klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-09 21:29:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

Log non schema adhering chat message before save to DB

Browse Source
This commit is contained in:
Debanjum
2025-06-11 11:51:41 -07:00
parent 71763684a9
commit 30ced1d86c
1 changed files with 15 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
src/khoj/processor/conversation/utils.py
View File

@@ -20,7 +20,7 @@ import yaml
from langchain_core.messages.chat import ChatMessage from langchain_core.messages.chat import ChatMessage
from llama_cpp import LlamaTokenizer from llama_cpp import LlamaTokenizer
from llama_cpp.llama import Llama from llama_cpp.llama import Llama
from pydantic import BaseModel from pydantic import BaseModel, ConfigDict, ValidationError, create_model
from transformers import AutoTokenizer, PreTrainedTokenizer, PreTrainedTokenizerFast from transformers import AutoTokenizer, PreTrainedTokenizer, PreTrainedTokenizerFast
from khoj.database.adapters import ConversationAdapters from khoj.database.adapters import ConversationAdapters
@@ -362,7 +362,6 @@ def message_to_log(
"""Create json logs from messages, metadata for conversation log""" """Create json logs from messages, metadata for conversation log"""
default_khoj_message_metadata = { default_khoj_message_metadata = {
"intent": {"type": "remember", "memory-type": "notes", "query": user_message}, "intent": {"type": "remember", "memory-type": "notes", "query": user_message},
"trigger-emotion": "calm",
} }
khoj_response_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S") khoj_response_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
@@ -377,6 +376,20 @@ def message_to_log(
khoj_log = merge_dicts(khoj_message_metadata, default_khoj_message_metadata) khoj_log = merge_dicts(khoj_message_metadata, default_khoj_message_metadata)
khoj_log = merge_dicts({"message": chat_response, "by": "khoj", "created": khoj_response_time}, khoj_log) khoj_log = merge_dicts({"message": chat_response, "by": "khoj", "created": khoj_response_time}, khoj_log)
# Validate message logs
# Only validates top-level fields, not nested fields, defined in ChatMessageModel
class StrictChatMessageModel(ChatMessageModel):
model_config = ConfigDict(extra="forbid", strict=True)
try:
StrictChatMessageModel(**human_log)
except ValidationError as e:
logger.error(f"Validation error in user chat message: {e}\nUser Message: {human_log}\n")
try:
StrictChatMessageModel(**khoj_log)
except ValidationError as e:
logger.error(f"Validation error in khoj chat message: {e}\nKhoj Message: {khoj_log}\n")
human_message = ChatMessageModel(**human_log) human_message = ChatMessageModel(**human_log)
khoj_message = ChatMessageModel(**khoj_log) khoj_message = ChatMessageModel(**khoj_log)
chat_history.extend([human_message, khoj_message]) chat_history.extend([human_message, khoj_message])