# Time#round in 1.9 **URL:** \ **Category:** ruby-talk\ **Created:** [30 August 2010 18:42 UTC](https://rubytalk.org/t/time-round-in-1-9/59831 "2010-08-30T18:42:09Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![7rans](https://avatars.discourse-cdn.com/v4/letter/7/a8b319/32.png) [@7rans](https://rubytalk.org/u/7rans)\ **Post date:** [30 August 2010 18:42 UTC](https://rubytalk.org/t/time-round-in-1-9/59831/1 "2010-08-30T18:42:09Z")
Anyone know about Time#round in 1.9. Apparently it exists but it seems to do nothing but eat up cpu cycles.   ruby-1.9.2-p0 \> t = Time.new    =\> 2010-08-30 14:34:03 -0400   ruby-1.9.2-p0 \> t.round(100000)    =\> 2010-08-30 14:34:03 -0400   ruby-1.9.2-p0 \> t.round(1000000)    =\> 2010-08-30 14:34:03 -0400 ---
**Author:** ![Kirk\_Haines2](https://yyz1.discourse-cdn.com/flex029/user_avatar/rubytalk.org/kirk_haines2/32/1992_2.png) [@Kirk\_Haines2](https://rubytalk.org/u/Kirk_Haines2)\ **Post date:** [30 August 2010 19:13 UTC](https://rubytalk.org/t/time-round-in-1-9/59831/2 "2010-08-30T19:13:48Z")
A voice was heard to utter: "Use the source, Luke! Errr, uh, Trans. Oh, and the docs!" /\* \* call-seq: \* time.round([ndigits]) -\> new\_time > **···** > > On Mon, Aug 30, 2010 at 12:42 PM, Intransition \ wrote: > > > Anyone know about Time#round in 1.9. Apparently it exists but it seems > > to do nothing but eat up cpu cycles. > > > > ruby-1.9.2-p0 \> t = Time.new > > =\> 2010-08-30 14:34:03 -0400 > > ruby-1.9.2-p0 \> t.round(100000) > > =\> 2010-08-30 14:34:03 -0400 > > ruby-1.9.2-p0 \> t.round(1000000) > > =\> 2010-08-30 14:34:03 -0400 > > \* > \* Rounds sub seconds to a given precision in decimal digits (0 digits > by default). > \* It returns a new time object. > \* \_ndigits\_ should be zero or positive integer. > \* > \* require 'time' > \* > \* t = Time.utc(2010,3,30, 5,43,"25.123456789".to\_r) > \* p t.iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* p t.round.iso8601(10) #=\> "2010-03-30T05:43:25.0000000000Z" > \* p t.round(0).iso8601(10) #=\> "2010-03-30T05:43:25.0000000000Z" > \* p t.round(1).iso8601(10) #=\> "2010-03-30T05:43:25.1000000000Z" > \* p t.round(2).iso8601(10) #=\> "2010-03-30T05:43:25.1200000000Z" > \* p t.round(3).iso8601(10) #=\> "2010-03-30T05:43:25.1230000000Z" > \* p t.round(4).iso8601(10) #=\> "2010-03-30T05:43:25.1235000000Z" > \* p t.round(5).iso8601(10) #=\> "2010-03-30T05:43:25.1234600000Z" > \* p t.round(6).iso8601(10) #=\> "2010-03-30T05:43:25.1234570000Z" > \* p t.round(7).iso8601(10) #=\> "2010-03-30T05:43:25.1234568000Z" > \* p t.round(8).iso8601(10) #=\> "2010-03-30T05:43:25.1234567900Z" > \* p t.round(9).iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* p t.round(10).iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* > \* t = Time.utc(1999,12,31, 23,59,59) > \* p((t + 0.4).round.iso8601(3)) #=\> "1999-12-31T23:59:59.000Z" > \* p((t + 0.49).round.iso8601(3)) #=\> "1999-12-31T23:59:59.000Z" > \* p((t + 0.5).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.4).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.49).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.5).round.iso8601(3)) #=\> "2000-01-01T00:00:01.000Z" > \* > \* t = Time.utc(1999,12,31, 23,59,59) > \* p (t + 0.123456789).round(4).iso8601(6) #=\> > "1999-12-31T23:59:59.123500Z" > \*/ > > Below that is a short, simple to follow implementation. You are > telling it to round to 100000 or 1000000 digits. That's probably not > what you mean? > > Kirk Haines ---
