# Licensed under a 3-clause BSD style license - see LICENSE.rst
import os
import warnings
import json
import copy
import re
from io import BytesIO
import astropy.units as u
import astropy.coordinates as coord
import astropy.table as tbl
import astropy.utils.data as aud
from collections import OrderedDict
import astropy.io.votable as votable
from astropy.io import ascii, fits
from pyvo import registry
from ..query import BaseQuery
from ..utils import commons
from ..utils import async_to_sync
from ..utils import schema
from . import conf
from ..exceptions import TableParseError, EmptyResponseError
__all__ = ['Vizier', 'VizierClass']
[docs]
@async_to_sync
class VizierClass(BaseQuery):
_str_schema = schema.Schema(str)
_schema_columns = schema.Schema([_str_schema],
error="columns must be a list of strings")
_schema_ucd = schema.Schema(_str_schema, error="ucd must be string")
_schema_column_filters = schema.Schema(
{schema.Optional(_str_schema): _str_schema},
error=("column_filters must be a dictionary where both keys "
"and values are strings"))
_schema_catalog = schema.Schema(
schema.Or([_str_schema], _str_schema, None),
error="catalog must be a list of strings or a single string")
def __init__(self, *, columns=["*"], column_filters={}, catalog=None,
keywords=None, ucd="", timeout=conf.timeout,
vizier_server=conf.server, row_limit=conf.row_limit):
"""
Parameters
----------
columns : list
List of strings
column_filters : dict
catalog : str or None
keywords : str or None
ucd : string
"Unified Content Description" column descriptions. Specifying
these will select only catalogs that have columns matching the
column descriptions defined on the Vizier web pages.
See https://vizier.cds.unistra.fr/vizier/vizHelp/1.htx#ucd and
https://vizier.cds.unistra.fr/vizier/doc/UCD/
vizier_server : string
Name of the VizieR mirror to use.
(This parameter's default is set from a configuration object.)
timeout : number
timeout for connecting to server
(This parameter's default is set from a configuration object.)
row_limit : int
Maximum number of rows that will be fetched from the result
(set to -1 for unlimited).
(This parameter's default is set from a configuration object.)
"""
super().__init__()
self.columns = columns
self.column_filters = column_filters
self.catalog = catalog
self._keywords = None
self.ucd = ucd
if keywords:
self.keywords = keywords
self.TIMEOUT = timeout
self.VIZIER_SERVER = vizier_server
self.ROW_LIMIT = row_limit
@property
def columns(self):
""" Columns to include. The special keyword 'all' will return ALL
columns from ALL retrieved tables. """
# columns need to be immutable but still need to be a list
return list(tuple(self._columns))
@columns.setter
def columns(self, values):
self._columns = VizierClass._schema_columns.validate(values)
@property
def column_filters(self):
"""
Filters to run on the individual columns. See the Vizier website
for details.
"""
return self._column_filters
@column_filters.setter
def column_filters(self, values):
self._column_filters = (
VizierClass._schema_column_filters.validate(values))
@property
def catalog(self):
"""
The default catalog to search. If left empty, will search all
catalogs.
"""
return self._catalog
@catalog.setter
def catalog(self, values):
self._catalog = VizierClass._schema_catalog.validate(values)
@property
def ucd(self):
"""
UCD criteria: see https://vizier.cds.unistra.fr/vizier/vizHelp/1.htx#ucd
Examples
--------
>>> Vizier.ucd = '(spect.dopplerVeloc*|phys.veloc*)'
"""
return self._ucd
@ucd.setter
def ucd(self, values):
self._ucd = VizierClass._schema_ucd.validate(values)
def _server_to_url(self, return_type='votable'):
"""
Not generally meant to be modified, but there are different valid
return types supported by Vizier, listed here:
https://vizier.cds.unistra.fr/doc/asu-summary.htx
HTML: VizieR
votable: votable
tab-separated-values: asu-tsv
FITS ascii table: asu-fits
FITS binary table: asu-binfits
plain text: asu-txt
"""
# Only votable is supported now, but in case we try to support
# something in the future we should disallow invalid ones.
assert return_type in ('votable', 'asu-tsv', 'asu-fits',
'asu-binfits', 'asu-txt')
if return_type in ('asu-txt',):
# I had a look at the format of these "tables" and... they just
# aren't. They're quasi-fixed-width without schema. I think they
# follow the general philosophy of "consistency is overrated"
# The CDS reader chokes on it.
raise TypeError("asu-txt is not and cannot be supported: the "
"returned tables are not and cannot be made "
"parseable.")
