def test_lazy_functions():
df = pl.DataFrame({
"a": ["foo", "bar", "2"],
"b": [1, 2, 3],
"c": [1.0, 2.0, 3.0]
})
out = df[[pl.count("a")]]
assert out[0] == 3
assert pl.count(df["a"]) == 3
out = df[[
pl.var("b"),
pl.std("b"),
pl.max("b"),
pl.min("b"),
pl.sum("b"),
pl.mean("b"),
pl.median("b"),
pl.n_unique("b"),
pl.first("b"),
pl.last("b"),
]]
expected = 1.0
assert np.isclose(out[0], expected)
assert np.isclose(pl.var(df["b"]), expected)
expected = 1.0
assert np.isclose(out[1], expected)
assert np.isclose(pl.std(df["b"]), expected)
expected = 3
assert np.isclose(out[2], expected)
assert np.isclose(pl.max(df["b"]), expected)
expected = 1
assert np.isclose(out[3], expected)
assert np.isclose(pl.min(df["b"]), expected)
expected = 6
assert np.isclose(out[4], expected)
assert np.isclose(pl.sum(df["b"]), expected)
expected = 2
assert np.isclose(out[5], expected)
assert np.isclose(pl.mean(df["b"]), expected)
expected = 2
assert np.isclose(out[6], expected)
assert np.isclose(pl.median(df["b"]), expected)
expected = 3
assert np.isclose(out[7], expected)
assert np.isclose(pl.n_unique(df["b"]), expected)
expected = 1
assert np.isclose(out[8], expected)
assert np.isclose(pl.first(df["b"]), expected)
expected = 3
assert np.isclose(out[9], expected)
assert np.isclose(pl.last(df["b"]), expected)
expected = 3
assert np.isclose(out[9], expected)
assert np.isclose(pl.last(df["b"]), expected)
def test_lazy_functions():
df = pl.DataFrame({
"a": ["foo", "bar", "2"],
"b": [1, 2, 3],
"c": [1.0, 2.0, 3.0]
})
out = df[[pl.count("a")]]
assert out["a"] == 3
assert pl.count(df["a"]) == 3
out = df[[
pl.var("b"),
pl.std("b"),
pl.max("b"),
pl.min("b"),
pl.sum("b"),
pl.mean("b"),
pl.median("b"),
pl.n_unique("b"),
pl.first("b"),
pl.last("b"),
]]
expected = 1.0
assert np.isclose(out.select_at_idx(0), expected)
assert np.isclose(pl.var(df["b"]), expected)
expected = 1.0
assert np.isclose(out.select_at_idx(1), expected)
assert np.isclose(pl.std(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(2), expected)
assert np.isclose(pl.max(df["b"]), expected)
expected = 1
assert np.isclose(out.select_at_idx(3), expected)
assert np.isclose(pl.min(df["b"]), expected)
expected = 6
assert np.isclose(out.select_at_idx(4), expected)
assert np.isclose(pl.sum(df["b"]), expected)
expected = 2
assert np.isclose(out.select_at_idx(5), expected)
assert np.isclose(pl.mean(df["b"]), expected)
expected = 2
assert np.isclose(out.select_at_idx(6), expected)
assert np.isclose(pl.median(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(7), expected)
assert np.isclose(pl.n_unique(df["b"]), expected)
expected = 1
assert np.isclose(out.select_at_idx(8), expected)
assert np.isclose(pl.first(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(9), expected)
assert np.isclose(pl.last(df["b"]), expected)
expected = 3
assert np.isclose(out.select_at_idx(9), expected)
assert np.isclose(pl.last(df["b"]), expected)
def test_filter_str() -> None:
# use a str instead of a column expr
df = pl.DataFrame({
"time": ["11:11:00", "11:12:00", "11:13:00", "11:14:00"],
"bools": [True, False, True, False],
})
q = df.lazy()
# last row based on a filter
result = q.filter(pl.col("bools")).select(pl.last("*")).collect()
expected = pl.DataFrame({"time": ["11:13:00"], "bools": [True]})
assert result.frame_equal(expected)
# last row based on a filter
result = q.filter("bools").select(pl.last("*")).collect()
assert result.frame_equal(expected)
def test_filter_str():
# use a str instead of a column expr
df = pl.DataFrame({
"time": ["11:11:00", "11:12:00", "11:13:00", "11:14:00"],
"bools": [True, False, True, False],
})
q = df.lazy()
# last row based on a filter
q.filter(pl.col("bools")).select(pl.last("*"))