def test_GIVEN_gridded_datagroups_WHEN_read_datagroups_THEN_data_returned_in_list(
self):
datagroup_1 = {
'variables': ['var1', 'var2'],
'filenames': ['filename1.nc'],
'product': None
}
datagroup_2 = {
'variables': ['var3'],
'filenames': ['filename2.nc'],
'product': 'cis'
}
var1 = make_from_cube(make_square_5x3_2d_cube())
var2 = make_from_cube(make_square_5x3_2d_cube())
var3 = make_from_cube(make_square_5x3_2d_cube())
get_data_func = MagicMock(side_effect=[var1, var2, var3])
get_var_func = MagicMock(side_effect=lambda f: {
'filename1.nc': ['var1', 'var2'],
'filename2.nc': ['var3']
}[f])
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
data = reader.read_datagroups([datagroup_1, datagroup_2])
assert_that(len(data), is_(3))
assert_that(data[0], is_(var1))
assert_that(data[1], is_(var2))
assert_that(data[2], is_(var3))
def stats_cmd(main_arguments):
"""
Main routine for handling calls to the statistics command.
:param main_arguments: The command line arguments (minus the stats command)
"""
from cis.stats import StatsAnalyzer
from cis.data_io.gridded_data import GriddedDataList
data_reader = DataReader()
data_list = data_reader.read_datagroups(main_arguments.datagroups)
analyzer = StatsAnalyzer(*data_list)
results = analyzer.analyze()
header = "RESULTS OF STATISTICAL COMPARISON:"
note = "Compared all points which have non-missing values in both variables"
header_length = max(len(header), len(note))
print(header_length * '=')
print(header)
print(header_length * '-')
print(note)
print(header_length * '=')
for result in results:
print(result.pprint())
if main_arguments.output:
cubes = GriddedDataList([result.as_cube() for result in results])
variables = []
filenames = []
for datagroup in main_arguments.datagroups:
variables.extend(datagroup['variables'])
filenames.extend(datagroup['filenames'])
history = "Statistical comparison performed using CIS version " + __version__ + \
"\n variables: " + str(variables) + \
"\n from files: " + str(set(filenames))
cubes.add_history(history)
cubes.save_data(main_arguments.output)
def test_GIVEN_multiple_datagroups_WHEN_read_datagroups_THEN_get_data_called_correctly(
self):
datagroup_1 = {
'variables': ['var1', 'var2'],
'filenames': ['filename1.nc'],
'product': None
}
datagroup_2 = {
'variables': ['var3', 'var4'],
'filenames': ['filename2.nc'],
'product': 'cis'
}
get_data_func = MagicMock(
return_value=make_regular_2d_ungridded_data())
get_var_func = MagicMock(side_effect=lambda f: {
'filename1.nc': ['var1', 'var2'],
'filename2.nc': ['var3', 'var4']
}[f])
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
data = reader.read_datagroups([datagroup_1, datagroup_2])
assert_that(get_data_func.call_count, is_(4))
assert_that(get_data_func.call_args_list[0][0],
is_((['filename1.nc'], 'var1', None)))
assert_that(get_data_func.call_args_list[1][0],
is_((['filename1.nc'], 'var2', None)))
assert_that(get_data_func.call_args_list[2][0],
is_((['filename2.nc'], 'var3', 'cis')))
assert_that(get_data_func.call_args_list[3][0],
is_((['filename2.nc'], 'var4', 'cis')))
def plot_cmd(main_arguments):
"""
Main routine for handling calls to the 'plot' command.
Reads in the data files specified and passes the rest of the arguments to the plot function.
:param main_arguments: The command line arguments
"""
from cis.plotting.formatted_plot import Plotter
from cis.data_io.data_reader import DataReader
data = DataReader().read_datagroups(main_arguments.datagroups)
# We have to pop off the arguments which plot isn't expecting so that it treats everything else as an mpl kwarg
main_arguments = vars(main_arguments)
_ = main_arguments.pop('command')
_ = main_arguments.pop("quiet")
_ = main_arguments.pop("verbose")
_ = main_arguments.pop("force_overwrite")
_ = main_arguments.pop("output_var", None)
layer_opts = [{
k: v
for k, v in d.items()
if k not in ['variables', 'filenames', 'product']
} for d in main_arguments.pop('datagroups')]
Plotter(data, layer_opts=layer_opts, **main_arguments)
def aggregate_cmd(main_arguments):
"""
Main routine for handling calls to the aggregation command.
:param main_arguments: The command line arguments (minus the aggregate command)
"""
import cis.exceptions as ex
from cis.data_io.gridded_data import GriddedDataList
if len(main_arguments.datagroups) > 1:
__error_occurred("Aggregation can only be performed on one data group")
input_group = main_arguments.datagroups[0]
data = DataReader().read_single_datagroup(input_group)
if isinstance(data, GriddedDataList):
logging.warning(
"The aggregate command is deprecated for GriddedData and will not be supported in future "
"versions of CIS. Please use 'collapse' instead.")
if any(v is not None for v in main_arguments.grid.values()):
raise ex.InvalidCommandLineOptionError(
"Grid specifications are not supported for Gridded aggregation."
)
output = data.collapsed(list(main_arguments.grid.keys()),
how=input_group.get("kernel", ''))
else:
output = data.aggregate(how=input_group.get("kernel", ''),
**main_arguments.grid)
output.save_data(main_arguments.output)
def test_GIVEN_multiple_variable_gridded_WHEN_read_data_THEN_GriddedDataList_returned(
self):
variables = ['var1', 'var2']
filenames = 'filename1'
product = None
gridded_data = make_square_5x3_2d_cube()
gridded_data.__class__ = GriddedData
get_data_func = MagicMock(return_value=gridded_data)
reader = DataReader(get_data_func=get_data_func)
data = reader.read_data_list(filenames, variables, product)
# Check the data read function is called correctly
assert_that(get_data_func.call_count, is_(2))
first_call_args = get_data_func.call_args_list[0][0]
second_call_args = get_data_func.call_args_list[1][0]
assert_that(first_call_args[0], is_([filenames]))
assert_that(first_call_args[1], is_(variables[0]))
assert_that(second_call_args[1], is_(variables[1]))
assert_that(first_call_args[2], is_(product))
# Check the data we got back is as expected
assert_that(data, instance_of(GriddedDataList))
assert_that(data[0].data.tolist(),
is_(make_square_5x3_2d_cube().data.tolist()))
assert_that(data[1].data.tolist(), is_(data[0].data.tolist()))
def col_cmd(main_arguments):
"""
Main routine for handling calls to the collocate ('col') command.
:param main_arguments: The command line arguments (minus the col command)
"""
from cis.collocation.col_framework import get_kernel
from cis.parse import check_boolean
# Read the sample data
missing_data_for_missing_sample = False
if main_arguments.samplevariable is not None:
sample_data = DataReader().read_data_list(
main_arguments.samplefiles, main_arguments.samplevariable,
main_arguments.sampleproduct)[0]
missing_data_for_missing_sample = True
else:
sample_data = DataReader().read_coordinates(
main_arguments.samplefiles, main_arguments.sampleproduct)
# Unpack the sample options
col_name, col_options = main_arguments.samplegroup.get(
'collocator', ('', {}))
kern_name, kern_options = main_arguments.samplegroup.get(
'kernel', ('', {}))
missing_data_for_missing_sample = check_boolean(
col_options.pop('missing_data_for_missing_sample',
str(missing_data_for_missing_sample)), logging)
kernel = get_kernel(kern_name)(**kern_options) if kern_name else None
for input_group in main_arguments.datagroups:
