From 263ebf1d386299f3eefd1eeb943a63c1029f0cce Mon Sep 17 00:00:00 2001
From: quant <kamil246@mail.ru>
Date: Sat, 20 Jun 2026 22:02:41 +0500
Subject: [PATCH] feat: exclude selected fields from json output

---
 src/data_profiling/profile_report.py | 26 ++++++++++-
 tests/issues/test_issue1680.py       | 68 ++++++++++++++++++++++++++++
 2 files changed, 93 insertions(+), 1 deletion(-)
 create mode 100644 tests/issues/test_issue1680.py

diff --git a/src/data_profiling/profile_report.py b/src/data_profiling/profile_report.py
index 8d68b8775..f973d9da1 100644
--- a/src/data_profiling/profile_report.py
+++ b/src/data_profiling/profile_report.py
@@ -73,6 +73,7 @@ def __init__(
         sortby: Optional[str] = None,
         sensitive: bool = False,
         explorative: bool = False,
+        excluded_fields: list[str] | None = None,
         sample: Optional[dict] = None,
         config_file: Optional[Union[Path, str]] = None,
         lazy: bool = True,
@@ -97,6 +98,7 @@ def __init__(
             Only available for pd.DataFrame
             sort_by: ignored if ts_mode=False. Order the dataset by a provided column.
             sensitive: hides the values for categorical and text variables for report privacy
+            excluded_fields: excludes selected fields from JSON generation
             config_file: a config file (.yml), mutually exclusive with `minimal`
             lazy: compute when needed
             sample: optional dict(name="Sample title", caption="Caption", data=pd.DataFrame())
@@ -110,6 +112,8 @@ def __init__(
 
         self._df_type = type(df)
 
+        self.excluded_fields = excluded_fields
+
         if config_file or minimal:
             if not config_file:
                 config_file = get_config("config_minimal.yaml")
@@ -453,7 +457,11 @@ def encode_it(o: Any) -> Any:
             if is_dataclass(o):
                 o = asdict(o)
             if isinstance(o, dict):
-                return {encode_it(k): encode_it(v) for k, v in o.items()}
+                return {
+                    encode_it(k): encode_it(v)
+                    for k, v in o.items()
+                    if (self.excluded_fields is None or k not in self.excluded_fields)
+                }
             else:
                 if isinstance(o, (bool, int, float, str)):
                     return o
@@ -502,6 +510,22 @@ def to_json(self) -> str:
 
         Returns:
             JSON string
+
+        To disable fields for output, pass a list with
+        of fields to exclude.
+        Example:
+            from data_profiling import ProfileReport
+            import pandas as pd
+
+            df = pd.DataFrame(np.random.rand(7, 3), columns=["a", "b", "c"])
+            fields = ["value_counts_without_nan",
+            "value_counts_index_sorted", "n_var"]
+            profile = ProfileReport(
+            df, title="Data Profiling Report",
+            excluded_fields=fields
+            )
+            report_json = profile.to_json()
+
         """
 
         return self.json
diff --git a/tests/issues/test_issue1680.py b/tests/issues/test_issue1680.py
new file mode 100644
index 000000000..14ae96c70
--- /dev/null
+++ b/tests/issues/test_issue1680.py
@@ -0,0 +1,68 @@
+"""
+Test for issue 1680.
+
+https://github.com/Data-Centric-AI-Community/fg-data-profiling/issues/1680
+"""
+import json
+
+import numpy as np
+import pandas as pd
+import pytest
+from typeguard import TypeCheckError
+
+from data_profiling import ProfileReport
+
+
+@pytest.fixture
+def data():
+    """Create a dataframe."""
+    np.random.seed(42)
+
+    return pd.DataFrame(np.random.rand(7, 3), columns=["a", "b", "c"])
+
+
+def test_empty_df(data):
+    """Check if the data frame is empty."""
+    if not len(data):
+        raise ValueError("Check the dataframe it is empty")
+
+
+# Checking excluded fields
+def test_excluded_fields(data):
+    """Check if there are excluded fields in the json."""
+    test_empty_df(data)
+
+    excluded_fields = [
+        "value_counts_without_nan",
+        "value_counts_index_sorted",
+        "n_var",
+        "mean",
+        "sum",
+        "mad",
+    ]
+
+    report = ProfileReport(data, excluded_fields=excluded_fields)
+    report_json = report.to_json()
+    data_json = json.loads(report_json)
+
+    def check_no_key(dict_recursion):
+        for key, value in dict_recursion.items():
+            if key in excluded_fields:
+                description = f"The field: {key} " f"that was excluded is present."
+                raise ValueError(description)
+            if isinstance(value, dict):
+                check_no_key(value)
+
+    check_no_key(data_json)
+
+
+@pytest.mark.parametrize("invalid_value", [12345, "abcde", {}, True])
+def test_invalid_excluded_fields(data, invalid_value):
+    """
+    Validate excluded_fields types.
+
+    Checks that passing invalid types
+    to excluded_fields causes a type error.
+    """
+    with pytest.raises(TypeCheckError):
+        ProfileReport(data, excluded_fields=invalid_value)  # type: ignore # noqa