BUG: StataWriter value_label encoding (#47199)

Author: eirki

doc/source/whatsnew/v1.5.0.rst

@@ -828,7 +828,7 @@ I/O
 - Bug in :func:`read_parquet` when ``engine="fastparquet"`` where the file was not closed on error (:issue:`46555`)
 - :meth:`to_html` now excludes the ``border`` attribute from ``<table>`` elements when ``border`` keyword is set to ``False``.
 - Bug in :func:`read_sas` returned ``None`` rather than an empty DataFrame for SAS7BDAT files with zero rows (:issue:`18198`)
--
+- Bug in :class:`StataWriter` where value labels were always written with default encoding (:issue:`46750`)
 
 Period
 ^^^^^^

pandas/io/stata.py

@@ -668,7 +668,9 @@ class StataValueLabel:
         Encoding to use for value labels.
     """
 
-    def __init__(self, catarray: Series, encoding: str = "latin-1") -> None:
+    def __init__(
+        self, catarray: Series, encoding: Literal["latin-1", "utf-8"] = "latin-1"
+    ) -> None:
 
         if encoding not in ("latin-1", "utf-8"):
             raise ValueError("Only latin-1 and utf-8 are supported.")
@@ -2250,7 +2252,7 @@ class StataWriter(StataParser):
     """
 
     _max_string_length = 244
-    _encoding = "latin-1"
+    _encoding: Literal["latin-1", "utf-8"] = "latin-1"
 
     def __init__(
         self,
@@ -2331,7 +2333,7 @@ def _prepare_non_cat_value_labels(
                     f"Can't create value labels for {labname}, value labels "
                     "can only be applied to numeric columns."
                 )
-            svl = StataNonCatValueLabel(colname, labels)
+            svl = StataNonCatValueLabel(colname, labels, self._encoding)
             non_cat_value_labels.append(svl)
         return non_cat_value_labels
 
@@ -3575,7 +3577,7 @@ class StataWriterUTF8(StataWriter117):
     >>> writer.write_file()
     """
 
-    _encoding = "utf-8"
+    _encoding: Literal["utf-8"] = "utf-8"
 
     def __init__(
         self,

pandas/tests/io/test_stata.py

@@ -1797,6 +1797,7 @@ def test_utf8_writer(self, version):
             "ᴐᴬᵀ": "",
         }
         data_label = "ᴅaᵀa-label"
+        value_labels = {"β": {1: "label", 2: "æøå", 3: "ŋot valid latin-1"}}
         data["β"] = data["β"].astype(np.int32)
         with tm.ensure_clean() as path:
             writer = StataWriterUTF8(
@@ -1807,11 +1808,16 @@ def test_utf8_writer(self, version):
                 variable_labels=variable_labels,
                 write_index=False,
                 version=version,
+                value_labels=value_labels,
             )
             writer.write_file()
             reread_encoded = read_stata(path)
             # Missing is intentionally converted to empty strl
             data["strls"] = data["strls"].fillna("")
+            # Variable with value labels is reread as categorical
+            data["β"] = (
+                data["β"].replace(value_labels["β"]).astype("category").cat.as_ordered()
+            )
             tm.assert_frame_equal(data, reread_encoded)
             reader = StataReader(path)
             assert reader.data_label == data_label