Clarify download_fileobj flush guidance in docs

boto3/s3/inject.py

@@ -805,6 +805,9 @@ def download_fileobj(
 
     This is a managed transfer which will perform a multipart download in
     multiple threads if necessary.
+    If you plan to read from the file-like object after the download completes
+    and before closing it, call ``Fileobj.flush()`` and ``Fileobj.seek(0)``
+    first.
 
     Usage::
 
@@ -871,6 +874,9 @@ def bucket_download_fileobj(
 
     This is a managed transfer which will perform a multipart download in
     multiple threads if necessary.
+    If you plan to read from the file-like object after the download completes
+    and before closing it, call ``Fileobj.flush()`` and ``Fileobj.seek(0)``
+    first.
 
     Usage::
 
@@ -920,6 +926,9 @@ def object_download_fileobj(
 
     This is a managed transfer which will perform a multipart download in
     multiple threads if necessary.
+    If you plan to read from the file-like object after the download completes
+    and before closing it, call ``Fileobj.flush()`` and ``Fileobj.seek(0)``
+    first.
 
     Usage::
 

docs/source/guide/s3-example-download-file.rst

@@ -36,6 +36,17 @@ object must be opened in binary mode, not text mode.
     with open('FILE_NAME', 'wb') as f:
         s3.download_fileobj('amzn-s3-demo-bucket', 'OBJECT_NAME', f)
 
+If you need to read from the same file object before closing it, call
+``flush()`` and then ``seek(0)`` first:
+
+.. code-block:: python
+
+    with open('FILE_NAME', 'wb+') as f:
+        s3.download_fileobj('amzn-s3-demo-bucket', 'OBJECT_NAME', f)
+        f.flush()
+        f.seek(0)
+        data = f.read()
+
 
 Like their upload cousins, the download methods are provided by the 
 S3 ``Client``, ``Bucket``, and ``Object`` classes, and each class provides 
@@ -52,4 +63,3 @@ The download method's ``Callback`` parameter is used for the same purpose
 as the upload method's. The upload and download methods can both invoke the 
 same ``Callback`` class.
 
-