My feature branch to issue #19129 (read_json and orient='table' With Numeric Column)

doc/source/whatsnew/v3.0.0.rst

@@ -63,6 +63,7 @@ Other enhancements
 - :meth:`DataFrame.plot.scatter` argument ``c`` now accepts a column of strings, where rows with the same string are colored identically (:issue:`16827` and :issue:`16485`)
 - :class:`DataFrameGroupBy` and :class:`SeriesGroupBy` methods ``sum``, ``mean``, ``median``, ``prod``, ``min``, ``max``, ``std``, ``var`` and ``sem`` now accept ``skipna`` parameter (:issue:`15675`)
 - :class:`Rolling` and :class:`Expanding` now support aggregations ``first`` and ``last`` (:issue:`33155`)
+- :func:`read_json` with ``orient="table"`` now correctly restores non-string column names when reading JSON data, ensuring that column names retain their original types as specified in the schema (:issue:`19129`).
 - :func:`read_parquet` accepts ``to_pandas_kwargs`` which are forwarded to :meth:`pyarrow.Table.to_pandas` which enables passing additional keywords to customize the conversion to pandas, such as ``maps_as_pydicts`` to read the Parquet map data type as python dictionaries (:issue:`56842`)
 - :meth:`.DataFrameGroupBy.transform`, :meth:`.SeriesGroupBy.transform`, :meth:`.DataFrameGroupBy.agg`, :meth:`.SeriesGroupBy.agg`, :meth:`.SeriesGroupBy.apply`, :meth:`.DataFrameGroupBy.apply` now support ``kurt`` (:issue:`40139`)
 - :meth:`DataFrameGroupBy.transform`, :meth:`SeriesGroupBy.transform`, :meth:`DataFrameGroupBy.agg`, :meth:`SeriesGroupBy.agg`, :meth:`RollingGroupby.apply`, :meth:`ExpandingGroupby.apply`, :meth:`Rolling.apply`, :meth:`Expanding.apply`, :meth:`DataFrame.apply` with ``engine="numba"`` now supports positional arguments passed as kwargs (:issue:`58995`)

pandas/io/json/_table_schema.py

@@ -366,17 +366,29 @@ def parse_table_schema(json, precise_float: bool) -> DataFrame:
         :class:`Index` name of 'index'  and :class:`MultiIndex` names starting
         with 'level_' are not supported.
 
+        To handle cases where column names are non-string types (e.g., integers),
+        all column names are first converted to strings when constructing the DataFrame.
+        After applying the correct data types using `astype(dtypes)`, the column names
+        are restored to their original types as specified in the schema.
+        This ensures compatibility with `to_json(orient="table")` while maintaining
+        the integrity of non-string column names.
+
     See Also
     --------
     build_table_schema : Inverse function.
     pandas.read_json
     """
     table = ujson_loads(json, precise_float=precise_float)
-    col_order = [field["name"] for field in table["schema"]["fields"]]
+    col_order = [
+        field["name"] if isinstance(field["name"], str) else str(field["name"])
+        for field in table["schema"]["fields"]
+    ]
     df = DataFrame(table["data"], columns=col_order)[col_order]
 
     dtypes = {
-        field["name"]: convert_json_field_to_pandas_type(field)
+        field["name"]
+        if isinstance(field["name"], str)
+        else str(field["name"]): convert_json_field_to_pandas_type(field)
         for field in table["schema"]["fields"]
     }
 
@@ -388,6 +400,15 @@ def parse_table_schema(json, precise_float: bool) -> DataFrame:
 
     df = df.astype(dtypes)
 
+    # Convert column names back to their original types
+    original_types = {
+        str(field["name"])
+        if not isinstance(field["name"], str)
+        else field["name"]: field["name"]
+        for field in table["schema"]["fields"]
+    }
+    df.columns = [original_types[col] for col in df.columns]
+
     if "primaryKey" in table["schema"]:
         df = df.set_index(table["schema"]["primaryKey"])
         if len(df.index.names) == 1: