mirror of
https://github.com/hwchase17/langchain.git
synced 2025-06-22 06:39:52 +00:00
community[minor]: Add ability to specify file encoding and json encoding for FileChatMessageHistory (#24258)
Description: Add UTF-8 encoding support Issue: Inability to properly handle characters from certain languages (e.g., Korean) Fix: Implement UTF-8 encoding in FileChatMessageHistory --------- Co-authored-by: Eugene Yurtsev <eugene@langchain.dev> Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
This commit is contained in:
parent
020cc1cf3e
commit
07715f815b
@ -1,6 +1,6 @@
|
|||||||
import json
|
import json
|
||||||
from pathlib import Path
|
from pathlib import Path
|
||||||
from typing import List
|
from typing import List, Optional
|
||||||
|
|
||||||
from langchain_core.chat_history import (
|
from langchain_core.chat_history import (
|
||||||
BaseChatMessageHistory,
|
BaseChatMessageHistory,
|
||||||
@ -11,21 +11,33 @@ from langchain_core.messages import BaseMessage, messages_from_dict, messages_to
|
|||||||
class FileChatMessageHistory(BaseChatMessageHistory):
|
class FileChatMessageHistory(BaseChatMessageHistory):
|
||||||
"""Chat message history that stores history in a local file."""
|
"""Chat message history that stores history in a local file."""
|
||||||
|
|
||||||
def __init__(self, file_path: str) -> None:
|
def __init__(
|
||||||
|
self,
|
||||||
|
file_path: str,
|
||||||
|
*,
|
||||||
|
encoding: Optional[str] = None,
|
||||||
|
ensure_ascii: bool = True,
|
||||||
|
) -> None:
|
||||||
"""Initialize the file path for the chat history.
|
"""Initialize the file path for the chat history.
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
file_path: The path to the local file to store the chat history.
|
file_path: The path to the local file to store the chat history.
|
||||||
|
encoding: The encoding to use for file operations. Defaults to None.
|
||||||
|
ensure_ascii: If True, escape non-ASCII in JSON. Defaults to True.
|
||||||
"""
|
"""
|
||||||
self.file_path = Path(file_path)
|
self.file_path = Path(file_path)
|
||||||
|
self.encoding = encoding
|
||||||
|
self.ensure_ascii = ensure_ascii
|
||||||
|
|
||||||
if not self.file_path.exists():
|
if not self.file_path.exists():
|
||||||
self.file_path.touch()
|
self.file_path.touch()
|
||||||
self.file_path.write_text(json.dumps([]))
|
self.file_path.write_text(
|
||||||
|
json.dumps([], ensure_ascii=self.ensure_ascii), encoding=self.encoding
|
||||||
|
)
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def messages(self) -> List[BaseMessage]: # type: ignore
|
def messages(self) -> List[BaseMessage]: # type: ignore
|
||||||
"""Retrieve the messages from the local file"""
|
"""Retrieve the messages from the local file"""
|
||||||
items = json.loads(self.file_path.read_text())
|
items = json.loads(self.file_path.read_text(encoding=self.encoding))
|
||||||
messages = messages_from_dict(items)
|
messages = messages_from_dict(items)
|
||||||
return messages
|
return messages
|
||||||
|
|
||||||
@ -33,8 +45,12 @@ class FileChatMessageHistory(BaseChatMessageHistory):
|
|||||||
"""Append the message to the record in the local file"""
|
"""Append the message to the record in the local file"""
|
||||||
messages = messages_to_dict(self.messages)
|
messages = messages_to_dict(self.messages)
|
||||||
messages.append(messages_to_dict([message])[0])
|
messages.append(messages_to_dict([message])[0])
|
||||||
self.file_path.write_text(json.dumps(messages))
|
self.file_path.write_text(
|
||||||
|
json.dumps(messages, ensure_ascii=self.ensure_ascii), encoding=self.encoding
|
||||||
|
)
|
||||||
|
|
||||||
def clear(self) -> None:
|
def clear(self) -> None:
|
||||||
"""Clear session memory from the local file"""
|
"""Clear session memory from the local file"""
|
||||||
self.file_path.write_text(json.dumps([]))
|
self.file_path.write_text(
|
||||||
|
json.dumps([], ensure_ascii=self.ensure_ascii), encoding=self.encoding
|
||||||
|
)
|
||||||
|
Loading…
Reference in New Issue
Block a user