Issue 4812: Junk in the decimals namespace (original) (raw)

Created on 2009-01-02 22:31 by rhettinger, last changed 2022-04-11 14:56 by admin. This issue is now closed.

Messages (7)
msg78893 - (view)	Author: Raymond Hettinger (rhettinger) *	Date: 2009-01-02 22:31
In r59144 , a bunch of internal-use constants were dumped in the main namespace. These all need to be prefixed with an underscore. They should be fixed right-away before people start using them. Since they are externally undocumented and since the internal notes describe them as being only for internal-use, I think this can go in as a bugfix.
msg78894 - (view)	Author: Raymond Hettinger (rhettinger) *	Date: 2009-01-02 22:35
Mark, it looks like r59144 merely moved these around and that they existed beforehand. Can we go ahead and get them renamed? They are defeating the extensive efforts we've made to keep that namespace free of anything that isn't part of the documented spec.
msg78900 - (view)	Author: Mark Dickinson (mark.dickinson) *	Date: 2009-01-02 23:09
I assume you're referring to Dec_0 and friends? Fixed in the trunk in r68182. Leaving open until I get the chance to merge it to the other branches.
msg78901 - (view)	Author: Raymond Hettinger (rhettinger) *	Date: 2009-01-02 23:22
Thanks. BTW, my preferred new names are _One _Zero and _NegativeOne. Names like dec_n1 are cryptic in places other than where they're defined.
msg78902 - (view)	Author: Mark Dickinson (mark.dickinson) *	Date: 2009-01-02 23:34
Fine. Will re-rename in the morning. Are _Inf, _negInf, _NaN and _Infsign okay as they are?
msg78907 - (view)	Author: Raymond Hettinger (rhettinger) *	Date: 2009-01-03 00:01
Let's spell them out: _Infinity, _NegativeInfinity, _NaN, and _SignedInfinity.
msg78951 - (view)	Author: Mark Dickinson (mark.dickinson) *	Date: 2009-01-03 12:18
Done (r68191 through r68194).