feat: Add ruff in pre-commit hooks

Author: KShivendu

.pre-commit-config.yaml

@@ -28,3 +28,9 @@ repos:
       - id: isort
         name: "Sort Imports"
         args: ["--profile", "black"]
+
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.3.5
+    hooks:
+      # Run the linter.
+      - id: ruff

benchmark/dataset.py

@@ -1,4 +1,3 @@
-import gzip
 import os
 import shutil
 import tarfile

engine/base_client/__init__.py

@@ -6,3 +6,12 @@
 
 class IncompatibilityError(Exception):
     pass
+
+
+__all__ = [
+    "BaseClient",
+    "BaseConfigurator",
+    "BaseSearcher",
+    "BaseUploader",
+    "IncompatibilityError",
+]

engine/base_client/client.py

@@ -1,7 +1,6 @@
 import json
 import os
 from datetime import datetime
-from pathlib import Path
 from typing import List
 
 from benchmark import ROOT_DIR

engine/base_client/upload.py

@@ -1,6 +1,6 @@
 import time
 from multiprocessing import get_context
-from typing import Iterable, List, Optional, Tuple
+from typing import Iterable, List, Optional
 
 import tqdm
 

engine/base_client/utils.py

@@ -1,4 +1,4 @@
-from typing import Any, Iterable
+from typing import Iterable
 
 from dataset_reader.base_reader import Record
 

engine/clients/elasticsearch/__init__.py

@@ -1,3 +1,9 @@
 from engine.clients.elasticsearch.configure import ElasticConfigurator
 from engine.clients.elasticsearch.search import ElasticSearcher
 from engine.clients.elasticsearch.upload import ElasticUploader
+
+__all__ = [
+    "ElasticConfigurator",
+    "ElasticSearcher",
+    "ElasticUploader",
+]

engine/clients/elasticsearch/search.py

@@ -55,7 +55,7 @@ def search_one(cls, query: Query, top: int) -> List[Tuple[int, float]]:
             **{"num_candidates": 100, **cls.search_params},
         }
 
-        meta_conditions = cls.parser.parse(meta_conditions)
+        meta_conditions = cls.parser.parse(query.meta_conditions)
         if meta_conditions:
             knn["filter"] = meta_conditions
 

engine/clients/elasticsearch/upload.py

@@ -1,6 +1,6 @@
 import multiprocessing as mp
 import uuid
-from typing import List, Optional
+from typing import List
 
 from elasticsearch import Elasticsearch
 
@@ -46,8 +46,6 @@ def init_client(cls, host, distance, connection_params, upload_params):
 
     @classmethod
     def upload_batch(cls, batch: List[Record]):
-        if metadata is None:
-            metadata = [{}] * len(batch)
         operations = []
         for record in batch:
             vector_id = uuid.UUID(int=record.idx).hex

engine/clients/milvus/__init__.py

@@ -1,3 +1,9 @@
 from engine.clients.milvus.configure import MilvusConfigurator
 from engine.clients.milvus.search import MilvusSearcher
 from engine.clients.milvus.upload import MilvusUploader
+
+__all__ = [
+    "MilvusConfigurator",
+    "MilvusSearcher",
+    "MilvusUploader",
+]