def test_ref_missing_required():
spec_builder = datacraft.spec_builder()
spec_builder.add_ref('values', datacraft.builder.values([1, 2, 3]))
spec_builder.ref('points_at_nothing')
generator = spec_builder.build().generator(1)
with pytest.raises(datacraft.SpecException):
next(generator)
def test_keywords_are_now_valid(keyword):
spec_builder = datacraft.spec_builder()
spec_builder.values('field', 84)
mapping = {'field': keyword}
formula = '{{%s}}/2' % keyword
spec_builder.calculate('meaning_of_life', fields=mapping, formula=formula)
assert next(spec_builder.build().generator(1))['meaning_of_life'] == 42.0
def _build_csv_spec(field_name, **config):
base = {
"datafile": "weighted.csv"
}
base.update(config)
return datacraft.spec_builder() \
.add_field(field_name, datacraft.builder.weighted_csv(**base)) \
.build()
def test_calculate_missing_required(fields, refs, formula):
spec_builder = datacraft.spec_builder()
spec_builder.calculate('something_interesting',
fields=fields,
refs=refs,
formula=formula)
with pytest.raises(datacraft.SpecException):
next(spec_builder.build().generator(1))
def test_calculate_valid_from_builder():
spec_builder = datacraft.spec_builder()
spec_builder.values('field', 21)
mapping = {'field': 'a'}
formula = '{{a}} * 2'
spec_builder.calculate('meaning_of_life', fields=mapping, formula=formula)
assert next(spec_builder.build().generator(
1, enforce_schema=True))['meaning_of_life'] == 42.0
def test_weighted_csv_from_builder():
spec = datacraft.spec_builder() \
.weighted_csv('status', datafile='weighted.csv', column='status', weight_column='weight', headers=True) \
.to_spec()
gen = spec.generator(1, enforce_schema=True, data_dir=test_dir)
val = next(gen)
assert 'status' in val
assert val['status'].isnumeric()
def test_row_level_sampling():
builder = datacraft.spec_builder()
config = {"datafile": "test.csv", "headers": True, "sample_rows": "on"}
builder.config_ref('csvconfig', **config)
builder.csv('status', config_ref='csvconfig', column=1)
builder.csv('status_description', config_ref='csvconfig', column=2)
spec = builder.build()
with open(os.sep.join([test_dir, 'test.csv'])) as handle:
parts = [line.strip().split(',') for line in handle.readlines()]
valid_mappings = {parts[0]: parts[1] for parts in parts}
gen = spec.generator(10, data_dir=test_dir)
for i in range(10):
value = next(gen)
status = value['status']
assert status in valid_mappings
assert value['status_description'] == valid_mappings[status]
def _build_csv_spec(field_name, **config):
base = {"datafile": "test.csv", "headers": True, "column": 1}
base.update(config)
return datacraft.spec_builder() \
.add_field(field_name, datacraft.builder.csv(**base)) \
.build()
def test_random_range_invalid_data_type(field_type, data):
with pytest.raises(datacraft.SpecException):
builder = datacraft.spec_builder()
spec = builder.add_field('test', {'type': field_type, 'data': data}).build()
next(spec.generator(1))['test']
def test_random_range_parameterized(field_type, data, lower, upper):
builder = datacraft.spec_builder()
spec = builder.add_field('test', {'type': field_type, 'data': data}).build()
val = next(spec.generator(1, enforce_schema=True))['test']
assert lower <= float(val) <= upper
def test_ref_with_ref_name():
spec_builder = datacraft.spec_builder()
spec_builder.add_ref('values', datacraft.builder.values([1, 2, 3]))
spec_builder.ref('points_at_values', ref_name='values')
generator = spec_builder.build().generator(1)
assert next(generator) == {'points_at_values': 1}
def test_config_ref_in_refs():
spec_builder = datacraft.spec_builder()
spec_builder.refs().config_ref('test', key1='value1', key2='value2')
spec = spec_builder.build()
assert 'refs' in spec
assert spec['refs'].get('test') == {'type': 'config_ref', 'config': {'key1': 'value1', 'key2': 'value2'}}
def test_ref_with_data_as_name():
spec_builder = datacraft.spec_builder()
spec_builder.add_ref('values', datacraft.builder.values([1, 2, 3]))
spec_builder.ref('points_at_values_with_prefix', data='values', prefix='@')
generator = spec_builder.build().generator(1)
assert next(generator) == {'points_at_values_with_prefix': '@1'}
def builder():
return datacraft.spec_builder()