Added credentials option for iter_bucket. (#372)

* Added credentials option for iter_bucket.

This closes issue #259

* Applied PR Feedback

* Formatting changes

* explicitly construct session with session_kwargs

* fix bug, update unit tests

* remove test with tricky mocks

* isolate credentials test

Co-authored-by: Michael Penkov <m@penkov.dev>

smart_open/s3.py

@@ -655,8 +655,14 @@ def _accept_all(key):
     return True
 
 
-def iter_bucket(bucket_name, prefix='', accept_key=None,
-                key_limit=None, workers=16, retries=3):
+def iter_bucket(
+        bucket_name,
+        prefix='',
+        accept_key=None,
+        key_limit=None,
+        workers=16,
+        retries=3,
+        **session_kwargs):
     """
     Iterate and download all S3 objects under `s3://bucket_name/prefix`.
 
@@ -676,6 +682,11 @@ def iter_bucket(bucket_name, prefix='', accept_key=None,
         The number of subprocesses to use.
     retries: int, optional
         The number of time to retry a failed download.
+    session_kwargs: dict, optional
+        Keyword arguments to pass when creating a new session.
+        For a list of available names and values, see:
+        https://boto3.amazonaws.com/v1/documentation/api/latest/reference/core/session.html#boto3.session.Session
+
 
     Yields
     ------
@@ -716,8 +727,16 @@ def iter_bucket(bucket_name, prefix='', accept_key=None,
         pass
 
     total_size, key_no = 0, -1
-    key_iterator = _list_bucket(bucket_name, prefix=prefix, accept_key=accept_key)
-    download_key = functools.partial(_download_key, bucket_name=bucket_name, retries=retries)
+    key_iterator = _list_bucket(
+        bucket_name,
+        prefix=prefix,
+        accept_key=accept_key,
+        **session_kwargs)
+    download_key = functools.partial(
+        _download_key,
+        bucket_name=bucket_name,
+        retries=retries,
+        **session_kwargs)
 
     with _create_process_pool(processes=workers) as pool:
         result_iterator = pool.imap_unordered(download_key, key_iterator)
@@ -736,8 +755,13 @@ def iter_bucket(bucket_name, prefix='', accept_key=None,
     logger.info("processed %i keys, total size %i" % (key_no + 1, total_size))
 
 
-def _list_bucket(bucket_name, prefix='', accept_key=lambda k: True):
-    client = boto3.client('s3')
+def _list_bucket(
+        bucket_name,
+        prefix='',
+        accept_key=lambda k: True,
+        **session_kwargs):
+    session = boto3.session.Session(**session_kwargs)
+    client = session.client('s3')
     ctoken = None
 
     while True:
@@ -762,14 +786,14 @@ def _list_bucket(bucket_name, prefix='', accept_key=lambda k: True):
             break
 
 
-def _download_key(key_name, bucket_name=None, retries=3):
+def _download_key(key_name, bucket_name=None, retries=3, **session_kwargs):
     if bucket_name is None:
         raise ValueError('bucket_name may not be None')
 
     #
     # https://geekpete.com/blog/multithreading-boto3/
     #
-    session = boto3.session.Session()
+    session = boto3.session.Session(**session_kwargs)
     s3 = session.resource('s3')
     bucket = s3.Bucket(bucket_name)
 

smart_open/tests/test_s3.py

@@ -564,6 +564,27 @@ def test(self):
         self.assertEqual(sorted(keys), sorted(expected))
 
 
+#
+# This has to be a separate test because we cannot run it against real S3
+# (we don't want to expose our real S3 credentials).
+#
+@moto.mock_s3
+class IterBucketCredentialsTest(unittest.TestCase):
+
+    def test(self):
+        num_keys = 10
+        populate_bucket(num_keys=num_keys)
+        result = list(
+            smart_open.s3.iter_bucket(
+                BUCKET_NAME,
+                workers=None,
+                aws_access_key_id='access_id',
+                aws_secret_access_key='access_secret'
+            )
+        )
+        self.assertEqual(len(result), num_keys)
+
+
 @maybe_mock_s3
 class DownloadKeyTest(unittest.TestCase):
     def setUp(self):