On Sat, 21 Oct 2017 08:28:19 -0700, c...@zoffix.com wrote: > On Mon, 23 Nov 2015 19:40:36 -0800, raiph wrote: > > What I did > > ========== > > > > $/ := "Uhoh"; > > put "Foo".subst: /Foo/, -> $/ {$/}; > > put "Foo".subst: /Foo/, -> $/ {$/} for 1; > > put "Foo".subst: /Foo/, {$/}; > > put "Foo".subst: /Foo/, {$/} for 1; > > > > What I expected > > =============== > > > > FooFooFooFoo > > > > What I got > > ========== > > > > FooFooUhohUhoh > > > > Is it a bug? > > ============ > > > > The way it's working now might make sense from a purist pov but not a > > pragmatic or Perlish one imo. I'm hoping this will be obvious to > > whoever reads this so I'll defer giving a detailed rationale for > > calling this a bug in the anticipation that it won't be necessary. :) > > > > Does it look like any other bug? > > ================================ > > > > Yes, at least the previously resolved s/// bug #118705. > > > > See also http://www.perlmonks.org/?node_id=1148378 > > > Thank you for the report. This is now fixed. > > Fix: https://github.com/rakudo/rakudo/commit/738908be4d > Test: https://github.com/perl6/roast/commit/80a49c4324
I ended up reverting the tests and the fix. There was fallout from ~20 modules due to this and it feels to me having method forms mess with $/ is going against the grain for how the users tend to use these routines. I filed another Issue https://github.com/rakudo/rakudo/issues/1235 for this problem.