Add ruff rules for tryceratops (#559)

Author: cbornet

examples/evaluation/tru_shared.py

@@ -129,7 +129,7 @@ def get_recorder(
             feedback_mode=feedback_mode,
         )
     else:
-        raise Exception(f"Unknown framework: {framework} specified for get_recorder()")
+        raise ValueError(f"Unknown framework: {framework} specified for get_recorder()")
 
 
 def get_azure_chat_model(
@@ -152,7 +152,7 @@ def get_azure_chat_model(
             temperature=temperature,
         )
     else:
-        raise Exception(f"Unknown framework: {framework} specified for getChatModel()")
+        raise ValueError(f"Unknown framework: {framework} specified for getChatModel()")
 
 
 def get_azure_embeddings_model(framework: Framework):
@@ -169,7 +169,7 @@ def get_azure_embeddings_model(framework: Framework):
             temperature=temperature,
         )
     else:
-        raise Exception(
+        raise ValueError(
             f"Unknown framework: {framework} specified for getEmbeddingsModel()"
         )
 
@@ -190,7 +190,7 @@ def get_astra_vector_store(framework: Framework, collection_name: str):
             embedding_dimension=1536,
         )
     else:
-        raise Exception(
+        raise ValueError(
             f"Unknown framework: {framework} specified for get_astra_vector_store()"
         )
 
@@ -201,7 +201,9 @@ def execute_query(framework: Framework, pipeline, query):
     elif framework == Framework.LLAMA_INDEX:
         pipeline.query(query)
     else:
-        raise Exception(f"Unknown framework: {framework} specified for execute_query()")
+        raise ValueError(
+            f"Unknown framework: {framework} specified for execute_query()"
+        )
 
 
 # runs the pipeline across all queries in all known datasets

examples/notebooks/conftest.py

@@ -34,7 +34,7 @@ def try_delete_with_backoff(collection: str, sleep=1, max_tries=2):
     except Exception as e:
         max_tries -= 1
         if max_tries < 0:
-            raise e
+            raise
 
         logging.warning(f"An exception occurred deleting collection {collection}: {e}")
         time.sleep(sleep)

libs/colbert/ragstack_colbert/cassandra_database.py

@@ -20,6 +20,10 @@
 from .objects import Chunk, Vector
 
 
+class CassandraDatabaseError(Exception):
+    pass
+
+
 class CassandraDatabase(BaseDatabase):
     """
     An implementation of the BaseDatabase abstract base class using Cassandra as the
@@ -167,7 +171,7 @@ def add_chunks(self, chunks: List[Chunk]) -> List[Tuple[str, int]]:
             success_chunks.append((doc_id, chunk_id))
 
         if len(failed_chunks) > 0:
-            raise Exception(
+            raise CassandraDatabaseError(
                 f"add failed for these chunks: {failed_chunks}. "
                 f"See error logs for more info."
             )
@@ -198,9 +202,9 @@ async def _limited_put(
                     await self._table.aput(
                         partition_id=doc_id, row_id=row_id, vector=vector
                     )
-                return doc_id, chunk_id, embedding_id, None
             except Exception as e:
                 return doc_id, chunk_id, embedding_id, e
+            return doc_id, chunk_id, embedding_id, None
 
     async def aadd_chunks(
         self, chunks: List[Chunk], concurrent_inserts: Optional[int] = 100
@@ -269,7 +273,7 @@ async def aadd_chunks(
                 failed_chunks.append((doc_id, chunk_id))
 
         if len(failed_chunks) > 0:
-            raise Exception(
+            raise CassandraDatabaseError(
                 f"add failed for these chunks: {failed_chunks}. "
                 f"See error logs for more info."
             )
@@ -293,12 +297,12 @@ def delete_chunks(self, doc_ids: List[str]) -> bool:
         for doc_id in doc_ids:
             try:
                 self._table.delete_partition(partition_id=doc_id)
-            except Exception as exp:
-                logging.error(f"issue on delete of document: {doc_id}: {exp}")
+            except Exception:
+                logging.exception(f"issue on delete of document: {doc_id}")
                 failed_docs.append(doc_id)
 
