Bug #9595: Float#to_d inconsistent between 2.0.0 and 2.1.1

• Status: Rejected
• Assignee: mrkn (Kenta Murata)
• Target version: -
• ruby -v: ruby 2.1.1p76 (2014-02-24 revision 45161) [x86_64-darwin13.0]
• Tags: ext

[boone (Mike Boone)]

The following code produces different BigDecimal results in 2.0.0 vs. 2.1.1:

require 'big decimal'
require 'bigdecimal/util'
(1/1.3667).to_d

2.0.0 returns: #<BigDecimal:7fa7fc04dc28,'0.7316894709 885124E0',18(45)>
2.1.1 returns: #<BigDecimal:7fabd1a25768,'0.7316894709 88512E0',18(36)>

2.1.1 is a digit short. Related to #9305?

[boone (Mike Boone)]

I did come across this Github PR (https://github.com/ruby/ruby/pull/323) which changed the default precision on Float#to_d.

If 2.1.1 is the proper behavior and compatibility with the behavior of 2.0.0 is not desired, please close this bug.

Thanks.

[mrkn (Kenta Murata)]

• Category set to ext
• Status changed from Open to Assigned
• Assignee set to mrkn (Kenta Murata)

[jeremyevans0 (Jeremy Evans)]

• Status changed from Assigned to Rejected
• Backport deleted (1.9.3: UNKNOWN, 2.0.0: UNKNOWN, 2.1: UNKNOWN)

BigDecimal behavior has remained the same since 2.1, and 2.0's excessive precision was actually a bug, fixed in 1699396fe502d929ced019dc0dd3463a35fbf6d3.