**Author:** ![7rans](https://avatars.discourse-cdn.com/v4/letter/7/a8b319/32.png) [@7rans](https://rubytalk.org/u/7rans)\ **Post date:** [30 August 2010 19:45 UTC](https://rubytalk.org/t/time-round-in-1-9/59831/3 "2010-08-30T19:45:17Z")
> \> Anyone know about Time#round in 1.9. Apparently it exists but it seems > \> to do nothing but eat up cpu cycles. > > \> ruby-1.9.2-p0 \> t = Time.new > \> =\> 2010-08-30 14:34:03 -0400 > \> ruby-1.9.2-p0 \> t.round(100000) > \> =\> 2010-08-30 14:34:03 -0400 > \> ruby-1.9.2-p0 \> t.round(1000000) > \> =\> 2010-08-30 14:34:03 -0400 > > A voice was heard to utter: "Use the source, Luke! Errr, uh, Trans. > Oh, and the docs!" The docs would be nice, but ri has been broken for months[1] and I could not find any mention of the method online. But you are right I could have checked the source. > /\* > \* call-seq: > \* time.round([ndigits]) -\> new\_time > \* > \* Rounds sub seconds to a given precision in decimal digits (0 digits > by default). > \* It returns a new time object. > \* \_ndigits\_ should be zero or positive integer. > \* > \* require 'time' > \* > \* t = Time.utc(2010,3,30, 5,43,"25.123456789".to\_r) > \* p t.iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* p t.round.iso8601(10) #=\> "2010-03-30T05:43:25.0000000000Z" > \* p t.round(0).iso8601(10) #=\> "2010-03-30T05:43:25.0000000000Z" > \* p t.round(1).iso8601(10) #=\> "2010-03-30T05:43:25.1000000000Z" > \* p t.round(2).iso8601(10) #=\> "2010-03-30T05:43:25.1200000000Z" > \* p t.round(3).iso8601(10) #=\> "2010-03-30T05:43:25.1230000000Z" > \* p t.round(4).iso8601(10) #=\> "2010-03-30T05:43:25.1235000000Z" > \* p t.round(5).iso8601(10) #=\> "2010-03-30T05:43:25.1234600000Z" > \* p t.round(6).iso8601(10) #=\> "2010-03-30T05:43:25.1234570000Z" > \* p t.round(7).iso8601(10) #=\> "2010-03-30T05:43:25.1234568000Z" > \* p t.round(8).iso8601(10) #=\> "2010-03-30T05:43:25.1234567900Z" > \* p t.round(9).iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* p t.round(10).iso8601(10) #=\> "2010-03-30T05:43:25.1234567890Z" > \* > \* t = Time.utc(1999,12,31, 23,59,59) > \* p((t + 0.4).round.iso8601(3)) #=\> "1999-12-31T23:59:59.000Z" > \* p((t + 0.49).round.iso8601(3)) #=\> "1999-12-31T23:59:59.000Z" > \* p((t + 0.5).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.4).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.49).round.iso8601(3)) #=\> "2000-01-01T00:00:00.000Z" > \* p((t + 1.5).round.iso8601(3)) #=\> "2000-01-01T00:00:01.000Z" > \* > \* t = Time.utc(1999,12,31, 23,59,59) > \* p (t + 0.123456789).round(4).iso8601(6) #=\> > "1999-12-31T23:59:59.123500Z" > \*/ > > Below that is a short, simple to follow implementation. You are > telling it to round to 100000 or 1000000 digits. That's probably not > what you mean? I see, so it only rounds sub-seconds per digit. That's too bad, Facets' #round method could round to any second. I have to rename. Thanks. [1] /home/trans/.rvm/rubies/ruby-1.9.2-p0/lib/ruby/1.9.1/rdoc/ri/ driver.rb:909:in `load_method': undefined method `' for nil:NilClass (NoMethodError) > **···** > > On Aug 30, 3:13 pm, Kirk Haines \ wrote: > > > On Mon, Aug 30, 2010 at 12:42 PM, Intransition \ wrote: