def add_spex(input_path, database): """ Function for adding the SpeX Prism Spectral Library to the database. Parameters ---------- input_path : str Path of the data folder. database : h5py._hl.files.File The HDF5 database. Returns ------- NoneType None """ distance_url = 'https://people.phys.ethz.ch/~stolkert/species/distance.dat' distance_file = os.path.join(input_path, 'distance.dat') if not os.path.isfile(distance_file): urllib.request.urlretrieve(distance_url, distance_file) distance_data = pd.pandas.read_csv( distance_file, usecols=[0, 3, 4], names=['object', 'distance', 'distance_error'], delimiter=',', dtype={ 'object': str, 'distance': float, 'distance_error': float }) database.create_group('spectra/spex') data_path = os.path.join(input_path, 'spex') if not os.path.exists(data_path): os.makedirs(data_path) url_all = 'http://svo2.cab.inta-csic.es/vocats/v2/spex/' \ 'cs.php?RA=180.000000&DEC=0.000000&SR=180.000000&VERB=2' xml_file_spex = os.path.join(data_path, 'spex.xml') if not os.path.isfile(xml_file_spex): urllib.request.urlretrieve(url_all, xml_file_spex) table = parse_single_table(xml_file_spex) # name = table.array['name'] twomass = table.array['name2m'] url = table.array['access_url'] unique_id = [] for i, item in enumerate(url): if twomass[i] not in unique_id: xml_file_1 = os.path.join(data_path, twomass[i].decode('utf-8') + '.xml') if not os.path.isfile(xml_file_1): urllib.request.urlretrieve(item.decode('utf-8'), xml_file_1) table = parse_single_table(xml_file_1) name = table.array['ID'] name = name[0].decode('utf-8') url = table.array['access_url'] print_message = f'Downloading SpeX Prism Spectral Library... {name}' print(f'\r{print_message:<72}', end='') xml_file_2 = os.path.join(data_path, f'spex_{name}.xml') if not os.path.isfile(xml_file_2): urllib.request.urlretrieve(url[0].decode('utf-8'), xml_file_2) unique_id.append(twomass[i]) print_message = 'Downloading SpeX Prism Spectral Library... [DONE]' print(f'\r{print_message:<72}') h_twomass = photometry.SyntheticPhotometry('2MASS/2MASS.H') # 2MASS H band zero point for 0 mag (Cogen et al. 2003) h_zp = 1.133e-9 # (W m-2 um-1) for votable in os.listdir(data_path): if votable.startswith('spex_') and votable.endswith('.xml'): xml_file = os.path.join(data_path, votable) table = parse_single_table(xml_file) wavelength = table.array['wavelength'] # (Angstrom) flux = table.array['flux'] # Normalized units wavelength = np.array(wavelength * 1e-4) # (um) flux = np.array(flux) # (a.u.) error = np.full(flux.size, np.nan) # 2MASS magnitudes j_mag = table.get_field_by_id('jmag').value h_mag = table.get_field_by_id('hmag').value ks_mag = table.get_field_by_id('ksmag').value j_mag = j_mag.decode('utf-8') h_mag = h_mag.decode('utf-8') ks_mag = ks_mag.decode('utf-8') if j_mag == '': j_mag = np.nan else: j_mag = float(j_mag) if h_mag == '': h_mag = np.nan else: h_mag = float(h_mag) if ks_mag == '': ks_mag = np.nan else: ks_mag = float(ks_mag) name = table.get_field_by_id('name').value name = name.decode('utf-8') twomass_id = table.get_field_by_id('name2m').value twomass_id = twomass_id.decode('utf-8') try: sptype = table.get_field_by_id('nirspty').value sptype = sptype.decode('utf-8') except KeyError: try: sptype = table.get_field_by_id('optspty').value sptype = sptype.decode('utf-8') except KeyError: sptype = 'None' sptype = data_util.update_sptype(np.array([sptype]))[0].strip() h_flux, _ = h_twomass.magnitude_to_flux(h_mag, error=None, zp_flux=h_zp) phot = h_twomass.spectrum_to_flux(wavelength, flux) # Normalized units flux *= h_flux / phot[0] # (W m-2 um-1) spdata = np.vstack([wavelength, flux, error]) # simbad_id, distance = query_util.get_distance(f'2MASS {twomass_id}') simbad_id = query_util.get_simbad(f'2MASS {twomass_id}') if simbad_id is not None: simbad_id = simbad_id.decode('utf-8') dist_select = distance_data.loc[distance_data['object'] == simbad_id] if not dist_select.empty: distance = (dist_select['distance'], dist_select['distance_error']) else: distance = (np.nan, np.nan) else: distance = (np.nan, np.nan) if sptype[0] in ['M', 'L', 'T'] and len(sptype) == 2: print_message = f'Adding SpeX Prism Spectral Library... {name}' print(f'\r{print_message:<72}', end='') dset = database.create_dataset(f'spectra/spex/{name}', data=spdata) dset.attrs['name'] = str(name).encode() dset.attrs['sptype'] = str(sptype).encode() dset.attrs['simbad'] = str(simbad_id).encode() dset.attrs['2MASS/2MASS.J'] = j_mag dset.attrs['2MASS/2MASS.H'] = h_mag dset.attrs['2MASS/2MASS.Ks'] = ks_mag dset.attrs['distance'] = distance[0] # (pc) dset.attrs['distance_error'] = distance[1] # (pc) print_message = 'Adding SpeX Prism Spectral Library... [DONE]' print(f'\r{print_message:<72}') database.close()
def add_irtf(input_path: str, database: h5py._hl.files.File, sptypes: Optional[List[str]] = None) -> None: """ Function for adding the IRTF Spectral Library to the database. Parameters ---------- input_path : str Path of the data folder. database : h5py._hl.files.File Database. sptypes : list(str, ), None List with the spectral types ('F', 'G', 'K', 'M', 'L', 'T'). All spectral types are included if set to ``None``. Returns ------- NoneType None """ if sptypes is None: sptypes = ['F', 'G', 'K', 'M', 'L', 'T'] distance_url = 'https://people.phys.ethz.ch/~stolkert/species/distance.dat' distance_file = os.path.join(input_path, 'distance.dat') if not os.path.isfile(distance_file): urllib.request.urlretrieve(distance_url, distance_file) distance_data = pd.pandas.read_csv(distance_file, usecols=[0, 3, 4], names=['object', 'distance', 'distance_error'], delimiter=',', dtype={'object': str, 'distance': float, 'distance_error': float}) datadir = os.path.join(input_path, 'irtf') if not os.path.exists(datadir): os.makedirs(datadir) data_file = {'F': os.path.join(input_path, 'irtf/F_fits_091201.tar'), 'G': os.path.join(input_path, 'irtf/G_fits_091201.tar'), 'K': os.path.join(input_path, 'irtf/K_fits_091201.tar'), 'M': os.path.join(input_path, 'irtf/M_fits_091201.tar'), 'L': os.path.join(input_path, 'irtf/L_fits_091201.tar'), 'T': os.path.join(input_path, 'irtf/T_fits_091201.tar')} data_folder = {'F': os.path.join(input_path, 'irtf/F_fits_091201'), 'G': os.path.join(input_path, 'irtf/G_fits_091201'), 'K': os.path.join(input_path, 'irtf/K_fits_091201'), 'M': os.path.join(input_path, 'irtf/M_fits_091201'), 'L': os.path.join(input_path, 'irtf/L_fits_091201'), 'T': os.path.join(input_path, 'irtf/T_fits_091201')} data_type = {'F': 'F stars (4.4 MB)', 'G': 'G stars (5.6 MB)', 'K': 'K stars (5.5 MB)', 'M': 'M stars (7.5 MB)', 'L': 'L dwarfs (850 kB)', 'T': 'T dwarfs (100 kB)'} url_root = 'http://irtfweb.ifa.hawaii.edu/~spex/IRTF_Spectral_Library/Data/' url = {'F': url_root+'F_fits_091201.tar', 'G': url_root+'G_fits_091201.tar', 'K': url_root+'K_fits_091201.tar', 'M': url_root+'M_fits_091201.tar', 'L': url_root+'L_fits_091201.tar', 'T': url_root+'T_fits_091201.tar'} for item in sptypes: if not os.path.isfile(data_file[item]): print(f'Downloading IRTF Spectral Library - {data_type[item]}...', end='', flush=True) urllib.request.urlretrieve(url[item], data_file[item]) print(' [DONE]') print('Unpacking IRTF Spectral Library...', end='', flush=True) for item in sptypes: tar = tarfile.open(data_file[item]) tar.extractall(path=datadir) tar.close() print(' [DONE]') database.create_group('spectra/irtf') for item in sptypes: for root, _, files in os.walk(data_folder[item]): for _, filename in enumerate(files): if