BUG: Fix astype from float32 to string (#36464) (#36519)

Co-authored-by: Daniel Saxton <[email protected]>

Author: jorisvandenbossche

doc/source/whatsnew/v1.1.3.rst

@@ -47,6 +47,7 @@ Bug fixes
 - Bug in :class:`Series` constructor where integer overflow would occur for sufficiently large scalar inputs when an index was provided (:issue:`36291`)
 - Bug in :meth:`DataFrame.sort_values` raising an ``AttributeError`` when sorting on a key that casts column to categorical dtype (:issue:`36383`)
 - Bug in :meth:`DataFrame.stack` raising a ``ValueError`` when stacking :class:`MultiIndex` columns based on position when the levels had duplicate names (:issue:`36353`)
+- Bug in :meth:`Series.astype` showing too much precision when casting from ``np.float32`` to string dtype (:issue:`36451`)
 - Bug in :meth:`Series.isin` and :meth:`DataFrame.isin` when using ``NaN`` and a row length above 1,000,000 (:issue:`22205`)
 
 .. ---------------------------------------------------------------------------

pandas/_libs/lib.pyx

@@ -650,11 +650,12 @@ cpdef ndarray[object] ensure_string_array(
         Py_ssize_t i = 0, n = len(arr)
 
     result = np.asarray(arr, dtype="object")
+
     if copy and result is arr:
         result = result.copy()
 
     for i in range(n):
-        val = result[i]
+        val = arr[i]
         if not checknull(val):
             result[i] = str(val)
         else:

pandas/core/arrays/string_.py

@@ -199,11 +199,9 @@ def _from_sequence(cls, scalars, dtype=None, copy=False):
         if dtype:
             assert dtype == "string"
 
-        result = np.asarray(scalars, dtype="object")
-
         # convert non-na-likes to str, and nan-likes to StringDtype.na_value
         result = lib.ensure_string_array(
-            result, na_value=StringDtype.na_value, copy=copy
+            scalars, na_value=StringDtype.na_value, copy=copy
         )
 
         return cls(result)

pandas/tests/arrays/string_/test_string.py

@@ -336,3 +336,12 @@ def test_memory_usage():
     series = pd.Series(["a", "b", "c"], dtype="string")
 
     assert 0 < series.nbytes <= series.memory_usage() < series.memory_usage(deep=True)
+
+
+@pytest.mark.parametrize("dtype", [np.float16, np.float32, np.float64])
+def test_astype_from_float_dtype(dtype):
+    # https://github.com/pandas-dev/pandas/issues/36451
+    s = pd.Series([0.1], dtype=dtype)
+    result = s.astype("string")
+    expected = pd.Series(["0.1"], dtype="string")
+    tm.assert_series_equal(result, expected)

pandas/tests/series/methods/test_astype.py

@@ -1,3 +1,4 @@
+import numpy as np
 import pytest
 
 from pandas import Interval, Series, Timestamp, date_range
@@ -46,3 +47,11 @@ def test_astype_ignores_errors_for_extension_dtypes(self, values, errors):
             msg = "(Cannot cast)|(could not convert)"
             with pytest.raises((ValueError, TypeError), match=msg):
                 values.astype(float, errors=errors)
+
+    @pytest.mark.parametrize("dtype", [np.float16, np.float32, np.float64])
+    def test_astype_from_float_to_str(self, dtype):
+        # https://github.com/pandas-dev/pandas/issues/36451
+        s = Series([0.1], dtype=dtype)
+        result = s.astype(str)
+        expected = Series(["0.1"])
+        tm.assert_series_equal(result, expected)