From @masak

<thundergnat> std​: my $x; $x &&= 1; say $x;
<p6eval> std : OUTPUT«ok 00​:01 118m␤»
<thundergnat> rakudo​: my $x; $x &&= 1; say $x;
<p6eval> rakudo 51e3ff​: OUTPUT«No applicable candidates found to
dispatch to for 'infix​:<&&>'. Available candidates are​:␤​:(Mu $a, Mu
$b)␤␤ in main program body at line 1␤»
* masak submits rakudobug
<masak> thundergnat++
<thundergnat> rakudo​: my $x; $x ?&= 1; say $x;
<p6eval> rakudo 51e3ff​: OUTPUT«No applicable candidates found to
dispatch to for 'infix​:<?&>'. Available candidates are​:␤​:(Any $a, Any
$b)␤␤ in main program body at line 1␤»
<thundergnat> rakudo​: my $x; $x ?|= 1; say $x;
<p6eval> rakudo 51e3ff​: OUTPUT«No applicable candidates found to
dispatch to for 'infix​:<?|>'. Available candidates are​:␤​:(Any $a, Any
$b)␤␤ in main program body at line 1␤»
<thundergnat> rakudo​: my $x; $x ?^= 1; say $x;
<p6eval> rakudo 51e3ff​: OUTPUT«No applicable candidates found to
dispatch to for 'infix​:<?^>'. Available candidates are​:␤​:(Any $a, Any
$b)␤␤ in main program body at line 1␤»

From @pmichaud

Now fixed in 358d728. This fix also provides the reduction [?&], [?|],
etc. forms for 0- and 1-element lists.

Assigning to moritz++ for spectest coverage.

Pm

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

From @moritz

Basic tests added to S03-operators/short-circuit.t.

19​:52 < moritz_> rakudo​: my $x; $x &&= 5; say $x
19​:52 <+p6eval> rakudo a204ba​: OUTPUT«5␤»
19​:52 < moritz_> somebody please confirm that this is wrong
...
19​:53 < jnthn> moritz_​: Looks wrong to me.

I'm glad I haven't lost my sanity over it :-)

From @kyleha

This is an automatically generated mail to inform you that tests are now available in S03-operators/short-circuit.t

commit 519bef8c25fae4227ae9676f84da410b39291357
Author​: Moritz Lenz <moritz@​faui2k3.org>
Date​: Sat Sep 18 19​:54​:10 2010 +0200

  some basic tests for RT #​77864, &&=, ||= etc. One of them fails

diff --git a/S03-operators/short-circuit.t b/S03-operators/short-circuit.t
index 87d0254..a6388ef 100644
--- a/S03-operators/short-circuit.t
+++ b/S03-operators/short-circuit.t
@@ -14,7 +14,7 @@ it is closely related to || and && and //.
 
 # test cases by Andrew Savige
 
-plan 37;
+plan 39;
 
 {
     my $x = 1;
@@ -174,6 +174,17 @@ plan 37;
 # a rakudo regression
 ok (0 || 0 || 1), '0 || 0 || 1 is true';
 
+# RT #77864
+{
+    my $x;
+    $x &&= 5;
+    #?rakudo todo '77864'
+    nok $x.defined, '&&= leaves var on the right undefined';
+    my $y ||= 'moin';
+    is $y, 'moin', '||= on a fresh variable works';
+
+}
+
 done_testing;
 
 # vim: ft=perl6

From @pmichaud

On Sat Sep 18 10​:55​:52 2010, moritz wrote​:

Basic tests added to S03-operators/short-circuit.t.

19​:52 < moritz_> rakudo​: my $x; $x &&= 5; say $x
19​:52 <+p6eval> rakudo a204ba​: OUTPUT«5␤»
19​:52 < moritz_> somebody please confirm that this is wrong
...
19​:53 < jnthn> moritz_​: Looks wrong to me.

Rakudo is correct here -- see S03​:3979​:

  "If you apply an assignment operator to a container containing a
  type object (which is undefined), [...] the operation is defined
  in terms of the corresponding reduction operator, where the type
  object autovivifies to the operator's identity value."

The identity value for [&&] is True, so Rakudo has this correct.

By way of comparison, consider​: "my $x; $x *= 5;" which leaves $x
with the value of 5.

Pm

From @pmichaud

Updated test to match spec, resolving ticket.

Pm

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