# Then collocate each datagroup
data = DataReader().read_single_datagroup(input_group)
output = data.collocated_onto(
sample_data,
how=col_name,
kernel=kernel,
missing_data_for_missing_sample=missing_data_for_missing_sample,
**col_options)
output.save_data(main_arguments.output)
def test_GIVEN_no_matching_variables_found_overall_WHEN_read_data_THEN_raises_Error(
self):
variables = ['test?.hdf', '*.nc']
file_vars = ['sample_file.hdf', 'aeronet.lev20']
filenames = 'filename1'
get_data_func = MagicMock()
get_var_func = MagicMock(return_value=file_vars)
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
with self.assertRaises(ValueError):
reader.read_data_list(filenames, variables)[0]
def test_GIVEN_multiple_variable_mix_of_gridded_ungridded_WHEN_read_data_THEN_raises_TypeError(
self):
variables = ['var1', 'var2']
filenames = 'filename1'
product = None
gridded_data = make_square_5x3_2d_cube()
gridded_data.__class__ = GriddedData
ungridded_data = make_regular_2d_ungridded_data()
get_data_func = MagicMock(side_effect=[gridded_data, ungridded_data])
reader = DataReader(get_data_func=get_data_func)
with self.assertRaises(TypeError):
data = reader.read_data_list(filenames, variables, product)[0]
def test_GIVEN_no_matching_variables_for_wildcards_WHEN_read_data_THEN_no_Error(
self):
variables = ['aeronet.lev20', '*.nc', 'test?.hdf']
file_vars = ['aeronet.lev20', 'var2.hdf']
filenames = 'filename1'
get_data_func = MagicMock(
return_value=make_regular_2d_ungridded_data())
get_var_func = MagicMock(return_value=file_vars)
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
reader.read_data_list(filenames, variables)[0]
assert_that(reader._get_data_func.call_count, is_(1))
assert_that(reader._get_data_func.call_args_list[0][0][1],
is_('aeronet.lev20'))
def test_GIVEN_aliases_missing_WHEN_read_datagroups_THEN_read_OK_aliases_default_to_var_names(
self):
datagroup = {
'variables': ['var1'],
'filenames': ['filename1.nc'],
'product': None
}
var1 = make_from_cube(make_square_5x3_2d_cube())
get_data_func = MagicMock(side_effect=[var1])
get_var_func = MagicMock(side_effect=['var1'])
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
data = reader.read_datagroups([datagroup])
assert_that(data[0].var_name, is_('dummy'))
def evaluate_cmd(main_arguments):
"""
Main routine for handling calls to the evaluation command
:param main_arguments: The command line arguments (minus the eval command)
"""
from cis.evaluate import Calculator
data_reader = DataReader()
data_list = data_reader.read_datagroups(main_arguments.datagroups)
calculator = Calculator()
result = calculator.evaluate(data_list, main_arguments.expr,
main_arguments.output_var,
main_arguments.units,
main_arguments.attributes)
result.save_data(main_arguments.output)
def test_GIVEN_aliases_WHEN_read_datagroups_THEN_output_data_has_aliases(
self):
datagroup = {
'variables': ['var1'],
'filenames': ['filename1.nc'],
'product': None,
'aliases': ['alias1']
}
get_data_func = MagicMock(
return_value=make_from_cube(make_square_5x3_2d_cube()))
get_var_func = MagicMock(return_value=['var1'])
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
data = reader.read_datagroups([datagroup])
assert_that(data[0].alias, is_('alias1'))
def test_GIVEN_not_enough_aliases_WHEN_read_datagroups_THEN_raises_ValueError(
self):
datagroup = {
'variables': ['var1', 'var2'],
'filenames': ['filename1.nc'],
'product': None,
'aliases': ['alias1']
}
var1 = make_from_cube(make_square_5x3_2d_cube())
var2 = make_from_cube(make_square_5x3_2d_cube())
get_data_func = MagicMock(side_effect=[var1, var2])
get_var_func = MagicMock(side_effect=['var1', 'var2'])
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
with self.assertRaises(ValueError):
data = reader.read_datagroups([datagroup])
def test_GIVEN_wildcards_WHEN_read_data_THEN_matching_variables_identified(
self):
variables = ['*.nc', 'test?.hdf']
file_vars = [
'aeronet.lev20', 'var2.hdf', 'netcdf1.nc', 'netcdf3.nc',
'test.hdf', 'test1.hdf'
