EureCA / dspy /primitives /python_interpreter.py
tonneli's picture
Delete history
f5776d3
# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
# Licensed under the Apache License, Version 2.0 (the “License”);
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an “AS IS” BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
import ast
import difflib
import importlib
import typing
import inspect
from typing import (
Any,
Callable,
Dict,
Mapping,
List,
Optional,
Set,
Tuple,
TypeVar,
Union,
)
import builtins
class InterpreterError(ValueError):
r"""An error raised when the interpreter cannot evaluate a Python
expression, due to syntax error or unsupported operations.
"""
pass
class PythonInterpreter():
r"""A customized python interpreter to control the execution of
LLM-generated codes. The interpreter makes sure the code can only execute
functions given in action space and import white list. It also supports
fuzzy variable matching to receive uncertain input variable name.
.. highlight:: none
This class is adapted from the Camel adaptation https://github.com/camel-ai/
camel/blob/9a9d71874944e9736c55cdaed3df469a8becec05/camel/utils/python_interpreter.py
which adapts from the hugging face implementation `python_interpreter.py
<https://github.com/huggingface/transformers/blob/8f093fb799246f7dd9104ff44728da0c53a9f67a
/src/transformers/tools/python_interpreter.py>`_. The original license applies::
Copyright 2023 The HuggingFace Inc. team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied. See the License for the specific language governing
permissions and limitations under the License.
Camel's modifications:
"We have modified the original code to suit our requirements. We have
encapsulated the original functions within a class and saved the
interpreter state after execution. We have added support for "import"
statements, "for" statements, and several binary and unary operators. We
have added import white list to keep `import` statement safe. Additionally,
we have modified the variable matching logic and introduced the
:obj:`fuzz_state` for fuzzy matching."
DSPy's modifications:
"We expanded upon the Camel libraries modifications by adding additional
support for "Mapping" statements, "conditional" operators, and including
the "CodePrompt" and "TextPrompt" classes for code execution.
Modifications copyright (C) 2023 CAMEL-AI.org
Args:
action_space (Dict[str, Any]): A dictionary that maps action names to
their corresponding functions or objects. The interpreter can only
execute functions that are either directly listed in this
dictionary or are member functions of objects listed in this
dictionary. The concept of :obj:`action_space` is derived from
EmbodiedAgent, representing the actions that an agent is capable of
performing.
import_white_list (Optional[List[str]], optional): A list that stores
the Python modules or functions that can be imported in the code.
All submodules and functions of the modules listed in this list are
importable. Any other import statements will be rejected. The
module and its submodule or function name are separated by a period
(:obj:`.`). (default: :obj:`None`)
"""
def __init__(self, action_space: Dict[str, Any],
import_white_list: Optional[List[str]] = None) -> None:
self.action_space = action_space
self.state = self.action_space.copy()
self.fuzz_state: Dict[str, Any] = {}
self.import_white_list = import_white_list or []
def execute(self, code: str, state: Optional[Dict[str, Any]] = None,
fuzz_state: Optional[Dict[str, Any]] = None,
keep_state: bool = True) -> Any:
r""" Execute the input python codes in a security environment.
Args:
code (str): Generated python code to be executed.
state (Optional[Dict[str, Any]], optional): External variables that
may be used in the generated code. (default: :obj:`None`)
fuzz_state (Optional[Dict[str, Any]], optional): External variables
that do not have certain variable names. The interpreter will
use fuzzy matching to access these variables. For example, if
:obj:`fuzz_state` has a variable :obj:`image`, the generated
code can use :obj:`input_image` to access it. (default:
:obj:`None`)
keep_state (bool, optional): If :obj:`True`, :obj:`state` and
:obj:`fuzz_state` will be kept for later execution. Otherwise,
they will be cleared. (default: :obj:`True`)
Returns:
Any: The value of the last statement (excluding "import") in the
code. For this interpreter, the value of an expression is its
value, the value of an "assign" statement is the assigned
value, and the value of an "if" and "for" block statement is
the value of the last statement in the block.
"""
if state is not None:
self.state.update(state)
if fuzz_state is not None:
self.fuzz_state.update(fuzz_state)
try:
expression = ast.parse(code)
except SyntaxError as e:
error_line = code.splitlines()[e.lineno - 1]
raise InterpreterError(f"Syntax error in code at line {e.lineno}: {error_line}\nError: {e}")
result = None
for idx, node in enumerate(expression.body):
try:
line_result = self._execute_ast(node)
except InterpreterError as e:
if not keep_state:
self.clear_state()
msg = (f"Evaluation of the code stopped at node {idx}. "
f"See:\n{e}")
# More information can be provided by `ast.unparse()`,
# which is new in python 3.9.
raise InterpreterError(msg)
if line_result is not None:
result = line_result
if not keep_state:
self.clear_state()
return result
def clear_state(self) -> None:
r"""Initialize :obj:`state` and :obj:`fuzz_state`"""
self.state = self.action_space.copy()
