Bug #15712: DateTime#=== should be defined and compare date and time instead of just the date

• Status: Closed
• Assignee: -
• Target version: -
• ruby -v: ruby 2.7.0dev (2019-03-18 trunk 67296) [x86_64-darwin17]
• Backport: 2.4: UNKNOWN, 2.5: UNKNOWN, 2.6: UNKNOWN
• Tags: ext

[localhostdotdev (localhost .dev)]

DateTime#=== is inherited from Date#=== without overwriting the #=== method, this leads to DateTime#=== comparing only dates and ignore the time part.

DateTime.new(2001, 2, 3) === DateTime.new(2001, 2, 3, 12)
=> false

I think this is not the expected behavior but existing code might rely on this.

Date#===: https://ruby-doc.org/stdlib-2.1.9/libdoc/date/rdoc/Date.html#method-i-3D-3D-3D

Source: https://medium.com/@dvandersluis/an-rspec-time-issue-and-its-not-about-timezones-a89bbd167b86

[hsbt (Hiroshi SHIBATA)]

@localhostdotdev (localhost .dev)

I fixed an issue of our tracker. I deleted 15713-15716.

[nobu (Nobuyoshi Nakada)]

• Status changed from Open to Feedback

What version of DateTime do you use?

With the trunk I got:

require 'date'
dt1 = DateTime.new(2001, 2, 3)
dt2 = DateTime.new(2001, 2, 3, 12)
p dt1 === dt2 #=> true
p dt1 == dt2  #=> false

[localhostdotdev (localhost .dev)]

@nobu (Nobuyoshi Nakada) Not sure why I wrote false (maybe I confused === with == in my testing), I'm getting true on ruby 2.5.3, 2.6.1, 2.6.2 and ruby-head (2.7.0 (same as the one registered in the issue)).

rvm ruby-head do ruby -e 'require "date"; dt1 = DateTime.new(2001, 2, 3); dt2 = DateTime.new(2001, 2, 3, 12); puts "=== #{dt1 === dt2}"; puts "== #{dt1 == dt2 }"; puts RUBY_VERSION'

[matz (Yukihiro Matsumoto)]

I admit it is a bug. But we consider use of DateTime should be discouraged. So we keep this bug behavior for compatibility's sake.
We are going to describe obsoletion in the document.

Matz.

[jeremyevans0 (Jeremy Evans)]

• Status changed from Feedback to Closed

DateTime deprecation documentation was added to date: https://github.com/ruby/date/commit/58ca6e6a3ee20c72a77266e0f74920b12a06ee9d

Date#=== is describes the current behavior as only considering the date, and uses examples with DateTime, so I think the situation is fully documented and I'm closing this.