return "https://" + self.VIZIER_SERVER + "/viz-bin/" + return_type
@property
def keywords(self):
"""The set of keywords to filter the Vizier search"""
return self._keywords
@keywords.setter
def keywords(self, value):
self._keywords = VizierKeyword(value)
@keywords.deleter
def keywords(self):
self._keywords = None
[docs]
def find_catalogs(self, keywords, *, include_obsolete=False, verbose=False,
max_catalogs=None, return_type='votable'):
"""
Search Vizier for catalogs based on a set of keywords, e.g. author name
Parameters
----------
keywords : list or string
List of keywords, or space-separated set of keywords.
From `Vizier <https://vizier.cds.unistra.fr/doc/asu-summary.htx>`_:
"names or words of title of catalog. The words are and'ed, i.e.
only the catalogues characterized by all the words are selected."
include_obsolete : bool, optional
If set to True, catalogs marked obsolete will also be returned.
max_catalogs : int or None
The maximum number of catalogs to return. If ``None``, all
catalogs will be returned.
Returns
-------
resource_dict : dict
Dictionary of the "Resource" name and the VOTable resource object.
"Resources" are generally publications; one publication may contain
many tables.
Examples
--------
>>> from astroquery.vizier import Vizier
>>> catalog_list = Vizier.find_catalogs('Kang W51') # doctest: +REMOTE_DATA +IGNORE_WARNINGS
>>> catalog_list # doctest: +REMOTE_DATA +IGNORE_OUTPUT
OrderedDict([('J/ApJ/684/1143', </>), ('J/ApJ/736/87', </>) ... ])
>>> print({k:v.description for k,v in catalog_list.items()}) # doctest: +REMOTE_DATA +IGNORE_OUTPUT
{'J/ApJ/684/1143': 'BHB candidates in the Milky Way (Xue+, 2008)',
'J/ApJ/736/87': 'Abundances in G-type stars with exoplanets (Kang+, 2011)',
'J/ApJ/738/79': "SDSS-DR8 BHB stars in the Milky Way's halo (Xue+, 2011)",
'J/ApJ/760/12': 'LIGO/Virgo gravitational-wave (GW) bursts with GRBs (Abadie+, 2012)',
...}
"""
if isinstance(keywords, list):
keywords = " ".join(keywords)
data_payload = {'-words': keywords, '-meta.all': 1}
data_payload['-ucd'] = self.ucd
if max_catalogs is not None:
data_payload['-meta.max'] = max_catalogs
response = self._request(
method='POST', url=self._server_to_url(return_type=return_type),
data=data_payload, timeout=self.TIMEOUT)
if 'STOP, Max. number of RESOURCE reached' in response.text:
raise ValueError("Maximum number of catalogs exceeded. Try "
"setting max_catalogs to a large number and"
" try again")
result = self._parse_result(response, verbose=verbose,
get_catalog_names=True)
# Filter out the obsolete catalogs, unless requested
if include_obsolete is False:
for key in list(result):
for info in result[key].infos:
if (info.name == 'status') and (info.value == 'obsolete'):
del result[key]
return result
[docs]
def get_catalogs_async(self, catalog, *, verbose=False, return_type='votable',
get_query_payload=False):
"""
Query the Vizier service for a specific catalog
Parameters
----------
catalog : str, Resource, or list, optional
The catalog(s) that will be retrieved
Returns
-------
response : `~requests.Response`
Returned if asynchronous method used
"""
if not isinstance(catalog, (str, votable.tree.Resource)):
catalog = list(catalog)
data_payload = self._args_to_payload(catalog=catalog)
if get_query_payload:
return data_payload
response = self._request(
method='POST', url=self._server_to_url(return_type=return_type),
data=data_payload, timeout=self.TIMEOUT)
return response
[docs]
def query_object_async(self, object_name, *, catalog=None, radius=None,
coordinate_frame=None, get_query_payload=False,
return_type='votable', cache=True):
"""
Serves the same purpose as `query_object` but only
returns the HTTP response rather than the parsed result.
