def set_product_attributes(cube, product):
"""
Set attributes on an output cube of type matching a key string in the
improver.metadata.constants.attributes.DATASET_ATTRIBUTES dictionary.
Args:
cube (iris.cube.Cube):
Cube containing product data
product (str):
String describing product type, which is a key in the
DATASET_ATTRIBUTES dictionary.
Returns:
iris.cube.Cube:
Cube with updated attributes
"""
try:
original_title = cube.attributes["title"]
except KeyError:
original_title = ""
try:
dataset_attributes = DATASET_ATTRIBUTES[product]
except KeyError:
options = list(DATASET_ATTRIBUTES.keys())
raise ValueError("product '{}' not available (options: {})".format(
product, options))
updated_cube = amend_metadata(cube, attributes=dataset_attributes)
if STANDARD_GRID_TITLE_STRING in original_title:
updated_cube.attributes["title"] += " on {}".format(
STANDARD_GRID_TITLE_STRING)
return updated_cube
def test_convert_units(self):
"""Test amend_metadata updates attributes OK. """
changes = "Celsius"
cube = set_up_variable_cube(np.ones((3, 3), dtype=np.float32),
units='K')
result = amend_metadata(cube, units=changes)
self.assertEqual(result.units, "Celsius")
def test_attributes_deleted(self):
"""Test amend_metadata updates attributes OK. """
attributes = {'attribute_to_update': 'delete'}
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
attributes=attributes)
self.assertFalse('attribute_to_update' in result.attributes)
def test_basic(self):
"""Test that the function returns a Cube and the input cube is not
modified. """
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype)
self.assertIsInstance(result, Cube)
self.assertEqual(result.name(), 'new_cube_name')
self.assertNotEqual(self.cube.name(), 'new_cube_name')
def test_attributes_updated_and_added(self):
"""Test amend_metadata updates and adds attributes OK. """
attributes = {
'attribute_to_update': 'second_value',
'new_attribute': 'new_value'
}
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
attributes=attributes)
self.assertEqual(result.attributes['attribute_to_update'],
'second_value')
self.assertEqual(result.attributes['new_attribute'], 'new_value')
def test_cell_method_updated_and_added(self):
"""Test amend_metadata updates and adds a cell method. """
cell_methods = {
"1": {
"action": "add",
"method": "point",
"coords": "time"
}
}
cm = deepcopy(cell_methods)
cm["1"].pop("action")
expected_cell_method = iris.coords.CellMethod(**cm["1"])
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
cell_methods=cell_methods)
self.assertTrue(expected_cell_method in result.cell_methods)
def test_cell_method_deleted(self):
"""Test amend_metadata updates attributes OK. """
cell_methods = {
"1": {
"action": "delete",
"method": "point",
"coords": "time"
}
}
cm = deepcopy(cell_methods)
cm["1"].pop("action")
cell_method = iris.coords.CellMethod(**cm["1"])
self.cube.cell_methods = (cell_method, )
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
cell_methods=cell_methods)
self.assertEqual(result.cell_methods, ())
def test_coords_deleted_and_adds(self):
"""Test amend metadata deletes and adds coordinate. """
coords = {
self.threshold_coord: 'delete',
'new_coord': {
'points': [2.0]
}
}
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
coordinates=coords)
found_key = self.threshold_coord in [
coord.name() for coord in result.coords()
]
self.assertFalse(found_key)
self.assertArrayEqual(
result.coord('new_coord').points, np.array([2.0]))
def test_coords_updated(self):
"""Test amend_metadata returns a Cube and updates coord correctly. """
updated_coords = {
self.threshold_coord: {
'points': [2.0]
},
'time': {
'points': [1447896600, 1447900200]
}
}
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
coordinates=updated_coords)
self.assertArrayEqual(
result.coord(self.threshold_coord).points, np.array([2.0]))
self.assertArrayEqual(
result.coord('time').points, np.array([1447896600, 1447900200]))
def test_warnings_on_works(self, warning_list=None):
"""Test amend_metadata raises warnings """
updated_attributes = {'new_attribute': 'new_value'}
updated_coords = {self.threshold_coord: {'points': [2.0]}}
warning_msg_attr = "Adding or updating attribute"
warning_msg_coord = "Updated coordinate"
result = amend_metadata(self.cube,
name='new_cube_name',
data_type=np.dtype,
coordinates=updated_coords,
attributes=updated_attributes,
warnings_on=True)
self.assertTrue(
any(item.category == UserWarning for item in warning_list))
self.assertTrue(
any(warning_msg_attr in str(item) for item in warning_list))
self.assertTrue(
any(warning_msg_coord in str(item) for item in warning_list))
self.assertEqual(result.attributes['new_attribute'], 'new_value')
def process(output_data,
target_grid=None,
source_landsea=None,
metadata_dict=None,
regrid_mode='bilinear',
extrapolation_mode='nanmask',
landmask_vicinity=25000,
fix_float64=False):
"""Standardises a cube by one or more of regridding, updating meta-data etc
Standardise a source cube. Available options are regridding
(bi-linear or nearest-neighbour, optionally with land-mask
awareness), updating meta-data and converting float64 data to
float32. A check for float64 data compliance can be made by only
specifying a source cube with no other arguments.
Args:
output_data (iris.cube.Cube):
Output cube. If the only argument, then it is checked bor float64
data.
target_grid (iris.cube.Cube):
If specified, then regridding of the source against the target
grid is enabled. If also using landmask-aware regridding then this
must be land_binary_mask data.
Default is None.
source_landsea (iris.cube.Cube):
A cube describing the land_binary_mask on the source-grid if
coastline-aware regridding is required.
