Message246388
| Author |
martin.panter |
| Recipients |
BreamoreBoy, angad, berker.peksag, docs@python, martin.panter, orsenthil, r.david.murray |
| Date |
2015年07月07日.01:01:37 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<1436230898.68.0.849372579109.issue13456@psf.upfronthosting.co.za> |
| In-reply-to |
| Content |
Why do people want "response_class" to be part of the API? If so, more details about it may need to added, e.g. the following methods and attributes seem to be required: _read_status(), fp, close(), isclosed(), begin() and will_close.
The "debuglevel" attribute seems fairly redundant with the existing set_debuglevel() method.
Also, what is the point of adding the "default_port" attribute, if it cannot be modified? The only use case I can imagine is in a subclass that specifically does modify it. But I’m not sure it should be added at all.
So I am sorry, but I don’t see why any of the three additions in the patch should be made. IMO it would be better to explain that "response_class" is an internal implementation detail, or even drop it entirely from the doc string. |
|