diff options
author | Nicholas Clark <nick@ccl4.org> | 2009-09-26 17:34:09 +0100 |
---|---|---|
committer | Nicholas Clark <nick@ccl4.org> | 2009-09-26 17:51:18 +0100 |
commit | 325914f9861ab5df5b7d9511c4578256da4059d6 (patch) | |
tree | 40a12e9e76de26bded5dbe8081450a5faa46be2a /cpan | |
parent | d1f770803c95df0ac5e9e873daa3bb453e913fda (diff) | |
download | perl-325914f9861ab5df5b7d9511c4578256da4059d6.tar.gz |
Move Tie::RefHash from ext/ to cpan/
Diffstat (limited to 'cpan')
-rw-r--r-- | cpan/Tie-RefHash/lib/Tie/RefHash.pm | 274 | ||||
-rw-r--r-- | cpan/Tie-RefHash/t/rebless.t | 36 | ||||
-rw-r--r-- | cpan/Tie-RefHash/t/refhash.t | 331 | ||||
-rw-r--r-- | cpan/Tie-RefHash/t/storable.t | 63 | ||||
-rw-r--r-- | cpan/Tie-RefHash/t/threaded.t | 77 |
5 files changed, 781 insertions, 0 deletions
diff --git a/cpan/Tie-RefHash/lib/Tie/RefHash.pm b/cpan/Tie-RefHash/lib/Tie/RefHash.pm new file mode 100644 index 0000000000..f95bf41efd --- /dev/null +++ b/cpan/Tie-RefHash/lib/Tie/RefHash.pm @@ -0,0 +1,274 @@ +package Tie::RefHash; + +use vars qw/$VERSION/; + +$VERSION = "1.38"; + +use 5.005; + +=head1 NAME + +Tie::RefHash - use references as hash keys + +=head1 SYNOPSIS + + require 5.004; + use Tie::RefHash; + tie HASHVARIABLE, 'Tie::RefHash', LIST; + tie HASHVARIABLE, 'Tie::RefHash::Nestable', LIST; + + untie HASHVARIABLE; + +=head1 DESCRIPTION + +This module provides the ability to use references as hash keys if you +first C<tie> the hash variable to this module. Normally, only the +keys of the tied hash itself are preserved as references; to use +references as keys in hashes-of-hashes, use Tie::RefHash::Nestable, +included as part of Tie::RefHash. + +It is implemented using the standard perl TIEHASH interface. Please +see the C<tie> entry in perlfunc(1) and perltie(1) for more information. + +The Nestable version works by looking for hash references being stored +and converting them to tied hashes so that they too can have +references as keys. This will happen without warning whenever you +store a reference to one of your own hashes in the tied hash. + +=head1 EXAMPLE + + use Tie::RefHash; + tie %h, 'Tie::RefHash'; + $a = []; + $b = {}; + $c = \*main; + $d = \"gunk"; + $e = sub { 'foo' }; + %h = ($a => 1, $b => 2, $c => 3, $d => 4, $e => 5); + $a->[0] = 'foo'; + $b->{foo} = 'bar'; + for (keys %h) { + print ref($_), "\n"; + } + + tie %h, 'Tie::RefHash::Nestable'; + $h{$a}->{$b} = 1; + for (keys %h, keys %{$h{$a}}) { + print ref($_), "\n"; + } + +=head1 THREAD SUPPORT + +L<Tie::RefHash> fully supports threading using the C<CLONE> method. + +=head1 STORABLE SUPPORT + +L<Storable> hooks are provided for semantically correct serialization and +cloning of tied refhashes. + +=head1 RELIC SUPPORT + +This version of Tie::RefHash seems to no longer work with 5.004. This has not +been throughly investigated. Patches welcome ;-) + +=head1 MAINTAINER + +Yuval Kogman E<lt>nothingmuch@woobling.orgE<gt> + +=head1 AUTHOR + +Gurusamy Sarathy gsar@activestate.com + +'Nestable' by Ed Avis ed@membled.com + +=head1 SEE ALSO + +perl(1), perlfunc(1), perltie(1) + +=cut + +use Tie::Hash; +use vars '@ISA'; +@ISA = qw(Tie::Hash); +use strict; +use Carp qw/croak/; + +BEGIN { + local $@; + # determine whether we need to take care of threads + use Config (); + my $usethreads = $Config::Config{usethreads}; # && exists $INC{"threads.pm"} + *_HAS_THREADS = $usethreads ? sub () { 1 } : sub () { 0 }; + *_HAS_SCALAR_UTIL = eval { require Scalar::Util; 1 } ? sub () { 1 } : sub () { 0 }; + *_HAS_WEAKEN = defined(&Scalar::Util::weaken) ? sub () { 1 } : sub () { 0 }; +} + +BEGIN { + # create a refaddr function + + local $@; + + if ( _HAS_SCALAR_UTIL ) { + Scalar::Util->import("refaddr"); + } else { + require overload; + + *refaddr = sub { + if ( overload::StrVal($_[0]) =~ /\( 0x ([a-zA-Z0-9]+) \)$/x) { + return $1; + } else { + die "couldn't parse StrVal: " . overload::StrVal($_[0]); + } + }; + } +} + +my (@thread_object_registry, $count); # used by the CLONE method to rehash the keys after their refaddr changed + +sub TIEHASH { + my $c = shift; + my $s = []; + bless $s, $c; + while (@_) { + $s->STORE(shift, shift); + } + + if (_HAS_THREADS ) { + + if ( _HAS_WEAKEN ) { + # remember the object so that we can rekey it on CLONE + push @thread_object_registry, $s; + # but make this a weak reference, so that there are no leaks + Scalar::Util::weaken( $thread_object_registry[-1] ); + + if ( ++$count > 1000 ) { + # this ensures we don't fill up with a huge array dead weakrefs + @thread_object_registry = grep { defined } @thread_object_registry; + $count = 0; + } + } else { + $count++; # used in the warning + } + } + + return $s; +} + +my $storable_format_version = join("/", __PACKAGE__, "0.01"); + +sub STORABLE_freeze { + my ( $self, $is_cloning ) = @_; + my ( $refs, $reg ) = @$self; + return ( $storable_format_version, [ values %$refs ], $reg ); +} + +sub STORABLE_thaw { + my ( $self, $is_cloning, $version, $refs, $reg ) = @_; + croak "incompatible versions of Tie::RefHash between freeze and thaw" + unless $version eq $storable_format_version; + + @$self = ( {}, $reg ); + $self->_reindex_keys( $refs ); +} + +sub CLONE { + my $pkg = shift; + + if ( $count and not _HAS_WEAKEN ) { + warn "Tie::RefHash is not threadsafe without Scalar::Util::weaken"; + } + + # when the thread has been cloned all the objects need to be updated. + # dead weakrefs are undefined, so we filter them out + @thread_object_registry = grep { defined && do { $_->_reindex_keys; 1 } } @thread_object_registry; + $count = 0; # we just cleaned up +} + +sub _reindex_keys { + my ( $self, $extra_keys ) = @_; + # rehash all the ref keys based on their new StrVal + %{ $self->[0] } = map { refaddr($_->[0]) => $_ } (values(%{ $self->[0] }), @{ $extra_keys || [] }); +} + +sub FETCH { + my($s, $k) = @_; + if (ref $k) { + my $kstr = refaddr($k); + if (defined $s->[0]{$kstr}) { + $s->[0]{$kstr}[1]; + } + else { + undef; + } + } + else { + $s->[1]{$k}; + } +} + +sub STORE { + my($s, $k, $v) = @_; + if (ref $k) { + $s->[0]{refaddr($k)} = [$k, $v]; + } + else { + $s->[1]{$k} = $v; + } + $v; +} + +sub DELETE { + my($s, $k) = @_; + (ref $k) + ? (delete($s->[0]{refaddr($k)}) || [])->[1] + : delete($s->[1]{$k}); +} + +sub EXISTS { + my($s, $k) = @_; + (ref $k) ? exists($s->[0]{refaddr($k)}) : exists($s->[1]{$k}); +} + +sub FIRSTKEY { + my $s = shift; + keys %{$s->[0]}; # reset iterator + keys %{$s->[1]}; # reset iterator + $s->[2] = 0; # flag for iteration, see NEXTKEY + $s->NEXTKEY; +} + +sub NEXTKEY { + my $s = shift; + my ($k, $v); + if (!