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_precedence_with_between():
expected_expr = Expressions.or_(
Expressions.and_(Expressions.greater_than_or_equal("col_a", 1),
Expressions.less_than_or_equal("col_a", 2)),
Expressions.equal("col_c", 3))
conv_expr = Expressions.convert_string_to_expr(
"col_a between 1 and 2 or col_c=3")
assert expected_expr == conv_expr
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_between():
expected_expr = Expressions.and_(
Expressions.greater_than_or_equal("col_a", 1),
Expressions.less_than_or_equal("col_a", 2))
conv_expr = Expressions.convert_string_to_expr("col_a between 1 and 2")
assert expected_expr == conv_expr
def test_lte():
expected_expr = Expressions.less_than_or_equal("col_a", 1)
conv_expr = Expressions.convert_string_to_expr("col_a <= 1")
assert expected_expr == conv_expr
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'
}), (Expressions.is_null('a'), ~ds.field('a').is_valid(), {
'a': 'a'
})])
def test_simple(expr, dataset_filter, column_map):
translated_dataset_filter = get_dataset_filter(expr, column_map)
assert dataset_filter.equals(translated_dataset_filter)
@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_int_lt_eq(inc_man_spec, inc_man_file, val, expected):
assert InclusiveManifestEvaluator(inc_man_spec,
Expressions.less_than_or_equal("id", val)).eval(inc_man_file) == expected
@pytest.mark.parametrize("expression,expected", [
(Expressions.is_null("all_nulls"), True),
(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):
def test_integer_lt_eq(strict_schema, strict_file):
assert not StrictMetricsEvaluator(strict_schema, Expressions.less_than_or_equal("id", 29)).eval(strict_file)
assert not StrictMetricsEvaluator(strict_schema, Expressions.less_than_or_equal("id", 30)).eval(strict_file)
assert StrictMetricsEvaluator(strict_schema, Expressions.less_than_or_equal("id", 79)).eval(strict_file)
assert StrictMetricsEvaluator(strict_schema, Expressions.less_than_or_equal("id", 80)).eval(strict_file)
(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),
