Feature #3653: Diferential behaviour of positives and negatives ranges as subindex of string or arrays.

• Status: Rejected
• Assignee: matz (Yukihiro Matsumoto)
• Target version: 2.6

[dsilber (Daniel Silberschmidt)]

=begin
The expression '[a..b]' when applied to a string or to an array return the slice from the string/array from a to b.
for example [0..3] returns the first four letters from a string
irb(main):004:0> "carromato"[0..3]
=> "carr"
the slice [-4..-1] returns the last four letters from a string
irb(main):005:0> "carromato"[-4..-1]
=> "mato"
but if the length of the original string is smaller than the slice the results are conceptually very different depending on positive or negative indexes:
irb(main):006:0> "carromato"[0..24]
=> "carromato"
irb(main):007:0> "carromato"[-25..-1]
=> nil
I would like a simetric behaviour in the last case returning the whole word (and not nil). It would be more natural to have the same behaviour when exeding by left of by right of the string/array.
Does Anbody agree or disagree whith my position?
Daniel
=end

[shyouhei (Shyouhei Urabe)]

• Status changed from Open to Assigned
• Assignee set to matz (Yukihiro Matsumoto)

=begin

=end

[alexeymuranov (Alexey Muranov)]

I agree with this proposal. -Alexey.

[mame (Yusuke Endoh)]

• Description updated (diff)
• Target version set to 2.6

[matz (Yukihiro Matsumoto)]

• Status changed from Assigned to Rejected

I don't think the change does not worth the potential incompatibility.

Matz.