Default is None.
metadata_dict (dict):
Dictionary containing required changes that will be applied to
the metadata.
Default is None.
regrid_mode (str):
Selects which regridding techniques to use. Default uses
iris.analysis.Linear(); "nearest" uses Nearest() (Use for less
continuous fields, e.g precipitation.); "nearest-with-mask"
ensures that target data are sources from points with the same
mask value (Use for coast-line-dependant variables
like temperature).
extrapolation_mode (str):
Mode to use for extrapolating data into regions beyond the limits
of the source_data domain. Refer to online documentation for
iris.analysis.
Modes are -
extrapolate -The extrapolation points will take their values
from the nearest source point.
nan - The extrapolation points will be set to NaN.
error - A ValueError exception will be raised notifying an attempt
to extrapolate.
mask - The extrapolation points will always be masked, even if
the source data is not a MaskedArray.
nanmask - If the source data is a MaskedArray the extrapolation
points will be masked. Otherwise they will be set to NaN.
Defaults is 'nanmask'.
landmask_vicinity (float):
Radius of vicinity to search for a coastline, in metres.
Defaults is 25000 m
fix_float64 (bool):
If True, checks and fixes cube for float64 data. Without this
option an exception will be raised if float64 data is found but no
fix applied.
Default is False.
Returns:
iris.cube.Cube:
Processed cube.
Raises:
ValueError:
If source landsea is supplied but regrid mode not
nearest-with-mask.
ValueError:
If source landsea is supplied but not target grid.
ValueError:
If regrid_mode is "nearest-with-mask" but no landmask cube has
been provided.
Warns:
warning:
If the 'source_landsea' did not have a cube named land_binary_mask.
warning:
If the 'target_grid' did not have a cube named land_binary_mask.
"""
if (source_landsea and "nearest-with-mask" not in regrid_mode):
msg = ("Land-mask file supplied without appropriate regrid_mode. "
"Use --regrid_mode=nearest-with-mask.")
raise ValueError(msg)
if source_landsea and not target_grid:
msg = ("Cannot specify input_landmask_filepath without "
"target_grid_filepath")
raise ValueError(msg)
# Process
# Re-grid with options:
check_cube_not_float64(output_data, fix=fix_float64)
# if a target grid file has been specified, then regrid optionally
# applying float64 data check, metadata change, Iris nearest and
# extrapolation mode as required.
if target_grid:
regridder = iris.analysis.Linear(extrapolation_mode=extrapolation_mode)
if regrid_mode in ["nearest", "nearest-with-mask"]:
regridder = iris.analysis.Nearest(
extrapolation_mode=extrapolation_mode)
output_data = output_data.regrid(target_grid, regridder)
if regrid_mode in ["nearest-with-mask"]:
if not source_landsea:
msg = ("An argument has been specified that requires an input "
"landmask cube but none has been provided")
raise ValueError(msg)
if "land_binary_mask" not in source_landsea.name():
msg = ("Expected land_binary_mask in input_landmask cube "
"but found {}".format(repr(source_landsea)))
warnings.warn(msg)
if "land_binary_mask" not in target_grid.name():
msg = ("Expected land_binary_mask in target_grid cube "
"but found {}".format(repr(target_grid)))
warnings.warn(msg)
output_data = RegridLandSea(
vicinity_radius=landmask_vicinity).process(
output_data, source_landsea, target_grid)
target_grid_attributes = ({
k: v
for (k, v) in target_grid.attributes.items()
if 'mosg__' in k or 'institution' in k
})
output_data = amend_metadata(output_data,
attributes=target_grid_attributes)
# Change metadata only option:
# if output file path and json metadata file specified,
# change the metadata
if metadata_dict:
output_data = amend_metadata(output_data, **metadata_dict)
check_cube_not_float64(output_data, fix=fix_float64)
return output_data
def process(self,
cube_list,
new_diagnostic_name,
revised_coords=None,
revised_attributes=None,
expanded_coord=None):
"""
Create a combined cube.
Args:
cube_list (iris.cube.CubeList):
Cube List contain the cubes to combine.
new_diagnostic_name (str):
New name for the combined diagnostic.
revised_coords (dict or None):
Revised coordinates for combined cube.
revised_attributes (dict or None):
Revised attributes for combined cube.
expanded_coord (dict or None):
Coordinates to be expanded as a key, with the value
indicating whether the upper or mid point of the coordinate
should be used as the point value, e.g.
{'time': 'upper'}.
Returns:
iris.cube.Cube:
Cube containing the combined data.
Raises:
TypeError: If cube_list is not an iris.cube.CubeList.
ValueError: If the cubelist contains only one cube.
"""
if not isinstance(cube_list, iris.cube.CubeList):
msg = ('Expecting data to be an instance of iris.cube.CubeList '
'but is {}.'.format(type(cube_list)))
raise TypeError(msg)
if len(cube_list) < 2:
msg = 'Expecting 2 or more cubes in cube_list'
raise ValueError(msg)
# resulting cube will be based on the first cube.
data_type = cube_list[0].dtype
result = cube_list[0].copy()
for ind in range(1, len(cube_list)):
cube1, cube2 = (resolve_metadata_diff(
result.copy(),
cube_list[ind].copy(),
warnings_on=self.warnings_on))
result = self.combine(cube1, cube2)
if self.operation == 'mean':
result.data = result.data / len(cube_list)
# If cube has coord bounds that we want to expand
if expanded_coord:
result = expand_bounds(result, cube_list, expanded_coord)
result = amend_metadata(result,
new_diagnostic_name,
data_type,
revised_coords,
revised_attributes,
warnings_on=self.warnings_on)
return result
