Issue #16440 has been updated by shevegen (Robert A. Heiler).


I have no strong opinion either way but I can understand the 
assumption by st0012 to some extent. For example, I personally
always seem to think more about .include? than .cover?, largely
because I simply use .include? a lot more. I once even added
some .partial_include? method to Enumerable (or somewhere else,
I don't remember ... was years ago).

The other thing is DateTime, Date, and Time. Personally I'd love
if we could have just one-ring-to-rule-them-all one day, perhaps
in ruby 4.0 or so - I think that is a partial complaint by Stan,
in the sense of the behaviour he showed (but I am assuming this
here). But again, I have no real strong opinion either way.

Would be interesting to ask Stan Lo whether he knew about 
.cover? or not. :)

----------------------------------------
Bug #16440: Date range inclusion behaviors are inconsistent
https://bugs.ruby-lang.org/issues/16440#change-83306

* Author: st0012 (Stan Lo)
* Status: Open
* Priority: Normal
* Assignee: 
* Target version: 
* ruby -v: ruby 2.6.5p114 (2019-10-01 revision 67812) [x86_64-darwin19]
* Backport: 2.5: UNKNOWN, 2.6: UNKNOWN
----------------------------------------
It's weird that a Date range can include Time and DateTime objects that were converted from a Date object. But it can't include a newly generated DateTime object. For example:

```
may1 = Date.parse("2019-05-01")
may3 = Date.parse("2019-05-03")
noon_of_may3 = DateTime.parse("2019-05-03 12:00")
may31 = Date.parse("2019-05-31")

(may1..may31).include? may3 # => True
(may1..may31).include? may3.to_time # => True
(may1..may31).include? may3.to_datetime # => True
(may1..may31).include? noon_of_may3 # => False
```

Shouldn't the last case return `true` as well? 

Related Rails issue: https://github.com/rails/rails/issues/36175



-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>