BUG: Fix from_records() column reorder issue, if columns!=None use passed param (#59717)

doc/source/whatsnew/v3.0.0.rst

@@ -624,6 +624,7 @@ I/O
 - Bug in :meth:`DataFrame.to_stata` when writing :class:`DataFrame` and ``byteorder=`big```. (:issue:`58969`)
 - Bug in :meth:`DataFrame.to_string` that raised ``StopIteration`` with nested DataFrames. (:issue:`16098`)
 - Bug in :meth:`HDFStore.get` was failing to save data of dtype datetime64[s] correctly (:issue:`59004`)
+- Bug in :meth:`from_records` where columns parameter with numpy array data was not reordeing and filtering out the columns (:issue:`59717`)
 - Bug in :meth:`read_csv` causing segmentation fault when ``encoding_errors`` is not a string. (:issue:`59059`)
 - Bug in :meth:`read_csv` raising ``TypeError`` when ``index_col`` is specified and ``na_values`` is a dict containing the key ``None``. (:issue:`57547`)
 - Bug in :meth:`read_csv` raising ``TypeError`` when ``nrows`` and ``iterator`` are specified without specifying a ``chunksize``. (:issue:`59079`)

pandas/core/internals/construction.py

@@ -750,7 +750,8 @@ def to_arrays(
 
     elif isinstance(data, np.ndarray) and data.dtype.names is not None:
         # e.g. recarray
-        columns = Index(list(data.dtype.names))
+        if columns is None:
+            columns = Index(list(data.dtype.names))
         arrays = [data[k] for k in columns]
         return arrays, columns
 

pandas/tests/frame/methods/test_to_arrays.py

@@ -0,0 +1,34 @@
+import numpy as np
+from numpy import array
+
+import pandas._testing as tm
+from pandas.core.indexes.api import ensure_index
+from pandas.core.internals.construction import to_arrays
+
+
+def test_to_arrays():
+    # GH 59717
+    data = np.array(
+        [
+            ("John", 25, "New York", 50000),
+            ("Jane", 30, "San Francisco", 75000),
+            ("Bob", 35, "Chicago", 65000),
+            ("Alice", 28, "Los Angeles", 60000),
+        ],
+        dtype=[("name", "U10"), ("age", "i4"), ("city", "U15"), ("salary", "i4")],
+    )
+
+    columns = ["name", "salary", "city"]
+    indexed_columns = ensure_index(columns)
+
+    actual_arrays, actual_cols = to_arrays(data, indexed_columns)
+    expected_arrays = [
+        array(["John", "Jane", "Bob", "Alice"], dtype="<U10"),
+        array([50000, 75000, 65000, 60000], dtype="int32"),
+        array(["New York", "San Francisco", "Chicago", "Los Angeles"], dtype="<U15"),
+    ]
+
+    for actual, expected in zip(actual_arrays, expected_arrays):
+        tm.assert_numpy_array_equal(actual, expected)
+
+    assert actual_cols.equals(indexed_columns)