Uh oh!
There was an error while loading. Please reload this page.
Support blob streaming for file-like objects - #1914
Conversation
googlebot
commented
Jun 27, 2016
Thanks for your pull request. It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). 📝 Please visit https://cla.developers.google.com/ to sign. Once you've signed, please reply here (e.g.
|
linar-jether
commented
Jun 27, 2016
Signed it |
googlebot
commented
Jun 27, 2016
CLAs look good, thanks! |
| self._buffered_data = self._stream.read(size) | ||
| self._buffered_data = self._stream.read(size) \ | ||
| if not hasattr(self._stream, | ||
| 'closed') or not self._stream.closed else '' |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
linar-jether
commented
Jul 12, 2016
fixed according to comments |
tseaver
commented
Jul 12, 2016
@linar-jether thanks for the changes. The pylint failures on Travis aren't due to this PR: they released a new version last Friday, which introduced new checks that we now fail. The |
linar-jether
commented
Jul 13, 2016
All tests passed |
| import copy | ||
| import hashlib | ||
| from io import BytesIO | ||
| from io import BytesIO, UnsupportedOperation |
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Uh oh!
There was an error while loading. Please reload this page.
daspecster
commented
Jul 13, 2016
@linar-jether, cool! Do you think you could squash these commits down? Otherwise LGTM, but I defer to @tseaver/@dhermes for confirmation. |
742a95e to
43464b7Comparelinar-jether
commented
Jul 14, 2016
Sure, squashed all previous commits and separated the imports |
tseaver
commented
Jul 14, 2016
@linar-jether Thanks again for your work! |
* fix: add pyarrow version check for range support * add comment why we are making a separate constant --------- Co-authored-by: Tim Sweña (Swast) <swast@google.com>
* fix: resolve location reset issue in bigquery options * fix * fix * fix the test
When uploading file-like objects without a specified size, upload in chunks instead of exhausting the stream and keeping the data in-memory.
See #1909