def test_djs_laxisgen():
import numpy as np
from astropy.tests.helper import raises
from .. import djs_laxisgen
#
# 1d
#
assert (np.arange(4,dtype='i4') == djs_laxisgen((4,))).all()
#
# 2d
#
l = np.array([[0,0,0,0],[1,1,1,1],[2,2,2,2],[3,3,3,3]],dtype='i4')
assert (l == djs_laxisgen((4,4))).all()
assert (l.T == djs_laxisgen((4,4),iaxis=1)).all()
with raises(ValueError):
foo = djs_laxisgen((4,4),iaxis=2)
#
# 3d
#
l = np.zeros((4,4,4),dtype='i4')
l[1,:,:] = 1
l[2,:,:] = 2
l[3,:,:] = 3
assert (l == djs_laxisgen((4,4,4))).all()
assert (l.swapaxes(0,1) == djs_laxisgen((4,4,4),iaxis=1)).all()
assert (l.swapaxes(0,2) == djs_laxisgen((4,4,4),iaxis=2)).all()
with raises(ValueError):
foo = djs_laxisgen((4,4,4),iaxis=3)
#
# More d
#
with raises(ValueError):
foo = djs_laxisgen((4,4,4,4))
def test_filter_thru():
from .. import filter_thru
import numpy as np
from os.path import dirname, join
from astropy.io import fits
from astropy.tests.helper import raises
# with fits.open(join('pydl','pydlspec2d','spec2d','tests','t','spPlate-4055-55359-0020.fits')) as hdulist:
with fits.open(join(dirname(__file__), "t", "spPlate-4055-55359-0020.fits")) as hdulist:
flux = hdulist[0].data
npix = hdulist[0].header["NAXIS1"]
ntrace = hdulist[0].header["NAXIS2"]
crval1 = hdulist[0].header["COEFF0"]
cd1_1 = hdulist[0].header["COEFF1"]
assert flux.shape == (ntrace, npix)
loglam0 = crval1 + cd1_1 * np.arange(npix, dtype=flux.dtype)
waveimg = 10 ** (np.tile(loglam0, 20).reshape(flux.shape))
assert waveimg.shape == flux.shape
f = filter_thru(flux, waveimg=waveimg)
idl_data_file = join(dirname(__file__), "t", "filter_thru_idl_data.txt")
idl_data = np.loadtxt(idl_data_file, dtype="f", delimiter=",").T
assert f.shape == (20, 5)
assert np.allclose(f, idl_data, atol=1.0e-6)
#
# Test bad input.
#
with raises(ValueError):
f = filter_thru(flux)
with raises(ValueError):
f = filter_thru(flux, waveimg=waveimg, filter_prefix="sdss")
return
def test_filter_thru(self):
fname = get_pkg_data_filename('t/spPlate-4055-55359-0020.fits')
with fits.open(fname) as hdulist:
flux = hdulist[0].data
npix = hdulist[0].header['NAXIS1']
ntrace = hdulist[0].header['NAXIS2']
crval1 = hdulist[0].header['COEFF0']
cd1_1 = hdulist[0].header['COEFF1']
assert flux.shape == (ntrace, npix)
loglam0 = crval1 + cd1_1*np.arange(npix, dtype=flux.dtype)
waveimg = 10**(np.tile(loglam0, 20).reshape(flux.shape))
assert waveimg.shape == flux.shape
f = filter_thru(flux, waveimg=waveimg)
idl_data_file = get_pkg_data_filename('t/filter_thru_idl_data.txt')
idl_data = np.loadtxt(idl_data_file, dtype='f', delimiter=',').T
assert f.shape == (20, 5)
assert np.allclose(f, idl_data, atol=1.0e-6)
#
# Test bad input.
#
with raises(ValueError):
f = filter_thru(flux)
with raises(ValueError):
f = filter_thru(flux, waveimg=waveimg, filter_prefix='sdss')
return
def test_djs_laxisnum(self):
#
# 1d
#
assert (np.zeros((4,), dtype='i4') == djs_laxisnum((4,))).all()
#
# 2d
#
l = np.array([[0, 0, 0, 0],
[1, 1, 1, 1],
[2, 2, 2, 2],
[3, 3, 3, 3]],
dtype='i4')
assert (l == djs_laxisnum((4, 4))).all()
assert (l.T == djs_laxisnum((4, 4), iaxis=1)).all()
with raises(ValueError):
foo = djs_laxisnum((4, 4), iaxis=2)
#
# 3d
#
l = np.zeros((4, 4, 4), dtype='i4')
l[1, :, :] = 1
l[2, :, :] = 2
l[3, :, :] = 3
assert (l == djs_laxisnum((4, 4, 4))).all()
assert (l.swapaxes(0, 1) == djs_laxisnum((4, 4, 4), iaxis=1)).all()
assert (l.swapaxes(0, 2) == djs_laxisnum((4, 4, 4), iaxis=2)).all()
with raises(ValueError):
foo = djs_laxisnum((4, 4, 4), iaxis=3)
#
# More d
#
with raises(ValueError):
foo = djs_laxisnum((4, 4, 4, 4))
def test_rebin(self):
x = np.arange(40)
with raises(ValueError):
r = rebin(x, d=(10, 10))
with raises(ValueError):
r = rebin(x, d=(70, ))
with raises(ValueError):
r = rebin(x, d=(30, ))
x = np.array([[1.0, 2.0], [2.0, 3.0]])
rexpect = np.array([[1.0, 2.0], [1.5, 2.5], [2.0, 3.0], [2.0, 3.0]])
r = rebin(x, d=(4, 2))
assert np.allclose(r, rexpect)
rexpect = np.array([[1.0, 1.5, 2.0, 2.0], [2.0, 2.5, 3.0, 3.0]])
r = rebin(x, d=(2, 4))
assert np.allclose(r, rexpect)
rexpect = np.array([[1.0, 2.0], [1.0, 2.0], [2.0, 3.0], [2.0, 3.0]])
r = rebin(x, d=(4, 2), sample=True)
assert np.allclose(r, rexpect)
rexpect = np.array([[1.0, 1.0, 2.0, 2.0], [2.0, 2.0, 3.0, 3.0]])
r = rebin(x, d=(2, 4), sample=True)
assert np.allclose(r, rexpect)
x = np.arange(10)
rexpect = np.array([0.0, 2.0, 4.0, 6.0, 8.0])
r = rebin(x, d=(5, ), sample=True)
assert np.allclose(r, rexpect)
x = np.array([[1.0, 2.0, 3.0, 4.0], [2.0, 3.0, 4.0, 5.0],
[3.0, 4.0, 5.0, 6.0], [4.0, 5.0, 6.0, 7.0]])
rexpect = np.array([[2.0, 4.0], [4.0, 6.0]])
r = rebin(x, d=(2, 2))
assert np.allclose(r, rexpect)
rexpect = np.array([[1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 4.5],
[3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 6.5]])
r = rebin(x, d=(2, 8))
assert np.allclose(r, rexpect)
def test_filter_thru(self):
fname = get_pkg_data_filename('t/spPlate-4055-55359-0020.fits')
with fits.open(fname) as hdulist:
flux = hdulist[0].data
npix = hdulist[0].header['NAXIS1']
ntrace = hdulist[0].header['NAXIS2']
crval1 = hdulist[0].header['COEFF0']
cd1_1 = hdulist[0].header['COEFF1']
assert flux.shape == (ntrace, npix)
loglam0 = crval1 + cd1_1 * np.arange(npix, dtype=flux.dtype)
waveimg = 10**(np.tile(loglam0, 20).reshape(flux.shape))
assert waveimg.shape == flux.shape
f = filter_thru(flux, waveimg=waveimg)
idl_data_file = get_pkg_data_filename('t/filter_thru_idl_data.txt')
idl_data = np.loadtxt(idl_data_file, dtype='f', delimiter=',').T
assert f.shape == (20, 5)
assert np.allclose(f, idl_data, atol=1.0e-6)
