def test_integer_gt(schema, file):
assert not InclusiveMetricsEvaluator(
schema, Expressions.greater_than("id", 85)).eval(file)
assert not InclusiveMetricsEvaluator(
schema, Expressions.greater_than("id", 79)).eval(file)
assert InclusiveMetricsEvaluator(schema,
Expressions.greater_than("id",
78)).eval(file)
assert InclusiveMetricsEvaluator(schema,
Expressions.greater_than("id",
75)).eval(file)
def test_multiple_references(assert_all_bound):
expr = Expressions.or_(
Expressions.and_(Expressions.equal("x", 7),
Expressions.less_than("y", 100)),
Expressions.greater_than("z", -100))
assert_all_bound("Multiple references", Binder.bind(STRUCT, expr))
def test_and(schema, file):
assert not InclusiveMetricsEvaluator(
schema,
Expressions.and_(Expressions.less_than(
"id", 5), Expressions.greater_than_or_equal("id", 0))).eval(file)
assert InclusiveMetricsEvaluator(
schema,
Expressions.and_(Expressions.greater_than("id", 5),
Expressions.less_than_or_equal("id", 30))).eval(file)
def test_and(strict_schema, strict_file):
assert not StrictMetricsEvaluator(strict_schema,
Expressions.and_(Expressions.greater_than("id", 5),
Expressions.less_than_or_equal("id", 30))).eval(strict_file)
assert not StrictMetricsEvaluator(strict_schema,
Expressions.and_(Expressions.less_than("id", 5),
Expressions.greater_than_or_equal("id", 0))).eval(strict_file)
assert StrictMetricsEvaluator(strict_schema,
Expressions.and_(Expressions.less_than("id", 85),
Expressions.greater_than_or_equal("id", 0))).eval(strict_file)
def test_zero_record_file(strict_schema, empty):
exprs = [Expressions.less_than("no_stats", 5),
Expressions.less_than_or_equal("no_stats", 30),
Expressions.equal("no_stats", 70),
Expressions.greater_than("no_stats", 78),
Expressions.greater_than_or_equal("no_stats", 90),
Expressions.not_equal("no_stats", 101),
Expressions.is_null("no_stats"),
Expressions.not_null("no_stats")]
for expr in exprs:
assert StrictMetricsEvaluator(strict_schema, expr).eval(empty)
def test_missing_stats(strict_schema, missing_stats):
exprs = [Expressions.less_than("no_stats", 5),
Expressions.less_than_or_equal("no_stats", 30),
Expressions.equal("no_stats", 70),
Expressions.greater_than("no_stats", 78),
Expressions.greater_than_or_equal("no_stats", 90),
Expressions.not_equal("no_stats", 101),
Expressions.is_null("no_stats"),
Expressions.not_null("no_stats")]
for expr in exprs:
assert not StrictMetricsEvaluator(strict_schema, expr).eval(missing_stats)
def test_complex_expr():
expr = Expressions.or_(
Expressions.and_(Expressions.greater_than('a', 1),
Expressions.equal("b", "US")),
Expressions.equal("c", True))
translated_dataset_filter = get_dataset_filter(expr, {
'a': 'a',
'b': 'b',
'c': 'c'
})
dataset_filter = (((ds.field("a") > 1) & (ds.field("b") == "US")) |
(ds.field("c") == True)) # noqa: E712
assert dataset_filter.equals(translated_dataset_filter)
def test_or(assert_all_bound, assert_and_unwrap):
expr = Expressions.or_(Expressions.greater_than("z", -100),
Expressions.less_than("y", 100))
bound_expr = Binder.bind(STRUCT, expr)
assert_all_bound("Or", bound_expr)
or_ = assert_and_unwrap(bound_expr, Or)
left = assert_and_unwrap(or_.left, None)
# should bind z correctly
assert 2 == left.ref.field_id
right = assert_and_unwrap(or_.right, None)
# should bind y correctly
assert 1 == right.ref.field_id
def test_basic_simplification(assert_and_unwrap):
# Should simplify or expression to alwaysTrue
assert Expressions.always_true() == Binder.bind(
STRUCT,
Expressions.or_(Expressions.less_than("y", 100),
Expressions.greater_than("z", -9999999999)))
# Should simplify or expression to alwaysfalse
assert Expressions.always_false() == Binder.bind(
STRUCT,
Expressions.and_(Expressions.less_than("y", 100),
Expressions.less_than("z", -9999999999)))
bound = Binder.bind(
STRUCT,
Expressions.not_(Expressions.not_(Expressions.less_than("y", 100))))
pred = assert_and_unwrap(bound, None)
assert 1 == pred.ref.field_id
(Expressions.is_null("some_nulls"), True),
(Expressions.is_null("no_nulls"), False)])
def test_no_nulls(inc_man_spec, inc_man_file, expression, expected):
assert InclusiveManifestEvaluator(inc_man_spec, expression).eval(inc_man_file) == expected
def test_missing_column(inc_man_spec, inc_man_file):
with pytest.raises(ValidationException):
InclusiveManifestEvaluator(inc_man_spec, Expressions.less_than("missing", 5)).eval(inc_man_file)
@pytest.mark.parametrize("expression", [
Expressions.less_than("id", 5),
Expressions.less_than_or_equal("id", 30),
Expressions.equal("id", 70),
Expressions.greater_than("id", 78),
Expressions.greater_than_or_equal("id", 90),
Expressions.not_equal("id", 101),
Expressions.less_than_or_equal("id", 30),
