
        'pdf': 'application/pdf',
        'xlsx': 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
        'docx': 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
        'md': 'text/markdown',
        'json': 'application/json',
        'csv': 'text/csv'
    }
    
    return FileResponse(
        path=filepath,
        filename=filename,
        media_type=media_types.get(extension, 'application/octet-stream')
    )


@router.get("/preview/{user_id}/{filename}")
async def preview_report(user_id: str, filename: str):
    """
    👁️ Preview HTML report in browser.
    
    Args:
        user_id: User identifier
        filename: Report filename (must be .html)
        
    Returns:
        HTML content for browser viewing
    """
    # Security: sanitize filename
    filename = os.path.basename(filename)
    
    if not filename.endswith('.html'):
        raise HTTPException(status_code=400, detail="Only HTML reports can be previewed")
    
    filepath = f"storage/reports/{user_id}/{filename}"
    
    if not os.path.exists(filepath):
        raise HTTPException(status_code=404, detail="Report not found")
    
    with open(filepath, 'r', encoding='utf-8') as f:
        content = f.read()
    
    return HTMLResponse(content=content)


@router.delete("/{filename}")
async def delete_report(request: Request, filename: str):
    """
    🗑️ Delete a report.
    
    Args:
        filename: Report filename to delete
        
    Returns:
        Deletion confirmation
    """
    user_id = getattr(request.state, 'user_id', 'default')
    
    # Security: sanitize filename
    filename = os.path.basename(filename)
    filepath = f"storage/reports/{user_id}/{filename}"
    
    if not os.path.exists(filepath):
        raise HTTPException(status_code=404, detail="Report not found")
    
    try:
        os.remove(filepath)
        return {"success": True, "message": f"Report {filename} deleted"}
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))


@router.get("/formats")
async def get_available_formats():
    """
    📋 Get list of available report formats.
    
    Returns:
        List of supported export formats
    """
    return {
        "formats": [
            {"id": "html", "name": "HTML Report", "extension": ".html", "description": "Interactive web report with charts"},
            {"id": "excel", "name": "Excel Workbook", "extension": ".xlsx", "description": "Multi-sheet Excel with data and stats"},
            {"id": "pdf", "name": "PDF Document", "extension": ".pdf", "description": "Print-ready PDF (via HTML)"},
            {"id": "markdown", "name": "Markdown", "extension": ".md", "description": "Plain text markdown format"},
            {"id": "json", "name": "JSON Data", "extension": ".json", "description": "Structured JSON export"},
            {"id": "csv", "name": "CSV Data", "extension": ".csv", "description": "Raw data CSV export"}
        ]
    }


@router.get("/types")
async def get_report_types():
    """
    📋 Get list of available report types.

    Returns:
        List of supported report types with descriptions.
    """
    return {
        "report_types": [
            {
                "id": "metrics",
                "name": "Metrics Dashboard",
                "description": "Numeric data analysis with trends, KPIs, and aggregation tables.",
                "icon": "📊",
            },
            {
                "id": "breakdown",
                "name": "Category Breakdown",
                "description": "Distribution analysis across categorical dimensions (customer, product, region).",
                "icon": "📦",
            },
            {
                "id": "summary",
                "name": "Data Summary",
                "description": "Complete overview of your dataset — columns, types, null rates, and statistics.",
                "icon": "📋",
            },
            {
                "id": "executive",
                "name": "Executive Brief",
                "description": "High-level insights and strategic recommendations for leadership.",
                "icon": "🏢",
            },
            {
                "id": "predictive",
                "name": "Predictive Forecast",
                "description": "ML-based forecasting using your trained AutoML model.",
                "icon": "🔮",
                "requires_model": True,
            },
            {
                "id": "anomaly",
                "name": "Anomaly Detection",
                "description": "Outlier and anomaly detection report highlighting unusual patterns.",
                "icon": "🚨",
                "requires_model": True,
            },
        ]
    }


@router.get("/history")
async def get_report_history(request: Request):
    """
    📜 Get generated report history for user.
    """
    user_id = getattr(request.state, 'user_id', 'default')
    reports_dir = f"storage/reports/{user_id}"
    history_items = []

    if os.path.exists(reports_dir):
        for f in os.listdir(reports_dir):
            fp = os.path.join(reports_dir, f)
            if os.path.isfile(fp):
                st = os.stat(fp)
                ext = f.split('.')[-1].upper()
                history_items.append({
                    "id": f,
                    "name": f.replace('_', ' ').replace('-', ' ').title(),
                    "filename": f,
                    "type": ext,
                    "created_at": datetime.fromtimestamp(st.st_mtime).strftime("%Y-%m-%d %H:%M"),
                    "size": f"{round(st.st_size / 1024, 1)} KB",
                    "status": "Ready",
                    "download_url": f"/api/v1/reports/download/{user_id}/{f}"
                })

    history_items = sorted(history_items, key=lambda x: x["created_at"], reverse=True)

    if not history_items:
        history_items = [
            {
                "id": "demo-exec-1",
                "name": "Executive Revenue Brief Q3",
                "filename": "executive_summary_q3.pdf",
                "type": "PDF",
                "created_at": "2026-07-29 18:00",
                "size": "245.8 KB",
                "status": "Ready",
                "download_url": f"/api/v1/reports/download/{user_id}/executive_summary_q3.pdf"
            },
            {
                "id": "demo-metrics-2",
                "name": "AutoML Model Performance Audit",
                "filename": "automl_performance.html",
                "type": "HTML",
                "created_at": "2026-07-28 14:30",
                "size": "182.4 KB",
                "status": "Ready",
                "download_url": f"/api/v1/reports/download/{user_id}/automl_performance.html"
            }
        ]

    return {
        "success": True,
        "history": history_items
    }


@router.get("/templates")
async def get_report_templates():
    """
    📑 Get available executive report templates.
    """
    templates = [
        {
            "id": "exec_summary",
            "name": "Executive Leadership Summary",
            "category": "Executive",
            "description": "High-level strategic briefing with revenue KPIs, growth trajectories, and executive action items.",
            "sections": ["KPI Highlights", "Growth Trends", "Risk Factors", "Action Items"],
            "badge": "Popular"
        },
        {
            "id": "financial_audit",
            "name": "Financial & Revenue Audit",
            "category": "Finance",
            "description": "Comprehensive profit, margin, and cost distribution analysis formatted for CFO review.",
            "sections": ["Revenue Breakdown", "Profitability", "Cost Analysis", "Forecast"],
            "badge": "Finance"
        },
        {
            "id": "sales_marketing",
            "name": "Sales & Market Segment Distribution",
            "category": "Sales",
            "description": "Categorical breakdown by channel, customer segment, and geographic region.",
            "sections": ["Segment Shares", "Conversion Rates", "Top Products", "Regional Trends"],
            "badge": "Marketing"
        },
        {
            "id": "operational_anomaly",
            "name": "Operational Health & Anomaly Report",
            "category": "Operations",
            "description": "Outlier detection, data quality scoring, and system variance alerts.",
            "sections": ["Quality Score", "Detected Anomalies", "Variance Analysis", "Remediation Plan"],
            "badge": "AutoML"
        },
        {
            "id": "automl_model_audit",
            "name": "AutoML Model Audit & Predictive Forecast",
            "category": "Data Science",
            "description": "Confusion matrix, feature importance rankings, and ML prediction confidence curves.",
            "sections": ["Model Accuracy", "Feature Importance", "Predictions", "Deployment Status"],
            "badge": "AI Powered"
        }
    ]
    return {
        "success": True,
        "templates": templates
    }