BUG: Fix record_prefix ignored when record_path is empty (GH 62205)

- Modified early return optimization to exclude cases where record_prefix is provided
- Added proper prefix application using nested_to_record for consistent separator handling
- Added comprehensive test case to verify the fix
- Maintains backward compatibility and performance optimization for cases without record_prefix

Fixes #62205

Author: SumitkCodes

pandas/io/json/_normalize.py

@@ -524,7 +524,16 @@ def _pull_records(js: dict[str, Any], spec: list | str) -> list:
             # TODO: handle record value which are lists, at least error
             #       reasonably
             data = nested_to_record(data, sep=sep, max_level=max_level)
-        return DataFrame(data, index=index)
+        
+        result = DataFrame(data, index=index)
+        
+        # Apply record_prefix if provided, even when record_path is None
+        if record_prefix is not None:
+            # Use nested_to_record with prefix to properly handle separators
+            prefixed_data = nested_to_record(data, prefix=record_prefix, sep=sep, level=1)
+            result = DataFrame(prefixed_data, index=index)
+        
+        return result
     elif not isinstance(record_path, list):
         record_path = [record_path]
 

pandas/tests/io/json/test_normalize.py

@@ -569,6 +569,16 @@ def test_series_index(self, state_data):
         result = json_normalize(series, "counties")
         tm.assert_index_equal(result.index, idx.repeat([3, 2]))
 
+    def test_record_prefix_without_record_path(self):
+        # GH 62205: record_prefix should be applied even when record_path is None
+        data = Series([{"k": f"{i}", "m": "q"} for i in range(5)])
+        result = json_normalize(data, record_prefix="T")
+        expected = DataFrame({
+            "T.k": [f"{i}" for i in range(5)],
+            "T.m": ["q"] * 5
+        })
+        tm.assert_frame_equal(result, expected)
+
 
 class TestNestedToRecord:
     def test_flat_stays_flat(self):