Parameters
----------
object_name : str
The name of the identifier.
catalog : str or list, optional
The catalog(s) which must be searched for this identifier.
If not specified, all matching catalogs will be searched.
radius : `~astropy.units.Quantity` or None
A degree-equivalent radius (optional).
coordinate_system : str or None
If the object name is given as a coordinate, you *should* use
`~astroquery.vizier.VizierClass.query_region`, but you can
specify a coordinate frame here instead (today, J2000, B1975,
B1950, B1900, B1875, B1855, Galactic, Supergal., Ecl.J2000, )
cache : bool
Defaults to True. If set overrides global caching behavior.
See :ref:`caching documentation <astroquery_cache>`.
Returns
-------
response : `~requests.Response`
The response of the HTTP request.
"""
catalog = VizierClass._schema_catalog.validate(catalog)
if radius is None:
center = {'-c': object_name}
else:
radius_arcmin = radius.to(u.arcmin).value
cframe = (coordinate_frame if coordinate_frame in
["today", "J2000", "B1975", "B1950", "B1900", "B1875",
"B1855", "Galactic", "Supergal.", "Ecl.J2000"]
else 'J2000')
center = {'-c': object_name, '-c.u': 'arcmin', '-c.geom': 'r',
'-c.r': radius_arcmin, '-c.eq': cframe}
data_payload = self._args_to_payload(
center=center,
catalog=catalog)
if get_query_payload:
return data_payload
response = self._request(
method='POST', url=self._server_to_url(return_type=return_type),
data=data_payload, timeout=self.TIMEOUT, cache=cache)
return response
[docs]
def query_region_async(self, coordinates, *, radius=None, inner_radius=None,
width=None, height=None, catalog=None,
get_query_payload=False, cache=True,
return_type='votable', column_filters={},
frame='fk5'):
"""
Serves the same purpose as `query_region` but only
returns the HTTP response rather than the parsed result.
At least one of ``radius`` or ``width`` must be specified.
Parameters
----------
coordinates : str, `astropy.coordinates` object, or `~astropy.table.Table`
The target around which to search. It may be specified as a
string in which case it is resolved using online services or as
the appropriate `astropy.coordinates` object. ICRS coordinates
may also be entered as a string. If a table is used, each of
its rows will be queried, as long as it contains two columns
named ``_RAJ2000`` and ``_DEJ2000`` with proper angular units.
radius : convertible to `~astropy.coordinates.Angle`
The radius of the circular region to query.
inner_radius : convertible to `~astropy.coordinates.Angle`
When set in addition to ``radius``, the queried region becomes
annular, with outer radius ``radius`` and inner radius
``inner_radius``.
width : convertible to `~astropy.coordinates.Angle`
The width of the square region to query.
height : convertible to `~astropy.coordinates.Angle`
When set in addition to ``width``, the queried region becomes
rectangular, with the specified ``width`` and ``height``.
catalog : str or list, optional
The catalog(s) which must be searched for this identifier.
If not specified, all matching catalogs will be searched.
column_filters: dict, optional
Constraints on columns of the result. The dictionary contains
the column name as keys, and the constraints as values.
frame : str, optional
The frame to use for the request. It should be 'fk5', 'icrs',
or 'galactic'. This choice influences the the orientation of
box requests.
cache : bool
Defaults to True. If set overrides global caching behavior.
See :ref:`caching documentation <astroquery_cache>`.
