cast list of enums to string preemptively for polars (#7058)

## 📝 Summary

<!--
Provide a concise summary of what this pull request is addressing.

If this PR fixes any issues, list them here by number (e.g., Fixes
#123).
-->
Fixes #7032 . Pre-emptively cast list of enums to strings, sometimes
polars doesn't raise the error list of enums, and silently the output is
incorrect. This fixes it.

## 🔍 Description of Changes

<!--
Detail the specific changes made in this pull request. Explain the
problem addressed and how it was resolved. If applicable, provide before
and after comparisons, screenshots, or any relevant details to help
reviewers understand the changes easily.
-->

## 📋 Checklist

- [x] I have read the [contributor
guidelines](https://github.com/marimo-team/marimo/blob/main/CONTRIBUTING.md).
- [ ] For large changes, or changes that affect the public API: this
change was discussed or approved through an issue, on
[Discord](https://marimo.io/discord?ref=pr), or the community
[discussions](https://github.com/marimo-team/marimo/discussions) (Please
provide a link if applicable).
- [x] I have added tests for the changes made.
- [x] I have run the code and verified that it works as expected.

Author: Light2Dark

marimo/_plugins/ui/_impl/tables/polars_table.py

@@ -118,6 +118,15 @@ def to_json_str(
                             result = self._convert_time_to_string(
                                 result, column
                             )
+                        # https://github.com/marimo-team/marimo/issues/7032
+                        # Polars issue with ordering and write_json for enums, so we convert to strings
+                        elif isinstance(dtype, pl.List) and isinstance(
+                            dtype.inner, (pl.Enum, pl.Categorical)
+                        ):
+                            # Convert each element in the list to a string
+                            result = result.with_columns(
+                                pl.col(column.name).cast(pl.List(pl.String))
+                            )
                     return sanitize_json_bigint(result.write_json())
                 except (
                     BaseException
@@ -154,16 +163,6 @@ def to_json_str(
                                 result, column
                             )
                             converted_columns.append(column.name)
-                        # https://github.com/marimo-team/marimo/issues/5562
-                        elif isinstance(dtype, pl.List) and isinstance(
-                            dtype.inner, (pl.Enum, pl.Categorical)
-                        ):
-                            # Convert each element in the list to a string
-                            result = result.with_columns(
-                                pl.col(column.name).cast(pl.List(pl.String))
-                            )
-                            converted_columns.append(column.name)
-
                     if converted_columns:
                         LOGGER.info(
                             "Converted columns %s to safe values.",

tests/_plugins/ui/_impl/tables/test_polars_table.py

@@ -3,6 +3,7 @@
 import datetime
 import json
 import unittest
+from enum import Enum
 from math import isnan
 from typing import Any
 
@@ -995,3 +996,35 @@ def test_to_json_enum_list_supported(self) -> None:
 
         data_list = pl.DataFrame(data, schema={"A": pl.List(pl.Categorical())})
         data_list.write_json()
+
+    @pytest.mark.xfail(
+        reason="Polars does not properly order sliced data to json when enums in list"
+    )
+    def test_failing_enums_list(self) -> None:
+        import polars as pl
+
+        class MyEnum(Enum):
+            A = 1
+            B = 2
+            C = 3
+            D = 4
+
+        # Create 10 rows cycling through enum values B, C, D, A...
+        enum_names = [e.name for e in MyEnum]
+        rows = [
+            {"value": [enum_names[i % len(enum_names)]]} for i in range(1, 11)
+        ]
+
+        expected_first_five = '[{"value":["B"]},{"value":["C"]},{"value":["D"]},{"value":["A"]},{"value":["B"]}]'
+        expected_second_five = '[{"value":["C"]},{"value":["D"]},{"value":["A"]},{"value":["B"]},{"value":["C"]}]'
+
+        # Test without schema - works fine
+        df = pl.DataFrame(rows)
+        assert df[0:5].write_json() == expected_first_five
+        assert df[5:10].write_json() == expected_second_five
+
+        # Test with schema - second slice fails
+        schema = {"value": pl.List(pl.Enum(enum_names))}
+        df_schema = pl.DataFrame(rows, schema=schema)
+        assert df_schema[0:5].write_json() == expected_first_five
+        assert df_schema[5:10].write_json() == expected_second_five  # fails

tests/_plugins/ui/_impl/test_table.py

@@ -3,6 +3,7 @@
 
 import json
 from datetime import date
+from enum import Enum
 from typing import TYPE_CHECKING, Any
 from unittest.mock import patch
 
@@ -2399,3 +2400,41 @@ def hover_text(_row: str, _col: str, value: Any) -> str:
         "4": {"column_0": "hover:carrots"},
         "5": {"column_0": "hover:carrots"},
     }
+
+
+@pytest.mark.skipif(
+    not DependencyManager.polars.has(), reason="Polars not installed"
+)
+def test_polars_enums_in_list():
+    import polars as pl
+
+    class MyEnum(Enum):
+        A = 1
+        B = 2
+        C = 3
+        D = 4
+
+    # Create 10 rows cycling through enum values B, C, D, A...
+    enum_names = [e.name for e in MyEnum]
+    rows = [{"value": [enum_names[i % len(enum_names)]]} for i in range(1, 11)]
+
+    schema = {"value": pl.List(pl.Enum(enum_names))}
+    df = pl.DataFrame(rows, schema=schema)
+
+    table = ui.table(df, selection=None)
+
+    # First page
+    response = table._search(SearchTableArgs(page_size=5, page_number=0))
+    assert (
+        response.data
+        == '[{"value":["B"]},{"value":["C"]},{"value":["D"]},{"value":["A"]},{"value":["B"]}]'
+    )
+
+    # Second page
+    response_next_page = table._search(
+        SearchTableArgs(page_size=5, page_number=1)
+    )
+    assert (
+        response_next_page.data
+        == '[{"value":["C"]},{"value":["D"]},{"value":["A"]},{"value":["B"]},{"value":["C"]}]'
+    )