         if len(failed_docs) > 0:
-            raise Exception(
+            raise CassandraDatabaseError(
                 f"delete failed for these docs: {failed_docs}. "
                 f"See error logs for more info."
             )
@@ -313,9 +317,9 @@ async def _limited_delete(
         async with sem:
             try:
                 await self._table.adelete_partition(partition_id=doc_id)
-                return doc_id, None
             except Exception as e:
                 return doc_id, e
+            return doc_id, None
 
     async def adelete_chunks(
         self, doc_ids: List[str], concurrent_deletes: Optional[int] = 100
@@ -354,7 +358,7 @@ async def adelete_chunks(
                 failed_docs.append(doc_id)
 
         if len(failed_docs) > 0:
-            raise Exception(
+            raise CassandraDatabaseError(
                 f"delete failed for these docs: {failed_docs}. "
                 f"See error logs for more info."
             )

libs/e2e-tests/e2e_tests/llama_index/test_astra.py

@@ -136,7 +136,7 @@ def verify_document(document, expected_content, expected_metadata):
         # metadata is not returned by LlamaIndex
         # assert document.metadata == expected_metadata
     else:
-        raise Exception(
+        raise TypeError(
             "document is not of type NodeWithScore but of type " + str(type(document))
         )
 

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

@@ -233,7 +233,7 @@ def try_delete_with_backoff(collection: str, sleep=1, max_tries=5):
     except Exception as e:
         max_tries -= 1
         if max_tries < 0:
-            raise e
+            raise
 
         logging.warning(f"An exception occurred deleting collection {collection}: {e}")
         time.sleep(sleep)

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

@@ -14,9 +14,9 @@ def record_langsmith_sharelink(
         sharelink = LANGSMITH_CLIENT.share_run(run_id)
         record_property("langsmith_url", sharelink)
         logging.info(f"recorded langsmith link: {sharelink}")
-    except Exception as e:
+    except Exception:
         # runs may take a while to be discoverable
         if tries < 0:
-            raise e
+            raise
         time.sleep(5)
         record_langsmith_sharelink(run_id, record_property, tries - 1)

libs/knowledge-graph/ragstack_knowledge_graph/knowledge_graph.py

@@ -226,7 +226,7 @@ def insert(
                         ),
                     )
                 else:
-                    raise ValueError(f"Unsupported element type: {element}")
+                    raise TypeError(f"Unsupported element type: {element}")
 
             # TODO: Support concurrent execution of these statements.
             self._session.execute(batch_statement)

libs/knowledge-graph/tests/conftest.py

@@ -57,11 +57,11 @@ def db_session(cassandra_port: int) -> Session:
 def llm() -> BaseChatModel:
     try:
         from langchain_openai import ChatOpenAI
-
-        model = ChatOpenAI(model_name="gpt-4o", temperature=0.0)
-        return model
     except ValueError:
         pytest.skip("Unable to create OpenAI model")
+    else:
+        model = ChatOpenAI(model_name="gpt-4o", temperature=0.0)
+        return model
 
 
 class DataFixture:

libs/knowledge-store/ragstack_knowledge_store/math.py

@@ -26,11 +26,6 @@ def cosine_similarity(x: Matrix, y: Matrix) -> np.ndarray:
         )
     try:
         import simsimd as simd
-
-        x = np.array(x, dtype=np.float32)
-        y = np.array(y, dtype=np.float32)
-        z = 1 - np.array(simd.cdist(x, y, metric="cosine"))
-        return z
     except ImportError:
         logger.debug(
             "Unable to import simsimd, defaulting to NumPy implementation. If you want "
@@ -43,3 +38,8 @@ def cosine_similarity(x: Matrix, y: Matrix) -> np.ndarray:
             similarity = np.dot(x, y.T) / np.outer(x_norm, y_norm)
         similarity[np.isnan(similarity) | np.isinf(similarity)] = 0.0
         return similarity
+    else:
+        x = np.array(x, dtype=np.float32)
+        y = np.array(y, dtype=np.float32)
+        z = 1 - np.array(simd.cdist(x, y, metric="cosine"))
+        return z

libs/llamaindex/tests/unit_tests/test_import.py

@@ -14,7 +14,7 @@ def test_import():
 def check_no_import(fn: callable):
     try:
         fn()
-        raise Exception("Should have failed to import")
+        raise RuntimeError("Should have failed to import")
     except ImportError:
         pass