GeoSight MCP Server

""" Report generation tool for comprehensive analysis. """ from datetime import datetime from typing import Optional, List, Literal import structlog from geosight.tools.geocoding import resolve_location from geosight.tools.indices import calculate_ndvi, calculate_ndwi from geosight.tools.classification import detect_land_cover from geosight.tools.change_detection import detect_changes logger = structlog.get_logger(__name__) async def generate_report( start_date: str, end_date: str, latitude: Optional[float] = None, longitude: Optional[float] = None, location_name: Optional[str] = None, radius_km: float = 10.0, analyses: Optional[List[str]] = None, output_format: Literal["pdf", "html", "markdown"] = "markdown", report_title: Optional[str] = None, ) -> dict: """Generate comprehensive analysis report.""" try: lat, lon, display_name = await resolve_location( location_name=location_name, latitude=latitude, longitude=longitude, ) if analyses is None: analyses = ["ndvi", "land_cover"] logger.info( "generating_report", location=display_name, analyses=analyses, ) title = report_title or f"Satellite Analysis Report: {display_name}" # Run requested analyses results = {} if "ndvi" in analyses: results["ndvi"] = await calculate_ndvi( start_date=start_date, end_date=end_date, latitude=lat, longitude=lon, radius_km=radius_km, ) if "ndwi" in analyses: results["ndwi"] = await calculate_ndwi( start_date=start_date, end_date=end_date, latitude=lat, longitude=lon, radius_km=radius_km, ) if "land_cover" in analyses: results["land_cover"] = await detect_land_cover( date=end_date, latitude=lat, longitude=lon, radius_km=radius_km, ) if "change_detection" in analyses: results["change_detection"] = await detect_changes( date_before=start_date, date_after=end_date, latitude=lat, longitude=lon, radius_km=radius_km, ) # Generate report content if output_format == "markdown": report_content = generate_markdown_report( title, display_name, lat, lon, start_date, end_date, radius_km, results ) elif output_format == "html": report_content = generate_html_report( title, display_name, lat, lon, start_date, end_date, radius_km, results ) else: report_content = generate_markdown_report( title, display_name, lat, lon, start_date, end_date, radius_km, results ) summary = f"""📄 **Report Generated Successfully** 📍 **Location:** {display_name} 📅 **Period:** {start_date} to {end_date} 📐 **Area:** {(radius_km * 2) ** 2:.1f} km² 📊 **Analyses included:** {', '.join(analyses)} 📝 **Format:** {output_format.upper()} The report includes detailed analysis results with visualizations. """ return { "summary": summary, "report_content": report_content, "format": output_format, "title": title, "analyses_completed": list(results.keys()), "statistics": { "location": display_name, "latitude": lat, "longitude": lon, "start_date": start_date, "end_date": end_date, "radius_km": radius_km, }, } except ValueError as e: return {"summary": f"❌ **Error:** {str(e)}", "error": str(e)} except Exception as e: logger.error("report_generation_error", error=str(e), exc_info=True) return {"summary": f"❌ **Report generation failed:** {str(e)}", "error": str(e)} def generate_markdown_report( title: str, location: str, lat: float, lon: float, start_date: str, end_date: str, radius_km: float, results: dict, ) -> str: """Generate Markdown report.""" sections = [ f"# {title}", "", f"**Generated:** {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}", "", "## Executive Summary", "", f"This report presents satellite imagery analysis for **{location}** ", f"covering the period from {start_date} to {end_date}.", "", "## Study Area", "", f"- **Location:** {location}", f"- **Coordinates:** {lat:.4f}°N, {lon:.4f}°E", f"- **Analysis Area:** {(radius_km * 2):.1f} km × {(radius_km * 2):.1f} km ({(radius_km * 2) ** 2:.1f} km²)", f"- **Date Range:** {start_date} to {end_date}", "", ] if "ndvi" in results: ndvi = results["ndvi"] stats = ndvi.get("statistics", {}) sections.extend([ "## Vegetation Analysis (NDVI)", "", "The Normalized Difference Vegetation Index (NDVI) measures vegetation health and density.", "", f"- **Mean NDVI:** {stats.get('mean_ndvi', 'N/A'):.3f}" if isinstance(stats.get('mean_ndvi'), (int, float)) else "- **Mean NDVI:** N/A", f"- **Dense Vegetation:** {stats.get('dense_vegetation_pct', 0):.1f}%", f"- **Moderate Vegetation:** {stats.get('moderate_vegetation_pct', 0):.1f}%", f"- **Sparse Vegetation:** {stats.get('sparse_vegetation_pct', 0):.1f}%", "", ]) if "ndwi" in results: ndwi = results["ndwi"] stats = ndwi.get("statistics", {}) sections.extend([ "## Water Analysis (NDWI)", "", "The Normalized Difference Water Index (NDWI) identifies water bodies and moisture.", "", f"- **Water Coverage:** {stats.get('water_coverage_pct', 0):.1f}%", f"- **Water Area:** {stats.get('water_area_km2', 0):.2f} km²", "", ]) if "land_cover" in results: lc = results["land_cover"] stats = lc.get("statistics", {}) classes = stats.get("class_percentages", {}) sections.extend([ "## Land Cover Classification", "", "Machine learning classification of land cover types:", "", ]) for class_name, pct in sorted(classes.items(), key=lambda x: -x[1]): if pct > 0.5: sections.append(f"- **{class_name}:** {pct:.1f}%") sections.append("") if "change_detection" in results: cd = results["change_detection"] stats = cd.get("statistics", {}) sections.extend([ "## Change Detection", "", f"Analysis of changes between {start_date} and {end_date}:", "", f"- **Changed Area:** {stats.get('changed_area_km2', 0):.2f} km²", f"- **Change Percentage:** {stats.get('change_percentage', 0):.1f}%", "", ]) sections.extend([ "## Methodology", "", "This analysis uses Sentinel-2 satellite imagery processed through the GeoSight platform.", "Machine learning models are applied for land cover classification and object detection.", "", "## Disclaimer", "", "This report is generated automatically and should be verified with ground truth data.", "", "---", f"*Report generated by GeoSight MCP Server v1.0.0*", ]) return "\n".join(sections) def generate_html_report( title: str, location: str, lat: float, lon: float, start_date: str, end_date: str, radius_km: float, results: dict, ) -> str: """Generate HTML report.""" markdown_content = generate_markdown_report( title, location, lat, lon, start_date, end_date, radius_km, results ) # Simple markdown to HTML conversion html_content = f"""<!DOCTYPE html> <html> <head> <meta charset="UTF-8"> <title>{title}</title> <style> body {{ font-family: Arial, sans-serif; max-width: 800px; margin: 0 auto; padding: 20px; }} h1 {{ color: #2c3e50; border-bottom: 2px solid #3498db; }} h2 {{ color: #34495e; margin-top: 30px; }} ul {{ list-style-type: disc; }} .stats {{ background: #f8f9fa; padding: 15px; border-radius: 5px; }} </style> </head> <body> <pre>{markdown_content}</pre> </body> </html>""" return html_content