fix(linter): format and lint errors

Author: dineshpinto

src/flare_ai_rag/retriever/qdrant_retriever.py

@@ -22,13 +22,15 @@ def __init__(
     @override
     def semantic_search(self, query: str, top_k: int = 5) -> list[dict]:
         """
-        Perform semantic search by converting the query into a vector and searching in Qdrant.
+        Perform semantic search by converting the query into a vector
+        and searching in Qdrant.
 
         :param query: The input query.
         :param top_k: Number of top results to return.
         :return: A list of dictionaries, each representing a retrieved document.
         """
-        # Convert the query into a vector embedding using the SentenceTransformer instance.
+        # Convert the query into a vector embedding using the
+        # SentenceTransformer instance.
         query_vector = self.embedding_model.encode(query)
 
         # Search Qdrant for similar vectors.
@@ -41,11 +43,11 @@ def semantic_search(self, query: str, top_k: int = 5) -> list[dict]:
         # Process and return results.
         output = []
         for hit in results:
-            output.append(
-                {
-                    "text": hit.payload.get("text", ""),
-                    "score": hit.score,
-                    "metadata": {k: v for k, v in hit.payload.items() if k != "text"},
-                }
-            )
+            if hit.payload:
+                text = hit.payload.get("text", "")
+                metadata = {k: v for k, v in hit.payload.items() if k != "text"}
+            else:
+                text = ""
+                metadata = ""
+            output.append({"text": text, "score": hit.score, "metadata": metadata})
         return output

src/flare_ai_rag/router/config.py

@@ -1,17 +1,7 @@
 from dataclasses import dataclass
 
 from flare_ai_rag.openrouter.model import Model
-
-BASE_PROMPT = """You are a query router. Analyze the query below and classify it by returning exactly one of these values—without any additional text:\n\n
-
-    - **{answer_option}**: Use this if the query is clear, specific, and can be answered with factual information.\n
-    - **{clarify_option}**: Use this if the query is ambiguous, vague, or needs additional context.\n
-    - **{reject_option}**: Use this if the query is inappropriate, harmful, or completely out of scope.\n\n
-
-    Query: {query}\n\n
-
-    Return exactly one word: ANSWER, CLARIFY, or REJECT.
-"""
+from flare_ai_rag.router.prompts import BASE_PROMPT
 
 
 @dataclass(frozen=True)

src/flare_ai_rag/router/prompts.py

@@ -0,0 +1,13 @@
+from typing import Final
+
+BASE_PROMPT: Final = """
+You are a query router. Analyze the query below and classify it by returning exactly one of these values—without any additional text:\n\n
+
+    - **{answer_option}**: Use this if the query is clear, specific, and can be answered with factual information.\n
+    - **{clarify_option}**: Use this if the query is ambiguous, vague, or needs additional context.\n
+    - **{reject_option}**: Use this if the query is inappropriate, harmful, or completely out of scope.\n\n
+
+    Query: {query}\n\n
+
+    Return exactly one word: ANSWER, CLARIFY, or REJECT.
+"""

src/flare_ai_rag/router/router.py

@@ -1,3 +1,5 @@
+from typing import override
+
 from flare_ai_rag.openrouter.client import OpenRouterClient
 from flare_ai_rag.router.base_router import BaseQueryRouter
 from flare_ai_rag.router.config import RouterConfig
@@ -9,7 +11,7 @@ class QueryRouter(BaseQueryRouter):
     classify a query as ANSWER, CLARIFY, or REJECT.
     """
 
-    def __init__(self, api_key: str, config: RouterConfig):
+    def __init__(self, api_key: str, config: RouterConfig) -> None:
         """
         Initialize the router with an API key and model name.
         :param api_key: Your OpenRouter API key.
@@ -18,6 +20,7 @@ def __init__(self, api_key: str, config: RouterConfig):
         self.config = config
         self.client = OpenRouterClient(api_key=api_key)
 
+    @override
     def route_query(self, query: str) -> str:
         """
         Analyze the query using the configured prompt and classify it.

tests/test_generate_collection.py

@@ -1,4 +1,3 @@
-
 import pandas as pd
 import structlog
 from qdrant_client import QdrantClient
@@ -18,7 +17,7 @@ def generate_collection(
     client: QdrantClient,
     qdrant_config: QdrantConfig,
     collection_name: str,
-):
+) -> None:
     """Routine for generating a Qdrant collection for a specific CSV file type."""
     # Create the collection.
     create_collection(client, collection_name, qdrant_config.vector_size)
@@ -45,7 +44,7 @@ def generate_collection(
             # Compute the embedding for the document content.
             embedding = embedding_model.encode(content).tolist()
         except Exception as e:
-            logger.error(
+            logger.exception(
                 "Error encoding document.", filename=row["Filename"], error=str(e)
             )
             continue
@@ -73,7 +72,7 @@ def generate_collection(
         logger.warning("No valid documents found to insert.")
 
 
-def main():
+def main() -> None:
     # Load Qdrant config
     config_json = loader.load_json(config.input_path / "input_parameters.json")
     qdrant_config = QdrantConfig.load(config_json["qdrant_config"])

tests/test_qdrant_retriever.py

@@ -9,7 +9,7 @@
 logger = structlog.get_logger(__name__)
 
 
-def main():
+def main() -> None:
     # Load Qdrant config
     config_json = loader.load_json(config.input_path / "input_parameters.json")
     qdrant_config = QdrantConfig.load(config_json["qdrant_config"])