#
# Test bad input.
#
with raises(ValueError):
f = filter_thru(flux)
with raises(ValueError):
f = filter_thru(flux, waveimg=waveimg, filter_prefix='sdss')
return
def test_filter_thru():
from .. import filter_thru
import numpy as np
from os.path import dirname, join
from astropy.io import fits
from astropy.tests.helper import raises
# with fits.open(join('pydl','pydlspec2d','spec2d','tests','t','spPlate-4055-55359-0020.fits')) as hdulist:
with fits.open(join(dirname(__file__), 't',
'spPlate-4055-55359-0020.fits')) as hdulist:
flux = hdulist[0].data
npix = hdulist[0].header['NAXIS1']
ntrace = hdulist[0].header['NAXIS2']
crval1 = hdulist[0].header['COEFF0']
cd1_1 = hdulist[0].header['COEFF1']
assert flux.shape == (ntrace, npix)
loglam0 = crval1 + cd1_1 * np.arange(npix, dtype=flux.dtype)
waveimg = 10**(np.tile(loglam0, 20).reshape(flux.shape))
assert waveimg.shape == flux.shape
f = filter_thru(flux, waveimg=waveimg)
idl_data_file = join(dirname(__file__), 't', 'filter_thru_idl_data.txt')
idl_data = np.loadtxt(idl_data_file, dtype='f', delimiter=',').T
assert f.shape == (20, 5)
assert np.allclose(f, idl_data, atol=1.0e-6)
#
# Test bad input.
#
with raises(ValueError):
f = filter_thru(flux)
with raises(ValueError):
f = filter_thru(flux, waveimg=waveimg, filter_prefix='sdss')
return
def test_sdss_flagval(self):
val = sdss_flagval('TARGET', 'ROSAT_A')
assert val == 2**9
val = sdss_flagval('ANCILLARY_TARGET1', ['BLAZGX', 'ELG', 'BRIGHTGAL'])
assert val == 2310346608843161600
with raises(KeyError):
val = sdss_flagval('TARGET', 'ROSAT_Q')
with raises(KeyError):
val = sdss_flagval('ABADMASK', "ABADFLAG")
def test_sdss_flagval(self):
val = sdss_flagval('TARGET', 'ROSAT_A')
assert val == 2**9
val = sdss_flagval('ANCILLARY_TARGET1', ['BLAZGX', 'ELG', 'BRIGHTGAL'])
assert val == 2310346608843161600
with raises(KeyError):
val = sdss_flagval('TARGET', 'ROSAT_Q')
with raises(KeyError):
val = sdss_flagval('ABADMASK', "ABADFLAG")
def test_nw_cut_to_box(self):
colors = np.random.random((10, 10))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors)
colors = np.random.random((10, 10, 3))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors, origin=(1.0, 1.0))
boxed_colors = nw_cut_to_box(colors)
assert np.allclose(boxed_colors, colors)
def test_nw_cut_to_box(self):
colors = np.random.random((10, 10))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors)
colors = np.random.random((10, 10, 3))
with raises(ValueError):
boxed_colors = nw_cut_to_box(colors, origin=(1.0, 1.0))
boxed_colors = nw_cut_to_box(colors)
assert np.allclose(boxed_colors, colors)
def test_fchebyshev_split():
from .. import fchebyshev_split
import numpy as np
from astropy.tests.helper import raises
x = np.array([-1, -0.5, 0, 0.5, 1], dtype="d")
#
# Test order
#
with raises(ValueError):
f = fchebyshev_split(x, 0)
with raises(ValueError):
f = fchebyshev_split(x, 1)
#
# m = 2
#
f = fchebyshev_split(x, 2)
foo = np.ones((2, x.size), dtype="d")
foo[0, :] = (x >= 0).astype(x.dtype)
assert np.allclose(f, foo)
#
# m = 3
#
f = fchebyshev_split(x, 3)
foo = np.ones((3, x.size), dtype="d")
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
assert np.allclose(f, foo)
#
# m = 4
#
f = fchebyshev_split(x, 4)
foo = np.ones((4, x.size), dtype="d")
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = 2.0 * x ** 2 - 1.0
assert np.allclose(f, foo)
#
# m = 5
#
f = fchebyshev_split(x, 5)
foo = np.ones((5, x.size), dtype="d")
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = 2.0 * x ** 2 - 1.0
foo[4, :] = 4.0 * x ** 3 - 3.0 * x
assert np.allclose(f, foo)
#
# random float
#
f = fchebyshev_split(2.88, 3)
assert np.allclose(f, np.array([[1.00], [1.00], [2.88]]))
def test_nw_scale_rgb(self):
colors = np.random.random((10, 10))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors)
colors = np.random.random((10, 10, 3))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors, scales=(1.0, 1.0))
colors = np.ones((2, 2, 3))
scaled_colors = nw_scale_rgb(colors, scales=(2.0, 2.0, 2.0))
assert np.allclose(scaled_colors, 2.0)
def test_nw_scale_rgb(self):
colors = np.random.random((10, 10))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors)
colors = np.random.random((10, 10, 3))
with raises(ValueError):
scaled_colors = nw_scale_rgb(colors, scales=(1.0, 1.0))
colors = np.ones((2, 2, 3))
scaled_colors = nw_scale_rgb(colors, scales=(2.0, 2.0, 2.0))
assert np.allclose(scaled_colors, 2.0)
def test_read_table_yanny(self):
"""Test reading to an astropy Table.
"""
filename = self.data('test_table.par')
with raises(PydlutilsException):
t = Table.read(filename)
with raises(KeyError):
t = Table.read(filename, tablename='foo')
t = Table.read(filename, tablename='test')
assert isinstance(t.meta, OrderedDict)
assert t.meta['name'] == 'first table'
assert (t['a'] == np.array([1, 4, 5])).all()
assert (t['c'] == np.array([b'x', b'y', b'z'])).all()
def test_read_table_yanny(self):
"""Test reading to an astropy Table.
