DATA
6de0bf9f5b
Vendored deer-flow upstream (bytedance/deer-flow) plus prompt-injection hardening: - New deerflow.security package: content_delimiter, html_cleaner, sanitizer (8 layers — invisible chars, control chars, symbols, NFC, PUA, tag chars, horizontal whitespace collapse with newline/tab preservation, length cap) - New deerflow.community.searx package: web_search, web_fetch, image_search backed by a private SearX instance, every external string sanitized and wrapped in <<<EXTERNAL_UNTRUSTED_CONTENT>>> delimiters - All native community web providers (ddg_search, tavily, exa, firecrawl, jina_ai, infoquest, image_search) replaced with hard-fail stubs that raise NativeWebToolDisabledError at import time, so a misconfigured tool.use path fails loud rather than silently falling back to unsanitized output - Native client back-doors (jina_client.py, infoquest_client.py) stubbed too - Native-tool tests quarantined under tests/_disabled_native/ (collect_ignore_glob via local conftest.py) - Sanitizer Layer 7 fix: only collapse horizontal whitespace, preserve newlines and tabs so list/table structure survives - Hardened runtime config.yaml references only the searx-backed tools - Factory overlay (backend/) kept in sync with deer-flow tree as a reference / source See HARDENING.md for the full audit trail and verification steps.
79 lines
2.6 KiB
Python
79 lines
2.6 KiB
Python
"""Canonical serialization for LangChain / LangGraph objects.
|
|
|
|
Provides a single source of truth for converting LangChain message
|
|
objects, Pydantic models, and LangGraph state dicts into plain
|
|
JSON-serialisable Python structures.
|
|
|
|
Consumers: ``deerflow.runtime.runs.worker`` (SSE publishing) and
|
|
``app.gateway.routers.threads`` (REST responses).
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from typing import Any
|
|
|
|
|
|
def serialize_lc_object(obj: Any) -> Any:
|
|
"""Recursively serialize a LangChain object to a JSON-serialisable dict."""
|
|
if obj is None:
|
|
return None
|
|
if isinstance(obj, (str, int, float, bool)):
|
|
return obj
|
|
if isinstance(obj, dict):
|
|
return {k: serialize_lc_object(v) for k, v in obj.items()}
|
|
if isinstance(obj, (list, tuple)):
|
|
return [serialize_lc_object(item) for item in obj]
|
|
# Pydantic v2
|
|
if hasattr(obj, "model_dump"):
|
|
try:
|
|
return obj.model_dump()
|
|
except Exception:
|
|
pass
|
|
# Pydantic v1 / older objects
|
|
if hasattr(obj, "dict"):
|
|
try:
|
|
return obj.dict()
|
|
except Exception:
|
|
pass
|
|
# Last resort
|
|
try:
|
|
return str(obj)
|
|
except Exception:
|
|
return repr(obj)
|
|
|
|
|
|
def serialize_channel_values(channel_values: dict[str, Any]) -> dict[str, Any]:
|
|
"""Serialize channel values, stripping internal LangGraph keys.
|
|
|
|
Internal keys like ``__pregel_*`` and ``__interrupt__`` are removed
|
|
to match what the LangGraph Platform API returns.
|
|
"""
|
|
result: dict[str, Any] = {}
|
|
for key, value in channel_values.items():
|
|
if key.startswith("__pregel_") or key == "__interrupt__":
|
|
continue
|
|
result[key] = serialize_lc_object(value)
|
|
return result
|
|
|
|
|
|
def serialize_messages_tuple(obj: Any) -> Any:
|
|
"""Serialize a messages-mode tuple ``(chunk, metadata)``."""
|
|
if isinstance(obj, tuple) and len(obj) == 2:
|
|
chunk, metadata = obj
|
|
return [serialize_lc_object(chunk), metadata if isinstance(metadata, dict) else {}]
|
|
return serialize_lc_object(obj)
|
|
|
|
|
|
def serialize(obj: Any, *, mode: str = "") -> Any:
|
|
"""Serialize LangChain objects with mode-specific handling.
|
|
|
|
* ``messages`` — obj is ``(message_chunk, metadata_dict)``
|
|
* ``values`` — obj is the full state dict; ``__pregel_*`` keys stripped
|
|
* everything else — recursive ``model_dump()`` / ``dict()`` fallback
|
|
"""
|
|
if mode == "messages":
|
|
return serialize_messages_tuple(obj)
|
|
if mode == "values":
|
|
return serialize_channel_values(obj) if isinstance(obj, dict) else serialize_lc_object(obj)
|
|
return serialize_lc_object(obj)
|