polars to_dicts() instead of to_json for nan handling (#7297)

## 📝 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).
-->

handles nans and infs.

<img width="656" height="478" alt="image"
src="https://github.com/user-attachments/assets/12651df8-7b46-4342-8465-1d4d72138d6c"
/>

<img width="982" height="622" alt="image"
src="https://github.com/user-attachments/assets/34d208dd-bdf5-4f81-b5ba-21146cb74351"
/>

<img width="772" height="544" alt="image"
src="https://github.com/user-attachments/assets/a023c6df-0548-4de1-ad3e-aaf4278c9b0e"
/>

## 🔍 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/_messaging/msgspec_encoder.py

@@ -14,7 +14,11 @@
 import msgspec
 import msgspec.json
 
+from marimo import _loggers
 from marimo._dependencies.dependencies import DependencyManager
+from marimo._plugins.core.media import io_to_data_url
+
+LOGGER = _loggers.marimo_logger()
 
 
 def enc_hook(obj: Any) -> Any:
@@ -126,6 +130,39 @@ def enc_hook(obj: Any) -> Any:
             ):
                 return str(obj)
 
+    # Handle Pillow images
+    if DependencyManager.pillow.imported():
+        try:
+            from PIL import Image
+
+            if isinstance(obj, Image.Image):
+                return io_to_data_url(obj, "image/png")
+        except Exception:
+            LOGGER.debug("Unable to convert image to data URL", exc_info=True)
+
+    # Handle Matplotlib figures
+    if DependencyManager.matplotlib.imported():
+        try:
+            import matplotlib.figure
+            from matplotlib.axes import Axes
+
+            from marimo._output.formatting import as_html
+            from marimo._plugins.stateless.flex import vstack
+
+            if isinstance(obj, matplotlib.figure.Figure):
+                html = as_html(vstack([str(obj), obj]))
+                mimetype, data = html._mime_()
+
+            if isinstance(obj, Axes):
+                html = as_html(vstack([str(obj), obj]))
+                mimetype, data = html._mime_()
+                return {"mimetype": mimetype, "data": data}
+        except Exception:
+            LOGGER.debug(
+                "Error converting matplotlib figures to HTML",
+                exc_info=True,
+            )
+
     # Handle objects with __slots__
     slots = getattr(obj, "__slots__", None)
     if slots is not None:

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

@@ -13,9 +13,7 @@
 
 from marimo import _loggers
 from marimo._data.models import BinValue, ColumnStats, ExternalDataType
-from marimo._dependencies.dependencies import DependencyManager
 from marimo._output.data.data import sanitize_json_bigint
-from marimo._plugins.core.media import io_to_data_url
 from marimo._plugins.ui._impl.tables.format import (
     FormatMapping,
     format_value,
@@ -702,50 +700,3 @@ def __repr__(self) -> str:
         if rows is None:
             return f"{df_type}: {columns:,} columns"
         return f"{df_type}: {rows:,} rows x {columns:,} columns"
-
-    def _sanitize_table_value(self, value: Any) -> Any:
-        """
-        Sanitize a value for display in a table cell.
-
-        Most values are unchanged, but some values are for better
-        display such as Images.
-        """
-        if value is None:
-            return None
-
-        # Handle Pillow images
-        if DependencyManager.pillow.imported():
-            try:
-                from PIL import Image
-
-                if isinstance(value, Image.Image):
-                    return io_to_data_url(value, "image/png")
-            except Exception:
-                LOGGER.debug(
-                    "Unable to convert image to data URL", exc_info=True
-                )
-
-        # Handle Matplotlib figures
-        if DependencyManager.matplotlib.imported():
-            try:
-                import matplotlib.figure
-                from matplotlib.axes import Axes
-
-                from marimo._output.formatting import as_html
-                from marimo._plugins.stateless.flex import vstack
-
-                if isinstance(value, matplotlib.figure.Figure):
-                    html = as_html(vstack([str(value), value]))
-                    mimetype, data = html._mime_()
-
-                if isinstance(value, Axes):
-                    html = as_html(vstack([str(value), value]))
-                    mimetype, data = html._mime_()
-                    return {"mimetype": mimetype, "data": data}
-            except Exception:
-                LOGGER.debug(
-                    "Error converting matplotlib figures to HTML",
-                    exc_info=True,
-                )
-
-        return value

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

@@ -116,9 +116,6 @@ def to_json(result: pd.DataFrame) -> list[dict[str, Any]]:
                             if inferred_dtype == "date":
                                 result[col] = result[col].apply(str)
 
-                            result[col] = result[col].apply(
-                                self._sanitize_table_value
-                            )
                             # Cast bytes to string to avoid overflow error
                             if self._infer_dtype(col) == "bytes":
                                 result[col] = result[col].apply(str)

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

@@ -110,6 +110,11 @@ def to_csv_str(
             def to_json_str(
                 self, format_mapping: Optional[FormatMapping] = None
             ) -> str:
+                def to_json(result: pl.DataFrame) -> list[dict[str, Any]]:
+                    # Use to_dicts instead of write_json
+                    # Preserves certain types like nans, infs, -infs, etc.
+                    return result.to_dicts()
+
                 result = self.apply_formatting(format_mapping).collect()
                 try:
                     for column in result.get_columns():
@@ -127,7 +132,7 @@ def to_json_str(
                             result = result.with_columns(
                                 pl.col(column.name).cast(pl.List(pl.String))
                             )
-                    return sanitize_json_bigint(result.write_json())
+                    return sanitize_json_bigint(to_json(result))
                 except (
                     BaseException
                 ):  # Sometimes, polars throws a generic exception
@@ -169,7 +174,7 @@ def to_json_str(
                             ", ".join(f"'{col}'" for col in converted_columns),
                         )
 
-                    return sanitize_json_bigint(result.write_json())
+                    return sanitize_json_bigint(to_json(result))
 
             def _convert_time_to_string(
                 self, result: pl.DataFrame, column: pl.Series
@@ -189,7 +194,7 @@ def _cast_object_to_string(
                         # As of writing this, cast(pl.String) doesn't work
                         # for pl.Object types, so we use map_elements
                         column.map_elements(
-                            lambda v: str(self._sanitize_table_value(v)),
+                            lambda v: str(v),
                             return_dtype=pl.String,
                         )
                     )

marimo/this.py

@@ -1,3 +1,4 @@
+# Copyright 2025 Marimo. All rights reserved.
 print("""The marimo contract
 
 To use marimo means entering into a social contract.

tests/_messaging/test_enc_hook.py

@@ -0,0 +1,65 @@
+# Copyright 2025 Marimo. All rights reserved.
+
+import json
+
+import pytest
+
+from marimo._dependencies.dependencies import DependencyManager
+from marimo._messaging.msgspec_encoder import enc_hook
+
+
+def test_enc_hook() -> None:
+    """Test the enc_hook function."""
+    assert enc_hook(1) == 1
+    assert enc_hook("hello") == "hello"
+    assert enc_hook(3.14) == 3.14
+    assert enc_hook(True) is True
+    assert enc_hook(None) is None
+
+    if DependencyManager.numpy.imported():
+        import numpy as np
+
+        assert enc_hook(np.array([1, 2, 3])) == [1, 2, 3]
+        assert enc_hook(np.array([1, 2, 3])) == [1, 2, 3]
+
+
+@pytest.mark.skipif(
+    not DependencyManager.pillow.imported(),
+    reason="Pillow not installed",
+)
+def test_serialize_pillow_image() -> None:
+    from PIL import Image
+
+    img = Image.new("RGB", (10, 10), color="red")
+
+    result = enc_hook(img)
+
+    assert result is not None
+    assert result.startswith("data:image/png;base64,")
+
+
+@pytest.mark.skipif(
+    not DependencyManager.matplotlib.imported(),
+    reason="Matplotlib not installed",
+)
+def test_serialize_matplotlib_figure() -> None:
+    import matplotlib.pyplot as plt
+
+    fig, ax = plt.subplots()
+    ax.plot([1, 2, 3], [1, 2, 3])
+
+    # Serialize the figure
+    result = enc_hook(fig)
+    for obj in [fig, ax]:
+        result = enc_hook(obj)
+        assert isinstance(result, dict)
+
+        assert "mimetype" in result
+        assert "data" in result
+
+        assert "application/vnd.marimo+mimebundle" in result["mimetype"]
+        assert "image/png" in result["data"]
+
+        image_data = json.loads(result["data"])
+        assert "image/png" in image_data
+        assert image_data["image/png"].startswith("data:image/png;base64,")

tests/_plugins/ui/_impl/tables/snapshots/polars.csv

@@ -1,4 +1,4 @@
-strings,bool,int,large_int,float,datetime,date,struct,list,array,nulls,category,set,imaginary,time,duration,mixed_list,enum_list
-a,true,1,18446744073709551616,1.0,2021-01-01T00:00:00.000000,2021-01-01,"{""a"":1,""b"":2}","1,2",1,,cat,"{1, 2}",(1+2j),12:30:00.000000000,1d,"1,two","A,B,C"
-b,false,2,36893488147419103233,2.0,2021-01-02T00:00:00.000000,2021-01-02,"{""a"":3,""b"":4}","3,4",2,data,dog,"{3, 4}",(3+4j),13:45:00.000000000,315µs,"3.0,0.0","A,B,C"
-c,true,3,73786976294838206466,3.0,2021-01-03T00:00:00.000000,2021-01-03,"{""a"":5,""b"":6}","5,6",3,,mouse,"{5, 6}",(5+6j),14:15:00.000000000,2h 30m,2021-01-01 00:00:00.000000,"A,B,C"
+strings,bool,int,large_int,float,datetime,date,struct,list,array,nulls,category,set,imaginary,time,duration,nans,infs,mixed_list,enum_list
+a,true,1,18446744073709551616,1.0,2021-01-01T00:00:00.000000,2021-01-01,"{""a"":1,""b"":2}","1,2",1,,cat,"{1, 2}",(1+2j),12:30:00.000000000,1d,NaN,inf,"1,two","A,B,C"
+b,false,2,36893488147419103233,2.0,2021-01-02T00:00:00.000000,2021-01-02,"{""a"":3,""b"":4}","3,4",2,data,dog,"{3, 4}",(3+4j),13:45:00.000000000,315µs,NaN,-inf,"3.0,0.0","A,B,C"
+c,true,3,73786976294838206466,3.0,2021-01-03T00:00:00.000000,2021-01-03,"{""a"":5,""b"":6}","5,6",3,,mouse,"{5, 6}",(5+6j),14:15:00.000000000,2h 30m,NaN,inf,2021-01-01 00:00:00.000000,"A,B,C"