"""
filename = self.data('test_table.par')
with raises(PydlutilsException):
t = Table.read(filename)
with raises(KeyError):
t = Table.read(filename, tablename='foo')
t = Table.read(filename, tablename='test')
assert isinstance(t.meta, OrderedDict)
assert t.meta['name'] == 'first table'
assert (t['a'] == np.array([1, 4, 5])).all()
assert (t['c'] == np.array([b'x', b'y', b'z'])).all()
def test_nw_arcsinh(self):
colors = np.random.random((10, 10))
with raises(ValueError):
fitted_colors = nw_arcsinh(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
fitted_colors = nw_arcsinh(colors)
colors = np.random.random((10, 10, 3))
fitted_colors = nw_arcsinh(colors, nonlinearity=0)
assert (fitted_colors == colors).all()
colors = np.ones((2, 2, 3))
fac = np.arcsinh(9.0) / 9.0
fitted_colors = nw_arcsinh(colors)
assert np.allclose(fitted_colors, fac)
def test_nw_arcsinh(self):
colors = np.random.random((10, 10))
with raises(ValueError):
fitted_colors = nw_arcsinh(colors)
colors = np.random.random((10, 10, 5))
with raises(ValueError):
fitted_colors = nw_arcsinh(colors)
colors = np.random.random((10, 10, 3))
fitted_colors = nw_arcsinh(colors, nonlinearity=0)
assert (fitted_colors == colors).all()
colors = np.ones((2, 2, 3))
fac = np.arcsinh(9.0)/9.0
fitted_colors = nw_arcsinh(colors)
assert np.allclose(fitted_colors, fac)
def test_fchebyshev_split(self):
x = np.array([-1, -0.5, 0, 0.5, 1], dtype='d')
#
# Test order
#
with raises(ValueError):
f = fchebyshev_split(x, 0)
with raises(ValueError):
f = fchebyshev_split(x, 1)
#
# m = 2
#
f = fchebyshev_split(x, 2)
foo = np.ones((2, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
assert np.allclose(f, foo)
#
# m = 3
#
f = fchebyshev_split(x, 3)
foo = np.ones((3, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
assert np.allclose(f, foo)
#
# m = 4
#
f = fchebyshev_split(x, 4)
foo = np.ones((4, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = (2.0 * x**2 - 1.0)
assert np.allclose(f, foo)
#
# m = 5
#
f = fchebyshev_split(x, 5)
foo = np.ones((5, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = (2.0 * x**2 - 1.0)
foo[4, :] = (4.0 * x**3 - 3.0 * x)
assert np.allclose(f, foo)
#
# random float
#
f = fchebyshev_split(2.88, 3)
assert np.allclose(f, np.array([[1.00], [1.00], [2.88]]))
def test_fchebyshev_split(self):
x = np.array([-1, -0.5, 0, 0.5, 1], dtype='d')
#
# Test order
#
with raises(ValueError):
f = fchebyshev_split(x, 0)
with raises(ValueError):
f = fchebyshev_split(x, 1)
#
# m = 2
#
f = fchebyshev_split(x, 2)
foo = np.ones((2, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
assert np.allclose(f, foo)
#
# m = 3
#
f = fchebyshev_split(x, 3)
foo = np.ones((3, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
assert np.allclose(f, foo)
#
# m = 4
#
f = fchebyshev_split(x, 4)
foo = np.ones((4, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = (2.0*x**2 - 1.0)
assert np.allclose(f, foo)
#
# m = 5
#
f = fchebyshev_split(x, 5)
foo = np.ones((5, x.size), dtype='d')
foo[0, :] = (x >= 0).astype(x.dtype)
foo[2, :] = x
foo[3, :] = (2.0*x**2 - 1.0)
foo[4, :] = (4.0*x**3 - 3.0*x)
assert np.allclose(f, foo)
#
# random float
#
f = fchebyshev_split(2.88, 3)
assert np.allclose(f, np.array([[1.00], [1.00], [2.88]]))
def test_template_metadata(self):
with raises(Pydlspec2dException):
slist, metadata = template_metadata('/no/such/file.par')
inputfile = get_pkg_data_filename('t/test_template_metadata.par')
slist, metadata = template_metadata(inputfile)
assert metadata['object'] == 'gal'
assert not metadata['nonnegative']
def test_window_score_no_photo_resolve(self, monkeypatch):
monkeypatch.setenv('PHOTO_CALIB', '/fake/directory')
monkeypatch.delenv('PHOTO_RESOLVE', raising=False)
with raises(PhotoopException) as e:
window_score()
assert (str(e.value) ==
'You have not set the environment variable PHOTO_RESOLVE!')
def test_template_metadata(self):
with raises(Pydlspec2dException):
slist, metadata = template_metadata('/no/such/file.par')
inputfile = get_pkg_data_filename('t/test_template_metadata.par')
slist, metadata = template_metadata(inputfile)
assert metadata['object'] == 'gal'
assert not metadata['nonnegative']
def test_read_fits_polygons(self):
poly = mng.read_fits_polygons(self.poly_fits)
use_caps = np.array([31, 15, 31, 7, 31, 15, 15, 7, 15, 15,
15, 31, 15, 15, 15, 15, 15, 15, 31, 15],
dtype=np.uint32)
#
# Attribute access doesn't work on unsigned columns.
#
assert (poly['USE_CAPS'] == use_caps).all()
assert (poly['use_caps'] == use_caps).all()
with raises(AttributeError):
foo = poly.no_such_attribute
cm0 = np.array([-1.0, -0.99369437, 1.0, -1.0, 0.00961538])
assert np.allclose(poly.cm[0][0:poly.ncaps[0]], cm0)
assert poly[0]['NCAPS'] == 5
poly = mng.read_fits_polygons(self.poly_fits, convert=True)
assert poly[0].use_caps == 31
assert np.allclose(poly[0].cm, cm0)
assert poly[0].cmminf() == 4
#
# A FITS file might not contain IFIELD.
#
poly = mng.read_fits_polygons(self.no_id_fits)