$s->[2]) { + if (($k, $v) = each %{$s->[0]}) { + return $v->[0]; + } + else { + $s->[2] = 1; + } + } + return each %{$s->[1]}; +} + +sub CLEAR { + my $s = shift; + $s->[2] = 0; + %{$s->[0]} = (); + %{$s->[1]} = (); +} + +package Tie::RefHash::Nestable; +use vars '@ISA'; +@ISA = 'Tie::RefHash'; + +sub STORE { + my($s, $k, $v) = @_; + if (ref($v) eq 'HASH' and not tied %$v) { + my @elems = %$v; + tie %$v, ref($s), @elems; + } + $s->SUPER::STORE($k, $v); +} + +1; diff --git a/cpan/Tie-RefHash/t/rebless.t b/cpan/Tie-RefHash/t/rebless.t new file mode 100644 index 0000000000..4ae40f43fe --- /dev/null +++ b/cpan/Tie-RefHash/t/rebless.t @@ -0,0 +1,36 @@ +#!/usr/bin/perl -T -w + +BEGIN { + if( $ENV{PERL_CORE} ) { + chdir 't'; + @INC = '../lib'; + } +} + +use strict; + +use Tie::RefHash; + +{ + package Moose; + sub new { bless { }, shift }; + + package Elk; + use vars qw/@ISA/; + @ISA = "Moose"; +} + +$\ = "\n"; +print "1..2"; + +my $obj = Moose->new; + +tie my %hash, "Tie::RefHash"; + +$hash{$obj} = "magic"; + +print ( ( $hash{$obj} eq "magic" ) ? "" : "not ", "ok - keyed before rebless" ); + +bless $obj, "Elk"; + +print ( ( $hash{$obj} eq "magic" ) ? "" : "not ", "ok - still the same"); diff --git a/cpan/Tie-RefHash/t/refhash.t b/cpan/Tie-RefHash/t/refhash.t new file mode 100644 index 0000000000..d19f7d3593 --- /dev/null +++ b/cpan/Tie-RefHash/t/refhash.t @@ -0,0 +1,331 @@ +#!/usr/bin/perl -T -w +# +# Basic test suite for Tie::RefHash and Tie::RefHash::Nestable. +# +# The testing is in two parts: first, run lots of tests on both a tied +# hash and an ordinary un-tied hash, and check they give the same +# answer. Then there are tests for those cases where the tied hashes +# should behave differently to normal hashes, that is, when using +# references as keys. +# + +BEGIN { + if( $ENV{PERL_CORE} ) { + chdir 't'; + @INC = '../lib'; + } +} + +BEGIN { + unless ( eval { require Data::Dumper; 1 } ) { + print "1..0 # Skip -- Data::Dumper is not available\n"; + exit 0; + } +} + +use strict; +use Tie::RefHash; +use Data::Dumper; +my $numtests = 39; +my $currtest = 1; +print "1..$numtests\n"; + +my $ref = []; my $ref1 = []; + +package Boustrophedon; # A class with overloaded "". +sub new { my ($c, $s) = @_; bless \$s, $c } +use overload '""' => sub { ${$_[0]} . reverse ${$_[0]} }; +package main; +my $ox = Boustrophedon->new("foobar"); + +# Test standard hash functionality, by performing the same operations +# on a tied hash and on a normal hash, and checking that the results +# are the same. This does of course assume that Perl hashes are not +# buggy :-) +# +my @tests = standard_hash_tests(); + +my @ordinary_results = runtests(\@tests, undef); +foreach my $class ('Tie::RefHash', 'Tie::RefHash::Nestable') { + my @tied_results = runtests(\@tests, $class); + my $all_ok = 1; + + die if @ordinary_results != @tied_results; + foreach my $i (0 .. $#ordinary_results) { + my ($or, $ow, $oe) = @{$ordinary_results[$i]}; + my ($tr, $tw, $te) = @{$tied_results[$i]}; + + my $ok = 1; + local $^W = 0; + $ok = 0 if (defined($or) != defined($tr)) or ($or ne $tr); + $ok = 0 if (defined($ow) != defined($tw)) or ($ow ne $tw); + $ok = 0 if (defined($oe) != defined($te)) or ($oe ne $te); + + if (not $ok) { + print STDERR + "failed for $class: $tests[$i]\n", + "ordinary hash gave:\n", + defined $or ? "\tresult: $or\n" : "\tundef result\n", + defined $ow ? "\twarning: $ow\n" : "\tno warning\n", + defined $oe ? "\texception: $oe\n" : "\tno exception\n", + "tied $class hash gave:\n", + defined $tr ? "\tresult: $tr\n" : "\tundef result\n", + defined $tw ? "\twarning: $tw\n" : "\tno warning\n", + defined $te ? "\texception: $te\n" : "\tno exception\n", + "\n"; + $all_ok = 0; + } + } + test($all_ok); +} + +# Now test Tie::RefHash's special powers +my (%h, $h); +$h = eval { tie %h, 'Tie::RefHash' }; +warn $@ if $@; +test(not $@); +test(ref($h) eq 'Tie::RefHash'); +test(defined(tied(%h)) and tied(%h) =~ /^Tie::RefHash/); +$h{$ref} = 'cholet'; +test($h{$ref} eq 'cholet'); +test(exists $h{$ref}); +test((keys %h) == 1); +test(ref((keys %h)[0]) eq 'ARRAY'); +test((keys %h)[0] eq $ref); +test((values %h) == 1); +test((values %h)[0] eq 'cholet'); +my $count = 0; +while (my ($k, $v) = each %h) { + if ($count++ == 0) { + test(ref($k) eq 'ARRAY'); + test($k eq $ref); + } +} +test($count == 1); +delete $h{$ref}; +test(not defined $h{$ref}); +test(not exists($h{$ref})); +test((keys %h) == 0); +test((values %h) == 0); +$h{$ox} = "bellow"; # overloaded "" +test(exists $h{$ox}); +test($h{$ox} eq "bellow"); +test(not exists $h{"foobarraboof"}); +undef $h; +untie %h; + +# And now Tie::RefHash::Nestable's differences from Tie::RefHash. +$h = eval { tie %h, 'Tie::RefHash::Nestable' }; +warn $@ if $@; +test(not $@); +test(ref($h) eq 'Tie::RefHash::Nestable'); +test(defined(tied(%h)) and tied(%h) =~ /^Tie::RefHash::Nestable/); +$h{$ref}->{$ref1} = 'bungo'; +test($h{$ref}->{$ref1} eq 'bungo'); + +# Test that the nested hash is also tied (for current implementation) +test(defined(tied(%{$h{$ref}})) + and tied(%{$h{$ref}}) =~ /^Tie::RefHash::Nestable=/ ); + +test((keys %h) == 1); +test((keys %h)[0] eq $ref); +test((keys %{$h{$ref}}) == 1); +test((keys %{$h{$ref}})[0] eq $ref1); + +{ + # Tests that delete returns the deleted element [perl #32193] + my $ref = \(my $var = "oink"); + tie my %oink, 'Tie::RefHash'; + $oink{$ref} = "ding"; + test($oink{$ref} eq "ding"); + test(delete($oink{$ref}) eq "ding"); +} + +die "expected to run $numtests tests, but ran ", $currtest - 1 + if $currtest - 1 != $numtests; + +@tests = (); +undef $ref; +undef $ref1; + +exit(); + + +# Print 'ok X' if true, 'not ok X' if false +# Uses global $currtest. +# +sub test { + my $t = shift; + print 'not ' if not $t; + print 'ok ', $currtest++, "\n"; +} + + +# Wrapper for Data::Dumper to 'dump' a scalar as an EXPR string. +sub dumped { + my $s = shift; + my $d = Dumper($s); + $d =~ s/^\$VAR1 =\s*//; + $d =~ s/;$//; + chomp $d; + return $d; +} + +# Crudely dump a hash into a canonical string representation (because +# hash keys can appear in any order, Data::Dumper may give different +# strings for the same hash). +# +sub dumph { + my $h = shift; + my $r = ''; + foreach (sort keys %$h) { + $r = dumped($_) . ' => ' . dumped($h->{$_}) . "\n"; + } + return $r; +} + +# Run the tests and give results. +# +# Parameters: reference to list of tests to run +# name of class to use for tied hash, or undef if not tied +# +# Returns: list of [R, W, E] tuples, one for each test. +# R is the return value from running the test, W any warnings it gave, +# and E any exception raised with 'die'. E and W will be tidied up a +# little to remove irrelevant details like line numbers :-) +# +# Will also run a few of its own 'ok N' tests. +# +sub runtests { + my ($tests, $class) = @_; + my @r; + + my (%h, $h); + if (defined $class) { + $h = eval { tie %h, $class }; + warn $@ if $@; + test(not $@); + test(ref($h) eq $class); + test(defined(tied(%h)) and tied(%h) =~ /^\Q$class\E/); + } + + foreach (@$tests) { + my ($result, $warning, $exception); + local $SIG{__WARN__} = sub { $warning .