filename[-9:] != '_ext.fits': fitsfile = os.path.join(root, filename) spdata, header = fits.getdata(fitsfile, header=True) name = header['OBJECT'] sptype = header['SPTYPE'] if name[-2:] == 'AB': name = name[:-2] elif name[-3:] == 'ABC': name = name[:-3] spt_split = sptype.split() if item in ['L', 'T'] or spt_split[1][0] == 'V': print_message = f'Adding IRTF Spectral Library... {name}' print(f'\r{print_message:<70}', end='') simbad_id = query_util.get_simbad(name) if simbad_id is not None: simbad_id = simbad_id.decode('utf-8') dist_select = distance_data.loc[distance_data['object'] == simbad_id] if not dist_select.empty: distance = (dist_select['distance'], dist_select['distance_error']) else: simbad_id, distance = query_util.get_distance(name) else: distance = (np.nan, np.nan) sptype = data_util.update_sptype(np.array([sptype]))[0] dset = database.create_dataset(f'spectra/irtf/{name}', data=spdata) dset.attrs['name'] = str(name).encode() dset.attrs['sptype'] = str(sptype).encode() dset.attrs['simbad'] = str(simbad_id).encode() dset.attrs['distance'] = distance[0] dset.attrs['distance_error'] = distance[1] print_message = 'Adding IRTF Spectral Library... [DONE]' print(f'\r{print_message:<70}') database.close()
def add_spex(input_path: str, database: h5py._hl.files.File) -> None: """ Function for adding the SpeX Prism Spectral Library to the database. Parameters ---------- input_path : str Path of the data folder. database : h5py._hl.files.File The HDF5 database. Returns ------- NoneType None """ parallax_url = "https://home.strw.leidenuniv.nl/~stolker/species/parallax.dat" parallax_file = os.path.join(input_path, "parallax.dat") if not os.path.isfile(parallax_file): urllib.request.urlretrieve(parallax_url, parallax_file) parallax_data = pd.pandas.read_csv( parallax_file, usecols=[0, 1, 2], names=["object", "parallax", "parallax_error"], delimiter=",", dtype={"object": str, "parallax": float, "parallax_error": float}, ) database.create_group("spectra/spex") data_path = os.path.join(input_path, "spex") if not os.path.exists(data_path): os.makedirs(data_path) url_all = "http://svo2.cab.inta-csic.es/vocats/v2/spex/cs.php?" \ "RA=180.000000&DEC=0.000000&SR=180.000000&VERB=2" xml_file_spex = os.path.join(data_path, "spex.xml") if not os.path.isfile(xml_file_spex): urllib.request.urlretrieve(url_all, xml_file_spex) table = parse_single_table(xml_file_spex) # name = table.array['name'] twomass = table.array["name2m"] url = table.array["access_url"] unique_id = [] print_message = "" for i, item in enumerate(url): if twomass[i] not in unique_id: if isinstance(twomass[i], str): xml_file_1 = os.path.join(data_path, twomass[i] + ".xml") else: # Use decode for backward compatibility xml_file_1 = os.path.join( data_path, twomass[i].decode("utf-8") + ".xml" ) if not os.path.isfile(xml_file_1): if isinstance(item, str): urllib.request.urlretrieve(item, xml_file_1) else: urllib.request.urlretrieve(item.decode("utf-8"), xml_file_1) table = parse_single_table(xml_file_1) name = table.array["ID"] url = table.array["access_url"] if isinstance(name[0], str): name = name[0] else: name = name[0].decode("utf-8") empty_message = len(print_message) * " " print(f"\r{empty_message}", end="") print_message = f"Downloading SpeX Prism Spectral Library... {name}" print(f"\r{print_message}", end="") xml_file_2 = os.path.join(data_path, f"spex_{name}.xml") if not os.path.isfile(xml_file_2): if isinstance(url[0], str): urllib.request.urlretrieve(url[0], xml_file_2) else: urllib.request.urlretrieve(url[0].decode("utf-8"), xml_file_2) unique_id.append(twomass[i]) empty_message = len(print_message) * " " print(f"\r{empty_message}", end="") print_message = "Downloading SpeX Prism Spectral