Bug #14086: Interpreter seems to sneak in a nil value when a comma is missing on a nested array wrapped in parenthesis

• Status: Rejected
• Assignee: -
• Target version: -
• Backport: 2.3: UNKNOWN, 2.4: UNKNOWN

[mariogintili (Mario Gintili)]

I believe the title is self explanatory. But code is always louder than words:

$ ruby --version
ruby 2.2.4p230 (2015-12-16 revision 53155) [x86_64-darwin16]
$ irb
2.2.4 :001 > ([[2,2], [3,3], [4,4], [5,5], [6,6]])
 => [[2, 2], [3, 3], [4, 4], [5, 5], [6, 6]] 
2.2.4 :002 > ([[2,2] [3,3], [4,4], [5,5], [6,6]])
 => [nil, [4, 4], [5, 5], [6, 6]] 

the only difference between lines 1 and 2 in my REPL is a coma right after the first nested array. is this the expected behaviour?

[zverok (Victor Shepelev)]

is this the expected behaviour?

Yes.

Try this:

[2, 2][3, 3] # => nil

Why?
Because it is read as "Array of [2,2], call operator [] with arguments 3, 3 (which means fetch 3 elements starting from 3rd)". As this array doesn't have a 3rd element, the result of the operation is nil. So the first item of your larger array is the result of this calculation, e.g. nil.

[marcandre (Marc-Andre Lafortune)]

• Status changed from Open to Rejected