docs(examples): add pandera interoperability recipe (#169)

Author: naseem173

examples/09_pandera_recipe.py

@@ -0,0 +1,65 @@
+"""Recipe: validate a messy DataFrame with pandera, repair it with freshdata,
+then validate again to see what's fixed and what still needs manual repair.
+"""
+
+import pandas as pd
+import pandera.pandas as pa
+
+import freshdata as fd
+
+
+def main() -> None:
+    df = pd.DataFrame(
+        {
+            "Age": [25, 41, None, 33],
+            "Amount($)": ["$19.99", "$5.00", "$8.25", "n/a"],
+            "Status": [" shipped ", "PENDING", "shipped", "pending"],
+        }
+    )
+
+    raw_schema = pa.DataFrameSchema(
+        {
+            "Age": pa.Column(float, nullable=False),
+            "Amount($)": pa.Column(float, nullable=False),
+            "Status": pa.Column(str, nullable=False),
+        }
+    )
+
+    print("=== Step 1: validate the raw frame ===")
+    try:
+        raw_schema.validate(df)
+    except pa.errors.SchemaError as exc:
+        print(f"Validation failed, as expected:\n{exc}\n")
+
+    cleaned = fd.clean(df)
+
+    clean_schema = pa.DataFrameSchema(
+        {
+            "age": pa.Column(float, nullable=False),
+            "amount": pa.Column(float, nullable=False),
+            "status": pa.Column(str, pa.Check.isin(["shipped", "pending"]), nullable=False),
+        }
+    )
+
+    print("=== Step 2: validate the freshdata-cleaned frame ===")
+    try:
+        clean_schema.validate(cleaned)
+        print("Schema check passed after cleaning:")
+        print(cleaned)
+    except pa.errors.SchemaError as exc:
+        print(f"Still failing after cleaning (freshdata renamed/typed the columns, "
+              f"but doesn't impute nulls or normalize casing):\n{exc}\n")
+
+    print("=== Step 3: repair the remaining issues, then re-validate ===")
+    repaired = cleaned.copy()
+    repaired["age"] = repaired["age"].fillna(repaired["age"].median())
+    repaired["amount"] = repaired["amount"].fillna(repaired["amount"].median())
+    repaired["status"] = repaired["status"].str.lower()
+
+    clean_schema.validate(repaired)
+    print("Schema check passed after repair:")
+    print(repaired)
+
+
+if __name__ == "__main__":
+    main()

examples/README.md

@@ -19,5 +19,6 @@ python examples/01_missing_values.py
 | [`06_large_dataset.py`](06_large_dataset.py) | Cleaning a large synthetic dataset, with timing |
 | [`07_pandas_integration.py`](07_pandas_integration.py) | Dropping freshdata into an existing pandas workflow |
 | [`08_csv_automation.py`](08_csv_automation.py) | Batch CSV cleaning automation with audit logs |
+| [`09_pandera_recipe.py`](09_pandera_recipe.py) | Validating with pandera before and after freshdata cleaning |
 
 See the [documentation](https://freshcode-org.github.io/freshdata/) for full guides.