Optimize performance: lazy file handle, configurable LLM params, lancedb 0.30.2

- agent/a11y_agent.py: open qa_dataset.jsonl once in __init__ (buffering=1)
instead of open/close on every request; move json/datetime imports to
module level; add llm_temperature/max_tokens/top_p from settings;
pre-cache system prompts for both languages in __init__
- config.py: add llm_temperature, llm_max_tokens, llm_top_p fields
- database/vector_store_client.py: upgrade to lancedb 0.30.2 API —
list_tables().tables instead of table_names(); to_list() instead of
to_df().to_dict(); count_rows() instead of len(); remove unused asyncio
- requirements.txt: pin lancedb==0.30.2

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

agent/a11y_agent.py

@@ -1,5 +1,7 @@
 """A11y Expert - Main accessibility question-answering agent."""
 
+import json
+from datetime import datetime
 from typing import Optional, Generator
 from openai import OpenAI
 from langdetect import detect, LangDetectException
@@ -37,7 +39,14 @@ class A11yExpertAgent:
 
         settings = get_settings()
         self.model = settings.llm_model
-        self.system_prompt = get_system_prompt(language, expertise)
+        self.temperature = settings.llm_temperature
+        self.max_tokens = settings.llm_max_tokens
+        self.top_p = settings.llm_top_p
+        self._prompts = {
+            "pl": get_system_prompt("pl", expertise),
+            "en": get_system_prompt("en", expertise),
+        }
+        self._log_file = open("qa_dataset.jsonl", "a", encoding="utf-8", buffering=1)
 
         logger.info(f"A11yExpertAgent initialized (lang={language}, expertise={expertise}, stateless=True)")
     
@@ -48,6 +57,8 @@ class A11yExpertAgent:
                 self.vector_store.close()
             if hasattr(self.llm_client, 'close'):
                 self.llm_client.close()
+            if self._log_file and not self._log_file.closed:
+                self._log_file.close()
             logger.info("A11yExpertAgent resources closed")
         except Exception as e:
             logger.warning(f"Error closing A11yExpertAgent: {e}")
@@ -72,9 +83,8 @@ class A11yExpertAgent:
         
         logger.info(f"Detected language: {language}")
         
-        # Dynamically update system prompt based on detected language
-        current_system_prompt = get_system_prompt(language, self.expertise)
-            
+        current_system_prompt = self._prompts.get(language, self._prompts["en"])
+
         logger.info("Searching knowledge base...")
         context, sources = search_knowledge_base(question, self.vector_store, language=language)
 
@@ -89,9 +99,9 @@ class A11yExpertAgent:
             response_stream = self.llm_client.chat.completions.create(
                 model=self.model,
                 messages=messages,
-                temperature=0.3,
-                max_tokens=1500,
-                top_p=0.9,
+                temperature=self.temperature,
+                max_tokens=self.max_tokens,
+                top_p=self.top_p,
                 stream=True
             )
 
@@ -185,9 +195,6 @@ Remember to:
             language: Language of the conversation
         """
         try:
-            import json
-            from datetime import datetime
-
             qa_entry = {
                 "timestamp": datetime.now().isoformat(),
                 "question": question,
@@ -203,9 +210,7 @@ Remember to:
                 "model": self.model
             }
 
-            # Append to JSONL file (one JSON per line)
-            with open("qa_dataset.jsonl", "a", encoding="utf-8") as f:
-                f.write(json.dumps(qa_entry, ensure_ascii=False) + "\n")
+            self._log_file.write(json.dumps(qa_entry, ensure_ascii=False) + "\n")
 
             logger.debug(f"Logged Q&A pair to qa_dataset.jsonl")
 

config.py

@@ -54,6 +54,24 @@ class Settings(BaseSettings):
         default=None,
         description="Base URL for OpenAI-compatible API (optional)"
     )
+    llm_temperature: float = Field(
+        default=0.3,
+        ge=0.0,
+        le=2.0,
+        description="Sampling temperature for chat completions"
+    )
+    llm_max_tokens: int = Field(
+        default=1500,
+        ge=1,
+        le=16000,
+        description="Maximum tokens in chat completion response"
+    )
+    llm_top_p: float = Field(
+        default=0.9,
+        ge=0.0,
+        le=1.0,
+        description="Top-p nucleus sampling for chat completions"
+    )
     
     # Embeddings Configuration  
     embedding_model: str = Field(

database/vector_store_client.py

@@ -6,7 +6,6 @@ connection management and lazy table initialization.
 """
 
 import lancedb
-import asyncio
 from typing import List, Dict, Any, Optional
 from datetime import datetime
 from loguru import logger
@@ -89,7 +88,7 @@ class VectorStoreClient:
             LanceDB table or None if table doesn't exist yet
         """
         if self._table is None:
-            if self.table_name in self.db.table_names():
+            if self.table_name in self.db.list_tables().tables:
                 logger.debug(f"Opening existing table: '{self.table_name}'")
                 self._table = self.db.open_table(self.table_name)
             else:
@@ -170,7 +169,7 @@ class VectorStoreClient:
             logger.info(f"Adding {len(valid_docs)} documents to '{self.table_name}'")
             
             # Create table on first insert or open existing
-            if self.table_name not in self.db.table_names():
+            if self.table_name not in self.db.list_tables().tables:
                 self._table = self.db.create_table(self.table_name, data=valid_docs)
                 logger.info(f"✅ Created table '{self.table_name}' with {len(valid_docs)} docs")
             else:
@@ -220,9 +219,9 @@ class VectorStoreClient:
             if where:
                 query = query.where(where)
             
-            results = query.limit(top_k).to_df()
+            results = query.limit(top_k).to_list()
             logger.debug(f"Found {len(results)} documents")
-            return results.to_dict("records")
+            return results
         except Exception as e:
             logger.error(f"Search failed: {e}")
             return []
@@ -236,14 +235,14 @@ class VectorStoreClient:
         """
         if self.table is None:
             return 0
-        return len(self.table)
+        return self.table.count_rows()
     
     def get_statistics(self) -> Dict[str, Any]:
         """Get database statistics."""
         if self._db is None:
             self.connect()
         
-        if self.table_name not in self._db.table_names():
+        if self.table_name not in self._db.list_tables().tables:
             logger.warning(f"Table '{self.table_name}' does not exist yet")
             return {
                 "total_documents": 0,
@@ -350,9 +349,9 @@ class VectorStoreClient:
             if where_clause:
                 query = query.where(where_clause)
             
-            results = query.limit(top_k).to_df()
+            results = query.limit(top_k).to_list()
             logger.debug(f"Found {len(results)} documents with filters")
-            return results.to_dict("records")
+            return results
         except Exception as e:
             logger.error(f"Search with filters failed: {e}")
             return []

requirements.txt

@@ -1,7 +1,7 @@
 gradio==6.1.0
 huggingface-hub>=0.26.0
 openai
-lancedb
+lancedb==0.30.2
 loguru
 langdetect
 diskcache