def test_step_resource_transform():
source = Package("data/package/datapackage.json")
target = transform(
source,
steps=[
steps.resource_update(name="data", title="It's our data"),
steps.resource_remove(name="data2"),
steps.resource_add(name="data2", path="data2.csv"),
steps.resource_transform(
name="data",
steps=[
steps.cell_set(field_name="description", value="Zeroed"),
steps.cell_set(field_name="amount", value=0),
],
),
steps.resource_transform(
name="data2",
steps=[
steps.cell_set(field_name="comment", value="It works!"),
],
),
],
)
assert target.resource_names == ["data", "data2"]
assert target.get_resource("data").read_rows() == [
{"id": "A3001", "name": "Taxes", "description": "Zeroed", "amount": 0},
{"id": "A5032", "name": "Parking Fees", "description": "Zeroed", "amount": 0},
]
assert target.get_resource("data2").read_rows() == [
{"parent": "A3001", "comment": "It works!"},
{"parent": "A3001", "comment": "It works!"},
{"parent": "A5032", "comment": "It works!"},
]
def test_step_cell_set():
source = Resource(path="data/transform.csv")
source.infer()
target = transform(
source,
steps=[
steps.cell_set(field_name="population", value=100),
],
)
assert target.schema == source.schema
assert target.read_rows() == [
{
"id": 1,
"name": "germany",
"population": 100
},
{
"id": 2,
"name": "france",
"population": 100
},
{
"id": 3,
"name": "spain",
"population": 100
},
]
def test_step_table_write(tmpdir):
path = str(tmpdir.join("table.json"))
# Write
source = Resource("data/transform.csv")
transform(
source,
steps=[
steps.cell_set(field_name="population", value=100),
steps.table_write(path=path),
],
)
# Read
resource = Resource(path=path)
assert resource.read_rows() == [
{
"id": 1,
"name": "germany",
"population": 100
},
{
"id": 2,
"name": "france",
"population": 100
},
{
"id": 3,
"name": "spain",
"population": 100
},
]
def test_step_table_validate():
source = Resource("data/transform.csv")
target = transform(
source,
steps=[
steps.cell_set(field_name="population", value="bad"),
steps.table_validate(),
],
)
assert target.schema == {
"fields": [
{
"name": "id",
"type": "integer"
},
{
"name": "name",
"type": "string"
},
{
"name": "population",
"type": "integer"
},
]
}
with pytest.raises(FrictionlessException) as excinfo:
target.read_rows()
error = excinfo.value.error
assert error.code == "step-error"
assert error.note.count('type is "integer/default"')
def test_step_table_validate():
source = Resource(path="data/transform.csv")
source.infer(only_sample=True)
target = transform(
source,
steps=[
steps.cell_set(field_name="population", value="bad"),
steps.table_validate(),
],
)
assert target.schema == source.schema
with pytest.raises(exceptions.FrictionlessException) as excinfo:
target.read_rows()
error = excinfo.value.error
assert error.code == "step-error"
assert error.note.count('type is "integer/default"')
def test_step_cell_set():
source = Resource(path="data/transform.csv")
target = transform(
source,
steps=[
steps.cell_set(field_name="population", value=100),
],
)
assert target.schema == {
"fields": [
{"name": "id", "type": "integer"},
{"name": "name", "type": "string"},
{"name": "population", "type": "integer"},
]
}
assert target.read_rows() == [
{"id": 1, "name": "germany", "population": 100},
{"id": 2, "name": "france", "population": 100},
{"id": 3, "name": "spain", "population": 100},
]