= $_[0] }; + $result = scalar(eval $_); + if ($@) + { + die "$@:$_" unless defined $class; + $exception = $@; + } + + foreach ($warning, $exception) { + next if not defined; + s/ at .+ line \d+\.$//mg; + s/ at .+ line \d+, at .*//mg; + s/ at .+ line \d+, near .*//mg; + s/(uninitialized value)( within)? [\$@%].*? in /$1 in /g; + } + + my (@warnings, %seen); + foreach (split /\n/, $warning) { + push @warnings, $_ unless $seen{$_}++; + } + $warning = join("\n", @warnings); + + push @r, [ $result, $warning, $exception ]; + } + + return @r; +} + + +# Things that should work just the same for an ordinary hash and a +# Tie::RefHash. +# +# Each test is a code string to be eval'd, it should do something with +# %h and give a scalar return value. The global $ref and $ref1 may +# also be used. +# +# One thing we don't test is that the ordering from 'keys', 'values' +# and 'each' is the same. You can't reasonably expect that. +# +sub standard_hash_tests { + my @r; + + # Library of standard tests on keys, values and each + my $STD_TESTS = <<'END' + join $;, sort keys %h; + join $;, sort values %h; + { my ($v, %tmp); $tmp{$v}++ while (defined($v = each %h)); dumph(\%tmp) } + { my ($k, $v, %tmp); $tmp{"$k$;$v"}++ while (($k, $v) = each %h); dumph(\%tmp) } +END + ; + + # Tests on the existence of the element 'foo' + my $FOO_TESTS = <<'END' + defined $h{foo}; + exists $h{foo}; + $h{foo}; +END + ; + + # Test storing and deleting 'foo' + push @r, split /\n/, <<"END" + $STD_TESTS; + $FOO_TESTS; + \$h{foo} = undef; + $STD_TESTS; + $FOO_TESTS; + \$h{foo} = 'hello'; + $STD_TESTS; + $FOO_TESTS; + delete \$h{foo}; + $STD_TESTS; + $FOO_TESTS; +END + ; + + # Test storing and removing under ordinary keys + my @things = ('boink', 0, 1, '', undef); + foreach my $key (map { dumped($_) } @things) { + foreach my $value ((map { dumped($_) } @things), '$ref') { + push @r, split /\n/, <<"END" + \$h{$key} = $value; + $STD_TESTS; + defined \$h{$key}; + exists \$h{$key}; + \$h{$key}; + delete \$h{$key}; + $STD_TESTS; + defined \$h{$key}; + exists \$h{$key}; + \$h{$key}; +END + ; + } + } + + # Test hash slices + my @slicetests; + @slicetests = split /\n/, <<'END' + @h{'b'} = (); + @h{'c'} = ('d'); + @h{'e'} = ('f', 'g'); + @h{'h', 'i'} = (); + @h{'j', 'k'} = ('l'); + @h{'m', 'n'} = ('o', 'p'); + @h{'q', 'r'} = ('s', 't', 'u'); +END + ; + my @aaa = @slicetests; + foreach (@slicetests) { + push @r, $_; + push @r, split(/\n/, $STD_TESTS); + } + + # Test CLEAR + push @r, '%h = ();', split(/\n/, $STD_TESTS); + + return @r; +} diff --git a/cpan/Tie-RefHash/t/storable.t b/cpan/Tie-RefHash/t/storable.t new file mode 100644 index 0000000000..6c28b77a54 --- /dev/null +++ b/cpan/Tie-RefHash/t/storable.t @@ -0,0 +1,63 @@ +#!