File size: 18,121 Bytes
5374a2d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 |
import os
import yaml
import json
import copy
import logging
from typing import Callable, Any, Dict, List
from pydantic import BaseModel, ValidationError
from pydantic._internal._model_construction import ModelMetaclass
from .logging import logger
from .callbacks import callback_manager, exception_buffer
from .module_utils import (
save_json,
custom_serializer,
parse_json_from_text,
get_error_message,
get_base_module_init_error_message
)
from .registry import register_module, MODULE_REGISTRY
class MetaModule(ModelMetaclass):
"""
MetaModule is a metaclass that automatically registers all subclasses of BaseModule.
Attributes:
No public attributes
"""
def __new__(mcs, name, bases, namespace, **kwargs):
"""
Creates a new class and registers it in MODULE_REGISTRY.
Args:
mcs: The metaclass itself
name: The name of the class being created
bases: Tuple of base classes
namespace: Dictionary containing the class attributes and methods
**kwargs: Additional keyword arguments
Returns:
The created class object
"""
cls = super().__new__(mcs, name, bases, namespace)
register_module(name, cls)
return cls
class BaseModule(BaseModel, metaclass=MetaModule):
"""
Base module class that serves as the foundation for all modules in the EvoAgentX framework.
This class provides serialization/deserialization capabilities, supports creating instances from
dictionaries, JSON, or files, and exporting instances to these formats.
Attributes:
class_name: The class name, defaults to None but is automatically set during subclass initialization
model_config: Pydantic model configuration that controls type matching and behavior
"""
class_name: str = None
# NOTE: do not set "validate_assignment" to True, otherwise infinite recursion will occur when validating the model.
model_config = {"arbitrary_types_allowed": True, "extra": "allow", "protected_namespaces": (), "validate_assignment": False}
def __init_subclass__(cls, **kwargs):
"""
Subclass initialization method that automatically sets the class_name attribute.
Args:
cls (Type): The subclass being initialized
**kwargs (Any): Additional keyword arguments
"""
super().__init_subclass__(**kwargs)
cls.class_name = cls.__name__
def __init__(self, **kwargs):
"""
Initializes a BaseModule instance.
Args:
**kwargs (Any): Keyword arguments used to initialize the instance
Raises:
ValidationError: When parameter validation fails
Exception: When other errors occur during initialization
"""
try:
for field_name, _ in type(self).model_fields.items():
field_value = kwargs.get(field_name, None)
if field_value:
kwargs[field_name] = self._process_data(field_value)
# if field_value and isinstance(field_value, dict) and "class_name" in field_value:
# class_name = field_value.get("class_name")
# sub_cls = MODULE_REGISTRY.get_module(cls_name=class_name)
# kwargs[field_name] = sub_cls._create_instance(field_value)
super().__init__(**kwargs)
self.init_module()
except (ValidationError, Exception) as e:
exception_handler = callback_manager.get_callback("exception_buffer")
if exception_handler is None:
error_message = get_base_module_init_error_message(
cls=self.__class__,
data=kwargs,
errors=e
)
logger.error(error_message)
raise
else:
exception_handler.add(e)
def init_module(self):
"""
Module initialization method that subclasses can override to provide additional initialization logic.
"""
pass
def __str__(self) -> str:
"""
Returns a string representation of the object.
Returns:
str: String representation of the object
"""
return self.to_str()
@property
def kwargs(self) -> dict:
"""
Returns the extra fields of the model.
Returns:
dict: Dictionary containing all extra keyword arguments
"""
return self.model_extra
@classmethod
def _create_instance(cls, data: Dict[str, Any]) -> "BaseModule":
"""
Internal method for creating an instance from a dictionary.
Args:
data: Dictionary containing instance data
Returns:
BaseModule: The created instance
"""
processed_data = {k: cls._process_data(v) for k, v in data.items()}
# print(processed_data)
return cls.model_validate(processed_data)
@classmethod
def _process_data(cls, data: Any) -> Any:
"""
Recursive method for processing data, with special handling for dictionaries containing class_name.
Args:
data: Data to be processed
Returns:
Processed data
"""
if isinstance(data, dict):
if "class_name" in data:
sub_class = MODULE_REGISTRY.get_module(data.get("class_name"))
return sub_class._create_instance(data)
else:
return {k: cls._process_data(v) for k, v in data.items()}
elif isinstance(data, (list, tuple)):
return [cls._process_data(x) for x in data]
else:
return data
@classmethod
def from_dict(cls, data: Dict[str, Any], **kwargs) -> "BaseModule":
"""
Instantiate the BaseModule from a dictionary.