self.fuzz_state = {}
# ast.Index is deprecated after python 3.9, which cannot pass type check,
# but is still necessary for older versions.
@typing.no_type_check
def _execute_ast(self, expression: ast.AST) -> Any:
if isinstance(expression, ast.Assign):
# Assignment -> evaluate the assignment which should
# update the state. We return the variable assigned as it may
# be used to determine the final result.
return self._execute_assign(expression)
elif isinstance(expression, ast.Attribute):
value = self._execute_ast(expression.value)
return getattr(value, expression.attr)
elif isinstance(expression, ast.AugAssign):
return self._execute_augassign(expression)
elif isinstance(expression, ast.BinOp):
# Binary Operator -> return the result value
return self._execute_binop(expression)
elif isinstance(expression, ast.Call):
# Function call -> return the value of the function call
return self._execute_call(expression)
elif isinstance(expression, ast.Compare):
return self._execute_condition(expression)
elif isinstance(expression, ast.Constant):
# Constant -> just return the value
return expression.value
elif isinstance(expression, ast.Dict):
# Dict -> evaluate all keys and values
result: Dict = {}
for k, v in zip(expression.keys, expression.values):
if k is not None:
result[self._execute_ast(k)] = self._execute_ast(v)
else:
result.update(self._execute_ast(v))
return result
elif isinstance(expression, ast.Expr):
# Expression -> evaluate the content
return self._execute_ast(expression.value)
elif isinstance(expression, ast.For):
return self._execute_for(expression)
elif isinstance(expression, ast.FormattedValue):
# Formatted value (part of f-string) -> evaluate the content
# and return
return self._execute_ast(expression.value)
elif isinstance(expression, ast.FunctionDef):
self.state[expression.name] = expression
return None
elif isinstance(expression, ast.If):
# If -> execute the right branch
return self._execute_if(expression)
elif isinstance(expression, ast.Import):
# Import -> add imported names in self.state and return None.
self._execute_import(expression)
return None
elif isinstance(expression, ast.ImportFrom):
self._execute_import_from(expression)
return None
elif hasattr(ast, "Index") and isinstance(expression, ast.Index):
# cannot pass type check
return self._execute_ast(expression.value)
elif isinstance(expression, ast.JoinedStr):
return "".join(
[str(self._execute_ast(v)) for v in expression.values])
elif isinstance(expression, ast.List):
# List -> evaluate all elements
return [self._execute_ast(elt) for elt in expression.elts]
elif isinstance(expression, ast.Name):
# Name -> pick up the value in the state
return self._execute_name(expression)
elif isinstance(expression, ast.Return):
return self._execute_ast(expression.value)
elif isinstance(expression, ast.Subscript):
# Subscript -> return the value of the indexing
return self._execute_subscript(expression)
elif isinstance(expression, ast.Tuple):
return tuple([self._execute_ast(elt) for elt in expression.elts])
elif isinstance(expression, ast.UnaryOp):
# Binary Operator -> return the result value
return self._execute_unaryop(expression)
else:
# For now we refuse anything else. Let's add things as we need
# them.
raise InterpreterError(
f"{expression.__class__.__name__} is not supported.")
def _execute_assign(self, assign: ast.Assign) -> Any:
targets = assign.targets
result = self._execute_ast(assign.value)
for target in targets:
self._assign(target, result)
return result
def _assign(self, target: ast.expr, value: Any):
if isinstance(target, ast.Name):
self.state[target.id] = value
elif isinstance(target, ast.Tuple):
if not isinstance(value, tuple):
raise InterpreterError(f"Expected type tuple, but got"
f"{value.__class__.__name__} instead.")
if len(target.elts) != len(value):
raise InterpreterError(
f"Expected {len(target.elts)} values but got"
f" {len(value)}.")
for t, v in zip(target.elts, value):
self.state[self._execute_ast(t)] = v
else:
raise InterpreterError(f"Unsupported variable type. Expected "
f"ast.Name or ast.Tuple, got "
f"{target.__class__.__name__} instead.")