Returns
-------
response : `requests.Response`
The response of the HTTP request.
"""
if frame not in ('galactic', 'fk5', 'icrs'):
raise ValueError("Only the 'galactic', 'icrs', and 'fk5' frames are supported by VizieR")
catalog = VizierClass._schema_catalog.validate(catalog)
center = {}
columns = []
# Process coordinates
if isinstance(coordinates, (commons.CoordClasses, str)):
target = commons.parse_coordinates(coordinates).transform_to(frame)
if not target.isscalar:
center["-c"] = []
for pos in target:
if frame == 'galactic':
glon_deg = pos.l.to_string(unit="deg", decimal=True, precision=8)
glat_deg = pos.b.to_string(unit="deg", decimal=True, precision=8,
alwayssign=True)
center["-c"] += ["G{}{}".format(glon_deg, glat_deg)]
else:
ra_deg = pos.ra.to_string(unit="deg", decimal=True, precision=8)
dec_deg = pos.dec.to_string(unit="deg", decimal=True,
precision=8, alwayssign=True)
center["-c"] += ["{}{}".format(ra_deg, dec_deg)]
columns += ["_q"] # Always request reference to input table
else:
if frame == 'galactic':
glon = target.l.to_string(unit='deg', decimal=True, precision=8)
glat = target.b.to_string(unit="deg", decimal=True, precision=8,
alwayssign=True)
center["-c"] = "G{glon}{glat}".format(glon=glon, glat=glat)
else:
ra = target.ra.to_string(unit='deg', decimal=True, precision=8)
dec = target.dec.to_string(unit="deg", decimal=True, precision=8,
alwayssign=True)
center["-c"] = "{ra}{dec}".format(ra=ra, dec=dec)
elif isinstance(coordinates, tbl.Table):
if (("_RAJ2000" in coordinates.keys()) and ("_DEJ2000" in
coordinates.keys())):
center["-c"] = []
sky_coord = coord.SkyCoord(coordinates["_RAJ2000"],
coordinates["_DEJ2000"],
unit=(coordinates["_RAJ2000"].unit,
coordinates["_DEJ2000"].unit))
for (ra, dec) in zip(sky_coord.ra, sky_coord.dec):
ra_deg = ra.to_string(unit="deg", decimal=True,
precision=8)
dec_deg = dec.to_string(unit="deg", decimal=True,
precision=8, alwayssign=True)
center["-c"] += ["{}{}".format(ra_deg, dec_deg)]
columns += ["_q"] # Always request reference to input table
else:
raise ValueError("Table must contain '_RAJ2000' and "
"'_DEJ2000' columns!")
else:
raise TypeError("Coordinates must be one of: string, astropy "
"coordinates, or table containing coordinates!")
if radius is not None:
if inner_radius is None:
_, unit_str, o_radius = _parse_angle(radius)
center["-c.r" + unit_str] = str(o_radius)
else:
unit, unit_str, i_radius = _parse_angle(inner_radius)
o_radius = coord.Angle(radius).to_value(unit)
center["-c.r" + unit_str] = f"{i_radius},{o_radius}"
elif width is not None:
unit, unit_str, w_box = _parse_angle(width)
h_box = w_box if height is None else coord.Angle(height).to_value(unit)
center["-c.b" + unit_str] = f"{w_box}x{h_box}"
else:
raise Exception(
"At least one of radius, width/height must be specified")
# Prepare payload
data_payload = self._args_to_payload(center=center, columns=columns,
catalog=catalog, column_filters=column_filters)
if get_query_payload:
return data_payload
response = self._request(
method='POST', url=self._server_to_url(return_type=return_type),
data=data_payload, timeout=self.TIMEOUT, cache=cache)
return response
[docs]
def query_constraints_async(self, *, catalog=None, return_type='votable',
cache=True, get_query_payload=False,
**kwargs):
"""
Send a query to Vizier in which you specify constraints with
keyword/value pairs.
See `the vizier constraints page
<https://vizier.cds.unistra.fr/vizier/vizHelp/syntax.htx>`_ for details.
