From 08628610a8ccb53e832f0e0dc70d8949b8610b9b Mon Sep 17 00:00:00 2001
From: huazhuang80-star <295584745+huazhuang80-star@users.noreply.github.com>
Date: Mon, 22 Jun 2026 22:35:40 +0800
Subject: [PATCH] refactor: deprecate root PII scrubber

---
 scrubber.py                | 20 +++++-----------
 tests/test_pii_scrubber.py | 48 ++++++++++++++++++--------------------
 2 files changed, 29 insertions(+), 39 deletions(-)

diff --git a/scrubber.py b/scrubber.py
index b93b6a43..b5ac7d0b 100644
--- a/scrubber.py
+++ b/scrubber.py
@@ -1,16 +1,8 @@
 """
-Legacy PII scrubber — preserved for reference.
-Active anonymization logic has moved to app/ai-service/services/pii_scrubber.py.
-"""
-import re
-
-def scrub_pii(text: str) -> str:
-    """Remove emails, phone numbers, and IDs from text using regex.
-
-    This is a legacy module kept for documentation purposes.
-    """
-    text = re.sub(r'[\w\.-]+@[\w\.-]+', '[REDACTED_EMAIL]', text)
-    text = re.sub(r'\+?\d[\d\s\-]{7,}\d', '[REDACTED_PHONE]', text)
-    text = re.sub(r'\b\d{4,}\b', '[REDACTED_ID]', text)
-    return text
+Deprecated root PII scrubber module.
 
+The active anonymization implementation lives in
+app/ai-service/services/pii_scrubber.py as PIIScrubberService.
+This module intentionally no longer exposes the legacy regex-only scrub_pii
+function so new code does not depend on the deprecated implementation.
+"""
diff --git a/tests/test_pii_scrubber.py b/tests/test_pii_scrubber.py
index 7240af7e..f163732c 100644
--- a/tests/test_pii_scrubber.py
+++ b/tests/test_pii_scrubber.py
@@ -1,36 +1,34 @@
-import json
-import difflib
-from scrubber import scrub_pii
+import ast
+from pathlib import Path
 
 
-def load_json(path):
-    with open(path, "r") as f:
-        return json.load(f)
+ROOT = Path(__file__).resolve().parents[1]
+ROOT_SCRUBBER = ROOT / "scrubber.py"
+AI_SERVICE_SCRUBBER = ROOT / "app" / "ai-service" / "services" / "pii_scrubber.py"
 
 
-inputs = load_json("tests/fixtures/pii_inputs.json")
-expected = load_json("tests/fixtures/expected_outputs.json")
+def _function_names(path: Path) -> set[str]:
+    tree = ast.parse(path.read_text(encoding="utf-8"))
+    return {node.name for node in ast.walk(tree) if isinstance(node, ast.FunctionDef)}
 
 
-def test_pii_scrubbing():
+def _class_method_names(path: Path, class_name: str) -> set[str]:
+    tree = ast.parse(path.read_text(encoding="utf-8"))
+    for node in ast.walk(tree):
+        if isinstance(node, ast.ClassDef) and node.name == class_name:
+            return {
+                child.name
+                for child in node.body
+                if isinstance(child, ast.FunctionDef)
+            }
+    return set()
 
-    for inp, exp in zip(inputs, expected):
 
-        result = scrub_pii(inp["input"])
+def test_root_scrubber_no_longer_exports_legacy_scrub_function():
+    assert "scrub_pii" not in _function_names(ROOT_SCRUBBER)
 
-        if result != exp["expected"]:
 
-            diff = "\n".join(
-                difflib.unified_diff(
-                    [exp["expected"]],
-                    [result],
-                    fromfile="expected",
-                    tofile="actual",
-                    lineterm=""
-                )
-            )
+def test_ai_service_pii_scrubber_is_canonical_implementation():
+    methods = _class_method_names(AI_SERVICE_SCRUBBER, "PIIScrubberService")
 
-            print("\nRegression Detected:")
-            print(diff)
-
-        assert result == exp["expected"]
+    assert "anonymize" in methods