Library... [DONE]" print(f"\r{print_message}") h_twomass = photometry.SyntheticPhotometry("2MASS/2MASS.H") # 2MASS H band zero point for 0 mag (Cogen et al. 2003) h_zp = 1.133e-9 # (W m-2 um-1) for votable in os.listdir(data_path): if votable.startswith("spex_") and votable.endswith(".xml"): xml_file = os.path.join(data_path, votable) table = parse_single_table(xml_file) wavelength = table.array["wavelength"] # (Angstrom) flux = table.array["flux"] # Normalized units wavelength = np.array(wavelength * 1e-4) # (um) flux = np.array(flux) # (a.u.) error = np.full(flux.size, np.nan) # 2MASS magnitudes j_mag = table.get_field_by_id("jmag").value h_mag = table.get_field_by_id("hmag").value ks_mag = table.get_field_by_id("ksmag").value if not isinstance(j_mag, str): j_mag = j_mag.decode("utf-8") if not isinstance(h_mag, str): h_mag = h_mag.decode("utf-8") if not isinstance(ks_mag, str): ks_mag = ks_mag.decode("utf-8") if j_mag == "": j_mag = np.nan else: j_mag = float(j_mag) if h_mag == "": h_mag = np.nan else: h_mag = float(h_mag) if ks_mag == "": ks_mag = np.nan else: ks_mag = float(ks_mag) name = table.get_field_by_id("name").value if not isinstance(name, str): name = name.decode("utf-8") twomass_id = table.get_field_by_id("name2m").value if not isinstance(twomass_id, str): twomass_id = twomass_id.decode("utf-8") # Optical spectral type try: sptype_opt = table.get_field_by_id("optspty").value if not isinstance(sptype_opt, str): sptype_opt = sptype_opt.decode("utf-8") sptype_opt = data_util.update_sptype(np.array([sptype_opt]))[0] except KeyError: sptype_opt = None # Near-infrared spectral type try: sptype_nir = table.get_field_by_id("nirspty").value if not isinstance(sptype_nir, str): sptype_nir = sptype_nir.decode("utf-8") sptype_nir = data_util.update_sptype(np.array([sptype_nir]))[0] except KeyError: sptype_nir = None h_flux, _ = h_twomass.magnitude_to_flux(h_mag, error=None, zp_flux=h_zp) phot = h_twomass.spectrum_to_flux(wavelength, flux) # Normalized units flux *= h_flux / phot[0] # (W m-2 um-1) spdata = np.column_stack([wavelength, flux, error]) simbad_id = query_util.get_simbad(f"2MASS {twomass_id}") if simbad_id is not None: if not isinstance(simbad_id, str): simbad_id = simbad_id.decode("utf-8") par_select = parallax_data[parallax_data["object"] == simbad_id] if not par_select.empty: parallax = ( par_select["parallax"].values[0], par_select["parallax_error"].values[0], ) else: parallax = (np.nan, np.nan) else: parallax = (np.nan, np.nan) print_message = f"Adding spectra... {name}" print(f"\r{print_message:<72}", end="") dset = database.create_dataset(f"spectra/spex/{name}", data=spdata) dset.attrs["name"] = str(name).encode() if sptype_opt is not None: dset.attrs["sptype"] = str(sptype_opt).encode() elif sptype_nir is not None: dset.attrs["sptype"] = str(sptype_nir).encode() else: dset.attrs["sptype"] = str("None").encode() dset.attrs["simbad"] = str(simbad_id).encode() dset.attrs["2MASS/2MASS.J"] = j_mag dset.attrs["2MASS/2MASS.H"] = h_mag dset.attrs["2MASS/2MASS.Ks"] = ks_mag dset.attrs["parallax"] = parallax[0] # (mas) dset.attrs["parallax_error"] = parallax[1] # (mas) print_message = "Adding spectra... [DONE]" print(f"\r{print_message:<72}") database.close()