def _execute_call(self, call: ast.Call) -> Any:
callable_func = self._execute_ast(call.func)
args = [self._execute_ast(arg) for arg in call.args]
kwargs = {
keyword.arg: self._execute_ast(keyword.value)
for keyword in call.keywords
}
if isinstance(callable_func, ast.FunctionDef):
old_state = self.state.copy()
for param_name, arg_value in zip([param.arg for param in callable_func.args.args], args):
self.state[param_name] = arg_value
result = None
for stmt in callable_func.body:
result = self._execute_ast(stmt)
if isinstance(stmt, ast.Return):
break
self.state = old_state
return result
return callable_func(*args, **kwargs)
def _execute_augassign(self, augassign: ast.AugAssign):
current_value = self.state[augassign.target.id]
increment_value = self._execute_ast(augassign.value)
if not (isinstance(current_value, (int, float)) and isinstance(increment_value, (int, float))):
raise InterpreterError(f"Invalid types for augmented assignment: {type(current_value)}, {type(increment_value)}")
if isinstance(augassign.op, ast.Add):
new_value = current_value + increment_value
elif isinstance(augassign.op, ast.Sub):
new_value = current_value - increment_value
elif isinstance(augassign.op, ast.Mult):
new_value = current_value * increment_value
elif isinstance(augassign.op, ast.Div):
new_value = current_value / increment_value
#TODO - any other augassign operators that are missing
else:
raise InterpreterError(f"Augmented assignment operator {augassign.op} is not supported")
self._assign(augassign.target, new_value)
return new_value
def _execute_subscript(self, subscript: ast.Subscript):
index = self._execute_ast(subscript.slice)
value = self._execute_ast(subscript.value)
if not isinstance(subscript.ctx, ast.Load):
raise InterpreterError(
f"{subscript.ctx.__class__.__name__} is not supported for "
"subscript.")
if isinstance(value, (list, tuple)):
return value[int(index)]
if index in value:
return value[index]
if isinstance(index, str) and isinstance(value, Mapping):
close_matches = difflib.get_close_matches(index,
list(value.keys()))
if len(close_matches) > 0:
return value[close_matches[0]]
raise InterpreterError(f"Could not index {value} with '{index}'.")
def _execute_name(self, name: ast.Name):
if name.id in dir(builtins):
return getattr(builtins, name.id)
if isinstance(name.ctx, ast.Store):
return name.id
elif isinstance(name.ctx, ast.Load):
return self._get_value_from_state(name.id)
else:
raise InterpreterError(f"{name.ctx} is not supported.")
def _execute_condition(self, condition):
if isinstance(condition, ast.BoolOp):
if isinstance(condition.op, ast.And):
results = [self._execute_ast(value) for value in condition.values]
return all(results)
elif isinstance(condition.op, ast.Or):
results = [self._execute_ast(value) for value in condition.values]
return any(results)
else: #TODO - add any other BoolOps missing
raise InterpreterError(f"Boolean operator {condition.op} is not supported")
elif isinstance(condition, ast.Compare):
if len(condition.ops) > 1:
raise InterpreterError("Cannot evaluate conditions with multiple operators")
if len(condition.ops) > 1:
raise InterpreterError(
"Cannot evaluate conditions with multiple operators")
left = self._execute_ast(condition.left)
comparator = condition.ops[0]
right = self._execute_ast(condition.comparators[0])
if isinstance(comparator, ast.Eq):
return left == right
elif isinstance(comparator, ast.NotEq):
return left != right
elif isinstance(comparator, ast.Lt):
return left < right
elif isinstance(comparator, ast.LtE):
return left <= right
elif isinstance(comparator, ast.Gt):
return left > right
elif isinstance(comparator, ast.GtE):
return left >= right
elif isinstance(comparator, ast.Is):
return left is right
elif isinstance(comparator, ast.IsNot):
return left is not right
elif isinstance(comparator, ast.In):
return left in right
elif isinstance(comparator, ast.NotIn):
return left not in right
else:
raise InterpreterError("Unsupported condition type")
def _execute_if(self, if_statement: ast.If):
result = None
if self._execute_condition(if_statement.test):
for line in if_statement.body:
line_result = self._execute_ast(line)
if line_result is not None:
result = line_result
else:
for line in if_statement.orelse:
line_result = self._execute_ast(line)
if line_result is not None:
result = line_result
return result
def _execute_for(self, for_statement: ast.For):
result = None
for value in self._execute_ast(for_statement.iter):
self._assign(for_statement.target, value)
for line in for_statement.body:
line_result = self._execute_ast(line)
if line_result is not None:
result = line_result
return result
def _execute_import(self, import_module: ast.Import) -> None:
for module in import_module.names:
self._validate_import(module.name)
alias = module.asname or module.name
self.state[alias] = importlib.import_module(module.name)
def _execute_import_from(self, import_from: ast.ImportFrom):
if import_from.module is None:
raise InterpreterError("\"from . import\" is not supported.")
for import_name in import_from.names:
full_name = import_from.module + f".{import_name.name}"
self._validate_import(full_name)
imported_module = importlib.import_module(import_from.module)
alias = import_name.asname or import_name.name
self.state[alias] = getattr(imported_module, import_name.name)
def _validate_import(self, full_name: str):
tmp_name = ""
found_name = False
for name in full_name.split("."):
tmp_name += name if tmp_name == "" else f".{name}"
if tmp_name in self.import_white_list:
found_name = True
return
if not found_name:
raise InterpreterError(f"It is not permitted to import modules "
f"than module white list (try to import "
f"{full_name}).")
