Handle \n, dedupe json cleaning into single function for reusability

Use placeholder for newline in json object values until json parsed
and values extracted. This is useful when research mode models outputs
multi-line codeblocks in queries etc.

src/khoj/processor/conversation/anthropic/anthropic_chat.py

@@ -14,9 +14,9 @@ from khoj.processor.conversation.anthropic.utils import (
     format_messages_for_anthropic,
 )
 from khoj.processor.conversation.utils import (
+    clean_json,
     construct_structured_message,
     generate_chatml_messages_with_context,
-    remove_json_codeblock,
 )
 from khoj.utils.helpers import ConversationCommand, is_none_or_empty
 from khoj.utils.rawconfig import LocationData
@@ -98,7 +98,7 @@ def extract_questions_anthropic(
 
     # Extract, Clean Message from Claude's Response
     try:
-        response = remove_json_codeblock(response)
+        response = clean_json(response)
         response = json.loads(response)
         response = [q.strip() for q in response["queries"] if q.strip()]
         if not isinstance(response, list) or not response:

src/khoj/processor/conversation/google/gemini_chat.py

@@ -14,9 +14,9 @@ from khoj.processor.conversation.google.utils import (
     gemini_completion_with_backoff,
 )
 from khoj.processor.conversation.utils import (
+    clean_json,
     construct_structured_message,
     generate_chatml_messages_with_context,
-    remove_json_codeblock,
 )
 from khoj.utils.helpers import ConversationCommand, is_none_or_empty
 from khoj.utils.rawconfig import LocationData
@@ -93,7 +93,7 @@ def extract_questions_gemini(
 
     # Extract, Clean Message from Gemini's Response
     try:
-        response = remove_json_codeblock(response)
+        response = clean_json(response)
         response = json.loads(response)
         response = [q.strip() for q in response["queries"] if q.strip()]
         if not isinstance(response, list) or not response:

src/khoj/processor/conversation/openai/gpt.py

@@ -12,9 +12,9 @@ from khoj.processor.conversation.openai.utils import (
     completion_with_backoff,
 )
 from khoj.processor.conversation.utils import (
+    clean_json,
     construct_structured_message,
     generate_chatml_messages_with_context,
-    remove_json_codeblock,
 )
 from khoj.utils.helpers import ConversationCommand, is_none_or_empty
 from khoj.utils.rawconfig import LocationData
@@ -95,8 +95,7 @@ def extract_questions(
 
     # Extract, Clean Message from GPT's Response
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         response = json.loads(response)
         response = [q.strip() for q in response["queries"] if q.strip()]
         if not isinstance(response, list) or not response:

src/khoj/processor/conversation/utils.py

@@ -442,9 +442,9 @@ def reciprocal_conversation_to_chatml(message_pair):
     return [ChatMessage(content=message, role=role) for message, role in zip(message_pair, ["user", "assistant"])]
 
 
-def remove_json_codeblock(response: str):
+def clean_json(response: str):
-    """Remove any markdown json codeblock formatting if present. Useful for non schema enforceable models"""
+    """Remove any markdown json codeblock and newline formatting if present. Useful for non schema enforceable models"""
-    return response.removeprefix("```json").removesuffix("```")
+    return response.strip().replace("\n", "").removeprefix("```json").removesuffix("```")
 
 
 def defilter_query(query: str):

src/khoj/processor/tools/run_code.py

@@ -12,8 +12,8 @@ from khoj.database.models import Agent, KhojUser
 from khoj.processor.conversation import prompts
 from khoj.processor.conversation.utils import (
     ChatEvent,
+    clean_json,
     construct_chat_history,
-    remove_json_codeblock,
 )
 from khoj.routers.helpers import send_message_to_model_wrapper
 from khoj.utils.helpers import timer
@@ -111,8 +111,7 @@ async def generate_python_code(
     )
 
     # Validate that the response is a non-empty, JSON-serializable list
-    response = response.strip()
+    response = clean_json(response)
-    response = remove_json_codeblock(response)
     response = json.loads(response)
     codes = [code.strip() for code in response["codes"] if code.strip()]
 

src/khoj/routers/helpers.py

@@ -90,9 +90,9 @@ from khoj.processor.conversation.openai.gpt import converse, send_message_to_mod
 from khoj.processor.conversation.utils import (
     ChatEvent,
     ThreadedGenerator,
+    clean_json,
     construct_chat_history,
     generate_chatml_messages_with_context,
-    remove_json_codeblock,
     save_to_conversation_log,
 )
 from khoj.processor.speech.text_to_speech import is_eleven_labs_enabled
@@ -334,8 +334,7 @@ async def aget_relevant_information_sources(
         )
 
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         response = json.loads(response)
         response = [q.strip() for q in response["source"] if q.strip()]
         if not isinstance(response, list) or not response or len(response) == 0:
@@ -413,8 +412,7 @@ async def aget_relevant_output_modes(
         )
 
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         response = json.loads(response)
 
         if is_none_or_empty(response):
@@ -475,8 +473,7 @@ async def infer_webpage_urls(
 
     # Validate that the response is a non-empty, JSON-serializable list of URLs
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         urls = json.loads(response)
         valid_unique_urls = {str(url).strip() for url in urls["links"] if is_valid_url(url)}
         if is_none_or_empty(valid_unique_urls):
@@ -527,8 +524,7 @@ async def generate_online_subqueries(
 
     # Validate that the response is a non-empty, JSON-serializable list
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         response = json.loads(response)
         response = [q.strip() for q in response["queries"] if q.strip()]
         if not isinstance(response, list) or not response or len(response) == 0:
@@ -801,8 +797,7 @@ async def generate_excalidraw_diagram_from_description(
         raw_response = await send_message_to_model_wrapper(
             query=excalidraw_diagram_generation, user=user, tracer=tracer
         )
-        raw_response = raw_response.strip()
+        raw_response = clean_json(raw_response)
-        raw_response = remove_json_codeblock(raw_response)
         response: Dict[str, str] = json.loads(raw_response)
         if not response or not isinstance(response, List) or not isinstance(response[0], Dict):
             # TODO Some additional validation here that it's a valid Excalidraw diagram

src/khoj/routers/research.py

@@ -11,9 +11,9 @@ from khoj.database.models import Agent, KhojUser
 from khoj.processor.conversation import prompts
 from khoj.processor.conversation.utils import (
     InformationCollectionIteration,
+    clean_json,
     construct_iteration_history,
     construct_tool_chat_history,
-    remove_json_codeblock,
 )
 from khoj.processor.tools.online_search import read_webpages, search_online
 from khoj.processor.tools.run_code import run_code
@@ -99,8 +99,7 @@ async def apick_next_tool(
         )
 
     try:
-        response = response.strip()
+        response = clean_json(response)
-        response = remove_json_codeblock(response)
         response = json.loads(response)
         selected_tool = response.get("tool", None)
         generated_query = response.get("query", None)