def test_ambiguous_format(tmpdir):
@data_factory('b', identifier=df.has_extension('spam'), priority=34)
def reader1(filename):
return Data()
@data_factory('a', identifier=df.has_extension('spam'), priority=34)
def reader2(filename):
return Data()
@data_factory('c', identifier=df.has_extension('spam'), priority=22)
def reader3(filename):
return Data()
filename = tmpdir.join('test.spam').strpath
with open(filename, 'w') as f:
f.write('Camelot!')
# Should raise a warning and pick the highest priority one in alphabetical
# order
with warnings.catch_warnings(record=True) as w:
factory = df.find_factory(filename)
assert len(w) == 1
assert str(w[0].message) == "Multiple data factories matched the input: 'a', 'b'. Choosing 'a'."
assert factory is reader2
def test_ambiguous_format(tmpdir):
@data_factory('b', identifier=df.has_extension('spam'), priority=34)
def reader1(filename):
return Data()
@data_factory('a', identifier=df.has_extension('spam'), priority=34)
def reader2(filename):
return Data()
@data_factory('c', identifier=df.has_extension('spam'), priority=22)
def reader3(filename):
return Data()
filename = tmpdir.join('test.spam').strpath
with open(filename, 'w') as f:
f.write('Camelot!')
# Should raise a warning and pick the highest priority one in alphabetical
# order
with warnings.catch_warnings(record=True) as w:
factory = df.find_factory(filename)
assert len(w) == 1
assert str(w[0].message) == "Multiple data factories matched the input: 'a', 'b'. Choosing 'a'."
assert factory is reader2
def test_basedata(tmpdir):
# Regression test for a bug that caused load_data to fail if a data
# factory returned a BaseData (but not Data) subclass, due to the
# LoadLog expecting a Data object
class BigData(BaseCartesianData):
def get_kind(self):
pass
def compute_histogram(self):
pass
def compute_statistic(self):
pass
def get_mask(self):
pass
@property
def shape(self):
pass
@property
def main_components(self):
return []
@data_factory('bigdata',
identifier=df.has_extension('bigdata'),
priority=34)
def reader(filename):
return BigData()
filename = tmpdir.join('test.bigdata').strpath
with open(filename, 'w') as f:
f.write('Camelot!')
d = df.load_data(filename)
assert isinstance(d, BigData)
class RegionData(Data):
def to_subset(self, wcs):
print("thing with to_subset: ",self)
preg = [reg.to_pixel(wcs)
if hasattr(reg, 'to_pixel')
else reg
for reg in self['regions']]
subsets = [reg_to_roi(reg) for reg in preg]
return subsets
@data_factory('DS9 Region File', has_extension('reg'), default='reg')
def ds9_region(filename):
reg = regions.read_ds9(filename)
comp = Component(np.ones(len(reg), dtype='bool'))
data = RegionData(label='Regions: {0}'.format(os.path.split(filename)[-1]),
regions=reg)
return data
@layer_action(label='Convert regions to subset')
def layer_to_subset(selected_layers, data_collection):
# loop over selected layers
for layer in selected_layers:
if isinstance(layer, RegionData):
from glue.core import Data
from glue.core.coordinates import coordinates_from_wcs
from glue.config import data_factory
from glue.core.data_factories import has_extension
import pyfits as fits
from stwcs.wcsutil import HSTWCS
@data_factory('Hubble Image', has_extension('fits fit'), default='fits fit')
def hubble_data(filename):
"""
Data loader customized for 'typical' hubble fits files
This function extracts groups of (SCI, ERR, and DQ) extensions
from a file. Each is retuned as a glue Data object
HSTWCS objects are used to parse wcs.
