From @RandalSchwartz

  use Data​::Dumper;
  my @​dogs = ( 'Fido', 'Wags' );
  my %kennel = (
  First => \$dogs[0],
  Second => \$dogs[1],
  );
  $dogs[2] = \%kennel;
  my $mutts = \%kennel;

  print "CORRECT​:\n";
  print Data​::Dumper->new(
  [\@​dogs, \%kennel, $mutts],
  [qw($dogs $kennel $mutts)])->Purity(1)->Dump;

  print "WRONG​:\n";
  print Data​::Dumper->new(
  [\%kennel, \@​dogs, $mutts],
  [qw($kennel $dogs $mutts)])->Purity(1)->Dump;

Output​:

  CORRECT​:
  $dogs = [
  'Fido',
  'Wags',
  {
  'First' => '',
  'Second' => ''
  }
  ];
  $dogs->[2]{'First'} = \$dogs->[0];
  $dogs->[2]{'Second'} = \$dogs->[1];
  $kennel = $dogs->[2];
  $mutts = $dogs->[2];
  WRONG​:
  $kennel = {
  'First' => \'Fido',
  'Second' => \'Wags'
  };
  $dogs = [
  ${$kennel->{'First'}},
  ${$kennel->{'Second'}},
  {}
  ];
  $dogs->[2] = $kennel;
  $mutts = $kennel;

The WRONG is wrong because the $dogs->[0] is set to a separate copy of
the 'Fido' string, and no longer is associated with $kennel->{First}.
Ditto for 'Wags'.

I'm not sure how to fix this... you really need to dump $dogs before
$kennel here so you can take a reference to the array element in
place. Or it has to be patched up afterward. I'm turning Purity on,
which is usually pretty good about constructing the patchups. But it
wasn't sufficient here.

Site configuration information for perl 5.00503​:

Configured by root at Sat May 15 13​:27​:02 PDT 1999.

Summary of my perl5 (5.0 patchlevel 5 subversion 3) configuration​:
  Platform​:
  osname=linux, osvers=2.0.35, archname=i686-linux
  uname='linux halfdome.holdit.com 2.0.35 #10 mon nov 23 19​:37​:42 pst 1998 i686 unknown '
  hint=recommended, useposix=true, d_sigaction=define
  usethreads=undef useperlio=undef d_sfio=undef
  Compiler​:
  cc='cc', optimize='-O2', gccversion=egcs-2.91.60 19981201 (egcs-1.1.1 release)
  cppflags='-Dbool=char -DHAS_BOOL -I/usr/local/include'
  ccflags ='-Dbool=char -DHAS_BOOL -I/usr/local/include'
  stdchar='char', d_stdstdio=define, usevfork=false
  intsize=4, longsize=4, ptrsize=4, doublesize=8
  d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
  alignbytes=4, usemymalloc=n, prototype=define
  Linker and Libraries​:
  ld='cc', ldflags =' -L/usr/local/lib'
  libpth=/usr/local/lib /lib /usr/lib
  libs=-lnsl -lndbm -lgdbm -ldb -ldl -lm -lc -lposix -lcrypt
  libc=, so=so, useshrplib=false, libperl=libperl.a
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-rdynamic'
  cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'

Locally applied patches​:
 

@​INC for perl 5.00503​:
  /usr/lib/perl5/5.00503/i686-linux
  /usr/lib/perl5/5.00503
  /usr/lib/perl5/site_perl/5.005/i686-linux
  /usr/lib/perl5/site_perl/5.005
  .

Environment for perl 5.00503​:
  HOME=/home/merlyn
  LANG (unset)
  LANGUAGE (unset)
  LD_LIBRARY_PATH (unset)
  LOGDIR (unset)
  PATH=/home/merlyn/bin​:/usr/local/bin​:/bin​:/usr/bin
  PERL_BADLANG (unset)
  SHELL=/bin/tcsh

From @chorny

Similar results on 5.12.0.

