Expose AWS Region to HDF5IO

CHANGELOG.md

@@ -1,5 +1,10 @@
 # HDMF Changelog
 
+## HDMF 3.12.3 (Upcoming)
+
+### Bug fixes
+- Exposed `aws_region` to `HDF5IO` and downstream passes to `h5py.File`. @codycbakerphd [#1040](https://github.com/hdmf-dev/hdmf/pull/1040)
+
 ## HDMF 3.12.2 (February 9, 2024)
 
 ### Bug fixes

src/hdmf/backends/hdf5/h5tools.py

@@ -63,15 +63,21 @@ def can_read(path):
             {'name': 'file', 'type': [File, "S3File", "RemFile"],
              'doc': 'a pre-existing h5py.File, S3File, or RemFile object', 'default': None},
             {'name': 'driver', 'type': str, 'doc': 'driver for h5py to use when opening HDF5 file', 'default': None},
+            {
+                'name': 'aws_region',
+                'type': str,
+                'doc': 'If driver is ros3, then specify the aws region of the url.',
+                'default': None
+            },
             {'name': 'herd_path', 'type': str,
              'doc': 'The path to read/write the HERD file', 'default': None},)
     def __init__(self, **kwargs):
         """Open an HDF5 file for IO.
         """
         self.logger = logging.getLogger('%s.%s' % (self.__class__.__module__, self.__class__.__qualname__))
-        path, manager, mode, comm, file_obj, driver, herd_path = popargs('path', 'manager', 'mode',
+        path, manager, mode, comm, file_obj, driver, aws_region, herd_path = popargs('path', 'manager', 'mode',
                                                                                        'comm', 'file', 'driver',
-                                                                                       'herd_path',
+                                                                                       'aws_region', 'herd_path',
                                                                                        kwargs)
 
         self.__open_links = []  # keep track of other files opened from links in this file
@@ -92,6 +98,7 @@ def __init__(self, **kwargs):
         elif isinstance(manager, TypeMap):
             manager = BuildManager(manager)
         self.__driver = driver
+        self.__aws_region = aws_region
         self.__comm = comm
         self.__mode = mode
         self.__file = file_obj
@@ -117,6 +124,10 @@ def _file(self):
     def driver(self):
         return self.__driver
 
+    @property
+    def aws_region(self):
+        return self.__aws_region
+
     @classmethod
     def __check_path_file_obj(cls, path, file_obj):
         if isinstance(path, Path):
@@ -134,13 +145,13 @@ def __check_path_file_obj(cls, path, file_obj):
         return path
 
     @classmethod
-    def __resolve_file_obj(cls, path, file_obj, driver):
+    def __resolve_file_obj(cls, path, file_obj, driver, aws_region=None):
         path = cls.__check_path_file_obj(path, file_obj)
 
         if file_obj is None:
             file_kwargs = dict()
             if driver is not None:
-                file_kwargs.update(driver=driver)
+                file_kwargs.update(driver=driver, aws_region=bytes(aws_region, "ascii"))
             file_obj = File(path, 'r', **file_kwargs)
         return file_obj
 
@@ -757,6 +768,9 @@ def open(self):
             if self.driver is not None:
                 kwargs.update(driver=self.driver)
 
+                if self.driver == "ros3" and self.aws_region is not None:
+                    kwargs.update(aws_region=bytes(self.aws_region, "ascii"))
+
             self.__file = File(self.source, open_flag, **kwargs)
 
     def close(self, close_links=True):

tests/unit/test_io_hdf5_streaming.py

@@ -89,6 +89,12 @@ def test_basic_read(self):
         with get_hdf5io(s3_path, "r", manager=self.manager, driver="ros3") as io:
             io.read()
 
+    def test_basic_read_with_aws_region(self):
+        s3_path = "https://dandiarchive.s3.amazonaws.com/blobs/11e/c89/11ec8933-1456-4942-922b-94e5878bb991"
+
+        with get_hdf5io(s3_path, "r", manager=self.manager, driver="ros3", aws_region="us-east-2") as io:
+            io.read()
+
 # Util functions and classes to enable loading of the NWB namespace -- see pynwb/src/pynwb/spec.py