/usr/bin/perl -T -w + +BEGIN { + if( $ENV{PERL_CORE} ) { + chdir 't'; + @INC = '../lib'; + } +} + +BEGIN { + unless ( eval { require Storable; 1 } ){ + print "1..0 # Skip -- Storable is not available\n"; + exit 0; + } +} + +use strict; + +use Tie::RefHash; + +use Storable qw/dclone nfreeze thaw/; + +$\ = "\n"; +print "1..24"; + +sub ok ($$) { + print ( ( $_[0] ? "" : "not " ), "ok - $_[1]" ); +} + +sub is ($$$) { + print ( ( ( $_[0] eq $_[1] ) ? "" : "not "), "ok - $_[2]" ); +} + +sub isa_ok ($$) { + ok( eval { $_[0]->isa($_[1]) }, "the object isa $_[1]"); +} + +tie my %hash, "Tie::RefHash"; + +my $key = { foo => 1 }; +$hash{$key} = "value"; +$hash{non_ref} = "other"; + +foreach my $clone ( \%hash, dclone(\%hash), thaw(nfreeze(\%hash)) ){ + + ok( tied(%$clone), "copy is tied"); + isa_ok( tied(%$clone), "Tie::RefHash" ); + + my @keys = keys %$clone; + is( scalar(@keys), 2, "one key in clone"); + my $key = ref($keys[0]) ? shift @keys : pop @keys; + my $reg = $keys[0]; + + ok( ref($key), "key is a ref after clone" ); + is( $key->{foo}, 1, "key serialized ok"); + + is( $clone->{$key}, "value", "and is still pointing at the same value" ); + + ok( !ref($reg), "regular key is non ref" ); + is( $clone->{$reg}, "other", "and is also a valid key" ); +} + + diff --git a/cpan/Tie-RefHash/t/threaded.t b/cpan/Tie-RefHash/t/threaded.t new file mode 100644 index 0000000000..7e4fa1a7f0 --- /dev/null +++ b/cpan/Tie-RefHash/t/threaded.t @@ -0,0 +1,77 @@ +#!/usr/bin/perl -T -w + +BEGIN { + if( $ENV{PERL_CORE} ) { + chdir 't'; + @INC = '../lib'; + } +} + +use strict; + + +BEGIN { + # this is sucky because threads.pm has to be loaded before Test::Builder + use Config; + eval { require Scalar::Util }; + + if ( $^O eq 'MSWin32' ) { + print "1..0 # Skip -- this test is generally broken on windows for unknown reasons. If you can help debug this patches would be very welcome.\n"; + exit 0; + } + if ( $Config{usethreads} and !$Config{use5005threads} + and defined(&Scalar::Util::weaken) + and eval { require threads; "threads"->import } + ) { + print "1..14\n"; + } else { + print "1..0 # Skip -- threads aren't enabled in your perl, or Scalar::Util::weaken is missing\n"; + exit 0; + } +} + +use Tie::RefHash; + +$\ = "\n"; +sub ok ($$) { + print ( ( $_[0] ? "" : "not " ), "ok - $_[1]" ); +} + +sub is ($$$) { + print ( ( ( ($_[0]||'') eq ($_[1]||'') ) ? "" : "not "), "ok - $_[2]" ); +} + +tie my %hash, "Tie::RefHash"; + +my $r1 = {}; +my $r2 = []; +my $v1 = "foo"; + +$hash{$r1} = "hash"; +$hash{$r2} = "array"; +$hash{$v1} = "string"; + +is( $hash{$v1}, "string", "fetch by string before clone ($v1)" ); +is( $hash{$r1}, "hash", "fetch by ref before clone ($r1)" ); +is( $hash{$r2}, "array", "fetch by ref before clone ($r2)" ); + +my $th = threads->create(sub { + is( scalar keys %hash, 3, "key count is OK" ); + + ok( exists $hash{$v1}, "string key exists ($v1)" ); + is( $hash{$v1}, "string", "fetch by string" ); + + ok( exists $hash{$r1}, "ref key exists ($r1)" ); + is( $hash{$r1}, "hash", "fetch by ref" ); + + ok( exists $hash{$r2}, "ref key exists ($r2)" ); + is( $hash{$r2}, "array", "fetch by ref" ); + + is( join("\0",sort keys %hash), join("\0",sort $r1, $r2, $v1), "keys are ok" ); +}); + +$th->join; + +is( $hash{$v1}, "string", "fetch by string after clone, orig thread ($v1)" ); +is( $hash{$r1}, "hash", "fetch by ref after clone ($r1)" ); +is( $hash{$r2}, "array", "fetch by ref after clone ($r2)" ); |