mcp-stargazing

import json import os import time from typing import List, Dict, Any from astropy.table import Table from astroquery.simbad import Simbad from astroquery.vizier import Vizier import astropy.units as u from astropy.coordinates import SkyCoord def download_messier_objects() -> List[Dict[str, Any]]: print("Downloading Messier Catalog via Simbad...") # Configure Simbad to retrieve relevant fields # id(M) is deprecated, just use main_id or assume query order # flux(V) is deprecated in favor of 'V' custom_simbad = Simbad() custom_simbad.add_votable_fields('otype', 'flux(V)') # Simbad might still accept flux(V) or we should use 'V' # Actually, let's stick to simple defaults + otype + flux # Note: Simbad API changes frequently. # Let's try to remove 'id(M)' which caused the crash. messier_objects = [] # M1 to M110 identifiers = [f"M {i}" for i in range(1, 111)] # Query in batches to be safe, or one big query # Simbad query_objects can handle a list try: table = custom_simbad.query_objects(identifiers) if table is None: print("Error: No results from Simbad for Messier objects.") return [] for row in table: # Parse RA/Dec # Simbad returns RA/Dec as strings usually "HH MM SS", "DD MM SS" # Column names might be 'RA_d' or 'DEC_d' if we asked for degrees, or just 'RA'/'DEC' # Let's check typical defaults. Simbad default is usually 'RA' and 'DEC' in sexagesimal. ra_key = 'RA' if 'RA' in table.colnames else 'ra' dec_key = 'DEC' if 'DEC' in table.colnames else 'dec' if ra_key not in table.colnames: # Fallback: maybe 'RA_2000' or similar? # Let's print columns if we fail print(f"Columns found: {table.colnames}") raise KeyError("RA") ra_str = row[ra_key] dec_str = row[dec_key] coord = SkyCoord(f"{ra_str} {dec_str}", unit=(u.hourangle, u.deg)) # Parse Magnitude (Flux V) # Column name changed to 'FLUX_V' or just 'V' mag = 99.9 if 'FLUX_V' in row.colnames: mag = float(row['FLUX_V']) elif 'V' in row.colnames: mag = float(row['V']) # Parse Object Type # Column might be 'OTYPE' or 'OTYPE_S' or similar otype = "Unknown" if 'OTYPE' in table.colnames: otype = str(row['OTYPE']) elif 'otype' in table.colnames: otype = str(row['otype']) # Parse Name (Main ID) # Default main_id name = "Unknown" if 'MAIN_ID' in table.colnames: name = str(row['MAIN_ID']) elif 'main_id' in table.colnames: name = str(row['main_id']) messier_objects.append({ "name": name, "type": otype, "ra": float(coord.ra.deg), "dec": float(coord.dec.deg), "magnitude": mag, "catalog": "Messier" }) print(f"Successfully fetched {len(messier_objects)} Messier objects.") return messier_objects except Exception as e: print(f"Failed to download Messier objects: {e}") return [] def download_ngc_objects(max_mag: float = 10.0) -> List[Dict[str, Any]]: print(f"Downloading NGC Objects (Mag < {max_mag}) via Vizier...") # Use Vizier to query NGC 2000.0 Catalog (VII/118) # We want objects brighter than max_mag v = Vizier(columns=['NGC', 'Type', 'RA2000', 'DE2000', 'Mag', 'Const'], row_limit=10000) # Get enough rows # Filter: Mag < max_mag # Note: Vizier filters are strings like "< 10" try: # Catalog VII/118: NGC 2000.0 (Sky Publishing, ed. Sinnott 1988) # It seems requesting specific columns is filtering out the main ID column if we don't guess it right? # Or maybe Vizier is returning a different subset. # Let's try NOT specifying columns initially to see what we get by default. # This usually returns the default view. v = Vizier(row_limit=10000) catalogs = v.query_constraints(catalog='VII/118', Mag=f"<{max_mag}") if not catalogs: print("Error: No results from Vizier for NGC objects.") return [] table = catalogs[0] # Determine column names ngc_col = 'NGC' if 'NGC' in table.colnames else 'Name' # Fallback type_col = 'Type' if 'Type' in table.colnames else 'Otype' mag_col = 'Mag' if 'Mag' in table.colnames else 'V' if ngc_col not in table.colnames: # Try one more fallback: maybe it's just index? # But VII/118 definitely has 'NGC' column usually. print(f"NGC Columns found: {table.colnames}") return [] # Fail gracefully # Now re-query with the exact columns we want + hidden coordinates # We must use the exact names we found. cols = [ngc_col, type_col, mag_col, 'Const', '_RAJ2000', '_DEJ2000'] v = Vizier(columns=cols, row_limit=10000) catalogs = v.query_constraints(catalog='VII/118', Mag=f"<{max_mag}") table = catalogs[0] ngc_objects = [] for row in table: # Skip Messier objects (duplicates) # NGC objects that are also Messier usually have standard mapping, but simple check: # We will dedup later by position if needed, or just let them be. # Actually, let's keep them, or maybe filter known duplicates later. name = f"NGC {row[ngc_col]}" ra = float(row['_RAJ2000']) dec = float(row['_DEJ2000']) mag_val = 99.9 if mag_col in row.colnames and row[mag_col] is not None and str(row[mag_col]).strip() != '': mag_val = float(row[mag_col]) otype = str(row[type_col]) ngc_objects.append({ "name": name, "type": otype, "ra": ra, "dec": dec, "magnitude": mag_val, "catalog": "NGC" }) print(f"Successfully fetched {len(ngc_objects)} NGC objects.") return ngc_objects except Exception as e: print(f"Failed to download NGC objects: {e}") return [] def main(): messier = download_messier_objects() ngc = download_ngc_objects(max_mag=8.0) # Conservative limit for "Showpiece" # Merge all_objects = messier + ngc # Deduplicate based on position (very simple check) # Some NGC objects ARE Messier objects (e.g. M1 = NGC 1952) # We prefer the "M" name. # Create a spatial index or just simple distance check? # For < 500 objects, simple O(N^2) is fine or just check if we have a mapping. # Let's just save them all for now, duplicates aren't fatal. # Actually, let's remove NGC items if they are close to an existing Messier item. final_list = [] final_list.extend(messier) messier_coords = [SkyCoord(m['ra'], m['dec'], unit="deg") for m in messier] print("Deduplicating...") duplicates = 0 for n in ngc: n_coord = SkyCoord(n['ra'], n['dec'], unit="deg") is_dup = False # Check against all Messier # Threshold: 0.1 degree (6 arcmin) for idx, m_coord in enumerate(messier_coords): sep = n_coord.separation(m_coord) if sep.deg < 0.1: is_dup = True break if not is_dup: final_list.append(n) else: duplicates += 1 print(f"Removed {duplicates} duplicates (NGC objects that are Messier objects).") print(f"Total objects: {len(final_list)}") # Save to file output_path = os.path.join(os.path.dirname(__file__), '../src/data/objects.json') output_path = os.path.abspath(output_path) with open(output_path, 'w') as f: json.dump(final_list, f, indent=2) print(f"Saved to {output_path}") if __name__ == "__main__": main()