Agreed. Although not all list assignments seem to do it, and you don;t need to be using either join or x or print?

I'm pretty sure that this is an over zealous optimisation -- reusing an existing stack frame or some such.

By tying the hash, the problem disappears.

P:\test>perl
sub TIEHASH { return bless {}, 'main'; } 
sub STORE{ $_[0]->{$_[1]} = $_[2]; } 
sub FETCH{ $_[0]->{$_[1]} } 
sub CLEAR{ undef %{$_[0]} }
tie %h, 'main';
print %h = ( 1,1,1,1);
^Z
1111
[download]

I tried using B::Concise to isolate a difference between a bizarre result and an expected one.

Normal

P:\test>perl -we"print map 1, %h = (1,1,1,1);"
Name "main::h" used only once: possible typo at -e line 1.
1111
P:\test>perl -MO=Concise -we"print map 1, %h = (1,1,1,1);"
Name "main::h" used only once: possible typo at -e line 1.
i  <@> leave[t1] vKP/REFC ->(end)
1     <0> enter ->2
2     <;> nextstate(main 1 -e:1) v ->3
h     <@> print vK ->i
3        <0> pushmark s ->4
f        <|> mapwhile(other->g)[t4] lK/1 ->h
e           <@> mapstart lK/2 ->f
4              <0> pushmark s ->5
-              <1> null lK/1 ->5
g                 <$> const(SPECIAL Null)[t9] s ->f
d              <2> aassign[t3] lKMS ->e
-                 <1> ex-list lKP ->a
5                    <0> pushmark s ->6
6                    <$> const(SPECIAL Null)[t5] s ->7
7                    <$> const(SPECIAL Null)[t6] s ->8
8                    <$> const(SPECIAL Null)[t7] s ->9
9                    <$> const(SPECIAL Null)[t8] s ->a
-                 <1> ex-list lK ->d
a                    <0> pushmark s ->b
c                    <1> rv2hv[t2] lKRM*/1 ->d
b                       <#> gv s ->c
-e syntax OK
[download]

Bizarre

P:\test>perl -we"print %h = (1,1,1,1);"
Name "main::h" used only once: possible typo at -e line 1.
Use of uninitialized value in print at -e line 1.
111
P:\test>perl -MO=Concise -we"print %h = (1,1,1,1);"
Name "main::h" used only once: possible typo at -e line 1.
e  <@> leave[t1] vKP/REFC ->(end)
1     <0> enter ->2
2     <;> nextstate(main 1 -e:1) v ->3
d     <@> print vK ->e
3        <0> pushmark s ->4
c        <2> aassign[t3] lKS ->d
-           <1> ex-list lKP ->9
4              <0> pushmark s ->5
5              <$> const(SPECIAL Null)[t4] s ->6
6              <$> const(SPECIAL Null)[t5] s ->7
7              <$> const(SPECIAL Null)[t6] s ->8
8              <$> const(SPECIAL Null)[t7] s ->9
-           <1> ex-list lK ->c
9              <0> pushmark s ->a
b              <1> rv2hv[t2] lKRM*/1 ->c
a                 <#> gv s ->b
-e syntax OK
[download]

The only difference I can see is entirely due to the map. Exclude that, and the rest is identical.

So, a bug? If it is, it's been arounf since at least 5.6.1!