[Issue 5849] std.random.dice is better as a range
d-bugmail at puremagic.com
d-bugmail at puremagic.com
Wed Jan 4 14:58:11 PST 2012
http://d.puremagic.com/issues/show_bug.cgi?id=5849
bearophile_hugs at eml.cc changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #1060|0 |1
is obsolete| |
--- Comment #4 from bearophile_hugs at eml.cc 2012-01-04 14:58:04 PST ---
Created an attachment (id=1066)
Version 0.5 of the fast dice
> You may want to package contributions as pull requests. Thanks!
Thank you. I think the current code is not yet fit for a pull request, there
are some things to be done/solved first:
- Create statistical unittests.
- Is popLast() overkill/not useful enough?
- Is the Range correct now?
- Is uint[] enough for mAlias?
- ddoc comments are missing still.
- Currently the Dice constructor lacks a pre-condition: I think the input
probabilities must sum to 1.0. But is this necessary? An alternative
design is to accept any array of numbers (integers too), and normalize
their sum to 1.0 inside the Dice constructor itself.
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
More information about the Digitalmars-d-bugs
mailing list