def test_outer():
out = outer([1, 2], [1, 2, 3])
assert_frame_equal(out, DataFrame([[1, 2, 3], [2, 4, 6]]))
out = outer([1, 2], [1, 2, 3], fun=paste0)
assert_frame_equal(out, DataFrame([["11", "12", "13"], ["21", "22",
"23"]]))
def test_data_context():
df = DataFrame(dict(a=[1, 2], b=[3, 4]))
with data_context(df) as _:
out = expandgrid(f.a, f.b)
assert_frame_equal(
out,
DataFrame({
"a": [1, 1, 2, 2],
"b": [3, 4, 3, 4],
}),
)
def test_expandgrid():
df = expandgrid([1, 2], [3, 4])
assert_frame_equal(
df,
DataFrame({
"[1, 2]": [1, 1, 2, 2],
"[3, 4]": [3, 4, 3, 4],
}),
)
def test_group_split_works_with_subclasses_implementing_group_by_ungroup():
# test_that("group_split() works with subclasses implementing group_by()
# / ungroup()", {
class TibbleGrouped1(TibbleGrouped):
...
df = TibbleGrouped1.from_groupby(DataFrame(dict(x=[1, 2, 2])).groupby("x"))
out = group_split.list(df, f.x)
assert len(out) == 2
assert out[0].equals(df.iloc[[0], :])
assert out[1].equals(df.iloc[[1, 2], :].reset_index(drop=True))
def test_rowcolnames():
df = DataFrame(dict(x=[1, 2, 3]))
assert colnames(df) == ["x"]
assert rownames(df).tolist() == [0, 1, 2]
df = DataFrame([1, 2, 3], index=["a", "b", "c"])
assert colnames(df) == [0]
assert rownames(df).tolist() == ["a", "b", "c"]
df = colnames(df, ["y"])
assert_iterable_equal(df.columns, ["y"])
df = colnames(df, ["y"], nested=False)
assert_iterable_equal(df.columns, ["y"])
assert_iterable_equal(colnames(df, nested=False), ["y"])
df = rownames(df, ["a", "b", "c"])
assert_iterable_equal(df.index, ["a", "b", "c"])
df = tibble(a=tibble(x=1, y=1), b=tibble(u=2, v=3), z=2)
df = df >> colnames(["c", "d", "w"], nested=True)
assert_iterable_equal(df.columns, ["c$x", "c$y", "d$u", "d$v", "w"])
def test_transform_default():
@func_factory("transform", "x")
def double(x):
return x * 2
# scalar
out = double(3)
assert out[0] == 6
out = double(np.array([1, 2], dtype=int))
assert_iterable_equal(out, [2, 4])
@func_factory("transform", "x")
def double(x):
return x * 2
out = double([1, 2])
assert_iterable_equal(out, [2, 4])
# default on series
x = Series([2, 3], index=["a", "b"])
out = double(x)
assert isinstance(out, Series)
assert_iterable_equal(out.index, ["a", "b"])
assert_iterable_equal(out, [4, 6])
# default on dataframe
x = DataFrame({"a": [3, 4]})
out = double(x)
assert isinstance(out, DataFrame)
assert_iterable_equal(out.a, [6, 8])
# default on seriesgroupby
x = Series([1, 2, 1, 2]).groupby([1, 1, 2, 2])
out = double(x)
assert isinstance(out, SeriesGroupBy)
assert_iterable_equal(out.obj, [2, 4, 2, 4])
assert out.grouper.ngroups == 2
# on tibble grouped
x = tibble(x=[1, 2, 1, 2], g=[1, 1, 2, 2]).group_by("g")
out = double(x)
# grouping variables not included
assert_iterable_equal(out.x.obj, [2, 4, 2, 4])
x = tibble(x=[1, 2, 1, 2], g=[1, 1, 2, 2]).rowwise("g")
out = double(x)
assert isinstance(out, TibbleRowwise)
assert_frame_equal(out, out._datar["grouped"].obj)
assert_iterable_equal(out.x.obj, [2, 4, 2, 4])
assert_iterable_equal(out.group_vars, ["g"])
def test_rownames_preserved():
df = DataFrame(dict(x=[1, 2]), index=["a", "b"])
df = mutate(df, y=2)
assert df.index.tolist() == ["a", "b"]
def test_non_syntactic_grouping_variable_is_preserved():
df = DataFrame({"a b": [1]}) >> group_by("a b") >> select()
assert df.columns.tolist() == ["a b"]
df = DataFrame({"a b": [1]}) >> group_by(f["a b"]) >> select()
assert df.columns.tolist() == ["a b"]