]
should_match = ['netcdf1.nc', 'netcdf3.nc', 'test1.hdf']
filenames = 'filename1'
get_data_func = MagicMock(
return_value=make_regular_2d_ungridded_data())
get_var_func = MagicMock(return_value=file_vars)
reader = DataReader(get_data_func=get_data_func,
get_variables_func=get_var_func)
reader.read_data_list(filenames, variables)[0]
assert_that(reader._get_data_func.call_count, is_(len(should_match)))
for i in range(len(should_match)):
assert_that(reader._get_data_func.call_args_list[i][0][1],
is_(should_match[i]))
def test_GIVEN_single_variable_ungridded_WHEN_read_data_THEN_GriddedData_returned(
self):
variables = 'var1'
filenames = 'filename1'
product = None
get_data_func = MagicMock(
return_value=make_regular_2d_ungridded_data())
reader = DataReader(get_data_func=get_data_func)
data = reader.read_data_list(filenames, variables, product)[0]
# Check the data read function is called correctly
assert_that(get_data_func.call_count, is_(1))
call_args = get_data_func.call_args_list[0][0]
assert_that(call_args[0], is_([filenames]))
assert_that(call_args[1], is_(variables))
assert_that(call_args[2], is_(product))
# Check the data we got back is as expected
assert_that(data, instance_of(UngriddedData))
assert_that(data.data.tolist(),
is_(make_regular_2d_ungridded_data().data.tolist()))
def subset_cmd(main_arguments):
"""
Main routine for handling calls to the subset command.
:param main_arguments: The command line arguments (minus the subset command)
"""
import cis.exceptions as ex
if len(main_arguments.datagroups) > 1:
__error_occurred("Subsetting can only be performed on one data group")
data = DataReader().read_single_datagroup(main_arguments.datagroups[0])
subset = data.subset(**main_arguments.limits)
if subset is None:
# Constraints exclude all data.
raise ex.NoDataInSubsetError(
"No output created - constraints exclude all data")
subset.save_data(main_arguments.output)
def collapse_cmd(main_arguments):
"""
Main routine for handling calls to the collapse command.
:param main_arguments: The command line arguments (minus the collapse command)
"""
from cis.data_io.ungridded_data import UngriddedDataList
if len(main_arguments.datagroups) > 1:
__error_occurred("Collapse can only be performed on one data group")
input_group = main_arguments.datagroups[0]
data = DataReader().read_single_datagroup(input_group)
if isinstance(data, UngriddedDataList):
logging.error(
"The collapse command can only be performed on gridded data. "
"Please use 'aggregate' instead.")
output = data.collapsed(main_arguments.dimensions,
how=input_group.get("kernel", ''))
output.save_data(main_arguments.output)
def read_data_list(filenames, variables, product=None, aliases=None):
"""
Read multiple data objects from a list of files. Files can be either gridded or ungridded but not a mix of both.
:param filenames: The filenames of the files to read. This can be either a single filename as a string, a comma
separated list, or a :class:`list` of string filenames. Filenames can include directories which will be expanded to
include all files in that directory, or wildcards such as ``*`` or ``?``.
:type filenames: string or list
:param variables: One or more variables to read from the files
:type variables: string or list
:param str product: The name of the data reading plugin to use to read the data (e.g. ``Cloud_CCI``).
:param aliases: List of aliases to put on each variable's data object as an alternative means of identifying them.
:type aliases: string or list
:return: A list of the data read out (either a :class:`GriddedDataList` or :class:`UngriddedDataList` depending on
the type of data contained in the files)
"""
from cis.data_io.data_reader import DataReader, expand_filelist
try:
file_set = expand_filelist(filenames)
except ValueError as e:
raise IOError(e)
if len(file_set) == 0:
raise IOError("No files found which match: {}".format(filenames))
return DataReader().read_data_list(file_set, variables, product, aliases)
