# Time class instance **URL:** \ **Category:** IronRuby\ **Created:** [April 24, 2009, 10:56pm UTC](https://www.ruby-forum.com/t/time-class-instance/166415 "2009-04-24T22:56:29Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [April 24, 2009, 10:56pm UTC](https://www.ruby-forum.com/t/time-class-instance/166415/1 "2009-04-24T22:56:29Z")
Hi Time class instance in IronRuby act as static. Given following code: t = Time.new t.utc t.utc? # should be true behave t.localtime t.utc? # should be false “t” should convert itself into UTC type after method “utc” being called. There are three methods must modifying the receiver; utc, gmtime, and localtime It also effect to Time.rfc2882 formating as you can run MRI vs IronRuby at - [http://gist.github.com/101229](http://gist.github.com/101229) I have propose my fix at - [http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d](http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d) File changes: • Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/ironruby- tags/library/time/rfc2822\_tags.txt • Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/rubyspec/ core/time/utc\_spec.rb • Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/ FileOps.cs • Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/ TimeOps.cs • Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/ Initializers.Generated.cs • Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Zlib/zlib.cs • Merlin/Main/Languages/Ruby/Ruby/Builtins/RubyDateTime.cs • Merlin/Main/Languages/Ruby/Ruby/Ruby.csproj Note: Shri, this push should fix 23) Failure: test\_request\_unmodifed(TestGemRemoteFetcher) [test\_gem\_remote\_fetcher.rb:604]: Expected “Mon, 13 Apr 2009 13:22:16 -0700”, not “Mon, 13 Apr 2009 20:22:16 -0000” Thank you, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 3:44am UTC](https://www.ruby-forum.com/t/time-class-instance/166415/2 "2009-05-04T03:44:39Z")
Doesn’t look like anyone has reviewed this yet. To make sure it doesn’t get lost, can you also make a bug on codeplex for this? From: removed\_email\_address@domain.invalid [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of removed\_email\_address@domain.invalid Sent: Friday, April 24, 2009 1:56 PM To: removed\_email\_address@domain.invalid Subject: [Ironruby-core] Time class instance Hi Time class instance in IronRuby act as static. Given following code: t = Time.new t.utc t.utc? # should be true behave t.localtime t.utc? # should be false “t” should convert itself into UTC type after method “utc” being called. There are three methods must modifying the receiver; utc, gmtime, and localtime It also effect to Time.rfc2882 formating as you can run MRI vs IronRuby at - [101229’s gists · GitHub](http://gist.github.com/101229) I have propose my fix at - [http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d](http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d) File changes: \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/ironruby-tags/library/time/rfc2822\_tags.txt \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/rubyspec/core/time/utc\_spec.rb \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/FileOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/TimeOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Initializers.Generated.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Zlib/zlib.cs \* Merlin/Main/Languages/Ruby/Ruby/Builtins/RubyDateTime.cs \* Merlin/Main/Languages/Ruby/Ruby/Ruby.csproj Note: Shri, this push should fix 23) Failure: test\_request\_unmodifed(TestGemRemoteFetcher) [test\_gem\_remote\_fetcher.rb:604]: Expected “Mon, 13 Apr 2009 13:22:16 -0700”, not “Mon, 13 Apr 2009 20:22:16 -0000” Thank you, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 4:12am UTC](https://www.ruby-forum.com/t/time-class-instance/166415/3 "2009-05-04T04:12:58Z")
Hi, It would be nice to have anyone to review my code :-). bug file in at [http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021](http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021) Thanks, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 9:17am UTC](https://www.ruby-forum.com/t/time-class-instance/166415/4 "2009-05-04T09:17:23Z")
