feat(ruff): enable Refurb and Ruff rules (#22385)

.ruff.toml

@@ -175,10 +175,12 @@ select = [
   "E",    # pycodestyle
   "F",    # Pyflakes
   "FAST", # FastAPI
+  "FURB", # Refurb
   "I",    # isort
   "PERF", # Perflint
   "PGH",  # pygrep-hooks
   "PL",   # Pylint
+  "RUF",  # Ruff
   "TRY",  # tryceratops
   "UP",   # pyupgrade
 ]

authorization/hm-opal-client/opal_fetcher_postgres/provider.py

@@ -1,6 +1,7 @@
 # https://github.com/permitio/opal-fetcher-postgres
 
 import json
+from typing import Any, ClassVar
 
 import asyncpg
 from asyncpg.exceptions import DataError
@@ -38,12 +39,11 @@ class PostgresFetchEvent(FetchEvent):
 
 
 class PostgresFetchProvider(BaseFetchProvider):
-    RETRY_CONFIG = {
+    RETRY_CONFIG: ClassVar[dict[str, Any]] = {
         "wait": wait.wait_random_exponential(),
         "stop": stop.stop_after_attempt(10),
-        "retry": retry_unless_exception_type(
-            DataError
-        ),  # query error (i.e: invalid table, etc)
+        # query error (i.e: invalid table, etc)
+        "retry": retry_unless_exception_type(DataError),
         "reraise": True,
     }
 

data-storage/lance/src/main.py

@@ -69,7 +69,7 @@ def main() -> None:
     # nprobes:
     #   The number of probes determines the distribution of vector space.
     #   While a higher number enhances search accuracy, it also results in slower performance.
-    #   Typically, setting nprobes to cover 5–10% of the dataset proves effective in achieving high recall with minimal latency.
+    #   Typically, setting nprobes to cover 5 - 10% of the dataset proves effective in achieving high recall with minimal latency.
     #
     # refine_factor:
     #   Refine the results by reading extra elements and re-ranking them in memory.

machine-learning/feature-store/driver_features/example.py

@@ -9,7 +9,7 @@
 # production, you can use your favorite DWH, such as BigQuery. See Feast documentation
 # for more info.
 driver_hourly_stats = FileSource(
-    path="/Users/hongbo-miao/Clouds/Git/hongbomiao.com/machine-learning/feature-store/driver_features/data/driver_stats.parquet",  # noqa: E501
+    path="/Users/hongbo-miao/Clouds/Git/hongbomiao.com/machine-learning/feature-store/driver_features/data/driver_stats.parquet",
     timestamp_field="event_timestamp",
     created_timestamp_column="created",
 )

machine-learning/hm-langgraph/applications/chat-pdf/src/main.py

@@ -100,8 +100,8 @@ def retrieve_context(
         relevant_chunks = [chunks[idx] for idx in indices[0]]
         state["context"] = "\n".join(relevant_chunks)
         logger.info("Context retrieval completed")
-    except Exception as e:
-        logger.error(f"Error in retrieve_context: {str(e)}", exc_info=True)
+    except Exception:
+        logger.exception("Error in retrieve_context.")
         raise
     else:
         return state
@@ -128,8 +128,8 @@ def generate_answer(state: MessagesState) -> MessagesState:
             ],
         )
         state["answer"] = response.choices[0].message.content
-    except Exception as e:
-        logger.error(f"Error in generate_answer: {str(e)}", exc_info=True)
+    except Exception:
+        logger.exception("Error in generate_answer.")
         raise
     else:
         return state