On Wed Jul 04 06​:11​:31 2001, merlyn@​stonehenge.com wrote​:

use Data​::Dumper;
my @​dogs = \( 'Fido'\, 'Wags' \);
my %kennel = \(
  First => \\$dogs\[0\]\,
  Second =>  \\$dogs\[1\]\,
\);
$dogs\[2\] = \\%kennel;
my $mutts = \\%kennel;

print "CORRECT​:\\n";
print Data​::Dumper\->new\(
  \[\\@​dogs\, \\%kennel\, $mutts\]\,
  \[qw\($dogs $kennel $mutts\)\]\)\->Purity\(1\)\->Dump;

print "WRONG​:\\n";
print Data​::Dumper\->new\(
  \[\\%kennel\, \\@​dogs\, $mutts\]\,
  \[qw\($kennel $dogs $mutts\)\]\)\->Purity\(1\)\->Dump;

Output​:

CORRECT​:
$dogs = \[
          'Fido'\,
          'Wags'\,
          \{
            'First' => ''\,
            'Second' => ''
          \}
        \];
$dogs\->\[2\]\{'First'\} = \\$dogs\->\[0\];
$dogs\->\[2\]\{'Second'\} = \\$dogs\->\[1\];
$kennel = $dogs\->\[2\];
$mutts = $dogs\->\[2\];
WRONG​:
$kennel = \{
            'First' => \\'Fido'\,
            'Second' => \\'Wags'
          \};
$dogs = \[
          $\{$kennel\->\{'First'\}\}\,
          $\{$kennel\->\{'Second'\}\}\,
          \{\}
        \];
$dogs\->\[2\] = $kennel;
$mutts = $kennel;

The WRONG is wrong because the $dogs->[0] is set to a separate copy of
the 'Fido' string, and no longer is associated with $kennel->{First}.
Ditto for 'Wags'.

I'm not sure how to fix this... you really need to dump $dogs before
$kennel here so you can take a reference to the array element in
place. Or it has to be patched up afterward. I'm turning Purity on,
which is usually pretty good about constructing the patchups. But it
wasn't sufficient here.

--
Alexandr Ciornii, http​://chorny.net

From [Unknown Contact. See original ticket]

Similar results on 5.12.0.

On Wed Jul 04 06​:11​:31 2001, merlyn@​stonehenge.com wrote​:

use Data​::Dumper;
my @​dogs = \( 'Fido'\, 'Wags' \);
my %kennel = \(
  First => \\$dogs\[0\]\,
  Second =>  \\$dogs\[1\]\,
\);
$dogs\[2\] = \\%kennel;
my $mutts = \\%kennel;

print "CORRECT​:\\n";
print Data​::Dumper\->new\(
  \[\\@​dogs\, \\%kennel\, $mutts\]\,
  \[qw\($dogs $kennel $mutts\)\]\)\->Purity\(1\)\->Dump;

print "WRONG​:\\n";
print Data​::Dumper\->new\(
  \[\\%kennel\, \\@​dogs\, $mutts\]\,
  \[qw\($kennel $dogs $mutts\)\]\)\->Purity\(1\)\->Dump;

Output​:

CORRECT​:
$dogs = \[
          'Fido'\,
          'Wags'\,
          \{
            'First' => ''\,
            'Second' => ''
          \}
        \];
$dogs\->\[2\]\{'First'\} = \\$dogs\->\[0\];
$dogs\->\[2\]\{'Second'\} = \\$dogs\->\[1\];
$kennel = $dogs\->\[2\];
$mutts = $dogs\->\[2\];
WRONG​:
$kennel = \{
            'First' => \\'Fido'\,
            'Second' => \\'Wags'
          \};
$dogs = \[
          $\{$kennel\->\{'First'\}\}\,
          $\{$kennel\->\{'Second'\}\}\,
          \{\}
        \];
$dogs\->\[2\] = $kennel;
$mutts = $kennel;

The WRONG is wrong because the $dogs->[0] is set to a separate copy of
the 'Fido' string, and no longer is associated with $kennel->{First}.
Ditto for 'Wags'.

I'm not sure how to fix this... you really need to dump $dogs before
$kennel here so you can take a reference to the array element in
place. Or it has to be patched up afterward. I'm turning Purity on,
which is usually pretty good about constructing the patchups. But it
wasn't sufficient here.

--
Alexandr Ciornii, http​://chorny.net