[perl #119513] [BUG] Non-digit parts of a number (like the underscore _) aren't considered to be part of the number in repetition quantifiers in regexes in Rakudo (but are in STD.pm6)

[Carl Mäsak]

# New Ticket Created by  "Carl Mäsak" 
# Please include the string:  [perl #119513]
# in the subject line of all future correspondence about this issue. 
# <URL: https://rt.perl.org:443/rt3/Ticket/Display.html?id=119513 >


<yoleaux> 01:36Z <diakopter> masak: unexpected in rakudo:  say so "a"
x 1_0 ~~ /a ** 1_0/  says  False
<masak> r: say so "a" x 1_0 ~~ /a ** 1_0/
<camelia> rakudo 9faa30: OUTPUT«False␤»
<masak> r: say so "a" ~~ /a ** 1_0/
<camelia> rakudo 9faa30: OUTPUT«False␤»
<masak> r: say so "a_0" ~~ /a ** 1_0/
<camelia> rakudo 9faa30: OUTPUT«True␤»
<moritz> it's parsed as /[a ** 1] _0/
<masak> moritz: yes.
<moritz> masak: and TimToady++ already patched it in STD
<jnthn> So just needs somebody to nab the patch :)
* masak submits rakudobug
<masak> r: say so "a" ~~ / a ** 1e0 / # what about this one?
<camelia> rakudo 9faa30: OUTPUT«False␤»
<masak> r: say so "ae0" ~~ / a ** 1e0 /
<camelia> rakudo 9faa30: OUTPUT«True␤»
<moritz> masak: I'd think it would make sense to require a word
boundary after the number that follows a **
<moritz> number of reptitions are inherently integers, so I don't see
the point of allowing floating-point numbers here
<masak> moritz: I think I agree.