Any other extensions are ignored
"""
#assumption: relevant SCI/ERR/DQ arrays are
#grouped together, with SCI component first
result = []
hdulist = fits.open(filename, memmap=True)
label = filename.split('.')[0]
label = label.split('/')[-1].split('\\')[-1]
index = 0
with open(filename) as hfile:
hdr = hfile.read().splitlines()
hdr = map(float, hdr)
hdr = dict(CD1_1=hdr[0],
CD1_2=hdr[1] * (-1),
CD2_1=hdr[2],
CD2_2=hdr[3] * (-1),
CRVAL1=hdr[4],
CRVAL2=hdr[5],
CRPIX1=0,
CRPIX2=shp[0])
wcs = WCS(hdr)
return coordinates_from_wcs(wcs)
@data_factory('GIS TIFF', has_extension('tiff tif'), default='tif')
def read_tiff_metadata(filename):
""" Read a TIFF image, looking for .tfw metadata """
base, ext = os.path.splitext(filename)
data = np.flipud(np.array(Image.open(filename).convert('L')))
result = Data()
if os.path.exists(base + '.tfw'):
result.coords = tfw_to_coords(base + '.tfw', data.shape)
result.add_component(data, 'map')
return result
@data_factory('GIS Shapefile', has_extension('shx shp dbf'),
from glue.core import Data
from glue.core.coordinates import coordinates_from_wcs
from glue.config import data_factory
from glue.core.data_factories import has_extension
import pyfits as fits
from stwcs.wcsutil import HSTWCS
@data_factory('Hubble Image', has_extension('fits fit'), default='fits fit')
def hubble_data(filename):
"""
Data loader customized for 'typical' hubble fits files
This function extracts groups of (SCI, ERR, and DQ) extensions
from a file. Each is retuned as a glue Data object
HSTWCS objects are used to parse wcs.
Any other extensions are ignored
"""
#assumption: relevant SCI/ERR/DQ arrays are
#grouped together, with SCI component first
result = []
hdulist = fits.open(filename, memmap=True)
label = filename.split('.')[0]
label = label.split('/')[-1].split('\\')[-1]
index = 0
from __future__ import print_function
from glue.core import Data
from glue.core.coordinates import coordinates_from_header
from glue.config import data_factory
from glue.core.data_factories import has_extension
from glue.external.astro import fits
import numpy as np
@data_factory('GALFA HI cube', has_extension('fits fit'))
def _load_GALFAHI_data(filename, **kwargs):
def _get_cube_center(header, cubeshape):
ra = header['CRVAL1'] + header['CDELT1'] * (np.arange(cubeshape[2])+0.5 - header['CRPIX1']) ## degree
dec = header['CRVAL2'] + header['CDELT2'] * (np.arange(cubeshape[1])+0.5 - header['CRPIX2']) ## degree
return np.mean(ra), np.mean(dec)
# add the primary components
cube = fits.getdata(filename)
header = fits.getheader(filename)
header['CDELT3'] = header['CDELT3'] * (10**(-3)) # m/s --> km/s
cen_ra, cen_dec = _get_cube_center(header, cube.shape)
nn = filename.split('/')[-1]
# cube_name = '%s_RA%dDEC%d' % (nn[0:3], cen_ra, cen_dec)
cube_name = 'G_%d%+.2fradec_%.1fkm/s_%.1fa' % (cen_ra, cen_dec, header['CDELT3'], header['CDELT2']*60.)
data = Data()
data.coords = coordinates_from_header(header)
data.add_component(cube, cube_name)
data.label = cube_name
data_list = []
from __future__ import print_function
from os.path import basename
from collections import defaultdict
from glue.core import Data
from glue.core.coordinates import coordinates_from_header
from glue.config import data_factory
from glue.core.data_factories import has_extension
from glue.external.astro import fits
@data_factory('Generic FITS', has_extension('fits fit'))
def _load_fits_generic(filename, **kwargs):
hdulist = fits.open(filename)
groups = defaultdict(Data)
for extnum, hdu in enumerate(hdulist):
if not isinstance(hdu, fits.TableHDU) and\
hdu.data is not None:
shape = hdu.data.shape
if shape not in groups:
label = '{}[{}]'.format(
basename(filename).split('.', 1)[0],
'x'.join((str(x) for x in shape))
)
data = Data(label=label)
data.coords = coordinates_from_header(hdu.header)
groups[shape] = data
else:
data = groups[shape]
data.add_component(component=hdu.data,
from glue.core import Data
from glue.core.coordinates import coordinates_from_wcs
from glue.config import data_factory
from glue.core.data_factories import has_extension
import astropy.io.fits as fits
from astropy.wcs import WCS
import numpy as np
@data_factory('Herschel Image',
has_extension('fits fit fits.gz'),
default='fits fit fits.gz')
def herschel_data(filename):
"""
Data loader customized for Herschel fits files
This function extracts extension named 'image',
'error', 'coverage', etc
from a file. Each is returned as a glue Data object.
To handle PACS cubes, if ImageIndex extension is present
it is used to provide wavelengths
astropy.wcs.WCS objects are used to parse wcs.
Any other extensions are ignored
"""
hdulist = fits.open(filename, memmap=True, ignore_missing_end=True)
d = Data("data")