Add ruff rules for perflint (#548)

Author: cbornet

libs/colbert/ragstack_colbert/cassandra_database.py

@@ -338,15 +338,13 @@ async def adelete_chunks(
         """
 
         semaphore = asyncio.Semaphore(concurrent_deletes)
-        all_tasks = []
-
-        for doc_id in doc_ids:
-            all_tasks.append(
-                self._limited_delete(
-                    sem=semaphore,
-                    doc_id=doc_id,
-                )
+        all_tasks = [
+            self._limited_delete(
+                sem=semaphore,
+                doc_id=doc_id,
             )
+            for doc_id in doc_ids
+        ]
 
         results = await asyncio.gather(*all_tasks, return_exceptions=True)
 

libs/e2e-tests/e2e_tests/test_utils/astradb_vector_store_handler.py

@@ -116,14 +116,13 @@ def put_document(
         )
 
     def search_documents(self, vector: List[float], limit: int) -> List[str]:
-        results = self.collection.vector_find(
-            vector,
-            limit=limit,
-        )
-        docs = []
-        for result in results:
-            docs.append(result["document"])
-        return docs
+        return [
+            result["document"]
+            for result in self.collection.vector_find(
+                vector,
+                limit=limit,
+            )
+        ]
 
 
 class EnhancedAstraDBLlamaIndexVectorStore(
@@ -142,14 +141,13 @@ def put_document(
         )
 
     def search_documents(self, vector: List[float], limit: int) -> List[str]:
-        results = self.client.vector_find(
-            vector,
-            limit=limit,
-        )
-        docs = []
-        for result in results:
-            docs.append(result["document"])
-        return docs
+        return [
+            result["document"]
+            for result in self.client.vector_find(
+                vector,
+                limit=limit,
+            )
+        ]
 
 
 class AstraDBVectorStoreTestContext(VectorStoreTestContext):

libs/e2e-tests/e2e_tests/test_utils/cassandra_vector_store_handler.py

@@ -95,17 +95,15 @@ def put_document(
 
     def search_documents(self, vector: List[float], limit: int) -> List[str]:
         if isinstance(self.table, MetadataVectorCassandraTable):
-            results = self.table.ann_search(vector=vector, n=limit)
-            docs = []
-            for result in results:
-                docs.append(result["body_blob"])
-            return docs
+            return [
+                result["body_blob"]
+                for result in self.table.ann_search(vector=vector, n=limit)
+            ]
         else:
-            results = self.table.search(embedding_vector=vector, top_k=limit)
-            docs = []
-            for result in results:
-                docs.append(result["document"])
-            return docs
+            return [
+                result["document"]
+                for result in self.table.search(embedding_vector=vector, top_k=limit)
+            ]
 
 
 class EnhancedCassandraLlamaIndexVectorStore(

libs/knowledge-store/ragstack_knowledge_store/_utils.py

@@ -35,6 +35,7 @@ def strict_zip(*iterables):
             return
 
         iterators = tuple(iter(iterable) for iterable in iterables)
+        items = []
         try:
             while True:
                 items = []

libs/langchain/ragstack_langchain/graph_store/cassandra.py

@@ -77,13 +77,12 @@ def add_nodes(
         nodes: Iterable[Node],
         **kwargs: Any,
     ) -> Iterable[str]:
-        _nodes = []
-        for node in nodes:
-            _nodes.append(
-                graph_store.Node(
-                    id=node.id, text=node.text, metadata=node.metadata, links=node.links
-                )
+        _nodes = [
+            graph_store.Node(
+                id=node.id, text=node.text, metadata=node.metadata, links=node.links
             )
+            for node in nodes
+        ]
         return self.store.add_nodes(_nodes)
 
     @classmethod

pyproject.toml

@@ -58,12 +58,19 @@ warn_unused_ignores = true
 extend-include = ["*.ipynb"]
 
 [tool.ruff.lint]
+ignore = [
+    "PERF203", # Incorrect detection
+]
+
 select = [
     "E",
     "F",
+    "FLY",
     "I",
     "ISC",
     "N",
+    "NPY",
+    "PERF",
     "UP",
     "W",
 ]

scripts/generate-testspace-report.py

@@ -152,14 +152,12 @@ def generate_link_annotation(link):
 def parse_snyk_report(input_file: str):
     test_cases = []
     vulnerabilities = {}
-    files = []
 
     if os.path.isdir(input_file):
-        for f in os.listdir(input_file):
-            if f.endswith(".json"):
-                files.append(f)
+        files = [f for f in os.listdir(input_file) if f.endswith(".json")]
     else:
-        files.append(input_file)
+        files = [input_file]
+
     all_links = []
 
     for snykfile in files:
@@ -265,18 +263,19 @@ def parse_test_report(input_file: str):
                     )
 
                 properties = test_case.find("properties")
-                links = []
                 if properties:
-                    for prop in properties.iter("property"):
-                        if prop.get("name") == "langsmith_url":
-                            links.append(
-                                Link(
-                                    name="LangSmith trace",
-                                    url=prop.get("value"),
-                                    level="info",
-                                    description="",
-                                )
-                            )
+                    links = [
+                        Link(
+                            name="LangSmith trace",
+                            url=prop.get("value"),
+                            level="info",
+                            description="",
+                        )
+                        for prop in properties.iter("property")
+                        if prop.get("name") == "langsmith_url"
+                    ]
+                else:
+                    links = []
 
                 report_test_case = TestCase(
                     name=rewrite_name(test_case.get("name")),