Args:
data: Dictionary containing instance data
**kwargs (Any): Additional keyword arguments, can include log to control logging output
Returns:
BaseModule: The created module instance
Raises:
Exception: When errors occur during initialization
"""
use_logger = kwargs.get("log", True)
with exception_buffer() as buffer:
try:
class_name = data.get("class_name", None)
if class_name:
cls = MODULE_REGISTRY.get_module(class_name)
module = cls._create_instance(data)
# module = cls.model_validate(data)
if len(buffer.exceptions) > 0:
error_message = get_base_module_init_error_message(cls, data, buffer.exceptions)
if use_logger:
logger.error(error_message)
raise Exception(get_error_message(buffer.exceptions))
finally:
pass
return module
@classmethod
def from_json(cls, content: str, **kwargs) -> "BaseModule":
"""
Construct the BaseModule from a JSON string.
This method uses yaml.safe_load to parse the JSON string into a Python object,
which supports more flexible parsing than standard json.loads (including handling
single quotes, trailing commas, etc). The parsed data is then passed to from_dict
to create the instance.
Args:
content: JSON string
**kwargs (Any): Additional keyword arguments, can include `log` to control logging output
Returns:
BaseModule: The created module instance
Raises:
ValueError: When the input is not a valid JSON string
"""
use_logger = kwargs.get("log", True)
try:
data = yaml.safe_load(content)
except Exception:
error_message = f"Can not instantiate {cls.__name__}. The input to {cls.__name__}.from_json is not a valid JSON string."
if use_logger:
logger.error(error_message)
raise ValueError(error_message)
if not isinstance(data, (list, dict)):
error_message = f"Can not instantiate {cls.__name__}. The input to {cls.__name__}.from_json is not a valid JSON string."
if use_logger:
logger.error(error_message)
raise ValueError(error_message)
return cls.from_dict(data, log=use_logger)
@classmethod
def from_str(cls, content: str, **kwargs) -> "BaseModule":
"""
Construct the BaseModule from a string that may contain JSON.
This method is more forgiving than `from_json` as it can extract valid JSON
objects embedded within larger text. It uses `parse_json_from_text` to extract
all potential JSON strings from the input text, then tries to create an instance
from each extracted JSON string until successful.
Args:
content: Text that may contain JSON strings
**kwargs (Any): Additional keyword arguments, can include `log` to control logging output
Returns:
BaseModule: The created module instance
Raises:
ValueError: When the input does not contain valid JSON strings or the JSON is incompatible with the class
"""
use_logger = kwargs.get("log", True)
extracted_json_list = parse_json_from_text(content)
if len(extracted_json_list) == 0:
error_message = f"The input to {cls.__name__}.from_str does not contain any valid JSON str."
if use_logger:
logger.error(error_message)
raise ValueError(error_message)
module = None
for json_str in extracted_json_list:
try:
module = cls.from_json(json_str, log=False)
except Exception:
continue
break
if module is None:
error_message = f"Can not instantiate {cls.__name__}. The input to {cls.__name__}.from_str either does not contain a valide JSON str, or the JSON str is incomplete or incompatable (incorrect variables or types) with {cls.__name__}."
error_message += f"\nInput:\n{content}"
if use_logger:
logger.error(error_message)
raise ValueError(error_message)
return module
@classmethod
def load_module(cls, path: str, **kwargs) -> dict:
"""
Load the values for a module from a file.
By default, it opens the specified file and uses `yaml.safe_load` to parse its contents
into a Python object (typically a dictionary).
Args:
path: The path of the file
**kwargs (Any): Additional keyword arguments
Returns:
dict: The JSON object instantiated from the file
"""
with open(path, mode="r", encoding="utf-8") as file:
content = yaml.safe_load(file.read())
return content
@classmethod
def from_file(cls, path: str, load_function: Callable=None, **kwargs) -> "BaseModule":
"""
Construct the BaseModule from a file.
This method reads and parses a file into a data structure, then creates
a module instance from that data. It first verifies that the file exists,
then uses either the provided `load_function` or the default `load_module`
method to read and parse the file content, and finally calls `from_dict`
to create the instance.