def add_irtf(input_path: str, database: h5py._hl.files.File, sptypes: Optional[List[str]] = None) -> None: """ Function for adding the IRTF Spectral Library to the database. Parameters ---------- input_path : str Path of the data folder. database : h5py._hl.files.File Database. sptypes : list(str), None List with the spectral types ('F', 'G', 'K', 'M', 'L', 'T'). All spectral types are included if set to ``None``. Returns ------- NoneType None """ if sptypes is None: sptypes = ["F", "G", "K", "M", "L", "T"] parallax_url = "https://home.strw.leidenuniv.nl/~stolker/species/parallax.dat" parallax_file = os.path.join(input_path, "parallax.dat") if not os.path.isfile(parallax_file): urllib.request.urlretrieve(parallax_url, parallax_file) parallax_data = pd.pandas.read_csv( parallax_file, usecols=[0, 1, 2], names=["object", "parallax", "parallax_error"], delimiter=",", dtype={ "object": str, "parallax": float, "parallax_error": float }, ) datadir = os.path.join(input_path, "irtf") if not os.path.exists(datadir): os.makedirs(datadir) data_file = { "F": os.path.join(input_path, "irtf/F_fits_091201.tar"), "G": os.path.join(input_path, "irtf/G_fits_091201.tar"), "K": os.path.join(input_path, "irtf/K_fits_091201.tar"), "M": os.path.join(input_path, "irtf/M_fits_091201.tar"), "L": os.path.join(input_path, "irtf/L_fits_091201.tar"), "T": os.path.join(input_path, "irtf/T_fits_091201.tar"), } data_folder = { "F": os.path.join(input_path, "irtf/F_fits_091201"), "G": os.path.join(input_path, "irtf/G_fits_091201"), "K": os.path.join(input_path, "irtf/K_fits_091201"), "M": os.path.join(input_path, "irtf/M_fits_091201"), "L": os.path.join(input_path, "irtf/L_fits_091201"), "T": os.path.join(input_path, "irtf/T_fits_091201"), } data_type = { "F": "F stars (4.4 MB)", "G": "G stars (5.6 MB)", "K": "K stars (5.5 MB)", "M": "M stars (7.5 MB)", "L": "L dwarfs (850 kB)", "T": "T dwarfs (100 kB)", } url_root = "http://irtfweb.ifa.hawaii.edu/~spex/IRTF_Spectral_Library/Data/" url = { "F": url_root + "F_fits_091201.tar", "G": url_root + "G_fits_091201.tar", "K": url_root + "K_fits_091201.tar", "M": url_root + "M_fits_091201.tar", "L": url_root + "L_fits_091201.tar", "T": url_root + "T_fits_091201.tar", } for item in sptypes: if not os.path.isfile(data_file[item]): print( f"Downloading IRTF Spectral Library - {data_type[item]}...", end="", flush=True, ) urllib.request.urlretrieve(url[item], data_file[item]) print(" [DONE]") print("Unpacking IRTF Spectral Library...", end="", flush=True) for item in sptypes: tar = tarfile.open(data_file[item]) tar.extractall(path=datadir) tar.close() print(" [DONE]") database.create_group("spectra/irtf") print_message = "" for item in sptypes: for root, _, files in os.walk(data_folder[item]): for _, filename in enumerate(files): if filename[-9:] != "_ext.fits": fitsfile = os.path.join(root, filename) spdata, header = fits.getdata(fitsfile, header=True) spdata = np.transpose(spdata) name = header["OBJECT"] sptype = header["SPTYPE"] if name[-2:] == "AB": name = name[:-2] elif name[-3:] == "ABC": name = name[:-3] spt_split = sptype.split() if item in ["L", "T"] or spt_split[1][0] == "V": empty_message = len(print_message) * " " print(f"\r{empty_message}", end="") print_message = f"Adding spectra... {name}" print(f"\r{print_message}", end="") simbad_id = query_util.get_simbad(name) if simbad_id is not None: # For backward compatibility if not isinstance(simbad_id, str): simbad_id = simbad_id.decode("utf-8") par_select = parallax_data[parallax_data["object"] == simbad_id] if not par_select.empty: parallax = ( par_select["parallax"], par_select["parallax_error"], ) else: simbad_id, parallax = query_util.get_parallax( name) else: parallax = (np.nan, np.nan) sptype = data_util.update_sptype(np.array([sptype]))[0] dset = database.create_dataset(f"spectra/irtf/{name}", data=spdata) dset.attrs["name"] = str(name).encode() dset.attrs["sptype"] = str(sptype).encode() dset.attrs["simbad"] = str(simbad_id).encode() dset.attrs["parallax"] = parallax[0] dset.attrs["parallax_error"] = parallax[1] empty_message = len(print_message) * " " print(f"\r{empty_message}", end="") print_message = "Adding spectra... [DONE]" print(f"\r{print_message}") database.close()