Skip to content

A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.

fix(datasets): avoid exponential blow-up of nested struct sample values (#9506)

Kiran Gadhave0eecccd369c4
diff --git a/marimo/_plugins/ui/_impl/tables/narwhals_table.py b/marimo/_plugins/ui/_impl/tables/narwhals_table.py
index eba2594359..a2a82dc5ed 100644
--- a/marimo/_plugins/ui/_impl/tables/narwhals_table.py
+++ b/marimo/_plugins/ui/_impl/tables/narwhals_table.py
@@ -4,7 +4,9 @@ from __future__ import annotations
import datetime
import functools
import io
+import json
import math
+from enum import Enum
from functools import cached_property
from typing import TYPE_CHECKING, Any, Literal, cast
@@ -698,17 +700,22 @@ class NarwhalsTableManager(
# Sample 3 values from the column
SAMPLE_SIZE = 3
try:
- from enum import Enum
+
+ def _json_default(o: Any) -> str:
+ if isinstance(o, Enum):
+ return o.name
+ return str(o)
def to_primitive(value: Any) -> str | int | float:
- if isinstance(value, list):
- return str([to_primitive(v) for v in value])
- elif isinstance(value, dict):
- return str({k: to_primitive(v) for k, v in value.items()})
- elif isinstance(value, Enum):
+ if isinstance(value, Enum):
return value.name
- elif isinstance(value, (float, int)):
+ if isinstance(value, (int, float)):
return value
+ if isinstance(value, (list, dict)):
+ try:
+ return json.dumps(value, default=_json_default)
+ except (TypeError, ValueError):
+ return str(value)
return str(value)
if self.data[column].dtype == nw.Datetime:
diff --git a/tests/_plugins/ui/_impl/tables/test_narwhals.py b/tests/_plugins/ui/_impl/tables/test_narwhals.py
index 8ae25a4941..53a3c222a4 100644
--- a/tests/_plugins/ui/_impl/tables/test_narwhals.py
+++ b/tests/_plugins/ui/_impl/tables/test_narwhals.py
@@ -1535,6 +1535,77 @@ def test_get_sample_values(df: Any) -> None:
assert sample_values == ["b'bytes1'", "b'bytes2'", "b'bytes3'"]
[email protected](not HAS_DEPS, reason="polars not installed")
+def test_get_sample_values_nested_struct_is_bounded() -> None:
+ import polars as pl
+
+ def build_payload(row_idx: int, depth: int):
+ base = {
+ "kind": chr(65 + (row_idx % 26)),
+ "scores": [row_idx + 1, row_idx + 2, row_idx + 3],
+ "meta": {"city": "Zurich", "active": row_idx % 2 == 0},
+ }
+ if depth == 0:
+ return base
+ return {
+ "level": depth,
+ "items": [
+ base,
+ {
+ "branch": row_idx,
+ "child": build_payload(row_idx, depth - 1),
+ },
+ ],
+ "summary": {"row": row_idx, "depth": depth},
+ }
+
+ df = pl.DataFrame({"payload": [build_payload(i, 8) for i in range(3)]})
+ manager = NarwhalsTableManager.from_dataframe(
+ nw.from_native(df, eager_only=True)
+ )
+
+ t0 = time.perf_counter()
+ samples = manager.get_sample_values("payload")
+ elapsed = time.perf_counter() - t0
+
+ assert len(samples) == 3
+ assert elapsed < 1.0, (
+ f"get_sample_values took {elapsed:.2f}s (expected < 1s)"
+ )
+ for s in samples:
+ assert isinstance(s, str)
+ assert len(s) < 1_000_000
+
+
[email protected](not HAS_DEPS, reason="polars not installed")
+def test_get_sample_values_nested_struct_is_json() -> None:
+ import polars as pl
+
+ df = pl.DataFrame({"x": [{"a": 1, "b": [1, 2, 3]}]})
+ manager = NarwhalsTableManager.from_dataframe(
+ nw.from_native(df, eager_only=True)
+ )
+
+ samples = manager.get_sample_values("x")
+ assert len(samples) == 1
+ assert samples[0].startswith('{"')
+ parsed = json.loads(samples[0])
+ assert parsed == {"a": 1, "b": [1, 2, 3]}
+
+
[email protected](not HAS_DEPS, reason="polars not installed")
+def test_get_sample_values_nested_struct_with_non_json_leaf() -> None:
+ import polars as pl
+
+ df = pl.DataFrame({"x": [{"when": datetime.datetime(2026, 1, 1), "n": 1}]})
+ manager = NarwhalsTableManager.from_dataframe(
+ nw.from_native(df, eager_only=True)
+ )
+ samples = manager.get_sample_values("x")
+ assert len(samples) == 1
+ assert "2026" in samples[0]
+
+
@pytest.mark.skipif(not HAS_DEPS, reason="optional dependencies not installed")
@pytest.mark.parametrize(
"df",