From @finanalyst

finanalyst​: perl6​: my @​v=2,5; say ( map {; $^b => $^a }, @​v ).perl
p6eval​: rakudo 32fda1​: OUTPUT«Lexical '$b' not found␤current instr.​:
'_block14' pc 29 (EVAL_1​:0)␤»
p6eval​: ..elf 30299​: OUTPUT«Undefined subroutine &GLOBAL​::map called at
(eval 124) line 4.␤ at ./elf_h line 5881␤»
p6eval​: ..pugs​: OUTPUT«((\5 => \2),)␤»
finanalyst​: alpha​: my @​v=2,5; say ( map {; $^b => $^a }, @​v ).perl
p6eval​: alpha 30e0ed​: OUTPUT«[5 => 2]␤»

From @moritz

21​:24 <@​moritz_> rakudo​: say ({; $_ => 1}).WHAT
21​:24 <+p6eval> rakudo 43f110​: OUTPUT«Hash()␤»
21​:25 <@​moritz_> looks like a bug to me
21​:25 <@​colomon> yup.

I'd expect Block instead

From @moritz

Although it looks like a bug with twigils, it's actually a bug in the
hash vs. block distinction algorithm. Despite the leading ';' rakudo
interprets the construct as a hash.

The RT System itself - Status changed from 'new' to 'open'

From @pmichaud

Now fixed in 06723b4... assigning to moritz++ for test verification and
unfudging.

Pm

@moritz - Status changed from 'open' to 'resolved'