Expressions.is_null("id"),
Expressions.not_null("id")])
def test_missing_stats(inc_man_spec, inc_man_file_ns, expression):
assert InclusiveManifestEvaluator(inc_man_spec, expression).eval(inc_man_file_ns)
@pytest.mark.parametrize("expression, expected", [
(Expressions.less_than("id", 5), True),
(Expressions.greater_than("id", 5), False)])
def test_not(inc_man_spec, inc_man_file, expression, expected):
assert InclusiveManifestEvaluator(inc_man_spec, Expressions.not_(expression)).eval(inc_man_file) == expected
def test_integer_gt(strict_schema, strict_file):
assert not StrictMetricsEvaluator(strict_schema, Expressions.greater_than("id", 79)).eval(strict_file)
assert not StrictMetricsEvaluator(strict_schema, Expressions.greater_than("id", 78)).eval(strict_file)
assert not StrictMetricsEvaluator(strict_schema, Expressions.greater_than("id", 30)).eval(strict_file)
assert StrictMetricsEvaluator(strict_schema, Expressions.greater_than("id", 29)).eval(strict_file)
def test_gt():
expected_expr = Expressions.greater_than("col_a", 1)
conv_expr = Expressions.convert_string_to_expr("col_a > 1")
assert expected_expr == conv_expr
def test_not_conversion():
expr = Expressions.not_(Expressions.greater_than('a', 1))
translated_dataset_filter = get_dataset_filter(expr, {'a': 'a'})
assert (~(ds.field("a") > 1)).equals(translated_dataset_filter)
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
from iceberg.api.expressions import Expressions
from iceberg.parquet.dataset_utils import get_dataset_filter
import pyarrow.dataset as ds
import pytest
@pytest.mark.parametrize(
"expr, dataset_filter, column_map",
[(Expressions.greater_than('a', 1), ds.field('a') > 1, {
'a': 'a'
}),
(Expressions.greater_than_or_equal('a', 1), ds.field('a') >= 1, {
'a': 'a'
}), (Expressions.less_than('a', 1), ds.field('a') < 1, {
'a': 'a'
}),
(Expressions.less_than_or_equal('a', 1), ds.field('a') <= 1, {
'a': 'a'
}), (Expressions.equal('a', 1), ds.field('a') == 1, {
'a': 'a'
}), (Expressions.not_equal('a', 1), ds.field('a') != 1, {
'a': 'a'
}), (Expressions.not_null('a'), ds.field('a').is_valid(), {
'a': 'a'
def test_not(schema, file):
assert InclusiveMetricsEvaluator(schema, Expressions.not_(Expressions.less_than("id", 5))).eval(file)
assert not InclusiveMetricsEvaluator(schema,
Expressions.not_(Expressions.greater_than("id", 5))).eval(file)
@pytest.fixture(scope="session",
params=[
Operation.LT, Operation.LT_EQ, Operation.GT,
Operation.GT_EQ, Operation.EQ, Operation.NOT_EQ
])
def op(request):
yield request.param
@pytest.fixture(scope="session",
params=[
Expressions.always_false(),
Expressions.always_true(),
Expressions.less_than("x", 5),
Expressions.less_than_or_equal("y", -3),
Expressions.greater_than("z", 0),
Expressions.greater_than_or_equal("t", 129),
Expressions.equal("col", "data"),
Expressions.not_equal("col", "abc"),
Expressions.not_null("maybeNull"),
Expressions.is_null("maybeNull2"),
Expressions.not_(Expressions.greater_than("a", 10)),
Expressions.and_(Expressions.greater_than_or_equal("a", 0),
Expressions.less_than("a", 3)),
Expressions.or_(Expressions.less_than("a", 0),
Expressions.greater_than("a", 10)),
Expressions.equal("a", 5).bind(exp_schema.as_struct())
])
def expression(request):
yield request.param
def test_not(strict_schema, strict_file):
assert StrictMetricsEvaluator(strict_schema,
Expressions.not_(Expressions.less_than("id", 5))).eval(strict_file)
assert not StrictMetricsEvaluator(strict_schema,
Expressions.not_(Expressions.greater_than("id", 5))).eval(strict_file)
def test_int_gt(inc_man_spec, inc_man_file, val, expected):
assert InclusiveManifestEvaluator(inc_man_spec, Expressions.greater_than("id", val)).eval(inc_man_file) == expected
assert InclusiveManifestEvaluator(
inc_man_spec, expression).eval(inc_man_file) == expected
def test_missing_column(inc_man_spec, inc_man_file):
with pytest.raises(ValidationException):
InclusiveManifestEvaluator(inc_man_spec,
Expressions.less_than("missing",
5)).eval(inc_man_file)
@pytest.mark.parametrize("expression", [
Expressions.less_than("id", 5),
Expressions.less_than_or_equal("id", 30),
Expressions.equal("id", 70),
Expressions.greater_than("id", 78),
Expressions.greater_than_or_equal("id", 90),
Expressions.not_equal("id", 101),
Expressions.less_than_or_equal("id", 30),
Expressions.is_null("id"),
Expressions.not_null("id")
])
def test_missing_stats(inc_man_spec, inc_man_file_ns, expression):
assert InclusiveManifestEvaluator(inc_man_spec,
expression).eval(inc_man_file_ns)
@pytest.mark.parametrize("expression, expected",
[(Expressions.less_than("id", 5), True),
(Expressions.greater_than("id", 5), False)])
def test_not(inc_man_spec, inc_man_file, expression, expected):
