fix(python): don't allow duplicate columns in read_csv arg

pola-rs · Dec 26, 2022 · e9fb2e0 · e9fb2e0
1 parent 4aede96
commit e9fb2e0
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 12 deletions.
diff --git a/py-polars/Cargo.lock b/py-polars/Cargo.lock
diff --git a/py-polars/polars/utils.py b/py-polars/polars/utils.py
@@ -162,9 +162,13 @@ def handle_projection_columns(
             columns = None
         elif not is_str_sequence(columns):
             raise ValueError(
-                "columns arg should contain a list of all integers or all strings"
+                "'columns' arg should contain a list of all integers or all strings"
                 " values."
             )
+        if len(set(columns)) != len(columns):  # type: ignore[arg-type]
+            raise ValueError(
+                f"'columns' arg should only have unique values. Got '{columns}'."
+            )
     return projection, columns  # type: ignore[return-value]
 
 

diff --git a/py-polars/tests/unit/test_errors.py b/py-polars/tests/unit/test_errors.py
@@ -301,3 +301,13 @@ def test_epoch_time_type() -> None:
         pl.ComputeError, match="Cannot compute timestamp of a series with dtype 'Time'"
     ):
         pl.Series([time(0, 0, 1)]).dt.epoch("s")
+
+
+def test_duplicate_columns_arg_csv() -> None:
+    f = io.BytesIO()
+    pl.DataFrame({"x": [1, 2, 3], "y": ["a", "b", "c"]}).write_csv(f)
+    f.seek(0)
+    with pytest.raises(
+        ValueError, match=r"'columns' arg should only have unique values"
+    ):
+        pl.read_csv(f, columns=["x", "x", "y"])