assert len(poly) == 1
#
# A FITS file might contain exactly one polygon with exactly one cap.
#
poly = mng.read_fits_polygons(self.one_cap_fits, convert=True)
assert poly[0].ncaps == 1
def test_aesthetics():
from .. import aesthetics
from ... import Pydlspec2dException
import numpy as np
from astropy.tests.helper import raises
np.random.seed(137)
flux = np.random.rand(100)
ivar = np.random.rand(100)
#
# No bad
#
f = aesthetics(flux, ivar)
assert (f == flux).all()
#
# Bad points
#
ivar[ivar < 0.1] = 0.0
#
# Bad method
#
with raises(Pydlspec2dException):
f = aesthetics(flux, ivar, 'badmethod')
#
# Nothing
#
f = aesthetics(flux, ivar, 'nothing')
assert (f == flux).all()
def test_unwrap_objid(self):
objid = unwrap_objid(np.array([1237661382772195474]))
assert objid.skyversion == 2
assert objid.rerun == 301
assert objid.run == 3704
assert objid.camcol == 3
assert objid.firstfield == 0
assert objid.frame == 91
assert objid.id == 146
objid = unwrap_objid(np.array(['1237661382772195474']))
assert objid.skyversion == 2
assert objid.rerun == 301
assert objid.run == 3704
assert objid.camcol == 3
assert objid.firstfield == 0
assert objid.frame == 91
assert objid.id == 146
objid = unwrap_objid(np.array([587722984180548043]))
assert objid.skyversion == 1
assert objid.rerun == 40
assert objid.run == 752
assert objid.camcol == 5
assert objid.firstfield == 1
assert objid.frame == 618
assert objid.id == 459
with raises(ValueError):
objid = unwrap_objid(np.array([3.14159]))
def test_pcomp(self):
test_data_file = join(self.data_dir, 'pcomp_data.txt')
test_data = np.loadtxt(test_data_file, dtype='d', delimiter=',')
with raises(ValueError):
foo = pcomp(np.arange(10))
pcomp_data = test_data[0:20, :]
m = 4
n = 20
means = np.tile(pcomp_data.mean(0), n).reshape(pcomp_data.shape)
newarray = pcomp_data - means
foo = pcomp(newarray, covariance=True)
#
# This array is obtained from the IDL version of PCOMP.
# It is only accurate up to an overall sign on each column.
#
derived = test_data[20:40, :]
for k in range(m):
assert_allclose(abs(foo.derived[:, k]), abs(derived[:, k]), 1e-4)
coefficients = test_data[40:44, :]
coefficientsT = coefficients.T
for k in range(m):
assert_allclose(abs(foo.coefficients[:, k]),
abs(coefficientsT[:, k]), 1e-4)
eigenvalues = test_data[44, :]
assert_allclose(foo.eigenvalues, eigenvalues, 1e-4)
variance = test_data[45, :]
assert_allclose(foo.variance, variance, 1e-4)
#
# Test the standardization.
#
foo = pcomp(pcomp_data, standardize=True, covariance=True)
# for k in range(m):
# assert_allclose(abs(foo.derived[:, k]), abs(derived[:, k]), 1e-4)
# for k in range(m):
# assert_allclose(abs(foo.coefficients[:, k]),
# abs(coefficientsT[:, k]),
# 1e-4)
eigenvalues = test_data[46, :]
assert_allclose(foo.eigenvalues, eigenvalues, 1e-4)
variance = test_data[47, :]
assert_allclose(foo.variance, variance, 1e-4)
# assert_allclose(foo.derived[0, :], np.array([-1.64153312,
# -9.12322038,
# 1.41790708,
# -8.29359322]))
#
# Make sure correlation is working at least.
#
foo = pcomp(pcomp_data, standardize=True)
assert_allclose(
foo.eigenvalues,
np.array([
2.84968632e+00, 1.00127640e+00, 1.48380121e-01, 6.57156222e-04
]))
assert_allclose(
foo.variance,
np.array([
7.12421581e-01, 2.50319100e-01, 3.70950302e-02, 1.64289056e-04
]))
def test_sdss_flagname(self):
names = sdss_flagname('ANCILLARY_TARGET1', 2310346608843161600)
assert tuple(names) == ('BRIGHTGAL', 'BLAZGX', 'ELG')
names = sdss_flagname('ANCILLARY_TARGET1', 2310346608843161600,
concat=True)
assert names == 'BRIGHTGAL BLAZGX ELG'
with raises(KeyError):
names = sdss_flagname('ABADMASK', 123456789)
def test_sdss_flagname(self):
names = sdss_flagname('ANCILLARY_TARGET1', 2310346608843161600)
assert tuple(names) == ('BRIGHTGAL', 'BLAZGX', 'ELG')
names = sdss_flagname('ANCILLARY_TARGET1', 2310346608843161600,
concat=True)
assert names == 'BRIGHTGAL BLAZGX ELG'
with raises(KeyError):
names = sdss_flagname('ABADMASK', 123456789)
def test_pcomp(self):
test_data_file = join(self.data_dir, 'pcomp_data.txt')
test_data = np.loadtxt(test_data_file, dtype='d', delimiter=',')
with raises(ValueError):
foo = pcomp(np.arange(10))
pcomp_data = test_data[0:20, :]
m = 4
n = 20
means = np.tile(pcomp_data.mean(0), n).reshape(pcomp_data.shape)
newarray = pcomp_data - means
foo = pcomp(newarray, covariance=True)
#
# This array is obtained from the IDL version of PCOMP.
# It is only accurate up to an overall sign on each column.
#
derived = test_data[20:40, :]
for k in range(m):
assert_allclose(abs(foo.derived[:, k]), abs(derived[:, k]), 1e-4)
coefficients = test_data[40:44, :]
coefficientsT = coefficients.T
for k in range(m):
assert_allclose(abs(foo.coefficients[:, k]),
abs(coefficientsT[:, k]),
1e-4)
eigenvalues = test_data[44, :]
assert_allclose(foo.eigenvalues, eigenvalues, 1e-4)
variance = test_data[45, :]
assert_allclose(foo.variance, variance, 1e-4)
#
# Test the standardization.
#
foo = pcomp(pcomp_data, standardize=True, covariance=True)
# for k in range(m):
# assert_allclose(abs(foo.derived[:, k]), abs(derived[:, k]), 1e-4)
# for k in range(m):
# assert_allclose(abs(foo.coefficients[:, k]),
# abs(coefficientsT[:, k]),
# 1e-4)
eigenvalues = test_data[46, :]
assert_allclose(foo.eigenvalues, eigenvalues, 1e-4)
variance = test_data[47, :]
assert_allclose(foo.variance, variance, 1e-4)
# assert_allclose(foo.derived[0, :], np.array([-1.64153312,
# -9.12322038,
# 1.41790708,
# -8.29359322]))
#
# Make sure correlation is working at least.
#
foo = pcomp(pcomp_data, standardize=True)
assert_allclose(foo.eigenvalues, np.array([2.84968632e+00,
1.00127640e+00,
1.48380121e-01,
6.57156222e-04]))
assert_allclose(foo.variance, np.array([7.12421581e-01,
2.50319100e-01,
3.70950302e-02,
1.64289056e-04]))
def test_ManglePolygon(self):
#
# Zero caps
#
poly = mng.ManglePolygon()
assert np.allclose(poly.str, 4.0*np.pi)
assert poly.cmminf() is None
assert not poly.gzeroar()
assert np.allclose(poly.garea(), 4.0*np.pi)
