The agent engineering platform.
Python agentsaiai-agentsanthropicchatgptdeepagentsenterpriseframeworkgeminigenerative-ailangchainlanggraphllmmultiagentopen-sourceopenaipydanticpythonragtypescript
fix(langchain): normalize custom detector output to prevent KeyError in hash/mask strategies (#35651)
Giulio Leone46fdade7e68b
diff --git a/libs/langchain_v1/langchain/agents/middleware/_redaction.py b/libs/langchain_v1/langchain/agents/middleware/_redaction.pyindex 18c1bb17df..df93d7075f 100644--- a/libs/langchain_v1/langchain/agents/middleware/_redaction.py+++ b/libs/langchain_v1/langchain/agents/middleware/_redaction.py@@ -373,7 +373,25 @@ def resolve_detector(pii_type: str, detector: Detector | str | None) -> Detector]return regex_detector- return detector++ # Wrap the custom callable to normalize its output.+ # Custom detectors may return dicts with "text" instead of "value"+ # and may omit "type". Map them to proper PIIMatch objects so that+ # downstream strategies (hash, mask) can access match["value"].+ raw_detector = detector++ def _normalizing_detector(content: str) -> list[PIIMatch]:+ return [+ PIIMatch(+ type=m.get("type", pii_type),+ value=m.get("value", m.get("text", "")),+ start=m["start"],+ end=m["end"],+ )+ for m in raw_detector(content)+ ]++ return _normalizing_detector@dataclass(frozen=True)diff --git a/libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_pii.py b/libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_pii.pyindex 40c199591e..8738942562 100644--- a/libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_pii.py+++ b/libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_pii.py@@ -1,5 +1,6 @@"""Tests for PII detection middleware."""+import refrom typing import Anyimport pytest@@ -557,6 +558,57 @@ class TestCustomDetector:assert result is not Noneassert "[REDACTED_CONFIDENTIAL]" in result["messages"][0].content+ def test_custom_callable_detector_with_text_key_hash(self) -> None:+ """Custom detectors returning 'text' instead of 'value' must work with hash strategy.++ Regression test for https://github.com/langchain-ai/langchain/issues/35647:+ Custom detectors documented to return {"text", "start", "end"} caused+ KeyError: 'value' when used with hash or mask strategies.+ """++ def detect_phone(content: str) -> list[dict]: # type: ignore[type-arg]+ return [+ {"text": m.group(), "start": m.start(), "end": m.end()}+ for m in re.finditer(r"\+91[\s.-]?\d{10}", content)+ ]++ middleware = PIIMiddleware(+ "indian_phone",+ detector=detect_phone,+ strategy="hash",+ apply_to_input=True,+ )++ state = AgentState[Any](messages=[HumanMessage("Call +91 9876543210")])+ result = middleware.before_model(state, Runtime())++ assert result is not None+ assert "<indian_phone_hash:" in result["messages"][0].content+ assert "+91 9876543210" not in result["messages"][0].content++ def test_custom_callable_detector_with_text_key_mask(self) -> None:+ """Custom detectors returning 'text' instead of 'value' must work with mask strategy."""++ def detect_phone(content: str) -> list[dict]: # type: ignore[type-arg]+ return [+ {"text": m.group(), "start": m.start(), "end": m.end()}+ for m in re.finditer(r"\+91[\s.-]?\d{10}", content)+ ]++ middleware = PIIMiddleware(+ "indian_phone",+ detector=detect_phone,+ strategy="mask",+ apply_to_input=True,+ )++ state = AgentState[Any](messages=[HumanMessage("Call +91 9876543210")])+ result = middleware.before_model(state, Runtime())++ assert result is not None+ assert "****" in result["messages"][0].content+ assert "+91 9876543210" not in result["messages"][0].content+def test_unknown_builtin_type_raises_error(self) -> None:with pytest.raises(ValueError, match="Unknown PII type"):PIIMiddleware("unknown_type", strategy="redact")