resolve merge conflicts

Author: kukushking

awswrangler/athena/_cache.py

@@ -73,9 +73,10 @@ def sorted_successful_generator(self) -> list["QueryExecutionTypeDef"]:
             Returns successful DDL and DML queries sorted by query completion time.
         """
         filtered: list["QueryExecutionTypeDef"] = []
-        for query in self._cache.values():
-            if (query["Status"].get("State") == "SUCCEEDED") and (query.get("StatementType") in ["DDL", "DML"]):
-                filtered.append(query)
+        with self._lock:
+            for query in self._cache.values():
+                if (query["Status"].get("State") == "SUCCEEDED") and (query.get("StatementType") in ["DDL", "DML"]):
+                    filtered.append(query)
         return sorted(filtered, key=lambda e: str(e["Status"]["CompletionDateTime"]), reverse=True)
 
     def __contains__(self, key: str) -> bool:

awswrangler/distributed/ray/datasources/file_datasink.py

@@ -10,7 +10,7 @@
 from ray.data._internal.delegating_block_builder import DelegatingBlockBuilder
 from ray.data._internal.execution.interfaces import TaskContext
 from ray.data.block import Block, BlockAccessor
-from ray.data.datasource.datasink import Datasink
+from ray.data.datasource.datasink import Datasink, WriteResult
 from ray.data.datasource.filename_provider import FilenameProvider
 from ray.types import ObjectRef
 
@@ -21,7 +21,7 @@
 _logger: logging.Logger = logging.getLogger(__name__)
 
 
-class _BlockFileDatasink(Datasink):
+class _BlockFileDatasink(Datasink[str]):
     def __init__(
         self,
         path: str,
@@ -91,12 +91,12 @@ def write_block(self, file: Any, block: BlockAccessor) -> None:
     # and is meant to be used for singular actions like
     # [committing a transaction](https://docs.ray.io/en/latest/data/api/doc/ray.data.Datasource.html).
     # As deceptive as it may look, there is no race condition here.
-    def on_write_complete(self, write_results: list[Any], **_: Any) -> None:
+    def on_write_complete(self, write_results: WriteResult[str]) -> None:
         """Execute callback after all write tasks complete."""
         _logger.debug("Write complete %s.", write_results)
 
         # Collect and return all write task paths
-        self._write_paths.extend(write_results)
+        self._write_paths.extend(write_results.write_returns)
 
     def get_write_paths(self) -> list[str]:
         """Return S3 paths of where the results have been written."""

pyproject.toml

@@ -23,10 +23,13 @@ dependencies = [
     "botocore>=1.23.32,<2",
     "pandas>=1.2.0,<3.0.0",
     "numpy>=1.26,<3.0",
-    "pyarrow>=8.0.0,<19.0.0",
+    "pyarrow>=18.0.0,<19.0.0 ; sys_platform != 'darwin' or platform_machine != 'x86_64'",
+    # pyarrow 18 causes macos build failures
+    # https://github.com/ray-project/ray/pull/48446
+    "pyarrow>=8.0.0,<19.0.0 ; sys_platform == 'darwin' and platform_machine == 'x86_64'",
     "typing-extensions>=4.4.0,<5",
     "packaging>=21.1,<25.0",
-    'setuptools ; python_version >= "3.12"',
+    "setuptools ; python_version >= '3.12'",
 ]
 
 [project.optional-dependencies]
@@ -56,7 +59,7 @@ progressbar = ["progressbar2>=4.0.0,<5"]
 deltalake = ["deltalake>=0.18.0,<0.26.0"]
 geopandas = ["geopandas>=1.0.0,<2"]
 modin = ["modin>=0.31,<0.33"]
-ray = ["ray[default, data]>=2.30.0,<2.38.0"]
+ray = ["ray[default, data]>=2.45.0,<3"]
 
 [project.urls]
 Homepage = "https://aws-sdk-pandas.readthedocs.io/"