#
# One cap.
#
x = np.array([[0.0, 0.0, 1.0]])
cm = np.array([0.5])
poly = mng.ManglePolygon(x=x, cm=cm)
assert np.allclose(poly.garea(), np.pi)
#
# Bad inputs
#
with raises(ValueError):
poly = mng.ManglePolygon(weight=1.0)
#
# Multiple caps
#
x = np.array([[0.0, 0.0, 1.0],
[1.0, 0.0, 0.0],
[0.0, 1.0, 0.0]])
cm = np.array([1.0, 1.0, 1.0])
poly = mng.ManglePolygon(x=x, cm=cm, str=np.pi/2.0)
assert poly.ncaps == 3
assert poly.weight == 1.0
assert poly.use_caps == (1 << 3) - 1
poly = mng.ManglePolygon(x=x, cm=cm, weight=0.5)
assert poly.weight == 0.5
poly = mng.ManglePolygon(x=x, cm=cm, pixel=20)
assert poly.pixel == 20
poly = mng.ManglePolygon(x=x, cm=cm, use_caps=3, str=np.pi/2.0)
assert poly.use_caps == 3
poly2 = poly.copy()
assert poly2.use_caps == poly.use_caps
assert (poly2.cm == poly.cm).all()
assert np.allclose(poly2.str, np.pi/2.0)
x = np.array([[0.0, 0.0, 1.0],
[1.0, 0.0, 0.0]])
cm = np.array([1.0, 1.0])
poly = mng.ManglePolygon(x=x, cm=cm)
poly2 = poly.add_caps(np.array([[0.0, 1.0, 0.0], ]), np.array([1.0, ]))
assert poly2.ncaps == 3
assert poly2.use_caps == poly.use_caps
assert poly2.str == 1.0 # dummy value!
poly3 = poly.polyn(poly2, 2)
assert poly3.ncaps == 3
assert poly3.use_caps == poly.use_caps
assert np.allclose(poly3.x[2, :], np.array([0.0, 1.0, 0.0]))
poly3 = poly.polyn(poly2, 2, complement=True)
assert poly3.ncaps == 3
assert poly3.use_caps == poly.use_caps
assert np.allclose(poly3.cm[2], -1.0)
def test_sdss_name(self):
#
# Bad ftype
#
with raises(KeyError):
p = sdss_name('fooBar', 137, 4, 42)
for ftype in self.name_data:
assert sdss_name(ftype, 137, 4, 42, '301', 'r',
no_path=True) == self.name_data[ftype]
def test_write_ndarray_to_yanny_exceptions(self):
"""Make sure certain execptions are raised by the
write_ndarray_to_yanny function.
"""
test_data = self.test_data['tempfile1.par']
table = test_data['structures']['MYSTRUCT0']
dt = self.json2dtype(table['dtype'])
mystruct = np.zeros((table['size'],), dtype=dt)
for col in table['dtype']:
mystruct[col[0]] = np.array(table['data'][col[0]], dtype=col[1])
enums = {'new_flag': test_data['enums']['new_flag']}
with raises(PydlutilsException):
par = write_ndarray_to_yanny(self.data('test.par'), mystruct,
structnames='magnitudes', enums=enums)
with raises(PydlutilsException):
par = write_ndarray_to_yanny(self.temp('tempfile3.par'), mystruct,
structnames=('magnitudes', 'my_status'),
enums=enums)
def test_sdss_name(self):
#
# Bad ftype
#
with raises(KeyError):
p = sdss_name('fooBar', 137, 4, 42)
for ftype in self.name_data:
assert sdss_name(ftype, 137, 4, 42, '301', 'r',
no_path=True) == self.name_data[ftype]
def test_write_ndarray_to_yanny_exceptions(self):
"""Make sure certain execptions are raised by the write_ndarray_to_yanny function."""
mystruct = np.zeros((4,),
dtype=[('ra','f8'), ('dec','f8'), ('mag','f4',(5,)),
('flags','i4'), ('new_flag','|S5')])
mystruct['ra'] = np.array([10.0, 20.5, 30.75, 40.55],dtype=np.float64)
mystruct['dec'] = np.array([-5.1234, -10.74832, 67.994523, 11.437281],dtype=np.float64)
mystruct['mag'] = np.array([[0.0,1.0,2.0,3.0,4.0],
[5.1,6.2,7.3,8.4,9.5],
[22.123,23.95,22.6657,21.0286,22.9876],
[13.54126,15.37456,14.52647,12.648640,12.0218]],dtype=np.float32)
mystruct['flags'] = np.array([2**2, 2**4, 2**6, 2**8 + 2**3],dtype=np.int32)
mystruct['new_flag'] = np.array(['FALSE','TRUE','TRUE','FALSE'],dtype='|S5')
enums = {'new_flag':('BOOLEAN',('FALSE','TRUE'))}
with raises(PydlutilsException):
par = write_ndarray_to_yanny(os.path.join(self.data_dir,'test.par'),mystruct,structnames='magnitudes',enums=enums)
with raises(PydlutilsException):
par = write_ndarray_to_yanny(self.temp('tempfile3.par'),mystruct,structnames=('magnitudes','my_status'),enums=enums)
def test_spheregroup(self):
test_data_file = get_pkg_data_filename('t/spheregroup_data.txt')
test_data = np.loadtxt(test_data_file, dtype='d', delimiter=',')
# np.random.seed(137)
# Ngroup = 3
# N = 50
# spread = 0.05
linklength = 3.0 # degrees
# x0 = np.concatenate( ( np.random.normal(loc=1,scale=spread,size=(N,)),
# np.random.normal(loc=-1,scale=spread,size=(N,)),
# np.random.normal(loc=1,scale=spread,size=(N,)),
# )).reshape((N*Ngroup,))
# y0 = np.concatenate( ( np.random.normal(loc=1,scale=spread,size=(N,)),
# np.random.normal(loc=-1,scale=spread,size=(N,)),
# np.random.normal(loc=-1,scale=spread,size=(N,)),
# )).reshape((N*Ngroup,))
# z0 = np.concatenate( ( np.random.normal(loc=1,scale=spread,size=(N,)),
# np.random.normal(loc=-1,scale=spread,size=(N,)),
# np.random.normal(loc=0,scale=spread,size=(N,)),
# )).reshape((N*Ngroup,))
# foo = np.arange(N*Ngroup)
# np.random.shuffle(foo)
# x = x0[foo]
# y = y0[foo]
# z = z0[foo]
# r = np.sqrt(x**2 + y**2 + z**2)
# theta = np.degrees(np.arccos(z/r))
# phi = np.degrees(np.arctan2(y,x))
# ra = np.where(phi < 0, phi + 360.0,phi)
# dec = 90.0 - theta
# group = spheregroup(ra,dec,linklength)
#
# Reproduce IDL results
#
ra = test_data[0, :]
dec = test_data[1, :]
expected_ingroup = test_data[2, :].astype(np.int64)
expected_multgroup = test_data[3, :].astype(np.int64)
expected_firstgroup = test_data[4, :].astype(np.int64)
expected_nextgroup = test_data[5, :].astype(np.int64)
group = spheregroup(ra, dec, linklength)
assert (group[0] == expected_ingroup).all()
assert (group[1] == expected_multgroup).all()
assert (group[2] == expected_firstgroup).all()
assert (group[3] == expected_nextgroup).all()
#
# Exceptions
#
with raises(PydlutilsException):
group = spheregroup(np.array([137.0]), np.array([55.0]), linklength)
#
# warnings
#
with catch_warnings(PydlutilsUserWarning) as w:
group = spheregroup(ra, dec, linklength, chunksize=linklength)
