feat: update LlamaEdgeClient initialization with additional parameters and refactor QdrantStore setup

Author: Acuspeedster

app/main.py

@@ -1,14 +1,15 @@
 import os
 import uuid
 import shutil
+import json  # Move from within functions to top level
 from typing import Dict, List, Optional
 from dotenv import load_dotenv
 
 # Load environment variables from .env file
 load_dotenv()
 
 from fastapi import FastAPI, BackgroundTasks, HTTPException
-from fastapi.responses import JSONResponse
+from fastapi.responses import JSONResponse, PlainTextResponse  # Move from within function to top level
 from pydantic import BaseModel
 
 from app.prompt_generator import PromptGenerator
@@ -96,7 +97,6 @@ async def get_project_status(project_id: str):
         raise HTTPException(status_code=404, detail="Project not found")
 
     # Read status file
-    import json
     with open(status_file, 'r') as f:
         status = json.load(f)
 
@@ -132,7 +132,6 @@ async def mcp_compile_and_fix_rust(request: dict):
             output_text += f"[filename: {filename}]\n{content}\n\n"
 
         # Return as plain text
-        from fastapi.responses import PlainTextResponse
         return PlainTextResponse(content=output_text.strip())
     else:
         # For errors, we can still return JSON
@@ -298,7 +297,6 @@ async def handle_project_generation(
 
 def save_status(project_dir: str, status: Dict):
     """Save project status to file"""
-    import json
     with open(f"{project_dir}/status.json", 'w') as f:
         json.dump(status, f)
 

examples/run_mcp_server.py

@@ -17,13 +17,21 @@ def main():
     # Get LLM API base URL
     llm_api_base = os.getenv("LLM_API_BASE", "https://coder.gaia.domains/v1")
     llm_model = os.getenv("LLM_MODEL", "Qwen2.5-Coder-32B-Instruct-Q5_K_M")
+    llm_embed_model = os.getenv("LLM_EMBED_MODEL", "nomic-embed")
 
     # Get Qdrant connection details
     qdrant_host = os.getenv("QDRANT_HOST", "localhost")
     qdrant_port = int(os.getenv("QDRANT_PORT", "6333"))
 
-    llm_client = LlamaEdgeClient(api_key=api_key, api_base=llm_api_base, model=llm_model)
-    vector_store = QdrantStore(embedding_size=llm_embed_size, host=qdrant_host, port=qdrant_port)
+    # Set environment variables for QdrantStore
+    os.environ["QDRANT_HOST"] = qdrant_host
+    os.environ["QDRANT_PORT"] = str(qdrant_port)
+    
+    # Initialize LlamaEdgeClient with all required parameters
+    llm_client = LlamaEdgeClient(api_key=api_key, api_base=llm_api_base, model=llm_model, embed_model=llm_embed_model)
+    
+    # Initialize QdrantStore with just the embedding_size parameter
+    vector_store = QdrantStore(embedding_size=llm_embed_size)
     vector_store.create_collection("project_examples")
     vector_store.create_collection("error_examples")