Default to octet-stream if mimetype detection fails. #157
Conversation
googlebot
added
the
cla: yes
| @@ -425,7 +425,11 @@ def __init__(self, filename, mimetype=None, chunksize=DEFAULT_CHUNK_SIZE, | |||
| self._filename = filename | |||
| fd = open(self._filename, 'rb') | |||
| if mimetype is None: | |||
| (mimetype, encoding) = mimetypes.guess_type(filename) | |||
| # No mimetype provided, make a guess. | |||
| (mimetype, _) = mimetypes.guess_type(filename) | |||
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
|
Looks mostly good; thank you for the contribution! |
|
Ping! |
|
@postmasters: your commit appears to have a different author than it does a committer; is that as you intend it to be? @tmatsuo: want to be a second pair of eyes on this? |
|
@nathanielmanistaatgoogle I don't understand your comment. Do you see a different email address being used than namnguyen at google.com? The commit should come from that address. |
|
@postmasters: do you see the text "postmasters committed with Nam T. Nguyen 13 days ago" when you click on the "Commits" tab of this pull request? |
|
Yes. I don't understand what that line is really saying though. "Nam T. Nguyen" is me. |
|
Do a |
The function mimetypes.guess_type returns (None, None) if it fails to find a matching extension. In that case, we should just treat the media as application/octet-stream.
|
Got it. Done. |
|
Looks good to me too :) |
Default to octet-stream if mimetype detection fails.
|
When will this be reflected in the PyPI package? |
|
@Anmol-Singh-Jaggi: It should be reflected in the 1.5.0 PyPI package. |
|
Great! :D |
The function mimetypes.guess_type returns (None, None) if it fails to
find a matching extension. In that case, we should just treat the media
as application/octet-stream.