# w = recwarn.pop(PydlutilsUserWarning)
assert "chunksize changed to" in str(w[0].message)
def test_circle_cap(self):
with raises(ValueError):
x, cm = mng.circle_cap(90.0, np.array([[1.0, 2.0, 3.0, 4.0], ]))
xin = np.array([[0.0, 0.0, 1.0], ])
x, cm = mng.circle_cap(90.0, xin)
assert np.allclose(x, xin)
assert np.allclose(cm, 1.0)
radec = np.array([[0.0, 90.0], ])
x, cm = mng.circle_cap(90.0, radec)
assert np.allclose(x, xin)
assert np.allclose(cm, 1.0)
x, cm = mng.circle_cap(np.float32(90.0), radec)
assert np.allclose(x, xin)
assert np.allclose(cm, 1.0)
x, cm = mng.circle_cap(np.array([90.0, ]), radec)
assert np.allclose(x, xin)
assert np.allclose(cm, np.array([1.0, ]))
with raises(ValueError):
x, cm = mng.circle_cap(np.array([90.0, 90.0]), radec)
def test_read_mangle_polygons(self):
with raises(PydlutilsException):
poly = read_mangle_polygons(self.bad_ply)
poly = read_mangle_polygons(self.poly_ply)
assert len(poly.header) == 3
assert poly.header[0] == 'pixelization 6s'
assert len(poly) == 4
assert np.allclose(
poly[0].x[0, :],
np.array([0.0436193873653360, 0.9990482215818578, 0.0]))
assert poly[3].ncaps == 3
def test_read_ds_cooling(self):
with raises(ValueError):
logT, loglambda = read_ds_cooling('m-99.cie')
logT, logL = read_ds_cooling('m-15.cie')
assert np.allclose(logT[0:5], np.array([4.0, 4.05, 4.1, 4.15, 4.2]))
assert np.allclose(logL[0:5], np.array([-26.0, -24.66, -23.52,
-22.62, -22.11]))
logT = np.array([4.025, 4.125, 4.225, 4.325])
logT2, logL = read_ds_cooling('m-15.cie', logT)
assert np.allclose(logT2, logT)
assert np.allclose(logL, np.array([-25.33, -23.07, -22.04, -22.06]))
def test_read_mangle_polygons(self):
with raises(PydlutilsException):
poly = read_mangle_polygons(self.bad_ply)
poly = read_mangle_polygons(self.poly_ply)
assert len(poly.header) == 3
assert poly.header[0] == 'pixelization 6s'
assert len(poly) == 4
assert np.allclose(poly[0].x[0, :],
np.array([0.0436193873653360, 0.9990482215818578,
0.0]))
assert poly[3].ncaps == 3
def test_read_ds_cooling(self):
with raises(ValueError):
logT, loglambda = read_ds_cooling('m-99.cie')
logT, logL = read_ds_cooling('m-15.cie')
assert np.allclose(logT[0:5], np.array([4.0, 4.05, 4.1, 4.15, 4.2]))
assert np.allclose(logL[0:5],
np.array([-26.0, -24.66, -23.52, -22.62, -22.11]))
logT = np.array([4.025, 4.125, 4.225, 4.325])
logT2, logL = read_ds_cooling('m-15.cie', logT)
assert np.allclose(logT2, logT)
assert np.allclose(logL, np.array([-25.33, -23.07, -22.04, -22.06]))
def test_ManglePolygon(self):
with raises(ValueError):
poly = ManglePolygon(weight=1.0)
with raises(ValueError):
poly = ManglePolygon()
x = np.array([[0.0, 0.0, 1.0], [1.0, 0.0, 0.0], [0.0, 1.0, 1.0]])
cm = np.array([1.0, 1.0, 1.0])
poly = ManglePolygon(x=x, cm=cm, str=np.pi / 2.0)
assert poly.ncaps == 3
assert poly.weight == 1.0
assert poly.use_caps == (1 << 3) - 1
poly = ManglePolygon(x=x, cm=cm, weight=0.5)
assert poly.weight == 0.5
poly = ManglePolygon(x=x, cm=cm, pixel=20)
assert poly.pixel == 20
poly = ManglePolygon(x=x, cm=cm, use_caps=3, str=np.pi / 2.0)
assert poly.use_caps == 3
poly2 = poly.copy()
assert poly2.use_caps == poly.use_caps
assert (poly2.cm == poly.cm).all()
assert np.allclose(poly2.str, np.pi / 2.0)
def test_djs_maskinterp(self):
y = np.array([0.0, 1.0, 2.0, 3.0, 4.0], dtype=np.float64)
mask = np.array([0, 1, 0])
with raises(ValueError):
yi = djs_maskinterp(y, mask)
mask = np.array([0, 1, 0, 0, 0])
x = np.array([0.0, 0.5, 1.0], dtype=y.dtype)
with raises(ValueError):
yi = djs_maskinterp(y, mask, xval=x)
# 1-D case
yi = djs_maskinterp(y, mask)
assert np.allclose(y, yi)
# 2-D case
x = np.array([0.0, 0.5, 1.0, 1.5, 2.0], dtype=y.dtype)
x = np.vstack((x, x, x))
y = np.vstack((y, y, y))
mask = np.vstack((mask, mask, mask))
with raises(ValueError):
yi = djs_maskinterp(y, mask)
with raises(ValueError):
yi = djs_maskinterp(y, mask, axis=-1)
with raises(ValueError):
yi = djs_maskinterp(y, mask, axis=2)
yi = djs_maskinterp(y, mask, axis=0)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=0, xval=x)
assert np.allclose(y, yi)
mask[:, 1] = 0
mask[1, :] = 1
yi = djs_maskinterp(y, mask, axis=1)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=1, xval=x)
assert np.allclose(y, yi)
# 3-D case
x = np.dstack((x, x, x, x, x, x, x))
y = np.dstack((y, y, y, y, y, y, y))
mask = np.dstack((mask, mask, mask, mask, mask, mask, mask))
mask[:, :, :] = 0
mask[:, :, 5] = 1
yi = djs_maskinterp(y, mask, axis=0)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=0, xval=x)
assert np.allclose(y, yi)
mask[:, :, :] = 0
mask[:, 3, :] = 1
yi = djs_maskinterp(y, mask, axis=1)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=1, xval=x)
assert np.allclose(y, yi)
mask[:, :, :] = 0
mask[1, :, :] = 1
yi = djs_maskinterp(y, mask, axis=2)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=2, xval=x)
assert np.allclose(y, yi)
# 4-D case
y = np.random.random((2, 2, 2, 2))
with raises(ValueError):
yi = djs_maskinterp(y, (y > 0.5), axis=0)
def test_sdss_flagexist():
from os.path import dirname, join
from astropy.tests.helper import raises
import pydl.pydlutils.sdss
from ..set_maskbits import set_maskbits
pydl.pydlutils.sdss.maskbits = set_maskbits(maskbits_file=join(dirname(__file__),'t','testMaskbits.par'))
from .. import sdss_flagname
names = sdss_flagname('ANCILLARY_TARGET1',2310346608843161600)
assert tuple(names) == ('BRIGHTGAL', 'BLAZGX', 'ELG')
names = sdss_flagname('ANCILLARY_TARGET1',2310346608843161600,concat=True)
assert names == 'BRIGHTGAL BLAZGX ELG'
with raises(KeyError):
names = sdss_flagname('ABADMASK',123456789)
def test_write_ndarray_to_yanny_exceptions(self):
"""Make sure certain execptions are raised by the
write_ndarray_to_yanny function.