The approach looks good. I have some comments though… You should do a search for “DateTime” in the code base to find all the places that might need updating. For example, Yaml.DateTimeOps needs to be changed (is there no spec for it?), and Converter.IsIntegral might need to be. RubyDateTime should be a good .NET citizen. You should override Equals (and HashCode) - I suspect comparison might be broken without that since DateTime was a valuetype whereas RubyDateTime is not. Also ToString. You should define implicit operator converting from and to DateTime so that Ruby code can pass a Ruby Time object to a CLR method that expects DateTime. Would be nice to add constructors to RubyDateTime to match the contructors of DateTime so that you can directly create instances of RubyDateTime using “new RubyDateTime(args)” instead of using nested expressions like “new RubyDateTime(new DateTime(args))” which is harder to read. You should mark it as Serializable. Take a look at DateTime on [msdn.com](http://msdn.com) and see if there are any other interfaces or attributes you can propagate from DateTime to RubyDateTime. The comment for RubyDateTime confused me a bit. Instead of “RubyDateTime to encapsulates DateTime type which support instance’s variable”, should it not be “Encapsulates DateTime and supports mutating of instances (whereas DateTime is immutable)” Creating bugs for code reviews is extra process for you. We can try to avoid it until there are too many code reviews. So do send a reminder if no one replies to your changes within 48 hours. From: removed\_email\_address@domain.invalid [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of removed\_email\_address@domain.invalid Sent: Sunday, May 03, 2009 7:12 PM To: removed\_email\_address@domain.invalid Subject: Re: [Ironruby-core] Time class instance Hi, It would be nice to have anyone to review my code :-). bug file in at [http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021](http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021) Thanks, -Jirapong On May 4, 2009, at 8:43 AM, Jimmy S. wrote: Doesn’t look like anyone has reviewed this yet. To make sure it doesn’t get lost, can you also make a bug on codeplex for this? From: removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Sent: Friday, April 24, 2009 1:56 PM To: removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Subject: [Ironruby-core] Time class instance Hi Time class instance in IronRuby act as static. Given following code: t = Time.new t.utc t.utc? # should be true behave t.localtime t.utc? # should be false “t” should convert itself into UTC type after method “utc” being called. There are three methods must modifying the receiver; utc, gmtime, and localtime It also effect to Time.rfc2882 formating as you can run MRI vs IronRuby at - [101229’s gists · GitHub](http://gist.github.com/101229) I have propose my fix at - [http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d](http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d) File changes: \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/ironruby-tags/library/time/rfc2822\_tags.txt \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/rubyspec/core/time/utc\_spec.rb \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/FileOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/TimeOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Initializers.Generated.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Zlib/zlib.cs \* Merlin/Main/Languages/Ruby/Ruby/Builtins/RubyDateTime.cs \* Merlin/Main/Languages/Ruby/Ruby/Ruby.csproj Note: Shri, this push should fix 23) Failure: test\_request\_unmodifed(TestGemRemoteFetcher) [test\_gem\_remote\_fetcher.rb:604]: Expected “Mon, 13 Apr 2009 13:22:16 -0700”, not “Mon, 13 Apr 2009 20:22:16 -0000” Thank you, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 7:28pm UTC](https://www.ruby-forum.com/t/time-class-instance/166415/5 "2009-05-04T19:28:06Z")
Hello Shri, Thank you very much for code review. It is make sense about constructors and ToString methods. I will work on it. p.s. please delete bug #1021 for me. Thank you, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 9:26am UTC](https://www.ruby-forum.com/t/time-class-instance/166415/6 "2009-05-04T09:26:09Z")
Why did you need to disable two examples in rfc2822\_spec.rb? We should not be disabling tests that worked before without a very good reason. You added a test for Time.utc. That should actually be in the second describe block in the same file for Time#utc. The (grammatically correct) name of the examples should be “modifies the receiver” From: removed\_email\_address@domain.invalid [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of Shri B. Sent: Monday, May 04, 2009 12:15 AM To: removed\_email\_address@domain.invalid Subject: Re: [Ironruby-core] Time class instance The approach looks good. I have some comments though… You should do a search for “DateTime” in the code base to find all the places that might need updating. For example, Yaml.DateTimeOps needs to be changed (is there no spec for it?), and Converter.IsIntegral might need to be. RubyDateTime should be a good .NET citizen. You should override Equals (and HashCode) - I suspect comparison might be broken without that since DateTime was a valuetype whereas RubyDateTime is not. Also ToString. You should define implicit operator converting from and to