EDPDEV-1535 add commit download support (#474)

* EDPDEV-1535 add commit download support

* prevent automatic decompression when downloading .gz files

* fix non-gz file download

Author: David Caplan

solvebio/resource/apiresource.py

@@ -8,6 +8,7 @@
 import os
 import requests
 import tempfile
+import shutil
 
 try:
     from urllib import quote_plus
@@ -213,15 +214,22 @@ def download(self, path=None, **kwargs):
             path = os.path.join(tempfile.gettempdir(), filename)
 
         try:
-            response = requests.request(method='get', url=download_url)
+            # use streaming to prevent automatic decompression
+            response = requests.request(method='get', url=download_url, stream=True)
         except Exception as e:
             _handle_request_error(e)
 
         if not (200 <= response.status_code < 400):
             _handle_api_error(response)
 
         with open(path, 'wb') as fileobj:
-            fileobj.write(response._content)
+            if filename.endswith('.gz'):
+                # Don't automatically decompress gzipped files
+                shutil.copyfileobj(response.raw, fileobj)
+            else:
+                for chunk in response.iter_content(chunk_size=1024 * 8):
+                    if chunk:
+                        fileobj.write(chunk)
 
         return path
 

solvebio/resource/datasetcommit.py

@@ -3,6 +3,7 @@
 from .apiresource import ListableAPIResource
 from .apiresource import CreateableAPIResource
 from .apiresource import UpdateableAPIResource
+from .apiresource import DownloadableAPIResource
 from .solveobject import convert_to_solve_object
 from .task import Task
 
@@ -34,7 +35,7 @@ def follow_commits(task, sleep_seconds):
 
 
 class DatasetCommit(CreateableAPIResource, ListableAPIResource,
-                    UpdateableAPIResource):
+                    UpdateableAPIResource, DownloadableAPIResource):
     """
     DatasetCommits represent a change made to a Dataset.
     """