"""
test_data = self.test_data['tempfile1.par']
table = test_data['structures']['MYSTRUCT0']
dt = self.json2dtype(table['dtype'])
mystruct = np.zeros((table['size'], ), dtype=dt)
for col in table['dtype']:
mystruct[col[0]] = np.array(table['data'][col[0]], dtype=col[1])
enums = {'new_flag': test_data['enums']['new_flag']}
with raises(PydlutilsException):
par = write_ndarray_to_yanny(self.data('test.par'),
mystruct,
structnames='magnitudes',
enums=enums)
with raises(PydlutilsException):
par = write_ndarray_to_yanny(self.temp('tempfile3.par'),
mystruct,
structnames=('magnitudes',
'my_status'),
enums=enums)
def test_djs_maskinterp(self):
y = np.array([0.0, 1.0, 2.0, 3.0, 4.0], dtype=np.float64)
mask = np.array([0, 1, 0])
with raises(ValueError):
yi = djs_maskinterp(y, mask)
mask = np.array([0, 1, 0, 0, 0])
x = np.array([0.0, 0.5, 1.0], dtype=y.dtype)
with raises(ValueError):
yi = djs_maskinterp(y, mask, xval=x)
# 1-D case
yi = djs_maskinterp(y, mask)
assert np.allclose(y, yi)
# 2-D case
x = np.array([0.0, 0.5, 1.0, 1.5, 2.0], dtype=y.dtype)
x = np.vstack((x, x, x))
y = np.vstack((y, y, y))
mask = np.vstack((mask, mask, mask))
with raises(ValueError):
yi = djs_maskinterp(y, mask)
with raises(ValueError):
yi = djs_maskinterp(y, mask, axis=-1)
with raises(ValueError):
yi = djs_maskinterp(y, mask, axis=2)
yi = djs_maskinterp(y, mask, axis=0)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=0, xval=x)
assert np.allclose(y, yi)
mask[:, 1] = 0
mask[1, :] = 1
yi = djs_maskinterp(y, mask, axis=1)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=1, xval=x)
assert np.allclose(y, yi)
# 3-D case
x = np.dstack((x, x, x, x, x, x, x))
y = np.dstack((y, y, y, y, y, y, y))
mask = np.dstack((mask, mask, mask, mask, mask, mask, mask))
mask[:, :, :] = 0
mask[:, :, 5] = 1
yi = djs_maskinterp(y, mask, axis=0)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=0, xval=x)
assert np.allclose(y, yi)
mask[:, :, :] = 0
mask[:, 3, :] = 1
yi = djs_maskinterp(y, mask, axis=1)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=1, xval=x)
assert np.allclose(y, yi)
mask[:, :, :] = 0
mask[1, :, :] = 1
yi = djs_maskinterp(y, mask, axis=2)
assert np.allclose(y, yi)
yi = djs_maskinterp(y, mask, axis=2, xval=x)
assert np.allclose(y, yi)
# 4-D case
y = np.random.random((2, 2, 2, 2))
with raises(ValueError):
yi = djs_maskinterp(y, (y > 0.5), axis=0)
def test_unwrap_specobjid(self):
s = 4565636362342690816
d = unwrap_specobjid(np.array([s], dtype=np.uint64))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 0
d = unwrap_specobjid(np.array(['4565636362342690816']))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 0
si = 4565636362342690816 + 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64),
specLineIndex=True)
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.index == 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64),
run2d_integer=True)
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 700
assert d.line == 137
#
# Because 5-digit plate IDs...
#
s5 = 11258999466550599680
d = unwrap_specobjid(np.array([s5], dtype=np.uint64))
assert d.plate == 10000
assert d.mjd == 57346
assert d.fiber == 1
assert d.run2d == 'v5_10_0'
assert d.line == 0
#
# Exceptions
#
with raises(ValueError):
d = unwrap_specobjid(np.array([4565636362342690816],
dtype=np.int64))
def test_unwrap_specobjid(self):
s = 4565636362342690816
d = unwrap_specobjid(np.array([s], dtype=np.uint64))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 0
d = unwrap_specobjid(np.array(['4565636362342690816']))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 0
si = 4565636362342690816 + 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64))
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.line == 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64),
specLineIndex=True)
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 'v5_7_0'
assert d.index == 137
d = unwrap_specobjid(np.array([si], dtype=np.uint64),
run2d_integer=True)