def _execute_binop(self, binop: ast.BinOp):
left = self._execute_ast(binop.left)
operator = binop.op
right = self._execute_ast(binop.right)
if isinstance(operator, ast.Add):
return left + right
elif isinstance(operator, ast.Sub):
return left - right
elif isinstance(operator, ast.Mult):
return left * right
elif isinstance(operator, ast.Div):
return left / right
elif isinstance(operator, ast.FloorDiv):
return left // right
elif isinstance(operator, ast.Mod):
return left % right
elif isinstance(operator, ast.Pow):
return left**right
elif isinstance(operator, ast.LShift):
return left << right
elif isinstance(operator, ast.RShift):
return left >> right
elif isinstance(operator, ast.MatMult):
return left @ right
else:
raise InterpreterError(f"Operator not supported: {operator}")
def _execute_unaryop(self, unaryop: ast.UnaryOp):
operand = self._execute_ast(unaryop.operand)
operator = unaryop.op
if isinstance(operator, ast.UAdd):
return +operand
elif isinstance(operator, ast.USub):
return -operand
elif isinstance(operator, ast.Not):
return not operand
else:
raise InterpreterError(f"Operator not supported: {operator}")
def _get_value_from_state(self, key: str) -> Any:
if key in self.state:
return self.state[key]
elif key in self.fuzz_state:
return self.fuzz_state[key]
else:
raise InterpreterError(f"The variable `{key}` is not defined.")
class TextPrompt(str):
r"""A class that represents a text prompt. The :obj:`TextPrompt` class
extends the built-in :obj:`str` class to provide a property for retrieving
the set of keywords in the prompt.
Attributes:
key_words (set): A set of strings representing the keywords in the
prompt.
"""
@property
def key_words(self) -> Set[str]:
r"""Returns a set of strings representing the keywords in the prompt.
"""
from camel.utils import get_prompt_template_key_words
return get_prompt_template_key_words(self)
def format(self, *args: Any, **kwargs: Any) -> 'TextPrompt':
r"""Overrides the built-in :obj:`str.format` method to allow for
default values in the format string. This is used to allow formatting
the partial string.
Args:
*args (Any): Variable length argument list.
**kwargs (Any): Arbitrary keyword arguments.
Returns:
TextPrompt: A new :obj:`TextPrompt` object with the format string
replaced with the formatted string.
"""
default_kwargs = {key: '{' + f'{key}' + '}' for key in self.key_words}
default_kwargs.update(kwargs)
return TextPrompt(super().format(*args, **default_kwargs))
class CodePrompt(TextPrompt):
r"""A class that represents a code prompt. It extends the :obj:`TextPrompt`
class with a :obj:`code_type` property.
Attributes:
code_type (str, optional): The type of code. Defaults to None.
"""
def __new__(cls, *args: Any, **kwargs: Any) -> 'CodePrompt':
r"""Creates a new instance of the :obj:`CodePrompt` class.
Args:
*args (Any): Positional arguments.
**kwargs (Any): Keyword arguments.
Returns:
CodePrompt: The created :obj:`CodePrompt` instance.
"""
code_type = kwargs.pop('code_type', None)
instance = super().__new__(cls, *args, **kwargs)
instance._code_type = code_type
return instance
@property
def code_type(self) -> Optional[str]:
r"""Returns the type of code.
Returns:
Optional[str]: The type of code.
"""
return self._code_type
def set_code_type(self, code_type: str) -> None:
r"""Sets the type of code.
Args:
code_type (str): The type of code.
"""
self._code_type = code_type
def execute(
self, interpreter: Optional[PythonInterpreter] = None,
user_variable: Optional[Dict[str, Any]] = None
) -> Tuple[Any, PythonInterpreter]:
r"""Executes the code string by a given python interpreter.
Args:
interpreter (PythonInterpreter, optional): interpreter to be used
during code execution. (default: :obj:`None`)
user_variable (Optional[Dict[str, Any]]): variables that can be
used in the code, which applying fuzzy matching, such as images
or documents. (default: :obj:`None`)
Returns:
Tuple[Any, PythonInterpreter]: A tuple containing the execution
result and the used interpreter. The execution result
represents the value of the last statement (excluding "import")
in the code. This value could potentially be the desired result
of the LLM-generated code.
"""
# NOTE: Only supports Python code for now.
if not interpreter:
interpreter = PythonInterpreter(action_space=globals())
execution_res = interpreter.execute(self, fuzz_state=user_variable,
keep_state=True)
return execution_res, interpreter