fix: Skip Batches Expression for boolean literals (#21310)

pola-rs · Feb 18, 2025 · a3072d2 · a3072d2
1 parent eae47a4
commit a3072d2
Show file tree

Hide file tree

Showing 3 changed files with 38 additions and 3 deletions.
diff --git a/crates/polars-plan/src/plans/aexpr/predicates/skip_batches.rs b/crates/polars-plan/src/plans/aexpr/predicates/skip_batches.rs
@@ -200,8 +200,8 @@ fn aexpr_to_skip_batch_predicate_rec(
     if let Some(Some(lv)) = constant_evaluate(e, expr_arena, schema, 0) {
         if let Some(av) = lv.to_any_value() {
             return match av {
-                AnyValue::Null => Some(lv!(bool: false)),
-                AnyValue::Boolean(b) => Some(lv!(bool: b)),
+                AnyValue::Null => Some(lv!(bool: true)),
+                AnyValue::Boolean(b) => Some(lv!(bool: !b)),
                 _ => None,
             };
         }

diff --git a/py-polars/tests/unit/io/test_parquet.py b/py-polars/tests/unit/io/test_parquet.py
@@ -2869,3 +2869,13 @@ def test_nested_string_slice_utf8_21202() -> None:
         pl.scan_parquet(f).slice(1, 1).collect().to_series(),
         s.slice(1, 1),
     )
+
+
+def test_filter_true_predicate_21204() -> None:
+    f = io.BytesIO()
+
+    df = pl.DataFrame({"a": [1]})
+    df.write_parquet(f)
+    f.seek(0)
+    lf = pl.scan_parquet(f).filter(pl.lit(True))
+    assert_frame_equal(lf.collect(), df)
diff --git a/py-polars/tests/unit/io/test_skip_batch_predicate.py b/py-polars/tests/unit/io/test_skip_batch_predicate.py
@@ -8,7 +8,7 @@
 
 import polars as pl
 from polars.meta import get_index_type
-from polars.testing import assert_series_equal
+from polars.testing import assert_frame_equal, assert_series_equal
 from polars.testing.parametric.strategies import series
 
 if TYPE_CHECKING:
@@ -55,6 +55,31 @@ def assert_skp_series(
         raise
 
 
+def test_true_false_predicate() -> None:
+    true_sbp = pl.lit(True)._skip_batch_predicate({})
+    false_sbp = pl.lit(False)._skip_batch_predicate({})
+    null_sbp = pl.lit(None)._skip_batch_predicate({})
+
+    df = pl.DataFrame({"len": [1]})
+
+    out = df.select(
+        true=true_sbp,
+        false=false_sbp,
+        null=null_sbp,
+    )
+
+    assert_frame_equal(
+        out,
+        pl.DataFrame(
+            {
+                "true": [False],
+                "false": [True],
+                "null": [True],
+            }
+        ),
+    )
+
+
 def test_equality() -> None:
     assert_skp_series(
         "a",