~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
Python academiaanthropicarxivbravedeep-researchencryptionhome-automationhomeserverlocallocal-deep-researchlocal-llmmistralollamaopenaipubmedresearchresearch-toolretrieval-augmented-generationsearxngself-hosted
"""Security settings utilities.
Provides functions to load security settings from JSON defaults with
environment variable overrides. Used during app initialization before
the full settings system is available.
"""
import json
from functools import lru_cache
from pathlib import Path
from typing import Any, Optional, TypeVar, Union
from loguru import logger
from local_deep_research.utilities.type_utils import unwrap_setting
from local_deep_research.settings.manager import check_env_setting
T = TypeVar("T", int, float, str, bool)
# Path to security settings JSON
_SETTINGS_PATH = (
Path(__file__).parent.parent / "defaults" / "settings_security.json"
)
@lru_cache(maxsize=1)
def _load_security_settings() -> dict:
"""Load and cache security settings from JSON file.
Returns:
Dictionary of security settings, or empty dict on error.
"""
try:
if _SETTINGS_PATH.exists():
# nosemgrep: semgrep.rules.path-traversal-risk, reason: Fixed settings filename below the installed package __file__ directory
with open(_SETTINGS_PATH, "r", encoding="utf-8-sig") as f:
return json.load(f)
except Exception:
logger.warning(
f"Failed to load security settings from {_SETTINGS_PATH}"
)
return {}
def _convert_value(value: str, target_type: type, key: str) -> Optional[Any]:
"""Convert string value to target type with proper error handling.
Args:
value: String value to convert
target_type: Target type (int, float, str, bool)
key: Setting key for error messages
Returns:
Converted value or None if conversion fails
"""
try:
if target_type is bool:
# Handle boolean strings properly
return value.lower() in ("true", "1", "yes", "on")
if target_type is int:
return int(value)
if target_type is float:
return float(value)
return str(value)
except (ValueError, TypeError):
logger.warning(
f"Invalid value for {key}: '{value}' cannot be converted to {target_type.__name__}"
)
return None
def _validate_bounds(
value: Union[int, float],
min_value: Optional[Union[int, float]],
max_value: Optional[Union[int, float]],
key: str,
) -> Union[int, float]:
"""Validate that value is within min/max bounds.
Args:
value: Value to validate
min_value: Minimum allowed value (inclusive)
max_value: Maximum allowed value (inclusive)
key: Setting key for error messages
Returns:
Clamped value within bounds
"""
original = value
if min_value is not None and value < min_value:
value = min_value
logger.warning(
f"Value {original} for {key} is below minimum {min_value}, using {value}"
)
if max_value is not None and value > max_value:
value = max_value
logger.warning(
f"Value {original} for {key} is above maximum {max_value}, using {value}"
)
return value
def get_security_default(key: str, default: T) -> T:
"""Load a security setting with environment variable override.
Priority order:
1. Environment variable (LDR_SECURITY_<KEY>)
2. JSON defaults file value
3. Provided default
Environment variables are validated against min/max bounds defined
in the JSON settings file.
Args:
key: Setting key (e.g., "security.session_remember_me_days")
default: Default value if setting not found (also determines type)
Returns:
Setting value of same type as default
Example:
>>> get_security_default("security.session_remember_me_days", 30)
30 # or value from env/JSON
"""
settings = _load_security_settings()
setting_data = settings.get(key, {})
# Get min/max bounds from settings schema
min_value = (
setting_data.get("min_value")
if isinstance(setting_data, dict)
else None
)
max_value = (
setting_data.get("max_value")
if isinstance(setting_data, dict)
else None
)
# Check environment variable first (e.g., LDR_SECURITY_SESSION_REMEMBER_ME_DAYS)
env_key = f"LDR_{key.upper().replace('.', '_')}"
env_value = check_env_setting(key)
if env_value is not None:
converted = _convert_value(env_value, type(default), env_key)
if converted is not None:
# Validate bounds for numeric types
if isinstance(converted, (int, float)) and (
min_value is not None or max_value is not None
):
converted = _validate_bounds(
converted, min_value, max_value, env_key
)
return converted # type: ignore
# Load from JSON defaults
if key in settings:
return unwrap_setting(setting_data)
return default