This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author ncoghlan
Recipients berker.peksag, brett.cannon, eric.snow, ethan.furman, fijall, ncoghlan, python-dev, vstinner
Date 2016-09-10.17:24:54
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1473528294.97.0.410101132167.issue24254@psf.upfronthosting.co.za>
In-reply-to
Content
While I hadn't read the related thread at the point where I made that comment (so thank you for at least raising the question there), I'm still on the opposite side of the planet, so any decision made in less than 24 hours from proposal to resolution is necessarily "too fast" for global collaboration (and 48 hours is better).

I'm not particularly fond of __definition_order__ either, but one of the specific points raised in the PEP 520 discussions due to Inada-san's pending patch was whether or not making class namespaces ordered by default would eliminate the need for the attribute, and the conclusion was that it *wouldn't*. So the "We made class namespaces ordered, so now __definition_order__ is gone" sounded a lot like folks forgetting that part of the discussions, and instead getting rid of a feature that was there for multiple reasons, not just because __dict__ didn't preserve the order.
History
Date User Action Args
2016-09-10 17:24:55ncoghlansetrecipients: + ncoghlan, brett.cannon, vstinner, ethan.furman, fijall, python-dev, eric.snow, berker.peksag
2016-09-10 17:24:54ncoghlansetmessageid: <1473528294.97.0.410101132167.issue24254@psf.upfronthosting.co.za>
2016-09-10 17:24:54ncoghlanlinkissue24254 messages
2016-09-10 17:24:54ncoghlancreate