def test_configuration(self):
"""Test the configuration function."""
# Test getting of all config. and store original values to test on:
org = cfdm.configuration()
self.assertIsInstance(org, dict)
self.assertEqual(len(org), 3)
org_atol = org["atol"]
self.assertIsInstance(org_atol, float)
org_rtol = org["rtol"]
self.assertIsInstance(org_rtol, float)
org_ll = org["log_level"] # will be 'DISABLE' as disable for test
self.assertIsInstance(org_ll, str)
# Store some sensible values to reset items to for testing,
# ensure these are kept to be different to the defaults:
atol_rtol_reset_value = 7e-10
ll_reset_value = "DETAIL"
# Test the setting of each lone item:
cfdm.configuration(atol=atol_rtol_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], atol_rtol_reset_value)
self.assertEqual(post_set["rtol"], org_rtol)
self.assertEqual(post_set["log_level"], org_ll)
cfdm.configuration(atol=org_atol) # reset to org
cfdm.configuration(rtol=atol_rtol_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], org_atol)
self.assertEqual(post_set["rtol"], atol_rtol_reset_value)
self.assertEqual(post_set["log_level"], org_ll)
# don't reset to org this time to test change persisting...
# Note setting of previous items persist, e.g. atol above
cfdm.configuration(log_level=ll_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], org_atol)
self.assertEqual(
post_set["rtol"], atol_rtol_reset_value
) # since changed it above
self.assertEqual(post_set["log_level"], ll_reset_value)
# Test the setting of more than one, but not all, items simultaneously:
new_atol_rtol_reset_value = 5e-18
new_ll_reset_value = "DEBUG"
cfdm.configuration(
rtol=new_atol_rtol_reset_value, log_level=new_ll_reset_value
)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], org_atol)
self.assertEqual(post_set["rtol"], new_atol_rtol_reset_value)
self.assertEqual(post_set["log_level"], new_ll_reset_value)
# Test setting all possible items simultaneously (to originals):
cfdm.configuration(
atol=org_atol, # same as current setting, testing on 'no change'
rtol=org_rtol,
log_level=org_ll,
)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], org_atol)
self.assertEqual(post_set["rtol"], org_rtol)
self.assertEqual(post_set["log_level"], org_ll)
# Test edge cases & invalid inputs...
# ... 1. User might set '0' or 'True' in some cases, which is
# somewhat a risk area for error as 0 is Falsy & True a Bool:
cfdm.configuration(rtol=0, atol=0.0, log_level=0)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], 0.0)
self.assertEqual(post_set["rtol"], 0.0)
self.assertEqual(post_set["log_level"], "DISABLE") # DISABLE == 0
cfdm.configuration(log_level=True) # deprecated but valid value
# Deprecated True is converted to value 'WARNING' by log_level
self.assertEqual(cfdm.configuration()["log_level"], "WARNING")
# 2. None as an input kwarg rather than as a default:
cfdm.configuration(atol=None, log_level=None, rtol=org_rtol)
post_set = cfdm.configuration()
self.assertEqual(post_set["atol"], 0.0) # 0.0 as set above
self.assertEqual(post_set["rtol"], org_rtol)
self.assertEqual(post_set["log_level"], "WARNING") # as set above
# 3. Gracefully error with useful error messages with invalid inputs:
with self.assertRaises(ValueError):
cfdm.configuration(rtol="bad")
with self.assertRaises(ValueError):
cfdm.configuration(log_level=7)
# 4. Check invalid kwarg given logic processes **kwargs:
with self.assertRaises(TypeError):
cfdm.configuration(bad_kwarg=1e-15)
old = cfdm.configuration()
try:
cfdm.configuration(atol=888, rtol=999, log_level="BAD")
except ValueError:
self.assertEqual(cfdm.configuration(), old)
else:
raise RuntimeError(
"A ValueError should have been raised, but wasn't"
)
def test_Configuration(self):
"""Test the Configuration class."""
c = cfdm.configuration()
self.assertIsInstance(repr(c), str)
self.assertEqual(str(c), str(dict(**c)))
def test_configuration(self):
if self.test_only and inspect.stack()[0][3] not in self.test_only:
return
# Test getting of all config. and store original values to test on:
org = cfdm.configuration()
self.assertIsInstance(org, dict)
self.assertEqual(len(org), 3)
org_atol = org['atol']
self.assertIsInstance(org_atol, float)
org_rtol = org['rtol']
self.assertIsInstance(org_rtol, float)
org_ll = org['log_level'] # will be 'DISABLE' as disable for test
self.assertIsInstance(org_ll, str)
# Store some sensible values to reset items to for testing,
# ensure these are kept to be different to the defaults:
atol_rtol_reset_value = 7e-10
ll_reset_value = 'DETAIL'
# Test the setting of each lone item:
cfdm.configuration(atol=atol_rtol_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], atol_rtol_reset_value)
self.assertEqual(post_set['rtol'], org_rtol)
self.assertEqual(post_set['log_level'], org_ll)
cfdm.configuration(atol=org_atol) # reset to org
cfdm.configuration(rtol=atol_rtol_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], org_atol)
self.assertEqual(post_set['rtol'], atol_rtol_reset_value)
self.assertEqual(post_set['log_level'], org_ll)
# don't reset to org this time to test change persisting...
# Note setting of previous items persist, e.g. atol above
cfdm.configuration(log_level=ll_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], org_atol)
self.assertEqual(post_set['rtol'],
atol_rtol_reset_value) # since changed it above
self.assertEqual(post_set['log_level'], ll_reset_value)
# Test the setting of more than one, but not all, items simultaneously:
new_atol_rtol_reset_value = 5e-18
new_ll_reset_value = 'DEBUG'
cfdm.configuration(rtol=new_atol_rtol_reset_value,
log_level=new_ll_reset_value)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], org_atol)
self.assertEqual(post_set['rtol'], new_atol_rtol_reset_value)
self.assertEqual(post_set['log_level'], new_ll_reset_value)
# Test setting all possible items simultaneously (to originals):
cfdm.configuration(
atol=org_atol, # same as current setting, testing on 'no change'
rtol=org_rtol,
log_level=org_ll)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], org_atol)
self.assertEqual(post_set['rtol'], org_rtol)
self.assertEqual(post_set['log_level'], org_ll)
# Test edge cases & invalid inputs...
# ... 1. User might set '0' or 'True' in some cases, which is
# somewhat a risk area for error as 0 is Falsy & True a Bool:
cfdm.configuration(rtol=0, atol=0.0, log_level=0)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], 0.0)
self.assertEqual(post_set['rtol'], 0.0)
self.assertEqual(post_set['log_level'], 'DISABLE') # DISABLE == 0
cfdm.configuration(log_level=True) # deprecated but valid value
# Deprecated True is converted to value 'WARNING' by log_level
self.assertEqual(cfdm.configuration()['log_level'], 'WARNING')
# 2. None as an input kwarg rather than as a default:
cfdm.configuration(atol=None, log_level=None, rtol=org_rtol)
post_set = cfdm.configuration()
self.assertEqual(post_set['atol'], 0.0) # 0.0 as set above
self.assertEqual(post_set['rtol'], org_rtol)
self.assertEqual(post_set['log_level'], 'WARNING') # as set above
# 3. Gracefully error with useful error messages with invalid inputs:
with self.assertRaises(ValueError):
cfdm.configuration(rtol='bad')
with self.assertRaises(ValueError):
cfdm.configuration(log_level=7)
# 4. Check invalid kwarg given logic processes **kwargs:
with self.assertRaises(TypeError):
cfdm.configuration(bad_kwarg=1e-15)