Message201539
| Author |
eli.bendersky |
| Recipients |
Arfrever, christian.heimes, effbot, eli.bendersky, ezio.melotti, flox, georg.brandl, martin.panter, python-dev, r.david.murray |
| Date |
2013年10月28日.13:06:09 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<CAF-Rda_d5Hujr2E_Y_zGodNQ4gNvjjr8AMAamWgrRn0yjbV+ig@mail.gmail.com> |
| In-reply-to |
<1382965200.2.0.346789077267.issue14007@psf.upfronthosting.co.za> |
| Content |
On Mon, Oct 28, 2013 at 6:00 AM, Martin Panter <report@bugs.python.org>wrote:
>
> Martin Panter added the comment:
>
> The best way to work around it for me is just to ignore the warning. It
> doesn’t really worry me that much, I only noticed it while porting a
> program to Python 3 anyway. So if you don’t want to touch the 2.7 branch I
> can live with that :)
>
I'd rather not. 2.7.x is in maintenance mode - we'll fix serious bugs and
may change trivial things. But in this case, since the effects are hardly
serious we should consider the danger of mucking with code in a very stable
maintenance branch. While this issue may seem simple at the surface, it's
not. That's because ET in 3.3+ is quite different from ET in 3.2-
(including 2.7-). In 3.3 ET underwent major changes, and its testing
infrastructure was greatly improved too. Back-porting across the 3.3
boundary is risky because it's hard to foresee the effects (i.e.
interactions between the C accelerator and the Python module). |
|