Args:
path: The path of the file
load_function: The function used to load the data, takes a file path as input and returns a JSON object
**kwargs (Any): Additional keyword arguments, can include `log` to control logging output
Returns:
BaseModule: The created module instance
Raises:
ValueError: When the file does not exist
"""
use_logger = kwargs.get("log", True)
if not os.path.exists(path):
error_message = f"File \"{path}\" does not exist!"
if use_logger:
logger.error(error_message)
raise ValueError(error_message)
function = load_function or cls.load_module
content = function(path, **kwargs)
module = cls.from_dict(content, log=use_logger)
return module
# def to_dict(self, **kwargs) -> dict:
# """
# convert the BaseModule to a dict.
# """
# return self.model_dump()
def to_dict(self, exclude_none: bool = True, ignore: List[str] = [], **kwargs) -> dict:
"""
Convert the BaseModule to a dictionary.
Args:
exclude_none: Whether to exclude fields with None values
ignore: List of field names to ignore
**kwargs (Any): Additional keyword arguments
Returns:
dict: Dictionary containing the object data
"""
data = {}
for field_name, _ in type(self).model_fields.items():
if field_name in ignore:
continue
field_value = getattr(self, field_name, None)
if exclude_none and field_value is None:
continue
if isinstance(field_value, BaseModule):
data[field_name] = field_value.to_dict(exclude_none=exclude_none, ignore=ignore)
elif isinstance(field_value, list):
data[field_name] = [
item.to_dict(exclude_none=exclude_none, ignore=ignore) if isinstance(item, BaseModule) else item
for item in field_value
]
elif isinstance(field_value, dict):
data[field_name] = {
key: value.to_dict(exclude_none=exclude_none, ignore=ignore) if isinstance(value, BaseModule) else value
for key, value in field_value.items()
}
else:
data[field_name] = field_value
return data
def to_json(self, use_indent: bool=False, ignore: List[str] = [], **kwargs) -> str:
"""
Convert the BaseModule to a JSON string.
Args:
use_indent: Whether to use indentation
ignore: List of field names to ignore
**kwargs (Any): Additional keyword arguments
Returns:
str: The JSON string
"""
if use_indent:
kwargs["indent"] = kwargs.get("indent", 4)
else:
kwargs.pop("indent", None)
if kwargs.get("default", None) is None:
kwargs["default"] = custom_serializer
data = self.to_dict(exclude_none=True)
for ignore_field in ignore:
data.pop(ignore_field, None)
return json.dumps(data, **kwargs)
def to_str(self, **kwargs) -> str:
"""
Convert the BaseModule to a string. Use .to_json to output JSON string by default.
Args:
**kwargs (Any): Additional keyword arguments
Returns:
str: The string
"""
return self.to_json(use_indent=False)
def save_module(self, path: str, ignore: List[str] = [], **kwargs)-> str:
"""
Save the BaseModule to a file.
This method will set non-serializable objects to None by default.
If you want to save non-serializable objects, override this method.
Remember to also override the `load_module` function to ensure the loaded
object can be correctly parsed by `cls.from_dict`.
Args:
path: The path to save the file
ignore: List of field names to ignore
**kwargs (Any): Additional keyword arguments
Returns:
str: The path where the file is saved, same as the input path
"""
logger.info("Saving {} to {}", self.__class__.__name__, path)
return save_json(self.to_json(use_indent=True, default=lambda x: None, ignore=ignore), path=path)
def deepcopy(self):
"""Deep copy the module.
This is a tweak to the default python deepcopy that only deep copies `self.parameters()`, and for other
attributes, we just do the shallow copy.
"""
try:
# If the instance itself is copyable, we can just deep copy it.
# Otherwise we will have to create a new instance and copy over the attributes one by one.
return copy.deepcopy(self)
except Exception:
pass
# Create an empty instance.
new_instance = self.__class__.__new__(self.__class__)
# Set attribuetes of the copied instance.
for attr, value in self.__dict__.items():
if isinstance(value, BaseModule):
setattr(new_instance, attr, value.deepcopy())
else:
try:
# Try to deep copy the attribute
setattr(new_instance, attr, copy.deepcopy(value))
except Exception:
logging.warning(
f"Failed to deep copy attribute '{attr}' of {self.__class__.__name__}, "
"falling back to shallow copy or reference copy."
)
try:
# Fallback to shallow copy if deep copy fails
setattr(new_instance, attr, copy.copy(value))
except Exception:
# If even the shallow copy fails, we just copy over the reference.
setattr(new_instance, attr, value)
return new_instance
__all__ = ["BaseModule"]
|