assert d.plate == 4055
assert d.mjd == 55359
assert d.fiber == 408
assert d.run2d == 700
assert d.line == 137
#
# Because 5-digit plate IDs...
#
s5 = 11258999466550599680
d = unwrap_specobjid(np.array([s5], dtype=np.uint64))
assert d.plate == 10000
assert d.mjd == 57346
assert d.fiber == 1
assert d.run2d == 'v5_10_0'
assert d.line == 0
#
# Exceptions
#
with raises(ValueError):
d = unwrap_specobjid(
np.array([4565636362342690816], dtype=np.int64))
def test_ManglePolygon(self):
with raises(ValueError):
poly = ManglePolygon(weight=1.0)
with raises(ValueError):
poly = ManglePolygon()
x = np.array([[0.0, 0.0, 1.0],
[1.0, 0.0, 0.0],
[0.0, 1.0, 1.0]])
cm = np.array([1.0, 1.0, 1.0])
poly = ManglePolygon(x=x, cm=cm, str=np.pi/2.0)
assert poly.ncaps == 3
assert poly.weight == 1.0
assert poly.use_caps == (1 << 3) - 1
poly = ManglePolygon(x=x, cm=cm, weight=0.5)
assert poly.weight == 0.5
poly = ManglePolygon(x=x, cm=cm, pixel=20)
assert poly.pixel == 20
poly = ManglePolygon(x=x, cm=cm, use_caps=3, str=np.pi/2.0)
assert poly.use_caps == 3
poly2 = poly.copy()
assert poly2.use_caps == poly.use_caps
assert (poly2.cm == poly.cm).all()
assert np.allclose(poly2.str, np.pi/2.0)
def test_compose_prefix_unit():
x = u.m.compose(units=(u.m, ))
assert x[0].bases[0] is u.m
assert x[0].scale == 1.0
x = u.m.compose(units=[u.km], include_prefix_units=True)
assert x[0].bases[0] is u.km
assert x[0].scale == 0.001
x = u.m.compose(units=[u.km])
assert x[0].bases[0] is u.km
assert x[0].scale == 0.001
x = (u.km / u.s).compose(units=(u.pc, u.Myr))
assert x[0].bases == [u.pc, u.Myr]
assert_allclose(x[0].scale, 1.0227121650537077)
with raises(u.UnitsError):
(u.km / u.s).compose(units=(u.pc, u.Myr), include_prefix_units=False)
def test_gcirc(self):
np.random.seed(137)
#
# Start in radians
#
offset = 5.0e-6 # approx 1 arcsec
ra1 = 2.0 * np.pi * np.random.rand(100)
dec1 = np.pi / 2.0 - np.arccos(2.0 * np.random.rand(100) - 1.0)
ra2 = ra1 + offset
ra2 = np.where((ra2 > 2.0 * np.pi), ra2 - 2.0 * np.pi, ra2)
dec2 = np.where((dec1 > 0), dec1 - offset, dec1 + offset)
deldec2 = (dec2 - dec1) / 2.0
delra2 = (ra2 - ra1) / 2.0
sindis = np.sqrt(
np.sin(deldec2) * np.sin(deldec2) +
np.cos(dec1) * np.cos(dec2) * np.sin(delra2) * np.sin(delra2))
dis = 2.0 * np.arcsin(sindis)
#
# units = 0
#
d0 = gcirc(ra1, dec1, ra2, dec2, units=0)
assert np.allclose(d0, dis)
#
# units = 2
#
d0 = gcirc(np.rad2deg(ra1) / 15.0,
np.rad2deg(dec1),
np.rad2deg(ra2) / 15.0,
np.rad2deg(dec2),
units=1)
assert np.allclose(d0, np.rad2deg(dis) * 3600.0)
#
# units = 2
#
d0 = gcirc(np.rad2deg(ra1),
np.rad2deg(dec1),
np.rad2deg(ra2),
np.rad2deg(dec2),
units=2)
assert np.allclose(d0, np.rad2deg(dis) * 3600.0)
#
# Units = whatever
#
with raises(ValueError):
d0 = gcirc(ra1, dec1, ra2, dec2, units=5)
def test_median(self):
odd_data = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13],
dtype=np.float32)
even_data = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],
dtype=np.float32)
assert median(odd_data) == 7
assert median(odd_data, even=True) == 7
assert median(even_data) == 7
assert median(even_data, even=True) == 6.5
assert (median(odd_data, 3) == odd_data).all()
with raises(ValueError):
foo = median(np.ones((9, 9, 9)), 3)
odd_data2 = np.vstack(
(odd_data, odd_data, odd_data, odd_data, odd_data))
assert (median(odd_data2, 3) == odd_data2).all()
assert (median(odd_data2, axis=0) == odd_data).all()
assert (median(odd_data2, axis=1) == 7 * np.ones(
(odd_data2.shape[0], ), dtype=odd_data2.dtype)).all()
def test_sdss_astrombad_raises(self):
with raises(ValueError):
foo = sdss_astrombad(77, 32, 20)
with raises(ValueError):
foo = sdss_astrombad(-1, 1, 20)
with raises(ValueError):
foo = sdss_astrombad(2**17, 1, 20)
with raises(ValueError):
foo = sdss_astrombad(-2, 1, 20)
with raises(ValueError):
foo = sdss_astrombad(251, 1, 2**16)
with raises(ValueError):
foo = sdss_astrombad(np.array([77, 85, 251]), np.array([1]),
np.array([20, 15, 151]))
with raises(ValueError):
foo = sdss_astrombad(np.array([77, 85, 251]), np.array([1, 2, 3]),
np.array([20]))
def test_fpoly(self):
x = np.array([-1, -0.5, 0, 0.5, 1], dtype='d')
#
# Test order
#
with raises(ValueError):
f = fpoly(x, 0)
#
# m = 1
#
f = fpoly(x, 1)
assert (f == np.ones((1, x.size), dtype='d')).all()
#
# m = 2
#
f = fpoly(x, 2)
foo = np.ones((2, x.size), dtype='d')
foo[1, :] = x
assert np.allclose(f, foo)
#
# m = 3
#
f = fpoly(x, 3)
foo = np.ones((3, x.size), dtype='d')
foo[1, :] = x
foo[2, :] = x**2
assert np.allclose(f, foo)
#
# m = 4
#
f = fpoly(x, 4)
foo = np.ones((4, x.size), dtype='d')
foo[1, :] = x
foo[2, :] = x**2
foo[3, :] = x**3
assert np.allclose(f, foo)
#
# random float
#
f = fpoly(2.88, 3)
assert np.allclose(f, np.array([[1.00], [2.88], [8.29440]]))
def test_flegendre(self):
x = np.array([-1, -0.5, 0, 0.5, 1], dtype='d')
#
# Test order
#
with raises(ValueError):
f = flegendre(x, 0)
#
# m = 1
#
f = flegendre(x, 1)
assert (f == np.ones((1, x.size), dtype='d')).all()
#
# m = 2
#
f = flegendre(x, 2)
foo = np.ones((2, x.size), dtype='d')
foo[1, :] = x
assert np.allclose(f, foo)
#
# m = 3
#
f = flegendre(x, 3)
foo = np.ones((3, x.size), dtype='d')
foo[1, :] = x
foo[2, :] = 0.5 * (3.0 * x**2 - 1.0)
assert np.allclose(f, foo)
#
# m = 4
#
f = flegendre(x, 4)
foo = np.ones((4, x.size), dtype='d')
foo[1, :] = x
foo[2, :] = 0.5 * (3.0 * x**2 - 1.0)
foo[3, :] = 0.5 * (5.0 * x**3 - 3.0 * x)
assert np.allclose(f, foo)
#
# random float
#
f = flegendre(2.88, 3)
assert np.allclose(f, np.array([[1.00], [2.88], [11.9416]]))