DateTime so that Ruby code can pass a Ruby Time object to a CLR method that expects DateTime. Would be nice to add constructors to RubyDateTime to match the contructors of DateTime so that you can directly create instances of RubyDateTime using “new RubyDateTime(args)” instead of using nested expressions like “new RubyDateTime(new DateTime(args))” which is harder to read. You should mark it as Serializable. Take a look at DateTime on [msdn.com](http://msdn.com) and see if there are any other interfaces or attributes you can propagate from DateTime to RubyDateTime. The comment for RubyDateTime confused me a bit. Instead of “RubyDateTime to encapsulates DateTime type which support instance’s variable”, should it not be “Encapsulates DateTime and supports mutating of instances (whereas DateTime is immutable)” Creating bugs for code reviews is extra process for you. We can try to avoid it until there are too many code reviews. So do send a reminder if no one replies to your changes within 48 hours. From: removed\_email\_address@domain.invalid [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of removed\_email\_address@domain.invalid Sent: Sunday, May 03, 2009 7:12 PM To: removed\_email\_address@domain.invalid Subject: Re: [Ironruby-core] Time class instance Hi, It would be nice to have anyone to review my code :-). bug file in at [http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021](http://ironruby.codeplex.com/WorkItem/View.aspx?WorkItemId=1021) Thanks, -Jirapong On May 4, 2009, at 8:43 AM, Jimmy S. wrote: Doesn’t look like anyone has reviewed this yet. To make sure it doesn’t get lost, can you also make a bug on codeplex for this? From: removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) [[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)] On Behalf Of removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Sent: Friday, April 24, 2009 1:56 PM To: removed\_email\_address@domain.invalid[mailto:removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Subject: [Ironruby-core] Time class instance Hi Time class instance in IronRuby act as static. Given following code: t = Time.new t.utc t.utc? # should be true behave t.localtime t.utc? # should be false “t” should convert itself into UTC type after method “utc” being called. There are three methods must modifying the receiver; utc, gmtime, and localtime It also effect to Time.rfc2882 formating as you can run MRI vs IronRuby at - [101229’s gists · GitHub](http://gist.github.com/101229) I have propose my fix at - [http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d](http://github.com/Jirapong/ironruby/commit/937a8c4aea048dd20ddea8aa6e27b0055bf6907d) File changes: \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/ironruby-tags/library/time/rfc2822\_tags.txt \* Merlin/External.LCA\_RESTRICTED/Languages/IronRuby/mspec/rubyspec/core/time/utc\_spec.rb \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/FileOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Builtins/TimeOps.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Initializers.Generated.cs \* Merlin/Main/Languages/Ruby/Libraries.LCA\_RESTRICTED/Zlib/zlib.cs \* Merlin/Main/Languages/Ruby/Ruby/Builtins/RubyDateTime.cs \* Merlin/Main/Languages/Ruby/Ruby/Ruby.csproj Note: Shri, this push should fix 23) Failure: test\_request\_unmodifed(TestGemRemoteFetcher) [test\_gem\_remote\_fetcher.rb:604]: Expected “Mon, 13 Apr 2009 13:22:16 -0700”, not “Mon, 13 Apr 2009 20:22:16 -0000” Thank you, -Jirapong ---
**Author:** ![jirapong](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/90ced4/32.png) [@jirapong](https://www.ruby-forum.com/u/jirapong)\ **Post date:** [May 4, 2009, 9:03pm UTC](https://www.ruby-forum.com/t/time-class-instance/166415/7 "2009-05-04T21:03:48Z")
Hello Shri, Because the spec passed by accident, I was checked result by this code snippet require ‘time’ t1 = (Time.utc(1976, 8, 26, 14, 30) + 4 \* 3600) t2 = Time.rfc2822(“26 Aug 76 14:30 EDT”) puts “#{t1} == #{t2}? #{t1==t2}” MRI: Thu Aug 26 18:30:00 UTC 1976 == Fri Aug 27 01:30:00 +0700 1976 true IrunRuby: Thu Aug 26 18:30:00 Z 1976 == Thu Aug 26 18:30:00 Z 1976? true According to rfc2822 section - 3.3. Date and Time Specification ([http://www.ietf.org/rfc/rfc2822.txt](http://www.ietf.org/rfc/rfc2822.txt) ) Timezone of rfc2822 should display in (( “+” / “-” ) 4DIGIT format. Thank you, -Jirapong On May 4, 2009, at 2:25 PM, Shri B. wrote: > Why did you need to disable two examples in rfc2822\_spec.rb? We > should not be disabling tests that worked before without a very good > reason. > > You added a test for Time.utc. That should actually be in the second > describe block in the same file for Time#utc. The (grammatically > correct) name of the examples should be “modifies the receiver” > [Thanks for correction :-)]