Parameters
----------
catalog : str or list, optional
The catalog(s) which must be searched for this identifier.
If not specified, all matching catalogs will be searched.
kwargs : dict
Any key/value pairs besides "catalog" will be parsed
as additional column filters.
cache : bool
Defaults to True. If set overrides global caching behavior.
See :ref:`caching documentation <astroquery_cache>`.
Returns
-------
response : `requests.Response`
The response of the HTTP request.
Examples
--------
>>> from astroquery.vizier import Vizier
>>> # note that glon/glat constraints here *must* be floats
>>> result = Vizier(row_limit=3).query_constraints(catalog='J/ApJ/723/492/table1',
... GLON='>49.0 & <51.0', GLAT='<0.0') # doctest: +REMOTE_DATA
>>> result[result.keys()[0]].pprint() # doctest: +REMOTE_DATA
GRSMC GLON GLAT VLSR DelV ... alpha Note RD09 _RA.icrs _DE.icrs
deg deg km / s km / s ... deg deg
------------- ------ ------ ------ ------ ... ------ ---- ---- -------- --------
G049.49-00.41 49.49 -0.41 56.90 9.12 ... 0.73 i RD09 290.9536 14.4992
G049.39-00.26 49.39 -0.26 50.94 3.51 ... 0.16 i RD09 290.7682 14.4819
G049.44-00.06 49.44 -0.06 62.00 3.67 ... 0.17 i RD09 290.6104 14.6203
"""
catalog = VizierClass._schema_catalog.validate(catalog)
data_payload = self._args_to_payload(
catalog=catalog,
column_filters=kwargs,
center={'-c.rd': 180})
if get_query_payload:
return data_payload
response = self._request(
method='POST', url=self._server_to_url(return_type=return_type),
data=data_payload, timeout=self.TIMEOUT, cache=cache)
return response
def _args_to_payload(self, *args, **kwargs):
"""
accepts the arguments for different query functions and
builds a script suitable for the Vizier votable CGI.
"""
body = OrderedDict()
center = kwargs.get('center')
# process: catalog
catalog = kwargs.get('catalog') or self.catalog
if catalog is not None:
if isinstance(catalog, str):
body['-source'] = catalog
elif isinstance(catalog, list):
catalog = [item.name if hasattr(item, 'name') else item
for item in catalog]
body['-source'] = ",".join(catalog)
elif hasattr(catalog, 'name'):
# this is probably a votable Resource, but no harm in duck-typing on `name`
body['-source'] = catalog.name
else:
raise TypeError("Catalog must be specified as list, string, or Resource")
# take row_limit in account
body['-out.max'] = self.ROW_LIMIT
# process: columns
columns = kwargs.get('columns', copy.copy(self.columns))
if columns is not None:
columns = self.columns + columns
# filter columns to _unique_ columns, preserving order in python >3.6
# note that "set" does not preserve order, but dict.keys does
columns = list(dict.fromkeys(columns, ).keys())
# special keywords need to be treated separately
# keyword names that can mean 'all'
alls = ['all', '**']
if any(x in columns for x in alls):
columns_all = True
for x in alls:
if x in columns:
columns.remove(x)
body['-out.all'] = None
else:
columns_all = False
# process: columns - identify sorting requests
columns_out = []
sorts_out = []
for column in columns:
if column[0] == '+':
columns_out += [column[1:]]
sorts_out += [column[1:]]
elif column[0] == '-':
columns_out += [column[1:]]
sorts_out += [column]
else:
columns_out += [column]
# calculated keyword names that start with an underscore
columns_calc = []
for column in columns_out:
if column[0] == '_':
columns_calc.append(column)
for column in columns_calc:
columns_out.remove(column)
if columns_out and not columns_all:
body['-out'] = ','.join(columns_out)
if columns_calc:
body['-out.add'] = ','.join(columns_calc)
if len(sorts_out) > 0:
body['-sort'] = ','.join(sorts_out)
# process: maximum rows returned
row_limit = kwargs.get('row_limit') or self.ROW_LIMIT
if row_limit < 0:
body["-out.max"] = 'unlimited'
else:
body["-out.max"] = row_limit
# process: column filters
column_filters = self.column_filters.copy()
column_filters.update(kwargs.get('column_filters', {}))
for (key, value) in column_filters.items():
body[key] = value
# process: center
if center is not None:
for (key, value) in center.items():
body[key] = value
# add column metadata: name, unit, UCD1+, and description
body["-out.meta"] = "huUD"
# merge tables when a list is queried against a single catalog
body["-out.form"] = "mini"
# computed position should always be in decimal degrees
body["-oc.form"] = "d"
ucd = kwargs.get('ucd', "") + self.ucd
if ucd:
body['-ucd'] = ucd
# create final script starting with keywords
script = []
if (not isinstance(self.keywords, property)
and self.keywords is not None):
script += [str(self.keywords)]
# add all items that are not lists
for key, val in body.items():
if type(val) is not list:
if val:
script += ["{key}={val}".format(key=key, val=val)]
else:
script += [key]
# add list at the end
for key, val in body.items():
if type(val) is list:
script += ["{key}=<<====AstroqueryList".format(key=key)]
script += val
script += ["====AstroqueryList"]
# merge result
return "\n".join(script)
def _parse_result(self, response, *, get_catalog_names=False, verbose=False,
invalid='warn'):
"""
Parses the HTTP response to create a `~astropy.table.Table`.
Returns the raw result as a string in case of parse errors.
Parameters
----------
response : `requests.Response`
The response of the HTTP POST request
get_catalog_names : bool
(only for VOTABLE queries)
If specified, return only the table names (useful for table
discovery).
invalid : 'warn', 'mask' or 'exception'
(only for VOTABLE queries)
The behavior if a VOTABLE cannot be parsed. The default is
'warn', which will try to parse the table, but if an
exception is raised during parsing, the exception will be
issued as a warning instead and a masked table will be
returned. A value of 'exception' will not catch the
exception, while a value of 'mask' will simply always mask
invalid values.
Returns
-------
table_list : `astroquery.utils.TableList` or str
If there are errors in the parsing, then returns the raw results
as a string.
"""
if response.content[:5] == b'<?xml':
try:
return _parse_vizier_votable(
response.content, verbose=verbose, invalid=invalid,
get_catalog_names=get_catalog_names)
except Exception as ex:
self.response = response
self.table_parse_error = ex
raise TableParseError("Failed to parse VIZIER result! The "
"raw response can be found in "
"self.response, and the error in "
"self.table_parse_error. The attempted "
"parsed result is in "
"self.parsed_result.\n Exception: "
+ str(self.table_parse_error))
elif response.content[:5] == b'#\n# ':
return _parse_vizier_tsvfile(response.content, verbose=verbose)
elif response.content[:6] == b'SIMPLE':
return fits.open(BytesIO(response.content),
ignore_missing_end=True)
@property
def valid_keywords(self):
if not hasattr(self, '_valid_keyword_dict'):
file_name = aud.get_pkg_data_filename(
os.path.join("data", "inverse_dict.json"))
with open(file_name, 'r') as f:
kwd = json.load(f)
self._valid_keyword_types = sorted(kwd.values())
self._valid_keyword_dict = OrderedDict([(k, kwd[k])
for k in sorted(kwd)])
return self._valid_keyword_dict
def _parse_vizier_tsvfile(data, *, verbose=False):
"""
Parse a Vizier-generated list of tsv data tables into a list of astropy
Tables.
Parameters
----------
data : response content bytes
Bytes containing the vizier-formatted list of tables
"""
# http://stackoverflow.com/questions/4664850/find-all-occurrences-of-a-substring-in-python
split_indices = [m.start() for m in re.finditer(b'\n\n#', data)]
# we want to slice out chunks of the file each time
split_limits = zip(split_indices[:-1], split_indices[1:])
tables = [ascii.read(BytesIO(data[a:b]), format='fast_tab', delimiter='\t',
header_start=0, comment="#") for
a, b in split_limits]
return tables
def _parse_vizier_votable(data, *, verbose=False, invalid='warn',
get_catalog_names=False):
"""
Given a votable as string, parse it into dict or tables
"""
if not verbose:
commons.suppress_vo_warnings()
tf = BytesIO(data)
if invalid == 'mask':
vo_tree = votable.parse(tf, verify='warn', invalid='mask')
elif invalid == 'warn':
try:
vo_tree = votable.parse(tf, verify='warn', invalid='exception')
except Exception as ex:
warnings.warn("VOTABLE parsing raised exception: {0}".format(ex))
vo_tree = votable.parse(tf, verify='warn', invalid='mask')
elif invalid == 'exception':
vo_tree = votable.parse(tf, verify='warn', invalid='exception')
else:
raise ValueError("Invalid keyword for 'invalid'. "
"Must be exception, mask, or warn")
if get_catalog_names:
return OrderedDict([(R.name, R) for R in vo_tree.resources])
else:
table_dict = OrderedDict()
for t in vo_tree.iter_tables():
if len(t.array) > 0:
if t.ref is not None:
name = vo_tree.get_table_by_id(t.ref).name
else:
name = t.name
if name not in table_dict.keys():
table_dict[name] = []
table_dict[name] += [t.to_table()]
for name in table_dict.keys():
if len(table_dict[name]) > 1:
table_dict[name] = tbl.vstack(table_dict[name])
else:
table_dict[name] = table_dict[name][0]
return commons.TableList(table_dict)
def _parse_angle(angle):
"""
Returns the Vizier-formatted units and values for box/radius
dimensions in case of region queries.
Parameters
----------
angle : convertible to `astropy.coordinates.Angle`
Returns
-------
(unit, unit_str, value) : tuple
"""
angle = coord.Angle(angle)
if angle.unit is u.arcsec:
return u.arcsec, "s", angle.value
elif angle.unit is u.arcmin:
return u.arcmin, "m", angle.value
else:
return u.deg, "d", angle.to_value(u.deg)
class VizierKeyword(list):
"""Helper class for setting keywords for Vizier queries"""
def __init__(self, keywords):
file_name = aud.get_pkg_data_filename(
os.path.join("data", "inverse_dict.json"))
with open(file_name, 'r') as f:
kwd = json.load(f)
self.keyword_types = sorted(kwd.values())
self.keyword_dict = OrderedDict([(k, kwd[k]) for k in sorted(kwd)])
self._keywords = None
self.keywords = keywords
@property
def keywords(self):
"""
List or string for keyword(s) that must be set for the Vizier
object.
"""
return self._keywords
@keywords.setter
def keywords(self, values):
if isinstance(values, str):
values = list(values)
keys = [key.lower() for key in self.keyword_dict]
values = [val.lower() for val in values]
# warn about unknown keywords
for val in set(values) - set(keys):
warnings.warn("{val} : No such keyword".format(val=val))
valid_keys = [
key for key in self.keyword_dict.keys()
if key.lower() in list(map(str.lower, values))]
# create a dict for each type of keyword
set_keywords = OrderedDict()
for key in self.keyword_dict:
if key in valid_keys:
if self.keyword_dict[key] in set_keywords:
set_keywords[self.keyword_dict[key]].append(key)
else:
set_keywords[self.keyword_dict[key]] = [key]
self._keywords = OrderedDict(
[(k, sorted(set_keywords[k]))
for k in set_keywords])
@keywords.deleter
def keywords(self):
del self._keywords
def __repr__(self):
return "\n".join([x for key in self.keywords
for x in self.get_keyword_str(key)])
def get_keyword_str(self, key):
"""
Helper function that returns the keywords, grouped into appropriate
categories and suitable for the Vizier votable CGI.
Comma-separated is not valid!!!
"""
keyword_name = "-kw." + key
return [keyword_name + "=" + s for s in self.keywords[key]]
Vizier = VizierClass()